BEN-GURION UNIVERSITY OF THE NEGEV

FACULTY OF ENGINEERING SCIENCES

DEPARTMENT OF INDUSTRIAL ENGINEERING & MANAGEMENT

Aspects of Fashion/Trends in Web Design

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR THE M.Sc DEGREE (HUMAN FACTORS ENGINEERING PROGRAM)

By: Gili Korman Golander

Supervised by: Prof. Noam Tractinsky

Advisor: Ilanit Kabessa Cohen

APRIL 2012
 BEN-GURION UNIVERSITY OF THE NEGEV
FACULTY OF ENGINEERING SCIENCES
DEPARTMENT OF INDUSTRIAL ENGINEERING & MANAGEMENT

Aspects of Fashion/Trends in Web Design

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR THE M.Sc DEGREE (HUMAN FACTORS ENGINEERING PROGRAM)

By: Gili Korman Golander

Supervised by: Prof. Noam Tractinsky

Advisor: Ilanit Kabessa Cohen

Author:………………………………….. Date:………………….

Supervisor:………………………………… Date:…………………..

Chairman of Graduate Studies Committee:…………………….. Date:……………

APRIL 2012
 I

Abstract
Fashion is a multidimensional, cultural and social phenomenon characterized by

accelerated changes. Trend is a broader term pertaining to style and design, deriving

from a cultural, social, or technological direction, of which fashion is a manifestation.

People‟s predispositions towards trends vary; there are trendsetters and trend followers

with different degrees of adaptation delays. HCI research and practice, traditionally

dealing with usability, have expanded to user experience, emotions when using

interactive systems, and HCI design aesthetics. These aspects are relevant to the

relatively unexplored domain of fashion\trends in HCI.

This research focused on aspects of fashion/trends in web design. Our main goal

was to establish the premise that trends in web design exist. Therefore, I conducted a

three-part research: compiling a web design trend library, validating it, and testing

hypotheses regarding attitudes towards trends of people from different adopter groups.

First, I collected web design trend data from online trend reviews and academic

publications. I conducted several refinement iterations and identified archetypical

samples. The data includes trend characterization, common elements, dates, enabling

technologies, trend group association and samples, which were synthesized into a web

design trend “library”, supporting the existence of such trends.

Second, I validated the library with a 2-round Delphi study targeting web design

trend experts. In round one, 22 experts rated 25 current trends‟ up-to-dateness and

were asked about connections with broader cultural, social or technological trends.

They were also asked about additional/missing trends and web design trendsetters.

This yielded an initial ranking of the trends‟ up-to-dateness, a list of additional trends
 II

and a list of web design trendsetters. Four bipolar dimensions characterizing web

design trends were extracted. In round two, participants revised their up-to-dateness

rating for the 20 most up-to-date trends, and rated each trend on the four

aforementioned dimensions.

Third, I examined web design trend perception differences between people from

different adopter groups: trendsetters and followers. I presented 26 web design trends

from three trend life-cycle stages: faded, past-peak and current trends. Participants

rated the trends‟ up-to-dateness and how much they liked them. I measured the

participants‟ levels of trendsetting, fashion innovativeness and centrality of visual

product aesthetics. Participants were information systems students (classified a priori

as followers) and designers\design students (classified a priori as trendsetters). I

hypothesized that trendsetters will evaluate current trends as more up-to-date than

past-peak and faded trends. I also postulated that they will like current trends more

than past-peak and faded trends. Our hypothesis regarding trend followers was a

mirror image of the first. Supporting our a priori classification, results indicated that

designers indeed tended to be more trendsetting and to have higher levels of centrality

of visual product aesthetics than non-designers. The trend evaluation results supported

our hypotheses. People like trends they evaluate as current, but trendsetters identify

the more current trends better than trend followers. Different variables affect trend

perception at different trend life-cycle stages, but a design background influences

trend perception in all stages.

Keywords
Design, Fashion, Graphical User Interface, GUI, HCI, Human Computer

Interaction, Trends, UI, User Interface, Web Design

Acknowledgments
First and foremost I offer my sincerest gratitude to my supervisor, Prof. Noam

Tractinsky, who has supported me throughout my thesis with his never ending

patience and vast professional knowledge whilst allowing me the room to explore and

study in my own way. His dedication to this research, profound insights, and

thoughtful guidance were simply invaluable. I would also like to thank my advisor,

Ilanit Kabessa Cohen, for her unique contribution to this research and to my own

education, for her willingness to generously share her time, her design knowledge and

her trend research expertise. I attribute the level of my Masters degree to their

encouragement and effort. One simply could not wish for better or friendlier

supervisors. I consider it a real privilege and a genuine pleasure to have conducted this

research with this wonderful team of instructors.

I would also like to thank Prof. Hava Golander for sharing her time and

profound knowledge of Qualitative research with me. I have much gratitude to Dr.

Liat Pollack Basis who has helped me with the more complex aspects of the statistical

analysis in this research. I am also thankful to my fellow researchers and friends

Eleanor Eitam and Shani Avnet who have shared their knowledge and support with me

during this research. I thank all the participants who have participated in this research.

Last but not least, I would like to thank my beloved family who backed me up

throughout the duration of my Masters degree: my husband Amit, my daughters Yuli

and Libby, my parents Riki and Avner, and my in-laws Hava and Avraham. Thank

you for everything – I couldn‟t have done it without you.

Contents
1. Introduction ______________________________________________ 1
2. Background ______________________________________________ 3
2.1 Fashion _________________________________________________ 3
2.1.1 Defining Fashion __________________________________________________ 3
2.1.2 Perspectives on Fashion _____________________________________________ 5
2.2 Trend _________________________________________________ 9
2.2.1 Defining Trend ____________________________________________________ 9
2.2.2 Perspectives on Trends _____________________________________________ 10
2.3 Fashion and Trend ____________________________________________ 19
2.4 Fashion\Trend Life Cycle ______________________________________ 20
2.5 Why People Desire the Fashionable\Trendy _______________________ 32
2.6 A Close-Up on Trendsetters ____________________________________ 38
3. Research Approach _____________________________________________ 41
4. Web Design Trend Library __________________________________ 43
4.1 Methodology ________________________________________________ 44
4.2 Contents ________________________________________________ 46
5. Study 1 – Delphi Study _____________________________________ 48
5.1 Round 1 ________________________________________________ 48
5.1.1 Overview _______________________________________________________ 48
5.1.2 Method _________________________________________________________ 48
5.1.3 Results _________________________________________________________ 50
5.2 Round 2 ________________________________________________ 52
5.2.1 Overview _______________________________________________________ 52
5.2.2 Method _________________________________________________________ 53
5.2.3 Results _________________________________________________________ 55
6. Study 2 – Main Study ______________________________________ 62
6.1 Method ________________________________________________ 63
6.1.1 Participants ______________________________________________________ 63
6.1.2 Stimuli _________________________________________________________ 64
6.1.3 Procedure _______________________________________________________ 64
6.2 Results ________________________________________________ 66
6.2.1 Background Variables _____________________________________________ 66
6.2.2 Dependent Variables ______________________________________________ 68
6.2.3 Hypothesis Testing ________________________________________________ 69
7. Discussion _____________________________________________ 77
7.1 Study Limitations _____________________________________________ 78
7.2 Conclusions and Recommendations ______________________________ 79
Appendix 1: Web Design Trend Library ___________________________ 82
Appendix 2: Study 1 - Round 1 __________________________________ 85
Appendix 3: Study 1 - Round 2 __________________________________ 93
Appendix 4: Study 2 ____________________________________________ 103
References ____________________________________________ 130
List of Figures

Figure 1: Process model of communication (Shannon and Weaver, 1949) ________ 7

Figure 2: The Trend Spectrum ____________________________________ 10
Figure 3: Web 2.0 Design simplicity in web page (twitter.com Dec 2009) _______ 13
Figure 4: Retro\Vintage complexity in web page (sensisoft.com Dec 2009) ______ 14
Figure 5: Oscillating trends in women's clothing styles in the 20th century (Vejlgaard,
2007) _____________________________________________ 14
Figure 6: Glossy shoes (melissaplasticdreams.com Dec 2009) _______________ 15
Figure 7: Glossy smart phones (apple.com Dec 2009) ____________________ 15
Figure 8: Glossy web page (bacardi.com Dec 2009) _____________________ 15
Figure 9: Clean and minimal shoes (unitednude.com Dec 2009) _____________ 15
Figure 10: Clean and minimal music players (store.apple.com Dec 2009) _______ 16
Figure 11: Clean and minimal web page (google.com Dec 2009) _____________ 16
Figure 12: Messy Desk in web page (ernesthemingwaycollection.com Apr 2010) __ 17
Figure 13: Messy Desk in web page (dannwhittakercreative.com Apr 2010) _____ 17
Figure 14: Adopter categorization on the basis of innovativeness (Rogers, 2003) __ 25
Figure 15: The diamond-shaped trend model (Vejlgaard, 2007) ______________ 27
Figure 16: Fashion life cycles (Sproles and Burns, 1994 in Lynch and Strauss, 2007) 30
Figure 17: The Gartner hype cycle (Linden and Fenn, 2003) ________________ 30
Figure 18: The envelope of acceptable designs within a fashion (Eckert and Stacey,
2001) _____________________________________________ 33
Figure 19: Interest over time in “web 2.0 design”, obtained from “Google Insights for
Search” and limited to “Internet\Web Design and Development” category45
Figure 20: A screenshot of a page in the study – round 1 (“Cartoons\Mascots” current
trend) _____________________________________________ 50
Figure 21: Round 1 up-to-dateness rating of 25 current web design trends, ordered
from highest to lowest _________________________________ 51
Figure 22: A screenshot of a page in the study – round 2 (“Cartoons\Mascots” current
trend) ______________________________________ 54
Figure 23: Round 1 and 2 up-to-dateness rating of 20 current web design trends
(highest to lowest by round 2 scores) ________________________ 55
Figure 24: “Typography | Big and Bold” trend rated the most simple on the Simple-
Complex dimension and the most direct on the Direct-Subtle dimension _ 56
Figure 25: “Ornamental” trend rated the most complex on the Simple-Complex
dimension ______________________________________ 57
Figure 26: “Magazine Layouts” trend rated the most impersonal on the Impersonal-
Personal dimension ____________________________________ 57
Figure 27: “Sketches\Hand Drawn Illustrations” trend rated the most personal on the
Impersonal-Personal dimension ____________________________ 58
Figure 28: “Texture | Grain” trend rated the most subtle on the Direct-Subtle
dimension ______________________________________ 58
Figure 29: “Transparency” trend rated the most modern on the Modern-Nostalgic
dimension ______________________________________ 59
Figure 30: “Retro\Vintage” trend rated the most nostalgic on the Modern-Nostalgic
dimension _____________________________________________ 59
Figure 31: A screenshot of a page in the main study stage (“Huge Images” Current
Trend) _____________________________________________ 65
Figure 32: Background variable differences between designers and non-designers
(statistically significant differences noted by a star) _______________ 68
Figure 33: Trend up-to-dateness perception by designers and non-designers in the
trend life-cycle stages (faded, past-peak, current) ________________ 74
Figure 34: Trend liking by designers and non-designers in the trend life-cycle stages
(faded, past-peak, current) _______________________________ 76
List of Tables
Table 1: Fashion definitions ______________________________________ 4
Table 2: Changes in art styles since the Renaissance (Vejlgaard, 2007) _________ 18
Table 3: Trend group proportions in the population (Vejlgaard, 2007) _________ 27
Table 4: Four web design trends bipolar dimensions extracted from round 1 results 52
Table 5: The four web design trends bipolar dimensions‟ statistics ____________ 56
Table 6: Pearson correlations between estimated duration (years) and up-to-dateness,
simple-complex, impersonal-personal, direct-subtle and modern-nostalgic
dimensions _____________________________________________ 60
Table 7: Descriptive statistics, reliabilities (on the diagonal) and correlations between
the modified background variables ___________________________ 67
Table 8: Independent Samples T-Test analyses summary for background variables
(Statistically significant differences noted in bold) ________________ 67
Table 9: Pearson correlations summary for perceived up-to-dateness and level of
liking (** p < .001; * p < .01) ______________________________ 69
Table 10: Regression analysis for perceived up-to-dateness of faded trends
(statistically significant dependant variables noted in bold) ___________ 70
Table 11: Regression analysis for perceived up-to-dateness of past-peak trends
(statistically significant dependant variables noted in bold) ___________ 71
Table 12: Regression analysis for perceived up-to-dateness of current trends
(statistically significant dependant variables noted in bold) ___________ 71
Table 13: Regression analysis for level of liking of faded trends (statistically
significant dependant variables noted in bold) ___________________ 71
Table 14: Regression analysis for level of liking of past-peak trends (statistically
significant dependant variables noted in bold) ___________________ 72
Table 15: Regression analysis for level of liking of current trends
(statistically significant dependant variables noted in bold) ___________ 72
Table 16: Independent Samples T-Test analyses summary for perceived up-to-dateness
in the trend life-cycle stages (statistically significant differences noted in
bold) _____________________________________________ 74
Table 17: Independent Samples T-Test analyses summary for level of liking in the
trend life-cycle stages (statistically significant differences noted in bold) _ 76
 1

1. Introduction
Fashion is central to modern life: it drives the economy, mediates communication,

influences aesthetic taste, shapes identities, defines individuals and groups, and often

fulfils contrasting human needs and desires. The use of information technology (IT) to

create, manipulate and disseminate designs has been a major contributor to fashion‟s

expansion and popularization and to the accelerated pace of its lifecycle (Vejlgaard,

2007). At the same time, many aspects of HCI have been subjected to fashion-like

processes (Tractinsky, 2006). In recent years HCI research went through a paradigm

shift, transferring focus away from purely cognitive, usability oriented topics towards

broader aspects such as User Experience (Hassenzahl, 2003), Hedonics (Diefenbach

and Hassenzahl, 2008), and Aesthetics (Lavie and Tractinsky, 2004). Consequently,

research on fashion and trends in HCI appears timely and beneficial. Buxton (2007,

p.50) recognized the importance of style and fashion in the design of user experiences:

“Style and fashion are really important. This is obvious to people from consumer

products or haute couture. But it is not so well appreciated in the high-tech sector”.

I propose that fashion\trends have various manifestations in HCI. Within this

general proposition I focus our preliminary efforts on the context of web design. To

the best of our knowledge there are yet no systematic studies of this kind. I believe

that the ground for such research has been laid in recent years, as the user interface

technology has reached a maturity level that enables almost limitless design

opportunities. Yet, the visual language of web design is relatively new and is still

evolving. Thus, we are at an ideal point to start analyzing its grammar and meanings

while setting the ground for future research. Such research would eventually benefit

our understanding of the individual and societal effects of IT.

 2

In this research I set out to examine aspects of fashion/trends in Human Computer

Interface design and formulate a theoretical framework for evaluating them. Within

the field‟s large research space, I have decided to focus on aspects of fashion/trends in

web design. This thesis‟ main goal was to establish the premise that trends in web

design exist. Therefore, I conducted a three-part research: compiling a web design

trend library, validating it, and testing hypotheses regarding attitudes towards trends of

people from different adopter groups: trendsetters and followers.

I begin by exploring the meaning of fashion and of a related concept – trend – by

taking a multidisciplinary perspective that relies on the literature on trends, fashion,

and design research, as well on studies from the fields of sociology, psychology and

economics. I then discuss fashion and trends in web design, elaborate on the current

research framework, evaluate and test it empirically, and suggest potential further

research directions.
 3

2. Background
In this section I define and offer perspectives on fashion and trend, which are the

background concepts of this research, by reviewing the diverse academic literature on

trends, fashion, and design research, as well as on studies from the fields of sociology,

psychology and economics.

2.1 Fashion
“Fashion is not something that exists in dresses only. Fashion is in the sky, in the street;
fashion has to do with ideas, the way we live, what is happening.”
Coco Chanel

2.1.1 Defining Fashion

Fashion is a centuries-old phenomenon: the word fashion is dated back to the 14th

century, deriving from the Latin word facere (to make), and means "a prevailing

custom, usage or style" (Merriam-Webster). Fashion‟s key elements are modernity,

„westernity‟, adoption, change, context\time dependency, communication of social

meanings, and a close affinity to style (Wilson, 1985, Davis, 1992, Barnard, 2007,

Lynch and Strauss, 2007). Barnard (2007) points out that being "in fashion" is inherent

to the meaning of fashion, as in "a prevailing style". Blumer (1969) expands fashion‟s

scope from clothing and bodily adornments to other domains such as arts,

entertainment, medicine, management, politics and even science.

Many scholars have tried to reach an all encompassing definition for fashion; some

of these definitions are summarized in Table 1. We can see that the modernity theme is

echoed in the definitions of Wilson (1985), Davis (1992) and Barnard (2007). The

relation of fashion to Western societies is apparent in Barnard‟s (2007) definition.

Barnard goes further to claim that a classless society with no social structure and no

possibility or desire for upwards mobility has no need for fashion, and therefore
 4

suggests that the existence of fashion in a society is a good test for both its modernity

and 'westernity'. Change is inherent to fashion, as Davis (1992) reveals. Fashion serves

as a means of communication and contains meanings, according to Davis (1992) and

Barnard (2007). These meanings are social and cultural dependant in nature, and may

define people's social identification group (Barnard, 2007). Wilson (1985) points out

the affiliation of fashion to the domains of aesthetics, art, leisure and entertainment.

Lynch and Strauss (2007) stress the role of style at the core of fashion, citing style‟s

definition as a “characteristic or distinctive mode or method of expression,

presentation or conception in the field of some art” (Nystrom, 1928).

Definition Author Year

Fashion is a branch of aesthetics, of the art of modern society. It is Wilson 1985
also a mass pastime, a form of group entertainment, of popular
culture.
Fashion…refer[s] to some alteration in the code of visual Davis 1992
conventions by which we read meanings…into the clothes we and
our contemporaries wear.
Fashion is one of the ways in which people are constructed as Barnard 2007
members (and/or non-members) of cultural groups.
Modern, western, meaningful and communicative bodily Barnard 2007
adornments or dress.
Fashion can be defined as the prevailing style at any given time. Lynch & 2007
Strauss
Table 1: Fashion definitions

Trying to capture the multi-faceted nature of fashion, as described by the above

scholars, I propose that fashion is a modern, western, cultural and social phenomenon

that manifests in new and popular trends and styles, changing over time and serving

as a form of communication for ideas and meanings, thus capturing the spirit of the

times.
 5

2.1.2 Perspectives on Fashion

Researchers have tried to shed light on the phenomenon of fashion from a variety

of social, psychological and technological perspectives (cf. Lynch and Strauss, 2007).

Bellow I discuss some of the major fashion theories, most are dress related, but seem

generalizable to other fashion manifestations, following Blumer‟s (1969)

aforementioned statement.

Fashion as a Function of Capitalism

Marxist theories have claimed that fashion‟s function stems from capitalism‟s need

for perpetual expansion. According to this view, fashion‟s alleged function is to

encourage consumption, because without consumerism capitalism will collapse

(Wilson, 1985). These Marxist theories are criticized as being oversimplified and

over-deterministic: they see people as passively abiding to consumerism with no

autonomy over their lives, they can be extrapolated to diminish other pleasures like

music, movies, or literature as being mere manifests of consumerism, and finally there

is the contrary evidence of menswear which despite having a slower change pace than

womenswear, promotes less consumption and has proven to be a more economically

stable market (Wilson, 1985). Therefore, fashion is probably not merely a function of

capitalism - there is more to it.

Fashion as Art

Some see fashion as a decorative or applied art form, portraying an artistic

expression of the ideas of a period, while at the same time designed to be practical.

Others protest that the functional aspect of fashion negates viewing it as an art form

(Rhodes and Rawsthorn, 2003) or that, in general, people are less tolerant to pluralism

in fashion styles compared to other art forms (Wilson, 1985).

 6

Fashion as Self Expression

A psychological view of fashion as communication of expression sees fashion as

an expression of the self (Campbell, 1997 and Barnard, 2007). Both individuals and

cultural communities can use fashion to express or make „internal‟ ideas and beliefs

externally visible (Barnard, 2007). Campbell (1997) and Barnard (2007) however

object to this view. Barnard (2007) claims that since meaning is a product of

interaction between cultural values and the items of fashion, it cannot be self-

expression, because self-expression assumes that meaning existed prior to the

interaction. Instead, he suggests that through fashion, identity is being constructed and

reproduced. Campbell (1997) comments that fashion cannot be a means of self-

expression and at the same time a shared communication vehicle, remembering

fashion's communicative nature. He also notes fashion‟s additional functional role, in

contrast to music which is purely expressive. Finally, he notes that fashion may

express just a mood, whim or temporary need, not necessarily identity.

Fashion as Communication

Fashion‟s communicative nature has been further acknowledged: it is referred to as

a language (Eco, 1973 and Davis, 1992) and as cultural negotiation of meanings

(Barnard, 2007). Eco‟s (1973) view is that fashion is a language (code), albeit weaker

than linguistics because it is time dependant. Davis (1992) also views fashion as a

language whose key terms are fabric, texture, color, pattern, volume, silhouette and

context. He notes several main characteristics of the fashion code: ambiguity (ever

shifting meanings), high context dependency, high variability in its understanding by

different social groups, undercoding (with no clear interpretation rules, interpretation

is done by cues), being an aesthetic code intended for communicating unformulated

notions, and finally, code changes are not spontaneous (propagated by fashion
 7

stakeholders). Campbell (1997), on the other hand, objects to regarding fashion as a

language. He states that fashion has no grammar, does not allow a dialog and forms no

singular words. He further notes that the essence of what fashion language is

communicating is not clear: it might be status, identity, gender, etc. If the message

type is not clear an observer cannot be expected to decode the message. He points to

the fact the “speaker” and “listener” may not be equally conversant in this language,

and the "speaker” does not know if the message was understood. Finally, he reminds

us that the "message" is not always intended by the “speaker”.

The debate over whether fashion is a language is part of a larger one, concerning

the essence of fashion as a form of communication. Barnard (2007) points to two

common views of communication: sending and receiving of messages (process model)

or cultural negotiation of meanings. The communication process model (Fig. 1) is

rooted in telecommunications and was first presented by Shannon and Weaver (1949).

Barnard objects to framing fashion in this model, stating that the fashion elements'

framing is not clear. For example, is the designer the information source or the

transmitter? Who is the receiver – the fashion item buyer or the onlooker? What is

noise? And so on. Furthermore, it seems as if the receiver may influence the signal,

because designers are aware of market preferences.

Information Transmitter Signal Receiver Destination

Source (encoder) (decoder)

Noise

Figure 1: Process model of communication (Shannon and Weaver, 1949)

Barnard (2007) believes that meaning does not pre-exist the process of

communication, as he supports the view of communication as the cultural negotiation

of meanings. Davis (1992) seconds that view, arguing that the signal (meaning) is not
 8

stable, is often ambiguous and that a single signal may convey multiple meanings.

Campbell (1997) adds that in fashion, there is always a signal, unless the person is

unobserved.

Fashion as a Social and Cultural Phenomenon

Wilson (1985) cites a sociological theory viewing fashion as a manifestation of the

fight for status in Capitalist societies: when feudal life dissolved, fashion became an

arena of continuous social struggle of the individual to rise in status. This formed a

“fashion spiral”, when the rich abandon a fashion as soon as it becomes common with

the bourgeoisie or lower classes. Critiques argue that this theory is incomplete and

dated: it does not explain why a specific fashion replaces another nor does it fit the

contemporary fashion pattern of street fashion influence on high fashion.

Simmel‟s (1904) sociological theory views fashion as a means for externalizing

social stratification in open class societies. According to this theory, the elite‟s fashion

acquires prestige in the eyes of the lower classes which try to emulate it. Simmel‟s

notable sociological observation was that fashion balances the human needs for group

identification and individual differentiation, simultaneously. By following fashion,

social adaptation is satisfied, and at the same time, through the subtleties and changes

of fashion, self-differentiation is satisfied.

Blumer (1969) criticized Simmel's (1904) theory as being outdated. Instead, his

“Trickle-Across“ sociological theory (Blumer, 1969, Lynch and Strauss, 2007) views

fashion as a process of collective selection from competing models, each trying to

capture the most modern ideas, embodying the “spirit of the times” (Zeitgeist). The

choice of products, styles or ideas which becomes fashion is affected by changes in

society‟s taste and concerns.

 9

To summarize, fashion is a multi-faceted phenomenon which was analyzed from

many theoretical perspectives. These perspectives, independently and as a whole, help

us to better understand what fashion is. In this research I shall look at some

manifestations of fashion in the web design domain.

2.2 Trend
“Trend-spotting is a little like the ancient Roman art of divination. You stir the ashes. You
consult the entrails of birds. A pattern emerges, and perhaps even one that contains
unexpected meanings about where the culture is headed.”
Guy Trebay, New York Times

2.2.1 Defining Trend

Fashion is a specific current manifestation of a higher order phenomenon – the

trend. Trend is a more recent concept than fashion - its meaning as "general tendency"

is dated to the 19th century and first usages of the related terms “trend-setter” and

“trendy” are dated to the 20th century (Online Etymology Dictionary).Vejlgaard

(2007) adds that for most of the 20th century trend was used in the context of statistics

and economics, referring to the direction of a curve based on historical data. In the last

third of the 20th century it became common in the fashion industry, focusing on the

prediction of fashion changes. Through the fashion industry it filtered into many other

domains, referring mainly to design, style and taste1.

Vejlgaard (2007) proposes three common meanings for trends: product news (e.g.

new furniture trends), product development (e.g. the trend in new cars from a specific

manufacturer), and process of change (e.g. moving from trendsetters to mainstream).

He believes the meanings are connected: "a trend is a process of change that

(sometimes) comes about because of product development that (sometimes) results in

new products". Trend sociologists are interested in trend prediction. To them a trend is

1
Vejlgaard (2007) defines design as "individual designed objects" (e.g. furniture), style as "a certain
mix of designed objects" (e.g. minimalist style), and taste as "what we like".
 10

defined as something that is going to happen, in a certain way that will be accepted by

the average person. Vejlgaard (2007) defines trend as a noticeable change in a product,

which a person can detect trough the visual, auditory, tactile, taste, or smell sense.

2.2.2 Perspectives on Trends

As Vejlgaard (2007) aptly put it, although trends are a much talked about subject,

we seem to know fairly little about it. Therefore, trend research is still largely a "gray

area" - a young and not yet a fully established discipline, so the available literature

contains some theories and models that have not yet been put under academic scrutiny.

That said, I will present the current prominent voices in this field.

The Trend Spectrum

Trend is located on a spectrum, differentiated by lifespan and influence range as

depicted in Figure 2. The trend spectrum classes are based on Bell (2003), Vejlgaard

(2007), and Penn and Zalesne (2007).

Figure 2: The Trend Spectrum

Vejlgaard (2007) discusses three trend types: fads, trends and megatrends, as the

building blocks of the trend spectrum.

 11

 At the transient end of the spectrum is the Fad, a short term craze for a new

or innovative product. It might gain mainstream adoption (e.g. Crocs shoes)

but its lifespan is short, often less than a year. Fads are sometimes heavily

marketed by stakeholders (e.g. industry, magazines).

 Trend is often social and cultural in nature. It evolves over a long period of

time and its lifespan is often measured in years, so it is usually detectable

only a while after its incubation. Trends signal a direction towards

mainstream adoption.

 Megatrend is a major political or technological shift, affecting a large part

of society (e.g. the Internet). Megatrends have a long lifespan and a lasting

influence on society.

There are a few additional trend classes, not included in Vejlgaard‟s (2007) spectrum.

 Somewhere in the middle of the spectrum lies the term Microtrend - a

small niche trend adopted by a small percent of the population, yet

affecting the larger society (Penn and Zalesne, 2007). For example, post

retirement workers constitute a microtrend, being a growing sector in the

United States, bearing significant implications for society (Penn and

Zalesne, 2007), let alone for UI requirements.

 At the long-term end of the spectrum is Futurology, a research field

attempting to predict future developments in society based on past and

present trends (Bell, 2003).

Trend classes can be related, e.g. a trend may be a manifestation of a megatrend.

For example, the Nostalgia megatrend is apparent in trends such as listening to music
 12

on the move using large “old school” headphones, in the rush for vintage clothes, in a

slew of new movie adaptations of childhood classics such as “Alice in Wonderland” or

“Where the Wild Things are”, and in retro inspired web design (Fig. 4). Within this

spectrum, the current research focuses on the Trend class.

Trend Characteristics

Gladwell (1997) observed professional trend spotters (cool hunters) at work, and

deduced that "you have to be one [cool], to know one [cool]", meaning that trends are

identifiable only by connoisseurs, based on intuition. Conversely, Vejlgaard (2007)

proposes a more structured approach to trend analysis and prediction. He states that

trends share common patterns, occurring repeatedly when new trends emerge, making

them identifiable by suitably trained observers.

To facilitate trend identification, observation and forecasting, Vejlgaard (2007)

suggests some common trend patterns:

 Trends evolve over time, and within that time frame they can be observed

and forecasted.

 Trends are initiated by „trend creators‟, a tiny group of people who invent

new ideas, products or styles, and are first adopted by „trendsetters‟, a

somewhat larger group characterized by extreme openness to change and

innovations in style and taste. Therefore, by observing „trendsetters‟ one

can possibly identify new trend patterns. The more „trendsetters‟ adopt a

product and the more different types of trendsetters adopt a product, the

more it is likely to become a trend.

 Trends usually emerge in major cities with a large concentration of

trendsetters and a strong appeal for trendsetter visitors such as New York,
 13

Tokyo, London, etc. Therefore, it is not surprising that the Silicon Valley,

with its large concentration of technological trendsetters, and close

proximity to the major trendsetting cities of Los Angeles and San

Francisco, is the birth-place for many a technological trend.

 There is continuous product development early in the trend process, in

order to keep a high level of interest in the product by trendsetters. A

prolonged period of trendsetter interest in a product raises the probability of

its adoption by followers.

 New trends are often a reaction to what has become mainstream or has

been in the market for many years. Trendsetters will usually abandon a

style when it becomes too mainstream.

 Style changes often oscillate from one end of a style axis to the other. For

example, the simplicity of Web 2.0 Design style (past-peak trend, Fig. 3)

versus the complexity of Retro\Vintage style (current trend, Fig. 4).

Another example, taken from the fashion industry, is depicted in an

illustration of this pattern in women's fashion in the 20th century (Fig. 5).

Figure 3: Web 2.0 Design simplicity in web page (twitter.com Dec 2009)
 14

Figure 4: Retro\Vintage complexity in web page (sensisoft.com Dec 2009)

Figure 5: Oscillating trends in women's clothing styles in the 20th century (Vejlgaard,
2007)

 Trends often appear in multiple industries (also in Gladwell, 2000). For

example, the glossy trend is concurrently observed in shoes, smart phones,

and web pages (Figs. 6-8) and the clean and minimal trend is concurrently

observed in shoes, music players, and web pages (Figs. 9-11).

 15

Figure 6: Glossy shoes (melissaplasticdreams.com Dec 2009)

Figure 7: Glossy smart phones (apple.com Dec 2009)

Figure 8: Glossy web page (bacardi.com Dec 2009)

Figure 9: Clean and minimal shoes (unitednude.com Dec 2009)

 16

Figure 10: Clean and minimal music players (store.apple.com Dec 2009)

Figure 11: Clean and minimal web page (google.com Dec 2009)

 The product or style can be easily copied, imitated, and manipulated, e.g.

the messy desk trend (Figs. 12-13). Lynch and Strauss (2007) state that

mimetic behavior, i.e. imitation through observing and copying, is largely

responsible for spreading a new style, until it reaches the mass adoption

stage and becomes fashion. Dawkins (1989) notes that some memes are

better replicators than others, and therefore spread more widely in the

population and last longer.

 17

Figure 12: Messy Desk in web page (ernesthemingwaycollection.com Apr 2010)

Figure 13: Messy Desk in web page (dannwhittakercreative.com Apr 2010)

Trend Acceleration and Compression

The pace of trend changes has been accelerating for some time, mostly in the last

century. One factor contributing to the acceleration in the introduction of new trends

may be a rise in individualism (Vejlgaard, 2007). Maslow's Hierarchy of Needs theory

(1970) predicts, as Vejlgaard (2007) notes, that when people reach the top of the

hierarchy of needs, they would strive to fulfill the need for self-actualization - “to

become everything that one is capable of becoming” (Maslow, 1970, p.382). Self-

actualization has, by nature, an individual manifestation for different people.

Therefore, as more low order hierarchy needs are being fulfilled in western societies

during the last century, the more we witness a rise in individualism which, in turn,

may contribute to the acceleration in trends.

Other factors which may have contributed to the acceleration in the introduction of

new trends are globalization and the proliferation of information, as they had
 18

shortened the distance and time of a trend spread phase (Vejlgaard, 2007). A trend that

could take years to reach from cultural centers of the world to its remote corners, could

now be observed within days through the information media, and adopted fairly

quickly following that. In addition, many people are now world travelers to some

extent, and can bring back new ideas to other locations within a short period of time.

In addition to the acceleration in the pace of the introduction of new trends, there

seems to be a visible pattern of temporal compression – shortening of the duration of

trends. Vejlgaard (2007) demonstrates this through the consistent shortening in the

duration of artistic styles from the 14th to the 20th Century (Table 2). Vejlgaard

(2007) notes that after 1972 (the year in which the example data ends), several

different artistic styles co-exist simultaneously.

Table 2: Changes in art styles since the Renaissance (Vejlgaard, 2007)

 19

As mentioned above, much of this acceleration in the introduction of new trends is

due to the increased capabilities and usage of information technology. At the same

time, the increased plasticity of information technologies – especially those related to

the design of interactive systems – has enabled general trends to be increasingly

manifested in UI design, and more specifically in web design, as demonstrated in

figures 3,4,8 and 11-13 above.

2.3 Fashion and Trend

The term fashion seems very close to the term trend. Earlier discussion elaborated

on both fashion and trend as a constant process of change, in which novel ideas which

represent the spirit of the times, i.e. the Zeitgeist (which may manifest in dress,

products, style or messages), emerge, become adopted and spread through the

population through human agents such as the group referred to as Trendsetters before

gaining mainstream acceptance. Both fashion and trend are representations of a major

social, cultural phenomenon that has a strong economic impact, which has undergone

a major acceleration and compression in the 20-21st centuries.

Fashion seems to be a more established field of research, while trend research is

still an emerging field. In addition, fashion and trend research focus on different

aspects of the phenomenon, although some aspects overlap, such as the role of

novelty, adoption groups, etc.

It is quite challenging to accurately define and differentiate the concepts of fashion

and trend, but nevertheless I would like to suggest the following perspective. A trend

represents the broader inclination, the larger movement and lasts for a considerable

(but varied) duration from incubation to fading. Fashion, on the other hand, is a
 20

temporary manifestation of a trend, a detailed incarnation of the trend‟s core message,

what is fashionable at present.

This research pertains mainly to visual elements in user interfaces. For example,

one web design fashion\trend I identified in this research, which was prevalent a few

years ago, is Web 2.0 Design (see section 4). It is a past-peak visual trend which has

already reached the mainstream. The online sources discussing Web 2.0 Design which

were reviewed in this research used both terms: fashion and trend. Therefore, I believe

that in this scope the terms fashion and trend can be used interchangeably.

2.4 Fashion\Trend Life Cycle

Fashion\Trends as Social Epidemics

Fashion and trend life-cycle research usually deals with topics such as emergence

and adoption processes. Gladwell (2000) has researched the emergence of

fashion\trends in various domains, whether in dress, crime waves or books becoming

bestsellers. He sees these phenomena as social epidemics, proclaiming that ideas,

products, styles, messages and behaviors spread like viruses. Fashion\trends have a

basic pattern, he asserts, based on three characteristics: they are contagious, little

changes can have big effects, and the change happens suddenly rather than gradually

(i.e. the tipping point).

The contagiousness of ideas or messages, which Gladwell (2000) called the

Stickiness Factor, was researched by Heath and Heath (2007), trying to better

understand why some ideas catch on better than others. They found six dimensions

that make ideas more "sticky" (being memorable, promote spreading, motivate to act

upon) than others: simplicity, unexpectedness, concreteness, credibility, emotions and

stories. Simplicity is about focusing on the message core. Unexpectedness pertains to

 21

creating interest and catching an audience's attention, using aids such as surprise factor

and deliberately creating information gaps. Concreteness means depicting ideas in

sensually perceivable metaphors, such as visual. Credibility is achieved by guiding

people to project an idea unto their own personal experience, rather than using

statistics or a source of authority. Emotions (or at least the appropriate type of

emotion) assist in creating an interest in an idea. Stories help people act upon ideas by

creating mental schemas.

Gladwell (2000) states that epidemics grow in geometric progression, which

explains why sometimes small causal changes, may result in big effects. This

phenomenon, he asserts, is evident in many fields other than epidemiology.

Furthermore, based on research in psychology (Haney et al., 1973, Hartshorne and

May, 1971, in Gladwell, 2000) human behavior is sensitive to the time, place,

circumstances and conditions. So, whether a certain idea or a product will become a

trend is largely context dependant. Gladwell names this factor the Power of Context.

The Tipping Point, or the moment in time where a certain idea, behavior or product

becomes a ubiquitous trend, arrives suddenly rather than gradually, according to

Gladwell (2000). It must reach a critical mass, a boiling point, a threshold, before

catching on with the masses. Gladwell argues that tipping points exist in all epidemics,

including technological ones. As examples, Gladwell mentions fax machines which

were first sold in 1984 but became ubiquitous only in 1987 when enough people had

them, or cell phones, which were getting smaller and cheaper throughout the 1990s,

but became a personal must only in 1998.

 22

Models of Adoption and Diffusion

Several theories suggest how ideas, products, or innovations catch on and become

ubiquitous within the population. One of the first theories, by the sociologist Simmel

(1904), presented a Trickle-Down process. Simmel argued that fashions are made for,

and first adopted by, the elite and then the lesser in status begin to emulate it, until it

becomes too common and the elite abandons it to adopt a new fashion. As mentioned

in the previous discussion about fashion, this theory is criticized as outdated due to the

loosening of social class hierarchy in modern society and to contrary evidence

pointing to fashion emanating from all layers of society.

Blumer (1969), also cited before in the fashion discussion, saw fashion as a

process of collective selection from competing models or ideas. The items that are

selected in the process and become widely adopted in society are those that best

represent the “spirit of the times” (Zeitgeist). Lynch and Strauss (2007) name

Blumer‟s theory (1969) Trickle-Across, as fashion is said to develop among the broad

social sphere as a convergence of collective taste, and the elite are merely early

responders to it. Therefore this theory sees the elite‟s role in the fashion process as

messengers or mimetic agents rather than initiators.

Blumer‟s (1969) perspective of fashion as a continuous process of selection out of

competing models preceded the later construct of meme (Dawkins, 1989). Dawkins

defined meme as “a unit of cultural transmission, or unit of imitation” and argued that

memes (such as tunes, ideas, catch-phrases, dress fashion, etc.) propagate through the

population through imitation, and are the building block of an evolutionary process

akin to that of genes. However, the process of meme propagation is unique because the

memes are not simply imitated, but are often altered (mutating, blending) through it.

Lynch and Strauss (2007) state that Mimetic Behavior, i.e. imitation through observing
 23

and copying, is largely responsible for spreading a new style until it reaches mass

adoption stage and becomes fashion. Dawkins (1989) notes that some memes are

better replicators than others; they spread more widely in the population and last

longer. Thus, fashion\trends can be thought of as a mechanism in an evolutionary

process for ideas, manifested in products and styles.

A later theory, by Suzuki and Best (2003), presented a Bubble-Up process, in

which sub-culture groups introduce new fashions and ideas into mainstream. These

researchers cite Rogers' (1995 [2003]) diffusion theory, classifying different adoption

groups at different stages of a fashion's life-cycle. Rogers classifies a class of fashion

adopters called "early adopters", which are not necessarily of high status but have

access to innovation information, sufficient resources to risk experimentation, and a

willingness to break with tradition.

Diffusion Theory (Rogers, 2003) states that all people are part of the fashion\trend

process. According to this theory, the general population is comprised of distinct

groups of people, characterized by different attitudes towards innovation.

Innovativeness is defined as the “degree to which an individual or other unit of

adoption is relatively earlier in adopting new ideas than other members of the [social]

system” (Rogers, 2003, p. 22). An innovation may be adopted only consecutively, by

one group after the other and in a very specific order (Rogers, 2003, Gladwell, 1997,

Gladwell, 2000, Vejlgaard, 2007). These groups are known as "adopter categories"

(Roger, 2003) referring to innovation diffusion in general or "trend groups" referring

more specifically to trend diffusion (Vejlgaard, 2007).

 24

In the Adopter Model framework, Rogers (2003) proposed the widely used five

group adopter classification. This classification is based on the operationalization of

innovativeness as the time of adoption. The five adopter groups are thus characterized:

 Innovators - a very small group of people who are more adventurous,

venturesome and open to new ideas than the rest of the population. This

group is not necessarily at the consensus of the social system and may not

have the respect of other members in it, but its members import new ideas

to the system from outside its boundaries.

 Early Adopters - a larger group of people with a high rate of opinion

leaders. These people embrace change and are influenced by the

Innovators. Contrary to the Innovators, they usually enjoy the respect of

their peers. Therefore, they play a special and important role in the trend

adoption process: they are the Trendsetters. When Early Adopters choose

to adopt a certain idea, product or behavior, the other adopter groups are

more likely in turn to adopt it.

 Early Majority - a group of people, influenced by Early Adopters or

Trendsetters. They tend to deliberate for some time before adopting a new

idea, and are likely to adopt it just before the average member of the

system.

 Late Majority - a more skeptical group of people, who would wait until an

innovative idea was adopted by the Early Majority before adopting it,

therefore are likely to adopt it just after the average member of the system.

The Early Majority and Late Majority groups comprise the better part of

the population.
 25

 Laggards - a small group of very traditional people, last to adopt an

innovation.

This classification lacks a non-adopter class, referring to people who would not

adopt an innovation at all (e.g. the Amish in the United States) and therefore it is not

exhaustive. Rogers (2003) acknowledges this difficultly and suggests overcoming it by

combining a series of innovations into a composite innovativeness scale. The

innovation creators or originators are not included in this classification either.

Rogers (2003) demonstrated that adopter distribution in the general population

approaches normality when plotted over time on a frequency basis (Fig. 14). Adopter

group proportions are not symmetrical across the curve axis and their influence in the

diffusion process is disproportional to their size. Early Adopters, who comprise only

13.5% of the population, are said to be largely responsible for adoption decisions of

the other adopter groups.

Figure 14: Adopter categorization on the basis of innovativeness (Rogers, 2003)

Vejlgaard (2007) proposes a more granular and exhaustive classification,

comprised of eight adopter groups, whose influence sequence and proportional sizes

are depicted in the diamond-shaped trend model (Fig. 15). The trend groups vary on

their openness to change (in style, ideas, behavior). Furthermore, a specific trend does

not necessarily pertain to the entire population.

 26

 Trend Creators - a tiny group of people who invent new ideas or products

or styles. Vejlgaard (2007) notes that Trend Creators are not identical to

Rogers' (2003) Innovators, because Innovators do not create trends, they

simply are the first to adopt a trend, imported from outside the system.

 Trendsetters - a group of people characterized by extreme openness to

change and innovations in style and taste, who will adopt new trends first.

They feel positively about change when it takes place at regular intervals.

The Trend Followers, Early Mainstreamers, Mainstreamers and Late

Mainstreamers are a more refined classification of Rogers‟ (2003) Early

Majority and Late Majority groups.

 Trend Followers - similar to Trendsetters in their openness to change, but

need to see a few trendsetters adopting an innovation before attempting it.

 Early Mainstreamers - accept new ideas just before the majority does and it

becomes completely mainstream. They need to see a number of people

adopting an idea before trying it on their own.

 Mainstreamers - average in their acceptance of innovation, mimicking the

Early Mainstreamers.

 Late Mainstreamers - very hesitant of changes but eventually accept them.

 Conservatives - do not like big changes and will adopt something new only

when they cannot obtain their old preference in the market anymore.

 Anti-Innovators - a very small group of people who do not accept changes

at all (e.g. the Amish in the United States). They do not adopt trends, but

are included in the model for completeness.

 27

Figure 15: The diamond-shaped trend model (Vejlgaard, 2007)

Vejlgaard‟s (2007) assessment of trend group proportion in the population, based

on a survey, is presented here (Table 3).

Trend Group Percentage in Population

1 Trend Creators 1%
2 Trend Setters 5%
3 Trend Followers 10%
4 Early Mainstreamers 20%
5 Mainstreamers 40%
6 Late Mainstreamers 15%
7 Conservatives 10%
8 Anti-Innovators 1%
Table 3: Trend group proportions in the population (Vejlgaard, 2007)
 28

Vejlgaard (2007) proposes that the trend process and the diffusion process studied

by Rogers (2003) are different phenomena. He sees trend as a change in style that can

be sensed and observed. Changes in abstract entities such as ideas, knowledge and

technology, which were the focus of Rogers's research, may spread differently from

visual trends. He proposes that Rogers' model is somewhat outdated in relation to

current trend patterns, for several reasons. First, Rogers's adopter categories have a

socio-economic dimension, where earlier adopters like Innovators and Early Adopters

are described as more cosmopolitan, well traveled, socially active, wealthy, etc. while

later adopters such as the Late Majority and Laggards are described as less wealthy

and socially connected. According to Vejlgaard "trends emerge from all strata of

society"; Trend Creators can be poor or come from a sub-culture and Trendsetters and

Trend Followers may have average socio-economic status but live or work in

trendsetting locations. Second, Trendsetters are said to adopt new styles with some

regularity, which is not the case with Innovators. Third, since the 1960s when

diffusion research was held, a media “explosion” exposed a major part of the

population to design, style and new product information. In the 1960s there was much

less such media coverage so the role of opinion leaders, spreading this knowledge in

small communities, was much more prominent. Fourth, Rogers' research has focused

on small homogenous communities, while trends often involve a large heterogeneous

population. Fifth, the diffusion process takes place in a geographically limited area,

while trends do not have such limitation. Last, diffusion relies on face-to-face

communication and people knowing one another, while mere observation suffices for

the trend process.

Regardless, both Rogers (2003) and Vejlgaard (2007) propose diffusion models

that are similar in terms of classifying the population to different groups by their
 29

attitude towards innovation adoption, highlighting the group of Trendsetters and their

special role in the innovation diffusion process and pointing to its sequential nature.

Short and Long Term Patterns

Lynch and Strauss (2007) quote Sproles (1981) who classified fashion change

patterns into short and long term fashion cycles. A short term fashion cycle lasts along

the life time of a product, usually a few months to a year or two. A long term fashion

cycle follows an evolutionary style movement and can last even a century. Studying

these patterns enriches our understanding of the fashion\trend phenomenon.

One of the most influential short term fashion cycle theories is Rogers‟ (2003)

diffusion theory described earlier, focusing on people‟s receptiveness to innovation

and the rate at which they adopt it. Complementing it, Lynch and Strauss (2007) cite

Nystrom‟s (1928) work, studying the cyclical spread of fashion throughout a

population by examining style differences. Nystrom (1928) suggested that short term

fashion cycles can be better understood by plotting fashion acceptance

(operationalized by number of adopters) as a function of time, from fashion initiation

to abandonment and proposed three fashion patterns: Fad, Fashion and Classic.

Nystrom‟s (1928) fashion life-cycle patterns were illustrated by Sproles and Burns

(1994) as presented in Figure 16. The fashion bell curves are different in their slope

steepness (at both ends of the curve) representing acceptance and abandonment rates,

in their height representing peak popularity, and in their width representing the

duration of a fashion. Fad curves are characterized by a sharp rise and fall, a short

lifespan, and not gaining mass popularity. Classic curves are characterized by mass

popularity, a wide central part at the peak of adoption (representing prolonged peak

duration), and a very slow rate of abandonment. Nystrom (1928) believed that the

primary motivation that drove fashion diffusion across a population was imitation, but
 30

that adoption rate is affected by relative prosperity, availability of leisure time,

consumer education levels and technical innovativeness.

Figure 16: Fashion life cycles (Sproles and Burns, 1994 in Lynch and Strauss, 2007)

Since 1995, technology life-cycles have been portrayed by the Gartner Hype Cycle

(Linden and Fenn, 2003). The Gartner Hype Cycle is a graphic representation of the

maturity and hype (visibility) of technologies and applications (Fig. 17). It depicts

early stages in a technology's life-cycle, with the hype peaking when there is little

adoption and the technology is immature.

Figure 17: The Gartner hype cycle (Linden and Fenn, 2003)
 31

Lynch and Strauss (2007) cite several researches in the field of long term fashion

cycle. Kroeber (1919), an anthropologist who studied dress as a manifestation of

cyclic developments in history, uncovered a pattern of steady change over time in the

dimensions of dress beyond short-term fluctuations. The dress dimensions, as a facet

of style, were moving from one extreme to the other over a very long period of time

(often well beyond a century), like a pendulum. Kroeber named the duration of time it

took a style to move from one extreme to another periodicity. He hypothesized that

fashion change is the product of some higher-order complex social force, such as

unsettling due to political conflict or social upheaval. The idea of a long term

pendulum pattern in dress fashion is reminiscent of Vejlgaard‟s (2007) style

oscillations, depicted in Figure 5.

Robinson (1975) continued this line of research by expanding it beyond the field of

dress. He emphasized the end points of a style‟s “pendulum swing” as points of excess

or design extremes, beyond which fashion tends to trend back in the opposite direction

towards a central equilibrium point. In addition, he speculated that the primary drive

for fashion change was the psychological need for novelty.

Lowe and Lowe (1982) used multivariate statistics to reexamine Kroeber‟s (1919)

and Robinson‟s (1975) work. Their statistical analysis confirmed that the long term

periodicities found by Kroeber (1919) were mostly statistically significant, although

some random variations were found in those wave-like patterns. Therefore, they

concluded that long term fashion cycles include deterministic as well as stochastic

components, allowing some degree of predication, which deteriorates as the prediction

period extends further into the future. Lowe and Lowe (1982) isolated the following
 32

deterministic and stochastic factors affecting long term fashion cycles. The

deterministic factors affecting long term fashion cycles were:

 Inertia - responsible for pushing style changes in a slow and continuous

direction.

 Cultural Continuity - a countering force which resists and slows down

change.

 Rule of Esthetic Proportions - biasing the style change towards the

mainstream‟s style expectations.

The stochastic factors included Individual Innovation and Initiatives as well as

other socio-cultural influential events.

2.5 Why People Desire the Fashionable\Trendy

Academic literature provides several explanations for people‟s desire and search

for the fashionable\trendy. I shall review the prominent theories accounting for this

behavior.

Novelty and Familiarity

Fashion is a change in the code of visual convention (Davis, 1992) and trends are

observable changes in products, which will be adopted by trendsetters and later

accepted by the mainstream (Vejlgaard, 2007). Therefore novelty is at the core of all

that is fashionable\trendy. Fashion\trends answer a human hardwired need for novelty

– we are consistently motivated by it, attracted to it, adopt it, and eventually move on

to something newer after a brief period of stability (Lynch and Strauss, 2007).
 33

However, to become fashionable, a new product must also contain a degree of

familiarity, as fashion is often related to its immediate predecessor either by contrast

or by extension (Blumer, 1969). Therefore, the fashionable\trendy is often

characterized by a mix of novelty and familiarity (Bianchi, 2002, Hekkert et al., 2003).

This desired mixture is epitomized by the MAYA (Most Advanced Yet Acceptable)

design principal, which was first formed by the renowned industrial designer

Raymond Loewy (1951) and researched by Hekkert et al. (2003). This is echoed in

Eckert and Stacey‟s (2001) view that outdated or avant-garde designs will not be

fashionable at a given time (Fig. 18).

Figure 18: The envelope of acceptable designs within a fashion (Eckert and Stacey, 2001)

A key observation is that when the novelty facet of a product is socially shared (as

in fashion\trends) it has a potential to become eroded over time, as more people adopt

it. Novelty is especially important to Trendsetters, who actively seek it and are quick

to adopt as well as forgo a product for more novel ones, if no product innovations are

introduced. Therefore, in order to maintain the aspect of novelty fashion always

changes and new trends emerge.

Bianchi (2002) provides cognitive and economic reasoning to people's novelty

seeking. In cognitive terms, novelty makes us change our goods classification and
 34

recognition rules, which in turn, open up opportunities for new uses and

interconnections with other goods (e.g. Swatch watches). In economic terms, faced

with novelty in a product, we obtain a new potential for gains and losses that is

independent of its price or our income, which influences its perceived subjective

value. The source of novelty in a product often lies less in its utilitarian attributes and

more in its design attributes and connections with other products.

Identification and Differentiation

Fashion allows people to differentiate from the masses, while at the same time be

identified and accepted as part of social group(s) (Simmel, 1957, Bianchi, 2002, Lynch

and Strauss, 2007). Therefore, fashion eases the inherent tension between one‟s wish

to preserve individuality and a contradictory need to signal group membership (Lynch

and Strauss, 2007).

Recent studies support that observation. Barnard (2007) suggests that fashion is

one of the ways people are constructed as members (and/or non-members) of cultural

groups. Liu and Donath (2006), on the other hand, view fashion as a social signal. This

social signal has a cost (in terms of time, money, reputation etc.), it changes over time,

and its quality is implied by the scope of access to information of what is fashionable.

They claim that people who wish to signal high quality fashion need to differentiate

themselves from others by generating a new signal in a different form than the

currently common. Bianchi (2002) sums it up: "Sharing styles and imitating the

behavior of others increases recognizability and identification. Contrasting styles and

behavioral habits emphasizes novelty and distinctiveness, and breaks the predictability

and lack of information in a code that is too much repeated."

 35

Seeking Status

Fashion is sometimes a means for signaling social status (Lynch and Strauss,

2007, Simmel, 1957). Simmel‟s (1957) theory states that the elite use the most

fashionable items, and the lower classes try to emulate them in order to gain status,

until eventually the elite abandons the fashion, moving on to a new one. Lynch and

Strauss (2007) cite McCracken‟s (1985, 1988) revision to Simmel‟s theory, to the

extent that the process does not simply occur between the social strata, but within

culturally derived categories (such as gender or ethnicity) and that the imitation does

not necessarily include fashion as a whole and can pertain to individual fashion

elements.

According to Conspicuous Consumption theory, fashion as expressed in

extravagant displays of wealth and in elements that contain novelty would signal a

person of the highest status (Lynch and Strauss, 2007). In conclusion, it is plausible,

that some people desire the fashionable\trendy because they wish to gain higher status

in the eyes of other members of society.

Identity Construction

Fashion participates in the process of social identity construction (Davis, 1992).

The social identity is a dynamic, constantly evolving, combination of personal

attitudes and distinguishing characteristics. The process of social identity construction

is affected by social pressures, and often externalizes in fashion choices and attitudes.

The Material Self psychological concept views a person‟s material possessions as

an extension of the self; therefore buying/adoption behaviors may not be purely

rational as other motivations like impression management can come into play

(Mannetti et al., 2002). Mannetti et al. (2002) cite Goffman‟s (1959) view that in
 36

contemporary everyday life, the presentation of self is based on material possessions

(not necessarily expensive ones) as much as, if not more than, on the expression of

attitudes and opinions. The Self Concept theory discusses a combination of the

physical and mental self and includes a motivation to reach the ideal "self" within a

private or social context (Evans, 1989). Therefore, individuals who adopt a fashion

may feel defined by it or wish to be perceived by others as fashionable (Diefenbach

and Hassenzahl, 2008).

Pleasure

People are after the fashionable for pleasure. Wilson (1985) sees fashion as a mass

pastime and a form of group entertainment. Bianchi (2002) proposes that fashion‟s

typical contrast between the expected\familiar and the unexpected\novel propagates

pleasure. She also proposes that pleasure increases with exposure in products that are

considered addictive (such as fashion products), because past consumption increases

the ability to use the product effectively.

Schifferstein and Zwartkruis-Pelgrim (2008) name four facets of the Self Concept

that determine people‟s level of attachment to objects – the Diffuse Self, Private Self,

Collective Self and Public Self. The Diffuse Self strives for hedonic satisfaction

through various types of pleasure (sensory, aesthetic, familiarity), therefore it is likely

to be a major factor in the pleasure people take in searching, adopting, buying, owning

or using fashionable products, as fashion is a branch of aesthetics and offers a highly

sensual experience. The Private Self aims for individual achievement, the Public Self

looks for the approval of others, and the Collective Self searches for approval of

reference group. It is plausible that when someone is fashionable, her private self is

enhanced and others, as well as reference group members, may approve of this

fashionable quality and enhance the public and collective self, appropriately.
 37

Very similarly, Green and Jordan (1999) discuss four product pleasure types and

their relevance in the context of products: physiological, psychological, social and

ideological pleasures. Physio-pleasure derives from sensory stimulation, for example

the tactile pleasure of holding and touching a product. Psycho-pleasure pertains to the

satisfaction of accomplishing a task, and products can help with making that

experience smooth and enjoyable. Socio-pleasure refers to enjoyment from the

company of others. Products can facilitate socio-pleasure in a number of ways, for

example a new gadget may attract pleasurable attention to its user. Ideo-pleasure refers

to pleasure derived from theoretical sources, such as product aesthetics and the values

behind its design. All these pleasure types can be induced by following fashion.

Trendsetters and Followers

Finally, Trendsetters – influential people, preoccupied with the new and

innovative, who receive public recognition and media coverage – are constantly in the

process of selecting and disseminating fashionable\trendy products (Vejlgaard, 2007).

At the same time, Followers are growing more aware of such products through

increased exposure to product information and to trendsetters. Fashion answers both

groups‟ psychological needs and desires and given the accelerated rate of

fashion\trend changes, mostly in the 20th and 21st centuries (Vejlgaard, 2007), people‟s

choices are increasingly influenced and sometimes may even be motivated by fashion.
 38

2.6 A Close-Up on Trendsetters

Trendsetters are defined as a sociological class of adoption group who are early

consumers of fashions and trends (Suzuki and Best, 2003, Vejlgaard, 2007). As

discussed above, Trendsetters have the special role of messengers in the trend process

– they reach out to wider society and spread the new message. In fact, they seem to

perform a triple role: they select from an offering of products and ideas, they tweak

them to better suit the public taste, and they influence the choice of followers.

Rogers (2003) characterized Trendsetters (Early Adopters) from a socio-economic

standpoint as more educated, of higher in social status, and with a greater degree of

social mobility. From a personality standpoint, he characterized Trendsetters as less

dogmatic and fatalistic, more intelligent and rational, with a greater ability to handle

abstractions and cope with uncertainty, with higher aspirations and more favorable

attitude towards change and science than later adopters. From a communication

standpoint, he characterized them as more socially active, interconnected and

cosmopolite, having more contact with change agents, with greater exposure to mass

media, and seeking innovation information actively, resulting in having greater

knowledge of innovations than later adopters.

Gladwell (2000) claims that the Trendsetter group is comprised of Connectors,

Mavens and Salesmen, and that this small group of people is responsible for spreading

trends. A Connector is a person with a unique social skill of creating weak ties in

many social spheres. When a Connector person adopts an idea, she quickly spreads it

to the larger population. A Maven is a person with a special knack for obtaining and

analyzing market information, who is motivated to help people using her accumulated

data bank. When a Maven person passes on her information to other people, she
 39

affects their behavior and choices, thus contributing to the build-up of trends. A

Salesman is a person with a special ability to convince people to do something.

Therefore, the salesman's part in the trend creation process is to persuade people to

adopt the trend. He names the influence of these unique messengers in the trend

mechanism the Law of the Few. While Trendsetters adopt an innovation from the

Innovators, they translate it to be more palatable to the mainstream. They may alter the

message by leveling it (dropping irrelevant details), sharpening it (exaggerating some

details) and improving its assimilation chances (changing the message to fit into the

mental schemas of the general population).

Citing Watts' (2007) findings, Vejlgaard (2007) disputes Gladwell's (2000)

argument of the Law of the Few. Using computer simulations, Watts (2007) found that

ideas spread in human networks not because of the presence of influential few, but due

to a critical mass of easily influenced people who will adopt an idea after exposure to a

single adopter from their network.

Vejlgaard (2007) describes the Trendsetter personality as generally very curious,

unafraid to stand out from the masses (with a strong sense of individualism) and

explore new things, mostly possessing a strong visual sense, and actively seeking

change. He claims that Trendsetters are overrepresented in the following groups:

young people, designers, artists, wealthy people, celebrities, gay men and a few other

style-conscious subcultures. He notes that Trendsetters tend to interact with polysocial

groups, which refers to having social contacts with groups that are different from

themselves. This is different from most social groups who tend to be monosoical,

which refers to people who prefer to interact with likes.

 40

Trendsetters must have acute sensibility to the public taste, or their selection will

not be accepted by followers and they may lose their trendsetting status. Therefore,

Trendsetters are constantly in the process of selection of fashionable and trendy

products (Suzuki and Best, 2003). In addition, Trendsetters receive public recognition

and media coverage, so they have a strong incentive to continue looking for the

fashionable and trendy. In summary, Trendsetters have a drive, a need and incentives

to both identify potential trends and influence other people to adopt them.

Vejlgaard (2007) adds that for a trend to move beyond Trendsetters and be adopted

by the mainstream, the interest of Trendsetters must be maintained for a certain period

of time, by product development and continued interest of other Trendsetters and

trendsetting media.
 41

3. Research Approach
Our overall objective is to study aspects of fashion\trends in HCI. Since that is a

fairly wide scope, I focus this research effort on the field of web design trends, and on

three research paths: establishing the premise that trends in web design exist,

validating the research‟s web design trend library, and investigating whether there are

differences in how people of different adopter groups perceive web design trends. In

order to achieve these goals, I conducted a three-part research program as described

below.

Part 1: Building the web design trend library

The purpose of this part of the research was to systematically document and map

web design trends. As part of this aspect of the project, I have built a library of web

design trends. For each trend, the library includes visual examples, alongside their

characteristics and attributes such as names, dates, references, typical elements,

correspondence to technological advances, etc. The method and process of compiling

the web design trend library is detailed in section 4.

The product of this part of the study (i.e., the web design trend library) served two

purposes: it provided evidence for the existence of trends in web design and it was

used as stimuli in the next two parts of the study. In addition, by documenting these

web design trends I hope to be able to share them with the HCI research community

and with industry practitioners, and to possibly lay the ground for future research as

well as contribute to more informed UI design decisions.

 42

Part 2: Delphi study

The purpose of this part of the research was to validate the web design trend

library I have compiled in part 1 of the project and enrich our understanding of trends‟

context and characteristics. In addition, I wanted to check the web design trend library

for current trend completeness (does it include all of the most current web design

trends?) and compile a list of current web design trendsetters. I conducted this part of

the research as a two-round Delphi study, targeting an international group of web

design trend experts as participants. This part of the research is detailed in the section

5.

Part 3: Testing adopter groups’ reactions to design trends

The purpose of this part of the research was to establish that web design trends

indeed behave in a manner that resembles other trend and fashion-like phenomena. I

measured the perception and attitudes of people from different adopter groups towards

web design trends. I defined two user groups as Trendsetters and Followers, based on

conceptual classifications such as Rogers‟ (2003) adopter classification, and

Vejlgaard‟s (2007) trend groups. To operationalize adopter groups I used existing

scales such as the TDS-K trendsetting questionnaire (Batinic et al., 2008), the Domain

Specific Innovativeness scale (Goldsmith and Hofacker, 1991), and the Centrality of

Visual Product Aesthetics scale (Bloch et al., 2003), which were translated and

adapted for the web design domain. Next, I tested whether people who belong to

different adopter groups reacted differentially to web design trends that were at

various trend life-cycle stages (e.g., faded, past-peak, and current trends). This part of

the research is detailed in the section 6.

 43

4. Web Design Trend Library

The purpose of this part of the research was to systematically document and map

web design trends. As part of this aspect of the project, I have built a library of web

design trends. For each trend, the library includes many examples, alongside their

visual characteristics and other attributes. The web design trend library is used in this

research as a source for study stimuli. In addition, by documenting these web design

trends I hope to be able to share them with the HCI community and possibly lay the

ground for future research and practice.

The following quotes from online web design trend experts, provided us with some

initial indication that web design trends exist, before I had started to map them out.

“Trends rule the web design world. More so than with other areas of
design, the web is very transparent and in a matter of hours I can
click my way through the work of maybe hundreds of individual web
designers, without even realising it. This ease of access makes „trend
osmosis‟ something of an unfortunate inevitability amongst web
designers.” (Faddis, 2006)

“Take the world of fashion for example. Fashion is in one minute

and out the next because we get bored of looking at the same things
day after day. The same pattern will occur in web design.” (Meek,
2007)

“Most design trends come unexpectedly, evolve over time, become

pointless and finally disappear from the design landscape. This
holds particularly for web design, which is - just as every other
creative field - prone for over-hyping and over-usage of trends.
Being used excessively, trends lose their ability to communicate
information, express something unique or innovative and
consequently lose their visual appeal.” (Lennartz, 2008)
 44

4.1 Methodology
Data Collection

The web design trends were collected by searching the internet for web design

trend reviews in online design magazines and blogs as well as reviewing the limited

amount of academic literature on this topic. Academic literature on web design trends

was available mainly for faded web design trends, i.e. trends that have been

discontinued (Engholm 2001, 2002, 2007, Lialina 2005, 2007, 2010). Web design

trend classification decisions were reached through consensus by two judges from this

research team (Korman Golander and Kabessa Cohen).

Criteria

I used three criteria as preliminary evidence for a web design trend‟s existence:

 The first criterion was met once we had sufficient evidence from multiple

online sources such as web design trend reviews that the trend exists.

 After the trends were identified we conducted additional refinement

iterations to improve the coherency of each trend. Here, a second criterion

had to be met, referred to as trend coherence. This criterion is conceptually

similar to the idea of construct validation, namely convergence and

discriminability (i.e., by finding various web sites that are designed very

similarly, and which are distinguishable from other trends).

 The third criterion for the existence of trends required that in addition to

archetypical web design samples for each trend, which were identified

during the refinement iterations, other sites will be found that adhere to the

principles of the trend.

 45

Dating

The web design trend library specifies the date range in which the web design

trends were valid. Since dating any type of trend is a rather complex task, I regard

these date range as an estimate. Web design trends date range estimation was done by

combining date information from three sources:

1 The date range of the online web design trend reviews it was found in.

2 The specific trend‟s design adoption dates by selected websites (using the
“Internet Archive Wayback Machine” archive at http://www.archive.org).

3 The date range in which the trend‟s keyword was searched online, limited to
Internet\Web Design and Development category (using the “Google Insights
for Search” online search tool at http://www.google.com/insights/search).

For example, the trend “Web 2.0 Design” appeared in web design trend

reviews from 2005 to 2009 (Source 1), its adoption dates by selected websites

ranged from 2006 to 2009 (Source 2), and the search results retrieved using

“Google Insights for Search” yielded a date range from 2005 to 2010 (Source 3).

Since the graph obtained from source 3 (Fig. 19) shows a steady decline in the

interest in this trend since 2007 and that sources 1 and 2 date ranges end by 2009, I

dated this trend as applicable in the years 2005-2009.

Figure 19: Interest over time in “web 2.0 design”, obtained from “Google Insights for
Search” and limited to “Internet\Web Design and Development” category
 46

Trend Life-Cycle Stage Classification

Based on the date ranges resulting from the dating procedure described above, an

initial classification of the web design trends trend life-cycle stage was performed.

Web design trends that have been discontinued, i.e. their date range is utterly in the

past, were classified as Faded. Web design trends that are still visible today but show

indications that they are past their peak, i.e. their date range may extend till recently

but has begun a few years ago and are not mentioned in recent years‟ web design trend

reviews or often referred to as “overused” were classified as Past-Peak. For example,

the “Web 2.0 Design” trend, depicted above as dated to 2005-2009 was classified as

Past-Peak. Web design trends that are visible today, their date range extends till the

present and are mentioned in recent years‟ web design trend reviews were classified as

Current.

4.2 Contents
I have classified 42 web design trends that met the three criteria. 10 web design

trends were classified as Faded, 12 web design trends as Past-Peak and 20 as Current

web design trends. These findings suggest that the pace of web design trends may be7

accelerating, in accordance with the general recorded acceleration in fashion\trends

(Vejlgaard, 2007).

For each web design trend documented, I listed items such as:

 Date Range – estimation of the trend‟s date range

 Trend Group - a thematic category for the trend style

 Tech Drive - new technologies that supported the emergence of the trend
 47

 References - quotes, dates and links to online sources, mainly web design trend

reviews

 Elements – characteristic visual elements of the web design trend

 Search Trendline – a graph depicting the volume of online searches for the

trend‟s keywords, limited to the Internet\Web Design and Development

searches (obtained using the “Google Insights for Search” online search tool at

http://www.google.com/insights/search).

The web design trend library will be publicly available at the following

URL: http://hci.ise.bgu.ac.il/trends (to appear). An example of a trend‟s record in

the web design trend library is given in appendix 1.

 48

5. Study 1 – Delphi Study

The objective of this study was to validate the web design trend library‟s current

trends collection and enhance our understanding of their megatrend relations and

industry trendsetters. The study was conducted using a 2-round Delphi method, a

research method used to develop, identify, forecast and to validate in a wide variety of

research areas, through an iterative process which collects and distills the judgments of

experts (Skulmoski et al., 2007).

5.1 Round 1

5.1.1 Overview

In the first round of the Delphi study I examined how a group of 22 international

web design trend experts rate the up-to-dateness of the 25 current web design trends I

had collected in the web design trend library. In addition, for each current web design

trend I asked the experts if it is related to any cultural, social or technological

megatrends. Finally, I asked if there were any current web design trends missing from

the library and who are the current web design trendsetters.

5.1.2 Method

Participants

The target population in this study was web design trend experts. The literature

does not provide clear recommendations regarding the appropriate sample size of

Delphi studies. One source states a range of 4-171 participants (Skulmoski et al.,

2007) while another source states a range of 10-1685 participants (Akins et al., 2005).

I compiled a list of 60 international web design trend experts, with main criteria of
 49

having published one or more articles concerning web design trends in online blogs,

magazines or websites. 22 participants completed the first round of the study, which

represents a response rate of 37% and falls within the appropriate sample size range.

Stimuli

25 current web design trends were used in this round. These web design trends

were taken from the current trends section of the web design trend library I have

created. Each web design trend was represented by an image comprising of four

screenshots of web sites, typical to the web design trend. A detailed list and images of

the 25 stimuli used in this round is available in Appendix 2, section 3.

Procedure

Round one of this study was performed using an online web survey service

(http://www.surveymonkey.net) in October-November 2010. The target group of

international web design trend experts received an email with an explanation about the

survey, a call to participate in it and a link to the online survey (appendix 2, section 1).

In the first stage of this round of the study, following a brief introduction, the

participants were requested to enter their contact information: name and email. These

details were gathered so that I could contact them later on for the second round of the

study. In the next stage, participants were given short instructions for the study process

(appendix 2, section 2).

In the main stage of this round of the study, participants were presented with the

25 current web design trends stimuli. For each stimulus, the participants were

presented with a large image. They rated the trend‟s up-to-dateness on a 1 - 10 scale

and elaborated on possible relation to cultural, social and technological megatrends in

a space provided under the ratings scale (example in Fig. 20).

 50

Figure 20: A screenshot of a page in the study – round 1 (“Cartoons\Mascots” current trend)

In the final stage of this round, the participants were requested to answer three

additional open ended questions. The first question was about the possible existence of

additional current web design trends that were missing from the survey and their

possible megatrend relations. The second question was about identifying current web

design Trendsetters (be it individual designers, design agencies or commercial

organizations). The third question was aimed at retrieving any additional comments

the participants may have had.

5.1.3 Results

Means and standard deviations of current trend up-to-dateness rating were

calculated (appendix 2, section 4). This analysis yielded an initial up-to-dateness

 51

ranking of the 25 current web design trends used in this round (Fig. 21). The overall

up-to-dateness rating had a mean of 6.10, minimum of 3.55 and maximum of 8.14.

Figure 21: Round 1 up-to-dateness rating of 25 current web design trends, ordered from
highest to lowest

Participants‟ responses to the open-ended questions regarding current web design

trends that were missing from the study and current web design trendsetters were

analyzed and yielded some results that might be of interest to industry practitioners2.

Participants‟ responses to the open ended question regarding boarder trend

relations were analyzed by two judges, receiving decisions in consensus. It became

evident during the initial analysis that the participants were not able to relate the

current web design trends to broader cultural, social or political megatrends, possibly

because they do not have a trend researcher background. However, participants‟

responses did supply interesting information about each trend that was more closely

related to its stylistic attributes and its relations with other web design trends.

2
A list of 9 additional current web design trends that were missing from the study‟s trends and
suggested by the research participants was assembled along with links to example websites (appendix 2,
section 5). A list of 16 current web design trendsetters suggested by the research participants along with
links to their respective websites was also assembled (appendix 2, section 6). Of this list, 3 web design
trendsetters were commercial firms, 11 were designers, design agencies, or design publications and 2
were non-web designers.
 52

Therefore, for each current web design trend, I summarized the participants‟ responses

into several categories. Then, these categories were further reduced to fewer higher

order categories per trend. Finally, these higher order categories were grouped under

four bipolar dimensions that may be used to quantify web design trends stylistic traits

(Table 4).

1 Simple Complex
2 Impersonal Personal
3 Direct Subtle
4 Modern Nostalgic
Table 4: Four web design trends bipolar dimensions extracted from round 1 results

Finally, I validated and enriched the web design trend library (section 4) using the

above results. First, I re-classified five web design trends that our dating procedure

pointed to be current, but received up-to-dateness rating of under than 5 (on a 1-10

scale) as past-peak trends (Stationary, Messy Desk, Grunge, Texture | Bricks and

Splattered Backgrounds trends). Second, I added a summary of the expert comments

regarding the broader context of the trend, relating it to other trends, revealing design

influences, etc. to each current web design trend in the library.

5.2 Round 2
5.2.1 Overview

In the second round of the Delphi study I allowed the same group of 22

international web design trend experts to revise their up-to-dateness rating of the 20

most current web design trends identified in the first study, in light of the expert

group‟s average rating. In addition, for each current web design trend I asked the

experts to rate it on four bipolar dimensions extracted from the expert group‟s

responses to the open question regarding broader trend context in the first round.
 53

5.2.2 Method

Participants

Out of the 22 international web design trend experts who completed the first round

of the study, 11 participants took part in the second round of the study (9 completed

fully, and 2 partially), which represents a response rate of 50% and falls within the

appropriate sample size range.

Stimuli

The 20 most current web design trends that were identified in the first round of this

study were used in this round. These web design trends were taken from the current

trends section of the web design trend library I have created. Each web design trend

was represented by an image comprising of four screenshots of web sites, typical to

the web design trend. The 20 trend representation images that were used in this round

were identical for those used for these trends in round 1. A detailed list and images of

the 20 stimuli used in this round is available in appendix 3, section 2.

Procedure

Round two of this study was performed using the same online web survey service

(http://www.surveymonkey.net) in December 2010. The group of 22 international web

design trend experts, who completed the first round of this study, received an email

with an interim report of the first round findings and an invitation to participate in the

follow-up survey (appendix 3, section 1). Each participant received a unique survey

link that presented her personal up-to-dateness rating from the previous round.

Following a brief introduction, the participants were presented the 20 most current

web design trends, according to the results of the first round (five web design trends

which received up-to-dateness ratings of less than 5 were omitted). For each trend, the
 54

participants were given the opportunity to revise their previous up-to-dateness rating,

while reviewing the mean ratings obtained in the first round of the survey. In addition,

participants were asked to rate the 20 web design trends on the four bipolar

dimensions (Table 4) on a 1 – 10 scale (example in Fig. 22).

Figure 22: A screenshot of a page in the study – round 2 (“Cartoons\Mascots” current trend)
 55

5.2.3 Results

Means and standard deviations for the current web design trends up-to-dateness

rating were calculated (appendix 3, section 3). This analysis yielded a final up-to-

dateness ranking of the 20 current web design trends used in this round. Figure 23

displays round 1‟s and round 2‟s up-to-dateness rating for the 20 current web design

trends, ordered highest to lowest by round 2 scores. The correlation between the up-to-

dateness ratings obtained in the two rounds was high (0.96).

Figure 23: Round 1 and 2 up-to-dateness rating of 20 current web design trends (highest to
lowest by round 2 scores)

In addition, participants rated the 20 current web design trends used, on the four

bi-polar dimensions extracted from the first round (simple-complex, impersonal-

personal, direct-subtle, and modern-nostalgic), yielding web design characteristic

quantification (appendix 3, sections 4-7). The trends‟ distribution along these

dimensions is presented in Table 5 bellow.

 56

Dimension Mean Median Minimum Maximum

Simple (1) – Complex (10) 5.45 5.40 3.00 8.10
Impersonal (1) – Personal (10) 5.98 5.79 4.70 7.67
Direct (1) – Subtle (10) 4.70 4.72 2.44 7.10
Modern (1) – Nostalgic (10) 5.16 5.00 2.44 8.22
Table 5: The four web design trends bipolar dimensions‟ statistics

In order to visualize the web design trend dimensions, I present examples from

each dimension‟s extreme pole. The Simple-Complex web design trend dimension

extremes were “Typography | Big and Bold” web design trend (Fig. 24), which is

closest to the Simple pole (3.00) and “Ornamental” web design trend (Fig. 25), which

is closest to the Complex pole (8.11).

Figure 24: “Typography | Big and Bold” trend rated the most simple on the Simple-Complex
dimension and the most direct on the Direct-Subtle dimension
 57

Figure 25: “Ornamental” trend rated the most complex on the Simple-Complex dimension

The Impersonal-Personal web design trend dimension extremes were “Magazine

Layouts” web design trend (Fig. 26), which is closest to the Impersonal pole (4.70)

and “Sketches\Hand Drawn Illustrations” web design trend (Fig. 27), which is closest

to the Personal pole (7.67).

Figure 26: “Magazine Layouts” trend rated the most impersonal on the Impersonal-Personal
dimension
 58

Figure 27: “Sketches\Hand Drawn Illustrations” trend rated the most personal on the
Impersonal-Personal dimension

The Direct-Subtle web design trend dimension extremes were “Typography | Big

and Bold” web design trend (Fig. 24), which is closest to the Direct pole (2.44) and

“Texture | Grain” web design trend (Fig. 28), which is closest to the Subtle pole (7.10).

Figure 28: “Texture | Grain” trend rated the most subtle on the Direct-Subtle dimension

The Modern-Nostalgic web design trend dimension extremes were “Transparency”

web design trend (Fig. 29), which is closest to the Modern pole (2.22) and
 59

“Retro\Vintage” web design trend (Fig. 30), which is closest to the Nostalgic pole

(8.22).

Figure 29: “Transparency” trend rated the most modern on the Modern-Nostalgic dimension

Figure 30: “Retro\Vintage” trend rated the most nostalgic on the Modern-Nostalgic
dimension

The correlation between the date range estimation conducted by the researchers

(operationalized as estimated duration in years) and the up-to-dateness rating of the

web design trend experts was low (Table 6) but indicated that the date ranges
 60

estimation are overall in-line with the experts‟ up-to-dateness rating. In addition the

correlations between the date range estimation conducted by the researchers

(operationalized as estimated duration in years) and the four bi-polar dimension rating

described above were also low (Table 6). These correlations indicate that trends are

gravitating towards being more simple, personal, subtle and nostalgic, but that these

tendencies are relatively weak. The full table for this analysis is located in appendix 3,

section 8.

Perceived Simple- Impersonal- Direct- Modern-

Up-to- Complex Personal Subtle Nostalgic
Dateness
Est. Duration -0.186 0.077 -0.388 -0.170 -0.092

Table 6: Pearson correlations between estimated duration (years) and up-to-dateness, simple-
complex, impersonal-personal, direct-subtle and modern-nostalgic dimensions

Finally, I further validated and enriched the web design trend library (section 4)

using the above results. First, I added the up-to-dateness ratings obtained in this round

to each current web design trend in the web design trend library. In addition, the

current web design trends in the library were ordered by these ratings, from the highest

to the lowest. Second, I added the web design characteristic quantification of the four

bi-polar dimensions described above to each current web design trend in the library

(see example in appendix 1).

Study 2 results seem to support the premise that web design trends exist in a

number of ways. First, the web design trend experts that participated in the study were

supportive of the idea of trends in web design. They commented that the study had

“good coverage and questions about web design trends” and that I “did a great job in

summarize the most notable trends”. In their answers to broader trend context

questions, they would often refer to related or counter trends, for example regarding
 61

the Minimalism trend one expert noted that “this trend is everywhere - fashion, interior

design, etc. I think it's a reaction to the overly decorative older trends like collage and

grunge” and regarding the Cartoons\Mascots trend, another expert noted that

“illustration is big trend right now, and mascots are part of that”. In addition, being

able to achieve consensus within the expert group regarding the trends‟ up-to-dateness

rating, gives reason to believe that these are, in fact, visual styles that change over

time, which is congruent with the definition of a trend (section 2).

 62

6. Study 2 – Main Study

The objective of this study was to test the differences in web design trend

perception between people from different adopter groups: Trendsetters and Followers.

Since Trendsetters are defined as a sociological class of adoption group who are early

consumers of fashions and trends (Suzuki and Best, 2003, Vejlgaard, 2007) and

Trendsetter personality is generally very curious, unafraid to stand out from the

masses (with a strong sense of individualism) and explore new things, mostly

possessing a strong visual sense, and actively seeking change (Vejlgaard, 2007), I

hypothesized that they differ from Followers in terms of their perceptions of, and

attitudes towards the various trends. If such differences exist in the expected direction

then my argument about the existence of trends and about how different adopter

groups react to them will be supported. In line with the theory about adopter groups

(see section 2), I submit the following hypotheses:

H1a. Trendsetters will perceive current web design trends as more

up-to-date than non-current web design trends.

H1b. Trendsetters will like current web design trends more than non-

current web design trends.

H2a. Followers will perceive non-current web design trends as more

up-to-date than current web design trends.

H2b. Followers will like non-current web design trends more than

current web design trends.

 63

6.1 Method
6.1.1 Participants

While I did not consider it a problem to recruit a large sample of participants who

belong to the Followers adopter group, ensuring adequate representation of

Trendsetters was a challenge. Towards this end I made an effort to recruit people who

have a high likelihood of belonging to the small adopter group of Trendsetters (e.g.,

design practitioners and students).

Therefore, there were two target populations in this study. One was Ben-Gurion

University Information Systems undergraduate class students and the other was

Bezalel Academy for Arts and Design mailing-list members including students,

graduates and lecturers. The reasoning for this decision was to enable classifying

participants to two distinct groups of designers and non-designers and test for adopter

group association and differences in web design trend perception. Each participant

from the Ben-Gurion University student group was awarded 1 class grade bonus point

in return for participation in this study. The Bezalel Academy for Arts and Design

group members did not receive any compensation for their participation in this study.

274 participants completed the study (139 males and 135 females, age range from

20 to 66 years, mean age 28.91 years, age SD 8.059). Out of the 274 participants 93

were Ben-Gurion University students, 126 were Bezalel Academy for Arts and Design

students\grads and 55 were otherwise affiliated (since this study was conducted

through an online survey, available through a web link, it must have been passed on to

other people by the Bezalel Academy for Arts and Design mailing list members).

I classified the participants to two groups: designers and non-designers, based on

education, academic affiliation, and occupation. Out of the 274 participants, 110 were
 64

classified as non-designers, 152 were classified as designers. I was not able to classify

the remaining 12 participants, due to missing information, so their input was omitted

from the later analysis. Thus, the analysis refers to data from 262 participants (137

males and 125 females, age range from 20 to 66 years, mean age 28.63 years, age SD

7.895).

6.1.2 Stimuli

26 different web design trends from three trend life-cycle stages were used in this

study. Of the 26 web design trends, 10 were Faded, 6 were Past-peak and 10 were

Current. These trends were selected from the web design trend library (section 4),

based on their visual coherence, i.e. having a clear internal resemblance within the

trend‟s samples, and being as distinguishable as possible from other trends. Each web

design trend was represented by an image comprising of four screenshots of websites,

typical to the web design trend. Each one of the four web site screenshots used for

representing a web design trend was captured with the browser frame, in order to

make it clear to participants that these images are of websites. See appendix 4, section

1 for a detailed list and images of the stimuli used in this study.

6.1.3 Procedure

This study was performed using an online web survey website in June 2010. The

Ben-Gurion University students received an email with a custom link to the online

survey, sent through the online web survey system, so that they could be identified for

the purpose of rewarding them with the grade bonus point. The Bezalel Academy for

Arts and Design mailing list members received an email, sent through a Bezalel

academic staff member email account, with a generic open link to the online survey,

so these participants were anonymous. Both types of web links led to the same online
 65

web survey, the only difference being that participant identity was known only for the

Ben-Gurion University students.

In the first stage of the study, following a brief introduction, the participants were

requested to enter their background information: gender, age, occupation, field of

academic studies, and academic affiliation (if applicable). In the next stage,

participants were given instructions for the study process. Following that, there was a

short training stage, comprising of a single trend (Texture | Bricks), clearly marked as

such, but identical in content and form to the actual study phase.

In the main stage of the study, clearly marked as such, participants were presented

with the 26 web design trends stimuli in a random order. For each web design trend,

the participants were presented with a large image of the stimuli, so that they could

perceive it as best as possible. In the next page the same stimuli was presented again

using a smaller image to use as a reminder and participants were then requested to rate

the trend‟s up-to-dateness and the how much they like it on a Likert scale of 1 – 10

(example in Fig. 31). There was no time limit for evaluating the web design trends in

this study.

Figure 31: A screenshot of a page in the main study stage (“Huge Images” Current Trend)
 66

In the final stage of the study, the participants answered three multiple-item scales

related to adoption groups. Since I could not find in the relevant literature a

measurement tool that will allow classifying participants into innovation adoption

groups, I decided to use this combination of measurement tools: TDS-K which

measures Trendsetting (Batinic et al, 2008), DSI which measures Domain Specific

Innovativeness (Goldsmith and Hofacker, 1991) and CVPA which measures the

Centrality of Visual Product Aesthetics (Bloch et al, 2003). The TDS-K scale (Batinic

et al, 2008) was adapted to target Trendsetting in design. In order to update the TDS-K

(Batinic et al, 2008) scale to the online era, I added a new item to it which measures

the participant‟s level of updating through online design blogs/websites/forums. The

DSI scale (Goldsmith and Hofacker, 1991) was adapted to measure product fashion

innovativeness and its last item was altered to measure the participant‟s level of

interest in new things. The participants indicated their level of agreement with each of

the statements on a 1 – 7 scale. See appendix 4, section 2 for the original English

version and adapted Hebrew version for the three measurement scales used.

6.2 Results
6.2.1 Background Variables

I conducted a Factor Analysis with 3 extracted factors (appendix 4, section 3.1.3).

An oblique rotation method (Oblimin) was used, since TDS-K, DSI and CVPA were

expected to be correlated. The total variance explained by the 3 factors was 67.8%.

The TDS-K scale was found to be cohesive. Item 6 in the DSI scale (which I have

altered to measure the participant‟s interest in new things) was cross-loaded on another

factor. Item 11 in the CVPA scale (measuring the urge to buy well designed products)

was also cross-loaded on another factor.

 67

Following the Factor Analysis results, I removed item 6 from the DSI scale and

item 11 from the CVPA, as they were loaded on different factors. I believe it is

possible that a confounding variable may be at work in regards to item 6 of the DSI

scale: non-designers may have fashion innovativeness pertaining to tech gadgets while

designers may have fashion innovativeness pertaining to design. The reliabilities

(Chronbach‟s α) of the scales ranged from 0.91 to 0.94 (Table 7). The correlations

between the scales ranged from 0.58 to 0.74 (Table 7) Therefore, the rest of the

analysis was done on the shortened scales (excluding CVPA item 11 and DSI item 6).

Scale # of Items Mean (SD) TDS-K DSI CVPA

TDS-K 10 41.98 (15.43) 0.94 0.58 0.74
DSI* 5 14.00 (7.43) 0.91 0.58
CVPA* 11 56.08 (15.37) 0.94
Table 7: Descriptive statistics, reliabilities (on the diagonal) and correlations between the
modified background variables

This study‟s hypotheses pertain to two adopter groups: Trendsetters and Followers. In

order to be able to classify the designers group as Trendsetters and the non-designer

group as Followers, I examined their background variable differences, i.e.

Trendsetting (TDS-K), Domain Specific Innovativeness (DSI*) and Centrality of

Visual Product Aesthetics (CVPA*). I tested this by performing independent samples

T-Tests for these variables (appendix 4, section 3.1.6). A summary of these T-Tests

results is presented in Table 8 below.

Mean (SD) P
Designers Non-Designers
Trendsetting 4.77 (1.42) 3.39 (1.36) < 0.001
Domain Specific Innovativeness 2.82 (1.42) 2.77 (1.58) ns
Centrality of Visual Product Aesthetics 5.56 (1.17) 4.46 (1.44) < 0.001
Table 8: Independent Samples T-Test analyses summary for background variables
(Statistically significant differences noted in bold)
 68

According to the above T-Tests results, designers had statistically significant

higher Trendsetting (TDS-K) and Centrality of Visual Product Aesthetics (CVPA*)

scores than non-designers (p<0.001). Designers and non-designers did not differ along

the Domain Specific Innovativeness (DSI*) scores. These results show that designers

are more Trendsetting and have higher Centrality of Visual Product Aesthetics scores

than non-designers (Figure 32).

Figure 32: Background variable differences between designers and non-designers

(statistically significant differences noted by a star)

6.2.2 Dependent Variables

I calculated the up-to-dateness scores for each trend life-cycle stage (Faded, Past-

Peak, and Current) by averaging the up-to-dateness scores of the trends classified in

that stage for each participant. Similarly I calculated the level of liking scores for each

trend life-cycle stage. A Summary of the Pearson correlations (appendix 4, section

3.2.1) between these dependant variables: perceived up-to-dateness and level of liking

of web design trends in the different stages of the trend life-cycle are presented in
 69

Table 9. High correlations were found between up-to-dateness and level of liking of

the same trend stage (background highlighted on the diagonal in Table 9). Low

correlations were found between up-to-dateness and level of liking of different trend

stages. In addition, I noted higher correlations between up-to-dateness and level of

liking for past-peak and faded trends (bolded in Table 9), compared to lower

correlations between up-to-dateness and level of liking of current and past-peak/faded

trends (italic in Table 9). Frequencies and distributions of up-to-dateness and level of

liking ratings, in the trend life cycle stages, are depicted in appendix 4, section 4.

Current Trends Past Peak Trends Faded Trends

up-to- up-to- up-to-
liking liking liking
dateness dateness dateness
up-to-
0.74** 0.39** 0.17* 0.22** 0.11
Current dateness
Trends
liking 0.36** 0.39** 0.26** 0.28**
up-to-
Past 0.77** 0.53** 0.42**
dateness
Peak
Trends liking 0.48** 0.49**
up-to-
Faded 0.78**
dateness
Trends
liking
Table 9: Pearson correlations summary for perceived up-to-dateness and level of liking
(** p < .001; * p < .01)

6.2.3 Hypothesis Testing

This study‟s hypotheses pertain to people‟s perception of and attitudes towards

current and non-current web design trends. The current web design trends refer to web

design trends classified as Current, while the non-current web design trends refer to

web design trends classified as Past-Peak or Faded.

 70

I used regression analysis and analysis of variance (ANOVA) to test the

research hypotheses. I start with a regression analysis of the effects of the background

variables on the dependent variables for each of the trend groups separately. The

analyses results are reported in tables 10-15. There were separate analyses for up-to-

dateness and to liking as dependant variables. Independent variables were Design

Orientation (design_orientation - the subject‟s classification as designer or non-

designer based on education and occupation), Trendsetting (mean_tdsk - the subject‟s

TDS-K scale score), Domain Specific Innovativeness (mean_dsi_new - the subject‟s

DSI* scale score) and Centrality of Visual Product Aesthetics (mean_cvpa_new - the

subject‟s CVPA* scale score).

R2 = 0.16
Coefficients
Model Understadardized Standardized t sig
Coefficients Coefficients
B Std.Error Beta
(Constant) 4.72 0.32 14.92 0.000
mean_tsdk -0.33 0.09 -0.35 -3.71 0.000
mean_dsi_new 0.11 0.08 0.11 1.45 0.149
mean_cvpa_new 0.14 0.09 0.13 1.46 0.146
design_orientation -0.68 0.20 -0.23 -3.38 0.001
Table 10: Regression analysis for perceived up-to-dateness of faded trends (statistically
significant dependant variables noted in bold)

R2 = 0.15
Coefficients
Model Understadardized Standardized t sig
Coefficients Coefficients
B Std.Error Beta
(Constant) 7.05 0.30 23.45 0.000
mean_tsdk -0.16 0.08 -0.18 -1.93 0.055
mean_dsi_new 0.14 0.07 0.16 1.97 0.050
mean_cvpa_new 0.17 0.09 0.17 1.92 0.056
 71

design_orientation -0.88 0.19 -0.32 -4.59 0.000

Table 11: Regression analysis for perceived up-to-dateness of past-peak trends (statistically
significant dependant variables noted in bold)

R2 = 0.12
Coefficients
Model Understadardized Standardized t sig
Coefficients Coefficients
B Std.Error Beta
(Constant) 6.30 0.28 22.24 0.000
mean_tsdk -0.10 0.08 -0.12 -1.21 0.227
mean_dsi_new 0.05 0.07 0.06 0.69 0.490
mean_cvpa_new 0.24 0.08 0.27 2.92 0.004
design_orientation 0.55 0.18 0.21 3.02 0.03
Table 12: Regression analysis for perceived up-to-dateness of current trends (statistically
significant dependant variables noted in bold)

R2 = 0.14
Coefficients
Model Understadardized Standardized t sig
Coefficients Coefficients
B Std.Error Beta
(Constant) 4.04 0.26 15.52 0.000
mean_tsdk -0.20 0.07 -0.27 -2.81 0.005
mean_dsi_new 0.11 0.06 0.14 1.76 0.080
mean_cvpa_new -0.04 0.08 -0.04 -0.46 0.645
design_orientation -0.44 0.17 -0.18 -2.62 0.009
Table 13: Regression analysis for level of liking of faded trends (statistically significant
dependant variables noted in bold)

R2 = 0.23
Coefficients
Model Understadardized Standardized t sig
Coefficients Coefficients
B Std.Error Beta
(Constant) 6.15 0.34 18.29 0.000
mean_tsdk -0.09 0.09 -0.09 -0.97 0.334
mean_dsi_new 0.22 0.08 0.20 2.65 0.009
 72

mean_cvpa_new 0.91 0.10 0.08 0.92 0.359

design_orientation -1.41 0.22 -0.43 -6.57 0.000
Table 14: Regression analysis for level of liking of past-peak trends (statistically significant
dependant variables noted in bold)

R2 = 0.11
Coefficients
Model Understadardized Standardized t sig
Coefficients Coefficients
B Std.Error Beta
(Constant) 5.55 0.30 18.8 0.000
mean_tsdk -0.03 0.08 -0.04 -0.39 0.694
mean_dsi_new 0.08 0.07 0.09 1.16 0.245
mean_cvpa_new 0.20 0.09 0.22 2.33 0.021
design_orientation 0.40 0.19 0.15 2.14 0.033
Table 15: Regression analysis for level of liking of current trends (statistically significant
dependant variables noted in bold)

In general, the regression results for the up-to-dateness evaluations and for the

level of liking were very similar. The regression coefficient of Design Orientation was

statistically significant in all trend groups. It was positive for Current trends, meaning

that designers perceived the current trends as more up-to-date and they also liked them

more. The coefficient was negative for the Faded and Past-peak trends, indicating that

for non-designers, these trends were more likeable and appeared more up-to date. In

the group of Current trends, Centrality of Visual Product Aesthetics was positively

related to Up-to-dateness perceptions and level of liking. Trend perception for Faded

trends was found to be negatively dependant on the score of the Trendsetting scale.

That is, faded trends were perceived as more up-to-date and were more liked by

participants with low trendsetting scores. Finally, trend perception for Past-peak

trends was found to be positively dependant on Domain Specific Innovativeness.

Thus, the regression results provide support for the hypotheses that people with higher

trendsetting orientation would more correctly perceive current trends as up-to-date and
 73

would like those trends more than people who tend to be followers. Conversely,

followers perceived older trends as more up-to-date and they also preferred such

trends to a greater degree.

A more direct test of the hypotheses was conducted by a mixed-design analysis

of variance with trend life-cycle stages (Faded, Past-Peak, and Current) as a within-

subjects factor, and adopter groups (Trendsetters vs. Followers) as a between-groups

factor, for the two dependent variables (up-to-dateness and liking). The full tables of

this analysis are reported in appendix 4, section 3.3.1.

Up-to-dateness scores

The analysis revealed a statistically significant effect of trend life-cycle factor

(F(2,520) = 1014.54, p < 0.001). The difference between groups (designers vs. Non-

Designers) was also statistically significant (F(1,260) = 9.80, p < 0.002). Finally, there

was a significant trend group (Faded, Past Peak, Current) by population group

(Designers, Non-Designers) interaction effect (F(2,520)=56.50, p < 0.001). The

interaction is presented in Figure 33.

To examine the source of the interaction effect I tested the difference between

trends‟ perceived up-to-dateness by the designers and non-designers groups, in each of

the trend life-cycle stages (Faded, Past-Peak, and Current), using independent

samples T-Tests. The full tables of this analysis are presented in appendix 4, section

3.3.2. A summary of these T-Tests results is presented in Table 16.

Mean (SD) P
Designers Non-Designers
Current 7.87 (1.13) 7.19 (1.34) < 0.001
Past-Peak 7.66 (1.14) 6.75 (1.40) < 0.001
Faded 3.55 (1.26) 4.53 (1.50) < 0.001
 74

Table 16: Independent Samples T-Test analyses summary for perceived up-to-dateness in the
trend life-cycle stages (statistically significant differences noted in bold)

According to the above T-Tests results, designers gave higher perceived up-to-

dateness scores to Current Trends than non-designers. On the other hand, non-

designers gave higher perceived up-to-dateness score to Past-Peak and Faded trends

than designers. These results support hypotheses H1a and H2a.

Figure 33: Trend up-to-dateness perception by designers and non-designers in the trend life-
cycle stages (faded, past-peak, current)

Trend Liking scores

The analysis revealed a statistically significant effect of the trend life-cycle

factor (F(2,520) = 1086.60, p < 0.001). The difference between groups (designers and

non-designers) was also statistically significant (F(1,260) = 17.06, p < .001). Finally,

there was a significant trend group (Faded, Past Peak, Current) by population group

(Designers, Non-Designers) interaction effect (F(2,520)=72.57, p < 0.001). The

interaction is depicted in Figure 34.

I used independent samples T-Tests to examine the source of the interaction

between groups (designers and non-designers) and trend life-cycle stages (Faded,

Past-Peak, and Current). The tests were performed in each trend life-cycle stage. The
 75

full tables of this analysis are presented in appendix 4, section 3.3.3. A summary of

these T-Tests results is presented in Table 17.

 76

Mean (SD) P
Designers Non-Designers
Current 7.15 (1.22) 6.57 (1.36) < 0.001
Past-Peak 5.42 (1.54) 6.85 (1.33) < 0.001
Faded 2.75 (0.99) 3.50 (1.29) < 0.001
Table 17: Independent Samples T-Test analyses summary for level of liking in the trend life-
cycle stages (statistically significant differences noted in bold)

According to the above T-Tests results, designers liked Current Trends more

than non-designers. On the other hand, non-designers liked Past-Peak and Faded

trends more than designers. These results support hypotheses H1b and H2b.

Figure 34: Trend liking by designers and non-designers in the trend life-cycle stages (faded,
past-peak, current)
 77

7. Discussion
This research effort was focused on the field of web design trends and tried to

establish the premise that trends in web design exist. In order to achieve that, I

followed three research paths: compiling the web design trend library, validating it

using a Delphi study in which web design trend experts participated, and investigating

the differences between the perceptions of web design trends by people associated

with two adopter groups – Trendsetters and Followers.

The web design trend library creation offers two main contributions. First, it is a

first attempt to methodically expose and categorize trends in the web design domain.

This effort yielded new information concerning web design trends such as archetypical

samples, characteristics, date ranges, etc. which may be of use in further research and

to practitioners. In addition, it provides some initial support to the premise that trends

in web design exist. Second, it offers some methodological contributions such as the

proposed methods for identifying, refining, and classifying web design trends, as well

as estimating their date range.

Study 1, conducted using the Delphi method with the participation of web design

trend experts validated the web design trend library‟s Current section and enriched it

with new information such as up-to-dateness ranking, broader trend contexts, and a

characteristic quantification of four dimensions extracted in this research, which may

be of use in further research and to practitioners. In addition, it provides further

support to the premise that trends in web design exist.

Study 2 presented our main hypotheses that Trendsetters will have more positive

attitudes for Current web design trends (will perceive them as more up-to-date and
 78

will like them more) than Followers, and that Followers will have more positive

attitudes for non-current (Past-Peak, Faded) web design trends (will perceive them as

more up-to-date and will like them more) than Trendsetters. The study‟s results

support our hypotheses. Since designers had statistically significant higher

Trendsetting and Centrality of Visual Product Aesthetics scores than non-designers, I

classified designers as Trendsetters and non-designers as Followers. I found that

designers had more positive attitudes towards Current web design trends: they

perceived them as more up-to-date and liked them more than non-current (Past-Peak,

Faded) web design trends. Furthermore, non-designers had more positive attitudes

towards non-current (Past-Peak, Faded) web design trends: they perceived them as

more up-to-date and liked them more than Current web design trends. These results

were statistically significant.

The importance of the results of Study 2 is that they indicate that people like web

design trends they perceive as current, but that Trendsetters are able to identify the

current web design trends better than Followers. These findings, showing varied

perceptions of people from different adopter groups towards web design trends from

different life-cycle stages, further indicate to the existence of web design trends.

7.1 Study Limitations

As in any study, the studies performed in this research had limitations. The

limitations of studies 1 and 2 are as a result of the stimuli used in these studies, the

web design trend samples. These samples, although each includes multiple website

screenshots and were considered to be archetypical, are static, as opposed to the actual

websites which may have some dynamic content. In addition, I have only used

screenshots of the main pages of these websites.

 79

Another limitation concerning study 1 is that I had a sufficient number of expert

participants (22) for the first round, but a much smaller number of these experts (11)

participated in the second round. Although this number of expert participants still falls

within the suggested sample size for Delphi studies (see Study 1), it is desirable to

have more participants in advanced rounds.

A third limitation pertains to the difficulty in obtaining adequate representation

and categorization for Faded web design trends. Since these trends have already been

discontinued and have been replaced by newer design in most websites, I had to rely

on sources such as the “Internet Archive Wayback Machine” online archive

(http://www.archive.org), which does not provide consistent high quality archiving

(for example, I could not use many website archived samples that were lacking proper

formatting or images), or on websites that have had their design “frozen” in time, due

to owner abandonment of website or other reasons.

7.2 Conclusions and Recommendations

The findings of this research suggest that designers, being visual trendsetters, are

equipped with the characteristics that allow them to distinguish current from older web

design trends. Given that, they may have the ability to select the appropriate design for

a product, given its time to market and target consumers. For example, if the time to

market is short and the target consumers are trendsetters, it might be advisable to

apply a very up-to-date web design trend for a website product. Alternatively, if the

time to market is short and the target consumers are followers, it might be advisable to

apply a past-peak web design trend for a website product.

Vejlgaard (2007) reveals that the nature of a trend is often social and cultural. In

addition, trends often do not materialize out of thin air, but are manifestations of
 80

broader political or technological shifts known as Megatrends. Given that, it might be

a good idea to discuss the social and cultural aspects of design, during the design

decision making process.

Trend acceleration and compression seem to be a general phenomenon, occurring

in the 20th and 21st centuries (Vejlgaard, 2007). During this research‟s web design

library creation, I have identified 20 current trends, 12 past-peak trends and 10 faded

trends. While this may not be conclusive evidence, it is plausible that these are the

symptoms of acceleration in the web design domain, congruently with the more

general acceleration and compression phenomenon. Based on these findings, the rate

of appearance of new web design trends (and the required adjustments of existing

websites) may continue to accelerate in the foreseeable future.

The current research, examining aspects of web design trends, is preliminary in the

field of HCI. I can suggest additional avenues of research for aspects of the

fashion\trends in HCI. This research can be expanded and replicated in additional HCI

domains, for example aspects of trends in desktop applications, smartphone

applications, interaction facets, etc. Another research avenue would be to conduct a

more thorough examination of the role of technology in the emergence and lifecycle of

HCI trends. It would also be interesting to examine the effect of web design trends on

the evolution of specific website designs, taking a closer look at some websites and

their design changes along the years. Other possible future studies may expand the

Delphi study conducted in this research to test a population of non-experts. One study

may test if non-experts would come up with similar or different dimensions to

describe web design trends. Another study may test how non-experts rate current

trends on the four dimensions identified by the experts in this research. I believe that it

would be highly beneficial to identify and quantify the HCI design elements that
 81

contribute to fashionability\trendiness. An additional contribution could be the

identification of HCI trendsetters (similar to what I found for web design) and

conducting HCI trend forecasting. Finally, I believe that research aimed at drafting

best practices for integrating HCI trend research into the design process of interactive

products is in order.
 82

Appendix 1: Web Design Trend

Library
Web Design Trend Library – sample record from the current section

Minimalism

Dates: 2005-2010

Trend Group: Simplicity, Counter trend to Collage and Grunge

Tech Drive: Cross device design, faster response times for mobile web
 83

References:

“In recent months you have probably noticed a new trend…the minimalist look”
Creative UI, 2007
creativeui.com/2007/10/10/new-minimalism-in-web-design

“The minimalist movement was a response to the abuse our senses took from early
web pages when busy backgrounds, blink tags and rainbow colored text were the
norm. “Minimalism” is a term coined by the art and literature circles to describe a
movement toward extreme simplification of form and color. Minimalism, as it relates
to web design, focuses on usability, aesthetics and letting the content be the star.”
Think Vitamin, 2007
carsonified.com/blog/design/web-design-isms-7-surefire-styles-that-work/

“It seems that everyone has a different opinion of what qualifies as a minimalistic
design”
Vandelay Design, 2007
vandelaydesign.com/blog/design/minimalistic-web-design/

“Perhaps it‟s my obsession with the Senior Vice President of Industrial Design at
Apple, Jonathan Ive. Or maybe the success of the Google search page. Or possibly
my seething hatred for retina-burning animated banner adverts and MySpace profile
pages. But I‟m sure Web users, designers and developers alike have all agreed that
less is certainly more. And so, Minimalism seems to be the new black (and white) in
the Web design industry.”
Gotta Quirk, 2009
gottaquirk.com/2009/06/03/trend-of-note-minimalism

“The trend of minimalist designs doesn‟t only apply to products design, but also to
web design with grid-based designs getting popular.”
Design Reviver, 2009
designreviver.com/inspiration/interface-design-trends-for-2009

“Forget the old school minimal websites. Websites of 2010 will continue to feature
lots of white space but with bold typology and surprising color schemes. Not all
minimal websites will agree with the notion of black and white simplicity. Although
minimalism is by nature muted, it will also showcase fresh colors. Minimalism isn‟t
cold, it‟s warm and to the point.”
WDL, 2009
webdesignledger.com/tips/web-design-trends-for-2010

“As an adjunct to the typography trend, minimalism will rule the day. Designers will
be designing with type instead of building pretty images around it. This will lead to a
cleaner look for many websites, with the font choices and layouts doing most of the
work aesthetically.”
Helikopter Design, 2009
helikopterdesign.com/archives/web-design-trends-2010

“Together with rich typography comes the minimalism. This trend comes and goes
every once in a while but it seems that during the 2010 minimalism is going to be one
of the more prominent trends regarding the web design.”
UX Passion, 2010
 84

uxpassion.com/2010/01/design-trends-2010-web-design-trends-2010

“Web minimalist has come back as a trend in the summer of 2007”

Rainfall Daffinson, 2010
rainfall-daffinson.com/minimalism

Elements:

“Lots of whitespaces, smart and clean usage of fonts, carefully positioned minimalist
imagery will be main parts of this trend.
This also includes (usually) white backgrounds and spacious layouts.”
UX Passion, 2010
uxpassion.com/2010/01/design-trends-2010-web-design-trends-2010

Broader Context:

Note: repeated participant comments are marked in bold

 Technological (cross device design, mobile web faster response times)

 Reduce information overload
 Apple design influence
 Content centric
 Counter trend to decorative trends (Collage, Grunge)
 Short attention spans
 Typography (Swiss style influence)

Search Trendline:

Google Insight for Search @ December 8, 2010

 85

Appendix 2: Study 1 – Round 1

1. Invitation to participate in current web design trend expert survey

Subject: Web Design Trend Experts Survey

Dear <web design trend expert name>,

As a web design trend expert, we would like to ask for your cooperation in an
academic research concerning web design trends. This research is aimed at
identifying current web design trends and their relations to broader cultural, social or
technological trends.

If you agree to participate, please go to our online survey, in which we will present to
you a list of current web design trends and samples from each trend and ask you to
rate how up-to-date they are. We will also ask for your opinion about who today‟s
web design trendsetters are, and whether some trends are related to broader trends.
The online survey should take no more than 15 minutes of your time.

Upon conclusion of this study we will send you a report of our findings.

Survey link: http://www.surveymonkey.com/s/3RQFQKK

We greatly appreciate your cooperation!

Gili Korman Golander, Ben Gurion University, Israel

Professor Noam Tractinsky, Ben Gurion University, Israel
Ilanit Kabessa Cohen, Bezalel Academy of Arts and Design, Israel

2. Current web design trend expert survey study short instructions

 86

3. Stimuli: Images of 25 current web design trends

Cartoons\Mascots Collage

Oversized Headers Grunge

Huge Images\Image Backgrounds Magazine Layout

Messy Desk Texture | Grain

 87

Typography | Letterpress Minimalism

Nature/Green/Organic One Page Layout

Ornamental Oversized Footers

Texture | Wood Typography | Big and Bold

 88

Retro \Vintage Sketches\Hand Drawn Illustrations

Splattered Backgrounds Stationary

Clean Illustrations Texture | Bricks

Transparency Typography | Handwritten Fonts

 89

Watercolors
 90

4. Current Web Design Trends Up-to-Dateness Ranking

The table below is ordered by up-to-dateness score (highest to lowest). Each trend in
the chart and in the table is identified by a label. The table also includes an arbitrary
serial number for each trend (according to which it was presented in the survey).

Trend # Trend Name Mean SD

10 Minimalism 8.14 1.726
23 Transparency 7.82 1.790
3 Oversized Headers 7.73 1.723
12 One Page Layouts 7.64 2.105
8 Texture | Grain 7.50 2.177
9 Typography | Letterpress 7.45 1.845
5 Huge Images\Image Backgrounds 7.32 2.009
6 Magazine Layouts 7.18 1.918
16 Typography | Big and Bold 7.05 2.011
18 Sketches\Hand Drawn Illustrations 6.82 1.943
21 Clean Illustrations\Clean Illustration Backgrounds 6.68 2.124
14 Oversized Footers 6.41 2.343
1 Cartoons\Mascots 6.27 2.004
24 Typography | Handwritten Fonts 5.82 1.842
17 Retro \Vintage 5.68 2.212
13 Ornamental 5.59 1.869
25 Watercolors 5.59 2.016
2 Collage 5.50 2.559
15 Texture | Wood 5.41 2.197
11 Nature/Green/Organic 5.05 1.939
20 Stationary 4.50 2.263
7 Messy Desk 4.14 2.494
4 Grunge 4.00 2.182
22 Texture | Bricks 3.73 1.980
19 Splattered Backgrounds 3.55 1.896
 91

5. Additional Current Web Design Trends

Prominent trends that were repeated by multiple participants are marked in bold.
The links provided for each trend were suggested by the participants.

Trend Name Trend Examples

Video nike.com/jumpman23/m6
theskylinemusic.com
Animations without 20thingsilearned.com
Flash
Inventive interaction wonder-wall.com/#project/en
wearehuhes.com/People.html
journey-to-zero.com
80's influence gioseppo.com
duplostudio.com/web
3D elements and text comcasttown.com
effects
Organic layouts sietedefebrero.com
gregorywood.co.uk/journal/lets-drink-to-the-death-of-regret
gregorywood.co.uk/journal/top-5-reasons-to-learn-to-dive
Different mobile
layouts
Patterned elements designinstruct.com
galp.in
heyhush.com
thedesigncubicle.com
cafundoestudio.com.br/blog2010
Reality
 92

6. Current Web Design Trendsetters

Trendsetters include individuals, commercial firms, design agencies and design

publications. Prominent trendsetters that were mentioned by multiple participants are
marked in bold. Links to the current web design trendsetters websites are also detailed.

Trendsetting Commercial Firms

Trendsetter Name Trendsetter Website Lead Designer

Apple apple.com Jonathan Ive
Twitter twitter.com Douglas Bowman
37Signals 37signals.com Ryan Singer

Trendsetting Designers / Design Agencies / Design Publications

Trendsetter Name Trendsetter Website Trendsetter Type

Nick La WebDesignerWall.com Designer
ndesign-studio.com
Chris Coyier CSS-Tricks.com Designer
Smashing Magazine smashingmagazine.com Design Publication
Alen Grakalic CSSGlobe.com Designer
Christian “Kriesi” Budschedl kriesi.at Designer
Mike Kus mikekus.com Designer
Carsonified carsonified.com Design Agency
Clear Left clearleft.com Design Agency
Fantasy Interactive f-i.com Design Agency
Elliot Jay Stocks elliotjaystocks.com Designer
Eric Meyer meyerweb.com Designer

Non Web Design Trendsetters

Trendsetter Name Trendsetter Website Trendsetter Type

Damien Hirst damienhirst.com Artist
Alexander McQueen alexandermcqueen.com Fashion designer
 93

Appendix 3: Study 1 - Round 2

1. Invitation to participate in current web design trend expert survey

Subject: Web Design Trend Experts Survey - Follow-up

Dear <web design trend expert name>,

We are happy to provide an interim report with the results of the current web design
trend survey.

The survey, conducted in Oct-Nov 2010, is part of an academic research aimed at

identifying and characterizing current web design trends.

Subsequent to this survey, we are conducting a follow-up survey on current web

design trends.
We would greatly appreciate it if you could participate in it as well.
Link to the follow-up survey: http://www.surveymonkey.com/s/YLNYGBH

Thanks again for participating in the web design trend survey!

Gili Korman Golander, Ben Gurion University, Israel

Professor Noam Tractinsky, Ben Gurion University, Israel
Ilanit Kabessa Cohen, Bezalel Academy of Arts and Design, Israel
 94

2. Stimuli: Images of 20 current web design trends

Cartoons\Mascots Collage

Oversized Headers Huge Images\Image Backgrounds

Magazine Layout Texture | Grain

Typography | Letterpress Minimalism

 95

Nature/Green/Organic One Page Layout

Ornamental Oversized Footers

Texture | Wood Typography | Big and Bold

Retro \Vintage Sketches\Hand Drawn Illustrations

 96

Clean Illustrations Transparency

Typography | Handwritten Fonts Watercolors

 97

3. Current Web Design Trends Up-to-Dateness Dimension

The table and graph below are ordered by round 2‟s up-to-dateness rating score
(highest to lowest). Each trend in the graph and in the table is identified by a label.

Trend Name Mean SD

Minimalism 8.27 1.68
One Page Layouts 8.09 1.87
Texture | Grain 8.09 1.58
Typography | Big and Bold 8.09 2.02
Oversized Headers 7.91 1.51
Transparency 7.91 1.70
Huge Images\Image Backgrounds 7.82 1.72
Typography | Letterpress 7.73 1.56
Magazine Layouts 7.45 0.93
Clean Illustrations\Clean Illustration Backgrounds 7.18 1.60
Oversized Footers 6.73 2.33
Cartoons\Mascots 6.55 1.29
Sketches\Hand Drawn Illustrations 6.27 1.74
Typography | Handwritten Fonts 5.91 2.51
Retro \Vintage 5.82 2.40
Ornamental 5.64 2.62
Watercolors 5.64 1.21
Texture | Wood 5.55 2.02
Collage 5.27 2.72
Nature\Green\Organic 5.18 2.14
 98

4. Current Web Design Trends Simple-Complex Dimension

The graph and table below are ordered by simple-complex dimension score (lowest to
highest). Each trend in the graph and in the table is identified by a label.

Trend Name Mean SD

Typography | Big and Bold 3.00 1.87
Minimalism 3.22 2.33
One Page Layouts 3.67 2.74
Oversized Headers 4.20 2.30
Huge Images\Image Backgrounds 4.40 2.46
Texture | Wood 4.44 1.42
Cartoons\Mascots 5.00 1.55
Texture | Grain 5.00 3.02
Typography | Handwritten Fonts 5.22 2.44
Oversized Footers 5.33 2.55
Clean Illustrations\Clean Illustration Backgrounds 5.44 2.24
Typography | Letterpress 5.56 2.35
Transparency 5.67 2.29
Nature\Green\Organic 6.11 1.76
Collage 6.18 2.36
Watercolors 6.44 1.51
Sketches\Hand Drawn Illustrations 6.89 2.03
Retro \Vintage 7.11 1.45
Magazine Layouts 7.90 1.10
Ornamental 8.11 1.76
 99

5. Current Web Design Trends Impersonal -Personal Dimension

The graph and table below are ordered by impersonal-personal dimension score
(lowest to highest). Each trend in the graph and in the table is identified by a label.

Trend Name Mean SD

Magazine Layouts 4.70 2.00
Typography | Big and Bold 4.78 2.86
Ornamental 4.89 2.37
Oversized Footers 5.00 1.50
Transparency 5.11 2.15
Texture | Grain 5.30 2.95
Texture | Wood 5.33 2.18
Typography | Letterpress 5.33 2.35
Collage 5.64 2.77
Oversized Headers 5.70 1.83
Minimalism 5.89 2.62
Nature\Green\Organic 5.89 2.42
Retro \Vintage 6.33 2.40
Watercolors 6.33 2.12
Clean Illustrations\Clean Illustration Backgrounds 6.78 1.79
Typography | Handwritten Fonts 6.78 3.03
Cartoons\Mascots 7.27 1.74
Huge Images\Image Backgrounds 7.30 2.31
One Page Layouts 7.56 1.33
Sketches\Hand Drawn Illustrations 7.67 1.80
 100

6. Current Web Design Trends Direct -Subtle Dimension

The graph and table below are ordered by direct-subtle dimension score (lowest to
highest). Each trend in the graph and in the table is identified by a label.

Trend Name Mean SD

Typography | Big and Bold 2.44 2.19
Oversized Headers 2.50 2.42
Minimalism 2.67 1.73
Huge Images\Image Backgrounds 4.30 2.91
One Page Layouts 4.33 2.35
Cartoons\Mascots 4.45 2.46
Collage 4.55 1.86
Typography | Handwritten Fonts 4.67 2.40
Typography | Letterpress 4.67 2.92
Sketches\Hand Drawn Illustrations 4.67 2.55
Ornamental 4.78 2.33
Retro \Vintage 4.78 2.68
Clean Illustrations\Clean Illustration Backgrounds 5.00 2.45
Nature/Green/Organic 5.22 1.48
Texture | Wood 5.22 1.99
Watercolors 5.56 2.13
Magazine Layouts 5.60 1.96
Oversized Footers 5.67 2.45
Transparency 5.78 2.05
Texture | Grain 7.10 2.60
 101

7. Current Web Design Trends Modern -Nostalgic Dimension

The graph and table below are ordered by modern-nostalgic dimension score (lowest
to highest). Each trend in the graph and in the table is identified by a label.

Trend Name Mean SD

Transparency 2.22 0.97
Typography | Big and Bold 2.44 1.42
Minimalism 2.78 2.11
Oversized Headers 3.30 2.63
Oversized Footers 3.56 2.35
Magazine Layouts 3.70 1.34
Huge Images\Image Backgrounds 3.90 3.11
One Page Layouts 4.00 3.00
Texture | Grain 4.90 2.92
Clean Illustrations\Clean Illustration Backgrounds 5.00 2.24
Typography | Letterpress 5.00 2.83
Ornamental 5.33 3.12
Nature\Green\Organic 5.78 1.72
Cartoons\Mascots 5.82 1.94
Sketches\Hand Drawn Illustrations 6.22 1.48
Typography | Handwritten Fonts 6.33 1.80
Collage 6.91 2.02
Watercolors 7.33 1.58
Texture | Wood 7.44 1.67
Retro \Vintage 8.22 1.72
 102

8. Current Web Design Trends Correlation with Est. Duration

Correlations Input Data

Trend Name Perceived Est. Est. Simple- Impersonal- Direct- Modern-

Up-to- Date Dura- Complex Personal Subtle Nostalgic
Dateness Range tion
(years)
Cartoons\Mascots 6.55 2006-2010 5 5.00 7.27 4.45 5.82
Clean Illustrations\ 7.18 2006-2010 5 5.44 6.78 5.00 5.00
Clean Illustration
Backgrounds
Collage 5.27 2004-2010 7 6.18 5.64 4.55 6.91
Huge Images\ 7.82 2006-2010 5 4.40 7.30 4.30 3.90
Image Backgrounds
Magazine Layouts 7.45 2004-2010 7 7.90 4.70 5.60 3.70
Minimalism 8.27 2005-2010 6 3.22 5.89 2.67 2.78
Nature\Green\ 5.18 2004-2010 7 6.11 5.89 5.22 5.78
Organic
One Page Layouts 8.09 2008-2010 3 3.67 7.56 4.33 4.00
Ornamental 5.64 2006-2010 5 8.11 4.89 4.78 5.33
Oversized Footers 6.73 2006-2010 5 5.33 5.00 5.67 3.56
Oversized Headers 7.91 2006-2010 5 4.20 5.70 2.50 3.30
Retro \Vintage 5.82 2004-2010 7 7.11 6.33 4.78 8.22
Sketches\ 6.27 2007-2010 4 6.89 7.67 4.67 6.22
Hand Drawn
Illustrations
Texture | Grain 8.09 2008-2010 3 5.00 5.30 7.10 4.90
Texture | Wood 5.55 2004-2010 7 4.44 5.33 5.22 7.44
Transparency 7.91 2004-2010 7 5.67 5.11 5.78 2.22
Typography | 8.09 2004-2010 7 3.00 4.78 2.44 2.44
Big and Bold
Typography | 5.91 2008-2010 3 5.22 6.78 4.67 6.33
Handwritten Fonts
Typography | 7.73 2009-2010 2 5.56 5.33 4.67 5.00
Letterpress
Watercolors 5.64 2008-2010 3 6.44 6.33 5.56 7.33

Correlations Results

Perceived Simple- Impersonal- Direct- Modern-

Up-to- Complex Personal Subtle Nostalgic
Dateness
Est. Duration -0.186 0.077 -0.388 -0.170 -0.092
 103

Appendix 4: Study 2
1. Stimuli: Images of 26 web design trends

Current Trends (10)

Messy Desk Huge Images

Clean Illustrations Magazine Layout

Minimalism Nature\Green\Organic

Retro\Vintage One Page Layout

 104

Hand Drawn Illustrations Typography | Big and Bold

Past-Peak Trends (6)

Web 2.0 Glossy

Stock Photography Diagonal Stripes

Gradients Sunbursts
 105

Faded Trends (10)

Raw HTML Hyper Functionalism

Usability Functionalism Kilobyte Minimalism

Pixel Under Construction

Amateur Messy
 106

Starry Night Rainbow

 107

2. Measurement tools – TDS-K, DSI, CVPA

2.1. TDS-K - A Short Version of the Trendsetting Questionnaire

Source: Batinic et al., 2008

Items (English, original version):

1.1 I often read detailed articles about the latest ideas, trends, and
developments.
2.1 There are some domains where it is important for me to always be up to
date.
3.1 I like trying something new.
4.1 I often notice that I cope better than others with changing to new things.
5.1 I often take time to explain to my friends about the new things that they do
not know about.
6.1 Regarding the latest ideas, trends, and developments, I often notice that
others act in accordance with me.
7.1 I am more likely to tell my friends and acquaintances about the newest
ideas, trends, and developments than they are to tell me.
8.1 I have the impression that I am generally regarded by my friends and
acquaintances as a good source for advice concerning the newest ideas,
trends, and developments.
9.1 I usually provide my friends and acquaintances with lots of information
when we discuss the newest ideas, trends, and developments.

Items (Hebrew, adapted to this study):

 108

2.2. DSI – A Domain Specific Innovativeness Scale

Source: Goldsmith and Hofacker, 1991

Items (English, original version):

Items (Hebrew, adapted to this study):

 109

2.3. CVPA – Centrality of Visual Product Aesthetics

Source: Bloch et al, 2003

Items (English, original version):

Items (Hebrew, adapted to this study):

 110

3. Quantitative analysis results

3.1. Background Variables Analysis

3.1.1. Initial Reliability

TDS-K DSI CVPA

Cronbach‟s Alpha 0.94 0.91 0.94

3.1.2. Initial Pearson Correlations

TDS-K DSI CVPA

TDS-K 0.64 0.73
DSI 0.65

3.1.3. Factor Analysis

Factor 1 Factor 2 Factor 3

TDSK_1 0.66
TDSK_2 0.73
TDSK_3 0.46
TDSK_4 0.51
TDSK_5 0.72
TDSK_6 0.71
TDSK_7 0.85
TDSK_8 0.92
TDSK_9 0.90
TDSK_10 0.63
DSI_1 0.77
DSI_2 0.80
DSI_3 0.86
DSI_4 0.65
DSI_5 0.51
DSI_6 0.45
CVPA_1 0.55
CVPA_2 0.84
CVPA_3 0.83
CVPA_4 0.71
CVPA_5 0.72
CVPA_6 0.67
CVPA_7 0.63
CVPA_8 0.69
CVPA_9 0.68
CVPA_10 0.55
CVPA_11 0.50 0.49
CVPA_12 0.57
 111

3.1.4. Final Reliability

TDS-K DSI * CVPA*

Cronbach‟s Alpha 0.94 0.91 0.94

3.1.5. Final Pearson Correlations

TDS-K DSI* CVPA*

TDS-K 0.58 0.74
DSI* 0.58

New DSI scale after item 6 removal noted as DSI*

New CVPA scale after item 11 removal noted as CVPA*

3.1.6. Group Statistics and Independent Samples T-Tests

Group Statistics - Trendsetting (TDS-K)

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
tsdk non-designer 110 3.3900 1.35653 .12934
designer 152 4.7724 1.41876 .11508

Independent Samples T-Test - Trendsetting (TDS-K)

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. Error 95%
(2- Difference Difference Confidence
tailed) Interval of the
Difference
Lower Upper
tdsk Equal .000 .990 - 260 .000 -1.3824 .17438 - -
variances 7.928 1.7257 1.0390
assumed
Equal - 240.903 .000 -1.3824 .17312 - -
variances 7.985 1.7234 1.0413
not
assumed
 112

Group Statistics - Domain Specific Innovativeness (DSI*)

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
dsi* non-designer 110 2.7709 1.58383 .15101
designer 152 2.8211 1.41742 .11497

Independent Samples T-Test - Domain Specific Innovativeness (DSI*)

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. Error 95%
(2- Difference Difference Confidence
tailed) Interval of the
Difference
Lower Upper
dsi* Equal 2.623 .107 - 260 .788 -.05014 .18645 -.4173 .3170
variances .269
assumed
Equal - 218.891 .792 -.05014 .18980 -.4242 .3239
variances .264
not
assumed
 113

Group Statistics - Centrality of Visual Product Aesthetics (CVPA*)

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
cvpa* non-designer 110 4.4636 1.43666 .13698
designer 152 5.5574 1.17341 .09518

Independent Samples T-Test - Centrality of Visual Product Aesthetics (CVPA*)

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. Error 95%
(2- Difference Difference Confidence
tailed) Interval of the
Difference
Lower Upper
cvpa* Equal 10.108 .002 - 260 .000 -1.0938 .16152 - -
variances 6.772 1.4118 .7757
assumed
Equal - 205.137 .000 -1.0938 .16680 - -
variances 6.557 1.4226 .7649
not
assumed
 114

3.2. Dependant Variables Analysis

3.2.1. Pearson Correlations

Correlations
mean_ mean_ mean_ mean_ mean_ mean_
current_ current_ pastpeak_ pastpeak_ faded_ faded_
trend_ trend_ trend_ trend_ trend_ trend_
uptodate like uptodate like uptodate like
mean_ Pearson 1 .735** .389** .174** .218** .110
current_ Correlation
trend_ Sig. (2-tailed) .000 .000 .005 .000 .075
uptodate N 262 262 262 262 262 262
mean_ Pearson .735** 1 .357** .390** .257** .283**
current_ Correlation
trend_ Sig. (2-tailed) .000 .000 .000 .000 .000
like N 262 262 262 262 262 262
** ** ** **
mean_ Pearson .389 .357 1 .774 .533 .424**
pastpeak_ Correlation
trend_ Sig. (2-tailed) .000 .000 .000 .000 .000
uptodate N 262 262 262 262 262 262
** ** ** **
mean_ Pearson .174 .390 .774 1 .475 .488**
pastpeak_ Correlation
trend_ Sig. (2-tailed) .005 .000 .000 .000 .000
like N 262 262 262 262 262 262
mean_ Pearson .218** .257** .533** .475** 1 .777**
faded_ Correlation
trend_ Sig. (2-tailed) .000 .000 .000 .000 .000
uptodate N 262 262 262 262 262 262
** ** ** **
mean_ Pearson .110 .283 .424 .488 .777 1
faded_ Correlation
trend_ Sig. (2-tailed) .075 .000 .000 .000 .000
like N 262 262 262 262 262 262
**. Correlation is significant at the 0.01 level (2-tailed).
 115

3.3. Hypotheses Testing

3.3.1. General Linear Model (GLM)

GLM with Repeated Measures – Perceived Up-to-Dateness

Tests of Within-Subjects Effects

Measure:MEASURE_1
Source Type III df Mean F Sig.
Sum of Square
Squares
trend_up_to_datedness Sphericity 1894.373 2 947.186 1014.539 .000
Assumed
Greenhouse- 1894.373 1.906 994.107 1014.539 .000
Geisser
Huynh-Feldt 1894.373 1.927 983.227 1014.539 .000
Lower-bound 1894.373 1.000 1894.373 1014.539 .000
trend_up_to_datedness * Sphericity 112.992 2 56.496 60.513 .000
design_orientation Assumed
Greenhouse- 112.992 1.906 59.294 60.513 .000
Geisser
Huynh-Feldt 112.992 1.927 58.645 60.513 .000
Lower-bound 112.992 1.000 112.992 60.513 .000
Error Sphericity 485.479 520 .934
(trend_up_to_datedness) Assumed
Greenhouse- 485.479 495.457 .980
Geisser
Huynh-Feldt 485.479 500.939 .969
Lower-bound 485.479 260.000 1.867

Tests of Between-Subjects Effects

Measure:MEASURE_1
Transformed Variable: Average
Source Type III Sum of Squares df Mean Square F Sig.
Intercept 29986.934 1 29986.934 9463.054 .000
design_orientation 31.059 1 31.059 9.801 .002
Error 823.899 260 3.169
 116

GLM with Repeated Measures – Level of Liking

Tests of Within-Subjects Effects

Measure:MEASURE_1
Source Type III df Mean F Sig.
Sum of Square
Squares
trend_liking Sphericity Assumed 2001.996 2 1000.998 1086.604 .000
Greenhouse-Geisser 2001.996 1.976 1013.066 1086.604 .000
Huynh-Feldt 2001.996 1.999 1001.561 1086.604 .000
Lower-bound 2001.996 1.000 2001.996 1086.604 .000
trend_liking * Sphericity Assumed 133.702 2 66.851 72.568 .000
design_orientation Greenhouse-Geisser 133.702 1.976 67.657 72.568 .000
Huynh-Feldt 133.702 1.999 66.888 72.568 .000
Lower-bound 133.702 1.000 133.702 72.568 .000
Error Sphericity Assumed 479.033 520 .921
(trend_liking) Greenhouse-Geisser 479.033 513.806 .932
Huynh-Feldt 479.033 519.708 .922
Lower-bound 479.033 260.000 1.842

Tests of Between-Subjects Effects

Measure:MEASURE_1
Transformed Variable:Average
Source Type III Sum of Squares df Mean Square F Sig.
Intercept 22128.517 1 22128.517 6984.885 .000
design_orientation 54.031 1 54.031 17.055 .000
Error 823.695 260 3.168
 117

3.3.2. Group Statistics and Independent Samples T-Tests – Up-to-Dateness

Group Statistics – Perceived Up-to-Dateness - Current Trends

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
mean_current_ non-designer 110 7.1855 1.33532 .12732
trend_uptodate designer 152 7.8684 1.13473 .09204

Independent Samples T-Test – Perceived Up-to-Dateness - Current Trends

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. 95%
(2- Difference Error Confidence
tailed) Difference Interval of the
Difference
Lower Upper
mean_ Equal 2.420 .121 - 260 .000 -.68297 .15307 - -
current_ variances 4.462 .98439 .38154
trend_ assumed
uptodate Equal - 211.08 .000 -.68297 .15710 - -
variances 4.347 .99266 .37328
not
assumed
 118

Group Statistics – Perceived Up-to-Dateness - Past-Peak Trends

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
mean_pastpeak_ non-designer 110 7.6606 1.14114 .10880
trend_uptodate designer 152 6.7489 1.39364 .11304

Independent Samples T-Test – Perceived Up-to-Dateness - Past-Peak Trends

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. 95%
(2- Difference Error Confidence
tail Differen Interval of the
ed) ce Difference
Lower Upper
mean_ Equal 4.293 .039 5.629 260 .000 .91170 .16196 .59279 1.23062
pastpeak_ variances
trend_ assumed
uptodate Equal 5.811 255.999 .000 .91170 .15690 .60273 1.22067
variances
not
assumed
 119

Group Statistics – Perceived Up-to-Dateness - Faded Trends

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
mean_faded_ non-designer 110 4.5309 1.50283 .14329
trend_uptodate designer 152 3.5513 1.25861 .10209

Independent Samples T-Test - Perceived Up-to-Dateness – Faded Trends

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. 95% Confidence
(2- Differen Error Interval of the
tail ce Differen Difference
ed) ce Lower Upper
mean_ Equal 3.112 .079 5.727 260 .000 .97959 .17103 .64280 1.31638
faded_ variances
trend_ assumed
uptodate Equal 5.568 208.889 .000 .97959 .17594 .63276 1.32643
variances
not
assumed
 120

3.3.3. Group Statistics and Independent Samples T-Tests – Level of Liking

Group Statistics – Level of Liking - Current Trends

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
mean_current_ non-designer 110 6.5691 1.36124 .12979
trend_like designer 152 7.1533 1.22035 .09898

Independent Samples T-Test – Level of Liking - Current Trends

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. 95%
(2- Differen Error Confidence
tail ce Differen Interval of the
ed) ce Difference
Lower Upper
mean_ Equal 2.870 .091 - 260 .000 -.58420 .16039 - -
current_ variances 3.642 .90003 .26836
trend_ assumed
like Equal - 219.153 .000 -.58420 .16323 - -
variances 3.579 .90589 .26250
not
assumed
 121

Group Statistics – Level of Liking - Past-Peak Trends

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
mean_pastpeak_ non-designer 110 6.8515 1.32919 .12673
trend_like designer 152 5.4243 1.53543 .12454

Independent Samples T-Test – Level of Liking - Past-Peak Trends

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. 95% Confidence
(2- Differen Error Interval of the
tail ce Differen Difference
ed) ce Lower Upper
mean_ Equal 2.373 .125 7.849 260 .000 1.42717 .18183 1.06913 1.78522
pastpeak_ variances
trend_ assumed
like Equal 8.032 251.719 .000 1.42717 .17768 1.07724 1.77711
variances
not
assumed
 122

Group Statistics – Level of Liking - Faded Trends

Group Statistics
design_orientation N Mean Std. Deviation Std. Error Mean
mean_faded_ non-designer 110 3.5027 1.28645 .12266
trend_like designer 152 2.7520 .98640 .08001

Independent Samples T-Test - Level of Liking – Faded Trends

Independent Samples Test

Levene's t-test for Equality of Means
Test for
Equality of
Variances
F Sig. t df Sig. Mean Std. 95% Confidence
(2- Differen Error Interval of the
tail ce Differen Difference
ed) ce Lower Upper
mean_ Equal 6.113 .014 5.345 260 .000 .75075 .14045 .47419 1.02732
faded_ variances
trend_ assumed
like Equal 5.127 195.887 .000 .75075 .14644 .46194 1.03956
variances
not
assumed
 123

4. Web design trend perception graphs

4.1. Trend Names and Numbering

Trend Number Trend Name

1 Messy Desk
2 Huge Images
3 Clean Illustrations
4 Magazine Layout
5 Minimalism
6 Nature\Green\Organic
7 Retro\Vintage
8 One Page Layout
9 Hand Drawn Illustrations
10 Typography | Big and Bold
11 Web 2.0
12 Glossy
13 Stock Photography
14 Diagonal Stripes
15 Gradients
16 Sunbursts
17 Raw HTML
18 Hyper Functionalism
19 Usability Functionalism
20 Kilobyte Minimalism
21 Pixel
22 Under Construction
23 Amateur
24 Messy
25 Starry Night
26 Rainbow
 124

4.2. Trend Perception Graphs

Perceived Up-to-dateness of Web Design Trends – Scatter Plot

X values represent web design trend identifiers (ordered by trend number).

Y values represent the mean perceived up-to-dateness of web design trends.

Level of Liking of Web Design Trends – Scatter Plot

X values represent web design trend identifiers (ordered by trend number).

Y values represent the mean level of liking of web design trends.
 125

Perceived Up-to-Dateness of Web Design Trends

X values represent web design trend identifiers (ordered by trend number).

Y values represent the mean perceived up-to-dateness of web design trends.

Perceived Up-to-Dateness of Web Design Trends – Designers Only

X values represent web design trend identifiers (ordered by perceived up-to-dateness).

Y values represent mean perceived up-to-dateness of web design trends by designers.
 126

Perceived Up-to-dateness of Web Design Trends – Non-Designers Only

X values represent web design trend identifiers (ordered by perceived up-to-dateness).

Y values represent mean perceived up-to-dateness of web design trends by non-
designers.
 127

Current Trends – Cumulative Perceived Up-to-Dateness

X values represent the mean perceived up-to-dateness of current web design trends.
Y values represent the cumulative percent of people who rated the current web design
trends.

Current Trends – Cumulative Liking

X values represent the mean level of liking of current web design trends.
Y values represent the cumulative percent of people who rated the current web design
trends.
 128

Past-Peak Trends – Cumulative Perceived Up-to-Dateness

X values represent the mean perceived up-to-dateness of past-peak web design trends.
Y values represent the cumulative percent of people who rated the past-peak web
design trends.

Past-Peak Trends – Cumulative Liking

X values represent the mean level of liking of past-peak web design trends.
Y values represent the cumulative percent of people who rated the past-peak web
design trends.
 129

Faded Trends – Cumulative Perceived Up-to-Dateness

X values represent the mean perceived up-to-dateness of faded web design trends.
Y values represent the cumulative percent of people who rated the faded web design
trends.

Faded Trends – Cumulative Liking

X values represent the mean level of liking of faded web design trends.
Y values represent the cumulative percent of people who rated the faded web design
trends.
 130

References
Akins, R. B., Tolson H., Cole B. R. (2005). Stability of Response Characteristics of a
Delphi Panel: Application of Bootstrap Data Expansion. BMC Medical Research
Methodology. Vol 5, p. 37.

Barnard, M. (2007). Introduction; Fashion as Communication; Fashion Statements:

Communication and Culture. In M. Barnard (Ed.), 2007, Fashion theory – a reader.
London and New York: Routledge.

Batinic, B., Wolff, H. G., Haupt, C. M. (2008). Construction and Factorial Structure of
a Short Version of the Trendsetting Questionnaire (TDS-K): A Cross-Validation
Using Multigroup Confirmatory Factor Analyses. European Journal of
Psychological Assessment. Vol 24(2), 88-94.

Bell, W. (2003). Foundations of Futures Studies, vol. 1: History, Purposes and

Knowledge. Transaction Publishers, New Brunswick, USA.

Bianchi, M. (2002). Novelty, preferences, and fashion: when goods are unsettling.
Journal of Economic Behavior & Organization, Volume 47, Issue 1, January 2002,
Pages 1-18.

Bloch, P. H., Brunel, F. F., Arnold T.J. (2003). Individual Differences in the Centrality
of Visual Product Aesthetics: Concept and Measurement. Journal of Consumer
Research 2003 29:4, 551-565

Blumer, H. (1969). Fashion: From Class Differentiation to Collective Selection. In M.

Barnard (Ed.), 2007, Fashion theory – a reader. London and New York: Routledge.

Buxton, B. (2007). Sketching User Experiences: Getting the Design Right and the
Right Design. Morgan Kaufmann.

Campbell, C. (1997). When the Meaning is Not the Message: A Critique of

Consumption as Communication Thesis. In M. Barnard (Ed.), 2007, Fashion
theory – a reader. London and New York: Routledge.

Davis, F. (1992). Do Clothes Speak? What Makes Them Fashion? In M. Barnard

(Ed.), 2007, Fashion theory – a reader. London and New York: Routledge.

Dawkins, R. (1989). The Selfish Gene. Oxford New York: Oxford University Press.

Diefenbach, S. and Hassenzahl, M. (2008). Give Me a Reason: Hedonic Product

Choice and Justification. In CHI '08 Extended Abstracts on Human Factors in
Computing Systems. ACM, New York, NY, 3051-3056.

Eckert, C.M. and Stacey, M.K. (2001) “Designing in the context of fashion –
designing the fashion context”, in H.H.C.M. Christiaans and P. Lloyd (eds.)
“Design in Context: Proceedings of the 5th Design Thinking Research
Symposium”, Delft, Netherlands: Delft University Press.
 131

Engholm, I. (2001). Digital Design History and the Registration of Web Development.
In proceedings from the conference: "Preserving the Present for the Future -
Strategies for the Internet", The Royal Library, Copenhagen 18th -19th of June
2001.

Engholm, I. (2002). Digital Style History: The Development of Graphic Design on the
Internet. Digital Creativity, Vol 13, No. 4, pp. 193-211.

Engholm, I. (2007). Design History of the WWW: Website Development from the
Perspective of Genre and Style Theory. Artifact, Vol 1, No. 4, pp. 217-231.

Evans M. (1989). Consumer Behavior towards Fashion. European Journal of

Marketing, Volume 23, Number 7. Emerald Group Publishing Limited.

Faddis, M. (2006). The Death of Web Design Trends. Retrieved December 22, 2010,
from http://www.michaelfaddis.com/2006/05/the-death-of-web-design-trends.php

Fashion. (2010). In Merriam-Webster Online Dictionary. Retrieved April 30, 2010,

from http://www.merriam-webster.com/dictionary/fashion

Gladwell, M. (1997). The Coolhunt. The New-Yorker. March 17, 78-88.

Gladwell, M. (2000). The Tipping Point: How Little Things Can Make a Big
Difference. New York: Little, Brown & Company.

Goffman, E. (1959). The presentation of self in everyday life. Garden City, NY:
Doubleday.

Goldsmith R. E. and Hofacker C.F. (1991). Measuring Consumer Innovativeness.

Journal of the Academy of Marketing Science. Volume 19, Issue 3, pp. 209-221.

Green, W.S. and Jordan, P.W. (Eds.) (1999). Human Factors in Product Design:
Current Practice and Future Trends. Taylor and Francis.

Haney, C., Banks, C., Zimbardo, P. (1973). Interpersonal dynamics in a simulated

prison. International Journal of Criminology and Penology. Volume 1, pp. 69-97.
Academic Press.

Hartshorne, H. and May, M. (1928). Studies in the nature of character: Studies in

Deceit (Volume 1). New York: Macmillan.

Hassenzahl, M. (2003). The Thing and I: Understanding the Relationship between

User and Product. In Blythe M., Overbeeke C., Monk A.F. and Wright P.C. (Eds.)
Funology: From Usability to Enjoyment, pp. 31–42. Dordrecht: Kluwer.

Heath, C. and Heath, D. (2007). Made to Stick: Why Some Ideas Survive and Others
Die. New York: Random House.

Hekkert, P., Snelders, D., van Wieringen, P.C. (2003). „Most Advanced, Yet
Acceptable‟: Typicality and Novelty as Joint Predictors of Aesthetic Preference in
Industrial Design. British Journal of Psychology, Vol 94, Issue 1, pp. 111-124.
 132

Kroeber, A. (1919). On the Principle of Order in Civilization as Exemplified by

Changes in Fashion. American Anthropologist, New Series, 21(3):235-63.

Lavie, T. and Tractinsky, N. (2004). Assessing Dimensions of Perceived Visual

Aesthetics of web Sites. International Journal of Human-Computer Studies, Vol
60, Issue 3, March 2004, pp. 269-298.

Lennartz, S. (2008). Grunge Style in Modern Web Design. Smashing Magazine.

Retrieved December 22, 2010, from
http://www.smashingmagazine.com/2008/01/29/grunge-style-in-modern-web-
design

Lialina, O. (2005). The Vernacular Web. Retrieved December 11, 2010, from
http://art.teleportacia.org/observation/vernacular

Lialina, O. (2007). The Vernacular Web 2. Retrieved December 11, 2010, from
http://contemporary-home-computing.org/vernacular-web-2

Lialina, O. (2010). The Vernacular Web 3. Retrieved December 11, 2010, from
http://contemporary-home-computing.org/prof-dr-style

Linden, A. and Fenn, J. (2003). Understanding Gartner's Hype Cycles. Retrieved

February 1, 2011, from http://www.ask-force.org/web/Discourse/Linden-
HypeCycle-2003.pdf.

Liu, C. M. and Donath, J. S. (2006). Urbanhermes: Social Signaling with Electronic

Fashion. In Proceedings of the SIGCHI Conference on Human Factors in
Computing Systems (Canada, April 22 - 27, 2006). R. Grinter, T. Rodden, P. Aoki,
E. Cutrell, R. Jeffries, and G. Olson, Eds. CHI '06. ACM, New York, NY, 885-
888.

Lowe, J. and Lowe, E. (1982). Cultural Pattern and Process: A Study of Stylistic
Change in Women‟s Dress. American Anthropologist. New Series, 84(3):521-44.

Lynch, A. and Strauss, M. (2007). Changing Fashion: A Critical Analysis to Trend

Analysis and Cultural Meaning. Oxford and New York: Berg Publications.

Mannetti, L., Pierro, A. and Livi, S. (2002). Explaining Consumer Conduct: From
Planned to Self-Expressive Behavior. Journal of Applied Social Psychology, 2002,
32, 7, pp. 1431-1451. V. H. Winston & Son, Inc.

Maslow, A. (1970). The Hierarchy of Needs. In Motivation and Personality. pp. 56,
58. USA: Longman.

McCracken, G. (1985). The Trickle Down Theory Rehabilitated. In M. Solomon (Ed.),

The Psychology of Fashion. Lexington, MA: Lexington Books.

McCracken, G. (1988). Culture and Consumption: New Approaches to the Symbolic

Character of Consumer Goods and Activities. Bloomington: Indiana University
Press.
 133

Meek, L. (2007). 7 Surefire Web Design Styles that Work. Think Vitamin. Retrieved
December 22, 2010, from http://thinkvitamin.com/design/web-design-isms-7-
surefire-styles-that-work/

Nystrom, P. (1928). Economics of Fashion. New York: Roland Press Company.

Penn, M. J. and Zalesne, E. K. (2007). Microtrends: The Small Forces behind

Tomorrow's Big Changes. New York: Twelve.

Rogers, E. M. (1995). Diffusion of Innovation. 4th ed. New-York: Free Press.

Rogers, E. M. (2003). Diffusion of Innovation. 5th ed. New-York: Free Press.

Schifferstein and Zwartkruis-Pelgrim (2008). Consumer-Product Attachment:

Measurement and Design Implications. International Journal of Design. Vol. 2(3),
p. 1.

Shannon, C. and Weaver, W. (1949). The Mathematical Theory of Communication.

Champaign: University of Illinois Press.

Simmel, G. (1957). Fashion. International Quarterly, 10(1), October 1904, pp. 130-
155, reprinted in American Journal of Sociology, 62(6), May 1957, pp. 541-558.

Skulmoski, G. J., Hartman, F. T., Krahn, J. (2007). The Delphi Method for Graduate
Research. Journal of Information Technology Education. Vol 6, pp. 1-21.

Sproles, G. (1981). Analyzing Fashion Life Cycles – Principles and Perspective.

Journal of Marketing, 45: 116-24.

Sproles, G. and Burns, L. (1994). Changing Appearances: Understanding Dress in

Contemporary Society. New York: Fairchild Publications.

Suzuki, T. and Best, J. (2003). The Emergence of Trendsetters for Fashions and Fads:
Kogaru in 1990s Japan. The Sociological Quarterly. Volume 44, Number 1, pp.
61-79.

Tractinsky, N. (2006). Aesthetics in Information Technology: Motivation and Future

Research Directions. In Zhang, P. and Galletta, D. (Eds.) Human-Computer
Interaction in Management Information Systems: Foundations, pp. 330-347. M. E.
Sharpe, Inc.

Trend. (2010). Online Etymology Dictionary. Retrieved May 03, 2010, from
Dictionary.com website: http://dictionary.reference.com/browse/trend

Vejlgaard, H. (2007). Anatomy of a Trend. McGraw-Hill Professional.

Watts, J. D. (2007). The Idea is Intuitively. USA: Harvard Business Review.

Watts, J. D. (2007). The Widespread Propagation. USA: Harvard Business Review.

Wilson, E. (1985). Explaining it Away. In M. Barnard (Ed.), 2007, Fashion theory – a
reader. London and New York: Routledge.
 ‫‪II‬‬

‫הקבוצה ודירגו כל טרנד על ארבעת המימדים שהוזכרו לעיל‪ .‬סבב זה הניב מדרג עדכניות משופר‬

‫של הטרנדים וכן כימות תכונות הטרנד בהתייחס לארבע המימדים שזוהו‪.‬‬

‫שלישית‪ ,‬בדקנו הבדלים בתפישת טרנדים בעיצוב לאינטרנט בין אנשים המשתייכים‬

‫לקבוצות אימוץ חדשנות שונות‪ :‬קובעי טרנדים (‪ )Trendsetters‬ומאמצי טרנדים (‪.)Followers‬‬

‫הצגנו ‪ 22‬טרנדים בעיצוב לאינטרנט‪ ,‬משלושה שלבים במחזור הטרנד‪ :‬טרנדים שחלפו‪ ,‬טרנדים‬

‫ששיאם עבר‪ ,‬וטרנדים עכשוויים‪ .‬הנבדקים דירגו את מידת עדכניות הטרנדים ובאיזו מידה הם‬

‫אוהבים אותם‪ .‬מדדנו את נטית הנבדקים להיות קובעי טרנדים‪ ,‬את מידת החדשנות שלהם‬

‫באופנה‪ ,‬ואת מידת המרכזיות שהם מייחסים לאסתטיקה ויזואלית במוצרים‪ .‬הנבדקים היו‬

‫סטודנטים למערכות מידע (שסווגו אפריורית כמאמצי טרנדים) ומעצבים‪/‬סטודנטים לעיצוב‬

‫(שסווגו אפריורית כקובעי טרנדים)‪ .‬שיערנו שקובעי טרנדים יעריכו טרנדים עכשוויים כעדכניים‬

‫יותר וטרנדים ששיאם עבר ושחלפו כעדכניים פחות‪ .‬כמו כן שיערנו שהם יאהבו יותר טרנדים‬

‫עכשוויים מאשר טרנדים ששיאם עבר ושחלפו‪ .‬השערתנו לגבי מאמצי טרנדים הייתה הפוכה‪.‬‬

‫התוצאות תמכו בסיווג האפריורי של מעצבים כנוטים יותר להיות קובעי טרנדים ולייחס חשיבות‬

‫רבה יותר לאסטטיקה ויזואלית של מוצרים יחסית ללא‪-‬מעצבים‪ .‬תוצאות הערכת העדכניות של‬

‫הטרנדים והעדפות הנבדקים תמכו בהשערות המחקר‪ .‬אנשים אוהבים טרנדים שהם מחשיבים‬

‫לעדכניים‪ ,‬אך קובעי טרנדים מזהים טרנדים עדכניים יותר טוב ממאמצי טרנדים‪ .‬משתני רקע‬

‫שונים משפיעים על תפישת טרנדים בעיצוב לאינטרנט בשלבים שונים של מחזור הטרנד‪ ,‬אך רקע‬

‫ועיסוק בעיצוב משפיע על תפישת הטרנדים בכל שלבי מחזור הטרנד‪.‬‬

 ‫‪I‬‬

‫תקציר‬
‫אופנה היא תופעה תרבותית וחברתית רבת מימדים המאופיינת בשינויים מואצים‪ .‬טרנד‬

‫הוא מושג רחב יותר הקשור לסגנון ועיצוב‪ ,‬ונגזר ממגמה תרבותית‪ ,‬חברתית או טכנולוגית‪,‬‬

‫שאופנה היא ביטוי נקודתי שלה‪ .‬יחסם של אנשים לטרנדים משתנה; יש מובילי טרנדים ומאמצי‬

‫טרנדים בדרגות שהוי שונות‪ .‬המחקר והפרקטיקה בעיצוב מנשקי אדם מחשב שעסקו באופן‬

‫מסורתי בשימושיות‪ ,‬התרחבו להיבטים של חוויית משתמש‪ ,‬רגשות בשימוש במערכות‬

‫אינטראקטיביות‪ ,‬ואסתטיקה בעיצוב המנשק‪ .‬היבטים אלו רלבנטיים לתחום הלא חקור יחסית‬

‫של אופנה‪/‬טרנדים בעיצוב מנשקי אדם מחשב‪.‬‬

‫במחקר זה התמקדנו בהיבטים של אופנה‪/‬טרנדים בעיצוב אתרי אינטרנט‪ .‬מטרתנו‬

‫המרכזית הייתה לבדוק האם אכן קיימים טרנדים בעיצוב לאינטרנט‪ .‬לצורך כך נערך מחקר בן‬

‫שלושה חלקים‪ :‬בניית ספריית טרנדים של עיצוב לאינטרנט‪ ,‬אימותה‪ ,‬וחקירת ההבדלים בתפישת‬

‫טרנדים בעיצוב לאינטרנט בין אנשים מקבוצות אימוץ חדשנות שונות‪.‬‬

‫ראשית‪ ,‬ערכנו איסוף נתונים על טרנדים בעיצוב לאינטרנט‪ .‬מקורותינו היו סקירות‬

‫מקוונות של טרנדים בעיצוב אתרי אינטרנט וכן פרסומים אקדמיים‪ .‬נערכו מספר סבבים של עידון‬

‫המידע ואיתור דוגמאות עיצוב טיפוסיות של אתרי אינטרנט‪ .‬נתונים אלו כוללים אפיוני טרנדים‪,‬‬

‫אלמנטים שכיחים‪ ,‬תאריכי טרנדים‪ ,‬טכנולוגיות מסייעות‪ ,‬שיוך לקבוצות טרנדים ודוגמאות‪,‬‬

‫שעובדו ל"ספריה" של טרנדים בעיצוב אתרי אינטרנט מתקופות שונות‪ ,‬התומכת בקיומם של‬

‫טרנדים כאלו‪.‬‬

‫שנית‪ ,‬אימתנו את ספריית הטרנדים באמצעות מחקר דלפי בן שני סבבים‪ ,‬שפנה למומחי‬

‫טרנדים בעיצוב לאינטרנט‪ .‬בסבב הראשון‪ 22 ,‬מומחים העריכו את עדכניותם של ‪ 22‬טרנדים‬

‫עכשוויים בעיצוב לאינטרנט ונשאלו לגבי קשרים אפשריים של טרנדים אלו לטרנדים תרבותיים‪,‬‬

‫טכנולוגיים וחברתיים כלליים יותר‪ .‬בנוסף‪ ,‬הם נשאלו לגבי טרנדים נוספים‪/‬חסרים‪ ,‬ולגבי זהותם‬

‫של קובעי טרנדים בעיצוב לאינטרנט‪ .‬סבב זה הניב מדרג ראשוני של מידת העדכניות של הטרנדים‬

‫וכן רשימות של טרנדים עדכניים נוספים ושל קובעי טרנדים בעיצוב לאינטרנט‪ .‬ארבעה מימדים‬

‫לסיווג מאפייני טרנדים בעיצוב לאינטרנט תומצתו מהתוצאות‪ .‬בסבב השני ‪ 11‬מתוך המומחים‬

‫עדכנו את דירוג העדכניות שלהם ל‪ 20-‬הטרנדים העדכניים ביותר‪ ,‬בהינתן הדירוג הממוצע של‬
 ‫אוניברסיטת בן‪-‬גוריון בנגב‬
‫הפקולטה למדעי ההנדסה‬
‫המחלקה להנדסת תעשייה וניהול‬

‫היבטים של אופנה ‪ /‬טרנדים בעיצוב אתרי אינטרנט‬

‫חיבור זה מהווה חלק מהדרישות לקבלת תואר מגיסטר בהנדסה‬

‫מאת‪ :‬גילי קורמן גולנדר‬

‫מנחה‪ :‬פרופ' נעם טרקטינסקי‬

‫יועצת‪ :‬אילנית קבסה כהן‬

‫תאריך‪....................‬‬ ‫‪........................‬‬ ‫חתימת המחבר‬

‫תאריך‪.....................‬‬ ‫אישור המנחה‪/‬ים ‪.........................‬‬

‫אישור יו"ר ועדת תואר שני מחלקתית‪ .........................‬תאריך‪.....................‬‬

‫אפריל ‪1022‬‬ ‫אדר התשע"א‬

 ‫אוניברסיטת בן‪-‬גוריון בנגב‬
‫הפקולטה למדעי ההנדסה‬
‫המחלקה להנדסת תעשייה וניהול‬

‫היבטים של אופנה ‪ /‬טרנדים בעיצוב אתרי אינטרנט‬

‫חיבור זה מהווה חלק מהדרישות לקבלת תואר מגיסטר בהנדסה‬

‫מאת‪ :‬גילי קורמן גולנדר‬

‫מנחה‪ :‬פרופ' נעם טרקטינסקי‬

‫יועצת‪ :‬אילנית קבסה כהן‬

‫אפריל ‪1022‬‬ ‫אדר התשע"א‬