LATIHAN MEMAHAMI ANALISIS STATISTIK

TUGASAN KUMPULAN :
KUMPULAN RAUDATUL JANNAH
AHLI KUMPULAN :
1.

ROSNAH BINTI MD ZAIN

- KORELASI

2.

ROSAZWANI BT. HJ MUHAMMAD

- T-TEST BERKEMBAR

3.

JASNIDA BT. ABU BAKAR

- ANOVA

4.

NORAZLENA BT. AWANG

- REGRASI & KHI KUASA DUA

5.

FATEN NAJUA BT. AHMAD

- T-TEST

ANALISIS UJIAN KORELASI

TAJUK JURNAL :

Effect of Self-Esteem in the Relationship

between Stress and Substance Abuse among
Adolescents: A Mediation Outcome
Disediakan oleh :
ROSNAH BINTI MD ZAIN

METODS

Substance abuse was measured with the Drug Abuse

Screening Test

Stress was measured with the Perceived Stress Scale

Self-esteem was measured with the Rosenberg Self-esteem

scale.

ANALISIS KORELASI
ISSUES
1) What is the level of stress, self-esteem and substance abuse?
2) What is the relationship between stress, self-esteem and substance
abuse?
3) Does self-esteem mediate the relationship between stress and
substance abuse?

OBJECTIVE
The present study assessed the mediating effect of self-esteem in the
relationship between stress and substance abuse among adolescents.

HIPOTESIS
There is a relationship between self-esteem effects of stress and drugs
abuse among adolescents.

RESULT

INTERPRESTASI

The study ascertained a negative and large correlation (r=-.538,

p<.01) between stress and self-esteem, similar results (r=-

.536,p<.01) was found between self-esteem and substance abuse

and a positive and medium correlation (r=.360,p<.01) was found
between stress and substance abuse. Self-esteem partially
mediated the relationship between stress and substance abuse.
So, hypothesis null was accepted.

ANALISIS T-TEST BERKEMBAR

TAJUK JURNAL :
THE IMPACT OF CHILD CENTERED PLAY
THERAPY TRAINING ON ATTITUDE,
KNOWLEDGE AND SKILLS
Disediakan oleh :
ROSAZWANI BT. HJ. MUHAMMAD

JENIS UJIAN ANALISIS T-TEST BERKEMBAR

T-test Berkembar yang mana menggunakan mix method

iaitu qualitative dan quantitative
- Quantitative adalah menggunakan data yang dikutip
- Qualitative adalah menggunakan interview demi

mengukuhkan data yang dikutip

ANALISIS T-TEST BERKEMBAR

PERSOALAN
1. Apakah kesan latihan Child Centered Play Therapy
terhadap tingkah laku siswazah tentang kanak- kanak,
interaksi dengan kanak- kanak dan pengetahuan serta
kemahiran tentang play therapy?

2. Apakah persepsi siswazah tentang struktur dan kandungan dalam pengenalan

kursus play therapy?
OBJEKTIF
Untuk mengenal pasti kesan Child Centered Play Therapy siswazah dari segi
kelakuan, pengetahuan dan kemahiran.
HIPOTESIS
Tidak terdapat perbezaan terhadap kelakuan, pengetahuan dan kemahiran siswazah
dalam pra ujiandan pasca ujian.

KEPUTUSAN ANALISIS T-TEST BERKEMBAR

Table 5.1 Pre and Post Test Scores on the Revised Play Therapy AttitudeKnowledge- Skills Survey
(n = 13)

---------------------------------------------------------------------------------------Pretest
Subscale

Attitude

4.15

Knowledge
Skills

SD

Posttest
M

SD

.26

4.53

.26

2.34

.77

4.02

.52

2.38

.68

3.99

.41

----------------------------------------------------------------------------------------P < .001

INTERPRETASI ANALISIS T-TEST BERKEMBAR

Keputusan

Ujian T Berkembar ke atas tahap kelakuan,

pengetahuan dan kemahiran siswazah dalam pra ujian dan

pasca ujian menunjukkan (t (13) = -7.74, p<.001. Ini bermakna

hipotesis nul berjaya ditolak. Maka, keputusan yang diperolehi
menunjukkan terdapat perbezaan yang signifikan antara ujian

pra dan pasca ujian bagi aspek tersebut. Didapati bahawa

tahap kelakuan, pengetahuan dan kemahiran telah meningkat
selepas intervensi program ini.

ANALISIS UJIAN ANOVA

Tajuk Jurnal :
Comparing student success between
developmental math courses offered
online, blended, and face-to-face.
Disediakan oleh :
JASNIDA BT. ABU BAKAR

TYPE OF TEST
This

study used quantitative research methodologies

Comparisons

between learning environments on

continuous outcomes (unit tests, final exam, course
average) were made using one-way ANOVA, with
learning environment as the factor, with three levels
(Face-to-face, Blended, and Online).

Tukeys

HSD was applied following significant main

effects to identify pairwise differences.

ANALISIS UJIAN ANOVA

ISSUES
1. Does the type of learning environment relate to successful course performance as
measured by test, final exam, and course grade?
2. Does the effect of course performance depend upon attrition?
OBJECTIVE
This study used quantitative research methodologies to compare student success in
different learning environments
HYPOTESIS
There is a different type of learning environment relate to successful course
performance as measured by test, final exam, and course grade?
There is a different effect of course performance depend upon attrition?

RESULT ANOVA

Table 1

IACE Scores for All Participants by Learning Environment

N
All Subjects
Face-to-face
Blended
Online

167
58
46
63

Number
Correct
25.3
28.9
22.0
24.3

SD for Number
Correct
14.72
10.83
16.61
15.83

Percentage
52.6%
60.2%
45.8%
50.7%

SD for
Percentage
30.69
22.61
34.59
33.02

INTERPRETASI ANOVA

Note. The N column lists the number of subjects in the group. The
number correct is out of 48 problems. For Fall 2009, the success
rate across all 104 Intermediate Algebra courses at this institution
was 54.7%. The standard deviation (SD) for both the number correct
and percentage is provided.

Students in the blended learning environment had the lowest mean

scores on all assessments.

For illustrative purposes, effect sizes were calculated for the Course
Average comparisons. The effect size for the face-to-face vs. online
difference was 0.17 favoring the face-to-face environment. For the
face-to-face vs. blended comparison, the effect size was 0.53, also
favoring the face-to-face environment. The online vs. blended
comparison effect size was 0.31, favoring the online environment.

Table

2
Mean and Standard Deviation Percent Correct on Unit Tests, IACE, and Course Average
by Learning Environment (Complete Dataset N = 167)
Face-toface
(N = 58)

Blended (N = 46)

Online (N = 63)
F(2, 164)

Pairwise tests of
significanceh

70.1 (21.4)

67.9 (32.1)

77.6 (24.6)

2.21

0.113

88.7 (14.7)

69.5 (34.5)

75.0 (32.5)

6.54

0.002*

50.4 (23.1)

45.4 (32.2)

59.3 (30.3)

3.39

0.036*

67.3 (22.7)

55.6 (33.5)

64.2 (34.1)

2.00

0.138

83.4 (20.2)

62.2 (41.1)

78.2 (35.4)

5.68

0.004*

63.8 (26.9)

52.8 (38.4)

64.7 (37.7)

1.82

0.165

75.5 (27.0)

54.4 (43.3)

65.0 (43.5)

3.86

0.023*

F>B

60.2 (22.6)
68.1 (18.7)

45.8 (34.6)
54.5 (32.0)

50.7 (33.0)
63.9 (28.5)

3.13
3.40

0.046*
0.036*

F>B
F>B

Unit Test
#1a
F>B;F>O

Unit Test #2b

O>B

Unit Test
#3c

Unit Test #4d

F>B; O>B

Unit Test
#5e

Unit Test #6f

Unit Test
#7g

IACE
Course Average

Note. *p < 0.05 for main effect of Learning

Environment.aFactoring, bFunction notation and
operations, cRational Expressions, dRadicals, eComplex and
Imaginary Numbers, fSolving Quadratic Equations,
gParabolas and Circles, hPairwise significance tests via
Tukeys HSD, p < 0.05.

The number of students earning passing grades was not

significantly different between the three learning
environments: Face-to-face 59%, Blended 48%, and Online
65%, !2 (2, N = 167) = 3.26; p = 0.20.

Table 3

Mean and Standard Deviation Percent Correct on Unit Tests, IACE, and Course
Average by Learning Environment (Attrition Adjusted Dataset N=134)
Face-toface
(N = 54)

Blended (N = 32)

Online (N = 48)
F(2, 131)

72.2 (19.5)

79.3 (21.1)

85.7 (11.4)

7.53

0.001*

90.7 (8.8)

86.6 (12.8)

88.4 (10.4)

1.63

0.200

52.7 (21.5)

59.8 (26.1)

70.4 (21.5)

7.71

0.001*

70.4 (19.0)

72.2 (20.7)

78.2 (20.1)

2.07

0.130

87.5 (13.3)

83.9 (20.4)

92.4 (15.3)

2.84

0.062

67.2 (24.1)

73.9 (22.6)

81.6 (21.7)

5.04

0.008*

Pairwise tests of
significanceh

O>F

Unit Test
#1a

Unit Test #2b

O>F

Unit Test
#3c

Unit Test
#4d

Unit Test #5e

Unit Test
#6f

O>F

Unit Test 80.6

#7g
(19.7)

78.2
(28.3)

81.4
(31.9)

0.14

0.869

IACE

64.7
(16.0)

65.8
(19.4)

66.5
(19.1)

0.13

0.878

Course

71.6
(13.3)

73.1
(15.8)

78.1
(10.1)

3.37 0.038* O>F

Note. *p < 0.05 for main effect of Learning Environment.aFactoring,

bFunction notation and operations, cRational Expressions, dRadicals,
eComplex and Imaginary Numbers, fSolving Quadratic Equations,
gParabolas and Circles.hPairwise significance tests via Tukeys HSD, p <
0.05

One discrepancy among the age and gender controlled analyses was
that the Online vs. Face-to-face difference was no longer statistically
significant after control of multiple comparisons by Tukeys HSD
(difference = 5.3 SE = 2.8, t = 1.93; adjusted p-value = 0.135).

Among those students who completed the course, the number of

students earning a passing course grade was significantly different
between the three learning environments: Faceto-face 63%, blended
69%, and online 85%, !2 (2, N = 134) = 6.69; p = 0.04. However, more
students in the face-to-face environment completed the course (93%)
compared to students in the blended (70%) and online (76%) courses,
!2 (2, N = 167) = 10.01; p = 0.007.

ANALISIS KORELASI, REGRESI

& KHI KUASA DUA
TAJUK JURNAL :

BIG FIVE PERSONALITY TRAITS

A TOOL FOR MANAGING STRESS
(Tactful Management Research Journal,
Vol. 1, Issue 2, Nov. 2012)
Disediakan oleh :
NORAZLENA BT. AWANG (817917)

METOD
Jumlah

penduduk mengambil bahagian : 600 konstabel

polis Balai Polis Cawangan Jenayah Negeri Tamilnadu

jenis instrumen

i)

Police Stress Inventory(PSI) Pienaar &

Rothmann (2006)

ii)

NEO Five Factor Inventory ( NEO FFI)

Skala

Likert

Jadual 1: Profil Demografi

Jadual 2:
Perhubungan antara Tekanan Pekerjaan
dan Big Five Trait Personaliti

Small : r=.10 to .29

Medium : r=.30 to .49
Large : r=.50 to 1.0

Extraversion
Keterbukaan
Bersetuju
Sifat berhati-hati
Stress kerja

Keputusan ini menunjukkan bahawa jumlah tekanan kerja yang signifikan dengan extraversion
(r=0.849, kesan yang kuat), keterbukaan (r=0.886, kesan yang kuat), agreeableness (r=0.859,
Kesan yang kuat), sifat berhati-hati (r=0.639, kesan sederhana) dan dengan stres kerja
(r=0.401, kesan sederhana).

REGRESSION:
OBJEKTIF :
Menentukan faktor personaliti yang
menjadi peramal kepada tekanan kerja.
PERSOALAN :
Adakah terdapat pengaruh personaliti
terhadap tekanan kerja?
HIPOTESIS :
Sifat-sifat personaliti yang menjadi
faktor peramal kepada tekanan kerja

Tekanan
Kerja

Personaliti

Jadual 3 :

Analisis Regresi Antara Tekanan

Pekerjaan dan Big Five Trait Personaliti

Hasil kajian menunjukkan terdapat pengaruh antara traits personaliti

dalam perkhidmatan terhadap tekanan, iaitu (r(70)= .837, p>.05). Ini
bermakna, hipotesis nul berjaya diterima.

ANALISIS KHI KUASA DUA

OBJEKTIF :
Untuk mengenalpasti perkaitan antara
tekanan kerja dengan kelompok umur,
jantina, status perkahwinan, tempoh
berkhidmat dan pendapatan bulanan .
PERSOALAN :
Adakah terdapat perkaitan antara Profil
Demografi, personaliti terhadap tekanan
kerja?
HIPOTESIS :
Tidak Terdapat perkaitan antara tekanan
kerja dengan kelompok mengikut umur,
jantina, status perkahwinan, tempoh
berkhidmat dan pendapatan bulanan.

Profil Demografi

Tekanan
Kerja

Personaliti

Jadual 4 :
Khi Kuasa Dua antara Profil Demografi
dan Tekanan Pekerjaan

Dari jadual di atas menunjukkan bahawa semua nilai yang ketara

adalah kurang daripada 0.05 , oleh itu semua detail demografi
Konstabel Polis adalah signifikan antara tekanan pekerjaan.

Jadual 5 :
Khi Kuasa Dua antara Profil Demografi
dan Personaliti

Dari jadual di atas menunjukkan bahawa semua nilai

yang ketara kurang daripada 0.05, oleh itu semua
butir-butir demografi konstabel polis ketara kaitan
antara personaliti.

PERBINCANGAN & KESIMPULAN

Kajian ini bertujuan untuk mengetahui hubungan antara

personaliti dan tekanan kerja antara strategi menghadapi
penyakit.

Hipotesis 1 menyatakan bahawa semua profil demografi

konstabel polis signifikan dengan tekanan pekerjaan.

Hipotesis 2 menyatakan bahawa semua profil demografi

konstabel polis signifikan dengan personaliti.

Hipotesis 3 menyatakan bahawa tekanan dan personaliti

sifat-sifat pekerjaan yang positif.

PERBINCANGAN & KESIMPULAN

Personaliti dapat meramalkan tekanan pekerjaan. Semua ciri-ciri

personaliti yang sangat berguna untuk descript tekanan polis.

Selain itu, ciri-ciri personaliti yang lebih meramalkan kuasa

daripada mana-mana pembolehubah konteks lain. Keputusan ini
menyokong keperluan untuk mengambil kira ciri-ciri personaliti
menyedari dan bergerak ke arah tekanan polis.

Akhirnya penyelidikan dalam tekanan polis perlu pergi jauh ke

dalam pembolehubah kontekstual seperti permintaan pekerjaan,
kekurangan atau sumber dan tekanan polis, keperluan fizikal dan
psikologi, dan peranan kecerdasan emosi dalam mengurangkan
tekanan.

Kajian ini tertumpu di selatan Tamilnadu konstabel polis. Dalam

pelbagai masa depan penyelidikan perlu dilakukan di semua
dilakukan di semua peringkat peribadi polis termasuk pegawai
peringkat tertinggi.

Using the Students t-test

with extremely small
sample sizes
Disediakan oleh :

FATEN NAJUA BT. AHMAD

ANALISIS T-TEST
Using the Students t-test with
extremely small sample sizes
(Using the Students t-test with
extremely small sample sizes Volume
18, Number 10, August 2013)
Disediakan oleh :
FATEN NAJUA BT. AHMAD

METHOD

Simulations were conducted to determine the statistical power and Type I

error rate of the onesample and two-sample t-tests. Sampling was done
from a normally distributed population with a mean of 0 and a standard
deviation of 1.

One of the two distributions was shifted with a value of Dwith respect to 0.
A sample was drawn from the population, and submitted to either the onesample t-test with a reference value of zero, or to the two-sample t-test
for testing the null hypothesis that the populations have equal means.

The simulations were carried out for Ds between 0 (i.e., the null
hypothesis holds) and 40 (i.e., the alternative hypothesis holds with an
extremely large effect), and for N = 2, N = 3, and N = 5. In the twosample
t-test, both samples were of equal size (i.e., N = M). A p value below 0.05
was considered statistically significant. All analyses were two-tailed. Each
case was repeated 100,000 times.

OBJECTIVE
1. behavior of the Welch test and a rank-transformation prior to conducting the t-test (t-testR).
2. Simulations were conducted to determine the statistical power and Type I error rate of the one
sample and two-sample t-tests.

3. Sampling was done from a normally distributed population with a mean of 0 and a standard
deviation of 1. One of the two distributions was shifted with a value of D with respect to 0. A sample
was drawn from the population, and submitted to either the one-sample t-test.
ISSUES

1.The present simulation study estimated the type 1 error rate and statistical power of the one- and
two sample t tests for normally distributed populations and various distortions such as unequal
sample sizes , unequal variences, the combination of unequel sample sizes and unequel variences
and lognormal population distribution
HIPOTESIS
There is difference type one error rate and statistical power of the one- and two sample t tests for
normally distributed populations.

RESULTS

DISCUSSION

The present simulation study showed that there is no fundamental objection to using a
regular t-test with extremely small sample sizes. Even a sample size as small as 2 did not
pose problems. In most of the simulated cases, the Type I error rate did not exceed the
nominal value of 5%. A paired t-test is also feasible with extremely small sample sizes,
particularly when the within-pair correlation coefficient is high.

A high Type I error rate was observed for unequal variances combined with unequal sample
sizes (withthe smaller sample drawn from the high variance population), and for a onesample t-test applied to nonnormally distributed data. The simulations further clarified that
when the sample size is extremely small,

Type II errors can only be avoided if the effect size is extremely large. In other words,
conducting a t-test with extremely small samples is feasible, as long as the true effect size is
large

The fact that the t-test functions properly for extremely small sample sizes may come as no
surprise to the informed reader. In fact, William Sealy Gosset (working under the pen
nameStudent) developed the t-test especially for small sample sizes (Student, 1908; for
reviews see Box, 1987; Lehmann, 2012; Welch, 1958; Zabell, 2008), a condition where the
traditional ztest provides a high rate of false positives. Student himself verified his tdistribution on anthropometric data of 3,000 criminals, which he randomly divided into 750
samples each having a sample size of 4.

SEKIAN, TERIMA KASIH.