You are on page 1of 16

INT’L. J. PSYCHIATRY IN MEDICINE, Vol.

44(3) 183-198, 2012

PSYCHOMETRIC PROPERTIES OF THE TURKISH


VERSION OF THE 42 ITEM DEPRESSION ANXIETY
STRESS SCALE (DASS-42) IN A CLINICAL SAMPLE

LEVENT HEKIMOGLU
ZEREN OZTURK ALTUN
EMINE ZEYNEP KAYA
Bursa Yuksek Ihtisas Education & Research Hospital
NURAN BAYRAM

NAZAN BILGEL
Uludag University

ABSTRACT
Objective: To study the psychometric properties of the Turkish translation
of the Depression Anxiety and Stress Scale (DASS-42) in a clinical group.
Methods: Outpatients diagnosed with anxiety (n = 138; mean age = 44.5
years; 74.6% female) or depression (n = 112; mean age = 46.2 years; 77.7%
female) from the psychiatric outpatient clinic of a public hospital were eval-
uated. A group of non-clinical volunteers (n = 250; mean age = 37 years; 68%
female) served as a community group for comparison. The participants
completed the Turkish versions of the DASS-42, the Beck Depression
Inventory (BDI), and the Beck Anxiety Inventory (BAI). Results: The struc-
ture of the DASS-42 was analyzed in the clinical sample using principal com-
ponents extraction. The three-factor solution accounted for 56% of the total
variance, with eigenvalues of 17.6, 3.0, and 2.6. The range of factor loadings
was 0.55-0.85 for depression, 0.47-0.62 for anxiety, and 0.49-0.74 for stress.
The Cronbach alpha values for the DASS depression, anxiety, and stress

183

Ó 2012, Baywood Publishing Co., Inc.


doi: http://dx.doi.org/10.2190/PM.44.3.a
http://baywood.com
184 / HEKIMOGLU ET AL.

subscales were 0.94, 0.88, 0.94 respectively. The concurrent validity of the
DASS was satisfactory. The non-clincal participants scored lower on all three
subscales than the individuals in all of the clinical groups. Conclusion: The
Turkish version of the DASS-42 appears to be an excellent instrument for
measuring features of depression, hyperarousal, and tension in clinical groups.
(Int'l. J. Psychiatry in Medicine 2012;44:183-198)

Key Words: DASS-42 scale, depression, anxiety; stress, outpatient

INTRODUCTION

Anxiety and depression are conceptually distinct from one another, although
studies to quantify these constructs have often demonstrated a high degree of
intercorrelation between measures of anxiety and depression [1-3]. Clark and
Watson suggested in their tripartite model that anxiety and depression have both
shared and unique features [1]. In this model, anxiety and depression share
symptoms of elevated negative affect (distress, irritability), they differ with
respect to the absence of positive affect which is specific to depression, and the
presence of physiological hyperarousal which is specific to anxiety.
Traditional scales have failed to make clear distinctions between anxiety and
depression [4]. Studies have shown that the popular Hamilton scales for anxiety
and depression overlapped greatly in content and were correlated with one another
quite highly [5]. The Spielberger State-Trait Anxiety Inventory is sensitive both to
symptoms of depression and anxiety [6]. The Beck Anxiety Inventory has
limitations in encompassing some important features of anxiety such as worry,
agitation, and muscle tension [7].
The researchers S. H. Lovibond and P. F. Lovibond conducted psychometric
evaluations of a questionnaire that they had developed to assess the full range
of core symptoms of anxiety and depression while providing maximum dis-
crimination between the scales of anxiety and depression [8, 9]. Although the
researchers intended to develop a measure for anxiety and depression, a third
factor emerged from their analyses of scale structure, consisting of items relating
to difficulty relaxing, irritability, and agitation and these three scales of 14 items
each were named the Depression Anxiety Stress Scales (DASS) [9]. In DASS,
Depression is characterized by low positive affect, loss of self-esteem, and a sense
of hopelessness. Anxiety includes items that are primarily related to symptoms of
physical arousal, panic attacks, and fearfulness, while Stress is characterized by
persistent tension, irritability, and a tendency to overreact to stressful events.
During the development of the DASS, non-specific items that loaded on both
Depression and Anxiety factors (e.g., appetite and sleep disturbance, guilt, dif-
ficulty concentrating) were deliberately excluded [8, 9]. Studies addressing the
psychometric properties of the DASS with non-clinical and clinical samples have
TURKISH VERSION OF DEPRESSION ANXIETY STRESS SCALE / 185

suggested that the DASS has excellent internal consistency and convergent and
discriminant validity and provides a better separation of the features of anxiety and
depression relative to other existing measures [9-13].
The DASS instrument was translated into several languages and assessed for its
psychometric properties in these languages. Studies revealed good reliability and
validity [14-21]. The Turkish version of the DASS-42 instrument was developed
by Uncu, Bayram, and Bilgel [22]. The psychometric properties of the DASS-42
in Turkish were assessed on an adult student sample and showed good reliability
and validity [23]. High Cronbach alpha values such as 0.92, 0.86, and 0.88 for
depression, anxiety, and stress respectively, revealed high internal consistency
[23]. Construct validity measured by item-scale correlations ranged from 0.48 to
0.70 for depression, from 0.33 to 0.59 for anxiety, and from 0.43 to 0.70 for stress
and convergent validity was assessed by correlations between the DASS-42 and
Hospital Anxiety and Depression (HAD) scales and showed a good convergent
validity [23]. Confirmatory factor analysis (CFA) of the Turkish DASS-42 was
consistent with previous empirical findings; the depression, anxiety, and stress
scales represent legitimate constructs in their own right [23].
Although existing data provide strong support for the psychometric and con-
ceptual basis of the Turkish version of DASS-42, all the previous research was
conducted using non-patient adult student samples. The present study aimed to
replicate and expand on previous studies of the Turkish DASS-42 by examining
the psychometric properties of the instrument in clinical and community samples.
The study examined the reliability, factor structure, and the discriminatory
properties of the Turkish DASS-42 in a clinical sample and in a non-clinical adult
comparison group.

METHODS

Study Participants
The study participants were outpatients diagnosed with anxiety (n = 138; mean
age = 44.5 years; 74.6% female) or depression (n = 112; mean age = 46.2 years;
77.7% female) from the psychiatric outpatient clinic of a state hospital. A group of
non-clinical volunteers (n = 250; mean age = 37 years; 68% female) served as a
community group for comparison. The participants in the community sample
were selected from individuals accompanying outpatients from all the outpatient
clinics of the same hospital. All participants provided informed consent and were
required to be between the ages of 18 and 65 years. Participants with current
diagnoses of substance abuse, dependence, psychotic disorder, or bipolar disorder
were excluded from the study. Diagnoses in the patient group were based on
DSM-IV criteria made by a psychiatrist and to ensure that the participants in the
non-clinical community sample did not have any of the major forms of psycho-
pathology, they were interviewed by a clinical psychologist. The participants
186 / HEKIMOGLU ET AL.

completed the Turkish versions of the DASS-42, the Beck Depression Inventory
(BDI), and the Beck Anxiety Inventory (BAI) [23-25]. In order to evaluate the
test-retest reliability of the DASS-42, 25 patients were randomly selected for
re-administration of the DASS-42 2 weeks following their initial evaluation.

Instruments

The Turkish version of the DASS-42 instrument is the same as the original in
terms of item content and order and has good reliability and validity as has been
previously mentioned [22, 23]. DASS-42 measures current (within the past week)
symptoms of depression, anxiety, and stress. Each of the three scales consists of 14
items answered by using a 0-3 scale (0 = did not apply to me at all, 3 = applied to
me very much or most of the time), thus giving a range of possible scores for each
scale of 0-42. Scores considered to be in the normal range are 0-9 for depression,
0-7 for anxiety, and 0-14 for stress. Scores above these ranges indicate the degree
of the problem from mild to extreme.
The Turkish version of the BDI is the same as the original revised form (1979),
in terms of item content and order and has a good reliability and validity [24].
The Turkish version of the BAI is a 21-item scale designed to measure symptoms
of clinical anxiety and is the same as the original in terms of item content and order.
The Turkish version of BAI has been validated for the Turkish language [25].

RESULTS

Factor Analyses

We used an exploratory factor analysis to examine the structure of the DASS-42


in the entire clinical sample (N = 250), excluding the non-clinical controls. We used
principal components extraction with varimax rotation (Kaiser normalization) and
the number of factors was determined by eigenvalues. The three-factor solution
accounted for 56% of the total variance, with eigenvalues of 17.6, 3.0, and 2.6. The
range of factor loadings was 0.55-0.85 for depression, 0.47-0.62 for anxiety, and
0.49-0.74 for stress. The factor loadings for each item are displayed in Table 1.
Items with loadings of 0.40 or greater were considered to load on a particular
factor. The first factor corresponded to the DASS-Depression scale, with all 14
items that make up this scale loading on the first factor. Two items on the DASS-D
scale had a complex structure, also loading on the anxiety factor. The second
factor paralleled the DASS-Anxiety scale. Each of the 14 items loaded on this
factor, and no items had complex loadings. The final factor corresponded to the
DASS-Stress scale with all 14 items that make up this scale loading on the third
factor. Two items on the DASS-S scale had a complex structure, also loading on
the depression factor.
Table 1. Items and Factor Loadings for the Depression Anxiety Stress Scale (DASS-42) Items

Item and item no. (in parentheses) Factor 1 Factor 2 Factor 3

DASS-Depression Scale

I felt down-hearted and blue (26) .846 (.839) .163 .121


I could see nothing to be hopeful about (37) .809 .132 .207
I felt that life was meaningless (38) .809 .091 .231
I felt sad and depressed (13) .798 .256 .017
I felt I had lost interest in just about everything (16) .782 .450 .020
I felt that I had nothing to look forward to (10) .775 .360 .043
I couldn’t seem to experience any positive feeling at all (3) .735 .461 .051
I couldn’t seem to get any enjoyment out of the things I did (24) .730 .391 .097
I felt that life wasn’t worthwhile (21) .729 .078 .190
I just couldn’t seem to get going (5) .679 .264 .182
I was unable to become enthusiastic about anything (31) .677 .347 .025
I felt I was pretty worthless (34) .570 .018 .386
I felt I wasn’t worth much as a person (17) .569 .108 .269
I found it difficult to work up the initiative to do things (42) .547 .149 .393

DASS-Anxiety Scale

I was worried about situations in which I might make a fool of myself (40) .149 .623 .090
I felt terrified (36) .051 .614 .039
I felt scared without any good reason (20) .102 .598 .259
I was aware of the action of my heart in the absence of physical exertion (25) .331 .595 .231
TURKISH VERSION OF DEPRESSION ANXIETY STRESS SCALE

I was aware of dryness of mouth (2) .354 .579 .139


I found myself in situations which made me so anxious I was most relieved when they ended (9) .031 .531 .145
/ 187

I feared that I would be thrown by some trivial but unfamiliar task (30) .346 .525 .128
Table 1. (Cont’d.)

DASS-Anxiety Scale Factor 1 Factor 2 Factor 3

I had a feeling of shakiness (7) .331 .515 .234


I experienced breathing difficulty (4) .361 .505 .198
I perspired noticeably in the absence of high temperatures or physical exertion (19) .239 .498 .240
I had a feeling of faintness (15) .133 .494 .318
I felt I was close to panic (28) .008 .489 .238
188 / HEKIMOGLU ET AL.

I had difficulty in swallowing (23) .181 .488 .293


I experienced trembling (41) .184 .474 .166

DASS-Stress Scale

I found that I was very irritable (27) .409 .013 .740


I felt that I was using a lot of nervous energy (12) .437 .007 .715
I tended to over react to situations (6) .350 .022 .704
I was in a state of nervous tension (33) .369 .005 .699
I found it hard to calm down after something upset me (29) .366 .089 .682
I found myself getting impatient when delayed in any way (14) .061 .250 .658
I found it difficult to relax (8) .380 .105 .636
I found it hard to wind down (22) .381 .009 .623
I found myself getting upset rather easily (11) .287 .047 .606
I found myself getting upset by quite trivial things (1) .317 .043 .575
I felt I was rather touchy (18) .262 .060 .571
I was intolerant of anything that kept me from getting on with what I was doing (35) .077 .305 .511
I found myself getting agitated (39) .008 .129 .505
I found it difficult to tolerate interruptions to what I was doing (32) .005 .187 .492
Note: Principal components extraction with varimax rotation (Kaiser normalization). Factor loadings of ³ 0.40 are presented in bold-face type.
TURKISH VERSION OF DEPRESSION ANXIETY STRESS SCALE / 189

Correlations among factors were as follows: stress and depression factors were
correlated at 0.75; stress and anxiety were correlated at 0.72; and anxiety and
depression were correlated at 0.66. Correlations between the DASS scale scores
are shown in Table 2.
We built three models for confirmatory factor analyses. The first model
accepted depression, anxiety, and stress as independent factors, whereas the
second model conceived them as correlated factors. The third model was identical
to the second model, with the addition of correlated error. To assess the model fit,
the goodness-of-fit index (GFI), the comparative fit index (CFI), and the root
mean square error of approximation (RMSEA) were used [26-28]. The GFI and
CFI are normed indexes, with lower bounds of zero and increasing toward unity
with improved fit of the model. Higher values of the GFI and CFI indicate
improved fit. The RMSEA is an evaluation statistic that is relatively unaf-
fected by sample size, and is suitable for assessing models of differing complexity
[29]. There are no definitive cut-off points for an acceptable model fit when using
these indices, but the following criteria are frequently used to indicate the
goodness of fit for a particular model: GFI > 0.90; CFI > 0.90; and RMSEA < 0.05.
The fit statistics for the CFA models are presented in Table 3. It can be seen
that the best fitting results came from model three, which was identical to model
two, but permitted correlated error in addition. This model was associated
with the optimal fit according to all criteria, with high fit indices and a c2 value
that, although statistically significant, was substantially lower than those of the
other models.

Reliability and Concurrent Validity

To assess the internal consistency of the DASS-42 in the entire sample,


Cronbach alpha values were computed. The Cronbach alpha values for the DASS
Depression, Anxiety, and Stress subscales were 0.94, 0.88, and 0.94 respectively.
As noted above, 25 patients (12 diagnosed with anxiety disorder and 13 diagnosed
with depressive disorder) were re-administered the DASS-42 2 weeks following
their initial evaluation. The actual correlation coefficients for initial and secondary
evaluation for depression, anxiety, and stress scores were: r = 0.81, p < 0.001;
r = 0.64, p < 0.001; and r = 0.75, p < 0.001 respectively. The results of paired
t-tests were nonsignificant for all three DASS-42 scales, which indicated that all
three scales had a favorable temporal stability.
To examine the concurrent validity of the DASS-42 scales, correlations with
other measures of depression and anxiety were computed. Correlations between
the various DASS-42 Scales and the BDI and the BAI were computed in the entire
study group (N = 500). The DASS-A scale correlated most highly with the BAI
and the DASS-D scale correlated most strongly with BDI. The resulting corre-
lation coefficients are given in Table 4.
Table 2. Item-scale Correlations for the Turkish Version of DASS-42
Item-scale Item-scale Item-scale
DASS-Depression items correlations DASS-Anxiety items correlations DASS-Stress items correlations

DASS 3 no positive feeling .77 DASS 2 dryness of mouth .60 DASS 1 upset .63
DASS 5 not seem to get going .70 DASS 4 breathing difficulty .66 DASS 6 over-react .80
DASS 10 to look forward to .80 DASS 7 feeling of shakiness .67 DASS 8 difficult to relax .77
DASS 13 sad and depressed .78 DASS 9 anxious .45 DASS 11 getting upset easily .79
190 / HEKIMOGLU ET AL.

DASS 16 lost interest about .84 DASS 15 feeling of faintness .51 DASS 12 using nervous energy .83
everything
DASS 17 I felt I wasn’t worth much .47 DASS 19 perspired .59 DASS 14 getting impatient .56
as a person
DASS 21 wasn’t worthwhile .69 DASS 20 scared .50 DASS 18 feeling touchy .75
DASS 24 I couldn’t seem to get any .78 DASS 23 difficulty in swallowing .64 DASS 22 hard to wind down .80
enjoyment out of the things I did
DASS 26 downhearted and blue .82 DASS 25 awareness of heart .69 DASS 27 irritable .84
action
DASS 31 unable to be enthusiastic .71 DASS 28 panic .50 DASS 29 hard to calm down .75
DASS 34 worthless .52 DASS 30 fear of being thrown .56 DASS 32 difficult to tolerate .41
DASS 37 I could see nothing in the .81 DASS 36 terrified .47 DASS 33 state of nervous tension .83
future to be hopeful about
DASS 38 life was meaningless .79 DASS 40 worried .48 DASS 35 intolerant .48
DASS 42 difficult to work .48 DASS 41 trembling .64 DASS 39 agitated .40
TURKISH VERSION OF DEPRESSION ANXIETY STRESS SCALE / 191

Table 3. Fit Indices for CFA Models of DASS-42 Turkish Version

Model c2 df c2/df GFI CFI RMSEA

1. Independent factors 4672.01 819 5.71 .76 .75 .07


2. Correlated factors 3803.66 816 4.66 .82 .84 .06
3. Correlated factors and
correlated error 3020.04 807 3.74 .91 .94 .04

Table 4. Correlations of the DASS-42 Scales with BDI and BAI


Measures in the Entire Sample

Measure DASS-Stress DASS-Depression DASS-Anxiety

DASS-Depression .75* — —
DASS-Anxiety .72* .66* —
BDI .81* .86* .67*
BAI .75* .66* .89*
Note: *Correlation is significant at the 0.01 level (2-tailed).

Comparing DASS-42 Scores Across Groups

To compare DASS-42 scores across groups we conducted a series of one-way


analyses of variance (ANOVAs). As indicated in Table 5, the pattern of findings
was identical for DASS-42. For all three subscales, the overall ANOVAs were
significant. Two diagnostic groups (anxiety and depression), as well as the
non-clinical volunteers group, were compared. Individuals in the non-clinical
group scored lower on all three subscales than individuals in all of the clinical
groups. Patients in the depressive disorder condition tended to score highest on the
Stress and Depression subscales, whereas individuals in the anxiety disorder
group scored highest on the Stress and Anxiety subscales.

Influence of Demographic Variables on DASS Scores

To analyze the DASS scales relationships with demographic variables, inde-


pendent samples t-test, ANOVAs, and correlations were performed for the entire
sample (N = 500). The results of the correlations are shown in Table 6. Among
demographic variables, income was not significantly correlated to any of the
scales, whereas gender and marital status was significantly correlated to depres-
sion. The remaining variables of age and education were significantly correlated to
all of the scales.
192 / HEKIMOGLU ET AL.

Table 5. Comparisons of Diagnostic Groups and Controls on the DASS-42

Clinical diagnosis

Non-clinical
Anxiety Depression group
Subscale (N = 138) (N = 112) (N = 250) F

Depression
Mean 14.37 21.04 5.47 354.07*
SD 5.00 5.64 6.02
Anxiety
Mean 16.68 9.97 4.89 151.94*
SD 5.36 5.35 5.05
Stress
Mean 24.81 22.53 10.32 394.46*
SD 4.05 4.36 6.37
Note: p < 0.0001.

Table 6. Pearson Correlations between Demographic


Variables and DASS Scores

DASS

Demographic
variable Depression Anxiety Stress

Age .266** .148** .194**


Gender .130** .081 .029
Education –.498** –.368** –.442**
Marital status .099* .052 .059
Income –.004 –.070 .073
Note: *Correlation is significant at .05 level (two-tailed).
**Correlation is significant at .01 level (two-tailed).

Independent samples t-tests revealed that females obtained significantly higher


scores than males on the depression scale (M = 12.77, SD = 8.91 [females]; M =
10.23, SD = 8.34 [males]; t = –2.920, p = 0.004). The difference between males
and females on the anxiety (M = 7.79, SD = 6.46 [males]; M = 8.98, SD = 6.59
[females]; t = –1.810, p = 0.070) and stress scales did not achieve statistical sig-
nificance (M = 16.65, SD = 9.19 [males]; M = 17.21.7, SD = 8.45 [females]; t =
–.647, p = 0.518). The mean depression score was significantly higher among
married participants (M = 9.98, SD = 8.65 [single]; M = 12.42, SD = 8.81
TURKISH VERSION OF DEPRESSION ANXIETY STRESS SCALE / 193

[married]; t = –2.225, p = 0.027). In terms of anxiety and stress scores, no


significant differences were found.
The influence of the remaining demographic variables (age and education) on the
DASS Anxiety, Depression, and Stress Scales was tested through ANOVAs. The
results are presented in Table 7 and Table 8. Table 7 shows that with an increasing
age depression, anxiety, and stress scores were increasing, whereas with higher
educational attainment they were decreasing. Hence, in our study group most of
the primary school educated participants were aged 40 years of age and over
(73.0%)—we wanted to test if age is really an important demographic factor for
depression, anxiety, and stress or if it is influenced by educational attainment.
Therefore, we performed another ANOVA analysis which is shown in Table 8.

Table 7. ANOVA Results of Age and Education on the DASS Anxiety,


Depression, and Stress Scales

Mean ± SD F p

D
E Age groups 18-29 9.81 ± 8.68
P 30-39 9.71 ± 7.92 15.380 0.001
R 40-49 13.18 ± 8.96
E 50 and over 15.74 ± 8.49
S
S Education Primary School 17.00 ± 7.62
I High School 10.02 ± 7.89 85.372 0.001
O University or more 6.27 ± 7.25
N

Age groups 18-29 7.87 ± 6.45


A
30-39 7.56 ± 6.53 4.832 0.003
N
40-49 8.99 ± 6.73
X
50 and 0ver 10.29 ± 6.40
I
E
Education Primary school 11.20 ± 6.49
T
High school 7.97 ± 6.24 39.070 0.001
Y
University or more 5.14 ± 5.35

Age groups 18-29 16.04 ± 9.26


S 30-39 14.73 ± 8.29 10.923 0.001
T 40-49 17.55 ± 8.81
R 50 and over 20.25 ± 7.45
E
S Education Primary school 21.07 ± 7.34
S High school 16.03 ± 8.90 60.417 0.001
University or more 11.51 ± 6.82
Table 8. Distribution of Mean Depression, Anxiety, and Stress Scores According to Age Groups
and Education Together with ANOVA Results

Depression Anxiety Stress


Age groups Education (Mean ± SD) F (p) (Mean ± SD) F (p) (Mean ± SD) F (p)
Primary 15.77 ± 6.31 12.72 ± 6.43 24.81 ± 6.97
18-29 Secondary 9.22 ± 8.04 8.263 7.25 ± 6.14 9.802 14.10 ± 9.04 16.118
(N = 106) High 6.94 ± 9.16 (0.000) 5.75 ± 5.43 (0.000) 13.27 ± 7.55 (0.000)
194 / HEKIMOGLU ET AL.

Primary 17.57 ± 7.81 11.54 ± 6.72 21.43 ± 6.35


30-39 Secondary 8.69 ± 6.31 33.293 7.64 ± 6.18 11.981 14.84 ± 8.45 23.487
(N = 157) High 6.24 ± 6.29 (0.000) 5.19 ± 5.67 (0.000) 10.80 ± 6.55 (0.000)

Primary 17.04 ± 7.78 10.91 ± 6.45 20.46 ± 7.48


³ 40 Secondary 12.08 ± 8.91 27.094 8.91 ± 6.38 13.148 18.89 ± 8.66 17.366
(N = 237) High 5.36 ± 6.50 (0.000) 4.16 ± 4.43 (0.000) 10.72 ± 6.14 (0.000)

Depression Anxiety Stress


Education Age groups (Mean ± SD) F(p) (Mean ± SD) F(p) (Mean ± SD) F(p)
18-29 15.77 ± 6.31 12.72 ± 6.43 24.81 ± 6.97
Primary 30-39 17.57 ± 7.81 0.384 11.54 ± 6.72 0.804 21.42 ± 6.34 3.513
(N = 211) ³ 40 17.04 ± 7.77 (0.682) 10.91 ± 6.45 (0.449) 20.46 ± 7.48 (0.032)

18-29 9.22 ± 8.04 7.25 ± 6.13 14.10 ± 9.04


Secondary 30-39 8.68 ± 6.31 3.179 7.63 ± 6.18 1.067 14.83 ± 8.45 4.898
(N = 167) ³ 40 12.08 ± 8.90 (0.044) 8.91 ± 6.38 (0.346) 18.89 ± 8.66 (0.009)

18-29 6.94 ± 9.16 5.75 ± 5.43 13.27 ± 7.55


High 30-39 6.24 ± 6.29 0.349 5.19 ± 5.66 0.652 10.80 ± 6.55 1.722
(N = 122) ³ 40 5.36 ± 6.50 (0.706) 4.16 ± 4.43 (0.523) 10.72 ± 6.14 (0.183)
TURKISH VERSION OF DEPRESSION ANXIETY STRESS SCALE / 195

According to the results of Table 8, it can be said that in every age group
depression, anxiety, and stress scores were decreasing with higher educational
attainment; on the other side, we did not find statistically significant differences in
primary and high school educated participants’ depression, anxiety, and stress
scores according to different age groups. Among secondary school educated
participants, depression and stress scores were increasing with increasing age,
which was statistically significant. Therefore, we concluded educational attain-
ment as an important demographic variable on DASS scores.

DISCUSSION

The exploratory factor analysis to examine the structure of the DASS-42 in the
clinical sample (N = 250), excluding the non-clinical controls, suggested a three
factor solution (eigenvalues of 17.6, 3.0, and 2.6), accounting for 56% of the total
variance. Two previous studies with clinical samples have also suggested three
factor solutions. One of these studies obtained eigenvalues of 18.92, 4.24, and
2.06 and 60% of the total variance [4]. The other study indicated eigenvalues of
16.64, 3.96, and 2.53 and 55% of the variance [11]. Our results are similar to those
of the previous studies. Two items of the depression subscale (item 16 “I felt I had
lost interest in just about everything” and item 3 “I coudn’t seem to experience any
positive feeling at all”) loaded both on depression and anxiety subscales but the
factor loadings on the depression subscale were higher (0.782 versus 0.450 for
item 16 and 0.735 versus 0.461 for item 3). Two items of the stress subscale loaded
on both the depression and stress subscales (item 27 “I found that I was very
irritable” and item 12 “I felt that I was using a lot of nervous energy”) but the
factor loadings on the stress subscale were higher (0.740 versus 0.409 for item 27
and 0.715 versus 0.437 for item 12). Overall, the factor structure in the present
sample was comparable with that found in other studies [4, 11].
The results of our study provide strong support for the internal consistency of
the three DASS scales in a clinical sample both across and within the anxiety, and
depression groups. The Cronbach alpha values found for the DASS depression,
anxiety, and stress subscales were 0.92, 0.86, and 0.88 respectively. Antony et al.
[4] determined the Cronbach alpha values for depression, anxiety, and stress
subscales as 0.97, 0.92, and 0.95, whereas in a study by Brown et al. [11] they were
0.96, 0.89, and 0.93. The internal consistency of each scale of the Turkish version
of DASS was quite favorable.
In line with the findings of other studies [4, 9, 11, 12, 15], the concurrent validity
of the DASS was satisfactory since correlations between similar constructs (e.g.,
DASS Depression and BDI) were high and significant. We found that the
DASS-D scale correlated most highly with BDI (r = 0.86) and the DASS-A scale
correlated most highly with BAI (r = 0.89). A previous study with the Turkish
version of DASS on a non-clinical sample found moderate correlations between
196 / HEKIMOGLU ET AL.

depression (r = 0.64) and anxiety (r = 0.58) subscales of the DASS and the
Hospital Anxiety and Depression scale [23].
The comparison of DASS scores across two diagnostic groups (anxiety and
depression) as well as on non-clinical participants showed that patients in the
depressive disorder condition tended to score highest on the depression and stress
subscales, whereas patients in the anxiety disorder group scored highest on the
anxiety and stress subscales. Non-clincal participants scored lower on all three
subscales than individuals in all of the clinical groups. The results of these
comparisons provided additional support for validity of the DASS scales as
indices of hyperarousal/fearfulness, depression, and tension/stress/negative affect.
The present findings are also fairly consistent with the tripartite model of anxiety
and depression [1]. The DASS Depression scale appears to measure features that
are unique to depression (low positive effect); the DASS Anxiety scale measures
features suggested as being unique to anxiety (physical hyperarousal).
Furthermore, as it has been shown in the study of Szabó and Lovibond [30], the
DASS Stress scale is thought to capture a third, specific factor comparable in
status to Depression and Anxiety, which corresponds most closely to the
diagnostic category of GAD (worry/tension).
The DASS scales’ relationships with demographic variables revealed that
females had higher scores on depression than males, and similar results have been
obtainted from several previous studies [12, 31]. We found a strong negative
association between DASS scores and level of education. Explanation for this
finding could be various. For example, the increasing level of education could help
people to find better ways for coping with depression, anxiety, and stress;
furthermore, level of education could be an important factor for seeking and
getting professional help. This relationship between DASS scores and level of
education needs to be further studied.
In summary, the Turkish version of the DASS-42 appears to be an excellent
instrument for measuring features of depression, hyperarousal, and tension in
clinical and non-clinical groups. Future studies are needed to assess the sensitivity
of the Turkish DASS in measuring post-treatment changes for anxiety disorders
and depression.

ACKNOWLEDGMENTS
The authors would like to thank Mrs. Caroline J. Walker for her assistance in
proofreading and editing of this manuscript. This study has no conflict of interests.

REFERENCES
1. Clark LA, Watson D. Tripartite model of anxiety and depression: Psychometric
evidence and taxonomic implications. Journal of Abnormal Psychology 1991;100:
316-336.
TURKISH VERSION OF DEPRESSION ANXIETY STRESS SCALE / 197

2. Sanderson WC, Di Nardo PA, Rapee RM, Barlow DH. Syndrome comorbidity in
patients diagnosed with a DSM-III-R anxiety disorder. Journal of Abnormal Psy-
chology 1990;99:308-311.
3. Brown TA, Barlow DH. Comorbidity among anxiety disorders: Implications for
treatment and DSM-IV. Journal of Consulting and Clinical Psychology 1992;60:
835-844.
4. Antony MM, Bieling PJ, Cox BJ, Enns MW, Swinson RP. Psychometric properties
of the 42 item and 21 item versions of the Depression Anxiety Stress Scales in clinical
groups and a community sample. Psychological Assessment 1998;10(2):176-181.
5. Moras K, Di Nardo PA, Barlow DH. Distinguishing anxiety and depression: Reexam-
ination of the reconstructed Hamilton Scales. Psychological Assessment 1992;4:
224-227.
6. Bieling PJ, Antony MM, Swinson RP. The State Trait Anxiety Inventory, Trait
version: Structure and content re-examined. Behavior Research and Therapy 1998;
36(7-8):777-788.
7. Cox BJ, Cohen E, Direnfeld DM, Swinson RP. Does the Beck Anxiety Inventory
measure anything beyond panic attack symptoms? Behavior Research and Therapy
1996;34(11-12):949-954 .
8. Lovibond SH, Lovibond PF. Manual for the Depression Anxiety Stress Scales
(2ndE d). Sydney Australia Psychology Foundation of Australia, 1995.
9. Lovibond SH, Lovibond PF. The structure of negative emotional states: Comparison
of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and
Anxiety Inventories. Behavior Research and Therapy 1995;33(3):335-342.
10. Lovibond PF. Long-term stability of depression, anxiety and stress syndromes.
Journal of Abnormal Psychology 1998;107:520-526.
11. Brown TA, Chorpita BF, Korotitsch W, Barlow DH. Psychometric properties of
the Depression Anxiety Stress Scales (DASS) in clinical samples. Behavior Research
and Therapy 1997;35(1):79-89.
12. Crawford JR, Henry JD. The Depression Anxiety Stress Scales (DASS): Normative
data and latent structure in a large non-clinical sample. British Journal of Clinical
Psychology 2003;42:111-131.
13. Clara IP, Cox BJ, Enns MW. Confirmatory factor analysis of the Depression–
Anxiety–Stress Scales in depressed and anxious patients. Journal of Psychopatho-
logical Behavior 2001;23(1):61-67.
13. Bados A, Solanas A, Andres R. Psychometric properties of the Spanish version of
Depression, Anxiety and Stress Scales (DASS). Psicothema 2005;17(4):679-683.
15. Daza P, Novy DM, Stanley MA, Averill P. The Depression Anxiety Stress Scale-21:
Spanish translation and validation with a Hispanic sample. Journal of Psychopatho-
logical Behavior 2002;24(3):195-205.
16. de Beurs E, Van Dyck R, Marquenie LA, Lange A, Blonk RWB. De DASS: Een
vragenlijst voor het meten van depressie, angst en stress [The DASS: A questionnaire
for the measurement of depression, anxiety, and stress]. Gedragstherapie 2001;34:
35-53.
17. Nieuwenhuijsen K, de Boer AGEM, Verbeek, JHAMR, Blonk WB, van Dijk FJH.
The Depression Anxiety Stress Scales (DASS): Detecting anxiety disorder and depres-
sion in employees absent from work because of mental health problems. Occupational
and Environmental Medicine 2003;60:77-82.
198 / HEKIMOGLU ET AL.

18. Apóstolo JLA, Mendes AC, Azeredo ZA. Adaptation to Portuguese of the Depres-
sion, Anxiety and Stress Scale (DASS). Revista Latino-Americana de Enfermagem
2006;14(6):863-871.
19. Taouk M, Lovibond PF, Laube R. Psychometric properties of an Arabic version of the
Depression Anxiety Stress Scales (DASS21). Report for New South Wales Trans-
cultural Mental Health Centre, Cumberland Hospital, Sydney, 2001. Retrieved March
12, 2010 from http://www2.psy.unsw.edu.au/groups/dass//Arabic/Arabic.htm
20. Taouk M, Lovibond PF, Laube R. Psychometric properties of a Chinese version of the
short Depression Anxiety Stress Scales (DASS21). Report for New South Wales Trans-
cultural Mental Health Centre, Cumberland Hospital, Sydney, 2001. Retrieved March
12, 2010 from http://www2.psy.unsw.edu.au/groups/dass//Chinese/tmhc.htm
21. Ramli M, Rosnani S, Aidil Faszrul AR. Psychometric Profile of Malaysian version
of the Depressive, Anxiety and Stress Scale 42-item (DASS-42). Malaysian Journal
of Psychiatry 2012;21(1):3-9.
22. Uncu Y, Bayram N, Bilgel N. Job related affective well-being among primary health
care physicians. European Journal of Public Health 2006;17:514-519.
23. Bilgel N, Bayram N. Turkish version of the Depression Anxiety Stress Scale (DASS-
42): Psychometric properties. Noropsikiyatri Arsivi 2010;47:118-126.
24. Ulusoy M, Sahin NH, Erkmen H. Turkish version of the Beck Anxiety Inventory:
Psychometric properties. Journal of Cognitive Psychotherapy 1998;12:163-172.
25. Kapci EG, Uslu R, Turkcapar MH, et al. Beck Depression Inventory II: Evalua-
tion of the psychometric properties and cut-off points in a Turkish adult population.
Depression and Anxiety 2008; 25:104-110.
26. Joreskog KG, Sorbom D. Lisrel 8: User’s reference guide. Chicago: Scientific Soft-
ware International, 2001.
27. Bollen KA. Structural equations with latent variables. New York: Wiley, 1989.
28. Bentler PM. Comparative fit indexes in structural models. Psychological Bulletin
1990;107:238-246.
29. Browne MW, Cudeck R. Alternative ways of assessing model fit. In Bollen KA,
Long JS (Eds.), Testing structural equation models (pp. 136-162). Sage: Newbury
Park, CA, 1993.
30. Szabó M, Lovibond PF. The cognitive content and affective correlates of naturally
occuring worry episodes. Cognitive Therapy and Research 2002; 26:167-177.
31. Horwath E.,Weissman MM. Epidemiology of depression and anxiety disorders. In
Tsuang, MT, Tohen, M, & Zahner, GEP (Eds.), Textbook in psychiatric epidemiology
(pp. 317-344). Wiley: New York, 1995.

Direct reprint requests to:


Levent Hekimoglu, M.D., Family Medicine Specialist
Bursa Yuksek Ihtisas Education & Research Hospital
(Bursa Yuksek Ihtisas Egitim ve Arastirma Hastanesi)
16140 Bursa/Turkey
e-mail: drleventhekimoglu@gmail.com

You might also like