Anxiety Disorders
Measurement and Evaluation in Counseling and Development 2016, Vol. 49(1) 3 –33 © The Author(s) 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0748175615596783 mec.sagepub.com
Assessment, Development, and Validation
The Beck Depression Inventory–Second Edi- tion (BDI-II) is the current version of an instrument used for more than 50 years to determine the level of depression in individu- als who are at least 13 years of age (Beck, Steer, & Brown, 1996). Since introduction of the BDI-II in 1996, numerous studies have been published that explore the instrument’s psychometric characteristics. The purpose of this meta-analysis is to identify these studies and aggregate their results to better under- stand the BDI’s internal consistency, test– retest reliability, external (convergent) validity, structural validity (exploratory and confirmatory factor analyses), and diagnostic validity (also known as decision reliability; sensitivity, specificity, positive predictive value, negative predictive value) from clinical and nonclinical samples found in the extant literature. This information is needed to inform users of the BDI-II of expected psy- chometric properties, rather than relying on the singular study outcome presented in the manual. Finally, because the BDI-II is a crite- rion-referenced interpretive instrument and no norms have been published, an additional area of exploration included aggregation of means
and standard deviations for nonclinical sam- ples with the goal of better understanding nor- mal depression ratings and enhancing clinical interpretations.
The BDI-II is the most commonly taught and used instrument in counselor education training (Peterson, Lomas, Neukrug, & Bon- ner, 2014) and is commonly used in depres- sion outcome research (Muller & Erford, 2012). The BDI-II consists of 21 self-report items designed to measure the occurrence and severity of symptoms of depressive disorders as defined by the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition (DSM-IV; American Psychiatric Association, 1994). Beck, Steer, and Brown (1996) noted that although the BDI-II was based on the symptoms indicated by DSM-IV, the BDI-II is
596783MECXXX10.1177/0748175615596783Measurement and Evaluation in Counseling and DevelopmentErford et al. research-article2015
1Loyola University Maryland, Timonium, MD, USA 2Duquesne University, Pittsburgh, PA, USA 3University of Iowa, Iowa City, IA, USA
Corresponding Author: Bradley T. Erford, Loyola University Maryland, Timonium Graduate Center, 2034 Greenspring Drive, Timonium, MD 21093, USA. Email: berford@loyola.edu
Meta-Analysis of the English Version of the Beck Depression Inventory–Second Edition
Bradley T. Erford1, Erin Johnson2, and Gerta Bardoshi3
Abstract This meta-analysis reviewed 144 studies from 1996 to 2013 using the Beck Depression Inventory–Second Edition. Internal consistency was .89 and test–retest reliability .75. Convergent comparisons were robust across 43 depression instruments. Structural validity supported both one- and two-factor solutions and diagnostic accuracy varied according to sample and criterion cutoff.
Keywords Beck Depression Inventory, depression, meta-analysis
4 Measurement and Evaluation in Counseling and Development 49(1)
a depression screening instrument and should not be used alone to diagnosis depression. To interpret the results of the BDI-II, the raw scores of each of the 21 items, ranging from 0 to 3, are added to determine a total raw score. Beck, Steer, and Brown (1996) defined four criterion-referenced interpretive categories to indicate the severity of symptoms, including “minimal” (total raw scores of 0 to 13), “mild” (total raw scores of 14 to 19), “moderate” (total raw scores of 20 to 28), and “severe” (total raw scores of 29 to 63). No norm-refer- enced interpretive data is available.
The BDI-II is a revised version of the amended Beck Depression Inventory (BDI- IA; Beck, Rush, Shaw, & Emery, 1979), itself an improvement over the original Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961). Of the 21 items on the BDI-II, four differ from the BDI-IA (Beck, Steer, & Brown, 1996). BDI- IA items that were less sensitive in identifying typical DSM-IV depression symptoms, such as “Body Image Change,” “Work Difficulty,” “Weight Loss,” and “Somatic Preoccupation” were eliminated and replaced with “Agita- tion,” “Worthlessness,” “Loss of Energy,” and “Concentration Difficulty” in the BDI-II revi- sion. In addition, items measuring changes in sleep and appetite were revised to allow for recording both increases and decreases in these behaviors. Furthermore, for a symptom to be considered present, the amount of time an individual must have experienced the symptom was changed from 1 week in the BDI-IA to at least 2 weeks in the BDI-II to better reflect the DSM-IV diagnostic standard for major depressive disorder (Beck, Steer, Ball, & Ranieri, 1996; Beck Steer, & Brown, 1996).
Multiple studies have reported high inter- nal consistency of the BDI-II (Beck, Steer, Ball, & Ranieri, 1996; Beck, Steer, & Brown, 1996; Dozois, Dobson, & Ahnberg, 1998; Steer & Clark, 1997). To examine the psycho- metric properties of the BDI-II, Beck, Steer, and Brown (1996) administered the BDI-II to 500 outpatients (clinical standardization sam- ple) and a group of 120 college students (non- clinical standardization sample). The BDI-II
had high internal consistency in both the out- patient group (α = .92) and the college student group (α = .93). Also, using 26 individuals from the outpatient group, the BDI-II demon- strated good test–retest reliability (r = .93). These seemed to be slight improvements over the previous versions of the BDI. For exam- ple, in a study conducted by Beck, Steer, Ball, and Ranieri (1996), 140 individuals diagnosed with depression were administered multiple inventories, including the BDI-IA and the BDI-II. High internal consistencies were reported for both the BDI-IA (α = .89) and the BDI-II (α = .91). In a study comparing the BDI and the BDI-II, Dozois et al. (1998) administered both inventories to a sample of 1,022 undergraduate students and found that both the BDI (α = .89) and BDI-II (α = .91) had high internal consistency.
While significant differences were reported between average scores on the BDI, BDI-IA, and BDI-II, high correlations were found among all three versions of the instrument (Beck, Steer, Ball, & Ranieri, 1996; Beck, Steer, & Brown, 1996; Dozois et al., 1998). Both Beck, Steer, and Brown (1996) and Beck, Steer, Ball, and Ranieri (1996) reported a correlation coefficient of r = .93 between the BDI-IA and the BDI-II. Although the average total score of the BDI-II was found to be about two points higher than that of the BDI-IA (t = 6.03, p < .001), these averages fell within the same symptom severity range. The average scores of five items were significantly differ- ent between the BDI-IA and the BDI-II. Two of the differences were anticipated due to changes in item wording. Similarly, the BDI- II yielded significantly greater average total scores than the BDI (t = 19.33, p < .001), though the inventories again were highly cor- related (r = .93; Dozois et al., 1998).
Evidence of convergent validity was offered primarily by a correlation of r = .71 with the Revised Hamilton Psychiatric Rating Scale for Depression (n = 87; Beck, Steer, & Brown, 1996). The BDI-II also correlated r = .60 with the Beck Anxiety Inventory. Evi- dence of structural validity was provided pri- marily through two exploratory factor analytic studies. Using a principal-factor analysis with
Erford et al. 5
promax rotation, Beck, Steer, and Brown (1996) determined that two dimensions best explained responses from the 500 participant clinical sample: Somatic-Affective (Items 4, 10–13, and 15–21) and Cognitive (Items 1–3, 5–9, and 14). This process was replicated on the 120 student nonclinical sample with a somewhat different result: Cognitive-Affec- tive (Items 1–14 and 17) and Somatic (Items 15–16 and 18–20). These factor solutions are somewhat controversial given that: (a) the 120 participant student analysis was signifi- cantly underpowered, (b) the two factors that emerged correlated .66 and .62 within the clinical and nonclinical samples indicating very high interdimensional relationships, and (c) only the BDI-II total raw score is used for interpretation. That is, clinicians rarely com- pute and interpret subscale/factor scores. With all this information stemming primarily from a couple of studies published in the manual (Beck, Steer, & Brown, 1996), the present study explored the subsequent extant litera- ture for what is now known about the BDI-II’s score reliability and validity.
Method
Journal articles, dissertations, theses, and electronically available unpublished manu- scripts were included to insure that all avail- able published and unpublished studies could be located in an attempt to control for publica- tion bias. Studies included in this meta-analy- sis: (a) used the English version of the BDI-II, (b) used the second edition of the BDI (BDI- II; Beck, Steer & Brown, 1996), and (c) pro- vided some type of reliability or validity data. Various translated versions of the BDI were located in the extant literature while conduct- ing this meta-analysis including Arabic, Chi- nese, Dutch, Farsi, French, German, Greek, Hebrew, Hindi, Hungarian, Icelandic, Italian, Japanese, Korean, Nigerian, Palestinian, Por- tuguese, Persian, Serbian, Spanish, Taiwan- ese, and Turkish. Only studies using the English BDI-II were selected into this study. In addition, several studies provided correla- tions for subscales of the BDI or “brief ver- sions” of the BDI only, but not for the total
score (21 item) scale, and thus were elimi- nated. In summary, only studies using the English version, 21-item BDI-II were selected so that all comparisons and analyses would involve the same test version in the same lan- guage.
Search Strategies
Redundant search procedures were used to insure maximum attainment of candidate studies. These redundant procedures included an electronic search, followed by hand searches of the reference lists of selected and synthesis articles (e.g., meta-analyses, quali- tative syntheses). PsychINFO, ERIC, Aca- demic Search Premier, and MEDLINE articles from 1996 to 2013 were searched using the keywords “Beck Depression Inventory” in the text. We limited the search parameters to iden- tify English-language candidate studies. To locate additional candidate studies, we then examined the reference lists of selected stud- ies and synthesis articles on the BDI-II. The full text of each article was reviewed by the authors, selection criteria applied, and each selected article submitted to analysis. Each candidate article was reviewed by the first two authors and disagreements resolved through consensus.
Psychometric Variables Analyzed and Statistical Methods Used
Six primary variables were of interest in this psychometric meta-analysis: internal consis- tency, test–retest reliability, convergent corre- lations with other depression measures, structural validity (i.e., exploratory factor analysis [EFA], confirmatory factor analysis [CFA]), diagnostic validity across various cut- off scores and samples (e.g., percent of correct classifications, sensitivity, specificity, posi- tive predictive value, negative predictive value, area under the curve [AUC] estimates), and descriptive statistics (i.e., means and stan- dard deviations) from nonclinical samples.
All correlations were independent and only similar effect sizes on comparable study designs were combined (Erford, Savin-Murphy,
6 Measurement and Evaluation in Counseling and Development 49(1)
& Butler, 2010). Coefficient alpha (α) was used as the test statistic to represent internal consistency analyses. Pearson r was used as the effect size for test–retest reliability and convergent validity analyses. Weighting pro- cedures were used to correct for sample size bias when combining effect size estimates. While α and test–retest coefficients were directly weighted and analyzed, Pearson rs for the convergent analyses were transformed into z values using Hedges and Olkin’s (1985) for- mula, weighted by sample
size, summed, averaged, and then the grand z was back-transformed to r. Assessment of homogeneity of r was conducted using a ran- dom effects model, which assumes the selected studies were sampled from a larger set of studies, allowing for greater external generalizability.
Erford et al. (2010) provided a formula to calculate effect size standard errors and 95% confidence intervals (CI95). The 95% confi- dence interval was chosen to facilitate a simple assessment of whether an effect size was greater than zero. For example, if a 95% confi- dence interval for r = .40 is +.20, the range of .20 to .60 is obtained, and the null hypothesis of r = 0 can be rejected because the complete range for r is >0. Should r = .10 with a 95% confidence interval of ±.20, the range produced would be −.10 to .30. Because a segment of the CI95 range is less than 0, the null hypothesis cannot be rejected and r cannot be considered to be significantly greater than zero.
To estimate heterogeneity in effect sizes, Cochran’s heterogeneity statistic (Q) was used. Using the chi-square distribution, a p < .05 for the Q statistic indicates heterogeneity across studies, leading to rejection of the null hypothesis of homogeneity. The degree of inconsistency was also calculated (I2; Hig- gins, Thompson, Deeks, & Altman, 2003) and interpreted as follows: 0% means no inconsis- tency, 25% means low inconsistency, 50% means moderate inconsistency, 75% means high inconsistency, and 100% means total inconsistency (total heterogeneity). I2 values greater than 50% implies significant heteroge- neity, and should be analyzed for mediator and moderator variables.
Publication Bias
Multiple methods should be used to detect pub- lication bias (Beretvas, 2010). Three methods were used in the current study: Duval and Tweedie’s (2000; Richardson, Abraham, & Bond, 2012) trim and fill procedure, funnel plot analysis, and Rosenthal’s (1979) fail-safe N. All comparisons were within normal limits, so pub- lication bias was determined to be insignificant.
Results
Of candidate articles, 954 were identified through computerized searches and 50 more candidate articles were identified via hand searches, for a total of 1,004 candidate articles. Full text review of all candidate articles elimi- nated 860 articles that violated one or more inclusion criteria. Thus, (j =) 144 articles were forwarded to the analysis phase. Several articles used more than one sample (k) so the results that follow will involve analysis of k studies.
Internal Consistency
A total of 99 studies with a combined sample size of 31,413 participants reported coeffi- cient α results. When all studies were weighted and then averaged, α = .893 (CI95 = .881–.905). Study αs ranged from .75 (Nobles, 2011) to .96 (King, Colella, Faris, & Thompson, 2009). Next, the studies were parsed to allow calcu- lation of average αs according to clinical and nonclinical samples. In the clinical samples (k = 56, n = 13,447), α = .907 (CI95: .889–.925) with study αs ranging from .81 (Chau & Mill- ing, 2006; Foster, 2009) to .96. In the nonclin- ical samples (k = 43, n = 17,966), α = .881 (CI95: .867–.895) with study αs ranging from .75 (Nobles, 2011) to .94 (Arnau, Meagher, Norris, & Bramson, 2001; Kneipp, Kairalla, Stacciarini, & Pereira, 2009; Novy, Stanley, Averill, & Daza, 2001; Wolfe-Christensen et al., 2012).
Test–Retest Reliability
A total of 12 studies with a total sample size of 1,562 were weighted and then combined to
(1log[(1+r)/(1-r)]) 2
Erford et al. 7
yield a test–retest reliability coefficient (r tt ) of
.75 (CI95: .69–.81; mean and median time lapse of 6 weeks). Study r
tt s ranged from .44
(Cukrowicz & Joiner, 2007) to .98 (Leigh & Anthony-Tolbert, 2001). Again, the studies were parsed to allow calculation of average r
tt s according to clinical and nonclinical sam-
ples. Six clinical samples were located (n = 572) yielding a weighted average r
tt of .68
(CI95: .60–.76), with coefficients ranging from .52 (Straits-Tröster et al., 2000) to .91 (Coles, Gibb, & Heimberg, 2001). Six non- clinical samples were located (n = 990) yield- ing an average r
tt = .80 (CI95: .74–.86), with
coefficients ranging from .44 to .98.
External (Convergent) Validity
External (convergent) validity was estimated through the calculation of Pearson rs between the BDI-II and 43 other depression invento- ries. These comparisons yielded robust con- vergent rs ranging from .45 (Positive and Negative Symptoms Scale [PANSS]; Kay, Fiszbein, & Opler, 1987) to .88 (Glasgow Depression Scale for People with Learning Disabilities [GDS-LD]; Cuthill, Espie, & Cooper, 2003). These results are reported in Table 1 and all comparisons were r > 0. The vast majority of these instruments appeared in only one, two, or three studies, but three depression instruments warrant special men- tion because each provided more than three convergent comparisons: Center for Epide- miological Studies–Depression (CES-D; Rad- loff, 1977); Hamilton Depression Inventory (HAM-D; Hamilton, 1960); and Zung Depres- sion Rating Scale (ZDRS; Zung, Richards, Gables, & Short, 1965). The self-report CES-D was compared with the BDI-II in 11 studies (combined n = 3,209) resulting in a weighted average r of .72 (CI95 = .68–.76). Another self-report instrument, the ZDRS, was compared with the BDI-II in four studies (combined n = 762) resulting in a similar combined r = .74 (CI95 = .67–.81). The HAM-D is a clinician-report instrument that was compared with the BDI-II in eight studies (combined n = 1,393) resulting in a combined r = .53 (CI95 = .48–.58). None of the distribu-
tions of effect sizes displayed significant het- erogeneity, so all null hypotheses stemming from Cochran’s Q statistic were rejected. Likewise, all I2 computations were less than the 50% criterion for exploration of modera- tor and mediator variables.
Structural Validity
Exploratory Factor Analysis. Eighteen studies were located that provided EFA of the BDI-II (see Table 2) and these studies will be pre- sented according the number of factors emerg- ing from the respective EFAs (i.e., one through five factors). While a majority of the studies were conducted with a sufficient sam- ple size for the number of items composing the BDI-II, it should be noted that seven of the studies were underpowered (see Table 2), fail- ing to meet the 10:1 participants to items ratio suggested by Tabachnick and Fidell (2013) for stable factor solutions. It is evident that studies with diverse sample sizes, extraction and rotation procedures, and participant types yielded differing results.
Two large sample analyses supported a unidimensional construct for the BDI-II. This is an interesting result because each study used different sample types. The Kim, Pilko- nis, Frank, Thase, and Reynolds (2002) study assessed 831 adult clinical patients. The Segal, Coolidge, Cahill, and O’Riley (2008) study assessed 376 normal adults. Both sin- gle-factor solutions accounted for only about 37% of the item variance.
Eleven of the EFAs supported two-factor solutions, accounting for between 36% and 88% of item variance. Two of these studies (Arnau et al., 2001; Steer, Ball, Ranieri, & Beck, 1999) yielded a 2-factor solution identi- fying the same two dimensions reported in the original Beck, Steer, and Brown (1996) EFA for the clinical outpatient sample: Cognitive and Somatic-Affective. Seven of these 11 EFA studies (Bos et al., 2002; Contreras, Fer- nandez, Malcarne, Ingram, & Vaccarino, 2004; Dozois et al., 1998; Kneipp et al., 2009; Palmer & Binks, 2008; Patterson et al., 2011; Steer & Clark, 1997) yielded a 2-factor solution identifying two slightly different dimensions:
8
T a b
le 1
. C
o nv
er ge
nt V
al id
it y
St ud
ie s
W it h
th e
B ec
k D
ep re
ss io
n In
ve nt
o ry
–S ec
o nd
E di
ti o n
(B D
I- II)
.
T es
t N
am e
k n
r (C
I9 5)
St ud
ie s
A ss
es sm
en t
o f D
ep re
ss io
n In
ve nt
o ry
( A
D I)
D
ep re
ss io
n 1
96 .8
2 (. 62
–1 .0
0) M
o gg
e, S
te in
be rg
, F re
m o uw
, & M
es se
r (2
00 8)
A ut
o m
at ed
N eu
ro ps
yc ho
lo gi
ca l A
ss es
sm en
t M
et ri
cs (
A N
A M
) M
o o d
Sc al
e 2
27 0
.7 0
(. 58
–. 82
) Jo
hn so
n, V
in ce
nt , J
o hn
so n,
G ill
ila nd
, & S
ch le
ge l (
20 08
); R
o eb
uc k-
Sp en
ce r
et a
l. (2
00 6)
A do
le sc
en t
Ps yc
ho pa
th o lo
gy S
ca le
( A
PS )
M aj
o r
D ep
re ss
io n
su bs
ca le
s 1
13 1
.6 1
(. 44
–. 78
) V
an V
o o rh
is &
B lu
m en
tr it t
(2 00
7)
A do
le sc
en t
Ps yc
ho pa
th o lo
gy S
ca le
( A
PS )
D ys
th ym
ic D
is o rd
er s
ub sc
al es
1 13
1 .6
2 (. 45
–. 79
) V
an V
o o rh
is &
B lu
m en
tr it t
(2 00
7)
B A
SI S
32 1
30 .6
5 (. 28
–1 .0
0) B
af fig
e (2
00 6)
B ri
ef S
ym pt
o m
I nv
en to
ry (
B SI
) D
ep re
ss io
n 1
10 0
.7 4
(. 54
–. 94
) Si
m s
(2 01
0) B
ri ef
S cr
ee n
fo r
D ep
re ss
io n
(B SD
) 1
69 0
.7 4
(. 66
–. 82
) Li
pp s,
L o w
e, &
Y o un
g (2
00 7)
C A
T I-
D ep
re ss
io n
1 37
6 .5
8 (. 48
–. 68
) Se
ga l,
C o o lid
ge , C
ah ill
, & O
’R ile
y (2
00 8)
C ar
di ac
D ep
re ss
io n
Sc al
e (C
D S)
3 25
9 .7
5 (. 63
–. 87
) D
i B en
ed et
to , L
in dn
er , H
ar e,
& K
en t
(2 00
6) ; K
in g,
C o le
lla , F
ar is
, &
T ho
m ps
o n
(2 00
9) C
en te
r fo
r Ep
id em
io lo
gi ca
l S tu
di es
–D ep
re ss
io n
11 3,
20 9
.7 2
(. 68
–. 76
) B
ak er
( 20
09 ); B
us h,
N o va
ck , S
ch ne
id er
, & M
ad an
( 20
04 ); J ay
& J o hn
(2
00 4)
; L ip
ps , L
o w
e, &
Y o un
g (2
00 7)
; P yn
e, T
ri pa
th i,
W ill
ia m
s, &
Fo
rt ne
y (2
00 7)
; S eg
al , C
o o lid
ge , C
ah ill
, & O
’R ile
y (2
00 8)
; S he
an &
B
al dw
in (
20 08
); S
pi el
be rg
er , R
it te
rb an
d, R
eh ei
se r,
& B
ru nn
er (
20 03
) C
al ifo
rn ia
P sy
ch o lo
gi ca
l I nv
en to
ry -D
ep re
ss io
n (C
PI -D
) 2
1, 28
8 .7
8 (. 72
–. 84
) Ja
y &
J o hn
( 20
04 )
D ep
re ss
io n–
A rk
an sa
s Sc
al e
(D -A
R K
) 1
65 0
.8 3
(. 75
–. 91
) W
al te
r, M
er es
m an
, K ra
m er
, & E
va ns
( 20
03 )
D ep
re ss
io n
A nx
ie ty
S tr
es s
Sc al
e– D
ep re
ss io
n (D
A SS
) 1
28 3
.7 8
(. 66
–. 90
) R
ec to
r, K
am ka
r, C
as si
n, A
ye ar
st , &
L ap
o sa
( 20
11 )
C o m
pu te
r- ad
ap te
d T
es t
o f D
ep re
ss io
n (D
-C A
T )
2 3,
38 1
.7 9
(. 76
–. 82
) Fl
ie ge
e t
al . (
20 09
); F
lie ge
e t
al . (
20 05
)
D ep
re ss
iv e
Pe rs
o na
lit y
D is
o rd
er I nv
en to
ry
(D PD
I) 1
36 2
.8 0
(. 69
–. 91
) C
ha m
be rl
ai n
& H
up ri
ch (
20 11
)
D ia
gn o st
ic I nt
er vi
ew S
ch ed
ul e
D SM
-IV (
D IS
-I V
) 1
95 .4
9 (. 28
–. 70
) Sh
ea n
& B
al dw
in (
20 08
) D
ep re
ss iv
e In
te ns
it y
Sc al
e C
ir cl
es (
D IS
C )
1 11
4 .6
6 (. 47
–. 85
) T
ur ne
r- St
o ke
s, K
al m
us , H
ir an
i, &
C le
gg (
20 05
)
(c on
tin ue
d)
9
T es
t N
am e
k n
r (C
I9 5)
St ud
ie s
Em o ti o na
l A ss
es sm
en t
Sy st
em –D
ep re
ss io
n (E
A S)
1 12
0 .5
9 (. 41
–. 77
) M
it ch
el l (
20 05
)
Ed in
bu rg
h Po
st na
ta l D
ep re
ss io
n Sc
al e
(E PD
S) 1
15 0
.8 1
(. 65
–. 97
) B
ec k
& G
ab le
( 20
01 )
G ro
ss m
an -C
o le
D ep
re ss
io n
In ve
nt o ry
( G
C D
I) 1
10 1
.7 3
(. 53
–. 93
) C
o le
, G ro
ss m
an , P
ri lli
m an
& H
un sa
ke r
(2 00
3) G
er ia
tr ic
D ep
re ss
io n
Sc al
e (G
D S)
3 32
5 .7
9 (. 68
–. 90
) Lo
w &
H ub
le y
(2 00
7) ; R
o se
( 20
02 ); Y
o ch
im , M
ue lle
r, J un
e, &
S eg
al
(2 01
1) G
la sg
o w
D ep
re ss
io n
Sc al
e fo
r Pe
o pl
e w
it h
Le ar
ni ng
D is
ab ili
ti es
( G
D S-
LD )
1 38
.8 8
(. 56
–1 .0
0) C
ut hi
ll, E
sp ie
, & C
o o pe
r (2
00 3)
G lo
ba l M
o o d
Sc al
e– N
eg at
iv e
A ffe
ct (
G M
S) 1
14 0
.5 2
(. 35
–. 69
) M
o gg
e, S
te in
be rg
, F re
m o uw
& M
es se
r, (
20 08
) H
am ilt
o n
D ep
re ss
io n
In ve
nt o ry
( H
A M
-D )
8 1,
39 3
.5 3
(. 48
–. 58
) C
ah ill
e t
al . (
20 06
); C
o le
s, G
ib b,
& H
ei m
be rg
( 20
01 ); J o e,
W o o lle
y,
B ro
w n,
G ha
hr am
an lo
u- H
o llo
w ay
, & B
ec k
(2 00
8) ; K
im , P
ilk o ni
s,
Fr an
k, T
ha se
, & R
ey no
ld s
(2 00
2) ; K
o ns
ta nt
in id
is , M
ar ti ny
, B ec
h, &
K
as pe
r (2
01 1)
; M cC
al l,
R eb
o us
si n,
& C
o he
n (2
00 0)
; M o us
tg aa
rd
(2 00
5) ; W
ee ks
& H
ei m
be rg
( 20
05 )
Se lf-
re po
rt H
am ilt
o n
D ep
re ss
io n
In ve
nt o ry
(H
D I)
2 80
1 .8
0 (. 73
–. 87
) D
o zo
is (
20 03
); G
ib b,
B ut
le r,
& B
ec k
(2 00
3)
M ill
o n
M aj
o r
D ep
re ss
io n
an d
D ys
th ym
ia S
ca le
s (M
C M
I) M
aj o r
D ep
re ss
io n
Sc al
e 1
46 .6
8 (. 39
–. 97
) M
ar gu
er at
D eg
aa rd
( 20
05 )
M ill
o n
M aj
o r
D ep
re ss
io n
an d
D ys
th ym
ia S
ca le
s (M
C M
I) D
ys th
ym ia
S ca
le 1
46 .6
5 (. 36
–. 94
) M
ar gu
er at
D eg
aa rd
( 20
05 )
M aj
o r
D ep
re ss
io n
In ve
nt o ry
( M
D I)
1 51
.8 2
(. 54
–1 .0
0) K
o ns
ta nt
in id
is , M
ar ti ny
, B ec
h, &
K as
pe r
(2 01
1) M
in ne
so ta
M ul
ti ph
as ic
P er
so na
lit y
In ve
nt o ry
(M
M PI
) D
ep re
ss io
n Sc
al e
1 40
8 .5
5 (. 45
–. 65
) O
sm an
, K o pp
er , B
ar ri
o s,
G ut
ie rr
ez , &
B ag
ge (
20 04
)
N eu
ro bi
o lo
gi ca
l F un
ct io
ni ng
I nv
en to
ry (
N FI
) D
ep re
ss io
n Sc
al e
1 17
2 .7
6 (. 61
–. 91
) Se
el &
K re
ut ze
r (2
00 3)
Pe rs
o na
lit y
A ss
es sm
en t
In ve
nt o ry
( PA
I)
D ep
re ss
io n
su bs
ca le
1 96
.8 6
(. 66
–1 .0
0) M
o gg
e, S
te in
be rg
, F re
m o uw
, & M
es se
r (2
00 8)
Po si
ti ve
a nd
N eg
at iv
e Sy
m pt
o m
s Sc
al e–
D ep
re ss
io n
(P A
N SS
) 1
23 0
.4 5
(. 32
–. 58
) C
he m
er in
sk i,
B o w
ie , A
nd er
so n,
& H
ar ve
y (2
00 8)
(c on
tin ue
d)
T a b
le 1
. (c
o n
ti n
u e d
)
10
T es
t N
am e
k n
r (C
I9 5)
St ud
ie s
Pa ti en
t H
ea lt h
Q ue
st io
nn ai
re (
PH Q
-9 ; 9
it em
s) 2
41 2
.7 4
(. 64
–. 84
) H
ep ne
r, H
un te
r, E
de le
n, Z
ho u,
& W
at ki
ns (
20 09
); T
it o v
et a
l. (2
01 1)
Pr o fil
e o f M
o o d
St at
es (
PM S)
1 21
0 .6
0 (. 46
–. 74
) Jo
hn so
n, V
in ce
nt , J
o hn
so n,
G ill
ila nd
, & S
ch le
ge l (
20 08
) R
ey no
ld s
A do
le sc
en t
D ep
re ss
io n
Sc al
e (R
A D
S) 1
10 0
.8 4
(. 64
–1 .0
0) K
re fe
tz , S
te er
, G ul
ab , &
B ec
k (2
00 2)
SC L-
90 -R
( Sy
m pt
o m
C he
ck lis
t) D
ep re
ss io
n Sc
al e
2 29
8 .8
4 (. 72
–. 96
) Le
o na
rd so
n et
a l.
(2 00
3) ; S
te er
& B
al l (
19 97
)
St ru
ct ur
ed C
lin ic
al I nt
er vi
ew f o r
th e
D SM
-IV
(S C
ID )
2 71
2 .6
7 (. 60
–. 77
) Sp
ri nk
le e
t al
. ( 20
02 ); W
at so
n et
a l.
(2 00
8)
Sh o rt
er P
sy ch
o th
er ap
y an
d C
o un
se lin
g Ev
al ua
ti o n
(G en
er al
D ep
re ss
io n)
( SP
A C
ED )
1 10
8 .7
6 (. 57
–. 95
) H
al st
ea d,
L ea
ch , &
R us
t (2
00 7)
St at
e T
ra it A
nx ie
ty I nv
en to
ry –D
ep re
ss io
n (S
T A
I- D
) 1
40 8
.7 6
(. 66
–. 86
) O
sm an
, K o pp
er , B
ar ri
o s,
G ut
ie rr
ez , &
B ag
ge (
20 04
)
St at
e an
d T
ra it D
ep re
ss io
n (S
/T /D
) Sc
al e
(S -D
is
st at
e de
pr es
si o n)
2 25
1 .6
0 (. 47
–. 73
) Sp
ie lb
er ge
r, R
it te
rb an
d, R
eh ei
se r,
& B
ru nn
er (
20 03
)
St at
e an
d T
ra it D
ep re
ss io
n (S
/T /D
) Sc
al e
(T -D
is
t ra
it d
ep re
ss io
n) 3
36 4
.7 4
(. 64
–. 84
) C
o le
s, G
ib b,
& H
ei m
be rg
( 20
01 ); S
pi el
be rg
er , R
it te
rb an
d, R
eh ei
se r,
&
B ru
nn er
( 20
03 )
T ea
te D
ep re
ss io
n In
ve nt
o ry
( T
D I)
1 20
01 .7
3 (. 69
–7 7)
B al
sa m
o e
t al
. ( 20
13 )
Z un
g Se
lf- re
po rt
D ep
re ss
io n
Sc al
e (Z
SD S)
4 76
2 .7
4 (. 67
–. 81
) M
ay na
rd e
t al
. ( 20
10 ); M
o gg
e, S
te in
be rg
, F re
m o uw
, & M
es se
r (2
00 8)
; Sp
ie lb
er ge
r, R
it te
rb an
d, R
eh ei
se r,
& B
ru nn
er (
20 03
)
N ot
e. S
o m
e st
ud ie
s pr
es en
te d
m o re
t ha
n o ne
s am
pl e;
s o m
e in
st ru
m en
ts y
ie ld
ed m
o re
t ha
n o ne
d ep
re ss
io n-
re la
te d
su bs
ca le
.
T a b
le 1
. (c
o n
ti n
u e d
)
11
T a b
le 2
. Ex
pl o ra
to ry
F ac
to r
A na
ly si
s St
ud ie
s o f th
e B
ec k
D ep
re ss
io n
In ve
nt o ry
–S ec
o nd
E di
ti o n
(B D
I- II)
.
A rt
ic le
n an
d Sa
m pl
e D
es cr
ip ti o n
N um
be r
o f Fa
ct o rs
: F ac
to r
N am
es %
V ar
ia nc
e
A rn
au , M
ea gh
er , N
o rr
is , &
B ra
m so
n (2
00 1)
34 0
A du
lt s
2 fa
ct o rs
: C o gn
it iv
e, S
o m
at ic
-A ffe
ct iv
e 54
B o s
et a
l. (2
00 9)
68 5
Pr eg
na nt
o r
re ce
nt m
o th
er s
2 fa
ct o rs
: C o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
- an
xi et
y 42
C he
m er
in sk
i, B
o w
ie , A
nd er
so n,
& H
ar ve
y (2
00 8)
23 0
Pa ti en
ts w
it h
Sc hi
zo ph
re ni
a 3
fa ct
o rs
: D ys
ph o ri
a, P
sy ch
o so
m at
ic
& R
eg re
t N
R
C o nt
re ra
s, F
er na
nd ez
, M al
ca rn
e, I ng
ra m
, &
V ac
ca ri
no (
20 04
) 3,
81 3
U nd
er gr
ad ua
te s
2 fa
ct o rs
: C o gn
it iv
e- A
ffe ct
iv e
&
So m
at ic
30
D o zo
is , D
o bs
o n,
& A
hn be
rg (
19 98
) 1,
02 2
U nd
er gr
ad ua
te s
2 fa
ct o rs
C o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
- V
eg et
at iv
e 46
D um
, P ic
kr en
, S o be
ll, &
S o be
ll (2
00 8)
a 1 08
S ub
st an
ce u
si ng
o ut
pa ti en
ts 3
fa ct
o rs
: C o gn
it iv
e, A
ffe ct
iv e,
S o m
at ic
63 K
im , P
ilk o ni
s, F
ra nk
, T ha
se , &
R ey
no ld
s (2
00 2)
83 1
A du
lt p
at ie
nt s
1 fa
ct o r
37 K
ne ip
p, K
ai ra
lla , S
ta cc
ia ri
ni , &
P er
ei ra
( 20
09 )
a 1 08
L o w
in co
m e
w o m
en 2
fa ct
o rs
: C o gn
it iv
e- A
ffe ct
iv e
&
So m
at ic
88
Pa lm
er &
B in
ks (
20 08
) a 1
17 1
8– 21
y ea
r o ld
in ca
rc er
at ed
m al
es 2
fa ct
o rs
: C o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
36 Pa
tt er
so n
et a
l. (2
01 1)
67 1
H ep
at it is
C p
at ie
nt s
2 fa
ct o rs
: C o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
N R
R o dr
íg ue
z- G
ó m
ez , D
áv ila
-M ar
tí ne
z, &
C o lla
zo -
R o dr
íg ue
z (2
00 6)
. 41
0 Pu
er to
R ic
an e
ld er
ly 4
fa ct
o rs
( no
t na
m ed
) 52
Se ga
l, C
o o lid
ge , C
ah ill
, & O
’R ile
y (2
00 8)
37 6
N o rm
al a
du lt s
1 fa
ct o r
38 St
ee r,
B al
l, R
an ie
ri , &
B ec
k (1
99 9)
a N R
C lin
ic al
ly d
ep re
ss ed
p at
ie nt
s 2
fa ct
o rs
: C o gn
it iv
e, S
o m
at ic
-A ffe
ct iv
e N
R St
ee r
& C
la rk
( 19
97 )
a 1 70
U nd
er gr
ad s
2 fa
ct o rs
: C o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
N R
St ee
r, K
um ar
, R an
ie ri
, & B
ec k
(1 99
8) 21
0 A
do le
sc en
t o ut
pa ti en
ts 3
fa ct
o rs
C o gn
it iv
e, S
o m
at ic
-A ffe
ct iv
e,
a no
ns al
ie nt
f ac
to r
N R
St ee
r, R
is sm
ill er
, & B
ec k
(2 00
0) a 1
30 O
ld er
a du
lt in
pa ti en
ts 2
fa ct
o rs
: C o gn
it iv
e, N
o nc
o gn
it iv
e 37
Su br
am an
ia m
, H ar
re ll,
H un
tl ey
, & T
ra cy
( 20
09 )
a 1 45
A do
le sc
en ts
w it h
su bs
ta nc
e ab
us e
5 fa
ct o rs
: D -C
o gn
it iv
e, D
-A ffe
ct iv
e,
nu tr
it io
n, D
-s el
f- pe
rs ec
ut io
n,
pe ss
im is
m
61
V an
V o o rh
is &
B lu
m en
tr it t
(2 00
7) a 1
31 M
ex ic
an -A
m er
ic an
y o ut
h 2
fa ct
o rs
: C o gn
it iv
e- so
m at
ic , A
ffe ct
iv e
43
N ot
e. P
er ce
nt ag
e is
p er
ce nt
o f va
ri an
ce a
cc o un
te d
fo r
by t
he f ac
to rs
in t
he a
na ly
si s;
N R
= n
o ne
r ep
o rt
ed ; D
= d
ep re
ss io
n. a U
nd er
po w
er ed
, m ea
ni ng
t he
s am
pl e
si ze
is le
ss t
ha n
th e
re co
m m
en de
d 10
:1 p
ar ti ci
pa nt
t o it
em r
at io
.
12 Measurement and Evaluation in Counseling and Development 49(1)
Cognitive-Affective and Somatic. In the liter- ature, this model usually is referred to as the Steer and Clark (1997) 2-factor model, similar to the college student 2-factor model reported by Beck, Steer, and Brown (1996). In an underpowered study of 130 older adult patients, Steer, Rissmiller, and Beck (2000) named their two factors Cognitive and Non- cognitive. The final 2-factor EFA (VanVoorhis & Blumentritt, 2007) was also underpowered, assessing 131 Mexican American youth, and yielding an additional dimensional permuta- tion: Cognitive-Somatic and Affective.
Three of the 18 EFA studies presented a 3-factor model, although in one of these stud- ies (Steer, Kumar, Ranieri, & Beck, 1998) the third factor was not salient and the remaining two factors reflected the dimensions evident in the Beck, Steer, and Brown (1996) 2-factor model (i.e., Cognitive and Somatic-Affec- tive). One of these 3-factor solutions (Dum, Pickren, Sobell, & Sobell, 2008) assessed only 108 substance abusing outpatients, iden- tifying Cognitive, Affective, and Somatic dimensions, while the other 3-factor solution (Chemerinski, Bowie, Anderson, & Harvey, 2008), which was conducted on 230 patients with schizophrenia, presented Dysphoria, Psychosomatic, and Regret dimensions— which aligned closely with the Dum et al. dimensions.
Two other EFA studies presented 4-factor (Rodriguez-Gomez, Davila-Martinez, & Col- lazo-Rodriguez, 2006) and 5-factor solutions (Subramaniam, Harrell, Huntley, & Tracy, 2009). The Rodriguez-Gomez et al. study was conducted on 410 elderly Puerto Ricans and the factors were not named in the article. The Subramaniam et al. study was underpowered, assessing only 145 adolescents with substance use disorders, and yielded five factors named: Depression-Cognitive, Depression-Affective, Nutrition, Depression-Self-persecution, and Pessimism.
Confirmatory Factor Analysis. Twenty-one stud- ies were located that applied CFA to test vari- ous models of the BDI-II. In all, 49 tests of 13 different BDI-II models were reported and the available output statistics are summarized in
Table 3. Six of the 21 articles used samples that were underpowered. The default unidi- mensional model was tested in nine samples across six articles (Carmody, 2005; Hepner, Hunter, Edelen, Zhou, & Watkins, 2009; Joiner et al. 2001; Osman, Kopper, Barrios, Gutierrez, & Bagge, 2004; Seignourel, Green, & Schmitz, 2008; Titov et al., 2011), and only one of these samples was underpowered (Titov et al., 2011). Comparative fit index (CFI) estimates across these nine studies ranged from .83 to .93 (median = .86), while the root mean square error of approximation (RMSEA) ranged from .049 to .11 (median = .08). Adequate fitting models have CFIs of ≥.90, with .95 indicating an excellent fit (Dimitrov, 2012). RMSEAs of ≤.06 indicate an adequate fit of the model to the data. Only one of the nine samples had data that fit the models well. On a sample of 408 adolescent psychiatric patients, Osman et al. (2004) reported a CFI for the unidimensional factor of .93 and RMSEA of .049. On a sample of 582 treatment-seeking substance users, Sei- gnourel et al. (2008) reported a CFI for the unidimensional factor of .90, but the RMSEA was .091, and Joiner et al. (2001), using a sample of 1,404 Air Force Cadets reported an RMSEA of .057, but CFI of .86. All other sample data produced fit statistics below the threshold of adequacy.
Six studies across four articles (Kneipp et al., 2009; Thombs, Ziegelstein, Beck, & Pilote, 2008; Whisman, Perez, & Ramel, 2000; Wiebe & Penley, 2005) tested the Beck, Steer, and Brown (1996) 2-factor model of Cognitive and Somatic-Affective dimensions. Kneipp et al. actually tested three modified versions of Beck, Steer, and Brown’s 2-factor model and all fit the data equivalently. The other three studies’ data also fit the model adequately, with CFIs ranging from .91 to .94 and RMSEAs ranging from .05 to .07. Thus, the Beck, Steer, and Brown (1996) 2-factor model (Cognitive and Somatic-Affective) appears to be a robust model for both clinical and nonclinical samples.
Fourteen studies across 12 articles (Canel- Cinarbas, Cui, & Lauridsen, 2011; Carmody, 2005; Cole, Grossman, Prilliman, & Hunsaker,
13
T a b
le 3
. C
o nf
ir m
at o ry
F ac
to r
A na
ly si
s (C
FA )
St ud
ie s
o f th
e B
ec k
D ep
re ss
io n
In ve
nt o ry
–S ec
o nd
E di
ti o n
(B D
I- II)
.
A rt
ic le
D es
cr ip
ti o n
# F
ac to
rs : S
ca le
N am
es C
FI T
LI N
FI R
M SE
A SR
M R
C an
el -C
in ar
ba s,
C ui
, &
La ur
id se
n (2
01 1)
50 0
U .S
. s am
pl e
St ee
r &
C la
rk (
19 97
) 2-
fa ct
o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 9
.0 4
C ar
m o dy
( 20
05 )
50 2
U nd
er gr
ad ua
te s
1- fa
ct o r
.8 6
.1 1
St ee
r &
C la
rk (
19 97
) 2-
fa ct
o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.8 7
.1 0
O
sm an
e t
al . (
19 97
) 3-
fa ct
o r:
N
eg at
iv e
A ffe
ct , P
er fo
rm an
ce
D iff
ic ul
ty , S
o m
at ic
.8 8
.1 0
C o le
, G ro
ss m
an , P
ri lli
m an
, &
H un
sa ke
r (2
00 3)
a 1 01
P sy
ch ia
tr ic
p at
ie nt
s St
ee r
& C
la rk
( 19
97 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
D o zo
is , D
o bs
o n,
& A
hn be
rg
(1 99
8) 51
1 U
nd er
gr ad
ua te
s St
ee r
& C
la rk
( 19
97 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 0
.0 6
10 22
U nd
er gr
ad ua
te s
St ee
r &
C la
rk (
19 97
) 2-
fa ct
o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 1
.0 6
G ro
th e
et a
l. (2
00 5)
22 0
A fr
ic an
A m
er ic
an s
2- fa
ct o r:
C o gn
it iv
e- So
m at
ic ,
A ffe
ct iv
e .9
4 .0
4
H ar
ri s
& D
’E o n
(2 00
8) 20
2 M
en w
it h
ch ro
ni c
pa in
O sm
an e
t al
. ( 19
97 )
3- fa
ct o r:
N
eg at
iv e
A tt
it ud
e, P
er fo
rm an
ce
D iff
ic ul
ty , a
nd S
o m
at ic
E le
m en
ts
.9 3
.0 5
.0 6
27 9
W o m
en w
it h
ch ro
ni c
pa in
O sm
an e
t al
. ( 19
97 )
3- fa
ct o r:
N
eg at
iv e
A tt
it ud
e, P
er fo
rm an
ce
D iff
ic ul
ty , a
nd S
o m
at ic
E le
m en
ts
.9 4
.0 5
.0 5
H ep
ne r,
H un
te r,
E de
le n,
Z
ho u,
& W
at ki
ns (
20 09
) 24
0 R
es id
en ti al
s ub
st an
ce
us er
s 1-
fa ct
o r
.8 8
.0 99
St
ee r
& C
la rk
( 19
97 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 0
.0 87
3-
fa ct
o r:
C o gn
it iv
e, A
ffe ct
iv e,
So
m at
ic .9
1 .0
86
(c on
tin ue
d)
14
A rt
ic le
D es
cr ip
ti o n
# F
ac to
rs : S
ca le
N am
es C
FI T
LI N
FI R
M SE
A SR
M R
Jo e,
W o o lle
y, B
ro w
n,
G ha
hr am
an lo
u- H
o llo
w ay
, &
B ec
k (2
00 8)
a 1 33
A fr
ic an
-A m
er ic
an s
ui ci
de
at te
m pt
er s
St ee
r &
C la
rk (
19 97
) 2-
fa ct
o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 2
.0 6
Jo in
er e
t al
. ( 20
01 )
84 4
ad ul
ts 1-
fa ct
o r
.8 4
.8 2
.0 76
2- fa
ct o r:
H o pe
le ss
ne ss
d ep
re ss
io n,
G
en er
al d
ep re
ss io
n .8
7 .8
4 .0
73
1, 60
4 ad
ul ts
1 fa
ct o r
.8 3
.8 2
.0 74
2- fa
ct o r:
H o pe
le ss
ne ss
d ep
re ss
io n,
G
en er
al d
ep re
ss io
n .8
4 .8
3 .0
73
68 0
ad ul
ts 1-
fa ct
o r
be nc
hm ar
k .8
4 .8
1 .0
81
2-
fa ct
o r:
H o pe
le ss
ne ss
d ep
re ss
io n,
G
en er
al d
ep re
ss io
n .8
5 .8
2 .0
74
1, 40
4 A
ir F
o rc
e C
ad et
s am
pl e
1- fa
ct o r
.8 6
.8 4
.0 57
2- fa
ct o r:
H o pe
le ss
ne ss
d ep
re ss
io n,
G
en er
al d
ep re
ss io
n .8
7 .8
5 .0
56
K ne
ip p,
K ai
ra lla
, S ta
cc ia
ri ni
, &
P er
ei ra
( 20
09 )
a 2 00
L o w
in co
m e
w o m
en B
ec k
et a
l. (1
99 6)
2 -f
ac to
r:
C o gn
it iv
e, A
ffe ct
iv e-
So m
at ic
1 un
co rr
el at
ed w
Y 13
a ss
ig ne
d
.9 5
.9 8
.0 76
.0 60
B
ec k
et a
l. (1
99 6)
2 -f
ac to
r C
o gn
it iv
e, A
ffe ct
iv e-
So m
at ic
1 co
rr el
at ed
.9 6
.9 9
.0 66
.0 57
B
ec k
et a
l. (1
99 6)
2 -f
ac to
r C
o gn
it iv
e, A
ffe ct
iv e-
So m
at ic
2 co
rr el
at ed
.9 6
.9 9
.0 66
.0 57
G
en er
al -S
o m
at ic
, C o gn
it iv
e un
co rr
el at
ed .9
7 .9
9 .0
62 .0
51
G
en er
al -S
o m
at ic
, C o gn
it iv
e co
rr el
at ed
.9 8
.9 9
.0 53
.0 49
(c on
tin ue
d)
T a b
le 3
. (c
o n
ti n
u e d
)
15
A rt
ic le
D es
cr ip
ti o n
# F
ac to
rs : S
ca le
N am
es C
FI T
LI N
FI R
M SE
A SR
M R
O sm
an , B
ar ri
o s,
G ut
ie rr
ez ,
W ill
ia m
s, &
B ai
le y
(2 00
8)
a 1 67
C lin
ic al
a do
le sc
en ts
St ee
r et
a l.
(1 99
8) 3
-f ac
to r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
- A
ffe ct
iv e,
G ui
lt -P
un is
hm en
t
.9 2
.9 1
.0 49
.0 46
St
ee r
et a
l. (1
99 9)
2 -f
ac to
r:
C o gn
it iv
e, A
ffe ct
iv e-
So m
at ic
.9 2
.9 1
.0 47
.0 45
O sm
an , K
o pp
er , B
ar ri
o s,
G
ut ie
rr ez
, & B
ag ge
( 20
04 )
40 8
A do
le sc
en t
ps yc
hi at
ry
pa ti en
ts St
ee r
& C
la rk
( 19
97 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 5
.8 9
.0 42
D
o zo
is e
t al
. ( 19
98 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
- V
eg et
at iv
e
.9 5
.8 9
.0 42
St
ee r
et a
l. (1
99 8)
3 -f
ac to
r:
C o gn
it iv
e, S
o m
at ic
-A ffe
ct iv
e,
G ui
lt -P
un is
hm en
t
.9 4
.8 8
.0 46
1-
fa ct
o r
.9 3
.8 7
.0 49
Pa
tt er
so n
et a
l. (2
01 1)
67 1
H ep
at it is
C p
at ie
nt s
St ee
r &
C la
rk (
19 97
) 2-
fa ct
o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 3
Se ga
l, C
o o lid
ge , C
ah ill
, &
O ’R
ile y
(2 00
8) 37
6 ad
ul ts
St ee
r et
a l.
(2 00
0) 2
-f ac
to r:
N
o nc
o gn
it iv
e, C
o gn
it iv
e .6
9 .0
8
Se ig
no ur
el , G
re en
, &
Sc hm
it z
(2 00
8) 58
2 T
re at
m en
t- se
ek in
g su
bs ta
nc e
us er
s 1-
fa ct
o r
.9 0
.9 8
.0 91
St
ee r
& C
la rk
( 19
97 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 2
.9 8
.0 82
St
ee r
& C
la rk
( 19
97 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e
& S
o m
at ic
(m
o di
fie d)
.9 3
.9 8
.0 75
3-
fa ct
o r:
C o gn
it iv
e, A
ffe ct
iv e,
So
m at
ic .9
4 .9
9 .0
67
(c on
tin ue
d)
T a b
le 3
. (c
o n
ti n
u e d
)
16
A rt
ic le
D es
cr ip
ti o n
# F
ac to
rs : S
ca le
N am
es C
FI T
LI N
FI R
M SE
A SR
M R
St ee
r, B
al l,
R an
ie ri
, & B
ec k
(1 99
9)
a C lin
ic al
ly d
ep re
ss ed
p at
ie nt
s St
ee r
et a
l. (2
00 0)
2 -f
ac to
r:
C o gn
it iv
e, N
o nc
o gn
it iv
e .9
2 .0
6
St o rc
h, R
o be
rt i,
& R
o th
(2
00 4)
41 4
U nd
er gr
ad ua
te s
St ee
r &
C la
rk (
19 97
) 2-
fa ct
o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 5
.9 4
T ho
m bs
, Z ie
ge ls
te in
, B ec
k,
& P
ilo te
( 20
08 )
47 7
H o sp
it al
a cu
te m
yo ca
rd ia
l pa
ti en
ts B
ec k
et a
l. (1
99 6)
2 -f
ac to
r: S
o m
at ic
- A
ffe ct
iv e,
C o gn
it iv
e .9
2 .9
6 .0
7 .0
8
St
ee r
& C
la rk
( 19
97 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.9 0
.9 6
.0 8
.0 8
3-
fa ct
o r:
G en
er al
, S o m
at ic
, C
o gn
it iv
e .9
2 .9
6 .0
7 .0
7
3-
fa ct
o r:
G en
er al
, S o m
at ic
, C
o gn
it iv
e si
m pl
ifi ed
.9 2
.9 6
.0 7
.0 7
T it o v
et a
l. (2
01 1)
a 1 72
D ep
re ss
ed p
at ie
nt s
1- fa
ct o r
.8 9
.8 8
.0 8
B ec
k et
a l.
(1 99
6) 2
-f ac
to r:
C
o gn
it iv
e, S
o m
at ic
-A ffe
ct iv
e .9
4 .9
4 .0
6
W hi
sm an
, P er
ez , &
R am
el
(2 00
0) 57
6 U
nd er
gr ad
ua te
s St
ee r
& C
la rk
( 19
97 )
2- fa
ct o r:
C
o gn
it iv
e- A
ffe ct
iv e,
S o m
at ic
.8 0
.0 8
W ie
be &
P en
le y
(2 00
5) 50
0 A
du lt s
B ec
k et
a l.
(1 99
6) 2
-f ac
to r:
C
o gn
it iv
e, A
ffe ct
iv e-
So m
at ic
.9 1
.0 5
N ot
e. N
R =
n o t
re po
rt ed
; C FI
= c
o m
pa ra
ti ve
f it in
de x;
T LI
= T
uc ke
r– Le
w is
in de
x; N
FI =
n o rm
ed f it in
de x;
R M
SE A
= r
o o t
m ea
n sq
ua re
e rr
o r
o f ap
pr o xi
m at
io n;
S R
M R
= s
ta nd
ar di
ze d
ro o t
m ea
n sq
ua re
r es
id ua
l. a U
nd er
po w
er ed
; m ea
ni ng
t he
s am
pl e
si ze
is le
ss t
ha n
th e
re co
m m
en de
d 10
:1 p
ar ti ci
pa nt
t o it
em r
at io
.
T a b
le 3
. (c
o n
ti n
u e d
)
Erford et al. 17
2003; Dozois et al., 1998; Hepner et al., 2009; Joe, Woolley, Brown, Ghahramanlou-Hollo- way, & Beck, 2008; Osman et al., 2004; Pat- terson et al., 2011; Seignourel et al., 2008; Storch, Roberti, & Roth, 2004; Thombs et al., 2008; Whisman et al., 2000) tested the Steer and Clark (1997) 2-factor model of Cognitive- Affective and Somatic dimensions. Only two of these studies were underpowered (Cole et al., 2003; Joe et al., 2008). The data from these studies, in general, also fit the model adequately, with CFIs ranging from .80 to .99 (median = .925) and RMSEAs ranging from .04 to .10 (median = .075). Thus, the Steer and Clark (1997) 2-factor model (Cognitive- Affective and Somatic) also appears to be a robust model when tested on both clinical and nonclinical samples. One slight modification on the Steer and Clark (1997) model was reported. In a study of 408 adolescent psychi- atric patients, Dozois et al. (1998) proposed Cognitive-Affective and Somatic-Vegetative dimensional modifications and derived a CFI of .95 and RMSEA of .042—an excellent fit- ting model.
Four studies across three articles tested the 3-factor (Cognitive, Affective, Somatic) model proposed by Dum et al. (2008). Hepner et al. (2009) assessed a sample of 240 residen- tial substance users and reported a CFI of .91 and RMSEA of .086, while Seignourel et al. (2008) tested the same model on a sample of 582 treatment-seeking substance users and derived a CFI of .94 and RMSEA of .067. Finally, in two analyses by Thombs et al. (2008), the 3-factor Dum et al. model and a slight modification of that model both yielded CFIs of .92, TLIs of .94, and RMSEAs of .07. All totaled, these four studies across three articles indicated marginal to adequate fits of the data to the Dum et al. 3-factor model.
Several other proposed models were tested, but in only one or two studies in single articles. These new proposed models usually stemmed from EFA solutions and should be viewed ten- tatively pending further replication as they were possibly merely sample dependent mod- els. For example, Osman et al.’s (1997) pro- posed 3-factor model consisting of Negative Attitude, Performance Difficulty, and Somatic
Elements was tested by Harris and D’Eon (2008) on a sample of patients with chronic pain. Harris and D’Eon reported CFIs of .93 and .94 for samples of 202 adult men and 279 adult women with chronic pain, respectively, while the RMSEA was a consistent .05, indi- cating a good fit. Unfortunately, we were unable to retrieve Osman et al.’s (1997) CFA because Volume 19 of Journal of Psychopa- thology Behavior Assessment was removed from the Springer website. Likewise, Steer et al.’s (2000) 2-factor model of Cognitive and Noncognitive was tested on a sample of clini- cally depressed patients by Steer, Ball, et al. (1999), resulting in a CFI of .92 and RMSEA of .06, an adequate fitting result.
Diagnostic Validity
Recall that the BDI-II was not norm-refer- enced; it is a criterion-referenced instrument. Beck, Steer, and Brown (1996) suggested that interpretations of BDI-II total scores should fall in the following descriptive depression categories: 0 to 13 is minimal, 14 to 19 is mild, 20 to 28 is moderate, and 29 to 63 is severe. Diagnostic validity (decision reliabil- ity) results were located in 21 studies and are summarized in Table 4. Some studies reported results only for what the authors judged to be an optimal cut-off score for the sample, while other studies provided data for multiple cut- off scores so clinicians could judge what an optimal score would be, depending on which parameter was most important for their sam- ple and decision-making criterion. For exam- ple, some practitioners may be most interested in percentage of correct classification, while others are more concerned about instrument sensitivity. When studies reported multiple cut-off scores (e.g., 10–20) and associated output, we summarized the output by present- ing data for those cut-off scores that were most commonly used across studies (e.g., 11, 13, 15, 17, 19). This was done so that the reader can conduct more efficient and stan- dardized interstudy comparisons, although interested readers should retrieve the original articles to view all diagnostic validity cut-off score output.
18
T a b
le 4
. D
ia gn
o st
ic V
al id
it y
St ud
ie s
o f th
e B
ec k
D ep
re ss
io n
In ve
nt o ry
–S ec
o nd
E di
ti o n
(B D
I- II)
.
A rt
ic le
n +
S am
pl e
D es
cr ip
ti o n
C ut
o ff
Se ns
it iv
it y
Sp ec
ifi ci
ty PP
P N
PP %
C C
A U
C
A rn
au , M
ea gh
er , N
o rr
is , &
B ra
m so
n (2
00 1)
34 0
ad ul
ts 18
.9 4
.9 2
.5 4
.9 9
.9 2
.9 6
B ae
r et
a l.
(2 00
0) —
Sa m
pl e
1 40
d ep
re ss
ed a
nd 5
5 no
nd ep
re ss
ed 11
.8 6
.8 5
B
ae r
et a
l. (2
00 0)
— Sa
m pl
e 2
29 d
ep re
ss ed
a nd
1 6
no nd
ep re
ss ed
11 .9
6 .3
1
C o ch
ra ne
-B ri
nk , L
o fc
hy , &
S ak
in o fs
ky
(2 00
0) 55
E m
er ge
nc y
ro o m
p at
ie nt
s 30
1. 00
.5 5
.3 6
1. 00
D e
So uz
a, J o ne
s, &
R ic
ka rd
s (2
01 0)
50 H
un ti ng
to n’
s di
se as
e pa
ti en
ts 13
.8 3
.7 1
.4 8
.9 3
.8 6
D o zo
is , D
o bs
o n,
& A
hn be
rg (
19 98
) 1,
02 2
un de
rg ra
du at
es <
13 ; >
19 .8
1 .9
2 .9
1
10
1. 00
.6 6
.4 8
1. 00
Ea
ck , S
in ge
r, &
G re
en o (
20 08
) 28
8 di
st re
ss ed
m o th
er s
11 .8
8 .5
3 .4
9 .9
0
13
.8 4
.6 5
.5 5
.8 9
15 .7
8 .7
4 .6
1 .8
7
16
( o pt
.) .7
6 .7
7 .6
3 .8
6
17
.6 8
.7 9
.6 3
.8 3
19 .6
5 .8
4 .6
7 .8
2
Fl ie
ge e
t al
. ( 20
09 )
43 8
pa ti en
ts w
it h
m ed
ic al
c o nd
it io
ns N
R .5
9
H o pk
o e
t al
. ( 20
08 )
23 d
ep re
ss ed
a nd
9 n
o nd
ep re
ss ed
13 1.
00 .6
7 .8
9 .9
1
16
.9 6
.7 8
.9 2
.9 1
18 .9
6 .8
9 .9
6 .9
4
22
.9 2
1. 00
1. 00
.9 4
Jo
ne s
et a
l. (2
00 5)
17 4
ad ul
t pa
ti en
ts w
it h
ep ile
ps y
11 .9
6 .7
8 .4
2 .9
9 .9
4 K
at z,
K o pe
k, W
al dr
o n,
D ev
in s,
&
T o m
lin so
n (2
00 4)
60 c
an ce
r pa
ti en
ts 10
.9 2
.7 7
.4 8
13 .9
2 .9
0 .6
9
16
.7 3
1. 00
1. 00
K
re fe
tz , S
te er
, G ul
ab , &
B ec
k (2
00 2)
10 0
ps yc
hi at
ri c
pa ti en
ts 24
.7 4
.7 0
.7 6
.6 7
.7 2
.7 8
Lo o sm
an , S
ie ge
rt , K
o rz
ec , &
H o ni
g (2
01 0)
62 r
en al
d is
ea se
p at
ie nt
s 13
.7 5
.9 0
.7 5
.9 0
.9 0
(c on
tin ue
d)
19
A rt
ic le
n +
S am
pl e
D es
cr ip
ti o n
C ut
o ff
Se ns
it iv
it y
Sp ec
ifi ci
ty PP
P N
PP %
C C
A U
C
Lo w
& H
ub le
y (2
00 7)
11 9
ca rd
ia c
pa ti en
ts 10
1. 00
.7 5
.1 8
1. 00
.8 4
11
.8 3
.7 6
.1 7
.9 9
13 .8
3 .8
4 .2
3 .9
9
17
.6 7
.9 2
.3 1
.9 8
O
sm an
, B ar
ri o s,
G ut
ie rr
ez , W
ill ia
m s,
&
B ai
le y
(2 00
8) 16
7 cl
in ic
al a
do le
sc en
ts 10
.8 7
.5 7
.7 2
.7 7
Pe rr
y &
G ilb
o dy
( 20
09 )
25 6
pr is
o n
in m
at es
21 .6
6 .6
8 .6
7
Se ig
no ur
el , G
re en
, & S
ch m
it z
(2 00
8) 58
2 tr
ea tm
en t-
se ek
in g
su bs
ta nc
e us
er s
15 1.
00 .3
6 .1
4 1.
00
17
.9 5
.4 1
.1 4
.9 9
19 .9
3 .5
0 .1
6 .9
8
Sh ea
n &
B al
dw in
( 20
08 )
39 5
un de
rg ra
du at
es 18
.4 0
1. 00
1. 00
.9 0
Sp
ri nk
le e
t al
. ( 20
02 )
13 7
un de
rg ra
du at
es in
t re
at m
en t
9 1.
00
11
.9 8
13 .9
1
15
.8 5
17 .8
2
Su br
am an
ia m
, H ar
re ll,
H un
tl ey
, & T
ra cy
(2
00 9)
14 5
ad o le
sc en
ts w
it h
su bs
ta nc
e ab
us e
11 .7
7 .4
7 .6
0 .6
7
13
.6 7
.5 8
.6 2
.6 4
15 .6
6 .6
5 .6
6 .6
5
17
.6 0
.7 5
.7 1
.6 5
19 .5
5 .7
5 .6
9 .6
2
T ur
ne r-
St o ke
s, K
al m
us , H
ir an
i, &
C le
gg
(2 00
5) 11
4 br
ai n
in ju
ry p
at ie
nt s
14 .7
4 .8
0 .6
9 .8
4
N ot
e. P
PP =
p o si
ti ve
p re
di ct
iv e
po w
er ; N
PP =
n eg
at iv
e pr
ed ic
ti ve
p o w
er ; %
C C
= p
er ce
nt c
o rr
ec t
cl as
si fic
at io
n; A
U C
= A
re a
un de
r th
e cu
rv e.
T a b
le 4
. (c
o n
ti n
u e d
)
20 Measurement and Evaluation in Counseling and Development 49(1)
Interpretations of which cut-off score may be optimal can be complicated by use of diverse samples (e.g., clinical or nonclinical participants), sample sizes, and type of clini- cal condition present (e.g., depression, renal disease, substance use). Mathematically speaking, the lower the cut-off score the higher the sensitivity (true positive rate; the proportion of participants with depression correctly identified by the BDI-II). Likewise, the lower the cut-off score the lower the spec- ificity (true negative rate; the proportion of participants without depression not identified by the BDI-II). Some authors use AUC and response operator characteristics to determine optimal cut-offs in a given data set, others use percentage of correct classifications, while still others judge optimal decision making as the lowest difference between sensitivity and specificity in conjunction with lowest differ- ence between positive predictive power (PPP) and negative predictive power (NPP). In reviewing Table 4, only 11 of the 21 studies reported percentage of correct classification or AUC: three studies supplied both, four sup- plied only AUC data, and four studies reported only percentage of correct classification results.
In trying to generalize from the available extant literature, 14 of the 21 located studies reporting BDI-II diagnostic validity presented only results for what those authors considered to be optimal cut-off scores. One study pro- vided data, but never revealed the actual cut- off score used (Fliege et al., 2009). Optimal cut-off scores for the remaining six studies were determined by the authors of this meta- analysis using the sensitivity–specificity and PPP/NPP differentiation described earlier. Optimal cut-off scores across the 20 diagnos- tic validity studies reporting results (i.e., elim- inating Fliege et al., 2009, which did not report a cut-off score), whether reported or judged, ranged from 10 to 30, with a median cut-off value of 13. Five of the studies chose 13 as the optimal value, and three studies each chose 11 or 18 as the cut-off score for optimal identification.
So how accurate are clinicians and researchers likely to be if they use a BDI-II
cut-off score of 13? Nine diagnostic validity studies reported results for a BDI-II cut-off score of 13 and when combined, the results across studies indicate median sensitivity of .83, specificity of .76, PPP of .66, and NPP of .90, yielding an estimated percentage of accu- rate classification of very near 80%.
Nonclinical Sample Distribution Characteristics
A final area of analysis involved identification of nonclinical samples from the extant litera- ture that reported sample statistics (i.e., mean and standard deviation). While samples that provided statistics by gender were of primary interest, total sample averages were more prevalent and also collected. Only six samples reported means and standard deviations disag- gregated by gender (Contreras et al., 2004; Dunn et al., 2012; Lewandowski et al., 2006; Malatras & Israel, 2013; Segal et al., 2008; Steer & Clark, 1997). When weighted and combined, these six studies yielded a male sample size of 2,006 with a mean of 6.43 and standard deviation of 6.05. The combined female sample size was 3,560 with a mean of 7.71 and standard deviation of 6.23. Thus, females self-reported slightly higher (+1.3) BDI-II raw scores than males.
While only six studies reported sample sta- tistics by gender, 24 studies (including these six) reported total sample statistics. When combined and weighted, 13,723 participants comprised the grand nonclinical sample, yielding a grand mean of 8.39 and standard deviation of 6.87. It should be noted that this grand mean was slightly higher than both the male and female means reported earlier because the six gender differentiation studies were only a subset of the 24 aggregated stud- ies comprising the grand mean.
Discussion
The BDI-II is one of the most commonly used depression inventories used for clinical screening and outcome research (Muller & Erford, 2012; Peterson et al., 2014). The results of this meta-analysis of 144 studies
Erford et al. 21
exploring the psychometric characteristics of the BDI-II English version provide robust estimates of internal consistency for both clin- ical (α = .91, k = 56, n = 13,447) and nonclini- cal (α = .88, k = 43, n = 17,966) samples. These cumulative estimates are slightly lower than those reported by Beck, Steer, and Brown (1996) for their clinical (α = .92) and nonclin- ical (α = .93) standardization samples, but quite adequate for screening level purposes. These estimates exceed the α = .80 criterion suggested by Erford (2013) for screening level purposes and approach the α = .90 mini- mum criterion suggested for diagnostic evalu- ations. On the other hand, test–retest reliability estimates for the current meta-analysis were r
tt = .68 (k = 6; n = 572; Mdn = 6 weeks) for
the clinical samples and r tt = .80 (k = 6; n =
990; Mdn = 6 weeks) for the nonclinical sam- ples. These test–retest reliability estimates are far lower than the r
tt = .93 reported by Beck,
Steer, and Brown (1996), although that single study used a sample of only 26 clinical par- ticipants retested after only 1 week. Instability of a construct like depression over a 6-week evaluation period is not unusual, but clini- cians and researchers should be aware that the BDI-II stability is likely to drop substantially beyond a single week re-administration period.
In terms of validity, this meta-analysis explored external (convergent and diagnostic) and structural elements (EFA and CFA) of the BDI-II score validity. The convergent validity coefficients presented in Table 1 show very robust estimates across all 43 measures located in the extant literature. Lipsey and Wilson (2001) suggested that Pearson rs of .1 indicate a small effect size, .3 a moderate effect size, and .5 a large effect size. Thus, vir- tually all convergent instruments displayed large effect sizes. Not surprisingly, self-report instruments (e.g., CES-D, ZDRS, HDI) gen- erally displayed much higher correlations with the BDI-II than clinician-completed inventories (e.g., HAM-D, DIS-IV). This often-seen phenomenon is due to common methods variance (Erford, 2013); common methods, like two depression self-rating inventories, usually have higher correlations
than unlike methods, like comparing scores from self-report and clinician report instru- ments. Note that most of the self-report instru- ments listed in Table 1 correlated close to r = .70 with the BDI-II, while the clinician-report instruments were closer to r = .50. The excep- tion to this was a very strong correlation between BDI-II and SCID scores (r = .67, k = 2, n = 712) found by combining the Sprinkle et al. (2002) and Watson et al. (2008) study results. With respect to the diagnostic validity studies (summarized in Table 4), a general cut-off score of 13 ordinarily should lead to optimal correct classifications in diverse sam- ples of clinical/nonclinical participants, but clinicians and researchers will want to evalu- ate the relative importance of sensitivity and specificity before making that decision.
In addition to the original Beck, Steer, and Brown (1996) study, 18 studies were located that performed EFAs using various method- ologies on diverse clinical and nonclinical samples and thus obtaining varying outcomes (see Table 2). The vast majority of these stud- ies should have been conducted as CFAs to test the fit of the rationally defined and inter- preted unidimensional model, and the Beck, Steer, and Brown (1996) 2-factor model (Somatic-Affective and Cognitive). Instead, more than a dozen models now exist, stem- ming from diverse methodologies and sample characteristics, and only the Steer and Clark (1997) 2-factor model (i.e., Cognitive-Affec- tive and Somatic) has garnered much support in the literature. Not surprisingly, 11 of the 18 EFAs supported some permutation of a 2-fac- tor model, although the three built-in subcon- structs of cognitive, affective, and somatic depression symptoms rotated around the two resulting factors. Two EFAs converged on a unidimensional model, three on a 3-factor model (basically Cognitive, Affective, Somatic), and one study each on 4- and 5-fac- tor models. These studies probably would have made greater contributions had they been conducted as CFAs to test existing model fit to the data.
Surprisingly, of the 21 CFA studies located (see Table 3) only six tested the unidimen- sional model’s fit to the data, and data from
22 Measurement and Evaluation in Counseling and Development 49(1)
only one sample (Osman et al., 2004) fit the model well. In future research it is essential that all CFA studies test this unidimensional model since it yields the BDI-II total raw score that is used in nearly all client score interpretations, and that has the best support- ing psychometrics (i.e., reliability and conver- gent validity estimates). Both the Beck, Steer, and Brown (1996) and Steer and Clark (1997) 2-factor models fit the data well in 6 and 14 CFA studies, respectively, and for both clini- cal and nonclinical samples. While the under- lying dimensions of the BDI-II continue to be debated, few CFA studies have provided help- ful insights by starting with testing the unidi- mensional model, then proceeding to additional model tests using nested proce- dures, although Osman et al. (2004) and Sei- gnourel et al. (2008) came closest. Each tested the unidimensional model, followed by the Steer and Clark (1997) 2-factor model (Cog- nitive-Affective and Somatic), and finally fol- lowed by a 3-factor permutation. In each instance, large clinical samples were used, the unidimensional model fit the data reasonably well, and the Steer and Clark (1997) model offered only a slight improvement in fit indi- ces. In each instance, the 3-factor permutation model did not offer a superior fit to the Steer and Clark (1997) model. Unfortunately, nei- ther of these studies tested Beck, Steer, and Brown’s (1996) 2-factor model (Somatic- Affective and Cognitive).
A question long plaguing users of the BDI- II is whether substantial sex differences exist between average male and female respon- dents. Beck, Steer, and Brown (1996) reported a mean statistically significant difference of 3.2 raw score points between females (M = 23.61, SD = 12.31, n = 317) and males (M = 20.44, SD = 13.28, n = 183) in the clinical standardization sample (n = 500) and 4.5 raw score points between females (M = 14.55, SD = 10.74, n = 67) and males (M = 10.04, SD = 8.23, n = 53) in the nonclinical standardiza- tion sample (n = 120). Given that the BDI-II is a criterion-referenced instrument with sug- gested interpretive ranges of 0 to 13 (mini- mal), 14 to 19 (mild), 20 to 28 (moderate), and 29 to 63 (severe), significant sex differences
of 3 to 5 points could lead to overidentifica- tion of females with depression or underiden- tification of males with depression. For example, a female mean of 14.55 on the non- clinical standardization sample means that at least half of the females had at least a mild degree of depression.
But the results of the current meta-analysis indicate that the sex differences, while still statistically significant, are probably much less than 3 to 5 points. By combining the non- clinical sample means and standard deviations across the six studies supplying disaggregated male and female data, the difference appears to be much closer to 1.3 BDI-II raw score points (females: M = 7.71, SD = 6.23, n = 3,560; males: M = 6.43, SD = 6.05, n = 2,006). Notice also that the means in these six studies are substantially lower than the male and female results reported originally by Beck, Steer, and Brown (1996). Thus, it can be con- cluded that while sex differences do exist, these differences probably lead to only slight overidentification of females and underidenti- fication of males in nonclinical samples. Counselors and researchers should take this into account when using the BDI-II for screen- ing purposes.
Even though the BDI-II was designed and is interpreted as a unidimensional criterion- referenced instrument, is it possible to use the instrument in a norm-referenced capacity? There are significant challenges in this regard. BDI-II score distributions, by design, approach a positive skew because it is assumed that most people will display mini- mal symptoms of depression. So the post hoc analyses and explanations that follow should be interpreted with caution because of the potential inaccuracy skewing can have on mean and standard deviation computation. However, recall from the Results section that combining the nonclinical sample means and standard deviations across 24 located studies (n = 13,723) resulted in a grand mean of 8.39 and a grand standard deviation of 6.87. Note that the grand mean of 8.39 is higher than either the male or female means reported ear- lier, because the male/female results came from a small subsample of six studies from
Erford et al. 23
the larger 24 sample set. In a positively skewed distribution, scores are somewhat truncated below the mean, but often appear reasonably normally distributed for scores above the mean. As such, if the grand mean is 8.39 and SD = 6.87, one standard deviation above the mean (i.e., z = +1.00) is a raw score of 15.26, which converts to a percentile rank of 84. Likewise, two standard deviations above the mean (i.e., z = 2.00) is a raw score of 22.13, which converts to a percentile rank of 98. This also means that the optimal cut-off score of 13 discussed earlier in the diagnostic validity studies reflects a z = .67, which falls at the 75th percentile.
These results are speculative, but given a sample size of 13,723 nonclinical participants, sampling error is probably quite small. In addition, interpretations should account for probable minor sex differences meaning that the grand female mean is probably closer to 9, and the grand male mean closer to 7.7. Such normative adjustments could prevent current overidentification of females and underidenti- fication of males. Furthermore, if norms are forthcoming in a new edition of the BDI, non- linear transformations should be used to cor- rect for the positively skewed distribution characteristics, should such skewing be evi- dent in standardization sample data.
Study Limitations
This comprehensive meta-analysis of BDI-II psychometric characteristics involved rigor- ous and conservative methodological consid- erations, including specific inclusion criteria, random effects modeling, multiple assess- ments for publication bias and variance homo- geneity, and use of standardized outcome measures. The methodology contained in this article also represents an interesting innova- tion and application of meta-analysis to the study of a single counseling instrument. Basi- cally, we have conducted an omnibus litera- ture review of all the reliability and validity score results and applied meta-analytic meth- odology to allow a combinatorial analysis and description. This methodological innovation can be replicated on other instruments to help
educate counselors and counseling research- ers about the most efficient use of the instru- ment.
While the rigor and methodological inno- vation are evident, a number of study limita- tions certainly exist and merit consideration. For example, we only selected studies of the English version of the BDI-II, reasoning that slight to significant variations in translations into other languages could infuse error into an aggregated analysis. Additionally, even proper translation of an instrument from a primary to a target language does not ensure similar psy- chometric properties, making the existence of bias a possibility (Gudmundsson, 2009). Other researchers are encouraged to explore the psychometric characteristics of various other language translations of the BDI-II and compare their results to these meta-analytic findings to judge the psychometric equiva- lence of translated versions. Likewise, the 144 articles selected into this meta-analysis all assessed the second edition of the BDI (BDI- II) published in 1996 by Beck, Steer, and Brown. Hundreds more studies likely exist in the literature conducted on the original BDI and BDI-IA, but variations in items constitut- ing those earlier versions again may have affected the instruments’ psychometric char- acteristics. Thus, choosing to analyze only studies using the English version of the BDI- II provided a standardized result, but on only a fraction of the available studies available on all versions of the BDI.
Nearly one-third of the EFA and CFA stud- ies were underpowered, and many of the con- vergent validity comparisons in Table 1 were conducted on only one or two studies or sam- ple sizes of lower than 100 participants. More convergent studies using these instruments and the BDI-II are necessary to increase con- fidence and power of the results. Also, the BDI-II has reached a point of development where additional EFA studies are not needed, and researchers should focus efforts on CFA, especially of the primary models previously generated (single factor model; Steer & Clark’s [1997] 2-factor model; Beck, Steer & Brown’s [1996] 2-factor model), and using nested analyses to directly compare model fit.
24 Measurement and Evaluation in Counseling and Development 49(1)
There was also a great deal of missing information in the diagnostic validity studies. The most helpful diagnostic validity studies reported sensitivity, specificity, PPP, negative predictive power, percentage of correct clas- sifications, and AUC estimates for a wide range of cut-off values. The least helpful stud- ies simply provided a portion of this output information for a single cut-off value deemed “optimal” by the researchers, even though the researchers often failed to mention their crite- ria for judging the value of some cut-off scores to be optimal. Results of EFA, CFA, and diagnostic validity studies are always “sample dependent,” and different researchers conducting similar (or different) analyses on varied samples will naturally obtain some- what different results. And the smaller the samples used, the greater the chances that substantial differences in results will occur.
A limitation of meta-analysis as a meth- odology is that variable results across diverse samples and studies are combined using a single metric (e.g., r
ic , r
tt , r). Heterogeneous
groupings can lead to analytic strengths and limitations. Theoretically, meta-analysis treats all similar studies as replications, thus allowing combinatorial analysis. However, this theoretical assumption is only true to the extent that samples share important charac- teristics in common or if the combined results are not affected by sample heteroge- neity.
Given the breadth of studies included in this meta-analysis, it seems that the BDI-II might be appropriate to use in various con- texts and among various populations, pending confirmation that the construct of depression exists in other cultures, the nomological net- work of such a construct is the same in other cultures, and that either the English version was used or that the translated version was properly translated, adapted, and empirically tested according to international best practice standards. Presently, the BDI-II has been translated into more than 20 languages and is currently used in Europe, the Middle East, Asia, and Latin America (Wang & Gorenstein, 2013). In our review of the literature, it was unclear whether these translated language
versions conformed to parallel translation best practices, including back-translation (AERA/ APA/NCME, 1999, Standard 9.7). If appro- priate translation procedures were not used and construct equivalence not established, slightly to significantly altered psychometric characteristics of other language versions may result, thus making comparisons with the English-version BDI-II problematic. Addi- tional research assessing the cross-cultural stability of the BDI-II factorial structure would help to further establish it internation- ally in both clinical and research settings (Far- aci & Tirrito, 2013).
The BDI-II is increasingly being used with adolescents, so we were surprised to encoun- ter so few convergent validity studies using instruments other than self-report versions. Adolescents usually have parents/guardians and teachers who can provide convergent per- spectives and this is common practice when assessing depression and anxiety in younger children. A needed area of future research is establishing convergent validity of the BDI-II with parent and teacher versions of depression scales rather than relying totally on adolescent self-report estimates. And this segues into a final concern related to the BDI-II: all items are “obviously” related to depression and no validity scales/items have been included. Thus, the examiner is completely dependent on the examinee for the truthfulness and accu- racy of the results. This is a valid criticism of any self-report instrument without a validity scale. Examinees with depression who do not want to be identified with depression mini- mize their responses to the items, leading to underidentification; examinees without depression who want to be identified with depression inflate their responses to the items, leading to overidentification. In addition, cul- tural differences in the reporting of depressive symptomatology may lead to additional dis- crepancies in identification (Probst, Laditka, Moore, Harun, & Powell, 2007). Although a screening instrument cannot replace clinical judgment, inclusion of a validity scale or at least several validity items on a future version of the BDI-II could help address these con- cerns.
Erford et al. 25
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
References
References marked with an asterisk were included in the meta-analysis. American Educational Research Association
(AERA), American Psychological Association (APA), & National Council on Measurement in Education (NCME). (1999). Standards for educational and psychological test- ing (3rd ed.). Washington, DC: American Psychological Association.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.
Arnau, R. C., Meagher, M. W., Norris, M. P., & Bramson, R. (2001). Psychometric evalu- ation of the Beck Depression Inventory-II with primary care medical patients. Health Psychology, 20, 112–119. doi:10.1037//0278- 6133.20.2.112
Baer, L., Jacobs, D. G., Meszler-Reizes, J., Blais, M., Fava, M., Kessler, R., . . . O’Laughlen, J. (2000). Development of a brief screening instrument: The HANDS. Psychotherapy and Psychosomatics, 69, 35–41.
Baffige, F. N. (2006). The validity and sensitivity of the Behavior and Symptom Identification Scale. Dissertation Abstracts International: Section B: The Sciences and Engineering, 67(1-B), 529.
Baker, L. I. (2009). Convergent and discrimi- nant validity of biculturalism and depression among northern plains American Indians. Dissertation Abstracts International: Section B: The Sciences and Engineering, 69(12-B), 7801.
Balsamo, M., Romanelli, R., Innamorati, M., Ciccarese, G., Carlucci, L., & Saggino, A. (2013). The state-trait anxiety inventory: Shadows and lights on its construct validity. Journal of Psychopathology and Behavioral Assessment, 35, 475–486. doi:10.1007/ s10862-013-9354-5
Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York, NY: Guilford Press.
Beck, A. T., Steer, R. A., Ball, R., & Ranieri, W. F. (1996). Comparison of Beck Depression Inventories–IA and –II in psychiatric outpa- tients. Journal of Personality Assessment, 67, 588–597.
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory (2nd ed.). San Antonio, TX: Psychological Corporation.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571.
Beck, C. T., & Gable, R. K. (2001). Further valida- tion of the Postpartum Depression Screening Scale. Nursing Research, 50, 155–164.
Becker, M. L. (2000). Assessing depression in women: Is the BDI-II biased? Dissertation Abstracts International: Section B: The Sciences and Engineering, 61(2-B), 1130.
Beretvas, S. N. (2010). Meta-analysis. In G. R. Hancock & R. O. Mueller (Eds.), The review- er’s guide to quantitative methods in the social sciences (pp. 255–263). New York, NY: Routledge.
Blom, E., Henje, L., Jan-Olov, S., & Eva Ingvar, M. (2010). The differentiation between depressive and anxious adolescent females and controls by behavioural self-rating scales. Journal of Affective Disorders, 122, 232–240. doi:10.1016/j.jad.2009.07.006
Bos, S. C., Pereira, A. T., Marques, M., Maia, B., Soares, M. J., Valente, J., . . . Azevedo, M. H. (2009). The BDI-II factor structure in pregnancy and postpartum: Two or three factors? European Psychiatry, 24, 334–340. doi:10.1016/j.eurpsy.2008.10.003
Brothers, B. M., Hae-Chung, Y., Strunk, D. R., & Andersen, B. L. (2011). Cancer patients with major depressive disorder: Testing a biobehav- ioral/cognitive behavior intervention. Journal of Consulting & Clinical Psychology, 79, 253– 260. doi:10.1037/a0022566
Brown, D. R. (2008). Assessment of spirituality in counseling: The relationship between spiritu- ality and mental health. Dissertation Abstracts International Section A: Humanities and Social Sciences, 69(6-A), 2154.
Bush, B. A., Novack, T. A., Schneider, J. J., & Madan, A. (2004). Depression following traumatic brain injury: The validity of the
26 Measurement and Evaluation in Counseling and Development 49(1)
CES-D as a brief screening device. Journal of Clinical Psychology in Medical Settings, 11, 195–201.
Cahill, J., Barkham, M., Twigg, E., Rees, A., Stiles, W. B., Hardy, G. E., & Evans, C. (2006). Convergent validity of the CORE measures with measures of depression for clients in cognitive therapy for depression. Journal of Counseling Psychology, 53, 253–259. doi:10.1037/0022-0167.53.2.253
Canel-Cinarbas, D., Cui, Y., & Lauridsen, E. (2011). Cross-cultural validation of the Beck Depression Inventory-II across U.S. and Turkish samples. Measurement and Evaluation in Counseling and Development, 44, 77–91. doi:10.1177/0748175611400289
Carleton, R. N., Gosselin, P., & Asmundson, G. J. G. (2010). The Intolerance of Uncertainty Index: Replication and extension with an English sample. Psychological Assessment, 22, 396–406. doi:10.1037/a0019230
Carmody, D. P. (2005). Psychometric characteristics of the Beck Depression Inventory-II with col- lege students of diverse ethnicity. International Journal of Psychiatry in Clinical Practice, 9, 22–28. doi:10.1080/13651500510014800
Chamberlain, J., & Huprich, S. K. (2011). The Depressive Personality Disorder Inventory and current depressive symptoms: Implications for the assessment of depressive personality. Journal of Personality Disorders, 25, 668– 680. doi:10.1521/pedi.2011.25.5.668
Chau, P. M., & Milling, L. S. (2006). Impact of dysphoria and self-consciousness on percep- tions of social competence: Test of the depres- sive realism hypothesis. Clinical Psychologist, 10, 99–108. doi:10.1080/13284200600939900
Chemerinski, E., Bowie, C., Anderson, H., & Harvey, P. D. (2008). Depression in schizo- phrenia: Methodological artifact or distinct fea- ture of the illness? Journal of Neuropsychiatry and Clinical Neurosciences, 20, 431–440.
Cochrane-Brink, K. A., Lofchy, J. S., & Sakinofsky, I. (2000). Clinical rating scales in suicide risk assessment. General Hospital Psychiatry, 22, 445–451.
Cole, J. C., Grossman, I., Prilliman, C., & Hunsaker, E. (2003). Multimethod valida- tion of the Beck Depression Inventory-II and Grossman-Cole Depression Inventory with an inpatient sample. Psychological Reports, 93, 1115–1129.
Coles, M. E., Gibb, B. E., & Heimberg, R. G. (2001). Psychometric evaluation of the Beck
Depression Inventory in adults with social anxiety disorder. Depression & Anxiety, 14, 145–148.
Colis, M. J., Steer, R. A., & Beck, A. T. (2006). Cognitive insight in inpatients with psychotic, bipolar, and major depressive disorders. Journal of Psychopathology & Behavioral Assessment, 28, 242–249. doi:10.1007/ s10862-005-9012-7.
Contreras, S., Fernandez, S., Malcarne, V. L., Ingram, R. E., & Vaccarino, V. R. (2004). Reliability and validity of the Beck depres- sion and anxiety inventories in Caucasian Americans and Latinos. Hispanic Journal of Behavioral Sciences, 26, 446–462. doi:10.1177/0739986304269164
Cukrowicz, K. C., & Joiner, T. E., Jr. (2007). Computer-based intervention for anxious and depressive symptoms in a non-clinical popu- lation. Cognitive Therapy and Research, 31, 677–693. doi:10.1007/s10608-006-9094-x
Cuthill, F. M., Espie, C. A., & Cooper, S. A. (2003). Development and psychometric properties of the Glasgow Depression Scale for people with a learning disability. Individual and career supplement versions. British Journal of Psychiatry: The Journal of Mental Science, 182, 347–353.
Davidson, M. A., Tripp, D. A., Fabrigar, L. R., & Davidson, P. R. (2008). Chronic pain assess- ment: A seven-factor model. Pain Research & Management: The Journal of the Canadian Pain Society, 13, 299–308.
De Souza, J., Jones, L. A., & Rickards, H. (2010). Validation of self-report depression rating scales in Huntington’s disease. Movement Disorders, 25, 91–96. doi:10.1002/ mds.22837
Di Benedetto, M., Lindner, H., Hare, D. L., & Kent, S. A. (2005). Cardiac Depression Visual Analogue Scale for the brief and rapid assess- ment of depression following acute coro- nary syndromes. Journal of Psychosomatic Research, 59, 223–229. doi:10.1016/j.jpsy- chores.2005.06.070
Di Benedetto, M., Lindner, H., Hare, D. L., & Kent, S. (2006). Depression following acute coronary syndromes: A comparison between the Cardiac Depression Scale and the Beck Depression Inventory II. Journal of Psychosomatic Research, 60, 13–20. doi:10.1016/j.jpsychores.2005.06.003
Dimitrov, M. D. (2012). Statistical methods for validation of assessment scale data in
Erford et al. 27
counseling and related fields. Alexandria, VA: American Counseling Association.
Downs, W. R., & Rindels, B. (2004). Adulthood depression, anxiety, and trauma symptoms: A comparison of women with nonabusive, abu- sive, and absent father figures in childhood. Violence and Victims, 19, 659–671.
Dozois, D. J. A. (2003). The psychometric characteristics of the Hamilton Depression Inventory. Journal of Personality Assessment, 80, 31–40.
Dozois, D. J. A., Dobson, K. S., & Ahnberg, J. L. (1998). A psychometric evaluation of the Beck Depression Inventory. Psychological Assessment, 10, 83–90.
Dum, M., Pickren, J., Sobell, L. C., & Sobell, M. B. (2008). Comparing the BDI-II and the PHQ-9 with outpatient substance abusers. Addictive Behaviors, 33, 381–387. doi:10.1016/j.add- beh.2007.09.017
Dunn, M. J., Rodriguez, E. M., Barnwell, A. S., Grossenbacher, J. C., Vannatta, K., Gerhardt, C. A., & Compas, B. E. (2012). Posttraumatic stress symptoms in parents of children with cancer within six months of diagnosis. Health Psychology, 31, 176–185. doi:10.1037/ a0025545
Duval, S., & Tweedie, R. (2000). Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta- analysis. Biometrics, 56, 455–463.
Eack, S. M., Singer, J. B., & Greeno, C. G. (2008). Screening for anxiety and depression in com- munity mental health: The Beck anxiety and depression inventories. Community Mental Health Journal, 44, 465–474. doi:10.1007/ s10597-008-9150-y
Erford, B. T. (2013). Assessment for counselors (2nd ed.). Belmont, CA: Cengage Wadsworth.
Erford, B. T., Savin-Murphy, J. A., & Butler, C. (2010). Conducting a meta-analysis of counseling outcome research: Twelve steps and practical procedures. Counseling Outcome Research and Evaluation, 1, 19-43. doi:10.1177/2150137809356682
Falck, R. S., Wang, J., Carlsin, R. G., & Siegal, H. A. (2006). Prevalence and correlates of current depressive symptomatology among a commu- nity sample of MDMA users in Ohio. Addictive Behaviors, 31, 90–101. doi:10.1016/j.add- beh.2005.04.017
Faraci, P., & Tirrito, A. (2013). Fifty years studying the Beck Depression Inventory (BDI) factorial stability without consistent results: A further
contribution. Clinical Neuropsychology, 10, 274–279.
Farran, B. (2004). Predictors of academic pro- crastination in college students. Dissertation Abstracts International: Section B: The Sciences and Engineering, 65(3-B), 1545.
Fliege, H., Becker, J., Walter, O. B., Bjorner, J. B., Klapp, B. F., & Rose, M. (2005). Development of a Computer-Adaptive Test for Depression (D-CAT). Quality of Life Research, 14, 2277–2291. doi:10.1007/ s11136-005-6651-9
Fliege, H., Becker, J., Walter, O. B., Rose, M., Bjorner, J. B., & Klapp, B. F. (2009). Evaluation of a computer-adaptive test for the assessment of depression (D-CAT) in clinical application. International Journal of Methods in Psychiatric Research, 18, 23–36.
Forand, N. R., & DeRubeis, R. J. (2013). Pretreatment anxiety predicts patterns of change in cognitive behavioral therapy and medications for depression. Journal of Consulting and Clinical Psychology, 81, 774– 782. doi:10.1037/a0032985
Forand, N., Gunthert, K., Cohen, L., Butler, A., & Beck, J. (2011). Preliminary evidence that anxiety is associated with accelerated response in cognitive therapy for depression. Cognitive Therapy and Research, 35, 151– 160. doi:10.1007/s10608-010-9348-5
Foster, T. W. (2009). Depression, anxiety, and attitude toward retirement as predictors of wellness for workers nearing retirement. Dissertation Abstracts International Section A: Humanities and Social Sciences, 69(7-A), 2615.
Gartner, A. M. (2012). The Elkins Distress Inventory: Development of a brief biopsycho- social battery for the assessment of pain and psychological distress in a chronic pain popu- lation. Dissertation Abstracts International: Section B: The Sciences and Engineering, 72(11-B), 7045.
Gibb, B. E., Butler, A. C., & Beck, J. S. (2003). Childhood abuse, depression, and anxiety in adult psychiatric outpatients. Depression & Anxiety, 17, 226–228. doi:10.1002/da.10111
Grothe, K. B., Dutton, G. R., Bodenlos, J., Ancona, M., Jones, G. N., & Brantley, P. J. (2005). Validation of the Beck Depression Inventory-II in a low-income African American sam- ple of medical outpatients. Psychological Assessment, 17, 110–114. doi:10.1037/1040- 3590.17.1.110
28 Measurement and Evaluation in Counseling and Development 49(1)
Gudmundsson, E. (2009). Guidelines for translat- ing and adapting psychological instruments. Nordic Psychology, 61(2), 29–45.
Halstead, J. E., Leach, C., & Rust, J. (2007). The development of a brief distress measure for the evaluation of psychotherapy and counseling (sPaCE). Psychotherapy Research, 17, 656– 672. doi:10.1080/10503300701235322
Hamilton, M. (1960). A rating scale for depression. Journal of Neurology and Neuroscience, 23, 56–62.
Harari, M. J., Waehler, C. A., & Rogers, J. R. (2005). An empirical investigation of a theo- retically based measure of perceived well- ness. Journal of Counseling Psychology, 52, 93–103. doi:10.1037/0022-0167.52.1.93
Harrington, J. A., & Blankenship, V. (2002). Ruminative thoughts and their relation to depression and anxiety. Journal of Applied Social Psychology, 32, 465–485.
Harris, C. A., & D’Eon, J. L. (2008). Psychometric properties of the Beck Depression Inventory- Second Edition (BDI-II) in individuals with chronic pain. Pain, 137, 609–622. doi:10.1016/j.pain.2007.10.022
Hedges, L. V., & Olkin, I. (1985). Statistical meth- ods for meta-analysis. Orlando, FL: Academic Press.
Heimlich, T. E. (1999). Coping, depression and anxiety as predictors of functioning in COPD. Dissertation Abstracts International: Section B: The Sciences and Engineering, 59(9-B), 5085.
Hepner, K. A., Hunter, S. B., Edelen, M. O., Zhou, A. J., & Watkins, K. (2009). A comparison of two depressive symptomatology measures in residential substance abuse treatment clients. Journal of Substance Abuse Treatment, 37, 318–325. doi:10.1016/j.jsat.2009.03.005
Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring inconsistency in meta-analysis. British Medical Journal, 327, 557–560. doi:10.1136/bmj.327.7414.557
Hopko, D. R., Bell, J. L., Armento, M. E., Robertson, S. M. C., Hunt, M. K., Wolf, N. J., & Mullane, C. (2008). The phenomenology and screening of clinical depression in cancer patients. Journal of Psychosocial Oncology, 26, 31–51. doi:10.1300/J077v26n01_03
Jay, M., & John, O. P. (2004). A depressive symp- tom scale for the California Psychological Inventory: Construct validation of the CPI- D. Psychological Assessment, 16, 299–309. doi:10.1037/1040-3590.16.3.299
Joe, S., Woolley, M. E., Brown, G. K., Ghahramanlou-Holloway, M., & Beck, A. T. (2008). Psychometric properties of the Beck Depression Inventory-II in low-income, African American suicide attempters. Journal of Personality Assessment, 90, 521–523. doi:10.1080/00223890802248919
Johnson, D. R., Vincent, A. S., Johnson, A. E., Gilliland, K., & Schlegel, R. E. (2008). Reliability and construct validity of the Automated Neuropsychological Assessment Metrics (ANAM) mood scale. Archives of Clinical Neuropsychology, 23, 73–85. doi:10.1016/j.acn.2007.10.001
Joiner, T. E., Jr., Steer, R. A., Abramson, L. Y., Alloy, L. B., Metalsky, G. I., & Schmidt, N. B. (2001). Hopelessness depression as a distinct dimension of depressive symptoms among clinical and non-clinical samples. Behaviour Research and Therapy, 39, 523–536.
Jones, J. E., Hermann, B. P., Woodard, J. L., Barry, J. J., Gilliam, F., Kanner, A. M., & Meador, K. J. (2005). Screening for major depression in epilepsy with common self-report depression inventories. Epilepsia (Series 4), 46, 731–735. doi:10.1111/j.1528-1167.2005.49704.x
Katz, M. R., Kopek, N., Waldron, J., Devins, G. M., & Tomlinson, G. (2004). Screening for depression in head and neck cancer. Psycho- oncology, 13, 269–280. doi:10.1002/pon.734
Kay, S. R., Fiszbein, A., & Opler, L. A. (1987). The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophrenia Bulletin, 13, 261–276.
Kim, Y., Pilkonis, P. A., Frank, E., Thase, M. E., & Reynolds, T. F. (2002). Differential func- tioning of the Beck Depression Inventory in late-life patients: Use of item response theory. Psychology and Aging, 17, 379–391. doi:10.1037/0882-7974.17.3.379
King, K. M., Colella, T. J. F., Faris, P., & Thompson, D. R. (2009). Using the Cardiac Depression Scale in men recovering from coro- nary artery bypass surgery. Journal of Clinical Nursing, 18, 1617–1624. doi:10.1111/j.1365- 2702.2009.02800.x
Kneipp, S. M., Kairalla, J. A., Stacciarini, J., & Pereira, D. (2009). The Beck Depression Inventory II factor structure among low- income women. Nursing Research, 58, 400–409.
Koby, D. G. (2009). Somatic symptoms in adult anxiety disorders and response to cognitive behavioral therapy. Dissertation Abstracts
Erford et al. 29
International: Section B: The Sciences and Engineering, 69(8-B), 5035.
Konstantinidis, A., Martiny, K., Bech, P., & Kasper, S. (2011). A comparison of the Major Depression Inventory (MDI) and the Beck Depression Inventory (BDI) in severely depressed patients. International Journal of Psychiatry in Clinical Practice, 15, 56–61. doi :10.3109/13651501.2010.507870
Krefetz, D. G., Steer, R. A., Gulab, N. A., & Beck, A. T. (2002). Convergent validity of the Beck Depression Inventory-II with the Reynolds Adolescent Depression Scale in psychiatric inpatients. Journal of Personality Assessment, 78, 451–460.
Leigh, I. W., & Anthony-Tolbert, S. (2001). Reliability of the BDI-II with deaf persons. Rehabilitation Psychology, 46, 195–202. doi:10.1037/0090-5550.46.2.195
Leonardson, G. R., Daniels, M. C., Ness, F. K., Kemper, E., Mihura, J. L., Koplin, B. A., & Foreyt, J. P. (2003). Validity and reliability of the general well-being schedule with northern plains American Indians diagnosed with type 2 diabetes mellitus. Psychological Reports, 93, 49–58.
Leventhal, A. M., Chasson, G. S., Tapia, E., Miller, E. K., & Pettit, J. W. (2006). Measuring hedonic capacity in depression: A psychomet- ric analysis of three anhedonia scales. Journal of Clinical Psychology, 62, 1545–1558. doi:10.1002/jclp.20327
Lewandowski, K. E., Barrantes-Vidal, N., Nelson- Gray, R. O., Clancy, C., Kepley, H. O., & Kwapil, T. R. (2006). Anxiety and depres- sion symptoms in psychometrically identified schizotypy. Schizophrenia Research, 83, 225– 235. doi:10.1016/j.schres.2005.11.024
Lindsay, W. R., & Lees, M. S. (2003). A compari- son of anxiety and depression in sex offenders with intellectual disability and a control group with intellectual disability. Sexual Abuse: Journal of Research and Treatment, 15, 339– 345. doi:10.1177/107906320301500409
Lindsay, W. R., & Skene, D. D. (2007). The Beck Depression Inventory II and the Beck Anxiety Inventory in people with intellectual disabilities: Factor analyses and group data. Journal of Applied Research in Intellectual Disabilities, 20, 401–408. doi:10.1111/j.1468- 3148.2007.00380.x
Lipps, G. E., Lowe, G. A., & Young, R. (2007). Validation of the Beck Depression Inventory-II
in a Jamaican university student cohort. West Indian Medical Journal, 56, 404–408.
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.
Loosman, W. L., Siegert, C. E. H., Korzec, A., & Honig, A. (2010). Validity of the Hospital Anxiety and Depression Scale and the Beck Depression Inventory for use in end-stage renal disease patients. British Journal of Clinical Psychology, 49, 507–516. doi:10.134 8/014466509X477827
Low, G. D., & Hubley, A. M. (2007). Screening for depression after cardiac events using the Beck Depression Inventory-II and the Geriatric Depression Scale. Social Indicators Research, 82, 527–543. doi:10.1007/s11205-006-9049-3
Malatras, J. W., & Israel, A. C. (2013). The influ- ence of family stability on self-control and adjustment. Journal of Clinical Psychology, 69, 661–670. doi:10.1002/jclp.21935
Mantere, O., Isometsä, E., Ketokivi, M., Kiviruusu, O., Suominen, K., Valtonen, H. M., . . . Leppämäki, S. (2010). A prospective latent analyses study of psychiatric comorbidity of DSM-IV bipolar I and II disorders. Bipolar Disorders, 12, 271–284. doi:10.1111/j.1399- 5618.2010.00810.x
Marguerat Degaard, F. (2005). The inspiration process: Its source and its role. Dissertation Abstracts International: Section B: The Sciences and Engineering, 66(3-B), 1725.
Maynard, D.-M. B., Campbell, M. H., Devonish, D., Leon, T., Emmanuel, M. K., & Roberti, J. W. (2010). Psychometric properties and factorial structure of the State Trait Anxiety Inventory- Trait Version in Caribbean university students. Journal of Psychology in Africa, 20, 7–12.
McCall, W. V., Reboussin, B. A., & Cohen, W. (2000). Subjective measurement of insom- nia and quality of life in depressed inpa- tients. Journal of Sleep Research, 9, 43–48. doi:10.1046/j.1365-2869.2000.00186.x
Mckenzie, J. S. (1998). An exploration of the relationship between forgiveness, anger, and depression with prisoners in a drug abuse treatment program. Dissertation Abstracts International: Section B: The Sciences and Engineering, 58(8-B), 4507.
Miller, J. D., Tant, A., & Bagby, R. M. (2010). Depressive personality disor- der: A comparison of three self-report measures. Assessment, 17, 230–240. doi:10.1177/1073191109356537
30 Measurement and Evaluation in Counseling and Development 49(1)
Mitchell, K. (2005). Assessing depression in patients with neurologic disease: Concurrent validation and cluster analysis of the emotional assessment system. Dissertation Abstracts International: Section B: The Sciences and Engineering, 66(5-B), 2833.
Mogge, N. L., Steinberg, J. S., Fremouw, W., & Messer, J. (2008). The Assessment of Depression Inventory (ADI): An appraisal of validity in an outpatient sample. Depression & Anxiety, 25, 64–68. doi:10.1002/da.20247
Moore, E. L. (2005). Neuropsychological function- ing and mild traumatic brain injury: The impact of anxiety sequelae. Dissertation Abstracts International: Section B: The Sciences and Engineering, 66(6-B), 3420.
Moustgaard, A. K. (2005). Mindfulness-based cog- nitive therapy (MBCT) for stroke survivors: An application of a novel intervention. Dissertation Abstracts International: Section B: The Sciences and Engineering, 65(11-B), 6054.
Muller, B. E., & Erford, B. T. (2012). Choosing assessment instruments for depression out- come research with school-aged youth. Journal of Counseling & Development, 90, 208-220.
Nobles, W. (2011). Understanding the presence of gerotranscendence among diverse racial and ethnic older adults in Florida. Dissertation Abstracts International: Section B: The Sciences and Engineering, 72(5-B), 3118.
Novy, D. M., Stanley, M. A., Averill, P., & Daza, P. (2001). Psychometric comparability of English- and Spanish-language measures of anxiety and related affective symptoms. Psychological Assessment, 13, 347–355. doi:10.1037//1040-3590.13.3.347
Osman, A., Barrios, F. X., Gutierrez, P. M., Williams, J. E., & Bailey, J. (2008). Psychometric properties of the Beck Depression Inventory-II in nonclinical adoles- cent samples. Journal of Clinical Psychology, 64, 83–102. doi:10.1002/jclp.20433
Osman, A., Downs, W. R., Barrios, F. X., Kopper, B. A., Gutierrez, P. M., & Chiros, C. E. (1997). Factor structure and psychometric character- istics of the Beck Depression Inventory-II. Journal of Psychopathology Behavior Assessment, 19, 359–376.
Osman, A., Kopper, B. A., Barrios, F., Gutierrez, P. M., & Bagge, C. L. (2004). Reliability and validity of the Beck Depression Inventory-II with adolescent psychiatric inpatients. Psychological Assessment, 16, 120–132. doi:10.1037/1040-3590.16.2.120
Palmer, E. J., & Binks, C. (2008). Psychometric properties of the Beck Depression Inventory-II with incarcerated male offenders aged 18–21 years. Criminal Behaviour and Mental Health, 18, 232–242. doi:10.1002/cbm.701
Patterson, A. L., Morasco, B. J., Fuller, B. E., Indest, D. W., Loftis, J. M., & Hauser, P. (2011). Screening for depression in patients with hepatitis C using the Beck Depression Inventory-II: Do somatic symptoms compro- mise validity? General Hospital Psychiatry, 33, 354–362. doi:10.1016/j.genhosppsych. 2011.04.005
Pedersen, S. S., Yagensky, A., Smith, O. R. F., Yagenska, O., Shpak, V., & Denollet, J. (2009). Preliminary evidence for the cross- cultural utility of the type D personality con- struct in the Ukraine. International Journal of Behavioral Medicine, 16, 108–115. doi:10.1007/s12529-008-9022-4
Perry, A. E., & Gilbody, S. (2009). Detecting and predicting self-harm behaviour in prisoners: A prospective psychometric analysis of three instruments. Social Psychiatry & Psychiatric Epidemiology, 44, 853–861. doi:10.1007/ s00127-009-0007-7
Peterson, C. H., Lomas, G. I., Neukrug, E. S., & Bonner, M. W. (2014). Assessment use by counselors in the United States: Implications for policy and practice. Journal of Counseling & Development, 92, 90–98. doi:10.1002/ j.1556-6676.2014.00134.x
Probst, J. C., Laditka, S. B., Moore, C. G., Harun, N., & Powell, M. P. (2007). Race and ethnicity differences in reporting of depressive symp- toms. Administration and Policy in Mental Health, 34, 519–529.
Proudfoot, J., Goldberg, D., Mann, A., Everitt, B., Marks, I., & Gray, J. A. (2003). Computerized, interactive, multimedia cognitive-behavioural program for anxiety and depression in general practice. Psychological Medicine, 33, 217– 227. doi:10.1017/S0033291702007225
Pyne, J. M., Tripathi, S., Williams, D. K., & Fortney, J. (2007). Depression-free day to util- ity-weighted score: Is it valid? Medical Care, 45, 357–362.
Radloff, L. S. (1977). The CES-D scale: A self report depression scale for research in the general population. Applied Psychological Measurements, 1, 385–401.
Rector, N. A., Kamkar, K., Cassin, S. E., Ayearst, L. E., & Laposa, J. M. (2011). Assessing excessive reassurance seeking in the anxiety
Erford et al. 31
disorders. Journal of Anxiety Disorders, 25, 911–917. doi:10.1016/j.janxdis.2011.05.003
Resnick, H., Acierno, R., Waldrop, A. E., King, L., King, D., Danielson, C., . . . Kilpatrick, D. (2007). Randomized controlled evalua- tion of an early intervention to prevent post- rape psychopathology. Behaviour Research and Therapy, 45, 2432–4247. doi:10.1016/j. brat.2007.05.002
Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university stu- dents’ academic performance: A systematic review and meta-analysis. Psychological Bulletin, 138, 353–387.
Rodríguez-Gómez, J. R., Dávila-Martínez, M. G., & Collazo-Rodríguez, L. C. (2006). Factor structure of the Beck Depression Inventory– Second Edition (BDI-II) with Puerto Rican elderly. Puerto Rico Health Sciences Journal, 25, 127–132.
Roebuck-Spencer, T. M., Yarboro, C., Nowak, M., Takada, K., Jacobs, G., Lapteva, L., . . .Bleiberg, J. (2006). Use of computerized assessment to predict neuropsychological functioning and emotional distress in patients with systemic lupus erythematosus. Arthritis & Rheumatism, 55, 434–441.
Rose, J. W. (2002). Validity and reliability of the geriatric depression scale with the elderly resid- ing in assisted living facilities. Dissertation Abstracts International Section A: Humanities and Social Sciences, 62(9-A), 2981.
Rosenthal, R. (1979). The file drawer problem and tolerance for null results. Psychological Bulletin, 86, 638–641.
Sales, E., Greeno, C., Shear, M. K., & Anderson, C. (2004). Maternal caregiving strain as a mediator in the relationship between child and mother mental health problems. Social Work Research, 28, 211–223.
Schulenberg, S. E., & Yutrzenka, B. A. (2001). Equivalence of computerized and conventional versions of the Beck Depression Inventory-II (BDI-II). Current Psychology, 20, 216–230.
Seel, R. T., & Kreutzer, J. S. (2003). Depression assessment after traumatic brain injury: An empirically based classification method. Archives of Physical Medicine and Rehabilitation, 84, 1621–1628.
Segal, D. L., Coolidge, F. L., Cahill, B. S., & O’Riley, A. A. (2008). Psychometric prop- erties of the Beck Depression Inventory-II (BDI-II) among community-dwelling older adults. Behavior Modification, 32, 3–20. doi:10.1177/0145445507303833
Seignourel, P. J., Green, C., & Schmitz, J. M. (2008). Factor structure and diagnos- tic efficiency of the BDI-II in treatment- seeking substance users. Drug and Alcohol Dependence, 93, 271–278. doi:10.1016/j.dru- galcdep.2007.10.016
Senior, A. C., Rhoades, H. M., Novy, D. M., Kunik, M. E., Wilson, N. L., & Stanley, M. A. (2007). Utility of telephone assessments in an older adult population. Psychology and Aging, 22, 392–397. doi:10.1037/0882-7974.22.2.392
Shean, G., & Baldwin, G. (2008). Sensitivity and specificity of depression questionnaires in a college-age sample. Journal of Genetic Psychology, 169, 281–292.
Sims, V. (2010). Utilization of the Brief Symptom Inventory with homeless men: A conver- gent validity study of the depressive symp- toms dimension. Dissertation Abstracts International: Section B: The Sciences and Engineering, 71(4-B), 2741.
Spielberger, C. D., Ritterband, L. M., Reheiser, E. C., & Brunner, T. M. (2003). The nature and measurement of depression. International Journal of Clinical and Health Psychology, 3, 209–234.
Sprinkle, S. D., Lurie, D., Insko, S. L., Atkinson, G., Jones, G. L., Logan, A. R., & Bissada, N. N. (2002). Criterion validity, severity cut scores, and test-retest reliability of the Beck Depression Inventory-II in a university coun- seling center sample. Journal of Counseling Psychology, 49, 381–385. doi:10.1037//0022- 0167.49.3.381
Steer, R. A., & Ball, R. (1997). Further evidence for the construct validity of the Beck Depression Inventory-II with psychiatric outpatients. Psychological Reports, 80, 443–446.
Steer, R. A., Ball, R., Ranieri, W. F., & Beck, A. T. (1999). Dimensions of the Beck Depression Inventory-II in clinically depressed outpa- tients. Journal of Clinical Psychology, 55, 117–128.
Steer, R. A., Cavalieri, T. A., Leonard, D. M., & Beck, A. T. (1999). Use of the Beck Depression Inventory for primary care to screen for major depression disorders. General Hospital Psychiatry, 21, 106–111.
Steer, R. A., & Clark, D. A. (1997). Psychometric characteristics of the Beck Depression Inventory-II with college students. Measurement and Evaluation in Counseling and Development, 30, 128–137.
Steer, R. A., Clark, D. A., Beck, A. T., & Ranieri, W. F. (1999). Common and specific dimensions of
32 Measurement and Evaluation in Counseling and Development 49(1)
self-reported anxiety and depression: The BDI-II versus the BDI-IA. Behaviour Research and Therapy, 37, 183–190.
Steer, R., Clark, D., Kumar, G., & Beck, A. (2008). Common and specific dimensions of self- reported anxiety and depression in adoles- cent outpatients. Journal of Psychopathology and Behavioral Assessment, 30, 163–170. doi:10.1007/s10862-007-9060-2
Steer, R. A., Kumar, G., Ranieri, W. F., & Beck, A. T. (1998). Use of the Beck Depression Inventory-II with adolescent psychiatric out- patients. Journal of Psychopathology and Behavioral Assessment, 20, 127–137.
Steer, R. A., Ranieri, W. F., Kumar, G., & Beck, A. T. (2003). Beck Depression Inventory-II items associated with self-reported symptoms of ADHD in adult psychiatric outpatients. Journal of Personality Assessment, 80, 58–63.
Steer, R. A., Rissmiller, D. J., & Beck, A. T. (2000). Use of the Beck Depression Inventory-II with depressed geriatric inpatients. Behaviour Research and Therapy, 38, 311–318.
Stice, B. D. (2008). Depression among injured workers: Measurement and prediction. Dissertation Abstracts International: Section B: The Sciences and Engineering, 68(8-B), 5594.
Storch, E. A., Roberti, J. W., & Roth, D. A. (2004). Factor structure, concurrent validity, and internal consistency of the Beck Depression Inventory–Second Edition in a sample of col- lege students. Depression and Anxiety, 19, 187–189. doi:10.1002/da.20002
Strain, J. D., & Shuff, I. M. (2010). Psychological well-being and level of outness in a popula- tion of male-to-female transsexual women attending a national transgender conference. International Journal of Transgenderism, 12, 230–240. doi:10.1080/15532739.2010.5 44231
Straits-Tröster, K., Fields, J. A., Wilkinson, S. B., Pahwa, R., Lyons, K. E., Koller, W. C., & Tröster, A. I. (2000). Health-related qual- ity of life in Parkinson’s disease after palli- dotomy and deep brain stimulation. Brain and Cognition, 42, 399–416. doi:10.1006/ brcg.1999.1112
Subramaniam, G., Harrell, P., Huntley, E., & Tracy, M. (2009). Beck Depression Inventory for depression screening in substance- abusing adolescents. Journal of Substance Abuse Treatment, 37, 25–31. doi:10.1016/j. jsat.2008.09.008
Sulkowski, M. L., Storch, E. A., Geffken, G. R., Ricketts, E., Murphy, T. K., & Goodman, W. K. (2008). Concurrent validity of the Yale–Brown Obsessive–Compulsive Scale– Symptom checklist. Journal of Clinical Psychology, 64, 1338–1351. doi:10.1002/ jclp.20525
Swan, A., Watson, H. J., & Nathan, P. R. (2009). Quality of life in depression: An important outcome measure in an outpatient cognitive- behavioural therapy group programme? Clinical Psychology & Psychotherapy, 16, 485–496. doi:10.1002/cpp.588
Tabachnick, B. G., & Fidell, L. S. (2013). Using multivariate statistics (6th ed.). Boston, MA: Allyn and Bacon.
Thombs, B. D., Ziegelstein, R. C., Beck, C. A., & Pilote, L. (2008). A general factor model for the Beck Depression Inventory-II: Validation in a sample of patients hospital- ized with acute myocardial infarction. Journal of Psychosomatic Research, 65, 115–121. doi:10.1016/j.jpsychores/2008.02.027
Titov, N., Dear, B. F., McMillan, D., Anderson, T., Zou, J., & Sunderland, M. (2011). Psychometric comparison of the PHQ-9 and BDI-II for mea- suring response during treatment of depression. Cognitive Behaviour Therapy, 40, 126–136. doi:10.1080/16506073.2010.550059
Turner-Stokes, L., Kalmus, M., Hirani, D., & Clegg, F. (2005). The Depression Intensity Scale Circles (DISCs): A first evaluation of a simple assessment tool for depression in the context of brain injury. Journal of Neurology, Neurosurgery & Psychiatry, 76, 1273–1278. doi:10.1136/jnnp.2004.050096
VanVoorhis, C., & Blumentritt, T. (2007). Psychometric properties of the Beck Depression Inventory-II in a clinically-identi- fied sample of Mexican American adolescents. Journal of Child & Family Studies, 16, 789– 798. doi:10.1007/s10826-006-9125-y
Vieira, Da, Silva Magalhaes, P., Tavares Pinheiro, R., Lessa Horta, B., Amaral Tavares Pinheiro, K., Azevedo Da Silva, R. (2008). Validity of the Beck Depression Inventory in the postpartum period. International Journal of Psychiatry in Clinical Practice, 12, 81–84. doi:10.1080/13651500701330775
Vikan, A., Hassel, A. M., Rugset, A., Johansen, H. E., & Moen, T. (2010). A test of shame in outpatients with emotional disorder. Nordic Journal of Psychiatry, 64, 196–202. doi:10.3109/08039480903398177
Erford et al. 33
Walter, L. J., Meresman, J. F., Kramer, T. L., & Evans, R. B. (2003). The Depression-Arkansas Scale: A validation study of a new brief depres- sion scale in an HMO. Journal of Clinical Psychology, 59, 465–481. doi:10.1002/ jclp.10137
Watson, D., O’Hara, M. W., Chmielewski, M., McDade-Montez, E. A., Koffel, E., Naragon, K., & Stuart, S. (2008). Further validation of the IDAS: Evidence of convergent, dis- criminant, criterion, and incremental valid- ity. Psychological Assessment, 20, 248–259. doi:10.1037/a0012570
Weeks, J. W., & Heimberg, R. G. (2005). Evaluation of the psychometric properties of the Beck Depression Inventory in a non- elderly adult sample of patients with general- ized anxiety disorder. Depression and Anxiety, 22, 41–44. doi:10.1002/da.20068
Wetherell, J., & Gatz, M. (2005). The Beck Anxiety Inventory in older adults with generalized anxiety disorder. Journal of Psychopathology & Behavioral Assessment, 27, 17–24. doi:10.1007/s10862-005-3261-3
Wetherell, J. L., Gatz, M., & Craske, M. G. (2003). Treatment of generalized anxiety disorder in older adults. Journal of Consulting and Clinical Psychology, 71, 31–40. doi:10.1037/0022- 006X.71.1.31
Williams, M. T., Wetterneck, C. T., Thibodeau, M. A., & Duque, G. (2013). Validation of the Yale-Brown Obsessive-Compulsive Severity Scale in African Americans with obsessive-compulsive disorder. Psychiatry Research, 209, 214–221. doi:10.1016/j.psy- chres.2013.04.007
Whisman, M. A., Perez, J. E., & Ramel, W. (2000). Factor structure of the Beck Depression Inventory-Second Edition (BDI-II) in a stu- dent sample. Journal of Clinical Psychology, 56, 545–551.
Wiebe, J. S., & Penley, J. A. (2005). A psycho- metric comparison of the Beck Depression Inventory-II in English and Spanish. Psychological Assessment, 17, 481–485. doi:10.1037/1040-3590.17.4.481
Wolfe-Christensen, C., Fedele, D. A., Kirk, K., Phillips, T. M., Mazur, T., Mullins, L. L., . . . Wisniewski, A. B. (2012). Degree of external genital malformation at birth in children with a disorder of sex development and subsequent caregiver distress. Journal of Urology, 188, 1596–1600. doi:10.1016/j.juro.2012.02.040
Yochim, B. P., Mueller, A. E., June, A., & Segal, D. L. (2011). Psychometric properties of the Geriatric Anxiety Scale: Comparison to the Beck Anxiety Inventory and Geriatric Anxiety Inventory. Clinical Gerontologist, 34, 21–33. doi:10.1080/07317115.2011.524600
Yoo, S. H., Matsumoto, D., & LeRoux, J. A. (2006). The influence of emotion recognition and emotion regulation on intercultural adjust- ment. International Journal of Intercultural Relations, 30, 345–363. doi:10.1016/j.ijin- trel.2005.08.006
Wang, Y. P., & Gorenstein, C. (2013). Psychometric properties of the Beck Depression Inventory-II: A comprehensive review. Revista Brasileira de Psiquiatria, 35(4), 416–431. doi:10.1590/1516- 4446-2012-1048
Zenger, N. L. (2010). Malingering of depression in a prison population: Evaluating the utility of the Beck Depression Inventory-II. Dissertation Abstracts International: Section B: The Sciences and Engineering, 70(9-B), 5855.
Zung, W. W., Richards, M. S., Gables, C., & Short, M. (1965). A self-rating depression scale. Archives of General Psychiatry, 12, 63–70.
Author Biographies
Bradley T. Erford, PhD, is a professor in the School Counseling Program of the Education Spe- cialties Department in the School of Education at Loyola University Maryland.
Erin Johnson is a doctoral student in the Coun- selor Education and Supervision program at Duquesne University.
Gerta Bardoshi, PhD, is an assistant professor in the School Counseling Program in the Department of Counseling at the University of Iowa.