Wiltink et al. BMC Psychiatry  (2017) 17:377 
DOI 10.1186/s12888-017-1545-2
RESEARCH ARTICLE Open Access
Mini - social phobia inventory (mini-SPIN):
psychometric properties and population
based norms of the German version
Jörg Wiltink1*, Sören Kliem2,3, Matthias Michal1, Claudia Subic-Wrana1, Iris Reiner1, Manfred E. Beutel1,
Elmar Brähler1 and Rüdiger Zwerenz1
Abstract
Background: A short screening for social anxiety disorder is useful in clinical and epidemiological contexts.
However, the German version of the short form of the Social Phobia Inventory (mini-SPIN) has not been evaluated
yet. Therefore, our aim was to determine reliability, validity and population based norms of the German mini-SPIN.
Methods: The mini-SPIN was evaluated in a clinical (N = 1254) and in a representative community sample (N = 1274).
Clinical diagnoses, the Patient Health Questionnaire depression (PHQ-9) and somatization modules (PHQ-15), the
Generalized Anxiety Disorder Scale (GAD-7), the Liebowitz Social Anxiety Scale (LSAS), and the Short-Form-12 Health
Survey (SF-12) were used in the clinical sample. In the community sample, participants filled out socio-demographic
and health related questions and short versions of the PHQ (PHQ-2, GAD-2, panic item). Internal consistency, test-retest
reliability, sensitivity to change, discriminant validity, and convergent validity were examined. Receiver operating
characteristic curve analyses were performed to determine cut-off scores. Population based norms were computed from
the community sample.
Results: We found internal consistencies between 0.80 and 0.83. Test-retest correlation was Rho = 0.61; sensitivity
to change was comparable to the LSAS. Correlations indicated good convergent and discriminant validity of the
mini-SPIN. Strict measurement invariance can be assumed regarding age and gender. Receiver operating characteristic
curve analysis suggested a cut-off of 6 or higher for a probable diagnosis of SAD.
Conclusions: The German version of the mini-SPIN is a reliable and valid instrument. Its brevity makes it valuable for
screening and assessing changes of social anxiety in clinical and epidemiological studies.
Keywords: Social anxiety disorder, Mini social phobia inventory, Mini-SPIN, Reliability, Validity, Cut-off, Norms
Background intense fear or anxiety. The fear/anxiety is out of propor-
According to DSM-5 (Diagnostic and statistical manual tion to actual threat. The fear/anxiety/avoidance has
of mental disorders, 5th edition) social anxiety disorder lasted 6 months, leads to significant distress or func-
(SAD) is marked by fear of situations where the individ- tional impairment, is not due to a medical condition/
ual is exposed to scrutiny by others; this may include drug or another mental disorder and either unrelated to
interaction, observation or performance situations. The existing medical conditions [1].
fears will act in a way or show anxiety that will lead to The average 12-month prevalence of SAD in the
being negatively evaluated [1]. Social situations almost German population is 2% [2, 3] and 7.4% in the US
always provoke anxiety and are avoided or endured with population [4]. Women are more likely than males to
develop SAD; mean age of onset is between age 10 and
16.6 years [5, 6]. It is a chronic and disabling disorder
* Correspondence: joerg.wiltink@unimedizin-mainz.de
1 often accompanied by comorbid depression, personalityDepartment of Psychosomatic Medicine and Psychotherapy, University
Medical Center of the Johannes Gutenberg University Mainz, Mainz, disorders, other anxiety disorders or substance abuse [5].
Germany Keller indicates that only a minority of patients with
Full list of author information is available at the end of the article
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 2 of 10
SAD attain full remission within 8 years [7]. Mistaken as a representative community sample regarding its a) reli-
shyness, SAD is often not recognized and therefore un- ability (internal consistency, test-retest reliability) and b)
treated [5, 7]. Because SAD remains undiagnosed – even discriminant and convergent aspects of validity. Further,
in psychosomatic outpatient and consultation-liaison ser- we wanted c) to determine cut-off scores for the detec-
vices – valid screening instruments are urgently needed tion of social anxiety, and d) to determine population
[8]. There are several valid questionnaires available asses- based norms.
sing social anxiety (performance anxiety and/or anxiety in
interactions); e.g. Liebowitz Social Anxiety Scale, LSAS Methods
[9–11], Social Phobia Scale, SPS [12], Social Interaction Study 1 (clinical sample)
Anxiety Scale, SIAS [12]. SIAS and SPS each consists of Participants
20 items, the LSAS consists of 24 items for the assessment A total of N = 1254 patients have been treated in the in-
of anxiety and 24 items for the assessment of avoidance. patient and day hospital units of the Department of Psy-
All of these instruments are relatively long and therefore chosomatic Medicine and Psychotherapy of the University
not feasible in settings with the need of brief orientation Medical Center of the Johannes Gutenberg University,
on symptoms (e.g. in general practice). Mainz between August 2010 and March 2015. Data were
Connor et al. [13] derived a short form with three routinely collected according to the German law of data
items from the 17-item self-administered Social Phobia protection (130a BDSG) and in accordance with the
Inventory (SPIN, [14]; German version [15]). guidelines in the Declaration of Helsinki.
Its three items are supposed to discriminate between The mean age of patients was 38.5 (Standard Devi-
individuals with generalized social anxiety disorder and ation, SD 13.2) ranging from 16 to 78 years. 61% were
controls: “Fear of embarrassment causes me to avoid female. 61% lived in a partnership. 48% had at least high
doing things or speaking to people”, “I avoid activities in school education. About one half of the sample was
which I am the centre of attention”, and “Being embar- employed, 7.8% were on pension, 19.5% were un-
rassed or looking stupid are among my worst fears”. The employed, and the others reported schooling, part-time
5-point-Likert rating scale ranges from 0=“not at all” to work or being responsible for household. The majority
4 = “extremely”. Using a cut-off score of 6 (range 0–12), of 94% of the patients held German nationality.
the English version of the mini-SPIN has demonstrated Most of the patients were diagnosed with a depressive
sensitivity of 89% and specificity of 90% for detecting disorder (81.3%), 28.3% with somatoform disorder, 21.8%
generalized social anxiety disorder [13, 16, 17]; psycho- with agoraphobia/panic disorder, 12.9% with generalized
metric properties of several translations in other lan- anxiety disorder, 9.6% with eating disorder, and 8.3%
guages have been demonstrated: Finnish [18], Spanish with social anxiety disorder. Furthermore, 17% of the pa-
[19], Portuguese [20, 21]. tients were diagnosed with a personality disorder. Mean
The 17-item SPIN has been translated into German duration of the inpatient or day hospital treatment was
and translated back by systematic techniques to ensure 48 (SD 19) days.
the original meaning of the items. It was translated into
German by a team of clinical psychology researchers and Measures
translated back by a bilingual clinical psychologist. Fi- In this inpatient and day hospital sample, mental disorders
nally, the back-translated version of the German SPIN were clinically assessed by psychotherapists according to
was reviewed and consensually approved by a team [15]. ICD-10 (International Statistical Classification of diseases,
The three items of the German version of the mini-SPIN 10th edition [23]). Diagnoses were approved by the senior
are identical with the three corresponding items of the physicians or psychologists in regular supervisions.
German translation of the 17-item SPIN. Patients are routinely assessed at the beginning and at
While the psychometric properties of the 17-item the end of their treatment by several questionnaires includ-
SPIN have been assessed in an earlier community survey ing measures on anxiety, depression and quality of life.
[22], reliability and validity of the German short form Depression was measured by the Patient Health Ques-
(mini-SPIN) are unknown. tionnaire (PHQ-9 [24, 25]). Examples for items of the
Compared to the longer 17-item SPIN and other exist- PHQ-9 are: “Little interest or pleasure in doing things?” or
ing scales a very short form of the questionnaire with “Poor appetite or overeating.” (0 = “not at all”, 1 = “several
sufficient psychometric properties is particularly less days”, 2 = “over half the days”, and 3 = “nearly every day”).
time consuming during assessment, thus more cost- Psychometric qualities of the PHQ-9 are comparable to
efficient and applicable in clinical (e.g. general practice) clinical interviews [26]. Internal consistency of the PHQ-9
and scientific contexts (e.g. community surveys). was good (Cronbach’s alpha = 0.88) [27, 28]. In a meta-
Therefore, the aim of this study was to evaluate the analysis with more than 5000 participants in a primary
German translation of the mini-SPIN in a clinical and in care setting including 17 validation studies Gilbody et al.
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 3 of 10
(2007) found a sensitivity of 92% and a specificity of 80% für Umfragen, Methoden und Analysen; independent ser-
for the detection of major depression (cut-off > = 10) [29]. vice for surveys, methods and analyses in market and social
Anxiety was screened with the GAD-7 (Generalized Anx- research), which is an institute for demographic research.
iety Disorder Scale, GAD-7 [30, 31]); e.g. “Trouble relaxing” The German law of data protection (§ 30a BDSG, Bundes-
(0 = “not at all”, 1 = “several days”, 2 = “over half the days”, datenschutzgesetz) was regarded and written consent was
and 3 = “nearly every day”). Internal consistency of the obtained. Ethics were weighted to the interests of the public
GAD-7 can be rated as good (Cronbach alpha = 0.89) [32]. and individuals concerned following 1823 (BGB, Bundesge-
A sum score of 10 and more indicates generalized anxiety setzbuch) of the Civil Code of Law and in accordance with
with a good sensitivity (89%) and specificity (82%) [31]. the guidelines in the Declaration of Helsinki. All data were
Somatic symptoms were assessed with the PHQ-15 of collected by the end of 2006. Data assessment was based on
the Patient Health Questionnaire [27, 33]. The question- 129 sample areas which represented the different socioeco-
naire contains the 15 most common complaints covering nomic structures of Germany. Households were selected
the main DSM-IV criteria for the diagnosis of randomly. The members of the households fulfilling the pre-
somatization disorder. Examples for items are: “Stomach defined inclusion criteria were also selected by random pro-
pain” or “Dizziness” (0 = “not bothered a lot”, 1 = “both- cedure. Participants were included when German was the
ered a little”, 2 = “bothered a lot”). The internal native language and when they were 14 or more years of
consistency of the PHQ-15 was good (Cronbach alpha = age. Firstly, 2157 addresses were attempted following a ran-
0.89) [33]. For the PHQ-15 van Ravesteijn et al. found a dom procedure; 2079 of the addresses were valid. Selected
sensitivity of 78% and a specificity of 71% for the detection persons were tried to contact for three times. (for detailed
of a somatoform disorder [34]. description of the data collection cf. [36]).
The Liebowitz Social Anxiety Scale (LSAS [9, 10]) was This survey was independent from the 2002 survey
used to assess intensity of fear in 24 social situations (e.g. assessing the psychometric properties of the German 17-
“Participating in small groups – having a discussion item SPIN. 1287 persons between 14 and 90 years agreed
with a few others”; 0 = “none”, 1 = “mild”, 2 = “moderate”, to participate (61.9% of valid addresses). All participants
3 = “severe” fear or anxiety) and their avoidance in were contacted by trained interviewers in their homes.
this situation (0 = “never”, 1 = “occasionally”, 2 = “often”, Self-rating questionnaires were presented. Interviewers of-
3 = “usually”) by self-report. The LSAS demonstrates fered help in case of difficulties to understand single ques-
good internal consistency for the total score (Cronbach tions. 13 subjects did not complete the mini-SPIN validly.
alpha 0.96) [11]. Considering sensitivity and specificity Therefore, 1274 participants were included into further
Mennin et al. (2002) identified a cut-off score of 30 for the analysis. The sample was representative for the German
probable diagnosis of a social anxiety disorder and a cut- population in terms of age, gender, and education.
off score of 60 for a generalized social anxiety disorder [9]. Mean age was 48.8 (SD 18.2) ranging from 14 to
Subjective quality of life was assessed with the German 90 years. 54.2% were female. 54% of the participants
version of the Short-Form-12 Health Survey (SF-12) as a were married, and 61% lived in a partnership. A total of
common, reliable and valid instrument for evaluating 88% had less than high school education. Household in-
various aspects of health status. It examines two main come was mostly (75%) higher than Euro 1250 per
components by eight health-related concepts: The ‘physical month. One third of the sample was employed, whereas
health component’ (PHC; e.g. “Pain interferes with normal 31% were on pension and 6% were unemployed. The
work”; 1 = “extremely”, 2 = “quite a bit”, 3 = “moderately”, majority of 97% held German nationality. A total of 19%
4 = “a little bit”) consists of the subscales ‘physical function- were residents of the Eastern states of Germany.
ing,’ ‘role-physical,’ ‘bodily pain’ and ‘general health’; the
‘mental health component’ (MHC; e.g. “Felt calm and Measures
peaceful”; 1 = “none of the time”, 2 = “a little of the time”, Participants filled out standardised self-report inventories
3 = “some of the time”, 4 = “a good bit of the time”, and on socio-demographic (e.g. age, gender, income) and
5 = “most of the time”) contains the subscales ‘mental health related questions (e.g. weight, height, health behav-
health’, ‘role-emotional,’ ‘social functioning’ and ‘vitality’. As iour, health care utilization; for detailed description of the
unit of measurement the total sum for both scales is assessment cf. [36]). In addition to the mini-SPIN, we used
calculated [35]. The reliability of the SF-12 was judged as the German version of the Patient Health Questionnaire
satisfactory to good [35]. (PHQ) to assess generalized anxiety with the two screen-
ing items of the GAD-7 [30, 31, 37]: “Feeling nervous,
Study 2 (community sample) anxious or on edge”, “Not being able to stop or control
Participants worrying”. (0=“not at all”, 1 = “several days”, 2 = “over half
A representative German community survey was conducted the days”, and 3 = “nearly every day”). The internal
by the USUMA GmbH (Unabhängige Serviceeinrichtung consistency of the two items was good (Cronbach alpha =
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 4 of 10
0.82) [31]. A sum score of 3 and more (range 0–6) indi- <35 years; Group 2: males 35 to 50 years; Group 3: males
cates generalized anxiety with good sensitivity (86%) and 51 to 64 years; Group 4: males >64 years; Group 5: fe-
specificity (83%) [31]. Panic was assessed with the screen- males <35 years; Group 6: females 35 to 50 years; group
ing question of the PHQ [38]: “In the last 4 weeks, have 7: females 51 to 64 years; Group 8: females >64 years.
you had an anxiety attack – suddenly feeling panic or In the case of partial measurement invariance (one or
fear?”. Item sensitivity for detecting a panic disorder is more model parameters identified, that were found to be
very good (93%), with a moderate specificity of 78% [38]. variant across samples), we followed the recommenda-
Depression was measured using the two-item depres- tion of Byrne et al. (1989) to only conduct further invari-
sion module of the PHQ [39]: “Little interest or pleasure ance tests, when a minimum of two parameters per
in doing things”, “Feeling down, depressed, or hopeless” invariance test were found [40] (e.g., at least two factor
0 = “not at all”, 1 = “several days”, 2 = “over half the loadings equivalent in metric invariance tests). If multi-
days”, and 3 = “nearly every day”. The internal variate normality assumption was violated, we used the
consistency of the PHQ-2 was good (Cronbach alpha = Satorra and Bentler’s (2001) scaling method [41]. We
0.83). For the detection of major depressive disorder, a used a series of increasingly stringent model comparison
cut-off score of three has a sensitivity of 87%, and a spe- steps to assess the factorial invariance of the mini-SPIN.
cificity of 78%. Sensitivity for the detection of any de- First, weak invariance was tested. This is necessary for
pressive disorder was 79%, specificity 86% [39]. unbiased comparison of structural relationships (e.g.,
correlation coefficients, structural [path] coefficients) be-
Statistical analyses tween latent constructs in different groups. Second,
Means, standard deviations, skewness and kurtosis were strong invariance was tested which allows the compari-
calculated for each item of the mini-SPIN. Additionally, we son of means of the latent construct between groups.
determined the corrected item-scale correlation for each Lastly, strict invariance was tested which allows un-
item and Cronbach’s alpha for the scale [28]. For the popu- biased decisions in screening processes that depend on
lation based norms we used cumulated percentages of the the expression of a construct, resulting in different error
sum score of the scale separately for age and gender. Be- rates (e.g., sensitivity, specificity) for different groups
cause scores are not normally distributed (especially in the (see fig. 1 for further details of the different measure-
community sample) non-parametric analyses (Mann-Whit- ment models).
ney tests, Spearman-Rho correlations) were performed. We used scaled CFI (comparative fit index) differences
To test sensitivity to change, we calculated pre- to (ΔCFI) as well as scaled RMSEA (Root Mean Square Error
post-intervention within group effect sizes (ESpre-post) for of Approximation) differences (ΔRMSEA) to compare the
the mini-Spin total score and the LSAS total score using difference stages of measurement invariance. As recom-
the clinical sample. We subsequently compared the rele- mended by Chen (2007), a change of .010 in ΔCFIscaled,
vant ESpre-post for the two measures and checked for any supplemented by a change of ΔRMSEAscaled = 0.015, was
significant differences. ESpre-post were calculated by regarded as indicative of non-invariance [42].
standardizing pre-post/pre-follow-up mean differences To evaluate the goodness of fit of the relevant model
for each intervention group by the standard deviation in general, we follow the recommendation of Hu & Ben-
(SD) of the difference. tler (1999): A CFI > .900 was supposed for an adequate
To determine optimal cut-offs, sensitivity, and specifi- and a CFI > .950 for a good model fit [43]. Regarding the
city we used receiver operating characteristic (ROC) RMSEA a value of RMSEA < .050 were supposed for a
curves. We applied three criteria for these analyses. The close fit, values between.050 and .080 for a reasonably
clinical diagnosis of social anxiety was used. Being aware close fit, and values > .080 represent an unacceptable
of the relatively low number of diagnoses (8.3%, also see model fit. These analyses were conducted using the sta-
[8]) we additionally used internationally validated cut-offs tistics software R (Version 3.2.5, [44]), with R Package
of the LSAS as criteria [9]. A cut-off of 60 indicated a gen- lavaan [45].
eralized social anxiety and a cut-off of 30 a social anxiety.
These statistical computations were done with SPSS
Statistics 23. Results
Level of significance was defined by p < .05; larger ef- Study 1 (clinical sample)
fects (p < .01, p < .001) were reported additionally. We Internal consistencies
did not perform alpha-adjustment because of the ex- Cronbach’s alpha of the three items for the clinical sam-
ploratory nature of the analyses. ple was 0.83. Table 1 displays the item characteristics
To examine the levels of measurement invariance, a and internal consistencies of the three items of the mini-
multi-group confirmatory factor analysis was conducted SPIN. For better reading we reported scale means not
(MGCFA) using these group variable: Group 1: males sum scores (c.f. Table 1).
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 5 of 10
Convergent and discriminant validity
In order to determine validity, the mini-SPIN was corre-
lated with questionnaires covering similar (convergent
validity) and divergent constructs (discriminant validity).
The mini-SPIN was closely related to social anxiety
measured with the LSAS (Rho = 0.704, p < 0.001). It was
also related to depression (PHQ-9, Rho = 0.485, p <
0.001), generalized anxiety (GAD-7, Rho = 0.455, p <
0.001) and somatization (PHQ-15, Rho = 0.266, p <
0.001). Lower scores in the mini-SPIN were associated
with a higher mental health component regarding qual-
ity of life (MHC, Rho = −0.391, p < 0.001). There was no
relation to the physical health component (PHC, Rho =
−0.058, p = 0.070).
Sensitivity and specificity
We determined sensitivity and specificity of different
cut-offs of the mini-SPIN for different criteria: clinical
Fig. 1 Explanation of the different models regarding measurement
invariance analysis. Notes: Weak Invariance (Model0): λ1 = 1; λ _group A diagnosis of a social anxiety disorder, generalized social2
=…= λ2_group H; λ3_group A =…= λ3group H. Strong Invariance anxiety based on LSAS >60, and social anxiety based on
(Model1): α1_group A = 0; τ1_group A =…= τ1_group H; τ2_group LSAS >30. For the criterion, clinical diagnosis of social
A=…= τ2_group H; τ3_group A =…= τ3_group H; + weak invariance. anxiety disorder, we could analyze N = 1012 patients. A
Strict Invariance (Model2): Var (ε1_group A) =…= Var (ε1_group H); Var total of N = 87 (8.6%) were diagnosed with a social anx-
(ε2_group A) =…= Var (ε2_group H); Var (ε3_group A) =…= Var
(ε3_group H); + weak and strong invariance. Strict Invariance (Model2b): iety disorder. Mean age of them was 30 (SD = 9) years
Var (ε1_group A) =… = Var (ε1_group H) ≠ Var (ε1_group C); Var and N = 36 (41%) were female. For the two LSAS based
(ε2_group A) =… = Var (ε2_group H); Var (ε3_group A) =… = Var criteria we could analyze N = 1007 patients. A total of
(ε3_group H); + weak and strong invariance N = 734 had a LSAS >30 (mean age 39 years, SD = 13;
female N = 459, 63%); N = 405 had a LSAS >60 (mean
Test-retest reliability age 38 years, SD = 13; female N = 252, 62%).
The correlation between the mini-SPIN at beginning Results can be found in Table 2.
and the end of the treatment after a mean of 48 days of
treatment was Rho = 0.61 (p < 0.001). Study 2 (community sample)
Internal consistencies
Cronbach’s alpha of the three items in the community
Sensitivity to change sample was 0.80. Table 1 gives an overview.
For the mini-Spin total score we found a pre- to post-
intervention ES of 0.37 (95%-CI [0.31; 0.44]). For the Correlates of the mini-SPIN
LSAS we found a similar ESpre-post = 0.33 (95%-CI [0.26; The mini-SPIN was unrelated to age (Rho = 0.015,
0.40]). Thus, sensitivity to change seems comparable be- p = 0.600). However, participants reaching the cut-off of
tween both measures. at least 6 points were significantly younger (z = −2.03,
Table 1 Item and scale characteristics of the mini-SPIN
Original item (Connor et al. [14]) German translation (Stangier & Steffens [15]) Inpatient/ day hospital sample Representative community
(N = 1082) sample (N = 1274)
M SD Skew Kurt rit M SD Skew Kurt rit
Fear of embarrassment causes me Aus Angst vor Verlegenheit vermeide ich 1.62 1.26 0.30 −1.00 0.74 0.32 0.63 2.13 4.48 0.65
to avoid doing things or speaking es, bestimmte Dinge zu tun oder Personen
to people. anzusprechen
I avoid activities in which I am the Ich vermeide Aktivitäten, durch die ich im 1.93 1.29 0.03 −1.02 0.78 0.56 0.87 1.49 1.53 0.65
centre of attention Mittelpunkt der Aufmerksamkeit stehe
Being embarrassed or looking Sich zu schämen oder dumm zu wirken, 1.66 1.34 0.30 −1.10 0.79 0.32 0.66 2.46 6.79 0.69
stupid are among my worst fears gehört zu meinen schlimmsten Ängsten
Total Scale (mean) 1.74 1.12 0.18 −0.85 0.83a 0.40 0.61 1.86 3.62 0.80a
M=mean, SD = standard deviation, Skew = skewness, Kurt = kurtosis, rit = corrected item scale correlation,
a Cronbach’s alpha; possible answers: 0 = “not at all”
to 4 = “extremely”
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 6 of 10
Table 2 Sensitivity and specificity of the mini-SPIN in a clinical sample
Cut-off Clinical diagnosis of SAD (N = 87)1 LSAS > 60 ‘generalized social anxiety’ (N = 405)2 LSAS > 30 ‘social anxiety’ (N = 734)2
Sensitivity specificity sensitivity specificity sensitivity specificity
1 1.000 .116 .993 .163 .960 .264
2 .989 .170 .993 .251 .947 .421
3 .966 .253 .978 .365 .896 .560
4 .920 .357 .938 .518 .816 .718
5 .874 .464 .881 .643 .714 .824
6 .851 .568 .800 .756 .604 .897
7 .690 .654 .694 .832 .495 .930
8 .552 .762 .541 .914 .354 .960
9 .448 .830 .415 .952 .257 .971
10 .310 .902 .262 .980 .155 .985
11 .218 .936 .180 .992 .105 .996
12 .092 .964 .096 .995 .056 .996
Bold type = cut-off with best balance regarding sensitivity and specificity; missing data: 1N = 221, 2N = 247
p < .05). The frequency (days per week) of alcohol intercepts, and item residual variances for a unidimensional
consumption (Rho = 0.000, p = 0.994) and the Body measurement model using the entire sample.
Mass Index (Rho = −0.048, p = 0.084) were unrelated
to the mini-SPIN. It was significantly related to a bad Population based norms
subjective health status (Rho = 0.151, p < 0.001), depres- Table 4 provides a detailed illustration of the population
sion (Rho = 0.374, p < 0.001) and generalized anxiety based norms of the mini-SPIN. We display norms separ-
(Rho = 0.378, p < 0.001). ately for gender (female, male) and age groups (<30, 31–40,
Female (compared to male) participants and par- 41–50, 51–60, 61–70, >70 years). Percentiles (cumulative
ticipants with panic attacks reported higher scores percentages) are displayed for the sum scores of the scale.
in the mini-SPIN (z = 3.736, p < 0.001 resp. z = 8.470,
p < 0.001). Discussion
Our aim was to evaluate the German version of the
Factorial invariance three item short form of the Social Phobia Inventory
A baseline model (Model 0), which simultaneously esti- (min-SPIN) in a clinical and in a representative commu-
mated all model parameters constraining all factor loadings nity sample.
to be invariant across aforementioned groups resulted in Taking into account the shortness of the scale we
excellent model fit (CFIscaled = 1.0; RMSEAscaled = 0.000 found good internal consistencies (Cronbach’s alpha 0.80
[CI: 0.000, 0.000]). Strong invariance was examined by to 0.83). Our results are therefore in the range of com-
comparing Model 0 with Model 1 (see Table 3), which con- parable studies of the mini-SPIN in other languages (e.g.
strained all item intercepts to be invariant across groups. [16, 19, 46]. Thus, there is no need for a revision of the
ΔCFI were below the cut-off recommended by Chen. Fur- mini-SPIN including different items to enhance internal
thermore, the model fit was excellent (CFIscaled = 0.996; consistency, as Aderka et al. (2013) have suggested [47].
RMSEAscaled = 0.020 [CI: 0.000, 0.058]). Therefore, weak Test-retest reliability (Rho = 0.61) was somewhat lower
invariance can be assumed. Strict Invariance was examined than in comparable studies (e.g. [16]). However, we
by comparing Model 1 with Model 2a, which constrained assessed the mini-SPIN before and after treatment in
all item residual variances to be invariant across groups, our clinical sample. As patients were treated in our in-
resulting in a considerable worsening of model fit (ΔCFI = patient or day hospital setting treatments did not differ
−0.028; supplemented by ΔRMSEA= 0.021). Subsequently, significantly. The therapeutic orientation of the multi-
one item residual variance in one group were freed, the modal treatment setting was psychodynamic including
resulting Model 2b exhibited an exactable difference in fit cognitive-behavioral and psychoeducational elements,
compared with Model 1 (ΔCFI =−0.007; ΔRMSEA = art therapy, body-oriented therapy, relaxation therapy
+0.004). Furthermore, the model fit was excellent (CFI = and physical therapy. However, patients were diagnosed
0.989; RMSEA= 0.024 [0.000, 0.048]). Thus, strict invari- with a broad spectrum of mental disorders leading to a
ance can be assumed for the mini-SPIN regarding age and heterogeneous psychotherapeutic outcome especially re-
gender. Figure 2 illustrates the factor loadings, factor garding social anxiety, which can be expected to lower
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 7 of 10
Table 3 Measurement invariance of the mini-SPIN
χ2scaled df CFI ΔCFI RMSEA ΔRMSEA Measurement Invariance Test
a
Model 0 weak invariance 5.63 14 1.000 – 0.0 – √
Model 1 strong invariance 25.65 27 .996 −.004 .020 +.020 √
Model 2a strict invariance 61.00 48 .968 −.028 .041 +0.21 x
Model 2b Strict invariance (partial) 51.40 47 .989 −.007 .024 +.004 √
df = degrees of freedom; CFI = Comparative Fit Index; ΔCFI = differences between models (0 and 1, 1 and 2a; 1 and 2b) regarding CFI; RMSEA = root mean square
of approximation; ΔRMSEA = differences between models (0 and 1, 1 and 2a; 1 and 2b) regarding RMSEAa = ΔCFI ≤ −.010 supplemented by ΔRMSEA ≥ .015 indicates
non-invariance. √ marks invariance
the association between test and retest by reducing so- In our community study relations of the mini-SPIN to
cial anxiety in a part of the sample. Regarding the sensi- anxiety (GAD-2) and depression PHQ-2) were of
tivity to change, we found comparable pre- to post- medium height and comparable to the correlations
intervention effect sizes for the mini-SPIN total score found in the clinical sample. The positive relations be-
and the LSAS total score. The effect sizes were only tween mini-SPIN and a bad subjective health status and
small [48]. This might be due to the fact that reducing female sex were plausible and compatible with findings
social anxiety was not the primary goal of most of the of previous studies [5]. In contrast to other epidemio-
inpatient or day hospital treatments. logical studies in our sample the mini-SPIN was unre-
In our clinical sample, we found evidence of a good lated to age (e.g. [49]) in correlational analyses. This
construct validity of the mini-SPIN. It was strongly re- finding was caused by a small variance in this sample
lated with measures of the same construct (LSAS) as an due to a majority (80%) of participants only reaching 2
aspect of convergent validity. Correlations of the mini- points or less in the mini-SPIN. However, comparing
SPIN with scales assessing different symptoms (PHQ-9, participants above the defined criterion (mini-SPIN > = 6
GAD-7, PHQ-15) were also significant but somewhat points) with those below, patients suffering from social
lower. The mentally disabling character of social anxiety anxiety were significantly younger. Moreover, there is
[5] is reflected by the positive correlation of the mini- evidence from the literature, that social anxiety is related
SPIN with the mental health component (MHC). How- to alcohol abuse (e.g. [5]). In our study the frequency
ever, as expected, the mini-SPIN was unrelated to the of alcohol consumption (average of 1.82 days per
physical health component (PHC). The lacking relation week, SD 2.04) was unrelated to the mini-SPIN. Prob-
to PHC and the lower correlations to symptoms reflect- ably the lack of relation between mini-SPIN and fre-
ing different disorders and MHC can be interpreted as quency of alcohol consumption is due to a) disparate
an aspect of discriminant validity. reasons for the consumption (e.g. to cope inad-
equately with social anxiety or a sociable life style
with frequent but moderate alcohol consumption and
not abuse) or compared to other studies due to b)
differing assessment of alcohol consumption (not tak-
ing into account quantity, frequency of and functional
impairment following from alcohol consumption).
Furthermore, evidence of strict measurement invari-
ance by sex and age and the associated possibility of un-
biased comparison of means, correlation coefficients,
path coefficients within SEM (Structural equation mod-
eling) as well as the possibility of undistorted screening
decisions between aforementioned groups, appear to be
explicitly relevant.
To examine diagnostic accuracy of the mini-SPIN we
assessed sensitivity and specificity for the three criteria
(study 1). We confirmed a cut-off of 6 for the German
version of the mini-SPIN (e.g. [13, 16, 17] to be best bal-
anced regarding sensitivity and specificity for the clinical
diagnosis of SAD and generalized social anxiety (LSAS >
60) as criteria (c.f. Table 2). For social anxiety deter-
Fig. 2 Factor loadings, intercepts, and residual variances for a mined by a lower cut-off (LSAS > 30) sensitivity and spe-
unidimensional measurement model using the entire sample
cificity were best balanced at a mini-SPIN cut-off of 4.
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 8 of 10
Table 4 Population based norms of the mini-SPIN (representative community sample, N = 1274)
Cumulative %
Score female male
<=30y 31-40y 41-50y 51-60y 61-70y >70y <=30y 31-40y 41-50y 51-60y 61-70y >70y
N = 122 N = 125 N = 120 N = 115 N = 127 N = 82 N = 119 N = 87 N = 92 N = 97 N = 115 N = 73
0 59,0 52,8 59,2 48,7 50,4 50,0 63,0 59,8 60,9 67,0 61,7 60,3
1 66,4 60,8 64,2 63,5 64,6 64,6 73,9 73,6 77,2 76,3 73,9 75,3
2 77,0 70,4 77,5 78,3 78,7 76,8 84,0 86,2 84,8 86,6 81,7 80,8
3 88,5 83,2 87,5 85,2 91,3 84,1 94,1 92,0 91,4 91,8 88,7 89,0
4 92,6 88,8 89,2 91,3 96,1 93,9 96,6 96,6 92,4 94,8 93,0 94,5
5 94,3 89,6 95,8 96,5 99,2 100 98,3 100 95,7 97,9 94,8 97,3
6 96,7 92,8 96,7 99,1 100 100 100 100 96,7 100 96,5 98,6
7 97,5 97,4 96,7 99,1 100 100 100 100 97,8 100 98,3 100
8 98,4 99,2 99,2 100 100 100 100 100 100 100 99,1 100
9 99,2 99,2 100 100 100 100 100 100 100 100 100 100
10 99,2 99,2 100 100 100 100 100 100 100 100 100 100
11 100 100 100 100 100 100 100 100 100 100 100 100
12 100 100 100 100 100 100 100 100 100 100 100 100
Noteworthy, is the fact, that a sufficient sensitivity is ac- and assessing changes of social anxiety in clinical and epi-
companied by relatively low specificity in all criteria, es- demiological studies.
pecially for the clinical diagnosis. The reasons for the
low specificity might be the relatively seldom (8.3%) clin- Abbreviations
ical diagnosis of SAD (e.g. [8]) and the lack of a non- BDSG: Bundesdatenschutzgesetz; BGB: Bundesgesetzbuch; CFI: Comparative
clinical comparison group. fit index; DSM-5: Diagnostic and statistical manual of mental disorders, 5thedition; ES: Effect size; GAD-2: Generalized Anxiety Disorder Scale, short
Based on data of a large community sample (N = 1274) version; GAD-7: Generalized Anxiety Disorder Scale; ICD-10: International
we were able to determine population based norms for Statistical Classification of diseases, 10th edition; Kurt: kurtosis;
different age groups and sex. LSAS: Liebowitz Social Anxiety Scale; M: Mean; MGCFA: Multi-groupconfirmatory factor analysis; MHC: Mental health component of the SF-12;
The strengths of our studies are the large sample sizes Mini-SPIN: mini Social Phobia Inventory; PHC: physical health component of
allowing determination of cut-offs and population based the SF-12; PHQ-15: Patient Health Questionnaire, somatization module; PHQ-
norms. Our results are a) somewhat limited by the lack 2: Patient Health Questionnaire, depression module, short version; PHQ-9: Patient Health Questionnaire, depression module; RMSEA: Root Mean
of a non-clinical comparison group to assess diagnostic Square Error of Approximation; ROC: receiver operating characteristic;
accuracy, and b) by the lack of a standardized clinical SAD: Social Anxiety Disorder; SD: Standard deviation; SEM: Structural
interview to ensure the diagnosis of SAD. equation modeling; SF-12: Short-Form-12 Health Survey; SIAS: SocialInteraction Anxiety Scale; Skew: skewness; SPIN: Social Phobia Inventory;
Further studies on the mini-SPIN should address diag- SPS: Social Phobia Scale; USUMA: Unabhängige Serviceeinrichtung für
nostic accuracy and especially include a non-clinical Umfragen, Methoden und Analysen; independent service for surveys,
comparison group and standardized diagnostic of SAD. methods and analyses in market and social research
Despite its limitations, the results of our two validation
studies encourage the use of the German mini-SPIN in Acknowledgements
We thank all study participants for their willingness to provide their data for
different settings. Its brevity, its easy interpretation and this research project and we are indebted to all coworkers for their
its reasonable psychometric properties make it suitable enthusiastic commitment.
as a screening instrument in clinical (e.g. primary care)
and also in study contexts (e.g. psychotherapy trials, epi- Funding
demiological studies). It can also be used as an easy to This research did not receive any specific grant from funding agencies in the
public, commercial, or not-for-profit sectors.
apply follow-up measure in clinical studies or during in-
patient and outpatient psychotherapy.
Availability of data and materials
The authors confirm that, for approved reasons, some access restrictions
apply to the data underlying the findings. Due to ethical restrictions, the
Conclusions data cannot be made publicly available, but it is available upon request. The
data set contains identifying participant information, which is not suitable for
The German version of the mini-SPIN is a reliable and public deposition. The request should be directed to the corresponding
valid instrument. Its brevity makes it valuable for screening author.
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 9 of 10
Authors’ contributions und Konsildienst unterdiagnostizierte Angsterkrankung? Psychother
All authors had full access to all data in the study and take responsibility for Psychosom Med Psychol. 2010;60:111–7. doi:10.1055/s-0029-1220692.
the integrity of the data, the accuracy of the data analysis, and the decision 9. Mennin DS, Fresco DM, Heimberg RG, Schneier FR, Davies SO, Liebowitz MR.
to submit for publication. All authors have approved the final manuscript. Screening for social anxiety disorder in the clinical setting: using the
Study concept and design: MEB, EB, MM, IR, CSW, JW, RZ; Drafting of the Liebowitz social anxiety scale. J Anxiety Disord. 2002;16:661–73.
manuscript: SK, JW, RZ; Critical revision of the manuscript for important 10. Stangier U, Heidenreich T. Die Liebowitz Soziale Angst- Skala (LSAS). In
intellectual content: All authors; Statistical analysis: SK, JW. Collegium Internationale Psychiatriae Scalarum (Hrsg.), Internationale Skalen
für Psychiatrie. Weinheim, Beltz; 2004.
Ethics approval and consent to participate 11. Heimberg RG, Horner KJ, Juster HR, Safren SA, Brown EJ, Schneier FR,
Study 1: There was no need for written consent because the study analyzed Liebowitz MR. Psychometric properties of the Liebowitz social anxiety scale.
clinical data obtained by clinical standard assessment (i.e., not within the Psychol Med. 1999;29:199–212.
context of an epidemiological or clinical study). This procedure complies 12. Mattick RP, Clarke JC. Development and validation of measures of social
with the hospital laws of Rhineland-Palatinate [50]. phobia scrutiny fear and social interaction anxiety. Behav Res Ther. 1998;36:
Study 2: A nationwide survey representative of the German general population 455–70. doi:10.1016/S0005-7967(97)10031-6.
was conducted by an institute for demographic research (USUMA GmbH = 13. Connor KM, Kobak KA, Churchill LE, Katzelnick D, Davidson JR. Mini-SPIN: A
Unabhängige Serviceeinrichtung für Umfragen, Methoden und Analysen; brief screening assessment for generalized social anxiety disorder. Depress
independent service for surveys, methods and analyses in market and social Anxiety 2001; 14:137–140.
research) according to the German law of data protection (§ 30a BDSG) and 14. Connor KM, Davidson JRT, Churchill LE, Sherwood A, Foa E, Weisler RH.
with written consent. In case of minor participants their parents provided Psychometric properties of the social phobia inventory (SPIN): new self-
consent on their stead. Previously ethics were weighted to the respective rating scale. Brit J Psychiatr. 2000;176:379–86.
interests of the public and of the individuals concerned following 1823 (BGB) of 15. Stangier U, Steffens M. Social Phobia Inventory (SPIN) – Deutsche Fassung.
the Civil Code of Law and in accordance with the guidelines in the Declaration Frankfurt am Main: Psychologisches Institut der Universität Frankfurt am
of Helsinki. Main; 2002.
16. Seeley-Wait E, Abbott MJ, Rapee RM. Psychometric properties of the mini-
Consent for publication social phobia inventory. Prim Care Companion J Clin Psychiatr. 2009;11:231–
Not applicable. 6. doi:10.4088/PCC.07m00576.
17. Weeks JW, Spokas ME, Heimberg RG. Psychometric evaluation of the mini-
Competing interests social phobia inventory (mini-SPIN) in a treatment-seeking sample. Depress
The authors declare that they have no competing interests. Anxiety. 2007;24:382–91. doi:10.1002/da.20250.
18. Ranta K, Kaltiala-Heino R, Rantanen P, Marttunen M. The mini-social phobia
inventory: psychometric properties in an adolescent general population
Publisher’s Note sample. Compr Psychiatry. 2012;53:630–7. doi:10.1016/j.comppsych.2011.07.007.
Springer Nature remains neutral with regard to jurisdictional claims in 19. Garcia-Lopez L, Moore HT. Validation and diagnostic efficiency of the mini-
published maps and institutional affiliations. SPIN in Spanish-speaking adolescents. PLoS One. 2015;10(8):e0135862.
doi:10.1371/journal.pone.0135862.
Author details 20. Osório FL, Crippa JA, Loureiro SR. A study of the discriminative validity of a
1Department of Psychosomatic Medicine and Psychotherapy, University screening tool (MINI-SPIN) for social anxiety disorder applied to Brazilian
Medical Center of the Johannes Gutenberg University Mainz, Mainz, university students. Eur Psychiatr. 2007;22:239–43. doi:10.1016/j.eurpsy.2007.
Germany. 2Criminological Research Institute of Lower Saxony, Hannover, 01.003.
Germany. 3Institute of Psychology, Technical University of Braunschweig, 21. Osório FL, Crippa JA, Loureiro SR. Further study of the psychometric
Braunschweig, Germany. qualities of a brief screening tool for social phobia (MINI-SPIN) applied to
clinical and nonclinical samples. Perspect Psychiatr Care. 2010;46:266–78.
Received: 9 May 2017 Accepted: 16 November 2017 doi:10.1111/j.1744-6163.2010.00261.x.
22. Sosic Z, Gieler U, Stangier U. Screening for social phobia in medical in- and
outpatients with the German version of the social phobia inventory (SPIN). J
References Anxiety Disord. 2008;22:849–59. doi:10.1016/j.janxdis.2007.08.011.
1. American Psychiatric Association. Diagnostic and statistical manual of 23. Dilling H, Mombour W, Schmidt MH. Internationale Klassifikation psychischer
mental disorders (5th ed.). Arlington, VA, American Psychiatric Publishing; Störungen ICD 10 Kapitel V (F). Bern, Huber: Klinisch-diagnostische Leitlinien;
2013. 1991.
2. Wittchen HU, Jacobi F. Die Versorgungssituation psychischer Störungen in 24. Löwe B, Gräfe K, Zipfel S, Witte S, Loerch B, Herzog W. Diagnosing ICD-10
Deutschland. Eine klinisch-epidemiologische Abschätzung anhand des depressive episodes: superior criterion validity of the patient health
Bundes-Gesundheitssurveys 1998. Bundesgesundheitsbl Gesundheitsforsch questionnaire. Psychother Psychosom. 2004;73:386–90. doi:10.1159/000080393.
Gesundheitsschutz. 2001;44:993–1000. 25. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression
3. Fehm L, Beesdo K, Jacobi F, Fiedler A. Social anxiety disorder above and severity measure. J Gen Intern Med. 2001;16:606–13.
below the diagnostic threshold: prevalence, comorbidity and impairment in 26. Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version
the general population. Soc Psychiatry Psychiatr Epidemiol. 2008;43:257–65. of PRIME-MD: the PHQ primary care study. Primary care evaluation of mental
doi:10.1007/s00127-007-0299-4. disorders. Patient health questionnaire. JAMA. 1999;282(18):1737–44.
4. Kessler RC, Petukhova M, Sampson NA, Zaslavsky AM, Wittchen HU. Twelve- 27. Gräfe K, Zipfel S, Herzog W, Löwe B. Screening psychischer Störungen mit
month and lifetime prevalence and lifetime morbid risk of anxiety and dem "Gesundheitsfragebogen für Patienten (PHQ-D)". Ergebnisse der
mood disorders in the United States. Int J Methods Psychiatr Res. 2012;21: deutschen Validierungsstudie. Diagnostica. 2004;50:171–81.
169–84. doi:10.1002/mpr.1359. 28. Cronbach LJ. Coefficient alpha and the internal structure of tests.
5. Wittchen HU, Fehm L. Epidemiology and natural course of social fears and Psychometrika. 1951;16(3):297–334.
social phobia. Acta Psychiatr Scand. 2003;108(Suppl):4–18. 29. Gilbody S, Richards D, Brealey S, Hewitt C. Screening for depression in medical
6. Beesdo-Baum K, Knappe S, Fehm L, Höfler M, Lieb R, Hofmann SG, Wittchen settings with the patient health questionnaire (PHQ): a diagnostic meta-
HU. The natural course of social anxiety disorder among adolescents and analysis. J Gen Intern Med. 2007;22:1596–602. doi:10.1007/s11606-007-0333-y.
young adults. Acta Psychiatr Scand. 2012;126(6):411–25. doi:10.1111/j.1600- 30. Spitzer RL, Kroenke K, Williams JB, Löwe B. A brief measure for assessing
0447.2012.01886.x. generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166:1092–7.
7. Keller MB. The lifelong course of social anxiety disorder: a clinical doi:10.1001/archinte.166.10.1092.
perspective. Acta Psychiatr Scand. 2003;108(Suppl):85–94. 31. Kroenke K, Spitzer RL, Williams JB, Monahan PO, Löwe B. Anxiety disorders
8. Wiltink J, Haselbacher A, Knebel A, Tschan R, Zwerenz R, Michal M, Subic- in primary care: prevalence, impairment, comorbidity, and detection. Ann
Wrana C, Beutel ME. Soziale Phobie – eine im psychosomatischen Ambulanz- Intern Med. 2007;146:317–25.
Wiltink et al. BMC Psychiatry  (2017) 17:377 Page 10 of 10
32. Löwe B, Decker O, Müller S, Brähler E, Schellberg D, Herzog W, Herzberg PY.
Validation and standardization of the generalized anxiety disorder screener
(GAD-7) in the general population. Med Care. 2008;46:266–74. doi:10.1097/
MLR.0b013e318160d093.
33. Kroenke K, Spitzer RL, Williams JB. The PHQ-15: validity of a new measure
for evaluating the severity of somatic symptoms. Psychosom Med. 2002;64:
258–66.
34. van Ravesteijn H, Wittkampf K, Lucassen P, van de Lisdonk E, van den
Hoogen H, van Weert H, Huijser J, Schene A, van Weel C, Speckens A.
Detecting somatoform disorders in primary care with the PHQ-15. Ann Fam
Med. 2009;7:232–8. doi:10.1370/afm.985.
35. Bullinger M, Kirchberger I. Der SF-36 Fragebogen zum Gesundheitszustand.
Göttingen, Hogrefe Verlag: Handbuch für die deutschsprachige
Fragebogenversion; 1998.
36. Wiltink J, Tschan R, Michal M, Subic-Wrana C, Eckhardt-Henn A, Dieterich M,
Beutel ME. Dizziness: anxiety, health care utilization and health behaviour.
Results from a representative German community survey J Psychosom Res.
2009;66:417–24. doi:10.1016/j.jpsychores.2008.09.012.
37. Löwe B, Spitzer RL, Zipfel S, Herzog W. PHQ-D Gesundheitsfragebogen für
Patienten (German Version of the Patient Health Questionnaire). Pfizer:
Karlsruhe; 2002.
38. Löwe B, Grafe K, Zipfel S, Spitzer RL, Herrmann-Lingen C, Witte S, Herzog W.
Detecting panic disorder in medical and psychosomatic outpatients:
comparative validation of the hospital anxiety and depression scale, the
patient health questionnaire, a screening question, and physicians
diagnosis. J Psychosom Res. 2003;55:515–9.
39. Löwe B, Kroenke K, Grafe K. Detecting and monitoring depression with
a two-item questionnaire (PHQ-2). J Psychosom Res. 2005;58:163–71.
doi:10.1016/j.jpsychores.2004.09.006.
40. Byrne BM, Shavelson RJ, Muthén B. Testing for the equivalence of factor
covariance and mean structures: the issue of partial measurement
invariance. Psychol Bull. 1989;105(3):456–66.
41. Satorra A, Bentler PMA. Scaled difference chi-square test statistic for
moment structure analysis. Psychometrika. 2001;66(4):507–14.
42. Chen FF. Sensitivity of goodness of fit indexes to lack of measurement
invariance. Struct Equ Model. 2007;14(3):464–504.
43. LT H, Bentler PM. Cutoff criteria for fit indexes in covariance structure
analysis: conventional criteria versus new alternatives. Struct Equ Model.
1999;6(1):1–55.
44. Core Team R. R: a language and environment for statistical computing.
Vienna, Austria: R Foundation for Statistical Computing; 2016. URL https://
www.R-project.org/
45. Rosseel Y. Lavaan: an R package for structural equation modeling. J Stat
Softw. 2012;48(2):1–36.
46. Fogliati VJ, Terides MD, Gandy M, Staples LG, Johnston L, Karin E, Rapee RM,
Titov N, Dear BF. Psychometric properties of the mini-social phobia
inventory (mini-SPIN) in a large online treatment-seeking sample. Cogn
Behav Ther. 2016;45(3):236–57. doi:10.1080/16506073.2016.1158206.
47. Aderka IM, Pollack MH, Simon NM, Smits JA, Van Ameringen M, Stein MB,
Hofmann SG. Development of a brief version of the social phobia inventory
using item response theory: the mini-SPIN-R. Behav Ther. 2013;44:651–61.
doi:10.1016/j.beth.2013.04.011.
48. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed.
Hillsdale: Lawrence Erlbaum Associates; 1988.
49. Jacobi F, Höfler M, Strehle J, Mack S, Gerschler A, Scholl L, Busch MA, Maske
U, Hapke U, Gaebel W, Maier W, Wagner M, Zielasek J, Wittchen HU. Mental
disorders in the general population: study on the health of adults in
Germany and the additional module mental health (DEGS1-MH). Nervenarzt. Submit your next manuscript to BioMed Central 
2014;85(1):77–87. doi:10.1007/s00115-013-3961-y.
50. Landeskrankenhausgesetz (LKG) vom 28. November 1986. (GVBl. S. 342), and we will help you at every step:
zuletzt geändert durch Gesetz vom 19. Dezember 2014 (GVBl. S. 302), BS •  We accept pre-submission inquiries 
2126–3. http://landesrecht.rlp.de/jportal/?quelle=jlink&query=KHG
+RP&psml=bsrlpprod.psml Accessed 14 Nov 2017. •  Our selector tool helps you to find the most relevant journal
•  We provide round the clock customer support 
•  Convenient online submission
•  Thorough peer review
•  Inclusion in PubMed and all major indexing services 
•  Maximum visibility for your research
Submit your manuscript at
www.biomedcentral.com/submit