Measuring the “Blue” Face of Narcissism: Psychometric Properties of Ukrainian Translation of the Hypersensitive Narcissism Scale in Non-Clinical Groups

Aim. Little research has focused on hypersensitive narcissism in Slavic countries. One of the reasons for that is the absence of suitable inventories. The present study aims to make the Ukrainian adapted translation of the Hypersensitive Narcissism Scale (HSNS) and to check for its psychometric properties in a non-clinical sample. Methods. To prepare materials, we first conducted a double-blind translation procedure with further linguistic analysis. The following two empirical studies to collect data for statistical analysis were then made. Results. The data shows adequate internal consistency and scale validity, as well as a three-factor structure (obtained with the principal components analysis [PCA]), in line with the theoretical background. However, the confirmatory factor analysis (CFA) and Cronbach’s alphas analysis yield the unidimensional scale to be the best fit for the Ukrainian version. The data also support evidence for maladap tive social relationship patterns of people with high HSNS scores. Conclusion. We proved vulnerable and grandiose narcissism are qualitatively different phenomena and showed how a cultural context of narcissism manifestation may be reflected through social interactions and self-attitude. HSNS in Ukrainian is a reliable and valid tool for complex psychological personality research among non-clinical adult samples. Practical application . The proposed translation of the HSNS is adapted to be used for research in Ukrainian culture, with Ukrainian-speaking respondents and psychologists’ and psychotherapists’ clients. personality meta-traits, Internet social network activity, and interpersonal relationship measures.


Introduction
T he term narcissism is conceptualised in both clinical (as a personality disorder) and social-personality (as a personality trait) aspects that tend to be different and even opposite in many senses (Miller & Campbell, 2008). The trait level of narcissism manifests in two forms: grandiose and vulnerable (Jauk et al.; Miller et al., 2011Miller et al., , 2017Zajenkowski & Szymaniak, 2019).
Psychologists study grandiose narcissism more widely as first thought to be a "mental 'prototype' of NPD (narcissistic personality disorder)" (Miller & Campbell, 2008, p. 470). However, it soon proved to have a much more positive "face" as it is linked to positive personality traits like extraversion, assertiveness, achievement striving, self-discipline (Miller & Campbell, 2008), positive affect, well-being (Sedikides et al., 2004), and explicit selfesteem (Hyatt et al., 2018).
Grandiose and vulnerable narcissism, although sharing a "common dark core" (Jauk & Kaufman, 2018, p. 131), do not show a strong linear correlation (Jauk & Kaufman, 2018;Miller et al., 2017;Sengul et al., 2015). Thus, it is not enough to measure the level of overt narcissism as covert narcissism is qualitatively different.
The HSNS (Hendin & Cheek, 1997) is a 10-item unidimensional questionnaire built on Henry Murray's concept of narcissism (Murray, 1938) for measuring a person's feeling of being neglected, anxious, and sensitive to others' opinion, that is a characteristic of vulnerable narcissism (Rogoza et al., 2018). It is reliable (.75 Cronbach alpha coefficient), has a single-factor structure, and a zero correlation with Narcissistic Personality Inventory (NPI) score (Hendin & Cheek, 1997).
The HSNS was created in English and later translated and validated in several languages (namely, there exist Italian [Fossati et al., 2009], Spanish [Ripoll et al., 2010], Turkish [Sengul et al., 2015], Russian [Nesterova, 2017]; in some research, a Polish version of the scale is also used [Czarna et al., 2014], however, without a full validation). This research aimed to create a reliable, validated Ukrainian version of the HSNS, and the following article is dedicated to the description of its psychometric properties.

Materials and Methods
The linguistic and cultural adaptation of HSNS in Ukraine had several stages.
Firstly, we made the double-blind translation of the scale from English to Ukrainian. The Ukrainian native speakers with an advanced English level provided two direct translations of the original HSNS into Ukrainian. Analysing these versions, we have created one test translation.
Secondly, two English native speakers, social and behavioural sciences professionals with an advanced Ukrainian level, made independent reverse translations of the test back into English. We have compared those two results with the original HSNS and chose the most suitable wordings of each statement in the Ukrainian version.
To ensure the correct linguistic form for each statement, we discussed the Ukrainian version with the Ukrainian linguist. This process led to the final test version, which we used in further studies (the Ukrainian version is available upon request). Eventually, we used this Ukrainian translation to obtain empirical data on the test's reliability and validity through Study 1 and Study 2.
Finally, we tested the psychometric properties of the Ukrainian version of HSNS. In Study 1, the factorial structure of the vulnerable narcissism as a construct was measured by HSNS. We then checked the general (internal), test-retest reliability, and discriminant validity. We conducted Study 2 for a criterion validity review, namely, for correlations of the HSNS with self-concept, personality meta-traits, Internet social network activity, and interpersonal relationship measures.

Participants and Procedure
To test the factorial structure, reliability, and discriminant validity of the HSNS, we collected a non-probabilistic sample of 969 Ukrainian native speakers (708 females (73.07%), 261 males). The majority of the participants (707 people) were contacted via social networks where they voluntarily acceded to the online questionnaire. The rest 262 respondents were undergraduate students from public and private universities, recruited via the paper-and-pencil form. Participants' age ranged from 17 to 60 years (M = 24.76, SD = 9.47); 404 people (41.69%) had higher education; 670 people (69.14%) were married. In four weeks, we recruited 117 respondents out of those 969 for the second trial to fill in the forms again.

Measures
The Hypersensitive Narcissism Scale (Hendin & Cheek, 1997) is a 10-item self-report measure using the Likert scale from 1 (very uncharacteristic or untrue, strongly disagree) to 5 (very characteristic or true, strongly agree) concerning respondents' personality features and attitudes describing covert narcissism. We obtained the total hypersensitive narcissism score by calculating a simple sum of all item ratings.
Narcissistic Personality Inventory (Ames et al., 2006) is a 16-double--item self-report scale, retrieved from a 40-statement original NPI by Robert Raskin and Howard Terry (1988), which total score shows the level of grandiose narcissism. The 16-item scale's reliability is satisfactory (α = .72) (Ames et al., 2006). For the present study, we used the Ukrainian adaptation of NPI-16, which consists of 14 items (α = .69) (Pylat et al., manuscript in preparation).
Narcissistic Admiration and Rivalry Questionnaire -Short (NARQ-S) is a short 6-item 6-point Likert-scale measure with two subscales showing narcissists' social strategies: admiration, which is a "tendency to approach social admiration utilising self-promotion (assertive self-enhancement)", and rivalry, a "tendency to prevent social failure through self-defence (antagonistic self-protection)" (Back et al., 2013(Back et al., , p. 1015. A simple sum of each statement grade indicates a total score for admiration and rivalry. In the current study, we used a NARQ-S Ukrainian translation by Nataliya Pylat and Inna Haletska (2018). The reliability test showed satisfactory results (α = .71 for admiration, α = .56 for rivalry subscales).
Morris Rosenberg's (1965) self-esteem scale (RSE) is a 10-statement 4-point Likert-scale and one of the most widely used explicit self-report measures of self-esteem (Bosson et al., 2008). We used the Ukrainian translation of the questionnaire; in this research, Cronbach's alpha coefficient for the scale is high (α = .83).

Participants and Procedure
We involved three more samples in a research to provide a criterion validity assessment with the measures described below. Sample

Measures
We used HSNS in the tested Ukrainian translation and a short version of the NPI in Ukrainian for vulnerable and grandiose narcissism assessment.
For Sample 1, we used the Twenty Statements Test (TST) by Manfred Kuhn and Thomas McPartland (1954) to measure self-concept. Each respondent filled in 20 statements starting with "I am…" (in Ukrainian). We analysed findings dividing statements into two types, following Nataliya Ivanova's approach (2009). The first type of statements describes a person's self-attitude to identify "reflexive" self-characteristics (e.g., "I am smart", "I am a creative person", etc.). The statements of the second type denote a person's social role-either interpersonal ("I am a good friend"), family ("I am a mother") or labour-professional ("I am a student", "I am a teacher") for "social" self-characteristics (Kletsyna, 2004).
For Sample 2, we used the Big Five Locator (Howard et al., 1996) in the Ukrainian validated translation (Burlachuk & Korolev, 2000). The questionnaire is based on the theory of 5 personality meta-traits of Robert McCrae and Paul Costa (1990). It has five subscales to measure neuroticism, extraversion, agreeableness, openness to experience, and conscientiousness; five pairs of the opposite characteristics in each need to be self-evaluated on a 5-point Likert scale.
Additionally, respondents have filled in a 12-question inventory about their Internet social network activity. There were questions concerning a person's activity in general observation of the content posted on the Internet social network (measured with the general number of personal profiles in popular social media, the frequency of checking these profiles, the time usually spent on social media), social interaction activity (measured by the frequency of person making posts in social networks or commenting on the publications of other people), and self-presentation activity (the amount and the frequency of personal content posted, like reporting on life events, publicly showing photos and videos with the self, writing posts with a personal attitude towards something, etc.), four questions to each aspect of the activity. Further coding in per cent was made for evaluating the measure of activity in a social network a person has in all three parts and in general.
Relationship Closeness Inventory (RCI) (Berscheid et al., 1989) is a self--reported questionnaire where a respondent is asked to think about the closest person for him/her. It includes three subscales: frequency (the amount of time spent with the nearest person), diversity (the number of daily activities (in the list of 38) that are made together), and strength (the influence a close person has on actions, decisions, and plans), that are characteristics of person's relationships.

Results
We have started with the structural analysis of the theoretical construct of vulnerable narcissism using principal component analysis.

Table 1
Factor loadings and total percent of variance (after varimax normalised rotation) for each factor of the HSNS (Ukrainian)

Item
Factor 1 "Hypersensitivity to judgments" Factor 2 "Social detachment" Factor 3 "Self-absorption" 2. My feelings are easily hurt by ridicule or the slighting remarks of others.
. Thus, in this research, a three-factor structure of the hypersensitive narcissism construct looked the most suitable. In confirmatory factor analysis, both models were found to be significant and well-fit (see Table 5). To choose the best questionnaire structure of the Ukrainian version of HSNS, we conducted a further internal consistency analysis.
Means, standard deviations, skewness and kurtosis, inter-item correlations, and Cronbach's alpha coefficients for the HSNS items on the material of Study 1 are displayed in Table 6. The total average inter-item correlation is r = .14.
The HSNS composite score's internal consistency in the non-clinical sample was α = .61, a minimum satisfactory but slightly lower from the original version and other language versions (see Table 2 for comparison). We have checked the internal consistency, additionally deleting items with weaker inter-item correlations (No. 4,6,8,10). However, we did not obtain a significantly higher Cronbach's alpha score for HSNS total score (the highest is for 6-items α = .63).
Later, we checked for the internal consistency of each of the three factors. However, the unsatisfactory Cronbach's alpha coefficients for two of them were obtained (α = .49 for "social detachment" and α = .36 for "self--absorption"), although one factor which probably is the crucial aspect of the measured concept showed better scores (α = .71 for "hypersensitivity to judgments"). Finally, we chose the 10-item version as the most suitable in Ukrainian translation. The HSNS is a unidimensional scale (as it was constructed from the very beginning), and three dimensions of hypersensitive narcissism cannot be the questionnaire subscales, just the sides of one phenomenon reflecting its complex structure.
The correlation between the Ukrainian versions of HSNS and NPI was strongly negative (r = -.15, p < .001) (see Table 3). For a more profound understanding of the links between grandiose and vulnerable narcissism, we have studied the correlations of each HSNS item with the NPI total score (see Table 7). Notably, only statements included in the "hypersensitivity to judgments" dimension (2,3,7,9) were all negatively correlated to NPI score. Albeit "self-absorption" (items 4, 5, and 10) and "social detachment" (items 1, 6, and 8) dimensions of hypersensitive narcissism did not have such clear connections. Remarkably, two of three scores of the "self--absorption" dimension did not have significant correlations with NPI (p > .05), as well as one from the "egoism" dimension. Moreover, the correlation coefficients for items 5 and 10 were positive yet non-significant.
Source: own research.
We then calculated Cronbach's alphas and inter-item correlations for each sample. The results were somewhat higher for two samples (α = .65 and r = .15 for Sample 2, and α = .63 and r = .14 for Sample 3). Although the results may seem disputable (α = .54 and r = .10 for Sample 1), due to the small number of participants (N = 51), it cannot be taken as enough representative.
A tendency of negative correlation of HSNS and NPI scores appeared in all samples. However, only for the Sample 2 these links were significant (r = -.25, p < .01), albeit no significant connections were obtained in Sample 1 (r = -.15, p > .05) and Sample 3 (r = -.03, p > .05).
On the data of Sample 1, we found out that HSNS total score was negatively correlated to the amount of self-related characteristics in person's explicit self-image (r = -.29, p < .05). Among the social roles, there was a negative correlation with community roles (r = -.30, p < .05) and positive with professional roles (r = .33, p < .05). It is even more illustrative result in comparison to NPI total score correlations: the higher grandiose narcissism a person had, the more self-related (r = .49, p < .001) and profession-related (r = .31, p < .05) roles and the less family-related roles (r = -.51, p < .001) were there in his/her self-image (see Table 8).
The results of the Sample 2 study proved that there were connections between personality meta-traits and HSNS measures. Particularly, the higher vulnerable narcissism was, the higher neuroticism (r = .37, p < .001), the lower extraversion (r = -.34, p < .001), openness to experience (r = -.24, p < .01) and conscientiousness (r = -.24, p < .01) were observed. The NPI total score had directly opposite correlations with the listed scales (see Table 4). Moreover, a person's Internet social network activity is strongly related to narcissistic levels, both grandiose and vulnerable. HSNS grew bigger in parallel with the tendency to be an active social network observer (r = .26, p < .01), but with lowering of the interaction (r = -.16, p < .05) and self-presentation (r = -.18, p < .05) activities. On the contrary, grandiose narcissism was related to the higher activity in self-presentation (r = .29, p < .001) and interaction (r = .24, p < .001), as well as the higher total SNA score (see Table 9).
According to the findings on Sample 3 data, hypersensitive narcissism is related to numerous interpersonal problems. Namely, the higher the HSNS score was, the more significant interpersonal issues could be observed: the IIP-32 subscales were all significantly correlated to the HSNS score, with only the "Self-sacrificing -Overly nurturant (too caring)" subscale being comparably less connected (r = .16, p < .05 comparing to r = .31, p < .001 for other subscales) (see Table 10). Vulnerably narcissistic people have troubles with assertiveness (r = .31) and giving support to others (r = .34), mainly building relationships through control and domineering (r = .41); feel hard to be socially engaged (r = .38) and are usually cold and distant (r = .41); cannot open their feeling to other people, but if they do it, they tend to spread too much (r = .38); feel being dependent and exploited by others (r = .44). They are also more likely to have less diverse relationships with their close person (r = -.18, p < .05).

Discussion
Based on two studies, the double-translation of the 10-item self-report HSNS in non-clinical samples of Ukrainian-speaking respondents has satisfactory internal reliability, moderate retest reliability in the one-month interval, and adequate validity indicators.
The Remarkably, reliability coefficients are somewhat lower in Eastern European parts compared to the Western respondents. The reason for it may be language specificity. Although we used a double-blind translation procedure, the meaning of some questionnaire items may be quite lingua-specific and cultural-specific, which can lower the homogeneity of the tool in Slavic languages, like Polish, Russian, or Ukrainian.
Language and cultural similarities (and differences) exist between the countries compared to the USA, where the preliminary validation took place. One of the most prominent differences of the named countries is the level of individualism: it is the lowest in Ukraine and significantly higher in the USA compared to all Easter European societies (moreover, taking other dimensions into account, the USA is named "one of the most individualist cultures in the world") (Hofstede Insights, 2021). Such cultural differences may considerably influence the exact ways of narcissism manifestation. Items with the lowest inter-item correlations (4, "I dislike sharing the credit of achievement with others"; 6, "I feel that I am temperamentally different from most people"; 8, "I easily become wrapped up in my own interests and forget the existence of others"; 10, "I am secretly "put out" or annoyed when other people come to me with their troubles, asking me for my time and sympathy" [Hendin & Cheek, 1997, p.592]) have strong individualistic meaning. They cannot represent a collectivistic style where "We" (not "I") defines a self-image much more (Hofstede Insights, 2021). Additionally, American culture is much more "driven by competition, achievement, and success, with success being defined by the "winner" or "best-in-the-field" (Hofstede Insights, 2021), which is also an essential narcissistic feature. Therefore, the whole HSNS may not be sensitive enough to sociocultural specificity. It generally corresponds to the statement that narcissism scores (but mostly in grandiose form) are higher in individualistic cultures than in collectivistic cultures (Vater et al., 2018).
Despite this, the received results on reliability (both internal consistency and retest stability) are moderate. A possible reason for it can be a small number of the items in an original questionnaire: "if alpha is very low, the test is either too short" (Nunnally & Bernstein, 1994, p.252). Alphas above 0.6 can be considered as acceptable (Pallant, 2001), namely for the 10-item scales (Loewenthal & Lewis, 2018).
Murray's (1938, as cited in Hendin & Cheek, 1997) conception of narcissism is the background of HSNS. He described its covert form by the tendencies of being exploitative, self-enhancing, and self-aggrandising, combined with feelings of anxiety and hypersensitivity to other people's opinions. Unless Holly Hendin and Jonathan Cheek (1997) did not provide any data on the phenomenon structure, the "vulnerable narcissism" concept was expanded and clarified within the years of studies, adding numerous characteristics. Traits of being focused on the self and worried about the self-image remain basic. Andrea Fossati et al. (2009) reproduced a two-factor structure using an Italian version of HSNS, reflecting those two aspects of hypersensitive narcissism manifestation. However, Carmen Ripoll et al. (2010) have obtained a three-factor structure in a Spanish-speaking sample. Besides hypersensitivity to judgment, they interpreted the self-focus as divided into two sides: namely self-absorption or egocentrism and egoism as a difficulty to share. In the Ukrainian sample, we received a very akin three-factor scale structure. We argue that the items included in the "egoism" factor do not reflex "a doctrine that individual self-interest is the actual motive of all conscious action" (Merriam-Webster, 2021). This factor (items 1, 6, and 8) is more likely to describe the detachment domain traits, e.g., social withdrawal and intimacy avoidance (Miller et al., 2013(Miller et al., , 2017. Thus, we have named the factor "social detachment", following Graham Room's definition as of a "discontinuity in relationships with the rest of society" (1999, p. 171). In contrast to hypersensitivity to judgment (where a person stresses what other people think of him/her) and to self-absorption (which means to be focused on personal problems only), the "social detachment part" of the construct reflects the way people build cooperation with others. People scoring high on the vulnerable narcissism scale tend to maintain distance and avoid social contacts to protect themselves from sharing other people's problems. The source of social detachment is probably the narcissistic lack of empathy, mainly their inability to identify with other people's emotional distress (Luchner & Tantleff-Dunn, 2016). Daniel Lapsley and Paul Stey (2011) claim that this empathy dysfunction appeared because of specific conditions in early childhood when parents did not provide an adequate empathic mirroring to a child. Therefore, in adulthood, such a person builds a life around this unsatisfied need for unconditioned love. He/she is looking for a positive reflection of others and has an oversensitive reaction to being rejected or criticised. At the same time, they have nothing to give back; thus, when others ask for support, they experience it as strained by their "miserable problems." These three described factors are not the HSNS subscales, as two of them do not withstand a demand for internal consistency. HSNS is a unidimensional scale, and we recommend using it that way in a Ukrainian translation, according to Cronbach's alphas and CFA analysis. Despite this, these three spheres of hypersensitive narcissism construct fully describe a "blue face" of narcissism, and it is significantly different from its "bright face" (Rogoza et al., 2018).
Namely, our study revealed that the grandiose and vulnerable sides of narcissism are generally different and in many aspects, opposite phenomena. HSNS strongly negatively correlates with NPI scores. In clinical conditions, positive correlations of these two forms of narcissism appear The reported results, thus, may partly be explained by the differences between "eastern" and "western" mentality: Aline Vater et al. (2018), comparing narcissistic traits of people from East and West Germany, found out that those grown in the East have lower grandiose narcissism but higher self-esteem, and "by trend" lower scores of vulnerable narcissism, compared to individuals from the West. The characteristics given to Ukrainian society reflect the need to save good relationships when putting "I" at the first stage is culturally atypical. It may be a reason for such a gap between "apparent" (grandiose) and "hidden" (vulnerable) self-reported narcissism; however, this issue needs further detailed study.
On the results of the conducted analysis, we may describe the image of a highly vulnerably narcissistic person. It includes being sensitive to other people's judgments, explicitly low self-esteem, and an ability to be hostile and use aggressive behaviours to protect oneself. The current findings prove the Ukrainian translation of the HSNS to have high criterion validity. , which is proof of the Ukrainian version of the HSNS discriminant validity. Additionally, the vulnerable narcissism score is strongly positively related to rivalry as the social strategy of complimentary self--image maintenance in a self-protecting, aggressive manner (the same reported by Back et al., 2013;Rogoza et al., 2018). To protect their "fragile sense of self" (Kaufman et al., 2020, p. 5), highly vulnerably narcissistic people are more likely to use antagonistic forms, devaluate other people, be hostile to diminish the threat to the ego that they feel others may provide being better than they are. As a result, social conflicts appear (Back et al., 2013). That's why Rogoza et al. (2018) call rivalry strategy a "dark face of narcissism." One possible explanation of the aggressive self-defensive strategy can be found in Guido Veronese et al. : It appears that the more individuals try to defend themselves from threats to their identity by raising the shield of narcissistic grandiosity, the more the self feels threatened by the failure of this strategy of "positive self-description." Subjects become trapped in a vicious cycle that does not allow them to distance themselves from the continuous battle between their need to define themselves positively and humiliating attacks on their self-esteem (2015, p. 40).
The current research reveals that, compared to people with grandiose narcissism, those who score high for hypersensitive form build their self--attitude, not from the "I"-position (negative correlation with self-focused personal characteristics), but the social roles, mainly professional. Their self--image consists of labour functions, which means that they perceive themselves basically for what they do, not for who they are. We perceive this result as supporting Nancy McWilliams (2011) psychoanalytic narcissism interpretation: after being a parent's narcissistic extension in childhood, a person is used to treating oneself from the role function -in this case, from what he/she does to earn money and for being socially wanted. One more specific result is the negative correlation between covert narcissism score and community roles in self-concept. We explain it by bad communal traits (like agreeableness, cooperativeness, generousness, etc.) (Fukushima & Hosoe, 2011).
Such self-esteem deficits are probably a source of interpersonal problems that are more expressed among people with higher HSNS scores. Kelly Dickinson and Aaron Pincus (2003) reported that vulnerable narcissistic individuals are usually highly distressed, have traits of avoidant personality and numerous interpersonal problems. Additionally, in the current research, the higher the covert narcissism score is, the less assertive, more intrusive, and more dependent a person is. High scores of vulnerable narcissism correlate with difficulties in maintaining close relationships due to "considerable fears of relating to others, lack of confidence in their ability to initiate and maintain social relationships, and fears of being disappointed or ashamed of their needs within relationships" (Dickinson & Pincus, 2003, p. 201).
We have also found evidence supporting a specific Internet social media activity depending on the covert narcissism level. Namely, compared to those who score higher for grandiose narcissism, the higher the vulnerable narcissism is, the less active social media behaviour in communication and self-demonstrating aspects are. Although some researchers (e.g., Ksinan & Vazsonyi, 2016;Liu & Baumeister, 2016) claim that the inability to find positive feedback in the real world leads to more active distant communication, in the case of vulnerable narcissism, it is only partly true. Higher HSNS scores connect to more active observations, but not being "a speaker" and "a reporter" in social networks. The explanation could be found in the works of Joshua Miller et al. (2017) and Czarna et al. (2018), who suggest that shame is one of the main emotional experiences of vulnerable narcissists. Shame drives the interpersonal relationships of narcissists (Dickinson & Pincus, 2003). It shows the same behavioural patterns in the interpersonal sphere online and offline. Therefore, highly narcissistic people can neither fulfil their needs in the real world nor on social media.
In summary, the Ukrainian version of HSNS is a reliable and valid psychological tool, ready to be applied in further research for non-clinical samples. The HSNS promises to be a valuable addition to the scales measuring complex personality constructs among adults.

Limitations and Further Research
This research has some limitations. Firstly, there are sample composition imperfections. In particular, the significant number of young participants (age less than 30 in all four studies is the average) and women (more than 2/3 in all groups were ladies). It is also notable that we involved only a non-clinical sample, although many research studies state numerous links of covert narcissism and the symptoms of mental disorders. There also were group respondents' disproportions that could affect the results. Secondly, we used some of the questionnaires for comparison without their proper adaptation in Ukrainian. Thirdly, self-report measures typically give socially desired but not always realistic data; thus, they are not ideal for making exact conclusions. Finally, reliability coefficients were expected to be higher than 0.61-0.65, although the reason for it is probably a small items' number in an original questionnaire. Future research would be well served by better research groups balancing, involving participants not only in testing but in clinical interviews, examining the differences between clinical and non-clinical samples.
Source: own research.