Predicting Borderline Personality Features on the Basis of Alexithymia and Attitude toward Mother
Razavi Int J Med. 2015 May; 3(2): e25027.
DOI: 10.5812/rijm.3(2)2015.25027
Published online 2015 May 2.
Research Article
Seyede Fateme Sajadi 1,2,*; Yadolla Zargar 1; Marziye Khodahemati 3; Zahra Sahraeeyan 4; Seyede Forugh Sajadi 5
1Department of Psychology, Shahid Chamran University, Ahvaz, IR Iran 2Young Researchers and Elite Club, Shiraz, IR Iran
3Department of Psychology, Shiraz University, Shiraz, IR Iran
|
4Department of Psychology, Allame Tabatabaee University, Tehran IR Iran
5Department of Health, Bushehr University of Medical Sciences, Bushehr, IR Iran
: Seyede Fateme Sajadi, Department of Psychology, Shahid Chamran University, Ahvaz, IR Iran. Tel: +98-9173046244, E-mail: f-sajadi@mscstu.scu.ac.ir
Received: October 30, 2014; Revised: January 11, 2015; Accepted: February 8, 2015
1. Background
Borderline Personality Disorder (BPD) is a severe men- tal disorder characterized by pervasive pattern of insta- bility of moods, interpersonal relationships, self-image and behavior (1). The distinguishing symptoms of BPD are malfunctions in a wide range of neurobehavioral systems, emotional manifests (e.g. marked reactivity), behavioral inhibitions (e.g. impulsivity), cognitions (e.g. paranoia or disruption during serious distress), interper- sonal performances (e.g. fear of rejection) and disrupted interpersonal relationships, which are considered as the core of diagnosing mental disorders and distress associ- ated with the disorder (2). Although the etiology of BPD is not well known, many theoretical models emphasize on having negative experiences with primary caregivers. Adverse childhood experiences can help to explain why patients with BPD have difficulty in having attachments, are afraid of being rejected and rely less on others com- pared with controls (3).
According to Linehan’s biosocial theory (1993), symptoms of BPD are caused by two factors rooted in biosocial factors of the individual’s childhood; one of these factors is invalidating environment and the other is biological background of emotional instability. Invalidating en- vironment occurs when the closest people to the child (especially parents) frequently criticize, humiliate and punish the child’s awareness of his inner experiences (thoughts and feelings) and attribute these thoughts and feelings to undesirable characteristics such as laziness or ignorance. The fourth edition of Diagnostic and Statisti- cal Manual of Mental Disorder says if clinicians find out that frequent behavioral patterns in children are a part of their character not a set of behaviors that children show in particular situations, a diagnosis of personality disor- der can be considered for them. However, possibility of this diagnosis for children with borderline personality features provides the appropriate treatment (4).
Also, according to the Linehan’s theory, BPD is first an emotional regulation disorder. It is believed that emotional dysregulation is caused by intense emotional reac- tions, effective emotional experiences and lack of skills to manage difficult emotions. Alexithymia is a structure as- sociated with the ability to regulate emotions. Alexithymia means deficits in emotional processing that involves im- paired emotion recognition, expressing feelings toward others and externally oriented style of thinking (5). Beren- baum (1996) also stated in his theory that when people can not identify their emotions, they may fail to use the feed- back that emotions generate. Therefore, he believed that as emotional dysregulation is one of the main character- istics of BPD and as the inability to identify emotions is as- sociated with inability to regulate them, BPD is associated with alexithymia. To better understand the overlapping association between alexithymia and BPD, it is necessary to investigate their infrastructure relationship. One com- mon flaw that exists in both conditions is emotional dys- regulation characterized by intense emotional reactions, severe and extreme emotional experience and lack of skills to manage strong emotions (6).
Webb and McMurran (6) reported that alexithymia was the only predictor of borderline personality features in a sample of 134 students. Based on the results of this study, the association between these two clinical syn- dromes that involves detecting, identifying, understand- ing and communicating with the emotions of others can strengthen a person’s ability to control these emotions. Researches have shown that emotions act as a feedback system through which people can regulate their behav- ior and interpersonal relationships (7).
2. Objectives
The present study was performed due to lack of the domestic researches on the association between alexi- thymia and attitude toward mother with borderline personality features in a nonclinical sample of under 18-year-old people. Achieving the objectives of the study, from theoretical perspective, would increase knowledge on BPD, and from practical point of view, can be used to identify specific patterns of borderline personality. Therefore, this study aimed to help identify further variables affecting borderline personality features of high school students.
3. Patients and Methods
This was a correlation-descriptive study, investigating the association between addiction potential, suicide ide- ation and educational performance with borderline per- sonality features in high school students in Shiraz, Fars province, Southern Iran. The data obtained by question- naires was analyzed with canonical correlation analysis using SPSS-18 (IBM Corporation, New York). Canonical cor- relation is similar to multiple regression analysis a com- pound of predict variables is applied to predict criterion variables. The difference lies in the number of criterion variables; in multiple regression there only exists one criterion variable, while in canonical analysis, there is more t3h.1a.nPoanrteic(8ip).ants
|
Students from four regions in Shiraz, studying in first to third grades of high school in educational year of 2012 – 2013 (Iranian year of 1391 – 1392) were considered to partic- ipate in the questionnaire surveys. The sample included 300 students (half and half males-females) chosen with multistage random sampling. First, two girl’s schools and two boy’s schools were chosen randomly, then two classes of each were selected by random and finally half of the students of each class were randomly cho- sen to answer the questionnaires. The age range of the participants was 14 to 18 years, with an average of 15.72 and standard deviation of 0.99. Overall, 45% of subjects studied in the first grade, 41% in the second grade and 14% in the third grade of high school. The average CGPA (Cumulative Grade Point Average) of students in sam- ple was 17.14, SD = 1.92.
Borderline Personality Features Scale for Children (BPFS-C: Crick, Murray-Close, and Woods, 2005): This is a 24-item self-report questionnaire, which assesses bor- derline personality features in children and adolescents aged 9 – 17 years (9). This measure was adopted from the BOR (borderline) scale of the personality assessments scale (PAI; Morey, 1991), modified for use with youth. BPFS_C is scored on the 5-point Likert scale with respons- es ranging from 1 “not at all true” to 5 “always true” to valuate affective instability, identity problems, negative relationships and self-harm (10). The BPFS-C has shown good internal consistency in 12 months study by Crick et al. (9), performed on 400 students aged 10 – 12 years, (α > 0.76) as well as criterion validity (11) and construct validity (9). Prior research in Iran examining the 22-item instruments with a large community sample (n = 400) of boys and girls in high school showed a high consistency (α > 0.84) (12). In the current study, Cronbach’s α was 0.83. Toronto Alexithymia Scale (TAS-20): This is a 20-item self-descriptive statement. Participants were rated on the 5-point (strongly disagree to strongly agree) Likert Scale, each using the five-point Likert scale. The TAS-20 includes three dimensions: DIF, DDF and EOT (13). The TAS-20 has been demonstrated to have good psychometric proper- ties. The internal consistency of each subscale (α = 0.83,0.77 and 0.73 for the DIF, DDF and EOT subscales respec- tively) and the TAS-20 total score (α = 0.82) (14). Cronbach’s α in Iranian sample of students was 0.87 for total scale and 0.85, 0.84 and 0.80 for DIF, DDF and EOT respectively (15). In the current study, Cronbach’s α was 0.95 for total scale a19n9d20).75, 0.54 and 0.40 for DDF, DIF and EOT respectively.
|
This is a 25-item self-report scale, which measures the severity of problems in the child-mother relationship from the child point of view. The items are scored on the 7-point Likert scale ranging from 1 (rarely or none of them) to 7 (most or all of the time). Items are both positively and negatively worded to reduce the response bias, which positive items are reverse scored. High score is the indicator of severe problem in child-mother relationship (16). Cronbach’s α of the scale ranged between 0.93 and 0.97 (17). Cronbach’s α in the Iranian sample was 0.85 (12) and in the current study was 0.75.
4. Results
Descriptive statistics are reported for main study variables in Table 1. As seen in Table 1, the Mean (SD) scores obtained by the sample (n = 300) on variables of borderline personal- ity features was 58.98 (1.244), alexithymia was 59.153 (9.95) and child’s attitude toward mother was 84.290 (11.91).
To investigate the association between borderline personality features with alexithymia and child’s attitude toward mother in the total scale and subscales, a matrix for Pear- son correlation coefficient was calculated. Bivariate corre- lations between the studies variables is given in Table 2.
Results from correlational analyses revealed significant relationships between borderline personality features’ subscales and child’s attitude toward mother, and diffi- culty identifying feelings and difficulty describing feel- ings. Alpha coefficients ranged from 0.1 to 0.5 (Table 2).
Focusing on the main study variables, as shown in Table 2, significant positive relationships were found between borderline personality features dimensions, child atti- tudes toward mother, difficulty identifying feelings and difficulty describing feelings. No significant relation- ships were found between external oriented thinking style, borderline personality features and child’s attitude toward mother. The Pearson correlation demonstrated that high scores of difficulty identifying feelings had the most association with high identity problems’ scores.
In the current study, to investigate variables’ relation- ships, canonical correlation analysis was used.
| Table 1. Mean and Standard Deviation and Range for Main Study Variables | ||||
| Variables | Mean | SD | Minimum | Maximum |
| Borderline personality features | 58.983 | 1.244 | 31.00 | 98.00 |
| Affective instability | 15.263 | 3.613 | 6.00 | 24.00 |
| Identity problems | 14.816 | 4.190 | 6.00 | 25.00 |
| Negative relationships | 14.283 | 3.922 | 6.00 | 26.00 |
| Self-harm | 14.620 | 4.120 | 6.00 | 28.00 |
| Alexithymia | 59.153 | 9.959 | 31.00 | 88.00 |
| Difficulty identifying feelings | 20.90 | 6.17 | 7.00 | 34.00 |
| Difficulty describing feelings | 15.32 | 4.10 | 5.00 | 25.00 |
| External oriented thinking style | 22.92 | 3.79 | 8.00 | 34.00 |
| Child attitude toward mother | 84.290 | 11.916 | 51.00 | 122.00 |
| Table 2. Pearson Correlation for Research Variables a | ||||||||
| Variables | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| Borderline personality features | ||||||||
| Affective instability | – | |||||||
| Identity problems | 0.47 b | – | ||||||
| Negative relationships | 0.50 b | 0.47 b | – | |||||
| Self-harm | 0.44 b | 0.49 b | 0.53 b | – | ||||
| Alexithymia | ||||||||
| Difficulty identifying feelings | 0.35 b | 0.45 b | 0.39 b | 0.34 b | – | |||
| Difficulty describing feelings | 0.37 b | 0.31 b | 0.30 b | 0.27 b | 0.56 b | – | ||
| External oriented thinking style | -0.01 | -0.03 | -0.04 | 0.01 | -0.03 | 0.12 c | – | |
| Child attitude toward mother | 0.26 b | 0.22 b | 0.30 b | 0.29 b | 0.18 b | 0.15 b | -0.002 | – |
| a (n = 300).
b Correlations are significant at P < 0.001. c Correlations are significant at P < 0.05. |
||||||||
Alexithymia (difficulty identifying feelings, difficulty describing feelings and external oriented thinking style) and child’s attitude toward mother are considered as predictors of borderline personality features (affective instability, identity problems, negative relationships and self-harm) to study the joint multivariate relation- ship between these two classes of variables. The results of multivariate test of significance for canonical correlation full model are presented in Table 3. Wilks lambda (P < 0.001) being statistically significant, explains a relation- ship between difficulty identifying feelings, difficulty describing feelings, external oriented thinking style, child’s attitude toward mother and affective instability, identity problems, negative relationships and self-harm (Table 3). Lambda (λ) is the sign of unexplained variance, consequently 1- λis the full model effect size in r2 matrix. Accordingly, the effect size of three canonical correlation function equals 1 – 0.55 = 0.45. The effect size is the joint variance between two classes of variables that the full model can explain; therefore, the obtained model in this study explains 45% of variance between alexithymia (dif- ficulty identifying feelings, difficulty describing feelings and external oriented thinking style) and child’s attitude toward mother with borderline personality features (af- fective instability, identity problems, negative relation- ships and self-harm).
In addition to the mentioned method, researcher can test the significance level by dimension reduction analy- sis (Table 5). The test of significance results of cumulative effect of functions 1 to 4 is presented in Table 5, first row. The test checks whether the structure of functions is sig- nificant. As mentioned, cumulative effect of functions 1 to 4 (full model) is statistically significant (P < 0.001), but the rest of cumulative effect is not significant. Only the first function explains a significant amount of joint vari- ance between two classes of variables.
Results explain a significant relationship between two classes of variables and only the first function explains a significant variance. To find out the role of each vari- able in functions, standard and structural coefficients of variables are considered. Table 6 presents standard coef- ficients, structural coefficients and square structural co- efficient for dependent and independent variables in the first canonical function.
Following Alpert and Peterson (1972), only variables with minimum structural coefficients of 0.3 are inter- preted. Therefore, data presented in Table 6 shows that in the first function, difficulty identifying feelings (SC = 0.61), child’s attitude toward mother (SC = 0.42), dif- ficulty describing feelings (SC = 0.29) and external ori- ented thinking style (SC= 0.06) have respectively more important roles in linear structure of predictor vari- ables. Regarding dependent variables, identity problems (SC = 0.41), negative relationships (SC = 0.35), affective in- stability (SC = 0.30) and self-harm (SC = 0.18) play a role in linear structure of dependent variables.
More specifically, borderline personality features (affec- tive instability, identity problems,negative relationships and self-harm) are predicted by alexithymia (difficulty identifying feelings, difficulty describing feelings and ex- ternal oriented thinking style) and child’s attitude toward mother. Moreover, canonical R square coefficient (R2 C) is 33%, which determines amount of joint variance between two canonical classes of independent and dependent vari- ables. Furthermore, based on the standard coefficients presented in Table 6, for each the standard deviation incre- ment in affective instability, identity problems, negative relationships and self-harm, the first canonical function score increases as 0.30, 0.41, 0.35 and 0.18 respectively. For each unit, increase in standard deviation of difficulty iden- tifying feelings, difficulty describing feelings, external ori- ented thinking style and child’s attitude toward mother, the score of first canonical function increases to 0.61, 0.29, 0.06 and 0.42 respectively.
|
|||||||||||||||||||||||||||||||
In canonical correlation analysis, there is no convenient way to test significance level of functions separately. One way to investigate the issue is to consider the amount of variance that explains each function. As shown in Table 4, canonical correlations square (R2 C) of functions are 0.33, 0.03, 0.01 and 0.00 respectively Regarding the find- ings by Sherry et al. (18), functions explaining less than 10% variance are laid away and are not interpreted, then only the first function explaining 33% of joint variance is accepted and other functions are not interpreted.
However the number of functions obtained in canonical analysis is equal to the number of variables in the smallest class (dependent or independent). In this re- search, as a result of having four dependent and independent variables, four functions are acquired (Table 4).
|
Table 4. Functions Obtained From Canonical Correlation Analysis |
|||||
| Root Number | Eigenvalue | Percent | Cumulative Percent | Canonical Correlation | Square Correlation |
| 1 | 0.504 | 90.7 | 90.7 | 0.578 | 0.335 |
| 2 | 0.323 | 5.83 | 96.6 | 0.177 | 0.031 |
| 3 | 0.016 | 2.92 | 99.5 | 0.126 | 0.016 |
| 4 | 0.002 | 0.44 | 100 | 0.049 | 0.002 |
|
Table 5. Results for Dimension Reduction Analysis of Canonical Functions |
|||||
| Roots | Wilks L | F | DF1 | DF2 | P |
| 1 To 4 | 0.632 | 9.03 | 16 | 892 | P < 0.001 |
| 2 To 4 | 0.950 | 1.66 | 9 | 713 | 0.095 |
| 3 To 4 | 0.981 | 1.37 | 4 | 588 | 0.242 |
| 4 To 4 | 0.997 | 0.72 | 1 | 295 | 0.394 |
|
Table 6. Standard, Structural and Square Structural Coefficients of Research Variables |
|||
| Variables | Standard Coefficient | Structural Coefficient | Square Structural Coefficient |
| Affective instability | 0.307 | 0.768 | 0.589 |
| Identity problems | 0.414 | 0.820 | 0.672 |
| Negative relationships | 0.354 | 0.807 | 0.6511 |
| Self-harm | 0.189 | 0.722 | 0.521 |
| R2 C | 0.33 | ||
| Difficulty identifying feelings | 0.619 | 0.870 | 0.756 |
| Difficulty describing feelings | 0.299 | 0.702 | 0.492 |
| External oriented thinking style | 0.060 | 0.058 | 0.003 |
| Childs attitude toward mother | 0.422 | 0.585 | 0.342 |
5. Discussion
The present study aimed to investigate the role of alexi- thymia and attitude toward mother in predicting bor- derline personality features of high school students in Shiraz. The findings suggested that such prediction is possible. Canonical analysis produced a statistically sig- nificant function. The findings are explained based on cross loading that is the best way to interpret the conven- tional functions. Cross loadings of significant function represent that a pattern of high scores of alexithymia and attitude toward mother is correlated with a pattern of high scores of borderline personality features.
Such a finding fits well with Ridings (19), Webb and Mc- Murran (6) views on the association between alexithymia and borderline personality features. Since identifying and distinguishing different emotions experienced by these people are a part of borderline personality fea- tures, alexithymia may be conceptualized as a common feature of BPD (19). Therefore, difficulty in identifying, distinguishing, understanding and communicating with feelings and emotions would weaken the ability of an individual with BPD to regulate his or her emotions (6). Emotional dysregulation is a core feature of BPD and alexithymia is correlated with all aspects of emotional dysregulation including impulsivity and negative emo- tion. For example, alexithymia has a significant positive association with subscales of impulsivity control difficul- ties and difficulty in handling purposeful behavior that is related to an individual’s willingness to engage in impul- sive behaviors. Moreover, alexithymia is correlated with subscales of rejection of negative emotion and limited access to emotion regulation strategies, both of which are subscales representing an individual’s willingness to engage in negative emotions and are prominent features of borderline personality (19).
In line with Cohen et al. (20), Bandelow et al. (21), and Links et al. (22), we demonstrated that attitude toward mother is associated with borderline personality fea- tures. They approved the hypotheses of the present study since they proved that patients with borderline disorder have experienced adverse parenting practices and lack of care associated with affection and have negative at- titudes toward their parents. Furthermore, in keeping with Bradley (23), separation from mother before the age of five years has greater frequency in BPD patients. How- ever, separation from mother does not lead to BPD symp- toms per se. It is difficult for children to understand the reason of separation and they may blame themselves for the absence of mother. This could have a negative impact on children’s mental image of himself or herself and oth- ers (24). More specifically, dysfunctional parenting prac- tices and mostly parents’ dissatisfaction with children are strong predictors of BPD symptoms. In fact, mother’s self-righteous dissatisfaction with child can increase the risk of early separation and make the child feel worthless. In addition, these parental variables in individuals expe- riencing separations have not reduced symptoms of BPD. Early separations with attachment anxiety (fear of being abandoned) and avoidance (fear of intimacy) are closely correlated, both of which have a great correlation with BPD symptoms (25). Maternal interpersonal problems (including difficulty in having intimacy, blaming others and being angry with others) overlap with borderline symptoms. Therefore, these properties along with mod- eling or having potential genetic background can have adverse effects on children (26).
Overall, we did find support for linear combination of psychological variables including alexithymia and at- titude toward mother explains the borderline person- ality features in high school students. Despite its large sample size and noteworthy findings, there are several limitations to the current study possibly impacted our findings. firstly, our sample included 14 – 18 years old high school students, so the results should be generalized cau- tiously. secondly, TAS-20 was the only scale used to mea- sure alexithymia, which can be considered as a limita- tion since this scale does not contain daydreaming and fantasy components, these components were excluded in 1994 due to high correlation with social desirability and low correlation with the total scale (27), and the lack of reliability of this subscale has been also observed in sev- eral other studies (28). In conclusion, it is suggested to perform further studies with a closer look at these com- ponents to clarify the finding of this study better.
Acknowledgements
The authors are grateful to all students and schools who participated in this investigation. This research has been responded the support of Shahid Chamran University of Ahvaz (Department of Psychology).
Authors’ Contributions
Seyede Fateme Sajadi: analysis and interpretation of data, drafting of the manuscript and statistical analysis. Yadolla Zargar study supervision. Marziye Khodahemati: study concept and design. Zahra Sahraeeyan administra- tive, technical and material supports. Seyede Forugh Sa- jadi: data acquisition.
References
- Neacsiu AD, Lungu A, Harned MS, Rizvi SL, Linehan Impact of dialectical behavior therapy versus community treatment by
bitoy radnedrlMineentpael rHseoanlathli.ty disorder traits in young adults. Personal–
|
|
- Carver CS, Scheier MF. Origins and functions of positive and negative affect: A control-process view. Psychological Review.
|
|
- D1S9cei9el0anv;c9ae7rs(]1.A):.19–35.
|
- Crick NR, Murray-Close D, Woods Borderline personality fea- tures in childhood: a short-term longitudinal study.
|
|
- Sharp C, Ha C, Michonski J, Venta A, Carbone Borderline person- ality disorder in adolescents: evidence in support of the Childhood Interview for DSM-IV Borderline Personality Disorder in a sample of adolescent inpatients. (6):765–74.
|
|
- Chang B, Sharp C, Ha The criterion validity of the Borderline Personality Features Scale for Children in an adolescent inpa-
- SajtacdomiSeFs. o. f borderline personality disorder in high school students
. Ahvaz: Shahid Chamran University of Ahvaz; 2013.
|
- Evren C, Cinar O, Evren Relationship of alexithymia and dissocia- tion with severity of borderline personality features in male sub- stance-dependent inpatients. 2012; (6):854–9.
|
- Tull MT, Medaglia E, Roemer An investigation of the con- struct validity of the 20-Item Toronto Alexithymia Scale
2005;59(2):77–84.
|
|
|
- Besharat Reliability and factorial validity of a Farsi version of the 20-item Toronto Alexithymia Scale with a sample of Iranian students. (1):209–20.
|
- Dwairy Parenting styles and mental health of Palestinian- Arab adolescents in Israel. 2004; (2):233–52.
- Grotevant HD, Carlson
lONDON: Guilford Press; 1989.
|
|
- Sherry A, Henson RK. Conducting and interpreting canonical correlation analysis in personality research: a user-friendly
- Rdiisdoirndgesr:LeEx.p.laining the link between secondary psychopathy and
. Dayton: University of Dayton; 2011.
|
- Cohen P, Chen H, Gordon K, Johnson J, Brook J, Kasen Socio- economic background and the developmental course of schizo-
2008;20(2):633–50.
- Bandelow B, Krause J, Wedekind D, Broocks A, Hajak G, Ruther
|
|
- Early traumatic life events, parental attitudes, family history, and birth risk factors in patients with borderline personality disorder and healthy controls. 2005; (2):169–79.
|
|
- Links PS, Steiner M, Offord DR, Eppel Characteristics of bor- derline personality disorder: a Canadian study.
|
- Bradley The relationship of early maternal separation to bor- derline personality in children and adolescents: a pilot study.
borderline personality disorder. Behav Res Ther. 53:47–54.
ment relationships: A construct revisited. In: Cassidy J, Shaver PR
experts on emotional experience, expression, and acceptance in
2014;
|
|
- Lazarus SA, Cheavens JS, Festa F, Rosenthal Interpersonal functioning in borderline personality disorder: A systematic review of behavioral and laboratory-based assessments.
|
|
- Veague HB, Hooley Enhanced sensitivity and response bias for male anger in women with borderline personality disorder.
|
|
- Haugaard Recognizing and Treating Uncommon Behavioral and Emotional Disorders in Children and Adolescents Who have been Severely Maltreated: Bipolar Disorders.
2004; (2):131–8.
|
- Lang S, Stopsack M, Kotchoubey B, Frick C, Grabe HJ, Spitzer C, et Cortical inhibition in alexithymic patients with borderline personality disorder. 2011; (2-3):227–32.
- Webb D, McMurran Emotional intelligence, alexithymia and
6
- Bretherton I, Munholland Internal working models in attach-
|
editors.
. New York: Guilford Press; 1999. p. 89–111.
|
- Crawford TN, Cohen PR, Chen H, Anglin DM, Ehrensaft Early maternal separation and the trajectory of borderline personal-
2009; (3):1013–30.
|
- Bezirganian S, Cohen P, Brook The impact of mother-child interaction on the development of borderline personality disor- der. 1993; (12):1836–42.
|
- Bagby RM, Parker JD, Taylor The twenty-item Toronto Alexi- thymia Scale–I. Item selection and cross-validation of the factor
1994;
|
|
- Kiyotaki Y, Yokoyama K. Relationships of eating disturbances to alexithymia, need for social approval, and gender identity among Japanese female undergraduate
Razavi Int J Med. 2015;3(2):e25027
Copyright © 2015, Razavi Hospital. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 Internation- al License (http://creativecommons.org/licenses/by-nc/4.0/) which permits copy and redistribute the material just in noncommercial usages, provided the original work is properly cited.