The role of perfectionism in psychological health: A study of adolescents in Pakistan

This study investigated the effects of perfectionism on psychological health and explored the adaptive and maladaptive dimensions of perfectionism in the Eastern culture of Pakistan. Demographic data were also analyzed. Participants were 323 university students (144 males & 179 females), who completed the Frost Multidimensional Perfectionism Scale (Frost, Marten, Lahart & Rosenblate, 1990), the Perceived Stress Scale (Cohen, Kamarck & Mermelstein, 1983), the Psychological Well-being Measure (Ryff, 1989) and the Hospital Anxiety and Depression Scale (Zigmond & Snaith, 1983). Results indicate that overall perfectionism has a significant positive relationship with psychological distress and non-significant negative relationship with psychological wellbeing. Three dimensions of perfectionism, personal standards (PS), parental expectations (PE) and organization (OG), have a significant positive relationship with psychological well-being, similarly other three dimensions – concern over mistake (CM), parental criticism (PC) and doubt about actions (DA) – have a strong positive relationship with psychological distress. Results of Independent sample t-tests indicate that there is a significant difference between high and low overall perfectionism groups on stress, depression, anxiety, psychological distress and two sub-scales of psychological well-being environmental mastery and purpose in life. Moreover, individuals having high scores on CM, PC and DA also have high scores on depression, stress, anxiety and psychological distress as compared to low scorers on the same scales, whereas high scored PS, PE and OG groups showed high level of psychological well-being, environmental mastery and purpose in life. Finally, analyses of demographic data indicate that only age has a significant effect on DA and age by gender interaction has a significant effect on PE. Potential implications for educational and counseling purposes are discussed.


Perfectionism and psychological well-being
There are only a few studies regarding the association between perfectionism and psychological well-being. For example, Robert (1996) found that adaptive perfectionists had lower levels of procrastination than non-perfectionists and perfectionism was a significant predictor of achievement motivation. Moreover, as the level of perfectionism increases the level of depression decreases and that of self-esteem increases (Denise, Michael & Robert, 2000). A study of 261 college students investigated the relationship among parenting experiences, adaptive and maladaptive perfectionism. Results suggested that adaptive and maladaptive perfectionism had a different relationship with depression as maladaptive perfectionism mediates the relationship between harsh parenting and depression proneness and perfectionist parenting leads to adaptive perfectionism and has an inverse relationship with depression proneness (Enns, Cox & Clara., 2002). It was found from study of 123 tenth-through twelfth-grade students that students' personal standards were significant predictors of academic achievement and achievement motivation moreover as personal standards increased, the levels of self-esteem also increased and depression decreased (Accordino, Accordino & Slaney, 2000).

Dimensions of perfectionism
Previous empirical researches have provided contradictory results about the positive and negative dimensions of perfectionism. For example, the study of children suggests that dimensions of perfectionism may be relevant variables in, and differential predictors of, maladjustment and distress as self-oriented perfectionism interacted with social stress to predict anxiety, whereas, self-oriented perfectionism interacted with achievement stress and with social stress to predict depression Europe's Journal of Psychology 128 (Hewitt, Caelian, Flett, Sherry, Collins & Flynn, 2002). According to Koivula, Hassmén and Fallby (2002), the relation between self-esteem and perfectionism differs depending on which dimensions of self-esteem and perfectionism are considered.
Athletes with a high self-esteem based on a respect and love for themselves had more positive patterns of perfectionism, whereas athletes who have a self-esteem that is dependent on competence aspects showed a more negative type of perfectionism. Furthermore, negative dimensions of perfectionism were related to higher levels of cognitive anxiety and lower levels of self-confidence. In a clinical study of social phobic patients high scores on two perfectionism dimensionsconcern over mistakes and doubts about actions -were associated with increased social anxiety, trait anxiety, and general psychopathology (Juster et al., 1995). Stoeber and Rambow (2007), revealed from a study of adolescent school students that negative reactions to imperfection were related to fear of failure, somatic complaints, and depressive reactions whereas striving for perfection was related to hope of success, motivation for school, and school achievement. In addition, striving for perfection showed a negative correlation with depressive symptoms and was associated with adaptive traits. Two dimensions of perfectionism -pure personal standards and maladaptive evaluative concerns -have different associations with psychological distress and positive affect. Doubts about actions has a close association with anxiety (Frost et al., 1993), concern over mistakes, doubts about actions, and parental criticism were associated depression and negative affect Minarik & Ahrens, 1996), whereas pressure from others to be perfect has a correlation with stress, increased depression and negative affect (Cheng, 2001;Dunkley, Zuroff, & Blankstein, 2003). In a sample of Australian university students, doubts about actions showed a positive association with psychological distress, e.g. stress, depression, anxiety and negative affect (Robert & Teresa, 2002).
In addition, maladaptive dimensions of perfectionism have a positive relation with depression (Blatt, 1995;Murray & Brian, 1999), stress (Chang, Watkin & Banks., 2004) and have a negative association with self-esteem (Kenneth., Jeffrey & Robert, 1998). A pure personal standard has a significant negative correlation with psychopathology, and a positive correlation with well-being. In contrast, maladaptive evaluative concerns was positively related to psychopathology and negatively related to well-being (DiBartolo, Yen Li & Frost, 2008). Some previous findings indicate that adaptive dimensions of perfectionism have a positive association with personal control, resourcefulness, self-esteem, and adaptive learning strategies (Flett, Hewitt, Blankstein, & Dynin, 1994;Flett, Hewitt, Blankstein, & O'Brien, 1991).

Rationale of present study
The present research was conducted to redefine the trait conceptually and explore whether it is adaptive or maladaptive in an Eastern culture. It explored the role of perfectionism and its dimensions in mental health in terms of psychological wellbeing and psychological distress. A student sample was used for this research because the university is an important place to study the correlates of perfectionism as it presents both a challenging and stressful environment. Hypotheses: 1. There are significant relationships among perfectionism, psychological wellbeing and psychological distress.
2. Perfectionism has a significant positive relationship with psychological distress compared to psychological well-being.
3. There are significant effects of perfectionism and dimensions of perfectionism on sub-scales of psychological well-being and psychological distress.
4. There are age and gender effects on overall perfectionism and dimensions of perfectionism, psychological well-being and psychological distress.

Participants
The sample included 323 students (144 males and 179 females), from GC University, Lahore and KC University, Lahore, Pakistan. The age of participants ranged from 16 to 30 with a mean of 19.40 years and a standard deviation of 2.63 years, and 48% were from intermediate classes and 52% were from Masters Classes. All of the participants indicated Pakistani nationality, they were unmarried and Muslim by religion. Of the total sample, 66.3% were belonged to single family system and 33.7% were from joint family system. Social status breakdown was as follow: Lower (19%), Middle (63%), and Upper (18%).

Measures
Frost Multidimensional Perfectionism Scale  was used to assess the overall perfectionism and the dimensions of perfectionism. comprising six items. Psychometric properties indicated that MPS is a highly reliable and valid measure as reported internal consistency (reliability) is .90 (Parker & Adkins, 1995), and criterion-related validity has been tested through correlations between MPS subscales and measures of psychological symptoms (Frost et al., 1993.   and the Burns Perfectionism Scale (Burns, 1980). Organization subscale was omitted to calculate the overall perfectionism score because this scale scores are not required to obtain the total perfectionism score. For the present sample coefficient alpha was .78 for Overall perfectionism, whereas for dimensions of perfectionism coefficient alpha ranged between .70-.83.
The Perceived Stress Scale (Cohen, Kamarck & Mermelstein, 1983) is a widely used measure to assess the amount of stress in one's life rather than in response to a specific stressor. There are three versions of the scale, with 4-items, 10-items, or 14items. The 10-item version has maximum reliability (Cronbach's alpha, .78) as tested by Cohen and Williamson (1988) and also convergent validity with other assessments of stress (Cohen et al, 1983), so PSS-10 was used for the present research to evaluate stress. Ratings are based on a five-point scale ranging from never (0) to very often (5) and positively-phrased items 4, 5, 7, and 8, are coded in reverse. For the present sample Coefficient alpha was .72.
The Psychological Well-being Measure (Ryff, 1989) was used to measure psychological well-being. It consists of 18 items, three for each of the six distinct subscales. Responses are recorded on a six point Likert scale ranging from 1 (Strongly disagree) to 6 (Strongly agree). The dimensions of well-being are: Self-acceptance, Positive Relations with Others, Autonomy, Environmental Mastery, Purpose in Life and Personal Growth, having Cronbach alpha measures of . 82, .86, .77, .79, .73 and .84 respectively as reported by Ryff and Keyes (1995). For the present sample alpha coefficient ranged between .73-.84 for sub-scales of psychological well-being scale.
For the total psychological well-being scale the Alpha value was .80.

131
The Hospital Anxiety and Depression Scale (Zigmond & Snaith, 1983), was used to assess levels of anxiety and depression. It comprises statements which rate experiences over the past week. The 14 statements are relevant to either generalized anxiety (7 statements) or depression (7 statements). Even-numbered questions relate to depression and odd-numbered questions relate to anxiety.
Responses are scored on a scale from 3 to 0. The two subscales, anxiety and depression have been found to be independent measures. Many studies confirmed that it is also a valid measure for community samples as reported by Snaith (2003). It is considered to have a good factor structure, homogeneity, inter correlations and internal consistency as the reported Cronbach alpha for anxiety was .80 and for depression .76 (Mykletun, Stordal & Dahl, 2001). For the present sample coefficient alphas were .73 for both sub-scales.

Procedure
The questionnaires were administered in classroom settings at both Universities.
Ethical permission was obtained from both Universities before data collection. All participants in the study participated voluntarily. The researcher visited each class prior to class time and informed the students about the purpose of study. Participants filled out first the subject consent form then completed some demographic variables e.g., age, gender, education, marital status, income level and family status, and after that the five questionnaires for measuring Perfectionism, Psychological wellbeing, Stress, Depression and Anxiety. The task took approximately 40 min. Due to incomplete responses 23 questionnaires were excluded from the analysis through list wise deletion and overall response rate was 99%.

Results
Pearson Product Moment correlations were computed between the main variables to explore their relationship. Results in Table 1 indicate that overall perfectionism has a significant positive relationship with psychological distress and all its subscales.
Moreover, overall perfectionism also shows a non-significant negative relationship with psychological well-being and its three subscales (autonomy, self-acceptance and positive relations with others) and a significant positive relationship with its other three subscales (environmental mastery, purpose in life and personal growth).
Concern over mistakes shows a significant positive relationship with psychological distress and with all its subscales. It has a non-significant negative relationship with psychological well-being in terms of significant negative relationship with (environmental mastery), insignificant negative relationship with (autonomy, self-Europe's Journal of Psychology 132 acceptance, and positive relations with others) and non-significant positive relationship with (purpose in life and personal growth. Personal standards indicate a non-significant negative relationship with psychological distress and all its subscales. Moreover it also shows a significant positive relationship with psychological well-being and with its three sub scales (environmental mastery, purpose in life and personal growth). It also indicates a positive relationship with other three subscales of psychological well-being (autonomy, self-acceptance, positive relations with others) but this relationship was not significant.
Parental expectations shows a very weak negative relationship with psychological distress and with all its subscales whereas it indicates a significant positive relationship with psychological well-being and its two sub scale (environmental mastery and purpose in life) and non-significant positive relationship with (autonomy, selfacceptance, positive relations with others and personal growth).
Parental criticism has a significant positive relationship with psychological distress and all its subscales and a non-significant negative relationship with psychological well-being in terms of significant negative relationship with (autonomy and selfacceptance) and a non-significant positive relationship with (environmental mastery, purpose in life, positive relations with others and with personal growth).
Similarly doubts about actions also have a significant positive relationship with psychological distress and with all its subscales and a significant negative relationship with psychological well-being and all with its subscales.
Organization indicates a non significant negative relationship with psychological distress and all its subscales. Moreover it also indicates a significant positive relationship with psychological well-being and all its subscales Seven separate sets of independent sample t-test were computed to ascertain the effects of dimensions of perfectionism on the subscales of psychological well-being and psychological distress.  Table 2 indicates significant differences between high and low overall perfectionism groups on psychological distress and its all sub scales (depression, stress and anxiety) as well as on two subscales of psychological well-being (environmental mastery and purpose in life). Results also shows non-significant differences between high and low overall perfectionism groups on psychological well-being and its four subscales (autonomy, self-acceptance, positive relations with others and personal growth).
Results indicated a significant difference in psychological distress, stress, depression and anxiety scores for low(11-25) and high (26-44) concern over mistakes groups, low (4-12) and high (13-20) doubts about actions groups, and low (4-10) and high (11-57) parental criticism groups whereas high and low groups of these three dimension did not show significant differences on psychological well-being and any subscale of psychological well-being (see Table 3, 4 & 5).   has high scores on psychological well-being and its three subscales (environmental mastery, purpose in life and self-acceptance), whereas high and low scored organization groups indicate significant differences on psychological well-being (environmental mastery, purpose in life and personal growth). Furthermore, high and low scored groups on these three dimensions of perfectionism did not show any significant differences on psychological distress and its all subscales and any other scales of psychological well-being.

The role of perfectionism in psychological health
137 Multivariate analyses of variances were conducted to determine age and gender differences in perfectionism, psychological well-being and psychological distress.
The results, reported in Table 9, reveal that gender did not have a significant effect on any dimension of perfectionism, overall perfectionism, psychological well-being and psychological distress. Age and Gender by age interaction also did not have significant effect on any dimension of perfectionism, overall perfectionism, psychological well-being and psychological distress. It was found though that Age only has a significant effect on doubt about actions, F (3, 323) = 6.21, p < .05 and the Age by Gender interaction has a significant effect on parental expectations, F (3, 323) = 0.67, p < .05.

Discussion
The present study examined the role of perfectionism in mental health. There were two basic objectives of the study: to explore the adaptive and maladaptive dimensions of perfectionism in eastern culture of Pakistan and to examined the effects of perfectionism on mental health in terms of psychological well-being and psychological distress.
Regarding first objective of present research the results indicate that overall perfectionism has a clear association with mental health reflected in significant positive relationships with psychological distress and non-significant negative relationships with psychological well-being. These results confirm some previous findings that overall perfectionism has a positive association with measures of psychological distress Frost et al., 1993) and negative relationships with positive psychological constructs e.g. achievement motivation, self-esteem (Frost & Henderson, 1991).
The sub dimensions concern over mistake, parental criticism and doubt about actions have a positive relationship with the maladaptive personality construct e.g.
psychological distress, depression, stress and anxiety and showed non-significant negative relationship with psychological well-being and its some subscales. This is in line with previous findings that concern over mistakes and doubts about actions have a close relationship with psychopathology Yoshihisa & Yasukszu, 1999), concern over mistakes has a strong association with trait anger (Sinclair, Czech, Joyner & Munkasy, 2006) and ample amount of research indicated a positive relationship between maladaptive dimensions of perfectionism and psychological distress (Aldea & Rice, 2006;Chang & Rand, 2000;Connor., et al., 2002;Hewitt, Flett, & Ediger, 1996;Wyatt & Gilbert, 1998).
Personal standards, parental expectations and organization have a significant positive relationship with the adaptive personality construct e.g. psychological wellbeing, and show a negative relationship with psychological distress. This is again in line with previous findings that high personal standards and organization are associated with adaptive traits  and both these dimensions have a negative correlation with psychological distress (Lynd-Stevenson & Hearne, 1999).
These findings highlight the fact that overall perfectionism and its three dimensions of concern over mistake, parental criticism and doubt about actions are negative and other three dimensions e.g. personal standards, parental expectations and organization are positive in a Pakistani context.
Regarding the effects of overall perfectionism and sub-dimensions of perfectionism on psychological well-being and psychological distress, results revealed that overall perfectionism has a significant effect on two subscales of psychological well-being environmental mastery and purpose in life, as individuals having low scores on overall perfectionism have high scores on environmental mastery and purpose in life.
Moreover, results also showed that high level of overall perfectionism also indicates high level of psychological distress, stress, depression and anxiety.
Personal standard has a significant effect on psychological well-being and two subscales of psychological well-being: environmental mastery and purpose in life, as individuals with high scores on personal standards showed high scores on environmental mastery and purpose in life and psychological well-being. Similarly, The role of perfectionism in psychological health 139 individuals having high scores on parental expectations also showed high scores on environmental mastery, purpose in life, self-acceptance and psychological wellbeing whereas there is a significant difference in mean of high and low organization groups on environmental mastery, purpose in life, personal growth and psychological well-being. Furthermore high and low scored groups of personal standards, parental expectations and organization did not indicate significant differences on depression, stress and anxiety and psychological distress. The results also revealed that individuals having low scores on concern over mistakes, doubts about actions and parental criticism showed lower scores on stress, depression, anxiety and psychological distress whereas high and low scored groups of these three dimensions did not shows any significant differences on psychological wellbeing and any subscale of psychological well-being. Thus, these three dimensions of perfectionism seem unhealthy or maladaptive. These results support the notion that overall perfectionism is an unhealthy personality trait and it has negative effect on the mental health of Pakistani university students. These results also indicate that personal standards, parental expectations and organization have a positive effect on the mental health of Pakistani university students whereas other three dimensions of perfectionism -concern over mistake, doubt about actions and parental criticism -have negative effect on their mental health.
In line with previous findings (Parker & Adkins, 1995;1996;Robert & Teresa, 2002, Sinclair, Czech, Joyner & Munkasy, 2006, the present study confirms that gender did not have a significant effect on any dimension of perfectionism or overall perfectionism. Age has a significant effect only on doubt about actions and age by gender interaction has a significant effect just on the parental expectations dimension of perfectionism.

Limitations and suggestions
Present research has certain limitations as well. First of all the present sample consists of university student from only two universities of Pakistan, so generalization of these results is difficult. Larger samples from more universities of Pakistan should be used for further research. Furthermore, different types of samples can be used in order to obtain a broader picture of the relationships between these variables. Secondly, these results are based on self-report measures alone, which might have biased the results as self-report measures have some limitations e.g. unreliable answers of respondents, negative affectivity bias, and extreme response style. In light of this, other methods of data collection can be used in further research. Finally, this is cross sectional research whereas traits like perfectionism, psychological well-being and psychological distress should also be studied longitudinally.

Conclusion
It is concluded that perfectionism has a negative role for the mental health of Pakistani university students because overall perfectionism has strong positive relationship with psychological distress rather than psychological well-being. It only has three adaptive dimensions -personal standards, parental expectations and organization -as these dimensions have a strong positive association with psychological well-being and higher scores on these dimensions are associated with high scores on some subscales of psychological well-being. Similarly, it has three maladaptive dimensions -concern over mistakes, parental criticism and doubts about actions -since these dimensions have a strong positive correlation with psychological distress and individuals having high scores on these dimensions also show high scores on subscales of psychological distress e.g. stress depression and anxiety. The cultural factor might be important in these findings and needs further exploration for consistency with results in other Eastern cultures. These results could be very helpful for the development of a theory of perfectionism and for clinical and counseling purposes. In this regard, overall perfectionism and its three maladaptive dimensions can be addressed in clinical interventions and the counseling of students having psychological distress could focus on these issues. Finally, it is hoped that the conclusions of the present article will facilitate a clearer understanding of the relationship between perfectionism and psychological health in the Eastern culture of Pakistan.