Perfectionism: A Good Predictor of Stress and Anxiety Personality research has revealed a number of interactions between traits and clinically-significant mental health issues. For example, neuroticism has been shown to be predictive of anxiety and depressive disorders, while introversion is a common trait among those suffering from social phobias (reviewed by Bienvenu et al., 2004). While some these traits may be refractory to clinical intervention, insights into relationships between lower-order personality dimensions and clinically-significant psychological problems may open up new avenues for treatment. Among the more interesting personality traits is perfectionism, because it has been linked to eating, anxiety, depressive, and obsessive-compulsive disorders, in addition to personal self-efficacy and achievement (Stairs, Smith, Zapolski, Combs, & Settles, 2011). To better understand the clinical relevance of perfectionism the findings of several studies will be reviewed here.

A large (N = 731) study examined the prevalence of the big five personality domains among individuals suffering from a range of disorders (Bienvenu et al., 2004). Neuroticism was universally predictive of these disorders (p ? .01), while conscientiousness, a trait similar to perfectionism, was minimally associated. The most common facet of conscientiousness among the disorders was self-discipline, which showed a significant relationship with simple phobia, agoraphobia, generalized anxiety disorder (GAD), and major depressive disorder (MDD) (p ? .002). The conscientiousness facets of competence, striving for achievement, and deliberation, however, were only associated with social phobia. These findings suggest neuroticism and conscientiousness are distinct traits, as Stairs et al. (2011) have concluded.

Stairs and colleagues (2011) identified nine distinct traits among 15 different instruments designed to quantify perfectionism and then combined these into a single instrument called the Measure of Constructs of Perfectionism (M-CUP). Notably, perfectionism was considered a facet of conscientiousness with an internal consistency of .79 (Stairs et al., 2011, p. 148). The facets of conscientiousness from the study by Bienvenu et al. (2004) that seem to match the traits underlying perfectionism, as defined by Stairs et al. (2011), are self-discipline and deliberation (order, details and checking).

Gnilka and colleagues (2012) reviewed early theories discriminating between adaptive and maladaptive perfectionism. The latter representing an inability to experience positive affect because of an interminable discrepancy between their standards and perceived performance. When this definition was applied to college students, a discrepancy subscale correlated significantly with anxiety, self-controlling behaviors, distancing, accepting responsibility, and escape-avoidance behaviors (p < .05). In a separate study, Dittner and colleagues (2011) examined the relationship between negative perfectionism and stress in college students over the course of a semester. Negative perfectionism scores were from a conditional acceptance subscale, which was based on the facets of poor academic performance, ineffective pursuit of goals, higher levels of psychological distress, and low self-esteem. Significant correlations were obtained between negative perfectionism and depression at the beginning of the semester (p < .01), and physical fatigue (p < .05), depression (p < .05), and negative perfectionism (p < .01) at the end of the semester.

The above research findings clearly show a relationship between perfectionism and psychological distress or illness. Featuring prominently are anxiety and related disorders. To build upon this body of work the instrument developed and tested by Stairs et al. (2011) will be used to further investigate the relationship between perfectionism and anxiety, and perfectionism and stress. The hypotheses are as follows:

1. The order subscale will be significantly associated with anxiety, based on the findings of Bienvenu et al. (2004).

2. Given the findings of Dittner et al. (2011), the order subscale will be inversely correlated with stress.

3. The high standards subscale will have a positive, but weak association with anxiety and stress, based on the findings of Bienvenu et al. (2004) and Dittner et al. (2011), respectively.

4. The reactivity to mistakes subscale would probably be classified as maladaptive (Gnilka, Ashby, & Noble, 2012) or negative (Dittner, Rimes, & Thorpe, 2011) perfectionism; therefore, a significant relationship will exist between this subscale and anxiety and stress.

5. The perceived pressure from others subscale does not align well with any of the scales discussed above, because it could be adaptive or maladaptive and is not a facet of conscientiousness as defined by Bienvenu et al. (2004); however, individuals with high scores on this subscale would probably experience low levels of anxiety and little to no stress. Accordingly, perceived pressure from others will be weakly correlated with anxiety, but not stress.

Methods

Participants

Primarily young adults (363 females, 82 males, Mage = 25.81 years, SD = 9.39, range: 18-66 years) participated in the study. Recruitment occurred in the Personality unit or via a link to the survey sent by students in the Personality unit. All participants...

Only four of nine subscales were used: (1) high standards, (2) order, (4) reactivity to mistakes, and (5) perceived pressure from others. The number of items per subscale was 6, 9, 6, and 7, respectively, for a total of 28. A 7-point Likert scale was used, where 1 = "Strongly disagree" and 7 = "Strongly agree"; therefore, the higher the score for a given scale, the greater the confidence that a participant would have that trait.
Anxiety and stress was quantified using the Depression Anxiety and Stress Scale (DASS), Short-Form (Lovibond & Lovibond, 1995). Of the three subscales available, only two were used in the current study: (1) anxiety and (2) stress. Each subscale had 7 items, for a total of 14. A 4-point Likert scale was used, where 0 = "Did not apply to me at all" and 3 = "Applied to me very much, or most of the time." A high score on the anxiety scale would predict increased autonomic arousal, situational anxiety, and subjective experience of anxious affect; whereas, a high score on the stress scale would predict increased chronic, non-specific arousal and worry.

Procedure

The data obtained from the two instruments were analyzed by HPS307/791 staff. Individual scores were calculated for each subscale by summing the scores for each item within a subscale.

Results

The study participants were primarily female (82%) and young adults (N = 445, Mage = 25.81, SD = 9.39), recruited from the Personality unit in the Department of Psychology. As described in the Methods section, all subscales for the M-CUP and DASS were structured so that high scores would predict the existence of that trait in an individual. In general, participants were experiencing low and moderate levels of anxiety (Manxiety = 7.47, SD = 8.36) and stress (Mstress = 16.70, SD = 9.31), respectively (Table 1, top). The subscales for perfectionism were well above midrange, with order (Morder = 46.37, SD = 10.14) and high standards (Mstandards = 32.96, SD = 7.16) having the highest means. The other two subscales, pressure from others (Mpressure = 23.98, SD = 8.29) and reactivity to mistakes (Mmistakes = 30.04, SD = 9.39), were also above midrange.

To determine whether there was a relationship between perfectionism and anxiety or stress, Pearson correlation coefficients were calculated (Table 1, bottom). A significant relationship between anxiety and the two perfectionism subscales of order (r (443) = .090, p > .05) and high standards (r (443) = .033, p > .05) was not found; however, anxiety levels were correlated with pressure from others (r (443) = .223, p < .001) and reactivity to mistakes (r (443) = .415, p < .001), such that higher anxiety levels predicted greater felt pressure from others and reactivity to mistakes. Although the magnitude of the correlation between stress and the two perfectionism subscales of order (r (443) = .135, p < .01) and high standards (r (443) = .199, p < .001) was low, statistical significance was reached. Since stress and anxiety were strongly and positively correlated (r (443) = .537, p < .001), the magnitude and sign of correlation coefficients for each perfectionism subscale were comparatively similar between anxiety and stress. Accordingly, the relationship between stress and pressure from others (r (443) = .319, p < .001), and reactivity to mistakes (r (443) = .465, p < .001), were strong and also significant. Overall, the perfectionism traits of felt pressure from others and reactivity to mistakes were stronger predictors of anxiety and stress.

Table 1.

M-CUP and DASS subscale scores and the relationship between traits of perfectionism and anxiety and stress

M

SD

Possible Sub-Scale Range

Anxiety

7.47

8.36

0 -- 42

Stress

16.70

9.31

0 -- 42

Order

46.37

10.14

7 -- 63

High Standards

32.96

7.16

7 -- 42

Pressure from others

23.98

8.29

7 -- 42

Reactivity to mistakes

30.04

9.39

7 -- 49

Correlations (r)

Anxiety

Stress

Order

.090

.135*

High standards

.033

.199**

Pressure from others

.223**

.319**

Reactivity to mistakes

.415**

.465**

Notes: * = significant at p < .01; ** = significant at p < .001; anxiety and stress were significantly correlated (r = .537, p < .001).

Discussion

The current findings confirm in part the proposed hypotheses. We predicted the order subscale of the M-CUP would have a significant positive relationship with anxiety, but a weak inverse relationship with stress. None of these predictions were supported. Instead order was not associated with anxiety, but was associated with stress. We based the hypothesis in part on the findings of Bienvenu et al. (2004), who found the conscientiousness facets of self-discipline and deliberation were good predictors of anxiety disorders.…