R*vised Synthesis - R*search Skills

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=uecd20

Measurement and Evaluation in Counseling and Development

ISSN: 0748-1756 (Print) 1947-6302 (Online) Journal homepage: https://www.tandfonline.com/loi/uecd20

Doctoral Student Perfectionism and Emotional Well-Being

Randall M. Moate, Philip B. Gnilka, Erin M. West & Kenneth G. Rice

To cite this article: Randall M. Moate, Philip B. Gnilka, Erin M. West & Kenneth G. Rice (2019) Doctoral Student Perfectionism and Emotional Well-Being, Measurement and Evaluation in Counseling and Development, 52:3, 145-155, DOI: 10.1080/07481756.2018.1547619

To link to this article: https://doi.org/10.1080/07481756.2018.1547619

Published online: 01 Apr 2019.

Submit your article to this journal

Article views: 4703

View related articles

View Crossmark data

Citing articles: 11 View citing articles

ASSESSMENT, DEVELOPMENT, AND VALIDATION

Doctoral Student Perfectionism and Emotional Well-Being

Randall M. Moatea, Philip B. Gnilkab, Erin M. Westa, and Kenneth G. Ricec

aUniversity of Texas at Tyler, Tyler, TX, USA; bVirginia Commonwealth University, Richmond, VA, USA; cGeorgia State University, Atlanta, GA, USA

ABSTRACT This study examined the association between perfectionism and perceived stress, life satisfaction, positive emotions, and negative emotions among a national sample of 528 doctoral students. Latent profile analysis based on a measure of perfectionism supported a 3-class model (i.e., adaptive per- fectionists, nonperfectionists, and maladaptive perfectionists). Adaptive per- fectionists had the lowest levels of perceived stress and negative emotions and the highest levels of positive emotions and life satisfaction. Maladaptive perfectionists had the highest levels of perceived stress and negative emotions along with the lowest levels of negative emotions.

KEYWORDS Doctoral students; perfectionism; life satisfaction; stress; emotional well-being

Doctoral students are a unique student population within higher education who face a difficult journey toward completion of their degree (Smith, Maroney, Nelson, Abel, & Abel, 2006). It is widely accepted that the attrition rate for doctoral students in the United States is approximately 50% (Nettles & Millett, 2006). The approximate half of doctoral students who do not complete their degree programs might experience feelings of failure and shame at not meeting their per- sonal goals, in addition to coping with the financial ramifications of only partially completing their degree program (Doran, Kraha, Marks, Ameen, & El-Ghoroury, 2016). The other approxi- mate half of doctoral students who do obtain their doctorate must then contend with the chal- lenge of finding work in competitive job markets, transitioning into professionally demanding positions, and beginning to pay back student loans (National Science Foundation, 2016). Thus, the doctoral student experience is typified by a long and stressful journey where success is not guaranteed and with few assurances of immediate employment and security on completion.

Students must complete a series of intensive academic tasks (e.g., completing requisite course work, passing competency examinations, completing a dissertation) to obtain a doctorate. In add- ition to focusing on their demanding studies, doctoral students must balance other life responsi- bilities, such as family demands, job responsibilities, and managing finances (Smith et al., 2006). Juggling the rigorous demands of a doctoral program with other life responsibilities can cause doctoral students to experience stress and negative emotions, making it challenging for individu- als to maintain a sense of well-being (Stubb, Pyh€alt€o, & Lonka, 2011).

Stress is a frequent experience among doctoral students that can hinder their ability to achieve and maintain a sense of well-being. Stress is experienced by individuals when they perceive an imbalance between the coping resources they have available to meet a demand, and the level of threat that demand poses (Hobfoll, 1989). Considering the intellectual, emotional, and time-inten- sive responsibilities doctoral students juggle, it is not hard to imagine that they often perceive having inadequate resources to cope with their educational and personal demands. Indeed,

CONTACT Randall M. Moate [email protected] Department of Psychology and Counseling, University of Texas at Tyler, 3900 University Blvd., HPR 212, Tyler, TX 75799, USA. � 2018 Association for Assessment and Research in Counseling (AARC)

MEASUREMENT AND EVALUATION IN COUNSELING AND DEVELOPMENT 2019, VOL. 52, NO. 3, 145–155 https://doi.org/10.1080/07481756.2018.1547619

researchers (e.g., Kurtz-Costes, Helmke, & €Ulk€u-Steiner, 2006; Toews, Lockyer, Dobson, & Brownell, 1993) have found that doctoral students experience high levels of stress, exhaustion, and negative emotions. This is concerning given that chronic stress, exhaustion, and negative emotions have been identified as factors detrimental to personal wellness, which can lead to burn- out and abandonment of work and studies (Perepiczka & Balkin, 2010; Stubb et al., 2011).

Although doctoral students experience stress that can reduce their wellness, some are more successful in managing their stress levels and maintaining adequate levels of emotional wellness. Researchers have focused on organizational factors (e.g., student selection process, program struc- ture, community of the program, effective mentorship, and faculty advisors) and other factors (e.g., relationships with significant others, family responsibilities, support systems, employment responsibilities and financial strain, and time constraints) that affect doctoral attrition rates (Smith et al., 2006). Additional researchers have studied important demographic variables such as gender (Kurtz-Costes et al., 2006) and race (Ellis, 2001) as a means of understanding differences in stress and satisfaction among doctoral students. However, few studies have explored how indi- vidual personality characteristics might influence the well-being of doctoral students (Hyun, Quinn, Madon, & Lustig, 2006). This is surprising, as personality characteristics have been found to be an important factor in influencing how individuals perceive experiences such as stress (Ghorpade, Lackritz, & Singh, 2007). Perfectionism is a personality construct that warrants inves- tigation in doctoral students, as it has accounted for differences in multiple outcomes with differ- ent populations such as college professors (e.g., Moate, Gnilka, West, & Bruns, 2016) and undergraduate students (Rice, Lopez, & Richardson, 2013).

Perfectionism

Perfectionism is a personality construct that has been studied across various settings, populations, and facets of functioning (e.g., Gnilka, Ashby, & Noble, 2013; Rice & Ashby, 2007; Stoeber & Otto, 2006). Recent empirical work has supported Parker’s (1997) tripartite model of perfection- ism (Moate et al., 2016; Rice et al., 2013; Wang, Permyakova, & Sheveleva, 2016). Adaptive per- fectionists set and maintain high standards for themselves, yet are able to take pride in their work and practice self-acceptance when falling short of achieving their high standards. Similarly, maladaptive perfectionists also set high personal standards but experience self-criticalness when they are unable to meet their high standards. Nonperfectionists differ from both adaptive and maladaptive perfectionists in that they have lower personal standards (Stoeber & Otto, 2006).

In general, researchers have tended to find positive emotional outcomes linked with adaptive perfectionism and negative emotional outcomes linked with maladaptive perfectionism. Adaptive perfectionism has been associated with greater levels of satisfaction in life (Gnilka et al., 2013; Park & Jeong, 2015), more positive family relationships (DiPrima, Ashby, Gnilka, & Noble, 2011; Martin & Ashby, 2004), and more problem-focused coping strategies (Gnilka, Ashby, & Noble, 2012; Prud’homme et al., 2017). In contrast to the relative benefits of adaptive perfectionism, mal- adaptive perfectionism has been associated with poor emotional well-being, such as increased lev- els of depression (Rice & Ashby, 2007; Wang et al., 2016), burnout (Hill & Curran, 2015; Moate et al., 2016), unhealthy coping processes (Gnilka et al., 2012; Prud’homme et al., 2017), anxiety (Ashby & Bruner, 2005; Gnilka et al., 2012), perceived stress (Ashby et al., 2012; Dunkley, Zuroff, & Blackstein, 2003; Rice, Leever, Christopher, & Porter, 2006), and lower levels of satisfaction with life (Dunkley et al., 2003; Gnilka et al., 2013).

This Study

This study explored, through a large national sample of doctoral students, whether different types of perfectionism (e.g., adaptive, maladaptive, and nonperfectionists) were differentially associated

146 R. M. MOATE ET AL.

with indicators of emotional well-being, such as perceived stress, satisfaction with life, and posi- tive and negative emotions. Perfectionism was used as a measure to perform a latent profile ana- lysis (LPA) of the data, which classified groups of perfectionists and nonperfectionists. Based on the findings from previous researchers (Ashby et al., 2012; Moate et al., 2016; Rice & Ashby, 2007; Wang et al., 2016), we hypothesized a three-class solution (i.e., adaptive, nonperfectionists, maladaptive) would be found. If a three-class model fit the data, we also hypothesized the follow- ing: (a) adaptive perfectionists would experience better emotional well-being in the form of lower levels of perceived stress and negative emotions, and higher levels of satisfaction with life and positive emotions than nonperfectionists and maladaptive perfectionists; (b) nonperfectionists would experience more stress and negative emotions, and lower satisfaction with life and negative emotions than adaptive perfectionists; and (c) maladaptive perfectionists would experience signifi- cantly poorer emotional well-being in the form of higher levels of stress and negative emotions, and significantly lower levels of satisfaction with life and positive emotions than adaptive perfec- tionists and nonperfectionists.

Method

Procedure

Participants were recruited through invitation e-mails sent through various channels (e.g., list- servs, student associations, faculty members, word of mouth). Once participants provided informed consent, they were asked to fill out the demographic form and survey instruments online. Participants received no incentive from the researchers for their participation. University internal review board approval was obtained for this study.

Participants

A total of 528 students completed an online survey. There were 358 women, 167 men, and 3 who declined to provide their gender. Ages ranged from 21 to 59 (M¼ 30.5, SD¼ 8.66). Of those reporting, 76.6% identified as non-Hispanic White (n¼ 405), 4.7% African American (n¼ 25), 3.4% Asian American (n¼ 18), 2.6% Hispanic American (n¼ 14), 1.3% Native American (n¼ 7), 8.9% chose other (n¼ 47), and 2.5% declined to answer (n¼ 13). With regard to relationship sta- tus, 33.6% identified as married (n¼ 178), 21.9% identified as in a relationship (n¼ 116), and 39.1% identified as not in a relationship (n¼ 207). In addition, 90.4% identified as heterosexual (n¼ 478). Participants were pursuing doctorates in a variety of fields, with medicine (18%), social and behavioral sciences (15%), nursing (14%), law (12%), education (12%), and natural and phys- ical sciences (11%), among the most studied areas.

Instruments

The Almost Perfect Scale–Revised (APS–R; Slaney, Rice, Mobley, Trippi, & Ashby, 2001) is a self- report inventory consisting of 23 items and three subscales: High Standards, Order, and Discrepancy. Stoeber and Otto (2006) concluded that the Order subscale did not contribute to the classification of perfectionism and is consistent with previous study designs (e.g., Gnilka et al., 2013; Rice & Ashby, 2007). The APS–R uses a 7-point Likert-type scale where responses range from 1 (strongly disagree) to 7 (strongly agree). The High Standards subscale (7 items) measures high personal standards and performance expectations, and the Discrepancy subscale (12 items) measures the negative reaction experienced when there is an incongruity between an individual’s high standards and actual performance. Concurrent validity and reliability have been supported (see Slaney et al., 2001).

MEASUREMENT AND EVALUATION IN COUNSELING AND DEVELOPMENT 147

The Satisfaction with Life Scale (SWLS; Diener, Emmons, Larsen, & Griffin, 1985) is a 5-item self-report measure that assesses perceived satisfaction with life. Individuals assess this perceived satisfaction using a 7-point Likert-type scale with responses ranging from 1 (strongly disagree) to 7 (strongly agree). The SWLS demonstrates good external validity by holding strong negative cor- relations with other clinical measures of distress. Additionally, the SWLS has demonstrated good convergent validity with other scales measuring perceived well-being. When examined over sev- eral weeks, test–retest reliability was .82, and Cronbach’s coefficient alpha was .87 (Diener et al., 1985).

The Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983) is a self-report inven- tory that measures perceived stress through 14 items. The inventory uses a 5-point Likert-type scale with scores ranging from 0 (never) to 4 (very often). Cohen et al. (1983) demonstrated good reliability including Cronbach coefficient alphas ranging from .84 to .86 and test–retest reliability of .85 (2 days) and .55 (6 weeks) in two different samples (college students and a community sample). Concurrent validity was established with positive associations between the impact of various life stressors (rs¼ .17–.35), social anxiety (r¼ .37), and depression (r¼ .65). The factor structure of the PSS has been demonstrated in previous studies (Martin, Kazarian, & Breiter, 1995).

The Scale of Positive and Negative Experience (SPANE; Diener et al., 2010) is a 12-item self-report inventory designed to assess for subjective feelings of well-being and ill-being using a 5-point Likert-type scale from 1 (very rarely or never) to 5 (very often or always). The SPANE asks participants to report how often they have had different feelings over the past 4 weeks such as “afraid,” “joyful,” and “contented.” It has two subscales: Positive Emotions (SPANE-P) and Negative Emotions (SPANE-N). Diener et al. (2010) noted Cronbach coeffi- cient alphas of .87 for the SPANE-P and .81 for the SPANE-N, suggesting good reliability. Factor analyses have confirmed the arrangement of two subscales and independence from one another. Concurrent validity was demonstrated with other measures of positive and negative emotions (e.g., flourishing, life satisfaction, positive and negative affect).

Due to a clerical error, one of the SPANE-N items was not used for the first 316 participants who participated in the study and was used by the remaining sample (n¼ 213). An average SPANE-N score was calculated based on either five items or the full six items. To determine if there was a significant difference in the total average score between participants who completed five items or six items, a t test was conducted with the SPANE-N as the outcome. No differences were found between participants’ scores who completed five or six items, t(513)¼ 0.10, p> .05. Therefore, the average SPANE-N score was used for the study.

Results

Descriptive statistics, Cronbach alpha coefficients, and bivariate correlations among study scales are reported in Table 1. Cronbach alpha coefficients for the scores ranged from .80 to .95 and were consistent with previous studies (Gnilka et al., 2013; Rice & Ashby, 2007). Discrepancy and

Table 1. Descriptive Statistics and Correlations for Study Subscales.

M SD a 1 2 3 4 5 6 7 8

1. APS–R Standards 42.80 5.12 .80 — 2. APS–R Discrepancy 43.86 17.65 .95 .05 — 3. Perceived Stress 18.20 6.41 .88 .00 .51� –.41� .22� — 4. Life Satisfaction 24.78 6.19 .86 .10� –.39� .56� –.23� –.61� — 5. SPANE-Positive 3.68 0.69 .91 .09� –.41� .48� –.13� –.70� .66� — 6. SPANE-Negative 2.70 0.69 .82 –.01 .41� –.38� .18� .73� –.56� –.70� —

Note. N¼ 528. APS–R¼Almost Perfect Scale–Revised; Perceived Stress¼ Perceived Stress Scale; Life Satisfaction¼ Satisfaction with Life Scale; SPANE¼ Scale of Positive and Negative Experiences.�p < .05.

148 R. M. MOATE ET AL.

standards were essentially orthogonal (r¼ .05). Study correlations were comparable to other research findings (e.g., Gnilka et al., 2013; Rice & Ashby, 2007; Wang et al., 2016).

LPA was conducted with Mplus (Version 7.11; Muth�en & Muth�en, 1998–2013) using a robust maximum likelihood estimator (MLR) and full information maximum likelihood estimation under the assumption that the data are missing at random. For this study, covariance coverage ranged between .94 to 1.00, which allowed for reliable model convergence. Analyses were based on two measured variables from the APS–R (Standards and Discrepancy) as continuous indica- tors of a latent class variable. Five thousand random sets of starting values were used and after 100 iterations, 500 optimizations were used in the final stage. The final stage log-likelihood values were replicated across the optimizations in the final stage, and all models converged on proper solutions.

Model testing was started by setting a single-class, latent profile model, given the chance that no additional class might exist. Next, model comparisons were explored by comparing a k class model to a k – 1 class model. Given we hypothesized that there were three subtypes of perfection- ism, a total of six latent profile models were tested. To determine which model fit the data best, several fit indexes were used. The first of these fit indexes were the Bayesian information criterion (BIC; Schwartz, 1978) and adjusted Bayesian information criterion (aBIC; Sclove, 1987). Relatively smaller numbers of BIC and aBIC indicate a better fit. Entropy was used to determine classification accuracy with higher scores (range from 0–1) indicating better accuracy. The Lo–Mendell–Rubin test (LMR; Lo, Mendell, & Rubin, 2001) and the bootstrap likelihood ratio test (BLRT; McLachlan & Peel, 2000) were also used. Both of these indexes determine statistically significant improvement of fit by adding one more class to the model being tested (e.g., two vs. three; four vs. five). Finally, we considered the size of the smallest class when making a decision. Lubke and Neale (2006) suggested that when adding an additional class of small size (i.e., less than 1.0% or n< 25), specific justification needs to be provided if included.

LPA results appear in Table 2. The BIC and aBIC values decreased for all models. Entropy was higher for the two- and six-class models than for the other models. Comparisons of k versus k – 1 class models noted equivalent findings for the LMR up to the three-class model suggesting a significant improvement over the k – 1 class models. The four- and higher class models resulted in a nonsignificant LMR. The BLRT suggested significant improvement up to the four-class model but ran into convergence issues at the five- and higher class models. Finally, it was noted that the four-class model yielded a class with only 4 participants. It was determined based on the- ory, interpretability, and LMR results that a three-class model best fit the data.

The first class was called adaptive perfectionists with a population estimate of 58.1% and had the lowest levels of discrepancy (M¼ 32.80, SD¼ 11.50) and high levels of high standards (M¼ 45.31, SD¼ 3.41). The second class was called nonperfectionists with a population estimate of 13.1%. Individuals in this group were characterized as having very low levels of high standards (M¼ 34.12, SD¼ 3.41) and moderate levels of discrepancy (M¼ 40.76, SD¼ 11.50). The third class (population estimate of 28.8%) was called maladaptive perfectionists. Individuals in this

Table 2. Fit Indexes and Entropy for One- to Six-Class Models.

Model BIC aBIC LMR

p Value BLRT

p Value Entropy

1-class model 7,777.83 7,765.13 N/A N/A N/A 2-class model 7,685.16 7,662.94 < .0001 < .0001 0.874 3-class model 7,646.13 7,614.39 0.104 < .0001 0.750 4-class model 7,631.06 7,589.80 0.264 < .0001 0.793 5-class model 7,626.82 7,576.03 0.430 < .0001a 0.769 6-class model 7,606.73 7,515.59 0.035 < .0001a 0.817

Note. BIC¼ Bayesian information criterion; aBIC¼ sample size adjusted BIC; LMR¼ Lo–Mendell–Rubin Test; BLRT¼ bootstrap likelihood ratio test.

aConvergence issues.

MEASUREMENT AND EVALUATION IN COUNSELING AND DEVELOPMENT 149

group had very high levels of discrepancy (M¼ 63.08, SD¼ 11.50) and high standards (M¼ 45.43, SD¼ 3.41).

Next, several sets of analyses using the DCON command in Mplus (Lanza, Tan, & Bray, 2013) were conducted to determine whether the means of perceived stress, life satisfaction, positive emotions, and negative emotion differed across the three latent classes (see Table 3). The DCON command provides an omnibus test plus individual comparisons between the different latent classes on each outcome variable separately (see Asparouhov & Muth�en, 2013; Lanza et al., 2013, for a detailed discussion of this approach). All four omnibus tests were significant (p< .001), so specific tests for each variable were conducted. In regard to perceived stress, adaptive perfection- ists had lower scores when compared with nonperfectionists, W¼ v2(1, N¼ 500)¼ –13.23, p< .001, d¼ –.47, and maladaptive perfectionists, W¼ v2(1, N¼ 500)¼ –161.56, p< .001, d¼ –1.27. Nonperfectionists had lower scores when compared to maladaptive perfectionists, W¼ v2(1, N¼ 500)¼ –34.91, p< .001, d¼ –.84.

The same directions of effects held for negative emotions. Adaptive perfectionists had lower scores when compared to nonperfectionists, W¼ v2(1, N¼ 515)¼ –7.39, p< .01, d¼ –.37, and maladaptive perfectionists, W¼ v2(1, N¼ 515)¼ –103.87, p< .001, d¼ –1.03. Nonperfectionists had lower scores when compared to maladaptive perfectionists, W¼ v2(1, N¼ 515)¼ –21.38, p< .001, d¼ –.67.

In regard to life satisfaction, adaptive perfectionists had higher scores when compared to non- perfectionists, W¼ v2(1, N¼ 519)¼ 28.91, p< .001, d¼ .77, and maladaptive perfectionists, W¼ v2(1, N¼ 519)¼ 96.57, p< .001, d¼ 1.02. No significant differences were found between maladaptive perfectionists and nonperfectionists, W¼ v2(1, N¼ 519)¼ 2.30, p> .05, d¼ .22.

In contrast to life satisfaction, significant differences were found for positive emotions between all three classes. Adaptive perfectionists had the highest scores when compared with nonperfec- tionists, W¼ v2(1, N¼ 513)¼ 19.66, p< .001, d¼ .60, and maladaptive perfectionists, W¼ v2(1, N¼ 513)¼ 95.22, p< .001, d¼ 1.23. Nonperfectionists had higher scores when compared to mal- adaptive perfectionists, W¼ v2(1, N¼ 513)¼ 7.54, p< .01, d¼ .67.

Discussion

The APS–R subscale correlations with perceived stress, life satisfaction, positive emotions, and negative emotions were similar to those correlations in past studies of perfectionism (e.g., Rice & Ashby, 2007; Wang et al., 2016). Overall, mean scores of study variables were similar to those of past studies involving undergraduate students with a few exceptions. First, although Standards scores were similar to undergraduate populations (e.g., Gnilka et al., 2013), doctoral students had

Table 3. Latent Profile Means and Standard Deviations for Perceived Stress, Life Satisfaction, Positive Emotions, and Negative Emotions.

Adaptive Perfectionist Nonperfectionist

Maladaptive Perfectionists

(AP; n¼ 307) (NP; n¼ 69) (MP; n¼ 152)

Variable M SD M SD M SD Class Comparisons

High standards 45.31 3.41 34.12 3.41 45.43 3.41 discrepancy 32.80 11.50 40.76 11.50 63.08 11.50 Stress 15.50 5.57 18.06 5.21 22.68 5.75 MP>NP>AP Life satisfaction 27.09 4.85 22.70 6.38 21.29 6.41 AP > (NP¼MP) Positive emotions 3.92 0.61 3.54 0.65 3.27 0.69 AP>NP>MP Negative emotions 2.47 0.61 2.70 0.62 3.13 0.67 MP>NP>AP

Note. High standards¼Almost Perfect Scale Revised (APS–R) High Standards subscale; Discrepancy¼APS–R Discrepancy sub- scale; Stress¼ Perceived Stress Scale; Life Satisfaction¼ Satisfaction with Life Scale; Positive Emotions¼ Scale of Positive and Negative Experiences (SPANE) Positive Experiences subscale; Negative Emotions¼ SPANE Negative Experiences subscale. All class comparisons between the four classes had p values < .05 except between NP and MP on Life Satisfaction.

150 R. M. MOATE ET AL.

slightly higher discrepancy mean scores than studies of undergraduate students (e.g., Rice & Ashby, 2007). One possibility for this finding could be that doctoral students perceive their fellow peers as having high standards and performing well, thus encouraging them to inflate their own standards even if those standards do not match their ability level.

Parker’s (1997) tripartite model of perfectionism (i.e., adaptive perfectionists, nonperfectionists, maladaptive perfectionists) found among doctoral students in this study was similar to findings in previous research on perfectionism (e.g., Moate et al., 2016; Rice et al., 2013; Wang et al., 2016). In addition to classifying the doctoral student sample into three classes, we also attempted to show how these classes are differentially related to various psychological outcomes. A consist- ent pattern of differences between the three groups was found in regard to perceived stress, life satisfaction, positive emotions, and negative emotions. Adaptive perfectionists who demonstrated high levels of standards reported lower levels of perceived stress and negative emotions and greater levels of positive emotions and life satisfaction than nonperfectionists and maladaptive perfectionists. Adaptive perfectionists appeared better able to maintain a sense of well-being than maladaptive or nonperfectionists during their doctoral degree programs. This finding is congruent with earlier studies (e.g., Park & Jeong, 2015; Rice & Ashby, 2007) suggesting that adaptive per- fectionists experience greater levels of satisfaction with life. This group also reported the lowest levels of perceived stress similar to previously conducted studies (Ashby et al., 2012; Moate et al., 2016). Additionally, adaptive perfectionists experienced high levels of positive emotions and the lowest levels of negative emotions. These findings suggest that adaptive perfectionists might thrive amidst the intellectual rigor of doctoral programs.

In regard to adaptive perfectionism, Dunkley, Blankstein, and Berg (2012) suggested a connec- tion with motivation to succeed and conscientiousness might offer some insight into why doctoral students with adaptive perfectionism were able to thrive. Having a high motivation to succeed in a demanding environment might help adaptive perfectionists to find meaning in the process of continually striving for excellence, and to experience this pursuit as pleasurable (Hamachek, 1978). Heightened conscientiousness could help adaptive perfectionists remain vigilant in moni- toring their stress levels and elicit self-care behaviors that reduce minor stressors before they become chronic. Thus, adaptive perfectionists might have natural tendencies that make them more likely to experience well-being during their doctoral programs that maladaptive or nonper- fectionist doctoral students do not possess.

Another important finding of this study is that doctoral students who were maladaptive per- fectionists reported a markedly lower quality of life than adaptive perfectionists and nonperfec- tionists. Consistent with previous research (e.g., Ashby et al., 2012; Dunkley et al., 2003; Gnilka et al., 2012; Rice & Ashby, 2007), maladaptive perfectionists reported more perceived stress, negative emotions, and the least amount of positive emotions compared to the other two classes, although no differences were found between nonperfectionists and maladaptive perfectionists in life satis- faction. Maladaptive perfectionists might experience higher levels of stress due to intense self-crit- icalness when they fail to meet their high personal standards. Due to the long and stressful road of completing a doctoral program, perpetual feelings of self-criticalness could erode maladaptive perfectionists’ capacity to find meaning in what they are doing and experience less satisfaction in their lives. Maladaptive perfectionists appear to be at the greatest risk of experiencing ill-being during their degree programs, which might make completing an already difficult journey even more challenging.

Implications for Professional Practice

This study offered an initial look into how perfectionism could influence the emotional well-being of doctoral students. The outcomes associated with the three-class model (i.e., adaptive, nonper- fectionists, maladaptive) found in this study suggests a relationship between perfectionism and

MEASUREMENT AND EVALUATION IN COUNSELING AND DEVELOPMENT 151

the psychological functioning of doctoral students. In general, doctoral students who were adap- tive perfectionists were the most well off, whereas maladaptive perfectionists were the least well off among the three classes. These findings have implications for counselors and faculty advisors working with doctoral students.

When working with doctoral students, counselors should bear in mind that not all types of perfectionism might be harmful; in fact, adaptive perfectionism seems to be beneficial to this population. Counselors who wish to assess for perfectionism can use the APS–R, a brief 23-item instrument that delineates adaptive and maladaptive perfectionists and nonperfectionists. Accurate assessment of perfectionism might be useful in conceptualizing whether a client is being hampered or benefitted by his or her perfectionistic traits, and in tailoring interventions that are well suited to the needs of adaptive and maladaptive perfectionists.

Despite these relative benefits, adaptive perfectionist doctoral students are not immune to feel- ing stressed, overwhelmed, or burned out during their degree programs. When working with adaptive perfectionists, counselors and faculty advisors might want to used strength-based and solution-focused approaches that use what these individuals are already doing well. One example of this could be harnessing adaptive perfectionists’ natural tendency to enjoy striving for excel- lence in demanding situations (Hamachek, 1978). Instead of allowing clients to perseverate on where they have come up short, counselors can help emphasize a sense of accomplishment and acceptance in trying one’s best while pursuing excellence in a demanding environment, such as a doctoral program.

Faculty advisors and counselors should be aware that maladaptive perfectionists are at higher risk for experiencing emotional distress during their doctoral programs. This group suffers from chronic internal self-criticism about their performance, and might feel like they are constantly failing to measure up to academic and personal goals. When attempting to build rapport with maladaptive perfectionists, it is important to keep in mind that these individuals might have diffi- culty forming strong working alliances, as as suggested by Gnilka, Rice, Ashby, and Moate’s (2016) findings with master’s-level counseling students. Thus, faculty advisors and counselors might want to anticipate these tendencies and make extensive efforts to set reasonable expecta- tions at the beginning of the doctoral program and to convey additional warmth and empathy.

Several studies have investigated treatment modalities for maladaptive perfectionists that might be useful to counselors. One such study by Rice, Neimeyer, and Taylor (2011) suggested that coherence therapy might be helpful in reducing self-criticalness symptoms of maladaptive perfec- tionism. A counselor using coherence therapy could begin by helping his or her clients identify and explore important situations and environments in which intense self-criticalness occurs. Next, the counselor could help clients explore alternative positive emotions that could be con- nected to those environments and situations enhancing clients’ awareness of how their self-crit- icalness impinges on their lives. Finally, as clients become increasingly successful with adopting alternative positive emotions to replace their self-criticalness, their maladaptive perfectionism might gradually subside. Counselors might also want to consider other treatment options such as guided self-help (Pleva & Wade, 2007) and cognitive behavioral therapy (Egan, Wade, Shafran, & Antony, 2014; Radhu, Daskalakis, Arpin-Cribbie, Irvine, & Ritvo, 2012), which have been identi- fied as helpful treatment modalities for individuals with perfectionistic concerns.

Limitations and Future Research

This study has several limitations that should be noted. First, all data were self-report, which could have affected some of the findings. For example, latent profile results might differ from future studies using other multidimensional measures of perfectionism, and outcome implications might not generalize to other work that operationalizes emotional well-being with indicators dif- ferent from those in this study. Second, this study also used a cross-sectional design limiting the

152 R. M. MOATE ET AL.

ability to make predictions; future studies should consider longitudinal designs to allow for pre- dictive inferences on various psychological outcomes. Another future research possibility would be to investigate whether individuals shift from one class to another during their time in graduate school and if the various psychological outcomes change as well. Although this study used a large sample across multiple different doctoral programs in the United States, future studies might want to specifically focus on specific types of doctoral programs. Finally, over three fourths of the sample self-identified as non-Hispanic White. Future researchers should focus on obtaining larger samples of ethnic minorities to replicate and extend these findings. In addition, this study used only one measure of perfectionism (i.e., APS–R) to create classes. Future researchers might want to consider replicating this study using measures from other competing theories of perfectionism such as Frost, Marten, Lahart, and Rosenblate (1990) and Hewitt and Flett (1991).

Conclusion

In summary, we found support for a three-class model of perfectionism within a large national sample of doctoral students. Our findings suggested that adaptive perfectionists were the least stressed, had the fewest negative emotions, and were very satisfied with life while experiencing many positive emotions. The most distressed group were the maladaptive perfectionists, who reported very high levels of stress and negative emotions, little satisfaction with life, and a large amount of negative emotions. Nonperfectionists fell between adaptive and maladaptive perfection- ists in stress, positive emotions, and negative emotions while having comparable life satisfaction with maladaptive perfectionists. These findings provide an important foundation from which to increase the understanding of how perfectionism plays a role in the emotional well-being of doc- toral students.

Notes on Contributors

Randall M. Moate is in the Department of Psychology and Counseling at the University of Texas at Tyler, Tyler, Texas.

Philip B. Gnilka is in the Department of Counselor Education and Special Education at Virginia Commonwealth University, Richmond, Virginia.

Erin M. West is in the Department of Psychology and Counseling at the University of Texas at Tyler, Tyler, Texas.

Kenneth G. Rice is in the Department of Counseling and Psychological Services at Georgia State University, Atlanta, Georgia.

References

Ashby, J. A., & Bruner, L. P. (2005). Multidimensional perfectionism and obsessive compulsive behaviors. Journal of College Counseling, 8(1), 31–40. https://doi.org/10.1002/j.2161-1882.2005.tb00070.x

Ashby, J., Noble, C., & Gnilka, P. (2012). Multidimensional perfectionism, depression, and satisfaction with life: Differences among perfectionists and tests of a stress-mediation model. Journal of College Counseling, 15, 130–143.

Asparouhov, T., & Muth�en, B. (2013). Auxiliary variables in mixture modeling: 3-step approaches using Mplus. Retrieved from http://www.statmodel.com/download/3stepOct28.pdf

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24(4), 385–396.

Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49(1), 71–75.

MEASUREMENT AND EVALUATION IN COUNSELING AND DEVELOPMENT 153

Diener, E., Wirtz, D., Tov, W., Kim-Prieto, C., Choi, D. W., Oishi, S., & Biswas-Diener, R. (2010). New well-being measures: Short scales to assess flourishing and positive and negative feelings. Social Indicators Research, 97(2), 143–156. https://doi.org/10.1007/s11205-009-9493-y

DiPrima, A., Ashby, J., Gnilka, P., & Noble, C. (2011). Family relationships and perfectionism in middle-school students. Psychology in the Schools, 48(8), 815–827.

Doran, J. M., Kraha, A., Marks, L. R., Ameen, E. J., & El-Ghoroury, N. H. (2016). Graduate debt in psychology: A quantitative analysis. Training and Education in Professional Psychology, 10(1), 3–13. https://doi.org/10.1037/ tep0000112

Dunkley, D. M., Blankstein, K. R., & Berg, J. (2012). Perfectionism dimensions and the five-factor model of per- sonality. European Journal of Personality, 26(3), 233–244. https://doi.org/10.1002/per.829

Dunkley, D. M., Zuroff, D. C., & Blankstein, K. R. (2003). Self-critical perfectionism and daily affect: Dispositional and situational influences on stress and coping. Journal of Personality & Social Psychology, 84, 234–252. https:// doi.org/10.1037//0022-3514.84.1.234

Egan, S. J., Wade, T. D., Shafran, R., & Antony, M. M. (2014). Cognitive-behavioral treatment of perfectionism. New York: The Guilford Press.

Ellis, E. M. (2001). The impact of race and gender on graduate school socialization, satisfaction with doctoral study, and commitment to degree completion. Western Journal of Black Studies, 25, 30–45.

Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14, 449–468. https://doi.org/10.1007/BF01172967

Ghorpade, J., Lackritz, J., & Singh, G. (2007). Burnout and personality evidence from academia. Journal of Career Assessment, 15(2), 240–256. https://doi.org/10.1177/1069072706298156

Gnilka, P., Ashby, J., & Noble, C. (2012). Multidimensional perfectionism and anxiety: Differences among individ- uals with perfectionism and tests of a coping-mediation model. Journal of Counseling & Development, 90, 427–436.

Gnilka, P., Ashby, J., & Noble, C. (2013). Adaptive and maladaptive perfectionism as mediators of adult attachment styles and depression, hopelessness, and life satisfaction. Journal of Counseling & Development, 91, 78–86.

Gnilka, P., Rice, K., Ashby, J., & Moate, R. (2016). Adult attachment, multidimensional perfectionism, and the alli- ances among counselor supervisees. Journal of Counseling & Development, 94, 285–296.

Hamachek, D. (1978). Psychodynamics of normal and neurotic perfectionism. Psychology, 15, 27–33. Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and social contexts: Conceptualization, assessment,

and association with psychopathology. Journal of Personality & Social Psychology, 60, 456–470. https://doi.org/ 10.1037//0022-3514.60.3.456

Hill, A. P., & Curran, T. (2015). Multidimensional perfectionism and burnout. Personality and Social Psychology Review, 20, 269–288. https://doi.org/10.1177/1088868315596286

Hobfoll, S. E. (1989). Conservation of resources. A new attempt at conceptualizing stress. The American Psychologist, 44(3), 513–524.

Hyun, J., Quinn, B., Madon, T., & Lustig, S. (2006). Graduate student mental health: Needs assessment and utiliza- tion of counseling services. Journal of College Student Development, 47(3), 247–266. https://doi.org/10.1353/csd. 2006.0030

Kurtz-Costes, B., Helmke, A. L., & €Ulk€u-Steiner, B. (2006). Gender and doctoral studies: The perceptions of PhD students in an American university. Gender & Education, 18(2), 137–155.

Lanza, S. T., Tan, X., & Bray, B. C. (2013). Latent class analysis with distal outcomes: A flexible model-based approach. Structural Equation Modeling, 20(1), 1–26. https://doi.org/10.1080/10705511.2013.742377

Lo, Y., Mendell, N. R., & Rubin, D. B. (2001). Testing the number of components in a normal mixture. Biometrika, 88(3), 767–778. https://doi.org/10.1093/biomet/88.3.767

Lubke, G., & Neale, M. C. (2006). Distinguishing between latent classes and continuous factors: Resolution by maximum likelihood. Multivariate Behavioral Research, 41(4), 499–532. https://doi.org/10.1207/ s15327906mbr4104_4

Martin, J. L., & Ashby, J. S. (2004). Perfectionism and fear of intimacy: Implications for relationships. The Family Journal, 12(4), 368–374. https://doi.org/10.1177/1066480704267279

Martin, R., Kazarian, S., & Breiter, H. (1995). Perceived stress, life events, dysfunctional attitudes, and depression in adolescent psychiatric inpatients. Journal of Psychopathology and Behavioral Assessment, 17, 81–95. https:// doi.org/10.1007/BF02229205

McLachlan, G., & Peel, D. (2000). Finite mixture models. New York: John Wiley & Sons. Moate, R. M., Gnilka, P. B., West, E. M., & Bruns, K. L. (2016). Stress and burnout among counselor educators:

Differences between adaptive perfectionists, maladaptive perfectionists, and nonperfectionists. Journal of Counseling & Development, 94(2), 161–171. https://doi.org/10.1002/jcad.12073

Muth�en, L., & Muth�en, B. (1998–2013). Mplus (Version 7.11). Los Angeles: Muth�en & Muth�en. National Science Foundation. (2016). Survey of earned doctorates: Doctorate recipients from U.S. universities 2014.

Retrieved from http://www.nsf.gov/statistics/srvydoctorates/#sd

154 R. M. MOATE ET AL.

Nettles, M., & Millett, C. (2006). Three magic letters: Getting to Ph.D. Baltimore, MD: John Hopkins University Press.

Park, H., & Jeong, D. Y. (2015). Psychological well-being, life satisfaction, and self-esteem among adaptive perfec- tionists, maladaptive perfectionists, and nonperfectionists. Personality & Individual Differences, 72, 165–170. https://doi.org/10.1016/j.paid.2014.08.031

Parker, W. D. (1997). An empirical typology of perfectionism in academically talented children. American Educational Research Journal, 34(3), 545–562.

Perepiczka, M., & Balkin, R. (2010). Relationship between wellness and age, matriculation, and relationships status of counselor education doctoral students. The Journal of Humanistic Counseling, Education and Development, 49(2), 203–216. https://doi.org/10.1002/j.2161-1939.2010.tb00098.x

Pleva, J., & Wade, T. D. (2007). Guided self-help versus pure self-help for perfectionism: A randomised controlled trial. Behaviour Research and Therapy, 45(5), 849–861.

Prud’homme, J., Dunkely, D. M., Bernier, E., Berg, J., Ghelerter, A., & Starrs, C. J. (2017). Specific perfectionism components predicting daily stress, coping, and negative affect six months and three years later. Personality & Individual Differences, 111, 134–138. https://doi.org/10.1016/j.paid.2017.01.034

Radhu, N., Daskalakis, Z., Arpin-Cribbie, C., Irvine, J., & Ritvo, P. (2012). Evaluating a web-based cognitive-behav- ioral therapy for maladaptive perfectionism in university students. Journal of American College Health, 60(5), 357–366. https://doi.org/10.1080/07448481.2011.630703

Rice, K., & Ashby, J. (2007). An efficient method for classifying perfectionists. Journal of Counseling Psychology, 54(1), 72–85. https://doi.org/10.1037/0022-0167.54.1.72

Rice, K. G., Leever, B. A., Christopher, J., & Porter, J. D. (2006). Perfectionism, stress, and social (dis)connection: A short-term study of hopelessness, depression, academic adjustment among honor students. Journal of Counseling Psychology, 53, 524–534. https://doi.org/10.1037/0022-0167.53.4.524

Rice, K. G., Lopez, F. G., & Richardson, C. M. (2013). Perfectionism and performance among STEM students. Journal of Vocational Behavior, 82(2), 124–134.

Rice, K. G., Neimeyer, G. J., & Taylor, J. M. (2011). Efficacy of coherence therapy in the treatment of procrastin- ation and perfectionism. Counseling Outcome Research and Evaluation, 2, 126–136. https://doi.org/10.1177/ 2150137811417975

Schwartz, G. (1978). Estimating the dimension of a model. Annals of Statistics, 6, 461–464. Sclove, S. L. (1987). Application of model-selection criteria to some problems in multivariate analysis.

Psychometrika, 52(3), 333–343. https://doi.org/10.1007/BF02294360 Slaney, R., Rice, K., Mobley, M., Trippi, J., & Ashby, J. (2001). The Revised Almost Perfect Scale. Measurement &

Evaluation in Counseling and Development, 34, 130–145. Smith, R., Maroney, K., Nelson, K., Abel, A., & Abel, H. (2006). Doctoral programs: Changing high rates of attri-

tion. The Journal of Humanistic Counseling, Education and Development, 45, 17–31. https://doi.org/10.1002/j. 2161-1939.2006.tb00002.x

Stoeber, J., & Otto, K. (2006). Positive conceptions of perfectionism: Approaches, evidence, challenges. Personality and Social Psychology Review, 10(4), 295–319. https://doi.org/10.1207/s15327957pspr1004_2

Stubb, J., Pyh€alt€o, K., & Lonka, K. (2011). Balancing between inspiration and exhaustion: PhD students’ experi- enced socio-psychological well-being. Studies in Continuing Education, 33(1), 33–50. https://doi.org/10.1080/ 0158037X.2010.515572

Toews, J. A., Lockyer, J. M., Dobson, D. J., & Brownell, A. K. (1993). Stress among residents, medical students, and graduate science (Msc/PhD) students. Academic Medicine, 68, 46–48.

Wang, K. T., Permyakova, T. M., & Sheveleva, M. S. (2016). Assessing perfectionism in Russia: Classifying perfec- tionists with the short almost perfect scale. Personality & Individual Differences, 92, 174–179. https://doi.org/10. 1016/j.paid.2015.12.044

MEASUREMENT AND EVALUATION IN COUNSELING AND DEVELOPMENT 155