Perfectionism is characterized by the setting of unreasonably high standards and the resulting self-criticism when those standards are not reached (Burns 1980). In early psychological theories of personality and psychopathology, perfectionism was described as obsessional neurosis (Freud 1959), and characterized by exacting standards, meticulous living, and unfulfilled expectations (Horney 1950). The construct was also characterized as irrational and dysfunctional, highlighted by the catastrophic nature of failing to meet unrealistic standards (Ellis 1962), and the all-or-nothing attitude in which falling short of expectations is interpreted as a failure (Beck 1976).

Perfectionism has been associated with a variety of negative outcomes including procrastination (Flett et al. 1992), stress (Hewitt and Flett 2002), shame and guilt (Fedewa et al. 2005), low self-esteem (Ashby and Rice 2002), and interpersonal problems (Flett et al. 2001). It has been associated with more severe psychopathology including eating disorders (Goldner et al. 2002), obsessive-compulsive disorder (Frost and Steketee 1997), depression (Hewitt and Flett 1991a), social anxiety (Alden et al. 2002), and suicide (Hewitt et al. 1994). It has also been associated with environments characterized by high standards, including religion or spirituality (Ashby and Huffman 1999).

Several approaches to measuring perfectionism have emerged, and most have utilized multiple dimensions to conceptualize the construct. An early approach addressed the intrapersonal dimensions of perfectionism including excessively high standards, the level of concern over mistakes, a sense of doubt about the quality of one’s performance, concern over parent’s expectations and evaluations, an overemphasis on precision, order, and organization (Frost et al. 1990).

Another early approach addresses the interpersonal dimensions of perfectionism including self-oriented perfectionism, other-oriented perfectionism, and socially prescribed perfectionism (Hewitt and Flett 1991b). Other approaches have made a distinction between the adaptive and maladaptive dimensions of perfectionism. The Dual Process Model of Perfectionism made a distinction between the pursuit of success, excellence, approval, and satisfaction; and the avoidance of failure, mediocrity, disapproval, and dissatisfaction (Slade and Owens 1998). The Almost Perfect Scale (APS; Slaney et al. 2002) identified three aspects of perfectionism (high standards, order, and discrepancy) that are associated with adaptive and maladaptive dimensions of perfectionism.

Religiosity

Responding to a distressing relationship between religiosity and prejudicial attitudes, Allport developed a theory and accompanying scale for measuring intrinsic and extrinsic religious orientation to shed some light on religion and prejudice (Allport and Ross 1967). According to Allport’s view, individuals with an intrinsic orientation are characterized by the internalization of their religious beliefs, which serve as their primary motivation. Individuals with an extrinsic orientation are characterized by their use of religion for selfish and utilitarian ends such as security, solace, socialization, distraction, or status. Individuals who displayed both intrinsic and extrinsic qualities were described as indiscriminately proreligious, and individuals who displayed neither intrinsic nor extrinsic qualities were described as indiscriminately antireligious or nonreligious.

It was thought that religious individuals with an extrinsic orientation would adopt the authoritarian and ethnocentric attitudes often associated with dogmatic religious traditions while religious individuals with an intrinsic orientation would reject these negative attitudes and behaviors. This was confirmed in the investigation with some additional findings that found that indiscriminately pro-religious individuals were characterized by a consistent response style which was attributed to an “undifferentiated cognitive disposition” (pp. 441–442).

For example, the religion as a whole is good while a minority group as a whole is bad, and it was discovered that these individuals were the most prejudiced of all. This was attributed to the tendency to overgeneralize and stereotype, indicating that they were relatively inflexible in their response to religious teachings, practices, and the need to make fine distinctions in a complex environment. Thus, a portion of the pathological nature of religion was attributable to individual differences and a rigid response to the religious standards.

Perfectionism and Religiosity

Two studies have examined the specific relationship between perfectionism and religiosity, and help clarify the nature of the relationship by taking the subdimensions of the constructs into account. Citing the high standards and expectations as the common denominator between the two constructs, perfectionism was compared to religiosity in 242 undergraduate college students (Ashby and Huffman 1999). The revised edition of the APS was used to measure both adaptive and maladaptive dimensions of perfectionism and religiosity was measured by asking about the level of religious activity and an intrinsic-extrinsic religiosity measure. Increased religious activity and intrinsic religiosity were positively associated with the adaptive dimensions of perfectionism (high standards and order). Maladaptive perfectionism was not related to intrinsic or extrinsic religiosity.

Similarly, in a survey of 500 Christian college students, a significant positive correlation was found between maladaptive perfectionism and problematic spiritual functioning characterized by instability and disappointment with one’s relationship with God (Thelander 2002). Significant small negative correlations were also found between maladaptive perfectionism and mature aspects of spiritual functioning (awareness of God and acceptance of spiritual disappointments). Taken together, these two studies suggest that adaptive perfectionism can be associated with adaptive religiosity and maladaptive perfectionism can be associated with maladaptive religiosity.

The connection between perfectionism and religiosity seems likely because of their shared high standards. When the constructs are compared using adaptive and maladaptive subdimensions, the relationship is clarified further. It appears, then, that the relationship between perfectionism and religiosity can best be described by accounting for the subdimensions of the construct. Specifically, it appears that the adaptive and maladaptive distinction in both religiosity and perfectionism helps explain how religious standards in well-meaning individuals can be associated with negative outcomes (maladaptive perfectionism). This leads to the primary research question at hand in this investigation which looks at why this established relationship exists using a mediational variable.

Psychological Inflexibility

The idea of a mediational variable that could explain why this relationship exists emerged from research in which a rigid and inflexible personality style was regularly implicated in the literature examining both perfectionism and religiosity. Psychological inflexibility refers to a rigid and inflexible style of responding to unwanted private experiences (thoughts, feelings, bodily sensations). It involves an unwillingness to experience difficult thoughts, emotions, or physical sensations and often results in maladaptive efforts to avoid these experiences (Abramowitz et al. 2009; Hayes et al. 2006). Psychological inflexibility has been found to be associated with poor overall health, depression, negative affect, anxiety, social phobia, posttraumatic stress disorder, self-harm, and alcohol dependence (Hayes et al. 2006). Psychological inflexibility has also been proposed as the mechanism by which some individuals develop anxiety disorders in the context of normative fears, such as a traumatic incident, while others do not (Olatunji et al. 2007).

Conclusion

Research in perfectionism and religiosity has identified the importance of accounting for adaptive and maladaptive dimensions of the constructs. When the adaptive and maladaptive dimensions are accounted for, it appears that adaptive perfectionism is related to adaptive religiosity and maladaptive perfectionism is related to maladaptive religiosity. The purpose of this investigation is to build on the existing knowledge of an established relationship between perfectionism and religiosity by looking closely at the subdimensions of the constructs and then examining how those relationships are mediated by psychological inflexibility. This study attempts to answer the following questions: (1) What is the relationship between the adaptive and maladaptive dimensions of perfectionism and the adaptive and maladaptive dimensions of religiosity?, (2) What is the relationship between psychological inflexibility and the adaptive and maladaptive dimensions of perfectionism and religiosity?, and (3) If the established relationships from the literature hold in this investigation, does psychological inflexibility mediate those relationships?

Method

Procedures

Participants were recruited from undergraduate general/introductory psychology classes and the general campus. The recruitment information provided a link to an anonymous online survey. The survey took approximately 30 minutes to complete. Measures were administered in random order for each participant to counterbalance for the effects of the order of administration.

Participants

Undergraduate college students, over the age of 18, from an urban university in the western United States were asked to participate. In total, 421 individuals accessed the online survey, but 45 participants did not complete all of the questions. Inspection of the data did not reveal any systematic reasons for the incomplete data, and it was assumed that the incomplete cases were randomly distributed throughout the data set. These cases were removed from the data set resulting in 376 total participants that were included in the analyses. Complete participant characteristics are provided in Table 1.

Table 1 Participant characteristics
Full size table

Measures

Demographics

The demographics questionnaire included questions about sex, age, marital status, education, ethnicity/race, current religious affiliation, any past religious affiliation, the importance of religion, and religious activity.

Perfectionism

Information about perfectionism was obtained using the APS-R (Slaney et al. 2001). The APS-R takes into account the body of research that has identified both positive and negative dimensions of perfectionism and provides specific measures of the defining features of perfectionism (high standards and extreme self-criticism). It is made up of 23 items that are responded to on a Likert scale with 7 scale points ranging from “strongly disagree” to “strongly agree” (Mobley et al. 2005). The APS-R contains three subscales: (a) high standards (7 items), (b) discrepancy (12 items), and (c) order (four items).

The high standards and order subscales are associated with the aspects of positive perfectionism and the discrepancy subscale is associated with the aspects of negative perfectionism. The subscales can be used to distinguish between adaptive and maladaptive perfectionists with adaptive perfectionists scoring high on high standards and low on discrepancy and maladaptive perfectionists scoring high on both high standards and discrepancy (Ashby and Kottman 1996). Internal consistency of the APS-R has ranged from .85 to .92. Additionally, the APS-R has been shown to correlate with the expected outcomes of perfectionism including depression, self-esteem, and GPA (Slaney et al. 2002). Cronbach’s alpha for the current sample was .87 for high standards, .88 for order, and .94 for discrepancy. The APS-R was chosen because it covers both adaptive and maladaptive dimensions of perfectionism in a clear manner. The APS-R also has a rigorous body of psychometric support and has been widely used with the population of interest in this investigation.

Religiosity

Information about religiosity was gathered using the Age Universal Religious Orientation Scale (AUROS; Gorsuch and Venable 1983). This 20-item measure contains two subscales designed to measure intrinsic and extrinsic (I-E) religious orientation. It is a modified version of the I-E Religious Orientation Scale (Allport and Ross 1967) updated for use with children and adults. While the idea of religious orientation purports to measures religious orientation or motivation, it has been widely used as a general indicator of overall religiosity or religiousness. The AUROS is completely interchangeable with the original scale. Nineteen of the 20 items are scored on a 5-point Likert scale with 1 indicating “strong disagreement” and 5 indicating “strong agreement.” The other item is a measure of church attendance where 1 indicates “a few times a year” and 5 indicates “more than once a week.”

Internal consistency ranged from .66 to .73. Cronbach’s alpha for the current sample was .93 for intrinsic religiosity and .69 for extrinsic religiosity. The AUROS was chosen as a measure of religiosity because it provides both a general indication of religiosity and differentiates between two types of religiosity: intrinsic religiosity characterized by the internalization of religious beliefs that serve as primary motivation; and extrinsic religiosity characterized by the use of religion for selfish and utilitarian ends such as security, solace, socialization, distraction, or status. It was hoped that these subtypes of religiosity would provide further understanding of the individual difference apparent in the literature examining the relationship between perfectionism and religiosity. The age universal version was chosen because it used language that was accessible and clear for a diverse group of religious affiliations.

Psychological Inflexibility

Information about psychological inflexibility was gathered using the Acceptance and Action Questionnaire–II (AAQ-II; Bond et al. 2009). The AAQ-II is a revised version of the original Acceptance and Action Questionnaire designed to measure the unidimensional construct of psychological flexibility, as described in Acceptance and Commitment Therapy (ACT). The ACT theory focuses on positive attributes so higher scores indicate higher flexibility. In this study, the variable of interest is the inflexibility of the participants, so lower scores will indicate greater inflexibility. The scale is comprised of 10 items that yield a single factor solution (psychological flexibility). Normative research with over 3,000 participants resulted in internal consistency ranging from .76 to .87 and test-retest reliability ranging from .78 to .80. The AAQ-II has also demonstrated concurrent validity with theoretically similar scales as demonstrated by negative associations with depression, anxiety, stress, and overall psychological distress. Cronbach’s alpha for the current sample was .85. The AAQ-II was chosen because it has proven to be a good measure of overall psychological inflexibility. This construct was implicated in the perfectionism and religiosity literature, but as a byproduct of other analyses and it was never directly measured. The AAQ-II is unique as it provides a direct measure of psychological inflexibility.

Results

Table 2 is a summary of the descriptive statistics for the measures used in this investigation. Visual inspection of the score distributions indicated that each measure approximated the normal distribution and skewness statistics are also reported. Correlations were calculated between internal and external religious orientation and the three dimensions of perfectionism (high standards, order, and discrepancy). Two significant relationships emerged. There was a small positive correlation between internal religious orientation and high standards perfectionism (r = .26, p < .01, two-tailed), and a small positive correlation between external religious orientation and discrepancy perfectionism (r = .13, p < .05, two-tailed).

Table 2 Descriptive statistics for measures of perfectionism, religious orientation, and psychological inflexibility
Full size table

Correlations were also calculated between psychological inflexibility and the dimensions of perfectionism and religiosity. Psychological inflexibility was measured by the AAQ-II, and lower scores reflect higher inflexibility, so a negative correlation would indicate a positive association between the two variables. There was a significant negative correlation between psychological inflexibility and external religious orientation (r = −.20, p < .01, two-tailed), and a negative correlation between psychological inflexibility and discrepancy perfectionism (r = −.54, p < .01, two-tailed).

The test for mediation as proposed by Baron and Kenny (1986) includes three steps in which four conditions must be met. This is not a stepwise or hierarchical process, and each regression analysis is performed separately. Six tests of mediation were performed to examine the mediating role played by psychological inflexibility between religious orientation and perfectionism. Internal religious orientation was the independent variable with a separate test for each dimension of perfectionism (high standards, order, discrepancy) as the dependent variable, and external religious orientation was the independent variable with a separate test for each dimension of perfectionism as the dependent variable.

Only one test was significant, psychological inflexibility mediated the relationship between an external religious orientation and the discrepancy dimension of perfectionism (see Table 3). In step one, a significant effect was found between the independent variable, external religious orientation, and the mediator, psychological inflexibility. As already seen in the correlational analyses, higher levels of external religious orientation predict higher levels of psychological inflexibility. In step two, the independent variable, external religious orientation, predicted discrepancy perfectionism, β = .13, t(374) = 2.43, p < .05. Psychological inflexibility also explained a significant portion of variance in discrepancy perfectionism, R 2 = .02, F(1, 374) = 5.897, p < .05. In step three, the mediator, psychological inflexibility, had a significant effect on the dependent variable, discrepancy perfectionism, β = −.54, t(373) = −12.14, p < .001, and the effect of the independent variable, external religious orientation, on the dependent variable, discrepancy perfectionism, was diminished and no longer significant, β = .02, t(373) = .34, p > .05.

Table 3 Test for mediation: psychological inflexibility as a mediator between external religious orientation and discrepancy perfectionism
Full size table

This successful test of mediation supports the hypothesis that psychological inflexibility helps account for the relationship between external religious orientation and discrepancy, or maladaptive, perfectionism. It has already been noted there is a significant correlation between external religious orientation and discrepancy perfectionism. In this correlation, higher levels of external orientation predict higher levels of discrepancy perfectionism. Successful mediation can explain why this correlation exists, in this case because of higher levels of psychological inflexibility.

Discussion

A review of the literature suggested the relationship between perfectionism and religiosity is further clarified when the subdimensions of the constructs are accounted for, and the results of this study were consistent with this idea. Intrinsic or internal religiosity, the adaptive dimension of religiosity, was positively related to the adaptive dimension of perfectionism, high standards. This is consistent with previous research (Ashby and Huffman 1999).

Extrinsic or external religiosity, the maladaptive dimension of religiosity, was positively related to the maladaptive dimension of perfectionism, discrepancy. This is also consistent with previous research that found a relationship between maladaptive perfectionism and problematic spiritual functioning (Thelander 2002). As suggested in the literature, considering the adaptive/maladaptive dimensions of these two constructs helps understand the nature of the relationship.

These results were complemented by the relationships that emerged between psychological inflexibility, a maladaptive construct, and the maladaptive dimensions of religiosity (external) and perfectionism (discrepancy). To understand why this relationship might exist, tests of mediation were performed with the only significant result emerging with psychological inflexibility as a mediator between the maladaptive dimensions of religiosity and perfectionism, external religiosity and discrepancy perfectionism. This suggested that individuals with an external religious orientation were more likely to display increased levels of maladaptive perfectionism because of their psychological inflexibility.

Religiosity is typically associated with high standards, and these results would suggest that intrinsic religiosity is characterized by high standards that have been internalized and are a source of motivation for religious behavior. Certainly, any time an individual sets high standards, there will be a discrepancy between the standards and the actual performance. For individuals with an extrinsic religious orientation, this discrepancy between standards and performance appears to be unacceptable and results in maladaptive perfectionism.

This result is consistent with literature that suggests a rigid and inflexible style of interaction with high standards. This suggests that psychological inflexibility can account for the relationship between the two variables, that when individuals with an external religious orientation adopt the high standards of a religion, their level of psychological inflexibility may predict the development of maladaptive perfectionism. The presence of maladaptive perfectionism is an indicator that the religious activity may be serving an important role or purpose in that individual’s life such as security or distraction, and the unmet standards may detract from that goal. It is also possible that individuals with an extrinsic religious orientation may only engage portions of the religious doctrine and not fully understand the role of high standards in religion in that the commandment to be “perfect” could be a gradual and cumulative process.

The role of psychological inflexibility as a mediator raises an interesting question of whether or not the inflexibility is a global cause of the extrinsic religious orientation and the maladaptive perfectionism, or if it is possible to have some individual adopt an extrinsic religious orientation and still interact flexibly with the religious standards. This question was not answered directly in this investigation, but the small-to-moderate correlations are suggestive that there are also other variables that could explain the relationship. Additionally, the idea of psychological inflexibility is not necessarily a global personality characteristic. For example, individuals with obsessive-compulsive disorder may demonstrate severe inflexibility in their response to contamination obsessions while at the same time they are able to respond flexibly to comorbid conditions or other difficult life events.

The results of this investigation are important because they demonstrate the need to account for the adaptive/maladaptive dimensions and clarify the relationship between perfectionism and religiosity, specifically that adaptive religiosity is related to adaptive perfectionism and maladaptive religiosity is related to maladaptive perfectionism. Both constructs have subdimensions leading to more complex and subtle relationships between the variables.

It is therefore important to consider the nature of an individual’s religious orientation and dimensions of perfectionism before drawing any conclusions about possible psychopathology. Theoretically, this is similar to the idea of performance enhancing anxiety versus anxiety disorders, or appropriate sadness and grieving versus clinical depression. Emotions or thoughts that are typically evaluated as “negative” (anxiety, sadness) can lead to negative outcomes, but it is not a definitive relationship. For example, this study has shown that high standards are not inherently bad, but when combined with psychological inflexibility, they can functionally lead to maladaptive perfectionism. In applied work, identifying the functional role of personality in an individual’s presentation would be crucial to an accurate understanding of diagnosis and treatment.

Perfectionism, particularly maladaptive perfectionism when this has been accounted for in the research, has been linked to numerous problematic outcomes and psychopathology. The results of this investigation suggest that the relationship between maladaptive religiosity and maladaptive perfectionism can be linked to similar outcomes. This has important implications in the understanding of religiosity and its possible negative effects. In particular, a major focus in the religiosity literature has been to look for possible negative effects of religiosity on mental health.

Religiosity has been associated with both positive and negative psychological outcomes (Wulff 1996). Recognition of the role of the adaptive and maladaptive dimensions clarifies the inconsistency in the research and also points to the idea of individual differences in how one interacts with a religion. This points to the need for a mediator to explain why that relationship exists, which in this case, is individual differences in psychological inflexibility.

The identification of psychological inflexibility as a mediator between external religiosity and maladaptive perfectionism is an important finding of this investigation. In addition to empirical evidence for a relationship between perfectionism and religiosity using the adaptive and maladaptive dimensions, a possible causal factor for that relationship has been suggested. This provides empirical support for the conclusion from the literature that individual differences would likely account for this relationship, in this case individual differences in psychological inflexibility.

This also counters claims that the high standards of religion can lead to psychopathology, instead, it is the level of flexibility with which an individual interacts with those high standards. Not only should further theoretical work be sure to include analysis of the subdimensions, but mediating variables should be examined as well. There is limited work in the etiology of perfectionism, and this mediational study has provided insight into the cause of maladaptive perfectionism. The mediational findings may also have important implications for research investigating the relationships between specific environments and any negative outcomes. This is especially true for perfectionism as it has been examined in several different contexts including sports and exercise, school environments, intimate relationships, and professional achievement.

The mediational findings may be most valuable in an applied setting. Clinicians can point to an underlying cause for pathological perfectionism in a religious population. This provides invaluable understanding in the conceptualization of a problem, and provides specific guidance on how to intervene. To address maladaptive perfectionism in a religious population, this study would suggest that increasing psychological flexibility would be the appropriate intervention.

Indeed, the construct of psychological inflexibility as defined in this study, is a part of ACT, which is a psychosocial intervention designed to target rigid and inflexible interactions with internal events (thoughts, emotions, and physical sensations) (Hayes et al. 2006). ACT has been proven effective at treating several psychological disorders that are characterized by this overall psychological rigidity including anxiety, depression, substance abuse, obsessive-compulsive disorder, and chronic pain (Hayes et al. 2006). The results of this study are thus complementary with the empirical research on ACT. The ACT research would suggest that psychological inflexibility is an important variable in many types of psychopathology, and this investigation confirmed its role as a mediator in a specific psychopathology, maladaptive perfectionism.

Limitations and Future Directions

While the results of this investigation are significant, it is important to acknowledge the limitations of this research. This study was conducted with a sample of convenience from a college student population. Thus, the results may not be applicable to the general population. Additionally, the majority of the respondents in this study endorsed “Latter-day Saint” (84.3%) for religious affiliation. As such, these results may need to be replicated with other denominations or more heterogeneous religious samples to demonstrate relevance with the general religious population.

However, these findings are consistent with previous research that found relationships between intrinsic religiosity and adaptive perfectionism (Ashby and Huffman 1999). This study used the same measures of religiosity and perfectionism in a sample of 242 undergraduate college students in the Midwest, with similar results. The consistency of the results in these two studies suggests that the findings of this investigation may be generalized to more heterogeneous religious groups. The consistency of results of this study are encouraging, but it would be important to replicate these findings with diverse cultural and age groups.

It would also be important to replicate the mediational findings, as these are the first of their kind to be demonstrated empirically with perfectionism. Additionally, this was a correlational design. The test of mediation does allow for the implication of causation, but the results would be more robust if replicated in a controlled experimental design. A controlled experimental design to test for the effects of psychological inflexibility on maladaptive perfectionism would lead to more robust results and encourage additional work in this line of research.

The theoretical implications of this research suggests continued examination of the role of psychological flexibility in the relationship between religiosity and mental health, and the relationship between perfectionism and environments characterized by high standards. This investigation also informs further research into the treatment of maladaptive perfectionism. Treatments that address this problem, especially in religious populations, should take into account the role of psychological inflexibility in the development of pathology. Treatment research of perfectionism in general would be merited, as well as treatment research that draws on the findings of this investigation. Specifically, the findings of this investigation would suggest the application of ACT for the treatment of maladaptive perfectionism. If an intervention designed to target psychological inflexibility was successful at improving flexibility and reducing maladaptive perfectionism, this would provide additional experimental support for the mediational findings of this investigation.