کمال گرایی خوب، بد، و یا هر دو؟ مدل های بررسی ساختار کمال گرایی

Perfectionism has been conceptualized as a personality variable that underlies a variety of psychological difficulties. Recently, however, theorists and researchers have begun to distinguish between two distinct types of perfectionism, one a maladaptive form that results in emotional distress, and a second form that is relatively benign, perhaps even adaptive. In this study, we compared varying models of the perfectionism construct using the best known measures of perfectionism. In a sample of 198 students, three competing models of perfectionism were examined using confirmatory factor analysis. A model that incorporated two factors, one corresponding to maladaptive perfectionism and the other adaptive perfectionism, was a better fit to the data than a unitary perfectionism model. We also examined the relations of the two types of perfectionism, Maladaptive Evaluative Concerns and Positive Striving, to a well known measure of psychological distress. Maladaptive Evaluative Concerns was more strongly associated with depression, anxiety, stress, and test taking anxiety. Overall, this study supports the validity of a distinction between two types of perfectionism, and points to the importance of this duality for measurement and research on perfectionism.

Theoretical and research interest in the personality construct of perfectionism has grown markedly over the last decade (see Shafran & Mansell, 2001 for a review). There has been a particular focus on the negative correlates and consequence of perfectionism, including concurrent psychopathology and the perfectionist's future vulnerability to distress (Antony et al., 1998, Hewitt and Flett, 1991b, Hewitt and Flett, 1993, Hewitt et al., 1996 and Hewitt et al., 1991). Moreover, clinical writers suggest that perfectionism leads to a chronic sense of failure, indecisiveness, procrastination, and shame (Burns, 1980, Hamachek, 1978, Hollender, 1965 and Pacht, 1984). An as yet unresolved issue is whether or not perfectionism always, and uniformly, leads to difficulties for the individual. Certainly, the deleterious consequences of perfectionism have been emphasized since the construct was first examined carefully (Frost et al., 1990, Hamachek, 1978, Hewitt and Flett, 1991a, Hewitt and Flett, 1991b, Hollender, 1965 and Pacht, 1984). However, dissenting views and recent research suggest that perfectionism may contain both adaptive and maladaptive features (e.g., Bieling et al., 2003 and Enns et al., 2001) and, interestingly, there is additional recent evidence from a psychometric and theoretical standpoint that two forms of perfectionism exist (Cox et al., 2002, Frost et al., 1993, Hill et al., 1997 and Slade and Owens, 1998). A distinction between adaptive or “healthy” perfectionism and maladaptive “unhealthy” perfectionism is important because it raises a number of conceptual and pragmatic issues about the construct itself. These include the most fundamental definition of perfectionism and construction of appropriate measurement tools that capture the underlying construct(s). The two most common measures of perfectionism were derived from a clinical perspective and view perfectionism as a personality feature that is problematic and in need of modification. These two scales, both called the Multi-dimensional Perfectionism Scale were developed over a decade ago by two separate teams of researchers (Frost et al., 1990 and Hewitt and Flett, 1991a). Paul Hewitt and his colleagues view perfectionism as a multidimensional construct with intrapersonal and interpersonal content. Their scale has three dimensions; self-oriented perfectionism (setting of excessive personal standards and stringently evaluating one's behavior), other oriented-perfectionism (expectations of perfection that one has for others) and socially prescribed perfectionism (a perception that other people expect one to be perfect and that others are harsh, punitive judges). Frost and his colleagues also focused on high, rigid standards in their measure but went on to suggest that a critical component of perfectionism is a tendency to be critical of one's own behavior (Frost et al., 1990). In addition the Frost Scale also assesses the origin of perfectionism, parental expectations and parental criticism. These two perfectionism scales have been associated with a wide variety of psychopathology (Shafran & Mansell, 2001) as well as negative behavioral consequences (e.g., Alden et al., 1994, Bieling and Alden, 1997 and Bieling et al., 2003). However, unlike the developers of these scales, earlier theorists such as Hamachek (1978) argued that some aspects of perfectionism might be adaptive since they foster excellence and striving to meet important goals. Hamachek termed this more adaptive form of perfectionism “normal perfectionism” and termed the maladaptive form “neurotic perfectionism”. It is also noteworthy that in the larger culture outside of clinical and personality psychology, perfectionism is often tolerated, perhaps encouraged, due to the perception that perfection is associated with important rewards in domains such as sports, business, science, and academics. An argument that might explain the lay perception of perfectionism has been advanced by Slade and Owens (1998) who distinguish conceptually between a normal/healthy form of perfectionism and a pathological form. Based on a behavioral model, the authors suggest that healthy perfectionism is related to positive reinforcement. In other words, perfectionism that leads to achievement of high standards and, ultimately, rewards for achieving those standards may not lead to psychological distress. In contrast, maladaptive perfectionism is seen to be related to negative reinforcement and self-defeating behaviors (Slade & Owens, 1998). For instance, being concerned about how others evaluate the self, self-doubt, and worry about making mistakes may lead to problematic behaviors (i.e., task avoidance, excessive checking, or behaviors aimed at regaining self-worth) that reduce negative outcomes (Slade & Owens, 1998). This reconceptualization of perfectionism into adaptive and maladaptive components can be partially tested by carefully examining existing measures of perfectionism for both positive and negative components. Although few studies have explicitly set out to validate the distinction between positive or adaptive perfectionism and negative or maladaptive perfectionism there is some evidence for this conceptual distinction. One of the first studies to shed light on this issue was performed by Frost and his colleagues who were interested in the relationships between the two multi-dimensional perfectionism measures described earlier (Frost et al., 1993). These researchers found that some subscales of the two perfectionism scales were positively correlated with positive affect and not correlated with negative affect. Moreover, factor analysis of the nine subscales of the Frost and Hewitt measures revealed a two-factor solution with five subscales loading on one factor, and four subscales loading on a second factor. The authors termed the first factor Maladaptive Evaluative Concerns and the second factor Positive Striving. The authors concluded that Positive Striving represented the “adaptive aspect of personal motivation” (Frost et al., 1993; p. 125). More recently, Cox and colleagues conducted a similar but more thoroughgoing analysis of items from both the HMPS and FMPS using confirmatory factor analytic methods (Cox et al., 2002). After refining item sets and subscales, they too found evidence of a higher order two-factor solution that appeared to correspond with an adaptive/maladaptive model (Cox et al., 2002). However, they did not directly compare this adaptive/maladaptive model with other plausible models of the perfectionism construct, and they did not include all of the HMPS and FMPS subscales in their higher order analyses. Other correlational and behavioral validation studies appear to also provide evidence for a distinction between adaptive and maladaptive perfectionism. Hill and colleagues (1997) carried out a study to understand the placement of the perfectionism construct within the taxonometric network of the Big Five: neuroticism, extraversion, openness, agreeableness, and conscientiousness (Costa & McCrae, 1990). In a sample of undergraduates, the authors found that self-oriented perfectionism was associated with numerous facets of conscientiousness, and negatively associated with the vulnerability facet of neuroticism. Other-oriented perfectionism was associated with more agency; that is more self-confidence and competitiveness, whereas socially-prescribed perfectionism was associated only with neuroticism, particularly the depression facet (Hill et al., 1997). This study suggests that the scales of the Hewitt MPS have both positive and negative connotations in terms of the Big Five. Self-oriented perfectionism appeared to have largely adaptive consequences whereas socially-prescribed perfectionism had more negative consequences. Emerging studies also suggest that this distinction between adaptive and maladaptive perfectionism has at least some behavioral validity. Adaptive perfectionism appears to lead to fewer self-defeating behaviors in evaluation situations, a task focus as opposed to rumination about performance on a task, and less vulnerability to negative affectivity (Bieling et al., 2003, Enns et al., 2001 and Rheaume et al., 2000). Taken together, both theoretical explications and at least some empirical data suggest that perfectionism may have both adaptive and maladaptive features. However existing studies leave open the question of whether or not perfectionism is better described as a unitary construct, or whether it is best described as having two underlying dimensions, adaptive and maladaptive. Another question is whether the two forms of perfectionism are independent of one another or whether they are correlated, positively or negatively. Also, not fully understood to this point is the extent to which the two common perfectionism scales assess the same underlying construct or whether they are best seen as assessing two distinct constructs based on their item content.