تفاوت های فردی در اضطراب آمار: نقش کمال گرایی، تعلل و اضطراب خصلتی

Recent research detected small but significant associations between perfectionism and statistics anxiety [Onwuegbuzie, A., & Daley, C. (1999). Perfectionism and statistics anxiety. Personality and Individual Differences, 26, 1089–1102]. The current study re-examined this relationship whilst simultaneously controlling for trait anxiety and procrastination, variables which are known to be associated with perfectionism and which also have a likely association with statistics anxiety. Measures of all four constructs were obtained by questionnaire from a sample of 93 students, and multiple regression analyses were employed. Statistics anxiety constituted the criterion variable whilst perfectionism, trait anxiety and procrastination were regarded as predictor variables. Results indicated very modest links between inter-personal perfectionism and components of statistics anxiety, whereas intra-personal perfectionism, trait anxiety and procrastination were each found to have good predictive utility. It was concluded that aspects of ego-involvement such as fear of failure and evaluation concern, which are thought to pervade each of these predictors, may be responsible for statistics anxiety.

Increased research attention to individual differences in perfectionism over the past two decades has brought with it a proliferation of definitions and conceptualisations of the construct. For example, Hamachek (1978) distinguished between normal and neurotic perfectionism, Norman, Davies, and Nicholson (1998) differentiated between adaptive and maladaptive perfectionism, Terry-Short, Owens, Slade, and Dewey (1995) discriminated positive from negative perfectionism, and Adkins and Parker (1996) separated active from passive perfectionism. As well as the traditionally negative emphasis, these two-dimensional characterisations implicitly acknowledged positive aspects of perfectionism, in particular achievement-striving and self-actualisation (Frost, Marten, Lahart, & Rosenblate, 1990). Multidimensional conceptualisations of perfectionism have also emerged in the last few years. For example, Hewitt and Flett (1991) identified three types of neurotic perfectionism—self-oriented, other-oriented, and socially prescribed. Self-oriented perfectionism involves setting unrealistic goals for oneself, stringently evaluating oneself against their attainment, selectively attending to failure and over-generalising it, and engaging in all-or-nothing thinking. Other-oriented perfectionism is similar except that it is directed towards others; that is, the efforts of others are stringently evaluated against unrealistic standards. Finally, socially prescribed perfectionism defines a set of beliefs people have that others expect perfection from them, hold unrealistic standards for them, and will evaluate them stringently. An alternative, multidimensional, model has been offered by Frost et al. (1990). They identified six components which, when combined, provide a total perfectionism score. The most important components are those related to the setting of high standards and excessive concern over mistakes. Frost, Heimberg, Holt, Mattia, and Neubauer (1993) compared their own measure of perfectionism with that of Hewitt and Flett (1991) and found considerable overlap. A factor analysis of all nine sub-scales yielded a two-factor solution, the first reflecting maladaptive evaluation concerns and the second positive achievement striving. It seems that however one chooses to conceptualise perfectionism, setting unrealistic goals and negatively evaluating the self in response to inevitable failure are central to the construct. One reason why the study of perfectionism has become increasingly popular may be attributable to its consistent links with a wide variety of psychological disturbances and distress. For example, neurotic perfectionism has been linked with anorexia nervosa (Cooper, Cooper, and Fairburn, 1985), depression (Blatt, 1995 and Hewitt & Dyck, 1986), suicide ideation (Baumeister, 1990 and Hewitt et al., 1997); alcoholism (Hoge & McCarthy, 1983) and depressive affect (Higgins, Bond, Klein, & Strauman, 1986). More recent work has found links between self-oriented perfectionism and burnout in competitive junior tennis players (Gould, Tuffey, Udry, & Loehr, 1997), and between Hewitt and Flett's (1991) three components of perfectionism and cynicism and exhaustion among career mothers (Mitchelson & Burns, 1998). In addition, perfectionism has been linked with procrastination among students (Solomon & Rothblum, 1984), and with depression and anxiety among those striving for academic achievement. Of particular interest to the current study is the association between perfectionism and anxiety. Antony, Purdon, Huta, and Swinson (1998) found higher levels of socially prescribed perfectionism among anxiety patients (panic disorder, obsessive compulsive disorder and social phobia) than among non-clinical controls. Likewise, in a non-clinical student sample, Saboonchi and Lundh (1997) reported positive associations between expressions of social anxiety and socially prescribed perfectionism. Sullivan, Bulik, Fear, and Pickering (1998) also reported higher anxiety in socially prescribed perfectionists. The mechanisms which mediate the relationship between perfectionism and these various pathological outcomes are not well understood. A stress-diathesis model was put forward and tested by Hewitt, Flett, and Ediger (1996). They found that perfectionism was predictive of depression over time, but only if stressful conditions prevailed. According to Hewitt and Flett (in press), perfectionism influences the aversiveness and duration of stress, and may be responsible for generating it in the first instance. For example, when encountering an ego-involving failure experience, perfectionists experience distress more intensely than non-perfectionists. This is because their sense of self-worth is equated with perfect performance (Pacht, 1984), because they tend to over-generalise failure, and because they engage in negative ruminations about the self. Frost et al. (1997) reported higher levels of concern and negative affect over mistakes among perfectionists than non-perfectionists. Because of their self-imposed excessively high standards, no performance can ever be seen as satisfying. Constantly identifying episodes of failure may in itself be stress-generating. At a sub-clinical level, perfectionism has also been associated with statistics anxiety (Onwuegbuzie & Daley, 1999). Statistics anxiety has been defined simply as anxiety that occurs as a result of encountering statistics in any form and at any level (Onwuegbuzie, DaRos, & Ryan, 1997), and has been found to negatively affect learning (Onwuegbuzie & Seaman, 1995). Because a basic grounding in statistics forms an important component of many different programs at university level, related anxiety can compromise a successful university experience and may result in academic under-achievement (Zeidner, 1991). Although research into statistics anxiety has been conducted for about twenty years now (e.g. Roberts & Saxe, 1982), Zimmer and Fuller (1996) could find no studies linking personality to statistics, with Onwuegbuzie and Daley's (1999) study apparently being the first to establish such a link. Like perfectionism, statistics anxiety also emerged as multidimensional. In particular, Cruise, Cash, and Bolton (1985) identified six components, namely worth of statistics, interpretation anxiety, test and class anxiety, computational self-concept, fear of asking for help, and fear of statistics teachers. Using Hewitt and Flett's (1991) three-dimensional model of perfectionism and Cruise et al's measure of statistics anxiety, Onwuegbuzie and Daley (1999) found that other-oriented and socially prescribed perfectionism positively predicted statistics anxiety in a graduate sample. Relationships were significant for interpretation anxiety, computational self-concept, and fear of asking for help. These findings are in keeping with the view that perfectionists are vulnerable when exposed to tasks that can illuminate failure. In examining the limitations of their study, Onwuegbuzie and Daley suggested that, compared with perfectionism, other personality dimensions might play an important role in predicting statistics anxiety. The current study is designed to address this possibility. Two established personality factors, procrastination and trait anxiety, will be examined. The reason for including trait anxiety is fairly obvious given that those who score high on this dimension have a tendency to selectively attend to environmental threat (Eysenck, McLeod, & Mathews, 1987) and to preferentially interpret ambiguous stimuli in a threatening rather than non-threatening manner (Calvo, Eysenck, & Castillo, 1997). These cognitive biases are likely to result in trait anxious individuals experiencing more anxiety than non-anxious counterparts (Derakshan & Eysenck, 1997). In addition, there is direct empirical evidence supporting a link between perfectionism and trait anxiety (Flett, Hewitt, & Dyck, 1989). Accordingly, any examination of factors which predict statistics anxiety should, at the very least, include trait anxiety as a control variable. Procrastination is another trait which has been found to be associated with perfectionism (Burka & Yuen, 1983, Flett et al., 1995, Hamachek, 1978, Missildine, 1963 and Pacht, 1984). It is defined as an irrational tendency to delay tasks that should be completed (Lay, 1986), and those exhibiting academic procrastination have reported problematic levels of anxiety (Rothblum, Solomon, & Murakami, 1986). It has been estimated that between 40% (Solomon & Rothblum, 1984) and 95% (Ellis & Knaus, 1977) of students engage in academic procrastination. Fear of failure has been suggested as one factor underlying procrastination (Flett et al., 1991 and Rothblum, 1990), whilst Saddler and Buley (1999) have argued that concerns about negative evaluation underpin the construct. Such factors are also closely associated with perfectionism (Flett et al., 1995). Empirical evidence of a positive link between socially prescribed perfectionism and the magnitude and frequency of academic procrastination has been provided by Flett, Blankstein, Hewitt, and Koledin (1992). Given that procrastination is associated with high levels of anxiety, and with fear of failure and negative social evaluation, its similarity to and potential overlap with perfectionism is clear. Therefore, as in the case of trait anxiety, in order to obtain a complete and accurate account of the relationship between perfectionism and statistics anxiety it is necessary also to include a measure of procrastination in the study. In sum, the purpose of the study was to assess the relationship between perfectionism and statistics anxiety in a manner similar to Onwuegbuzie and Daley (1999), but to include the additional dimensions of trait anxiety and procrastination in order to remove their effects from any such relationship.