The dimensions of perfectionism. Cognitive Therapy and Research, 14 5 , CM If I do not set the highest standards for myself, I am likely to end up as a second-rate person. PS My parents wanted me to be the best at everything. PE Even when I do something very carefully, I often feel that it is not quite right.
|Published (Last):||25 March 2009|
|PDF File Size:||1.79 Mb|
|ePub File Size:||2.47 Mb|
|Price:||Free* [*Free Regsitration Required]|
Perfectionism is defined as the desire to achieve the highest standards of performance, in combination with unduly critical evaluations of one's performance. This study was designed to analyse the psychometric proprieties of the Portuguese version of the Frost Multidimensional Perfectionism Scale through confirmatory factor analysis in two different samples calibration with adolescent athletes and validation with adolescent athletes.
All constructs presented good internal consistency, convergent and discriminant validity. A multi-group analysis exposed the cross validity of the model.
The correlations between perfectionism, fear of failure and sport anxiety measures revealed evidence of its concurrent validity, providing evidence for the proposed model. Overall, the results of the present study provided evidence for these instruments' validity and reliability, however further investigations are required. To assess these various dimensions of perfectionism Frost et al. A similar self-reported measure of perfectionism was designed by P.
Theoretically, meaningful relationships have been observed between the subscales of the two instruments. Consistent patterns of correlations highlighting the functional nature of the Frost-MPS and Hewitt-MPS susbcales in sport domain are also evident in the literature. Conroy, Willow, and Metzler investigated how aspects of perfectionism in athletes relate to the fears of failure. Findings suggested that the perfectionist concerns dimension of perfectionism displays close links to fears of failure.
Similar positive correlations between perfectionism concerns over mistakes with all five fears of failure were identified in Kaye, Conroy, and Fifer study. Conroy et al. Similar positive correlations between perfectionism concerns over mistakes with all five fears of failure were identified in Kaye et al.
Sagar and Stoeber showed that concern over mistakes predicted higher levels of all fears of failure. According to Hall et al. These authors concluded that overall perfectionism underpins achievement anxiety as it emerged as a consistent predictor of cognitive anxiety prior to performance.
Kawamura, Hunt, Frost, and DiBartolo demonstrated that increased concern over mistakes and doubts about action were associated with higher levels of anxiety symptoms. Similar findings were reported on an investigation into running addiction by Coen and Ogles , arguing that athletes who had a tendency to express doubts about the quality of their performances reported significantly higher trait anxiety.
In the same way, Martinent, Campo, and Ferrand , confirmed previous results regarding strong associations between perfectionism subscales and specific dimensions of state anxiety.
In sum, these findings suggest that perfectionism may promote the use of maladaptive motivational cognitions during the performance process, which will possibly threaten the athlete's self-worth and eventually might contribute to the prediction of multidimensional state anxiety in sport. Some perfectionism theorists e.
On the other hand, and according to Hewitt, P. Hewitt, Flett, Besser, Sherry, and McGee : "it has been demonstrated empirically that individuals with a high level of commitment to one perfectionistic goal also tend to have elevated levels of commitment to a wide range of other perfectionistic goals and that the greater the importance placed on being perfect in many domains, the greater the depressive symptomatology in perfectionists".
This is the main reason why there are still a great amount of uses of the original Frost-MPS in several domains and languages. Therefore, it was our intention to analyse the psychometric proprieties of the Frost Multidimensional Perfectionism scale in its Portuguese version in the universe of Portuguese athletes, in order to provide a psychometrically valid and reliable instrument.
More specifically, we intended to use the data collected to analyse the six-factor structure as it was proposed by the Frost-MPS authors. Furthermore, into a more refined analysis, we tested the model to determine its internal consistency and convergent validity; tested the invariance of the structure with a cross-validation strategy; and explored the concurrent validity with a sample of Portuguese athletes.
The study encompassed a total of adolescent athletes in two convenience samples, with an age range from 12 to 18 years old. The first sample calibration , used for confirmatory factor analysis purposes, with athletes 94 female, male had a mean age of The study was reviewed by the University Ethics Board prior to data collection. Letters and parental consent forms were sent home to parents for participants under the age of 18 informing them of the nature of the study and requesting their permission for their child's participation in the study.
All participants, including minors, signed consent forms. The Portuguese version of the Frost et al. This item questionnaire generates an overall perfectionism score as well as scores for six subscales that reflect specific domains of perfectionism: 1 concern over mistakes, 2 doubts about actions, 3 personal standards, 4 parental expectations, 5 parental criticism, and 6 organization.
The total perfectionism score is the sum of all subscales except organization Frost et al. Participants were asked to rate how strongly they believed each of the 14 aversive consequences of failure were likely to occur to them after failing. Items were answered on a five-point Likert scale from 1 do not believe at all to 5 truly believe. A composite fear of failure score can be derived by summing all 14 items.
The Portuguese version of the Sport Anxiety Scale SAS-2 Smith, Smoll, Cumming, and Grossbard translated and adapted by Cruz and Gomes has tree subscales somatic anxiety, worry, and concentration disruption each consist of five items. Satisfactory psychometric proprieties of this instrument were obtained in Dias, Cruz, and Fonseca study. A composite anxiety score can be derived by summing all 15 items.
Data analysis, were performed using AMOS The maximum likelihood ML method was used. Following discovery of unsatisfactory fit, modification indices and standardized estimates were examined to evaluate for alternative models or probable item removals. Univariate skewness and kurtosis of items were examined, as well as multivariate normal distribution using Mardia's coefficient of multivariate kurtosis West, Finch, and Curran Discriminant validity was established when AVE for each construct exceeded the squared correlations between that construct and any other Hair et al.
We also considered Cheung and Rensvold suggestion that a difference of CFI of less than or equal to. Evidence of the concurrent validity was obtained by studying the association among constructs measures that, theoretically, should be related, confirming, empirically, these relationships between the constructs Campbell and Fiske On the data there were no missing values and it was approximately univariately normal, since items with absolute values of skewness lower than 3 and kurtosis lower than 7 did not deviate enough from the normal distribution Kline Results of Mardia coefficient revealed that data violated the assumption of a multivariate Gaussian distribution P.
Bentler and Wu , and a bootstrapping procedure of Bollen and Stine was applied to adjust the p value of the chi-square statistic. Formerly, not all estimated factor loadings exceed the cut-off point of. Due to the lack of support from CFA performed, post hoc model adjustments were conducted in an effort to develop a better fitting model and further analyses became exploratory in the sense that they focused on the detection of misfitting parameters in the originally hypothesized model Byrne Since the results in the original model first-order model indicated that not all items did load significantly on its construct, in the interest of scientific parsimony, all scale items that showed unacceptable factor loadings were removed Byrne This process resulted in the removal of item 10 and 16, from the original model.
Examination of the modification indices MI , suggested that an improved model resulted in the elimination of the items 4, 14, 15, 17, 18, 24, 26 and 29, following the intent of Chartrand, Robbins, Morril, and Boggs to create "pure measures of each factor" p. We tried to keep the number of items eliminated from the model as small as possible and to leave at least three per factor, in accordance with the recommendations of Hatcher According to Byrne large MI argue the presence of factor-cross loadings i.
Another type of method effect that can trigger error covariance is a high degree of overlap in item content. This redundancy occurs when an item, although worded differently, essentially asks the same question or is related to another factor Byrne This suggested that allowing the two errors to correlate would significantly improve model fit. Examining items 7 "I am a neat person" and 31 "I am an organized person" , it was clear that they were attempting to measure perfectionism related to Organization.
Given the redundancy between both items we decided to correlate the two errors involved. At this point of model adjustments, and besides the acceptable fit indices showed, the model was presenting some validity concerns e. As reported in Table 2 , standardized items loadings and the z-values provided evidence that the items accurately captured their respective factorsAnderson and Gerbing All composite reliability values exceeded the recommended minimum.
Furthermore, all constructs were considered to exhibit discriminant validity because all AVE values exceeded the appropriate square factor correlations. Overall, the measurement model was within the required criteria and showed good psychometric proprieties. Moreover, since there were no significant differences in the CFI values for all model comparisons, the results demonstrated the model's invariance in both samples, indicating that the factorial structure of the scale was stable in the two independent samples Cheung and Rensvold The results presented in Table 4 , using the validation sample, revealed that all dimensions of perfectionism and SAS-2 sub-scales and total scores were positively correlated, except for somatic anxiety factor and organization factor.
All aspects of perfectionism showed positive correlations with the different fears of failure dimensions, except for organization factor. Perfectionism total scores were also positively correlated with fear of failure total score. The purpose of this study was to test the factorial validity in a Portuguese sport setting of the Frost Multidimensional Per fectionism scale Frost-MPS , originally developed by Frost et al. The confirmatory factorial analysis performed on the Frost-MPS, to a sample of athletes, presented lack of support from CFA performed.
Even though the original factor structure of the scale was confirmed in our study, several problems were detected, namely items with unacceptable factor loadings in their different subscales, reliability issues and convergent validity problems.
The lack of strong statistical support for the original model led us to a respecification and a reestimation of the model. Although confirmatory factor analysis continued to be used, it should be clearly acknowledged that these analyses are exploratory in the sense that they focus on the detection of misfitting parameters in the originally hypothesized model in order to investigate an alternative structure to better represent the sample of athletes.
The factor analysis using the re-specified model showed an acceptable fit for the Portuguese athletes' sample and confirmed a first-order construct and, as well as, a second-order construct. The first-order construct showed composite reliability, convergent validity and discriminant validity. However, in the re-specification process, doubts about actions' factor possesses only two items and it its recognised to be a problematic issue for some time.
The correlation coefficients between the six subscales of the Portuguese version of the Frost-FPM were somehow analogous to the ones obtained by Frost et al. Concern over mistakes presented the highest correlations with all the subscales. Instead, organization was not strongly associated with all the subscales, except for the personal standards sub-scale.
According to Frost and Henderson organization and personal standards subscales reflect a positive feature of perfectionism and consequently presented a strong correlation. Furthermore, these two subscales capture the first component of perfectionism, namely the tendency to aspire to the highest standards of performance, without the explicit meaning of what violations of those standards might mean to the self Frost et al.
On the other hand, and similar to Frost et al. Frost-MPS's concurrent validity has been ascertained and, as expected, scores have exhibited appropriate convergent validity with measures of sport anxiety and fear of failure Conroy et al. Likewise, a positive correlation was found in our study with all dimensions of sport anxiety, providing additional evidence for convergent validity. Correlations between all dimensions of perfectionism and fear of failure reported medium to strong positive correlations and are similar to previous studies Conroy et al.
Regarding the re-specification done, it must be recognised that previous investigations of the factor structure have arrived at divergent solutions i. Further research is necessary in order to further establish factorial validity and reliability, namely factorial invariance analysis and exploratory factor analysis. One limitation of this study has to do with the fact that our sample is different from the one that was used by the original authors. The sample of the original authors and most of the studies conducted to inspect the psychometric proprieties were composed by university students, whereas the Portuguese sample originated from adolescents' sports leagues and federations.
In our study, a post hoc model modification was used, refining the measurement instrument. This procedure helped to identify factorial ambiguous items that can be removed from the model Markland and Ingledew According to Hofmann , this approach does not comprise the integrity of the a priori model, since the model remains essentially the same; it simply has fewer indicators of its factors.
However, it should be unequivocally acknowledged that, since we adopted an exploratory model generation approach and by itself represents a departure from the hypothetic-deductive model testing ideal, which structural equation modelling was principally designed in the first place P.
An evaluation of the factor structure of the frost multidimensional perfectionism scale
Personality and Individual Differences, 24 4. ISSN Despite its increasing popularity in personality and clinical research, the FMPS has also drawn some criticism for its factorial instability across samples. The present article argues that this instability may be due to an overextraction of components. Whereas all previous analyses presented six-factor solutions for the FMPS items, a reanalysis with Horn's parallel analysis suggested only four or five underlying factors.
Quick Links Menus
Frost Multidimensional Perfectionism Scale: the portuguese version. Corrected item-total correlations ranged from. The scale test-retest reliability suggested a good temporal stability with a test-retest correlation of. A principal component analysis with Varimax rotation was performed and based on the Scree plot, two robust factorial structures were found four and six factors. The principal component analyses, using Monte Carlo PCA for parallel analyses confirmed the six factor solution.
Frost multidimensional perfectionism scale
Perfectionism is defined as the desire to achieve the highest standards of performance, in combination with unduly critical evaluations of one's performance. This study was designed to analyse the psychometric proprieties of the Portuguese version of the Frost Multidimensional Perfectionism Scale through confirmatory factor analysis in two different samples calibration with adolescent athletes and validation with adolescent athletes. All constructs presented good internal consistency, convergent and discriminant validity. A multi-group analysis exposed the cross validity of the model. The correlations between perfectionism, fear of failure and sport anxiety measures revealed evidence of its concurrent validity, providing evidence for the proposed model.
Start managing your projects on the OSF today.
We'd like to understand how you use our websites in order to improve them. Register your interest. Although this scale is in common use, its psychometric parameters still remain in dispute. The ORG organization subscale was removed after evaluation. The evaluation of the specific relationships of each dimension of the adapted version of the FMPS with shame, guilt, worry, rumination, and indecisiveness revealed that when controlling for the variances of the remaining dimensions, significant relationships occurred almost only in the case of the CM, DAA and PS scales, suggesting that they constitute the three core facets of perfectionism as measured by the FMPS.