The construct of perfectionism is related to many important outcome variables. However, the term “perfectionism” has been defined in many different ways, and items comprising the different existing scales appear to be very different in content. The overarching aim of the present set of studies was to help clarify the specific unidimensional constructs underlying what is called “perfectionism”. First, trained raters reliably sorted items from existing measures of perfectionism into nine dimensions. An exploratory factor analysis, followed by a confirmatory factor analysis on an independent sample, resulted in a 9 scale, 61 item measure, called the Measure of Constructs Underlying Perfectionism (MCUP). The nine scales were internally consistent and stable across time, and they were differentially associated with relevant measures of personality and psychosocial functioning in theoretically meaningful ways.

Perfectionism has long been recognized as a factor that is central to understanding psychological disorders, as it is significantly higher in the majority of psychological disorders compared to the general population. The construct of perfectionism was examined in this research by exploring differences between positive and negative perfectionism. The literature to date has focused almost exclusively on perfectionism as a maladaptive construct, with little research examining if perfectionism can be a positive factor. The implication of the study was to determine if some factors identified may be potentially important in future treatments targeted towards perfectionism, as there is some evidence to suggest that perfectionism may predict poorer response to standard cognitive behavioural treatment. This research compared three different groups; (i) a clinical group with diagnoses of anxiety and depression (n = 40); (ii) a group of athletes (n = 111) and (iii) a student control group (n = 101). The research consisted of 5 studies. In Study 1, evidence was found for the validity, consistency of factor structure and internal consistency of the Positive and Negative Perfectionism Subscale (PANPS; Terry-Short, Owens, Slade, & Dewey, 1995). In Study 2, clinical participants with a range of diagnoses were found to have significantly higher overall perfectionism and negative perfectionism compared to athletes and controls. Rigidity predicted higher positive perfectionism. Dichotomous thinking accounted for a large proportion of variance in negative perfectionism, and was argued to be an important factor distinguishing between positive and negative perfectionism. In Study 3, positive perfectionism was found to relate to faster performance time in athletes competing in triathlons, and negative perfectionism was not found to impede sporting performance. / The Big Five personality domains were investigated in Study 4, and Agreeableness was found to be a significant predictor of negative perfectionism in the clinical group. In comparing clinical and athlete groups, athletes had significantly lower Neuroticism, and higher Extraversion and Conscientiousness. Study 5 was a clinical descriptive study that examined motivation to change and cognitions about failure in a select sample of clinical participants with extreme high scores on negative perfectionism and athletes with extreme low scores. The clinical participants reported many negative consequences, yet despite this recognition, the majority reported they did not wish to change perfectionism. Also, as the level of negative perfectionism increased, the degree of diagnostic comorbidity increased. It was concluded that it may be more useful to distinguish between positive achievement striving and negative perfectionism rather than positive and negative perfectionism. Clinical implications were outlined which included targeting dichotomous thinking and resistance to change in the development of treatments for perfectionism.

El objetivo de la investigación fue evaluar las propiedades psicométricas de la Escala Multidimensional de Perfeccionismo de Belloch, Carrasco y Perpiña en estudiantes de universidades públicas en Lima Metropolitana. Para esto, se desarrolló una investigación de tipo instrumental contando con la participación de 420 estudiantes (49.76% mujeres y 50.23% hombres), donde el 1.19% de personas eran menor o igual a 18 años, el 95.71% de estudiantes tenían entre 19 y 25 años, mientras que 2.61% tenían 26 o más años. De acuerdo con los resultados del análisis factorial confirmatorio, los datos se ajustan a menor índice mejor ajuste (Brooke, Russell & Price, 1988); Rock Mim Residuals (RMR≤.08), Goodness of Fit Index (GFI≥.90), Adjusted Goodness of Fit Index (AGFI≥.90), Chi cuadra (c²) e Índice de parsimonia (PRATIO). Con ello, el modelo de 4 dimensiones (miedo a los errores, influencias paternas, expectativas de logro y organización) se encuentra relacionado entre sí, pues la consistencia interna resultó aceptable (ω =.87 a .93) para cada dimensión. Para una validación convergente, empleamos la escala de Escala de Metas de Estudio (EME) en donde se evidenció que existe correlación positiva y negativa entre factores De esta manera, se concluye que la Escala Multidimensional de Perfeccionismo es un instrumento con adecuadas propiedades psicométricas para medir el perfeccionismo en universitarios peruanos. / The objective of the research was to evaluate the psychometric properties of the Multidimensional Perfectionism Scale of Belloch, Carrasco and Perpiña in students of public universities in Metropolitan Lima. For this, an instrumental type investigation was developed with the participation of 420 students (49.76% women and 50.23% men), where 1.19% of people were less than or equal to 18 years, 95.71% of students were between 19 and 25 years, while 2.61% were 26 or older. According to the results of the confirmatory factor analysis, the data fit the lower the better fit index (Brooke, Russell & Price, 1988); Rock Mim Residuals (RMR≤.08), Goodness of Fit Index (GFI≥.90), Adjusted Goodness of Fit Index (AGFI≥.90), Chi square (c²) and Parsimony Index (PRATIO). With this, the 4-dimensional model (fear of mistakes, parental influences, expectations of achievement and organization) is related to each other, since the. The internal consistency was acceptable (ω = .87 to .93) for each dimension. For a convergent validation, we used the Scale of Study Goals (EME) scale where it was evidenced that there is a positive and negative correlation between factors In this way, it is concluded that the Multidimensional Perfectionism Scale is an instrument with adequate psychometric properties to measure perfectionism in Peruvian university students. / Tesis