Tripartite Typology of Perfectionism among Psychology Students

: The objective of the research was to find out whether it is possible to identify the perfectionist typology according to Parker with three groups namely functional perfectionists, dysfunctional perfectionists and non-perfectionists. The research group was composed of full-time university students of single-subject psychology (N = 172) with age range of 18 to 25 years. We used the Multidimensional Perfectionism Scale (F-MPS) to measure perfectionism. Based on the non-hierarchical cluster analysis (k-means) of the items in the F-MPS questionnaire, we identified three types of perfectionists, with the first cluster of students identified as functional perfectionists (N=51.74%), second as dysfunctional perfectionists (N=17,44%) and third cluster of non-perfectionists (N=30.81%). Identification of dysfunctional perfectionists (N=17.44%) among psychology students accentuates the need to focus on the issue. Dysfunctional perfectionism adversely affects the quality of the study and also the performance of the future profession itself.


Introduction
The issue of perfectionism is currently quite a frequent topic especially abroad, but in Slovakia this phenomenon is still a rather new and unexplored topic.This term is often associated with "exaggerated detailism" or perceived as "a desirable phenomenon that is the privilege of successful people".However, there is a certain type of perfectionism that actually hinders achievement of success.The research is focused on the identification of three types of perfectionism in psychology students.The pressure of postmodern society very often stimulates dysfunctional perfectionism, which can have a negative effect on the practice of the psychologist Through our research, it is possible to design an e-learning program aimed at stimulating adaptive perfectionism.It is a relatively permanent personality trait that is nowadays considered by the professional literature and by scientists to be a multidimensional construct (Hewitt & Flett, 1991;Parker, 1997).For determining perfectionism as a phenomenon which can act in both a positive or negative way on the individual or not act at all, we found the inspiration in the research of typology of perfectionists made by Parker (1997), which was the first to identify functional and dysfunctional perfectionists and nonperfeccionists (Codd, 2010).These findings became the basis for our research in which we decided to verify the possibility of identifying the typology of perfectionists among psychology students.In the context of the typology itself, our research may be included in typological procedures because, based on previous findings of Parker (1997), which will serve as a comparative model, we will try to identify and analyse individual types of perfectionists (Parker, 1997;Stoeber & Stroeber, 2009).

Multidimensional concepts of perfectionism
The arrival of the 90s in the last century was revolutionary for the rise of perfectionism.The one-dimensional perception of perfectionism, pushing its negative core, was substituted by a multidimensional concept that presents both the positive and negative aspects of perfectionism (Frost et al., 1990).Among the most well-known representatives of the comprehensive perfectionism are Hewitt and Flett (1991).They define perfectionism as "the pursuit of perfection, and they consider extreme perfectionists as those who want to be perfect in all aspects of life" (Hewitt & Flett, 1991, p. 32).This means a tendency to prescribe high standards based on disparate degrees of perfectionism, depending on a certain amount of effort to achieve that standard, as well as a response to failure.The magnitude of the collision between what is really achievable and the determined can be considered as the interface factor between normal and pathology (Winter, 2006).Frost et al. (1990) is also an important supporter of a multidimensional construct, who considers the decisive characteristic of perfectionism to be "setting of extremely high goals" (Codd, 2010).However, attempting to be the best in itself conceals also pitfalls in the form of a higher sensitivity to life problems (Winter, 2006).This fact is also supported by Hamachek's idea (as cited by Parker, 2000), because perfectionism also has a positive dimension that helps to develop personality and is not strictly linked to psychological and somatic disorders.The same way Chang et al. (2004) also perceive perfectionism as a multidimensional phenomenon, and mention its five basic attributes, which occur to varying degrees and not just in perfectionists: 1) "High personal standards 2) Self-criticism 3) Doubts about the effectiveness of own actions 4) Efforts to meet social expectations 5) Extreme focus on organization and order" (Chang et al., 2004, p. 97).

The typology of perfectionism
In connection with the reported classification of multidimensional perfectionism, there are researches (Hawkins, 2005;Hawkins et al., 2006;Park & Jeong, 2015;Parker, 1997;Rice & Dellwo, 2002) which, by cluster analysis of the dimensions of the Frost Multidimensional Scale of Perfectionism (Frost et al., 1990), extend the construct to the tripartite typology of perfectionist individuals (Hawkins, 2005).
In his study of academically gifted students using F-MPS, Parker (1997) identified two clusters of perfectionists -functional and dysfunctional, and one cluster of non-perfectionists.The first cluster was described as non-perfectionists who scored low in dimensions of PS (Personal Standards), PE (Parental Expectations), and O (Organization) as well as Low Total Perfectionism Score (P), which is the sum of scores for each dimension.The second cluster was characterized as healthy or functional perfectionists, because they achieved low scores in CM (Concern over Mistakes), PC (Parental Criticism) and DA (Doubt about actions), medium score in PS (Personal Standards) and high scores in O (Organization).Students in the third cluster were referred to as dysfunctional perfectionists because they achieved the highest scores in CM (Concern over Mistakes), PS (Personal Standards), PE (Parental Expectations), PC (Parental Criticism), DA (Doubt about actions) and also the highest total score (P) (Hawkins, 2005;Landa & Bybee, 2007;Parker & Stumpf, 1995).F-MPS has also been used in other studies (Hawkins et al., 2006;Rice & Dellwo, 2002;Rice & Mirzadeh, 2000) concerning university students which have determined similar results with two clusters of perfectionists (adaptive, maladaptive) and one cluster of non-perfectionists.
Although the perfectionist typology was confirmed in each of the studies, representation of the F-MPS dimension in each cluster included some differences.Higher levels of PS (Personal Standards) were noted in a cluster of dysfunctional perfectionists in the studies of Parker (1997) and Rice and Mirzadeh (2000), both in functional and dysfunctional clusters in the study (Hawkins et al., 2006), while Rice and Dellwo (2002) reported that the cluster of functional perfectionists achieved the highest score in given dimension.These studies (Hawkins, Watt & Sinclair, 2006;Rice & Dellwo, 2002;Rice & Lapsley, 2001;Rice & Mirzadeh, 2000) consistently report the highest score in the O (organization) dimension in the cluster of functional perfectionists.The cluster of dysfunctional perfectionists generally scores the highest in the CM, DA, PE and PC dimensions, although Rice and Dellwo (2002) found in their study that a cluster of functional perfectionists had a higher PE (Parental expectation) score than PC (Parental criticism), indicating that healthy perfectionists perceive their parents as those who maintain high expectations in their success and at the same time perceive them as uncritical.In an Australian study with high school students (Hawkins, 2005), the results showed that both clusters of functional and dysfunctional perfectionists had the highest PS scores (Personal standards), similarly to the Hawkins et al. (2006) study.In their study, Park and Jeong (2015) used a tripartite model of perfectionism.They examined the psychological well-being, life satisfaction and self-esteem across the three groups: adaptive perfectionists, maladaptive perfectionists and nonperfectionists.They focused on college students (N = 200) from South Korea.With cluster analysis, they confirmed the existence of three groups, based on findings from previous studies (Parker, 1997(Parker, , 2000)).By analysing the results, among other things, they found that adaptive perfectionists and non-perfectionists had higher scores of life satisfaction and self-confidence than maladaptive perfectionists (Park & Jeong, 2015).These studies (Hawkins, 2005;Hawkins et al., 2006;Parker, 1997) provide sufficient support for the tripartite typology of functional, dysfunctional perfectionism and non-perfectionism through the F-MPS scale.Some of the issues remain unresolved, and it is important to verify the existing findings and to explore the concept of typology of perfectionists in different populations and to compare them with various variables (Chang, 2000;Hawkins et al., 2006).
Perfectionism and college students J. J. Arnett (2007), proposed a new concept of "emerging maturity", which should more accurately characterize the period in which the university students are set and he situated it approximately within the range of 18 to 25 years.It was found that the emerging maturity in our conditions is characterized by a low level of stability and the search for own "self" (Codd, 2007;Radchuk et al., 2019).Lisa Lee (2007) examined, among other things, the relationship between perfectionism and age on a group of students (N = 301, of which 42 were men) with an average age of 18,26 years.The study did not reveal any correlations between the F-MPS dimensions and age.Research of college students points to the relationship with perfectionism, in particular, it was found that the rate of perfectionism decreases with age (Chang, 2002;Landa & Bybee, 2007;Parker & Stumpf, 1995).

Research group
The research group included university full-time students of of single-subject psychology (N = 172) with the age range from 22 to 25.The selection of the research group was realized through deliberate selection.The criterion for inclusion in the study was the study in the 1st or 2nd year of the Master's study.The original number of respondents (N = 262) was reduced by excluding students that were not comparable in terms of achievements and study orientation.Due to the insufficient number of men in the research (N = 30), we decided to exclude them and target only the female population.The F-MPS questionnaire was given to students of the first and second grade of psychology at UCM in Trnava, UK in Bratislava and the St Elizabeth University.Data collection took place between July 2018 and November 2018.Students participated in our research voluntarily and anonymously by completing the questionnaire, which is further detailed in the following section.We tried to balance the research group in terms of the socio-economic status of the students, the focus of the study, the results achieved and the age of participants.

Research method
We used the Multidimensional Perfectionism Scale to measure perfectionism, based on a multidimensional concept of perfectionism (F-MPS) consisting of 35 items (Frost et al., 1990), with the respondent recording the responses on a 5-degree Likert scale from 1 = very atypical to 5 = very typical.The scale consists of the following six dimensions: 1. Concern over Mistakes (CM) is a dimension defined as a susceptibility to react negatively to own mistakes and at the same time tendency to perceive errors as total failure.CM consists of nine items; 2. Doubts over Action (DA) is a dimension defined as an inclination for the feeling that the intended intentions do not go forward in future.DA consists of four items; 3. Personal Standards (PS) is a dimension defined as a tendency to set high goals and then to evaluate them based on their fulfilment.PS consists of seven items; 4. Organization (O) is a dimension defined as an emphasis on the system and order in organizing our own affairs.O consists of six items.Organization is the only dimension that is not counted in the overall score;

Parental Criticism (PC) is a dimension defined as an inclination
to perceive criticism from parents to their children as exaggerated.The PC consists of four items; 6. Parental Expectations (PE) is a dimension defined as an inclination to perceive the expectation of parents from their children as too high.The PE consists of five items (Frost et al., 1990).
Through F-MPS, you can quantify overall scores for perfectionism as well as for individual dimensions.Reliability in terms of internal consistency of the F-MPS questionnaire was determined by the Cronbach alpha coefficient and it shows a high internal consistency (α = 0.83).We have translated the questionnaire from the available Czech version of the questionnaire into the Slovak language and then subjected this to bilingual verification.

Methods of data analysis
With the way of our organization, our research is included in the typological procedures, because on the basis of the previous findings (Hawkins et al., 2006, Park & Jeong, 2015;Parker, 1997) and procedures in this field, we will try to identify and analyse individual types of perfectionists, and is also one of the statistical procedures.In order to identify the types of perfectionists in psychology students according to their perfectionist selfconsideration, we used cluster analysis.We used non-hierarchical cluster analysis (k-means) on Frost's Multidimensional Perfectionist scale items.We also subjected the F-MPS to a reliability test (Cronbach alpha values are given in connection with research methods).

Results
For the purpose of clarity, we present a separate table and graph under the research question RQ1: Is it possible to identify functional perfectionists, dysfunctional perfectionists and non-perfectionists in our research group of psychology students?
As part of the study we attempted to identify the perfectionist typology based on the Parker (1997)    Through analysis of variance (ANOVA) (see Table 2), which is part of the cluster analysis, we have found that the individual variables we used in the cluster analysis were needed to create clusters, with it being significant in all areas (p<0.005).The observed minimum distance between cluster cores is D = 41.701.Based on the fact that cluster analysis identified three types of perfectionists: functional, dysfunctional perfectionists and nonperfectionists, and on the F-MPS dimension values that represent individual clusters in accordance with the results of Parker 1 , we can conclude that it is possible to identify functional perfectionists, dysfunctional perfectionists and non-perfectionists in our research group of psychology students.

Interpretation of results
In order to meet the objective of our study, we have identified through non-hierarchical cluster analysis whether functional perfectionists, dysfunctional perfectionists and non-perfectionists can be identified in our research group of psychology students (RQ1).Compliant with the Parker (1997) criterion, we created three clusters for which the following findings were characteristic.The first cluster, which we identified as functional perfectionists, consisted of 89 respondents (51.74%).Individuals in this cluster were characterized by low values in dimensions: Concerns over Mistakes (CM), Doubt about actions (DA), Parental Criticism (PC), Parental expectations (PE); medium values in the dimension: Personal Standards (PS) and high values in dimension: Organization (O).Their overall perfectionism score reached average values.According to Frost (as cited by Hawkins et al., 2006), these are individuals who do not perceive their own mistakes as their overall failure, and yet believe that they will achieve the set intentions in future.Such individuals can be characterized as having a positive perfectionism because with the help of it they can motivate themselves and have proper self-evaluation (Spitzer, 2018).The criticism of parents is perceived as appropriate, not exaggerated, and their expectations are considered to be adequate.They set reasonable goals for their abilities, achieving them in a systematic and highly organized way.Based on Parker and Stumpf (1995), they are characterized by a higher level of perfectionist efforts, which include higher scores in the dimensions of Personal standards and Organization, and a low level of perfectionist fears which include higher scores in dimensions of Concern over Mistakes, Doubt about actions, Parental expectations and Parental Criticism.It follows that in our cluster 1 there are individuals who were identified as healthy perfectionists (Parker & Stumpf, 1995).For comparison, Parker (1997Parker ( , 2000)), in his research, describes functional perfectionists as people who focus on goals they can actually achieve.Parker (1997Parker ( , 2000) ) further states that functional perfectionists in their actions are not hesitant because they do not restrain themselves from possible mistakes in their performance and do not think that their parents would criticize them for the performance.We find this a satisfactory result because the values found by us compared to Parker (1997) are shown to match those of functional perfectionists.
In the second cluster there were a total of 30 psychology students (17.44%), who can be described as dysfunctional perfectionists.Individuals in this cluster were characterized by high, at the same time, the highest values in the following dimensions: Concern over mistakes (CM), Doubt about actions (DA), Parental criticism (PE), Personal standards (PS) and Parental expectations; low and at the same time lowest values in dimension of Organization (O).Their overall perfectionism scores were high or highest values.Frost et al. (1990) states that for dysfunctional individuals it is characteristic that they feel pressure from other people who have high expectations of their performance.These individuals experience anxiety from their failure.When comparing values measured by us with the values of other studies (Chang et al., 2004, Hawkins, 2005;Hawkins et al., 2006;Rice & Dellwo, 2002, Rice & Mirzadeh, 2000), dysfunctional perfectionists were characterized by extremely high values in all F-MPS dimensions.However, the group consisted of mathematically gifted children, so it is ambiguous whether the groups are comparable.Portešová (2002) found that a common population has a higher tendency to dysfunctional perfectionism.For dysfunctional perfectionists in Parker's (1997) group, compared with our findings, the highest score was characteristic in the dimensions of the Concern over mistakes, Personal standards, Parental expectations, Parental criticism, Doubt about actions, and the highest overall perfectionist score.As a result of over-criticism towards the perspective of parents, eating disorders and depression may occur in individual subjects (Chang et al., 2004;Park & Jeong, 2015;Rice & Mirzadeh, 2000;Spitzer, 2018;Stoeber & Stoeber;2009;Stoeber et al., 2009).In line with our findings, the results (Parker, 1997), which point to the highest values in the dimensions of Concern over mistakes, Doubt about actions, Parental expectations and Parental criticism are especially present in the cluster of dysfunctional perfectionists.The dimension of Personal standards shows unstable values in multiple clusters, as shown, for example, by study by Hawkins et al. (2006), in which both functional and dysfunctional types exhibited high values in personal standards.Within these studies, our findings of the dysfunctional cluster values are compliant, based on compliance in all dimensions except for Organization and Personal standards which are ambiguous, as has been suggested by our considerations and comparisons.Nevertheless, it is necessary to emphasize that the work of a psychologist for a person with dysfunctional perfectionism may in part be dangerous, because in the case of failures at work, dependence and various disorders may occurtu (De Cuyper et al., 2015).
The third cluster consisted of 53 respondents (30.81%).We have identified this cluster as non-perfectionists in our research.Individuals in this cluster were characterized by low values in all dimensions: Concerns about Mistakes (CM), Doubt about actions (DA), Parental Criticism (PC), Organization (O), Personal Standards (PS) and Parental expectations (PE).Their overall perfectionism score was low, or rather with lowest values.Parker's (1997) cluster of non-perfectionists, in line with ours, achieved a low score in the dimensions of Personal Standards, Parental expectations and Organization as well as a low overall score of perfectionism.However, it does not give values in dimensions of Concern over Mistakes, Doubt about actions and Parental Criticism.This low value of Personal standards and Organization does not coincide with the definition of perfectionism that considers these two dimensions to be crucial, which implies that these are individuals who are not perfectionism oriented.Parker and Stumf (1995) identified non-perfectionists who were characterized by a low level of perfectionist efforts for which a low score in the dimension of Personal standards and Organization is characteristic.A low score in all F-MPS dimensions was confirmed the same way as in the research (Landa & Bybee, 2007), which however worked with a typology different to the tripartite one but with an identified group of non-perfectionists.In the cluster of nonperfectionists in the research (Frost et al., 1990), the lowest total perfectionism scores and lowest values in all dimensions were also confirmed.Interestingly, the non-perfectionists in our group are more focused on organization and systematicity than dysfunctional perfectionists, but these differences are negligible and other findings do not correspond with them.
In Table no.3 we can see that there are differences between the types of perfectionists we have identified with other research findings (Parker, 1997), which can be caused by the size of the groups being compared, the unequal representation of men and women or the children and adults in research, or the fact that the study was performed on academically or otherwise gifted students.Some differences could also have been caused by the cluster analysis process, with other differences resulting from individual differences between groups as well as cultural differences.We cannot rule out the differences in the education system in Slovakia and in America, from which the surveyed research groups originate.As we know, there is no free education in America, so student demands can also increase as a result of this pressure.It is also possible to anticipate higher student performance requirements in these countries.

Conclusion
In our territory, the issue of perfectionism is still under-developed.The very identification of the fact that there are also dysfunctional perfectionist individuals among the university students, namely 17.44%, points to the importance of raising our population's awareness of the phenomenon.As dysfunctional perfectionism can lead to burnouts, depressions, anxiety disorders, social phobias and addictions (Chang, 2002;De Cuyper et al., 2015;Landa & Bybee, 2007;Parker, 2000;Stroeber & Stroeber, 2009;Stoeber et al., 2009;Wang et al., 2009) enough attention needs to be paid to this issue.In practice, our findings could be applied as part of programs to develop the psychologist's key competencies, thereby avoiding the emergence of other negative aspects that psychologists encounter very often in the performance of their work.It would be appropriate to use relaxation techniques, KBT therapy elements already during the study of psychology (Chang, 2000;Spitzer, 2018;Wang et al., 2009;Winter, 2006), which could potentially inhibit the dysfunctional perfectionism itself.
In our research, we also identified several limitations that need to be considered when interpreting the results of the study.Our research has focused on identifying the typology of perfectionists.The identified values, especially in the dysfunctional type of perfectionists, were borderline with the functional type of perfectionists, so these types can only be considered approximate in their division.We consider it necessary to emphasize that by designating dysfunctional perfectionists, we do not classify a given group of students as clinically pathological, but as a possible indicator for the occurrence of disorders.Based on these facts, which limit our findings, we recommend treating our results carefully as part of further comparisons in other research.One of the limitations we consider important is that we were working with a self-review questionnaire, which could result in results being distorted because the participants tried to respond in a desirable way.During the administration and evaluation of creativity tests, distortions could have occurred, despite our attempts to avoid any distortion.As part of our research, we did not focus on eliminating transient factors on the part of the participants.In future it would be advisable to administer the tests at the second meeting.In the context of our research group, which was made up exclusively of women, we consider limiting the impossibility of generalizing these results to the entire population of students, but only to the students of psychology.Similarly, the lower number of students involved in research and their unequal representation within individual university colleges, as well as low representation within individual types of perfectionists, may somewhat distort our results.
In the future, we would like to extend our concept to the issue of burnout, because we expect that dysfunctional perfectionism characterized by high demands, exaggerated criticism and expectations can greatly influence the emergence of burnout itself, especially in the field of helping professions.We will extend our research to the personality itself in order to differentiate individual personality traits leading to dysfunctional perfectionism and burnout.We assume that we can determine a statistically significant relationship between individual personality dimensions and dysfunctional perfectionism and burnout.

Figure no. 1 :
Figure no.1:Distribution of average values of dimensions in each of the clusters Cluster 1 -Functional perfectionists (51,74%): 89 respondents (of the total number of 172) constitute the group of functional perfectionists.In the dimensions of Concern over Mistakes, Parental Criticisms and Parental Expectations they show low average score (M CM = 25; M DA = 10; M PC = 10; M PE = 14) and in the dimension of Personal Standards they show medium average score (M PS = 21) and in the organization dimension they achieved the highest average score (M O = 26) compared to other types.Their overall average perfectionism score is average (M P = 82).Cluster 2 -Dysfunctional perfectionists (17,44%): 30 respondents (of the total number of 172) constitute the group of dysfunctional perfectionists.In the dimensions of Concern over Mistakes, Doubt about actions, Parental Criticisms, Personal Standards and Parental Expectations they show the highest average score mainly in comparison with other types (M CM = 33; M DA = 16; M PC = 14; M PS = 28, M PE = 18) and in organization dimension they show the lowest average score (M O = 17).Their overall average perfectionism score reaches high values (M P = 109) and is the highest among all types of perfectionists.Cluster 3 -Non-perfectionists (30,81%): 53 respondents (of the total number of 172) constitute the group of non-perfectionists.They achieve the lowest average score in all F-MPS dimensions (M CM = 17; M DA = 9; M PC = 8; M PS = 17, M O = 18; M PE = 10) and especially in comparison with research which, on the group of academically gifted students, identified the three types of perfectionists, as shown in Table no.1.The table shows that there are three clusters of perfectionism in our research group.The distribution of individual values of perfectionism is also shown in figure no. 1.

Table no .1
Average scores for each dimension F-MPS Source: Authors own conception

Table no . 2 :
Significance of the analysis of variance (ANOVA)

Table no . 3: Comparison
of percentage representations of individual types of perfectionists in previous research and in current research