The textbook chapter outlines a number of different factors that can influence an adolescent’s achievement outside their own abilities (e.g. psychological beliefs, school and home str

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/232515819

An investigation of perfectionism, mental health,

achievement, and achievement motivation in adolescents

Article  in  Psychology in the Schools · November 2000

DOI: 10.1002/1520-6807(200011)37:63.0.CO;2-O

CITATIONS

147 READS

926

3 authors, including:

Michael P. Accordino

Springfield College

24 PUBLICATIONS   352 CITATIONS   

SEE PROFILE

Robert B Slaney

Pennsylvania State University

67 PUBLICATIONS   4,192 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Michael P. Accordino on 27 January 2022.

The user has requested enhancement of the downloaded file.

AN INVESTIGATION OF PERFECTIONISM, MENTAL HEALTH, ACHIEVEMENT, AND ACHIEVEMENT MOTIVATION IN ADOLESCENTS

denise b. accordino

West Springfield Public Schools

michael p. accordino

Springfield College

robert b. slaney

The Pennsylvania State University

This study examined the relationship of perfectionism with measures of achievement and achieve- ment motivation and mental health aspects of depression and self-esteem in high school students. Participants were 123 tenth- through twelfth-grade students. Results of multiple regression analy- ses indicated that students’personal standards were significant predictors of academic achievement. Students’ personal standards also significantly predicted achievement motivation. Analyses of the relationship between perfectionism and depression and self-esteem found that as students’ person- al standards increased, their levels of depression decreased and self-esteem increased. Furthermore, when students experienced a discrepancy between their personal standards and actual performance, their depression levels increased and self-esteem decreased. © 2000 John Wiley & Sons, Inc.

The topic of perfectionism has been receiving increased research attention recently. Researchers have viewed some aspects of perfectionism as a healthy component of the human condition (Ashby, Bieschke, & Slaney, 1997; Ashby & Huffman, 1999; Ashby & Kottman, 1996; Ashby, Kottman, & Schoen, 1997; Parker & Adkins, 1995a, 1995b; Slaney, Ashby, & Trippi, 1995; Slaney, Mobley, Rice, Trippi, & Ashby, 1998). Parker and Adkins (1995b) point out how Adler’s claim that striving for per- fection in life is inherent is suggestive of self-actualization as construed by Maslow (1970). Ac- cording to Maslow’s view, striving for perfection through self-actualization is an indication of the absence of neurosis rather than evidence of its presence (Parker & Adkins, 1995a; Parker, 1997).

On the other hand, Horney (1950), Ellis (1962), Hollender (1965), and Missildine (1963) have all hypothesized that perfectionists may experience psychological problems. For example, there has been support that certain aspects of perfectionism are associated with depression (Alden, Bieling, & Wallace, 1994; Blatt, 1995; Hewitt & Dyck, 1986). Theoretically, people who are perfectionistic strive to be perfect and minor mistakes can be catastrophic—the reason being that mistakes tend to produce high levels of discrepancy, which may lead to depression in some circumstances (Frost, Marten, Lahart, & Rosenblate, 1990). Furthermore, when perfectionists feel that they can never meet their own expectations, self-criticism and eventual depression may be possible outcomes. In addi- tion, when students experience a discrepancy between high standards and substandard performance, they may develop low self-esteem and may evaluate themselves poorly (Burka & Yuen, 1983; Pacht, 1984). It should not be concluded, however, that perfectionism is the cause of depression and low self-esteem.

Empirical research in the area of perfectionism is limited, especially in adolescents. While some studies of perfectionism have included children, predominantly focusing on the gifted and academ- ically talented population, most of the empirical studies have utilized college students or have been directed toward particular adult clinical populations (Parker, 1997). Furthermore, empirical research on perfectionism’s link with achievement has only utilized gifted and academically talented samples (Parker & Mills, 1996; Roberts & Lovett, 1994). For example, Roberts and Lovett (1994) found that gifted students demonstrate higher levels of perfectionism and more negative affective and physio-

Psychology in the Schools, Vol. 37(6), 2000 © 2000 John Wiley & Sons, Inc.

535

Correspondence to: Denise Burdick Accordino, 26 Central St., 4th floor, West Springfield, MA 01089-2777.

logical stress reactions following failure compared to their nongifted peers. Bransky (1989) also assessed mentally gifted adolescents and reported that academic perfectionism, defined as “perfec- tionistic thinking and behavior specifically related to academic expectations, activities, and perfor- mances” (p. 7), was explained by high levels of self-expectations and a tendency toward shame following failure regardless of effort.

Perfectionism is a multidimensional construct that includes possessing high standards for per- sonal performance and striving for order (Frost et al., 1990; Slaney et al., 1995). Slaney et al. de- scribe this aspect as “Standards.” Another aspect is “Discrepancy,” which refers to perceptions of in- consistency between one’s standards and one’s performance (Slaney, Rice, & Ashby, in press). Factor analytic studies using college student samples (Frost, Heimberg, Holt, Mattia, & Neubauer, 1993; Slaney et al., 1995; Suddarth, 1996) have further suggested that there are two higher order dimen- sions of perfectionism. One factor is referred to as adaptive or healthy while the other factor has been termed maladaptive or unhealthy. Hamachek (1978) posits a related theory of perfectionism that in- cludes two ends of the continuum: normal and neurotic. Individuals with normal perfectionism tend to derive pleasure from striving to meet challenging but attainable goals. Moreover, they possess per- sonal standards for performance and can apply these standards in a flexible manner according to the requirements of a given situation and still feel satisfied if their performance is not exact. Neurotic perfectionists, on the other hand, are characterized as having high levels of anxiety and a strong fear of failure. They rarely feel that their efforts are good enough. Neurotic perfectionists are unable to experience pleasure from their efforts because they seldom feel their accomplishments are good enough to warrant that feeling. According to Hamachek, normal perfectionism contributes to greater achievement and enhanced motivation. Based on this theory, it would seem likely that higher achievement and achievement motivation would be associated with the normal or healthy dimension of perfectionism.

In a related study, Braver (1996) used a sample of 336 college students to examine the rela- tionship between perfectionism as measured by subscales of the Almost Perfect Scale–Revised (APS-R; Slaney, Mobley, Trippi, Ashby, & Johnson, 1996) and academic achievement as measured by self-reported grade point average and SAT scores. Results indicated that endorsing high standards was positively and significantly associated with college grade point average (r 5 .31, p , .001, d 5 .65) and SAT scores (r 5 .24, p , .001, d 5 .55); both of these relationships produced effect sizes in the large range (Cohen, 1988). Order and discrepancy, however, were not found to be significant. Wade (1997) also studied 246 undergraduates and explored the relationship between measures of perfectionism and status as an adult child of an alcoholic using the APS-R; results indicated corre- lations between college grade point average and standards, order, and discrepancy of .43 ( p , .001, d 5 .96), .12, and 2.00, respectively. The obtained effect size for standards was in the large range (Cohen, 1988).

In adolescents, it is apparent that certain mental health variables tend to influence their achieve- ment and achievement motivation in addition to perfectionism. Depression (Dotan, 1990; Hollon, 1970) and self-esteem (Coopersmith, 1967) are two such variables. While there is some evidence in the literature on how depression and self-esteem affect achievement and achievement motivation (Blechman, McEnroe, Carella, & Audette, 1986; Harter, 1983; Learner & Kruger, 1997), limited re- search exists on the relationship between perfectionism, achievement, achievement motivation, and aspects of mental health such as depression and self-esteem in adolescents. As a result, the purpose of the present study is to address two research questions:

1. What is the relationship between independent perfectionism variables of standards and dis- crepancy and dependent academic achievement (GPA) and achievement motivation vari- ables of mastery, work orientation, and competitiveness.

2. What is the relationship between independent perfectionism variables of standards and dis- crepancy and dependent mental health variables of depression and self-esteem?

536 Accordino, Accordino, and Slaney

Method

Participants

The participants were 123 tenth- through twelfth-grade students. Twelve percent (n 5 15) of the sample was comprised of 10th graders. Within the 10th grade there were 13.3% (n 5 2) females and 86.7% (n 5 13) males. Thirty-eight percent (n 5 47) of the sample was comprised of 11th graders. Females accounted for 72.3% (n 5 34) while males accounted for 27.7% (n 5 13) of the sample. The remaining 50% (n 5 61) of the sample were 12th graders. Females accounted for 45.9% (n 5 28) and males accounted for 54.1% (n 5 33). Students ranged from 14 to 21 years of age (M 5 17.22, SD 5 1.25). Regarding education services, 64.2% (n 5 79) received regular education. Of students receiving special education services, 15.4% (n 5 19) received gifted support/enrichment services, 10.6% (n 5 13) received learning support services, 6.5% (n 5 8) received emotional sup- port services, and 3.3% (n 5 4) received speech and language support. In terms of ethnicity, there were a total of 117 (95.2%) Whites, 3 (2.4%) African Americans, 2 (1.6%) Hispanic Americans, and 1 (0.8%) Native American. The sample consisted primarily of White, middle-class, and rural stu- dents.

Procedures

Parental and child informed consent forms along with a description of the study were sent to 885 tenth- through twelfth-grade students at eight junior/senior high schools within Pennsylvania and Virginia. Potential participants were informed that the purpose of the study was to examine the relationship between achievement and various personality traits. Consent included permission to ac- cess student records to obtain GPA. Participants were then given an envelope containing five self- report measures and a demographic questionnaire. To ensure confidentiality, the measures were numbered prior to distribution and participants were instructed not to include their name or other identifying information other than that requested on the demographic questionnaire. Counterbal- anced packets of self-report measures were administered on site and took approximately 20 min to complete. The measures were counterbalanced so that the correlations would not be inflated due to the contiguity of the instruments (Rude & Burnham, 1993).

Approximately 14% (n 5 123) of the solicited participants returned the signed consent forms and participated in the study. Participants were instructed to enclose completed measures in the en- velope. Upon returning the envelope, the investigator requested the participant’s name which was then eliminated from the list of participants. At that time, the participant’s previously calculated cu- mulative GPA based on academic records was placed in the participant’s envelope.

Measures

Work and Family Orientation Questionnaire. The Work and Family Orientation Question- naire (WOFO; Helmreich & Spence, 1978; Spence & Helmreich, 1983) is a multidimensional mea- sure of achievement motivation and attitudes toward family and career. The scale, which was first developed in 1978 by Helmreich and Spence, was revised in 1983 and now consists of 19 items that assess achievement motives and 9 items that assess personal concerns (e.g., educational aspirations, salary, prestige). Items are answered on a 5-point Likert scale ranging from 1 (strongly agree) to 5 (strongly disagree). This scale is inversely rated: lower scores indicate higher achievement motiva- tion. The possible range of scores for each subscale is from 19 to 95. The scale consists of three fac- tors of achievement motivation: (a) Mastery, (b) Work Orientation, and (c) Competitiveness. The Mastery factor contains items reflecting a “preference for difficult, challenging tasks and for meet- ing internally prescribed standards of performance excellence” (Spence & Helmreich, 1983, p. 41). An example of an item from the Mastery scale is “I prefer to work in situations that require a high level of skill.” The Work Orientation factor contains items reflecting “the desire to work hard and to

Perfectionism, Mental Health, Achievement 537

do a good job of what one does” (Spence & Helmreich, 1983, p. 41); it also represents a dimension of effort. An example of an item from this scale is “I find satisfaction in working as hard as I can.” The Competitiveness factor describes “the desire to win and be better than others” in interpersonal situations (Spence & Helmreich, 1983, p. 41). An example of an item from the Competitiveness scale is “It is important for me to perform better than others on a task.” The coefficients of internal con- sistency for the subscales range from .61 for Mastery to .76 for Competitiveness. Evidence of con- struct validity was also found (Spence & Helmreich, 1983). Lastly, the WOFO items are suitable for a high school aged population (J.T. Spence, personal communication, July 27, 1997).

Cumulative GPA. Grade point average (GPA) is the operationalization of the construct of achievement. It is defined as the mean of grades earned in all subject areas. Students’numerical grades from ninth grade to the present year were used and the grading scale was from 0 to 100: 92–100 5 A, 84–91 5 B, 76–83 5 C, 70–75 5 D, and 0–69 5 F. All high schools surveyed reported using weight- ed grades when determining grade point averages for the students. Weighted grades were determined by assigning more weight to a difficult course (e.g., Honors English, Advanced Math) than a general course (e.g., Basic English). An average was used because a measure of academic performance based on several courses or a full year or more of course work seemed more reliable than a measure based on performance in a single course. In describing GPA for each grade level, 10th graders (n 5 15) had a mean of 87.63 and a standard deviation of 2.06, 11th graders (n 5 47) had a mean of 89.89 and a stan- dard deviation of 1.14, and 12th graders (n 5 61) had a mean of 86.77 and a standard deviation of 1.04. GPAs were not controlled for across years or number of courses completed.

Almost Perfect Scale–Revised. The APS-R (Slaney et al., 1996) provided the measure of per- fectionism in this study. The APS-R contains 59 items; there is also a short form consisting of 23 items. The participant responds to one of seven choices ranging from 1 (strongly disagree) to 7 (strongly agree). Due to the wording of the items, higher scores on this scale indicate higher levels of perfectionism. The APS-R consists of three subscales: (a) Standards, (b) Order, and (c) Discrep- ancy. Only the Standards and Discrepancy scales were used in this study. The Standards scale re- flects the level of standards people set for themselves and is composed of seven items (items 1, 5, 8, 12, 14, 18, 22). The Discrepancy scale reflects the amount of distress people feel in regard to their personal standards and is composed of 12 items (items 3, 6, 9, 11, 13, 15, 16, 17, 19, 20, 21, 23). The short form of the APS-R was used in this study; therefore the remaining 36 items were not adminis- tered. The following are examples of items taken from the APS-R: “I try to do my best at everything I do” (Standards scale) and “My performance rarely measures up to my standards” (Discrepancy scale). There is no absolute cut-off for low or high Standards or Discrepancy scores currently. The coefficients of internal consistency for the subscale scores are as follows: Standards alpha 5 .87 and Discrepancy alpha 5 .92 (Suddarth, 1996). Test–retest correlations have not been reported for this version. Results of a confirmatory factor analysis revealed item pure factor loadings ranging from .49 to .86 indicating excellent convergent validity (Slaney et al., 1998). Because the APS-R is an in- strument whose scores have not been validated in the adolescent population, three readability indices were calculated. The Flesch–Kincaid analysis (Microsoft Word 2000, 1999) yielded a grade level for reading of 5.5. The SMOG index (calculated by adding 3 to the square root of the number of items with 3 or more syllables per 100 words) yielded an average reading level of 5.985. The Raygor Read- ability Index provided the third reading grade level. The Raygor index recommends basing the in- dex on three 100-word passages. The reader should be cautioned that the APS-R consists of only two 100-word groups. Nonetheless, the computation resulted in a valid grade index of 4 (Readence, Bean, & Baldwin, 1998).

Reynolds Adolescent Depression Scale. The Reynolds Adolescent Depression Scale (RADS; Reynolds, 1986b) is designed to assess symptomatology associated with depression rather than to

538 Accordino, Accordino, and Slaney

provide a diagnosis of a specific and definitive depressive disorder (Reynolds, 1986a). It provides a thorough sampling of depressive symptoms that are included in the Diagnostic and Statistical Man- ual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994) and other nosologies for children ages 13 to 18 (Note: while the age range of participants in this study was 14 to 21, it was believed by the authors that this instrument would adequately measure depression in this sample.) The RADS consists of 30 items and the client responds to one of four choices ranging from 1 (al- most never) to 4 (most of the time). The possible range of scores is from 30 to 120. Lower scores in- dicate less depressive symptomatology. An example of an item from this scale is “I feel sorry for my- self.” The internal consistency estimate reported is .92 and a test–retest coefficient of .80 is reported for a 6-week interval (Reynolds, 1986a). Content validity for the RADS has been established. Cri- terion-related and construct validity have also been demonstrated by the RADS’ correlations with other self-report depression measures and related constructs (Reynolds, 1986b).

Rosenberg Self-Esteem Scale. The Rosenberg Self-Esteem Scale (RSE; Rosenberg, 1965) is the most widely used unidimensional measure of global self-concept in adolescents (Blascovich & Tomaka, 1991). The RSE consists of 10 items and is scored on a Likert scale ranging from 1 (strong- ly agree) to 4 (strongly disagree). The possible range of scores is from 10 to 40. Low scores on this scale are indicative of high self-esteem. Reliability estimates for the total RSE scale using alpha co- efficients range from .77 (Dobson, Goudy, Keith, & Powers, 1979) to .88 (Fleming & Courtney, 1984). Considerable support for convergent and discriminant validity has been demonstrated with the RSE (Blascovich & Tomaka, 1991).

Demographic Questionnaire. The Demographic Questionnaire (DQ) is an instrument devel- oped by the first author to obtain the following demographic information: (a) age, (b) grade in school, (c) gender, (d) race, (e) type of support services currently received, (f ) retentions, (g) parental sta- tus, and (h) GPA. These characteristics were obtained to provide descriptive information about the sampling design and aid in generalizing results to the population; however, they were not included in the analyses or used to exclude any participant from taking part in the study.

Data Analysis

Six multiple regression analyses were conducted. The first four regression analyses looked at the relationship between independent predictor variables of dtandards and fiscrepancy and depen- dent variables of achievement (GPA) and achievement motivation variables of mastery, work orien- tation, and competitiveness. Two additional analyses were also conducted to assess how well inde- pendent variables of standards and discrepancy predicted dependent mental health variables of depression and self-esteem. An alpha of .05 was used for all statistical tests. Given that six statisti- cal tests were run, the adjusted alpha used to reject the null hypothesis was .05/6 or a 5 .008.

A power analysis was conducted to determine the necessary sample size to reject the null hy- pothesis as well as to determine the probability that each statistical test would produce statistically significant results (Cohen, 1988; Cohen & Cohen, 1983). Using the appropriate tables in Cohen and Cohen (1983), it was determined that, in order to achieve a power level of .90 and detect an effect size as small as .11 (small to medium range), a sample size of n 5 117 was needed to reject the null hypothesis at alpha of .05. The sample size obtained for the statistical analyses ranged from n 5 113 to n 5 119; therefore, the power level for the analyses was approximately .90.

The Variance Inflation Factor (VIF) method was used to detect the presence of multicollinear- ity. The VIF method measures how much the variances of the estimated regression coefficients are inflated compared to when the predictor variables are not linearly related (Neter, Kutner, Nachtstein, & Wasserman, 1996). In this study, individual VIF values greater than 10 were considered indica- tive of multicollinearity because it suggests that independent variables have highly shared elements.

Perfectionism, Mental Health, Achievement 539

No variables had individual VIF values greater than 2, indicating that multicollinearity was most likely not present.

Results

Internal Consistency of Measures

Internal consistency was assessed on all of the measures. Cronbach alphas were obtained for each; results are as follows:

1. Almost Perfect Scale – Revised A. Standards scale: Cronbach’s alpha 5 .87 (n 5 122) B. Discrepancy scale: Cronbach’s alpha 5 .90 (n 5 120)

2. Rosenberg Self-Esteem Scale Cronbach’s alpha 5 .78 (n 5 121) 3. Reynolds Adolescent Depression Scale Cronbach’s alpha 5 .91 (n 5 116) 4. Work and Family Orientation scale

A. Mastery scale: Cronbach’s alpha 5 .35 (n 5 121) B. Work Orientation scale: Cronbach’s alpha 5 .86 (n 5 122) C. Competitiveness scale: Cronbach’s alpha 5 .76 (n 5 122)

Most

Collepals.com Plagiarism Free Papers

Are you looking for custom essay writing service or even dissertation writing services? Just request for our write my paper service, and we'll match you with the best essay writer in your subject! With an exceptional team of professional academic experts in a wide range of subjects, we can guarantee you an unrivaled quality of custom-written papers.

Get ZERO PLAGIARISM, HUMAN WRITTEN ESSAYS

Why Hire Collepals.com writers to do your paper?

Quality- We are experienced and have access to ample research materials.

We write plagiarism Free Content

Confidential- We never share or sell your personal information to third parties.

Support-Chat with us today! We are always waiting to answer all your questions.