Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

An Examination of the Relationship of the 16PF Fifth Edition to a Multidimensional Model of Self-Esteem:

An Examination of the Relationship of the 16PF Fifth Edition to a Multidimensional Model of... The regression equation needed to predict the self-esteem score from the Sixteen Personality Factor Questionnaire (16PF) Fifth Edition is not available for hand scoring. This score is derived from the Coopersmith’s Self-Esteem Inventory (SEI), which is based on a unitary model of self-esteem. The SEI may be subject to gender and extroversion biases. Moreover, there is a gap in the literature on the 16PF fifth edition and self-esteem. For these reasons, we investigated the relationship of the Culture-Free Self-Esteem Inventory–II Form AD (CFSEI-II), a multidimensional model of self-esteem, using multiple regression to the scales of the 16PF. Based on the adult sample of 234 in this study, we developed and made available to users of the 16PF split sample–validated regression equations for this multidimensional model of self-esteem. Keywords Sixteen Personality Factor Questionnaire Fifth Edition, self-esteem, model, personality, statistical analysis The Sixteen Personality Factor Questionnaire Fifth Edition to how similar or dissimilar they perceive themselves to be in (16PF), released in 1994 (Institute for Personality and Ability relationship to the item (Stark, Chernyshenko, Drasgow, & Testing [IPAT], 2009), is the latest edition of the personality Williams, 2006). These 16 primary factors are considered test, which was devised to measure normal personality func- primary personality traits (Cattell & Schuerger, 2003). tions (Cattell, Eber, & Tatsuoka, 1970). Since its origin, Primary personality traits contribute to five second-order or, developers have revised the 16PF 5 times, updating content, more commonly, global traits (see Table 1). Thus, the scales norms, and/or to make other improvements, all of which of the 16PF, 16 primary and 5 global, reflect an underlying added to its overall psychometric qualities (IPAT, 2009). multitiered conceptualization of personality. Furthermore, There are also validity scales, such as impression manage- Cattell theorized third-order factors, as well, and only ment. Reviewers have consistently rated the 16PF one of the recently has reliable evidence of a two-factor solution been most used and researched personality tests (Dancer & Woods, established (Cattell & Mead, 2008). 2006; Walsh & Betz, 1995). This ubiquity may be traced to The 16PF Fifth Edition may be administered via paper its empirical roots, multitiered trait structure (including Big and pencil or computer. A number of computer-generated Five personality traits) of normal-range personality, and reports can be derived from the scale scores, including the availability in more than 30 languages (Cattell & Schuerger, Personal Career Development Profile (Clark & Blackwell, 2003). 2007). Historically, the 16PF has been a popular personality The 16PF was originally constructed by factor analyzing measure because of the multiple special scores that can be descriptions of personality based on English-language adjec- calculated from combinations of various factor scores. tives (Cattell et al., 1970; IPAT, 2009). The factor analysis One special score that can be calculated from the 16PF yielded 16 primary factors (see Table 1). One of these fac- Fifth Edition is a measurement of unitary self-esteem. Using tors, Reasoning (B), is a proxy for measuring cognitive or reasoning ability. All items and associated traits are assumed University of Idaho, Moscow, USA to exist on continua and thus are conceptually bipolar. For Kent State University, OH, USA example, descriptors for the primary factor Warmth (A) North Dakota State University, Fargo, USA range from “reserved” to “attentive to others.” This underly- Corresponding Author: ing structure (i.e., each trait and its associated items existing James S. Korcuska, North Dakota State University, Dept. 2625, P.O. Box along a continuum) assumes a dominance response process. 6050, Fargo, ND 58108-6050, USA. In a dominance process, a person responds to an item relative Email: james.korcuska@ndsu.edu Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License (http://www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open Table 1. Means and Standard Deviations for 16PF Fifth Edition Researchers continue to debate whether self-esteem is a and Culture-Free Self-Esteem Inventory–II. unitary or multidimensional construct. Many professionals conceptualize self-esteem as multifaceted (Battle, 1992; a a Scale n M SD Luhtanen & Crocker, 1992; Rosenberg, Schooler, Schoenbach, 16PF Fifth Edition & Rosenberg, 1995). Battle’s (1992) multidimensional model A Warmth 234 6.3 1.8 includes dimensions of self-esteem based on (a) overall per- B Reasoning 233 6.2 1.7 ceptions of self-worth, (b) perceptions of relationship quality, C Emotional Stability 234 5.6 1.8 and (c) evaluation of intimate sense of self-worth. Thus, evalu- E Dominance 234 5.2 2.0 ations of self-worth refer to perceptions of worthiness as a per- F Liveliness 234 6.0 1.9 son in comparison with others. The intimate sense of self-worth G Rule-Consciousness 234 5.4 1.9 dimension refers to the evaluation of self-worth independent H Social Boldness 233 5.7 2.0 of others’ evaluations of it. In addition, Rosenberg et al. (1995) I Sensitivity 234 6.4 1.8 found a stronger correlation of global self-esteem to psycho- L Vigilance 234 5.3 2.0 logical well-being than with a specific or unitary conceptual- M Abstractness 234 5.7 1.9 ization, further indicating that a multidimensional model of N Privateness 234 5.0 2.1 self-esteem might hold more utility than a unitary conceptual- O Apprehension 234 5.9 1.5 ization. Intimate evaluation of self-worth might have a stron- Q1 Open to Change 234 5.7 2.1 ger relationship to the perception of mental health and for the Q2 Self-Reliance 234 5.3 1.9 expectation of positive treatment outcomes than a sense of Q3 Perfectionism 234 5.0 2.0 social self-esteem or more general sense of self-worth (Battle, Q4 Tension 234 5.5 1.8 1992). Extroversion (EX) 234 6.1 1.9 There are several reasons for examining how the 16PF Anxiety (AX) 234 5.6 1.7 Fifth Edition relates to multidimensional models of self- Tough-Mindedness (TM) 234 4.7 2.0 Independence (IN) 234 5.5 1.9 esteem. Today’s mental health environment, Health Self-Control (SC) 233 5.1 1.7 Maintenance Organizations (HMOs), and insurance compa- Culture-Free Self-Esteem Inventory–II Form AD nies provide less time and money for counselors to use mul- General 234 52.3 8.3 tiple assessment tools. Clinicians must use assessment tools Personal 234 51.6 9.9 efficiently and economically. If valid special scores for a Social 234 54.4 8.9 multidimensional model of self-esteem for the 16PF Fifth Total 234 52.3 8.0 Edition can be developed, its use would be enhanced, increas- ing efficiency in the data-gathering process. Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition. To date, there has been limited research using the current Values rounded to single decimal place. The 16PF scores are reported in Sten scores. A Sten has a mean of 5.5 and a standard deviation of 2.0. The edition of the 16PF Fifth Edition with unitary models of self- Culture-Free Self-Esteem Inventory–II Form AD (CFSEI) is reported using esteem, a difference between the fifth edition and earlier ver- T scores. sions (Rieke & Conn, 1994). Rieke and Conn examined the relationship between the 16PF Fifth Edition and a unitary model of self-esteem. In a sample of 318 largely non-White a regression equation, combinations of 16PF Fifth Edition subjects (56%), higher scores were found on the global fac- factor scores are entered to predict a measure of self-esteem tors of IN (Independence) and EX (Extroversion) with a based on the Coopersmith’s Self-Esteem Inventory (SEI; lower score on AX (Anxiety) being correlated to higher self- Coopersmith, 1981). Currently, the regression equation nec- esteem. Correlations at p < .05 for global factors were AX essary to predict the SEI score from the 16PF is not available (Anxiety) = −.64, IN (Independence) = .49, EX (Extroversion) for hand scoring. According to sources at the Institute for = .45, and SC (Self-Control) = .26. The primary factors asso- Personality and Ability Testing (IPAT), there is no intention ciated with higher self-esteem on the current edition included of making this formula available apart from their computer correlations on A (Warmth) = .34, C (Emotional Stability) = scored reports (S. Conn & Institute for Personality and Ability .64, H (Social Boldness) = .54, and negative correlations on Testing Staff, personal communication, August 24, 1994). N (Privateness) = −.20 and Q1 (Openness to Change) = .28, The formula used to calculate this computer-generated spe- along with lower scores on O (Apprehension) = .58, and M cial score was derived using a cross-validated multiple regres- (Abstractness) = .44. These results, however, warrant cau- sion procedure described by Rieke and Conn (1994). The tion. The model of self-esteem upon which the Coopersmith’s regression equation that resulted from their work provided a SEI is built has shown biases in gender (Chapman & Mullis, r of .72 using the five global factors and an r of .76 for the 2002) and extroversion (Francis, 1997). primary factors. Rieke and Conn (1994), citing Kawash, In addition to the limited research on unitary models of Clews, and Keating (1989), chose a unitary model of self- self-esteem and the recent edition of the 16PF, a search of esteem, an important choice as we were interested in multi- ERIC and PsychInfo found no other studies examining the dimensional constructs of self-esteem. Pietrzak et al. 3 relationship between the 16PF Fifth Edition and other mod- edition was revised using modern psychometric procedures els of self-esteem. Thus, an extension of the research exam- and norm collecting procedures to update language and ining the relationship of the 16PF Fifth Edition to multiple improve the psychometric qualities of the tool. The 16PF dimensions of self-esteem is appropriate. The primary pur- contains 16 bipolar scales and several validity scales. Fifteen pose of this study was to examine the relationship of the of the scales measure personality traits and one scale mea- 16PF Fifth Edition to the four dimensions of self-esteem sures cognitive or reasoning abilities (see earlier discussion). measured by the Culture-Free Self-Esteem Inventory–II The stability coefficients for the personality and validity Form AD (CFSEI-II) with an adult sample using multiple scales range from r = .69 to r = .86 for a 2-week period with regression. A secondary purpose of the study was to begin a range of r = .56 to r = .79 for a 2-month period. The internal developing split sample–validated regression equations for consistency of the personality and validity scales ranged this multidimensional model of self-esteem and make them from r = .66 to r = .87 (Rieke & Conn, 1994). available to users of the 16PF Fifth Edition. In doing so, In addition to the 16 primary factors, 5 global factors can users would have a choice of self-esteem models. be obtained. The global factors are calculated from multiple regression equations developed from second-order factor analyses of the primary factors (Cattell et al., 1970; Rieke & Method Conn, 1994). While no internal consistency indices can be calculated for the global factors, the stability of these scales Participants ranges from r = .84 to r = .91 over a 2-week period, and r = The sample used in this study consisted of 234 volunteers’ .70 to r = .82 over a 2-month period (Rieke & Conn, 1994). responses. Of these, 109 (47%) were from the Southern Indiana, Southern Illinois, or Northern Kentucky area, 58 CFSEI-II. The CFSEI-II is a widely used multidimensional (24%) were from the Northeastern Ohio or Northwestern measure of self-esteem. James Battle’s theory of self-esteem Pennsylvania area, and 67 (29%) were from Southern (Battle, 1992) guided the instrument’s development. Admin- Florida. Participants’ age ranged from 18 years to 69 years. istration time is between 10 and 15 min. The adult form of The mean age was 27.9 years (SD = 10). Two participants the CFSEI-II has 40 items. Respondents’ yes or no item included in this study did not report their age. choices fall into either a high or low self-esteem group. Items The participants were largely female (71%; n = 165; form three scales measuring aspects of self-esteem (General males; 29%; n = 69). The primary researcher solicited par- Self-Esteem, Social Self-Esteem, and Personal Self-Esteem) ticipants through minority organizations in the community and a scale to detect “faking good.” The three self-esteem and various class levels at multiple university locations. scale scores can also be summed to obtain a global measure These procedures resulted in the reported racial heritage of of self-esteem (Battle, 1992). The General Self-Esteem score the sample participants consisting of 82% (n = 193) White serves as a measure of an adult’s “overall perceptions of non-Hispanics, 9% (n = 20) African Americans, 2% (n = 5) worth” (p. 3). The Social Self-Esteem score is an assessment Asian Americans, 6% (n = 15) Hispanics, and 0.4% (n = 1) of that “aspect of self-esteem that refers to individuals’ per- Native American. Participants reported educational levels ceptions of the quality of their relationships with peers” ranging from an 11th-grade education to doctorates. The (p. 3). The Personal Self-Esteem score was designed to mea- mean educational level was 14.9 years (2 years of college) sure “the aspect of self-esteem that refers to individuals’ with a standard deviation of 2.4 years. One person did not most intimate perceptions of self-worth” (p. 4). Summing report his or her educational level. these scores yields a cumulative, or more global, measure of There were 59% (n = 138) undergraduate and graduate stu- overall self-esteem. dents and 41% (n = 96) community members (not enrolled in The stability of the CFSEI-II scale scores was reported as classes) in the sample. A diverse sample of university partici- r = .56 for the Social score, r = .78 for the Personal score, r = pants was solicited from undergraduate programs in art, music, .82 for the General score, and r = .82 for the Total score over accounting, business management, counseling, elementary a 4-week period (Battle, 1977b). Research regarding the education, secondary education, and special education. The validity of the CFSEI-II has been supportive. There is evi- counseling participants were graduate students. No remunera- dence of concurrent validity with the Coopersmith’s’ SEI, tion or other incentive was provided to the participants. with correlations of .71 to .80 (Battle, 1992). A number of studies have examined the validity of the CFSEI-II scores as correlates of depression (Brooke, 1995). For example, the Materials Beck Depression Inventory (overall: −.55; male: −.53, 16PF. The 16PF Fifth Edition is an empirically derived per- females: −.56; Battle, 1978). Other validity research has sonality test devised to measure normal personality func- included content analysis, factor analysis, and the relationship tions. It was originally constructed by factor analyzing of CFSEI-II scores to external criteria, including its relation- English-language adjectives used to describe personality ship to teachers’ ratings and the Minnesota Multiphasic (Cattell, 1989; Cattell et al., 1970; IPAT, 2009). The fifth Personality Inventory (Battle, 1977a). There is inadequate 4 SAGE Open support for the claim, however, that the CFSEI-II is culture Table 2. Correlation Between16PF Fifth Edition and Culture- Free Self-Esteem Inventory–II (N = 234). free (Brooke, 1995). Culture-Free Self-Esteem Inventory–II Form Procedure 16PF Fifth Edition Total General Social Personal The primary author collected the data for this study as part of a group of studies on the 16PF Fifth Edition. Sample size A Warmth .16* .19* .15* .04 was preestablished with a minimum of 170 participants. The B Reasoning .03 .05 .02 −.02 sample of this size allows for 10 or more cases per variable C Emotional Stability .63* .47* .37* .63* when calculating multiple regressions, and is the ratio experts E Dominance .34* .36* .23* .18* recommend (Hair, Babin, & Anderson, 2009; Janda, 1998; F Liveliness .24* .18* .31* .15* Nunnally & Bernstein, 1994). G Rule-Consciousness .02 −.03 .07 .04 The administration packets included the 16PF, CFSEI-II, H Social Boldness .51* .51* .36* .29* demographics sheet, and other instruments placed in an I Sensitivity −.04 −.01 −.05 −.07 L Vigilance −.26* −.20* −.18* −.23* envelope in counterbalanced order. Before opening the M Abstractness −.18* −.11 −.21* −.18 administration envelope, each participant read or heard a N Privateness −.27* −.33* −.18* −.09 description of this study. The researchers or research assis- O Apprehension .34* .36* .23* .18* tants distributed 260 packets. Sixty percent of the partici- Q1 Open to Change .31* .30* .15* .22* pants completed the materials in a group setting and the other Q2 Self-Reliance −.27* −.21* −.34* −.16* 40% did so at home. Participants returned the packets to one Q3 Perfectionism −.00 .02 −.03 −.02 of the researchers or a research assistant. Of this number, Q4 Tension −.23* −.12 −.09 −.35* participants returned 253 packets. The returned packets Extroversion (EX) .39* .38* .37* .19* resulted in 234 usable 16PF Fifth Edition and CFSEI-II sur- Anxiety (AX) −.34* −.19* −.18* −.44* vey instruments that were included in this study for analysis. Tough-Mindedness (TM) −.12 −.16* −.02 −.04 There was a 90% return rate. Independence (IN) .42* .45* .27* .23* The materials were hand scored separately and checked Self-Control (SC) .01 −.01 .01 .03 for accuracy twice, sometimes thrice, by different members of the research team. To estimate the degree of scoring error, Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition. *p < .05. researchers randomly selected 10 protocols and rescored them by hand. No errors were found in these 10 protocols. for each randomly generated subsample. Subsample regres- Data Analysis sion models may then be compared with each other, as well as 2 2 to the overall model using R , adjusted R , and the standard Statistical analyses were performed using SYSTAT 9.0 for error of measurement. The goal is to determine which indi- Windows (SPSS, 2007). First, an analysis of the correlations vidual variables across models demonstrate the highest level between the 16PF Fifth Edition and CFSEI-II was conducted of similarity, which is determined by beta weight. These (see Table 2). Next, a forward stepwise multiple regression variables govern the composition of the overall model. procedure with a .15 entry and .15 removal level was uti- We divided the total sample (N = 234) into two groups by lized, paralleling methods in previous research on the 16PF generating a random number for each case using SYSTAT (Rieke & Conn, 1994). Unlike the cross-validation method 9.0’s pseudo-random number generator. This number had a utilized by Rieke and Conn (1994), we utilized split sample normal distribution with a mean of 0 and a standard devia- validation to validate the regression models. Our purpose tion of 1. The sample was then split into two groups at the was to examine the relationship of the 16PF Fifth Edition to mean. The first of these groups contained 52% (n = 121) of a measure of multidimensional model of self-esteem and not the cases and the second contained 48% (n = 113) of the to directly compare the results of this study with those of cases. For convenience, these subsamples were called Split 1 Rieke and Conn. Therefore, we did not rigidly follow their and Split 2, respectively. Although other ratios are possible, analytic protocol. we chose the one most frequently employed, which is to split We followed guidelines for stepwise multiple regression the sample into near equal parts. model development and validation described by Hair, We calculated initial equations using a stepwise proce- Anderson, Tatham, and Black (1998; Hair et al., 2009), which dure, including the total sample (see Table 3). These initial utilizes split sample validation or split-half validation. In this equations were then compared with two equations calculated validation procedure, a full sample regression model is com- using the split samples. Predictors that entered into each of pared across two subsample regression models. To create the the three equations are then examined to see if (a) the same subsamples, the sample is divided into half, typically using a ran- predictors entered into each of the three equations and (b) the dom number generator. Next, regression models are estimated Pietrzak et al. 5 Table 3. Global Factor Regression Results of CFSEI From 16PF. 2a Scale group r r F EX AX TM IN SC Total score Total .70 .48 72.81** −.62 .18 .71 Split 1 .70 .48 38.05** −.55 .70 .16 Split 2 .71 .48 35.89** −.66 .23 .69 Final .68 .46 98.45** −.58 .62 General score Total .60 .35 31.92** .09 −.41 .16 .60 Split 1 .64 .39 26.65** −.40 .68 .17 Split 2 .58 .31 13.74** .15 −.40 .18 .53 Final .59 .34 −.42 .59 Social score Total .47 .21 16.05** .21 −.28 .19 .36 Split 1 .57 .30 13.58** .24 −.36 .26 .47 Split 2 .42 .15 5.75** .20 −.24 .16 .31 Final .47 .21 16.05** .21 −.28 .19 .36 Personal score Total .65 .41 40.85** −.16 −.72 .18 .61 Split 1 .60 .34 16.59** −.21 −.68 .15 .61 Split 2 .69 .46 33.27** −.72 .23 .55 Final .63 .39 50.71** −.65 .19 .53 Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition; CFSEI = Culture-Free Self-Esteem Inventory; EX = Extroversion; AX = Anxiety; TM = Tough-Mindedness; IN = Independence; SC = Self-Control. a 2 This represents the adjusted r . Total N = 234, Split 1 n = 121, Split 2 n = 113, and Final N = 234. **p < .001. standardized beta weight for the predictors had the same sign more, lower than the standardization sample of the 16PF (positive or negative) across the three equations. A final Fifth Edition on N (Privateness) and TM (Tough- equation was constructed using only those predictor vari- Mindedness). These differences are similar to those of the ables that were stable across all three regression models. The sample reported by Rieke and Conn (1994) in their work final overall equation (if one can be constructed) was deter- with the 16PF Fifth Edition and self-esteem. The manual for mined when it demonstrated statistically significant improve- the CFSEI-II (Battle, 1992) contains the means and standard ment in prediction over chance. Statistical significance of the deviations for the various scales and total score based upon final equation was evaluated using the F statistic. Practical the standardization sample. The descriptive statistics for the significance was examined using the adjusted r , which pro- current sample was similar to that of the CFSEI-II, with all vides an estimate of the amount of shared variance between scores within one half standard deviation of the mean. a set of predictors and the dependent variable. A criterion of The correlations between the 16PF Fifth Edition and the an adjusted r reaching at least .50 was set, so that the equa- CFSEI-II appear in Table 2. These data indicate that three of tion had at least 50% shared variance with the variable being four global factor scales (EX, AX, and IN) correlated moder- predicted. This level of shared variance is considered as the ately or low to CFSEI-II total and subscales. Two of the pri- minimum acceptable value for alternate forms of measure- mary scales comprising the global factor scales AX (Anxiety) ment of any given construct (Cicchetti, 1994; Hair et al., and Q4 (Tension) had negative correlations with self-esteem: 2009; Janda, 1998). As these scores increased, scores on CFSEI-II decreased. Results of the multiple regression equations from the global factors are summarized in Table 3. In predicting the Results CFSEI-II Total score, the primary factors of AX (Anxiety) The means and standard deviations for the 16PF Fifth Edition and IN (Independence) were stable for all three equations. and the CFSEI-II are provided in Table 1. These results indi- The equations entered both factors and each in the same cate that the sample used in this study was one half standard direction. The final equation reached statistical, but not prac- deviation, or more, higher than the standardization sample of tical significance. In predicting the CFSEI-II General scale the 16PF Fifth Edition on A (Warmth), B (Reasoning), F score, the global factors of AX (Anxiety) and IN (Liveliness), I (Sensitivity), and EX (Extroversion). In addi- (Independence) were stable for all three equations. Each fac- tion, the sample was at least one half standard deviation, or tor was entered by the equations and in the same direction. 6 SAGE Open The final equation was statistically significant but failed to (Extroversion) global factor did not contribute centrally to reach practical significance. The three equations predicting the prediction of global self-esteem. Furthermore, the only the CFSEI-II Social scale score entered the global factors of consistent observed role for EX (Extroversion) was to predi- EX (Extroversion), AX (Anxiety), TM (Tough-Mindedness), cate the CFSEI-II Social scale score, the scale measuring a and IN (Independence) in the same direction. The final equa- person’s perceptions of the quality of his or her relationships. tion was statistically significant, but failed to reach practical However, these differences may have been due to the differ- significance. In predicting the CFSEI-II Personal scale score, ences between the ways the CFSEI-II and the SEI measure the global factors of AX (Anxiety), TM (Tough-Mindedness), self-esteem. The SEI may have a bias toward extroversion and IN (Independence) were stable for all three equations. (Francis, 1997). There are correlations, however, between Each was entered by the equations and in the same direction. EX and all scales of the CFSEI-II (see Table 2), albeit small The final equation was statistically significant, but not prac- to moderate ones. Participants in our study were mostly tically significant. female (71%). At least two studies indicate a gender bias in The multiple regression results from the primary factors the items comprising the SEI (Chapman & Mullis, 2002; are summarized in Table 4. Prediction of the CFSEI-II Total Francis & James, 1998). To clarify these findings, we sug- scale score revealed the primary factors C (Emotional gest further research. Stability), H (Social Boldness), and Q1 (Openness to In examining the relationship of the 16PF Fifth Edition to Change) as stable in all three equations. The final equation other facets of self-esteem measured by the CFSEI-II, several was both statistically and practically significant. In predict- findings are worth mentioning. The 16PF Fifth Edition global ing the CFSEI-II General score, the primary factors C factor equations for the CFSEI-II Total scale score and General (Emotional Stability) and H (Social Boldness) were stable score had the same predictor variables of AX (Anxiety) and IN for all three equations. Each was entered and in the same (Independence). It is unclear from this research if this is due to direction. The final equation was statistically significant but the overlap of these constructs as measured by the CFSEI-II. not practically significant. No final equation was constructed Alternately, this finding may indicate that these constructs for the prediction of the CFSEI-II Social scale score because share an underlying dimension with the 16PF Fifth Edition. no primary factors were stable across all three equations. Still, it seems possible that there are differences between the Prediction of the CFSEI-II Personal scale score showed that constructs. This possibility is suggested by the finding that primary factor C (Emotional Stability) was stable for all the amount of variance accounted for is approximately 35% in three equations. The equations entered each factor and in the the case of the General scale score, while the Total score shares same direction. The final equation was statistically signifi- approximately 46% of the variance. Further research is neces- cant but failed to reach practical significance. sary and may clarify this issue. The regression equation for CFSEI-II Social scale score added other 16PF Fifth Edition factors beyond the inverse Discussion relationship between AX (Anxiety) and IN (Independence). The primary purpose of this study was to explore the rela- In the equation for the Social scale, the 16PF factors of EX tionship of the 16PF Fifth Edition to a multidimensional (Extroversion) and TM (Tough-Mindedness) were added. model of self-esteem with an adult sample. Although all of This suggested that while the Social scale shared the lower the final regression models based on the global factors AX (Anxiety) and greater IN (Independence) factors found reached statistical significance, they did not exceed the crite- in the other dimensions of self-esteem, the factors of ria set for practical significance. However, the global factor increased social participation and a more tough-minded equation for CFSEI-II Total scale score did approach practi- stance on issues were included. The regression equations for cal significance (adjusted r = .46). The prediction of the the CFSEI-II Personal score were similar to the equations for CFSEI-II Personal scale score had the next highest adjusted the Social scale, minus the social participation factor. r (.39). The third highest prediction of the CFSEI-II General Prediction of three of the four scores of the CFSEI-II was scale score had an adjusted r (.34). The CFSEI-II Social possible using the primary factors of the 16PF Fifth Edition. scale score had the lowest adjusted r (.21) of all the global However, only one of the final multiple regression equations factor equations. exceeded 50% shared variance with the CFSEI-II. This final Notably, all of the final equations shared the same central regression equation was for the CFSEI-II Total score com- elements of lower scores on AX (Anxiety) coupled with posed of the 16PF primary factors of C (Emotional Stability), higher scores on IN (Independence). This seems to suggest H (Social Boldness), and Q1 (Openness to Change; adjusted that the multiple regression model points to AX (Anxiety) r = .53). Rieke and Conn (1994) also found these primary fac- and IN (Independence) being the salient factors in construct- tors related to overall self-esteem on the SEI. In their analysis, ing a multidimensional model of self-esteem using the 16PF they also found A (Warmth), M (Abstractness), N (Privateness), Fifth Edition. These findings are consistent with previous and O (Apprehension) as significant predictors. While factor A research using the 16PF Fifth Edition (Rieke & Conn, 1994). (Warmth) did appear in one of this study’s equations, it did not Unlike the findings reported by Rieke and Conn, the EX consistently appear across the three regression equations Pietrzak et al. 7 Table 4. Primary Factor Regression Results of CFSEI From 16PF. 2 a Scale group R A − R F Predictors Total scale A C E G H I L M N Q1 Q2 Q3 Total .74 .54 39.68** .40 .11 .27 −.09 −.12 −.13 .20 Split 1 .75 .54 20.69** −.17 .36 .12 .45 −.15 .18 .12 Split 2 .80 .60 22.34** .54 .11 .20 −.25 −.13 .16 .15 −.24 Final .72 .51 80.77** .62 .20 .10 General scale C E F H L M N O Q1 Q2 Q3 Total .65 .40 23.35** .27 .14 −.09 .30 −.12 −.11 .20 Split 1 .69 .45 17.50** .23 .14 .33 −.17 .20 −.22 Split 2 .65 .39 11.15** .42 .23 −.13 −.22 Final .60 .35 64.47** .35 .39 Social scale C F H I M Q1 Q2 Q4 Total .52 .44 37.09** .13 .21 −.11 −.30 .24 −.16 Split 1 .61 .35 13.69** .24 −.18 −.33 .29 −.27 Split 2 .48 .20 10.60** .43 .18 .20 Final Personal scale A C H N Q2 Q4 Total .67 .44 37.09** −.10 .59 .15 .13 −.14 Split 1 .64 .39 16.30 .50 .30 .20 .17 −.19 Split 2 .72 .51 115.85** .72 Final .40 157.61** .64 Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition; CFSEI = Culture-Free Self-Esteem Inventory; R = multiple R; A − R = adjusted multiple R-square; F = F-ratio of the regression equation; Total = total sample; Split 1 = Split Validation Group 1; Split 2 = Split Validation Group 2; Final = final split sample–validated equation. Predictors are the letter names for the primary scales of the 16PF Fifth Edition. Unable to calculate. **p < .01. predicting the CFSEI-II Total score. These differences may (Apprehension) has been related to self-esteem in research be due to variability in the samples used or dissimilarities in with previous editions of the 16PF (Cattell, 1989), the cur- the measurement of self-esteem between the SEI and CFSEI. rent edition of the 16PF (Rieke & Conn, 1994), and is a cen- To clarify these differences, we suggest further research. tral component of calculation of the global factor AX Consistent with the global factor results, the CFSEI-II (Anxiety). Therefore, we expected factor O (Apprehension) Social scale score was the most difficult to predict. No pri- to take a more central role in predicting at least global self- mary factors were identified by the regression equations to esteem. However, factor O (Apprehension) was present in construct a stable prediction. The regression equations for only one of the three equations predicting the CFSEI-II the CFSEI-II General and Personal scores accounted for General scale score. In addition, it did not play a role in any approximately 35% and 40% shared variance, respectively. of the other regression equations. This finding is different The primary factor C (Emotional Stability) was the most from previous studies (Rieke & Conn, 1994), which used dif- consistent predictor of all facets of self-esteem as measured ferent samples, measures of self-esteem, and validation by the CFSEI-II. Primary factor C (Emotional Stability) was methods. Based on these results, we suggest further research the only consistent predictor for the CFSEI-II Personal self- to clarify this issue. esteem score. The results of this study indicate, as previous A secondary goal of this study was to make available split research has suggested, that factor C (Emotional Stability) is sample–validated equations for the user of the 16PF Fifth consistently related to various aspects of self-esteem (Rieke Edition. We anticipate that this will enable these users to apply & Conn, 1994). Primary factor Q1 (Openness to Change) the 16PF Fifth Edition to their setting using an alternative was the predictor present in the CFSEI-II Total score equa- model of self-esteem. The following regression equations serve only as experimental measures. Further research to tion that was not present in the CFSEI-II General score equation. examine their reliability and validity is obviously needed; Perhaps the most interesting finding in this study was the however, they may serve as an entry point for examining the lack of primary factor O (Apprehension) in any of the final relationship between a multidimensional model of self-esteem and the 16PF Fifth Edition. Table 5 provides the final regres- equations, although it correlated significantly with the sions of the CFSEI-II from the 16PF Fifth Edition. Descriptive CFSEI-II Total scale and its subscales (see Table 2). Factor O 8 SAGE Open Table 5. Final Regression Equations of the CFSEI From 16PF (N = 234). a b Scale M SD SEE 16PF equation Global factor score equations Total 6.0 1.08 1.18 (−.55 × AX) + (.53 × IN) + 6.21 General 6.0 0.97 1.36 (−.41 × AX) + (.52 × IN) + 5.41 Social 6.4 0.84 1.60 (.20 × EX) + (−.30 × AX) + (.17 × TM) + (.34 × IN) + 4.19 Personal 5.8 1.26 1.55 (−.77 × AX) + (.19 × TM) + (.56 × IN) + 6.16 Primary factor score equations Total 6.1 1.16 1.18 (.46 × C) + (.24 × H) + (.10 × Q1) + 1.56 General 5.9 0.99 1.34 (.32 × C) + (.32 × H) + 2.32 Personal 5.8 1.26 1.54 (.70 × C) + 1.91 Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition; CFSEI = Culture-Free Self-Esteem Inventory; AX = Anxiety; IN = Independence; EX = Extroversion; TM = Tough-Mindedness. Represents the standard error of the estimate. 16PF letter names used are those from the 16PF Fifth Edition Technical Manual (Rieke & Conn, 1994). statistics are there, as well. The regression equations are References reported using Sten scores for both the predictor and depen- Arlot, S., & Celisse, A. (2010). A survey of cross-validation pro- dent variables as is traditional when using the 16PF Fifth cedures for model selection. Statistics Survey, 4, 40-79. Edition (Rieke & Conn, 1994). doi:10.1214/09-SS054. Retrieved from http://projecteuclid. org/euclid.ssu/1268143839 Battle, J. (1977a). A comparison of two self-report inventories. Limitations and Recommendations Psychological Reports, 41, 159-160. Battle, J. (1977b). Test-retest reliability of the Canadian Self-Esteem The limitations of this study simultaneously offer interpreta- Inventory for Adults. Perceptual & Motor Skills, 44, p. 38. tive caution and research opportunities. First, the sample, Battle, J. (1978). Relationship between self-esteem and depression. although racially and ethnically diverse, was biased heavily Psychological Reports, 42, 745-746. toward females (71% of the sample). Future research should Battle, J. (1992). Culture-free self-esteem inventories: Examiners include larger and more representative samples. We suggest manual (2nd ed.) Austin, TX: Pro-Ed. greater consideration of the variables gender and self- Brooke, S. L. (1995). Critical analysis of the culture-free self-esteem esteem. Along these lines, other measures of self-esteem inventories. Measurement and Evaluation in Counseling and might be utilized or compared with the CFSEI-II. Third, Development, 27, 248-252. multiple regression is sensitive to the type of validation uti- Cattell, H. (1989). The 16PF: Personality in depth. Champaign, IL: Institute for Personality and Ability Testing. lized, especially data splitting (Arlot, & Celisse, 2010). Cattell, H., & Mead, A. (2008). The Sixteen Personality Factor Furthermore, with any multiple regression study, there is no Questionnaire (16PF). In G. Boyle, G. Matthews, & D. substitution for multiple replication studies to refine the Saklofske (Eds.), The SAGE handbook of personality finding of any given study. Fourth, as with any assessment theory and assessment: Volume 2—Personality measure- tools, the application and interpretation of the 16PF Fifth ment and testing (pp. 135-160). London, England: SAGE. Edition and CFSEI-II need to take place in the context of doi:10.4135/9781849200479.n7 sound clinical practice marked by an appropriate under- Cattell, H., & Schuerger, J. M. (2003). Essentials of 16PF assess- standing of assessment. Clinicians should use assessment ment. Hoboken, NJ: John Wiley. materials with proper training and experiences. Furthermore, Cattell, R., Eber, H., & Tatsuoka, M. (1970). Handbook for the for clinicians to reach meaningful and useful interpretations 16PF. Champaign, IL: Institute for Personality and Ability of assessment data, they should integrate such information Assessment. in the context of other relevant client data, such as family Chapman, P. L., & Mullis, A. K. (2002). Readdressing gen- der bias in the Coopersmith Self-Esteem Inventory-Short and cultural influences. Form. The Journal of Genetic Psychology, 163, 403-409. doi:10.1080/00221320209598692 Declaration of Conflicting Interests Cicchetti, D. V. (1994). Guidelines, criteria, and rules of thumb for The author(s) declared no potential conflicts of interest with respect evaluating normed and standardized assessment instruments in to the research, authorship, and/or publication of this article. psychology. Psychological Assessment, 6, 284-290. Clark, W. L., & Blackwell, T. L. (2007). 16PF® (5th edi- Funding tion) Personal Career Development Profile: Test review. The author(s) received no financial support for the research and/or Rehabilitation Counseling Bulletin, 50, 247-250. doi:10.1177 authorship of this article. /00343552070500040601 Pietrzak et al. 9 Coopersmith, S. (1981). Self-esteem inventories. Palo Alto, CA. concepts, different outcomes. American Sociological Review, Consulting Psychologist Press. 60, 141-156. Dancer, L. J., & Woods, S. A. (2006). Higher-order factor structures SPSS. (2007). SPSS for Windows [Version 16.0]. Chicago, IL: and intercorrelations of the 16PF5 and FIRO-B. International Author. Journal of Selection and Assessment, 14, 385-391. Stark, S., Chernyshenko, O. S., Drasgow, F., & Williams, B. Francis, L. J. (1997). Coopersmith’s model of self-esteem: Bias A. (2006). Examining assumptions about item respond- toward the stable extravert? The Journal of Social Psychology, ing in personality assessment: Should ideal point methods 137, 139-142. be considered for scale development and scoring? Journal Francis, L. J., & James, D. (1998). Is there gender bias in the of Applied Psychology, 91, 25-39. doi:10.1037/0021- short form Coopersmith Self-Esteem Inventory? Educational 9010.91.1.25 Research, 40, 83-89. Walsh, W., & Betz, N. (1995). Tests and assessment (3rd ed.). Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Englewood Cliffs, NJ: Prentice Hall. Multivariate data analysis (5th ed.). Upper Saddle River, NJ: Prentice Hall. Author Biographies Hair, J. F., Babin, B. J., & Anderson, R. E. (2009). Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall. Dale Pietrzak, Ed.D., LPC-MH (SD), CCMHC is the Director of Institute for Personality and Ability Testing. (2009). 16PF fifth edi- Institutional Research and Assessment at the University of Idaho. tion manual. Champaign, IL: Author. He has been a clinician, faculty member and and/or administrator Janda, L. (1998). Psychological testing: Theory and application. since 1987. Boston, MA: Allyn & Bacon. Betsy Page is an Associate Professor of Counselor Education and Kawash, G., Clews, J., & Keating, L. (1989). An examination of Supervision at Kent State University. Her interests include group self-esteem in the context of general personality functioning. work, suicide assessment, and clinical supervision. Multivariate Experimental Research, 9, 27-33. Luhtanen, R., & Crocker, J. (1992). A Collective Self-Esteem James S. Korcuska, Ph.D., LPC, NCC is an associate professor at Scale: Self-evaluation of one’s social identity. Personality and North Dakota State University in Counselor Education and Social Psychology Bulletin, 18, 302-318. Supervision. He also coordinates the program’s clinical training Nunnally, J., & Bernstein, I. (1994). Psychometric theory (3rd ed.). experiences and teaches advanced assessment. New York, NY: McGraw-Hill. Amber Bach Gorman Ph.D, works full time at the North Dakota Rieke, M. L., & Conn, S. R. (1994). Psychological adjustment and State University Counseling Center as a mental health counselor self-esteem. In S. Conn & M. Rieke (Eds.), 16PF fifth edition and clinical supervisor. She is also an adjunct lecturer in Counselor technical manual (pp. 143-162). Champaign, IL: Institute for Education and Supervsion at North Dakota State University and Personality and Ability Testing. Minnesota State University – Moorhead’s Counseling and Student Rosenberg, M., Schooler, C., Schoenbach, C., & Rosenberg, F. Affairs program. (1995). Global self-esteem and specific self-esteem: Different http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png SAGE Open SAGE

An Examination of the Relationship of the 16PF Fifth Edition to a Multidimensional Model of Self-Esteem:

Loading next page...
 
/lp/sage/an-examination-of-the-relationship-of-the-16pf-fifth-edition-to-a-0LUTSni5z6

References (30)

Publisher
SAGE
Copyright
Copyright © 2022 by SAGE Publications Inc, unless otherwise noted. Manuscript content on this site is licensed under Creative Commons Licenses.
ISSN
2158-2440
eISSN
2158-2440
DOI
10.1177/2158244015611453
Publisher site
See Article on Publisher Site

Abstract

The regression equation needed to predict the self-esteem score from the Sixteen Personality Factor Questionnaire (16PF) Fifth Edition is not available for hand scoring. This score is derived from the Coopersmith’s Self-Esteem Inventory (SEI), which is based on a unitary model of self-esteem. The SEI may be subject to gender and extroversion biases. Moreover, there is a gap in the literature on the 16PF fifth edition and self-esteem. For these reasons, we investigated the relationship of the Culture-Free Self-Esteem Inventory–II Form AD (CFSEI-II), a multidimensional model of self-esteem, using multiple regression to the scales of the 16PF. Based on the adult sample of 234 in this study, we developed and made available to users of the 16PF split sample–validated regression equations for this multidimensional model of self-esteem. Keywords Sixteen Personality Factor Questionnaire Fifth Edition, self-esteem, model, personality, statistical analysis The Sixteen Personality Factor Questionnaire Fifth Edition to how similar or dissimilar they perceive themselves to be in (16PF), released in 1994 (Institute for Personality and Ability relationship to the item (Stark, Chernyshenko, Drasgow, & Testing [IPAT], 2009), is the latest edition of the personality Williams, 2006). These 16 primary factors are considered test, which was devised to measure normal personality func- primary personality traits (Cattell & Schuerger, 2003). tions (Cattell, Eber, & Tatsuoka, 1970). Since its origin, Primary personality traits contribute to five second-order or, developers have revised the 16PF 5 times, updating content, more commonly, global traits (see Table 1). Thus, the scales norms, and/or to make other improvements, all of which of the 16PF, 16 primary and 5 global, reflect an underlying added to its overall psychometric qualities (IPAT, 2009). multitiered conceptualization of personality. Furthermore, There are also validity scales, such as impression manage- Cattell theorized third-order factors, as well, and only ment. Reviewers have consistently rated the 16PF one of the recently has reliable evidence of a two-factor solution been most used and researched personality tests (Dancer & Woods, established (Cattell & Mead, 2008). 2006; Walsh & Betz, 1995). This ubiquity may be traced to The 16PF Fifth Edition may be administered via paper its empirical roots, multitiered trait structure (including Big and pencil or computer. A number of computer-generated Five personality traits) of normal-range personality, and reports can be derived from the scale scores, including the availability in more than 30 languages (Cattell & Schuerger, Personal Career Development Profile (Clark & Blackwell, 2003). 2007). Historically, the 16PF has been a popular personality The 16PF was originally constructed by factor analyzing measure because of the multiple special scores that can be descriptions of personality based on English-language adjec- calculated from combinations of various factor scores. tives (Cattell et al., 1970; IPAT, 2009). The factor analysis One special score that can be calculated from the 16PF yielded 16 primary factors (see Table 1). One of these fac- Fifth Edition is a measurement of unitary self-esteem. Using tors, Reasoning (B), is a proxy for measuring cognitive or reasoning ability. All items and associated traits are assumed University of Idaho, Moscow, USA to exist on continua and thus are conceptually bipolar. For Kent State University, OH, USA example, descriptors for the primary factor Warmth (A) North Dakota State University, Fargo, USA range from “reserved” to “attentive to others.” This underly- Corresponding Author: ing structure (i.e., each trait and its associated items existing James S. Korcuska, North Dakota State University, Dept. 2625, P.O. Box along a continuum) assumes a dominance response process. 6050, Fargo, ND 58108-6050, USA. In a dominance process, a person responds to an item relative Email: james.korcuska@ndsu.edu Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License (http://www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open Table 1. Means and Standard Deviations for 16PF Fifth Edition Researchers continue to debate whether self-esteem is a and Culture-Free Self-Esteem Inventory–II. unitary or multidimensional construct. Many professionals conceptualize self-esteem as multifaceted (Battle, 1992; a a Scale n M SD Luhtanen & Crocker, 1992; Rosenberg, Schooler, Schoenbach, 16PF Fifth Edition & Rosenberg, 1995). Battle’s (1992) multidimensional model A Warmth 234 6.3 1.8 includes dimensions of self-esteem based on (a) overall per- B Reasoning 233 6.2 1.7 ceptions of self-worth, (b) perceptions of relationship quality, C Emotional Stability 234 5.6 1.8 and (c) evaluation of intimate sense of self-worth. Thus, evalu- E Dominance 234 5.2 2.0 ations of self-worth refer to perceptions of worthiness as a per- F Liveliness 234 6.0 1.9 son in comparison with others. The intimate sense of self-worth G Rule-Consciousness 234 5.4 1.9 dimension refers to the evaluation of self-worth independent H Social Boldness 233 5.7 2.0 of others’ evaluations of it. In addition, Rosenberg et al. (1995) I Sensitivity 234 6.4 1.8 found a stronger correlation of global self-esteem to psycho- L Vigilance 234 5.3 2.0 logical well-being than with a specific or unitary conceptual- M Abstractness 234 5.7 1.9 ization, further indicating that a multidimensional model of N Privateness 234 5.0 2.1 self-esteem might hold more utility than a unitary conceptual- O Apprehension 234 5.9 1.5 ization. Intimate evaluation of self-worth might have a stron- Q1 Open to Change 234 5.7 2.1 ger relationship to the perception of mental health and for the Q2 Self-Reliance 234 5.3 1.9 expectation of positive treatment outcomes than a sense of Q3 Perfectionism 234 5.0 2.0 social self-esteem or more general sense of self-worth (Battle, Q4 Tension 234 5.5 1.8 1992). Extroversion (EX) 234 6.1 1.9 There are several reasons for examining how the 16PF Anxiety (AX) 234 5.6 1.7 Fifth Edition relates to multidimensional models of self- Tough-Mindedness (TM) 234 4.7 2.0 Independence (IN) 234 5.5 1.9 esteem. Today’s mental health environment, Health Self-Control (SC) 233 5.1 1.7 Maintenance Organizations (HMOs), and insurance compa- Culture-Free Self-Esteem Inventory–II Form AD nies provide less time and money for counselors to use mul- General 234 52.3 8.3 tiple assessment tools. Clinicians must use assessment tools Personal 234 51.6 9.9 efficiently and economically. If valid special scores for a Social 234 54.4 8.9 multidimensional model of self-esteem for the 16PF Fifth Total 234 52.3 8.0 Edition can be developed, its use would be enhanced, increas- ing efficiency in the data-gathering process. Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition. To date, there has been limited research using the current Values rounded to single decimal place. The 16PF scores are reported in Sten scores. A Sten has a mean of 5.5 and a standard deviation of 2.0. The edition of the 16PF Fifth Edition with unitary models of self- Culture-Free Self-Esteem Inventory–II Form AD (CFSEI) is reported using esteem, a difference between the fifth edition and earlier ver- T scores. sions (Rieke & Conn, 1994). Rieke and Conn examined the relationship between the 16PF Fifth Edition and a unitary model of self-esteem. In a sample of 318 largely non-White a regression equation, combinations of 16PF Fifth Edition subjects (56%), higher scores were found on the global fac- factor scores are entered to predict a measure of self-esteem tors of IN (Independence) and EX (Extroversion) with a based on the Coopersmith’s Self-Esteem Inventory (SEI; lower score on AX (Anxiety) being correlated to higher self- Coopersmith, 1981). Currently, the regression equation nec- esteem. Correlations at p < .05 for global factors were AX essary to predict the SEI score from the 16PF is not available (Anxiety) = −.64, IN (Independence) = .49, EX (Extroversion) for hand scoring. According to sources at the Institute for = .45, and SC (Self-Control) = .26. The primary factors asso- Personality and Ability Testing (IPAT), there is no intention ciated with higher self-esteem on the current edition included of making this formula available apart from their computer correlations on A (Warmth) = .34, C (Emotional Stability) = scored reports (S. Conn & Institute for Personality and Ability .64, H (Social Boldness) = .54, and negative correlations on Testing Staff, personal communication, August 24, 1994). N (Privateness) = −.20 and Q1 (Openness to Change) = .28, The formula used to calculate this computer-generated spe- along with lower scores on O (Apprehension) = .58, and M cial score was derived using a cross-validated multiple regres- (Abstractness) = .44. These results, however, warrant cau- sion procedure described by Rieke and Conn (1994). The tion. The model of self-esteem upon which the Coopersmith’s regression equation that resulted from their work provided a SEI is built has shown biases in gender (Chapman & Mullis, r of .72 using the five global factors and an r of .76 for the 2002) and extroversion (Francis, 1997). primary factors. Rieke and Conn (1994), citing Kawash, In addition to the limited research on unitary models of Clews, and Keating (1989), chose a unitary model of self- self-esteem and the recent edition of the 16PF, a search of esteem, an important choice as we were interested in multi- ERIC and PsychInfo found no other studies examining the dimensional constructs of self-esteem. Pietrzak et al. 3 relationship between the 16PF Fifth Edition and other mod- edition was revised using modern psychometric procedures els of self-esteem. Thus, an extension of the research exam- and norm collecting procedures to update language and ining the relationship of the 16PF Fifth Edition to multiple improve the psychometric qualities of the tool. The 16PF dimensions of self-esteem is appropriate. The primary pur- contains 16 bipolar scales and several validity scales. Fifteen pose of this study was to examine the relationship of the of the scales measure personality traits and one scale mea- 16PF Fifth Edition to the four dimensions of self-esteem sures cognitive or reasoning abilities (see earlier discussion). measured by the Culture-Free Self-Esteem Inventory–II The stability coefficients for the personality and validity Form AD (CFSEI-II) with an adult sample using multiple scales range from r = .69 to r = .86 for a 2-week period with regression. A secondary purpose of the study was to begin a range of r = .56 to r = .79 for a 2-month period. The internal developing split sample–validated regression equations for consistency of the personality and validity scales ranged this multidimensional model of self-esteem and make them from r = .66 to r = .87 (Rieke & Conn, 1994). available to users of the 16PF Fifth Edition. In doing so, In addition to the 16 primary factors, 5 global factors can users would have a choice of self-esteem models. be obtained. The global factors are calculated from multiple regression equations developed from second-order factor analyses of the primary factors (Cattell et al., 1970; Rieke & Method Conn, 1994). While no internal consistency indices can be calculated for the global factors, the stability of these scales Participants ranges from r = .84 to r = .91 over a 2-week period, and r = The sample used in this study consisted of 234 volunteers’ .70 to r = .82 over a 2-month period (Rieke & Conn, 1994). responses. Of these, 109 (47%) were from the Southern Indiana, Southern Illinois, or Northern Kentucky area, 58 CFSEI-II. The CFSEI-II is a widely used multidimensional (24%) were from the Northeastern Ohio or Northwestern measure of self-esteem. James Battle’s theory of self-esteem Pennsylvania area, and 67 (29%) were from Southern (Battle, 1992) guided the instrument’s development. Admin- Florida. Participants’ age ranged from 18 years to 69 years. istration time is between 10 and 15 min. The adult form of The mean age was 27.9 years (SD = 10). Two participants the CFSEI-II has 40 items. Respondents’ yes or no item included in this study did not report their age. choices fall into either a high or low self-esteem group. Items The participants were largely female (71%; n = 165; form three scales measuring aspects of self-esteem (General males; 29%; n = 69). The primary researcher solicited par- Self-Esteem, Social Self-Esteem, and Personal Self-Esteem) ticipants through minority organizations in the community and a scale to detect “faking good.” The three self-esteem and various class levels at multiple university locations. scale scores can also be summed to obtain a global measure These procedures resulted in the reported racial heritage of of self-esteem (Battle, 1992). The General Self-Esteem score the sample participants consisting of 82% (n = 193) White serves as a measure of an adult’s “overall perceptions of non-Hispanics, 9% (n = 20) African Americans, 2% (n = 5) worth” (p. 3). The Social Self-Esteem score is an assessment Asian Americans, 6% (n = 15) Hispanics, and 0.4% (n = 1) of that “aspect of self-esteem that refers to individuals’ per- Native American. Participants reported educational levels ceptions of the quality of their relationships with peers” ranging from an 11th-grade education to doctorates. The (p. 3). The Personal Self-Esteem score was designed to mea- mean educational level was 14.9 years (2 years of college) sure “the aspect of self-esteem that refers to individuals’ with a standard deviation of 2.4 years. One person did not most intimate perceptions of self-worth” (p. 4). Summing report his or her educational level. these scores yields a cumulative, or more global, measure of There were 59% (n = 138) undergraduate and graduate stu- overall self-esteem. dents and 41% (n = 96) community members (not enrolled in The stability of the CFSEI-II scale scores was reported as classes) in the sample. A diverse sample of university partici- r = .56 for the Social score, r = .78 for the Personal score, r = pants was solicited from undergraduate programs in art, music, .82 for the General score, and r = .82 for the Total score over accounting, business management, counseling, elementary a 4-week period (Battle, 1977b). Research regarding the education, secondary education, and special education. The validity of the CFSEI-II has been supportive. There is evi- counseling participants were graduate students. No remunera- dence of concurrent validity with the Coopersmith’s’ SEI, tion or other incentive was provided to the participants. with correlations of .71 to .80 (Battle, 1992). A number of studies have examined the validity of the CFSEI-II scores as correlates of depression (Brooke, 1995). For example, the Materials Beck Depression Inventory (overall: −.55; male: −.53, 16PF. The 16PF Fifth Edition is an empirically derived per- females: −.56; Battle, 1978). Other validity research has sonality test devised to measure normal personality func- included content analysis, factor analysis, and the relationship tions. It was originally constructed by factor analyzing of CFSEI-II scores to external criteria, including its relation- English-language adjectives used to describe personality ship to teachers’ ratings and the Minnesota Multiphasic (Cattell, 1989; Cattell et al., 1970; IPAT, 2009). The fifth Personality Inventory (Battle, 1977a). There is inadequate 4 SAGE Open support for the claim, however, that the CFSEI-II is culture Table 2. Correlation Between16PF Fifth Edition and Culture- Free Self-Esteem Inventory–II (N = 234). free (Brooke, 1995). Culture-Free Self-Esteem Inventory–II Form Procedure 16PF Fifth Edition Total General Social Personal The primary author collected the data for this study as part of a group of studies on the 16PF Fifth Edition. Sample size A Warmth .16* .19* .15* .04 was preestablished with a minimum of 170 participants. The B Reasoning .03 .05 .02 −.02 sample of this size allows for 10 or more cases per variable C Emotional Stability .63* .47* .37* .63* when calculating multiple regressions, and is the ratio experts E Dominance .34* .36* .23* .18* recommend (Hair, Babin, & Anderson, 2009; Janda, 1998; F Liveliness .24* .18* .31* .15* Nunnally & Bernstein, 1994). G Rule-Consciousness .02 −.03 .07 .04 The administration packets included the 16PF, CFSEI-II, H Social Boldness .51* .51* .36* .29* demographics sheet, and other instruments placed in an I Sensitivity −.04 −.01 −.05 −.07 L Vigilance −.26* −.20* −.18* −.23* envelope in counterbalanced order. Before opening the M Abstractness −.18* −.11 −.21* −.18 administration envelope, each participant read or heard a N Privateness −.27* −.33* −.18* −.09 description of this study. The researchers or research assis- O Apprehension .34* .36* .23* .18* tants distributed 260 packets. Sixty percent of the partici- Q1 Open to Change .31* .30* .15* .22* pants completed the materials in a group setting and the other Q2 Self-Reliance −.27* −.21* −.34* −.16* 40% did so at home. Participants returned the packets to one Q3 Perfectionism −.00 .02 −.03 −.02 of the researchers or a research assistant. Of this number, Q4 Tension −.23* −.12 −.09 −.35* participants returned 253 packets. The returned packets Extroversion (EX) .39* .38* .37* .19* resulted in 234 usable 16PF Fifth Edition and CFSEI-II sur- Anxiety (AX) −.34* −.19* −.18* −.44* vey instruments that were included in this study for analysis. Tough-Mindedness (TM) −.12 −.16* −.02 −.04 There was a 90% return rate. Independence (IN) .42* .45* .27* .23* The materials were hand scored separately and checked Self-Control (SC) .01 −.01 .01 .03 for accuracy twice, sometimes thrice, by different members of the research team. To estimate the degree of scoring error, Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition. *p < .05. researchers randomly selected 10 protocols and rescored them by hand. No errors were found in these 10 protocols. for each randomly generated subsample. Subsample regres- Data Analysis sion models may then be compared with each other, as well as 2 2 to the overall model using R , adjusted R , and the standard Statistical analyses were performed using SYSTAT 9.0 for error of measurement. The goal is to determine which indi- Windows (SPSS, 2007). First, an analysis of the correlations vidual variables across models demonstrate the highest level between the 16PF Fifth Edition and CFSEI-II was conducted of similarity, which is determined by beta weight. These (see Table 2). Next, a forward stepwise multiple regression variables govern the composition of the overall model. procedure with a .15 entry and .15 removal level was uti- We divided the total sample (N = 234) into two groups by lized, paralleling methods in previous research on the 16PF generating a random number for each case using SYSTAT (Rieke & Conn, 1994). Unlike the cross-validation method 9.0’s pseudo-random number generator. This number had a utilized by Rieke and Conn (1994), we utilized split sample normal distribution with a mean of 0 and a standard devia- validation to validate the regression models. Our purpose tion of 1. The sample was then split into two groups at the was to examine the relationship of the 16PF Fifth Edition to mean. The first of these groups contained 52% (n = 121) of a measure of multidimensional model of self-esteem and not the cases and the second contained 48% (n = 113) of the to directly compare the results of this study with those of cases. For convenience, these subsamples were called Split 1 Rieke and Conn. Therefore, we did not rigidly follow their and Split 2, respectively. Although other ratios are possible, analytic protocol. we chose the one most frequently employed, which is to split We followed guidelines for stepwise multiple regression the sample into near equal parts. model development and validation described by Hair, We calculated initial equations using a stepwise proce- Anderson, Tatham, and Black (1998; Hair et al., 2009), which dure, including the total sample (see Table 3). These initial utilizes split sample validation or split-half validation. In this equations were then compared with two equations calculated validation procedure, a full sample regression model is com- using the split samples. Predictors that entered into each of pared across two subsample regression models. To create the the three equations are then examined to see if (a) the same subsamples, the sample is divided into half, typically using a ran- predictors entered into each of the three equations and (b) the dom number generator. Next, regression models are estimated Pietrzak et al. 5 Table 3. Global Factor Regression Results of CFSEI From 16PF. 2a Scale group r r F EX AX TM IN SC Total score Total .70 .48 72.81** −.62 .18 .71 Split 1 .70 .48 38.05** −.55 .70 .16 Split 2 .71 .48 35.89** −.66 .23 .69 Final .68 .46 98.45** −.58 .62 General score Total .60 .35 31.92** .09 −.41 .16 .60 Split 1 .64 .39 26.65** −.40 .68 .17 Split 2 .58 .31 13.74** .15 −.40 .18 .53 Final .59 .34 −.42 .59 Social score Total .47 .21 16.05** .21 −.28 .19 .36 Split 1 .57 .30 13.58** .24 −.36 .26 .47 Split 2 .42 .15 5.75** .20 −.24 .16 .31 Final .47 .21 16.05** .21 −.28 .19 .36 Personal score Total .65 .41 40.85** −.16 −.72 .18 .61 Split 1 .60 .34 16.59** −.21 −.68 .15 .61 Split 2 .69 .46 33.27** −.72 .23 .55 Final .63 .39 50.71** −.65 .19 .53 Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition; CFSEI = Culture-Free Self-Esteem Inventory; EX = Extroversion; AX = Anxiety; TM = Tough-Mindedness; IN = Independence; SC = Self-Control. a 2 This represents the adjusted r . Total N = 234, Split 1 n = 121, Split 2 n = 113, and Final N = 234. **p < .001. standardized beta weight for the predictors had the same sign more, lower than the standardization sample of the 16PF (positive or negative) across the three equations. A final Fifth Edition on N (Privateness) and TM (Tough- equation was constructed using only those predictor vari- Mindedness). These differences are similar to those of the ables that were stable across all three regression models. The sample reported by Rieke and Conn (1994) in their work final overall equation (if one can be constructed) was deter- with the 16PF Fifth Edition and self-esteem. The manual for mined when it demonstrated statistically significant improve- the CFSEI-II (Battle, 1992) contains the means and standard ment in prediction over chance. Statistical significance of the deviations for the various scales and total score based upon final equation was evaluated using the F statistic. Practical the standardization sample. The descriptive statistics for the significance was examined using the adjusted r , which pro- current sample was similar to that of the CFSEI-II, with all vides an estimate of the amount of shared variance between scores within one half standard deviation of the mean. a set of predictors and the dependent variable. A criterion of The correlations between the 16PF Fifth Edition and the an adjusted r reaching at least .50 was set, so that the equa- CFSEI-II appear in Table 2. These data indicate that three of tion had at least 50% shared variance with the variable being four global factor scales (EX, AX, and IN) correlated moder- predicted. This level of shared variance is considered as the ately or low to CFSEI-II total and subscales. Two of the pri- minimum acceptable value for alternate forms of measure- mary scales comprising the global factor scales AX (Anxiety) ment of any given construct (Cicchetti, 1994; Hair et al., and Q4 (Tension) had negative correlations with self-esteem: 2009; Janda, 1998). As these scores increased, scores on CFSEI-II decreased. Results of the multiple regression equations from the global factors are summarized in Table 3. In predicting the Results CFSEI-II Total score, the primary factors of AX (Anxiety) The means and standard deviations for the 16PF Fifth Edition and IN (Independence) were stable for all three equations. and the CFSEI-II are provided in Table 1. These results indi- The equations entered both factors and each in the same cate that the sample used in this study was one half standard direction. The final equation reached statistical, but not prac- deviation, or more, higher than the standardization sample of tical significance. In predicting the CFSEI-II General scale the 16PF Fifth Edition on A (Warmth), B (Reasoning), F score, the global factors of AX (Anxiety) and IN (Liveliness), I (Sensitivity), and EX (Extroversion). In addi- (Independence) were stable for all three equations. Each fac- tion, the sample was at least one half standard deviation, or tor was entered by the equations and in the same direction. 6 SAGE Open The final equation was statistically significant but failed to (Extroversion) global factor did not contribute centrally to reach practical significance. The three equations predicting the prediction of global self-esteem. Furthermore, the only the CFSEI-II Social scale score entered the global factors of consistent observed role for EX (Extroversion) was to predi- EX (Extroversion), AX (Anxiety), TM (Tough-Mindedness), cate the CFSEI-II Social scale score, the scale measuring a and IN (Independence) in the same direction. The final equa- person’s perceptions of the quality of his or her relationships. tion was statistically significant, but failed to reach practical However, these differences may have been due to the differ- significance. In predicting the CFSEI-II Personal scale score, ences between the ways the CFSEI-II and the SEI measure the global factors of AX (Anxiety), TM (Tough-Mindedness), self-esteem. The SEI may have a bias toward extroversion and IN (Independence) were stable for all three equations. (Francis, 1997). There are correlations, however, between Each was entered by the equations and in the same direction. EX and all scales of the CFSEI-II (see Table 2), albeit small The final equation was statistically significant, but not prac- to moderate ones. Participants in our study were mostly tically significant. female (71%). At least two studies indicate a gender bias in The multiple regression results from the primary factors the items comprising the SEI (Chapman & Mullis, 2002; are summarized in Table 4. Prediction of the CFSEI-II Total Francis & James, 1998). To clarify these findings, we sug- scale score revealed the primary factors C (Emotional gest further research. Stability), H (Social Boldness), and Q1 (Openness to In examining the relationship of the 16PF Fifth Edition to Change) as stable in all three equations. The final equation other facets of self-esteem measured by the CFSEI-II, several was both statistically and practically significant. In predict- findings are worth mentioning. The 16PF Fifth Edition global ing the CFSEI-II General score, the primary factors C factor equations for the CFSEI-II Total scale score and General (Emotional Stability) and H (Social Boldness) were stable score had the same predictor variables of AX (Anxiety) and IN for all three equations. Each was entered and in the same (Independence). It is unclear from this research if this is due to direction. The final equation was statistically significant but the overlap of these constructs as measured by the CFSEI-II. not practically significant. No final equation was constructed Alternately, this finding may indicate that these constructs for the prediction of the CFSEI-II Social scale score because share an underlying dimension with the 16PF Fifth Edition. no primary factors were stable across all three equations. Still, it seems possible that there are differences between the Prediction of the CFSEI-II Personal scale score showed that constructs. This possibility is suggested by the finding that primary factor C (Emotional Stability) was stable for all the amount of variance accounted for is approximately 35% in three equations. The equations entered each factor and in the the case of the General scale score, while the Total score shares same direction. The final equation was statistically signifi- approximately 46% of the variance. Further research is neces- cant but failed to reach practical significance. sary and may clarify this issue. The regression equation for CFSEI-II Social scale score added other 16PF Fifth Edition factors beyond the inverse Discussion relationship between AX (Anxiety) and IN (Independence). The primary purpose of this study was to explore the rela- In the equation for the Social scale, the 16PF factors of EX tionship of the 16PF Fifth Edition to a multidimensional (Extroversion) and TM (Tough-Mindedness) were added. model of self-esteem with an adult sample. Although all of This suggested that while the Social scale shared the lower the final regression models based on the global factors AX (Anxiety) and greater IN (Independence) factors found reached statistical significance, they did not exceed the crite- in the other dimensions of self-esteem, the factors of ria set for practical significance. However, the global factor increased social participation and a more tough-minded equation for CFSEI-II Total scale score did approach practi- stance on issues were included. The regression equations for cal significance (adjusted r = .46). The prediction of the the CFSEI-II Personal score were similar to the equations for CFSEI-II Personal scale score had the next highest adjusted the Social scale, minus the social participation factor. r (.39). The third highest prediction of the CFSEI-II General Prediction of three of the four scores of the CFSEI-II was scale score had an adjusted r (.34). The CFSEI-II Social possible using the primary factors of the 16PF Fifth Edition. scale score had the lowest adjusted r (.21) of all the global However, only one of the final multiple regression equations factor equations. exceeded 50% shared variance with the CFSEI-II. This final Notably, all of the final equations shared the same central regression equation was for the CFSEI-II Total score com- elements of lower scores on AX (Anxiety) coupled with posed of the 16PF primary factors of C (Emotional Stability), higher scores on IN (Independence). This seems to suggest H (Social Boldness), and Q1 (Openness to Change; adjusted that the multiple regression model points to AX (Anxiety) r = .53). Rieke and Conn (1994) also found these primary fac- and IN (Independence) being the salient factors in construct- tors related to overall self-esteem on the SEI. In their analysis, ing a multidimensional model of self-esteem using the 16PF they also found A (Warmth), M (Abstractness), N (Privateness), Fifth Edition. These findings are consistent with previous and O (Apprehension) as significant predictors. While factor A research using the 16PF Fifth Edition (Rieke & Conn, 1994). (Warmth) did appear in one of this study’s equations, it did not Unlike the findings reported by Rieke and Conn, the EX consistently appear across the three regression equations Pietrzak et al. 7 Table 4. Primary Factor Regression Results of CFSEI From 16PF. 2 a Scale group R A − R F Predictors Total scale A C E G H I L M N Q1 Q2 Q3 Total .74 .54 39.68** .40 .11 .27 −.09 −.12 −.13 .20 Split 1 .75 .54 20.69** −.17 .36 .12 .45 −.15 .18 .12 Split 2 .80 .60 22.34** .54 .11 .20 −.25 −.13 .16 .15 −.24 Final .72 .51 80.77** .62 .20 .10 General scale C E F H L M N O Q1 Q2 Q3 Total .65 .40 23.35** .27 .14 −.09 .30 −.12 −.11 .20 Split 1 .69 .45 17.50** .23 .14 .33 −.17 .20 −.22 Split 2 .65 .39 11.15** .42 .23 −.13 −.22 Final .60 .35 64.47** .35 .39 Social scale C F H I M Q1 Q2 Q4 Total .52 .44 37.09** .13 .21 −.11 −.30 .24 −.16 Split 1 .61 .35 13.69** .24 −.18 −.33 .29 −.27 Split 2 .48 .20 10.60** .43 .18 .20 Final Personal scale A C H N Q2 Q4 Total .67 .44 37.09** −.10 .59 .15 .13 −.14 Split 1 .64 .39 16.30 .50 .30 .20 .17 −.19 Split 2 .72 .51 115.85** .72 Final .40 157.61** .64 Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition; CFSEI = Culture-Free Self-Esteem Inventory; R = multiple R; A − R = adjusted multiple R-square; F = F-ratio of the regression equation; Total = total sample; Split 1 = Split Validation Group 1; Split 2 = Split Validation Group 2; Final = final split sample–validated equation. Predictors are the letter names for the primary scales of the 16PF Fifth Edition. Unable to calculate. **p < .01. predicting the CFSEI-II Total score. These differences may (Apprehension) has been related to self-esteem in research be due to variability in the samples used or dissimilarities in with previous editions of the 16PF (Cattell, 1989), the cur- the measurement of self-esteem between the SEI and CFSEI. rent edition of the 16PF (Rieke & Conn, 1994), and is a cen- To clarify these differences, we suggest further research. tral component of calculation of the global factor AX Consistent with the global factor results, the CFSEI-II (Anxiety). Therefore, we expected factor O (Apprehension) Social scale score was the most difficult to predict. No pri- to take a more central role in predicting at least global self- mary factors were identified by the regression equations to esteem. However, factor O (Apprehension) was present in construct a stable prediction. The regression equations for only one of the three equations predicting the CFSEI-II the CFSEI-II General and Personal scores accounted for General scale score. In addition, it did not play a role in any approximately 35% and 40% shared variance, respectively. of the other regression equations. This finding is different The primary factor C (Emotional Stability) was the most from previous studies (Rieke & Conn, 1994), which used dif- consistent predictor of all facets of self-esteem as measured ferent samples, measures of self-esteem, and validation by the CFSEI-II. Primary factor C (Emotional Stability) was methods. Based on these results, we suggest further research the only consistent predictor for the CFSEI-II Personal self- to clarify this issue. esteem score. The results of this study indicate, as previous A secondary goal of this study was to make available split research has suggested, that factor C (Emotional Stability) is sample–validated equations for the user of the 16PF Fifth consistently related to various aspects of self-esteem (Rieke Edition. We anticipate that this will enable these users to apply & Conn, 1994). Primary factor Q1 (Openness to Change) the 16PF Fifth Edition to their setting using an alternative was the predictor present in the CFSEI-II Total score equa- model of self-esteem. The following regression equations serve only as experimental measures. Further research to tion that was not present in the CFSEI-II General score equation. examine their reliability and validity is obviously needed; Perhaps the most interesting finding in this study was the however, they may serve as an entry point for examining the lack of primary factor O (Apprehension) in any of the final relationship between a multidimensional model of self-esteem and the 16PF Fifth Edition. Table 5 provides the final regres- equations, although it correlated significantly with the sions of the CFSEI-II from the 16PF Fifth Edition. Descriptive CFSEI-II Total scale and its subscales (see Table 2). Factor O 8 SAGE Open Table 5. Final Regression Equations of the CFSEI From 16PF (N = 234). a b Scale M SD SEE 16PF equation Global factor score equations Total 6.0 1.08 1.18 (−.55 × AX) + (.53 × IN) + 6.21 General 6.0 0.97 1.36 (−.41 × AX) + (.52 × IN) + 5.41 Social 6.4 0.84 1.60 (.20 × EX) + (−.30 × AX) + (.17 × TM) + (.34 × IN) + 4.19 Personal 5.8 1.26 1.55 (−.77 × AX) + (.19 × TM) + (.56 × IN) + 6.16 Primary factor score equations Total 6.1 1.16 1.18 (.46 × C) + (.24 × H) + (.10 × Q1) + 1.56 General 5.9 0.99 1.34 (.32 × C) + (.32 × H) + 2.32 Personal 5.8 1.26 1.54 (.70 × C) + 1.91 Note. 16PF = Sixteen Personality Factor Questionnaire Fifth Edition; CFSEI = Culture-Free Self-Esteem Inventory; AX = Anxiety; IN = Independence; EX = Extroversion; TM = Tough-Mindedness. Represents the standard error of the estimate. 16PF letter names used are those from the 16PF Fifth Edition Technical Manual (Rieke & Conn, 1994). statistics are there, as well. The regression equations are References reported using Sten scores for both the predictor and depen- Arlot, S., & Celisse, A. (2010). A survey of cross-validation pro- dent variables as is traditional when using the 16PF Fifth cedures for model selection. Statistics Survey, 4, 40-79. Edition (Rieke & Conn, 1994). doi:10.1214/09-SS054. Retrieved from http://projecteuclid. org/euclid.ssu/1268143839 Battle, J. (1977a). A comparison of two self-report inventories. Limitations and Recommendations Psychological Reports, 41, 159-160. Battle, J. (1977b). Test-retest reliability of the Canadian Self-Esteem The limitations of this study simultaneously offer interpreta- Inventory for Adults. Perceptual & Motor Skills, 44, p. 38. tive caution and research opportunities. First, the sample, Battle, J. (1978). Relationship between self-esteem and depression. although racially and ethnically diverse, was biased heavily Psychological Reports, 42, 745-746. toward females (71% of the sample). Future research should Battle, J. (1992). Culture-free self-esteem inventories: Examiners include larger and more representative samples. We suggest manual (2nd ed.) Austin, TX: Pro-Ed. greater consideration of the variables gender and self- Brooke, S. L. (1995). Critical analysis of the culture-free self-esteem esteem. Along these lines, other measures of self-esteem inventories. Measurement and Evaluation in Counseling and might be utilized or compared with the CFSEI-II. Third, Development, 27, 248-252. multiple regression is sensitive to the type of validation uti- Cattell, H. (1989). The 16PF: Personality in depth. Champaign, IL: Institute for Personality and Ability Testing. lized, especially data splitting (Arlot, & Celisse, 2010). Cattell, H., & Mead, A. (2008). The Sixteen Personality Factor Furthermore, with any multiple regression study, there is no Questionnaire (16PF). In G. Boyle, G. Matthews, & D. substitution for multiple replication studies to refine the Saklofske (Eds.), The SAGE handbook of personality finding of any given study. Fourth, as with any assessment theory and assessment: Volume 2—Personality measure- tools, the application and interpretation of the 16PF Fifth ment and testing (pp. 135-160). London, England: SAGE. Edition and CFSEI-II need to take place in the context of doi:10.4135/9781849200479.n7 sound clinical practice marked by an appropriate under- Cattell, H., & Schuerger, J. M. (2003). Essentials of 16PF assess- standing of assessment. Clinicians should use assessment ment. Hoboken, NJ: John Wiley. materials with proper training and experiences. Furthermore, Cattell, R., Eber, H., & Tatsuoka, M. (1970). Handbook for the for clinicians to reach meaningful and useful interpretations 16PF. Champaign, IL: Institute for Personality and Ability of assessment data, they should integrate such information Assessment. in the context of other relevant client data, such as family Chapman, P. L., & Mullis, A. K. (2002). Readdressing gen- der bias in the Coopersmith Self-Esteem Inventory-Short and cultural influences. Form. The Journal of Genetic Psychology, 163, 403-409. doi:10.1080/00221320209598692 Declaration of Conflicting Interests Cicchetti, D. V. (1994). Guidelines, criteria, and rules of thumb for The author(s) declared no potential conflicts of interest with respect evaluating normed and standardized assessment instruments in to the research, authorship, and/or publication of this article. psychology. Psychological Assessment, 6, 284-290. Clark, W. L., & Blackwell, T. L. (2007). 16PF® (5th edi- Funding tion) Personal Career Development Profile: Test review. The author(s) received no financial support for the research and/or Rehabilitation Counseling Bulletin, 50, 247-250. doi:10.1177 authorship of this article. /00343552070500040601 Pietrzak et al. 9 Coopersmith, S. (1981). Self-esteem inventories. Palo Alto, CA. concepts, different outcomes. American Sociological Review, Consulting Psychologist Press. 60, 141-156. Dancer, L. J., & Woods, S. A. (2006). Higher-order factor structures SPSS. (2007). SPSS for Windows [Version 16.0]. Chicago, IL: and intercorrelations of the 16PF5 and FIRO-B. International Author. Journal of Selection and Assessment, 14, 385-391. Stark, S., Chernyshenko, O. S., Drasgow, F., & Williams, B. Francis, L. J. (1997). Coopersmith’s model of self-esteem: Bias A. (2006). Examining assumptions about item respond- toward the stable extravert? The Journal of Social Psychology, ing in personality assessment: Should ideal point methods 137, 139-142. be considered for scale development and scoring? Journal Francis, L. J., & James, D. (1998). Is there gender bias in the of Applied Psychology, 91, 25-39. doi:10.1037/0021- short form Coopersmith Self-Esteem Inventory? Educational 9010.91.1.25 Research, 40, 83-89. Walsh, W., & Betz, N. (1995). Tests and assessment (3rd ed.). Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Englewood Cliffs, NJ: Prentice Hall. Multivariate data analysis (5th ed.). Upper Saddle River, NJ: Prentice Hall. Author Biographies Hair, J. F., Babin, B. J., & Anderson, R. E. (2009). Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall. Dale Pietrzak, Ed.D., LPC-MH (SD), CCMHC is the Director of Institute for Personality and Ability Testing. (2009). 16PF fifth edi- Institutional Research and Assessment at the University of Idaho. tion manual. Champaign, IL: Author. He has been a clinician, faculty member and and/or administrator Janda, L. (1998). Psychological testing: Theory and application. since 1987. Boston, MA: Allyn & Bacon. Betsy Page is an Associate Professor of Counselor Education and Kawash, G., Clews, J., & Keating, L. (1989). An examination of Supervision at Kent State University. Her interests include group self-esteem in the context of general personality functioning. work, suicide assessment, and clinical supervision. Multivariate Experimental Research, 9, 27-33. Luhtanen, R., & Crocker, J. (1992). A Collective Self-Esteem James S. Korcuska, Ph.D., LPC, NCC is an associate professor at Scale: Self-evaluation of one’s social identity. Personality and North Dakota State University in Counselor Education and Social Psychology Bulletin, 18, 302-318. Supervision. He also coordinates the program’s clinical training Nunnally, J., & Bernstein, I. (1994). Psychometric theory (3rd ed.). experiences and teaches advanced assessment. New York, NY: McGraw-Hill. Amber Bach Gorman Ph.D, works full time at the North Dakota Rieke, M. L., & Conn, S. R. (1994). Psychological adjustment and State University Counseling Center as a mental health counselor self-esteem. In S. Conn & M. Rieke (Eds.), 16PF fifth edition and clinical supervisor. She is also an adjunct lecturer in Counselor technical manual (pp. 143-162). Champaign, IL: Institute for Education and Supervsion at North Dakota State University and Personality and Ability Testing. Minnesota State University – Moorhead’s Counseling and Student Rosenberg, M., Schooler, C., Schoenbach, C., & Rosenberg, F. Affairs program. (1995). Global self-esteem and specific self-esteem: Different

Journal

SAGE OpenSAGE

Published: Oct 14, 2015

Keywords: Sixteen Personality Factor Questionnaire Fifth Edition; self-esteem; model; personality; statistical analysis

There are no references for this article.