Personality and Individual Dierences 41 (2006) 663673

www.elsevier.com/locate/paid

The Student Readiness Inventory and the Big Five: Examining social desirability and college academic performance
Christina Hamme Peterson, Alex Casillas, Steven B. Robbins
ACT, Inc., Iowa City, IA 52243-0168, United States Received 31 October 2005; received in revised form 13 February 2006 Available online 5 May 2006
*

Abstract This study compared the relative power of a Big Five measure, the Big Five Inventory (BFI), and a measure of academic-related factors, the Student Readiness Inventory (SRI), for predicting college academic performance (GPA) using a sample of 468 college students from 2- and 4-year institutions. The extent to which social desirability inuenced construct and predictive validity was also examined. Both the SRI and the BFI were signicant predictors of college GPA, but hierarchical regression results showed that the SRI scales accounted for a larger proportion of variance (range = 22% to 29%) than the BFI scales (range = 3% to 9%). As expected, the impact of social desirability on the relations between the two instruments, as well as between each instrument and college GPA, was minimal. Further evidence of the construct validity of the SRI was provided by the emergence of moderate to strong relations between the BFI (Extraversion, Neuroticism, and Conscientiousness) and the SRI (Social Activity, Emotional Control, and Academic Discipline), respectively. 2006 Elsevier Ltd. All rights reserved.
Keywords: College academic performance; Big Five; Student Readiness Inventory

Corresponding author. Tel.: +1 319 337 1211; fax: +1 319 341 2284. E-mail address: steve.robbins@act.org (S.B. Robbins).

0191-8869/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.paid.2006.03.006

664

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

1. Introduction With 6-year graduation rates averaging as low as 34% at some institutions (Swail, 2004), there is some concern that students are ill prepared for college. To circumvent academic diculties, early identication of at-risk students has become critical and has centered on standardized achievement tests and high school GPA. However, certain groups are disadvantaged by standardized test scores (Jencks & Phillips, 1998; Sackett, Schmitt, Ellingson, & Kabin, 2001), and use of schoolbased performance indicators do not take into account dierences across schools in expectation and performance (Bassiri & Schulz, 2003; Tam & Sukhatme, 2003), nor high school grade ination (Ziomek & Svec, 1995). Standardized tests have been demonstrated as valid methods for predicting college outcomes (ACT, 1997; Willingham, Lewis, Morgan, & Ramist, 1990); however, these account for only a portion of the variance and should be augmented by the assessment of other relevant attributes (Sternberg, 1986). Research has demonstrated the contribution of noncognitive factors in predicting college student academic success (Musgrave-Marquart, Bromley, & Dalley, 1997; Wolfe & Johnson, 1995). Similarly, numerous nonacademic programs aimed at increasing retention and academic performance are recommended for college use (e.g., Keels, 2004; Martin, Swartz-Kulstad, & Madson, 1999). Postsecondary institutions planning early identication and intervention programs require clear noncognitive assessments to customize their interventions to at-risk student characteristics and needs (Noonan, Sedlacek, & Veerasamy, 2005; Sternberg, 1986). Big Five measures have yielded some results in predicting academic success, through relations of performance with conscientiousness (Busato, Prins, Elshout, & Hamaker, 2000; MusgraveMarquart et al., 1997) and agreeableness (Fritzche, McIntire, & Yost, 2002). These ndings, however, oer only broad factors to serve as a basis for intervention and may not be as predictive as more narrow facets of personality (Lounsbury, Sundstrom, Loveland, & Gibson, 2002; Paunonen & Ashton, 2001). To address this issue, Le, Casillas, Robbins, and Langley (2005) developed the Student Readiness Inventory (SRI), a measure of psychosocial and academic-related factors (PSFs), found in a recent meta-analysis to be predictive of academic performance (Robbins et al., 2004). The SRI is comprised of 10 scales that measure a range of academic-related constructs (see Table 1). Research found that SRI constructs were relatively independent of standardized achievement test scores (Robbins, Allen, Casillas, Peterson, & Le, in press), and oered a viable alternative to forecasting academic performance. In a longitudinal study, the SRI constructs, demographics, and traditional predictive factors were examined to determine what role each played in the prediction of college performance (Robbins et al., in press). With a sample of 14,464 students from 23 2-year and 25 4-year institutions, incremental predictive validity models were run. After controlling for SES, race/ethnicity, gender and dierences between postsecondary institutions, PSFs were signicant predictors of academic performance above all other factors in the models. This has important implications for institutions aiming to enhance student academic success. With only high school GPA and achievement test scores to serve as diagnostic tools, college counselors and advisors are limited in their identication of at-risk students and the customization of their interventions. The SRI PSFs oer a skills and personality based interposing avenue which, coupled with cognitive interventions, may yield better results.

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673 Table 1 SRI scale denitions and sample items Scale Academic discipline Denition The amount of eort one puts into schoolwork and the degree to which a student is hardworking and conscientious The belief in ones ability to perform well in school Ones commitment to staying in college and getting a degree Attentiveness to others feelings and exibility in resolving conicts with others Ones responses to strong feelings The extent to which one strives to follow through on commitments and obligations The strength of ones eorts to achieve objectives and end goals Ones comfort in meeting and interacting with other people Ones feelings of connection and involvement with the college community The extent to which one believes one knows how to assess an academic problem, organize a solution, and successfully complete academic assignments Sample item I consistently do my school work well

665

Academic self-condence Commitment to college Communication skills

I achieve little for the amount of time I spend studying A college education will help me achieve my goals Im willing to compromise when resolving a conict I have a bad temper It is important for me to nish what I start I bounce back after facing disappointment or failure I avoid activities that require meeting new people I feel part of this college

Emotional control General determination

Goal striving Social activity Social connection

Study skills

I summarize important information in diagrams, tables, or lists

This study addresses the construct validation of the SRI. In Le et al. (2005), the SRI dimensions were supported using factor analysis. This study is designed to assess SRI constructs through their relations to the Big Five and the degree to which they predict academic achievement above the Big Five factors. We hypothesize that several SRI scales will relate strongly to the Big Five: Social Activity to Extraversion, Emotional Control to Neuroticism, and Academic Discipline to Conscientiousness. Based on Paunonen and Ashtons (2001) nding that more narrow facets of personality are stronger predictors of behavior than the Big Five, we also hypothesize that the SRI constructs will account for greater variance in predicting academic success than the Big Five. In addition, we explore the impact of socially desirable responding on the SRI. A concern in selfreport assessment is the potential for response distortion. Studies have found distortion when participants are asked to fake-good or bad responses (Douglas, McDaniel, & Snell, 1996; Viswesvaran & Ones, 1999). Similarly, there is evidence of distortion in noncognitive measures of constructs such as goal orientation (Tan & Hall, 2005) and the Big Five (Barrick & Mount, 1996). Although some would argue that this bias can result in validity decay (cf. Douglas et al., 1996), there is growing evidence that the susceptibility to faking intrinsic to some self-report measures has little impact on construct or criterion validity. Researchers have shown that social desirability

666

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

has little inuence on personality measure factor structures (Ellingson, Smith, & Sackett, 2001), selection measure construct validity (Smith & Ellingson, 2002), and prediction of performance (Barrick & Mount, 1996). Based on these ndings, we expect that there may be evidence of response distortion due to social desirability in the SRI, but this will not impact the scales construct or criterion validity.

2. Methods 2.1. Participants A total of 468 college students from ve institutions [three 2-year (38.1%) and two 4-year institutions (61.9%)] volunteered to participate in the study (average participation rate based on eligible students was 23%). Institutions ranged in selectivity from open enrollment (n = 178) to traditional (n = 148) and selective (n = 142). Participants ages ranged from 17 to 65, with a mean age of 21.2 years (SD = 7.9 years; median = 18 years). The sample was 73.3% female and 83.8% Caucasian. We created an SES index using a standardization of the combination of fathers education level, mothers education level, and annual family income. SES index range was 2.99 to 4.67 (M = .24, SD = 1.89). Of these students, only 359 had a college GPA, thus analyses featuring GPA are based on this subsample. Students from the subsample ranged in age from 17 to 55 years (M = 20.9 years, SD = 7.3 years, median = 18 years) and were 74.9% female and 84.0% Caucasian. 2.2. Measures The Big Five Inventory (BFI; Benet-Martinez & John, 1998; John & Srivastava, 1999) is a 44item instrument that assesses the Big Five (Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness). BFI items were developed from denitions of expert and observer ratings of personality, and veried by subsequent factor analysis (John, 1990). Items are scored using a 5point, Likert scale that ranges from disagree strongly to agree strongly. BFI scales include 810 items each, and have demonstrated moderate to high internal consistency reliability (alpha range = .79 to .88; median = .82) as well as substantial convergent/discriminant relations with other Big Five instruments (John & Srivastava, 1999). The MarloweCrowne Social Desirability Scale (SDS) is a 33-item true/false inventory designed to tap social desirability without pathology-related content. Internal reliability estimates (KR-20 or coecient alpha) have ranged from the low 70s to the high 80s and temporal stability has been reported at .89 (Barger, 2002; Crowne & Marlowe, 1960). The Student Readiness Inventory (SRI) is a 108-item inventory comprising 10 scales (Table 1) that measure students academic-related personality facets and skills. The inventory was designed to assist postsecondary institutions in identifying and intervening with students at risk for drop out or poor performance. Items are scored using a 6-point, Likert scale that ranges from strongly disagree to strongly agree. Scales range from 10 to 12 items and have demonstrated moderate to high internal consistency reliabilities (alpha range = .81 to .87; median = .84), as well as incremental validity over demographic, institutional, and standardized achievement variables (Le et al., 2005; Robbins et al., in press).

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

667

2.3. Procedure Test materials were shipped to participating institutions after institutions received clearance from their respective institutional review boards. College representatives were instructed to administer the survey to students within the rst 6 weeks of the Fall 2004 semester. Participants provided informed consent in which they were assured that all data would be stored securely and condentially. Following test administration, institutions were asked to collect and return test materials and to provide students cumulative GPA. A follow-up social desirability survey was sent to participating students via mail.

3. Results Table 2 features descriptive statistics and internal consistency of all study scales. Means and standard deviations for the BFI were generally consistent with those obtained in previous research (e.g., Benet-Martinez & John, 1998), with students in this sample scoring slightly higher on Extraversion, Agreeableness, and Conscientiousness (d = .37, .49, and .48, respectively). Similarly, means and standard deviations for the SRI were consistent with those obtained for a normative

Table 2 Descriptive statistics and coecient alpha of study variables Measure/scale (# of items) BFI Neuroticism (8) Extraversion (8) Openness (10) Agreeableness (9) Conscientiousness (9) MarloweCrowne Social Desirability Scale (33) SRI Commitment to college (10) Goal striving (10) Academic discipline (10) General determination (11) Study skills (12) Communication skills (10) Social activity (10) Social connection (11) Academic self-condence (12) Emotional control (12) GPA Score range 840 840 1050 945 945 033 1060 1060 1060 1166 1272 1060 1060 1166 1272 1272 04.0 M 22.5 28.0 36.8 36.9 35.3 18.6 53.9 51.5 48.3 59.4 54.3 51.7 43.2 50.1 53.7 50.7 3.10 SD 6.5 6.5 6.4 5.6 5.6 6.1 7.4 6.1 8.6 5.8 9.4 6.0 9.7 9.2 10.4 10.0 .74 a .83 .84 .80 .83 .81 .83 .90 .85 .89 .87 .88 .84 .88 .83 .88 .86

Note. N = 468. M, mean; SD, standard deviation; a, Cronbachs coecient alpha. BFI, Big Five Inventory; SRI, Student Readiness Inventory.

668

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

sample (Robbins et al., in press), with the current sample scoring slightly higher on General Determination and Communication Skills (d = .42 and .32, respectively). The mean and standard deviation obtained for the SDS also were somewhat higher than those obtained in recent research using college students (e.g., Ballard, 1992; Loo & Thorpe, 2000) (d = .49), but lower than those obtained in forensic samples, in which scores tend to be higher than those of nonforensic samples (Andrews & Meyer, 2003). All alphas fell in the moderate to high range, with the BFI scales between .80 and .84 (median = .83), the SRI scales between .83 and .90 (median = .88), and the SDS at .83, all of which are consistent with previous research. Correlations between social desirability and the BFI scales ranged from .17 (Extraversion, Openness) to .59 (Agreeableness), with a median of .42. Correlations between social desirability and the SRI scales ranged from .21 (Social Connection, Academic Self-Condence) to .56 (Emotional Control), with a median of .39. Moreover, there was no relation (r = .01) between social desirability and cumulative GPA. Table 3 features correlations between the BFI and SRI and between the SRI and GPA with and without the eects of social desirability. As expected, there were several convergent relations between the BFI and SRI: BFI Neuroticism was strongly related to SRI Emotional Control and, to a lesser extent, SRI Social Activity. BFI Extraversion was strongly related to SRI Social Activity. BFI Agreeableness was strongly related to SRI Communication Skills and, to a lesser extent, Emotional Control and General Determination. BFI Conscientiousness was strongly related to the SRIs Academic Discipline, General Determination, Goal Striving, and, to a lesser extent, Study Skills. Further, relations between BFI Openness and SRI scales were generally small (range = .10 to .34, median = .26). Thus, these data suggest that, with the exception of openness, the SRI taps a variety of personality domains. The eects of social desirability, as noted by the dierence between the left (zero-order correlations) and right (social desirability partialed out) columns of each set of correlations in Table 3, are relatively small (range = .01 to .21, median = .06). The two scales that evidenced the highest
Table 3 Correlations of the Student Readiness Inventory with the Big Five Inventory and Cumulative GPA with and without Social Desirability SRI scales Commitment to college Goal striving Academic discipline General determination Study skills Communication skills Social activity Social connection Academic self-condence Emotional control BFI scales N .20 .43 .26 .20 .20 .27 .51 .36 .36 .67 (.12) (.31) (.11) (.04) (.02) (.11) (.45) (.30) (.31) (.58) E .16 .26 .16 .13 .18 .20 .73 .45 .17 .17 (.12) (.21) (.10) (.07) (.12) (.14) (.72) (.42) (.14) (.09) O .14 .31 .15 .20 .33 .26 .30 .25 .34 .23 (.10) (.27) (.09) (.14) (.29) (.21) (.27) (.22) (.32) (.16) A .31 .44 .41 .50 .39 .67 .35 .39 .16 .61 (.22) (.27) (.24) (.35) (.17) (.57) (.22) (.34) (.05) (.41) C .44 .60 .66 .64 .50 .37 .33 .26 .43 .47 (.38) (.50) (.59) (.56) (.37) (.21) (.23) (.18) (.38) (.27) .27 .25 .50 .30 .17 .11 .06 .09 .34 .08 (.27) (.27) (.55) (.33) (.18) (.12) (.06) (.09) (.34) (.09) GPAa

Note. N = 468. Social desirability has been partialed out from correlations in parentheses. Correlations P.13 are signicant (p 6 .01). Correlations P j.50j are in bold. a n = 359.

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

669

levels of dierences between the zero-order correlations and the correlations with social desirability partialed out were agreeableness (dierence range = .06 to .21, median = .14) and neuroticism (dierence range = .05 to .18, median = .11), which is consistent with the literature (e.g., Graziano & Tobin, 2002). This suggests that social desirability had little impact on relations between the BFI and SRI. Six SRI scales evidenced signicant relations with GPA, with Academic Discipline showing the strongest relation. The dierences between the zero-order correlations of SRI scales with GPA, and the same set of correlations with social desirability partialed out, were minimal to nonexistent (dierence range = .00 to .05, median = .01) If the dierences were of greater magnitude it may
Table 4 Summary of stepwise regressions using demographics, the BFI, and the SRI for predicting college GPA Step/variables Version A Step 1. Demographics Age Ethnicity SES Step 2. BFI scales Extraversion Agreeableness Conscientiousness Step 3. SRI scales Academic discipline Study skills Academic self-condence B SE B b DR2 Total Adj. R2

.019 .110 .060

.006 .034 .022

.184 .172 .155 .070*

.014 .019 .050

.007 .009 .008

.122 .143 .361 .092*

.050 .015 .014

.007 .005 .004

.565 .180 .184 .221* .347

Version B Step 1. Demographics Age Etnicity SES Step 2. SRI scales Academic discipline Study skills Academic self-condence Step 3. BFI scales Agreeableness

.019 .110 .060

.006 .034 .022

.184 .172 .155 .070**

.047 .014 .015

.006 .005 .004

.544 .167 .203 .286*

.027

.010

.200 .026* .347

Note. N = 359. BFI, Big Five Inventory. SRI, Student Readiness Inventory. * p 6 .05.

670

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

have indicated that social desirability acted as a suppressor of the relation between the SRI scales and GPA (see Smith, Ager, & Williams, 1992, for a discussion on suppression). This suggests that, as hypothesized, social desirability did not play a role in the relations between the SRI and academic performance. Finally, Table 4 features two sets of hierarchical regressions demonstrating the incremental predictive power of the SRI scales. For version A, we rst entered demographic variables (age, gender, ethnicity, and SES) to predict cumulative GPA. This accounted for approximately 7% of the variance. Second, we added the Big Five variables, which accounted for approximately 9% of additional variance [F(5, 339) = 7.42, p < .001]. Last, we added the 10 SRI scales. This step accounted for approximately 22% of additional variance [F(10, 329) = 11.76, p < .001]. In combination, demographics, the BFI, and SRI scales accounted for approximately 35% of the variance in cumulative GPA (based on adjusted R-squared estimates). All variables in the model were entered using a stepwise procedure and those featured in Table 4 were signicant predictors. To understand the relative contributions of each set of scales more clearly, we redid this analysis reversing the order in which the BFI and SRI scales were entered. For version B, we entered the SRI scales after demographic variables. This accounted for approximately 29% of the additional variance [F(10, 334) = 14.86, p < .001]. Then, we added the BFI scales, which accounted for approximately 3% of additional variance [F(5, 329) = 2.79, p = .02]. Although both are significant predictors of academic performance, the SRI accounted for a larger proportion of variance (range = 22% to 29%) than the BFI (range = 3% to 9%).

4. Discussion This study extends the construct validity of the SRI in three ways by: (1) establishing a pattern of convergent/discriminant relations with the Big Five, (2) demonstrating that the SRI scales are not any more vulnerable to socially desirable responding than other self-report measures, and (3) showing incremental validity over the Big Five in predicting academic performance. As expected, several of the SRI scales showed clear convergent relations to the Big Five (e.g., Emotional Control to Neuroticism, Social Activity to Extraversion, Communication Skills to Agreeableness, Academic Discipline to Conscientiousness), suggesting that they tap personality variance also tapped by Big Five inventories. Other SRI scales (e.g., Social Connection) showed more multifaceted patterns, providing evidence of discriminant validity. It remains to be seen whether these patterns replicate in future research. Further, the study revealed low to moderate relations between the SRI scales and Openness, suggesting that the SRI does not tap Openness to the same degree as the other Big Five domains. Although the literature linking Openness to academic performance is sparse, some ndings have related it to academic achievement (e.g., Asendorpf & Van Aken, 2003). Thus, a weakness of the SRI is that it does not tap this personality characteristic. Also as expected, this study showed that the SRI is not more vulnerable to the eects of selfpresentation than other self-report measures. For instance, the relations between SRI scales and social desirability (range = .21 to .56, median = .39) were in the same range and magnitude as those of the BFI scales (range = .17 to .59, median = .42). Moreover, correlations between the SRI and BFI evidenced relatively small dierences when the eects of social desirability were par-

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

671

tialed out (dierence range = .01 to .21, median = .06), and social desirability did not play a role in the relations between the SRI and GPA. Altogether, these results seem consistent with the literature regarding the small to negligible eects of social desirability on self-report measures of personality (e.g., Barrick & Mount, 1996; Ellingson et al., 2001). Finally, we found that the SRI scales provided incremental validity in predicting academic performance, as measured by cumulative GPA, when controlling for demographics and the Big Five. Indeed, in our regression analyses, the SRI scales accounted for a larger proportion of variance than the BFI scales, regardless of the order in which the scales were entered into the analysis. Further, in the model in which the SRI scales were entered rst, the subsequent entry of the BFI scales into the model only accounted for an additional 3% of the variance. Similar to the ndings of Paunonen and Ashton (2001), in which facet-level scales of the Big Five were better predictors than domain-level ones, our results demonstrate the tradeo between delity (i.e., measuring one set of characteristics in depth) and bandwidth (broadly measuring many characteristics, with some loss of detail). The results suggest that the SRI, which is designed to tap academic-related facets, is a stronger predictor of GPA than the BFI, which is designed to measure much broader personality domains. This study had several strengths and limitations. Strengths were use of well-known measures, such as the BFI, as well as use of the SRI, which is based on existing theoretical and empirical literature on academic performance and persistence and is normed on a large, representative sample. Also, the study had a reasonable sample size including students from several 2- and 4-year institutions. In terms of limitations, our sample included a larger proportion of women and a smaller proportion of minorities than is typical of institutional enrollments. Although this is a common issue in psychological and educational research, ndings would be more generalizable with a more representative sample. Further, because of the cross-sectional nature of the study, we did not obtain information about retention, an important outcome. Another weakness of the study was that approximately 23% of students did not have a college GPA. Although this portion did not appear substantially dierent from those students with a GPA, it would be easier to generalize the regression ndings with complete data. Finally, we are aware that the SDS has been criticized (e.g., Barger, 2002). To improve on the measurement of this construct, the study could have used multiple measures of social desirability, thus limiting the eects of measurement error from any one measure. One of our primary interests is the predictive utility of noncognitive variables as measured by the SRI on academic performance. We believe that using instruments specically tailored to measure academic-related criteria (versus those designed to measure broad domains) provide higher utility. By applying our ndings, we hope to improve the technology available for identifying students relative strengths and weaknesses, particularly for those students who may be at risk of poor performance. In future research, we hope to identify dierences in predictive utility of the SRI based on institution-type and student ability. In addition, we hope to evaluate interventions aimed at specic SRI constructs for their eectiveness in improving student outcomes. References
ACT, Inc. 1997. ACT assessment technical manual. Iowa City, IA: Act, Inc. Andrews, P., & Meyer, R. G. (2003). MarloweCrowne Social Desirability Scale and short form C: Forensic norms. Journal of Clinical Psychology, 59, 483492.

672

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

Asendorpf, J. B., & Van Aken, M. A. G. (2003). Validity of big ve personality judgments in childhood: A 9-year longitudinal study. European Journal of Personality, 17, 117. Ballard, R. (1992). Short forms of the MarloweCrowne Social Desirability Scale. Psychological Reports, 71, 11551160. Barger, S. D. (2002). The MarloweCrowne aair: Short forms, psychometric structure, and social desirability. Journal of Personality Assessment, 79, 286305. Barrick, M. R., & Mount, M. K. (1996). Eects of impression management and self-deception on the predictive validity of personality constructs. Journal of Applied Psychology, 81, 261272. Bassiri, D., & Schulz, E. M. (2003). Constructing a universal scale of high school course diculty. Journal of Educational Measurement, 40, 147161. Benet-Martinez, V., & John, O. P. (1998). Los Cinco Grandes across cultures and ethnic groups: Multitrait multimethod analyses of the Big Five in Spanish and English. Journal of Personality and Social Psychology, 75, 729750. Busato, V. V., Prins, F. J., Elshout, J. J., & Hamaker, C. (2000). Intellectual ability, learning style, personality, achievement motivation and academic success of psychology students in higher education. Personality and Individual Dierences, 29, 10571068. Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349354. Douglas, E. F., McDaniel, M. A., & Snell, A. F. (1996, August). The validity of non-cognitive measures decays when applicants fake. Paper presented at the annual meeting of the Academy of Management, Cincinnati, OH. Ellingson, J. E., Smith, D. B., & Sackett, P. R. (2001). Investigating the inuence of social desirability on personality factor structure. Journal of Applied Psychology, 86, 122133. Fritzche, B. A., McIntire, S. A., & Yost, A. P. (2002). Holland type as a moderator of personalityperformance predictions. Journal of Vocational Behavior, 60, 422436. Graziano, W. G., & Tobin, R. M. (2002). Agreeableness: Dimension of personality or social desirability artifact? Journal of Personality, 70, 695727. Jencks, C., & Phillips, M. (1998). The blackwhite test score gap. Washington, DC: Brookings Institutions. John, O. P. (1990). The Big Five factor taxonomy: Dimensions of personality in the natural language and in questionnaires. In L. A. Pervin (Ed.), Handbook of personality: Theory and research (pp. 66100). New York: Guilford. John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality (2nd ed., pp. 102138). Guilford: New York. Keels, C. L. (2004). Keeping students aoat. Black Issues in Higher Education, 21, 3239. Le, H., Casillas, A., Robbins, S., & Langley, R. (2005). Motivational and skills, social, and self-management predictors of college outcomes: Constructing the Student Readiness Inventory. Educational and Psychological Measurement, 65, 482508. Loo, R., & Thorpe, K. (2000). Conrmatory factor analyses of the full and short versions of the MarloweCrowne Social Desirability Scale. Journal of Social Psychology, 140, 628635. Lounsbury, J. W., Sundstrom, E., Loveland, J. L., & Gibson, L. W. (2002). Broad versus narrow personality traits in predicting academic performance of adolescents. Learning and Individual Dierences, 14, 6575. Martin, W. E., Swartz-Kulstad, J. L., & Madson, M. (1999). Psychosocial factors that predict the college adjustment of rst-year undergraduate students: Implications for college counselors. Journal of College Counseling, 2, 121133. Musgrave-Marquart, D., Bromley, S. P., & Dalley, M. B. (1997). Personality, academic attribution, and substance use as predictors of academic achievement in college students. Journal of Social Behavior and Personality, 12, 501511. Noonan, B. M., Sedlacek, W. E., & Veerasamy, S. (2005). Employing noncognitive variables in admitting and advising community college students. Community College Journal of Research and Practice, 29, 463469. Paunonen, S. V., & Ashton, M. C. (2001). Big ve factors and facets and the prediction of behavior. Journal of Personality and Social Psychology, I81, 524539. Robbins, S., Allen, J., Casillas, A., Peterson, C. H., & Le, H. (in press). Unraveling the dierential eects of motivational and skills, social and self-management measures from traditional predictors of college outcomes: A prospective study. Journal of Educational Psychology.

C.H. Peterson et al. / Personality and Individual Dierences 41 (2006) 663673

673

Robbins, S., Lauver, K., Le, H., Davis, D., Langley, R., & Carlstrom, A. (2004). Do psychosocial and study skill factors predict college outcomes? A meta-analysis. Psychological Bulletin, 130, 261288. Sackett, P. R., Schmitt, N., Ellingson, J. E., & Kabin, M. B. (2001). High-stakes testing in employment, credentialing and higher education. American Psychologist, 56, 302318. Smith, D. B., & Ellingson, J. E. (2002). Substance versus style: A new look at social desirability in motivating contexts. Journal of Applied Psychology, 87, 211219. Smith, R. L., Ager, J. W., & Williams, D. L. (1992). Suppressor variables in multiple regression/correlation. Educational and Psychological Measurement, 52, 1729. Sternberg, R. J. (1986). What would better intelligence tests look like? In Measures in the college admission process (pp. 146150). New York: College Entrance Examination Board. Swail, W. S. (2004). Legislation to improve graduation rates could have the opposite eect. The Chronicle of Higher Education, 5(20), B16, January. Tam, M. S., & Sukhatme, U. (2003). The importance of high school quality in university admissions decisions. College and University, 78, 38. Tan, J. A., & Hall, R. J. (2005). The eects of social desirability bias on applied measures of goal orientation. Personality and Individual Dierences, 38, 18911902. Viswesvaran, C., & Ones, D. S. (1999). Meta-analyses of fakability estimates: Implications for personality measurement. Education and Psychological Measurement, 59, 197210. Willingham, W. W., Lewis, C., Morgan, R., & Ramist, L. (1990). Predicting college grades: An analysis of institutional trends over two decades. Princeton, NJ: Educational Testing Service. Wolfe, R. N., & Johnson, S. D. (1995). Personality as a predictor of college performance. Educational and Psychological Measurement, 55, 177185. Ziomek, R. L., & Svec, J. C. (1995). High school grades and achievement: Evidence of grade ination (ACT Research Rep. No. 95-3). Iowa City, IA: ACT, Inc.