Testing the Relationship Between a
Cognitive Ability Test and Player Success:
The National Football League Case
Arthur J. Adams & Frank E. Kuzmits
University of Louisville, USA
The purpose of this study was to examine the relationship between the Wonderlic Personnel Test (WPT) and subsequent National Football League (NFL) success. The WPT is a well known and established measure of cognitive ability that is used by the NFL in their annual combine, a pre-draft assessment of player mental and physical skills. The authors investigated the relationship between WPT scores and NFL success for all players drafted at 3 different offensive positions over a recent 6-year period. Using multiple measures of NFL player success, we found no evidence of any relationship. Accordingly we question the usefulness of the WPT as a combine measure. However given that other psychological constructs (for example; aggression, leadership, coachability, and self-confidence) have shown a relationship with athletic success, it seems wise for the NFL to consider expanding their mental assessments at the combine to include higher level cognitive measurements and also to examine the current research on peak performance that may provide for potential improvements in the NFL player selection process.
Identifying the psychological variables that help athletes achieve athletic success continues to be an important goal of researchers in the field of sports and athletics (Beilock & McConnell, 2004; Hays & Kenkel, 2006). The field of sports psychology has grown markedly in the past two decades, and the discovery of psychological constructs that help athletes achieve peak performance, defined as the “superior use of human potential” (Privette, 1981) and “outstanding accomplishments” (Jackson & Roberts, 1992) has led to advancements in the understanding of success in a wide variety of athletic endeavors. One research study involving peak performance and successful athletes showed that the psychological profile of peak performing athletes included high self-confidence, energy, feelings of control, concentration, positive attitudes, determination, and commitment (Krane & Williams, 2006).
Predicting Athletic Success
The development of psychological profiles of athletic success also enhances the development of selection models that assist in choosing successful athletes. In a research review of personnel selection practices in athletics, Humara (2002) found a number of psychological constructs related to various athletic endeavors, including aggression, leadership, coachability, and self-confidence. Further, affiliation and conformity may predict athletic performance; e.g.; an athlete low in conformity and affiliation may not perform well in team sports or under an autocratic coach. Based on his review, Humara (2002) concluded that in the selection of athletes, in addition to an assessment of an athlete’s past performance and bio (physiological) data, administrators should make greater use of psychological assessments, including the Athletic Motivation Inventory (AMI), the Test of Attentional and Interpersonal Style (TAIS), and the Profile of Mood States (POMS).
Cognitive Ability as a Psychological Construct
While the examination of psychological constructs and their relationships to athletic performance is not new, the role of cognitive ability and its part in athletic success, particularly in the realm of predicting athletic success, has received scant attention in the literature. Cognitive ability, also called general mental ability (GMA), remains one of the most widely used predictors of job performance (Brody, 1992; Jensen, 1986; Schmidt & Hunter, 1998). Several definitions have been put forth for cognitive ability. One approach views cognitive ability as the ability to reason, solve problems, understand complex ideas, and learn from experience (Gottfredson, 1997). Another conceptual framework views cognitive ability as a single general ability, labeled g, which includes multiple abilities such as memory, learning, cognitive speed, retrieval ability, and visual and auditory perception (Carroll, 1993). Finally, a simple definition of the concept puts cognitive ability as “the ability to learn” (Schmidt, 2002).
General Cognitive Ability, Job Performance, and Training Success
As a predictor of human behavior, cognitive ability has long been of interest to scientific investigators, with research dating back to WWI when the U.S. Army used a paper and pencil version of the Binet intelligence test to screen and classify conscripts. (Ree & Earles, 1992). In the workplace, cognitive ability has been demonstrated to be one of the strongest and most consistent predictors of job performance in traditional employment settings (Campbell, 1990; Hunter & Hunter, 1984; Schmidt & Hunter, 1998; Thorndike, 1986). Further, the predictive power of cognitive ability extends across multiple job families, including managers, salespersons, law enforcement, service workers, trades and crafts, industrial workers and vehicle operators (Hunter & Hunter, 1984). In addition cognitive ability predicts job performance for military as well as civilian occupational classes (Hunter, 1986; Ree, Earles & Teachout, 1994). However the predictive power of cognitive ability becomes stronger in more complex employment situations, such as managerial jobs and jobs facing continual and unexpected change (Hunter & Hunter, 1984; LePine, Colquitt, & Erez, 2000).
Cognitive Ability, Wonderlic Personnel Test (WPT) and the NFL
One widely used measure of cognitive ability is the Wonderlic Personnel Test (WPT). Material contained on the Wonderlic company website claims that “Since 1937, more than 120 million people at thousands of organizations worldwide have taken the WPT”…and that the test “Can be used to match job candidates or employees to jobs in which they will be effective and satisfied. WPT information helps shorten training time” (Wonderlic Inc., 2006).
The WPT measures the two principal dimensions of general intelligence: fluid intelligence (intellectual development resulting from biological factors, for example, heredity) and crystallized intelligence (intellectual development resulting from education, experience, and acculturation). Test items include word and number comparisons, disarranged sentences, and geometric figures and problems that require mathematical and logical problem solving (Furnham, Rawles, & Iqbal, 2006). Sample questions are available at www.wonderlic.com.
Wonderlic advises employers to gauge the utility of test scores by comparing test takers’ scores to norms for the relevant occupational level. For example test score norms (The maximum score is 50) for attorneys, bookkeepers, and food service workers are 29.67, 23.36, and 16.31, respectively. Accordingly Wonderlic suggests that employers seek a minimum test score of 24 for bookkeepers. Wonderlic also recommends certain test scores for broad occupational groups, e.g., 28 and above for upper level management, 20 to 26 for general clerical workers and first line supervisors, and 10 to 17 for workers who operate simple process equipment and perform routine jobs. Wonderlic also cautions employers that employment tests be used in conjunction with other job related performance predictors (Wonderlic, 2002).
As part of the NFL draft each spring, the NFL and team personnel conduct a series of tests, termed the combine, for the most promising professional prospects among college football players. The weeklong series of tests are primarily physical tests (e.g. speed, agility, and strength) and a single mental exercise, the WPT. The NFL began giving combine invitees the WPT in the early 1970s, when coach Tom Landry of the Dallas Cowboys stated that smarter players were better players and that the test results would provide important information in the NFL draft selection process (Mirabile, 2004). Since then the WPT has become a required measure during the combine, with great media attention given to test scores, which range from 0 to 50. For example a Google search of ‘2006 NFL combine and the Wonderlic personnel test’ resulted in 10,500 hits, with several of the top 20 hits focusing on star University of Texas quarterback Vince Young’s (reportedly) very low score of six (Dougherty, 2006).
As a predictor of success in the NFL, the WPT is not without controversy. NFL teams themselves question the validity of the WPT and also worry about the possibility of cheating (Mulligan, 2004). Some teams minimize the significance of the test except for extreme outliers (Merron, 2002). Still others see the WPT as a sinister NFL tool to influence a player’s marketability. As one sports writer puts it, “It’s (the WPT) a manipulative tool in the NFL’s strategy of misinformation in the run-up of head fakes and spin moves to a draft selection” (Roberts, 2006).
Predictably the Wonderlic organization supports the use of the test as part of the combine exercises. Quotes from a Wonderlic press release include:
The coaches understand that players have to be smart and think quickly to succeed on the field, and the closer they are to the ball the smarter they need to be… Whether you are hiring a mailroom clerk or a CEO, a defensive lineman or a quarterback, intelligence is an accurate determiner of success (Wonderlic Inc., 2005).
NFL and WPT Research
In spite of the NFL’s continued use of the WPT as a measure of cognitive ability, no scientific support of the WPT as a predictor of NFL success has been provided by either the NFL or the Wonderlic organization (Lyons, Michel & Hoffman, 2005). Further, a review of the literature uncovered only two studies that have examined the ability of the WPT to predict professional football success. In the first study the WPT results of 261 NFL players selected in the 2002 draft were correlated with NFL performance for 2002 and 2003. Performance data were position specific; for example performance criteria for the quarterback position included the NFL quarterback rating (discussed in a later section); criteria for the running back position included yards gained, pass receptions and yards, and touchdowns. The results failed to show a statistical relationship between the WPT and NFL performance for either year or both years combined. The WPT also failed to predict overall draft order for 2002 and 2003 (Lyons, Michel & Hoffman, 2005). In the second study the relationship between WPT scores and NFL performance was examined for 82 quarterbacks drafted in the years 1989-2004. WPT scores predicted neither first-year NFL performance (passing efficiency) nor first-year salary (Mirabile, 2004).
The purpose of this research is to further investigate whether the WPT has the ability to predict NFL performance. Specifically we examine the relationship between the WPT and multiple measures of success at three NFL offensive positions (quarterback, running back, and wide receiver) over a multi-year period, 1999 through 2004. Results of this research should be of interest to NFL and team personnel in determining whether there is statistical justification for requiring combine participants to complete the WPT. In a broader sense this research may also encourage greater interest in the use of psychological construct testing, whether it be the WPT or other types of cognitive ability tests, in not only professional football but also other athletic endeavors (e.g.; baseball, basketball) and at other levels (e.g.; Division I and II college and university athletics).
Selecting the quarterback (QB) position for inclusion in this research was based upon three factors. First, performance data for the QB position were believed to be more objective and readily obtainable as compared to other positions such as guards, tackles, and centers. Second, one may surmise that the QB position requires greater cognitive ability than other positions, as the quarterback must handle the ball on the vast majority of offensive plays and must make play-calling strategic decisions within seconds; for example, by reading the defense and changing plays at the line of scrimmage. Third, the relationship between the WPT and quarterback performance had previously been investigated; however this research will define success in broader terms. We also examined running backs (RB) and wide receivers (WR) – the other primary so-called “skill” positions that involve handling the ball on offense. Further, RBs and WRs met the Wonderlic criterion that “…the closer they are to the ball the smarter they need to be (Wonderlic, Inc., 2005).”
A total of 68 NFL QBs were included in the study. This number represented all combine invitees who were drafted by the NFL from 1999 through 2004. The study also included data for 86 drafted RBs and 152 drafted WRs for the same 6-year period. (Typically the number of players drafted per year was slightly more than half of those participating in the combine). Table 1 shows the number of draftees by position for the years under study.
First published in 1937, the WPT is a 50-item, timed 12-minute test. The test yields a single score that reflects the total number of questions answered correctly. Originally adopted from the Otis Self-Administering Test of Mental Ability (Beck, 1986), the test includes multiple choice and short answer questions designed to measure verbal, numerical, general knowledge, and spatial relationship abilities. Few people complete the test because of time constraints.
The great majority of research concerning the validity and reliability of the WPT has been performed in traditional employment settings. In six employment categories (supervision, blue collar, general office, data processing, engineering, and professionals) the Wonderlic organization reports validities ranging from .22 to .67. Validity studies criteria included a variety of measures, primarily job performance rankings, productivity and production records, and supervisory ratings. The highest validities were in the professional group (e.g.; insurance agents, medical personnel, police officers), with validities ranging from .39 to .67. The lowest validities were in the general office group (e.g.; bank tellers and typists) with validities ranging from .27 to .35. The WPT also correlates highly with the other tests of cognitive ability widely used in employment settings, including the Weschler Adult Intelligence Test (WAIS) (Weschler, 1955), the WAIS-Revised (WAIS-R) (Bell, Matthews, Lassiter, & Leverett, 2002; Dodrill, 1981; Dodrill & Warner, 1988; Edinger, Shipley, Watkins, & Hammett, 1985; Hawkins, Faraone, Seidman, & Tsaung, 1990; Weschler, 1981), and the General Aptitude Test Battery (GATB) (Wonderlic, Inc., 2002). Various forms of reliability estimates for the WPT (test-retest, longitudinal, alternate form, and internal consistency) range from .82 to .95 (Dodrill & Warner, 1988; Furnham & Petrides, 2003; McKelvie, 1989; Wonderlic Inc., 2002).
WPT data for individual combine participants are not made public by either the NFL or Wonderlic, Inc. Wonderlic, Inc. does make available a single mean WPT score for all combine prospects and also mean scores for each position. Nonetheless WPT scores achieved by combine participants are widely available on the Internet. WPT scores for each combine participant included in this study were retrieved from nfldraftscout.com (“Rankings,” n.d.), a website devoted to current and historical information on the NFL, NFL teams, and NFL players. Author discussions with nfldraftscout.com representatives indicated that Wonderlic scores were obtained from NFL team personnel, player scouts, and agents. For each position included in this study, a random sample of players’ reported Wonderlic scores were checked and verified against at least two Internet sources; no disparities were found.
Player performance information was obtained from nfl.com (“Players,” n.d.), and includes regular season games only (post-season play is excluded). This website contains career statistics for all players included in this sample. NFL salary data were collected from usatoday.com. According to the usatoday.com website (“Sports,” n.d.), salary data are obtained from player agents and NFL Players Association research documents. Nfl.com and usatoday.com are owned, respectively, by the National Football League and the Gannett Co. As with the WPT scores, performance data collected for each position were checked and verified against other Internet sources.
Correlation analysis was used to determine if the WPT is linearly related to any one of several possible measures of performance of drafted players. Treating the professional player prospect WPT score as the independent variable, the following possible dependent variables were examined:
Draft order. The players are picked in the NFL draft through several rounds. If the WPT is related to a player’s talent and/or potential, there should be a negative correlation between WPT and draft order, as the “best” players should have a high WPT and a low (selected early) draft order number while the less promising prospects should have a low WPT and a high draft order number.
Salary. Hypothetically players are paid according to their promise and performance on the field, so we anticipate a positive relationship between WPT and salary if indeed the WPT is a performance predictor. We consider the WPT and salaries for each of the first three years in the NFL. (Note however that for players drafted in 2004, there is only one year of salary and other data described below. Similarly for players drafted in 2003, as we have data for their first and second but not their third year.)
On-field performance measures. Quarterback performance was defined by applying the NFL’s quarterback rating system. According to nfl.com, this official NFL statistic (in use since 1973) is based on a formula that takes into account four aspects of passing performance: pass completion rate, average yards gained per attempt, percentage of interceptions, and percentage of passes that result in touchdowns. Again if the WPT predicts job performance, there should be a positive correlation for WPT with this on-field measure. As with salary we consider the first three years of data for QB rating. For RBs the success measure was average yards per carry; for WRs the success measure was average yards per reception.
Games played. Though a less complex measure than the others, it is possible to argue that WPT should be positively related to the number of games a player appears in per season, as a higher performing player should make more game appearances than a lower performing one and vice versa. As is the case for the other variables, we will analyze a player’s first three years in the NFL. (One potential drawback of this measure is that it does not consider actual playing time per game. Conceivably a “game played” could involve simply one play, several plays, or the entire game. However the NFL does not make playing time per game data available, and neither do other websites or professional football statistics sources.)
We will consider the correlations for the 6-year sample period separately for each position (QB, RB, and WR). This first analysis involves only combine participants who were drafted by an NFL team. We will also conduct two-sample t-tests to see if WPT scores differ depending on whether a combine participant is drafted or not drafted. Each position will be treated separately for the comparisons of means (i.e., there will be three t-tests).
Sample Summary – Correlations by Position
Table 2 presents correlations for each WPT-success measure pair and for each player position over the 6-year time period. Sample sizes are 68 QBs, 86 RBs, and 152 WRs. For instance the first correlation shown is -.138, representing the WPT-draft order correlations for all 68 QBs drafted over the 6-year study period; the correlations for RBs are -.028 and for WRs the result is .086. None of these is significant at the .05 level; in fact only 2 of the 30 correlations in Table 2 generate a p-value < .05, an outcome that is consistent with a random chance model. Further one of these two correlations (-.259 for WR-Year 2 Salary) is in the “wrong” direction, assuming Year 2 Salary should have a positive association with WPT scores.
(It is possible to argue that the three draft order correlations in Table 2 should be determined via a nonparametric procedure since draft order could be interpreted as an ordinal number. This was done; yielding values for Spearman’s rho little changed from the parametric correlations reported in Table 2: -.123, -.020, and .084, respectively.)
Sample Summary – Drafted versus Non-drafted Players
A second way to approach the question of how the WPT relates to NFL success is to compare the WPT scores of drafted players against those who were not drafted on a position-by-position basis. See Table 3 for WPT results for QBs, RBs, and WRs over the 6-year period of the study.
For QBs the average WPT for drafted players was 2.15 points higher than that of non-drafted players; however a two-sample t-test shows this difference is not statistically significant (at α = .10). WPT scores for drafted versus non-drafted RBs and WRs were nearly identical. The difference between the two groups for RBs was .18 point (p = .85) and the difference for WRs was .05 point (p = .94). Therefore for these three “skill” positions we fail to find that drafted players have significantly higher WPT scores than players who were not drafted. If assertions made by Wonderlic, Inc. were in fact true and believed to be true by the NFL teams selecting players, then we would expect drafted players to have higher WPT scores than those not drafted, but who play the same position.
Viewing the results as a whole we conclude that for the years, positions, and performance criteria we examined in this study, the WPT is unrelated to a player’s draft order, salary, games played in, and on-field position-specific success measures. Further overall tests of drafted players versus those not chosen in the draft failed to reveal any meaningful differences with respect to scores obtained on the WPT.
Although it may have utility in more general employment situations, the WPT appears to lack validity in this athletic oriented setting; this despite its advocates and some perhaps self-serving statements from Wonderlic Inc. Although (see limitations below) there may be multiple variables interacting and even masking the effects of other variables, we would still expect to find more correlations above and beyond random chance, or a significant difference in drafted versus non-drafted means if the WPT had any merit.
Given the research that has been published on the relationship between various psychological constructs and athletic performance discussed earlier and the obvious monetary risks of false positives and false negatives when selecting professional football players, it is surprising that the NFL has not adopted a more sophisticated approach to the measurement of cognitive ability and other psychological measures for combine participants. While the validity and reliability of the WPT in traditional employment settings has been established, the instrument does not appear to have utility in the professional football arena.
The results of this study raise serious questions about the degree of sophistication employed by the NFL in its player selection process. While the actual selection process used by NFL teams remains confidential and cloaked in a degree of secrecy, it could be assumed that NFL teams could benefit from using contemporary human resource selection models to the choosing of player personnel. While the selection process is obviously constrained by NFL draft rules and requirements, the drafting of a player essentially remains an employee hiring process, one in which an organization attempts to predict how an employee will perform in the future. Contemporary human resource management models for hiring employees stress the importance of job analysis, “fit” between the person and the organization, and the development and validation of selection instruments (Heneman & Judge, 2005). Efforts to predict success are not new to collegiate athletes and other sports contests; for example, Olympic events (Humara, 2000; Spieler, Czech, Joyner & Munksay, 2007), but there is no evidence, at least in the literature, that the NFL undertakes the level of personnel selection research found in non-professional sports. Assuming such research is absent among the professional football ranks, NFL teams could well profit from utilizing employee selection models that are commonplace in traditional business settings and by examining the potential utility of the human performance research that has taken place in non-professional athletic environments.
This study has several limitations. First both WPT predictor and performance data were collected from secondary sources: nfldraftscout.com, usatoday.com, and nfl.com. While the data were collected from such sources, they are nonetheless considered valid sources for professional football statistics. Second, although we examined six draft classes, the question remains as to whether the results are generalizable to other draft classes. A historical analysis of the WPT over a greater time period might add insight into the ability to generalize the predictability of the WPT. Third, we examined performance variables over a three-year period. Conceivably, a player might not ‘blossom’ until after several years of experience; on the other hand, the average NFL career lasts less than four years (Zaslow, 2002).
In addition the performance criterion ‘games played’ might be moderated by the quality of a player’s competition for playing time. That is, an excellent recently drafted QB, RB, or WR may see limited playing time because an experienced, superior player is currently performing for the team; conversely, a lesser quality recently drafted player may see considerable playing time for another team because he is the best available from a weaker roster of current players.
The relationship between the WPT scores and several aspects of skilled player performance was examined for NFL combine participants who were drafted in the period 1999-2004. In general the findings suggest that WPT scores are not related to draft order, salary, games played, or QB/RB/WR rating. Further the WPT scores do not appear to be related to whether or not a player is drafted. Thus “smarter” as measured by the WPT does not seem to translate into “better.”
Future research should continue to focus on the WPT as a predictor of player performance by examining additional draft classes. Further the predictive merits of the WPT should also be examined for other NFL positions, both offensive and defensive. Beyond the WPT, the NFL may improve the overall validity of the combine process by including a test battery comprised of additional psychological constructs, for example, those that have been shown to correlate with peak athletic performance. In addition the other combine predictors in current use (running, jumping, and strength tests) should be analyzed to determine their relationships with success measures.
Beck, M. D. (1986). The Otis and Otisennon tests: Their contributions. Educational Measurement: Issues and Practice, 5, 12-18.
Beilock, S. L., & McConnell, A. R. (2004). Stereotype threat and sport: Can athletic performance be threatened? Journal of Sport & Exercise Psychology, 26, 597-609.
Bell, N. L., Matthews, T. D., Lassiter, K. S., & Leverett, J. P. (2002). Validity of the Wonderlic Personnel Test as a measure of assessment. North American Journal of Psychology, 4, 113-121.
Brody, N. (1992). Intelligence. New York: Academic Press.
Campbell, J. P. (1990). Modeling the performance prediction problem in industrial and organizational psychology. In M. D. Dunnette, and L. M. Hough (Eds.), Handbook of industrial and organizational psychology (2nd ed.), Vol. 1 (pp. 688-732). Palo Alto, CA: Consulting Psychologists Press.
Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. Cambridge: Cambridge University Press.
Dodrill, C. B. (1981). An economic model for the evaluation of general intelligence in adults. Journal of Consulting and Clinical Psychology, 49, 668-673.
Dodrill, C. B., & Warner, M. H. (1988). Further studies of the Wonderlic Personnel Test as a brief measure of intelligence. Journal of Consulting and Clinical Psychology, 56, 145- 147.
Dougherty, P. (2006). Will Wonderlic cause teams to wonder about Young? Retrieved March 1, 2006 from http://www.usatoday.com/sports/football/draft/2006-03-01- young-wonderlic_x.htm
Edinger, J. D., Shipley, R. H., Watkins, C. E., & Hammett, E. B. (1985). Validity of the Wonderlic Personnel Test as a brief IQ measure in psychiatric patients. Journal of Consulting and Clinical Psychology, 53, 937-939.
Furnham, A, & Petrides, K. V. (2003). Trait emotional intelligence and happiness. Social Behavior and Personality, 31, 815-824.
Furnham, A., Rawles, R., & Iqbal, S. (2006). Personality, intelligence, and proofreading. Personality and Individual Differences, 41, 1457-1467.
Gottfredson. L. (1997). Why g matters: The complexity of everyday life. Intelligence, 24, 79-132.
Hawkins, K. A., Faraone, S. V., People, J. R., Seidman, L. J., & Tsuang, M. T. (1990). WAIS-R validation of the Wonderlic Personnel Test as a brief measure in a psychiatric sample. Psychological Assessment, 2, 198-201.
Hays, K. F., & Kenkel, M. B. (2006). Peak performance in sport: Identifying ideal performance states and developing athletes’ psychological skills. Professional Psychology: Research and Practice, 37, 233-243.
Heneman, H. G. & Judge, T. A. (2005). Staffing Organizations. New York: McGraw Hill.
Humara, M. (2000). The relationship between anxiety and performance: A cognitive- behavioral perspective. Athletic Insight, 1. Retrieved February 7, 2007 from http://athleticinsight.com/Vol1Iss2/Cognitive_Behavioral_Anxiety.htm
Humara, M. (2002). Personnel selection in athletic programs. Athletic Insight, 2. Retrieved February 1, 2007, from http://athleticinsight.com/Vol2Iss2/Personnel.htm
Hunter, J. E. (1986). Cognitive ability, cognitive aptitudes, job knowledge, and job performance. Journal of Vocational Behavior, 29, 340-362.
Hunter, J. E. & Hunter, R. F. (1984). Validity and utility of alternative predictors of job performance. Psychological Bulletin, 96, 72-98.
Jackson, S. A. & Roberts, G. C. (1992). Positive performance states of athletes: Toward a conceptual understanding of peak performance. The Sport Psychologist, 6, 156-171.
Jensen, A. R. (1986). G: Artifact or Reality? Journal of Vocational Behavior, 29, 301-331.
Krane, V., & Williams, J. M. (2006). Psychological characteristics of peak performance. In J. M. Williams (Ed.), Applied sport psychology: Personal growth to peak performance, 207-227. New York: McGraw-Hill.
LePine, J. A., Colquitt, J. A., & Erez, A. (2000). Adaptability to changing work task contexts: Effects of general cognitive ability, conscientiousness, and openness to experience. Personnel Psychology, 53, 563-593.
Lyons, B. D., Michel, J. W., & Hoffman, B. J. (2005). A preliminary investigation between the Wonderlic and NFL performance. Proceedings of the 20th annual meeting of the Society for Industrial and Organizational Psychology, April 2005, Los Angeles, CA.
McKelvie, S. J. (1989). The Wonderlic Personnel Test: Reliability and validity in an academic setting. Psychological Reports, 65, 161-162.
Merron, J. (2002). Taking your Wonderlics. Retrieved February 3, 2006 from http://www.espn.go.com/page2/s/closer/020228.html.
Mirabile, M. P. (2004). Intelligence and football: Testing for differentials in collegiate quarterback passing performance and NFL compensation. The Sport Journal. Retrieved January 9, 2006, from http://www.thesportjournal.org/2005Journal/Vol8- No2/mac-mirabile.asp.
Mulligan, M. (2004). Wonderlic scores have NFL teams wondering. The Chicago Sun-Times, April 22, p. S4. Players. (n.d.). Retrieved June 25, 2005 from http://www.nfl.com
Privette, G. (2004). The phenomenon of peak performance in sports. International Journal of Sport Psychology, 12, 51-60. Rankings. (n.d.). Retrieved June 11, 2005 from http://www.nfldraftscout.com
Ree, M. J. & Earles, J. A. (1992). Predicting training success: Not much more than g. Personnel Psychology, 44, 321-332.
Ree, M. J., Earles, J. A., & Teachout, M. S. (1994). Predicting job performance: Not much more than g. Journal of Applied Psychology, 79, 518-524.
Roberts, S. (2006). But can Wonderlic scramble or pass? The New York Times, March 26, p. 8.
Schmidt, F. L. (2002). The role of general cognitive ability and job performance: Why there cannot be a debate. Human Performance, 15, 187-210.
Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262-274.
Spieler, M., Czech, D. R., Joyner, A. R., & Munkasy, B. (2007). Predicting athletic success: Factors contributing to the success of NCAA Division I-AA collegiate football players. Athletic Insight, 9. Retrieved August 9, 2007 from www.athleticinsight.com/Vol9Iss2/PredictingSuccess.htm
Sports. (n.d.). Retrieved May 12, 2005 from http://www.usatoday.com
Thorndike, R. L. (1986). The role of general ability in prediction. Journal of Vocational Behavior, 29, 332-339.
Weschler, D. (1955). Weschler Adult Intelligence Scale. New York: The Psychological Corporation.
Weschler, D. (1981). Weschler Adult Intelligence Score-Revised. New York: The Psychological Corporation.
Wonderlic, Inc. (2002). Wonderlic personnel test & scholastic level exam user’s manual. Libertyville, IL: Wonderlic, Inc.
Wonderlic, Inc. (2005). NFL testing provides valuable lesson for all employees. Retrieved November 15, 2005 from http://www.wonderlic.com/Promotion/NFL_press_release/asp.
Wonderlic, Inc. (2006). Wonderlic products. Retrieved January 22, 2006, from http://www.wonderlic.com/products/product.asp?prod_id=4.
Zaslow, J. (2002). Athletes must tackle big career transitions. Retrieved June 10, 2006 from http://www.careerjournal.com/columnists/movingon/20020628-movingon.html.
Correspondence concerning this article should be sent to Dr. Arthur J. Adams, College of Business, University of Louisville, Louisville, KY 40292, USA, firstname.lastname@example.org, Phone 502 852 4778, Fax 502 852 7557