You're Perfect for the Job:

The Use and Validity of Pre-employment Personality Tests

Cari Adams

Abstract

In an effort to examine the validity of pre-employment personality tests, two identical personality questionnaires based on the Big Five Personality Model were administered to each participant, one that tested personality profiles and the other that tested job applicant personality. A situational judgment test (SJT) was administered along with the personality tests to measure actual behaviors in the workplace as indicated by the participants. A sample of 112 students from a liberal arts university located in the midwestern United States participated in the study. Results showed that students were likely to vary their scores between the two tests according to their perception of the ideal employee (p<0.001). The study also examined differences in gender and the order in which the tests were administered. In addition, correlations between scores for specific variables on both of the personality tests and the same variables on the SJT were observed. Because many companies employ the use of personality tests in deciding which employees to hire, it is crucial that these tests are valid and reliable in measuring certain characteristics and predicting workplace behaviors.

     For years personality tests, such as questionnaires, inkblots, color boards, and drawings, have been used to discover individuals' characteristics and likelihood to behave in specific ways in a variety of instances. Although certain personality tests seem to be more scientifically based than others, these types of assessments are taken by tens of millions of people each year (Paul, 2004). Also, according to Paul (2004), personality testing has become a $400-million industry and is growing at an eight to ten percent rate.  In numerous cases, these tests have simply been used by therapists, psychologists, and psychiatrists to assess and analyze their clients in order to determine characteristics and behaviors that may identify certain disorders or mental illnesses. In other occasions, they are utilized by American courts to determine custody battles or one's capability to commit a certain crime. In other instances yet, these types of tests have been used with children to diagnose learning and behavioral problems that might exist (Paul, 2004).

     It is also widely known that personality tests have been and are currently used by numerous companies for the purposes of hiring and promoting their employees (Bradley & Hebert, 1997). Sizeable companies known to use personality tests for these specific reasons include Walmart, Hewlett Packard, Target, and one of the most shocking, the National Football League (NFL) (Cullen, 2006; NFL, 2001; Camara & Merenda, 2000). Other companies, such as General Motors Corporation, have used personality tests to identify the personality types of their employees in order to aid co-workers and managers in their understanding and ability to work with other employees (Bradley & Hebert, 1997; Shuit, 2003).

     These tests are substantially controversial, given that, in particular cases, they contain highly personal and seemingly unreasonable questions, such as 'Are you a dog or a cat?' which is listed on the test that NFL prospects are administered during scouting (NFL, 2001). Furthermore, employees may feel that the answers to such questions are not privy to their employers' knowledge (Paul, 2004). An incident exemplifying this controversy involved a court case in which applicants of Target stores sued the company for denying them employment based on the results of their personality tests. The applicants won when the court declared that the questions on the test were irrelevant to the requirements of the jobs for which they were applying (Camara & Merenda, 2000).

     Another factor behind the controversy surrounding personality tests, is that, specifically in the area of employment, the use of these tests has been questioned as to how well they can predict certain behaviors in the workplace and how valid the results attained are. Not only has the discussion of pre-employment personality tests been focused on the validity of the tests' measurements of certain variables, but research has also addressed issues pertaining to the ability of applicants to fake their scores on such tests by elevating their scores on seemingly desirable characteristics and depressing their scores on seemingly undesirable characteristics. Furthermore, myriad studies have been focused on the effect of gender differences on the tests, specifically in relation to the notion that these assessments may contain biases affecting the scores based on gender characteristics. Complete validity of these tests, especially in instances of their use for hiring and promoting, is of paramount importance due to the fact that, in many cases, decisions are made based largely upon the indicative scores individuals receive as a result of their answers to the assessments.

     The following study will be focused on the validity of pre-employment personality tests, specifically pertaining to the ability of participants to fake their scores on personality tests in order to obtain a job. Comparisons related to gender will also be reviewed as related to the concept of characteristics that are more common to men or women. Important is to note, however, that this study will remain within the confines of the university environment, and, therefore, results and implications of this research may only be applicable to similar populations.

History of Personality Tests

     Attempts to discover personality traits and characteristics began in the nineteenth-century with proponents of phrenology, which is the science of the mind. According to this discipline, the bumps on a person's head could indicate a person's defining traits. As research in science and technology was furthered, so was the decline of the popularity of phrenology. At a similar time, Hermann Rorschach was working on his idea of a personality test, which consisted of a series of inkblots, some portraying dark shapes and others showing colorful images. According to Rorschach, the images that individuals saw within the inkblots indicated characteristics about their personalities. Although much research has disproved the theory behind the Rorschach inkblots, they continue to be used presently with mentally ill patients, professional athletes, and even on some Roman Catholic priests (Paul, 2004).

     The first actual personality inventory is known to have been in response to the need to identify mentally stable soldiers before sending them into battle during World War I to prevent shell shock. This test was called the Psychoneurotic Inventory and proved to be a template for the numerous sorts of personality inventories devised in the following years. One such test was the Minnesota Multiphasic Personality Inventiory (MMPI) created in the 1930's at the University of Minnesota to sort mentally ill patients into diagnostic categories. The MMPI was revised in 1989, resulting in the MMPI-2, which is used currently for a variety of measures, including mentally ill patients and employees (Paul, 2004; Butcher, Hooley, & Mineka, 2005). Although uses of personality tests have varied to a great degree throughout the past, one of the most widespread uses for the tests presently is for hiring and promoting employees (Paul, 2004).

Types and Validity of Pre-employment Personality Tests

     One common type of personality assessment in the workplace is the Myers-Briggs Type Inventory (MBTI). Based on the theories of psychologist Carl Jung, the MBTI has been available since 1943 (Shuit, 2003). Questions on the MBTI measure peoples' personalities based on four different sets of opposites, which include: introvert/extravert, sensing/intuiting, thinking/feeling, and judging/perceiving (Shuit, 2003). According to Jungian theory, extraversion is defined by an outward orientation, with thoughts and feelings being focused on objective conditions, while introversion is characterized by an inward focus, with thoughts and feelings being focused on inward experiences (Girelli & Stake, 1993).  Jungian theory further posits that sensing and intuiting are both ways that individuals may tend to perceive their environment, and, to this end, the MBTI is largely focused in determining the predominant method in which each individual assimilates the world (Girelli & Stake, 1993). Essentially, those who are sensory types tend to focus on specific details, while personalities that are predominantly intuitive are more inclined to see the big picture (Cloninger, 2008). Finally, according to Jungian theory, the thinking and feeling variables differ in such ways that thinking types primarily base decisions upon logic, while feeling types tend to base their judgments on an effort to increase their sensation of positive emotions (Cloninger, 2008; Girelli & Stake, 1993).

     Although judging/perceiving was not a part of Jung's original theory, it was added on the MBTI to measure individuals' preferences in interacting with the outside world. Individuals who identify with judging characteristics are likely to thrive in a planned, systematic, and decided lifestyle, while those who tend to have more perceiving traits might prefer a more flexible and spontaneous way of life (Bradley & Hebert, 1997). By measuring the levels of each of the eight opposing variables, the MBTI focuses on discovering dominant personality traits and strengths, and may distinguish specific weaknesses.

     Disagreements have arisen from the use of this specific assessment for hiring and promoting due to the question of its validity (Shuit, 2003). One major factor in the argument of the MBTI's validity is its forced choice format (Girelli & Stake, 1993). In order for individuals to be defined as one characteristic instead of another, it is essential that the characteristics be entirely bipolar, meaning that a person cannot be one while simultaneously being any part of the other (Girelli & Stake, 1993). A study performed by Cowen (1989) to measure the bipolarity of such variables administered a test to participants measuring for the same characteristics as the MBTI. Unlike the MBTI, however, this test utilized a Likert scale for each separate trait, instead of the MBTI's forced choice format. Cowen (1989) found that when the participants were able to identify themselves as a certain amount of one and a certain amount of the other, in some cases, individuals took on characteristics of what Jung's theory states must be complete opposites. To be more specific, several of the participants scored high on scales measuring Sensing, but these same individuals also scored high on measures of Intuition (Cowen, 1989).

     A more recent study using Likert scales to measure MBTI dimensions performed by Girelli and Stake (1993) had similar findings. The results of the study showed that 26.3% of the participants characterized themselves as having both traits on at least one of the MBTI dimensions that were tested. Furthermore, the researchers found that the Thinking and Feeling scale scores were significantly correlated in the positive direction (Girelli & Stake, 1993). Such findings tend to indicate that although the MBTI is used quite often, its results may not be valid or reliable.

     Other tests commonly used in the work environment have typically been based on McCrae and Costa's theory of the Big 5 Personality Model (Kubinger & Litzenberger, 2006; Thumin & Barclay, 1993). According to Kubinger and Litzenberger (2006), the Big 5 were developed from a series of re-analyses performed on eight different samples which continuously resulted in the existence of five main personality factors. Unlike the MBTI, the personality factors are set in a continuum form with individuals receiving either a higher or lower score for each feature. The variables measured according to the Big 5 Personality Model are as follows: extraversion, agreeableness, neuroticism, openness, and conscientiousness (Truxillo, Bauer, Campion, & Paronto, 2006; Cloninger, 2008).

     The measure of extraversion according to the Big 5 Personality Model is based on the individual's sociability. Individuals scoring high in extraversion also rate themselves highly in optimism, self-assuredness, and cheerfulness (Truxillo et al., 2006; Kubinger & Litzenberger, 2006). The Big 5 measures agreeableness according to individuals' friendliness and avoidance of hostility toward others. Those who score lower on measures of agreeableness are more likely to be distrustful, unsympathetic, and uncooperative (Cloninger, 2008). According to the Big 5 Personality Model, neuroticism is measured against emotional stability, meaning that individuals who score highly in this variable are more likely to be bothered by negative emotions. Furthermore, participants who receive high scores in neuroticism tend to be easily thrown out of balance emotionally; however, this sense of neuroticism does not relate to excessive cases, such as those that are clinically diagnosed (Cloninger, 2008; Truxillo et al., 2006). Openness, on the other hand, is related to levels of interest and excitement about participating in new experiences (Truxillo at al., 2006). Finally, according to the model, conscientiousness is a measure of people's self-discipline and control. Individuals who score high in this trait are typically concerned with planning and organizing (Truxillo et al., 2006).   

     Numerous tests and studies have been performed to test the validity and reliability of the Big 5 Personality Model, specifically with the notion that it is commonly used by myriad companies to test potential employees� job performance and attitudes (Truxillo et al., 2006; Kubinger & Litzenberger, 2006; Thumin & Barclay, 1993). According to Kubinger and Litzenberger (2006), one specific study performed analyses on the self-ratings of 256 participants and found that the data could be narrowed down into five main variables, all of which showed striking similarities to those variables that construct the Big 5 Personality Model.

      Another type of personality test that has been increasingly used to measure potential employees' personality types is a situational judgment test (SJT) (Biderman, McDaniel, & Nguyen, 2005). On SJTs, applicants are asked to respond to a series of work-related scenarios based on how they would react in each specific situation. The two types of SJTs are knowledge format and behavioral tendency format. In the knowledge format test, test-takers are asked to choose what the best/worst response would be in a given scenario, while in the behavioral tendency format, test-takers are asked to choose which response they would be most likely/least likely to have in a given scenario (Biderman, et al., 2005).

Applicants' Likelihood to Lie on Personality Tests  

     Although it is apparent that pre-employment personality tests, such as those previously described, have been studied by myriad researchers for their internal validity according to their abilities to measure specific variables, other factors affecting the validity of pre-employment personality tests have also been taken into consideration. To this end, the discussion will now turn to a focus on the ability of applicants to 'fake good' or lie about positive characteristics on these tests for the purposes of obtaining a job.

     In a study by Thumin and Barclay (1993), two different groups of participants were administered identical personality tests with one group's instructions being to try to discover their personality type, while it was the other group's notion that the scores would be used in hiring them for a job. To ensure validity, a large sample size was utilized, controlling for variables such as age, gender, and education. However, due to limitations in the ability to simulate an actual job applicant personality test, individuals who were truly applying for jobs comprised the entirety of one group, while the other group consisted solely of students enrolled in evening classes at a midwestern university. Important to note is that 94% of the students were working full-time and were, therefore, largely accommodated with the work industry, while 54% of the job applicants were part-time students. Consequently, it appears that the two groups may have been matched reasonable well, in that they all had similar lifestyles focused on higher education and successful careers (Thumin & Barclay, 1993).

     Findings of the study corresponded with the researchers' hypotheses that applicants for white collar/professional positions would score significantly higher on scales indicating confidence, ambition, sociability, organization, logic, consideration, and honesty than students attempting to discover their personality types (Thumin & Barclay, 1993). Furthermore, Thumin & Barclay (1993) found that the applicants' scores were significantly lower on scales measuring depression, irritability, and reservation than the scores of the students. Although there were significant differences in the levels of desired and undesirable characteristics, the overall personality profiles of both the applicants and the students were surprisingly similar. Both groups tended to score themselves high on desirable characteristics (although applicants scored themselves excessively high), while portraying themselves as having lower levels of undesirable qualities (although applicants scores were exceptionally low). To this end, it could be considered that both groups of stable, educated participants were able to assess the personality test in similar ways, no matter the situation (Thumin & Barclay, 1993). However, it is also of paramount importance to discuss the results that showed the job applicant group as having smaller standard deviations on seven of the 11 scales than the students. Such findings strongly indicate that individuals are more willing to identify themselves with undesirable characteristics when nothing is at consequence (Thumin & Barclay, 1993).

     In a study examining applicants' abilities to lie about how they would potentially act in given situations on SJT's, Biderman et al. (2005) instructed half of the participants to first answer honestly and then take it a second time 'faking good', while they told the other half to first 'fake good' and then answer honestly. They also administered both types of SJTs to the participants. By comparing the results of the honest answers and fake answers, Biderman et al. (2005) found that the participants were able to 'fake good' answers when the behavioral tendency format was used, no matter which order the participants were told to 'fake good' or be honest. However, on the knowledge response format, the ability to fake seemed to depend on the order in which the participants were asked to be honest/fake. The participants were only able to 'fake good' by elevating their scores on desirable characteristics and depressing their scores on undesirable characteristics when they were administered the SJT they were told to fake on before the SJT on which they were to be honest. This example might demonstrate that SJTs with a knowledge response format may be slightly more valid than SJTs with a behavioral tendency format (Biderman et al., 2005).    

     Numerous attempts have also been made to discover what types of characteristics applicants are most likely to lie about on personality tests in order to get hired. In an attempt to discover what traits are commonly faked on personality inventories, Kubinger (2002) tested 60 participants, of which were divided into two separate groups. The first group was provided with instructions to respond with the answer that best describes them, while the second group was told to 'try to get a good score.' Kubinger (2002) found a significant difference in 6 of the 8 characteristics tested between the two groups. Characteristics that were faked to get a good score included extraversion, dominance, nurturance, self-monitoring, achievement, and frustration resistance (Kubinger, 2002).

     Further studies have been performed to examine if applicants are likely to lie about certain characteristics based on the specific jobs for which they are applying. Dalen, Roberts, & Stanton (2001) demonstrated that participants were able to fake their answers, when they were given the job's description, the job's person specification, and even when they were only given the job title. However, Dalen et al. (2001) also found that there was a common faking strategy among the participants; the participants seemed to elevate their scores to unreasonable degrees on such characteristics as persuasive, controlling, social confidence, innovative, forward planning, detail conscious, conscientious, relaxed, tough minded, and optimistic. The stereotype created by the participants' fake answers also included moderately elevated scores on characteristics such as outgoing, practical, artistic, achieving, and decisive. Characteristics that tended to be depressed included worrying and emotional control. While it is important to note that the participants did have the ability to fake on the test, it is also noteworthy that the participants set up a stereotype for what they believed the perfect candidate would be characterized by. Surprisingly, the stereotype created by the participants failed to match the ideal profile for the job (Dalen et al., 2001). Even when the participants were provided highly sophisticated information about the job's requirements, they continued to be unable to match the ideal profile for the job (Dalen et al., 2001).

Gender Differences on Pre-employment Personality Tests

          The discussion related to personality tests in the work environment is occasionally focused on evaluating gender differences due to the questionable presence of predictive biases, which could be related to the specific success of women or men on such tests. Such is known to have existed in the past due to the fact that numerous companies practiced the use of scoring tests based on gender-specific norms until the enforcement of the Civil Rights Act in 1991, which prohibited any type of score adjustment on such tests (Saad & Sackett, 2002).

     According to Saad and Sackett (2002), in past studies, women have consistently scored higher than men on dependability measures, while men have typically scored higher than women in degrees of dominance and influence. In another study by Thumin and Barclay (1993), the answers of students and job applicants were compared on the basis of investigating differences due to 'faking good' by the job applicants. In addition to a high level of distinction between numerous characteristics between the two groups, differences in gender were also found to be significant. Men in both groups tended to score higher on scales measuring aggression, determination, power, and self-drive. Women, on the other hand, produced exceptionally higher scores than men on scales examining the presence of consideration, thoughtfulness, warmness, and sympathy. Furthermore, men outscored women in such traits as logic, rationality, and decisiveness, while women consistently scored higher in characteristics such as shyness and quietness (Thumin & Barclay, 1993).

Summary of Literature and Hypotheses

     Numerous conclusions can be deduced after reviewing the research pertaining to the use and validity of pre-employment personality tests. While many different types of personality tests are currently being used in the workplace, past studies have shown that in many cases, individuals are able to exaggerate their scores to fit their image of the ideal employee, even if their image of this employee is not an actual fit for the job. Furthermore, research has shown that when participants are administered two personality tests, one with the attempt of discovering their personalities and the other as a pre-employment test, the order in which the tests are taken may have an influence on the individuals' answers. Past studies have also indicated large discrepancies in certain characteristics on pre-employment personality tests due to gender differences. It appears that men have tended to portray more dominance and leadership characteristics than women, while women have tended to score higher on scales measuring care and concern for others than men.   

     In the present study, participants' ability and likelihood to 'fake good' on pre-employment personality tests will be examined. In the study, two identical personality tests will be given to participants, one with the instructions to attempt to find their true personalities, and the other with the directions to fill it out as if it is a pre-employment personality test. Due to the overall findings in the reviewed research of high levels of validity in the Big 5 Personality Model, the personality tests will present questions focused on measuring levels of extraversion, agreeableness, neuroticism, openness, and conscientiousness.  Furthering the initial study by Biderman et al. (2005) on how the order in which these two tests are administered might affect the scores, certain participants will be asked to take the test solely to discover their personalities first, while other participants will be instructed to take the pre-employment personality test initially. Moreover, to re-examine the honesty with which participants will answer the pre-employment personality test, a behavioral SJT consisting of specific workplace situations in which extraversion, agreeableness, neuroticism, openness, and conscientiousness are measured will be administered to participants after the personality tests. Important to note is that participants will be asked to answer all questions on the SJT completely honestly.

     Based on the findings of previous research studies, the following hypotheses were proposed: (H1) participants' scores on the pre-employment test will be higher on scales measuring extraversion, agreeableness, openness, and conscientiousness and lower on scales measuring neuroticism as compared to the personality profile assessment; (H2) participants who take the pre-employment test first will score themselves higher on extraversion, agreeableness, openness, and conscientiousness and lower on neuroticism than if they take the personality inventory first; (H3) men are more likely to have higher extraversion, agreeableness, openness and conscientiousness on the pre-employment personality test than women; and (H4) participants' scores to all five variables on the assessment for personality only will have a higher correlation with measurements of the same variables on the SJT than scores on the pre-employment personality tests.

     To rationalize the theory behind H1, according to the previous research, it appears that participants frequently 'fake good' by elevating desirable scores and depressing undesirable scores according to their idea of the ideal job candidate. Other studies have shown that characteristics typically portrayed by participants' idea of the ideal candidate include traits similar to those of the Big 5 Personality Model (excluding neuroticism, which is predicted to have lower scores). The relationship between variables in H2 is predicted based on previous findings that participants are more likely to be honest when they take a personality test solely to discover their personality profiles before taking the pre-employment personality test. Furthermore, it seems as though there is a high likelihood for participants to be honest on the pre-employment test when they have just recently answered the exact same questions truthfully. Because the participants will almost certainly be able to remember the answers from the first personality test, they will be completely aware that they are lying on the pre-employment personality test when they take it second. On the other hand, participants who take the pre-employment personality test first will have no honest answers to compare to, allowing them to feel less like they are blatantly lying.

     H3 is predicted on the basis that men are more likely to 'fake good' than women due to societal pressures to succeed in their careers. To be more specific, it seems that the concept of men being the main supporter of their families has been institutionalized, even though more recently, women have become major influences in the American workforce. To this end, it appears as though men might be more likely to exaggerate their answers in order to obtain a job than women may be. Finally, H4, which is focused on comparing the scores of variables tested by the personality test and pre-employment personality test with scores of the same variables tested by the SJT, is predicted on the basis that participants' faked scores from the pre-employment personality test will not correlate with the behaviors that they indicate on the SJT. To be more specific, it is predicted that individuals who indicate high scores of positive traits on the pre-employment personality tests will not predict behaving in ways that exemplify high levels of these same positive traits on the SJT.

 Method

Participants

     A total of 112 students were selected from a small liberal arts university located in the midwestern region of the United States. Of the participants, 57 were men and 54 were women, with one participant failing to indicate gender. In addition, 72% of the participants were between the ages of 18 and 21, 22% were between 22 and 25, 4% were between 26 and 30, and 1% were older than 31. Of the sample, 44% of students were business majors, 19% were social science majors, 11% were education majors, 4% were science majors, and 17% indicated that their majors were other than those listed. One student failed to indicate an age group, and seven students failed to indicate their majors. The main ethnicity comprising students at the university is Caucasian; however, there exists a relatively sizeable degree of African American students. Other ethnicities comprise a much smaller portion of the population.

Materials

     A survey based on McCrae and Costa's theory of the Big 5 Personality Model (Cloninger, 2008) was designed to measure participants' levels of extraversion, agreeableness, neuroticism, openness, and conscientiousness. On one personality test, directions indicated that participants should base their answers on an effort to discover their personality profiles, see Appendix A. On an identical copy of the aforementioned personality test, instructions indicated that participants should answer questions as they would if the test were a pre-employment personality assessment for a job, see Appendix B. The order of the two tests varied, as the personality profile was first for half of the participants and the pre-employment personality test was first for the other half.

     The final portion of the survey consisted of a work environment related situational judgment test (SJT), see Appendix C, also based on McCrae and Costa's Big 5 Personality Model (Cloninger, 2008). Eleven questions were asked measuring the individual's likelihood to experience certain feelings in a real life work environment related situation that described the individual as being assigned an important project to work on with ten other co-workers. Two questions were asked based on the individuals' likelihood to feel certain ways indicating levels of extraversion, agreeableness, neuroticism, and conscientiousness, and three questions were asked testing the participants' probability to feel certain ways as related to their levels of openness.

Procedure

     Before its administration, the survey was field tested for errors and validity. Due to the characteristics of the sample, the variable of age was unable to be controlled for; however, gender and academic majors were able to be controlled for. Furthermore, participants were selected using a convenience sample, and each student participated in the survey only one time. Important to note is that of the sample, 46.4% of participants were administered the personality test solely to discover their personality profiles first, while 53.6% of individuals were given the test as if it were a pre-employment personality test first. Although instructions were written on each part of the survey, verbal instructions on how to complete each part of the survey were also given to all participants. The survey took students approximately 10 minutes to finish, and any questions students had about the purpose of the research were answered after its completion.

     A specific score for each personality test was given to all participants based on their self reported levels of extraversion, agreeableness, neuroticism, openness, and conscientiousness. Comparisons were then made using paired samples t-tests to examine differences in the scores of the two tests. Scores were also compared using t-tests between gender and order in which the tests were taken. Further analysis of the data collected included correlations examining the relationship between scores on the two personality tests and answers to the SJT.

Results

Hypothesis 1    

     In order to determine differences in scores relating to the personality profile assessment as compared to the pre-employment personality test, paired samples t-tests were run. The analysis showed that extraversion scores (M = 16.54, SD = 2.28) on the pre-employment personality test were significantly higher than extraversion scores (M = 15.48, SD = 2.62) on the personality profile test, with t (107) = -4.66, p < 0.001. Similarly, agreeableness scores (M = 19.91, SD = 2.38) were significantly higher on the pre-employment personality test than on the personality profile measure (M = 18.37, 2.46), with t (110) = -6.46, p < 0.001. Neuroticism scores (M = 11.15, SD = 3.07) tended to be significantly lower on the pre-employment personality test than on the personality profile test (M = 13.15, SD = 3.19), t (109) = 6.77, p < .001. Furthermore, analyses on scales of openness found that scores on the pre-employment test (M = 11.99, SD = 2.23) were significantly less than those on the personality profile (M = 12.94, SD = 2.57), with t (108) = 4.55, p < 0.001. Levels of conscientiousness, on the other hand, were significantly higher on the pre-employment test (M = 21.50, SD = 2.58) than on the personality profile (M = 19.80, SD = 2.75), with t (110) = -6.94, p < 0.001. In summary, the results found that participants rated themselves as having higher levels of extraversion, agreeableness, and conscientiousness and lower levels of neuroticism and openness on the pre-employment test as compared to scores on the personality profile assessment.

 Table 1

Statistics for Extraversion, Agreeableness, Neuroticism, Openness, and Conscientiousness Scores on Personality Profile and Pre-employment Personality Assessments

Table 2

Paired Samples t-test for Extraversion, Agreeableness, Neuroticism, Openness, and Conscientiousness scores between the Personality Profile Test and the Pre-employment Personality Test

Figure 1. Line graph of means for Extraversion, Agreeableness, Neuroticism, Openness, and Conscientiousness on the Personality Profile Test and the Pre-employment Personality Test

Hypothesis 2    

     To assess differences in levels of extraversion, agreeableness, neuroticism, openness, and conscientiousness on the pre-employment personality tests in terms of the order in which the personality profile test and the pre-employment test were taken, an independent samples t-test was used. Analyses showed that extraversion levels on the pre-employment personality test were significantly higher when the first test taken was the personality profile test (M = 17.12, SD = 1.81) as compared to levels of extraversion when the first test taken was the pre-employment personality test (M =16.12, SD = 2.55), with t (105.14) = 2.41, p = 0.02. Similarly, when the participants took the personality profile test first, their scores for agreeableness on the pre-employment test were significantly higher (M = 20.71, SD = 2.17) than they were for participants who took the pre-employment test first (M = 19.23, SD = 2.36), t (109) = 3.40, p = 0.001. Scores for neuroticism on the pre-employment test decreased significantly when the personality assessment was taken first (M = 9.88, SD = 2.78), as compared to when the pre-employment test was taken first (M = 12.20, SD = 2.89), t(109) = -4.28, p < 0.001. Means of openness and conscientiousness did not differ on the pre-employment tests at significant levels based on the order in which the two tests were administered.

Table 3

Statistics for Extraversion, Agreeableness, Neuroticism, Openness, and Conscientiousness Scores on the Pre-employment Personality Test Based on the First Test Taken

 Table 4

Independent Samples t-test for levels of Extraversion, Agreeableness, Neuroticism, Openness, and Conscientiousness on the Pre-employment Personality Test based on which Test was taken First

Hypothesis 3    

     In order to assess the differences in levels of extraversion, agreeableness, neuroticism, openness, and conscientiousness on the pre-employment personality test between men and women, an independent samples t-test was used. Analyses showed no significant differences between the two genders on levels of extraversion, agreeableness, neuroticism, openness, and conscientiousness on the pre-employment personality test.

            Table 5

            Statistics for Extraversion, Agreeableness, Neuroticism, Openness, and Conscientiousness Scores on the Pre-employment Personality Test Based on Gender

Table 6

Independent Samples t-test for levels of Extraversion, Agreeableness, Neuroticism, Openness, and Conscientiousness on the Pre-employment Personality Test based on Gender

 Hypothesis 4    

     To assess the relationship between variables on the personality test and the same variables on the SJT, correlations were used. For extraversion, the SJT scores had a positive correlation with both the personality profile, with r (107) = 0.47, p < 0.001, and the pre-employment test scores at significant levels, r (105) = 0.32, p = 0.001. However, for agreeableness, only the scores on the pre-employment test had a significant correlation with the scores on the SJT, r (107) = 0.19, p = 0.05). While the personality assessment scores for agreeableness did have a positive correlation with the SJT, it was relatively weak and not at a significant level, r (108) = 0.18, p = 0.06. A moderate correlation was found between levels of neuroticism on the personality profile test and the SJT, r (107) = 0.24, p = 0.01, whereas no significant correlation was found for this variable on the pre-employment personality test and the SJT, r (107) = 0.18, p = 0.06). Similarly, a correlation between levels of openness on the personality profile assessment and the SJT was found at a significant level, r (107) = 0.24, p = 0.01, but a significant correlation did not exist for openness on the pre-employment personality test and the SJT, r (106) = 0.13, p = 0.19. Significant correlations did not exist between conscientiousness levels on the personality profile test and the SJT, r (108) = 0.07, p = 0.47, nor were they present between conscientiousness scores on the pre-employment test and the SJT, r (107) = 0.03, p = 0.73.

 Table 7

            Correlations Between Extraversion Levels on the Personality Profile and Pre-employment Tests and Levels on the SJT

Table 8

Correlations Between Agreeableness Levels on the Personality Profile and Pre-employment Tests and Levels on the SJT

Table 9

Correlations Between Neuroticism Levels on the Personality Profile and Pre-employment Tests and Levels on the SJT

Table 10

Correlations Between Openness Levels on the Personality Profile and Pre-employment Tests and Levels on the SJT

Table 11

Correlations Between Conscientiousness Levels on the Personality Profile and Pre-employment Tests and Levels on the SJT

Discussion

    Most of the findings of the study are congruent with the initial hypothesis that individuals will attempt to 'fake good' on a pre-employment personality test. The study found that students had significantly higher scores on scales measuring extraversion, agreeableness, and conscientiousness when they were applying for a job as compared to their scores on these same scales when they were attempting to discover their personality type. Similarly, students indicated significantly lower neuroticism scores on the pre-employment personality test than they did on the personality profile assessment.

     However, unlike the prediction of the initial hypothesis, participants indicated significantly lower levels of openness on the pre-employment test than they did on the personality profile test. Such findings may be due to the result that the students had a particular stereotype of the ideal employee that they were trying to meet, comparable to the findings of Dalen, et al., 2001. It appears that according to the students' belief, the ideal employee may not have characteristics related to openness, and may be more likely to want to focus on his/her current projects, uninterested in getting involved with other projects. While such a belief may not seem ideal to today's managers, it might appear desirable to students who are forced to work on specific projects that professors assign, and not given a variety of options to get involved in new projects that are more interesting to them.

     Another interesting finding related to H1 is that while the means for extraversion, agreeableness, neuroticism, openness, and conscientiousness were all significantly different at the 0.05 level between the personality profile test and the pre-employment personality test, by examining the mean plots, it is apparent that the means of each of the variables do follow relatively similar patterns. For instance, on both tests, participants indicated moderate levels of extraversion, high levels of agreeableness, low levels of neuroticism, low levels of openness, and high levels of conscientiousness. Although these levels were significantly increased or decreased on the pre-employment test based on the variable, they remained at similar patterns in terms of low, moderate, and high, comparable to the findings of Thumin and Barclay (1993). Such information may indicate that the participants were not completely faking on the pre-employment personality test, but rather exaggerating their true answers.

     Implications of the findings related to H1 are that individuals are likely to indicate answers that may not necessarily be true on a pre-employment personality test based on their notion of the characteristics of an ideal employee. However, it is also important to note, that participants may not be completely lying on such tests, but rather exaggerating their levels of characteristics to fit their belief of what the ideal profile would be. Therefore, while it appears that the validity of pre-employment personality tests may be compromised due to the ability of potential employees to lie on such assessments, it may be taken into consideration that scores are simply exaggerated compared to the applicant's true characteristics.

     Furthermore, by examining students' tendency to indicate lower levels of openness on the pre-employment test than the personality profile test, it appears that they might believe employees should not be open to new experiences in the workplace. However, such a thought may not be an ideal approach to working in today's rapidly changing workplace. Therefore, it would be recommendable that professors encourage students to be flexible and adaptable to new projects in the academic world to prepare them for involvement in new experiences in a work environment.

     H2, which examined scores relative to the order in which the tests were taken, was rejected in the case of all of the variables. Contrary to the prediction of the initial hypothesis, participants were more likely to exaggerate their scores on the pre-employment test when the personality profile was taken first, rather than when the pre-employment test was taken first. These results are unlike those of Biderman et al. (2005), who found that participants were only able to 'fake good' when they took the pre-employment test first rather than the personality profile.

     Such findings might indicate that when participants in this study took the personality profile test, they were able to identify their honest scores and wanted to improve them by increasing or decreasing their scores on seeming desirable and undesirable characteristics. The absence of significant differences in the variables of openness and conscientiousness may show that participants did not need to take the personality profile test first to be able to increase or decrease their original score, but rather they were able to do such without first identifying an honest score. The finding that the order of tests exerted a relevant impact on the scores related to extraversion, agreeableness, and neuroticism may imply to future researchers that the order upon which the two types of tests are administered should be taken into account as an important experimental factor affecting the levels of each of these three variables.

     The absence of significant differences on the pre-employment personality test based on gender may be strongly indicative that past societal beliefs that the man must support the household are rapidly diminishing. Men were not found to report higher levels of characteristics that the study found participants believed to be ideal, nor were they likely to indicate lower levels of variables that individuals seemingly thought were undesirable. Such findings may tend to relate to the ever increasing presence of women in the workplace and the reliance on women's income to support the household. Implications for managers who administer pre-employment personality tests to potential employees are that expected differences in terms of exaggerated scores for men should not be taken into account when analyzing results.

     The predictions made in H4 were consistent with the findings related to neuroticism and openness. However, it was found for extraversion that both the personality test levels and the pre-employment scores correlated with this variable's levels on the SJT. For agreeableness, the opposite of what was predicted in the hypothesis was actually found, in that a significant correlation existed between the pre-employment levels and the SJT, but not between the personality profile levels and the SJT. Alternatively, conscientiousness levels on the SJT were not correlated with levels on the pre-employment personality test nor on the personality profile test.

     Such findings may indicate that individuals are more aware of how they might feel or behave in situations that elicit neuroticism, openness, and extraversion than they are in situations that involve agreeableness and conscientiousness. In the same way, the correlations that exist for extraversion and agreeableness levels on the pre-employment test and the SJT may be explained in that the image individuals intend to portray on the pre-employment personality test might actually resemble the behaviors and feelings they would initially have in the workplace (as measured by the SJT). While such behaviors and feelings might not correspond to their true personalities, it appears as though their true characteristics may alter slightly due to the expectations of specific environments, such as that of the workplace. The absence of correlations between the SJT and both tests for conscientiousness tends to indicate that individuals may not be aware of how they would feel or respond in situations eliciting conscientiousness, in or out of the workplace.

     These findings strongly implicate that managers should not expect employees to accurately self-report their feelings or behaviors pertaining to conscientiousness. Furthermore, the absence of correlations between the pre-employment test and the SJT for neuroticism, openness, and conscientiousness, strongly implies that individuals will not truly feel or behave in the workplace in the way that they predict for these variables on a pre-employment test. For agreeableness, there did exist a correlation between the pre-employment personality test and the SJT, which may seem to suggest that such a variable could be accurately predicted on a pre-employment test. However, the longevity of such feelings and behaviors may be questionable due to the lack of correlation between agreeableness levels on the personality profile and the SJT. Such information may indicate that individuals may begin a job behaving in a way that conforms to the expectations of the workplace, but their true feelings and behaviors may present themselves in time. Given that correlations existed between both of the tests and the SJT for extraversion, it is suggested that scores measuring this variable may be highly accurate in predicting behaviors and feelings in the workplace.

Limitations of the Study

     Limitations of the present study include that the pre-employment personality test was only a simulation, and did not actually involve applicants or a real job. Therefore, while assumptions can be made about the true nature of applicants' responses on a real pre-employment personality test based on this research, the findings may only apply to simulations of pre-employment tests. Also, the sample of participants for this study consisted solely of students at a midwestern university in the United States. Consequently, the results may have been affected by factors relating to the values and beliefs of these students specifically, which may not correspond with the larger populations' values and beliefs. Furthermore, due to the probability that many college students have yet to experience the professional workplace, their notion of the ideal employee may be completely different from individuals who have been professional workers.

     Another limitation related to the sole involvement of college students is that most of the participants were of traditional college age, meaning that only 5% of participants were older than 26. It appears that older participants might have been less likely to 'fake good' on the pre-employment personality test, given that they might be more thoughtful of the consequences of lying on such a test. Additionally, a significantly large portion of the students at the midwestern university are Caucasian, with a relatively smaller portion being African American. Although other ethnicities are present, the populations of each are much less. To this end, it is important to note that the findings of the study are limited to such a population and may not apply to other cultures and ethnicities.

Implications for Future Research    

     Future research related to the topic of the validity of personality tests should examine participants' treatment of personality characteristics more specific than just the five variables investigated in the present study. Also, similar studies could be performed on the actual personality tests that are given by such companies as Target, Wal-Mart, and the NFL to measure the ability of applicants to 'fake good' for specific types of jobs or careers. Another interesting topic of research in relation to personality tests might be to compare the pre-employment personality tests of employees when they first started with personality tests of the same employees one year, five years, and ten years later. Such research might show a decrease in behaviors and feelings that the employee might have originally thought to be desirable in the workplace due to the increase in job security and comfort in the work environment. Also, it might be interesting to compare the characteristics, behaviors, and feelings as indicated by employees in a pre-employment test to their actual characteristics, behaviors, and feelings as portrayed in the workplace immediately upon hire, one year, five years, and ten years after being hired.

Conclusions

     The most salient results of the research showed that applicants are able to exaggerate on personality tests in a way that most often elevates seemingly desirable characteristics and depresses seemingly undesirable characteristics. Since individuals' careers are increasingly being established and determined based on the results of these tests, it appears that in the future managers should heed great caution when using these tests for the purposes of hiring employees. However, it does appear that one potential use for these tests in the workplace might be for identifying which employees will work well together on group projects.

References

Biderman, M., McDaniel, M., & Nguyen, N. (2005, December). Effects of response instructions on faking a situational judgment test. International Journal of Selection & Assessment, 13, 250-260. Retrieved November 24, 2007, from Academic Search Premier database.

Bradley, J., & Hebert, F. (1997). The effect of personality type on team performance. The Journal of Management Development, 16, 337-353.  Retrieved November 28, 2007, from ProQuest Psychology Journals database.

Butcher, J., Hooley, J., & Mineka, S. (2005). Abnormal Psychology. Upper Saddle River, NJ: Pearson Education, Inc.

Camara, W., & Merenda, P. (2000). Using personality tests in preemployment screening: Issues raised in Soroka v. Dayton Hudson Corporation. Psychology, Public Policy, and Law, 6, 1164-1186. Retrieved November 19, 2007, from PsycARTICLES database.

Cloninger, S. (2008). Theories of Personality: Understanding Persons. Upper Saddle River, NJ: Pearson Education, Inc.

Cowen, D.A. (1989). An alternative to the dichotomous interpretation of Jung's psychological functions: Developing more sensitive measurement technology. Journal of Personality Assessment, 53(3), 459-471. Retrieved March 15, 2009, from Academic Search Premier database.

Cullen, L. (2006). SATS for j-o-b-s. Time, 167, 89-89. Retrieved December 4, 2007, from Academic Search Premier database.

Dalen, L., Roberts, A., & Stanton, N. (2001). Faking personality questionnaires in personnel selection. The Journal of Management Development, 20, 729-741. Retrieved November 28, 2007, from ProQuest Psychology Journals database.

Girelli, S., & Stake, J. (1993). Bipolarity in Jungian Type Theory and the Myers-Briggs Type Indicator. Journal of Personality Assessment, 60, 290-301. Retrieved November 30, 2007, from Academic Search Premier database.

Kubinger, K. (2002). On faking personality inventories. Psychologische Beitrage, 44, 10-17. Retrieved November 28, 2007, from ProQuest Psychology Articles database.

Kubinger, K., & Litzenberger, M. (2006). On the shortcomings of a mono-methodical approach to personality assessment. Individual Differences Research, 4, 139-158. Retrieved December 4, 2007, from Academic Search Premier database.

NFL combine is now part Vince Lombardi and part Sigmund Freud. (2001, February 15). Newswise. Retrieved April 1, 2008, from http://www.newswise.com/articles/view/?id=NFLDRAFT.SMI.

Paul, A. (2004). The Cult of Personality: How Personality Tests Are Leading Us to Miseducate Our Children, Mismanage Our Companies, and Misunderstand Ourselves. New York, NY: Free Press.

Saad, S., & Sackett, P. (2002). Investigating differential prediction by gender in employment-oriented personality measures. Journal of Applied Psychology, 87, 667-674. Retrieved December 5, 2007, from PsycARTICLES database.

Shuit, D. (2003). At 60, Myers-Briggs is still sorting out and identifying people's types. Workforce Management, 82, 72-74. Retrieved November 30, 2007, from Academic Search Premier database.

Thumin, F., & Barclay, A. (1993). Faking behavior and gender differences on a new personality research instrument. Consulting Psychology Journal: Practice and Research, 45, 11-22. Retrieved December 4, 2007, from PsycARTICLES database.

Truxillo, D., Bauer, T., Campion, M., & Paronto, M. (2006). A field study of the role of big five personality in applicant perceptions of selection fairness, self, and the hiring organization. International Journal of Selection & Assessment, 14, 269-277. Retrieved November 24, 2007, from Academic Search Premier database.

Appendix A

Thank you for your participation. Participation is on a voluntary basis, and you may withdraw at any time. All information provided will remain confidential and will be used solely for research purposes. Please feel free to contact the researcher, Cari Adams at caadams@mckendree.edu, or the faculty advisor, Tami Eggleston at tegglest@mckendree.edu, at any time.

Please fill this survey out as if you are trying to discover what personality type you are.

Circle your answer:

Gender:       Male                Female

Age: 18-21               22-25               26-30               31+

Major:         Business          Social Science             Education        Science            Other

  Strongly                                                             Strongly

  Disagree      Disagree   Undecided     Agree       Agree

1.) You are well aware of your inner feelings.                     1                 2                 3                4                5

2.) When in a group of people, you are typically the                      

focus of everyone's attention.                                       1                 2                 3                4                5

3.) You naturally take the lead in a group of people.          1                 2                 3                4                5

4.) You are talkative.                                                           1                 2                 3                4                5

5.) You are unsure of yourself with new people.                 1                 2                 3                4                5

6.) You often feel that people are talking about you.          1                 2                 3                4                5

7.) You feel that nearly everyone is a good person                        

deep down.                                                                    1                 2                 3                4                5

8.) You are often smiling.                                                    1                 2                 3                4                5

9.) People around you often annoy you.                              1                 2                 3                4                5

10.) You are able to laugh at yourself.                                 1                 2                 3                4                5

11.) You often feel worried or anxious.                              1                 2                 3                4                5

12.) You tend to feel vulnerable when experiencing                      

 peer-pressure.                                                               1                 2                 3                4                5

                                                                                        Strongly                                                             Strongly

                                                                                        Disagree      Disagree   Undecided     Agree       Agree

12.) You tend to feel vulnerable when experiencing                      

 peer-pressure.                                                               1                 2                 3                4                5

13.) You easily become emotional.                                      1                 2                 3                4                5

14.) In stressful situations you are often the person                       

who keeps calm.                                                           1                 2                 3                4                5

15.) When under pressure you think about all that

can go wrong.                                                               1                 2                 3                4                5

16.) You enjoy seeing and creating different types

of art.                                                                            1                 2                 3                4                5

17.) You often find yourself daydreaming.                                     1                 2                 3                4                5

18.) You prefer to follow a strict daily routine.                   1                 2                 3                4                5

19.) You tend to be spontaneous.                                        1                 2                 3                4                5

20.) You are willing to work hard to achieve

your goals.                                                                    1                 2                 3                4                5

21.) You feel that you are much farther ahead in

life compared to other people your age.                       1                 2                 3                4                5

22.) You easily give up if you cannot solve a

hard problem.                                                               1                 2                 3                4                5

23.) You often forget to perform tasks you had

planned to do.                                                              1                 2                 3                4                5

24.) You feel that you have set high goals for yourself.      1                 2                 3                4                5

Appendix B

Thank you for your participation. Participation is on a voluntary basis, and you may withdraw at any time. All information provided will remain confidential and will be used solely for research purposes. Please feel free to contact the researcher, Cari Adams at caadams@mckendree.edu, or the faculty advisor, Tami Eggleston at tegglest@mckendree.edu, at any time.

Please fill this survey out as if it is a pre-employment personality test for your dream job.

Circle your answer:

Gender:       Male                Female

Age: 18-21               22-25               26-30               31+

Major:         Business          Social Science             Education        Science            Other

  Strongly                                                             Strongly

  Disagree      Disagree   Undecided     Agree       Agree

1.) You are well aware of your inner feelings.                     1                 2                 3                4                5

2.) When in a group of people, you are typically the                      

focus of everyone's attention.                                       1                 2                 3                4                5

3.) You naturally take the lead in a group of people.          1                 2                 3                4                5

4.) You are talkative.                                                           1                 2                 3                4                5

5.) You are unsure of yourself with new people.                 1                 2                 3                4                5

6.) You often feel that people are talking about you.          1                 2                 3                4                5

7.) You feel that nearly everyone is a good person                        

deep down.                                                                    1                 2                 3                4                5

8.) You are often smiling.                                                    1                 2                 3                4                5

9.) People around you often annoy you.                              1                 2                 3                4                5

10.) You are able to laugh at yourself.                                 1                 2                 3                4                5

11.) You often feel worried or anxious.                              1                 2                 3                4                5

12.) You tend to feel vulnerable when experiencing                      

 peer-pressure.                                                               1                 2                 3                4                5

                                                                                        Strongly                                                             Strongly

                                                                                        Disagree      Disagree   Undecided     Agree       Agree

12.) You tend to feel vulnerable when experiencing                      

 peer-pressure.                                                               1                 2                 3                4                5

13.) You easily become emotional.                                      1                 2                 3                4                5

14.) In stressful situations you are often the person                       

who keeps calm.                                                           1                 2                 3                4                5

15.) When under pressure you think about all that

can go wrong.                                                               1                 2                 3                4                5

16.) You enjoy seeing and creating different types

of art.                                                                            1                 2                 3                4                5

17.) You often find yourself daydreaming.                                     1                 2                 3                4                5

18.) You prefer to follow a strict daily routine.                   1                 2                 3                4                5

19.) You tend to be spontaneous.                                        1                 2                 3                4                5

20.) You are willing to work hard to achieve

your goals.                                                                    1                 2                 3                4                5

21.) You feel that you are much farther ahead in

life compared to other people your age.                       1                 2                 3                4                5

22.) You easily give up if you cannot solve a

hard problem.                                                               1                 2                 3                4                5

23.) You often forget to perform tasks you had

planned to do.                                                              1                 2                 3                4                5

24.) You feel that you have set high goals for yourself.      1                 2                 3                4                5

Appendix C

Work Environment Related Portion

You have been assigned to work on an important project that may cost your company millions of dollars if it fails. Ten of your co-workers will be working on the project with you.

Please answer the following questions as if you are in the scenario above:

                                                                                        Strongly                                                             Strongly

                                                                                        Disagree      Disagree   Undecided     Agree       Agree

1.) You are glad to be given a new and exciting

project to work on.                                                         1                 2                 3                4                5                                                       

2.) You would rather focus on your usual daily

tasks than work on this project.                                                 1                 2                 3                4                5                                           

3.) You would rather work on this project by yourself.      1                 2                 3                4                5

4.) You will most likely take on a leadership role

in the group.                                                                   1                 2                 3                4                5

5.) You will begin the project by developing a plan

for each step.                                                                  1                 2                 3                4                5

6.) The project must have a timeline for you to follow.       1                 2                 3                4                5

7.) You may have trouble approaching other people

in the group with an idea.                                              1                 2                 3                4                5

8.) You will probably come up with most of the ideas

for the project.                                                               1                 2                 3                4                5

9.) You will easily become angry with other group

members if they do not agree with your ideas.              1                 2                 3                4                5

10.) You feel that the other group members' work

cannot be trusted.                                                         1                 2                 3                4                5

11.) You will most likely become very depressed

if the project fails.                                                        1                 2                 3                4                5