Intentional response distortion on personality tests: using eye-tracking to understand response processes when faking

Intentional response distortion or faking among job applicants completing measures such as personality and integrity tests is a concern in personnel selection. The present study aimed to investigate whether eye-tracking technology can improve our understanding of the response process when faking. In an experimental within-participants design, a Big Five personality test and an integrity measure were administered to 129 university students in 2 conditions: a respond honestly and a faking good instruction. Item responses, response latencies, and eye movements were measured. Results demonstrated that all personality dimensions were fakeable. In support of the theoretical position that faking involves a less cognitively demanding process than responding honestly, we found that response times were on average 0.25 s slower and participants had less eye fixations in the fake good condition. However, in the fake good condition, participants had more fixations on the 2 extreme response options of the 5-point answering scale, and they fixated on these more directly after having read the question. These findings support the idea that faking leads to semantic rather than self-referenced item interpretations. Eye-tracking was demonstrated to be potentially useful in detecting faking behavior, improving detecting rates over and beyond response extremity and latency metrics.     

Response bias in self-report measures of general health

This study set out to examine the susceptability of two widely used, self-report, psychiatric inventories to response set bias. Subjects were requested to either fake well (physically and psychologically fit), fake psychologically ill, fake physically ill or respond honestly on the General Health Questionnaire (GHQ) and the Langner Scale of Distress (L-22). Predictably subjects who faked well had significantly lower scores than either of the other two experimental groups, however there was no difference between the Fake Well and Control group on the GHQ total or the Somatic and Depression scales. Subjects who faked psychologically ill showed 4 out of 11 significant differences between those who faked physically ill. All of the comparisons between the two Fake Ill groups, and the Well and Control groups, were statistically significant the former reporting higher incidences of distress than the latter. The results are discussed in terms of questionnaire design, subject motivation and alternative methods of assessing mental health.     

Can defendants with mental retardation successfully fake their performance on a test of competence to stand trial?

The purpose of this study was to determine whether criminal defendants with mental retardation can feign poor performance on a test of competence to stand trial. Four groups of adjudicated criminal defendants were given a test of competence to stand trial. In the experimental condition, defendants with mental retardation (MR) and mentally typical defendants (non-MR) were given instructions to fake their performance on the test. In the control, MR defendants and non-MR defendants took the test under standard conditions. The results indicated that both the MR and non-MR fakers scored significantly lower than the MR and non-MR controls. There was no significant difference in the scores of the MR and non-MR fakers. The results suggest that defendants with mental retardation may be able to successfully fake their performance in certain circumstances. Implications of these findings are discussed.     

The influence of demand characteristics and expectancies in the measurement of salivary response

The sensitivity of salivary response measures to demand and expectancies was studied. One hundred thirty-one college students were tested for increases or decreases in salivation under one of three conditions: (a) when told to imagine the taste of another food more or less palatable than a target food; (b) when given a placebo pill that, they were told, would alter their salivation; and (c) when given direct instructions to alter their salivation. The type of instructions subjects received altered their salivation response. Specifically, subjects told to increase salivation had higher levels than those told to decrease salivation. These results suggest that salivation measures, which have been viewed by some as refractory to demand and related effects, can be controlled. Additional results of the study call for further validation of salivation as a measure of hunger and perceived food palatability.     

Innocent but proven guilty: Eliciting internalized false confessions using doctored‐video evidence

More powerful computers and affordable digital‐video equipment means that desktop‐video editing is now accessible and popular. In two experiments, we investigated whether seeing fake‐video evidence, or simply being told that video evidence exists, could lead people to believe they committed an act they never did. Subjects completed a computerized gambling task, and when they returned later the same day, we falsely accused them of cheating on the task. All of the subjects were told that incriminating video evidence existed, and half were also exposed to a fake video. See‐video subjects were more likely to confess without resistance, and to internalize the act than told‐video subjects, and see‐video subjects tended to confabulate details more often than told‐video subjects. We offer a metacognitive‐based account of our results. Copyright © 2008 John Wiley & Sons, Ltd.     

Integrity Testing and Deviance: Construct Validity Issues and the Role of Situational Factors

Two studies were conducted to investigate relationships among several types of preemployment integrity tests and situational factors which could play a role in employee deviance. In one study, three integrity measures were administered to subjects who were told to assume the role of job applicant for a position in either a large, small, or unspecified organization. Moral development stage, social desirability, and demographic data were collected. While there was some construct consistency across the different integrity measures, the expected relation between honesty and moral development was not found. There was also no difference in integrity test scores as a function of the organizational size manipulation. In a second laboratory study in which subjects were asked to play the role of job incumbents, both size and perceived equity of organizational climate were experimentally manipulated. The dependent measure was a scale of likelihood of engaging in organizationally proscribed behaviors. While again no effect was found for organizational size, those individuals who perceived themselves as employed in an organization in which employees are mistreated tended to report greater likelihood of engaging in organizational deviance. Implications for integrity testing in personnel selection are discussed.     

Special Abilities Assessment in Children

The faking-detection validity and incremental validity of response latencies to Minnesota Multiphasic Personality inventory (MMPI) items was investigated using an analog research design. One hundred undergraduates were assigned at random to five groups: each group received different faking instructions (standard, fake good, fake bad, fake good with incentive, fake bad with incentive). All subjects completed a computer-administered version of the MMPI. Content-determined response deviance scores and latencies of responses to Subtle and Obvious scale items were determined for each subject. The principal findings suggest that response latencies may have greater faking good detection ability than responses deviance scores, and that response latencies have statistically significant incremental validity for both the detection of faking good and faking bad, when latencies are used with response deviance scores obtained from Subtle and Obvious scales.     

THE SUSCEPTIBILITY OF OVERT AND COVERT INTEGRITY TESTS TO COACHING AND FAKING

Abstract— Although previous research has indicated that faking can affect integrity test scores, the effects of coaching on integrity test scores have never been examined We conducted a between-subjects experiment to assess the effects of coaching and faking instructions on an overt and a covert integrity test Coaching provided simple rules to follow when answering test items and instructions on how to avoid elevated validity scale scores There were five instruction conditions "just take,""fake good,""coach overt,""coach covert," and coach both All subjects completed both overt and covert tests and a measure of intelligence Results provided strong evidence for the coachability of the overt integrity test, over and above the much smaller elevation in the faking condition The covert test apparently could be neither coached nor faked successfully Scores on both integrity tests tended to be positively correlated with intelligence in the coaching and faking conditions We discuss the generalizability of these results to other samples and other integrity tests, and the relevance of the coachability of integrity tests to the ongoing debate concerning the prediction of counterproductive behavior     

Using Personality Testing for Personnel Selection: Faking and Test-Taking Inductions

This research assessed whether warning subjects that faked responses could be detected would reduce the amount of faking that might occur when using a personality test for selection of police officers. Also, personality test subscales which best differentiated honest from dissimulated responses were determined. Subjects (N=120) were randomly assigned to a straight-take (that is, respond honestly), fake good, or modified fake good group. Both fake good groups were instructed to respond to the test so as to appear favourably for the job; additionally, the modified fake good group was warned that faking could be detected and could reduce hiring chances. Multivariate analyses revealed significant differences on the Denial and Deviation subscales between the three conditions (p <0.01). The pattern of differences suggested that the threat of faking detection reduced faking. Potential application of these findings in personnel selection was discussed.     

POTENTIAL VERSUS ACTUAL FAKING OF A BIODATA FORM: AN ANALYSIS ALONG SEVERAL DIMENSIONS OF ITEM TYPE

Two studies examined faking of a 25-item biodata questionnaire. The first study investigated potential and actual faking of the form using three groups: a group told to make themselves look as good as possible, a group told to complete the form honestly, and a group completing the instrument in a real selection situation. Subjects were 58 current employees and 231 job applicants. Results indicated that subjects could fake the instrument when instructed to do so. Also, some faking appeared to be occurring in practice, although results depended upon the composition of the comparison group. Only eight items appeared to be fakable, and only three of these seemed to be faked in practice. In Study 2, 26 business majors rated the biodata items on eight dimensions of item type. Results showed that the three items faked in practice were less historical, objective, discrete, verifiable, and external than other items, and were more job relevant.     

The good,the bad and the mad: Response bias in self-report measures

This study set out to examine the susceptibility of five extensively used, self-report measures to response set bias. Subjects were requested either to fake good, (give a good impression), fake bad, (give a bad impression), fake mad, (give an impression of mental instability) or respond honestly. Subjects who faked good had significantly higher Extraversion, Lie and Social Desirability scores but lowest Neuroticism, Psychoticism and Social Anxiety scores. Subjects who faked bad had significantly lower Extraversion and higher Psychoticism and Social Anxiety scores. Fake mad subjects scored higher on Self-Monitoring and Locus of Control. Four of the eight scales showed significant differences between subjects faking bad and those faking mad. The results are discussed in terms of questionnaire design and respondent's motivation.     

The effects of episodic and semantic memory on the identification of sentences masked by white noise

Two studies examine how episodic and semantic memory affect subjects’ abilities to repeat sentences masked by white noise. Subjects first hear a list of 70 sentences that are not masked. Subjects in the framework conditions are told prior to hearing the list that all sentences refer to a contextual framework concerning a deserted island. Subjects in the no-framework groups are not given this information. Subjects in the framework-after condition are given this information only after hearing the list of sentences. Subjects then perform a white-noise identification task. The results indicate that framework subjects are able to identify both old and new framework-related information better than other subjects. Subjects in the no-framework and framework-after conditions identify old information better than control subjects who do not participate in an acquisition phase. Emphasis is placed on the interdependence of episodic and semantic memory, including conditions leading to such interdependence.     

A psychological test battery to detect prison inmates who fake insanity or mental retardation

A contrasted groups design was used to investigate the accuracy with which a test battery could detect persons faking mental disorder. The MMPI and Bender Gestalt were used in combination with a Malingering Scale that was developed for initial validation. One hundred adult males comprised five groups of 20 subjects each. Noncriminal psychiatric inpatients and mental retardates were compared with prison inmates who were given a financial incentive to successfully fake ‘insanity’ or mental retardation, and with inmate controls. Based on discriminant analyses, 92 to 95% of subjects were correctly classified as either faking or not faking. Although cross-validation is needed, this study reduced many of the analog factors which have comprised the generalizability of previous findings.     

Test Sophistication and Faking in the TAT Situation

This study investigated the effects of conscious altering of response (faking) and test sophistication on TAT performance. Thirty-six naive undergraduates and 36 TAT-sophisticated graduate students and clinicians wrote stories to all 20 TAT cards under one of two conditions: standard, and instructions to “fake” hostility. The four groups were then compared by Analysis of Variance on a number of quantitative response measures. Effects of the two experimental variables, sophistication and faking, varied with the response measure, but, in general, faking instructions led to response changes, and sophisticated subjects performed differently than naive ones. Some sex differences were also apparent. It was concluded that sophistication is a factor in how a subject fakes, the sophisticated subjects generally doing a better job of it. The implications for the effect of stimulus pull were discussed and it was concluded that the pull value of the stimulus is not affected by a peripheral response set but is affected by sophistication.     

Factors Determining National Stereotypes

Summary

In order to test various explanations of the fact that perceived competence appears to correlate positively with altruism, Ss were given false feedback as to their competence on a “visual creativity” test and then asked to help with a second task, writing creative color names. Three groups of Ss who were told they were competent were given reduced cost, increased happiness, and increased responsibility as reasons for helping; the other three groups were given no explanations, but were told that their competence was high, or low, or given no information. The three explanation groups helped more than the other three groups, and the reduced cost group helped more than the other five. The three dependent measures of helping (percentage of helpful Ss, duration of help, and magnitude of help) proved differentially sensitive to treatment and sex differences. Only the latter two measures showed greater helping in the reduced cost group; only the magnitude measure showed significantly greater helping for males than females.     

The use of subtle items in detecting deception

When subjects attempt to fake psychopathology on the MMPI, scores on subtle subscales tend to be lower than those of nonfaking subjects. Our study hypothesized that this paradox comes about because the subtle subscales have no predictive validity, but their face validity for psychopathology is the opposite of the keyed direction for psychopathology. Subjects who attempt to fake psychopathology do so on the basis of item content and thus achieve lower rather than higher scores. Three groups of 80 undergraduates took the MMPI under regular, faking-good, or faking-bad instructions. As expected, faking-bad subjects scored significantly lower than regular subjects on the 100 most subtle items, and this was due to their responses to those. 73 of the items whose face validity was misleading. The results are consistent with other work showing valid uses of subtle items in detecting deception.     

The halo effect revisited: Forewarned is not forearmed

The purpose of the present research was to determine if forewarning subjects about the halo effect eliminated the effect or made people aware of its impact. The research was a replication and extension of R. E. Nisbett and T. D. Wilson's (Journal of Personality and Social Psychology, 1977, 35, 250–256) experiment on the halo effect. Subjects viewed an interview of a college instructor who was either warm or cold and then rated his physical appearance, mannerisms, and French accent. Some subjects were told to introspect about their cognitive processes while viewing the interview and were told that the purpose of the study was to see how aware they were of the determinants of their ratings. Other subjects were also told specifically what the halo effect was and given a motivation either to show it or not to show it. In addition, a measurement technique different from that used by Nisbett and Wilson was utilized to allow a more precise measurement of awareness, and a new cover story was used to make the task more involving and important to subjects. Despite these attempts to eliminate the halo effect (or, at a minimum, to make people aware of it), the results indicated that subjects in all conditions were very susceptible to it. That is, subjects who viewed the warm version of the interview rated the instructor's appearance, mannerisms, and accent significantly higher than did those who viewed the cold version, even when informed and forewarned about this effect. The forewarning and introspection instructions also had no impact on subject's awareness of the halo effect. Subjects in all conditions indicated that their liking for the instructor had had only a minimal effect on their ratings.     

Testing the Compatibility Test: How Instructions, Accountability, and Anticipated Regret Affect Prechoice Screening of Options

Subjects screened a set of jobs, retaining those for which they wished to apply and rejecting those that were no longer under consideration. In Experiment 1, subjects who indicated the jobs for which they would apply/not apply screened out fewer jobs than those with instructions to reject/not reject or those with instructions simply to screen (control). There were no differences between the reject and control conditions. Experiment 2 used a design similar to that of Experiment 1, but subjects were made accountable for their screening judgments. The reject-apply discrepancy remained, but the accountability manipulation made the subjects more stringent in their screening compared to those who were not accountable for their judgments. In Experiment 3, subjects were told to consider either the regret resulting from retaining a bad option (regret bad) or the regret from rejecting a good option (regret good). Subjects in the regret bad condition rejected more jobs than did subjects in the regret good condition, but not more than subjects in the control condition. As predicted by image theory, the normal screening process appears to be to screen out the bad options rather than screen in the good options. This is demonstrated by screening in the control condition being similar to screening under the reject instructions (Experiment 1) and under regret bad instructions (Experiment 3), since these conditions were shown to focus attention on the bad options. Copyright 1999 Academic Press.