Forewarned is forearmed: Using AI to detect SDR tendencies as personnel selection tools

Current practices of personnel selection often use questionnaires and interviews to assess candidates’ personality, but the effectiveness of both approaches can be hampered if social desirable responding (SDR) occurs. Detecting biases like SDR is important to ensure valid personnel selection for any organization, yet current instruments for assessing SDR are either inefficient or insufficient. In this paper, we propose a novel approach to appraise job applicants’ SDR tendency by employing Artificial Intelligence (AI)-based techniques. Our study extracts thousands of image and voice features from the video presentation of 91 simulated applicants to train two deep learning models for predicting their SDR tendency. The result shows that our two models, namely the Deep Image Model and Deep Voice Model, can predict SDR tendency with 82.55% and 88.89% accuracy rate, respectively. The Deep Voice Model moreover outperformed the baseline model built on a popular deep learning algorithm ResNet by 4.35%. These findings suggest that organizations can use AI driven technologies to assess job applicants’ SDR tendency during recruitment and improve the performance of their personnel selection.

Self-Report Imagery Questionnaire Scores and Subtypes of Social-Desirable Responding

Allbutt, Ling, and Shafiullah (2006) and Allbutt, Shafiullah, and Ling (2006) found that scores on self-report measures of visual imagery experience correlate primarily with the egoistic form of social-desirable responding. Here, three studies are reported which investigated whether this pattern of findings generalized to the ratings of imagery vividness in the auditory modality, a new version of the Vividness of Visual Imagery Questionnaire ( Marks, 1995 ), and reports of visual thinking style. The measure of social-desirable responding used was the Balanced Inventory of Desirable Responding (BIDR; Paulhus, 2002 ). Correlational analysis replicated the pattern seen in our earlier work and of the correlations with the egoistic bias, the correlation with vividness of visual imagery was largest and significant, the correlation with visual thinking style next largest and approached significance, and the correlation with vividness of auditory imagery was the smallest and not significant. The size of these correlations mirrored the extent to which the three aspects of imagery were valued by participants.

Brief Report. The Effect of the Accountability Instruction and Two Job Types on the Big Five Scores

Summary: Self-report questionnaires are susceptible to social desirability, and applicants can adapt their answers to job requirements. To combat social-desirable responding, subjects are asked to answer spontaneously and not think too long about their answer. Hofstee (1996) faces this problem in a different way by proposing to ask the applicant to answer accountably, in other words, to prove answers through future behavior. In the first study 46 subjects filled in the Big Five under the spontaneous and accountability conditions. Contrary to expectations, there was a small but significant effect. If subjects were asked to give answers they would have to account for, they scored higher on conscientiousness and emotional stability. In the second study subjects filled in the Big Five for two jobs differing in the extent to which the applicant has to manage people or systems. In line with expectations, there was an effect of autonomy but contrary to expectations not of conscientiousness and extroversion. The practical consequences of the accountability instruction for the validity of personality questionnaires and of job types for norms are discussed.