The Development Of A Prediction Model For Till Operator Job Performance

The purpose of this study was to develop a selection model that would predict the performance of Till Operators (other names used for a similar role across the globe are; Checkout Operators or Cashier) in a large retail chain of stores before they are hired. The retail chain is located in Zimbabwe. The focus was therefore on the process of developing a credible job analysis approach that would lead to the development of a selection model that could be used to select Till Operators.

Executive summary of key findings

• Overall, the results show that General Mental Ability (GMA) predicts the performance of Till Operators, though the relationship is weak when contextual variables are not taken into consideration. Cognitive ability (GMA) was positively related to performance as predicted (r = 0.16, p <0.05).
• When moderated mediation path analysis was employed to explore the effects of contextual variables, the results changed significantly. GMA contribution to Till Operator performance increased substantially
• Moderated mediation path analysis showed the dominance of contextual factors over person variables.
• The overall results show that contextual variables of service climate and interactional justice dwarf to insignificance the influence of individual variables of GMA and adaptive selling.
• Further analysis shows that for the current Till Operators already in the system, contextual variables of interactional justice and service climate contribute (54%) to performance compared to individual-level predictors at 46%. This means that neglecting contextual variables would result in misspecification of the prediction model for the incumbents.
• When predicting performance for prospective Till Operators, adaptive selling behaviour contributes the most (64.5%), followed by GMA (31.1%) and, lastly, agreeableness (4.4%).

• This project was anchored on a rigorous job analysis approach and the results from that analysis can not be ignored. Job analysis descriptor ratings of importance from accurate raters were positively and significantly correlated with validities of measures of these descriptors (r = 0.87, CI 0.516 to 0.988, p < 0.05) compared to ratings from the whole sample before data clean-up of careless/insufficient effort raters (r = 0.56, CI -0.739 to 0.995, p > 0.05).
• The approach used in this study can be used when developing selection models for all high headcount role such as Bank Tellers, Call Centres staff, Sales Staff, Insurance Sales Staff e.tc.

Project Leaders & Authors

1. Memory Nguwi - Managing Consultant & Registered Occupational Psychologist at Industrial Psychology Consultants (Pvt) Ltd mnguwi@ipcconsultants.com
2. Douglas Zvomuya - Registered Occupational Psychologist and Associate Consultant at Industrial Psychology Consultants (Pvt) Ltd