Toward Stable Predictions of Apprentices’ Training Success
Can Artificial Neural Networks Outperform Linear Predictions?
Abstract
Mechanical (statistical) predictions have proven to be useful in personnel selection. However, such predictions require the use of an algorithm to aggregate different predictor scores. The identification of such an algorithm requires analyzing predictor and criterion data obtained from previous applicants. The present manuscript compared predictions made by two different statistical methods: artificial neural networks (ANNs) and multiple regression analysis. Therefore, three consecutive cohorts of apprentices (n = 322, 217, and 118) were examined. Algorithms were derived from one cohort and applied to more recent cohorts. It was shown that ANNs outperformed linear predictions in a cross-validation within the cohorts. However, applying trained ANNs to other samples resulted in a predictive power which was worse than most of the linear predictions. Thus, we conclude that ANNs should only be used as selection algorithm if their validity in different cohorts has been confirmed.
References
1973). Information theory and an extension of the maximum likelihood principle. In , Proceedings of the 2nd International Symposium on Information Theory (pp. 267–281). Budapest: Akademiai Kiado.
(2007). Statistical judgment formation in personnel selection: A study in military aviation psychology. Military Psychology, 19, 119–136.
(1969). Mechanical Comprehension Test (MCT). New York, NY: The Psychological Corporation.
(1995). Neural networks for pattern recognition. Oxford, UK: Oxford University Press.
(1995). Alternative techniques for predicting success in air controller school. Military Psychology, 7, 207–219.
(1999). Modeling career counselor decisions with artificial neural networks: Predictions of fit across a comprehensive occupational map. Journal of Vocational Behavior, 54, 196–207.
(1983). Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum.
(1993). An application of the theory of neural computation to the prediction of workplace behavior: An illustration and assessment of network analysis. Personnel Psychology, 46, 503–524.
(1989). On the approximate realization of continuous mappings by neural networks. Neural Networks, 2, 183–192.
(1998). Predicting naval aviator flight training performance using multiple regression and artificial neural networks. International Journal of Aviation Psychology, 8, 121–135.
(2000). Clinical versus mechanical prediction: A meta-analysis. Psychological Assessment, 12, 19–30.
(2006). Neuronale Netze: Nichtlineare Methoden der statistischen Urteilsbildung in der psychologischen Eignungsdiagnostik
([Neural networks: Nonlinear methods of judgment formation in personnel selection] . Zeitschrift fuer Personalpsychologie, 5, 4–15.2006). Vergleich kriteriumsbezogener Validitäten verschiedener Intelligenztests zur Vorhersage von Ausbildungserfolg in Deutschland. Ergebnisse einer Metaanalyse
([Comparing criterion-related validities of different intelligence tests for the prediction of training success in Germany: A meta-analysis] . Zeitschrift für Personalpsychologie, 5, 145–162.1983). Wilde Intelligenz Test: Ein Strukturdiagnostikum
([Wilde Intelligence Test: A diagnostic tool for intelligence structure] . Göttingen, Germany: Hogrefe.1982). Judgement under uncertainty: Heuristics and biases. Cambridge, UK: Cambridge University Press.
(2006). Improving the quality of proficiency assessment: The German standardization approach. Psychology Science, 48, 85–98.
(1992). Neuronale Netze. Grundlagen, Anwendungen, Beispiele
([Neural networks. Basics, applications, examples] . München, Germany: Oldenbourg.2005). Investigating nonlinear conscientiousness-job performance relations for clerical employees. Human Performance, 18, 199–212.
(1990). Eignungsdiagnostik als prognostische Hilfe bei der Auswahl, Platzierung und Entwicklung von Führungskräften
([Proficiency assessment as a prediction tool in selection, placement, and development of managerial staff] . In Management Diagnostik (pp. 32–46). Göttingen, Germany: Hogrefe.2008). Effects of personality characteristics on knowledge, skill, and performance in servicing retail customers. International Journal of Selection and Assessment, 16, 272–280.
(2008). The relationship of age to ten dimensions of job performance. Journal of Applied Psychology, 93, 392–423.
(2001). Künstliche Neuronale Netze: Überblick, Einsatzmöglichkeiten und Anwendungsprobleme. In , Handbuch Data Mining im Marketing: Knowledge discovery in marketing databases. Braunschweig, Germany: Vieweg.
(2000). Applying artificial neural network models to clinical decision making. Psychological Assessment, 12, 40–51.
(2006). R: A language and environment for statistical computing (version 2.3.1). Vienna, Austria: R Foundation for Statistical Computing.
(1991). Research in the neural networks/expert system area. Decision Line, 22, 11–14.
(1991). Predicting training success: Not much more than g. Personnel Psychology, 44, 321–332.
(2003). International validity generalization of GMA and cognitive abilities: A European community meta-analysis. Personnel Psychology, 56, 573–605.
(2006). Neural networks in organizational research: Applying pattern recognition to the analysis of organizational behavior. Washington, DC: American Psychological Association.
(1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262–274.
(2004). Prädiktion von Ausbildungserfolg: Allgemeine Intelligenz (g) oder spezifische kognitive Fähigkeiten?
([The prediction of training success: General intelligence (g) or specific cognitive abilities?] . Zeitschrift für Personalpsychologie, 3, 147–158.2005). Manual Expertensystem Verkehr
([Manual for an expert system for traffic] . Mödling: Schuhfried GmbH.1978). Estimating the dimension of a model. Annals of Statistics, 6, 461–464.
(1999). Application of two neural network paradigms to the study of voluntary employee turnover. Journal of Applied Psychology, 84, 177–185.
(2001). Thinking differently: Assessing nonlinearities in the relationship between work attitudes and job performance using a Bayesian neural network. Journal of Occupational and Organizational Psychology, 74, 47–61.
(2005). Diagnostische Unterscheidbarkeit unfallfreier und mehrfach unfallbelasteter Kraftfahrer mit Hilfe nicht-linearer Auswertemethoden
([Differentiation between accident-free drivers and drivers with multiple accidents using nonlinear statistical methods] . Zeitschrift für Verkehrssicherheit, 51, 82–86.2004). Improvements in personnel selection with neural networks: A pilot study in the field of aviation psychology. International Journal of Aviation Psychology, 14, 103–115.
(2002). Modern applied statistics with S. New York, NY: Springer.
(1996). Understanding neural networks as statistical tools. The American Statistician, 50, 284–293.
(1995). Cognitive ability as a moderator of the relationship between personality and job performance. Journal of Management, 21, 1129–1139.
(