The Navy Survey on Reenlistment and Quality of Service: Using Choice-Based Conjoint to Quantify Relative Preferences for Pay and Nonpay Aspects of Naval Service

Abstract

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES):Center for Naval Analyses 4825 Mark Center Drive Alexandria, Virginia 22311-1850 PERFORMING ORGANIZATION REPORT NO.:CRMD0008416.A2 SPONSORING AGENCY NAME(S) AND ADDRESS(ES):Assistant Deputy Chief of Naval Operations, Manpower and Personnel (NIB) 2 Navy Annex Washington, DC 20380-1775 SUPPLEMENTARY NOTES: N/A ABSTRACT: (U)The primary tasking for this project was to develop a choice-based conjoint (CBC) model of Sailors' preferences for reenlistment incentives and quality-of-service factors to learn more about how compensation-based reenlistment incentives compare with non-compensation factors in influencing reenlistment decisions. In response to this tasking, we designed the Navy Survey on Reenlistment and Quality of Service (NSRQOS), and sent it to approximately 9,000 first-term Sailors. The survey results indicate that, even with several measures of pay included in the survey, nonpay factors play a substantial role in guiding Sailors' reenlistment intentions.(U) The secondary tasking was to demonstrate how CBC survey data and models can be used to analyze personnel issues. Based on a review of the relevant literature, we concluded that CBC surveys are most appropriate for short-term applications in which the decisions of interest are made on the basis of competitive differences between a few well-known attributes. Therefore, the best Navy personnel applications of the CBC methodology would be analyses of proposed policies that entail well-defined trade-offs between a few characteristics. Furthermore, use of CBC results should focus on the relative effects of different characteristics; results should not focus on absolute changes in predicted reenlistment rates or program participation rates. SUBJECT TERMS:Reenlistment behavior; reenlistment intentions; reenlistment incentives; pay elasticity; quality of service; preference survey; choice-based conjoint (CBC); discrete-choice modeling; hierarchical Bayes estimation; disk by mail; survey response rates