Paired Comparison Choice Data Approach in Addressing the Issue of Teacher Force Recruitment in Ghana
Eric Nyarko *
Department of Statistics, School of Physical and Mathematical Sciences, College of Basic and Applied Sciences, University of Ghana, Ghana.
Isaac Kwame Baidoo
Department of Statistics, School of Physical and Mathematical Sciences, College of Basic and Applied Sciences, University of Ghana, Ghana.
Kwabena Doku-Amponsah
Department of Statistics, School of Physical and Mathematical Sciences, College of Basic and Applied Sciences, University of Ghana, Ghana.
*Author to whom correspondence should be addressed.
Abstract
The shortage of public sector teachers in remote and rural areas is one of the main challenges facing policy makers in the education sector, in both developing and developed countries. This study sought to analyze the preferences of public sector teachers, and how they would respond to various job incentive packages associated with working in a rural area or remote location. A discrete choice experiment approach was used to capture respondent’s preferences. Under the indifference assumption of equal choice probabilities, the models were calibrated using the multinomial logit model. The magnitude of estimates from the multinomial logit model and the marginal effects indicate that, for rural area job preferences, public sector teachers highly value non-financial incentive packages. This was generally supported by the subgroup models by gender. The provision of housing in a rural area was highly valued by female teachers to their male counterparts. The levels of salary resulted in disutility of rural area job preferences.
Keywords: Stated preference modeling, job preferences, teacher’s recruitment, job location