The determinants of training participation, a multilevel approach: Evidence from PIAAC

Abstract

This chapter uses the first round of the Programme for the International Assessment of Adult Competencies (PIAAC) survey data to find out the determinants of training participation of workers for the 24 countries that participated in the survey. Two measures are used in quantifying training participation: (1) whether workers participated in any training during the last 12 months and (2) the number of training modes of participation (number of types of trainings workers attended during the last 12 months). Logistic and Poisson multilevel models were used to model the two measures respectively. Both models show similar findings for the fixed effects. The results of the random slopes models show that heterogeneity exists across the 24 countries, indicating that the effects of covariates on workers’ training participation and the number of training modes vary across countries. The magnitude of correlations of these random slopes differs between the logistic and Poisson models, indicating that the associations of these random effects are not totally in agreement between training participation and the number of training modes.