Cross-cultural construct validation of the learning transfer system inventory in Tawian

Abstract

Human Resource Development has historically made a large contribution to Taiwan's economic growth. Organizations in Taiwan perceive training as one of their top priorities to maintain their competitive advantage. Although many organizations have spent considerable money on training, there has been a lack of a comprehensive, generalizable, psychometrically-sound instrument to investigate transfer of learning in Taiwan. This study cross-culturally validated the constructs of the Learning Transfer System Inventory (LTSI) for use in Taiwan. By doing so, HRD practitioners in Taiwan can benefit by having an instrument to diagnose intervening variables and improve individual performance. The LTSI was translated through a rigorous forward-backward translation process including qualitative, quantitative, and pilot evaluations with feedback loops. A heterogeneous sample 583 trainees from 20 different organizations and 71 different training were surveyed. The results showed that 15 factors were valid for use in Taiwan constituting two training domains, Training in Specific and Training in General. The reliabilities of the 15 validated factors ranged from .65 to .92, and only one of those were less than .75. This study also assessed transfer system characteristics across different situational variables (organizational type, organization, and training type) and individual variables (gender, age, education, job type, hours of training experience in current organization, years of total job experience, and years of job experience in current organization). The results suggested that non-profit organizations appeared to have a stronger transfer system than other types of organizations. Similar types of organizations may have similar but not identical transfer systems. The results also suggested that the situational variables were the true source of variance rather than the individual variables. This study found that when the quality of translation is controlled, scales with low reliability can create translation problems which, in turn, influence the factor structure. This suggested that the reliability of scales should always be examined before translating an instrument. Finally, this study has created many opportunities for understanding transfer processes, both cross-culturally and in Taiwan. It also provides a valid, generalizable, culturally appropriate instrument to help unleash human expertise and establish HRD accountability in Taiwan. Future research directions are provided.