Lost in translation

Abstract

Purpose In this paper, empirical evidence is presented regarding the translation of the learning tactics inventory (LTI) instrument that measures learning versatility in entrepreneurs with four main learning strategies: acting, thinking, feeling or access to others. The purpose of this paper is to show how translating instruments from other languages and for cross-cultural studies is not sufficient to achieve instrument validity, and the use of structural equation modeling can help to strengthen the process. Design/methodology/approach After using iterative and multi-technique strategies that involved close translation and adaptation, structural equation modeling was also performed to validate whether relationships exist among the constructs and their variables using a confirmatory analysis. Findings After careful translation, the Spanish version of the LTI instrument does not measure the intended constructs. Such evidence was uncovered by contrasting the different dimensions of the English and Spanish translated versions. Research limitations/implications Instruments in cross-cultural studies require more than translation strategies to adapt the instrument to the new context. This paper shows that, with the use of structural equation modeling, constructs may change in different international contexts and how misinterpretations of the instrument can occur if additional validity tests are ignored. Originality/value Consistent with the extant literature, the findings suggest that, when studying a complex phenomenon such as learning through a survey developed in a different country and language, cultural factors should be explained to maintain construct validity. Hence, in entrepreneurship and management research, instruments such as the LTI need to be validated with confirmatory analysis to accurately reflect the different learning strategies of entrepreneurs across cultures.