The Colombian textile sector, which represents 3% of the gross domestic product, has staff turnover problems. When a worker quits, an immediate replacement is needed in order to not affect productivity. Companies often recruit people without training or even experience. In this paper is analyses job turnover of businesses in the textile sector in Barranquilla (Colombia) through internal, external and contextual factors with their respective indicators. The investigation is quantitative with a correlational scope and a cross-sectional design. The methodology consisted of reliability testing (Cronbach's alpha and McDonald's omega) and construct validity (exploratory factor analysis) for the measurement instrument. This was followed by a descriptive analysis, with measures in relation to central tendency and dispersion for each factor indicator, concluding a confirmatory factor analysis with maximum likelihood estimates to observe causality, covariation and incidence relationships. The results established that the relevant indicators are commitment and satisfaction, sickness and company flexibility. The strongest covariance is between internal and contextual factors. Consistent with the theory, it was possible to statistically validate the theoretical model applied and a tool to measure job turnover.