Abstract (eng)
The population on earth continues to grow steadily, but on average, it is getting older and older. Moreover, companies face, besides an older workforce, the problem of a shortage of skilled workers. Also, employees tend to switch jobs more frequently compared to the past. Therefore understanding employee turnover is of high interest to companies. Research on this topic has been going on for several decades, defining different types of turnover, such as voluntary turnover or avoidable turnover, theories on how it comes to turnover have been developed, and determinants that can be used to predict employee turnover have been investigated. However, start-ups were left out of this research. The present study wants to investigate whether already tested determinants in established companies show similar significances and correlations as in start-ups. Due to the special characteristics of start-ups, such as the enormous growth, further determinants, such as the funding volume and the number of patents and trademarks, are examined. For this purpose, two different data sets were used and merged. One dataset contained company information on start-ups founded since 2017 and the other public LinkedIn profiles. These datasets were processed and linked using Python, resulting in a data set with 1296 start-ups. This sample was analyzed using multiple regression analyses. The results of the analyses show that the number of patents and trademarks have no significant influence on the employee turnover of a startup (r = -.41 p > .1), whereas the funding volume has a significant influence (r = -.065, p < .01). A possible reason for the influence of the funding volume could be that it offers employees security in the sense of punctual payment. Nevertheless, only a small part of the variance in the variable employee turnover is explained by the model (r2 = .138). Finally, this study provides initial insights into the topic of employee turnover in start-ups and opens up many further research opportunities. Thus, the present model can be expanded by including additional determinants, which have already been proven to be good predictors for established companies, such as job satisfaction or organizational commitment.