Cluster analysis and the classification tree technique are applied to
investigate the relationship between the individual characteristics of Italian temporary
help agency (THA) workers and their probability of achieving a temporary
job. The application aims to show some advantages of these techniques with
respect to traditional econometric tools. Sketches of the most common profiles
among Italian THA workers are obtained as a result. Besides the typical THA
worker pointed out by previous studies (young male workers, with a medium–high
level of education, living in the Northern regions), two new profiles have been identified:
the first comprising male manual workers with previous job experience,
whose average age is over 30 and whose educational level is low; the second comprising
young female workers with a medium–high level of education, working in
the service sector or in the public sector. The results are compared with the more
usual logit analysis and show their robustness.
