Application of a Logit Model to Classify the Unemployed in Terms of Their Chance of Employment

The aim of this article is to classify the unemployed into groups with similar odds ratio of entering employment. This classification was based on the logit model. Groups proposed in statistical studies according to age or education may be heterogeneous, having in mind the causes of de-registration from the office. Grouping according to the characteristics of the unemployed was performed, which differs from that used by labour offices and public statistics (Statistics Poland). The unemployed deregistered from the Poviat Labour Office in Szczecin (Poland) in 2017 (n = 19,931) were examined. Taking up a job was considered as an enabling event. The study used the rarely used coding of the explanatory variable: –1, 0, 1, which allows comparing the odds of the analysed subgroups with the odds of the group average. In this way, subgroups of unemployed people with similar chances of taking up a job were distinguished.