3.4.3. Multidimensional Logistic Regression: Model 3 (H-PS)

The multinomial logistic regression analyses identified two predictors that explain cluster membership: type of contract and occupational category (see Table 5). The results show that the model has a good fit (−2 log LR = 68.14, X<sup>2</sup> = 38.70, df = 8, *<sup>p</sup>* ≤ 0.001). The probability of having high wellbeing and high performance is greater for people with a temporary contract and for highly qualified professionals or managers. The probability of having high wellbeing and low performance is greater for people with a permanent contract and people who do unqualified manual work. Finally, the probability of having low wellbeing and low performance is greater for people with a temporary contract and for unqualified manual workers or technicians/administrative workers.

Comparing Clusters 1 (high levels) and 3 (low levels), highly qualified professionals or managers are more likely to be in Cluster 1, whereas unqualified manual workers or technicians/administrative workers are more likely to be in Cluster 3.
