3.4.4. Multidimensional Logistic Regression: Model 4 (E-PS)

The multinomial logistic regression analyses identified five predictors that explain cluster membership: the organization's sector, gender, type of contract, age, and occupational category (see Table 6). The results show that the model has a good fit (−2 log LR = 358.37, X2 = 60.39, df = 16, *p* ≤ 0.001). The probability of having high wellbeing and high performance is greater in the production sector, women, people between 35-50 years old, people with a temporary contract, and managers. The probability of having high wellbeing and low performance is greater in the production sector, women, people under 35 years old, with a permanent contract, and who do unqualified manual work. Finally, the probability of having low wellbeing and low performance is greater for the services sector, men, people over 50 years old, with a permanent contract, and who do unqualified manual work.

Comparing Clusters 1 (high levels) and 3 (low levels), the production sector, women, people with a temporary contract, between 35–50 years old, and managers are more likely to be in Cluster 1, whereas the services sector, men, people with a permanent contract, over 50 years old, and who do unqualified manual work are more likely to be in Cluster 3.
