Skip to main content

Table 4 Fixed-effects ordered logit models for self reported health status by gender

From: Employment status and perceived health condition: longitudinal data from Italy

  Male Female
  M1 M2 M1 M2
  coeff. (s.e) coeff. (s.e) coeff. (s.e) coeff. (s.e)
Ref: Permanent     
Temporary -0.288 (0.145) -0.331 (0.234) 0.080 (0.163) 0.159 (0.247)
First-job seekers -0.801 (0.208) -0.974 (0.278) -0.052(0.220) 0.066 (0.314)
Unemployed -0.423 (0.212) -0.371 (0.326) -0.160 (0.245) -0.456 (0.385)
2008 -0.001 (0.083) -0.041 (0.092) -0.013 (0.065) -0.018 (0.071)
2010 0.140 (0.165) 0.171 (0.183) -0.040 (0.128) -0.026 (0.134)
Temp. × 2008   0.186 (0.265)   -0.119 (0.278)
Temp. × 2010   -0.114 (0.299)   -0.120 (0.291)
FJ seek. × 2008   0.586 (0.289)   -0.316 (0.311)
FJ seek. × 2010   -0.071 (0.330)   0.011 (0.371)
Unem. × 2008   0.041 (0.342)   0.911 (0.425)
Unem. × 2010   -0.115 (0.352)   -0.159 (0.457)
Hausman test Rej. Rej. Rej. Rej.
Log-lik -1,985.44 -1,981.11 -2,005.71 -1,999.54
Observations 18,603 18,603 19,179 19,179
  1. Coefficients statistically significant at the 10% level are in bold. Model specification also include the category for individuals out of the labor force, family income, family income square, family wealth, family wealth square, regional unemployment rate, age square. The Hausman test rejects the null hypothesis at the 5% level. Estimation results of the RE model are not presented here for brevity and are available upon request.