Regression analysis of recalculated total salaries - All full-time employees Both males and women Males Women Slope Standard deviation Significance Slope |Standard dev. Significan Slope Standard dev. Significance Constant 12.330 0.03 12.268 0.07 . .. 12.311 0.04 Gender Male=0 Woman=1 -0.037 0.0: - Job title 2) Directors 0.440 0,01 0.430 0.02 0.445 0.01 Experts 0.210 0.00 0.162 0.01 . .. 0.226 Tech and specialized workers 0.01 0.004 0.01 not significant 0.040 0.02 -0.012 0.01 Office workers not significant 0.160 0.01 -0.158 0.06 -0.155 Service- sales- and retail workers 0.01 -0.244 0.01 0.224 0.02 . . . -0.246 0.01 Industrial and unspecialized workers Age Age in years 0.017 0.00 0.020 0.00 Age 0.016 0.00 0.000 0.00 0.000 1.00 0.000 0.00 Period working Period working in years 0.001 0.00 0.001 0.00 0.001 Period working 0.0 0.000 0.00 . . . 0.000 0.00 . . . 0.000 0.00 Influence of work hours Overtime hours 0.009 0.00 0.008 0.00 0.010 0.00 Work hours worked outside if normal daily working hours 0.002 0.00 0.001 0.00 0.002 0.00 -.. R (Adjusted R-) 0.801 (0.800) 0.766 (0.764) 0.818 Quantity/Number 0.817) 3740 926 2814 Footnotes: There is a significant difference among groups * P = 0.05 ** P S 0.01 *** p 5 0.001 Difference between the salary of men and women are e-0.064 - 1 = -0.036 : Women have 3.6% lower salaries than men "Effect coding were used to examine the impact of the job group. For the job group 5 variables are made. One for each group other than unspecialized workers et al. Variable for directors is coded. Directors are given the value 1. Unspecialized workers are given the value -1 and other groups are given the value 0 The variable for experts are given 1 for those who work as experts, unspecialized workers are given the value -1 and others are given the value 0 Slope coefficients can be interpretet as a deviation from unweighted average of job groups Constant 0,440 indicates that directors had a 55.3%% higher owerall salary than unweighted average of all job groups (20.44-1 = -0.553 ) Both constants and variable overtime working hours are included. Working hours out of regulare working hours are estimated from strain payments