Suppose a researcher estimated the relationship between salary, gender, and age among a group consisting of male and female workers but ignored the fact that, on average, male workers have more work experience than females. The estimated regression of salary on gender and age is
i = 21354.83 + 239.45 i + 93.17 i
(15252.9) (95.6) (29.58)
Wherei = salary of a worker, i = 1 if the worker is male and 0 if the worker isfemale, and i = the employee’s age. Standard errors of the coefficients are in parentheses. When experience was included in the regression, the estimated regression is
i = 21177.75 + 226.27 i + 89.73 i + 443.41 i
(116111.3) (1186.8) (34.64) (47.7)
Wherei = the years of work experience of the worker.
Comparing the two estimated regressions, does there appear to be salary discrimination by gender? Discuss the implications of omitting the experience variable in the first regression.