This report analyzes the impact of a training program by examining its effect on the real annual earnings and its likeliness to earn a non-zero wage. To do so, we use linear and logistic regression models, respectively. We use a dataset from the National Supported Work (NSW) Demonstration (1975-1978)12 for the same and build models using the AIC Forward criterion based on the statistical significance and reasonableness. We then find out that the variables treat(taking training), age, marital status, education, age given marital status, and age given training have significant associations with annual incomes, suggesting that taking the training might be a very important factor associated with an increase in wages. Also, we find that non-zero wages are associated with race, age, marital status, race given marital status, age given no degree and age given training suggesting that training is not a significant factor associated with non-zero wages by itself. Thus, it can be understood that the training program as a factor does seem to have a very strong association with an increase in wages. However, it is also found that it does not have much association with non-zero wages and is associated with other demographic factors and educational qualifications.
Data Source: https://ids-702-f20.github.io/Course-Website/project/team-project-01.html
Teammates: Maobin Guo, Pranav Manjunath, Xinyi Pan
Training and Wages