Research | StackOverflow Salary Survey: Have a computer degree = high salary? It doesn't exist.

Anyone but God must use the data to speak. ”

I'm working on an article that discusses the correlation between professionalism and pay in the software industry. And I've got a guess as to why professionalism affects pay, how it affects pay, and how professionalism produces preferences. Next, the only thing I need to do is to use data statistics to demonstrate.

First, I got a 2017-year stack Overflow survey data (https://insights.stackoverflow.com/survey/2017) to clean it, It then inserts a number of variables that were previously studied that could affect the final compensation.

One of the toughest problems in data analysis is to understand which variables need to be tested and which variables are not tested and which are control variables. For example, I could test whether "using PHP at Work" would increase the salary, but if I did not take into account the respondents ' country, then we might not be able to prove the extent to which PHP has been affected by the salary.

When I test a professional branch, I need to simplify the dataset and increase my sensitivity to the changes in the professional branch. So I chose to use only the data from professional Web developers from the United States, and then compare the salaries of professional front-end and back-end developers with the entire stack developer, and analyze the impact of professionalism on pay.

During the analysis, I incrementally added the following variables: experience, education, Web Developer type. However, the analysis results are not satisfactory. I thought that having a formal education would have a positive effect on wages, but something unexpected happened.

I am a web developer, responsible for the development of the whole stack. I've been thinking that if I could be more professional and get a degree in computer Science (computerscience, hereinafter called CS) instead of a liberal arts degree, I would probably make more money. My brother has a CS degree, and my salary is at two different levels, and I can only do the same. So I told everyone I was interested in programming to say: CS Professional undergraduate degree is very valuable, have it, you can at least with other people generate 20,000 of dollars permanent salary gap, and this is my conservative estimate.

So you can imagine how shocked I was when I compared CS with other majors. The result is that there is no significant difference between the salaries of graduates from different professions.

I think, "My analysis is certainly wrong." In my first analysis, I mixed CS graduate engineers, mathematicians and information technology graduates. Clearly, CS professional needs to be subdivided into different categories.

Then I had another round of analysis.

But the result is still: there is no significant difference between different majors.

Read the original