Stack Overflow 2017 survey results: developers who use spaces earn more
This is from a certain point of view “holy war” among software developers, this topic has become the subject of many debates and jokes. I use spaces and never thought about the importance of this moment. But today we are publishing the initial data of the Stack Overflow 2017 developer survey and some analysts believe that this choice is more important than I expected.
Spaces make more money than tabs
28,657 respondents were interviewed who expressed their preference for tabs or spaces and considered themselves professional developers (students and former programmers were not counted). In this group, 40.7% use tabs and 41.8% spaces (17.5% use both methods). Of these, 12,426 left information about their salary.
Data analysis leads us to an interesting conclusion. Developers who use spaces for indentation make more money than those who use tabs, even if they have the same amount of experience:
')
The average developer who uses spaces has a salary of 59,140 dollars, while a developer using tabs has a salary of 43,750 dollars. (Note that all results were converted to US dollars from the respondent’s currency). The results of the developers who chose the option "Both", as a rule, are indistinguishable from those who chose "Taba".
This is a fun result, but of course it is not the final one. When I first discovered this, I assumed that the result could be different if sorted by country or programming language. For example, it is possible that developers in countries with low GDP per capita are more likely to use tabs and, therefore, developers usually have lower wages.
We decided to find out whether this effect takes place in each of the countries separately. Selected countries with the highest number of respondents participating in the survey:
This effect was observed to the least extent in Europe, and the largest gap was recorded in India, but, nevertheless, it is manifested in each country.
As another hypothesis, we know that developers from different fields often use different spaces (for example, DevOps developers are more likely to use spaces, and mobile developers are more likely to use tabs), often because they use different editors and languages. Participants in the survey also left information about which programming languages they use ( Python , Javascript , etc.) and in what area they work (web developer, embedded developer, etc.).
As you can see above, this effect is observed in each subgroup of developers. We obtained a similar result even when filtering developers by country or by a certain range of experience. Please note that respondents can choose several languages, so some of these groups overlap each other.
I did a number of other examinations of possible causes (such as education level or company size) and saw the same results: the gaps beat the tabs in each group. Now that the raw data is available, I urge you to check the possible factors yourself.
If we checked all the factors that we suspect could have a salary effect, what effect would the selection of tabs / spaces have in reality?
To answer this question, I approach linear regression, predicting salary based on the following factors:
The use of spaces instead of tabs leads to an increase in wages by 8.6% (confidence interval (6%, 10.4%), P-value <10 ^ -10). In other words, the use of spaces instead of tabs is equivalent to an additional 2.4 years of experience .
So ... this is definitely an amazing result that I did not expect to see. And it is impressively stable even when sorted into multiple segments. As an exercise, I tried to study the data on a variety of other factors missed in this article, but this effect was present everywhere.
Correlation is not a causal link, and we can never be sure that we have considered all the factors in the data set. If you are a statistician or analyst, I recommend that you download the initial survey data and study it yourself. Here you will find the code to play the analysis. In any case, we would be interested to hear your hypotheses.
Spaces make more money than tabs
28,657 respondents were interviewed who expressed their preference for tabs or spaces and considered themselves professional developers (students and former programmers were not counted). In this group, 40.7% use tabs and 41.8% spaces (17.5% use both methods). Of these, 12,426 left information about their salary.
Data analysis leads us to an interesting conclusion. Developers who use spaces for indentation make more money than those who use tabs, even if they have the same amount of experience:
')
The average developer who uses spaces has a salary of 59,140 dollars, while a developer using tabs has a salary of 43,750 dollars. (Note that all results were converted to US dollars from the respondent’s currency). The results of the developers who chose the option "Both", as a rule, are indistinguishable from those who chose "Taba".
This is a fun result, but of course it is not the final one. When I first discovered this, I assumed that the result could be different if sorted by country or programming language. For example, it is possible that developers in countries with low GDP per capita are more likely to use tabs and, therefore, developers usually have lower wages.
We decided to find out whether this effect takes place in each of the countries separately. Selected countries with the highest number of respondents participating in the survey:
This effect was observed to the least extent in Europe, and the largest gap was recorded in India, but, nevertheless, it is manifested in each country.
As another hypothesis, we know that developers from different fields often use different spaces (for example, DevOps developers are more likely to use spaces, and mobile developers are more likely to use tabs), often because they use different editors and languages. Participants in the survey also left information about which programming languages they use ( Python , Javascript , etc.) and in what area they work (web developer, embedded developer, etc.).
As you can see above, this effect is observed in each subgroup of developers. We obtained a similar result even when filtering developers by country or by a certain range of experience. Please note that respondents can choose several languages, so some of these groups overlap each other.
I did a number of other examinations of possible causes (such as education level or company size) and saw the same results: the gaps beat the tabs in each group. Now that the raw data is available, I urge you to check the possible factors yourself.
If we checked all the factors that we suspect could have a salary effect, what effect would the selection of tabs / spaces have in reality?
To answer this question, I approach linear regression, predicting salary based on the following factors:
- Tabs / Spaces
- A country
- Years of experience
- Developer Type and Programming Language
- Education Level (Bachelor, Master, Doctorate)
- Do they contribute to open source
- Will they program as a hobby
- Company size
The use of spaces instead of tabs leads to an increase in wages by 8.6% (confidence interval (6%, 10.4%), P-value <10 ^ -10). In other words, the use of spaces instead of tabs is equivalent to an additional 2.4 years of experience .
Conclusion
So ... this is definitely an amazing result that I did not expect to see. And it is impressively stable even when sorted into multiple segments. As an exercise, I tried to study the data on a variety of other factors missed in this article, but this effect was present everywhere.
Correlation is not a causal link, and we can never be sure that we have considered all the factors in the data set. If you are a statistician or analyst, I recommend that you download the initial survey data and study it yourself. Here you will find the code to play the analysis. In any case, we would be interested to hear your hypotheses.
Source: https://habr.com/ru/post/331026/
All Articles