StackOverflow Developer Survey

Analysing Data

Gleaning info from SO surveys

StackOverflow conducts a survey every year, polling its users on topics ranging from education, programming languages to preferences on pronouncing ‘GIF’.

The survey offers a treasure trove of information that can show trends in career paths and skills. This is especially useful for firms in the business of talent acquisition and online education, where they can help answer questions like :

We shall analyse the survey results to see if we can glean useful information

The zip file consists of the schema csv that states the survey questions asked and the survey results csv in wide format.

The schema shows that the responses are identified using Respondent as the primary key and there were 153 questions in the survey. The questions can be broadly classified into the following groups:

The responses ranging from Not at all important to Very important

The responses ranging from Not at all important to Very important

The responses ranging from Not at all satisfied to Very satisfied

The responses ranging from No influence at all to I am the final decision maker

Some of the survey questions allowed for more than one choice. These are represented as ; delimited strings.

~51,000 people took part in the survey.

The respondents mainly hail from US, India and Western Europe(GB, DE)

Majority of the respondents are professional developers, employed full time. We see a full spectrum in terms of years of coding experience, though most have less than 5 years experience in coding.

Most of the respondents are either open for new opportunities or actively searching for new jobs.

Interestingly, people actively looking for job seem to be spending the same time as those passively open to opportunities.

Generally, people feel that they are underpaid in this sector. It is worth exploring more in detail (perhaps together with other surveys and stats) to establish this. This skew does not exist for those who are happy where they are.

More than 50% said they do not maintain any form of CV or resume online. This holds true even for those who are actively looking for a job.

In the survey respondents were asked to choose the programming languages they preferred to work on and were allowed more than one point. Here we pool the results to detect correlations.

Comparing the languages they are currently using and those they wish to use, we see that Python has the potential to become the next big thing.

The analysis is a snapshot of one year, and is meant to showcase what can be done using the publicly available data. We can summarize that:

If we can analyse the complete survey series it may offer a much richer picture of the patterns .

Related posts:

Microsoft Windows 10 Pro 64 is almost as cool as Nintendo 64…. Microsoft FQC-08930. Delivery Method: DVD; Operating System Windows 10 Pro (64-Bit) Specs. In the Box: (1) Microsoft FQC-08930. Delivery…

When we use our computers to do something, it has become an accepted practice to analyze the results, then decide what’s working and what’s not. However, with deep learning, researchers are able to…

They were rounded up, arrested, and the pictures of them in skirts are how we remember them. They are proud as they are dragged away by police. They show up in full leather to parade for the right to…