Future of the Software Industry
For this analysis, we shall be using Python(Jupyter Notebook) and the Stack Overflow Developers Survey Result 2019 data set. You can get the dataset right (HERE).
Now that the greetings are out of the way, let's start with our analysis without further ado.
The questions that we ask this dataset can be varied for it is a feature-rich dataset with the information collected from nearly 90000 developers across the world. Let us get to know the current trends of the industry to predict the possibilities the future has.
Importing the basics necessities:
We have already imported the libraries we are going to use and the .csv file containing the data. For starters, we see that there are many features in this dataset. Let's get a more detailed look at the features present so that we can get rid of the ones we don't require.
We can get rid of the columns we are not going to use for our analysis.
NOTE: This is my personal choice, YOU as an analyst can definitely work with completely different features to reach a similar or maybe even different kind of conclusion.
Now that our data is clean of all the noise we can finally focus on the features we are interested to work with. Below we have the features left in the data object.
Lets start our analysis finally, i have already completed mine so i will show you step by step how i reached my conclusion.
EXPLORATORY DATA ANALYSIS
NOTE: This is very significant to understand the market trend. 74% of our devs are working in the industry and only 0.41% of our devs have retired. We get a fair idea about the age group we are dealing with. These numbers won't change much even in the next year if all conditions are normal.
Note: We see that the stereotype of the software industry being dominated by men being more apparent through this distribution. 91% of our devs are men that bring light onto the future trends regarding gender distribution of the industry.
Note: A majority of our developers have worked with HTML/CSS; JAVASCRIPT; PHP etc these kinds of languages. This shows that a large no. of our devs are working or have worked in the web dev sector. Be it for hobby or industry purposes, web development has been flourishing pretty well.
Note: We observe a shift in the trends of preference of programming language when asking our devs what languages they wanna pursue next year. We notice that the industry is ever-changing with Python being the hot favorite to learn/work with by our devs. We also notice the emergence of a new language that was nowhere in the top 10 in the past. The programing Language GO.
Note: We notice that people in this industry mainly our devs don't necessarily prefer using Instagram and its likes. They tend to either avoid social media or use more informative and useful ones like Reddit, Youtube, and Whatsapp.
Note: We notice the trend that the vast majority of our developers have been coding for 5 and more years. With some being very experiences with 15–20 years under their belt. The software industry experts and enthusiasts are the ones using Stackoverflow and providing the survey.
EASE OF OUR SURVEY
Note: We notice that the majority of the survey takers amidst almost 90000 candidates have considered the 20 mins long survey moderately easy. Therefore Stack Overflow can use the same survey next year with a bunch of tweaks here and there if need be to get details from our devs to ascertain the change in trends.
CONCLUSION
We can safely conclude that the Stack Overflow Survey data is a good dataset with lots of features to work on but knowingly we choose these features to keep our analysis simple and compact. We got a fair idea about how the survey ease is fine and it's okay to keep the same survey in the future with some tweaks. About how Python is in growing demand and the hottest language of 2020. About how the no. of developers who haven't completed higher education are ever-increasing (BOTH FOR HOBBY AND PROFESSIONAL AREAS). About how many of our surveyors are Web Developers and about how Geographical Location affects the likelihood of finding more developers among many other things. I’ll write a proper Medium Article later on regarding my Analysis and link it. Hope you liked my analysis. Thanks a lot.