Poll Results: R has a big lead, but Python is gaining
KDnuggets Poll results show that R has a big lead among data scientists and data miners, but Python is slowly gaining.
By Gregory Piatetsky, Dec 18, 2013. comments
Latest KDnuggets Poll asked data scientists and analytics professionals:
The results, based on 562 voters, indicate that R has a solid lead, and was used by about 77% of the voters. Python was used by about 32% of voters, and 24% used other languages. Note: some people used multiple languages, so the percentages add up to more than 100%.
However, we note that more R users (26%) are moving to Python than vice versa, and fewer R users (66%) are staying with it than Python users (79%). On the other hand, users of other tools are moving more to R than Python.
The poll gave options to keep using R/Python and switching between R/Python/Other, and I summarized the results in the following chart.
The sizes of the circles for R, Python, and "Other" correspond to the number of people using these tools, and sizes of the arrows are approximately proportional to the number of people who voted for that transition. Percentages on the arrows are relative to the base - e.g. 26% of those who used R said they switched to Python.
I realize that I am not engineer Minard and in process of doing this chart I gained even more appreciation for his map of Napoleon Retreat from Moscow, but my chart conveys the polls results faithfully.
Here is the table with the full poll results.
|If used either R or Python for data analysis/data mining in 2013, did you switch, for some tasks [562 voters total]
|Kept using R (286)||50.7%|
|From R to Python (115)||20.5%|
|From R to other tools (not Python) (33)||5.9%|
|Kept using Python (143)||25.4%|
|From Python to R (32)||5.7%|
|From Python to other tools (not R) (7)||1.2%|
|From other tools (not Python) to R (59)||10.5%|
|From other tools (not R) to Python (30)||5.3%|
|Did not use Python in 2013 (75)||13.3%|
|Did not use R in 2013 (68)||12.1%|
This poll was prompted by a recent blog Python Displacing R As The Programming Language For Data Science which has generated a lot of discussion and pushback, including
- Is Python really supplanting R for data work?, by Sharon Machlis
- Relative usage of R vs. python by Hadley Wickham