The results of our 2018 SAS, R, or Python survey have been released! Click here to see our latest analysis.
As some of you may know, last month we conducted a flash survey of our network of analytics professionals and data scientists to try to quantify their preference for either SAS or R. After we released the initial results, as well as a quick run-down of how emotive respondents were in their responses and some of the humorous comments that we received, we decided to dig in a bit deeper and see if we could spot any changes in SAS vs. R preferences compared to our survey results from last year.
This deeper dive analysis for our 2015 sample shows some interesting trends:
Comparing 2015’s results to 2014’s, there is an increase in the preference of R in all regions. R is now favored over SAS in both the West Coast (like last year) and the Northeast, while SAS still holds the majority of votes in the Mountain, Midwest, and Southeast regions.
Last year, when assessing preferences by industry, R only held the majority in Tech/Telecom/Gaming companies. This year, R is favored in Tech/Telecom again, as well as Academia, Consulting, Healthcare/Pharmaceuticals, and Marketing Services companies. SAS and R are preferred equally in Government and Corporate/Other. SAS continues to have a strong presence among those working in Financial Services and Retail/Consumer Packaged Goods companies.
To see 2014 SAS vs. R preferences by industry please see our deeper dive analysis from last year.
It seems that SAS is still the preferred tool for many professionals who hold Bachelor’s or Master’s degrees, but those with PhD’s favored R over SAS. By comparison, last year, R was not quite even with SAS for PhD respondents.
Similar to last year’s results, respondents with 5 or less years of experience were the most likely to favor R. However, the percentage of respondents that preferred R increased for both 6-15 years of experience and 16+ years of experience as well. For example, last year 29.7% of those with 6-15 years’ experience preferred R, vs. 40.2% this year.
This year, we also separated data scientists from other predictive analytics professionals (PAPs). Although they accounted for only 13% of the sample, data scientists overwhelmingly favor R over SAS, while overall, other predictive analytics professionals prefer SAS.
The conversation (or debate) around SAS vs. R has always been lively, but has become especially spirited over the past few years with the increased attention to Big Data, data science, and the increase in the number of tools that data experts have to manage and analyze data. It will be very interesting to see how these trends evolve over the next year, and I look forward to reading all of your comments and insights in the comments below.
Thanks to everyone who participated in our flash survey, and keep your eyes on your inbox for more opportunities to participate!
Thanks for this, Linda!
Interesting. And it makes total sense. Most of the people coming into data science have free access to R, while SAS costs thousands. Even if an individual uses SAS in their PHDs or Master programs, once they are out, if their place of employment doesn’t pay for SAS, R becomes the platform of choice. I think R will become the de facto standard at some point.
Ironically, the charts in the article seem to be built in MS Excel. It would be interesting to know if Linda used SAS or R for the underlying analysis? 🙂
I’ve found that our analysts use both and have a preference based on the situation or desired output. As a result, many may use both for an analysis!
I totally agree with this insight. Both R and SAS has its strengths and weakness. However from my short experience in financial services industry( I fall in that 0-5 year experience) SAS is heavily used for tasks that require heavy data extraction and crunching since it communicate very well with most RDBMS. Once the data is ready for analysis, R does much better job due to its excellent graphics and statistical packages. I use both tools on daily basis
The results make a lot of sense .
In Financial Services SAS makes a lot of sense due to extensive financial tools libraries .In data science and esp big data analytic R is always the tool of choice due to a number of algorithmic libraries and ready to access database access functionalities.
A survey based on Burtch Works’ own “network of analytics professionals and data scientists”?? Unscientific survey means worthless results.
The charts suggest “Momentum” of SAS and the surge up the S curve for R.
SAS is preferred by the longer established people and organisations – years of experience is particularly marked. The shift to R in the last year shows the advantage of a lower barrier to entry. SAS I often see in magazine articles and at exhibitions; R I heard about a year ago, installed 6 months ago at a 90 minute seminar and am now training up in.
This does not appear to be a valid sample(i.e. never heard of Linda’s firm, randomly drawn from a known population, etc.). Otherwise this would be interesting and potentially useful.
From a cost point of view, does the cost of an additional R resource outperform the cost of a SAS license for an existing group of analysts? Short term projects might have the benefit of an additional SAS license but for long term I would say R is the way to go 🙂