The big news from the SAS world this summer was the release, on May 28, of the SAS University Edition, which brings the effective price for a single user edition of SAS down from around $10,000 to $0. It does most of the things that statistical analysts need, is not a student edition, does not require academic status, and runs locally using a browser interface (SAS Studio). Clearly SAS is making a bid for the mindshare of the analytic community hooked on open-source software such as R.
Gregory Piatetsky-Shapiro, publisher of the KD-Nuggets data mining newsletter, published a poll yesterday (8/20) that shows a dramatic shift in program preference from 2013 among the data science community. In both 2013 and 2104 his poll asked “What programming/statistics languages you used for an analytics / data mining / data science work in 2014?” SAS has substantially increased its share of the responses (respondents can select more than one response):
SAS: From 21% (2013) to 36% (2014)
SQL: From 37% to 31%
Python: From 39% to 35%
R: From 61% to 49%
SAS users tend to be a more self-contained bunch — 58% said they used only SAS (compared to 20% for R, 14% for Python, and 5% for SQL). This is consistent with a recent poll from a Strata conference, which found two types of users – those who stuck mainly to one environment (usually SAS or Excel), and those who used multiple languages (R, Python, etc.).
Now that SAS is much easier to get, Statistics.com has added a course in SAS programming to its Programming for Data Science certificate program.