Abstract:
|
This talk is based on "Data Science: A Three Ring Circus or a Big Tent?." (JCGS 26(4) pp. 784–785), a comment that accompanied the publication of Donoho's "50 Years of Data Science" (JCGS 26(4) pp. 745-766). The comment is co-authored with Hadley Wickham and was one of the top 5 JCGS downloads for the year. First, we look at the possible connection between the historically narrow boundaries of academic statistics and the uneven representation of Statistics departments in the leadership of data science initiatives. Second, we propose data- and software-oriented skills that should have a place in modern statistical training. Finally, we elaborate on how software can be a valid medium for intellectual work, as it is the most natural language in which to express the abstractions of applied statistics.
|