Thursday, December 20, 2018

Data Science: The End of Statistics? #NotYet #Statistics

I came across an interesting blog called Normal Deviate today. Though the author has ceased to post since 2013, it has an interesting post: Data Science: The End of Statistics? that set me thinking about Statistics. Despite its title, the author is not writing about data science bringing about the "End of Statistics". The author bemoans the fact that the "very fact that people can talk about data science without even realizing there is a field already devoted to the analysis of data — a field called statistics — is alarming". The author also complain that that so-called “data scientists” are ignoring biases, not computing standard errors, not stating and checking assumption and so on. He/she states there there seems to be more of an emphasis on computer science to process large quantities of data to the detriment of statistics. The point seems to be that statisticians need more computer science so that they can "do serious computing" and to be able to understand and use "programming languages".

As a Lecturer who teaches statistics in a School of Computing, I don't consider myself as either a statistician or a computer scientist. I am a Marine Biologist by qualification who over the years has picked a few skills in statistics and programming. Nevertheless, I try to do my best to both get students to learn about analysing data with statistics, and to get them excited about data. In one class a few months ago when finishing up on the topic of Analysis of Variance (ANOVA) one student asked out loud "Who makes up this shit?"! Good question - who needs statistics? Charles Wheelan in his book (Naked Statistics) gives us plenty of excellent reasons why we should study statistics:
  • To summarize huge quantities of data
  • To make better decisions
  • To answer important social questions
  • To recognise patterns that can refine how we do everything from selling diapers to catching criminals
  • To catch cheaters and professional criminals
  • To evaluate effectiveness of policies, programs, drugs, medical procedures, and other innovations
  • And to spot the scoundrels who use these very same powerful tools for nefarious ends

The End of Statistics? Not yet!

No comments:

Post a Comment