The backlash against Big Data is continuing apace. It was supposed to be the answer to almost everything on a large scale but a report in the Washington Post suggests that Stateside there’s some disillusionment in the education market. There are now calls for the implementation of “small data” again.
The crux of the article is that Big Data can establish what’s going on at a given time on a large scale but it can’t establish correlation. So (and this is our example rather than theirs) if students wearing black happen to perform particularly well on a particular day, the data analysis tools can’t say whether that dress choice alone has led to people achieving high marks or whether it’s just a coincidence. A human being probably could.
The suggestion is that Big Data alone can’t fix the education system, it needs a lot more. We tend to agree; good teaching can be instinctive as well as methodical, and “reading” the students in front of you is best supported by, rather than dictated to, by a data feed.