Statistical Modeling, Causal Inference, and Social Science complains about a newspaper article claiming to be based on “computer analysis”:
A computer analysis? Regression analysis, even statistical or economic analysis, would give at least some vague notion of what was done, but the term computer analysis is about as uninformative as saying that the analysis was done inside an office building.
The story in question was probably an example of what is called “computer-assisted reporting” In journalism-speak, this means an investigative reporter with at least some basic statistical knowledge performed some analyses on a file of raw data obtained from some source.
CAR Report, a blog for practitioners of this sort of journalism, acknowledged the post, but missed what I think is the essential criticism.
Why are journalists still using this archaic term? It emerged in the early 1980s when doing some basic calculations on a spreadsheet or searching for something on the Internet was still a specialist skill for the solitary newsroom technology enthusiast.
It would be better to stop bragging about using a computer because it seems kind of silly in an era when almost all reporting is somehow “computer assisted”. Instead, the handful of good reporters doing this sort of work should be explaining what first-hand statistical analysis they are doing to second-guess official presentations of data. The fact that they used a computer to do that analysis should be a given.