Clearly you’ve never worked with data. Data is always tainted in some form and requires some cleaning or normalizing. Bias has to be taken into account and stripped away as best it can.
I agree with you. It is not the best source of data.
I would not use it to base a conclusion upon.
However, if someone is interested in the subject and inspired by it, forms a hypothesis as Scott Albers did, the NEXT STEP is to then do some cross-comparison.
a) Here’s my hypothesis. But, it was based upon questionable data.
b) I must seek out other sources to build with and use the inspiration for my hypothesis not as CORE data (because it does not fit the criteria for core data) but rather as SUPPLEMENTAL DATA.
I’d never stop someone from investigating a hypothesis but I might encourage someone to continue if I feel they could use additional materials than what they initially presented.
What I *personally* think about it doesn’t matter. But I’d never stop someone investigating a curiousity further.