Your ability to evaluate outside information and theories for accuracy. For example, the question you answered earlier about the fictional BacoNation fast food sandwich checked your Evidence Evaluation. The question reported that the BacoNutcase ad campaign produced a $16 million increase in BacoNation sales, but the sales of the sandwich had fluctuated by nearly as much in the recent past without an ad campaign. Since the connection between the ad campaign and the sales increase was dubious, the question awarded full points if you answered that the ads were only somewaht likely to have driven the sales up.
That just seems wrong. This was the data:
March: $55.0 million
April: $43.8 million
May: $59.4 million
June: $49.6 million
July: $46.1 million
August: $54.9 million
September: $44.5 million
October: $60.5 million
The a priori chance of ads increasing sales is high. Bayesian updating increases the chance quite significantly since the result is higher than any one in the past 7 months (albeit quasi-tied with May). How is it not very likely that the increase is due to the ads? Simply saying ‘it fluctuated by this much before’ seems to be misunderstanding Bayesian probability.
Took the test -
That just seems wrong. This was the data:
March: $55.0 million
April: $43.8 million
May: $59.4 million
June: $49.6 million
July: $46.1 million
August: $54.9 million
September: $44.5 million
The a priori chance of ads increasing sales is high. Bayesian updating increases the chance quite significantly since the result is higher than any one in the past 7 months (albeit quasi-tied with May). How is it not very likely that the increase is due to the ads? Simply saying ‘it fluctuated by this much before’ seems to be misunderstanding Bayesian probability.
I don’t think there’s any reason to write this in the clear without rot13.