ld111134 wrote:The other issue with the Tribune's reviews is that virtually every restaurant reviewed receives a 2- or 3-star rating, which reduces the utility of the ratings (like children in Lake Wobegone who are all "above average"). Contrast this with reviews in Time Out, where a good number of establishments are rated 1- or 2-stars out of 5.
I haven't studied the subject in detail, but what you describe doesn't sound out of order to me. On the Tribune's scale of 1 to 4, assuming a normal distribution (well, to be precise, approximately normal, since the scale has upper and lower bounds of 1 and 4), 2.5 would be the mean, and if one can't award 2.5 stars, wouldn't most of the places rated get 2 or 3 stars? This doesn't seem to me to be a Lake Wobegonish tendency for all of the children to be above average. Rather, it seems to me to be most of the numbers clustered around the average.
Simililarly, on Time Out's scale of 1 to 5, the mean would be 3. If Time Out's average rating is not 3, but rather less than 3, it means that Time Out reports on more lower-quality places than does the Tribune. Comparing the two scales isn't very meaningful (to me, at least), any more than it would be meaningful to say something weighs less when measured in kilos than when measured in pounds. More meaningful would be an analysis of all of the Tribune's ratings, on its 1 to 4 scale, and all of Time Out's ratings, on its 1 to 5 scale, to see if both approximate a normal distribution, within the defined boundaries of each (1 to 4 for the Tribune, 1 to 5 for Time Out).
"Your swimming suit matches your eyes, you hold your nose before diving, loving you has made me bananas!"