Game Reviews Don't Matter, Study Finds

Next Generation has an article up looking at a report from SIG, on the correlation between game reviews and sales. Their findings indicate that, while reviews obviously do have some effect on games sold, there just isn't that much of a correlation. From the article: "He said he doubted that publishers and PRs would stop caring about review scores, especially as they matter a lot with consumers who compare games from the same sub-genre — say, basketball games. But he said that, as with last year's report, the report's findings are unlikely to be popular. 'We received a lot of attention but the stats do not lie,' he said."

Studies don't matter, game review finds

[ucahg]

n/t

Re:Reviews Prevent Me From Buying

[Anonymous Coward]

You might (want to) check and (make sure) your keybard (is working) correctly.

Re:They do not lie

[Jerf]

Properly-created statistics never lie. They just tell useless truths, and then even if you're competent to judge the value of a statistic, it's extremely rare to be told the full statistic.

(That is, "70% of murderers hate their mothers" is not useful unless you know the definition of "murderer", the definition of "hate", even the definition of "mother" (would a step-mother who raised the murderer since he was two be a "mother"?), and how they were sampled. Even in this simple example, you can see why statistics are rarely useful when reported in the news, and this is a useless, contrived example; the real world is far more complicated.)