4 ms·
Off-hand, I would say you can't use chi-square unless you rolled at least 360 pairs of dice. Chi-square assumes a normal distribution and the rule-of-thumb you
by baking 13y ago
Off-hand, I would say you can't use chi-square unless you rolled at least 360 pairs of dice. Chi-square assumes a normal distribution and the rule-of-thumb you need at least ten hits and ten misses before the sample distribution approaches normal. Source: just helped my daughter pass her AP Stat mid-term.
- keithpeter 13y agoWe are getting there (250 rolls now), I'll put the extra throws up. Your finding may come as a surprise to many geneticists however...
- ronaldx 13y agoIt's common practice that the 'expected' values of a chi-squared test should be 5 or higher, with adjacent groups to be merged until this is the case (at UK A-level this is strictly required). This is a slightly conservative correction and Wikipedia agrees that no correction is required in this case: https://en.wikipedia.org/wiki/Pearson%27s_chi-squared_test#Problems https://en.wikipedia.org/wiki/Pearson%27s_chi-squared_test#P...