You need to know about H_{0}/H_{1} testing, if only because it is so widespread (and anyway it is the basis of the Edexcel exam). You should also have some awareness that many statisticians consider the whole “null hypothesis” approach unsound. For a simple explanation, click here.

For example, if you toss a coin and stop tossing when you find a “statistically significant” deviation of the number of heads from 50%, then you are *more likely than not* eventually to find a “statistically significant” deviation *even if* the coin is fair. This result is of major practical importance in marketing, where tests of whether one approach is better than another *usually* give “false positives”: see here.

### Like this:

Like Loading...

*Related*