2 thoughts on “Do equality and high average school performance really go hand in hand?”

  1. Posted 30/01/2014 at 13:12 | Permalink

    Erm… so are any of the results statistically significant?

    Also, you selectively provide t-stat, coefficients, and SE for some data sets and not for others. I’m sure you did this innocently but it looks suspicious.

    P-values and coefficients for all these data sets would make your argument stronger (or would it?!). As it is, these look like damn lies and t-statistics.

  2. Posted 31/01/2014 at 15:57 | Permalink

    The reason not all graphs provide the t-statistics is just that Stata doesn’t automatically do that for two-way graphs, in contrast to added-variable plots. But I note in the text whether the relationships are significant for Graphs 1 and 5. Didn’t spell it out for Graph 4, but I stated that there is ‘a strong positive relationship’, by which I of course meant that the relationship is significant.

    But here you go again: Graph 1: statistically significant at 10% level with t-stat of 1.92 (as I state in the text – ‘weakly significant’). Graph 2: you can see the t-values underneath (significant at 1% level). Graph 3: same as Graph 2. Graph 4: significant (at 1% level – t-stat: 6.7). Graph 5: not significant (as I state in the text).

    Again, the point about the post was not to provide evidence of causal relationships, but to question whether equality is related to achievement. It’s not, which would have been the case even if none of the results had been significant.

Comments are closed.