Today we carried out our first statistical test, Spearman's rank (or rho). We discussed probability, significance and type 1 and 2 errors, and why psychology generally uses a significance level of 0.05 or below. Spearman's rho is the test to use when the hypothesis predicts a correlation between two sets of data and when the data is ordinal or interval (definitely not nominal). You worked through two examples and the worksheets are here and here and all instructions for carrying out the test are on the first sheet.
The ppt on probability and significance is here (slides 9 onwards).
The ppt on probability and significance is here (slides 9 onwards).
We also concluded that there is no correlation between the expense and tastiness of milk chocolate (though I should point out that Spearman's rho becomes increasingly unusable with small sample sizes like these).
No comments:
Post a Comment