Exercise

SE of the slope

The previous exercise generated 10 different slopes under a model of no (i.e., null) relationship between the explanatory and response variables. Now repeat the null slope calculations 1000 times to derive a null sampling distribution for the slope coefficient. The null sampling distribution will be used as a benchmark against which to compare the original data. Then, you'll calculate the mean and standard deviation of the null sampling distribution for the slope.

Instructions 1/3

undefined XP
    1
    2
    3

Simulate 500 slopes.

  • Use specify() to specify Foster vs. Biological (same formula as for a linear regression).
  • Use hypothesize(), to set a null hypothesis of "independence".
  • Use generate() to generate 500 replicates (reps) of type "permute".
  • Use calculate() to calculate the summary statistic "slope".