Sign Test Flashcards
Sign test
determines whether the number of positive and negative values in a sample is likely to be due to chance
How do you evaluate the distribution’s tail?
compare alpha to the probability of obtaining any value as extreme or more extreme than the one obtained in the data
- use binomial distribution table to find likelihoods of the value obtained and more extreme (for two tailed evaluation, both tails of the distribution are added)
- compare p-value to alpha value α
One-tailed evaluation
for a directional hypothesis, only one tail of the distribution is evaluated
Two-tailed evaluation
for a non-directional hypothesis, both tails of the distribution are evaluated
Sign test step 1
ensure that conditions are met for using binomial distribution
Sign test step 2
formulate clear, specific null and alternate hypotheses
Sign test step 3
set alpha value
Sign test step 4
gather data and calculate number of pluses and minuses obtained
Sign test step 5
determine the probability of getting an outcome this extreme or any more extreme
- extreme: favors alternate hypothesis
- non-directional H1: do two-tailed evaluation
- directional H1: do a one-tailed evaluation
Sign test step 6
If the probability calculated in Step 5 is:
≤ α, reject H0 and accept H1
> α, retain (or “fail to reject”) H0
Sign test step 7
Write out a clear, specific conclusion that is consistent with the wording of your original hypotheses
eg, “Reject H0 and conclude that a romantic partner’s scent affects sleep efficiency.”