Binomial sign test Flashcards
When is this test used?
Nominal level data
Repeated measures/ matched participants design
How to remember what test to use?
Nick Reads books
Nominal/repeated measures/ binomial sign test
Step 1
Indicate the direction of difference from control to experimental condition with a + (if increase) or - (if decrease)
What happens if the participant is unchanged from control to experimental condition?
Ignore their score completely, don’t give them a sign
Calculated value of X
The number of times the least frequent sign (+ or -) shows up
Critical value of X
Taken from a table
N = number of participants who showed a difference (consistent ps discared)
Two tailed hypothesis at p<0.05
If calculated value is larger than critical value
the results are NOT significant at this significance level
So the probability the results are due to chance factors is higher than 5%
If critical value is smaller than calculated value
The results are significantly different
So the probability the results are due to chance is smaller than 5%
Accept the alternative hypotheses if
The critical value is smaller than the calculated value
Rhyme to remember significance
Chi is high and so is Rho
All the rest are low= calculated value must be lower than critical value in this case to be significant