Statistical Testing^ Flashcards
Statistical Testing
p < 0.05
findings have less than 5% difference
due to real difference
make result significant
Nominal (Sign Tests)
Test of Difference
- Ind. Groups
= chi-squared
- Repeated Matched
= sign test
Test of Correlation
= chi-squared
Ordinal (Sign Tests)
Test of Difference
- Ind. Groups
= mann-whitney
- Repeated Matched
= wilcoxon
Test of Correlation
= spearson’s rho
Interval (Sign Tests)
Test of Difference
- Ind. Groups
= unrelated t-test
- Repeated Matched
= related t-test
Test of Correlation
= pearson’ r
Nominal Info
separated categories
Ordinal Info
ranked data
Interval
precise data collected scientifically
What’s the point of statistical testing?
allows to accept or reject a hypothesis and accept the null hypothesis
to be confident the IV cause the change in the DV not chance
Conditions to use Sign Test
repeated measures (do both conditions)
looking for a difference
data must be nominal (separate categories)
How do we work out the S value?
see for each participant if there has been a negative or a positive change
tally the neg and pos
remove the no change participant/s
S = total of least frequent sign
What is S value?
the total of the least frequent sign
The Rule Of R
if the statistical test has an r in it, the calculated value has to be equal or more than the critical value to be significant
What do we need to know for the critical value?
Significance level desired (0.05 or 5%)
Number of participants (n value)
Whether the hypothesis is direction or not
The calculated value need to be equal or lower than the critical value to be significant
p value of 0.05
there is 5% probability your results are due to chance
