statistical tests Flashcards
levels of measurement
- nominal data
- ordinal data
- interval data
nominal data
- data consists of numbers of parts
- that might fall into different categories
- a person can be placed in one category only + not the other
ordinal data
- data can be placed in rank order from lowest to highest
- data is concerned w/ the order that the data can be presented in
- ordinal scale can consist of measurements that are of unequal intervals
interval data
- data has fixed + even intervals
- units of data are fixed throughout the range
which two are the most common levels of measurement?
nominal + ordinal
example of nominal data
how many males + how many females are at a football match?
example of ordinal data
the time taken for every year 7 student to complete a 200m race - timings will be put into rank order
examples of interval data
- height of 20 year 8 girls in a PE class
- weight of patients who are suffering from a disorder
non-parametric tests
- chi-squared
- spearman’s rho
- mann whitney
- wilcoxon
parametric tests
- pearson’s r
- related t-test
- unrelated t-test
why are parametric tests better than non-parametric tests?
parametric tests are:
- more robust + powerful than non-parametric tests
- rely on actual data collected (rather tan just examining the rank order of the data)
- more likely to detect if the data is significant or not
3 factors = parametric test
- interval levels of measurement
- nominal distribution
- variance of data = data should have similar variance/ spread of scored
–> look at standard deviation + dispersion = see if they are similar
how do you decide which test to use?
- correlation, test of difference or association?
- research design = independent measures, repeated measures, matched parts
- level of measurement = nominal, ordinal, interval
tests for correlation
- spearman’s rho
- pearson’s r
tests for test of difference
- chi-squared
- mann whitney
- wilcoxon
- unrelated t-test
- related t-test
acronym for deciding which test to use
Carrots Should Come
Mashed With Swede
Under Roast Potatoes
chi-squared
nominal
test of difference
independent measures
non-parametric
OR
nominal
test of association/ correlation
non-parametric
sign test
nominal
test of difference
repeated measures/ matched parts
non-parametric
mann whitney u test
ordinal
test of difference
independent measures
non-parametric
wilcoxon
ordinal
test of difference
repeated measures/ matched parts
non-parametric
spearman’s rho
ordinal
test of association/ correlation
non-parametric
unrelated t-test
interval
test of difference
independent measures
parametric
related t-test
interval
test of difference
repeated measures/ matched pairs
parametric
pearson’s r
interval
test of association/ correlation
parametric
