Non parametric tests Flashcards
Shapiro wilk test
a measure of how well the
sample quantiles fit the standard normal quantiles.
The bigger the
statistic,
the more likely the model is not correct.
Remedies for when criterias for parametric test fail
subject the data to normality checks as described, then transforming
the data if necessary.
choose one of the non-parametric tests.
legitimate normalization of data.
Instead of actual scores, a TRANSFORMATION of these scores can be used for
parametric tests.
Criteria for paarmetric tests
These criteria include satisfying the assumptions of outliers, linearity, normality,
homoscedasticity (i.e., homogeneity of variances in different groups being compared), sphericity
(i.e., equality of variances of the differences between all combinations of related groups) to name a
few.
Parametric model
When the data is generated from process
(model) that is known except for finite
number of unknown parameters
The sign test
test for the central location of a population.
basic non-parametric test that can be applied when the conditions for the
single sample t-test are not met.
perform a binomial test (or
use the normal distribution approximation when the sample is sufficiently large) on the
signs of the data elements
Mann Whitney U test
non-parametric test for unrelated samples of scores.