A researcher is selecting a statistical test to analyse performance scores on a standardized cognitive task. To justify using a parametric test (such as a ttt-test) rather than a non-parametric alternative, which combination of criteria must be satisfied?
The data must be at least ordinal level, the population from which samples are drawn must be normally distributed, and there must be homogeneity of variance.
The data must be nominal level, there must be a related design, and the population distribution must be skewed.
The data must be interval or ratio level, the population from which samples are drawn must be normally distributed, and there must be homogeneity of variance.
The data must be interval or ratio level, the population distribution can be of any shape, and there must be a large difference in variance between conditions.
644 exam-style questions on OCR A Level Psychology Research methods and techniques. Each one has a worked solution and a mark scheme showing where the marks go.