Statistical
Download as PPT, PDF0 likes268 views
1. The document discusses the assumptions of parametric and non-parametric statistics when analyzing experimental data. 2. For parametric tests, the key assumptions are random selection of subjects, normally distributed data, and equal variances between groups. 3. Non-parametric tests only require random selection and equal variances if measuring on an ordinal scale or higher.
Statistical
- 1. STATISTICAL TREATMENT OF THE DATA 1. Differences between means of groups 2. Testing nul hypothese 3. Significance levels
- 2. The proper use of parametric statistics includes not only measurements but also a variety of assumptios regarding certain characteristics of both the collected data and the population from which the sample has been drawn. Parametric and non parametric statistics
- 3. Assumption of parametric test Three general assumptions 1. The selection of subjects from the population was random and independent, I.e every subject in the sample has an equal chance of being chosen from the populatin and the selection of one subject in no way influenced the sampling of any other subject.
- 4. Assumption of parametric test Three general assumptions 2. The observations were drawn from noramally distributed populations. The experimental data, when plotted in terms of their frequency of occurrence, shoud have the shape of a normal distribution, that is, the score would be normally distributed within the population.
- 5. Assumption of parametric test Three general assumptions 3. The value of each set of scores or group of scores must be comparable; this is known as homogeneity of varince Parametric test tend to be used for hard dependent variable
- 6. Assumption of non-parametric test 1. The selection of subjects from the population was random and independent. 2. If the scale of measurement is ordinal or higher then homogeneity of variance must also be assumed. The data are more subjective or soft BACK TO HOME