Statistical Tests

PARAMETRIC
1. About
  1. Level or scale of measurement is of equal interval or ratio scaling
  2. The data are normally distributed OR can be transformed
    1. If the variables are counts or derived they must be transformed
  3. Uses actual observations or measurements
  4. Often compare means and variances of interval scales
    1. When comparing variances, each need to be homogenous
  5. No loss of information
  6. ROBUST if the sample size is LARGE
  7. Greater statistical power than nonparametric
2. TESTS
  1. t-test
    1. One sample t-test
    2. Independent sample t-test
    3. Paired sample t-test
  2. ANOVA
    1. One way ANOVA
    2. Multi-factorial ANOVA
      1. Hierarchical (nested) design
      2. Two-way crossed layout
        With replication
        Without replication
NONPARAMETRIC
1. About
  1. Used on Nominal or Ordinal data in ADDITION to Interval and Ratio
  2. Distribution FREE
  3. Used on observations, measurements or values represented by assigned ranks
  4. Compare medians of interval or ordinal scales
  5. Counts or derived variables can be analysed directly
  6. Loss of information
  7. Small sample sizes can be used as normality is not a prerequisite
  8. Have LESS statistical power than parametric equivalents
2. TESTS
  1. 2 independent samples
    1. Mann-Whitney U
    2. Kolmogorov-Smirnov Z test
  2. K Independent Samples
    1. Kruskal Wallis Test
    2. Median Test
  3. 2 Related Samples
    1. Wilcoxon Test
  4. K related samples
    1. Friedman Test

Next up

Statistical Tests

Description

Resource summary

Similar

	Created by jessica_weyell over 11 years ago