Parametric and Equivalent Nonparametric Procedures
Parametric procedures can be appropriate when sample size is large enough to describe an entire population and the samples' data distribution is similar to the population’s distribution, usually a normal distribution. When a sample size is not large enough to describe an entire population or a sample does not meet certain assumptions required for a parametric procedure, nonparametric procedures are used.
In this assignment, you will create an infographic illustrating parametric and equivalent nonparametric procedures.
Step 1. Research
Research the assumptions tests related to the three major statistical tests from this course: t tests, Pearson’s r, and ANOVA. Note: ANOVA is often considered robust enough to be used with nonparametric data.
Step 2. Illustrate
With an infographic, illustrate which assumptions must be met to use the parametric procedures and what could be done if they are not met (i.e., resample, acquire a larger sample, revert to a nonparametric procedure)
In the infographic, show the nonparametric alternatives/equivalents to the parametric procedures when assumptions are not met. Provide a brief description of each nonparametric alternative.
Step 3. Submit
Submit the infographic.
Name:
PARAMETRIC AND EQUIVALENT NONPARAMETRIC PROCEDURES
In statistical analysis, ensuring that the underlying assumptions of parametric procedures are met is crucial for reliable results. This infographic presents the key assumptions associated with three major statistical tests: t tests, Pearson’s r, and ANOVA. Additionally, it illustrates potential strategies to address assumption violations, including resampling, increasing sample size, and transitioning to nonparametric alternatives. Explore the informative graphic to understand the prerequisites for valid parametric analyses and the available options when assumptions pose challenges.
Parametric Procedures
T Tests, Pearson’s r (Correlation), and ANOVA
Nonparametric Procedures
Resampling methods, non-parametric alternatives, and dealing with small samples
T Tests Assumptions
Resampling methods
T-tests assume that the data follows a normal distribution. This means that the values of the variable being studied are symmetrically distributed around the mean, forming a bell-shaped curve.
Homogeneity of variance, also known as homoscedasticity, assumes that the variances of the two groups being compared are roughly equal (Kelter, R. (2021). In other words, the spread or dispersion of scores in one group is similar to the spread in the other group.
What if Not Met?