I am compiling information for a quantitative mini-project for my research class. This class is for my doctorate degree and is asynchronous learning so I have dependent upon what I can teach myself via the internet! Any help will be greatly appreciated.

I am conducting a quasi-experimental study. My problem is that state standardized test scores have declined after covid. I have 250 data points of archival test scores (given to me).

Here are my Research Questions & Hypotheses:

1. What difference exists, if any, between pre-covid and post-covid standardized test scores?

H₀: No statistically significant difference exists between students' pre-covid and students' post covid standardized test scores in a middle school math class.

H₁:A statistically significant difference exists betweenstudents' pre-covid and students' post covid standardized test scores in a middle school math class.

2. What difference exists, if any, in student test scores between rural students and urban students in post-covid learning environments?

H₀: No statistically significant difference exists betweenrural students and urban students in post-covid learning environments

H₁: A statistically significant difference exists betweenrural students and urban students in post-covid learning environments

I had stated that I would use a t-test or ANOVA to analyze the data and determine if the differences between the groups are large enough to be considered significant.

Here are the assignment instructions: (In Bold)

Explain the procedure's non-parametric equivalent and when it might be used.

I have this so far:

A non-parametric alternative for the t-test is the Mann-Whitney U test, also known as the Wilcoxon rank-sum test. This test is used to determine whether two independent samples come from the same population based on their ranks rather than assuming normal distribution. The Mann-Whitney U test is particularly useful when the data is not normally distributed, and the sample sizes are small. Another non-parametric alternative is the Wilcoxon signed-rank test, which is used when the data is paired, such as before-and-after measurements. This test is based on the ranks of the differences between pairs and can be used when the assumptions of the paired t-test are not met.

Determine the extent to which you think you will be able to collect sufficient data to use a parametric or non-parametric version of the test. Which assumptions tests must be performed or met if the procedure's parametric version can be used?

Describe the assumptions tests which must be met. For example, is the sample large enough? Is it normally distributed? Does it meet the requirement of equal variances? If not, what limitations would this introduce to your study? IS one of the statistical test below appropriate for your design.

1. t-test (paired/dependent or two sample/independent)
2. ANOVA
3. Pearson's r

Should the data not meet assumptions testing, what are the non-parametric alternatives for these tests? Does Excel support them in its Analysis ToolPak?

How do you analyze data in Excel? For example, if rows are not sorted by the testable groups, will you need to resort the rows? Do you need to clean up any other aspects of the data?