Study 1 Example Report

Most Recent Update: 2022-12-15 12:21:13.

The Study 1 Example Report is now available.

The Example Report provides a brief demonstration of an appropriate analysis for each of the required Study 1 analyses in Project B. Complete instructions for your Study 1 report are available here.

The Study 1 Example Report

The Study 1 Example Report document demonstrates many of the elements you would need to complete, including:

  • data ingest and merging
  • cleaning the data
  • codebook, analytic tibble and data summary
  • Analysis A: Compare two means/medians using paired samples
  • Analysis B: Compare two means/medians using independent samples
  • Analysis C: Compare 3-6 means/medians using independent samples
  • Analysis D: Create and analyze a \(2 \times 2\) table
  • Analysis E: Create and analyze a \(J \times K\) table, where \(2 \leq J \leq 5\) and \(3 \leq K \leq 5\)
  • Session Information

for a sample of 53 observations from a 2015 class survey of 431 students and TAs from that year. Since the sample size is much smaller than what you will be using (which will have a minimum of 250 complete cases) some issues are more problematic here than you might have to deal with in your work.

Study 1 Example Report Documents

The raw data files for the Study 1 Example Report are described in the R Markdown and HTML files above and are found on the 431-data website. They are named:

  • projectB-study1-demo-survey-2015a.xlsx (first 20 observations, all 21 columns)
  • projectB-study1-demo-survey-2015b.csv (remaining 33 observations, some of the 21 columns)
  • projectB-study1-demo-survey-2015c.csv (remaining 33 observations, rest of the 21 columns)

The final data set after merging for the Study 1 Example Report includes 53 observations on 21 columns. The subjects are identified with a code called s_id in each of the raw data files.

Caveats

  1. The Example Report does not precisely follow the requirements from Study 1 for the Research Question and Conclusions sections, because that is a big part of your job.

  2. The Example Report has a much smaller data set than you will have, so I don’t hold myself in the Example Report to the standards for cell sizes and numbers of observations that you will need to meet in your Study 1.

  3. In the Example Report, I produce all 5 analyses (you will do only 4) and I also run many different possible analyses in each case, when you will instead select one. For instance, in comparing paired samples, the demonstration shows code for confidence intervals based on the t distribution and the bootstrap, while your Project B Study 1 will include just one of these, should you choose to do that analysis.

  4. Outside of those caveats, these demonstrations are meant to be accurate and reflect the level of detail I’m looking for, although it is certainly possible to expand on these demonstrations in your work, if that seems helpful. I believe the Example Report to be largely accurate, but many eyes will find some issues, eventually.

If people find corrections and would be good enough to post them to Campuswire (under the projectB folder, please) I will attempt to reflect those edits in revisions. If I do something in the Example Report that appears to be in conflict with the instructions for Study 1, then (a) please let me know, and (b) treat the Study 1 instructions as your guidepost in completing your work.