Peer review feedback - Githubissues

You do a good job of justifying your decision to use self-reported data, and I think this adds to the power of your analysis. The first step in your analysis is a good starting point to lay out basic information, but variables in Table 1 could have clearer headings, eg. V1 is not very indicative. The interactive Table 2 is clear and effective, though it is not clear to me why you display the objects of search in Table 3; it does not seem relevant to any other part of the analysis. Graphs 1 and 2 are clear and well-labelled, but not connected to the research question explicitely. You show creativity with the maps and the different dimensions conveyed by these: Maps 1-3 are very effective and Maps 4-6 tie your findings nicely together. Your conclusion is sound and overall you find a good balance of depth and breadth considering the wordcount.

A minor code improvement could be to make a function to count NA's in the data supplied. Another improvement would be to use a relational database to store your data, at least for the first half of your project, instead of using tidyverse which is at times difficult to follow and lengthy. This would limit your data files and improve legibility. Further, it would be helpful to have a data-collection and cleaning section before the analysis to improve organization. This would make your plotting chunks easier to follow. Nonetheless the code runs without issues.

MichellePapadakis / MY472_PSF

Peer review feedback #1