mini project deliverable 1 + instructions

[iciarfernandez]

Hi all,

I have written a draft for the first deliverable of the Mini Data Analysis project, and added overall instructions for the whole project. As discussed in #21, the hope with the project is for students to practice on their own independent data analysis while being provided some sort of framework / guidance to minimize open-endness and define the scope of the project (also to avoid some students spending too much time on it, and others too little). @Vincenzo suggested the following overall "topics" for the deliverables:

1. Rmd and some dplyr exploration.
2. Tables (through tidy data manipulation)
3. Graphs (perhaps plotting those tables they made in Milestone 2, and this would also be useful for making the point that it's almost always better to present a graph than a table.)

I'd appreciate any feedback on this deliverable, and please don't hesitate to ask me any questions! Also, where I note in Task 3 that they can see an example, I'm working on that still - just a separate brief document that they can have a look at for extra guidance.

P.S. Ignore the worksheet 3a that I added to this branch by mistake, still working on that!! Sorry

[wvictor14]

This looks reasonable to me.

Perhaps some indicators of where we expect students to write their comments/code/answers? something like:

[wvictor14]

Question about milestone 2. "Tables", does the milestone focus more on dplyr data wrangling, specifically with the summarize and group_by functions? Does it also include any visualization with ggplot2? Is there anything else I should include in it? Do you want me to include "tidy data", which is taught on Oct 6th, and this milestone is due on Oct 9th?

I'm thinking for milestone 2, I'm going to build on the "4 research questions" that we asked students to define in milestone 1

[dy-lin]

Looks great Iciar (and Victor)!

I'm thinking for milestone 2, I'm going to build on the "4 research questions" that we asked students to define in milestone 1

I like this idea, but how are you planning on doing this since their research questions could differ quite drastically?

Perhaps some indicators of where we expect students to write their comments/code/answers?

I think that we should let the students know, that it should be clear which tasks they are choosing to do for those choose X out of Y questions. i.e., explicitly write which task they are doing before their code chunk, label the code-chunk, or title the plot(s)-- should make marking a bit easier for us

Also, are the datasets from UBC-MDS large enough for them to draw significant conclusions if they choose to do model-fitting?

[vincenzocoia]

@wvictor14:

Question about milestone 2. "Tables", does the milestone focus more on dplyr data wrangling, specifically with the summarize and group_by functions? Does it also include any visualization with ggplot2? Is there anything else I should include in it? Do you want me to include "tidy data", which is taught on Oct 6th, and this milestone is due on Oct 9th?

It might be a good idea to get them to pivot their data. They should be able to do it by the end of the week, and working on it as part of Milestone 2 sounds reasonable to me.

[wvictor14]

@wvictor14:

Question about milestone 2. "Tables", does the milestone focus more on dplyr data wrangling, specifically with the summarize and group_by functions? Does it also include any visualization with ggplot2? Is there anything else I should include in it? Do you want me to include "tidy data", which is taught on Oct 6th, and this milestone is due on Oct 9th?

It might be a good idea to get them to pivot their data. They should be able to do it by the end of the week, and working on it as part of Milestone 2 sounds reasonable to me.

Sounds good thanks