andreashandel/madacourse-rmarkdown

Course page for Modern Applied Data Analysis

This is the RMarkdown based version of the Modern Applied Data Analysis course.

As of January 2024, the site has been switched to Quarto and this repository is outdated and for archival purposes only.

Here is the current, Quarto based repository.

madacourse-rmarkdown's Issues

Resource Suggestions for MADA course

Hello Dr. Handel!

I watched some of these videos when we were first learning how to code for this class, and I found them to be somewhat informative and helpful. They are youtube videos that just go over how to do some basic stuff in R, and why R is so important. They are nice because you can listen to them while you code, and also watch the examples which in turn helps with coding practice. I think they could be useful to your course at least as a beginner's guide, offering additional support to what you have already provided.

https://www.youtube.com/watch?v=fDRa82lxzaU

https://www.youtube.com/watch?v=_V8eKsto3Ug

Thanks!

Issue opening tabs for topics at the top of course site

This could just be an error with my computer/mouse, but whenever I hover over the tabs at the top of the course site, the "subtopics" under that tab don't open. When I try to click on them the subtopics disappear. In order to open them I have to right click on the topic (ie Modules, or exercises), then click on the subtopic I want. Again, could just be an error with my computer.

Beaudry_Typo,Suggestion,Website_Issues

Hello,

I went to find some typos in The Art of Data Science. I found a typo in section 2. 6 Applying the Epicycle of Analysis Process. At the end of the second sentence they have a space between “are” and the colon. There should not be a space here.

In terms of suggestions for improving this course, I snuck ahead to section 5a – Types of Data. You went over two kinds of variables that are common: quantitative and qualitative. I would suggest that you also formally introduce the concept of “mixed-methods.” Although I have yet to do a mixed-methods study myself, they are fairly common and important to public health. I have listed below several articles I think that may work as reading example of these types of studies, if you had decided to include this concept in your course.

1. Creswell JW, Fetters MD, Ivankova NY. Designing a mixed methods study in primary care. Annals of Family Medicine 2004;2:7-12
2. Mumtaz, Z, Nykiforuk, C, Bhatti A, Ataullahjan A, Salway S. The role of social geography on Lady Health Workers’ mobility and effectiveness in Pakistan. Social Science and Medicine 2013: 91:48-57.
3. Mumtaz Z, Cutherell, A and Bhatti, A. Saving mothers and newborns in communities: strengthening community midwives to provide high quality essential newborn and maternal care in Baluchistan, Pakistan in a financially sustainable manner. BMC Pregnancy and Childbirth 2014;14:131
   You’ll note that two of the papers are by the same first author. She was a guest lecturer in a course I took on the foundation of public health research for my MS, and was expert on (you guessed it) mixed methods research!

Also, I am not sure if this is an issue on my end (as I have been having computer problems lately) but the top of the pages to some of the modules (i.e., all of module 4 and 5) are covered by the course menu….

Thank you,
Megan Beaudry

Prefer video to guide how to do something

Dear Dr. Handel,

In some modules, you write down some steps to do something such as clone to github by gitkraken, reference management in Markdown, creating introduction page and so on.

I am new to Github and R markdown. Sometimes, the guidance is not step by step to do it. Thus, it took me 3-4 hours to figure out and I am still not sure what I understand is true or not. For example, reference management in Markdown. I prefer to learn through a short video. For example, a video show how to make a citation in R markdown and automatically produce a word document with reference.

I only have one suggestion to improve this course. Overall, it is good.

Minor Broken Icon and Suggestion for Links under Communication

Hi Dr. Handel,

Under module 5c: Data Import to R, the emoji icon is not working.

I found it very hard to find anything typo-wise. I think this course website is very nicely laid out and easily navigable. As a student, I always appreciate having all the information at my fingertips.

My only suggestion: Would it be overkill to maybe add actual links under the Communications section to UGA ELC and the class GitHub? It would be nice to just have a quick link to these since I often switch computers.

Also I know this should probably go under the Project discussion board, but since I'm assuming other classmates will look at these issues too, will you be expanding the Project information at all? I'm assuming for Part 1, we submit the data we chose to our new project directory, but it would be nice to have that clarified on the 5c assessment page.

Thank you!

Megan Robertson

just a typo

under course communication tab:
"I prefer you use the discussion boards instead of email to ask questions"
-->
"I prefer you to use the discussion boards instead of email to ask questions"

Clarification on advantages of using Gitraken vs RStudio's Git client

In the first course module, we are encouraged to download a Git client (such as GitRaken) to work with Git repositories. I was wondering if you could provide clarification on the advantages of a Git client such as GitRaken, over using RStudio's basic Git client. I'm asking only because I tried using GitRaken but found it confusing, so I defaulted back to using RStudio's basic Git client. But I feel like I must be 'missing out' on something by using RStudio instead of GitRaken--I just haven't figured out yet what that 'missing out' is. Could you provide some insight on the major advantages of GitRaken over simply using RStudio's Git client?

Minor Typo in "Modules" drop-down menu!

Good evening,

I noticed on the MADA course GitHub there is a minor typo under the "Modules" tab. More specifically, on the "Modules" drop-down menu, the first module reads "1. Coures and Tools Introduction", but I believe you intended to say "1. Course and Tools Introduction."

Respectfully,

Paige Duffin

Github Setup Instructions

Hi Dr. Handel! This is a small suggestion, but I keep geeting tripped up with the Github instructions on the introducing Ourselves Page.

Here's the current text:
"Under the Options tab in settings, scroll down to Github Pages and switch it to master from none. Don’t choose a theme."

Is it possible to change it to "switch from none to master". For some reason, I can never read this blurb right and think that "none" is the correct setting. Thinking through it, I see why master works at this stage in the course.

Issue for MADA course

Some things can be unclear and seem tedious perhaps including explanations of why certain things are important would help.

Typo in 1a. Course Introduction

In the section titled "Course Philosophy", in the last paragraph (in Bold) the first line contains the word "challengin" where it is missing the second "g".

Otherwise i'm enjoying the course so it was hard to find issues!

Best wishes
Jonny

Introducing Ourselves Exercise-Github Setup: repo must be public when using GitKraken free.

Greetings,

When working through the exercise Introducing Ourselves under the Github Setup subheading, second bullet point; students are instructed to create their first repo for the assignment. The instructions read "Choose if you want the repository to be private or public." These instructions are not an issue if the user has installed Gitkraken Pro or has the Student Developer Pack, however when using GitKraken Free students will be unable to access the repo until a public designation has been made.

Though this would not likely be an issue for any experienced Github/Kraken users, it would be helpful to provide this information for first time users when working on their introductory project. Thanks!

Subset selection vs LASSO

I recently looked at this methods paper again in my internship so thought I'd post this here for us to think about.

Hastie T, Friedman J, Tibshirani R. Best subset, forward stepwise, or LASSO? Stat Sci, 2020; 34(5) 579-592.

On the subset selection page, it says this.

Most of the criticism is directed at doing subset selection through AIC, F-test or other criteria, not using CV. I haven't yet seen a convincing article that explains that subset selection is equally bad if one uses CV (if you find one, please let me know!)

Maybe we should update it to reference this paper: https://www.stat.cmu.edu/~ryantibs/papers/bestsubset-sts.pdf. Through lots of simulations (all available in supplement) they show that best subset / forward selection are similar and perform best in high SNR situations if variables are uncorrelated. LASSO tends to win if SNR is low, relaxed LASSO is pretty much always a good choice (even when it isn't best, which it often is, it never does too bad).

In some of these three authors' other work, they tend to say (mostly based on empirical/anecdoctal evidence) that ridge works best for many correlated small effects; LASSO/EN works best for (relatively) large, sparse, and potentially correlated effects; and forward selection works best for large, sparse, uncorrelated effects, and dominates when SNR is high. My personal bias is that most of the problems we work on are not orthogonal design, typically low SNR so I almost always prefer EN.

Identifying important course information

Hi Dr. Handel,

I really appreciated the skills I've been able to pick up in this course so far and I look forward to the rest of the semester. One of the things I've struggled most with in this course so far is knowing where to find all important information and when assignments are due. For example, I've gone by the schedule under General Information but I failed to see that parts of assignments are due are different times under the exercise description. Or an issue shared by several classmates was appropriately participating in class discussions. I know this can be solved by carefully reading all the relevant course information but I wanted to share the thing I was struggling with in navigating the course.

Thanks,
Emily

suggestions for MADA course

Hi Dr. Handel,

Thanks so much for giving this opportunity to provide feedback on the course site. It looks like the site has been combed through and is looking great! I have just a few suggestions and lots of praise to share.

Is there a way to center in the second line of the University Honor Code on the Syllabus Menu?

When working through the Wine ratings exercise, the file would only ever save on my device as a .txt file. Eventually, I copied and pasted the code into a new Rmd file and ran it from there. Not sure if anyone else had a problem with this.

As we create R markdown files for our own notes, I've been hoping to add in notes on functions and other tools in R. How do we go about writing these functions so that they appear with the font and grey background as they do in your site?

In terms of praise - thank you for making the website to user-friendly and accessible! I'm grateful for your friendly language and clear instructions. Having an online platform, it's some times hard to tell tone. Thank you for including emotions (and emojis!) to help us understand how we should be reading along - for providing a voice to the text.

Thank you!
Megan

Top of page doesn't show when using half screen view (Amanda Blubaugh- GitHub Issues Exercise)

Greetings,

I have found that when looking at the modules, it is easiest for me to have the website with the module open on half of the screen and have R studio or the recommended reading on the opposite half of the screen.

Only when the screen is halved, the very top of each module is no longer visible and is covered by the navigation bar on the top. When the page is open full screen it isn't an issue. I am not sure how to resolve this. Could it just be my browser? I mainly use Chrome.

Best,

Amanda Blubaugh

Frederick_Clarity in Module 5a

In this section you list explicitly for some chapters that we will need to know them for the quiz, other chapters we won't, and some you don't say anything about at all.
Specifically, Module 5a section "Working with different data types in R." For chapters 25 and 27, there isn't anything said about how much of these we will need to know for the quizzes, and with the amount of reading in the chapters for this module I'd like to know how much focus I should be putting on these as well.

##Assignment for Module 4
##Julia Frederick

Link on Website for Tree fitting exercise not working?

I have tried to follow the link to the next exercise on the website page (https://andreashandel.github.io/MADAcourse/Exercise_Tree_Fitting.html) without luck.

I keep getting the 404 page not found response. I will keep trying to figure it out, as it may be something on my end, but I wanted to mention it just in case.

Amanda

Page title error

The 6b assessment page for "Data wrangling and tidyverse", has the title of the previous week's section "Data types and getting data assessment".

RStudio Clarification

One thing that took me awhile to figure out because I've never used R/RStudio before was the RStudio markdown view. Maybe some clarification on this or a link out would be nice. Specifically, in RStudio in the top left panel where the script is written there is a button that says R Script that can be toggled to different ones such as R Markdown. This was confusing. Then at the same time when in the RMarkdown view the differences between all the different "Run" selections. Such as "Run selected lines, run current chunk..." I'm unsure of which is better to use in different situations, and I've mostly just been using "Run All" because it's been giving me the most reliable results.

--

On the website, the header has today's date for every page. I don't know if it's supposed to be the date that it was edited because I know you're changing a lot of stuff, or if it's updating in real time.

--

This is unrelated to the website...but the discussion board on ELC is horrible. I don't know if it's because I don't know how to use it or what. But I can't view the thread when I'm attempting to reply so it's very irritating to respond to people. It doesn't really foster conversation very well.