Peer review

[mnybe1]

Hi group 1!

• Great work, the usage section was clear and easy to follow and the make file seemed to run well up until the knitr script because I didn't have the 'bookdown' package - maybe this could be included in your packages list at the start of your readme. Once I installed bookdown, everything ran well! All scripts were executed individually and there were now problems. The messages in your scripts were a nice touch!

• I'm not sure whether if is common practice or not, but maybe it would be useful to have a script within your make file that also installs the packages? This is not a major thing, but would be 1 less thing to run.

• Scripts ran quickly and vectorization was effectively used. I also found the comments within your script really helpful to follow along with what was going on.

• I like how there are separate folders within your data folder for the raw and processed data, however I feel that the linear models folder would be better to be not included within the data folder, i.e. maybe in a separate rds folder within your repo. This would just make your linear models slightly easier to locate.

• Interesting research questions! Your facet plot displays your data well and clearly helps to explore your research question. Diagnostic plots are great to see! Although I'm not sure whether they would be needed in your final report as long as you explain them in the text or maybe they could be supplementary data. Good use of kable for the tables!

• Overall great work! Looking forward to seeing your dashboard!

[dy-lin]

Thanks for the feedback @mnybe1! we will address these changes for milestone04 issue #40:

• add bookdown to the list of packages
• add a script to install necessary packages if not installed, and add this into the Makefile

Future work:

• move all .rds files into a separate rds directory --> moved to issue Future work #41

Since the data/linear_models is coded into many scripts at this point and I don't want to break anything mid-milestone04, we will address this at the end if we have time.

I personally think that putting it in the data directory makes sense to me-- and I don't really think anyone will really need to 'find it', so to me, making it 'easier to find' is not a high priority, as the USAGE section of the README tells you the exact path and commands to run.

[dy-lin]

I'm not sure whether if is common practice or not, but maybe it would be useful to have a script within your make file that also installs the packages? This is not a major thing, but would be 1 less thing to run.

I am also not sure if this is common practice, but I have written a script scripts/install.R which I have set as a target in the Makefile:

.PHONY: all clean install

all: clean install docs/milestone3.html docs/milestone3.pdf

# install required packages
install:
	Rscript scripts/install.R