I just spent the last two full days in an #R learning sprint with @awunderground. A few thoughts and reflections to follow.
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ
-
In this case, I felt a learning 'sprint', with someone who really knows what he's doing would be my best bet. For a couple of lunches and covering his time, it was *totally* worth it.
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ
I sent him a quick list of things I wanted to cover: -Set up the basic R workspace (eg R script vs R project) -Read & write data (CSV, Excel, and Stata) -Reshape -Run simple tabs (means, variances, etc.) -Working with strings and number formats -Switch bw data sets
-
-
-Dataviz: --standard lines, bars, scatterplot, bubbleplot ----coloring different data values --small multiples/trellis/facet --more advanced charts: box&whisker, beeswarm, raincloud --save graphs --maps [we didn't get to this]
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
I also found it extremely useful to have a project/dataset to work through. Fortunately, I have a fairly fun project I've been delaying working on, and this provided a good opportunity.
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
Also, R has multiple ways to accomplish the same task. This is both frustrating but also fairly freeing as there doesn't seem to be a necessarily "right" way to accomplish a specific goal.
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
IMO, the
@rstudio workspace is much better than other tools (eg, Stata, SAS, Fortran97). It provides better search, formatting options, etc.ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
-
I really don't like the default graphic options in R. The gray background and white gridlines just don't thrill me. I know these kinds of styles can be subjective, so my critique may be off here, but I just don't get why so many tools have such poor defaults.
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
The core difference in my thinking (historically a Stata/SAS coder and big Excel user) is that data need to be in a long format, rather than wide. If I can get over that conceptual hump, I will be miles ahead. (Fortran is somewhat similar, but my Fortran skills are rudimentary).
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
The value of being able to use very function in very similar ways--eg, function(dataframe, argument1, argument2)--is incredibly useful. I almost don't need to think about it.
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
With respect to
@awunderground (incredibly awesome) instruction, I think the thing I valued most was being shown how to do something, then try to do it on my own, but then being immediately helped when I got stuck.ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
This may sound like a super minor thing, but it's not. Struggling for a few minutes is one thing, but struggling for hours is totally another. Aaron's approach of letting me try for a bit and then jumping in when it was clear I was stuck was immensely helpful.
ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
He also shared a variety of helpful resources, which I also recommend:
@urbaninstitute R guide: https://ui-research.github.io/r-at-urban/graphics-guide.html … CRAM task views: https://cran.r-project.org/web/views/Graphics.html …@awunderground personal page: https://awunderground.github.io/ggplot2-themes/@rstudio Community:https://community.rstudio.com/ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
Also,
@hadleywickham book is clearly a gem: https://amzn.to/2BuOKV2 (he was on#ThePolicyVizPodcast too: https://policyviz.com/podcast/episode-69-hadley-wickham/ …)ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ -
In sum (finally), I'm deeply indebted to
@awunderground for sitting down with me for nearly 20 hours to dive deep into R. I can't say I'm a true R coder yet, but feel armed to use it going forward. </thread>ಈ ಥ್ರೆಡ್ ತೋರಿಸಿ
ಸಂವಾದದ ಮುಕ್ತಾಯ
ಹೊಸ ಸಂವಾದ -
-
-
Interesting thread (
up and down) about @jschwabish’s priorities for an#rstats learning sprint with@awundergroundhttps://twitter.com/jschwabish/status/1075150270290649088 … - ಸಂವಾದದ ಮುಕ್ತಾಯ
ಹೊಸ ಸಂವಾದ -
ಲೋಡಿಂಗ್ ಸಮಯ ಸ್ವಲ್ಪ ತೆಗೆದುಕೊಳ್ಳುತ್ತಿರುವಂತೆನಿಸುತ್ತದೆ.
Twitter ಸಾಮರ್ಥ್ಯ ಮೀರಿರಬಹುದು ಅಥವಾ ಕ್ಷಣಿಕವಾದ ತೊಂದರೆಯನ್ನು ಅನುಭವಿಸುತ್ತಿರಬಹುದು. ಮತ್ತೆ ಪ್ರಯತ್ನಿಸಿ ಅಥವಾ ಹೆಚ್ಚಿನ ಮಾಹಿತಿಗೆ Twitter ಸ್ಥಿತಿಗೆ ಭೇಟಿ ನೀಡಿ.