Making RMD nice and publisable

[itchyshin]

@elmacartney - I have gone through the RMD - all the reasons why it did not knit was not due to technical reasons which I thought was the case. Everything was solved just by reading the warnings. Anyway, there are a few things to do to make it nicer than what we have. So I will go through these reasons and how you can improve your RMD skills - a few points for us to remember to discuss:

1. General setting ups of RMD
2. Run as you go - chunk by chunk (cache = TRUE)
3. Understanding Rmarkdown arguments (useful link for us to look at https://kbroman.org/knitr_knutshell/pages/Rmarkdown.html)
4. How to put alluvial plots on panels (tabs)?
5. When to save and why (note many problems were due to eval = FALSE and why)
6. Try to use more functions rather than repeating (I should do this as well - it makes life easier when we need to change something)
7. Header orders and also chunk orders (there were a few problems for these)

It is knitted for now but more to do to make it good enough for publication. Anyway, we are nearly there and a great effort.

[itchyshin]

the list of items to improve the HTML - making it accessible to the interested -especially keeping in mind students. Also, things to do to get this finished.

• [x] @elmacartney - add more explanations (see examples sent by @itchyshin earlier). Also, create a list of questions etc as she goes to discuss with @itchyshin
• [x] @elmacartney add VCV functionality to leave-1-out
• [x] @mlagisz - please create the list of figures which you cannot read the labels etc so we can fix it - if you can do this you can go ahead and do or @elmacartney and @itchyshin to sort it out (I do not think there are many of these tho).
• [x] @elmacartney the risk of bias section seems to be misplaced - don't we need to do a meta-regression or something to check this statistically - just % tables are listed (which seems like it should be a part of Data Exploration section).
• [x] @itchyshin - create a function for contrasts and implement an example
• [x] @elmacartney - once @itchyshin finishes the previous item, implement it for all the meta-regression analyses where applicable
• [x] @mlagisz to read the HTML once we are all done
• [x] @elmacartney - if contrasts show some new insights, this needs to be transferred to the main MS

[elmacartney]

@mlagisz Are you able to please check the HTML? I think all figures are clear now, but I can't seem to get the text visible in the visualisation of missing data figure. If you know how to do this, could you please add to the RMD code? Thanks!

[mlagisz]

@elmacartney I had a look - its just too much data to make it clear and reducing text will make it unreadable anyway. Since your aim (and figure title) is just to present data for selected predictors, I suggest to only keep study ID and a few columns with predictors used in the analyses - this will make it readable (i.e. no need to show the whole dataset here). Also this will be no longer needed below the figure (currently, a bit unclear): "Note that most missing data are for different descriptive statistics as papers are often report results using different types of descriptive statistics (i.e., SD, SE, mean, median etc)." Hope it makes sense.

[mlagisz]

Overall, it looks great! I will have a closer look at the text a bit later.

[elmacartney]

Great thanks!

I'm just making a final few changes and writing the cover letter, then will let you know when it's ready for you to go through everything :)

On Thu, Oct 28, 2021 at 3:24 PM Malgorzata Lagisz @.***> wrote:

Overall, it looks great! I will have a closer look at the text a bit later.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/elmacartney/EE_stress_MA/issues/24#issuecomment-953491240, or unsubscribe https://github.com/notifications/unsubscribe-auth/ATGFCZQNK5S5PAXYIJVOC63UJDGALANCNFSM5ESGQVWA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[mlagisz]

@elmacartney I made some small changes to the text directly on github version of Rmd file - I did not re-knit this.

Stress type coded as "MS" can be intuitively interpreted as "Multiple Sclerosis" - I recommend going explicitly for "maternal separation" when changing numeric codes to human readable ones.

I mostly expanded some figure descriptions and spacing. Some figure labels were shifted to odd position on the right - I think this happens when you do not finish the previous line with double white space (indicator of end of the line for the markdown - I tried to fix it).

I find figure numbering a bit confusing - some figures have numbers, other not, or the numbers are repeated, sometimes extra explanation is provided , for others it is not - maybe have some uniform numbering system with letter S and where relevant followed by info in brackets on fig number in the main text?

The figures with multiple panels (Fig.4, 5, 6, 7) are not really readable.

[itchyshin]

@elmacartney - I have gone through the RMD and fixed a few things and re-knitted and pushed it. I think it looks good to me. Please check it again and it is good to go. We will probably not need to meet specifically for this for now. Excellent to get to where we are in a relatively short time!