jamovi

Hi,

I teach statistics to undergrad biology students. I used to teach my students using R commands, under RStudio, to perform import/export data, data wrangling, statistical tests, and plot graphs.

For some students, they feel overwhelmed by too much programming. Then I learned about jamvoi, The spreadsheet-like user interface is exactly missed in R.

However, I have a general question. If I want to do something beyond what the menu options can provide, e.g. make a barplot of two groups with error bars, what would you recommend me to do? Should I write R codes via the Rj editor?

Thanks, Eric.

yup, two options eric.

you can either use the Rj editor, or if it's something you'd like to make it easy for other people to do, you could write a module (like the other modules in the jamovi library).

https://dev.jamovi.org/

cheers

jonathon

Just to note that barplots with error bars are discouraged by, for example, Nature journals because they break a number of important rules : they don't show the data, and they don't show the distribution. In jamovi you can plot your data, add violin plots to show shape and boxplots to show summary statistics. Much preferable.

You say:

barplots with error bars are discouraged by, for example, Nature journals because they break a number of important rules : they don't show the data, and they don't show the distribution.

I would *love* to have the reference to this article to pass on to people I know.

Any chance you can provide the reference?

P.

Hey @PeteD,

I don't know if these readings may be of interest:
https://www.nature.com/articles/nmeth.2813
https://www.scielo.br/j/rsbmt/a/KfzwDkz ... df&lang=en
https://bpspubs.onlinelibrary.wiley.com ... 2/prp2.369

Cheers,
Maurizio

Just to note the existence of rainbow plots in the module, which are certainly the most informative way of plotting your data. Have a look here

https://pubmed.ncbi.nlm.nih.gov/31069261/