The collation and evalutation of student records and performance indicators is helpful when trying to determine factors most effecting overall outcome and course performance. The code here written provides some basic structure an experimenter may wish to use to perform such an analysis.
All data processing was performed in the R programming language and using the RStudio IDE. If starting from scratch with R, it is recommened to use RStudio for troubleshooting purposes and its comprehensive support.
The following packages are needed for the code to run as-written:
- tidyverse
- readODS
- effsize
- dplyr
These packages are not required by the code as-written but are often called and generally helpful to have available at the experimenter's discretion:
- ggplot2
The above packages require these resources be installed. It is possible that this list is not exhaustive, but any experimenter using RStudio will be told which packages are necessary if the scripts fail to build.
- libxml2
- libxml2-dev
- openssl
- libssl-dev
- libcurl4-openssl-dev
- rtools1
The data is aggregated with some third-party software, I believe Moodle in this case. While a handy utility, it does not provide data that is directly usable in R without some pathological coding. I found it simpler to pre-process the data from Moodle directly. For the workflow, this means:
- sending the data through preprocessing.r
- using testscripting.r for analysis
Footnotes
-
May only be necessary to avoid warning messages about not being able to compile from sourcecode ↩