search

jepusto / scdhlm

An R package for estimating between-case standardized mean difference effect sizes for single-case designs. The package implements the methods proposed in Hedges, Pustejovsky, & Shadish (2012, 2013) and Pustejovsky, Hedges, & Shadish (2014).
http://jepusto.github.io/scdhlm/
4 stars 1 forks source link

Add 3-level models to the app #81

Closed manchen07 closed 2 years ago

manchen07 commented 2 years ago

I'm sorry that I did not have time to figure out the predicted lines for the RMBB and CMB designs in the app. To be specific, the predicted lines for Thiemann2001 look fine if we, after fitting a model and calculating effect sizes, copy and paste the syntax from the app and run the code in R. But the lines look distorted in the app. And I'm not sure what's the best way to get the predicted lines for each cluster in CMB design.

Another issue is with the variablesLoaded in server.R, line 126. I'm not sure how to update this with the 3-level models.

After you review, could you please let me know if I missed anything or if there is anything I can work on? Happy to talk anytime via zoom if needed. And sorry about the conflicts between main and update-app branch.

manchen07 commented 2 years ago

Thanks for editing my badly written code!! I'll continue working on this today.

codecov-commenter commented 2 years ago

Codecov Report

Merging #81 (375039a) into main (2038402) will increase coverage by 1.90%. The diff coverage is 93.69%.

:exclamation: Current head 375039a differs from pull request most recent head b60f059. Consider uploading reports for the commit b60f059 to get more accurate results

@@            Coverage Diff             @@
##             main      #81      +/-   ##
==========================================
+ Coverage   54.68%   56.58%   +1.90%     
==========================================
  Files           8        8              
  Lines         726      774      +48     
==========================================
+ Hits          397      438      +41     
- Misses        329      336       +7
Impacted Files Coverage Δ
R/preprocess-function.R 88.65% <89.83%> (-3.99%) :arrow_down:
R/graphing-functions.R 97.26% <98.07%> (-0.89%) :arrow_down:

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 2038402...b60f059. Read the comment docs.

manchen07 commented 2 years ago

I did some further debugging. The predicted lines in the app seem good using example data or uploaded data. The tests for preprocess_SCD() and graph_SCD() look fine as well.