In neuropsychology/psycho.R: Efficient and Publishing-Oriented Workflow for Psychological Science

library(knitr)

Psychologists are, for historical reasons, familiar with categorical predictors. These are reffered to as "effects" of different conditions upon an outcome variable. Interestingly, one of the first topic that is taught to first year students, in experimental psychology, is "interactions". When I started doing statistics, after a bachelor in psychology, there was, to me, no such things as nested effects or variables. You tested a model (often directly the ANOVA) with the interaction or without.

Let's take interest in the relationship between life satisfaction, sex and concealing (the tendency to suppress or hide one's emotions)

The Data

library(tidyverse)
library(psycho)

df <- psycho::affective %>% 
  select(Salary, Life_Satisfaction, Adjusting)

summary(df)

Primitive Way

anova_results <- aov(Life_Satisfaction ~ Sex * Concealing, data=df)
analyze(anova_results)

Regression

fit <- lm(Life_Satisfaction ~ Sex * Concealing, data=df)
summary(fit)

analyze(anova(fit)) 

First, note that running an anova on the linear model produces EXACTLY the same results. But we can to have a look at the model itself, which is richer than the ANOVA.

analyze(fit) 

Again, three lines of "effect". One could think these a similar to the one

Visualize

fit <- lm(Life_Satisfaction ~ Adjusting / Concealing, data=df)
summary(fit)
df %>% 
  refdata(c("Concealing", "Adjusting"), length.out = 5) %>% 
  get_predicted(fit, .) %>% 
  ggplot(aes(x=Adjusting, y=Life_Satisfaction_Predicted, alpha=Concealing, group=Concealing)) +
  geom_line()

Contribute

Of course, these reporting standards should change, depending on new expert recommandations or official guidelines. The goal of this package is to flexibly adaptive to new changes and good practices evolution. Therefore, if you have any advices, opinions or such, we encourage you to either let us know by opening an issue, or even better, try to implement them yourself by contributing to the code.

Credits

This package helped you? Don't forget to cite the various packages you used :)

You can cite psycho as follows:

• Makowski, (2018). The psycho Package: An Efficient and Publishing-Oriented Workflow for Psychological Science. Journal of Open Source Software, 3(22), 470. https://doi.org/10.21105/joss.00470

Previous blogposts

• APA Formatted Bayesian Correlation
• Fancy Plot (with Posterior Samples) for Bayesian Regressions
• How Many Factors to Retain in Factor Analysis
• Beautiful and Powerful Correlation Tables
• Format and Interpret Linear Mixed Models
• How to do Repeated Measures ANOVAs
• Standardize (Z-score) a dataframe
• Compute Signal Detection Theory Indices

neuropsychology/psycho.R documentation built on Jan. 25, 2021, 7:59 a.m.