When running factor analysis, one often needs to know how many components to retain. Fortunately, many methods exist to statistically answer this issue. Unfortunately, there is no consensus on which method is the best. Therefore, the n_factors()
function, available in the psycho package, performs a method agreement procedure. In other words, it runs all the routines and returns the number of factors with the higher consensus.
# devtools::install_github("neuropsychology/psycho.R") # Install the last psycho version if needed library(tidyverse) library(psycho) results <- attitude %>% psycho::n_factors() print(results)
We can have an overview of all values by using the summary
method.
summary(results)
knitr::kable(summary(results), digits=2)
And, of course, plot it :)
plot(results)
The plot shows the number of methods (in yellow), the Eigenvalues (red line) and the cumulative proportion of explained variance (blue line).
For more details, we can also extract the final result (the optimal number of factors) for each method:
psycho::values(results)$methods
knitr::kable(psycho::values(results)$methods)
We can also provide a correlation matrix, as well as changing the rotation and the factoring method.
df <- psycho::affective cor_mat <- psycho::correlation(df) cor_mat <- cor_mat$values$r results <- cor_mat %>% psycho::n_factors(rotate = "oblimin", fm = "mle", n=nrow(df)) print(results) plot(results)
This package helped you? Don't forget to cite the various packages you used :)
You can cite psycho
as follows:
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.