textTopicsTest: This function tests the relationship between a single topic...

In text: Analyses of Text using Transformers Models from HuggingFace, Natural Language Processing and Machine Learning

This function tests the relationship between a single topic or all topics and a variable of interest. Available tests include correlation, t-test, linear regression, binary regression, and ridge regression. (EXPERIMENTAL - under development)

Description

This function tests the relationship between a single topic or all topics and a variable of interest. Available tests include correlation, t-test, linear regression, binary regression, and ridge regression. (EXPERIMENTAL - under development)

Usage

textTopicsTest(
  model,
  pred_var,
  group_var = NULL,
  control_vars = c(),
  test_method = "linear_regression",
  multiple_comparison = "fdr",
  load_dir = NULL
)

Arguments

model	(data.frame) The model returned from textTopics().
pred_var	(string) Variable of interest for linear or binary regression
group_var	(string) Grouping variable for t-test
control_vars	(list) Control variables for linear or binary regression
test_method	(string) Choose between "correlation", "t-test", "binary_regression", "linear_regression" or "ridge_regression"
multiple_comparison	Method for correction of multiple tests (e.g., "fdr", "bonferroni").
load_dir	(string) if specified, the function returns the precomputed analysis from the directory, otherwise leave blank

Value

Metadata and results of the test such as estimate, t-value, p-value, and variable name.

text documentation built on Sept. 11, 2024, 7:22 p.m.