independent_samples: Independent Samples Data Analysis

In HelikarLab/DBERlibR: Automated Assessment Data Analysis for Discipline-Based Education Research

Independent Samples Data Analysis

Description

This function automatically cleans the datasets (e.g., converting missing values to "0), binds treatment-control group datasets, check assumptions, and then runs the Independent Samples T-test (parametric) and Mann–Whitney U test (nonparametric) to help you examine the difference between the groups. R scripts and their outputs are as follows (just pay attention to the outputs since the codes are automatically run back-end by the function).

Usage

independent_samples(
  treat_csv_data,
  ctrl_csv_data,
  m_cutoff = 0.15,
  m_choice = FALSE,
  key_csv_data
)

Arguments

treat_csv_data	This function requires a csv file with treatment group data. Its name (e.g., "data_treat_post.csv") can be passed as an argument. Make sure to set the folder with the data file(s) as the working directory.
ctrl_csv_data	This function requires a csv file with control group data. Its name (e.g., "data_ctrl_post.csv") can be passed as an argument. Make sure to set the folder with the data file(s) as the working directory.
m_cutoff	This package will treat skipped answers as incorrect. However, too many skipped answers may skew the results of the data analysis. User can can provide a cutoff for the proportion of skipped answers. For example, if the user enters 0.1, students who skipped more than 10 percent of the answers will be excluded from the data analysis to prevent skewed results. The default of 0.15 is commonly applied as a rule of thumb.
m_choice	This package is capable of handling multiple-choice answers for the convenience of users. If users want to use a csv data file with multiple-choice answers, they should put m_choice = TRUE and provide another csv file that contains answer keys using the argument of key_csv_data.
key_csv_data	This function requires a csv file that contains answer keys if m_choice = TRUE. The loaded answer keys will change the multiple- choice answers to a binary format of 1 (correct) and 0 (incorrect).

Value

This function returns a tibble() including the following information:

• n_students_deleted: Number of students deleted from the data for analysis based on the percentage obtained via the argument of m_cutoff

• descriptive_statistics: Descriptive statistics

• boxplots: Boxplots - visual presentation of the descriptive statistics

• shapiro_wilk_test: Shapiro-Wilk test results to determine normality

• normal_qq_plot: The normal q-q plot to visually inspect the normality

• independent_samples_t_test_equal: Results of the independent samples t-test with equal variances assumed

• independent_samples_t_test_unequal: Results of the independent samples t-test with unequal variances assumed

• mann_whitney_u_test: Results of the Mann-Whitney U test

Examples

# Run the following codes directly in the console panel. The plots
# generated through the link above may be displaced depending on the screen
# resolution.
independent_samples(treat_csv_data =
         system.file("extdata", "data_treat_post.csv", package = "DBERlibR"),
         ctrl_csv_data =
         system.file("extdata", "data_ctrl_post.csv", package = "DBERlibR"),
         m_cutoff = 0.15, m_choice = FALSE, key_csv_data = NULL)

HelikarLab/DBERlibR documentation built on Sept. 20, 2023, 12:37 p.m.