Add statistical test results to summary data

Description

Calculates and appends p-values with optional statistical details to a summary table based on variable types and group comparisons. Handles both continuous and categorical variables with appropriate statistical tests.

Usage

add_p(
  summary,
  digit = 3,
  asterisk = FALSE,
  add_method = FALSE,
  add_statistic_name = FALSE,
  add_statistic_value = FALSE
)

Arguments

Value

A data frame merged with statistical test results, containing: - Variable names - Summary - Formatted p-values - Optional method names/codes - Optional statistic names/values

Examples

# `summary` is a data frame processed by `add_var()` and `add_summary()`:
data <- add_var(iris, var = c("Sepal.Length", "Species"), group = "Species")
summary <- add_summary(data)

# Add statistical test results
result <- add_p(summary)

Add summary statistics to a add_var object

Description

This function generates summary statistics for variables from a data frame that has been processed by add_var(), with options to format outputs.

Usage

add_summary(
  data,
  add_overall = TRUE,
  continuous_format = NULL,
  norm_continuous_format = "{mean} ± {SD}",
  unnorm_continuous_format = "{median} ({Q1}, {Q3})",
  categorical_format = "{n} ({pct})",
  binary_show = "last",
  digit = 2
)

Arguments

Value

A data frame containing summary statistics with the following columns:

• variable: Variable name

• Overall (n=X): Summary statistics for all data, if add_overall=TRUE

• Group-specific columns named ⁠[group] (n=X)⁠ with summary statistics

Examples

# `data` is a data frame processed by `add_var()`:
data <- add_var(iris, var = c("Sepal.Length", "Species"), group = "Species")
# Add summary statistics
result <- add_summary(data, add_overall = TRUE)
result <- add_summary(data, continuous_format = "{mean}, ({SD})")

Prepare variables for add_summary

Description

This function processes a dataset for statistical analysis by categorizing variables into continuous and categorical types. It automatically handles normality checks, equality of variances checks, and expected frequency assumptions checks.

Usage

add_var(data, var = NULL, group = "group", norm = "auto", center = "median")

Arguments

Value

A modified data frame with an attribute 'add_var' containing a list of categorized variables and their properties:

• var: List of categorized variables:

  • valid: All valid variable names after checks

  • continuous: Sublist of continuous variables (further divided by normality/equal variance)

  • categorical: Sublist of categorical variables (further divided by ordered/expected frequency)

• group: Grouping variable name

• overall_n: Total number of observations

• group_n: Observation counts per group

• group_nlevels: Number of groups

• group_levels: Group level names

• norm: Normality check method used

Examples

data <- add_var(iris, var = c("Sepal.Length", "Species"), group = "Species")

Test for Equality of Variances

Description

Performs Levene's test to assess equality of variances between groups.

Usage

equal_test(data, var, group, center = "median")

Arguments

Value

Logical value:

• TRUE: Variances are equal, p-value more than 0.05

• FALSE: Variances are unequal or an error occurred during testing

Methodology for Equality of Variances

Levene's test is the default method adopted in SPSS, the original Levene's test select center = mean, but here select center = median for a more robust test

Examples

equal_test(iris, "Sepal.Length", "Species")

Format p-values with significance markers

Description

Formats p-values as strings with specified precision and optional significance asterisks.

Usage

format_p(p, digit = 3, asterisk = FALSE)

Arguments

Value

Character of formatted p-value or asterisks.

Examples

format_p(0.00009, 4)
format_p(0.03, 3)
format_p(0.02, asterisk = TRUE)

Perform normality test on a variable

Description

Conducts normality tests for a specified variable, optionally by group. Supports automatic testing and interactive visualization.

Usage

normal_test(data = NULL, var = NULL, group = NULL, norm = "auto")

Arguments

Value

A logical value:

• TRUE: data are normally distributed

• FALSE: data are not normally distributed

Methodology for p-values

Automatically selects test based on sample size per group:

• n < 3: Too small, assuming non-normal

• (3, 50] Shapiro-Wilk test

• (50, 1000]: D'Agostino Chi2 test, instead of Kolmogorov-Smirnov test

• n > 1000: Show p-values, plots QQ plots and prompts for decision

Examples

normal_test(iris, "Sepal.Length", "Species", norm = "auto")
normal_test(iris, "Sepal.Length", "Species", norm = TRUE)

Check Sample Size Adequacy for Chi-Squared Test

Description

This function determines if a contingency table meets the expected frequency assumptions for a valid chi-squared test. It categorizes the data into "not_small", "small", or "very_small" based on sample size and expected frequencies.

Usage

small_test(data, var, group)

Arguments

Value

A character string with one of three values:

• "not_small": Sample size more than or euqal to 40 and all expected frequencies more than or euqal to 5

• "small": Sample size more than or euqal to 40, all expected frequencies more than or euqal to 1 and at least one <5, only for 2*2 contingency tables

• "very_small": Other conditions, including sample size <40 or any expected frequency <1

Examples

df <- data.frame(
  category = factor(c("A", "B", "A", "B")),
  group    = factor(c("X", "X", "Y", "Y"))
)
small_test(data = df, var = "category", group = "group")