Package {ThesiStats}

Version:	1.1.0
Title:	Statistical Tools for Quantitative Theses
Description:	Provides helpers for the analyses that quantitative theses in the social and behavioral sciences repeat: renaming and scoring items, recoding Likert responses, cleaning sociodemographic variables written in Spanish (age, academic term, degree and university), descriptive statistics, univariate and multivariate normality checks, omega reliability from ordinal confirmatory factor models, correlation matrices in table format, and two-group or several-group comparisons with effect sizes.
License:	GPL (≥ 3)
URL:	https://github.com/jventural/ThesiStats
BugReports:	https://github.com/jventural/ThesiStats/issues
Encoding:	UTF-8
Depends:	R (≥ 4.1.0)
Imports:	broom, dplyr, ggplot2 (≥ 3.4.0), grid, gridExtra, gtable, lavaan, psych, readr, rlang, semTools, stats, stringdist, stringi, stringr, tibble, tidyr, utils
Suggests:	readxl, WRS2
LazyData:	true
RoxygenNote:	7.3.2
NeedsCompilation:	no
Packaged:	2026-09-23 18:18:12 UTC; PC
Author:	José Ventura-León [aut, cre]
Maintainer:	José Ventura-León <jventuraleon@gmail.com>
Repository:	CRAN
Date/Publication:	2026-10-05 15:30:07 UTC

ThesiStats: Statistical Tools for Quantitative Theses

Description

Helpers for the analyses that quantitative theses in the social and behavioral sciences repeat: renaming and scoring items, recoding Likert responses, cleaning sociodemographic variables, descriptive statistics, normality checks, reliability, correlations and group comparisons with effect sizes.

Author(s)

Maintainer: José Ventura-León jventuraleon@gmail.com (ORCID)

See Also

Useful links:

• https://github.com/jventural/ThesiStats

• Report bugs at https://github.com/jventural/ThesiStats/issues

Compare Two Groups with Welch's t Test and Cohen's d

Description

For each variable in cols, compares the two groups defined by group_var with Welch's t test and reports the means and standard deviations of each group, the t statistic, its degrees of freedom, the p value and Cohen's d with a verbal interpretation.

Usage

Calcule_Comparative(data, cols, group_var, Robust = FALSE)

Arguments

Details

The two groups are taken in the order of levels(factor(data[[group_var]])), the same order used by stats::t.test(), so the means, standard deviations, the difference and the sign of d always refer to the same group. The interpretation uses |d| > .80 "Grande", > .50 "Mediano", > .30 "Pequeno" and "Trivial" otherwise.

Value

A data frame with one row per variable and the columns Variables_interes, ⁠<group 1>(SD1)⁠, ⁠<group 2>(SD2)⁠, t, gl, p, d_cohen and Interpretacion.

Examples

set.seed(1)
df <- data.frame(
  grupo = rep(c("Mujer", "Varon"), each = 50),
  ansiedad = c(rnorm(50, 20, 4), rnorm(50, 18, 4)),
  depresion = c(rnorm(50, 15, 3), rnorm(50, 15, 3))
)
Calcule_Comparative(df, cols = c("ansiedad", "depresion"), group_var = "grupo")

Recode Specified Values in Selected Columns

Description

Applies recoding rules, written as two-sided formulas "old" ~ "new", to one or more columns of a data frame, replacing exact matches of each old value with the new one.

Usage

Correct_category(df, cols, ...)

Arguments

Value

df with every occurrence of each old value replaced by the new value in the selected columns. Other columns are unchanged.

Examples

df <- data.frame(
  Q1 = c("Yes", "No", "yes", "No"),
  Q2 = c("Maybe", "maybe", "No", "Yes")
)
Correct_category(df, Q1:Q2, "yes" ~ "Yes", "maybe" ~ "Maybe")

Omega Reliability of One Factor

Description

Fits a one-factor confirmatory model to the items in vars, treating them as ordinal (WLSMV estimator) and returns the composite reliability (omega) computed by semTools::compRelSEM().

Usage

Fiabilidad(vars, data)

Arguments

Value

A named numeric value: the omega coefficient of the factor.

See Also

calcula_omega_all() for several factors at once.

Examples

set.seed(123)
n <- 300
eta <- rnorm(n)
items <- as.data.frame(sapply(1:4, function(j) as.numeric(cut(
  0.7 * eta + rnorm(n, 0, 0.7), c(-Inf, -1.5, -0.5, 0.5, 1.5, Inf)))))
names(items) <- paste0("ANS", 1:4)

Fiabilidad(vars = names(items), data = items)

Q-Q Plot of Mahalanobis Distances with Mardia's Coefficients

Description

Plots the squared Mahalanobis distances of the observations against the quantiles of a chi-square distribution (a check of multivariate normality) and overlays a table with Mardia's skewness and kurtosis tests from mardia_test().

Usage

Multivariate_plot(data, xmin = 30, xmax = 40, ymin = 2, ymax = 7)

Arguments

Value

A ggplot object.

Examples

set.seed(1)
df <- as.data.frame(matrix(rnorm(200 * 4), ncol = 4))
Multivariate_plot(df, xmin = 8, xmax = 14, ymin = 1, ymax = 5)

Omega Reliability of Several Factors

Description

Applies Fiabilidad() to every factor of a list of item names and returns the omega coefficient of each one.

Usage

calcula_omega_all(extracted, data)

Arguments

Value

A data frame with the columns Variables (factor name) and Omega.

Examples

set.seed(123)
n <- 300
eta <- matrix(rnorm(n * 2), n, 2) %*% chol(0.7 * diag(2) + 0.3)
items <- as.data.frame(sapply(1:8, function(j) as.numeric(cut(
  0.7 * eta[, ceiling(j / 4)] + rnorm(n, 0, 0.7),
  c(-Inf, -1.5, -0.5, 0.5, 1.5, Inf)))))
names(items) <- c(paste0("ANS", 1:4), paste0("DEP", 1:4))

factores <- extract_items(c("Ansiedad: ANS1, ANS2, ANS3, ANS4",
                            "Depresion: DEP1, DEP2, DEP3, DEP4"))
calcula_omega_all(factores, items)

Correlation Matrix of a Range of Columns

Description

Computes the Spearman or Pearson correlation matrix of the columns that go from columna_inicial to columna_final, optionally with Winsorized Pearson correlations and significance marks, and returns it in the lower triangular format used in thesis tables.

Usage

calcular_correlaciones(
  data,
  columna_inicial,
  columna_final,
  method = c("spearman", "pearson"),
  winsorize = FALSE,
  show_pval = FALSE
)

Arguments

Value

A list with correlation (data frame with the correlations in the lower triangle, "-" on the diagonal and NA above it) and p_values (the p values in the same format).

Examples

set.seed(1)
x <- rnorm(80)
df <- data.frame(ansiedad = x, depresion = 0.5 * x + rnorm(80),
                 estres = 0.3 * x + rnorm(80))
calcular_correlaciones(df, "ansiedad", "estres", method = "pearson",
                       show_pval = TRUE)

Frequencies and Percentages of Categorical Variables

Description

For each variable in columnas, counts the observations of each category and its percentage of the total.

Usage

calcular_porcentajes(data, columnas)

Arguments

Value

A named list with one tibble per variable, holding the categories, their counts (n) and percentages (Porcentaje).

Examples

df <- data.frame(sexo = c("F", "M", "F", "F", "M"),
                 ciclo = c(1, 2, 2, 3, 3))
calcular_porcentajes(df, c("sexo", "ciclo"))

Descriptive Statistics of a Range of Columns

Description

Computes, with psych::describe(), the mean, standard deviation, minimum, maximum, skewness and kurtosis of the columns from start_col to end_col, plus the mean as a percentage of the maximum.

Usage

calculate_descriptives(data, start_col, end_col)

Arguments

Value

A data frame with the columns Variables, Media, DE, Min., Max., g1 (skewness), g2 (kurtosis) and ⁠%⁠, rounded to two decimals.

Examples

set.seed(1)
df <- data.frame(ansiedad = rnorm(50, 20, 4), depresion = rnorm(50, 15, 3))
calculate_descriptives(df, "ansiedad", "depresion")

Catalog of Peruvian University Degrees

Description

Reference list of undergraduate degrees and their faculties, used by normalize_carreras() to standardize free-text degree names.

Usage

carreras_peruanas

Format

A tibble with 93 rows and 2 columns:

Carrera

Standard name of the degree.

Facultad

Faculty the degree belongs to.

Source

Compiled by the package author from the degree offer of Peruvian universities.

Clean Age Strings into Fractional Years

Description

Standardizes a character column with ages written in Spanish (for example "19 años", "1 año 6 meses", "03 meses" or "2.5"): extracts years and months, converts months to fractions of a year and returns a numeric column. Rows without any digit are removed and reported with a message.

Usage

clean_edad(df, col_name = "Edad", round_decimals = 2)

Arguments

Value

df without the rows that had no digits, and with col_name replaced by the age in years.

Examples

df <- data.frame(Edad = c("19 años", "1 año 6 meses", "03 meses", "2.5",
                          "sin dato"))
clean_edad(df)

Convert Age Strings to Numeric Years

Description

Parses a character column with ages written in Spanish (for example "2 años", "un año" or "3,5") into numeric years. Unlike convert_age_to_years_months(), months are not added.

Usage

convert_age_to_years(
  df,
  col_name = "Edad",
  round_decimals = 2,
  drop_missing = FALSE
)

Arguments

Value

A list with cleaned_df (the data frame with col_name converted) and verify_df (a tibble pairing each original string with its value).

Examples

df <- data.frame(Edad = c("20 años", "un año", "3,5", "sin dato"))
res <- convert_age_to_years(df)
res$verify_df

Convert Age Strings with Years and Months to Numeric Years

Description

Parses a character column with ages written in Spanish (for example "2 años", "6 meses", "1 año 3 meses" or "3,5") into numeric years, expressing months as fractions of a year.

Usage

convert_age_to_years_months(
  df,
  col_name = "Edad",
  round_decimals = 2,
  drop_missing = FALSE
)

Arguments

Value

A list with cleaned_df (the data frame with col_name converted) and verify_df (a tibble pairing each original string with its value).

Examples

df <- data.frame(Edad = c("2 años", "6 meses", "1 año 3 meses", "3,5"))
res <- convert_age_to_years_months(df)
res$verify_df

Format Likert Response Options as a Single String

Description

Turns a data frame of Likert response options and their scores into one string of the form "0. Option; 1. Option; ...", ordered by score. This is the format read by remplace_alternative_response().

Usage

convert_to_expresions(df)

Arguments

Value

A character string.

Examples

likert <- data.frame(
  Alternativas = c("Nunca", "A veces", "Siempre"),
  score = c(0, 1, 2)
)
convert_to_expresions(likert)

Detect Likert Response Options and Assign Scores

Description

Looks in a data frame for the response options listed in likert_levels (ignoring case, accents and extra spaces), keeps those that appear, orders them as in likert_levels and assigns each one a score starting at 0 or 1.

Usage

detect_expression_Likert(
  df,
  start_zero = TRUE,
  likert_levels = c("Completamente en desacuerdo", "En desacuerdo", "Me es indiferente",
    "De acuerdo", "Completamente de acuerdo")
)

Arguments

Value

A tibble with the columns Alternativas (the options found, as an ordered factor) and score. A warning lists the options of likert_levels that do not appear in df.

Examples

respuestas <- data.frame(
  P1 = c("De acuerdo", "En desacuerdo", "Me es indiferente"),
  P2 = c("Completamente de acuerdo", "De acuerdo", "En desacuerdo")
)
suppressWarnings(detect_expression_Likert(respuestas))

Lower Triangular Format for a Correlation Matrix

Description

Numbers the rows ("1. Name") and columns (1, 2, ...) of a square matrix, puts "-" on the diagonal and NA above it, the usual layout of a correlation table in a thesis.

Usage

diag_aba_na(matriz)

Arguments

Value

A character matrix with the lower triangle of matriz.

Examples

m <- round(cor(mtcars[, 1:4]), 2)
diag_aba_na(m)

Summary Statistics of an Age Column

Description

Computes the mean, standard deviation, minimum and maximum of an age column.

Usage

edad_stat(obj, columna)

Arguments

Value

A one-row data frame with Media, DesviacionEstandar, Minimo and Maximo.

Examples

df <- data.frame(Edad = c(18, 20, 22, 25, 30))
edad_stat(df, Edad)
edad_stat(df, "Edad")

Kruskal-Wallis Test with Epsilon Squared

Description

Runs a Kruskal-Wallis test and adds its effect size, epsilon squared \epsilon^2 = H / (n - 1), with a verbal interpretation.

Usage

epsilon_cuadrado_kruskal(data, formula)

Arguments

Details

n counts the cases with non-missing outcome and group, the same cases used by the test. The interpretation is "Grande" for \epsilon^2 >= .50, "Mediano" >= .30, "Pequeño" >= .10 and "No significativo" otherwise.

Value

A tibble with the test results (statistic, p.value, parameter, method), Epsilon and Interpretación.

Examples

df <- data.frame(
  grupo = rep(c("A", "B", "C"), each = 5),
  puntaje = c(3, 4, 5, 4, 3, 7, 8, 6, 7, 8, 10, 12, 11, 9, 10)
)
epsilon_cuadrado_kruskal(df, puntaje ~ grupo)

Extract the Items of Each Factor from Text Lines

Description

Reads lines of the form "Factor: Item1, Item2, Item3" and returns a named list with the items of each factor.

Usage

extract_items(text, prefix = "")

Arguments

Value

A named list: one character vector of item names per factor.

Examples

extract_items(c("Ansiedad: ANS1, ANS2, ANS3", "Depresion: DEP1, DEP2"))

Add the Sum Score of Each Factor to a Data Frame

Description

Reads lines of the form "Factor: Item1, Item2, ..." and adds to df one column per factor with the row sum of its items.

Usage

generate_and_apply(df, text_lines, new_name = NULL)

Arguments

Value

df with one new column per factor (the row sum of its items, ignoring missing values). The item columns are converted to numeric.

Examples

df <- data.frame(A1 = c(1, 2, 3), A2 = c(2, 2, 2), B1 = c(0, 1, 1))
generate_and_apply(df, c("Ansiedad: A1, A2", "Estres: B1"))

Generate dplyr Code that Adds Factor Sum Scores

Description

Reads text with one line per factor ("Factor: Item1, Item2, ...") and returns, as a string, the dplyr code that adds the row sum of each factor to a data frame. The code can be printed with cat() and pasted into a script.

Usage

generate_code(text, name = "df_new_renombrado")

Arguments

Value

A character string with R code.

Examples

codigo <- generate_code("Ansiedad: A1, A2, A3\nEstres: B1, B2", name = "datos")
cat(codigo)

Boxplots of Several Variables

Description

Draws one boxplot per variable, each in its own panel with its own scale, to inspect the distribution and outliers of the scores.

Usage

grafico_boxplots(data, cols)

Arguments

Value

A ggplot object.

Examples

set.seed(1)
df <- data.frame(ansiedad = rnorm(60, 20, 4), depresion = rexp(60, 0.2))
grafico_boxplots(df, c("ansiedad", "depresion"))

Mardia's Test of Multivariate Normality

Description

Computes Mardia's multivariate skewness and kurtosis tests with psych::mardia() and reports whether each one is compatible with multivariate normality (p >= .05).

Usage

mardia_test(data)

Arguments

Value

A data frame with the columns Test, Statistic, p.value (formatted, "p < .001" when p <= .001) and Result ("YES" when normality is not rejected).

Examples

set.seed(1)
df <- as.data.frame(matrix(rnorm(200 * 3), ncol = 3))
mardia_test(df)

Shapiro-Wilk Normality Test for Several Variables

Description

Applies the Shapiro-Wilk test to each variable and classifies it as "Normal" (p >= .05) or "No-normal".

Usage

normality_test_SW(data, variables)

Arguments

Value

A tibble with the columns Variables, Shapiro-Wilk (the W statistic), p.value (formatted, "p < .001" when p < .001) and Normality.

Examples

set.seed(1)
df <- data.frame(ansiedad = rnorm(60), depresion = rexp(60))
normality_test_SW(df, c(ansiedad, depresion))

Normalize Peruvian University Degree Names

Description

Matches free-text degree names (for example "psicologia", "Ing. de Sistemas") against the catalog carreras_peruanas with fuzzy string matching, and adds a column with the standardized name (and optionally the faculty).

Usage

normalize_carreras(
  df,
  col_name = "Carrera",
  max_dist = 0.15,
  facultad = FALSE,
  manual_path = NULL,
  force_match = FALSE,
  fallback_dist = 0.4,
  remove_unmatched = FALSE
)

Arguments

Value

df with the column ⁠<col_name>_norm⁠ (and ⁠<col_name>_facultad⁠ when facultad = TRUE) placed after col_name. Messages report the values that could not be normalized.

Examples

df <- data.frame(Carrera = c("psicologia", "Derecho", "zzz"))
normalize_carreras(df)

Normalize the Academic Term (Ciclo) to an Integer

Description

Converts free-text academic terms in Spanish ("3er ciclo", "V", "quinto", "10") into integers. Rows that cannot be converted are removed and listed in a message.

Usage

normalize_ciclo(df, col_name = "Ciclo")

Arguments

Value

df without the rows that could not be converted, and with col_name as an integer.

Examples

df <- data.frame(Ciclo = c("3er ciclo", "V", "quinto", "10", "no se"))
normalize_ciclo(df)

Normalize the Length of a Relationship to Months

Description

Converts free-text durations in Spanish ("1 año y 6 meses", "2 semanas", "un año y medio", "8") into months.

Usage

normalize_tiempo_relacion(df, col_name = "Tiempo_Relacion", remover = TRUE)

Arguments

Value

A data frame with the new column ⁠<col_name>_norm⁠ (months).

Examples

df <- data.frame(Tiempo_Relacion = c("1 año y 6 meses", "2 semanas", "8"))
normalize_tiempo_relacion(df)

Normalize Peruvian University Names to Their Acronyms

Description

Matches free-text university names or acronyms against the catalog universidades_peruanas with Jaro-Winkler distance and adds a column with the standard acronym.

Usage

normalize_universidades(df, col_name)

Arguments

Value

df with the column ⁠<col_name>_norm⁠ (the acronym, or NA when there is no match within a distance of .15) placed after col_name.

Examples

df <- data.frame(U = c("Universidad Nacional Mayor de San Marcos", "unmsm"))
normalize_universidades(df, "U")

Detect and Score Several Blocks of Likert Items

Description

For each block of items, detects the response options with detect_expression_Likert(), builds the score mapping with convert_to_expresions() and replaces the text responses by their scores with remplace_alternative_response().

Usage

process_likert_blocks(df, specs)

Arguments

Value

df with the items of every block scored.

Examples

df <- data.frame(ANS1 = c("Nunca", "Siempre", "A veces"),
                 ANS2 = c("A veces", "Nunca", "Siempre"))
process_likert_blocks(df, list(
  list(prefix = "ANS", n_items = 2, start_zero = TRUE,
       levels = c("Nunca", "A veces", "Siempre"))
))

Replace Likert Text Responses with Their Scores

Description

Converts text responses into numbers in one or more blocks of item columns. Each block is described by its first and last column and by a string that maps each response option to its score ("0. Nunca; 1. A veces; ...", the format returned by convert_to_expresions()).

Usage

remplace_alternative_response(df, columnas_valores_entradas)

Arguments

Value

df with the responses of the listed columns replaced by their scores; responses not found in the mapping become NA.

Examples

df <- data.frame(P1 = c("Nunca", "Siempre"), P2 = c("A veces", "Nunca"))
remplace_alternative_response(
  df, list(list(c("P1", "P2"), "0. Nunca; 1. A veces; 2. Siempre"))
)

Rename Columns by Position

Description

Renames the columns at the positions given in columns with new_names.

Usage

rename_columns(df, new_names, columns)

Arguments

Value

df with the columns renamed.

Examples

df <- data.frame(a = 1:2, b = 3:4, c = 5:6)
rename_columns(df, c("Edad", "Sexo"), 1:2)

Rename a Block of Item Columns with Two Prefixes

Description

Renames the consecutive columns from inicio to final: the first n_items1 as prefix1 1, 2, ... and the next n_items2 as prefix2 1, 2, ... If the counts are not given, the block is split in halves.

Usage

rename_items2(
  df,
  prefix1 = "COPE",
  prefix2 = "E",
  inicio = NULL,
  final = NULL,
  n_items1 = NULL,
  n_items2 = NULL
)

Arguments

Value

df with the block renamed.

Examples

df <- data.frame(p1 = 1, p2 = 2, p3 = 3, p4 = 4, p5 = 5)
rename_items2(df, prefix1 = "ANS", prefix2 = "DEP", n_items1 = 3)

Rename a Block of Item Columns with Three Prefixes

Description

Like rename_items2() with three groups of items. If the counts are not given, the block is split in thirds.

Usage

rename_items3(
  df,
  prefix1 = "COPE",
  prefix2 = "E",
  prefix3 = "F",
  inicio = NULL,
  final = NULL,
  n_items1 = NULL,
  n_items2 = NULL,
  n_items3 = NULL
)

Arguments

Value

df with the block renamed.

Examples

df <- data.frame(p1 = 1, p2 = 2, p3 = 3, p4 = 4, p5 = 5, p6 = 6)
rename_items3(df, prefix1 = "A", prefix2 = "B", prefix3 = "C")

Rename a Block of Item Columns with One Prefix

Description

Renames the consecutive columns from inicio to final as prefix1 followed by 1, 2, 3, ...

Usage

rename_items_only(
  df,
  prefix1 = "COPE",
  inicio = NULL,
  final = NULL,
  n_items1 = NULL
)

Arguments

Value

df with the block renamed.

Examples

df <- data.frame(id = 1:2, p1 = 1:2, p2 = 3:4, p3 = 5:6)
rename_items_only(df, prefix1 = "ANS", inicio = "p1", final = "p3")

Mann-Whitney U Test with the Probability of Superiority

Description

Runs a Mann-Whitney U test (Wilcoxon rank-sum) between two groups and adds the probability of superiority, PS = U / (n_1 n_2): the probability that a random case of the first group scores higher than a random case of the second.

Usage

u_mann_whitney_superioridad(data, formula, alternative = "two.sided")

Arguments

Details

PS = .50 means no effect. The size is judged on the distance from .50 in either direction, using max(PS, 1 - PS): at least .71 is "Grande", at least .64 "Mediano", at least .56 "Pequeño" and below that "No efecto".

Value

A tibble with the test results (statistic is U for the first group, p.value, method, alternative), PSest and Interpretación.

Examples

set.seed(1)
df <- data.frame(grupo = rep(c("A", "B"), each = 30),
                 puntaje = c(rnorm(30, 10), rnorm(30, 11)))
u_mann_whitney_superioridad(df, puntaje ~ grupo)

Catalog of Peruvian Universities

Description

Reference list of Peruvian universities and their acronyms, used by normalize_universidades().

Usage

universidades_peruanas

Format

A tibble with 99 rows and 2 columns: the full name of the university (Nombre) and its acronym (second column).

Source

Compiled by the package author from public lists of Peruvian universities.

Unique Values of Several Columns

Description

Lists, sorted, the distinct values that appear in the selected columns, useful to check the categories before recoding them with Correct_category().

Usage

validation_categoria(df, cols)

Arguments

Value

A sorted vector with the distinct values.

Examples

df <- data.frame(Q1 = c("Si", "No", "si"), Q2 = c("No", "Tal vez", "Si"))
validation_categoria(df, Q1:Q2)