The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

Introduction to mariposa

library(mariposa)
library(dplyr)

What is mariposa?

mariposa (Marburg Initiative for Political and Social Analysis) is a comprehensive R package for professional survey data analysis. It covers the entire workflow — from importing SPSS, Stata, SAS, and Excel files through label management, recoding, and standardization to statistical analysis with survey weights and publication-ready output.

Every statistical result is validated against SPSS v29, so researchers migrating from SPSS can trust their numbers.

Key Features

The Example Dataset

mariposa includes survey_data, a synthetic survey of 2,500 respondents with demographics, attitudes, and a sampling weight:

data(survey_data)
glimpse(survey_data)
#> Rows: 2,500
#> Columns: 16
#> $ id                    <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 1…
#> $ age                   <dbl> 68, 58, 48, 46, 71, 73, 60, 48, 28, 30, 20, 58, …
#> $ gender                <fct> Female, Male, Male, Female, Male, Female, Male, …
#> $ region                <fct> East, West, West, West, West, East, East, West, …
#> $ education             <ord> Intermediate Secondary, Academic Secondary, Acad…
#> $ income                <dbl> 3500, 4800, 3500, 2600, 3000, 5200, 3200, NA, 37…
#> $ employment            <fct> Retired, Employed, Employed, Employed, Retired, …
#> $ political_orientation <int> 2, 3, 3, 5, 1, NA, NA, 3, 4, 4, 2, 2, 1, 3, 3, 1…
#> $ environmental_concern <int> 3, 5, 3, 2, NA, NA, 5, 4, 2, 4, 3, 5, 4, 4, 3, 3…
#> $ life_satisfaction     <int> 4, 3, 2, 2, 4, 4, 3, 3, 4, 3, 1, 1, 2, 5, 3, 2, …
#> $ trust_government      <int> 3, 4, 1, 1, 2, 1, 3, 3, 4, 3, NA, 4, 3, 3, 2, 2,…
#> $ trust_media           <int> 3, 3, 3, 2, 4, 4, 4, 1, 4, 2, 2, 1, 1, 3, 3, 3, …
#> $ trust_science         <int> 2, 4, 4, 1, 3, 5, 5, 3, 3, 4, 4, 3, 5, 4, 3, 4, …
#> $ sampling_weight       <dbl> 1.2690774, 0.8926824, 1.0424119, 1.0024385, 1.02…
#> $ stratum               <fct> East_Old, West_Old, West_Middle, West_Middle, We…
#> $ interview_mode        <fct> Face-to-face, Face-to-face, Online, Telephone, T…

All examples in this guide use this dataset.

Five-Minute Tour

Here is a complete analysis workflow showing what mariposa can do:

1. Explore the Data

# Find variables related to "trust"
find_var(survey_data, "trust")
#>   col             name                                    label
#> 1  11 trust_government Trust in government (1=none, 5=complete)
#> 2  12      trust_media      Trust in media (1=none, 5=complete)
#> 3  13    trust_science    Trust in science (1=none, 5=complete)
# Descriptive statistics with survey weights
survey_data %>%
  describe(age, income, life_satisfaction, weights = sampling_weight)
#> 
#> Weighted Descriptive Statistics
#> -------------------------------
#>           Variable     Mean Median       SD Range  IQR Skewness Effective_N
#>                age   50.514     50   17.084    77   25    0.159      2468.8
#>             income 3743.099   3500 1423.966  7200 1900    0.725      2158.9
#>  life_satisfaction    3.625      4    1.152     4    2   -0.499      2390.9
#> ----------------------------------------
# Frequency table
survey_data %>%
  frequency(education, weights = sampling_weight)
#> Frequency: education [Weighted]
#>   4 categories, N valid = 2516, missing = 0
#> Use summary() for detailed output.

2. Transform Variables

# Create age groups
survey_data <- rec(survey_data, age,
  rules = "18:29=1 [Young]; 30:49=2 [Middle]; 50:99=3 [Older]",
  suffix = "_group", as_factor = TRUE)

# Build a trust scale
survey_data <- survey_data %>%
  mutate(m_trust = row_means(., trust_government, trust_media, trust_science,
                             min_valid = 2))

3. Compare Groups

# t-test with survey weights
survey_data %>%
  t_test(life_satisfaction, group = gender, weights = sampling_weight)
#> t-Test: life_satisfaction by gender [Weighted]
#>   t(2391.3) = -1.069, p = 0.285 , g = -0.043 (negligible), N = 2436
# ANOVA across education levels
result <- survey_data %>%
  oneway_anova(life_satisfaction, group = education, weights = sampling_weight)
result
#> One-Way ANOVA: life_satisfaction by education [Weighted]
#>   F(3, 2432) = 65.333, p < 0.001 ***, eta2 = 0.075 (medium), N = 2437

Every result has a detailed view with summary():

summary(result, descriptives = FALSE)
#> Weighted One-Way ANOVA Results
#> ------------------------------
#> 
#> - Dependent variable: life_satisfaction
#> - Grouping variable: education
#> - Weights variable: sampling_weight
#> - Confidence level: 95.0%
#>   Null hypothesis: All group means are equal
#>   Alternative hypothesis: At least one group mean differs
#> 
#> 
#> --- life_satisfaction ---
#> 
#> 
#> Weighted ANOVA Results:
#> -------------------------------------------------------------------------------- 
#>          Source Sum_Squares   df Mean_Square      F p_value sig
#>  Between Groups     241.130    3      80.377 65.333   <.001 ***
#>   Within Groups    2992.019 2432        1.23                   
#>           Total    3233.149 2435                               
#> -------------------------------------------------------------------------------- 
#> 
#> Assumption Tests:
#> ---------------- 
#>  Assumption Statistic df1  df2 p_value sig
#>       Welch    62.636   3 1216   <.001 ***
#> 
#> Effect Sizes:
#> ------------ 
#>           Variable Eta_Squared Epsilon_Squared Omega_Squared Effect_Size
#>  life_satisfaction       0.075           0.073         0.073      medium
#> 
#> 
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05
#> 
#> Effect Size Interpretation:
#> - Eta-squared: Proportion of variance explained (biased upward)
#> - Epsilon-squared: Less biased than eta-squared
#> - Omega-squared: Unbiased estimate (preferred for publication)
#> - Small effect: eta-squared ~ 0.01, Medium effect: eta-squared ~ 0.06, Large effect: eta-squared ~ 0.14
#> 
#> Post-hoc tests: Use tukey_test() for pairwise comparisons

4. Post-Hoc Analysis

# Which education groups differ?
tukey_test(result)
#> Tukey HSD Post-Hoc Test by education [Weighted]
#>   life_satisfaction: 6 comparisons, 5 significant (p < .05)
#> Use summary() for the full comparison table.

5. Measure Relationships

survey_data %>%
  pearson_cor(age, income, life_satisfaction, weights = sampling_weight)
#> Pearson Correlation: 3 variables [Weighted]
#>   age x income:                  r = -0.005, p = 0.828  
#>   age x life_satisfaction:       r = -0.029, p = 0.150  
#>   income x life_satisfaction:    r = 0.450, p < 0.001 *** 
#>   1/3 pairs significant (p < .05), N = 2201

6. Build Models

survey_data %>%
  linear_regression(life_satisfaction ~ age + income + m_trust,
                    weights = sampling_weight)
#> Linear Regression: life_satisfaction ~ age + income + m_trust [Weighted]
#>   R2 = 0.201, adj.R2 = 0.200, F(3, 2109) = 177.02, p < 0.001 ***, N = 2113

Compact vs. Detailed Output

Every analysis function in mariposa provides two output levels:

You can toggle individual sections in the detailed output:

result <- survey_data %>%
  t_test(life_satisfaction, group = gender, weights = sampling_weight)

# Compact
result
#> t-Test: life_satisfaction by gender [Weighted]
#>   t(2391.3) = -1.069, p = 0.285 , g = -0.043 (negligible), N = 2436

# Detailed
summary(result)
#> Weighted t-Test Results
#> -----------------------
#> 
#> - Grouping variable: gender
#> - Groups compared: Male vs. Female
#> - Weights variable: sampling_weight
#> - Confidence level: 95.0%
#> - Alternative hypothesis: two.sided
#> - Null hypothesis (mu): 0.000
#> 
#> 
#> --- life_satisfaction ---
#> 
#>   Male: mean = 3.598, n = 1149.0
#>   Female: mean = 3.648, n = 1287.0
#> 
#> Weighted t-test Results:
#> -------------------------------------------------------------------------------- 
#>         Assumption t_stat       df p_value mean_diff        conf_int sig
#>    Equal variances -1.070 2434.609   0.285     -0.05 [-0.142, 0.042]    
#>  Unequal variances -1.069 2391.291   0.285     -0.05 [-0.142, 0.042]    
#> -------------------------------------------------------------------------------- 
#> 
#> Effect Sizes:
#> ------------ 
#>           Variable Cohens_d Hedges_g Glass_Delta Effect_Size
#>  life_satisfaction   -0.043   -0.043      -0.043  negligible
#> 
#> 
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05
#> 
#> Effect Size Interpretation:
#> - Cohen's d: pooled standard deviation (classic)
#> - Hedges' g: bias-corrected Cohen's d (preferred)
#> - Glass' Delta: control group standard deviation only
#> - Small effect: |effect| ~ 0.2
#> - Medium effect: |effect| ~ 0.5
#> - Large effect: |effect| ~ 0.8

# Detailed, skip effect sizes
summary(result, effect_sizes = FALSE)
#> Weighted t-Test Results
#> -----------------------
#> 
#> - Grouping variable: gender
#> - Groups compared: Male vs. Female
#> - Weights variable: sampling_weight
#> - Confidence level: 95.0%
#> - Alternative hypothesis: two.sided
#> - Null hypothesis (mu): 0.000
#> 
#> 
#> --- life_satisfaction ---
#> 
#>   Male: mean = 3.598, n = 1149.0
#>   Female: mean = 3.648, n = 1287.0
#> 
#> Weighted t-test Results:
#> -------------------------------------------------------------------------------- 
#>         Assumption t_stat       df p_value mean_diff        conf_int sig
#>    Equal variances -1.070 2434.609   0.285     -0.05 [-0.142, 0.042]    
#>  Unequal variances -1.069 2391.291   0.285     -0.05 [-0.142, 0.042]    
#> -------------------------------------------------------------------------------- 
#> 
#> 
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05

Grouped Analysis

All functions support dplyr::group_by() for subgroup analysis:

survey_data %>%
  group_by(region) %>%
  describe(income, life_satisfaction, weights = sampling_weight)
#> 
#> Weighted Descriptive Statistics
#> -------------------------------
#> 
#> Group: region = East
#> --------------------
#> ----------------------------------------
#>           Variable     Mean Median       SD Range  IQR Skewness Effective_N
#>             income 3760.687   3600 1388.321  7200 1700    0.721       421.9
#>  life_satisfaction    3.623      4    1.203     4    2   -0.558       457.4
#> ----------------------------------------
#> 
#> Group: region = West
#> --------------------
#> ----------------------------------------
#>           Variable     Mean Median       SD Range  IQR Skewness Effective_N
#>             income 3738.586   3500 1433.325  7200 1900    0.727      1738.1
#>  life_satisfaction    3.625      4    1.139     4    2   -0.481      1934.8
#> ----------------------------------------
survey_data %>%
  group_by(region) %>%
  t_test(income, group = gender, weights = sampling_weight)
#> [region = 1]
#> t-Test: income by gender [Weighted]
#>   t(431.2) = 1.674, p = 0.095 , g = 0.158 (negligible), N = 450
#> [region = 2]
#> t-Test: income by gender [Weighted]
#>   t(1740.2) = 0.009, p = 0.993 , g = 0.000 (negligible), N = 1751

Quick Reference

Data Import & Export

Function Purpose
read_spss(), read_por() Import SPSS files with tagged NA support
read_stata() Import Stata files
read_sas(), read_xpt() Import SAS files
read_xlsx() Import Excel files with label reconstruction
write_spss() Export to SPSS with label/missing roundtripping
write_stata() Export to Stata
write_xpt() Export to SAS transport format
write_xlsx() Export to Excel (data, codebook, frequencies)

Label Management

Function Purpose
var_label() Get/set variable labels
val_labels() Get/set value labels
find_var() Search variables by name or label
to_label() Labelled → factor
to_character() Labelled → character
to_numeric() Factor/labelled → numeric
to_labelled() Factor/character → labelled
set_na() Declare values as missing
unlabel() Strip all label metadata
copy_labels() Restore labels after dplyr operations
drop_labels() Remove unused value labels

Data Transformation

Function Purpose
rec() Recode with string syntax (ranges, reverse, median split)
to_dummy() One-hot encoding / dummy variables
std() Z-standardization (sd, 2sd, mad, gmd methods)
center() Mean-centering (grand-mean, group-mean)
row_means() Row-wise means with min_valid threshold
row_sums() Row-wise sums
row_count() Count specific values per row
pomps() Percent of Maximum Possible Scores (0–100)

Descriptive Statistics

Function Purpose
codebook() Interactive HTML data dictionary
describe() Numeric summaries (mean, sd, median, range, skewness)
frequency() Frequency tables with valid/cumulative percent
crosstab() Cross-tabulations with row/column/cell percentages

Hypothesis Testing

Function Purpose
t_test() Independent and one-sample t-tests
oneway_anova() One-way ANOVA
factorial_anova() Multi-factor ANOVA with Type III SS
ancova() ANCOVA with estimated marginal means
mann_whitney() Mann-Whitney U test
kruskal_wallis() Kruskal-Wallis H test
wilcoxon_test() Wilcoxon signed-rank test
friedman_test() Friedman test
binomial_test() Exact binomial test
chi_square() Chi-square test of independence
fisher_test() Fisher’s exact test
chisq_gof() Chi-square goodness-of-fit
mcnemar_test() McNemar’s test for paired proportions

Post-Hoc & Effect Sizes

Function Purpose
tukey_test() Tukey HSD pairwise comparisons
scheffe_test() Scheffe pairwise comparisons
levene_test() Test for homogeneity of variances
dunn_test() Dunn’s post-hoc for Kruskal-Wallis
pairwise_wilcoxon() Pairwise Wilcoxon for Friedman
phi() Phi coefficient
cramers_v() Cramer’s V
goodman_gamma() Goodman-Kruskal gamma

Scale Analysis

Function Purpose
reliability() Cronbach’s Alpha with item statistics
efa() Exploratory Factor Analysis (PCA/ML, Varimax/Oblimin/Promax)

Regression

Function Purpose
linear_regression() Linear regression with SPSS-style output
logistic_regression() Logistic regression with odds ratios

Weighted Statistics

Function Purpose
w_mean(), w_median(), w_sd(), w_var() Central tendency and spread
w_se(), w_quantile(), w_iqr(), w_range() Precision and distribution
w_skew(), w_kurtosis(), w_modus() Shape and mode

Guides

Explore the full documentation:

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.