Getting Started with exametrika

In exametrika: Test Data Engineering

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.width = 7,
  fig.height = 5,
  out.width = "100%"
)

Overview

The exametrika package provides comprehensive Test Data Engineering tools for analyzing educational test data. Based on the methods described in Shojima (2022), this package enables researchers and practitioners to:

• Analyze test response patterns and item characteristics
• Classify respondents using various psychometric models
• Investigate latent structures in test data
• Examine local dependencies between items
• Perform network analysis of item relationships

Installation

# Install from CRAN
install.packages("exametrika")

# Or install the development version from GitHub
if (!require("devtools")) install.packages("devtools")
devtools::install_github("kosugitti/exametrika")

Data Format

library(exametrika)

Data Requirements

Exametrika accepts both binary and polytomous response data:

• Binary data (0/1): 0 = incorrect, 1 = correct
• Polytomous data: ordinal response categories or multiple score levels
• Missing values: NA values supported; custom missing value codes can be specified

Data Formatting

The dataFormat() function processes input data before analysis:

# Format raw data for analysis
data <- dataFormat(J15S500)
str(data)

Sample Datasets

The package includes sample datasets from Shojima (2022). The naming convention is JxxSxxx where J = number of items and S = sample size.

| Dataset | Items | Examinees | Type | Use Case | |---------|------:|----------:|------|----------| | J5S10 | 5 | 10 | Binary | Quick testing | | J5S1000 | 5 | 1,000 | Ordinal | GRM examples | | J12S5000 | 12 | 5,000 | Binary | LDLRA examples | | J15S500 | 15 | 500 | Binary | IRT, LCA examples | | J15S3810 | 15 | 3,810 | Ordinal (4-point) | Ordinal LRA | | J20S400 | 20 | 400 | Binary | BNM examples | | J20S600 | 20 | 600 | Nominal (4-cat) | Nominal Biclustering | | J35S500 | 35 | 500 | Ordinal (5-cat) | Ordinal Biclustering | | J35S515 | 35 | 515 | Binary | Biclustering, network models | | J35S5000 | 35 | 5,000 | Multiple-choice | Rated LRA | | J50S100 | 50 | 100 | Binary | Small sample testing |

Basic Statistics

Test Statistics

TestStatistics(J15S500)

Item Statistics

ItemStatistics(J15S500)

Classical Test Theory

CTT(J15S500)

Next Steps

• Item Response Theory: IRT and GRM models
• Latent Class and Rank Analysis: LCA and LRA
• Biclustering: Biclustering, Ranklustering, and IRM
• Network Models: BNM, LDLRA, LDB, and BINET

Reference

Shojima, Kojiro (2022) Test Data Engineering: Latent Rank Analysis, Biclustering, and Bayesian Network (Behaviormetrics: Quantitative Approaches to Human Behavior, 13), Springer.

exametrika documentation built on Aug. 20, 2026, 5:08 p.m.