JASP


JASP is a free and open-source program for statistical analysis supported by the University of Amsterdam. It is designed to be easy to use, and familiar to users of SPSS. It offers standard analysis procedures in both their classical and Bayesian form. JASP generally produces APA style results tables and plots to ease publication. It promotes open science via integration with the Open Science Framework and reproducibility by integrating the analysis settings into the results. The development of JASP is financially supported by sponsors, several universities, and research funds.
WWW Official Site

Overview

In recognition of Bayesian pioneer Sir Harold Jeffreys, JASP stands for Jeffreys’s Amazing Statistics Program.

Analyses

JASP offers frequentist inference and Bayesian inference on the same statistical models. Frequentist inference uses p-values and confidence intervals to control error rates in the limit of infinite perfect replications. Bayesian inference uses credible intervals and Bayes factors to estimate credible parameter values and model evidence given the available data and prior knowledge.The following analyses are available in JASP in comparison to SPSS:
JASP 0.95.xSPSS 31JASP 0.95.xSPSS 31
AnalysisClassicClassicBayesianBayesian
Acceptance Sampling: Attribute and Variable Sample Plans✓X
ANCOVA, repeated ANOVA, MANOVA and non-parametrics✓✓
Audit: tools for the auditing of organisations e.g. Benfords Law✓X✓X
BFpack, BFF, Bain,✓X
BSTS - Bayesian structural time series✓X
Circular / Directional Statistics - analysis of directions, often angles✓XXX
Cochrane Meta-Analyses including database query from within JASP✓X✓X
Descriptives including multiple modules for plot building ✓
Distributions: >40 discrete and continuous ones✓X✓X
Equivalence T-Tests : Independent, Paired, One-Sample✓X✓X
Factor Analysis including score export to data functionality✓✓ / AMOSXX
Frequencies ✓✓✓
JAGS ✓
LearnStats, esci ✓X✓X
Machine Learning: Regression, Classification, Cluster, Prediction / Time Series✓✓XX
Meta-Analysis for Multilevel/Multivariate/SEM ✓✓X
Mixed Models✓✓✓X
Network✓✓✓X
Power Analysis / Sample Size PlanningXX
PROCESS ✓✓✓X
Time Series Analysis: Descriptives, Stationarity, ARIMA, Spectral Analysis, Prophet, Predictive AnalyticsX✓✓X
Quality Control ✓XX
Regression / Correlation: r, Rho, Tau, linear, logistic, generalized linear, export residual functionality✓✓
Reliability ✓✓X
Structural Equation Modeling inkl. Partial Least Squares, Latent Growth & MIMIC✓AMOSXX
Summary StatisticsXX✓X
Survival Analyses ✓✓XX
T-Tests: Independent, Paired, One-Sample ✓✓✓
Visual Modeling: Automated Plotting, Linear, Mixed, Generalized Linear✓✓XX

Other features

  • Accessibility features
  • integrated help files and annotated data library examples for many analyses.
  • R syntax editing and highlighting.
  • Extensive plot and formula editing capabilities.
  • Exports results as PDF or HTML; tables can also be copy pasted in LaTeX format.; plots as PNG, PPTX etc.; data can be exported as CSV.
  • Imports R, Excel, SAS and SPSS files etc..
  • Connects and syncs to SQL data bases, the Cochrane data base and the Open Science Framework.
  • Data filtering: Use either R code or a drag-and-drop GUI to select cases of interest.
  • Recode data with only one click.
  • Full data editing with one-click recoding; full undo / redo functionality.
  • Compute columns with R code or a drag-and-drop GUI to create new variables or compute them from existing ones or with simulated data.
  • Empty values settings per variable, per data set or globally.
  • Assumption checks via export and then plotting of residuals and/or per analyses via tests and plots.

Modules

JASP features seven common modules that are enabled by default:
  1. Descriptives: Explore the data with tables and plots.
  2. T-Tests: Evaluate the difference between two means.
  3. ANOVA: Evaluate the difference between multiple means.
  4. Mixed Models: Evaluate the difference between multiple means with random effects.
  5. Regression: Evaluate the association between variables.
  6. Frequencies: Analyses for count data.
  7. Factor: Explore hidden structure in the data.
JASP also features multiple additional modules that can be activated via the module menu:
  1. Acceptance Sampling: Methods for acceptance sampling and a quality control setting.
  2. Audit: Statistical methods for auditing. The audit module offers planning, selection and evaluation of statistical audit samples, methods for data auditing and algorithm auditing.
  3. Bain: Bayesian informative hypotheses evaluation for t-tests, ANOVA, ANCOVA, linear regression and structural equation modeling.
  4. Bayes Factor Functions
  5. BFpack
  6. BSTS: Bayesian take on linear Gaussian state space models suitable for time series analysis.
  7. Circular Statistics: Basic methods for directional data.
  8. Cochrane Meta-Analyses: Analyse Cochrane medical datasets.
  9. Distributions: Visualise probability distributions and fit them to data.
  10. Equivalence T-Tests: Test the difference between two means with an interval-null hypothesis.
  11. JAGS: Implement Bayesian models with the JAGS program for Markov chain Monte Carlo.
  12. Learn Bayes: Learn Bayesian statistics with simple examples and supporting text
  13. Learn Stats: Learn classical statistics with simple examples and supporting text.
  14. Machine Learning: Explore the relation between variables using data-driven methods for supervised learning and unsupervised learning. The module contains 19 analyses for regression, classification and clustering:
  15. *Regression
  16. *#Boosting Regression
  17. *#Decision Tree Regression
  18. *#K-Nearest Neighbors Regression
  19. *#Neural Network Regression
  20. *#Random Forest Regression
  21. *#Regularized Linear Regression
  22. *#Support Vector Machine Regression
  23. *Classification
  24. *#Boosting Classification
  25. *#Decision Tree Classification
  26. *#K-Nearest Neighbors Classification
  27. *#Neural Network Classification
  28. *#Linear Discriminant Classification
  29. *#Random Forest Classification
  30. *#Support Vector Machine Classification
  31. *Clustering
  32. *#Density-Based Clustering
  33. *#Fuzzy C-Means Clustering
  34. *#Hierarchical Clustering
  35. *#Model-based clustering
  36. *#Neighborhood-based Clustering
  37. *#Random Forest Clustering
  38. *Prediction
  39. Meta Analysis: Synthesise evidence across multiple studies. Includes techniques for fixed and random effects analysis, fixed and mixed effects meta-regression, forest and funnel plots, tests for funnel plot asymmetry, trim-and-fill and fail-safe N analysis.
  40. Network: Explore the connections between variables organised as a network. Network Analysis allows the user to analyze the network structure.
  41. Power: Conduct power analyses and sample size planning.
  42. Predictive Analytics: This module offers predictive analytics.
  43. Process: Implementation of Hayes' popular SPSS PROCESS module for JASP
  44. Prophet: A simple model for time series prediction.
  45. Quality Control: Investigate if a manufactured product adheres to a defined set of quality criteria.
  46. Reliability: Quantify the reliability of test scores.
  47. Robust T-Tests: Robustly evaluate the difference between two means.
  48. SEM : Evaluate latent data structures with Yves Rosseel's lavaan program.
  49. Summary statistics: Apply common Bayesian tests from frequentist summary statistics for t-test, regression, and binomial tests.
  50. Survival Analyses: non-parametric, semi-parametric, parametric
  51. Time Series: Time series analysis with Descriptives, Stationarity, ARIMA, Spectral Analysis.
  52. Visual Modeling: Graphically explore the dependencies between variables.
  53. R Console: Execute R code in a console.