SaNaSoft fsQCA
Free · v1.7Guided fuzzy-set QCA: calibration, necessary conditions, truth table, complex, parsimonious and intermediate solutions, robustness and predictive checks, and an APA-style Word report. Benchmarked against the R package QCA.
From raw questionnaire data to publication-ready results — in one guided workflow.
SaNaPLS takes you from an uploaded survey file to latent variable scores, bootstrapped path coefficients, fsQCA configurations, neural-network importance rankings and cIPMA — with every cut-off traced to a published source. Nothing you upload ever leaves your browser.
Quantitative Analysis — the SaNaPLS dashboard: survey data, PLS-SEM, fsQCA, ANN and cIPMA.
Thematic Analysis — SaNaQual, for interviews and open-ended answers, including follow-up interviews that explain your PLS-SEM results. Version 1.2.0: coding, themes, intercoder agreement, checklists and a Word report.
Statistics & Machine Learning — SaNaStat, for classical tests and explainable machine learning that adds to your PLS-SEM results. Version 1.0.0 (formerly SaNaML): data quality screening, data preparation (formulas, recoding, select cases, split file, weights, restructure), common statistical tests with effect-size confidence intervals and the usual post hoc tests, stepwise regression, layered cross-tables, rater agreement (kappa, ICC), factor analysis (EFA, CFA, ω, AVE, HTMT), mediation and moderation, ordinal, multinomial and count regression, multiple imputation, power analysis, HR analytics (survival, adverse impact, pay gaps) and marketing analytics (A/B tests, RFM, market basket, conjoint), explainable machine learning with fairness checks, a report builder, and projects you can version and share.
SaNaStat manuals: Getting started · Common Statistical Tests · Marketing Analytics · HR Analytics · practice datasets
No account. No install. No data leaves your device.
✓ Done
Latent variable scores
PLS-SEM path model, reliability and validity
Ready to use
fsQCA
Configurations for high and low outcomes
Ready to use
cIPMA (IPMA + NCA)
What matters most, and what's necessary
Why not just use what you have
Built to replace a folder of separate tools and licences
Nothing to install
Runs in any modern browser — Windows, Mac or Linux, no setup.
Your data stays yours
Every calculation runs on your machine. Nothing is uploaded to a server.
Works offline
After the first visit, keep running analyses without a connection.
One shared dataset
Upload once; every tool below reads the same file and the same constructs.
The workflow
Four stages, in order — the same ones on your dashboard
1
Prepare your data
Bring in a CSV, Excel or SPSS file, confirm what each column is, and screen the data once.
Import & screening
2
Describe
Summarise your sample, cross-tabulate demographics, check every item and score for normality, and explore which questions belong together with a factor analysis.
Demographics & descriptivesNormality testsExploratory factor analysis
3
Estimate your model
Calculate latent variable scores with PLS-SEM, then check reliability, validity and path significance with bootstrapping, and look for hidden groups of respondents with FIMIX-PLS and PLS-POS. Run correlations, linear regression and logistic regression on your uploaded data or on the PLS-SEM scores.
Latent variable scoresCorrelation & linear regressionLogistic regression
4
Supplementary analysis
Go beyond the path model: fsQCA configurations, neural-network importance rankings, and cIPMA — built on your estimated model or uploaded scores.
fsQCAArtificial neural networkcIPMA (IPMA + NCA)
Stand-alone apps
Only need fsQCA or ANN?
If you already have latent variable scores from your PLS-SEM software, you can use our classic single-purpose apps. They run in your browser and need no sign-in.
SaNaSoft-ANN
Free · v2.3Guided artificial neural network analysis for the PLS-SEM + ANN hybrid approach (Leong et al., 2025): ten networks, RMSE tables, sensitivity analysis, the PLS-SEM vs ANN comparison and a publication-ready report.
No guessing at thresholds
Every cut-off is traced to a published source
- Reliability and validity bands follow Hair et al.'s guidance, with the exact source named next to every number.
- Minimum sample size uses Kock & Hadaya's inverse-square-root method, matched value-for-value against established PLS-SEM software.
- fsQCA and NCA follow Ragin and Dul's published procedures, including robustness and predictive checks.
- The ANN stage follows Leong et al.'s published cross-validation pipeline.
- The Word report includes the figures too: the PLS-SEM path diagram, scree plot, heatmaps, regression and ROC plots.
- Every downloaded Word report ends with a "Criteria used" table listing every threshold applied and its source.
Reliability, validityHair et al.
Minimum sample sizeKock & Hadaya, 2018
Discriminant validityHenseler et al. (HTMT)
Regression checksHair et al. · Hosmer & Lemeshow
fsQCA / NCARagin · Dul
Neural networkLeong et al., 2025
Developed by
Built by PLS-SEM researchers, for PLS-SEM researchers
SN
Prof. S. Sabraz Nawaz
Professor in Management and IT
- Faculty
- Faculty of Management and Commerce
- University
- South Eastern University of Sri Lanka
- Location
- Oluvil, Sri Lanka
- sabraz@seu.ac.lk
- Phone
- +94 77 244 6699
GA
Prof. Ghazanfar Ali
Department of Management and Marketing
- School
- KFUPM Business School
- University
- King Fahd University of Petroleum & Minerals
- Location
- Saudi Arabia
- ghazanfarali.abbasi@kfupm.edu.sa
- Phone
- +966 50 785 6615