StatLabX powered by DOEpro 2.0.1

Loading R in your browser

Designed for the field

Free analysis of designed experiments for agricultural and allied research. Paste your data from Excel and get the ANOVA, means with C.D. and letters, assumption checks, post-hoc tests and a plain-language reading of every result, with R running in your browser.

  • Free and open source
  • On CRAN
  • Checked against Gomez & Gomez (1984)
Completely Randomised Design (CRD)Randomised Complete Block Design (RCBD)Latin Square Design (LSD)Factorial CRD (2-4 factors)Factorial RCBD (2-4 factors)Split Plot DesignStrip Plot (Split Block) DesignPooled analysis over environments - RCBD basePooled analysis over environments - CRD basePooled factorial over environments - RCBD base (2-4 factors)Pooled factorial over environments - CRD base (2-4 factors)
LSD (Fisher's protected)LSD (Bonferroni-adjusted)Tukey HSDDuncan's DMRTStudent-Newman-KeulsScheffeDunnettShapiro-Wilk on the residualsLevene's testBartlett's testBox-Cox profileDescriptive statisticsCorrelationRegressionPrincipal componentsCluster analysisFactor analysisPlain-language readingsHTML report
11experimental designs, each with the error terms its layout calls for
7ways to compare means, Dunnett's test against a control among them
6tools to explore data before the analysis of variance
0things to install: R runs inside this page

01The workflow

From spreadsheet
to publishable table

Four steps, in the order of the tabs. Every result comes with a short reading of what it shows, worded to match the evidence: a test that is not significant reads as insufficient evidence, never as proof of no difference.

  1. 01

    Load

    Paste straight from Excel, upload a CSV or start from a worked example. A data check finds decimal commas, units typed after numbers and labels that differ only in capitals, and corrects only what is certain.

  2. 02

    Explore

    Look before you test: descriptive statistics, correlation, regression, principal components, cluster and factor analysis, each with a reading of what the numbers and plots show.

  3. 03

    Analyse

    Choose the design and map its columns. The right error terms follow: split and strip plots get their strata, and analyses over seasons or locations treat the environment as random.

  4. 04

    Report

    The ANOVA, means with SE(m), SE(d), C.D. and letters, assumption checks, post-hoc tests and plots, then an interpretation and a report to download.

02The analyses

Everything a field
trial asks of you

Designs

11 designs, the right error terms

  1. Completely Randomised Design (CRD)
  2. Randomised Complete Block Design (RCBD)
  3. Latin Square Design (LSD)
  4. Factorial CRD (2-4 factors)
  5. Factorial RCBD (2-4 factors)
  6. Split Plot Design
  7. Strip Plot (Split Block) Design
  8. Pooled analysis over environments - RCBD base
  9. Pooled analysis over environments - CRD base
  10. Pooled factorial over environments - RCBD base (2-4 factors)
  11. Pooled factorial over environments - CRD base (2-4 factors)

Tables of means

Means you can publish

SE(m), SE(d) and the C.D. at the level you choose. Letters appear only when the F-test allows them, and a missing plot gets adjusted means rather than wrong ones.

V454.31a
V248.78b
V548.45b
V645.51c
V142.95c
V337.71d

Worked example (RCBD): C.D. (5%) 2.80 · CV 4.01%

Post-hoc

7 ways to compare

  • LSD (Fisher's protected)
  • LSD (Bonferroni-adjusted)
  • Tukey HSD
  • Duncan's DMRT
  • Student-Newman-Keuls
  • Scheffe
  • Dunnett

Assumptions

Checks before conclusions

Shapiro–Wilk on the residuals, Levene's and Bartlett's tests, the mean–variance slope and the Box–Cox profile. A transformation is suggested only when the checks support it.

Your data stay here

R runs on your own computer. Nothing you paste is uploaded anywhere.

Readings

Every result, explained

V4 has the highest mean (54.31) and V3 the lowest (37.71). Two Variety means must differ by more than the C.D., 2.80, to be declared different at the 5% level. Every other Variety mean is significantly lower than that of V4. Every other Variety mean is significantly higher than that of V3.

The reading under the worked example's table of means, as the app writes it.

Report

A report to keep

Download the analysis as one HTML page: the ANOVA, the tables of means, the assumption checks and the interpretation, ready to print or attach.

03The engine

R, running
right here

StatLabX runs R itself, compiled to WebAssembly, inside this page. There is no server: nothing to install, no account, no queue, and the data you analyse never leave your computer.

The first visit downloads about 59 MB of R and its packages; your browser usually keeps a copy, so later visits skip the download, though R still takes a little while to start.

R runtimeLoading
  1. Loading R
  2. Starting R in your browser
  3. Loading the packages the analyses use
  4. Installing and loading the packages
  5. Starting StatLabX
 

04The evidence

Checked against
the textbook

The worked examples of Gomez & Gomez, Statistical Procedures for Agricultural Research (1984), are reproduced exactly, with one deliberate difference: environments are treated as random, so treatments are tested against their interaction with the environment. Behind that, the package's tests check every statistic against an independent computation.

  • Completely randomised design
  • Randomised complete block design
  • Latin square
  • Split plot
  • Combined analysis over seasons

05The team

The people
behind it

Dr. M. Iqbal Jeelani

Scientist (Statistics)

Division of Agricultural Statistics, SKUAST-Kashmir

ORCID

Suggestions and bug reports are welcome: write to uzairkhan11w@gmail.com or open an issue on GitHub.

06Cite and install

Use it, cite it

How to cite

Shah, I. A., Khan, U. J. and Jeelani, M. I. (2026). DOEpro: Analysis of Designed Agricultural Experiments. R package version 2.0.1. doi:10.5281/zenodo.21399570

doi:10.5281/zenodo.21399570

This is the concept DOI: it always leads to the latest release.

In R

install.packages("DOEpro")
DOEpro::run_DOEpro()
CRAN GitHub

The same analyses as this site, as an R package (GPL-3).

Ready when you are

R is loading in the background while you read.