Multivariate analysis

Discriminant Analysis Online

Find the combinations of predictors that best separate known groups, and see how accurately the functions classify cases.

Free · No sign-upReplaces SPSS DiscriminantWorks with your data

Updated 13 October 2026 · By Bibin Xavier

Data 120 × 6

Paste from Excel, or drop a CSV, TSV or Excel file here.Upload file

Example data 120 rows, 6 variables. Data stays in this browser tab.

Plottext · 120 levels	Cultivartext · 3 levels	Leaf_Lengthnumeric	Leaf_Widthnumeric	Stem_Heightnumeric	Yield_kgnumeric
L001	Amber	5.34	3.25	18.4	1.19
L002	Amber	4.66	3.59	13.9	1.15
L003	Amber	5.14	3.39	16	1.4
L004	Amber	4.73	3.63	14.4	1.13
L005	Amber	5.14	2.59	8.5	1.18
L006	Amber	4.82	3.95	15.9	1.23
L007	Amber	4.91	3.23	15.1	1.19
L008	Amber	5.73	3.74	16.5	1.01

Variables and options

Predictors 4 selected · min 2

Grouping variable

Prior probabilities

Results update automatically when you change an option.

Results

120 rows · example data

Correctly classified87.5%

Discriminant functions2

Chance accuracy33.3%

Table 1

Eigenvalues and Wilks’ lambda

Function	Eigenvalue	% of variance	Canonical r	Wilks’ Λ	χ²	df	p
1 through 2	5.997	95.3	.926	.110	254.50	8	< .001
2	0.294	4.7	.477	.773	29.80	3	< .001

Note. The Wilks test in each row checks that function and all later ones.

Table 2

Standardized canonical discriminant coefficients

Predictor	Function 1	Function 2
Leaf_Length	0.459	-0.091
Leaf_Width	-0.705	0.720
Stem_Height	0.456	0.024
Yield_kg	0.400	0.569

Table 3

Group centroids

Group	Function 1	Function 2
Amber	-3.269	0.222
Basil	0.765	-0.738
Cedar	2.504	0.516

Table 4

Classification results

Actual group	Predicted Amber	Predicted Basil	Predicted Cedar	% correct
Amber	40	0	0	100.0
Basil	3	30	7	75.0
Cedar	0	5	35	87.5

Note. 87.5% of original grouped cases correctly classified.

Discriminant function scores

Interpretation

The first function accounts for 95.3% of the between-group variance (canonical r = .926) and is statistically significant. The functions classify 87.5% of cases correctly, compared with 33.3% expected by chance.

The first discriminant function was significant, Λ = .110, χ²(8) = 254.50, p < .001, canonical R = .93; 87.5% of cases were correctly classified.

How to cite this tool

Using these results in a thesis, journal article, report or class assignment? Please cite this free tool so others can find it. It keeps the site free.

APA 7

Xavier, B. (2026). Discriminant Analysis Online [Web application]. FreeStatisticalTools.com. https://freestatisticaltools.com/discriminant-analysis/

BibTeX

@software{xavier_discriminantanalysis_2026,
  author = {Xavier, Bibin},
  title = {Discriminant Analysis Online},
  publisher = {FreeStatisticalTools.com},
  year = {2026},
  url = {https://freestatisticaltools.com/discriminant-analysis/},
  note = {Web application}
}

Suggested sentence for your methods section: “The analysis was carried out with the free Discriminant Analysis Online tool on FreeStatisticalTools.com (Xavier, 2026).”

What is discriminant analysis?

Discriminant analysis finds the linear combinations of predictors that best separate known groups, then uses them to classify cases. It is the predictive counterpart of MANOVA.

When to use discriminant analysis

Use it to predict group membership from continuous predictors, or to describe which variables separate the groups. It complements MANOVA.

Assumptions

Multivariate normality within groups.
Equal covariance matrices across groups.
More cases in the smallest group than predictors.

How to use this tool

Enter your data. Paste columns from Excel, Google Sheets or SPSS with variable names in the first row, or upload a CSV or Excel file. The example data shows the expected layout.
Choose the variables. Select the predictors and grouping variable.
Set the options. Fill in prior probabilities.
Read the results. Results update automatically. Read the plain-English interpretation, then use Copy table to paste APA-style tables into Word and Copy for the results sentence.

Frequently asked questions

What does Wilks’ lambda show in discriminant analysis?

It tests whether a discriminant function separates the groups significantly. Smaller values mean better separation.

How do I judge classification accuracy?

Compare the percentage correctly classified with the accuracy expected by chance, which the tool reports.

What are standardized coefficients?

They show each predictor’s relative contribution to a function, on a common scale.

Related tools

Browse all categories

Data cleaning and preparation · Describe and explore · Compare means · Nonparametric tests · Correlation and regression · Bayesian statistics · Categorical data · Reliability and agreement · Multivariate analysis · Structural equation modeling · Econometrics and time series · Panel data and causal inference · Marketing and decision research · Survival, clinical and meta-analysis · Study planning and calculators · PLS-SEM