Brightstat.com has been completely redesigned as a **web application for statistics and data visualization**.

- Do statistics
**on tablets and mobile devices** **Better integration**on all devices**Interactive graphs****HTML5, CSS, Javascript****Faster computations****Presentation mode****secure connection****lots of bugfixes**

**Descriptive statistics**(Mean, Median, Mode, Dispersion, Range, Variance, etc.)**Non-parametric tests**(Mann-Whitney U test, Wilcoxon signed-rank test, Kruskal-Wallis test, Friedman test, sign test)**Parametric tests**(one sample t-test, two independent samples t-test, paired samples t-test, Analysis of variance univariate and multivariate (ANOVA and MANOVA), linear regression, logistic regression, reliability etc.)**Data visualization**(line plot, bar plot, area plot, scatter plot, box plot and histogram plot)**Data manipulation**- Filter, split and weight data
- etc.

more details

Unfortunately existing users need to create a new account. If you want to use the previous version of brightstat click here. The previous version will be available until end of 2015.

Brightstat lets you recode categorical variables, often coded as strings, into ascending numbers automatically. Just select 'Automatic Recode' in the 'Data Menu' and indicate the variables to be recoded. The old values will be stored as value labels for the new numeric variable.

In the 'Data Menu' there is 'Load Sample Datafile' in the first column 'Data File'. There you'll find a list of sample files (including files for all examples on Brightstat's website). So you can start discovering Brightstat right away.

Brightstat has a built in calculator which allows you to compute new variables with existing ones. If you have measured the weight (in kg) and height (in meters) of n subjects you can easily compute the body mass index (BMI) for all subjects. If your variables are named 'weight' and 'height' the formula would be: _weight/_height^2.

The sample file 'BodyFat' contains the variables 'Weight' (pounds) and 'Height' (inches), so the formula for calculating the BMI needs some adjustment in that case: (_Weight*0.45359237)/(_Height*2.54/100)^2.

Note: In the formula a variable name is always preceded by an underscore '_'.

Brightstat is not intended to be a data editor. Small changes can be done without problems but be aware that editing your data in Brighstat's data window may be very time consuming, especially with larger data files.

It is strongly recommended to prepare your data for Brightstat before uploading it into the database. A period '.' will be treated as a missing value.**How to prepare your data for Brightstat**

You can define a filter variable for your data. Analysis and graphs are then performed for selected cases only. Select 'Filter Data' from the menu and indicate which values of which variables should be included in the analysis.

You can split your datafile using a categorical variable. All tests and/or graphs are then performed for the individual categories of the split variable. Select 'Split Data' from the menu and indicate one or more split variables.

more details

Stricker, D. (2008). BrightStat.com: Free statistics online. *Computer Methods and Programs in Biomedicine, 92*, 135-143.

2015-11-24 v1.2.4

- Code for missing values can now be indicated for data upload. Minor optimization for reliability procedure: faster detection of constant variables that have to be eliminated from analysis.

2015-09-26 v1.2.3

- Minor bug fixes

2015-09-09 v1.2.2

- Type I sum of squares added for ANOVA and MANOVA procedure. While a model with type III sum of squares corrects every effect against all other effects, a type I sum of squares ANOVA model corrects every effect against all previously entered effects. The order of effects entering the model affects the outcome in an unbalanced multifactorial (non equal cell n) design.

2015-09-02 v1.2.1

- Minor bugfixes for Survival plot

2015-08-27 v1.2.0

- Kaplan Meier survival analysis has been added. Inclusive tests for equality of survival functions and statistic for linear trend. Four different plots (Survival, 1-Survival, Hazard [-ln(Survival)] and ln(Survival)) are available together with customizable confidence intervals.

more details

2015-09-01

- More graphs
- Graph Export as png or jpeg does not work on Windows
^{®}Phone

more details