Awesome

DABEST-Matlab

About

DABEST is a data analysis tool that is intended to make estimation statistics more accessible to scientific communities. Estimation statistics is a superior alternative to null hypothesis significance testing (NHST), in which effect size and confidence intervals are used to interpret results as opposed to dichotomous significance testing.

This code allows the user to visualize the data as scatterplots; calculates the effect size and confidence intervals of the difference between multiple groups; and plots the results on the same figure. This figure design allows for a visual inspection of the observed values distribution, and displays the differences between multiple groups of data.

Installation

DABEST-Matlab can be installed via MATLAB Central (https://www.mathworks.com/matlabcentral/fileexchange/65260-dabest-matlab) or GitHub (how to clone a repo: https://help.github.com/articles/cloning-a-repository/).

How to cite

Moving beyond P values: Everyday data analysis with estimation plots

Joses Ho, Tayfun Tumkaya, Sameer Aryal, Hyungwon Choi, Adam Claridge-Chang

Nature Methods 2019, 1548-7105. 10.1038/s41592-019-0470-3

Paywalled publisher site; Free-to-view PDF

Bugs

Please report any bugs on the Github issue tracker.

Contributing

All contributions are welcome; please read the Guidelines for contributing first.

We also have a Code of Conduct to foster an inclusive and productive space.

DABEST in other languages

DABEST is also available in R (dabestr) and Python (DABEST-Python).

Tutorial

Data format

Data should be in the csv file format and contain two columns with the headers: Identifiers and Values.

Identifiers are the labels of each data point, and Values are the data points (see the example below).

Note: All the sample data used in this tutorial are taken from S. Champely's anscombe2 dataset, and can be found in DABEST-Matlab/SampleData/.

Depending on the number of groups the data contain, the main function dabest produces various plots, and returns the key information as a table object.

1. Two groups

If the data have two different groups, dabest('TwoGroups_sample.csv') generates a two groups plot.

2. Paired

Running dabest('TwoGroups_sample.csv','Paired') generates a paired plot with the two groups data.

3. Multiple groups

If the number of groups is an even number, a multiple groups plot will be automatically generated by dabest('MultipleGroups_sample.csv') command.

4. Shared control

If there are more than two groups in the data, dabest('MultipleGroups_sample.csv') generates a shared control plot.

5. Merged groups

To combine two groups of data and compare to a third group, run dabest('MergedGroups_sample.csv','mergeGroups').

6. Multiple merged groups

For the data that contain more than three groups -and a number that is divisible by 3, dabest('MultipleMergedGroups_sample.csv','mergeGroups') generates a multiple merged groups plot.

7. Merged shared control

If the data contain more than three groups, dabest('MultipleMergedGroups_sample.csv','mergeGroups') automatically generates a second plot in which all the groups are compared to the merged shared control.