MatchMaker performs reiterative resampling procedures to produce random distributions of data. This program creates random pairings of dynamic data (such as data obtained from human behaviour) for a specifiable number of iterations, to create a random null distribution. The randoming procedure that MatchMaker uses to get to the null distribution is intriquitely specified, in order to maintain random integrity. This distribution can then be used as a baseline to compare true, observed data in order to statistically determine whether the structure in true data differs significantly from random.

In psychological research, an important question is how the behaviour of one group of humans compares to the behaviour of a different group of humans. However, statistically, any observed difference must be compared with differences between randomly selected measures of behaviour. MatchMaker allows for the creation of this random distribution in order to determine whether their observed human behaviour is truly different from random and therefore, meaningful. MatchMaker was used to create random distributions in Chapter 4 (in preparation for publication) of the doctoral dissertation, Back and Forth: Prediction and Interactive Alignment in Conversation, authored by Nida Latif in 2017.

MatchMaker is a Windows Forms Application developed entirely in C#. The user selects a folder in which all the files are located (a sample selection of such files are included in MatchMaker project files). Starting with the first file, each value in it is paired up with another random value from a different random file. This process is repeated for all values in every file in the selected folder. Once all values in every file is paired up, that concludes one trial. MatchMaker is able to repeat this process for the predetermined number of trials the experiment needs to run.

There are two files of results that are required to be printed. One is a spreadsheet detailing the average similarity and standard distribution of each individual trial, called the master file. The second file is a spreadsheet document detailing every random pairing for a randomly selected trial in the experiment. MatchMaker is able to accomplish this by using the Microsoft Excel 16.0 Object Library to create intriquite Excel spreadsheet documents programatically.

• Programming Language: C#
• IDE: Visual Studio 2017

Abdul Sadiq

This project is licensed under the MIT License - see the LICENSE.md file for details

Nida Latif - Thank you for allowing my inclusion in your research!