Data Analysis

Data format

The format of the data is a .csv file with the following headers: numHelped,crowdSize,sessionsCompleted,difficulty, and comfort. The rows must be numerical. Examples of this data can be found in the sample_data directory.

Analyzing the data

Place two files in a directory called data, named method_1.csv, and method_2.csv. Then, run the following command:

python analyze.py

The output will be the Matt Whitney U Test Results for the 5 questions, along with the statistical power:

+-----------------------------+
| Matt Whitney U Test Results |
+----------------+------------+-------------------------+
|        Question|   Statistic|                  P-Value|
+----------------+------------+-------------------------+
|   Number Helped|       XXX.X|                     X.XX|
|      Crowd Size|       XXX.X|                     X.XX|
|       Completed|       XXX.X|                     X.XX|
|      Difficulty|       XXX.X|                     X.XX|
|         Comfort|       XXX.X|                     X.XX|
+----------------+------------+-------------------------+
+-------------------+
| Statistical Power |
+----------------+--------+--------+
|        Question|    Beta|   Power|
+----------------+--------+--------+
|   Number Helped|    0.XX|    0.XX|
|      Crowd Size|    0.XX|    0.XX|
|       Completed|    0.XX|    0.XX|
|      Difficulty|    0.XX|    0.XX|
|         Comfort|    0.XX|    0.XX|
+----------------+--------+--------+

Generating Box-plot data

The script to generate the box plot is written in R. Please download the proper package here. Then, run the following command in the root directory:

Rscript gen_boxplot.r

It should save the box plots in a pdf file.