Group Chat Survey
This repository contains the data and analysis for the paper "Understanding User Perceptions of Security and Privacy for Group Chat: A Survey of Users in the US and UK". The paper can be found here.
Data Files
Filename | Contents |
---|---|
responses (raw).xlsx | Raw responses gathered from participants. Human readable. |
responses (coded).xlsx | Responses along with open coding results. Human readable. |
coding-report.txt | A high-level analysis of coded responses. |
responses-processed.csv | Participant multiple choice responses coded as a large array of boolean values. Used by the analysis scripts. |
question_mapping.cvs | Mapping between question Ids in response-processed.csv and question text. Used by the analysis scripts. |
Data Analysis
To run the data analysis scripts, you will need to have python
>=3.8 and the pipenv
package installed. Pipenv is used to handle all the dependencies. Once it is isntalled, do the following:
- Open the analysis directory
pipenv install
pipenv run python init.py
You can then run the various analysis scripts using pipenv run python script.py
. Output will be stored to analysis/output
. The scripts include:
Filename | Contents |
---|---|
init.py | Will regenerate csv files used by other analysis scripts. |
basic-overview.py | Script that looks at various splits (country, gender, age, usage) within the data. |
chi-squared.py | Script that looks at various non-parametric correlations. |
data-driving.py | Script exploring correlations within the data. |
k-means.py | script with k-means and k-modes analysis scripts. |