Paper title: Navigating Social Media Privacy: Awareness, Preferences, and Discoverability
Artifacts HotCRP Id: #18
Requested Badge: Reproduced
This artifact provides anonymous survey data and a Python notebook to reproduce the results presented in the paper. The specific commit at the time of publication is 81bf396.
We remove all survey metadata and provide only participants' answers, ensuring that no sensitive data is disclosed.
No special hardware requirements (can run on a local laptop).
We tested this artifact on MAC but it also can be run on Linux or Window as long as the platforms install Python 3.8.11 (recommended through pyenv) and Jupyter Notebook (recommended through VS Code).
The survey data (stored in raw) is around 700KB and running the Python notebook (artifact.ipynb) should complete within a few minutes.
We run the artifact using Python 3.8.11, which we recommend installing via pyenv.
In short, on MAC run the following to install Python 3.8.11
brew update
brew install pyenv
echo 'export PYENV_ROOT="$HOME/.pyenv"' >> ~/.zshrc
echo '[[ -d $PYENV_ROOT/bin ]] && export PATH="$PYENV_ROOT/bin:$PATH"' >> ~/.zshrc
echo 'eval "$(pyenv init - zsh)"' >> ~/.zshrc
exec "$SHELL"
pyenv install 3.8.11
pyenv local 3.8.11
If you are using Linux, replace the first two "brew" commands to
curl -fsSL https://pyenv.run | bash
The example above applies to Zsh as a shell. If you use other shells (e.g., Bash, Fish, etc.), please follow these instructions to set up your shell environment for pyenv.
Now that Python 3.8.11 is installed, we recommend setting up a virtual environment (venv) to install packages required to run the Python notebook.
Run the following to set up the venv
python -m venv artifact_venv
source artifact_venv/bin/activate
pip install -r requirements.txt
The packages required (requirements.txt) include numpy, pandas, matplotlib, statsmodels, and ipykernel.
Now that the venv is created, make sure to select it (artifact_venv) as your kernel in your notebook.
Run (click "Run All") the Python notebook (artifact.ipynb) to reproduce tables and plots in the paper.
We reproduce ALL results presented in the paper. Please refer to the notebook (artifact.ipynb) as we specify which code section reproduce which result.
To reproduce the statistical analysis in Section 4.4 (Correlations), 4.5 (Demographic and Usage Differences), and 4.6 (Platform-specific Differences), we used the statsmodels package to generate all correlation (4.4) and regression results (4.5 and 4.6). Alternatively, you can download the dataframe df_stat as csv after it was preprocessed in Demographic and usage differences (Section 4.5) df_stat.to_csv('stat_raw.csv', index=False) and use this csv file with your stat software (e.g., SPSS) to run statistical analysis. We verified that the results produced with statsmodels and SPSS, are identical.
We provide mapping of question numbers to questions asked below. For each platform, we asked: 1. if users have seen a privacy setting demo_cols_encode and general_usage_cols_encode in the notebook (artifact.ipynb).
privacy_features_seen_questions = {
"Facebook": {
"audience": "Q2.3",
"message": "Q2.5",
"ads": "Q2.7",
"activity_status": "Q2.9",
"account_suggestion": "Q2.11",
"connection_view": "Q2.13",
"profile": "Q2.15",
"search": "Q2.17"
},
"Instagram": {
"audience": "Q2.3",
"message": "Q2.5",
"ads": "Q2.7",
"activity_status": "Q2.9",
"video": "Q2.11",
},
"Twitter (X)": {
"audience": "Q2.3",
"message": "Q2.5",
"ads": "Q2.7",
"account_suggestion": "Q2.9",
"video": "Q2.11",
},
"LinkedIn": {
"audience": "Q2.3",
"message": "Q2.5",
"ads": "Q2.7",
"activity_status": "Q2.9",
"account_suggestion": "Q2.11",
"connection_view": "Q2.13",
"search": "Q2.15",
"profile": "Q2.17"
},
"TikTok": {
"audience": "Q2.3",
"message": "Q2.5",
"ads": "Q2.7",
"activity_status": "Q2.9",
"account_suggestion": "Q2.11",
"connection_view": "Q2.13",
"video": "Q2.15",
},
"Snapchat": {
"audience": "Q2.3",
"message": "Q2.5",
"ads": "Q2.7",
"activity_status": "Q2.9",
"account_suggestion": "Q2.11",
}
}
privacy_features_default_questions = {
"Facebook": {
"audience": "Q2.4",
"message": "Q2.6",
"ads": "Q2.8",
"activity_status": "Q2.10",
"account_suggestion": "Q2.12",
"connection_view": "Q2.14",
"profile": "Q2.16",
"search": "Q2.18"
},
"Instagram": {
"audience": "Q2.4",
"message": "Q2.6",
"ads": "Q2.8",
"activity_status": "Q2.10",
"video": "Q2.12",
},
"Twitter (X)": {
"audience": "Q2.4",
"message": "Q2.6",
"ads": "Q2.8",
"account_suggestion": "Q2.10",
"video": "Q2.12",
},
"LinkedIn": {
"audience": "Q2.4",
"message": "Q2.6",
"ads": "Q2.8",
"activity_status": "Q2.10",
"account_suggestion": "Q2.12",
"connection_view": "Q2.14",
"search": "Q2.16",
"profile": "Q2.18"
},
"TikTok": {
"audience": "Q2.4",
"message": "Q2.6",
"ads": "Q2.8",
"activity_status": "Q2.10",
"account_suggestion": "Q2.12",
"connection_view": "Q2.14",
"video": "Q2.16",
},
"Snapchat": {
"audience": "Q2.4",
"message": "Q2.6",
"ads": "Q2.8",
"activity_status": "Q2.10",
"account_suggestion": "Q2.12",
}
}
privacy_features_find_questions = {
"Facebook": {
"audience": "Q4.2",
"account_suggestion": "Q4.7",
"activity_status": "Q4.12",
"message": "Q4.17",
"profile": "Q5.2",
"connection_view": "Q5.7",
"ads": "Q5.12",
"search": "Q5.17"
},
"Instagram": {
"audience": "Q4.2",
"activity_status": "Q4.7",
"ads": "Q4.12",
"message": "Q4.17",
"video": "Q4.22",
},
"Twitter (X)": {
"audience": "Q4.2",
"account_suggestion": "Q4.7",
"message": "Q4.12",
"ads": "Q4.17",
"video": "Q4.22",
},
"LinkedIn": {
"audience": "Q4.2",
"account_suggestion": "Q4.7",
"message": "Q4.12",
"activity_status": "Q4.17",
"profile": "Q5.2",
"connection_view": "Q5.7",
"ads": "Q5.12",
"search": "Q5.17",
},
"TikTok": {
"audience": "Q4.2",
"account_suggestion": "Q4.12",
"message": "Q4.7",
"activity_status": "Q5.17",
"connection_view": "Q5.2",
"ads": "Q5.7",
"video": "Q5.12",
},
"Snapchat": {
"audience": "Q4.2",
"message": "Q4.17",
"ads": "Q4.22",
"activity_status": "Q4.12",
"account_suggestion": "Q4.7",
}
}
privacy_features_time_questions = {
"Facebook": {
"audience": "Q4.1",
"account_suggestion": "Q4.6",
"activity_status": "Q4.11",
"message": "Q4.16",
"profile": "Q5.1",
"connection_view": "Q5.6",
"ads": "Q5.11",
"search": "Q5.16"
},
"Instagram": {
"audience": "Q4.1",
"activity_status": "Q4.6",
"ads": "Q4.11",
"message": "Q4.16",
"video": "Q4.21",
},
"Twitter (X)": {
"audience": "Q4.1",
"account_suggestion": "Q4.6",
"message": "Q4.11",
"ads": "Q4.16",
"video": "Q4.21",
},
"LinkedIn": {
"audience": "Q4.1",
"account_suggestion": "Q4.6",
"message": "Q4.11",
"activity_status": "Q4.16",
"profile": "Q5.1",
"connection_view": "Q5.6",
"ads": "Q5.11",
"search": "Q5.16",
},
"TikTok": {
"audience": "Q4.1",
"account_suggestion": "Q4.11",
"message": "Q4.6",
"activity_status": "Q5.16",
"connection_view": "Q5.1",
"ads": "Q5.6",
"video": "Q5.11",
},
"Snapchat": {
"audience": "Q4.1",
"message": "Q4.16",
"ads": "Q4.21",
"activity_status": "Q4.11",
"account_suggestion": "Q4.6",
}
}