DuckDuckGo Filter Bubble Study (2018)

Python scripts used to analyse Google search results for the 2018 filter bubble study. This assumes Python version 2.7.

Usage

To count domain occurrences within organic links (i.e. excluding infoboxes):

$ python count_domains.py

To count variations of the search results and infoboxes:

$ python count_variations.py

To calculate the average differences (edit distances) of participants' search results:

$ python measure_difference.py

Dependencies

pyxDamerauLevenshtein >= 1.5

Reference

For the filter bubble study report, please see here:

• https://spreadprivacy.com/google-filter-bubble-study/

See also the spreadsheets containing all data from the study here:

• https://duckduckgo.com/download/duckduckgo-filter-bubble-study-2018_participants.xls
• https://duckduckgo.com/download/duckduckgo-filter-bubble-study-2018_raw-search-results.xls