Fairness in Data Analytics Flashcards
Learn the what, why and how of fairness in data analytics - based on the Coursera Google Data Analytics course
What is fairness?
Ensuring unbiased and inclusive data collection and analysis.
Why is fairness important?
Fairness prevents skewed results and ensures accurate, representative insights.
How to ensure fairness?
Consider all relevant data, identify biases, and test underrepresented groups.
Why consider all available data?
To avoid biased analysis and get a full understanding of the situation.
What happens if data is ignored?
Biased results and incomplete insights.
How to handle unexpected data?
Analyze it for relevance before discarding, to avoid missing key information.
What are surrounding factors?
External conditions that could influence the data and analysis.
Why identify surrounding factors?
They help provide context, making insights more accurate.
How to identify surrounding factors?
Look at external conditions, trends, and variables that might impact results.
What is self-reported data?
Data provided directly by participants about themselves.
Why use self-reported data?
It reduces observer bias and gives participants a voice.
How to use self-reported data fairly?
Combine it with other data types and separate it for context.
What is oversampling?
Increasing sample size of underrepresented groups for balance.
Why is oversampling important?
It helps ensure minority groups are fairly represented in the analysis.
How to use oversampling effectively?
Use it when certain groups are underrepresented, and adjust analysis accordingly.
