Drawing on his background in both psychology and data science, Tobias argues that there are 6 major sources of algorithmic bias (ranging from statistical artifacts and at least 6 distinct types of data issues to human biases of data scientists, users, as well as society at large) and that both data scientists and business users of algorithms (including managers and government agencies) can and need to contribute to fighting algorithmic bias. Tobias provides practical, proven techniques to effectively combat and eliminate bias and addresses both basic statistical algorithms and advanced techniques such as self-improving machine learning and artificial intelligence in a fun, laymen-friendly style. While some chapters are squarely aimed at data scientists, others address managers, government officials, policy makers, and philosophers by discussing questions such as under which circumstances a possibly biased algorithm should be used at all and if algorithms also can be a tool to fight biases so deeply ingrained in society that they already have become a self-fulfilling prophecy.