Web Books

Misrepresenting Data

Written by: Matt David
Reviewed by: Blake Barnhill

There is so much data you have access to within a company. Communicating accurate insights from it is challenging. This book covers the mental biases and common mistakes that people make when analyzing data. It then provides guidance on how to prevent and avoid these costly errors.

    Cognitive Biases

  • Confirmation Bias

    Confirmation bias negatively affects the accuracy of your analysis. Learn how to detect Confirmation Bias and how to avoid this cognitive bias.

  • Selection Bias

    Selection bias negatively affects your analysis. Learn how to detect Selection Bias and how to avoid this cognitive bias.

  • Survivorship Bias

    Survivorship bias negatively affects your analysis. Learn how to detect Survivorship Bias and how to avoid this cognitive bias.

    Analysis Mistakes

  • Statistic vs Distribution

    Distributions provide much more nuance to the data than a statistic does. Learn how to query to get distributions and then interpret them.

  • Overall vs Groups

    Overall statistics in data can be misleading because there may be distinct groups within the data that have very different statistics. Learn to avoid this analysis mistake.

  • Metric vs Metrics (In progress)

    A single Metric can be misleading. Learn how to use multiple Metrics to avoid misleading yourself and others. Learn more

  • Trends

    Trend in data can be misleading depending on what time frame you are looking at it. Learn how to accurately interpret trends.

  • Comparing to Historical Averages (In progress)

    Historical Averages can be misleading because they hide trends in the data. Learn to identify misleading historical statistics.

  • Relative vs Absolute Change

    Relative and Absolute changes can bias your interpretation of data you are analyzing. Learn to interpret them correctly.

    Experiment Design

  • Predicting Outcomes

    Predicting Outcomes prevents cognitive biases from affecting how your interpretation of the results of an experiment. Learn more.

  • Define Experiment Parameters

    Defining Experiment Parameters improves analysis and increases trust in results that are shared in an organization.

  • Review Outcomes

    Review Outcomes of Feature Releases to evaluate their impact and to create institutional knowledge. Learn how to interpret results accurately.