When we use data and evidence to make decisions, we often run into a common problem known as survivorship bias. Survivorship bias is an error in thinking caused by an omission of data from those who did not survive a particular test or experiment. It skews our perception of reality and can lead to faulty conclusions.
Survivorship Bias: What Is It?
Survivorship bias is the tendency to focus on those individuals or things that have “survived” an experience, while ignoring those that have not. For example, in World War II, military strategists analyzed the damage inflicted on warplanes after they returned from battle. They mistakenly attributed the lack of damage to the design of the aircraft, while failing to consider the many planes that had been lost in battle. This bias led to an incorrect conclusion about the effectiveness of their design, which ultimately skewed their strategy.
Survivorship bias is also common in the health industry. For example, when analyzing the results of a cancer study, researchers may mistakenly focus on those patients that survived the study, ignoring those that did not, thus skewing their view of the actual success rate of the treatment.
How to Avoid the Impact of Survivorship Bias
Avoiding the impact of survivorship bias begins with understanding it. Knowing how and why this bias exists helps us to look for it in our research and decision-making processes. Additionally, employing a systematic approach to data gathering and analysis can help reduce the impact of this type of thinking. Here are some tips for avoiding survivorship bias:
– Collect data from both survivors and non-survivors.
– Make sure the sample is representative of the entire population.
– Utilize a control group.
– Analyze data objectively.
– Take a holistic approach.
– Ask questions and seek alternative explanations.
Conclusion
Survivorship bias can be a dangerous problem when making decisions based on data, and should be avoided as much as possible. To ensure accurate results, it is important to take a balanced and objective approach to data collection and analysis. By doing so, we can reduce the impact of survivorship bias and make better-informed decisions.