Survivorship Bias Explained: 4 Examples of Survivor Bias

Survivorship bias (or survivor bias) is a cognitive fallacy in which, when looking at a given group, you focus only on examples of successful individuals (the “survivors”) in the selection process rather than the group as a whole (including the “non-survivors”).

This type of bias is prevalent in the field of business, in which students studying entrepreneurship concentrate on a handful of stories of successful startups rather than including in their research all of the startups that struggled or dissolved.

Survivorship bias leads to several major consequences in the decision-making process since it:

• Encourages overly optimistic thinking: When you look at only the successful subset of a particular situation, it encourages you to believe in a skewed reality in which circumstances are easier or more likely to work out than they actually are. This can cause you to take unnecessarily large risks in your life—whether financially or personally.
• Leaves out important voices: By limiting its view to only the positive results in a situation, survivorship bias leaves out the vital voices of those who have struggled to succeed.
• Suggests causation from correlation: If you look at all of the “success stories” of a particular group, you might start to notice patterns (or correlations) that offer a false sense of causation. For instance, if you look at several of the most successful startups or billionaires in business, you might begin to think the formula to becoming a high-power CEO begins with being a college dropout. This viewpoint fails to consider all the individuals who dropped out and weren’t successful.

Survivorship bias is a logical error that has played or can play a major role in many fields, including:

1. 1. Business success: Survivorship bias is a major component in the discussion of startups and mutual fund performance in business. MBA programs and other entrepreneur programs might discuss a handful of successful people and startups—such as Bill Gates’s Microsoft, Steve Jobs’s Apple, or Mark Zuckerberg’s Facebook—as benchmarks for how to begin a successful business. This education, however, ignores the many thousands of unsuccessful startups that occurred both before and after these successful few.
2. 2. Manufactured goods: A common saying in English in the United States is: “They don’t make them like they used to.” It describes goods like furniture, cars, or equipment as being higher quality in the past than in the present. This is an example of survivorship bias since all the examples of older goods are items that have lasted until the present day, while all the low-quality goods from the past have already broken down. In other words, only high-quality goods from the past exist now, but new goods of all qualities are still readily available.
3. 3. Portrayals of disability: Many portrayals of disability in the media fall into categories of survivorship bias, whereby they offer examples and imagery of people with disabilities outperforming others or whose circumstances changed. While these stories can be inspiring, they also focus only on a few cases and ignore the large majority of others who live with disabilities and struggle. The media can paint an overly optimistic picture that suggests those with disabilities don’t need recognition and support.
4. 4. World War II plane research: The prototypical example of survivorship bias comes from statistician Abraham Wald at Columbia University, who conducted research on WWII bomber planes to recommend places for reinforcement. His team reviewed the data from all returning bombers and identified the locations on the aircraft in which they underwent the most fire. Rather than recommend those locations as places for reinforcement, however, Wald recognized they were using a form of survivorship bias. Wald noted these locations were actually spots in which aircraft could sustain many bullet holes and still return, while the planes that sustained enemy fire in other locations were the ones that went down. His team recommended reinforcements to locations less represented in the data of returned planes—as a result, they made more effective predictions and saved many lives.

The best way to identify and address the survivorship cognitive bias is to ask yourself what’s missing. When looking at a given data set, consider any other possible data points that were on the “same path” as the successful data points but might not be present in the final data.

Practicing empathy can help you lead more effectively while building stronger relationships across the personal and professional facets of your life. Challenge your perceptions with the MasterClass Annual Membership and take lessons on emotional intelligence from Pharrell Williams, Roxane Gay, Gloria Steinem, Dr. Cornel West, Walter Mosley, Robert Reffkin, and Robin Arzón.