top of page

Navigating the Maze: Causation, Correlation, and the Survivorship Bias Effect

Updated: Apr 15

Have you ever read something from a motivation “expert” that starts with “the greats do this”? If so, you need to understand the difference between causation and correlation. Understanding the distinction between these concepts is vital for making informed decisions. First, let's discuss the differences between causation and correlation.

Causation vs. Correlation - Unraveling the Threads:

Causation implies a cause-and-effect relationship, asserting that one variable directly influences another. A classic example is the connection between smoking and lung cancer. The more cigarettes smoked, the higher the likelihood of developing lung cancer. In this scenario, smoking is considered a cause, and lung cancer is the effect.

On the other hand, correlation signifies a statistical association between two variables. If two variables change together, they exhibit correlation. However, correlation does not imply causation. Ice cream sales and drowning incidents, for instance, may correlate during the summer months, but one does not cause the other.

Survivorship Bias – How it Distorts Reality:

Survivorship bias—a subtle but strong distortion in our perceptions. This bias occurs when we focus only on surviving elements rather than considering the whole picture. For example, in a study on successful entrepreneurs, if we only analyze those who achieved success and neglect those who failed, we fall prey to survivorship bias. In this example we aren’t reviewing those that failed who may have done the exact same actions listed for those that succeeded, but were missing some other variable that may have prevented final success.

Cautions in Causal Connections:

Survivorship bias interconnects with causation and correlation, complicating the landscape. When analyzing causation, it's crucial to be vigilant against overlooking non-survivors—those instances or data points that did not result in the expected outcome. Failure to do so can lead to wrong conclusions (causations).

Consider a study exploring the correlation between a specific training program and athlete success. If only a successful athlete who completes the specific program is considered, survivorship bias creeps in, potentially attributing success solely to this specific training. However, a comprehensive analysis that includes both successful and unsuccessful cases paints a more accurate picture.

Navigating the Sea of Correlation:

Correlation, too, dances with survivorship bias. When studying the relationship between variables, it's essential to include the full spectrum of cases rather than just those that exhibit the desired correlation. By accounting for both survivors and non-survivors, we mitigate the impact of survivorship bias on our conclusions.

For example, looking at Steph Curry and saying that the reason he succeeded was because of his incredible self-discipline and shooting routines. While this is certainly a part of his incredible development this doesn’t mean that there haven’t been thousands of basketball players that have had great self-discipline and routines but yet didn’t reach the anywhere close to the same success. 

We can’t simply ignore that Steph Curry has a dad that played in the NBA, that was a great shooter in his own right, a mom that also played a division one college sport, he spent hours and hours around NBA players while he was young, and had great coaching and instruction from his dad while he was developing. In essence, his work ethic is absolutely great, but it is far from the only factor in his amazing skill set. Correlation, but NOT causation.

Conclusion - Crafting a Clearer Narrative:

In the tricky web of causation, correlation, and survivorship bias, creating perspective is essential. Understanding that correlation does not imply causation and being aware of survivorship bias ensures a more accurate interpretation of stories you may that start with “the greats do this.”

When navigating the causation and correlation maze, consider the broader context. Scrutinize not just the survivors but also the non-survivors to tell a more accurate narrative. By acknowledging the influences of survivorship bias, we build more informed decisions and interpretations, fostering a data-driven landscape that’s grounded by the understanding of causation and correlation.


Stuart Singer, M.Ed., and PsyD (ABD) is the Director of WellPerformance, a Mental Performance Coaching and Consulting practice, and the creator of the DoSo app . For more information regarding this topic, he can be contacted at or follow him on X: @wellperformance, or Instagram: @wellperformance

31 views0 comments


bottom of page