Selection Bias: How It Affects Our Lives

Human psychological bias affects us in many different ways. We have recently discussed how recency bias influences our trading. In this post, we’re going to talk about selection bias.

Selection bias occurs when a statistical sample is not representative of the entire population. This situation arises, for example, when a study is based on a small number of people, and/or the subjects are not truly randomized. It can also happen when the researchers subconsciously project their own beliefs onto the experiment design, thus leading to inaccurate results. In trading and portfolio management, the selection bias often manifests itself in the form of cherry-picking, i.e. the tendency for the portfolio managers to present results using samples that validate their views.

A recent article [1] examined the selection bias in a larger context. Specifically, it studied how the success narratives affect people’s decision making,

READ OUR POSTS

In this paper, we complement and build upon these previous efforts by focusing on the extent to which popular and widely-read success narratives containing plainly biased examples affect people’s decision-making in terms of what choices they make, how confident they are in those choices, and the justifications they provide around them. In our experiment, we observed a strikingly large difference in participants’ decisions depending on which condition they were assigned to: while 87% of people shown examples of successful graduates bet that graduates are more likely to succeed, only 32% of people shown examples of successful dropouts bet on graduates, a difference of 55 percentage points. Far from discounting the small sets of examples they acknowledged as biased, participants were substantially swayed by them. This strong effect on beliefs is caused by a weak manipulation that involves no outright deception.

Briefly, our decision making can be manipulated without difficulty. In fact, the authors pointed out that even with factual accurate data, its selective presentation can still lead us to a wrong conclusion,

There are many examples of the selective presentation of factually accurate data in polarizing domains like climate science, gun control, immigration, police violence, and social insurance programs. Given the size of the effect that we found with a manipulation that would clearly pass a conventional fact check, our results suggest that more attention should be paid to the prevalence and impact of presenting biased selections of factually accurate examples.

We believe that with the increasing influence of social media, selection bias will have a bigger impact on our decision making. Therefore it’s more important than ever to conduct investment research that is based on statistically valid samples and techniques.

References

[1] G. Lifchits, A. Anderson, DG. Goldstein, JM. Hofman, DJ. Watts, Success stories cause false beliefs about success, Judgment and Decision Making, Vol. 16, No. 6, November 2021, pp. 1439–1463