In the world of academia, researchers work with and rely on various testing, selection, and sampling methods to support their studies. Whether diving into datasets, analyzing information, or interpreting statistics, their work often revolves around uncovering patterns and drawing conclusions.
However, it is not uncommon to find biases sometimes subtly creeping into research – affecting findings, misleading decision-making, and weakening credibility in unintended ways. It can take place at any given stage of the research process while collecting, analyzing, and interpreting data or even during its publication and dissemination. Some common types of bias include researcher bias, information bias, selection bias, response bias, anchoring bias, survivorship bias and publication bias, among others.
Among these, survivorship bias is significant because it can create a skewed perception of reality by focusing only on successes while ignoring failures. In this article, we will explore what survivorship bias is and why it is essential, in addition to understanding its potentially far-reaching consequences in research.
Defining survivorship bias
We can define survivorship bias as a selection bias where researchers take into account only the sample (cases, entities or persons) that have successfully passed through a selection process. Here, groups or individuals who do not pass (non-survivors) the selection process are not considered.
Given that failures or unsuccessful outcomes are not factored in, survivorship bias leads researchers to focus only on a specific subset of a sample, basically the ones that did well. This typically leads to overestimating success rates or overlooking critical information about failures. In other words, researchers focusing only on the success stories inadvertently allow survivorship bias to distort their research findings and conclusions.
When only visible successes are studied, while invisible failures are ignored, it creates a skewed perception of reality and affects research outcomes. Let us understand the concept of survivorship bias with a simple, practical example. An aspiring athlete adopts a health or fitness regimen publicized by a neighbourhood club.
The individual makes the critical decision to join the club by reviewing feedback from current and past members she could contact. The glowing personal reviews, as well as the positive vibes being reflected by the club’s management, helped her to make this decision.
However, amidst all the upbeat and exaggerated observations, the individual fails to take into account cases of individuals who may have encountered unfavourable or negative experiences by following the club’s regimen. This sample or group could have faced injuries or other health issues but are not considered by the aspiring athlete; she goes on to take a cognitive shortcut that we now know as survivorship bias.
Why survivorship bias matters
Researchers must understand the consequences of survivorship bias in research.
- If survivorship bias is not taken into account, the research conclusions that are reached can lead to false optimism and the adoption of flawed approaches by the researchers.
- Moreover, since critical data points and important information are discounted as failed cases, survivorship bias can lead to the neglect or neglect of valuable insights.
- Importantly, across industries and sectors, the presence of survivorship bias can result in failure to achieve the desired or optimum results or outcomes due to biased or inconsistent findings or conclusions.
- Not fully considering potential risks and identifying probable fault lines can adversely impact decisions and policy-making, especially in the areas of business, education, health and medical sciences.
That is why researchers and students must take adequate measures to mitigate or prevent survivorship bias in their research endeavours.
Tips to prevent survivorship bias
While students and researchers need to acknowledge that there will be some degree of bias that will always creep in as they design and implement their research projects, it is equally imperative to be vigilant and avoid any potential for bias. As a first step, one must be aware of the different kinds of biases in research and, at the same time, equip oneself to take measures to mitigate these as far as possible. To reduce survivorship bias, researchers and decision-makers can take the following steps:
- Always consider both successes and failures in your study. For example, when analyzing why certain startups succeed, it is important to study why others fail. This provides a balanced perspective.
- Identify and include data that might not be immediately visible, such as cases excluded from a dataset due to dropout or closure. For instance, in the WWII aircraft example, data from missing planes was critical to avoid bias.
- Ensure your sample represents the entire population, not just a subset. For example, when analyzing employee retention, include those who left the company to identify factors contributing to turnover.
- Don’t just accept what’s visible; critically examine the underlying reasons behind successes and failures—question whether other perspectives or hidden data might change the conclusions.
- Implement methods like longitudinal studies or randomized sampling to ensure all relevant cases are included in the analysis.
By applying these strategies, researchers can better navigate the challenges of survivorship bias and produce more accurate, actionable insights.
Editage All Access is a subscription-based platform that unifies the best AI tools and services designed to speed up, simplify, and streamline every step of a researcher’s journey. The Editage All Access Pack is a one-of-a-kind subscription that unlocks full access to an AI writing assistant, literature recommender, journal finder, scientific illustration tool, and exclusive discounts on professional publication services from Editage.
Based on 22+ years of experience in academia, Editage All Access empowers researchers to put their best research forward and move closer to success. Explore our top AI Tools pack, AI Tools + Publication Services pack, or Build Your Own Plan. Find everything a researcher needs to succeed, all in one place – Get All Access now starting at just $14 a month!
