What Is Selection Bias? | Definition & Examples
Selection bias refers to situations where research bias is introduced due to factors related to the study’s participants. Selection bias can be introduced via the methods used to select the population of interest, the sampling methods, or the recruitment of participants. It is also known as the selection effect.
Due to this, the sample and the target population may differ in significant ways, limiting your ability to generalise your findings.
Selection bias may threaten the validity of your research, as the study population is not representative of the target population.
What is selection bias?
Selection bias occurs when the selection of subjects into a study (or their likelihood of remaining in the study) leads to a result that is systematically different to the target population.
Selection bias often occurs in observational studies where the selection of participants isn’t random, such as cohort studies, case-control studies, and cross-sectional studies. It also occurs in interventional studies or clinical trials due to poor randomization.
Selection bias is a form of systematic error. Systematic differences between participants and non-participants or between treatment and control groups can limit your ability to compare the groups and arrive at unbiased conclusions.
There are several potential sources of selection bias that can affect the study, either during the recruitment of participants or in the process of ensuring they remain in the study. These can include:
- Flawed procedure used to select participants, such as poorly defined inclusion and exclusion criteria
- External reasons that could explain why some participants want to participate in the study while others don’t
- Whether some participants are more likely to be selected than others
Types of selection bias
Selection bias is a general term describing errors arising from factors related to the population being studied, but there are several types of selection bias:
- Sampling bias or ascertainment bias occurs when some members of the intended population are less likely to be included than others. As a result, your sample is not representative of your population.
- Attrition bias occurs when participants who drop out of a study are systematically different from those who remain.
- Volunteer (or self-selection) bias arises when individuals decide entirely for themselves whether or not they want to participate in the study. Due to this, participants may differ from those who don’t – for example, in terms of motivation.
- Survivorship bias is a form of logical error that leads researchers who study a group to draw conclusions by only focusing on examples of successful individuals (the ‘survivors’) rather than the group as a whole.
- Non-response bias is observed when people who don’t respond to a survey are different in significant ways from those who do. Non-respondents may be unwilling or unable to participate, leading to their under-representation in the study.
- Undercoverage bias occurs when some members of your population are not represented in the sample. It is common in convenience sampling, where you recruit a sample that’s easy to obtain.
Examples of selection bias
Selection bias is introduced when data collection or data analysis is biased toward a specific subgroup of the target population.
If you only focus on current customers, their feedback is more likely to be positive than if you also included those who stopped shopping prior to checkout. While current customers had a positive enough experience to ultimately buy something, those who stopped shopping will have different insights. For example, they may be disappointed in the lack of service or the overall web design.
It is important to strive to represent your entire population in your sample. In this case, since a portion of the population is not included, your research runs the risk of undercoverage bias. In this case, results are often not generalizable.
Because of selection bias, study findings do not reflect the target population as a whole.
The cases consist of 200 women with cervical cancer who were referred to Mass General Hospital for treatment. The women were referred to from different parts of the state of Massachusetts. The cases were given a questionnaire asking about their income, education level, employment status, etc.
Control subjects were recruited by interviewers going door to door in the area around the hospital between 9:00 a.m. and 5:00 p.m.
This study runs the risk of selection bias because the two groups were selected in different ways. While the case group was recruited statewide, the control group was recruited from the area near the hospital, and only during typical working hours. This means that women in the control group were more likely to be unemployed. In other words, the control group participants were more likely to be selected if they had the exposure of interest (lower socioeconomic status).
How to avoid selection bias
Selection bias can be avoided as you recruit and retain your sample population.
- For non-probability sampling designs, such as observational studies, try to make the control group as comparable as possible to the treatment group. This method is called matching. Researchers match each treated unit with a non-treated unit of similar characteristics. This helps estimate the impact of a program or event for which it is not ethically or logistically feasible to randomise.
- In experimental research, selection bias can be minimised by proper use of random assignment, ensuring that neither researchers nor participants know which group each participant is assigned to. Otherwise, knowledge of group assignment can taint the data.
- Sampling bias can be avoided by carefully defining the target population and using probability sampling whenever possible. This ensures that all eligible participants have an equal chance of being included in the sample.
Other types of research bias
Frequently asked questions about selection bias
- What are common types of selection bias?
-
Common types of selection bias are:
- Sampling or ascertainment bias
- Attrition bias
- Volunteer or self-selection bias
- Survivorship bias
- Non-response bias
- Undercoverage bias
- Why is bias in research a problem?
-
Bias affects the validity and reliability of your findings, leading to false conclusions and a misinterpretation of the truth. This can have serious implications in areas like medical research where new forms of treatment are being evaluated.
- What is sampling bias?
-
Sampling bias occurs when some members of a population are systematically more likely to be selected in a sample than others.
Cite this Scribbr article
If you want to cite this source, you can copy and paste the citation or click the ‘Cite this Scribbr article’ button to automatically add the citation to our free Reference Generator.
Nikolopoulou, K. (2023, March 24). What Is Selection Bias? | Definition & Examples. Scribbr. Retrieved 24 November 2024, from https://www.scribbr.co.uk/bias-in-research/selection-bias-explained/
