Common sources of bias

One of the main problems with scientific studies is that bias (the conscious or unconscious influencing of the study and its results) can make them less dependable. bias can occur in a number of different ways and it is important for researchers to be aware of these and find ways to minimize bias. There are a great number of ways that bias can occur, these are a few common examples:

Recall bias

When survey respondents are asked to answer questions about things that happened to them in the past, the researchers have to rely on the respondents' memories of the past. Sometimes different types of events are more likely to be remembered than others, causing respondents to report those types of experiences more readily. This creates a form of bias called recall bias.

Selection bias

Research samples can sometimes under-represent certain people or groups, and over–represent others. This is called sample selection bias. The best way to select people for research is using the basis of chance, in other words, so that everyone in the population being investigated has an equal chance of being selected. This is called randomisation, because people are randomly selected to take part in the study.

Observation bias (also known as the Hawthorne Effect)

Observation bias occurs when participants in a study are aware that they are being observed by scientists and, either consciously or unconsciously, alter the way they act or the answers they give.

Confirmation bias

Confirmation bias is a type of bias that may occur during the interpretation of study data when researchers, consciously or unconsciously, look for information or patterns in their data that confirm the ideas or opinions that they already hold.

Publishing bias

Studies with negative findings (i.e. trials in which no significant results are found) are less likely to be submitted by scientists or published by scientific journals because they are perceived as less interesting. These 'negative' results are as important for understanding a scientific topic as significant results are but they are less likely to be published. This can skew our understanding of a topic because, for example, when carrying out a review or a meta-analysis on a new drug treatment, if this type of data is missing, it can make it seem like a drug is more or less effective than it actually is. This is called publishing bias.