Skip to main content icon/video/no-internet
icon backReturn to Entries

Bias

Bias is systematic error in data collected to address a research question. In contrast to random errors, which are randomly distributed and therefore even out across people or groups studied, biases are errors that are systematically related to people, groups, treatments, or experimental conditions and therefore cause the researcher to overestimate or underestimate the measurement of a behavior or trait. Bias is problematic because it can endanger the ability of researchers to draw valid conclusions about whether one variable causes a second variable (threats to internal validity) or whether the results generalize to other people (threats to external validity). Bias comes in many forms, including sampling bias, selection bias, experimenter expectancy effects, and response bias.

Sampling Bias

Human participants in studies generally represent a subset of the entire population of people whom the researcher wishes to understand; this subset of the entire population is known as the study sample. Unless a study sample is chosen using some form of random sampling in which every member of the population has a known, nonzero chance of being chosen to participate in the study, it is likely that some form of sampling bias exists. Even for surveys that attempt to use random sampling of a population via random-digit dialing, the sample necessarily excludes people who do not have phone service. Thus people of lower socioeconomic status—and who are therefore less likely to be able to afford phone service—may be underrepresented in samples generated using random-digit dialing. (A modern artifact of random-digit dialing that leads to bias in the other direction is that those with cell phones and many people with land-lines routinely screen calls and are less likely to answer the phone when the incoming call is from someone unknown.)

Researchers often rely on volunteers to participate in their studies, but there may be something different about those who volunteer to participate in studies and those who do not volunteer that systematically biases the sample. For example, people who volunteer to participate in studies of new treatments for psychological disorders may be more motivated to get better than those who do not volunteer, leading researchers to overestimate the effectiveness of a new treatment. Similarly, people who are selected to participate in surveys may choose not to respond to the survey. If there are systematic differences between responders and nonresponders, then the generalizability of the survey findings is limited.

Selection Bias

Selection bias is present when participants in different study conditions possess different characteristics at the start of a study that could influence the outcome measures in the study. Selection bias is presumed in quasi-experimental designs, in which participants are not randomly assigned to experimental conditions, as is often the case in educational research in which classrooms or classes receive different interventions. For example, if a researcher wanted to examine whether students learn more in introductory psychology classes that have a small number of students than they do in classes with a large number, a selection bias may exist if better students are more likely to choose classes that have fewer students. If students in smaller classes outperform students in larger classes, it will be unclear whether the performance difference is the result of smaller classes or because better students self-select into smaller classes.

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Read next

More like this

Sage Recommends

We found other relevant content for you on other Sage platforms.

close

Have you created a personal profile? Login or create a profile so that you can save clips, playlists and searches

close
close
close Signed In to profile You are signed in as:
close
Logged in Institution Institution