What are some ways a researcher can eliminate bias from subjects or participants in quantitative research? This solution reviews some of the common types of bias and threats to validity in quantitative research such as history, maturation, regression, selection, mortality, diffusion of treatment, testing, and instrumentation. It also discusses how to avoid these biases.© BrainMass Inc. brainmass.com October 10, 2019, 4:59 am ad1c9bdddf
When we refer to bias in quantitative research studies, we are often referring to threats to the internal validity of a study. Internal validity is the degree to which the results are accurate and the producedures of the experiment support the ability to draw correct assumptions or inferences about the results. So in order to eliminate bias for participants, we must first understand what types of bias can occur.
Potential Bias/Threats to Validity and Ways to Mitigate Them
1. History - If an experiment/study occurs over a longer period of time, participants may be exposed to different events or experiences that may influence them beyond the conditions of the experiment. For example, if you were conducting an experiment during 9/11, that event may change participants' beliefs and attitudes and bias your end results. To prevent this type of bias, it is helpful for the researcher to use both an experimental and control group that experience the same events. This may be achieved by selecting groups in the same organization or community.
2. Maturation - As a study is being conducted, the participants may mature or change during that time, again skewing the results. For example, if you were conducting a longer-term study of students over the course of a school year or several years it is likely that they will mature and change their attitudes and beliefs as a natural growth process. So, how can you show that the results of your study are due to the treatment or situation you are researching versus just the natural growth process? This is best managed by selecting participants who are the same age and would mature at the same pace throughout the experiment.
3. Regression - This bias occurs when researchers select participants that have extreme scores. For example, if we studied people with high anxiety and low anxiety scores only, it is natural that their scores will change over the course of the study because we are only looking at the extremes. As researchers we want ...
A description of some of the common types of bias in quantitative research and how to avoid or minimize them (includes references!).