Blog Week 1: Approaching and Minimizing Bias

Throughout week 1, I have focused on Bias in Clinical Research. Bias is “any tendency which prevents unprejudiced consideration of a question” (Pannucci & Wilkins, 2011). Some examples of bias include under or over-representing one side of the study, not reporting conflicts of interest, and inaccurately expressing the analysis.

In my study of clinical research participation, I need to minimize bias to produce the most accurate and well-informed results. Throughout this week’s blog, I will explain how I expect to counter bias in my study. Of course, bias is impossible to eliminate, but I will do my best to minimize it.

Confirmation bias is looking for data that supports your hypothesis and declining information that disagrees with you. This bias is relatively easy to minimize. I will include all discovered results (including all the outliers).

Selection bias is when the sample group studied doesn’t match the target group that the research will help. For example, if I only studied the people who agreed to participate in my study, then I would miss out on the people who disagreed to participate. To minimize this, I will track the amount of people who decline to participate.

Recall bias is when researchers don’t perfectly recall the past. This leads to inaccuracy when writing down results, analysis, or just notes. I will immediately write down what I observe after each interview or discussion with a patient. 

Finally, observer bias is when a researcher’s opinions or expectations change the study results. This bias stems from a researcher’s prior beliefs. To minimize observer bias, I will use a neutral and open mindset throughout my study.

In conclusion, bias is important to minimize when creating an accurate study. That is why I took measures to examine multiple types of bias.