Skip to main content

Common Pitfalls of Primary Research

OWL logo

Welcome to the Purdue OWL

This page is brought to you by the OWL at Purdue University. When printing this page, you must include the entire legal notice.



Copyright ©1995-2018 by The Writing Lab & The OWL at Purdue and Purdue University. All rights reserved. This material may not be published, reproduced, broadcast, rewritten, or redistributed without permission. Use of this site constitutes acceptance of our terms and conditions of fair use.


There are a few issues that researchers must confront all of the time. Here are some of the most common ones:

Over generalizing your results

It is impossible to make sweeping generalizations about groups of people based solely on a few interviews, observations, or surveys. You can find general patterns or trends, but should never assume that what you have found is what exists or what will always exist. In fact, it is hard to make concrete generalizations about any occurrence that relates to people because people themselves are dynamic and situations are always changing.

Biased methodology

If you create a biased survey or ask biased questions, you’ll get biased results. See the Creating good survey and interview questions section for tips on how to make your questions non-biased.

Correlation does not imply causation

Remember that just because two results have a relationship between them does not necessarily mean that one causes another to occur. For example, although video games and violent behaviors are shown to have a link, it has not been proven that video games cause violent behavior (instead, it could be that individuals who are predisposed toward violent activity are drawn to violent video games).

Not considering other related factors

It is very difficult to be able to study all the factors that relate to a specific group of people, an event, or an occurrence. Even so, if you do not include these factors within your primary research, they should still be considered when you begin to analyze your data. For example, if you are studying the parking issue on campus and look at the amount of cars being parked on campus vs. the student population, you are omitting other factors like the amount of commuter students, the number of faculty who drive, accessibility of public transportation, as well as many other factors.

Being able to know what data is valid

Some participants in your research may not take it seriously and will provide silly, inaccurate answers or engage in purposely aberrant behaviors. This most likely occurs with surveys that individuals complete but occasionally can occur during interviews or even with observations. These answers can throw off your entire research project, so it is very important that you examine your surveys or interviews for this type of erroneous information. If you find information that is highly questionable, it is best to not include it in your analysis of results. It is important to note that if a participant provides an answer that goes against your hypothesis, you should not just discount that response.

Reported behavior vs. actual behavior

How people report on their behavior might not actually be how they behave. People will often report their own behavior in a more positive light than it may actually be. For example, if you are surveying college students about their study habits, they may report that they study for more hours than they actually do.