Author: jdeboe (Page 1 of 4)

When talking about common survey methods, the two that first come to mind with researchers would be interviews and focus groups. However, these research approaches can often fail in reaching their primary goals. When most people think about interviews, the idea of jobs interviews could be the first thought that appears in their head. Keeping this mindset throughout a research interview would most likely lead to the interviewee being more nervous and less willing to tell the complete truth, since their main focus will be finding the “right” answer, instead of finding their answer. One possible way this could be reworded in order to put the participants more at ease would be to introduce them as questionnaires instead of interviews. 

Beyond that, the researchers themselves could also influence the responses from the participants. The way the questions are phrased often has an impact on the answers given. For example, a researcher could ask a question such as “Do you believe the new law that was passed is just?”, or they could ask it as “How do you feel about the new law that was passed?”. The first way may persuade a participant to say yes, while the second way excludes any bias. Many researchers may not realize it, but the way they ask a question can cause their responses to be less accurate.  

So, after a researcher conducts an interview and gets the responses, their study is complete, right? Not quite yet. Perhaps the most important part of conducting interviews is interpreting their results afterwards. Since interviews often provide more qualitative data than quantitative data, grouping similar responses is difficult, but crucial to seeing how many people answered a question a certain way. The struggle is finding a balance between putting similar responses in the same category, while not forcing a complex answer into a certain box. Failure at either one of these could cause many viewpoints to be ignored in a given study. Despite all the difficulties that come with interviews and focus groups, when done right, they can offer a lot of insight into the experiences and struggles of people from any given commnity. 

One of the main problems with data cleaning, as it is, is the connotation that the words give off. Data cleaning refers to the process of wiping away data that is different to the trend already being set up by the rest of the data. The name does not exactly give it justice for how crucial it is in the research process. It gives off the impression that data needs to be “fixed” or “altered” in order to be correct. The best way to phrase this practice would be to compare it to something more similar. Something such as “data-keeping” works better since it relates to housekeeping, which is a better description of what is actually being done during data cleaning. 

Unfortunately, the one difference between housekeeping and data cleaning is that data cleaning does not just remove “bad” data. The data cleaning process can often eliminate data that is significant because of its uniqueness. This could cause the results of a study to be less accurate to the real world. 

A possible solution for this would be to highlight the unique sets of data, instead of hiding them. This could be done by putting a spotlight on regional or even spatial differences. The researchers may not know why these differences are occurring, but that just leaves room for more research to be conducted on this topic in the future. This gives us more possibilities for future discoveries that we might not have had if the original study was “cleaned”. 

Conducting research, whether in the lab or out in the field, is not something that you can just dive right into. The complexity of experiments and all their variables is one factor that makes preparation a necessity in research studies. Add in ethics and the rules, and it makes sense how not just anybody could become a researcher. Many people starting off have no idea what criteria should be met. Thankfully many researchers have made different acronyms or memorable phrases to help ensure new researchers know what to look for in their early plans. One of them being EMRN: Engagement, Mechanics, Reasoning, and Novelty. 

Novelty, in this context, means coming up with an authentic plan that looks for deeper meanings in events. A great research study must come from an idea that is unique and significant to the subject matter. Conducting an experiment based on old ideas or unpopular ideas will not give people any reason to look at the results. This is most likely why many researchers would prefer to conduct fieldwork over lab experiments. Discoveries made on human patterns or interactions cannot always be found in a laboratory. Going out and talking to people is the best way to conduct a groundbreaking study. 

For the Mechanics part of the EMRN criteria, a researcher must make sure their plan is understandable and without fault. However, these “faults” can be from many different causes. For example, one of the biggest changes to research in the 21^st century is implementing rules for ethics.  Which is good for not only the participants of research studies, but the researchers as well, since mistreating volunteers in a study could lead to a shift in the results. But what about the other people who are a part of the study, such as the testers? They are not mentioned a lot, but what if they did not have ethics protection? Those same errors from the participants could still happen with the testers. This is why researchers need to be thorough in looking for “faults” in their studies, because they could easily come from an unsuspected source. 

Doing research for the first time is not easy. There will be challenges the person must overcome. But there are always resources to help early researchers. I know this from personal experience, I am soon to be conducting research for the first time too. Luckily, I know exactly where to go if I get lost.