We learned about these things through research. Someone somewhere did a study and found each of the above findings. While most of us accept the value of good nutrition for children and the dangers of heroin, at least some of us (especially Certs Sugarless Mint and Egg McMuffin fans) would quibble with the other two findings. Just how did the researchers do the studies that found those numbers (by the way, get in the habit of questioning how researchers find their results, even those you agree with)? Who knows, maybe they did the Croissanwich study at Burger King one morning and asked people which sandwich they preferred. The fact that the participants had already gone to Burger King for breakfast meant that they had at least some tolerance for Croissanwiches. What about the dentists? What percentage of them "recommended" candies of any kind? Which dentists participated in the study and what exactly do they prefer about sugarless gum over sugarless mints?
When we ask questions like these, we are questioning the methods the researchers employed in their studies. Research methods are a variety of techniques that people use when studying a given phenomenon. They are planned, scientific, and value-neutral. What that means is that good research methods don't "just happen." Instead, they are deliberately employed in a way that is designed to maximize the accuracy of the results. Assume, for example, that you want to do a survey to assess students' level of satisfaction with the parking situation on your campus. Because no one wants to take time to be interviewed, you are forced to start interviewing the people sitting in their cars waiting for parking places. Since they're bored anyway and have nothing better to do, they agree to answer your questions. Lo and behold, you find that all 50 people you talk to are VERY unhappy about the lack of parking spaces on campus.
Based on your research, can you say that the students at your university are dissatisfied with the parking situation? Even if the students in general really are unhappy, your study doesn't provide a strong answer to the question. What about the other 9,950 students that attend your school? What do they think? If they are all happy, then your 50 research participants represent less than one percent of the students. Ooops, major problem! Your sample was biased. You unnecessarily increased your chances of error by relying on a bogus sample.
A better way for you to do your survey would be to randomly select names from a list of all enrolled students (there will be more on random samples in a later section). You could ask them what they think about the parking situation. Because your sample was randomly selected, you would expect their answers to reflect what students as a whole felt about the parking situation. The use of random samples is just one way that researchers try to ensure that the answers they find are accurate.
Some common research techniques are surveys, experiments, and field research. No matter what technique you use, you will pay attention to certain things. The sample, or people you study, is one very important concern as we learned above. Studying the wrong group of people means that others can criticize your study. You often see this happen when magazines and other publications report the results of their monthly polls. Who answers these polls, anyway? The question other researchers have is usually: How representative of the rest of the population is the sample? Just because 37 out of 50 women surveyed by National Enquirer preferred chocolate to sex doesn't mean anyone else does (this was a real survey reported on 10/11/94!).
Another way that researchers try to increase accuracy in their findings is by trying to ask questions in a non-biased, easy-to-understand manner. How a question is worded can affect the responses that people provide. Because people react to subconscious cues contained within questions, for example, lawyers are not allowed to ask witnesses "leading" questions in court (i.e., questions that suggest an answer, such as "He was standing at the top of the stairs, wasn't he?"). The same applies to research; researchers have to be very careful not to "lead" their respondents or they may get biased responses. Likewise, asking questions that are difficult to understand will increase the chances that respondents provide biased answers. Just because I don't know what something means doesn't mean that I will tell you I don't understand; instead, I'll just smile and say that I agree that spelunkers should be banned from talking about their hobbies at elementary schools. My response, of course, would be biased because I don't know that spelunkers are cave explorers, and cave explorers are okay with me.
Recap Q1: Which of the following would be the best group to survey to get general attitudes?
A. religious leaders about their views toward abortion
C. members of the NRA (National Rifle Association) about their views regarding gun control.
D. the first person on each page of the Los Angeles telephone book about their views of the LAPD
You have now completed the section on "What are research methods?" Click here to return to the main menu.
Page last updated: April 15, 2008 --> --> -->