Introduction
Survey research is common in both science and daily life. Almost everyone has been exposed to survey research in one form or another. Data about the social world can be obtained in many ways, including observation, field studies, and experimentation. Two key methods for obtaining data—questionnaires and interviews—are survey research methods. Most of the social research conducted or published involves these two data collection methods.
In general, when using survey methods, the researcher gets information directly from each person (or respondent) by using self-report measurement techniques to ask people about their attitudes, behaviors, and demographics (statistical features of populations, such as age, income, race, and marital status), in addition to past experiences and future goals. In questionnaires, the questions are in written format and the subjects write down their answers. In interviews, the interviewer engages in one-to-one verbal communication with the respondent, either face-to-face or by means of a telephone. Both techniques are flexible and adaptable to the group of people being studied and the particular situation. Both can range from being highly structured to highly unstructured.
Strengths and Limitations
Questionnaires can be completed in groups or self-administered on an individual basis. They can also be mailed to people. They are generally less expensive than conducting interviews. Questionnaires also allow greater anonymity of the respondents. One drawback is that a questionnaire requires that the subjects understand exactly what the questions are asking. Also, people filling out the questionnaire may get bored or find it tedious to write down their answers. The survey researcher must therefore make sure that the questionnaire is not excessively long or complex.
In contrast, with an interview there is a better chance that the interviewer and subject will have good communication and that all questions will be understood. Telephone interviews are less expensive than face-to-face interviews; still, questionnaires tend to be less costly. In an interview, the respondent is presented with questions orally, whereas in the questionnaire, regardless of type or form, the respondent is presented with a written question. Each data collection device has pros and cons. The decision to use questionnaires versus interviews depends on the purpose of the study, the type of information needed, the size of the sample (the number of people who participate in a study and are part of a population), the resources for conducting the study, and the variable(s) to be measured. Overall, the interview is probably the more flexible device of the two.
Designing Bias-Free Questions
When creating a questionnaire, the researcher must give special thought to the wording of the questions. Researchers must avoid questions that would lead people to answer in a biased way, or ones that might be easily misinterpreted. For example, the questions “Do you favor eliminating the wasteful excesses in the federal budget?” and “Do you favor reducing the federal budget?” might well result in different answers from the same respondent.
Questions are either closed- or open-ended, depending on the researcher’s choice. In a closed-ended question, a limited number of fixed response choices are provided to subjects. With open-ended questions, subjects are able to respond in any way they like. Thus, a researcher could ask, “Where would you like a swimming pool to be built in this town?” as opposed to “Which of the following locations is your top choice for a swimming pool to be built in this town?” The first question allows the respondent to provide any answer; the second provides a fixed number of alternative answers from which the person must choose. Use of closed-ended questions is a more structured approach, allowing greater ease of analysis because the response alternatives are the same for everyone. Open-ended questions require more time to analyze and are therefore more costly. Open-ended questions, however, can provide valuable insights into what the subjects are actually thinking.
Clinical Interview
A specialized type of interview is the clinical, or therapeutic, interview. The specific goal of a particular clinical interview depends on the needs and the condition of the individual being interviewed. There is a distinction between a therapeutic interview, which attempts both to obtain information and to remedy the client’s problem, and a research interview, which attempts solely to obtain information about people at large. Because the clinical interview is a fairly unstructured search for relevant information, it is important to be aware of the factors that might affect its accuracy and comprehensiveness. Research on hypothesis confirmation bias suggests that it is difficult to search for unbiased and comprehensive information in an unstructured setting such as the clinical interview. In the context of the clinical interview, clinicians are likely to conduct unintentionally biased searches for information that confirm their early impressions of each client. Research on self-fulfilling prophecies suggests a second factor that may limit the applicability of interviews in general: the interviewer’s expectations may affect the behavior of the person being interviewed, and respondents may change their behavior to match the interviewer’s expectations.
Role of Scientific Method
Knowing what to believe about research is often related to understanding the scientific method. The two basic approaches to using the scientific method, the descriptive and the experimental research approaches, differ because they seek to attain different types of knowledge. Descriptive research tries to describe particular situations; experimental research tries to determine cause-and-effect relationships. Independent variables are not manipulated in descriptive research. For that reason, it is not possible to decide whether one thing causes another. Instead, survey research uses correlational techniques, which allow the determination of whether behaviors or attitudes are related to one another and whether they predict one another. For example, how liberal a person’s political views are might be related to that person’s attitudes about sexuality. Such a relationship could be determined using descriptive research.
Survey research, as a widely used descriptive technique, is defined as a method of collecting standardized information by interviewing a representative sample
of some population. All research involves sampling of subjects. That is, subjects must be found to participate in the research, whether that research is a survey or an experiment. Sampling is particularly important when conducting survey research, because the goal is to describe what a whole population is like based on the data from a relatively small sample of that population.
Kinsey Group Research
One famous survey study in the mid-1930s was conducted by Alfred Kinsey
and his colleagues. Kinsey studied sexual behavior. Until that time, most of what was known about sexual behavior was based on what biologists knew about animal sex; what anthropologists knew about sex among indigenous peoples in non-Western, nonindustrialized societies; or what Sigmund Freud learned about sexuality from his emotionally disturbed patients. Kinsey and his colleagues were the first psychological researchers to interview volunteers from mainstream American society about their sexual behaviors. The research was hindered by political investigations and threats of legal action. In spite of the harassment encountered by the scientists on the project, the Kinsey group published two reports: Sexual Behavior in the Human Male in 1948 and Sexual Behavior in the Human Female in 1953.
The findings of the Kinsey group have benefited the public immensely. As a result, it is now known that the majority of people (both men and women) interviewed by the Kinsey group masturbated at various times, but that more men than women said they masturbated. Data collected by the Kinsey group on oral-genital sexual practices have allowed researchers to discover that since the 1930s, attitudes toward oral-genital sex have become more positive. Their research also shocked the nation with the discovery that the majority of brides at that time were not virgins.
When scientific sampling techniques are used, the survey results can be interpreted as an accurate representation of the entire population. Although Kinsey and his associates helped to pave the way for future researchers to be able to investigate sexual behaviors and attitudes, there were some problems with the research because of its lack of generalizability. The Kinsey group’s research is still the largest study of sexual behavior ever completed. They interviewed more than ten thousand people; however, they did not attempt to select a random or representative sample of the population of the United States, which meant that the responses of middle-class, well-educated whites were overrepresented. There is also a problem with the accuracy of the respondents’ information, because of memory errors, exaggerations, or embarrassment about telling an interviewer personal, sensitive information. Despite these limitations, the interviewing conducted by Kinsey and associates made great strides for the study of sexuality and great strides for psychology in general.
Importance of Sampling Procedures
When research is intended to reveal very precisely what a population is like, careful sampling procedures must be used. This requires defining the population and sampling people from the population in a random fashion so that no biases will be introduced into the sample. To learn what elderly people think about the medical services available to them, for example, a careful sample of the elderly population is needed. Obtaining the sample only from retirement communities in Arizona would bias the results, because these individuals are not representative of all elderly people in the population.
Therefore, when evaluating survey data, a researcher must examine how the responses were obtained and what population was investigated. Major polling organizations such as the Gallup organization typically are careful to obtain representative samples of people in the United States. Gallup polls are frequently conducted to survey the voting public’s opinions about the popularity of a presidential candidate or a given policy. Many other surveys, however, such as surveys that are published in popular magazines, have limited generalizability because the results are based on people who read the particular magazine and are sufficiently motivated to complete and mail in the questionnaire. When Redbook, for example, asks readers to write in to say whether they have ever had an affair, the results may be interesting but would not give a very accurate estimate of the true extent of extramarital sexual activity in the United States. An example of an inaccurate sampling technique was a survey by Literary Digest (a now-defunct magazine) sampling almost ten million people in 1936. The results showed that Alfred Landon would beat Franklin D. Roosevelt by a landslide in that year’s presidential election. Although the sample was large, the results were completely inaccurate.
Early Survey Methods
One of the earliest ways of obtaining psychological information using descriptive techniques was through clinical interviewing. The early interviews conducted by Freud in the late 1800s were based on question-and-answer medical formats, which is not surprising, considering that Freud was originally a physician. Later, Freud relied on the less structured free-association technique. In 1902, Adolf Meyer developed a technique to assess a client’s mental functioning, memory, attention, speech, and judgment. Independent of the style used, all the early clinical interviews sought to get a psychological portrait of the person, determine the source of the problem, make a diagnosis, and formulate a treatment. More detailed studies of interviews were conducted in the 1940s and 1950s to compare and contrast interviewing styles and determine how much structure was necessary. During the 1960s, much research came about as a result of ideas held by Carl R. Rogers, who emphasized the interpersonal elements he thought were necessary for the ideal therapeutic relationship; among them are warmth, positive regard, and genuineness on the part of the interviewer.
In the 1800s and early 1900s, interviews were used mainly by psychologists who were therapists helping people with problems such as fear, depression, and hysteria. During that same period, experimental psychologists had not yet begun to use survey research methods. Instead, they used introspection to investigate their own thought processes. For example, experimental psychologist Hermann Ebbinghaus gave himself lists of pronounceable nonsense syllables to remember; he then tested his own memory and attempted to improve it methodically. Many experimental psychologists during this period relied on the use of animals such as dogs and laboratory rats to conduct behavioral research.
Evolution of Questionnaires
As mentioned above, one of the first attempts by experimental psychologists to study attitudes and behaviors by means of the interview was that of the Kinsey group in the 1930s. At about that same time, L. L. Thurstone, an experimental social psychologist, formalized and popularized the first questionnaire methodology for attitude measurement. Thurstone devised a set of questionnaires, or scales, that have been widely used for decades. He is considered by many to be the father of attitude scaling. Soon thereafter, Rensis Likert
made breakthroughs in questionnaire usage with the development of what are known as Likert scales. A Likert scale provides a series of statements to which subjects can indicate degrees of agreement or disagreement. Using the Likert technique, the respondent answers by selecting from predetermined categories ranging from “strongly agree” to “strongly disagree.” It is fairly standard to use five categories (strongly agree, agree, uncertain, disagree, strongly disagree), but more categories can be used if necessary. An example of a question using this technique might be, “Intelligence test scores of marijuana users are higher on the average than scores of nonusers.” The respondent then picks one of the five categories mentioned above in response. Likert scales have been widely used and have resulted in a vast amount of information about human attitudes and behaviors.
Bibliography
Abbott, Martin Lee, and Jennifer McKinney. Understanding and Applying Research Design. Hoboken: Wiley, 2013. Print.
Bordens, Kenneth S., and Bruce B. Abbott. Research Design and Methods: A Process Approach. 7th ed. Boston: McGraw-Hill, 2008. Print.
Coolican, Hugh. Research Methods and Statistics in Psychology. 5th ed. New York: Routledge, 2013. Print.
Cozby, Paul C. Methods in Behavioral Research. 10th ed. Boston: McGraw-Hill Higher Education, 2009. Print.
Dillman, Don, Jolene D. Smyth, and Leah Melani Christian. Internet, Mail, and Mixed-Mode Surveys: The Tailored Design Method. 3rd ed. Hoboken: Wiley, 2009. Print.
Hoyle, Rick H., Monica J. Harris, and Charles M. Judd. Research Methods in Social Relations. 7th ed. Fort Worth: Wadsworth, 2002. Print.
Saris, Willem, and Irmtraud N. Gallhofer. Design, Evaluation, and Analysis of Questionnaires for Survey Research. 2nd ed. Hoboken: Wiley, 2014. Digital file.
Stewart, Charles J., and William B. Cash, Jr. Interviewing Principles and Practices. 12th ed. Boston: McGraw-Hill Higher Education, 2008. Print.
No comments:
Post a Comment