Sampling Case Study

Problem: Researchers wanted to determine how aware adult Hispanics in the San Francisco area are of product warning messages and signs concerning cigarettes, alcoholic beverages, and other consumer products. The researchers needed to contact a sample of them to interview. A Solution: “Respondents were sampled using…random digit dialing. This procedure…avoids the use of directories with their inherent problem of the exclusion of unlisted numbers while guaranteeing equal probability of selection for all telephone numbers. Such an approach has been used in a number of telephone surveys with Hispanics and has been shown to be not only feasible but also very efficient… “A household was considered to be eligible if those answering the telephone self-identified as Hispanics. Within a given household, the respondent was selected by asking for the Hispanic resident who had most recently celebrated a birthday and who was between 18 and 65 years of age. Interviews were conducted in the language of preference of the respondent (English or Spanish) by bilingual, bicultural interviewers.” Your Opinions:

1.Note that using random digit dialing gives all telephone numbers, including unlisted numbers, an equal chance of being dialed. Does this procedure give all adult Hispanics in the San Francisco area an equal chance of being selected for the study? Why or why not? 2.Are there problems with using self-identification as a Hispanic as a criterion for eligibility to participate in the study? Explain your answer. 3.Speculate on why the interviewer asked to speak with the Hispanic resident between 18 and 65 years of age who had most recently celebrated a birthday. 4.Would you be interested in knowing whether some Hispanics who answered the phone refused to be interviewed? Why or why not? 1 Source/ reference: Marín, G. (1994). Self-reported awareness of the presence of product warning messages and signs by Hispanics in San Francisco. Public Health Reports, 109, 275-283. Submit your opinion to the case study to this threaded discussion.

Case Study 2 on Sampling
Using a Sample of Convenience1
Problem: Professors often use samples of college students as participants in their research because they are convenient. Using such samples can be problematic. A Solution: A researcher was interested in women’s preferences for married names (traditional – taking the husbands’ surnames and nontraditional – retaining their surnames or hyphenating the two surnames). The researchers surveyed 150 unmarried women students enrolled in an introductory psychology course. They asked the women to indicate what they would do, assuming that their husbands had no opinion on the matter. They were told “If you have not thought much about this issue, please consider it now and give your opinion even if you are not 100% certain.” They were also asked to give the reasons for their name choice. Almost 60% of the women chose to take the husband’s name. In their conclusion, the researchers noted that “one of the most striking results of this study is many respondents’ unquestioning attitudes toward” taking the husband’s surname as indicated by the fact that many gave responses such as “I don’t know, I guess it’s tradition,” or simply writing the work “tradition” as their reason. Your Opinions:

1. Because of the nature of the sample, this study provides information on what women plan to do. Do you think this is as interesting as gathering information on what a sample of married women have done? Why or why now? 2. The researchers apparently suspected that some participants in this sample may not have given the issue much thought before participating in the study and might not be certain of their opinion. In light of this, would it have been interesting to have the students rate the degree of certainty they had in their answers? Why or why not? 3. What is your opinion on having the women assume their husbands had no opinion? (Note that because of the nature of the sample, the researchers either needed to provide an assumption on this matter or allow the participants to make their own assumptions.) 4. Are you surprised by the “unquestioning attitudes toward name traditions”? Why or why not? 1 Source/ reference: Twenge, J.M. (1997). “Mrs. his name”: Women’s preferences for married names. Psychology of Women Quarterly, 21, 417-429. Submit your opinion to the case study to this threaded discussion.

Case Study 3 on Sampling
Using Quota Sampling1
Problem: Professors with limited resources usually find it difficult to obtain adequate samples of the general adult population for their research. Instead, many study samples of college students, who are readily available, even though there is no reason to believe that samples of college students are representative of the national population of adults. (Of course, often professors are interested in the behavior of college students only, in which case, using college students is not a problem.) A Solution: “In order to reach a wide range of people, the data were collected using a purposive, quota design. Students enrolled in a research methods class…were given course credit for collecting the data for this study. These research assistants were…instructed to recruit adults (not college students) that they knew in six age and gender quotas (male and female: 18-34, 35-49, and 50 and older) to complete questionnaires. Because the students were given their assignment in the weeks prior to spring break, most of them contacted adults from their home areas. The adult respondents completed the written self-administered questionnaires.. The research assistants returned a total of 566 questionnaires.” Your Opinions:

1.Speculate on why the research assistants were told to recruit adults who were “not college students.” 2.Do you think that the researcher achieved her goal of reaching a “wide range of people”? Why or why not? 3.Do you think that the sample is probably representative of the adult population of the United States? Why or why not? 4.In addition to gender and age, are there other variables for which quotas could have been established that would improve the sample? Explain your answer. 1 Source/ reference: Perse, E.M. (1996). Sensation seeking and the use of television for arousal. Communication Reports, 9, 37-48. Submit your opinion to the case study to this threaded discussion.

Case Study 4 on Sampling
Using an Incentive to Increase Response Rate1
Problem: When attempting to interview members of the general population, a substantial percentage of the sample often refuses to be interviewed. A Solution: Researchers sent a letter to a random sample of households indicating that an interviewer would be calling on them. As an incentive, the researchers included a gift-boxed ball point pen (imprinted with the words “University of Michigan” in gold) along with a note saying that the pen was a token of appreciation for participating in the upcoming interview. Drawbacks to the Solution: The cost of the study was increased by $3.98 ($3.75 for the pen and $0.23 for additional postage) per potential respondent contacted. In addition, the incentive might “alter the composition of the responding sample.” For example, it might be more effective gaining the participation of people from low socioeconomic status groups than higher socioeconomic groups. Furthermore, an incentive might alter the way in which respondents answered the questions, which asked about demographic variables, participants’ recollection of recent events in history, and their attitudes toward social issues. Your Opinions:

1.Do you think it would be better to send a letter offering a pen only to those who complete the interviews instead of giving away pens to all potential respondents before the interviewers contact them? Why or why not? 2.In your opinion, how serious is the possible drawback that the incentive might “alter the composition of the responding sample”? Explain your answer. 3.If you had received the gift pen, would you be likely to alter your answers to the questions because of it? For example, would you be more likely to try to “please the interviewer” by giving answers you think might please him or her? Why or why not? 4.Are there other relatively inexpensive gifts that might be used as incentives? If yes, name some. Are they likely to be as effective as a gift pen? Why or why not? 1 Source/ reference: Willimack, D.K., Schuman, H., Pennell, B., & Lepkowski, J.M. (1995). Effects of a prepaid nonmonetary incentive on response rates and response quality in a face-to-face survey. Public Opinion Quarterly, 59, 78-92. Submit your opinion to the case study to this threaded discussion.

Case Study 5 on Sampling
Obtaining a Sample of Psychotherapy Clients
Problem: Researchers wanted to survey “a large, diverse sample of psychotherapy patients” for a study of the patients’ positive and negative evaluations of being physically touched by psychotherapists. A Solution: The researchers decided to draw a sample of therapists and have them distribute anonymous questionnaires to their patients (neither the patients’ nor therapists’ names were to be placed on the questionnaires). “A list of therapists in a large Southern metropolitan area was generated from state professional association and telephone directories of therapists (and upon recommendation of therapists and respondents, expanded to include therapists in other states). Over 300 therapists were contacted. Of the 900 [questionnaires] distributed, approximately 125 were mailed to cities and mid-sized towns in other states in the Midwest, the Northeast, and the Southeast.” Two hundred thirty-one usable questionnaires were returned. Your Opinions:

1.In your opinion, was the researchers’ method of sampling likely to result in “a large, diverse sample” of patients? Why or why not? 2.Two hundred thirty-one of the 900 questionnaires were returned. Does this cause a sampling problem? Why or why not? 3.The researchers state that over 300 therapists were contacted, but they do not indicate how many agreed to distribute questionnaires. As a consumer of research, would you be interested in having this information? Why? 4.The therapists knew the purpose of the study and the contents of the questionnaires. Is it possible that some of the therapists felt threatened by the study even though patients’ responses were to be anonymous? Is this a serious threat to the study? Why or why not? 1 Source/ reference: Horton, J.A., Clance, P.R. Sterk-Elifson, C., Emshoff, J. (1995). Touch in psychotherapy: A survey of patients’ experiences. Psychotherapy, 32, 443-457. Submit your opinion to the case study to this threaded discussion.