CHAPTER 5 1

Subject:HistoryPrice:16.89 Bought3

Share With

CHAPTER 5 1. Distinguish between external and internal validity. 2. Recent investigations have revealed that educational experiences—especially at the college level influence the types of relationships that people share in families. Prepare and specify a procedure for obtaining matching subjects in control and experimental groups for an experiment designed to gather data on this subject. List the variables that are involved and specify the matching variables. 3. Design an experiment to test the impact of socioeconomic status (income, education, and occupational prestige) on the likelihood of married couples getting divorced. Specify the relative merits of the design that they propose, listing all of the relevant variables concerned. CHAPTER 6 4. Design a research study in order to determine the effect of motorcycle helmet laws on the motorcycle fatality rate. Use library resources in this project because much has been written on this topic. Due to the nature of the data, the units of analysis should probably be states; some states do not have helmet requirements. This provides an excellent opportunity for comparison. 5. Some critics of the modern American family scene have commented that the reason why the divorce rate is so high today is that the liberalization of applicable statutes (like nofault divorce) has made it "too easy" to get a divorce. Obviously, the causal implication here is that legislation like "no-fault" divorce statutes causes the divorce rate to be higher. Students are encouraged to think about how this hypothesis might be tested using correlational, quasi-experimental, or pre-experimental designs. What are the problems involved here? Is the hypothesis testable at all? 6. All college-age men and women are familiar with the concept "War on Drugs." Most are probably aware that much of the Drug Enforcement Administration's efforts focus on interdiction, referring to such strategies as intercepting drug shipments off the coast of Florida before they reach the mainland of the United States. Critics say that interdiction is a waste of time--that as long as there is demand for illegal drugs, there will be supply. Intercept one shipment, and it will be replaced by another. Think about this issue and to come up with an appropriate research design for testing the effectiveness of interdictive efforts to combat illegal drug sales. CHAPTER 7 7. Students enrolled in introductory research methods courses may confuse validity and reliability. In order to assist in clarifying these concepts, construct examples and illustrations of instances where a certain measuring instrument would be valid but unreliable and where another instrument would be reliable but invalid. CHAPTER 8 8. Distinguish between probability and nonprobability sampling, and explain the advantages and disadvantages of each. What sorts of social problems can be investigated with the use of the different sampling procedures? CHAPTER 10 SURVEY RESEARCH INTRODUCTION INTRODUCTION Survey research is an extremely useful approach to studying human behavior that is difficult to observe directly. ? Survey research has been a pivotal methodology central to the social sciences since World War II ? As with all research, survey research attempts to minimize the total error of measurement to gain understanding about relationships between and among variables ? There are four major methods used to elicit information from respondents: the mail questionnaire, the personal interview, and the telephone interview. ? ? More recently, web-blast (email questionnaire) & WinCATI INTRODUCTION In the conceptual, planning, and management phase, who to survey, by what method (telephone, mail, internet, in person), personnel requirements, and financial resources are considered ? Sampling is to be random and a representative estimator of the population parameters ? In designing the survey instrument, survey questions “structure and ordering” must elicit valid and reliable responses and measure what you want them to measure ? ? Open-end vs. Closed-end Question INTRODUCTION ? Pre-Testing the Instrument involves introducing the survey questionnaire into a sub-sample of the population of interest ? Researcher can determine if there are any problems in the instructions, wording of the questions, the order of the questions in terms of context effect, response set bias, or design and formatting of the questionnaire. ? Also, referred to as vetting the survey instrument INTRODUCTION ? ? Creating the codebook for data entry involves assigning unique variable names, variable descriptions, variable width, valid values, value label, and identifying codes for both valid values and missing data Data Analysis & Report Writing is the final stage of the survey process. It involves generating a univariate breakdown, which displays the variable name, description, values, value labels, and missing data. ? May include a bivariate analysis (compares independent variable with the dependent variable) or a multivariate analysis (compares more than two variables) ? May include tables, bar charts, histograms, and/or figures MAIL QUESTIONNAIRE MAIL QUESTIONNAIRE ? Mailed-out Questionnaire is an impersonal survey method, but can be useful ? Advantages – ? Low cost ? Reduction in biasing error ? Greater anonymity ? Time to consider answers and consult personal documents or other people ? Allows wide geographic accessibility ? Disadvantages – ? requires simple questions ? Allow no opportunity for probing ? No control over who fills out the questionnaire ? Low response rate MAIL QUESTIONNAIRE ? The difficulty of securing an acceptable response rate to mail questionnaires requires the use of various strategies that can be adopted to increase the rate of response, including: ? ? ? ? ? ? ? Sponsorship (e.g., U. S. Census Bureau, CDC) Inducements to respond Questionnaire formats and methods of mailing Cover letters Types of mailings (e.g., including prepaid envelop) Timing of mailings Follow-up procedures like the total design method (TDM) ? Questionnaire Construction ? Survey Implementation (e.g., standout from junk mail) and Multiwave follow-up ? ? Selection of respondents. Question: What constitutes an acceptable response rate? Limited by definition of sampling population, not easily resolved (e.g., nature of study & populace characteristic) TABLE 10.1 AVERAGE AND CUMULATIVE RESPONSE RATES TO FOUR MAILINGS Mailing Time Average Cumulative Response Rate Response Rate 1). First mailing 2). Postcard follow-up Week 1 Week 2 23.8% 18.2% 23.8% 42.0% 3). First replacement questionnaire Week 4 17.0% 59.0% 4). 2nd replacement questionnaire sent by certified mail Week 7 13.4% 72.4% Adapted from Donald A. Dillman, James A. Christensen, Edward H. Carpenter, and Ralph M. Brooks, “Increasing Mail Questionnaire Response: A Four-State Comparison,” American Sociological Review, 39 (1974): 755, and Donald A. Dillman and D. E. Moore, “Improving Response Rates to Mail Surveys: Results from Five Surveys,” paper presented at the annual meeting of the American Association for Public Opinion Research, Hershey, PA, 1983. Method Rank (High to Low) Techniques for Increasing Response Rate Follow-up 1 More than one follow-up/telephone can be used for follow-up. Inducement 2 Questionnaires containing a token monetary reward produce better results than ones without. However, the population and the type of questionnaire have to be considered. Sponsorship 3 4 People the respondent knows produce the best results. Method of return ? A regular stamped envelope produces better results than a business reply envelope. Format ? Esthetically pleasing cover; a title that will arouse interest; an attractive page format. ? • Nonreaders and non-writers are excluded from participation. • Interest in or familiarity with the topic under investigation is a major factor in determining the rate of return. • The better educated are more likely to return the questionnaires. • Professionals are more likely to return questionnaires. Introductory letter Selection of respondents An altruistic appeal seems to produce the best results. PERSONAL INTERVIEW SURVEYS PRINCIPLES OF INTERVIEWING The first step in the interviewing process is getting the respondent to cooperate and provide the desired information. ? Three factors help in motivating respondents to do this: ? ? They need to feel that their interactions with the interviewer will be pleasant and satisfying ? They need to see the investigation as being worthwhile ? Barriers to the interview in the respondents' minds need to be overcome. PRINCIPLES OF INTERVIEWING ? Probing is a technique used in the interviewing process designed to stimulate discussion and obtain more information. Probes have two major functions: ? They motivate the respondent to elaborate or clarify an answer or to explain the reasons behind the answer ? They help focus the conversation on the specific topic of the interview. ? In general, the less structured the interview, the more important probing becomes as an instrument for eliciting and encouraging further information. PERSONAL INTERVIEW SURVEYS ? Personal Interview (face-to-face) ? Schedule-Structured Interview (lest flexible) – number of questions and the wording of the question are identical for all respondents (e.g., no rewording, explanations, or clarifications) Automobile 1. Wanting to learn to drive 2. Getting a driver’s license 3. Wanting to use the family car 4. Using it too much 5. Keeping the car clean 6. Repairing the car 7. Driving someone else’s car 8. Want to own a car 9. The way you drive your own car 10. Other Never Only Once More Than Once Many Times PERSONAL INTERVIEW SURVEYS ? Personal Interview (face-to-face) ? ? ? ? ? Focused Interview (non-structured) ? Respondent known to be involved with a particular experience ? Situations have been analyzed prior to interview ? Interviewer guides question along specific topics related to research hypotheses ? Interview remains focused on subjects’ experiences regarding the situation under study Instructions to the interviewer: 1. Your task is to discover as many speci?c kinds of con?icts and tensions between child and parent as possible. The more concrete and detailed the account of each type of con?ict, the better. 2. Although there are 4 areas of possible con?ict that we want to explore (listed in question 3 below), you should not mention any area until after you have asked the ?rst two questions in the order indicated. 3. The ?rst question takes an indirect approach, giving you time to build up rapport with the respondent. What sorts of problems do teenagers have in getting along with their parents? (Possible probes: Do they always agree with their parents? Do any of your friends have “problem parents”?) What sort of disagreements do you have with your parents? (Possible probes: Do they cause you any problems? In what way do they try to restrict you? Do they like the same things you do?) Have you ever had any disagreement with either of your parents over: a) Using the family car? b) Friends of the same sex? c) Dating? PERSONAL INTERVIEW SURVEYS ? Nondirective Interview (most flexible) ?Researcher does not employ a schedule of prespecified, ordered questions ?Respondent is encouraged to relate their experiences reveal their opinions and attitudes as they see fit, with little or no direction from the researcher Personal interviews are extremely effective when dealing with heterogeneous populations and when detailed information is required. ? They provide good control of the interview situation. ? PERSONAL INTERVIEW VERSUS MAIL QUESTIONNAIRE ? Personal interviews offer certain advantages over mail questionnaires: ? Greater flexibility ? Control of the interview situation ? High response rate ? Collection of supplementary information (Fuller information) ? Personal interviews display three disadvantages: ? Higher cost ? Interviewer bias ? Lack of anonymity. TELEPHONE INTERVIEW SURVEYS TELEPHONE INTERVIEW SURVEYS ? The telephone survey is a semi-personal method of collecting information ? ? ? ? ? Until recently, this technique was regarded with great skepticism due to the risk of a serious sampling bias that economically underprivileged people were unlikely to have telephones. Today, over 93 percent of the population have telephones, so this bias is minimized. A technique known as random-digit dialing (RDD) permits random sampling in a telephone survey, thus eliminating the problems inherent in using telephone directories (unlisted numbers, disconnected phones, and so on). One of the latest developments in telephone surveys is Windows computer-assisted telephone interviewing (Win CATI). Recently, the telephone interview has become more popular among researchers due to the rising cost of personal interviews. TELEPHONE INTERVIEW SURVEYS ? Advantages ? Moderate Cost ? Speed - 1). Telephone interviewers can reach a large number of respondents in a short time. 2). Interviewers can code data directly into computers, which can later compile the data High response rate – 1). Telephone interviews provide access to people who are unlikely to reply by mail or will refuse a personal interview 2).Telephone interviews are most advantageous in terms of their speed, a high response rate, and applicability to heterogeneous populations. Quality - High-quality data can be collected when interviewers are centrally located and supervisors can ensure that questions are being asked correctly and answers recorded properly. ? ? TELEPHONE INTERVIEW SURVEYS ? Disadvantages ? Reluctance to discuss sensitive topics over the phone ? Respondents can terminate the interview before it is completed ? Interviewer cannot provide supplemental information about the respondents’ characteristics of environment TYPES OF TECHNOLOGY-BASED SURVEYS ? Computer-assisted survey information collection (CASIC) ? Used for the integration of the computer into the collection of data from respondents, transmitting of the data to remote sites, and the post-collection analysis and processing of survey data ? Computer-assisted personal interviewing (CAPI) ? Interviewer brings a portable computer, on which CAPI software is loaded and reads questions from it and enters the responses directly ? Allows respondent to be interviewed in-home, at business, in mall TYPES OF TECHNOLOGY-BASED SURVEYS ? Continuous Audience Response Technology (CART) ? Instant polling of focus groups where the principal mode to analysis is careful listening by the moderator ? Involving “a disciplined process, systematic steps, a defined protocol, verifiable results, and multiple feedback loops ? Email and Web Surveys (Using the Internet) ? Used independently as a text message or via a website, respectively, or in conjunction with one another ? Computer-assisted Telephone Interviewing (CATI) ? A hardware and software technology that is a convenient cost-effective method of obtaining interview data WINCATI RESEARCH FACLITIES EVALUATION OF THREE SURVEY METHODS Criterion •The Cost Survey method is one of the most important data collection methods in the social sciences. •It’s used extensively to collect information on numerous subject of research. •Due Personal Interview High High High Low Moderate Low High Low Moderate Applicability to Moderate High geographically dispersed populations Moderate Response rate Control of Interview situation to public demands for Applicability to High government accountability, heterogeneous it is widely emphasized. populations •Is a widely used tool of various government organizations. Mail Telephone Low High Collection of detailed information High Mode Moderate -rate Speed Low Low High SURVEY RESEARCH VIDEOS ? The Survey ? ? How to do a street intercept survey? From CR Market Surveys. ? ? http://www.youtube.com/watch?v=tS0DVtfmKS8 Truth in Survey Research ? ? http://www.youtube.com/watch?v=o3nHdx8Ja5g http://www.youtube.com/watch?v=anOCZ1Sw04g Survey Software Quest Manager-CATI in Focus ? http://www.youtube.com/watch?v=ezQGUxOvrnk