Coverage error - differences between the target population and the sampling frame; this is produced when the sampling frame does not include all members of the population ○e.g., when a telephone survey omits people without telephones ●Nonresponse error - differences in the characteristics of those who choose to respond toa survey and those who refuse or cannot be contacted because of an insufficient address, wrong phone number, never home, on vacation, etc… ●Measurement error - inaccurate responses associated with the respondent, the interviewer, the survey instrument, and the post survey data processing ●Face-to-Face Interviewing○appropriate when long interviews are necessary ○Computer-assisted personal interviewing (CAPI) - a computer program that can help minimize interviewer mistakes ○Advantage: ■the ability of an interviewer to clarify or restate questions that the respondent does not at first understand ■response rate is typically higher than telephone or mail surveys ●harder to shut the door than throw away an e-mail■more flexible (can probe for death)○Disadvantage: ■interviewers can introduce bias so interviewers must be carefully trained
■higher costs●Telephone interviewing ○random-digit dialing (RDD) - telephone numbers are chosen randomly ○computer-assisted telephone interviewing (CATI) - a set of computerized tools that aid interviewers and supervisors by automating various data-collection tasks○Advantages: ■reduced costs■quickest results■administration and staff supervision are much simpler ■more efficiency■more privacy■better response rate than self-administer ■more detailed response○Disadvantages:■response rates have been declining ■interview duration■complexity of the questions asked■more difficult to establish trust ○tailoring is difficult in telephone interviews due to the absence of visual cues and inadequate time ●Paper and Pencil Questionnaires ○usually mailed to respondents ○Advantages: ■less expensive than telephone and FTF ■takes less time to collect data than FTF ■more questions are left unanswered ■online or e-mailed questionnaires are options ■no interviewer influence■increased privacy and anonymity ○Disadvantages:■in order to get high return rates, researcher must reduce the cost forthe respondent, stress the importances, and have rewards ●e.g., postpaid return envelopes, enclosing small cash payments, ●e.g., making the survey visually appealing, easy to complete, sending a personalized cover letter ■may introduce non response bias due to response selectivity ●e.g., certain groups of people may not have strong enough writing skills or are not interested in the topic■no opportunity for clarification, probing for answers●usually yields the most reliable information when closed questionsare used●must be self-explanatory■coverage and nonresponse error could occur■
You've reached the end of your free preview.
Want to read all 26 pages?
- Spring '15
- researcher, external validity, internal validity