Please enable JavaScript.
Coggle requires JavaScript to display documents.
Survey Research, Types of Biases, A more personalized form of data…
-
Types of Biases
Recall bias: Pertains to respondents not adequately remembering their motivations, behaviors or memory of an event. Researcher can overcome recall bias by anchoring respondent's memory in specific events rather then asking them to recall from memory
Sampling bias: Can cause the respondent sample to be unrepresentative of the population and impacts the generalizability claims. Example is conducting telephone survey by calling random sample of publicly available telephone numbers. This excludes unlisted numbers, cell phone numbers and respondents who cannot answer the phone
**Social desirability bias: Tendency for respondents to spin the truth to portray themselves in a socially desirable manner. This impacts the validity of responses
Common method bias: refers to the amount of covariance shared between independent and dependent variables measured at the same point in time and using the same instrument. Statistical tests such as Harmon's single factor test
Non-response bias: Targeted respondents fail to return a survey and impacts the validity of the results. Response rates can be improved by offering incentives, advanced notification, follow up requests, endorsement or a respondent friendly questionnaire
A more personalized form of data collection and are conducted by trained interviewers. May contain space for observational notes and comments. Can be time consuming and resource intensive
-
Roles of Interviewer
-
Motivate respondents by demonstrating enthusiasm, communicate the importance of research and be attentive to respondents needs
Locate and enlist cooperation of respondents by selling the idea of participating in the study and work around respondents schedules
Prepare for interview by knowing interview process and survey method. Know the purpose of study and how responses will be stored and used. Should rehearse and time the interview prior to formal study
Observe quality of response and may supplement responses by using observations of gestures or body language
Conduct Interview
Prepare a kit containing cover letter, copies of survey instrument, photo identification and telephone numbers of respondents.
Follow the script and ask questions as written. Use probing techniques such as the silent probe, overt encouragement, ask for elaboration and reflection
Thank respondents for time and inform when to expect results. After leaving, write down notes or key observation
Types of Interviews
Face to Face : Interviewer works directly with respondent to ask questions and record their responses
Telephone Interview: Interviewers contact potential respondents over the phone using a random selection of a telephone directory and ask standard set of survey questions. Recent advance approach is a CATI(computer assisted telephone interview)
Focus Group: A small group of respondents are interviewed in a common place and time together. Used for exploratory research rather than descriptive or explanatory research
A research instrument consisting of a set of questions intended to capture responses from respondents in a standardized manner. Questions can be unstructured or structured. Unstructured ask respondents to provide responses in their own words. Structured allows for an answer to be selected from choices
Question Rules
-
-
-
Keep survey short, thank respondents, assure respondents of confidentiality and how responses will be recorded and reported and send out copy of final results
Types of Questionnaire
Self administered mail survey : same questionnaire is mailed to large number of people and respondents can complete survey at their convenience in a prepaid postage envelope. Surveys are inexpensive to administer but have low response rate
Group administered :Sample of respondents are brought together in a common place and time. Format is convenient to researcher and has a high response rate
Online/Web survey: Respondents receive an email requesting participation and a link to complete the survey. Surveys are inexpensive and results are instantly recorded and accessible