Chapter 5 : Survey Research

Chapter 5 : Survey Research

1. 8. Survey Methods

   • a) Mail surveys
     • i) Advantages
       • Quick, Convenient, Self-Administered (people do themselves), Best for highly personal or embarrassing topics
     • ii) Problem of Response Rate
       • People selected for sample don't complete & return the survey
       • Final sample may be biased - not representative of the population
       • Little control over how people respond to the questions (if someone dont understand the question)
   • b) Personal Interviews
     • i) You have the questions, you ask, you mark the answers for them
     • ii) Researchers gain more control over how survey is administered
     • iii) Interviewers can seek clarification of answers, ask questions
     • iv) Potential Problem: Interview Bias
       • Sometimes interviewer change question to a shorter version so structure become different
       • Occurs when interviewers record only selected portions of answers or changes wording of questions and answers
     • v) Interviews are costly; interviewers must be highly motivated, carefully trained, supervised
   • c) Telephone Interviews
     • i) Can be better than personal interviews because cannot see interviewer so participant is more truthful/honest
     • ii) Complete brief surveys efficiently and with greater access to population
     • iii) Random-digit dialing to select random samples
     • iv) Supervise interviewers easily (interviewer still in control)
     • v) Problems
       • d) Internet Surveys
         • e) Ways to Increase Response Rate
           • i) Questionnaire has a "personal touch" (use name, not dear student)
           • ii) Responding requires minimal effort (easy to access)
           • iii) Topic of survey is interesting to respondents (out of our control)
             • iv) Respondents identify with organization or sponsor of survey (you like the company so you support
         • i) Efficient, low cost, potential for very large samples
         • ii) Samples can be very diverse and access typically underrepresented samples
         • iii) Problems
           • Selection Bias: Access to Internet
           • Response Rate Bias: Willingness to respond
           • Lack of control over research environment cause is self administered
           • Privacy Issue
       • Selection Bias: no phone/multiple phone number/people reject call
         • Interviewer Bias: changes in survey questions & responses (how you ask the questions)
       • Response-Rate Bias: willingness to answer questions on phone

2. 9. Survey Research Designs

   • A plan for conducting a research project
     • Choose research method best suited for answering a particular question
     • c) Longitudinal Design
       • i) Same sample of individuals completes the survey at different points in time (ask you in first sem, second sem, third sem)
         • ii) Can assess how individuals change over time
       • iii) Responses from the sample are generalized to describe changes over time in the population (example: level of spirituality across one year)
       • iv) Attrition: people drop out of the study (sample no longer represents population from which it was selected)
       • v) Reactivity: respondents may strive to be consistent over time or become sensitized (alerted) to the topic [not being true, even if experience diff, I remember my previous answers was like that, I believe I should be consistent to respond based on how i responded before]
       • vi) Longitudinal surveys can't tell why people change over time (only correlations)
       • vii) How to overcome?
         • Attrition: Have a bigger sample size
         • Reactivity: Explain to participants and mention "answer according to current situation"
   • a) Cross-sectional Design
     • b) Successive Independent Samples Design
       • i) Example: student satisfaction (a yearly report of the university)
         • ii) Series of cross sectional designs over time
           • iii) A different sample from the population completes the survey each time
             • iv) Each sample is selected from the same population
       • v) Responses from each sample are used to describe the population at each point in time
         • vi) Compare survey responses from each sample to see how the population changes over time
           • vii) Can determine change over time (give survey to first sem students then next sem give to new first sem students to assess improvements)
       • viii) Can determine population change over time (if population is same)
         • ix) Cannot determine whether individuals change over time (cause asking different individuals over time)
           • x) Problem of noncomparable samples
             • If different populations are sampled each time, response may differ because of true changes over time or because different populations were sampled
     • i) Select sample from one or more populations at one time (results derived at that one time)
       • Choose population of interest
       • Use probabilty/convenience sampling (probability leads to more representative sample)
     • ii) Survey responses are used to
       • Describe population (descriptive statistics)
       • Make predictions for the population (correlations) at that one point of time
     • iii) If samples are drawn from different populations, compare the populations (Taylor's & Sunway)
     • iv) Cannot assess change over time with cross sectional designs (cannot see change in confidence level in first year student)