We offer consultation on all aspects of survey methodology. Brief consultations are offered at no charge (pending availability) to UNC students, faculty, and staff. We offer review and consultation to students working on dissertations, theses, or other research papers, and consult with students and faculty writing funding proposals. We do NOT provide consultation on class projects or assignments. For extended consultation, our time may be purchased on an hourly basis (by advance arrangement) through the Odum Institute Re-Charge Center. In certain situations and pending availability, hourly fee-for-service consultation may also be provided to organizations outside the UNC community.
Common topics of consultation include:
- Questionnaire review
- Question wording (respondent comprehension, unbiased wording)
- Question ordering
- Response formats (closed-ended questions, scales, response categories)
- Formatting self-administered questionnaires (mail questionnaires and Web surveys)
Please note that Teresa provides consultation on standardized survey instruments for quantitative statistical analysis of survey results. For consultation on instrument design for qualitative data analysis, see Paul Mihas.
Data Collection Modes and Methods
- Choice of data collection mode (mail, Web, telephone, mixed-mode, etc.)
- Methods for contacting respondents, number and type of contacts, etc.
- Maximizing response rate
- Surveys on sensitive topics
- Anonymous vs. confidential surveys
- Data capture
- Useability of computerized interviewing applications (e.g., Web surveys)
- How to's of survey administration
Pretesting and Cognitive Interviewing
- Where/how to pretest
- How much pretesting is necessary?
- Maximizing use of pretesting resources
- Conducting cognitive interviews to assess instrument quality and respondent comprehension
Sampling, Weighting and Estimation
- Determination of needed sample size (power calculations)
- Constructing weights for unequal probability samples that appropriate account for unequal probability sample selection, nonresponse and frame coverage errors
- Determining the appropriate variance estimation and point estimation techniques
- Imputing missing data or otherwise compensating for missing data in statistical analyses