Survey Methods & Advice
This is the print-friendly version of the following page:
[Return to original version]
The following advice has been collected from a number of definitive "Survey Development" books (provided below):
Essential steps for designing your survey:
- While designing your survey:
- Who/what are you surveying?
- What is your survey topic?
- What exactly are the research questions that you are trying to answer?
- What is your overall implementation plan?
- How many contacts do you plan to make?
- Will your survey be personalised to respondents?
- Will you be collecting group/point estimate figures or precise/complex ones?
- Which variables will be your primary ones in your research? Which ones will be secondary?
- Are you concerned with sub-group estimates?
- What is your data analysis plan?
- What will your data tables look like?
- If you have multiple goals, what are their priorities?
- While creating your survey:
- Identify your target population - What kind of information do you need?
- Find or use a sampling frame that lists your population.
- Choose a sample design:
- Ensure your design precisely describes how your population members will be chosen.
- Create an estimation plan:
- Computes values of the population parameters from your sample estimates.
- Choose a data collection method.
- Specify detailed data collection procedures to ensure quality in your data.
Back to Top
- Survey Cover Letter - Surveys should, at least, include a single page cover letter that contains:
- A clear and noticeable survey title.
- The dates that your survey is active.
- A short description of your survey and its purpose.
- Why your survey is important.
- A privacy statement.
- An additional ethics compliance statement or disclaimer (if applicable).
- The name and contact details for your project team.
- Writing survey questions - While writing survey questions:
- Use simple words.
- Do not be vague.
- Keep it short.
- Be specific.
- Do not talk down to your respondents.
- Avoid bias.
- Avoid objectional questions.
- Avoid hypothetical questions.
- Consider offering incentives to respondents.
Back to Top
References and Recommended Texts
NOTE: All links provided below will open a new browser tab or window.