Faunalytics worked with Animal Charity Evaluators to compile a set of best practices and advice for conducting surveys and other research. The recommendations relate to research on all animal issues, but there is a particular focus on the complexities of dietary research.
There are many aspects of survey design that can affect the validity and reliability of results. Ideally, organizations conducting studies and surveys should work with experts throughout the design and analysis processes to ensure the most accurate results possible. Faunalytics and Statistics Without Borders are two organizations that can help animal-focused individuals and groups get connected with experts in study design and statistical analysis.
If working with an expert is not a possibility, individuals and organizations can still improve their results by conducting their surveys thoughtfully. Here, we share some general areas of concern, and some of what we have learned about addressing each one. As some of these concepts may be confusing to those new to running studies, we also encourage you to reach out if you have any questions about implementation.
When creating a survey it is extremely important to choose questions wisely. To that end, Faunalytics and ACE have worked with other animal advocacy groups to compose a list of relevant questions. We have grouped the questions into categories, selected the most robust question formats, and spent considerable time altering question structures to be most useful. We hope that by sharing this question bank, we can help individuals and organizations design surveys with more confidence and less effort.
If you do use these questions on your own surveys, please let us know if our work has been helpful to you.
There are many different ways to measure diet. We cover a few of the self-report options—food frequency questionnaires and 24-hour recalls—and discuss which are more suited to intervention studies versus cross-sectional ones.
Social desirability bias is a tendency for responses to reflect what might be desired instead of truth, and it can affect even carefully designed studies. Well-designed studies take steps to lower the probability that this will occur, but sometimes it is not possible to eliminate all possibilities of such bias. To help deal with this problem, we provide a scale to measure the extent to which your survey questions are influenced by social desirability bias.
The term sample size refers to the number of participants in a study, or other units of observation. Sample size is important because it affects how confident we can be that the results of a study tell us something about reality. There are a number of free online tools you can use to calculate sample size.
The documents for the Survey Guidelines project were developed primarily by Jon Bockman, Allison Smith, Greg Boese, and Toni Adleberg of Animal Charity Evaluators and Kathryn Asher of Faunalytics. We’d like to thank Nick Cooney, Chris Monteiro, Che Green, Harish Sethu, Alex Felsinger, Michael Webermann, and the volunteers at Statistics Without Borders for providing feedback on draft versions of this project.