This section is based on an article in the Work Group Evaluation Handbook: Evaluating and Supporting Community Initiatives for Health and Development by Stephen B. Fawcett, Adrienne Paine-Andrews, Vincent T. Francisco, Jerry Schultz, Kimber P. Richter, R.K. Lewis, E.L. Williams, K.J. Harris, Jannette Berkley, Jacqueline L. Fisher, and Christine M. Lopez of the Work Group on Health Promotion and Community Development, University of Kansas, Lawrence, Kansas.
-
What are behavioral surveys?
-
Why use behavioral surveys?
-
When should you use behavioral surveys?
-
How do you conduct a behavioral survey?
What are behavioral surveys?
Asking people about their experiences can help your group understand community needs, identify strengths, and decide where to focus its work. Different types of surveys answer different questions. Behavioral surveys focus on the actions people report taking in their everyday lives.
A behavioral survey asks people about specific behaviors related to their physical, emotional, or mental well-being. Topics might include tobacco and nicotine use, sexual health practices, physical activity, eating patterns, or use of preventive services. A survey can explore both behaviors associated with health risks and behaviors that support health.
Constituent surveys of goals, process, and outcomes primarily ask about people's priorities, experiences, and perceptions of an initiative. Behavioral surveys ask what people report doing, often within a defined period. These are self-reports, not direct observations. Recall, question wording, privacy, and concerns about how answers will be judged can influence responses. Recognizing these limitations helps your group design a useful survey and interpret its findings appropriately.
Why use behavioral surveys?
Behavioral surveys can help your initiative understand how common particular behaviors are among respondents and identify priorities for action. When the sample and methods support it, the findings can also provide estimates for a defined population. These results are most useful when considered alongside community members' experiences and information about the conditions that shape their choices, such as access to services, affordable food, transportation, and safe places to be active.
Survey findings can also help people understand an issue and discuss possible responses. For example, a survey might identify cigarette or e-cigarette use among middle school students. Reporting the findings clearly can support conversations about prevention, support for quitting, and the environments young people encounter. Explain who participated, when the survey was conducted, and what the results can reasonably tell you. Numbers alone do not make a finding representative or conclusive.
Behavioral surveys can contribute to evaluation by showing whether reported behaviors change over time. Responses may be affected by difficulty remembering events, differing interpretations of questions, or a desire to give socially acceptable answers. Participation patterns can also affect results if people who respond differ from those who do not. Clear questions, appropriate sampling, and a private setting can help improve data quality, although they cannot eliminate every source of error.
Repeated surveys might show, for example, that the percentage of students reporting cigarette use has declined. That is useful evidence of a trend, but it does not by itself establish that your initiative caused the change. Consider other influences, changes in who participated, and additional evidence about the initiative's reach and implementation. Use the findings to recognize progress and identify questions that need further investigation.
When should you use behavioral surveys?
Plan surveys around the decisions you need to make and the time needed for meaningful change to occur. When possible, collect baseline information before an initiative begins, followed by surveys at appropriate intervals. Depending on your objectives, an annual survey or a longer interval may be more useful than frequent data collection. Consider available resources, participant burden, and existing surveys before setting a schedule. Keep the timing, questions, and sampling methods as consistent as possible so that comparisons over time are meaningful.
How do you conduct a behavioral survey?
Conducting a behavioral survey involves four main steps.
Obtain behavioral data
Begin by defining your objectives. Your initiative already has broader objectives; now identify the specific questions this survey should answer. For example, do you want to estimate how many people report a behavior, understand differences across groups, or track change over time? Involve people from the population you hope to survey in deciding what information is useful and how to ask for it respectfully.
These questions can help you define the survey's objectives:
- What do we need to learn, and which decisions will the findings inform?
- Whose behaviors do we want to understand, and how will we include people whose experiences might otherwise be missed?
- How will the information help us assess progress and improve our initiative?
Next, decide whether to use an existing questionnaire, adapt selected questions, or develop a new survey. When suitable, use established questions that have been tested with a population similar to yours. Behavioral questions often ask about frequency, quantity, or a specific event within a defined period. These response formats are different from agreement scales commonly associated with Likert measures. Use clear wording and response options that do not overlap and that cover the possible answers. Test the survey with a small group of intended participants before distributing it more widely.
The following examples illustrate possible topics and question formats; they are not a validated questionnaire. Select questions that match your objectives and participants' ages, languages, and experiences. Allow people to skip questions, and include a separate "Not sure" or "Prefer not to answer" option where appropriate. Do not treat either response as indicating that a behavior did not occur.
If your focus is substance use, you might ask about tobacco and nicotine products, alcohol, cannabis, or other substances relevant to local priorities. Ask about each substance separately and explain which products are included. For example, distinguish nicotine vaping from cannabis vaping rather than combining both under an undefined question about vaping.
Possible substance use topics include:
- Cigarettes
- Other tobacco and nicotine products, with separate questions for smokeless tobacco, nicotine pouches, and e-cigarettes
- Alcohol
- Cannabis, also called marijuana
- Other substances relevant to the survey's purpose, such as cocaine
A possible question is:
During the past 30 days, on how many days did you use cannabis, also called marijuana?
- 0 days
- 1 to 2 days
- 3 to 9 days
- 10 to 19 days
- 20 to 30 days
If your focus is adolescent sexual health and pregnancy prevention, you might ask about:
- Specific sexual behaviors relevant to the survey's purpose, using clear and age-appropriate definitions
- Use of contraception and access to sexual health information and services, addressed in separate questions
A possible question is:
The most recent time you had penis-in-vagina sex during the past 30 days, did you or your partner use any method to prevent pregnancy?
- Yes
- No
- Not sure
- I did not have penis-in-vagina sex during the past 30 days
- Prefer not to answer
If your focus is tobacco prevention and support for quitting, you might ask about:
- Cigarette use, with separate questions for other tobacco and nicotine products
- Attempts to quit and access to support for quitting
A possible question is:
During the past 30 days, on how many days did you smoke cigarettes?
- 0 days
- 1 to 2 days
- 3 to 5 days
- 6 to 9 days
- 10 to 19 days
- 20 to 29 days
- All 30 days
If your focus is eating patterns related to cardiovascular health, you might ask about consumption of:
- Specific foods and beverages relevant to your objectives, using clearly defined categories
- Fruits and vegetables, asked about separately
- Whole-grain foods
A possible question is:
Yesterday, how many times did you eat fruit? Include fresh, frozen, canned, or dried fruit. Do not count fruit juice. Count each meal or snack when you ate fruit as one time, regardless of the amount.
___0 times
___1 time
___2 or more times
Use these examples to clarify what each question measures. Days of use do not tell you the amount used on each day. Contraception use at the most recent sexual encounter does not establish consistent use throughout the month or use of a particular method. A question about fruit eaten yesterday does not measure portion sizes, usual diet, or overall nutritional adequacy. Add focused questions when your objectives require that information, while keeping the survey manageable.
If you are surveying young people, work with schools and community partners to develop an appropriate recruitment and administration plan. A school-based survey may miss young people who are absent or not enrolled, so describe that limitation and consider additional ways to include them. Before collecting data, confirm the required school approvals, parent or guardian permission procedures, and any applicable ethics review. Explain participation to young people in language they understand and obtain their agreement to participate. Make clear that participation is voluntary and questions may be skipped. Arrange a private setting, explain how responses will be protected, and promise anonymity only if the survey does not collect information that can identify respondents.
For a survey of the broader community, define the population you want to describe and develop a sampling plan suited to that population. Telephone, mail, online, and in-person methods each reach some people more readily than others. An open online link or a survey of people attending one event may provide useful feedback, but it should not automatically be treated as representative of the whole community. Seek help with sampling when population estimates are central to your purpose, and provide accessible formats and language options.
Before collecting new data, ask local school districts, health departments, and community organizations whether they have relevant survey findings. Existing data can reduce duplication and participant burden. Check when and how the data were collected, which population they describe, and whether the questions match your needs. Use public summaries or appropriately protected data, and follow any conditions governing access and use.
For broader context, explore the Centers for Disease Control and Prevention's Youth Risk Behavior Surveillance System (YRBSS) and Behavioral Risk Factor Surveillance System (BRFSS). YRBSS offers information about health-related behaviors among high school students, while BRFSS collects adult information about health behaviors, conditions, and access to care. Their questionnaires and documentation may also help you develop your survey. Check the population, survey year, and methods before comparing these findings with local results.
Tabulate the data
Review responses for missing information and inconsistencies, then calculate counts and percentages for the behaviors you defined. For a simple, unweighted percentage, divide the number of eligible respondents reporting a behavior by the number who provided a usable answer to that question, and multiply by 100. For example, 20 respondents reporting cigarette use out of 100 usable responses equals 20%. Do not count missing, "Not sure," or "Prefer not to answer" responses as "No." Explain how you handled these responses and any questions that did not apply to some participants. If the survey uses sampling weights or a complex design, obtain appropriate analysis support. Describe findings as applying to respondents unless the sampling and analysis support broader estimates.
Plot the data and share findings
Use clearly labeled charts to show patterns across time or groups. For example, you might graph the percentage of high school students reporting cigarette use during the past 30 days across several survey years. Keep the behavior definition, age or grade group, and reporting period consistent, and identify changes in questions or methods that could affect comparisons. Show sample sizes and uncertainty where appropriate. Protect privacy by avoiding reports about very small groups that could identify individuals. Share accessible summaries with participants, your initiative, and the community, and invite their perspectives on what the findings mean.
Use the data
Bring the findings to people who can help interpret them and act on them, including community members, program participants, staff, and local decision-makers. Explain both the findings and their limitations. Discuss which actions the evidence supports, what remains uncertain, and what additional information would help.
Ways to use the findings include:
- Identify needs and barriers to health. Survey results can help your group recognize where additional support or changes in services may be needed. For example, if respondents report less frequent use of contraception, explore the finding with young people and service providers. Consider access, cost, information, personal preferences, and other relevant circumstances before choosing a response. Avoid assuming that a single behavior measure explains pregnancy trends or that individual behavior is the only factor your initiative should address.
- Build public understanding of an issue. Explain results in plain language, with enough context to prevent misleading conclusions. If surveys show an increase in reported tobacco or nicotine use among students, specify the products, time period, and population measured. Use the findings to support constructive conversations about prevention and access to help, without labeling young people or particular communities as the problem.
- Assess progress and improve the initiative. Compare findings with your objectives and with other evidence, such as participation records, implementation information, and community feedback. Look for both intended and unexpected changes. Survey trends can help you decide what to strengthen or revise, but attributing those changes to the initiative requires an evaluation design and evidence that address other possible explanations.
Behavioral surveys are tools for learning and improvement. Agree on actions based on the findings, identify who will follow through, and tell participants how their input informed decisions. A useful survey does more than produce numbers: it helps your group respond to community priorities and assess whether its work is moving in the intended direction.