Tier II, Step 4 – Developing your instrument(s)

Once you have decided which assessment method(s) is appropriate (e.g., surveys or interviews) and defined your sample population, begin developing the appropriate instrument.

For a survey, develop a list of quantitative and qualitative questions that you are interested in asking. Below are some helpful tips to developing survey questions from this CDC Constructing Survey Questions guide. Click the link for additional tips on developing survey questions.

• Include background questions (age, gender identity, if they have been impacted by a specific disaster, etc.) as appropriate to help ensure that your respondents represent your target population.
• Closed-ended questions, such as rating scales, are popular ways of collecting information if you want to measure a person’s opinions, knowledge, or feelings. Closed-ended questions provide a list of choices from which the respondent can choose.
• Open-ended questions do not provide choices from which to select an answer. Instead respondents write an answer in their own words. Open-ended questions are a good way to ask respondents to explain their opinions and/or why they selected a particular response in a previous rating question.
• In each question, only ask about one topic or issue at a time.
• Do not use language that may be leading the respondent to a specific answer.

For interviews and focus groups, develop a protocol that includes the main questions that you are interested in knowing and some potential follow up questions to help “probe” participants if they are unsure how to answer. Below are some helpful tips from Harvard University on developing interview questions.

• Ask “how” questions rather than “why” questions to get stories of process rather than acceptable “accounts” of behavior. An example is, “How did the disaster impact your access to medical services?”
• Develop probes that will elicit more detailed and elaborate responses to key questions.
• Begin the interview with a “warm-up” question—something that the respondent can answer easily and at some length (though not too long). It doesn’t have to pertain directly to what you are trying to find out (although it might), but this initial rapport-building will put you more at ease with one another and thus will make the rest of the interview flow more smoothly.
• Think about the logical flow of the interview. What topics should come first? What follows more or less “naturally”? This may take some adjustment after several interviews.
• Difficult or potentially embarrassing questions should be asked toward the end of the interview when rapport has been established.
• The last question should provide some closure for the interview, and leave the respondent feeling empowered, listened to, or otherwise glad that they talked to you.

Keep in mind that surveys with closed-ended questions can be asked in an interview as well.

Helpful tip 1: If you do not need to know about a topic (e.g., sexually transmitted infections), do not include questions on that topic in your instrument. Similarly, if you are asking about the types of resources that your population may need, first check to ensure that you either have or can easily obtain those resources. For example, do not ask include a question asking if a participant needs a tent if you do not have a resource to help them get one.

Helpful tip 2: Ensure that your questions are at an appropriate reading level for your audience. Many word processing software programs, such as Microsoft Word, have a function to check the reading level of content. For a general U.S. population, a best practice is to use language between 6^th-8^th grade reading level.

There are three versions of pilot-tested questionnaires in this toolkit. Pilot testing means that the questions have been previously used in an assessment to determine if they work well in collecting the intended data. The 3-minute version is for a quick snapshot of current reproductive health needs in communities, the 5-minute version provides more comprehensive questions, and the 10-minute version provides the most in-depth assessment. You should select whichever questionnaire best meets of your assessment plan and the amount of time you believe you will have with each person.

The following table summarizes which topics and how many questions are included in each version. Keep in mind that reproductive age women will only answer Sections 1 and 4 if they are not pregnant or post-partum.

Note: Assessment times do not account for longer story sharing or provision of resources to participants. Please allow extra time for the “human” element of data collection.

Additional questions 

If you are interested in adding in some additional questions on access to WIC services, partner violence, access to sexually transmitted infections services, and access to mental health services, see Assessment Tools.

If the RHAD assessments do not fully capture your post-disaster assessment objectives, check out these resources for other similar questions that you can add to or utilize to tailor a jurisdiction-specific assessment.

• CDC Health Indicators for Disaster-Affected Pregnant Women, Postpartum Women, and Infants
• CDC Community Assessment for Public Health Emergency Response (CASPER) Toolkit
• CDC Reproductive Health Assessment Questionnaire for Conflict-Affected Women
• CDC Pregnancy Risk Assessment Monitoring System Questionnaires