Table 1 Sample of suggested survey design best practices and their application in the sedentary behaviour module development
Survey design best practice concept | Application to sedentary behaviour module |
|---|---|
Identify all of the concepts to be measured | Identified and defined sedentary behaviour and its types and domains. |
Modify writing of the questions to read at a grade 10 level | Adjusted previously validated questions to read at the grade 10 level by using appropriate language. |
Cover only one topic/concept per question | Each question posed focuses on one type/domain of sedentary behaviour (e.g., sedentary time, screen time, travel, reading). |
Break complex or conditional concepts into multiple questions | Recognizing that sedentary behaviour is engaged in differently during the week and weekend, each question includes a ‘per weekday’ and ‘per weekend day’ designation. |
Phrase questions positively and avoid negative structures (i.e. don’t say: How often do you not go out because you cannot afford it? Better phrasing would be: How often do your finances affect your ability to go out?) | No negative structures are used in asking about time spent sedentary. |
Avoid the use of leading or biased questions | There are no references to the sedentary behaviour of others in the questions. |
Create a logical order and flow to ensure the process feels like an interview rather than an interrogation | We begin by asking respondents to estimate total sedentary time, and then ask about specific sedentary behaviours. |
Design sensitive questions to ensure that their relevance is obvious or at least easy to explain | While none of the questions are ‘sensitive’, considerable thought was given to provide examples of pertinent activities under each question. |
Keep the questionnaire as short as possible to ensure high response rates and to avoid partially completed interviews | The module was designed to be modular and as brief as possible. Each question can be asked on its own and questions can be added or dropped in response to the final estimate of available time for the survey. |
Try to use consistent scales to avoid confusing respondents | Response options for all questions were standardized to reflect common formatting within the CHMS i.e. hours and minutes. |