Page 64 - ON THE WAY TO HEALTHIER SCHOOL CANTEENS - Irma Evenhuis

Chapter 4. Development of the Canteen Scan
improves the adaptation of the tool by practice. It reveals missing items, interpretation problems and gives insight in how long it takes to fill out the tool [162]. End-users were invited to fill out the online Canteen Scan using an iPad in their canteen. Meanwhile, respondents were asked to think out loud as they filled in the Canteen Scan. This cognitive interview technique ‘think aloud’ was used to understand respondent’s comprehensibility and to reveal areas for improvements [108, 163]. Although this method is time-consuming, subjective, and its validity questionable, in combination with other methods, it can support the development of new tools [108]. In addition, the researcher asked questions if their thoughts were not clear. Thereafter, the usability of the online Canteen Scan was assessed by the concepts comprehensibility, user-friendliness (i.e. easy to understand), feasibility (i.e. practically applicable), time-investment and overall satisfaction [161, 162]. Questions (answered on a 5-point Likert scale, ranging from 1 not feasible at all, to 5 very feasible) were asked to assess comprehensibility, user-friendliness and feasibility, structured within the five Canteen Scan elements; basic conditions (n=2), availability on display and vending machines (n=17), accessibility (n=28), and result and feedback (n=8), together with an overall opinion (n=3). In addition, questions were asked with respect to the investment of time (n=3), e.g. “The amount of time required to fill out the Canteen Scan was worth it” (5-point Likert scale: 1 totally disagree to 5 totally agree); the actual amount of time it took to fill out the Canteen Scan (minutes); and overall satisfaction (n=1) (“In general, how satisfied are you with the Canteen Scan”, 5-point Likert scale: 1. very unsatisfied to 5. very satisfied). Mean scores were calculated and the “think aloud” results were summarised per element of the Canteen Scan by NW, and checked by EV and CR. These results were discussed in the project team to improve the tool.
Table 4.1. Proposition and revisions per step and per construct of the Canteen Scan.
Element
Basic conditions
Availability of food and drinks - method
to classify products
Availability of food and drinks - method to assess quantity
Step 1: Development (expert meeting n=19; interviews n=6; second expert meeting n=10)
- 2 automatically assessed questions
- 2 multiple choice questions
Link the tool with the Dutch Branded Food database1
- Measuring relative shelf space
- Combination of counting facings and product numbers on displays, racks, coolers
Step 2: Content validity of paper scan (n=4)
- Link with the Dutch Branded Food database was evaluated positive
- A product list with common products was suggested,
to reduce time to enter a product
Step 3: Pilot testing the online scan
Translation into online scan
- 2 multiple choice questions were split into two different sets of questions.
- Two separate elements were created for “food and drink on display” and “food and drinks in vending
Pilot testing the online scan (n=4)
- Improve formulation of the questions
- Optimise the database
- Buttons were difficult to find - Added the option to adapt entered composite products.
- Difficult to fill in the Canteen Scan during opening hours as the assortment changes.
62
table continues