Page 99 - ON THE WAY TO HEALTHIER SCHOOL CANTEENS - Irma Evenhuis

amount of sent invitations, registrations, posts, reads were counted. Table S6.1 and S6.2 provides the questions.
Measures
School characteristics
Assessed school characteristics were: number of students, education streams at the schools (Vocational/Senior General/Pre-university education), existence of healthy food policy of the school (Yes/No/I do not know), organisation of the canteen (arranged by: external catering organisation/school), presence of vending machines (Yes/No), whether students purchase in the school surroundings such as supermarkets or snack bars (Yes/ No/I do not know), presence of a healthier school canteen team or action plan (No/No but intended/Yes). Information about the encouragement of drinking water (Yes/No) and availability of policy for a healthier school canteen (Yes/No) was retrieved from the Canteen Scan (on online tool to assess the availability/accessibility of food and drink products offered in the canteen, see Table 6.1) completed by the school canteen advisor.
Factors affecting implementation
The implementation plan aimed to change factors which hinder implementation, identified
by interviews with stakeholders. These perceived factors that can affect implementation,
were assessed by stakeholders with questions derived from the TDF [99] and the MIDI
[95]. In accordance with these models, both perceived individual factors, including determinants such as knowledge, self-efficacy, motivation and attitude and perceived environmental factors, including determinants such as need for support, innovation and 6 organisational support, were measured with a five point scale (from 1 = totally disagree, to
5 = totally agree) with 31 and 12 questions respectively. Determinants consisting of more than one item were tested on reliability with Cronbach’s Alpha and analysed separately if lower than p < 0.70 [186]. Table S6.1 provides this information.
Quality of implementation
To evaluate the quality of each implementation tool, different process evaluation concepts were measured quantitatively [101, 122]. Fidelity was measured by dose delivered and dose received. To assess dose delivered, the number of stakeholders provided with the tool by the school canteen advisors or researcher was recorded. Dose received was measured by asking whether stakeholders had received, read, and used the implementation tool. Participant satisfaction with each tool was measured on a 5-point Likert scale (from 1 = totally disagree, 5 = totally agree). Depending on the complexity of the tool, multiple questions were used. Reliability of composite concepts was tested with Cronbach’s Alpha and analysed separately if lower than p < 0.70 [186]. Open-ended questions in the stakeholders’ questionnaire and during an evaluation meeting collected additional information: an explanation of the satisfaction score; a short evaluation per tool; an overall evaluation, positive and negative experiences of the total implementation plan; and suggestions for improvements (Table S6.2). This qualitative data aimed to clarify the quantitative data.
In addition, objective online registered data about the delivery and use of each online tool were collected. For the questionnaires, the number of sent, started and completed questionnaires were registered automatically. For the online community, Facebook
97