This chapter provides the study aims, methods and results of the mediator questionnaire test–retest substudy.
Mediator questionnaire test–retest reliability study overview
The aims of the current study were to test the NAP SACC UK parent and nursery staff questionnaire mediators for (1) acceptability by examining response rates and missing data, (2) maximising the internal consistency of the scales using Cronbach’s α coefficients and (3) reliability by conducting test–retest analyses using weighted kappa coefficients and paired t-tests.
Development of the mediator questions
The mediator questions for parents and nursery staff (see Appendices 2 and 3) were based on the questionnaire items used in the Active for Life Year 5101 study and were adapted using the NAP SACC UK self-assessment form. The self-efficacy, motivation and knowledge questions were split into two sections: children’s nutrition/oral health and children’s PA. All the self-efficacy questions started with the same stem, ‘I feel able to’, and were followed by dietary, PA or oral health-related behaviours where the response options were: 1 – ‘Disagree a lot’, 2 – ‘Disagree a little’, 3 – ‘Not sure’, 4 – ‘Agree a little’ and 5 – ‘Agree a lot’. The same health-related behaviours were included in the motivation questions but used the stem, ‘I am motivated to’. The motivation response options were: 1 – ‘Never’, 2 – ‘Sometimes’, 3 – ‘I don’t know’, 4 – ‘Most of the time’ and 5 – ‘Always’. Multiple-choice questions were set for the knowledge items and varied in terms of having one or multiple correct response options.
Sample and recruitment
Nurseries from Bristol, UK, were recruited through postal invitations followed by an e-mail invitation 10 days later. Participating nursery managers recruited nursery staff and parents via e-mail. Parents were also recruited via an online advert on the survey forum of the UK-based parenting website (www.netmums.com). Inclusion criteria were nursery staff and parents or guardians who worked with or had 2- to 4-year-old children. Written informed consent was obtained from the nursery managers and online consent was gained from each participant prior to data collection commencing.
Data collection
Nursery managers were instructed to send a link to the online nursery staff questionnaire via e-mail to all nursery staff who worked with 2- to 4-year-olds. This was repeated for the parent questionnaire to parents who had 2- to 4-year-old children. Participants were asked to provide their e-mail address at the end of the questionnaire and those who did were automatically sent the questionnaire again 1 week later. They were sent a reminder e-mail a further 3 days later. Participants’ questionnaires were included if the second administration of the questionnaire was completed between 7 and 11 days after the first administration. Each participant was reimbursed with a £10 voucher on completion of the first and second administrations of the questionnaire.
Data analysis
Descriptive statistics were used to summarise the participant characteristics, response rates and missing data. Using the data from the first administration of the questionnaire, Cronbach’s α coefficients were calculated to determine the internal consistency of the four scales: nutrition self-efficacy, PA self-efficacy, nutrition motivation and PA motivation. Values of at least 0.7 were considered acceptable. 102 To assess test–retest reliability of the items, weighted kappa coefficients for ordinal variables103 were calculated. To interpret the kappa coefficient results, the cut-off points detailed by Landis and Koch104 were used: 0.00–0.20 = ‘slight’; 0.21–0.40 = ‘fair’; 0.41–0.60 = ‘moderate’; 0.61–0.80 = ‘substantial’; and 0.81–1.00 = ‘almost perfect’ agreement. For the knowledge questions, the percentage of correct answers was derived for each participant and paired t-tests were carried out on these continuous test and retest summed scores.