Globally, an estimated 38·3 million (5·6 %) children under 5 years of age were overweight in 2017( 1 ). Guidance and support for caregivers and childcare settings to provide healthy diets and physical activity opportunities have been identified as strategies to reduce the prevalence of obesity in children of pre-school age( 2 ). Parents of pre-school children can make certain foods available and accessible in the home environment to promote positive food behaviours( Reference Haire-Joshu and Nanney 3 – Reference Baranowski, Cullen and Baranowski 5 ) and parental encouragement and beliefs about physical activity are important predictors of children’s physical activity levels( Reference Trost, Sallis and Pate 6 , Reference Dempsey, Kimiecik and Horn 7 ). Various studies have reported that childcare policies have influenced children’s dietary intake and that pre-schools have a responsibility to assist parents in providing healthy food to children( Reference Mazarello Paes, Ong and Lakshman 8 ). Childcare staff can also influence the level of physical activity children engage in by encouraging them to be active( Reference Henderson, Grode and O’Connell 9 ). Early childhood caries is a global pandemic and the prevalence among children aged 3–5 years varies between different countries and continents( 10 ). Parents and pre-school staff need to supervise and be trained in tooth-brushing practices, together with reducing children’s consumption of sugary foods and drinks, to prevent the onset of early childhood caries( 10 ). Parental and family dental health habits influence their children’s oral health( Reference Castilho, Mialhe and Barbosa Tde 11 ). In the UK about 71 % of eligible 2-year-olds and 95 % of 3–4-year-olds receive government-funded early education in the UK( Reference National Statistics 12 ).
As parents’ and nursery (pre-school) staff’s encouragement have been associated with the quality of children’s diet, oral health and level of physical activity, interventions attempt to increase caregivers’ self-efficacy, motivation and knowledge to improve these behaviours( Reference Castilho, Mialhe and Barbosa Tde 11 , Reference Wright, Adams and Laforge 13 ). Self-efficacy, a strong predictor of health behaviour change( Reference Strecher, DeVellis and Becker 14 ), is defined as confidence in one’s ability to perform the target behaviour and is a construct of Bandura’s Social Cognitive Theory( Reference Wright, Adams and Laforge 13 , Reference Bandura 15 ). Motivation refers to one’s readiness to change a specific behaviour, which is defined as the degree to which a person feels a change is important( Reference Gunnarsdottir, Njardvik and Olafsdottir 16 , Reference Miller and Rollnick 17 ). Parental and nursery staff’s knowledge of healthy diets and physical activity may also help encourage children to engage in healthy eating and physical activity( Reference Mabiala Babela, Nika and Nkounkou Milandou 18 ). We are not aware of parent and/or nursery staff questionnaires which measure a combination of attitudes and knowledge towards pre-school children’s nutrition and physical activity. Thus the aims of the current study were to test the Nutrition and Physical Activity Self-Assessment for Child Care (NAP SACC) UK mediators for: (i) acceptability, by examining response rates and missing data; (ii) maximising the internal consistency of the scales using Cronbach’s α coefficients; and (iii) assessing the levels of test–retest reliability of individual items and scales using weighted κ coefficients, intraclass correlation coefficients (ICC) and paired t tests.
Methods
Sample
Nurseries from Bristol, UK were identified using the www.1bigdatabase.org.uk and recruited through postal invitations followed by an email invitation 10d later. Participating nursery managers (n 21) recruited nursery staff and parents via email. Parents were also recruited via an online advert on the survey forum of the UK-based parenting website www.netmums.com. Data were collected between November 2016 and January 2017. Inclusion criteria were nursery staff and parents or guardians who work with or have 2–4-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.
Study design
Nursery managers were instructed to send a link to the online nursery staff questionnaire via email to all nursery staff who worked with 2–4-year-olds. This was repeated for the parent questionnaire to parents who had 2–4-year-old children. Participants were asked to provide their email address at the end of the questionnaire; those who did were automatically sent the questionnaire again a week later. They were sent a reminder email a further 3d later. Participants’ questionnaires were included in the analyses if the second administration was completed between 7 and 11d after the first administration. Each participant was reimbursed with a £10 voucher on completion of the first and second administrations of the questionnaire.
Development of the mediator questions
The NAP SACC intervention was designed in the USA to improve the nutrition and physical activity environment, policies and practices in nursery settings( Reference Ward, Benjamin and Ammerman 19 ). The aim of the NAP SACC UK feasibility cluster-randomised trial was to assess the acceptability of the intervention, randomisation and study measures within the UK( Reference Kipping, Jago and Metcalfe 20 ). A set of potential mediator questions was created for the NAP SACC UK study to measure parents’ and nursery staff’s knowledge, motivation and self-efficacy towards children’s physical activity, oral health, nutrition and sedentary behaviours( Reference Kipping, Jago and Metcalfe 20 ). The mediator questions (see online supplementary material) were based on the questionnaire items used in the Active for Life Year 5( Reference Lawlor, Howe and Anderson 21 ) study and were adapted using the best practice of diet as recommended by the Children’s Food Trust( 22 ) and UK physical activity guidelines( 23 ). The self-efficacy, motivation and knowledge items were split into two sections: children’s nutrition/oral health and children’s physical activity. All the self-efficacy items started with the same stem, ‘I feel able to’, and were followed by dietary-, physical activity- 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 items 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.
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, Physical Activity Self-Efficacy, Nutrition Motivation and Physical Activity Motivation. Values of at least 0·7 were considered acceptable( Reference Bland and Altman 24 ). To assess test–retest reliability of the individual items, weighted κ coefficients for ordinal variables( Reference Cohen 25 ) were calculated. To interpret the κ coefficient results, the cut-offs detailed by Landis and Koch( Reference Landis and Koch 26 ) 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. A score was derived by calculating the total for each of the self-efficacy and motivation scales. For the knowledge items, the percentage of correct answers was derived for each participant. ICC were used to assess the test–retest agreement at scale level for each of the five scales, with an ICC>0·7 considered acceptable( Reference Nunnally 27 ). The sample size required for estimating an ICC of 0·8 with a 95 % CI ± 0·1 for two repeated measures was fifty participants( Reference de Vet, Terwee and Mokkink 28 ). Paired t tests were calculated on the continuous test and retest total self-efficacy, motivation and knowledge scale scores to determine whether the scores were higher at the test or retest administration. All analyses were carried out in the statistical software package Stata version 15 (2017).
Results
Participants
Eighty-two parents and sixty-nine nursery staff completed the first and second questionnaire administrations within 7–11d and were included in the analyses. Participants’ demographic characteristics are shown in Table 1. Most parents (43·9 %) were in the age group 31–35 years, whereas nursery staff were mainly in the 25–30 years age category (31·9 %). The majority of parents (41·5 %) and nursery staff (37·7 %) had a university degree. The Index of Multiple Deprivation scores of the twenty-one recruited nurseries ranged from 3·59 to 53·27.
Baseline characteristics of parents and nursery staff who completed two administrations of their respective questionnaires within an interval of 7–11d, Bristol, UK, November 2016–January 2017
GCSE, General Certificate of Secondary Education; GNVQ, General National Vocational Qualification.
Acceptability and missing data
The number of times that individuals clicked the consent button on the questionnaire link was 130 and 103 for parents and nursery staff, respectively; it was not possible to distinguish whether the same individuals clicked consent multiple times as they would not have provided any identifying information at this stage (email addresses). One hundred and two parents completed the first administration of the questionnaire and eighty-eight (86·3 %) completed it for the second administration. For the nursery staff questionnaire, eighty-six and seventy-four (86·0 %) participants completed the first and second administrations, respectively.
Seventy-three (89·0 %) and sixty-nine (84·1 %) of the parents completed all items in the first and second questionnaire administrations, respectively. The number of nursery staff completing all the items showed an increase from the first (n 57, 82·6 %) to the second (n 59, 85·5 %) administration. Thirty-eight (71·7 %) and thirty-four (64·2 %) of the fifty-three parental items had no missing data at test and retest administrations, respectively. Fifty-two (80·0 %) of sixty-five nursery staff questionnaire items had no missing data at both test and retest administrations.
Cronbach’s α coefficients
Tables 2 and 3 show the Cronbach α coefficients of each item for the test scale if the item is removed, as well as the α of the overall scale. The Nutrition Self-Efficacy scale showed an acceptable level of internal consistency (α=0·80) and the Physical Activity Self-Efficacy scale had the weakest internal consistency in the parent questionnaire but still at an acceptable level (α=0·73). The removal of item 17 relating to the provision of opportunities to walk to/from nursery would noticeably improve the internal consistency of the scale (α=0·81). The Nutrition Motivation scale showed a high level of internal consistency (α=0·86) and the Physical Activity Motivation scale demonstrated the highest overall Cronbach’s α (0·89). Unlike the equivalent item in the Physical Activity Self-Efficacy scale, the removal of item 37 had less of an increase on the internal consistency (α=0·92). The Nutrition Self-Efficacy and the Nutrition Motivation scales in the nursery staff questionnaire both had α coefficients of 0·89, which showed high levels of internal consistency. Both the Physical Activity Self-Efficacy and the Physical Activity Motivation scales also demonstrated high levels of internal consistency (α=0·91).
Cronbach’s α coefficients for the four scales in their questionnaire among parents who completed two administrations within an interval of 7–11d, Bristol, UK, November 2016–January 2017
Cronbach’s α coefficients for the four scales in their questionnaire among nursery staff who completed two administrations within an interval of 7–11d, Bristol, UK, November 2016–January 2017
Test–retest analyses
Test–retest analyses found that most of the weighted κ coefficients for individual items fell under the ‘moderate’ category for the parent (75·0 %) questionnaire and for the nursery staff (55·8 %) questionnaire (Table 4). The parent questionnaire scales demonstrated substantial levels of agreement (ICC=0·62 to 0·80). Overall the nursery staff questionnaire scales demonstrated good levels of test–retest reliability, apart from the Physical Activity Motivation (ICC=0·48) scale which can be in part explained by 50 % of the individual items displaying ‘fair’ test–retest reliability. Paired t tests found that self-efficacy, motivation and knowledge scale scores for parents were higher in the questionnaire’s second administration. Paired t tests showed strong evidence that the Nutrition Motivation (t=−2·91, df=81, P=0·00) and Knowledge (t=−3·22, df=81, P=0·00) scales were substantially higher at the retest administration. Similarly, the nursery staff’s scale scores were all higher in the questionnaire’s retest administration; however, there was no evidence that this increase was substantial.
Weighted κ coefficients of the items, intraclass correlation coefficients (ICC) and paired t tests of the test scales among parents and nursery staff who completed two administrations of their respective questionnaires within an interval of 7–11d, Bristol, UK, November 2016–January 2017
N/A, not applicable.
Discussion
In the present paper we found that our parental and nursery staff questionnaires on nutrition-, oral health- and physical activity-related self-efficacy, motivation and knowledge for pre-school children demonstrated high levels of acceptability, with most participants completing the second administration of the questionnaire. Eighty-four per cent of the parents and 86 % of the nursery staff participants completed all the items. When analysing the missing data further, no items were consistently unanswered by multiple participants or between the test and retest administrations of the questionnaires; this indicates that the items were seen to be appropriate.
The self-efficacy and motivation scales demonstrated acceptable and high levels of internal consistency. Removing the item on providing weekly opportunities to walk to/from nursery from the parent questionnaire would improve the internal consistency of the two physical activity scales. Our findings suggest that this item does not fit well within the Physical Activity Self-Efficacy and Physical Activity Motivation scales and could therefore affect the scores produced for these two scales. We advise removing this item from these scales or to include it as a separate item in the questionnaire.
The individual self-efficacy and motivation items demonstrated good levels of test–retest reliability, where over 50 % of the κ coefficients were categorised as ‘moderate’ for the parent and nursery staff questionnaires. A handful of items were found to have ‘fair’ and ‘slight’ agreement, which might suggest that participants do not understand the questions or are guessing the answers( Reference Vereecken, De Pauw and Van Cauwenbergh 29 ). Total scores for the self-efficacy, motivation and knowledge scales were derived for each participant and test–retest analyses were carried out using paired t tests. Among the parent population, there was a substantial difference between the test and retest responses for two of the scales. In terms of the Knowledge scale, no substantial test–retest difference was observed when the exact same items were answered by the nursery staff. Differences in the results between the parents and nursery staff may be the result of differences in participant age and education levels but this is unclear due to our limited sample size.
The test–retest correlations of the self-efficacy, motivation and knowledge scales ranged from 0·48 to 0·82 across both the parental and nursery staff questionnaires. Our findings are comparable with findings from the literature looking at similar topic areas and/or populations. In a study by Wright et al.( Reference Wright, Adams and Laforge 13 ), the 1-week test–retest reliability of parental self-efficacy scales relating to children’s physical activity and dietary behaviours ranged from 0·80 to 0·88. Cronbach’s α coefficients for the four scales ranged from 0·80 to 0·88 in two different participant samples. In a study by Whittaker and Cowley( Reference Whittaker and Cowley 30 ), the ICC of three parenting self-efficacy scales relating to children aged 1–4 years, including a play scale, ranged from 0·77 to 0·95 and the internal consistency ranged from 0·66 to 0·84. The Cronbach’s α coefficients and test–retest reliability of a seven-item effort motivation scale was 0·92 and 0·61 for teachers and 0·89 and 0·69 for parents of pre-school children( Reference Leung and Lo 31 ). Nutrition knowledge scales demonstrated test–retest reliability coefficients between 0·33 and 0·75 in a study by Vereecken et al.( Reference Vereecken, De Pauw and Van Cauwenbergh 29 ). The Cronbach’s α coefficients for four oral health-related knowledge, fatalism and self-efficacy measures ranged from 0·76 to 0·91 when measured in mothers of children aged 1–5 years( Reference Finlayson, Siefert and Ismail 32 ).
There are no currently existing questionnaires which measure parents’ and nursery staff’s self-efficacy, motivation and knowledge towards pre-school children’s nutrition, oral health and physical activity. Our analyses have demonstrated that the items and scales in our questionnaires are acceptable, internally consistent and reliable. A limitation in our paper and other similar studies is that the analyses were carried out in a single sample, therefore we cannot assume that our results would be reproduced when repeated using different populations. It is important to acknowledge that we were limited with our sample size and characteristics, which are not representative of the general population, and therefore it is uncertain whether these items would be deemed as acceptable to more diverse populations. In the UK, Level 6 qualifications for early years staff are degree level and include Qualified Teacher Status (QTS), Early Years Professional Status (EYPS), Early Years Teacher Status (EYTS) and other early years-related degree-level qualifications( 33 ). In England in 2016, 29 % of nursery staff had a minimum of a Level 6 qualification( 33 ) compared with our nursery staff sample where 46·4 % of individuals had a university degree or higher (minimum Level 6 qualification). Although the percentage of our nursery staff sample with a university degree was higher than the English average, we believe that this would be a problem only if internal consistency and test–retest reliability would be different in a group who had a lower level of educational achievement. However, we acknowledge that our nursery staff questionnaire results may not be generalisable to early years staff in other countries which have different requirements for early years staff qualifications. We recognise that our results may not be replicated if using paper-based or face-to-face versions of the questionnaires as opposed to the online versions used in the present study. This is important to consider in low- to middle-income countries where device and Internet access may not be available to administer tablet/web-based forms of the questionnaire. However, there is evidence to suggest that acceptability, internal consistency and test–retest reliability outcomes are comparable between paper-based and device/web-based forms of questionnaire administration( Reference Hohwu, Lyshol and Gissler 34 – Reference Kleinman, Leidy and Crawley 36 ). Due to the limitations stated above, caution needs to be taken when interpreting the magnitude of the results and deciding whether to remove certain items for use in studies.
Conclusions
The scales provided here are an acceptable and reliable method of assessing parents’ and nursery staff’s self-efficacy, motivation and knowledge about pre-school children’s diet, oral health and physical activity. The items in the questionnaire show low levels of missing data and good levels of acceptability, internal consistency and test–retest reliability. Overall our findings suggest that the questionnaires would be suitable measures in assessing parent and nursery staff levels of self-efficacy, motivation and knowledge.
Acknowledgements
Acknowledgements: The authors would like to thank all the nurseries, nursery staff and parents who took part in the study. They also thank Rowan Brockman for developing the mediator questionnaires as part of the NAP SACC UK feasibility study, Alison Horne for creating the online questionnaire on REDCap and Rhiannon Macefield for her advice on the study. The views expressed in this paper are those of the authors and not necessarily those of anyone in this acknowledgement list. Financial support: This work was supported in part by the GW4 BIOMED DTP (K.D., grant number MR/N0137941/1), awarded to the Universities of Bath, Bristol, Cardiff and Exeter from the Medical Research Council (MRC)/UKRI. This work was supported by the National Institute of Health Research (NIHR) Public Health Research Programme (PHR – 12/153/39). This work was supported by the Centre for the Development and Evaluation of Complex Interventions for Public Health Improvement (DECIPHer), a UKCRC Public Health Research Centre of Excellence: joint funding (R.K., J.W., grant number MR/KO232331/1) from the British Heart Foundation, Cancer Research UK, Economic and Social Research Council, Medical Research Council, the Welsh Government and the Wellcome Trust, under the auspices of the UK Clinical Research Collaboration (UKCRC), is gratefully acknowledged. This study was undertaken in collaboration with the Bristol Randomised Trials Collaboration (BRTC), a UKCRC Registered Clinical Trials Unit in receipt of NIHR CTU support funding. None of the funders are involved in the Trial Steering Committee, the data analysis, data interpretation, data collection or writing of the paper. The views expressed in this publication are those of the authors and not necessarily any of the funding bodies listed here. All funders had no role in the design, analysis or writing of this article. Conflict of interest: None. Authorship: J.W., C.M., R.K., A.P. and R.J. formulated the research questions, designed the study and advised on the statistical analyses. K.D. designed and carried out the study; advised on and carried out the statistical analyses; and wrote the first draft of the manuscript. All authors read and approved the final manuscript. Ethics of human subject participation: This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the Faculty of Health Sciences Research Ethics Committee at the University of Bristol (Ref: 41585). Written (online) informed consent was obtained from all participants.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1368980018004111
Open access
