Data Resource Profile: The South Australian Well-being and Engagement Collection (WEC)

Key Features

• To our knowledge, the Well-being and Engagement Collection (WEC) represents the largest population monitoring system of child and adolescent well-being in the world, and it continues to expand each year.

• The WEC was set up to support schools and the South Australian Department for Education to monitor student well-being and engagement over time, but also provides a powerful resource for conducting research into child and adolescent well-being.

• The WEC is conducted as an annual census, with all South Australian schools invited to participate. Almost 100 000 students completed the WEC in 2019, and 321 000 surveys have been collected since 2014, with many students participating over multiple cycles. Within the public education system, the use of a unique education identifier enables linkage over time at an individual child level (providing individual trajectories of well-being), as well as to administrative data on child-level demographics, family and area-level socioeconomic data, school absenteeism, suspensions and expulsions and academic achievement.

• The WEC measures a range of well-being and engagement constructs across four domains: Emotional Well-being, Engagement with School, Learning Readiness, and Health and Well-being out of school.

• Data are owned by the South Australian Department for Education, Australia. Enquires about accessing WEC data can be directed to [[email protected]].

Data resource basics

Mental health and well-being during childhood and adolescence have been shown to impact on health, educational attainment and employment in adulthood.^1–3 Although health and education systems worldwide have long recognized the importance of promoting student well-being,^4–6 population-wide monitoring of well-being remains uncommon. In this data resource profile, we describe an annual rolling population monitoring system that captures self-reported well-being and student engagement among children aged 8–18 years. Data are collected through the education system, and a unique identifier enables linkage over time at an individual child level, providing individual trajectories of well-being as well as linkage to other administrative data.

In 2012, the Telethon Kids Institute and the South Australian Department for Education (SA DfE), recognized the lack of student well-being data as a significant knowledge gap. With partnership funding from the Australian Research Council and the SA DfE, we embarked on the process of adapting and piloting a well-being tool. In conjunction, an online data collection system and data reporting products were developed to disseminate information on the well-being of student cohorts.⁷ In 2014, all primary and secondary schools (n = 721) in South Australia were invited to participate in the Well-being and Engagement Collection (WEC), and 189 schools participated. Since 2014, the WEC has built in scale and popularity, becoming a sustained annual collection with voluntary school participation. In 2019, a total of 95 973 Grade 4 to Grade 12 students from 526 schools completed the WEC, from a potential pool of 182 359 Grade 4 to Grade 12 students enrolled in 715 South Australian schools. Approximately 321 000 WEC surveys have been collected since 2014, with many students participating over multiple collection cycles.

The WEC measures a range of different well-being and engagement constructs across four domains: Emotional Well-being, Engagement with School, Learning Readiness and Health and Well-being out of school, using a combination of multi-item scales and single items (see Figure 1). The WEC survey is based on the Canadian Middle Years Development Instrument,⁸ which was adapted and validated for use in Australia,⁷ with scales and items added/removed from the survey over time to meet the needs of schools and the Department for Education. There are two versions of the survey: one for students in Grades 4 to 9 (approximately 8–14 years old), and a second for students in Grades 10 to 12 (approximately 15 to 18 years old), with the majority of items and scales collected for all students.

Figure 1

Wellbeing and engagement constructs measured in the 2019 WEC.

Table 1 presents information on the school-level participation rates for each of the WEC cycles from 2014 to 2019, disaggregated by school sector. In Australia, there are three types of schools (public, independent and Catholic). In South Australia, about two-thirds of children (65%) attend government/public schools, 18% attend Catholic schools and 17% attend independent schools.⁹ School-level participation rates have increased from 26% in 2014 to 74% in 2019, and have consistently been highest in the government/public school sector (89% in 2019), followed by the Catholic school sector (52% in 2019) and the independent school sector (19% in 2019). Within the public school sector, non-participating schools tended to be smaller than participating schools, and they were more likely to be located in remote areas of the state and to be classified as special schools (i.e. schools catering to students with special needs). Information on the characteristics of non-participating schools in the independent and Catholic school systems is unavailable.

Table 1

School-level participation rates by calendar year and school sector

		Participateda	Eligibleb	Participation rate
2014		189	721	26%
	Government	153	525	29%
	Catholic	26	103	25%
	Independent	10	93	11%
2015		366	719	51%
	Government	337	522	65%
	Catholic	17	103	17%
	Independent	12	94	13%
2016		500	717	70%
	Government	466	518	90%
	Catholic	26	103	25%
	Independent	8	96	8%
2017		498	714	70%
	Government	460	513	90%
	Catholic	20	103	19%
	Independent	18	98	18%
2018		523	716	73%
	Government	475	512	93%
	Catholic	30	102	29%
	Independent	18	102	18%
2019		526	715	74%
	Government	453	511	89%
	Catholic	53	101	52%
	Independent	20	103	19%

a

Information was provided by the SA Department for Education.

b

All schools in the state were considered eligible.

The number of students who have participated in the WEC each year has increased markedly over time (see Table 2). Between 2014 and 2016, Grade 6 to Grade 9 students were eligible to participate and the number of participating students increased more than 2-fold over this period. In 2017, Grade 4 to Grade 5 students were added to the eligibility criteria, and in 2019 the WEC was expanded to Grade 10 to Grade 12 students.

Table 2

Number of students who participated in the WEC by grade and year

	2014	2015	2016	2017	2018	2019
Grade 4	–	–	–	10,174	12 284	13 472
Grade 5	–	–	–	10 436	12 917	12 771
Grade 6	4450	8031	10 976	11 915	12 874	13 123
Grade 7	4453	7994	11 282	11 204	12 407	12 017
Grade 8	4640	7141	10 587	10 549	11 440	11 329
Grade 9	4077	6023	9986	9003	10 744	10 031
Grade 10	–	–	–	–	–	9299
Grade 11	–	–	–	–	–	7760
Grade 12	–	–	–	–	–	6171
Total	17 620	29 189	42 831	63 281	72 666	95 973

Numbers provided by the SA Department for Education. A small number of students who completed the survey but had missing year level have been omitted from this table. WEC, Well-being and Engagement Collection. See Table 3 and Supplementary Table S1, available as Supplementary data at IJE online for information on the number of eligible students and participation rates.

Data collected

Figure 1 provides a list of all constructs measured in the 2019 WEC survey and indicates which constructs are measured for Grades 10–12 students only. Detailed information including: (i) a description of all items and Likert response scales used in the 2019 WEC survey; (ii) psychometric properties of all scales; and (iii) a description of any changes to the items and response scales between 2014 and 2019, are provided in a technical report available on the SA Department for Education website.¹⁰ Scale scores are calculated by taking the average of all items within a scale, and are then categorized to low, medium and high well-being to create categorical variables that are used for school reporting purposes.

All primary and secondary schools in South Australia are invited to participate in the WEC, and schools are free to decide whether all or some of their Grade 4 to Grade 12 classes will complete the survey. Some of the most common reasons for classes not participating were that the class teacher: (i) didn’t know the survey was occurring; (ii) ran out of time during the 4-week window; (iii) believed the survey takes time away from learning; (iv) was concerned about students’ capacity to complete the survey; or (v) had wi-fi/connectivity issues when they tried to complete the survey. For participating classes, parents/guardians receive an information letter before the collection and are able to withdraw their child if they desire. Students are able to opt out before the collection or at any time while completing the survey. Some of the most common reasons for students not participating were: (i) concerns/anxiety about the types of questions; (ii) technical issues with survey; (iii) concerns about responses being kept confidential; and (iv) teachers having concerns about the child’s ability to participate and understand the questions. Students independently complete the survey over one to two class periods at school, with a teacher present in the room for any support if needed. Students’ responses are confidential and results are reported at an aggregate (group) level only. The survey takes approximately 25 to 45 min to complete via an online survey platform. All students have access to an electronic device (tablet or laptop) that can be used to access the platform. Some students bring their own electronic device to school, and computer rooms/additional devices are available for any students who do not have their own device. The survey is completed using a unique token for each student that allows linkage to school administrative records (see details below).

Before the WEC survey, participating schools from all sectors are asked to pre-populate demographic information about their students, including gender, age, Aboriginal or Torres Strait Islander status and postal code of residence. The provision of this demographic information varies markedly between the sectors. For instance, in the 2019 WEC, age was available for 98% government students, 83% of independent students and 40% of Catholic students, and postal code of residence was available for 97% of government students, 75% of independent and 1% of Catholic school students. Student complete questions about their gender, year level and Aboriginal status during the survey, but all other demographic information is missing in the dataset if it is not pre-populated. Government school students complete the WEC survey using their unique education identifier, which allows linkage to their school enrolment records containing additional child-level demographic characteristics such as non-English speaking background, refugee status, disability status and parent/guardian educational attainment and occupation.

Data resource use

The Department for Education introduced the WEC with two key aims. First, to enable schools, the department, other government agencies and the broader community to describe, understand and monitor the well-being and engagement of children and young people. Second, to provide a mechanism for children and young people’s views and experiences to feature in planning and decisions relating to them, with the data serving as a starting point for further engagement and reflection with students. The WEC data are used by schools, groups of schools (partnerships) and the SA Department for Education to monitor and track student well-being and engagement over time.

Annually, participating schools are provided with a school report containing current collection year results along with previous results. Additional reports are generated to assist particular areas within the SA DfE to understand the well-being of population groups of interest, for example students enrolled in flexible learning options, low attendance or disabilities

Student-level participation rates in the WEC can be estimated using the Australian schools census data,⁹ collated by the Australian Bureau of Statistics (ABS), which provides estimates of the number of school students per calendar year, grade level and school sector (see Table 3 and Supplementary Table S1, available as Supplementary data at IJE online). It is important to note that not all schools participate in the WEC, and within participating schools not all classes participate in the survey, so these figures overestimate the number of students who have an opportunity to participate in the survey. Linking the WEC records to the school enrolment census for government students provides an estimate of the student-level participation rate within participating schools, and provides the opportunity to explore the characteristics of students who do and do not participate in the WEC to understand the representativeness of the sample (see Supplementary Figure S1 and Supplementary Table S2, available as Supplementary data at IJE online for sociodemographic characteristics of participating and non-participating government students in the 2019 WEC).

Table 3

Student-level participation rates in WEC by school sector, grade and year

	Grade 4-5	Grade 6-7	Grade 8-9	Grade 10-12	Total
———————————————————————————2014 WEC collection ————————————————————————-
Participation rate %
Government	–	26.5%	27.3%	–	26.9%
Catholic	–	27.6%	22.0%	–	24.6%
Independent	–	10.7%	9.6%	–	10.1%
Total	–	23.7%	22.5%	–	23.1%
———————————————————————————2015 WEC collection————————————————————————-
Participation rate %
Government	–	57.4%	44.2%	–	51.0%
Catholic	–	18.8%	21.4%	–	20.2%
Independent	–	14.8%	20.0%	–	17.6%
Total	–	42.3%	34.4%	–	38.3%
—————————————————————————— 2016 WEC collection ————————————————————————-
Participation rate %
Government	–	82.3%	75.1%	–	78.8%
Catholic	–	24.9%	34.6%	–	30.0%
Independent	–	10.9%	14.5%	–	12.8%
Total	–	58.5%	54.1%	–	56.3%
—————————————————————————— 2017 WEC collection ————————————————————————-
Participation rate %
Government	67.3%	80.7%	71.4%	–	72.9%
Catholic	16.9%	22.7%	16.0%	–	18.4%
Independent	14.8%	21.9%	26.2%	–	21.3%
Total	50.3%	59.9%	51.5%	–	53.6%
—————————————————————————— 2018 WEC collection ————————————————————————-
Participation rate %
Government	80.4%	84.3%	77.1%	–	80.7%
Catholic	23.8%	25.2%	33.7%	–	27.7%
Independent	13.4%	22.4%	23.7%	–	20.2%
Total	60.3%	62.6%	57.6%	–	60.2%
—————————————————————————— 2019 WEC collection ————————————————————————-
Participation rate %
Government	78.3%	75.9%	69.8%	51.5%	67.5%
Catholic	44.0%	43.8%	34.9%	21.3%	34.1%
Independent	18.0%	23.2%	26.3%	18.4%	21.2%
Total	63.0%	60.3%	54.2%	39.0%	52.6%

Number of students who completed the WEC were provided by the SA Department for Education. Number of students enrolled in each grade x sector was taken from the Australian schools census data⁹ A small number of students who completed the survey but had missing year level or school sector information have been omitted from these calculation. Information on the number of students (enrolled and WEC participants) which formed the basis of the participation rate calculations is presented in Supplementary Table S1, available as Supplementary data at IJE online. WEC, Well-being and Engagement Collection.

From a research perspective, individual child level linkage of well-being data over time, as well as WEC data to other administrative datasets, is one of the key strengths of the WEC data. An example includes a recent study¹¹ which explored the association between children’s development in their first year of full-time school, collected through the 2009 Australian Early Development Census (AEDC),¹² and students’ social and emotional well-being 6 years later using the 2015 WEC data. After adjustment for a range of child- and family-level confounders, analyses showed that children who were vulnerable in their physical, social or emotional development at school entry had lower levels of life satisfaction and optimism and higher levels of sadness and worries in Grade 6, than children who were on track in their development. Children who were vulnerable in their early language and/or communication skills had higher levels of sadness and worries in Grade 6, but there were no differences in scores on the positive well-being indicators (life satisfaction and optimism).

Another recent study explored trajectories of depressive symptoms over a 4-year period in a sample of Grade 6 children who participated in the 2014 WEC (n = 3210) and were followed up in 2015 (Grade 7), 2016 (Grade 8) and 2017 (Grade 9).¹³ The aim of the study was to identify sex-specific trajectories of depressive symptoms in adolescents from Grade 6 to Grade 9 and to explore the protective effects of different types of social support (family, peer, teacher and school support). Four trajectory groups were identified, for both boys and for girls, but the shape of these trajectories differed between genders. Adjusted multinomial logistic regression analyses showed that family support, followed by peer belonging, were the strongest protective factors for girls. For boys, family support and peer belonging were also important protective factors but larger effects were observed for peer relationships than family support. This paper provides an example of the research that can be done by linking WEC records over time at an individual child level.

Other studies have used the WEC data to explore cross-sectional relationships between various aspects of student well-being. For example, a study using the 2015 WEC data explored the relationships between bullying, sleep and mental and physical health in a sample of 27 208 students ranging from 9 to 17 years of age.¹⁴ After adjusting for student age, gender and socioeconomic status, results suggested that students who were bullied had two or three times higher odds of going to bed after 10 pm, having fewer good nights’ sleep and reporting poor-fair physical health, and up to three times higher odds of having high levels of anxiety and sadness. Typically, researchers who collect data from school students on their physical and mental health recruit small and non-representative samples, making it difficult to generalize the results to the broader population and to explore effects for small population sub-groups. The large sample sizes in the WEC data are one of the key strengths of this dataset for research.

Future research studies exploring: (i) the association between social and emotional well-being and academic achievement; (ii) links between breakfast skipping, student well-being and school engagement; (iii) the social and emotional well-being of children suspended or expelled from school; (iv) the association between bullying, psychosocial outcomes and academic achievement; and (v) the prevalence of social and emotional well-being by age, gender, disability/chronic disease and socioeconomic position are currently under way. Existing data linkage studies¹⁵ plan to link the WEC data to information on child and maternal health, hospitalizations, child development, child maltreatment etc. This will provide opportunities to explore the relationships between a range of early life predictors and well-being in childhood and adolescence. For governments, education systems and other social service providers, the WEC offers a key outcome measure that can be used in evaluations of different policies and programmes such as physical activity, nutrition, sleep, mental health, anti-bullying and school engagement interventions.

Strengths and weaknesses

One of the key strengths of the WEC data is that government school students complete the survey after entering their unique education identifier, which allows linkage of well-being data over time at an individual child level (providing individual trajectories of well-being), as well as linkage of WEC data to other administrative datasets such as preschool attendance, school readiness, early literacy skills, school absenteeism, suspensions and expulsions, academic achievement and school dropout/completion. The linkage of WEC data to the school enrolment census provides a wealth of additional sociodemographic information on the student, allows exploration of any bias in the sample who complete the WEC and also allows for more fine-grained detail about student-level participation rates.

The ability to link WEC records over time for government students provides opportunities to explore the impacts of government policies and practices on student well-being and engagement, and to evaluate natural experiments. For example, in South Australia, Grade 7 has historically been the final year of primary school, but a new policy is currently being implemented so that Grade 7 will become the first year of high school, to align the SA school systems with other Australian states and territories.¹⁶ The linked WEC data could be used to explore trajectories of well-being and school engagement over the transition from primary school to high school, and to identify students with specific characteristics who are particularly vulnerable to disengaging with school at this time, to help guide the implementation of the new policy. Where programmes and policies are targeted at specific students, intervention and control groups can be defined, WEC data collected before the intervention can provide baseline information and post-intervention data can be used to follow them up. Where natural disasters/health crises affect students within a specific geographical region, WEC data before and after the event can be explored and compared with a matched control group of students living in unaffected communities.

One of the current weaknesses of the WEC data is the low school participation rates in the independent and Catholic school sectors. In 2019, 89% of all government schools participated in the 2019 WEC, compared with 52% of Catholic schools and 19% of independent schools. In Australia, children from more socially advantaged families and communities are more likely to attend independent or Catholic schools, and children living in more disadvantaged communities are most likely to attend public schools.¹⁷ As such, the low participation rates in the independent and Catholic schooling systems mean that the children living in more affluent communities are likely to be under-represented in the WEC data, leading to an underestimate of the socioeconomic inequalities in student well-being and engagement. Furthermore, the amount of missing sociodemographic information for the independent and Catholic students who participate in the WEC makes the data for these students somewhat less valuable from a research perspective (see Data Collected section for details).

A second weakness is the lower student-level participation rates for secondary students (see Table 3). Participation rates in the government and Catholic sectors were highest for primary school students and lowest for students in the senior years of high school (Grades 10–12). However, it is important to note that senior secondary school students have only been eligible to participate in the WEC since 2019. It is likely that the participation rates in the senior years will increase as the WEC becomes normalized as it has in the younger grades over time.

Data resource access

Ethics approval by an accredited ethics research committee is required to access the WEC data for research purposes. Applications to access de-identified WEC data can be made via the South Australian Department for Education website [www.education.sa.gov.au]. Researchers will need to describe the aims or objectives of their research, the methodology they will use, research benefits to children and young people, risk associated with the project, and a detailed description of the required data. Once approved, data will be transferred via an encrypted secure file transfer system, and data, syntax files and output files must be securely stored in line with protocols detailed in the Department for Education Research Application and the ethics application. Any reports, presentations or academic manuscripts using the data that have been accepted for publication need to be e-mailed to the Department for Education 30 days before being published.

Supplementary data

Supplementary data are available at IJE online.

Funding

The SA WEC is funded by the South Australian Department for Education, who manage the online collection portal, provide support to schools during the collection, produce a suite of reports for participating schools, education partnerships (school clusters) and communities, and manage all requests for access to data for research purposes.

Data availability

The WEC unit record file data used in this article is available via an application to the South Australian Department for Education [www.education.sa.gov.au]. Information about school and sector participation rates in the WEC is provided by the SA Department for Education who can be contacted about participation rates in future WEC cycles [[email protected]]. Student-level participation rates were estimated based on data tables (Table 43a) published by the Australian Bureau of Statistics, which are publicly available [www.abs.gov.au/statistics/people/education/schools/2019].

Acknowledgements

The authors would like to thank all of the schools and students who have participated in the Wellbeing and Engagement Collection since 2014. We would also like to acknowledge several colleagues in the SA Department for Education, namely Blair Grace, Alice Hawkes, Stacey Jones, Asad Ali and Marc Conboy and Jessica Strive, for assistance in collating information about the WEC implementation and participation numbers over time.

Ethics approval

Ethics approval was received from the University of Western Australia Human Research Ethics Committee (RA/4/20/6095) for Dr Gregory and Professor Brinkman to access the WEC data for this paper. Other co-authors are government employees and have direct access to WEC data through their positions in the Department for Education.

Conflicts of interest

The SA WEC is funded by the SA Department for Education and all SA schools (government, Catholic and Independent) can participate in the survey and receive school reports at no cost. The SA Department for Education work with other Australian jurisdictions and countries to measure student wellbeing on a fee for service basis.