Free School Meals Test of Change: Evaluation of expanded eligibility for S1-S3 pupils in receipt of Scottish Child Payment
An evaluation of the Free School Meals Test of Change, which expanded eligibility for free school meals to S1-S3 pupils in families receiving Scottish Child Payment in selected secondary schools
2. Methodology
Overview of approach
This evaluation used a mixed‑methods approach, combining qualitative interviews and observations, parent and pupil surveys, and monitoring data to understand how the Test of Change was implemented, experienced and perceived across participating local authorities.
Scoping phase
As part of a scoping phase from September to December 2025, the evaluation team reviewed background documents and conducted scoping interviews to strengthen their understanding of the Test of Change. Scoping interviews were conducted online and included the Test of Change lead in each participating local authority, as well as a catering lead and a senior teacher from each local authority. A summary of how the Test of Change was implemented in each LA is included in Appendix A.
Based on this, evaluators drafted a preliminary logic model (see Appendix B) for the Test of Change which was refined in a workshop with representatives from the Scottish Government and other key stakeholders. While the logic model was primarily used to structure the evaluation, it also reflects some of the intended outcomes of the wider programme. An evaluation framework was then created (see Appendix C). Both outputs were used to guide subsequent data collection as well as structure the analysis and reporting of findings.
Qualitative fieldwork
Main stage fieldwork took place from March to May 2026, with one final interview in Moray taking place in June. This consisted of in-person visits to 14 schools, where evaluators conducted depth interviews with head teachers or senior teachers involved in the Test of Change, small group discussions with catering staff, and depth interviews with newly eligible pupils. Throughout this report, ‘newly eligible pupils’ refers to pupils who became eligible for free school meals through receipt of Scottish Child Payment as part of the Test of Change. The term eligible pupils in the context of free school meals usually means pupils who are eligible as a result of the criteria set out within the Education (Scotland) Act 1980 (see Free School Meals eligibility). An observation exercise during lunchtime at each school was also carried out. Visits were conducted over the course of 1-2 days and interviews largely took place during school hours. Where participants were unable to take part during the visit, virtual follow up interviews were also offered.
As well as the school visits, virtual depth interviews were conducted with parents of newly eligible pupils and with the local authority leads for the Test of Change (these interviews with leads took place towards the end of the fieldwork period and focused on any changes since the scoping phase).
Sampling and recruitment
The evaluation team identified two schools per local authority to include as case studies in the evaluation. This took into account considerations such as including a mix of different local contexts, such as rurality, deprivation (based on the Scottish Index of Multiple Deprivation), and the number of local shops or food outlets nearby. However, choice was limited in some local authorities due to small numbers of schools participating or small numbers of pupils newly eligible as part of the Test of Change. The case study schools are intended to illustrate the range of experiences rather than be representative of all participating schools
In order to recruit participants, schools were provided with information sheets and privacy notices by the evaluation team. Tailored versions were created for staff, parents and pupils. Key contacts at each school shared these with the relevant audiences directly. Note that using schools as gatekeepers may have influenced participation or introduced self-selection bias. Parents were provided with a consent form to provide permission for pupils to take part (the pupils’ consent was also required), as well as choosing whether to share their own contact information in order to opt in to a parent interview. Key contacts arranged times for interviews to take place during the school day with staff and pupils. For parent interviews, an incentive consisting of a £35 Love2Shop voucher was shared.
It is worth noting that there were particular sampling and recruitment challenges in Shetland, due to especially low numbers of newly eligible pupils. (Note that this was, in part, due to recent Shetland-specific expansions to free school meal eligibility.) To mitigate this, all newly eligible pupils across multiple schools were invited to take part and virtual interviews were offered to provide more flexibility and a longer fieldwork period compared to an in-person visit. However, no pupils or parents came forward. Therefore, Shetland is not included in the sample for the main stage qualitative fieldwork.
By the later pupil interviews, few new themes were emerging within the achieved sample. This was less clearly the case among parent interviews.
|
Local authority |
Parent Interviews |
Pupil Interviews |
|---|---|---|
|
Total |
21 |
72 |
|
Aberdeen city |
3 |
8 |
|
Fife |
4 |
11 |
|
Glasgow |
4 |
11 |
|
Moray |
0 |
10 |
|
North Ayrshire |
5 |
11 |
|
Shetland |
0 |
0 |
|
South Lanarkshire |
4 |
11 |
|
Comhairle nan Eilean Siar |
1 |
10 |
|
Local authority |
LA leads |
Senior teachers |
Catering staff |
|---|---|---|---|
|
Total |
8 |
13 |
27 |
|
Aberdeen |
1 |
2 |
4 |
|
Fife |
1 |
1 |
4 |
|
Glasgow |
1 |
2 |
5 |
|
Moray |
1 |
2 |
5 |
|
North Ayrshire |
1 |
2 |
6 |
|
South Lanarkshire |
1 |
2 |
4 |
|
Shetland |
1 |
0* |
0* |
|
Comhairle nan Eilean Siar |
1 |
2 |
2 |
*Note that Shetland headteacher and catering staff fed into the research as part of the scoping phase, and indicated that little had changed throughout the Test of Change.
Design of research materials
All interviews were conducted using a semi-structured discussion guide, designed in close collaboration between the evaluation team and the Scottish Government. All discussion guides were based on the logic model and evaluation framework, but different versions of the discussion guide were developed for use with different audiences (for example for local authority leads, teaching staff, catering staff, pupils and parents). An observation guide was also developed to support evaluators to consistently and systematically observe lunchtime processes and settings across schools. Discussion guides can be found in Appendices D to H.
Parent and pupil surveys
A survey was conducted with 2,717 parents and 881 pupils. Fieldwork took place between 4 and 28 May 2026.
The parent survey was shared with all parents of pupils in S1, S2 or S3 at all schools participating in the Test of Change. A link to the survey was shared with parents via email by local authorities or schools. Survey questions covered awareness and experiences of the Test of Change, as well as perceived impacts since the Test of Change began.
The pupil survey was shared with all S1, S2 and S3 pupils at case study schools taking part in the evaluation only. Note that since pupils were only surveyed in case study schools, findings should not be interpreted as representative of all schools participating in the ToC and should be considered as indicative only. This approach was taken due to the additional support and buy-in required from schools to distribute the pupil survey. Where possible, schools administered the survey in class at a suitable time within the fieldwork period. In cases where this was deemed unfeasible, the survey link was emailed out to pupils (however this approach resulted in a very low response rate). There were some schools who were unable to facilitate sharing the pupil survey at all. This included Anderson High School, Sir E. Scott School Secondary, and Largs Academy. Survey questions covered lunchtime behaviours, views on school meals generally, awareness of and attitudes towards free school meals, and perceived impacts of getting a school meal.
The evaluation team did not receive information from all local authorities and schools on the number of people invited to take part in the surveys. Among the four local authorities for which invitation numbers were available, the estimated parent survey response rate was around 15%. Among the five schools for which equivalent information was available, the estimated pupil survey response rate was around 9%, although this varied considerably between schools. These figures should not be interpreted as overall response rates for the full survey samples
Table 2.3 below shows a breakdown of the sample for both the parent and pupil survey. This includes the percentage each group represented of the total achieved sample. This highlights some areas of skew. Notably, certain LAs are over-represented such as Fife for the parent survey and Comhairle nan Eilean Siar and South Lanarkshire for the pupil survey.
|
Subgroup |
Number of parents (% of sample achieved) |
Number of pupils (% of sample achieved) |
|---|---|---|
|
Total |
2,717 |
881 |
|
Man/boy |
311 (11%) |
428 (49%) |
|
Woman/girl |
2,363 (87%) |
425 (48%) |
|
Aberdeen City |
703 (26%) |
37 (4%) |
|
Comhairle nan Eilean Siar |
154 (6%) |
225 (26%) |
|
Fife |
1,139 (42%) |
60 (7%) |
|
Glasgow City |
30 (1%) |
48 (5%) |
|
Moray |
63 (2%) |
120 (14%) |
|
North Ayrshire |
347 (13%) |
26 (3%) |
|
Shetland Islands |
82 (3%) |
1 (*) |
|
South Lanarkshire |
196 (7%) |
358 (41%) |
|
Other |
3 (*) |
6 (1%) |
|
Minority ethnic |
424 (16%) |
70 (8%) |
|
Recorded having a religion |
1,037 (38%) |
265 (30%) |
|
Currently in S1 |
N/A |
225 (26%) |
|
Currently in S2 |
N/A |
295 (33%) |
|
Currently in S3 |
N/A |
349 (40%) |
|
Say they became eligible for FSM due to Test of Change |
265 (28%) |
N/A |
Secondary monitoring data
The evaluation team worked with the Scottish Government and local authorities to explore the availability of administrative data sources and the possibility of establishing a robust control group or identifying comparable data as part of the evaluation. However, this was not found to be feasible due to a lack of comparable data in LAs not involved in the Test of Change.
Local authorities were asked to provide data on the number of newly eligible pupils, the number of pupils who registered for FSM, the number of meals taken by newly eligible pupils, number of meals taken by S1-S3 pupils eligible under pre-pilot FSM criteria, and number of meals paid for by non-FSM pupils in S1-S3. Seven out of the eight participating LAs shared available data with the evaluation team in Spring 2026. Some high-level figures from this monitoring data have been included where possible. However, analysis was very limited due to a large amount of missing data and the fact that data was collected in different ways across LAs, making it difficult to compare
Limitations
Data analysed in this evaluation is predominantly qualitative in nature. The purpose of qualitative data is to explore the depth and breadth of experiences and is not intended to achieve a statistically representative sample of the wider population. Therefore, estimates of prevalence of experiences captured as part of qualitative interviews are inappropriate. However, where possible, qualitative findings are complemented with quantitative survey findings or monitoring data.
The evaluation cannot establish causality between the Test of Change and observed outcomes. Findings can, however, provide evidence consistent with the programme theory set out in the logic model. In particular, the evaluation did not directly measure changes in child poverty, health, attainment, attendance or behaviour attributable to the Test of Change. Findings in these areas relate primarily to participants’ reported or perceived impacts and should be interpreted as evidence about possible mechanisms and experiences rather than measured programme effects.
In terms of survey findings, it is important to note that not all participating local authorities or schools are equally represented in the final sample. When deciding to participate in the Test of Change, local authorities selected different numbers of schools to participate, ranging from two schools to eighteen schools. This meant that local authorities with more participating schools had higher numbers of respondents. Survey findings are not weighted and therefore should not be interpreted as reflecting the experiences of participating LAs equally. Additionally, when administering the pupil survey, some schools provided the survey link for pupils to complete in class, while others sent the link to pupils to complete at home. Generally, response rates were higher when the survey was completed in class, so some schools where the survey was sent home had lower pupil responses. In addition, the question used to identify parents whose child became newly eligible was asked only of parents who were already aware of the expansion. Findings reported for this subgroup therefore relate to aware, self-identified newly eligible families and should not be interpreted as representative of all newly eligible families.
There is also potential for self-selection and non-response bias. Families and pupils who were more engaged with school meals, more aware of the Test of Change, or who had particularly positive or negative experiences may have been more likely to respond. In addition, pupil responses were clustered within a relatively small number of case-study schools and the achieved sample varied substantially by school and local authority. Subgroup differences should therefore be treated as descriptive differences within the achieved sample rather than estimates for the wider pupil population.
Finally, it is worth noting that local authorities participating in the Test of Change had volunteered on the basis that they were confident that their schools could manage the increase in pupils getting school meals with existing staff and physical resources (kitchen space and equipment, dining space etc.). Due to this, findings relating to implementation capacity may overestimate readiness compared with a wider national rollout. This has important implications when interpreting findings on scalability.
Report conventions
In this report, ‘parents’ has been used to refer to all parents and carers, for brevity. Please note that, from this point onwards, free school meals will be referred to using the acronym ‘FSM’ and local authorities will be referred to as ‘LA’. ‘SCP’ stands for Scottish Child Payment.
The remainder of this report is structured as follows:
Chapter 3 examines planning and early implementation of the Test of Change.
Chapter 4 explores how participating local authorities and schools engaged newly eligible families and pupils.
Chapter 5 examines operational capacity to deliver the Test of Change and implications for scalability.
Chapter 6 summarises levels of free school meal uptake and the factors influencing uptake.
Chapter 7 explores reported and perceived impacts for parents and pupils.
Chapter 8 sets out the conclusions and lessons for any future rollout.
Chapter 9 draws together key findings from the evaluation.
This evaluation was carried out in accordance with the requirements of the international quality standard for Market Research, ISO 20252.
Contact
Email: socialresearch@gov.scot