Case Study 5: Randomised Control Trial (RCT)
The impact of summer schools on widening participation in HE
Project description and aim:
In 2019, the Education Policy Institute (EPI) and TASO conducted an evidence review of pre-entry interventions for HE by assessing the benefit of summer schools in boosting HE enrolment rates. The evidence review suggested that attendance at summer schools was correlated with increased aspiration and confidence related to HE; however, this evidence is correlational rather than causal. There is a lack of evidence definitively showing that summer schools are an effective intervention, particularly in increasing enrolment rates in HE for disadvantaged and underrepresented groups. To build a causal evidence base for summer schools, a group of researchers initiated a randomised controlled trial (RCT) in collaboration with eight partner HE providers to evaluate the impact of summer schools on aspirations and attitudes related to HE and on HE enrolment rates.
Methods:
This project involved eight partner HE providers (HEPs) in England. Four HEPs ran summer schools for students under the age of 16 (Year 9 and 10) and four HEPs ran summer schools for students over the age of 16 (Year 12). Applicants to each summer school were given participant information sheets that explained that the summer school they were applying to was part of a research project and what this entailed. At this point, applicants could choose whether they wanted to participate in the research project and were told that their decision would have no impact on whether they would be awarded a summer school place or not. All applicants meeting the summer school eligibility criteria (widening participation criteria) were randomly allocated to either the intervention group (they received a place) or the control group (they did not receive a place). This was possible because both summer schools were oversubscribed and, therefore, had more applicants than places available. Before and after the summer school, both intervention and control group students were asked to complete a survey on their attitudes and aspirations regarding HE. In the long term, the researchers used the HEAT to track whether applicants later enrolled in HE.
Key ethical considerations:
The eight partner HEPs included students from different age groups. The four HEPs targeting students under the age of 16 gave students the option to opt out of the study. Students and/or their parents/carers had to complete an opt-out form within two weeks if they did not want to participate in the research. It was deemed that these students were old enough to make their own decision; however, their parents could support the decision if required. The opt-out process was deemed ethical because all students would be required to give full consent at the point where their data was being collected, that is when they completed the questionnaire.
At this point, opt-out was chosen to allow students and parents not to be involved, but to avoid narrowing the pool of potential participants, as good coverage across the entire participant population is essential to ensure that analysis and findings from the data are accurate, unbiased and representative. If the researchers were to ask pupils and parents for fully informed consent at this point, only the most engaged pupils and parents, who actively take steps to opt-in would be included. The researchers would miss the overall story of differential education trends, because data under the opt-in model would reflect only a small, selective portion of the population.
For the four HEPs targeting students over the age of 16, opt-in consent was used at the initial application stage. Consent was not required from parents as the students were over 16. As part of their application to the summer school, students were required to read the participant information sheet and tick a box if they were happy to participate in the research.
Regardless of whether consent was given, all applicants were part of the randomisation process described above which decided their place on the summer school; however, those applicants who chose not to participate in the project were not tracked after the project or contacted to complete any surveys.
Full information was provided to students at the beginning of each survey and its submission was deemed active consent. Given the low risks of the survey and the relatively short time required for completion, consent from students, regardless of age, was deemed sufficient.
Explicit consent from students was also gained to track their future application(s) to HE.
Scientific limitations and recommendations for future research:
Some of the partner HEPs used local partner schools to recruit for their summer schools. These schools were only involved in the project via a third party and were not, therefore, directly engaged with the researchers. This posed difficulties for the researchers and the HEPs, who did not know whether students or parents had been fully informed about the nature of the project or whether the parental opt-out form was completed by parents or a school-based gatekeeper. However, as the core ethical consent was concerned with the submission of the survey and this did not require parental consent, it was not seen as ethically problematic.
The data for this research was gained from two sources and consent was gained in different ways for each. This led to some individual students’ data being only partial. For instance, applicants could decide to participate in the research project, which meant they could be contacted to complete surveys; however, they separately gave consent (or not) to have their long-term data tracked by HEPs using the HEAT database. Thus, in some cases, the researchers had permission to use the student data in the research project but did not have permission to track using HEAT. Conversely, some students did not agree to take part in the research project, but individual HEPs were able to track HE enrolment using HEAT.
In future iterations of the project, it would be simpler to gain consent to use both sets of data at the same time.
Legally, pupil data can be tracked using the HEAT database on a ‘public task’ basis; therefore, the HEPs and researchers did not require legal consent from students. However, ethically, they ensured that all students were informed of the data tracking and linkage, and the purpose of these activities, as part of the project.