Student Finance and Wellbeing Study (SFWS) Scotland 2023-2024: technical report

2. Sampling

Sample selection

The selection of students to take part in the study was done in two stages. In the first, we used a census approach to select the sample at institution level and invited all eligible Scottish universities (n=19) and colleges (n=26) to participate in the study.

We assumed a differential response rate for subgroups of HN/undergradates/Further Education students, and full-time/part-time students based on learning from SIES 2021-2022, lowered to account for the proposed web-only approach. Initially, 16 universities and 20 colleges agreed to take part in the study, with 4 colleges subsequently opting out mainly due to capacity or resourcing. A total of 16 universities and 16 colleges participated in the research, representing around 7 in 10 of all institutions in Scotland.

The second stage of sampling was to draw a random sample of students from those institutions that had agreed to take part to produce a representative sample of students in Scottish insitutions to be invited to take part in the study. The size of the sample for each provider was different and based on previous year’s data for the size of each cohort. This selection was carried out by the institutions who were sent a detailed protocol for how to randomly select a sample of students to be invited to participate in the survey.

Institutions were first asked to exclude from their sample any students that were not eligible to take part in the study. The sample of students included the following eligibility protocols:

• Domicile – all Scottish domiciled students only.
• Year of programme – all years of study.
• Mode of study – full-time and part-time. Part-time was defined as: For HE students – those studying at least 25% Full-Time Equivalent (FTE) or 30 or more credit points in the academic year up to 119 credits. For FE students – those studying 160 hours plus (160 hours - 320 hours, and 320 hours up to full time) / 4 credits as a minimum threshold for part-time activity on a recognised FE qualification (or sometimes known as substantial part-time FE courses).
• Length of course – standard length or extended courses.
• Level of study – students following any courses / across all subjects. This includes those studying on campus or those studying at a distance and includes sandwich students who are on their first year of study.

The exclusion criteria were:

• Non-Scottish domiciled students.
• Scottish domiciled students taught at the institution but registered elsewhere (not in Scotland).
• International students.
• Students under 16 years old.

The protocol then provided institutions with step-by-step instructions on how to draw a systematic random sample of eligible students from all years of study. The majority of participating institutions (n=30) were asked to draw the sample using this method, with two colleges being asked to take a census approach, inviting all eligible students to take part due to the smaller numbers of eligible students in their institution.

To comply with GDPR and data sharing permissions with students in Scottish institutions, it was not possible for student data to be passed to ScotCen without further permissions being granted by students. Institutions were therefore asked to distribute the survey directly to sampled students. This approach – used on SIES for first year students – minimised the possibility of delays in establishing data sharing permissions in a timely fashion, or the need for institutions to provide complete pseudonymised student data for sampling and distribution purposes.

The instructions specified the number of full-time and part-time students to be selected. Part-time students were oversampled relative to full-time students because part-time students had a lower response rate than full-time students in SIES 2021-2022. Additionally, in Scottish universities (excluding the OU), part-time students make up a minority of the total population (around 18%) but were a subgroup of interest to this study.

Sampling targets for institutions were calculated using the same approach as SIES 2021-2022. Overall targets for full- and part-time students were distributed according to the size of institutions and their proportions of full- and part-time students using population data from the 2021-2022 academic year provided by Higher Education Statistics Agency (HESA) and the Scottish Funding Council (SFC). The percentage of the estimated eligible population to be sampled, also known as the sampling fraction, was approximately 22% for participating colleges and 18% for participating universities.

Where institutions could sort their student records prior to drawing the sample, they were instructed to sort by level of study, FE, HN/Undergraduate, Postgraduate, year of study, sex, age – in that order. These characteristics were all of interest for subgroup analysis therefore sorting the sampling frame ensured a representative sample of students drawn by each institution.

During the sampling process ScotCen reviewed the overall profile of all students in their institution compared with that of selected students on these five subgroups. The most recent institution-level data from Higher Education Statistics Agency (HESA, 2021-2022)^[i] and Scotland Funding Council (SFC, 2021-2022)^[ii] were referred to as a baseline for comparison. For a small number of institutions with a considerable discrepancy between this and the profile of the sampled students or their current student population, ScotCen discussed this with the institutions to determine the rationale or whether the sample needed to be redrawn or target numbers of full- and part-time students reset.

Achieved sample profile

After the data were cleaned and edited, the achieved sample was 2,429 responses. This total was made up of 362 FE students, 1636 HNC, HND or other undergraduate students, and 411 postgraduate students.^[3]

Table 2.1 below sets out the profile of these three different study level student groups by the key cross break variables used in the analysis: study intensity (full or part time), age, sex, parental experience of higher education, accommodation type and area deprivation^[4]. This shows the inclusion of a wide range of students from different age groups and socio-economic backgrounds in the achieved sample for the study.

Missing data, such as those who responded ‘don’t know’ or ‘prefer not to say’ have been removed from the table, so each analysis grouping does not sum to the total number of respondents (2,429).

There were 4 students who stated their age as between 16 and 19 years old and that they were studying at postgraduate level. Full quality assurance checks were done on this data, so they have been included in the sample.

Percentages do not always sum to 100% due to rounding.

The report includes analysis of five student groups of interest to the study: care-experienced, parents, estranged, unpaid carers, and disabled students. It was not possible to split these into FE, HN and undergraduate, and postgraduate students due to low numbers, so analysis was undertaken on all student respondents in these five groups of interest regardless of their level of study to provide some high-level insight on their financial experiences whilst studying.

In summary as set out in Table 2.2 below:

• 110 respondents who identified as care experienced took part in the survey,
• 116 respondents reported they were estranged from their parents/ carers/ legal guardians,
• 429 respondents said they were student parents,
• 174 were unpaid carers; and
• 1003 reported that they had disabilities.

Given the intersectional nature of these experiences, students may have identified with one or more of the above.

Table 2.2 also sets out the profile of each of these five groups by the key cross break variables of study intensity, age, sex, parental experience of higher education, accommodation type and area deprivation.

Missing data, such as those who responded ‘don’t know’ or ‘prefer not to say’ have been removed from the table, so each analysis grouping does not sum to the total number of respondents in each category.

Percentages do not always sum to 100% due to rounding.

Asterisk (*) denotes figure of below 1%.

Living with parents includes living with other relatives, carers or legal guardians.