Secondary school-aged pupils in Scotland - mental wellbeing, relationships and social media: report

2. Methodology

To achieve the aim of this study, the Scottish Government added questions to Ipsos MORI's Young People in Scotland (YPIS) survey^[1]. YPIS is an omnibus survey, which means that it contains questions on a broad range of topics, and its content is provided by multiple clients. Findings based on other questions in the survey that relate to the mental wellbeing impact of COVID-19 mitigation measures (e.g., physical distancing) have previously been published [5], and were briefly summarised in the previous section.

Sample and procedure

The 2021 YPIS survey recruited a sample of 1,361 pupils from 50 Scottish local authority secondary schools (i.e., pupils aged between 11 and 18 years-old)^[2]. Pupils participating in the 2021 YPIS survey completed it online between 8^th February and 2^nd April 2021, with 75% (1016) of pupils completing it at home, and 25% (334) completing it at school^[3].

A breakdown of the sample by demographic characteristics^[4] is presented in Table 1. As shown in the table, due to categories of the characteristics being under- or over-sampled, the sample sizes for each category were weighted statistically before conducting analyses so that the sample was (statistically) representative of the Scottish population. The statistical weighting was based on gender identity, school year, urban or rural, and Scottish Index of Multiple Deprivation (SIMD) quintile^[5]. By recruiting and statistically weighting the sample in this way, the current study addressed limitations of other research conducted during the COVID-19 pandemic, as the sample can be treated as being nationally representative of Scottish secondary school-aged pupils.

Description of survey questions

As previously stated, the YPIS survey contains questions on a broad range of topics. The findings presented in this report focus specifically on questions relating to mental wellbeing, relationships, and social media. The specific questions are detailed below:

• Mental wellbeing. Mental wellbeing was assessed with the Warwick-Edinburgh Mental Wellbeing Scale [WEMWBS; 8], which comprises 14 items, and is a commonly used measure of mental wellbeing. For each item, participants indicate how often over the last two weeks a statement (e.g., 'I've been feeling relaxed') has applied to them, ranging from 'None of the time' (scored as 1) to 'All of the time' (scored as 5). The scores for all the items are then summed to give each participant a total score from 14 (lowest mental wellbeing) to 70 (highest mental wellbeing).
  • The interpretation of WEMWBS scores can be guided using the following values^[8]: 40 or less, probable depression; 41-44, possible depression; 45-59, average mental wellbeing; and 60 or more, high mental wellbeing. However, it should be noted that alternative thresholds have been used, and that these values are used in this report as an approximate guide to assist the interpretation of the findings.
  • After the weighting of the sample, participants who had not completed all WEMWBS items (n=346 unweighted, 25%) were removed from analyses of data collected using this scale. Information about how these omissions affected the weighted sample is reported in Annex 1.
• Loneliness. Loneliness was assessed by asking 'How often do you feel lonely?', with participants able to indicate 'Hardly ever or never', 'Sometimes', or 'Often or always'.
• Optimism. Optimism was assessed by asking participants to indicate the extent of their agreement with the following statement "Even if I am having a difficult time, I feel like I will be OK" ('Agree', 'Neither agree nor disagree', 'Disagree').
• Sleep (quantity). Participants were asked to indicate how much sleep (hours) they had had the previous night. If they could not provide an estimate to the nearest hour, they were asked to indicate an approximate amount from a list of three-hour ranges (e.g., 'Less than 3 hours', '3-5 hours', '6-8 hours').
• Family to talk to about problems. Participants were asked to indicate the extent of their agreement with the following statement "I have family members who I can talk to about my problems" ('Agree', 'Neither agree nor disagree', 'Disagree').
• Enjoyment of spending time with family. Participants were asked to indicate the extent of their agreement with the following statement "In general, I enjoy spending time with my family" ('Agree', 'Neither agree nor disagree', 'Disagree').
• Friends to talk to about problems. Participants were asked to indicate the extent of their agreement with the following statement "I have friends I can talk to about my problems" ('Agree', 'Neither agree nor disagree', 'Disagree').
• Problematic social media use. The Social Media Disorder Scale [SMDS; 8], which comprises nine items, was used to assess problematic use of social media. Participants were asked to indicate 'Yes' or 'No' in response to each item (e.g., 'During the past year, have you tried to spend less time on social media, but failed?'). If the participants answered 'Yes' to five or more questions, this was interpreted as them indicating problematic social media use.
  • After the weighting of the sample, participants who had not completed all SMDS items (n=551 unweighted, 40%) were removed from analyses of data collected using this scale. Information about how these omissions affected the weighted sample is reported in Annex 2.
• Social media and closeness to friends. Participants were also asked 'Does using social media make you feel closer to your friends, less close, or does it make no difference?'.

Participants were able to choose not to answer any of the questions above. As such, where proportions presented in figures do not total 100%, this is due to some participants choosing not to answer the question.

Limitations of the methodology

As the survey relies on self-report questions, the findings presented in this report are subject to limitations common across surveys. Such limitations include not knowing how accurately the pupils' responses reflect their actual experiences, and pupils who volunteer to complete the surveys potentially differing from those who choose not to complete it. Unfortunately, these caveats are common to almost all self-report surveys, and are, therefore, issues that should just be kept in mind.

A more specific limitation is that, due to the methodologies used in YPIS, SALSUS and HBSC differing (e.g., how the samples are statistically weighted), it is not possible to robustly compare the findings from these surveys. As such, while the average scores from the surveys can be compared, the conclusions drawn from such comparisons should be treated with caution.