Smoking, Drinking and Drug Use among Young People in England, 2023

The survey of Smoking, Drinking and Drug Use amongst Young people in England has been running since 1982. The survey takes place in secondary schools, with pupils in curriculum years 7 to 11 (mostly aged 11 to 15). Each survey since 1998 has included a core set of questions on smoking, drinking and drug use. In 2000, the survey questions changed to focus on smoking and drinking or on drug use in alternate years, but in 2016 the survey reverted back to including both drinking/smoking and drugs focused questions in one survey. However, only about half the pupils are asked the more detailed questions on smoking and drinking, with the other half asked more detailed questions on drug use. 

Since 2016 the surveys have been carried out by Ipsos UK on behalf of NHS England. The 2020 survey did not go ahead due to the Covid pandemic and was postponed to 2021. Surveys prior to 2016 were carried out by NatCen.

Further detail on the 2023 survey methodology including all changes is provided in the following sections.

In 2023, 13,387 pupils were surveyed between September 2023 and March 2024 across 185 schools, of which, 13,192 were included in the survey sample for this publication. A further 4,079 pupils from 27 schools were surveyed in volunteer schools. Data from volunteers will be made available in an anonymised dataset later in 2024

The questionnaire is available on the main publication page.

A2.1 Changes to questionnaire content in 2023

Ipsos UK were commissioned by NHS England to conduct the 2023 Smoking, Drinking and Drug Use among Young People in England survey. This biennial survey has been running since 1982 and provides vital information on smoking, drinking and drug use amongst young people, which is used by central and local government to better understand these behaviours and to develop policies, plan services and new initiatives, and to monitor and evaluate their impact. The survey has been conducted by Ipsos UK on behalf of NHS England since 2016.

The 2023 survey is the first one to be administered online rather than on paper. Designed to make the study easier for schools¹ to administer, the online survey approach is also easier for students to complete as they are automatically routed to sections that are relevant to them, rather than having to follow complex routing instructions, which in turn improves data quality. An additional benefit, of automatic routing is that it reduces the survey length; in this case to less than 30 minutes, compared to 40-50 minutes for the paper version.

Further information about the background to and development of the Smoking, Drinking and Drug Use survey, along with survey questionnaire is available on the NHS England website: https://digital.nhs.uk/data-and-information/areas-of-interest/public-health/smoking-drinking-and-drug-use-among-young-people-in-england

This appendix covers the 2023 Survey of Smoking, Drinking and Drug Use amongst Young People in England. The survey was targeted at secondary-aged students in England who attend mainstream school. In total, 13,387 students in England were surveyed between September 2023 and March 2024 across 185 sampled schools. The survey consisted of two components:

• A survey completed by students online in school lessons, supervised by a member of school staff.
• An online questionnaire completed by one teacher from each participating school, which included questions on aspects of the curriculum which focus on smoking, drinking and drug use, as well as information on the number of students that took part in the survey.

Further detail on the survey methodology is provided in the following sections.

^{(1) A pilot conducted in 2018 by Ipsos, demonstrated that when provided with the option of conducting surveys online or on paper, most schools opted for the online approach with 70% of student completing a questionnaire online and 30% on paper.}

A2.3 School discussions

The research team conducted discussions with six schools in May 2022 in order to validate the proposed approach for the role of the school in the survey process. The schools who participated in the discussions were part of the School Advisory Group, representing different areas of the country; both rural and urban, large and small schools, and a range of IT capacity. Overall Ipsos conducted eight interviews with a range of staff including Headteachers, Deputy headteachers, Heads of PSHE, Heads of Year, IT and admin staff.

Key areas of discussion

• How schools will be involved in the administration of the survey, and how to limit the burden on teaching staff.
• How anonymity can be guaranteed in all stages of the survey process (and if anonymity cannot be achieved, how this is managed).
• Understanding how familiar schools are with technology and exploring the possibility of setting a dual approach for administering the survey (web and paper).
• Checking the feasibility of the modular approach, both in terms of administration and survey length.

Key findings

Feedback confirmed that schools have the necessary technology to administer an online survey, and were supportive of a move away from paper-based survey, indicating that an online approach would indeed reduce the burden on schools.

The development interviews were valuable in terms of hearing from schools about their previous experience of administering online surveys. Many have virtual learning environments in place (e.g. Google Classroom) designed to share materials and links with students online, making the process easier.

A number of schools that took part in the interviews highlighted that they have safeguarding software in place to monitor students’ online activities whilst in school. This finding from the development stage informed the content of guidance sent out to schools which emphasised that device monitoring software should only be used at a high level to see if a student is on the survey website, not to view students’ responses to the survey.

The school discussions also highlighted the possibility that the survey could be blocked by the school firewall due to the terms used in the survey, e.g. names of illegal drugs. To address this issue, we designed a test link for teachers to trial in advance and shared instructions for IT staff on how to whitelist the survey link. The test link was about 5 minutes long and included a mix of questions containing words or phrases which potentially might be blocked so that these issues could be addressed prior to students taking survey.  

In the development stage we discussed having student-level, class level or school level links. Based on feedback from these initial discussions, it was agreed that school level links would be the best approach, as they would not require schools to distribute student-level links and would ensure anonymity.

Teachers stressed the importance of making students aware that their responses to the survey are anonymous, to ensure they complete the questionnaire honestly. The research team took steps to ensure that the student information leaflet and the start of the survey script highlighted the fact that student answers are confidential.

After consulting schools, the decision was made to eliminate mop-up sessions in 2023 because anonymity could be not fully guaranteed during mop-up sessions. Only a couple of students would take part in mop-up sessions if they were held, meaning that their survey completion timestamps could identify them. Eliminating mop-up sessions also reduced the burden on schools.

A2.4 Pilot stages

Two pilots were conducted as part of the survey development stage. The pilots were designed to:

• Test the survey questionnaire length.
• Test the look & feel of the online survey (inc. elements that have proved
  challenging for school IT in the past).
• Assess student’s understanding of key questions/ new questions.
• Review language used in the questionnaire (e.g. e-cigarettes).
• Gauge how schools ensure that students complete the survey to the end.
• Test how the survey will be administered in schools.
• Review the materials which are sent out to schools (E.g. Advance letter, instructions, privacy policy) for clarity.
• Gather feedback on the school reports.

Pilot 1:

Seven schools took part in the first stage of the pilot, which took place between 3^rd February and 10^th March 2023. Schools were recruited from a list of those who took part in the School Advisory Group or the 2021 Smoking, Drinking and Drug Use Survey among Young People in England.

The member of staff who agreed to take part in the pilot from each school was typically a PSHE teacher. In each school the teacher was asked to arrange a convenient time for the survey to take place with one class and were sent a link to the online student survey, along with the parent/carer notification letter, and leaflets for participating students and teachers. Follow up interviews were conducted with class teachers and other members of school staff after students had taken part.

Overall, the pilot schools feedback demonstrated that the set-up of the survey worked well, including the process of sending out advance letters to parents/carers via electronic mail and using the online test link to ensure the survey would not be blocked by firewalls. Teachers noted that they had shared the student link with students using email, Google classroom, whiteboards, and MS Teams. It took students around 15 to 30 minutes to complete the survey, which was considered ideal, as it meant that an entire lesson period would not be required for classes to take the survey. Teachers were confident that students had completed the full survey. They asked students to complete the survey in silence, and to treat the survey as if it were an exam. Students and teachers responded well to the informational materials provided, finding them easy to read and understand. It was reported that the student leaflet contained all the information needed for teachers as well, eliminating the need for a separate teacher information leaflet.

Pilot 2:

The second pilot was designed to test the recruitment process, survey set-up and materials with schools who had not previously engaged with the survey.

A random sample of 50 schools were sent an advance letter with information about the survey and inviting them to take part. Ipsos recruiters contacted sampled schools once the letter had been received. Seven schools took part in the second stage of the pilot, which took place between 24^th April and 9^th June 2023.

Key recruitment challenges were identified, namely schools being busy and low student response rates. These findings indicated the need to re-design the supporting materials to highlight incentives for schools (the individual school report and the PSHE Association lesson plan), and the importance of starting recruitment early.

Changes made to the survey process after both pilots:

• The email sent out to schools containing the test link, student link, and teacher link was redesigned into a grid format to help teachers identify the purpose of each link.
• The test link was changed to include questions which proved to be common drop-out points for students.
• Information leaflets for teachers and schools were redesigned to emphasise that the survey should not be completed on a mobile device, that it should be completed in exam conditions, and that teachers should use the test link. Schools were also given the option of receiving survey guidance materials via online copy or hard copy.
• Feedback from teachers suggested that the PSHE Association lesson plans would be welcomed by teachers. Therefore, a focus was placed on promoting the PSHE Association’s involvement, indicating how the survey can be incorporated into PSHE curriculum, as well as the usefulness of the school report and the PSHE lesson plans.
• For the most part, the questionnaire was found to be accessible (students could complete it without help) and appropriate for the age groups surveyed. However, a small number of students queried the language used to describe solvent use and queried some of the terms used to describe drugs. Based on this feedback questionnaire changes were made to ensure the question was easy to understand by students (See appendix C for changes).

A2.5 School Advisory Group

The School Advisory Group meeting took place on 19^th April 2023. During the session we asked teachers about the practicalities of administrating the survey and steps that we could take to reduce the burden on teachers.

The key takeaway was to give teachers as much notice about the survey as many plan their curriculum months in advance. Based on this feedback Ipsos sent out the initial letter to schools informing them of the 2023 survey in the summer term of 2022/23; the term before the fieldwork was due to take place. In response, around 100 schools expressed an interest in taking part. 

We also invited the School Advisory Group to feedback on the PSHE Association lesson plans. Overall, the group spoke very positively about the lesson plan. Their key recommendation was to send the initial lesson plan with the first mailing to allow teachers to incorporate the survey into their curriculum.

A2.6 Changes to Questionnaire Content in 2023

In 2021, there were two separate versions of the survey questionnaire, one with a focus on smoking and alcohol and the other with a focus on drug use. This design was used previously to reduce the questionnaire length as it took longer to complete on paper. In 2023, the switch to an online survey (and the ability to automatically route students’ past questions that were not applicable) provided the ability to combine the two questionnaires.

NHS England and Ipsos reviewed the 2021 questionnaire and agreed on questions which questions should be removed or added. The full list of changes can be found in appendix C.  

During the design of the questionnaire some changes were made with the aim of improving the data quality compared to the paper survey. These changes are outlined below:

• Soft checks were included in the online survey if students gave an answer which contradicts a previous answer given. For example, if they said their age was 13 but then said they were 16 when they first had a cigarette they would be asked if their response was correct and given the opportunity to change their answer.
• ‘Prefer not to say’ and ‘Don’t know’ response options were sometimes only shown if the ‘next’ button was clicked on without a response given, this was designed to encourage students to give a response to every question.
• On the test link and the teacher link a warning page was added to make it clear the link was for teachers to complete only. This change was made after the pilot when students were completing the test link or teacher link instead of the student link.

The survey population comprised students in Years 7 to 11 in secondary schools, or in an equivalent year group in middle and upper schools. At the time of sampling, most of these students were aged between 11 and 15 although some of those students in year 11 would have been 16 by the time they took part in the survey. The sample included almost all types of secondary school in both the maintained and independent sectors of education in England. Special schools and student referral units were excluded from the survey².

The survey uses a multi-stage probability design, in which first schools and then classes are selected using random methods. In 2023, 1,265 schools in England were sampled. In each of the participating schools three classes were then sampled, (one class from Years 7 and 8, and two classes from Years 9, 10 and 11).  All students in the sampled classes were eligible for the survey.  More classes were sampled from the older year groups to increase the precision of estimates (see section A3.1 on sampling classes for more information). This sampling design follows the design adopted for the first time in 2018, and repeated for 2021, of sampling three whole classes.

The majority of schools completed three classes to participate in the survey, while 24 schools only completed one or two classes, but it felt better to include these schools than exclude them and reduce the response rate (this was also the approach in 2018 and 2021).

A3.1 School sampling

The same sampling frame was used as in the 2021 survey. The aim of the sample was to achieve an equal number of schools in each region – 23 schools in each region. In order to do so, the number of schools selected in each region was based on the region-level response rate achieved in the 2021 survey.

The sampling frame was stratified by region, type of establishment (academy, voluntary, aided/controlled, community, foundation, independent); mixed or single sex; and average IMD score for all non-independent schools. In each region, schools (the primary sampling units) were selected at random from Get Information about Schools³, the Department for Education's register of educational establishments in England and Wales.

The probability of each school being selected was proportional to the numbers of students in Years 7 to 11, so that larger schools within each region had a higher chance of inclusion.

The numbers of schools selected in each region is shown in Table 1

Table 2 shows the number of schools in each stratum and the total number of selected schools allocated to each one.

Table 2: Number of schools in each stratum

^{(a) Students in Years 7 to 11: based on National Population Database for January 2022 and DfE population estimates for independent schools.}

A4.1 Contacting schools

1,686 schools were sampled to take part in the survey – 1,265 in the main sample and 421 in the reserve sample. All schools in the main sample were sent an advance letter in June 2023, explaining that they had been selected to take part in the survey and that they would be contacted at the start of the autumn term. The letter also contained information about the background of the survey and details of the incentives, which included a payment of £150, a personalised school report, and two PSHE Association lesson plans.

A reminder email was sent in July, explaining how to take part in the survey and also sharing the first PSHE Association lesson plan with schools, reminding them that a second PSHE lesson plan would be made available after taking part in the survey. Another letter was sent to sampled schools in September to inform them of the survey launch, which included information about how the survey works, what schools needed to do to participate, and what happens to the data that is collected. In total, 181 schools contacted the research team expressing interest in taking part in the survey.

A reminder email was sent to regions with low participation rates in the survey in January.

Recruiters contacted schools between September and February, inviting them to take part and, if they agreed, booked a date and time for three classes to take part in the survey.

A4.2 Encouraging schools to take part

In a change from the 2021 survey, the decision was made to contact all sampled schools earlier in the summer term to let them know that the survey was coming up and to give them an opportunity to express their interested in taking part. This decision was informed by feedback from the School Advisory Group.

All sampled schools were offered an incentive of £150 to take part and a school level report upon publication of the survey data, which provides comparisons between a schools’ aggregate results to key questions in the survey and the national averages. The PSHE Association was consulted in the development of the school report.

PSHE Association Lesson Plans

The PSHE Association is a membership association, supporting over 60,000 teachers with resources, training, guidance and advice. The PSHE Association designed two lesson plans to be distributed to schools, which meet the requirements of the national curriculum. The first lesson plan was sent out to all sampled schools in their original invitation letter, a choice that was again informed by input from the school advisory board.

The first lesson plan covered consent in relation to how data is collected, stored, and used, and the factors that affect someone’s decision to take part in research. Schools who completed the survey were sent the second lesson plan in addition to their incentive. This lesson plan covered how young people’s participation in research can be protective and empowering at a community level.

PLMR Communications Agency

PLMR are a leading education communication consultancy, working with academy trusts, independent schools, education charities, curriculum organisations and teacher unions. In order to raise the profile of the survey, PLMR reached out to their contacts to encourage them to take part in the survey and/or advertise the survey in their networks. PLMR also arranged for press releases about the survey to go out in local news coverage in order to raise awareness about the survey. For example, they created articles for Schools Weekly and TES (Times Educational Supplement) as well as local press releases focused on regions with a lower uptake. They also linked the survey promotion with media stories on vaping and engaged with journalists who had recently covered articles on young people and vaping.

Key recruitment messages

Recruiters were told to emphasise the importance of the study in informing policy and services that help support and protect young people (particularly in regard to the topical policy matter of vaping), the fact that the survey has been ethically reviewed and that the research team has partnered with the respected PSHE Association, the focus on minimising the burden placed on schools, and the incentive that was offered to schools (see below).

Recruiters were also asked to highlight the fact that this year’s survey was shorter than last year’s, meaning that the survey could be completed in 20-30 minutes.

A4.3 Carrying out the survey

In 2023, all schools were required to arrange for their students to complete the survey online. In previous years interviewers had gone in to schools and handed out paper questionnaires, but changes made in 2021 in response to the COVID-19 pandemic and visitor restrictions in schools, demonstrated that the teacher led approach had worked successfully.   

Once a school had agreed to take part they were then sent an email containing the online links to the student and teacher surveys, a test link for teachers to use several days prior to their class taking part to ensure that the linked worked with their school’s software, a student information leaflet, and a letter to share with parents/carers, informing them that their child will be taking part in the survey and giving them the opportunity to opt their child out of the survey. Students completed the questionnaire under the class teacher’s supervision in exam conditions.

A4.4 Withdrawing students from the survey

Schools are required to send out letters to parents/carers of students in selected classes to inform them when the survey will be taking place and to provide information about the survey content. The letter provides parents with an opportunity to withdraw their child from the survey.

In accordance with MRS guidelines⁵ cases, students were also provided with an opportunity to decline to take part (even though a responsible adult has given permission for their participation). In total, 27 students from sampled schools decided to opt out or were withdrawn out by their parents/carers⁶.

In total, 185 sampled schools took part in the survey out of the 1,265 eligible schools, providing a response rate of 15%. The reserve sample was not issued.

The response from selected students in participating schools was 90%⁷, yielding a total of 13,387 interviews. The product of the school and student rates gave an overall response of 13% (See Table 3 and Figure 2).

Table 3: School and Student Response, 1982 - 2023

Figure 2: SDD Overall response rate 1982-2023

Since the survey began in 1982, school response rates have continued to decrease. 2023 was the first year since 1999 where the response rate has increased. This is likely the result of a combination of factors including the involvement of the PSHE Association and PLMR, the move to an online survey, contacting schools earlier in the summer term and the interest in information on young student and vaping. Student response rates have remained high across the survey’s lifetime. 

Table 4 below shows the response rate by region which varied considerably. There is more discussion on the impact of this Appendix B

Table 4: Sampled school response by region

The number of schools who actively refused to take part was lower than in 2021, from 376 to 135 refusals. However, it’s important to consider the impact of COVID-19 in 2021; and the issues schools were facing at that time. Key reasons why schools refused to take part in 2023 are listed below.

Table 5: Reasons given by schools for not taking part in the survey

Overall, 945 schools did not respond to the invitation to participate in the survey; either to decline or accept.

As with previous years, the following checks were made to the data:

• Checking that filters were correctly followed.
• Checking ranges on consumption variables and age of first use variables.
• Checking whether answers given as ‘other’ answers could be back-coded into existing codes.
• Resolving inconsistencies between answers.

A dataset was then created for analysis purposes.

The sample is weighted to match the school population counts so that estimates from the survey are representative of the full school population. The school population counts were derived from the most recently completed DfE School Census.

The weighting was carried out in two stages. Because schools were allowed to include more than one class in a school year for the 2023 survey, but the sampling was based on only including one school year, the first stage was to produce a set of weights that adjusted the number of students in any school year for these school back to a single class. This was done by calculating an adjustment weight as 1 divided by the number of classes included, or using a nominal class size of 25 when the full school year was included. An adjustment weight was also applied to limit the class size for any other schools to 40.

The second stage was then to use cell weighting to adjust the weights from the first stage directly to the population counts for school year by gender separately for each region (Table 6).

Table 6: Population counts used for weighting

These weights were then rescaled to sum to the achieved sample size. Both unweighted and weighted bases are shown in the tables. The unweighted bases represent the number of students who responded, the weighted bases have no particular meaning per se, but can be used when combining estimates

Volunteer approach

During the development of the 2023 survey, the decision was taken to open-up the survey to schools who wanted to volunteer to take part, to test how many schools would potentially volunteer without dedicated recruiters contacting schools.

Promotional activities that were put in place, including the PLMR publicity campaign and promotion from the PSHE Association, resulted in the attention of secondary schools in England outside of the sample who contacted Ipsos to take part in the survey.

Schools that heard about the survey via these promotional activities, their local education authorities, other schools reached out to the research team expressing an interest in taking part in the survey. They were then sent a confirmation email by the research team which included a link to the student, teacher, and test surveys, as well as core materials like the parent/carer letter, student leaflet, and information about how to take part in the survey. Volunteer schools were asked to ensure that one unstreamed/ mixed ability class from each year group (years 7-11) took part in the survey.

Schools that expressed interest in taking part but did not complete the survey were followed up with two separate reminder emails. All communications with volunteer schools were handled by the research team and not by the recruitment team, in contrast with the procedure for sampled schools. This meant the communication with volunteer schools was more light touch as they did not have a dedicated recruiter to book them in and remind them to take part in the survey.

Data gathered from these volunteer schools would not be included in the national dataset, as the profile of volunteer schools was unknown. However, data from volunteer schools will be made available in an anonymised dataset later in 2024. Each school that participated received their own school-level report and the PSHE lesson plans—but no financial incentive.

Profile of volunteer schools

62 volunteer schools expressed an interest in taking part and were sent the links, but only 27 completed a session with at least one class. Table 7 below outlines the profile of volunteer schools who took part in the survey; highlighting that the take-up from volunteer schools was not consistent by region – although the achieved profile is not out of line with the national profile. By school funding type, there was a notable shortfall in independent schools volunteering to take part.

Table 7: Profile of volunteer schools

Table 8 outlines the information on the number of students from volunteer schools who took part in the survey. While all schools received written guidance on how to conduct the survey and the requirements regarding year group and class selection, we found that without the guidance of recruiters, there was a great deal of variation in the numbers taking part in each school. Some schools chose to include whole year groups, whereas only small numbers of students from one class took part in other schools. Volunteer schools were advised to conduct one classroom session with each year group, from Year 7 - Year 11, but as shown in Table 1.2, the average number of year groups participating in the survey in schools was three. 

Table 8: Number of completes per a school