Technical annex Download
10 Appendix
10.1 Survey materals
The following survey materials are available on the website:
All letters and text messages are available on request from the GP Patient Survey team at GPPatientSurvey@ipsos.com.
10.2 Experiments
As part of the 2026 GPPS fieldwork, Ipsos and NHS England undertook a series of experiments to try to improve cost-effectiveness while maintaining response rates, as well as examining any changes in sample profile, survey estimates and costs, compared with the main survey design.
Experiments were carried out using a factorial design. Factorial design involves testing interventions in combination, rather than one at a time. Interventions can then be analysed by combining all groups involving that particular intervention. In addition, this has the advantage of allowing the impact of a combination of interventions to be analysed (e.g. potentially demonstrating that a positive impact on response rates from the introduction of a second e-letter instead of the final paper mailing and a separate positive impact from sending an email still applies when the two are combined).
This led to a series of experiments on a sub-sample of GPPS respondents, testing the impact of contacting people in a variety of ways including:
-
via the NHS App, as first contact (replacing the initial letter) and as an additional contact: designed to increase response rates by sending an invitation (or reminder) through a legitimate source, as well as reducing survey costs.
-
sending all contacts via the NHS Notify service: this could be cheaper for the NHS, and benefit from the legitimacy of the NHS as the sender of communications.
-
replacing the initial letter with a text message as first contact: designed to reduce survey costs by decreasing the volume of printing and encouraging more people to take part online.
-
replacing e-letters, which are accessed via a text message, with a single text message: designed to reduce the perceived ‘burden’ of taking part by reducing the number of clicks required to enter the survey.
It was hoped that an additional experiment could be run replacing text message reminders with WhatsApp messages. However, there were a number of issues establishing the business account and subsequently the messages were blocked by Meta meaning that analysis for this group is not possible.
Experiment wording was also included in the online survey for a small group of the sample to support decisions about future changes made to the questionnaire. This included:
-
testing the inclusion of partial online responses: looking at the impact on response rates of adding an additional sentence to the survey introduction to support the potential inclusion of partial responses in future: ‘If you leave the questionnaire for any reason and want to come back to it before the survey fieldwork period closes, it will pick up where you left it. When the survey closes, your answers will be automatically submitted’.
-
testing the impact on trends of a change to an existing question: the wording at Q19 (‘Were you offered the following choices?’) was amended to include additional answer options and to clarify that this related to the ‘last appointment’. This was to understand whether trends could be presented for this question and to establish the potential impact on following questions (‘context effects’).
10.3 Allocation of sample to treatment groups
With the total sample of 2.97 million cases, roughly 150,000 cases were systematically selected to take part in the various experiments (using the method of random start and fixed interval). The sample was stratified by practice, with a ‘1 in n’ random selection. The 150,000 cases were randomly allocated to each of 13 treatment groups, with the rest of the sample receiving the ‘control’ approach.
10.4 Results
Analysis was carried out looking at the interventions across the different treatment groups (for example, looking at all the groups that were invited to take part through the NHS App) to look at the differences for each intervention compared with the groups that had not experienced that intervention. This concluded that:
-
adding an additional contact through the NHS App following a paper invitation to the survey had a small positive impact on the response rate (+0.9 percentage points).
-
there are a number of operational and cost considerations that need to be explored further in relation to using the NHS Notify service in order to deliver a survey of the scale of GPPS.
-
sending an RCS or SMS message in place of an initial paper invitation letter generated the largest estimated cost saving of all the approaches tested this year. However, it resulted in a small decrease in response rate (-0.7 percentage points).
-
it would be possible to replace the e-letter with a single text message, most effectively where it follows an earlier digital invitation rather than a physical letter to take part, with minimal impact on response rates.
Across these experiments, non-response bias analysis has shown that there are no differences in how respondents answer survey questions, or differences in demographic profiles, between the experiment groups and control.
In addition, analysis has been carried out to look at the differences in results between the experiment groups and the main survey across a range of key survey estimates and demographic profiles. This found that there would be no impact on the results from including the experiment groups within the overall responses. Responses to the experiment interventions have therefore been included within published results for 2026; they equate to 4.6% of the total number of returned questionnaires.
Our standards and accreditations
Ipsos’s standards and accreditations provide our clients with the peace of mind that they can always depend on us to deliver reliable, sustainable findings. Moreover our focus on quality and continuous improvement means we have embedded a ‘right first time’ approach throughout our organisation.
ISO 20252
This is the international market research specific standard that supersedes BS 7911/MRQSA and incorporates IQCS (Interviewer Quality Control Scheme). It covers the five stages of a Market Research project. Ipsos UK was the first company in the world to gain this accreditation.
Market Research Society (MRS) Company Partnership
By being an MRS Company Partner, Ipsos UK endorse and support the core MRS brand values of professionalism, research excellence and business effectiveness, and commit to comply with the MRS Code of Conduct throughout the organisation & we were the first company to sign our organisation up to the requirements and self-regulation of the MRS Code; more than 350 companies have followed our lead.
ISO 9001
International general company standard with a focus on continual improvement through quality management systems. In 1994 we became one of the early adopters of the ISO 9001 business standard.
ISO 27001
International standard for information security designed to ensure the selection of adequate and proportionate security controls. Ipsos UK was the first research company in the UK to be awarded this in August 2008.
The UK General Data Protection Regulation (GDPR) and the UK Data Protection Act (DPA) 2018
Ipsos UK is required to comply with the UK GDPR and the UK DPA. It covers the processing of personal data and the protection of privacy.
HMG Cyber Essentials
This is a government-backed scheme and a key deliverable of the UK’s National Cyber Security Programme. Ipsos was assessment-validated for Cyber Essentials certification in 2016. Cyber Essentials defines a set of controls which, when properly implemented, provide organisations with basic protection from the most prevalent forms of threat coming from the internet.
Fair Data
Ipsos UK is signed up as a “Fair Data” company, agreeing to adhere to 10 core principles. The principles support and complement other standards such as ISOs, and the requirements of Data Protection legislation.
For more information
3 Thomas More Square London E1W 1YW
t: +44 (0)20 3059 5000
About Ipsos Public Affairs
Ipsos Public Affairs works closely with national governments, local public services and the not-for-profit sector. Its c.200 research staff focus on public service and policy issues. Each has expertise in a particular part of the public sector, ensuring we have a detailed understanding of specific sectors and policy challenges. Combined with our methods and communications expertise, this helps ensure that our research makes a difference for decision makers and communities.