Improving our travel and tourism statistics: changes from July 2024

• From 1 July 2024, the Office for National Statistics (ONS) will change the data collection and methodology for estimating overseas travel and tourism; this will allow for more accurate and coherent statistics.

• The International Passenger Survey (IPS) departures data will be harmonised with the Civil Aviation Authority's (CAA) departing passenger survey data; we will use data collected from both sources in our estimates, and move to new interviewing locations at airports and at Dover seaport.

• The IPS arrivals survey will be stopped and the overseas expenditure data of Great Britain (GB) residents will be collected through the Great Britain Tourism Survey.

• From 1 April 2024, Northern Ireland (NI) travel and tourism data are being provided by the Northern Ireland Statistics and Research Agency (NISRA).

The Office for National Statistics (ONS) is changing the way data are collected on overseas travel and tourism. From 1 July 2024 we are moving away from the International Passenger Survey (IPS) being our main source of data, and will be using several different surveys and data collection methods.

We are stopping our IPS arrivals survey and transforming our IPS departures survey by harmonising it with the Civil Aviation Authority's (CAA) departing passengers survey. We will be using the Great Britain Tourism Survey to provide data on the overseas expenditure of Great Britain (GB) residents. Changes to how we collect data on those travelling into and out of the UK through Northern Ireland (NI) were implemented on 1 April 2024. All these changes are outlined in this article, along with the overall impact for our users and the next steps in this transformation project.

The first outputs produced using our transformed methods will be for Quarter 3 (July to Sept) 2024 and will be published at the beginning of 2025. Outputs for Quarter 1 (Jan to Mar) and Quarter 2 (Apr to June) of 2024 will be published later this year and will contain data collected and processed using our existing methods. 

These changes follow a statistical review that concluded with our Travel and tourism review: final report in May 2022. This review explains why these changes are required and outlined our vision for a hybrid approach to deliver travel and tourism statistics. We continue to research other areas outlined in the review that are not included in this report, such as alternative data sources and statistical modelling.

The implementation of our new data collection methods follows several years of research and planning by the ONS. Consultation with statistical and travel and tourism industry experts, including an expert panel working group set up by the ONS, has helped to share knowledge and refine our hybrid approach.

Newsletters, focus groups and presentations at national events are some of the ways that we continue to engage with our users. Quarterly project updates are sent out in GovDelivery email updates. Users can subscribe to our mailing list on the GovDelivery website to receive these updates.

Data collection is the first step in the statistical process. While we are ready to start the new sampling and collection methods on 1 July, some of our processing methodology is still being refined and will not be finalised until later this year.

Details about final methods will be shared at the beginning of 2025. This will accompany the first release of estimates using these new data collection procedures. It will explain the new methods in more detail and discuss the improvements they make to our estimates.

The changes we are introducing will make our data collection process more efficient. We expect outputs from our new methods to have greater accuracy and less uncertainty, better coherence with other measures of travel and tourism, and continue to provide relevant and timely data for our users. This was discussed in our May 2022 final report.

Estimates produced using the new approach will cover Quarter 3 2024 and will be published at the beginning of 2025. These statistics will be classified as official statistics in development (compared with our previously published outputs' classification as official statistics). They will be published as research into a new method for producing travel and tourism statistics and we will advise caution when using the data. This classification reflects that the new methods and outputs are of sufficient trustworthiness, quality and value to meet the standards set by the Code of Practice for Statistics, while they are still being developed and evaluated.

Statistics on international travel and tourism are produced by the Office for National Statistics (ONS), using the International Passenger Survey (IPS). The IPS is a regular sample survey of sea, air and Channel Tunnel passengers travelling between the UK and the rest of the world. Statistics on domestic tourism in Great Britain (GB) are produced by VisitEngland, Visit Wales and VisitScotland using the Great Britain Tourism Survey. Equivalent statistics are produced by the Northern Ireland Statistics and Research Agency (NISRA) for Northern Ireland (NI) using their Continuous Household Survey.

Previously, NI was included in the UK estimates of international travel and tourism produced by the ONS using the IPS, with passengers interviewed only at Belfast International Airport. Separately, NISRA compiled statistics based on its own passenger survey, Northern Ireland Passenger Survey (NIPS). Changes to the way that we collect data on visits to and from NI were implemented on 1 April 2024. More details are given in Section 5: Changes to data collection.

Detailed information on the current approach to measuring travel and tourism statistics can be found in Section 2 of our Travel and tourism review: final report, published May 2022.

We will continue to:

• publish regular outputs on the Office for National Statistics (ONS) website as shown on the ONS release calendar

• focus on international travel and tourism, overseas residents in the UK and UK residents abroad

• provide estimates on visits, nights, and spending, cross-tabulated by other characteristics such as "mode of travel" and "purpose of visit"  

• provide expenditure data on the contribution made by tourists and other travellers to the UK balance of payments

• publish a Travelpac of the data; this is a long file containing data on the key characteristics collected on travel to and from the UK

• publish up to date information on our methods and quality indicators for travel and tourism estimates

• regularly consult with topic experts and key users and support ad hoc data requests

The methods used to produce travel and tourism statistics are changing in three fundamental ways.

Firstly, changes to International Passenger Survey (IPS) data collection have already begun with more being introduced on 1 July 2024:   

• IPS departures data collection has been adapted, so it is similar to the Civil Aviation Authority (CAA) departing passenger survey, resulting in changes in the sample selection, size of sample and interviewing practices

• IPS arrivals data collection is being discontinued

• IPS was discontinued in Northern Ireland (NI) as of 1 April 2024

Secondly, the Office for National Statistics (ONS) will use household survey data to gain insights into recent international travel by Great Britain (GB) residents and their related spending while overseas. This is necessary because information is no longer being collected for arrivals into the UK and cannot be fully collected by the IPS departures data collection (as most spending occurs once they are abroad).

Thirdly, since 1 April 2024, travel and tourism data related to NI is being supplied by the Northern Ireland Statistics and Research Agency (NISRA).

Data collection will vary by the type of traveller being counted:

• travel activity of overseas residents and spending in GB will be collected by the harmonised CAA departing passenger survey and IPS departures survey

• international travel activity of GB residents and spending will be measured partly through questions on their departure from the UK on the IPS and CAA surveys and partly through questions about spend abroad via the Great Britain Tourism Survey

• visits abroad by residents of NI, and overseas residents travelling into and out of the UK through NI, will be collected and supplied by NISRA

Overseas residents visiting Great Britain

Data on overseas residents' completed trips to GB will continue to be collected by a passenger departure survey. The current IPS (departures) is being harmonised with the CAA departing passenger survey. This has involved extensive work to align core questions for overseas residents.

The data collected from both the IPS and CAA on departures will be combined by the ONS, to create one larger overseas residents dataset. This is expected to produce more precise estimates of residency, trip destination, trip purpose and expenditure of overseas residents. It will also increase confidence in estimates at sub-national level, an aspect of the statistics requested by users in our 2021 user consultation.   

The location of the IPS interviewers across airports is also changing, as they adopt the operational practices of the CAA survey interviewers. Collection of data through the IPS will move from interviewing passengers immediately after airport security checks to interviewing them at airport departure gates or lounges at smaller airports. Interviewers will move around the airport gates, selecting passengers for interviews according to a new sample design that will select departing flights. Data provided by the Official Airline Guide (OAG) will be used as the sampling frame for both surveys.

In 2023 the ONS ran trials to test aspects of the new design at Heathrow, Gatwick, Stansted, Manchester, Newcastle and Luton airports. The new design saw an improvement in engagement and response rates from departing passengers, with IPS interviewers reporting that respondents seemed more relaxed when interviewed at the gate rooms rather than after security checks.

Travel through our seaports, on Eurostar and the Channel Tunnel, will continue to be collected by the IPS departures survey as before, apart from at Dover. From 1 July onwards IPS interviews at Dover seaport will take place on the quayside rather than on the ferries. This is an operational change to reflect that IPS arrivals interviews are no longer occurring on return trips on these routes.

Visits abroad by residents of Great Britain

From 1 July 2024, the IPS will no longer interview UK residents arriving in the UK, at the completion of their visit abroad. Instead, data from residents of GB will be collected using a new hybrid approach.

Data on GB residents travelling abroad will now be collected, in part, by the departures surveys at airports, seaports and the Channel Tunnel. The month of return to GB for these respondents will be calculated by taking their date of departure and adding the number of nights they report they will spend outside of GB. The overall sample size of the surveys will be four times larger than the current IPS arrivals survey. We anticipate that this will increase precision and confidence in the estimates of GB residents travelling abroad.

An extra module of questions added into the departures surveys will collect most of the key information on GB residents leaving the country. The exception is post-departure expenditure (money spent while abroad). GB residents will know their travel destination and can report on pre-departure expenditure (such as flights and package holiday bookings), but at the departure gates they will be unable to report on spending that has yet to happen.

To capture post-departure expenditure by GB residents, an overseas arrivals module has been added to an online household panel survey. Respondents to the Great Britain Tourism Survey (GBTS) will report on trips taken overseas in the last 12 weeks and post-departure expenditure on these trips. The panel data will be used to derive a statistical model that predicts post-departure expenditure, given the characteristics of the residents and their trip. This model will be used to estimate the expenditure for the IPS departures survey respondents.

The approach for modelling post-departure expenditure is yet to be finalised, however multiple approaches are being considered and developed by the ONS. The final approach will be communicated to users in early 2025 and will include information about the level of quality of the expenditure data.  

Northern Ireland residents and overseas visitors visiting Northern Ireland

Our Travel and tourism review: final report, May 2022 recommended that travel and tourism statistics produced by the ONS and NISRA needed greater comparability and coherence. Following a period of review by the ONS, it was decided that all data on international tourism to and from NI should be provided by NISRA. Consequently, we stopped IPS data collection in NI on 1 April 2024.

NISRA uses four different data sources to compile international travel and tourism statistics for NI. Their Northern Ireland Passenger Survey (NIPS) collects data at all airports and seaports, whereas their Continuous Household Survey includes questions on international trips taken by NI residents. The other two data sources are run by the Central Statistics Office (CSO) of the Republic of Ireland; the Inbound Tourism Survey and the Household Travel Survey. The CSO provide NISRA with data from these surveys on movement across the land border between the two countries.

This change will produce greater consistency between NISRA's travel and tourism statistics and those used within the UK figures produced by the ONS. We also expect the quality of the data on international travel activity in NI to increase. Previously, the ONS relied on a small sample of data collected by the IPS at Belfast International Airport. The combined larger sample size of the surveys used by NISRA, as well as better representation of travel activity across the land border and at seaports, is likely to improve the precision and confidence of the statistics.   

NISRA only publishes tourism data once all four data sources for the reference period have been collected and aggregated. This has a knock-on effect for the data delivered to ONS, in that NISRA data for the quarter will not be available until five months after the end of the reference period. This will be too late for us to include this data in our next quarterly publication.

Following feedback received from our key stakeholders, the decision has been made to continue our quarterly publications on our current timescales, but to only produce estimates for GB. Estimates for the UK and NI will be published annually, the next being in our Travel Trends 2024 release, which is scheduled to publish in May 2025.

The timeliness of the data delivery from NISRA may improve in the future and fit in better with our publishing timetables. The ONS will therefore review this decision in 2025.

There are further aspects of the methodology for producing international travel and tourism estimates that are still being developed. These include:

• weighting 

• imputation 

• Reproducible Analytical Pipelines (RAP)    

• time series discontinuity analysis

Weighting and Imputation

The International Passenger Survey (IPS) currently uses a multi-stage weighting approach to account for aspects such as sample design, non-response, and known imbalances. The data are also calibrated to airport traffic totals using administrative data provided by the Civil Aviation Authority (CAA), Department for Transport (DfT), and Eurostar. Any missing data are imputed using values present in the data and an uplift index is calculated for London, to account for tourists tending to spend more there than elsewhere in the UK.  

Weighting and imputation for the improved IPS are being developed, ready for the first publication of estimates under the new design. Aspects of the old approaches will be retained, while others will need to be revised to account for the change in method. Details about the finalised weighting and imputation approaches will be shared with users in early 2025. 

Reproducible Analytical Pipelines

Reproducible Analytical Pipelines (RAP) are being introduced to our processing of travel and tourism data, to ensure our analysis is of high quality, reproducible, and sustainable. We are transforming the end-to-end data production process, from the data arriving at the Office for National Statistics (ONS) to the publication of estimates on the ONS website. Estimates from the new system will be produced more efficiently with less potential for human error, and this should reduce the time taken to prepare outputs for publication. More information can be found on our Analysis function RAP 2023 implementation plan web page.

Further information on the quality of the new outputs will be published later this year in an updated version of our Quality Methods and Information document (QMI), along with our publication schedule for Quarter 3 (July to Sept) 2024 and 2024 annual data.

Time series analysis for measuring discontinuity 

Data that are collected over time form a time series that allows for the historical comparison of estimates. Discontinuity, or a break in the series, is a risk when underlying data collection methods are changed. This was highlighted as a concern in our 2021 user consultation by key stakeholders who use the trends over time in their analysis of the latest reference period.

In response to these concerns, the ONS will be using time series analysis to provide users with assurance and confidence in the outputs from the new methods. The approach to deal with unexpected discontinuity and how much of it might be attributable to the new methods is still to be finalised. We plan to implement measured interventions to ensure IPS data users can analyse historical trends and perform comparisons with confidence.

Preliminary results of the time series analysis for core outputs, and how the new data compares with historical series, will be published alongside the first new outputs in early 2025.

Future publications cycle

We will carry on publishing regular outputs on overseas travel and tourism and intend to publish the first outputs using our new methods in early 2025. Over the summer of 2024, we will review our publication cycle and the range of travel and tourism outputs currently produced. We will consider the timeliness of outputs, the level of detail that we can confidently include in the data tables and the availability of the new processed data to feed into monthly and quarterly outputs. We will continue to engage with our main stakeholders to confirm their priorities regarding the published estimates and feed this into the decision-making process.

If you would like to stay updated with our ongoing travel and tourism work, and receive our regular project updates, you can subscribe to our mailing list on the GovDelivery website. For further information, or to feedback any views on this article, please email travel.and.tourism@ons.gov.uk.