Overview: Tourism Statistics Quarterly and Annual Publication QMI

Scope and Purpose

This data quality statement outlines the methodologies, quality assurance measures, and user guidance for the collection and dissemination of Tourism statistics by the Northern Ireland Statistics and Research Agency (NISRA). The report provides a comprehensive overview of how NISRA manages and reports tourism statistics along the dimensions outlined in the Quality Assurance Framework of the European Statistical System PDF (916KB) to support informed decision-making and public transparency. These statistics inform the NI Tourism Quarterly and Annual publication.

The Annual report comprises several data source components which provide a full picture of Tourism overnight trips, day visits and associated expenditure.

Detailed quality information for each data source is shown below,focusing on relevant dimensions that ensure the integrity and reliability of the publication in accordance with established standards.

Quality information from external sources such as the Central Statistics Office (CSO) and various administrative sources have also been detailed along these dimensions with links to the producers quality documentation cited where appropriate.

Notes for users:

  • The latest statistical report and associated tables are available here.
  • Caution should be taken when interpreting these data. Data are not directly comparable to previous NI Annual Tourism statistics due to methodological changes. Additionally, data are not adjusted for inflation.
  • The statistics are for the calendar year 2023 and mark the return to regular scheduling of NI Annual tourism statistics since the Covid 19 pandemic. The previous NI Annual Tourism figures were published for the 2019 calendar year in October 2020.
  • This break in series was due to the unprecedented impact of the Covid 19 pandemic on the collection of Tourism Statistics. During this period, in the absence of regular published data, NISRA produced an ‘Alternative Source’ document containing a wide range of tourism related statistics produced by both NISRA and other sources.
  • NI Annual Tourism Statistics report provides a variety of measures including estimates on the number of overnight trips in NI, the number of nights these overnight trips equate to and the associated expenditure including staying in local accommodation such as self-catering, hotels and other establishments.
  • The Annual Tourism Statistics will be of interest to tourism policy makers, public bodies, the tourism business community, economic commentators, academics, and the general public with an interest in Tourism and the local economy.
  • The next scheduled release of the NI Annual Tourism Statistics published on the NISRA website Spring 2025.The next scheduled release of NI Quarterly Tourism Statistics is October 2024.
  • The Northern Ireland Statistics and Research Agency wishes to thank the participating households and Ports for their co-operation in agreeing to take part in the surveys and for facilitating the collection of the relevant data. Thanks also to Central Statistics Office Ireland for the supply of CSO data.

Day trip data - New output for Q1 2024

Data on day trips were gathered from the Continuous Household Survey (CHS), which used a systematic random sampling technique. This sampling method ensures that each household had an equal chance of selection and data is representative of the Northern Ireland population. Data were collected continuously throughout the survey period April 2023- March 24. Day trips questions and definitions within the Continuous Household Survey have been harmonised with those in the CSO Household Travel Survey to allow coherence and comparability.

A “day trip” was defined as any trip taken by a household member that: Lasts less than 24 hours. A key aspect of the day trip is that the trip does not span overnight. It starts and ends on the same day. Involves travel away from home. The individual travels away from their usual place of residence for various purposes such a leisure, work or other activities.

Confidence intervals have been calculated for day trip data indicate the range within which the true population value is expected to lie, given the sample data.

For day trips, the confidence intervals represent a 95% confidence level, meaning there is a 95% probability that the actual population values fall within the intervals presented. These intervals allow users to assess the precision of the estimates and account for potential sampling variability.

The confidence intervals for Q2 2023 – Q1 2024 are as follows:

  • Trips: +/- 19%
  • Expenditure: +/- 37%

Overall Assessment of Quality

Relevance

The degree to which the statistical product meets user needs in both coverage and content.

The primary purpose of the quarterly and annual tourism statistics publications is to measure tourism in Northern Ireland, supporting the monitoring of the Tourism Strategy NI and the contribution of Tourism industries to the economy in Northern Ireland.

Strengths and Limitations: The publication consolidates a wide range of sources to illustrate tourism trends. Surveys are subject to sampling and non-sampling errors which should be considered when interpreting results. It should be noted that, due to Northern Ireland’s geography, visitors may exit through the Republic of Ireland, necessitating reliance on external (CSO) surveys which may be limiting in terms of alignment with specific organisational or contextual needs, yet is necessary for inclusion to achieve a complete picture of NI Tourism trends. The methodology and data collection processes of all external sources are detailed where appropriate in this document.

Accuracy and Reliability

The proximity between an estimate and the unknown true value.

The Tourism Annual is comprised almost exclusively by survey data and therefore survey design and sampling are key areas which impacts data reliability and accuracy of output. Below are the common types of errors impacting Tourism survey results.

Non-Sampling Error: Non-sampling errors include systematic and random errors. Efforts to minimize these errors include survey piloting, consistent interviewer training, and data weighting for non-response.

Sampling Error: Sampling error arises because estimates are based on samples rather than entire populations. For instance, a sample of those taking trips in Northern Ireland is used instead of interviewing everyone.

Total Error: Total error includes both non-sampling error and sampling error, representing the difference between sample estimates and true population values.

Revisions Based on Errors: Significant errors prompt revisions of the overview report and datasets, following the Code of Practice for Official Statistics. Further information on revisions is available on the NISRA website.

Users should note that some variables within the tourism statistics series are more error prone than others. For example, the estimate of the number of trips taken has more precision than expenditure figures. NISRA provides confidence intervals where appropriate to indicate the precision of the estimates, providing a range within which the true value is likely to fall, thereby giving an indication of the reliability of these statistics.

Timeliness and Punctuality

Timeliness refers to the time gap between publication and the reference period. Punctuality refers to the gap between planned and actual publication dates.

This publication is Quarterly and Annual.

Publication Delays and Methodological Considerations: The Tourism Industry was significantly impacted by the Covid 19 Pandemic, which also affected the collection of relevant Tourism Statistics in 2020/21 and 2021/22. Airport closures and subsequent social distancing resulted in a complete halt to inbound tourism data collection at airports. Social distancing regulations also shifted the data collection mode for Domestic Tourism (e.g. NISRAs Continuous household survey changed mode from face to face to telephone which, coupled with the drastic changes in travel behaviour, significantly sample size for 2020/21 and 2022. To maintain continuity during the pandemic, NISRA published alternative data sources publication to provide indication on happening in the industry Tourism Statistics Alternative Data Sources.

During this period, NISRA continued production of;

  • monthly Hotel occupancy statistics (NISRA_Hotel_Occupancy),
  • small service occupancy (NISRA_Small_service_Occupancy),
  • annual self catering survey (NISRA_Self_catering-Occupancy).

In December 2023, NISRA published estimates for 2022 (NISRA_Domestic_Tourism).

Recently (March 2024) NISRA also published a Tourism Satellite Account for NI providing an internationally recognised measure on the impact of tourism to the NI economy (NISRA_Tourism_Satellite_Account).

This annual publication, relates to 2023 and marks the first year following the pandemic where achieved annual sample size is akin to pre pandemic levels.

Since 2019, several methodological changes were implemented in data source components that contribute to the Annual report. Notably, the Central Statistics Office (CSO) revised the Household Travel Survey (HTS), and, following user feedback, the source for estimating external tourism moved from the Central Statistic Office’s Survey of Overseas Travellers to the CSO Inbound Tourism passenger survey. These combined changes have caused unprecedented delays in the regular reporting schedule. This year’s annual publication signifies a return to the standard reporting timeline. [See Comparability and Coherance for further information.]

Under normal circumstances, any delays in the release schedule are due to the requirement for additional quality assurance checks and are communicated to key stakeholders and announced online. All changes and delays are also explained in the publications documentation, adhering to the Code of Practice for Official Statistics. Updates are also made available on the GOV.UK release calendar. The tourism statistics revisions policy is accessible on the NISRA website.

Accessibility and Clarity

Accessibility is the ease with which users are able to access the data, also reflecting the format in which the data are available and the availability of supporting information. Clarity refers to the quality and sufficiency of the metadata, illustrations and accompanying advice.

Tourism statistics are available on the NISRA website in various formats, including CSV files and pivot tables. This year the report has been modernised and published in an accessible HTML report with interactive charts and data download options. The HTML has been developed to comply with guidance from World Wide Web Consortium (W3C), and specifically, their Web Accessibility Initiative and with UK Statistics Authority guidance as well as NISRA specific standards. This reporting year, all data has also been made available on the NISRA data portal.

Additional data requests can be directed to tourismstatistics@nisra.gov.uk.

Assessment of User Needs and Perceptions

The processes for finding out about users and uses, and their views on the statistical products.

Key users of this report include:

  • The Department for the Economy (DfE)
  • Tourism Northern Ireland (TNI)
  • Tourism Ireland (TIL)
  • Fáilte Ireland (FI)
  • Local councils, academics, students, researchers, consultants, the media, business organisations, and the general public.

Consultation is carried out regularly with key users to ensure the publication covers the main requirements for users. NISRA regularly and consistently engages with users to identify and address gaps between user needs and the tourism statistics provided. Feedback is solicited through various channels, including consultations and user group meetings, written engagement activities. Feedback is also invited through tourism publications and social media channels. Additionally, NISRA continually enhance user connections and networks by participating in tourism industry conferences and events. The 2023 publication has been updated and modernised into a HTML format in light of feedback and in order to comply with the Code of Practice’s requirement for continual improvement. Users have expressed a need for geographical breakdown of LGD figures which usually is published in a separate report in the quarter following the annual release. Due to the delay in the Annual reporting schedule, LDG data has been included in this Annual publication to ensure user need is met in a timely manner. Owing to intense organisational resource demands and following a detailed user outputs consultation review, the Visitor Attraction survey was paused and does not feature in the 2023 Annual Tourism Report.

Contact details for the responsible statisticians have been provided within the main report. Feedback can also be provided via this online user survey.

Coherence and Comparability

Coherence is the degree to which data that are derived from different sources or methods, but refer to the same topic, are similar. Comparability is the degree to which data can be compared over time and domain.

Tourism statistics in Northern Ireland have been recorded since 1959. In 2010, NISRA assumed responsibility for these statistics from the Northern Ireland Tourist Board. Improvements in collection and survey methods have been ongoing. Comparisons are made with similar data from the Republic of Ireland, the United Kingdom, and private occupancy surveys. Where possible all questions are harmonised (e.g. The CHS in 2023 marked the first year of complete module harmonisation with CSO’s Household Travel Survey).

The annual is comprised of a range of data sources, relating to inbound and domestic tourism visits and expenditure. It is necessary to include ROI tourism statistics which have NI inflows in order to ensure a complete picture of tourism in NI can be gauged. Since 2019, several methodological changes were implemented in the key components that contribute to the Annual report. Notably, the Central Statistics Office (CSO) revised the Household Travel Survey (HTS), and, following user feedback, the source for estimating external tourism moved from the Central Statistic Office’s Survey of Overseas Travellers to the CSO Inbound Tourism passenger survey. The day trip Tourism modules in CHS were also harmonised with CSO to improve comparability. Users should take these changes in into consideration and care should be taken when comparing these results to findings published prior to 2019. Due to the reasons outlined, this publication reports on 2023 data as a standalone publication and represents a break in series. Where appropriate, a dotted line has been added to time series data to highlight any key changes in methodology since 2011 and specifically between the previous (2019) and current (2023) reporting years.

Output Quality Trade-Offs

Trade-offs are the extent to which different aspects of quality are balanced against each other

Geographical breakdowns at LDG level is provided as per user request, however, it was not possible to provide further break down on LDG for some variables as CSO data does not provide data at the statistical geography level ‘LGD’.

Owing to intense organisational resource demands and following a detailed user outputs consultation review, the Visitor Attraction survey was paused and does not feature in the 2023 Annual Tourism Report. The 2023 report focuses on key user outputs:

  • Overnight Trips to NI (trips, nights, spend by Country of Residence/Reason for Visit/LGD)
  • Occupancy (Hotel, Small service accommodation and Self catering)
  • Economic Statistics (inc employee jobs, businesses by LGD)

To meet user demand for timely data, NISRA publishes results quarterly, incorporating any late returns and changes to grossing factors. This results in ongoing revisions as part of the trade-off between timeliness and accuracy. Due to the methodological changes implemented on ROI statistics and noting the expressed need for timely data from our key users, this publication reports on the key 2023 figures but cautions against comparisons with previous years. NISRA plan for a deeper methodological review for future time series reporting as a secondary workstream.

Confidentiality, Transparency, and Security

NISRA follows the Privacy and data confidentiality methods: a Data and Analysis Method Review (DAMR) in the collection and dissemination of these statistics.

Standard disclosure control methodologies ensure no identifiable information is published. NIPS, CHS, and CSO data are received by the statistical teams are anonymised (name/address information are not collected). In addition, the data represent a sample of the population and there is no way to identify the sample selected from the population.

The data are held within NISRA Department of Finance secure record management system, with restricted access and subject to approved record retention protocols. Responses are presented as weighted percentages and grossed figures, and analysis is only carried out if the sample size in the population sub-category is of sufficient size. Therefore, no additional disclosure control is required.

Statistical outputs are only seen by authorised staff prior to their publication and for 2023 Annual, there is no pre-release access on these statistics. The results are presented to prevent the identification of individuals or organisations, following the principles outlined in the Code of Practice for Official Statistics.

As part of the staff induction process and their ongoing personal development plan, staff regularly complete mandatory training on the Code of Practice and managing information

Performance Cost and Response Burden

The effectiveness, efficiency and economy of the statistical output.

The Code of Practice for Official Statistics requires that the cost burden on data suppliers be assessed relative to the benefits arising from the use of statistics. An estimated response burden on data suppliers is reported annually by NISRA at NI Statistical Surveys – Assessment of burden on businesses, households and individuals.

This document will be updated regularly in accordance with the publication of tourism statistics reports.

This document highlights all quality data available for tourism statistics in constructing the Northern Ireland Quarterly and Annual Tourism publications. Despite some limitations, these sources, taken together, are deemed appropriate for their intended use of measuring tourism progress, to inform marketing and for assessing the value of Tourism related expenditure in the NI Economy.

Detailed Background Quality and Methodology Information of Data Sources

The quarterly and annual tourism statistics publications are based on combination of data sources in order to build a complete picture of Tourism. Each of these sources are surveys that have different questionnaires and sampling modes, covering external tourism in NI (using ROI sources to provide a complete picture) and domestic tourism in NI. All data sources used in the Tourism Annual are prepared using the GSBPM process model.

Quality and methods information unique to each data source is detailed below. Further links to quality and methodology documentation (i.e. questionnaire content, survey design, sampling, exclusions, data validation methods, outliers, imputation, weighting, disclosure) are included where appropriate.

Methodological Considerations: Additional sources of data

While every attempt is made to capture information on all routes out of NI, it is recognised that some routes are missed. These include any residents or visitors departing from or arriving on cruise ships. Details on the capacity of cruise ships docking in NI ports are not included in the quarterly and annual reports.

Data Sources

Click on the buttons below to see more information on each data source.

Producer: NISRA

Relevance

Primary Purpose: To provide input to a measure on tourism in Northern Ireland. The NIPS accounts for approximately 30% of the full count of overnight visitors to NI. These are the primary means of monitoring progress of Tourism in NI.

Strengths: The Northern Ireland Passenger Survey (NIPS) has full coverage of all air and sea ports in NI. It also has good coverage of the flights/boats departing from these ports. As the interviews are carried out by trained interviewers face-to-face, the responses are checked at the time in person alongside in built survey validation checks and are of high quality.

Limitations: NIPS is subject to both sampling and non-sampling errors. About 95% of passengers exiting NI through NI ports have a chance of being sampled in the survey. The remainder are travelling at night (interviewing stops at 10pm and begins at 6am). Furthermore, the number of survey interviews on particular routes or for some main reason for visit (such as playing sports or a particular event), are sometimes small and consequently attract higher sampling errors. This applies to visits from countries with low visit numbers.

Due to the nature of Northern Ireland having a border with the Republic of Ireland - visitors from overseas may travel to Northern Ireland through RoI ports. This means that the NIPS itself is not a complete picture of all external visitors to NI. This presents limitations in showing overall trends without adding information from sources outside of NISRA’s control.

Survey Design: The NIPS is based on face-to-face interviews with a sample of passengers travelling via the principal airports and sea ports. Travellers passing through passport control as they leave NI are selected for interview. All interviews are voluntary and confidential.

Questionnaire: The NIPS questionnaire PDF 380KB is conducted via Computer Assisted Personal Interviewing (CAPI).

Short Questionnaire: Every fifth passenger sampled on a shift is required to fill in a ‘short questionnaire’ (i.e. questions on country of residence, number of overnights spent in NI, number of overnights spent in RoI, reason for visit to NI and/or RoI and total expenditure in NI and/or RoI).

Long Questionnaire: Every 20th passenger is asked the same ‘short questionnaire’ questions and additional questions regarding their trip to NI and/or RoI.

Sampling: The NIPS sample is stratified to ensure it is representative by mode of travel, port or route, and time of day. The NIPS uses a multi-stage sampling design. In the absence of a sampling frame of travellers, time periods at selected ports and routes are chosen at the first stage and travellers are then systematically selected at fixed intervals from a random start within these interviewing shifts or crossings at the second stage.

For more information see sample sizes and methodology.

Inclusions: The NIPS questionnaire is asked to every 5th (short) or 20th (long) passenger (for example it includes all children/babies etc)

Exclusions: The following groups are excluded from the Northern Ireland Passenger Survey

  1. passengers passing through passport control before 6am or after 10pm. Currently this would affect one boat from Belfast Port, one boat from Larne; four Flights from Belfast International Airport and two from Belfast City Airport.

  2. Overseas residents passing through NI en route to other destinations, but who do not stay overnight (it is estimated this does not affect very many passengers).

Data Validation methods: Numerous checks are built into the Computer Assisted Personal Interviewing (CAPI) program which acts as the first form of data validation and thus reducing the number of errors.

On return to the office, a comprehensive suite of validation and edit checks are carried out to clarify (and correct where necessary) any outstanding issues with the data. These include:

  1. checking zero spend, for example from the completed questionnaire

  2. checking high spend to ensure this has been coded correctly

  3. coding towns, countries, etc not included on the interviewers’ coding frames

  4. internal inconsistencies that have been identified, or flagged by the interviewer

Staff in both Central Survey Unit and Tourism Statistics Branch carry out validation check on the NIPS

Response outcome categories. There are 3 response outcomes (complete, partial and minimum) and non-response

  1. complete interview - all the questions applicable to respondent are answered

  2. partial interview - core questions are answered but responses to other questions are imputed

  3. minimum interview - insufficient information is obtained, but valid answers have been provided for nationality and residency to include in the data

  4. Non response - (a) where an interviewer attempts to make contact with the respondent but an interview is not achieved (e.g. Refusals, person was speaking on mobile phone) and (b) non-response due to no interviewers being available to carry out interview (also known as a ‘click’).

Outliers: Expenditure outliers for expenditure are identified based on the data distribution, the number of people in the expenditure group and average daily spend by reason for visit. Checks are built into the survey, and if daily expenditure based on reason for visit is above or below a set value based on previous years analysis; the respondent will be asked to explain the reason for high or low expenditure. This is then taken into account by Tourism Statistics Branch during quality assurance checks; and action taken if necessary i.e. if any one case has an impact of greater than 1% on the total expenditure value. These outliers are either weighted by 1 for expenditure only; or removed and the amounts inputted using a mean-value class imputation.

Imputation: Where the responses for the key items of interest are missing, the values are imputed on a topic by topic basis where the known values are similar to other records. Missing expenditure values are imputed based on known vaules and amounts inputted using a mean-value class imputation.

Weighting and Grossing: Once responses have been collected, the survey data are weighted to produce estimates. The method involves uplifting the data at each port to match the actual number of passengers passing through that port within the reporting period. Passenger traffic statistics are provided by the Civil Aviation Authority (CAA) and Stenaline and P&O ferries. NIPS weighting incorporates a number of stages to address different data accuracy requirements.

The weighting and grossing process can be accessed on the NI Passenger Survey - survey and analysis methodology Web page

While the Northern Ireland Passenger Survey collects information on visitors to Northern Ireland, it is recognised that this is an incomplete picture of those who visit. Other visitors may choose to travel through a port in the Republic of Ireland and travel over the border. To get the complete picture of all who visit NI, the results from the NIPS need to be combined with other sources.

Accuracy and Reliability

The NIPS sample design is complex. The design incorporates both clustering (which typically increases sampling error) and stratification (which reduces it). Examples of clustering employed on NIPS are the selection of the number of shifts at the different ports. Shifts are stratified so that all days of the week are equally represented by port (within the airports - the days of the week are split into ‘AM and PM’ shifts.)

Error: The total error is the difference between the estimate derived from the samples and the true population values. The errors tend to be made up of two types: non-sampling error and sampling error.

Non sampling Error: Sampling error (the difference between the estimates derived from the sample and the true population) arises because the variable estimates are based on samples rather than censuses. That is that a sample of those taking trips in NI as opposed to interviewing everyone.

Standard errors are calculated for the main NIPS variables and further information on the standard errors by main markets can be accessed on the Northern Ireland Passenger Survey Methodology PDF 193KB.

Timeliness/ punctuality

Northern Ireland Passenger Survey (i.e. All visitors who exit through NI ports) is published on an Annual basis and contributes on a Quarterly basis to the quarterly tourism statistics publication.

Date of previous Publication: Latest data for 2019 were published on 22nd October 2020. The lag was due to unprecedented impact of the Covid 19 pandemic on the collection of Tourism Statistics which were replaced in NI for most of this period with “Alternative Sources”. This publication marks the return to a regular publishing schedule.

Quarterly: Information from the NIPS is available approximately 12 weeks following the end of the period. However, Tourism Statistics Branch combines these data with a number of sources to create the quarterly publication and the publication is approximately 16 weeks after the end of the reporting quarter.

Annual: Annual publication contains more accurate and detailed estimates relating to the latest year, it is published approximately 5 and half months after the end of the year.

Coherence and Comparability

Tourism Statistics in Northern Ireland have been recorded from 1959 by Tourism Northern Ireland (formerly Northern Ireland Tourist Board). Over the period, improvements have been conducted regularly whether through collection methods, survey methods or additional sources.

In 2010, responsibility for the collection and production of Tourism Statistics in Northern Ireland was passed to Northern Ireland Statistics and Research Agency.

At this time, the design was changed in the Northern Ireland Passenger Survey. Click here for more information on the benefits and impacts of this change (PDF 52.4KB).

From April 2024 the Office for National Statistics Travel and Tourism Reform Project officially harmonised with the Northern Ireland Statistics and Research Agency (NISRA), standing down the International Passenger Survey (IPS) data collection in Northern Ireland and instead, will be using data provided by NISRA NIPS going forward. This harmonisation will allow for greater consistency between Northern Ireland’s travel and tourism statistics and those used within the UK figures produced by ONS.

NISRA also compares the patterns of results for the Republic of Ireland and the United Kingdom over the same period. The Tourism Satellite Account (which used data from NIPS) allowed the economic impact of Tourism to be benchmarked globally. Tourism Statistics Branch also compares trends to other sources, for example the Index of Services can give an indicator on how the tourism sector is performing in Northern Ireland.

Further Information on Northern Ireland Passenger Survey.

Producer: NISRA

Relevance

Primary Purpose: To provide input to a measure on domestic tourism in Northern Ireland and contribute to the overall measurement of Tourism in NI.

Strengths: The CHS is a systematic random sample of all households in NI and therefore results are representative of the populations. It is currently the most appropriate method to collect information on domestic tourism (overnight and day trips) trips taken by NI residents and meets key user needs. Due to the COVID-19 pandemic, in March 2020, NISRA suspended all face-to-face interviews, or CAPI (Computer Assisted Personal Interviewing), and the CHS mode moved to telephone interviewing, or CATI (Computer Assisted Telephone Interviewing). All interviews from April 2020 onwards conducted by telephone into a Blaise Questionnaire. Telephone to Blaise software allows quality checks to be carried out at the source and provides opportunities for the interviewer to explain any unknown terminology to the respondent.

Limitations: The CHS is subject to both sampling and non-sampling errors. Users should be aware that the ‘expenditure’ data is based on recall which may mean that it is an undercount. To attempt to address this, the questionnaire has been designed to probe interviewees on items that expenditure may have occurred (for example, accommodation, travel, entertainment).

Implications: As the household reference person (HRP) provides proxy answers on behalf of other members of the household, there is a chance that overnight trips may be recorded inaccurately for others, however all individuals in the household completing the individual section of the CHS check if the number of trips they have made matches the HRP’s answer and any discrepancies can be accounted for.

Survey Design: The survey is a carried out via telephone and completed by all those aged 16 and over in the household. Proxy responses are collected by those who do not complete the survey and children. Due to the nature of household surveys, response rates for younger people (particularly 16-24) and males tend to be lower. While weighting and grossing may help to overcome this problem, there is an element of sampling bias.

Questionnaire: For more information see The CHS questionnaire.

Sampling: The CHS sample is stratified to ensure that households are representative by geographical location within NI. The surveys are carried out continuously through the year, i.e. spread over the 12 months.

A ‘household reference person’ is asked to complete a section of the CHS on the household and then all people aged 16 and over complete an ‘individual’ section of the survey.

The tourism questions on overnight trips were added to the ‘individual’ section of the CHS in April 2010. In April 2015 to increase sample size and capture information on children in households taking overnight trips, the questions were moved to the household section. The household reference person completes the section on behalf of all those who live in the household.

From April 2016, this was further enhanced to gain more information from all members of the household. Individuals in the individual section of the CHS are asked about all trips taken by all members of the household to check this matches the information given by the household reference person. They are then asked to complete the same questions for any trips that did not match with the HRP’s answers.

In 2023, information was captured for 5.4 million overnight trips of which 2.1 million overnight trips were in Northern Ireland.

Exclusions: The following groups are excluded from the Continuous Household Survey -

  1. addresses not within the ‘Pointer list of domestic addresses’. Pointer is the address database for NI maintained by Land & Property Services. More information can be found here

  2. under 16s are not requested to fill in the form. To address this, the collection of data on domestic trips has moved from the individual section to the household section of the questionnaire. As of 2022, children were included in the NI tourism domestic overnight publication.

Data Validation Methods: Numerous checks are built into the Computer Assisted Telephone Interview (CATI) program which acts as the first form of data validation and thus reducing the number of errors.

On return to the office, a comprehensive suite of validation and edit checks are carried out to clarify (and correct where necessary) any outstanding issues with the data. These include:

  1. checking zero spend, for example from the completed questionnaire

  2. checking high spend to ensure this has been coded correctly

  3. coding towns, countries, etc not included on the interviewers’ coding frames

  4. internal inconsistencies that have been identified, or flagged by the interviewer

Staff in both Central Survey Unit and Tourism Statistics Branch carry out validation checks on the CHS.

Outliers: Outliers for overnight trips in Northern Ireland are for records identified as having any of the following: a weighted spend of over 2% of total spend, a record with a weighted number of nights over 2% of total trips or a household with a weighted number of trips over 2% of overall trips. Any record with such outliers will be given a weight of 1.

Imputation: Total expenditure is requested from all respondents and for those taking trips in NI, the expenditure is collected within categories, e.g. accommodation, travel and food & drink. In cases where expenditure is missing or coded 0 when there should be a cost (for example in commercial accommodation, or a trip when the person has been shopping but there is no cost), a mean value is calculated for the quarterly expenditure category and applied to the missing case.

Weighting: Once responses have been collected, the survey data are weighted to produce estimates. The method involves uplifting the number of people who responded to the CHS to the mid-year population estimates. The weighting and grossing is done by age groups (0-15, 16-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75+ and by sex (those aged 0-15 and 16-24 are not split by sex due to smaller sample sizes).

More information on the weighting process.

The number of NI residents staying in hotels can be compared to the occupancy survey results. There are no direct comparisons to overall overnight trips.

Since COVID-19 pandemic, the CHS data has been used to estimate the proportion of trips being taken by NI residents.

Accuracy and Reliability

NI residents Domestic tourism (overnights and day trips) are derived from the Continuous Household Survey (CHS). The CHS is one of the largest continuous surveys carried out in Northern Ireland. The survey is designed and administered by the Central Survey Unit (CSU), NISRA. It is based on a sample of the general population, aged 16 and over, resident in private households and has been running since 1983.

Sample: The CHS is based on a systematic random sample of addresses drawn from the NISRA Address register (NAR) The NAR is developed within NISRA and is primarily based on the Land and Property Services (LPS) Pointer database of domestic addresses. The Pointer address database for Northern Ireland is maintained by Land & Property Services (LPS), with input from Local Councils and Royal Mail. The addresses are sorted by district council and ward, so the sample is effectively stratified geographically. A starting point in the sampling frame is randomly selected and then every nth (fixed interval) is selected by counting through the frame. As the survey is asked to a representative sample of the Northern Ireland population there are no major coverage issues. People living in institutions are excluded.

Response rate: In 2022/23, 4,272 households took part in the survey. When ineligible addresses are discounted from the sampling frame this gives a survey response rate of 51.3%. Within those 4,272 households, a total of 4,960 individuals took part in the survey (1.2 individuals per household). This is higher than the previous 21/22 year overall response rate of 41%. See the CHS technical report PDF 2.0MB.

Due to the COVID-19 pandemic, NISRA suspended all face-to-face interviews, or CAPI (Computer Assisted Personal Interviewing), they were replaced by telephone interviewing, or CATI (Computer Assisted Telephone Interviewing), with all interviews from April 2020 onwards conducted by telephone.

Responses from the 2020/21 reporting year showed that switching to CATI and relying on sampled households to contact CSU negatively influenced response rates. From July 2021, with the lifting of some restrictions, NISRA implemented the ‘knock to nudge method’. This meant that interviewers could once again call at sampled addresses to encourage people to participate. Interviewers were not permitted to enter the property, only to collect contact information to complete the survey using CATI at an agreed time.

Both CAPI and CATI allows plausibility and consistency checks to be incorporated to improve data quality. Data are returned from the field via a Secure File Transfer Portal (SFTP). Data is downloaded onto secure network drives within NIGOV and processed through the CHS data management system. Data are coded and fully validated by CSU statistical staff.

The questionnaire is scripted in Blaise which allows CSU staff to define range and consistency checks for each question and to control routing throughout the questionnaire. The data are subject to further validation checks including treatment of outliers and detailed consistency checking including non-credible checks. Non-response bias and sampling error Non-response.

The CHS aspires to interview everyone aged 16 or over at a selected household (or collect information by proxy), but non-response, either by the household as a whole or by individuals within the household introduce non-response bias to the results. To counter this, data are weighted in order to produce estimates that better represent the population. The adjustment made to any data may be less than or greater than 1, but will generally be reasonably close to 1.

In 2021/22 a chi square goodness-of-fit test showed that the CHS sample was not representative of the population by age and sex when compared with the Population and Migration Estimates for Northern Ireland. To counter this, a range of different weights have been applied to account for how the various sample groups differ from that they represent. Cases have been weighted to adjust for age and sex. This reduces (but does not completely eliminate) error.

Sampling Error: The 95% confidence intervals for each estimate have been included in the data charts and tables which accompany the report. These confidence intervals represent the ranges either side of the CHS estimates which are 95% certain to include the true values for the population.

It is the nature of sampling variability that the smaller the group whose size is being estimated, the (proportionately) less precise that estimate is. Users should be aware of this when interpreting results particularly when looking at sub population or LGD data.

Confidence Interval

Year Trips confidence +/- Expenditure confidence +/-
2019 1.8% 8.4%
2022 2.6% 7.3%
2023 2.6% 7.9%

Timeliness and Punctuality

Quarterly: Information from the CHS is available approximately 6 weeks following the end of the period. Tourism Statistics Branch combines a number of sources to create the quarterly publication and the publication is approximately 16 weeks after the end of the reporting quarter. Due to issues with data quality due to the covid pandemic no quarterly publications have been produced since Quarter 3, 2019. The 2023 Annual Tourism Statistics marks a return to regular publication scheduling.

Annual: Annual publication contains more accurate and detailed estimates relating to the latest year, it is published usually approximately 6 months after the end of the year. Due to issues with data quality and sample sizes no publications were made for 2020 or 2021. 2023 marks the first year following the pandemic where achieved sample size is akin to pre pandemic levels.

Domestic Tourism 2022 (from CHS module) was published in December 2023 and “Alternative data sources” was published throughout the pandemic.

The final component of data required for the 2023 Tourism Annual release was received on 19th June. Purdah restrictions and quality assurance checks meant the report was published on the planned date 29th August, 10 weeks following receipt of data. The publication date was preannounced on the GOV.UK website.

Performance and Respondent Burden

The total cost for the inclusion of a question within the CHS is approximately £4,500. The overall cost of running the CHS cannot be provided as this is commercial-in-confidence

An annual survey control return is completed for the survey however NISRA has undergone a user review of outputs do to resource pressure, the outcome of this was the survey respondent burden survey was paused for 2023.

Further Information

Continuous Household Survey | Northern Ireland Statistics and Research Agency (nisra.gov.uk)

As NISRA currently do not carry out surveys in RoI, to gather information on trips taken by RoI residents in NI, NISRA depends on counterparts in CSO.

The following information details the quality information surrounding CSO Ireland Tourism data that combines with NI data to give a full picture of inbound tourism. The collection of this data provides information to Eurostat and NISRA believes this is currently the most robust method of collection for this.

Producer: CSO Ireland

Relevance

Primary Purpose: The purpose of the Inbound Tourism statistical series is to measure the travel patterns (trips, nights, purpose of trip, type of accommodation, expenditure, etc.) of foreign resident visitors to Ireland.

Scope: All foreign resident visitors departing Ireland on overseas air and sea routes. Foreign resident visitors who depart Ireland via Northern Ireland are not covered. Foreign resident cruise excursionists who visit Ireland are also not covered.

The source relates to Inbound travel patterns of foreign residents visiting Ireland, which involve overnight stays and same-day trips, and associated variables. For more information see Definition of statistical target concepts

The Inbound Tourism series is a new statistical series based on an updated sampling methodology and data collection process. The new methodology adopted by CSO represents a significant improvement on the previous methodology, particularly in terms of the sampling and a move to a fully digitised process, and this produces an enhanced measurement of inbound tourism in Ireland. In line with best practice, the CSO monitors the quality of its methodological processes and outputs of its statistical work. As part of the review process for this series, CSO conducted a range of validation exercises including a review of the raw passenger data by destination and the change to surveying departing passengers only (versus the previously applied process of surveying arriving and departing passengers).

Survey Design: Information is collected via the Passenger Survey, an ongoing survey of departing passengers at Cork, Dublin, Kerry, Knock and Shannon Airports, and at Dublin Port, Europort Rosslare and Ringaskiddy Port.

There are three reasons for the focus on departing passengers (as opposed to arriving passengers):

  1. EU Regulation on tourism statistics specifies that it is the date of departure, rather than the date of arrival, that determines the reference month for tourism flows. Surveying departing passengers aligns with this concept.

  2. The full details of a trip are not always known the moment a visitor first arrives in a country. This is particularly the case in relation to expenditure, where the visitor will not usually know exactly how much he/she will spend on their trip. Surveying departing passengers only, when the full trip details are known, avoids this ambiguity.

  3. From an operational perspective, it is easier to survey departing passengers (whilst waiting at an airport departure gate or waiting to board a vessel) than arriving passengers (who often depart directly the airport or port facilities).

The Passenger Survey is conducted by CSO enumerators and consists of enumerator-led interviews of a sample of departing passengers. The data is captured by the enumerators using of an electronic questionnaire hosted on tablets and mobile devices. The questionnaire captures information on the flight or sailing of the departing passenger, their residency, reason for travel, their length of stay in the country, the accommodation types they used and various expenditure information such as their fare cost, accommodation cost and day-to-day expenditure.

The Passenger Survey was first trialled in Dublin Airport in 2019 then, following a hiatus due to COVID-19, was fully implemented in Dublin Airport in 2022, before being rolled out to the other airports and seaports in early 2023.

Sample Design: The Passenger Survey employs a hybrid sample design. At Dublin Airport, where the main enumeration effort is made, a two-stage cluster sample approach is used. A sample of departing flights (clusters) is randomly selected. Then the enumerators interview a systematic sample of passengers preparing to board those flights. At Cork, Kerry, Knock and Shannon Airports, where flight departures are less frequent, a different approach is used. Enumerators are allocated rotating shifts that represent the full range of flight activity. The enumerators then attempt to interview a sample of passengers on all flights departing in their shifts. At the seaports, where sailings are even less frequent, a different approach again is taken. A systematic sample of departing sailings is selected and the enumerators then attempt to interview as many passengers as possible preparing to board these vessels.

Aggregation: The Inbound Tourism results are compiled by weighting the Passenger Survey returns to total departing airport and seaport passenger numbers reported monthly by the airport authorities and sea ferry operators respectively. There is a separate weighting scheme for each airport and seaport. The weighting scheme for Dublin Airport is the most complex, with passengers surveyed on a specific flight first being weighted to the total number of passengers on that flight then weighted again to represent the total passenger numbers departing to various countries. Non-transfer and transfer passengers are weighted separately. For the other airports, the passengers surveyed departing to the various destination countries are simply weighted to the total passengers departing to those countries. For the seaports, passenger returns departing on each specific ferry route are weighted to the total number of passengers departing on those ferry routes. All individual airport and seaport results are then combined to produce the Inbound Tourism results.

Timeliness and Punctuality:

Monthly: The Inbound Tourism series published May 2024 is a new statistical series based on an updated sampling methodology and data collection process. This new methodology represents a significant improvement on the previous methodology, particularly in terms of the sampling and a move to a fully digitised process, and this produces an enhanced measurement of inbound tourism to ROI, accounting for visitors to NI.

Further documentation relating to methodology and quality and an overview of the process model used for ROI Inbound Tourism can be found at CSO Inbound Tourism

Producer: Central Statistics Office, Ireland (CSO)

Primary Purpose: The primary purpose of the Household Travel Survey (HTS) is to measure domestic tourism of Irish residents within the Republic of Ireland. The survey is run by Central Statistics Office (CSO-Cork). The HTS is used by NISRA provide input to a measure on tourism in Northern Ireland. The HTS accounts for approximately 7% of the full count of overnight visitors to NI. These are the primary means of measuring Tourism in NI and providing monitoring data for Tourism Strategy NI.

Survey Design: The survey is a postal survey, with approximately 50% of all forms posted being returned to CSO for analysis.

As the survey is carried out by CSO, it is outside of NI’s remit. This means Tourism Statistics Branch may query results that may look unusual, but do not have sight of the intricate microdata/weighting methodology. It should be stated that the number of respondents to the large survey who do take a trip in NI are small, so it is expected there are large sampling errors.

Survey Design: The HTS is a paper questionnaire posted to a random sample of households throughout ROI. Respondents are requested to post their form back which includes information on trips taken within the last 4 weeks by everyone in that household.

The survey design changed in 2020 so NI data is collected on an all island basis where in the past it was seen as “outbound” travel.

Sampling: The HTS sample is selected randomly from the Electoral Register and stratified by District Electoral Division.

More information on sampling and methodology can be accessed at this link.

Exclusions: The following groups are excluded from the Household Travel Survey

  1. Residents travelling within the country with the intention of changing their usual residence.

  2. Persons who travel to work temporarily in institutions within the country.

  3. Persons who travel regularly or frequently between neighbouring localities to work or study.

  4. Nomads and persons without fixed residence.

  5. Armed forces on manoeuvre.

Data validation methods: While staff in CSO are responsible for data validation checks carried out on the HTS - NISRA TSB have a strong working relationship with CSO and both teams can inquire about any outliers or unexpected results. .

Imputation: Most missing data is either imputed or estimated. The principle variables imputed are expenditure incurred on the trip and also age. It is assumed that non-responding households have the same patterns as those households that respond. These procedures are carried out using PC SAS.

Weighting: Survey results are weighted to agree with household population estimates broken down by household type and region. These household population estimates are provided by the Quarterly National Household Results.

Quality documents for the HTS and methodological notes relating to this output can be found at CSO Household Travel Survey - Methods and CSO Household Travel Survey Quarter 4 and Year 2023

Accuracy and Reliability

Measurement error is not formally measured in the HTS. However CSO do note that the questionnaire structure may lead to error, they provide contact details for staff who are trained in the survey to aid any confusion. The data is returned by post and scanned in the office. Staff in the office check any missing data and are changed as necessary.

A series of edit programs are run to test for inconsistencies and missing values within the data - for example respondents indicating a trip was taken but no details provided, missing gender/age. Most missing data is imputed or estimates.

Users of the statistics should note that some of the variables within the tourism statistics range are more reliable than others e.g. the number of trips taken is a truer number than the expenditure taken during the trip. Expenditure is based on recall from the passengers whereas they will record the trip and nights. NISRA provide confidence intervals for trips and expenditure to display this. Confidence intervals are not available for the HTS however NISRA use confidence intervals calculated for the CHS and apply these to HTS estimates.

Timeliness/Punctuality

Date of Previous Publication Annual/Quarterly

Coherence and Comparability

The nearest comparisons can be made with the expenditure data from the Tourism and Travel frontier survey. The key item for comparison is that of expenditure by Irish residents while travelling abroad. Tables showing these results can be accessed at this link.

The HTS has been carried out since Q1 2000, however, due to a change in methodology there is a discontinuity in a the series relating to outbound and domestic travel from 2010 onwards and again in 2018. NISRA carry out checks year on year for the numbers of overnight trips to NI.

Assessment of User needs and perceptions

While NISRA may take views from users on the needs from the HTS and feedback to CSO, the survey belongs to CSO. This means that NISRA require information on the number of trips, nights and spend for monitoring NI Tourism trends and can request additional information but are not guaranteed to receive this.

Confidentiality, Transparency and Security

The confidentiality of all information provided by respondents to the CSO is guaranteed by law under the Statistics Act, 1993. All CSO office personnel become ‘Officers of Statistics’ on appointment and are liable to penalties under this Act if they divulge confidential information to any outside person or body. Every precaution is taken to ensure that there are no violations of this principle throughout the survey process. All forms are scanned and verified as soon as they are received in the office and the original forms are then stored in a secure and locked cabinet. Data is published in aggregate form only.

Further information on the Household Travel Survey may be found on the CSO website.

Producer: NISRA

The occupancy survey is a long established survey that gathers information on the number of rooms and beds available and occupied throughout NI hotels and B&Bs/guesthouses/guest accommodation on a monthly basis.

Relevance

Primary Purpose: The primary purpose of the Occupancy Survey is to give an early indicator on how the tourism industry in NI is performing. The occupancy survey provides information on how many rooms or beds in serviced accommodation are full over the number available.

Strengths: The occupancy survey has been ongoing since before NISRA assumed responsibility in 2010 which means that there is a strong time series (available back to 1997). It is a good early indication of how the tourism sector is performing in NI as these statistics are available earlier than the overall figures.

To offer accommodation in NI, establishments must be registered with Tourism Northern Ireland. This means that the stock file which NISRA draws a sample from is complete and reliable. It is an offence to offer accommodation without registering with TNI.

Limitations: In recent years the response rates to the occupancy survey are falling despite efforts by NISRA to improve them. This has led for an online occupancy survey to be developed from January 2023.

The occupancy survey only allows an indicator of how the hotels and B&Bs/guesthouses are performing, but does not account for NI visitors staying with friends or relatives or in alternative accommodation such as self-catering.

Survey Design: The Hotel occupancy survey is a census of all hotels in NI. The B&Bs/guest accommodation and guesthouses are randomly sampled and stratified by LGD and size of the establishment.

The occupancy survey is carried out by collecting data via an online methodology following a survey invite email. NI Direct also carry out follow up calls to those who have not responded in an attempt to raise the response rate. Establishments are also provided without instructions and contact details so the respondent can query anything they are unsure about.

Historically the random sample of Guesthouses, B&Bs and guest accommodation drawn from the stock provided by Tourism NI was extended to include good responders and those with 5+ rooms. Since January 2013, Tourism Statistics Branch has been selecting respondents on a stratified random sample basis larger establishments with 5+ rooms which weren’t randomly chosen during sample selection are also included. Those deemed as good responders (having responded 10+ months in the previous year) are no longer included if not randomly selected. Monitoring of the old and new systems has shown no bias.

Response Rate: The response rate for the hotel sector varies around 45% and of the sampled Guesthouses, B&Bs and Guest accommodation approximately 17% respond.

Users should note that TSB recognise that the response rate is low to the surveys and is working to improve this. NISRA has discussed whether it could make occupancy survey response a legal requirement, but this is in early stages. We have engaged with key users and Tourism Alliances to promote the need to complete the survey. TSB have worked hard to improve the response rate by making reminder calls and by modernising the completion method to online. TSB have discussed the issues with the Northern Ireland Hotel Federation, the NI bed & breakfast forum and with LGD tourism liaison officers and all organisations have encouraged their respective hotel, bed & breakfast, guesthouse and guest accommodation owners to respond to the NISRA survey.

Validation: NISRA carry out sense checks on the data by comparing month on month and year on year data.

NISRA uses an online data collection which allows quality checks at the point of input and should remove human error.

No data is imputed, instead if there is missing data, a further chase up call is completed to try and gain the information. If this is to no avail, the data is weighted to cover non-response.

Data is weighted by the number of bed-spaces, number of rooms and district council. Due to different response rates to different parts of the survey, there are different weights for arrivals, guests and room and bed space occupancy. (Some accommodation providers do not provide information on the country of residence of their arrivals/guests).

Other research companies carry out occupancy surveys through NI. While the results are slightly different to NISRA results - (tend to be higher occupancy rates) the trend remains the same. The reason the results are higher may be due to the sample deign, as these organisations tend to sample larger hotels with higher star ratings, whereas NISRA have a complete census of hotel responses.

Accuracy and Reliability

Error: Within non-sampling error there is systematic and random error. Systematic error occurs when data are biased in a certain direction. This was the case when the methodology of the occupancy survey forced good responders into the sample. Random error is the variation in sample data from the true values of the population which occurs by chance.

Sampling Error: Sampling error (the difference between the estimates derived from the sample and the true population) arises because the variable estimates are based on samples rather than censuses. That is that a sample of those taking trips in NI as opposed to interviewing everyone.

NISRA are working to develop a way to calculate standard errors or a measure to show how accurate the figures are. There are complications in doing this as there would be monthly measures, quarterly, year-to-date and annual figures. The respondents change on a monthly basis.

Sample sizes are provided in the published excel documents. (additional tables).

It is important for users to note that the occupancy surveys depend on an administrative source from Tourism Northern Ireland (stock files). TNI keep a record of all hotels, bed & breakfasts, guest accommodation and guesthouses in Northern Ireland. It is illegal to offer tourist accommodation without registering with TNI, so the record should include all establishments. TSB draw a sample from this stock file at the beginning of the year. However, throughout the year as new establishments open or as others close, TSB update the stock file for grossing up to. Each month, checks are carried out to compare total rooms and beds available by ‘size band by LGD’ of hotels and ‘guesthouses, guest accommodation and bed & breakfasts’. TSB and TNI will check these against each other for any discrepancies.

Further information on the quality assessment of these administrative sources can be found at this link

Timeliness and Punctuality

Monthly: NISRA publish monthly tables on occupancy rates approximately 6 weeks after the end of the reference month.

Quarterly: Overall quarterly figures are also produced at the same time as other estimates in the branch to produce a ‘full picture’ of how tourism is performing.

Annual: Annual publication contains more accurate and detailed estimates relating to the latest year, it is published approximately 5 and half months after the end of the year.

Coherence and Comparability

The Occupancy Surveys collect the same information as other occupancy surveys carried out by private companies. NISRA compares the results between them and while the actual rates are different - the trend is the same. NISRA assessed that results may be higher may be due to the sample deign, as these organisations tend to sample larger hotels with higher star ratings, whereas NISRA have a complete census of hotel responses.

Occupancy Surveys in Northern Ireland have been carried out since 1997 and therefore the results for Hotels can be compared month on month and year on year for a long timeframe. Due to a change in sampling methodology for the GHBB sector comparable results are only available for GHBB back to 2013.

Further information on the quality assessment of these administrative sources can be found at Tourism statistics data quality - administrative sources

Producer: NISRA

Relevance

Primary Purposes: The primary purpose of the Self-catering Survey is to give an indicator on how the self-catering sector in NI is performing. It provides information on how many units are available and occupied throughout the year.

Strengths: By collecting information on the self-catering industry there is a good complete picture of how the sector is performing.

The self-catering sector makes up a large part of accommodation available to visitors in certain parts of NI and often the sector performs better than B&Bs etc as a different market will be interested in this sector.

Limitations: The response rate is poor. This is not a complete picture of how tourism is performing in NI. The self-catering responses need to be looked at alongside the other accommodation.

Survey Design: The Self Catering Survey is issued by questionnaire electronically by email. 2023 marked the first year this was issued via Blaise software and was fully online in order to make the user experience as efficient as possible. Guidelines for completion and contact details are issued alongside the questionnaire for anyone who has queries or requires assistance to fill out the form.

The self-catering occupancy survey is a long established survey that gathers information on the number of units available and occupied throughout NI self-catering establishments on an annual basis.

The self-catering survey is carried out in collection of data through an online survey conducted on Blaise software. TSB also carries out chase up calls to those who have not responded in an attempt to raise the response rate. The survey is sent alongside instructions and contact details so the respondent can raise queries.

Sampling: All self-catering establishments are sent the survey form to take part in the survey (a census of self catering establishments).

Response Rate: The response rate for the self-catering survey is approximately 25%. Responding sample sizes are included within the annual tables online to give an indication of how reliable the occupancy statistics may be.

Data Validation Methods: NISRA carry out sense checks on the data by comparing year on year data.

Imputation: No data is imputed, instead if there is missing data the results are grossed up for non-response.

Weighting Data: is weighted by size of establishment and district council.

Annual: Published annually and contains estimates relating to the latest year, it is published approximately 5 and half months after the end of the year.

Accuracy and Reliability

The self catering data is collected via an online survey. In previous years it was conducted via post and email. The change in mode is for ease of completion for respondents and demonstrated a continuous commitment to improve and modernise data collection, reduce error and lift response rates.

Error: The total error is the difference between the estimate derived from the samples and the true population values. The errors tend to be made up of 2 types: non-sampling error and sampling error.

Non Sampling Error: Within non-sampling error there is systematic. Systematic error occurs when data are biased in a certain direction.

Sampling Error: The self catering occupancy is a census; however it has a low response rate and data is weighted to represent establishment size by district council; therefore allowing room for error.

NISRA are working to develop a way to calculate standard errors or a measure to show how accurate the figures are. There are complications in doing this for the self catering annual figures

There are no other self-catering surveys to compare the results against. However, Tourism Statistics Branch can compare to the overall tourism statistics range collected in the other surveys.

Annual Self Catering Occupancy Surveys in Northern Ireland have been carried out since 2004 and therefore the results can be compared year on year for a long timeframe (Seasonal Self Catering Occupancy Surveys were carried out between 1997 and 2003).

It is important for users to note that the self catering occupancy survey depends on an administrative source from Tourism Northern Ireland (stock files). TNI keep a record of all hotels, bed & breakfasts, guest accommodation and guesthouses in Northern Ireland. It is illegal to offer tourist accommodation without registering with TNI, so the record should include all establishments. TSB draw a sample from this stock file at the beginning of the year. Each year, checks are carried out to compare the numbers of self-catering establishments by LGD to see if any significant changes that look incorrect. TSB and TNI will check these against each other for any discrepancies.

Further information on the quality assessment of these administrative sources and the self catering survey can be found onTourism statistics data quality

COVID Impact on Tourism Industry and Tourism Statistics within NI

When COVID-19 caused a stop in face to face interviewing across NI, the result was that Tourism Statistics could not be produced.

The Continuous Household Survey (collecting data on NI residents overnight trips) moved to collection in a different format - data was collected through telephone calls rather than face to face interviewing. It is recognised that this could impact on the quality of the results as the survey was shorter to facilitate a telephone call interview.

However, on the other side it is unlikely there were as many trips being taken over this time as COVID restrictions did not allow them. When making telephone calls, response rates dropped so NISRA moved towards a “knock to nudge” process to increase them again (interviewers would phone from outside people’s houses to encourage a response). However, response rates were not back to the same levels as face to face interviewing.

The passenger surveys at ports in NI returned in October 2021, however COVID restrictions continued to impact responses due to mask wearing (interviewers did not have to interview someone without a mask), and some people were uncomfortable doing face to face interviewing. There were also issues as interviewing at ferries includes the interviewers getting on coaches and due to the enclosed spaces during COVID these were not seen as safe. Interviews at ports in Republic of Ireland recommenced in August 2022, returning at a slower pace with interviews commencing at larger ports first rather than all ports at once. Interviews were back to pre pandemic schedules in month year)

During COVID-19, NISRA explored alternative sources of data and have continued to publish these in a document for users to explore. These data use sources from other parts of NISRA to show trends in tourism from sources not for this purpose, for example the Nation Brands Index, economic sources and during COVID-19 the COVID opinion survey.

NISRA TSB surveys on occupancy in hotels and small service accommodation continued throughout the pandemic, but it was recognised that response rates dropped due to furloughed staff or properties that were empty with no one there to answer calls, etc.

Contact Details

Published by: Tourism Statistics, NISRA

Lead Statistician: Joanne Henderson

Telephone: 028 90 255163

Email: tourismstatistics@nisra.gov.uk

Accessibility contact

Please contact Dissemination Branch for assistance with accessibility requirements or alternative formats. Contact details are:

Email:

Telephone: +44 (0)300 200 7836

Dissemination Branch
NISRA
Colby House
Stranmillis Court
BELFAST
BT9 5RR