This case study investigates the use of diverse data sources to improve data quality and strengthen data-driven policies for the tourism sector in Türkiye. This includes the use of transaction data, health tourism data, and international transfer passenger data to strengthen data on tourism spend.
Using diverse data sources to improve official tourism statistics in Türkiye
Abstract
The 12th Development Plan, spanning from 2024 to 2028, defines the comprehensive policy framework for Türkiye, serves as a fundamental guide in harmony with Türkiye's key priorities and aspirations, and provides a strategic roadmap for the nation's development across critical sectors. In the Plan, a significant focus is placed on the advancement of tourism statistics as a part of the overarching tourism policy framework. In accordance with the policies outlined in the 12th Development Plan, the Turkish Statistical Institute (TurkStat) has intensified its efforts to improve tourism statistics. This includes work incorporating diverse data sources into the calculation system to improve the data quality of tourism revenues and strengthen data-driven policies for the tourism sector.
Since 2001, tourism income statistics have been estimated by TurkStat via the Departing Visitors Survey. The survey is conducted periodically at specific border crossings. Tourism statistics are calculated for international visitors departing from Türkiye, Turkish citizens living abroad and Turkish resident citizens traveling abroad.
Several projects have been planned and implemented within the framework of improving tourism statistics. In recent years, significant efforts have been made to leverage methodological innovations and additional data sources to enhance the accuracy of tourism statistics. As part of these efforts, the use of administrative records has been expanded to improve data quality and refine statistical outputs. Additionally, various research initiatives have been planned and executed to integrate demand-side data with supply-side data through comprehensive surveys.
In order to have more accurate estimations, TurkStat has conducted studies on various topics. These studies are: integrating Interbank Card Centre (ICC) data into the system, implementing Health Tourism Research, revising the sampling weight calculations, and including the expenses of international transfer passengers.
ICC is responsible for the operation and management of the card payment system in Türkiye, and therefore has the authority to collect and manage data on most credit card and debit card transactions in Türkiye. Any card transaction that is processed through the Turkish banking system will be part of ICC’s data.
As part of efforts to improve tourism statistics, a Health Tourism Survey was conducted. Health expenditures were collected from the demand side with the Departing Visitors Survey. In order to enhance data quality, it was decided to collect non-resident health expenditures from the supply side of health institutions.
According to the International Recommendations for Tourism Statistics, international transit passengers who do not enter the economic or legal territory of the country of reference (as is typically the case at airports) are not considered visitors. However, while in the international area (where they have entered the country’s economic but not legal territory), passengers can purchase goods from duty-free and non-duty-free shops. Such purchases should be considered tourism expenditure even though the purchaser is not a visitor to the country of reference.
Due to the law on the use of personal data, ICC, health institutions and airport authorities’ shares aggregated data with TurkStat, ensuring that personal data are not collected. The geographical scope of this study is specifically targeted to encompass Türkiye.
In addition to these studies, the calculations for sampling weights have been revised. Sampling weights are values assigned to observations in a dataset to ensure that metrics are representative of the entire population, correcting for sampling bias. The revised sampling weights are calculated based on nationality, type of visitor and the type of border gate exited. The aim of the revised weight calculation is to provide a more accurate measurement method.
Interbank Card Centre Datasets
ICC data are sent to TurkStat monthly using file transfer protocol and uploaded to databases by TurkStat’s IT teams. Afterwards, it is subjected to compliance checks by Tourism Statistics Group experts, such as whether there is a missing table or variable, or whether a newly added Merchant Category Code (MCC) is subject to tourism. In addition, transactions are split to distinguish between expenditure by international card holders who are resident in Türkiye and non-resident international card holders.
This project started after obtaining an agreement between the collaborating institutions, notably the Ministry of Culture and Tourism and the Central Bank of the Republic of Türkiye. The project is a collaborative effort between the TurkStat and the ICC, with ICC participating as the primary data provider. This project is carried out by TurkStat experts, and no funding has been requested or used. Within TurkStat, the initiative is conducted under the Tourism Statistics Group’s guidance. Data transfer is carried out within the scope of a protocol between TurkStat and the Central Bank. This protocol is signed for the long term and is automatically renewed every year.
Health Tourism Research
Supply side data are collected from health institutions via surveys conducted every three months. Each expert from the health institutions is assigned a username and password and is requested to input the required data into TurkStat's data entry program. Completion of the survey is mandatory under the Statistics Law. The data entered via the web questionnaire is analysed by experts from the Tourism Statistics Group. This project was launched in co-ordination with the Ministry of Health and is carried out by TurkStat experts without the need for additional funding.
International transfer passenger expenditure
International transfer passenger expenditure is collected and sent to TurkStat monthly via e-mail by airport authorities. Incoming data are analysed by Tourism Statistics Group experts. The data relates to passengers arriving from international lines and continuing on international lines that spend money in Türkiye's economic zones without entering the legal borders.
This project is a collaborative effort between TurkStat, the Ministry and airport authorities, with airport authorities acting as secondary data providers. The project is carried out by TurkStat experts, and no funds have been requested or used. Data transmission from airport authorities is co-ordinated with the Ministry within the framework of the protocol between TurkStat and the Ministry.
Sampling weights
Sampling weights are derived using passport records from border crossings. Within the scope of the protocol made between the General Directorate of Security (GDS) and the Ministry, the border figures are sent to the Ministry monthly in the defined distinctions (year, month, province name, province code, name of border gate, code of border gate, type of border gate, nationality). These figures are processed by the Ministry in the format requested by TurkStat and forwarded to TurkStat monthly. The data on same-day visitors and overnight visitors is sent directly to TurkStat by GDS via web service. The project is carried out on the border figures of GDS under the protocol between TurkStat and the Ministry. This project is carried out by TurkStat experts, and no funds have been requested or used.
Interbank Card Centre datasets
The ICC transmits data to TurkStat on a monthly basis, providing a table that segregates residence status and aggregates expenditure by sectors, based on the MCC. Data are kept at the level of domestic transactions with domestic cards, international transactions with domestic cards and domestic transactions with international cards, and is available for the origin of the card, MCC and physical/e-commerce distinction. Since ICC data before 2019 did not have sufficient detail, the use of ICC data are valid for 2019 and onwards.
To calculate spend by tourists, expenditure that is not directly related to tourism is excluded from the ICC data based on the MCC. As MCC is not a strict statistical classification, experts from TurkStat worked on converting these codes to tourism expenditure codes. In each period, new MCCs are examined and converting to tourism expenditure codes if required. Expenditure related to shopping and cash withdrawals via bank or credit cards are considered within the scope of tourism. The spend is converted into USD using the monthly exchange rate. The ICC data has been meticulously categorised by country and expenditure type relevant to tourism statistics.
Furthermore, a classification study was conducted to develop residency rules for the card usage data of international card holders residing in Türkiye. Internationally issued cards that have recorded at least five transactions on different days of each month, over a period of more than six months, are classified as resident cards and thus excluded from tourism income.
Statistics are produced from the Departing Visitors Survey and ICC data by country and country groups. The survey data are multiplied with sampling weights and ICC data are multiplied with card usage ratios by country and country groups. A query was incorporated into the Departing Visitor Survey to ascertain credit card usage ratios. The tourism spend is then calcuated by using a hybrid approach that combines the survey and ICC data. For country or country groups where the ICC value is higher than what is implied from the survey, ICC data are used instead of survey data. A comparative review was conducted between the hybrid approach and the Departing Visitors Survey results, which indicate that using the survey results alone may underestimate tourism spend, see Figure 1.
Health Tourism Research
The number of individuals who travelled to Türkiye for health purposes and the amount of their spending were collected through the Departing Visitors Survey. However, the expenditures covered by private insurances and/or other states and the health expenditures of visitors who organised their travels with package tours could not be captured in the existing surveys.
The General Directorate of Health Services and the General Directorate of Health Information Systems, both affiliated with the Ministry of Health, were contacted. The tax identification numbers, legal titles, and signboard titles of health facilities and intermediary institutions authorised by the Ministry of Health were requested, and the relevant information was promptly provided to TurkStat. Those who have received the International Tourism Authorisation Certificate from the Ministry of Health, those registered with a specific code in the Revenue Administration, and those who have a significant share in the service line they provide were included in the scope.
A hybrid approach combining the health expenditures obtained with the Health Tourism Research and Departing Visitors Survey data for health expenditures is included in the system. This method allows for the collection of data on visitors with significant health expenditures, which cannot be captured through traditional surveys, as well as for visitors whose payments are made by their governments or insurance companies, which would otherwise be unrecorded through the standard survey process.
International transfer passenger expenditure
Airport authorities in Türkiye have identified the number of visitors falling under this category through their administrative records. In addition, the average expenditures made in the specified areas were calculated by airport authorities in details and the expenditure values for transfer passengers were obtained. In this context, the statistics are published in three columns as tourism income obtained from visitors, tourism income obtained from transfer passengers and total tourism income, which is the sum of the two values. The results derived from the hybrid method continue to be included in the publications under statistics obtained from visitors. The dataset could be compiled from 2021 and onwards.
Sampling weights
Sampling weights are values assigned to observations in a dataset to ensure that metrics are representative of the entire population, correcting for sampling bias. The population for the Departing Visitors Survey is defined as the number of departing visitors recorded in the GDS passport data, while the sample size corresponds to the number of surveys administered. Calculations are conducted based on national and international requirements as well as available data sources. Prior to the improvement, the weight coefficient was determined based on nationality and visitor type (overnight for all relevant border gates, or same-day only for the relevant sea gate).
Under the new methodology, the sampling weights are derived based nationality, type of border gate (air, rail, road, water), and visitor type (overnight for all relevant border gates, or same-day for the relevant sea and land border gates). This has enhanced the accuracy and quality of representation, leading to improved data and statistical reliability. Additionally, the survey sample for same-day visitors has been expanded to better reflect their population, and a separate weight coefficient has been applied to same-day visitors compared to overnight visitors.
Thousand USD
Source: Turkish Statistical Institute
As the tourism sector is becoming increasingly significant both globally and in Türkiye, it is also increasingly important to measure. However, capturing tourism statistics is challenging due to their distribution across various economic activities. In this context, research and the integration of diverse data sources have become essential to enhance estimation accuracy. Innovative approaches are required in order to use different data sources. Both the widespread use of administrative records and the supply-side compilation of data from providers are critical in terms of improving data quality. To this end, processes have been outlined under four key areas to enhance tourism statistics and boost data quality. These are; incorporating Interbank Card Center data into the system, increasing data quality in health expenditure data, reconsidering the sampling weights and adding international transfer passengers to the tourism statistics.
One of the challenges encountered is the unknown nationality of credit card users. It is assumed that the country of issuance of the card corresponds to the user's country of residence. Therefore, for comparative purposes, transactions are analysed based on the country of residence rather than the nationality as recorded in the border survey. In the updated database, the country details are presented in accordance with the country of residence. A second challenge addressed in this initiative is the inclusion of e-commerce transactions within the data set. E-commerce transactions are only included if the service was received in Türkiye. To effectively identify such transactions, a thorough study was conducted on the relevant MCCs to include.
The ICC data has enabled the collection of expenditure data for countries that were previously unattainable through fieldwork, thereby enhancing the overall accuracy and quality of data. A further strength of the ICC data is that it remains available during for example pandemics, where it may not feasible to conduct surveys. During the COVID-19 pandemic, particularly during the second quarter of 2020 where the Departing Visitors Survey was not conducted, tourism income was calculated using solely the ICC data. A final advantage of working with the ICC data is how it can complement spend information from survey data. Expenditure can be underreported in surveys, as respondents may not remember their exact expenses or miss certain expenses. As such, the ICC data in combination with the surveys data may provide a more complete picture of actual spending.
Exploring additional data sources and ensuring their applicability in statistical processes play a vital role in increasing the quality of data and statistics. In this context; compiling data on the supply side through surveys and increasing the use of administrative records provide important contributions to the improvement of tourism statistics. These efforts enable more accurate and comprehensive tourism data production.
One of the most crucial aspects of this process was ensuring co-ordination between various institutions. While addressing technical issues, fostering communication between institutions or units was prioritised. Collaborations were established with different entities, and inter-institutional protocols were utilised more effectively.
The greatest challenge was encountered in converting administrative records into statistical formats. To address issues such as classification, standardisation, and other technical difficulties, working groups were formed, and opinions or support were obtained from different experts. Given TurkStat’s extensive experience with different administrative sources over the years, the expertise of TurkStat professionals has been instrumental in improving the quality of tourism statistics.
For further information please contact:
Ozgur Volkan Suer, Head of Tourism Statistics Group, Turkish Statistical Institute, volkan.suer@tuik.gov.tr;
Aslı Tasmaz Dadakoglu, Strategy and Budget Expert, General Directorate of Sectors and Public Investments, asli.tasmaz@sbb.gov.tr
This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.
© OECD 2025
Attribution 4.0 International (CC BY 4.0)
This work is made available under the Creative Commons Attribution 4.0 International licence. By using this work, you accept to be bound by the terms of this licence (https://creativecommons.org/licenses/by/4.0/).
Attribution – you must cite the work.
Translations – you must cite the original work, identify changes to the original and add the following text: In the event of any discrepancy between the original work and the translation, only the text of original work should be considered valid.
Adaptations – you must cite the original work and add the following text: This is an adaptation of an original work by the OECD. The opinions expressed and arguments employed in this adaptation should not be reported as representing the official views of the OECD or of its Member countries.
Third-party material – the licence does not apply to third-party material in the work. If using such material, you are responsible for obtaining permission from the third party and for any claims of infringement.
You must not use the OECD logo, visual identity or cover image without express permission or suggest the OECD endorses your use of the work.
Any dispute arising under this licence shall be settled by arbitration in accordance with the Permanent Court of Arbitration (PCA) Arbitration Rules 2012. The seat of arbitration shall be Paris (France). The number of arbitrators shall be one.
This document was produced with the financial support of the European Union. The views expressed herein can in no way be taken to reflect the official opinion of the European Union.