All individuals residing in Italy as far as participation in tourism is concerned.

All tourism trips of at least one overnight stay and all same-day visits abroad, made by the resident population outside the usual environment. Data refer to the population aged 15 and over even if all aged individuals are sampled.

• Since 1997 both for series on participation in tourism for personal purposes and for series on tourism trips.
• Since 2014 for series on same day visits abroad (every 3 years).

Different measures are applied to guarantee confidentiality.

1. The microdata exchange between Istat and private enterprises in charge of data collection, is made by Hypertext Transfer Protocol Secure (HTTPS) by using password for the access.
2. Identificative variables are stored separately as soon as the data collection ends.
3. Aggregated data and microdata are disseminated only after having made treatments such as the aggregation of disclosive information, cell suppression, analysis of single values.

The possibility to make personalised analyses on microdata files can be conducted in the ADELE Laboratory (Analisi di Dati ELEmentari Laboratory), which is a Research data centre (RDC) where researchers and academics are enable to conduct in a secure site their own statistical analyses on microdata produced by the Institute's surveys. Each analysis has to be in compliance with legislation concerning the confidentiality of personal data ("Code of conduct and professional practice applying to processing of personal data for statistical and scientific purposes" - Annex A.4, Legislative Decree no. 196 of 30th June 2003).

Main provisional data of 2014 were disseminated in 2015 (February), just the day before an important public event in the tourism sector called "International Tourism Exchange", held in Milan yearly. These data are collected in a Istat publication which is called the "Statistics Report" series, with comments and tables.

At national level, aggregated data and definitive microdata (MFR and micro.stat files for researchers and users) are disseminated, for free, by the end of November 2015. In particular, aggregated data contribute in filling in the Istat national data warehouse "I.stat", which is accessible on istat web site.

The schedule of statistical release dates at national level is yearly based and is publicly accessible.

The complete yearly press calendar is available on page: http://www.istat.it/en/information/journalists/press-calendar 

A regular report ("Statistical report") of the reference year with provisional data on trips and tourists, by year and quarters, is disseminated to the press and on the web site  on pages http://www.istat.it/it/archivio/148650 (Italian complete version) and  http://www.istat.it/en/archive/148657 (English synthetic version)

In 2015 there are two publication scheduled:

1) the yearly "Statistical report" with provisional data on trips and tourists by year and quarters disseminated on the web site in February (see link in concept 10.1);

2) a joined "Statistical report" with data from supply side and demand side disseminated on the web site in November.

Microdata files are released free of charge and in compliance with the principle of statistical secrecy and protection of personal data. There are 4 types:

• Public use files (mIcro.STAT), downloaded directly from the Istat website;
• Standard file, issued upon request with a valid reason for research purposes;
• Files for research purposes (MFR) issued to subjects belonging to universities or research bodies upon the presentation of a research project;
• Files for Sistan, accessible only by the statistical offices of the National Statistical System.

Standard files are being gradually replaced by MFR. In particular, until 2013, microdata on Tourism demand were disseminated as "standard files"; since 2014, microdata are produced as "mIcro.STAT files" and "Files for research purposes" (MFR files). Files for Sistan are yearly produced.

Through the Laboratory for Elementary Data Analysis (ADELE), researchers and academics can access to microdata files of all Istat surveys. Moreover, further special datasets (linked microdata) are available for access at the Laboratory. Also the access to the Laboratory is free of charge.

A volume on the methodology of the survey has been written in 2003 (http://www3.istat.it/dati/catalogo/20030717_01/, only  in Italian language).

An update version of the volume is going to be published in 2016.

Moreover, when aggregated data and microdata are disseminated, a methodological annex is always attached (the English version referred to 2009 is on page http://www.istat.it/en/archive/12948).

Documentation on methodology of the survey is also collected into the Istat's information system on quality SIQual (http://siqual.istat.it/SIQual/lang.do?language=UK, searching for keyword 'trips' as well on the page http://siqual.istat.it/SIQual/visualizza.do?id=0098401, by clicking on English version). There are information on the characteristics of all Istat's  surveys  as well as on activities performed to guarantee the quality of statistical information.

The main sources of error of the estimates are sampling errors, nonresponse errors (unit and item) and measurement errors.

As regard the first ones, they are assessed by statistical model; the thresholds are calculated for the main estimates (tourists, trips, nights and same day visits) and the main domains (geographical breakdown and main type of trips) and, when necessary, for other estimates.

Nonresponse errors due to units non-response are monitored by indicators in order to prevent and reduce them. This can be reached with actions made before the data collection starts (interviewing technique to avoid refusals, letters to households) and after data collection, by using calibration estimators (this is partially useful to avoid under-coverage errors).

Measurement errors and unit/item non-response errors, which are monitored by indicators and reports, are corrected by control and editing procedures. It is required that each non-responding household be replaced with a household having similar characteristics of the first one, in order to maintain as much as possible the sample representativeness and to minimize the impact of unit non-response. No more than 3 replaces are admitted.

The survey fulfils the geographical comparability because the sample is representative of all regions of the country.

The survey has been implemented in 1997 with the CATI technique. In 2014 CATI has been changed to CAPI. In this occasion other two important changes have been introduced:

• the reference period moved from three months to one;
• the data collection is included into the Household budget survey with a unique questionnaire asked in two different periods for each quarter (so called 'beginning interview' and 'final interview'). 

Despite all these changes, comparability of the series of the main data has been guaranteed over time at national level. In fact, the principal macrodata have been re-built by applying time series re-building methods. However, tables and microdata disseminated by Eurostat until 2013 will not be re-built, according with the Tourism unit of Eurostat. 

In 2001 and 2011, when the Census occurred, data series have been replaced, in order to guarantee the time series comparability. In these occasions microdata and aggregated data already sent to Eurostat are updated.

Data from demand side are compared with other two sources. In particular, data on the outbound flows (overnights and trips) are compared with data from National Bank of Italy, whereas data on domestic overnights spent in rented accommodation are compared with data from Istat supply side on nights spent by resident in Italy in collective accommodation. The comparison between demand side and supply side is limited to that type of collective accommodation which is similar.

First provisional results are generally disseminated in t+40 days from the last day of the reference period t.

Definitive data are transmitted to Eurostat in t+40 days + 4 months and disseminated at national level in t+40 days+8 months.

Yearly provisional results are revised when they are disseminated as definitive. The procedures of editing and imputation aim to obtain definitive estimates, adjusting more in depth single items. Deterministic and probabilistic methods are used.  The definitive estimates of trips, nights and tourists do not change much respect to the provisional ones, whereas there are no provisional results published for expenditure data. 

The municipalities' regitry office from which households are selected, covers the whole reference population. However, data might contain coverage errors as for information such as

1) over-coverage: non-existent or inhabited houses or population no longer living in the country/municipalities (due for instance to recent change of address or wrong inclusions (recent emigration); 

2) under-coverage errors depending on the time lag in registering new residents which lands to missed inclusions (recent immigration).