3.9.                  It is important to underline that the terminology, and thus the measurement, associated with the demand side depends on the form of tourism (see para. 2.41 above). In particular, trips and visits (see chap. II, sect. B.3), have different meanings for the different forms of tourism.

3.10.              For inbound tourism (as for all inbound travel), what are usually observed are “trips”: movements of non‑residents across international borders. From a tourism statistics perspective, there is only one notable exception: travel by a person who has entered the country as a non‑visitor but then engages in a secondary tourism activity, e.g., a person arriving in the country to work for a resident business who them takes a tourism trip while in the country, or a foreign diplomat who takes a trip within the country for personal reasons (see para. 2.45 above). If the observation takes a simple count of border crossings, it will not be possible to associate with such an individual all the other trips or visits he or she might have taken during the period of reference. Further, while the statistics might speak of “visitors”, what has actually been observed are “arrivals” (or “tourism trips”) (see chap. II, sect. B.3). Countries are encouraged to use precise and consistent terminology in all publications associated with the dissemination of tourism statistics data.

3.11.              In the case of domestic tourism, if data are collected using a household survey, the trips observed will be “round trips”, making it possible to associate individuals with each of the (round) trips they take during the period of reference. In this case, a distinction will be made between a trip and the person taking the trip, and the characteristics will be assigned unambiguously to either of these observation units.

3.12.              The same approach would apply to outbound tourism data when measured in a household survey. When measured at the border, however, only visits, fractions of outbound or of domestic trips will be observable. Usually, there is no time during border surveys to collect data on the person taking the trip. Although the distinction between “trip” and “person taking a trip” might be subtle, it should be borne in mind when comparing data drawn from different sources.

Box III.1 

Estimation (weighting): example of the Travel Survey of Residents of Canada 

Travel Survey of Residents of Canada estimates are produced based on survey data to which weights are applied, making it possible to inflate those data to agree with Canada’s non‑institutionalized population aged 18 years or over. The weights calculated to produce estimates are person, trip and person‑trip weights. 

The starting point in creating the person weights is the Labour Force Survey sub‑weight. The person weight is then adjusted to reflect (a) the subsampling of rotation groups on the Labour Force Survey, (b) subsampling of people (aged 18 years or over) within a household, (c) non‑response and (d) calibration to known control totals (age/sex groups, census metropolitan area totals). Person records from the second month of collection are also treated for recall bias. 

From the person weight, the person‑trip weight is derived by adjusting for (a) identical trips, (b) the ratio of declared to reported trips, (c) reported trips missing essential data and (d) trip‑level non‑response. Person‑trip weights are also treated for outliers, and records from the second month of collection are adjusted for recall bias. These weights are used to estimate trip volume. 

Finally, the trip weight is derived by dividing the person‑trip weight by the number of adults aged (18 years or over) in the household who accompanied the respondent on the trip. Trip weights are used to estimate expenditures.

_____________________________

Source: Statistics Canada (2013).

In this section: