Page tree
Skip to end of metadata
Go to start of metadata

4.67.                  Domestic tourism expenditure needs to be observed directly from the resident population by use of computer‑assisted telephone interviews (CATI) or other forms of household surveys. These can be designed either as specific operations or as a module of a larger household expenditure survey. The statistical challenges in designing household surveys has been discussed in chapter III (see sects. D.1 and D.2). Most issues covered in the previous section B.3 directly above, are also relevant here.

Box IV.6 

Domestic tourism statistics: example of the Philippines 

For the purposes of domestic tourism statistics, the Philippines conduct the Household Survey on Domestic Visitors (HSDV), which is a rider to the Labour Force Survey. The HSDV is to be conducted every three years as part of an omnibus/comprehensive survey. For the in‑between years, there could be a smaller HSDV survey providing core indicators in the form of a quarterly rider to the Labour Force Survey. The  National Statistical Office is also looking into the possibility of including a tourism module in its Family Income and Expenditure Survey.


Source: Fabian A., E. and Say Y., M. (2007).

4.68.                  Because a (sometimes large) proportion of the sampled households will not include a member who engaged in some form of tourism activity in the reference period, a large number of households must be selected to ensure that there is a net sample providing sufficient information. Since the interviews are conducted after visitors return to their usual environment, leaving the hurry and pressure of travel behind, they may be able to recall their expenditures more easily and more precisely by reviewing bills, invoices and credit‑card receipts (particularly if they receive a letter informing them of the interview in advance). In support of this exercise, it is recommended that visitors be asked to describe their trips, including the different places visited and the activities engaged in, so that expenditures can be connected more easily with each of these components and their addition can thereby yield a figure for total expenditure. 

4.69.                  An additional issue associated with domestic tourism expenditure, as compared with inbound tourism expenditure, is the fact that in many countries, visits of residents to vacation homes are frequent and thus frequently reported. Actual expenditure related to vacation homes and timeshares in the country of residence tends to fall into two broad categories: 

a.  Expenses related to ownership of the vacation home or timeshare that are more or less independent of any specific trip. Public utilities and maintenance fees fall in this category and might be reported in a different part of the expenditure survey. They should not be included in tourism expenditure (IRTS 2008, para. 4.7); 

b.  Expenses clearly associated with a specific trip to the vacation home, such as transportation and food. They should be reported within the specific tourism expenditure module. 

4.70.                  As noted earlier, domestic tourism expenditure includes expenditure by visitors within their economy of origin for and on outbound trips. Such expenses are principally related to international transportation, reservation services paid to domestic travel agencies, services provided by residents[1], and any tourism expenses incurred during the leg of the trip that occurs in the country of residence (see paras. 4.19-4.20). It is important to collect information on such outbound trip expenditures that occur before departure from the country of residence. This information, which is not necessarily covered by domestic travel surveys, can also be derived from surveys on outbound trips. 

4.71.                  So that more details may be obtained on certain aspects of visitor spending, subsamples of the selected population can be established according to a more detailed breakdown of expenditure. Such an approach has the advantage of allowing each respondent to provide the details relating to some but not necessarily all categories of expenditure, thereby making the interview procedure less burdensome while still permitting detailed information to be obtained. This approach assumes, however, that the size of the overall sample is sufficiently large to permit subsampling while still yielding reasonably precise information and allowing for the cross‑classification of observed variables (non‑monetary as well as monetary). As a result, the statistical design and grossing up procedures could be complex.

Box IV.7 

Trip imputation: Travel Survey of Residents of Canada 

In the Travel Survey of Residents of Canada, a trip imputation is used because of the subsampling strategy. First, respondents' trips are rostered and basic information is gathered. Then, a subset of those trips is selected for the collection of additional information which is referred to as the detailed trip information.

This strategy leaves all of the detailed trip information variables with missing values for unselected trips. This   information is imputed by means of a donor imputation strategy. For each   unselected trip (the "recipient"), one single trip (the "donor") is selected to impute all the detail variables. The   information from this donor is borrowed to fill in the missing values in the recipient. Therefore, it is important to select the donor carefully. Also,   note that the imputation is performed separately by collection month.

The process of finding a donor for the unselected trips resembles the way averages are calculated in the expenditures imputation stage. The donor is selected from a set of trips with similar characteristics to the recipient’s, called a "donor pool". Only trips from the same recall month and trip type (same‑day or overnight) as the recipient are considered to be part of the donor pool. These are mandatory matching categories.

In addition, trips with high reported total spending are excluded from the donor pools to avoid creating new outliers through imputation. After the donor pool is finalized, the final donor is selected from the pool using a distance function that indicates how far each of the donors in the pool is from the recipient in terms of some continuous variables. 

For same‑day trips, the donor whose trip distance is closer to the recipient's is chosen. For overnight trips, a more complex distance function, which incorporates the number of nights and the trip distance, is used to select the donor. Finally, after the donor is selected,   all its information is copied over to the recipient. A donor can be used more than once.

Approximately 13 percent of overnight trips and 25 percent of same‑day trips that were rostered but not explored in detail will have characteristics of other people’s trips assigned to them through this imputation procedure.


Source: Statistics Canada.

[1] It should be noted that, unlike the purchase of an imported good within the country of the outbound tourist, an imported service (which most often relates to transportation) is not part of tourism expenditure and hence should be excluded.