Survey data is collected on the International Passenger Survey via face to face interviews with passengers passing through ports and on routes into and out of the UK.
The IPS methodology involves conducting between 700,000 and 800,000 interviews a year, of which over 250,000 are used to produce estimates of Overseas Travel and Tourism patterns.
A multi stage sampling design is employed which involves sampling a port or route on a given day and within a given period of the day (this is referred to as a ‘shift’); and within the shift certain passengers passing an interview line are systematically chosen for interview at fixed intervals from a random start.
This involves sampling a port or route on a given day and within a given period of the day (this is referred to as a ‘shift’); and within the shift certain passengers passing an interview line are systematically chosen for interview at fixed intervals from a random start.
There are two fixed intervals applied on each shift.
The first is referred to as the ‘first-stage sampling rate’. It is the sample interval applied to all passengers crossing the interviewing line.
Each of the passengers contacted through application of the first-stage sampling rate is screened to assess whether or not he or she is migrating (if the contact is migrating, he/she will then be asked migration-specific questions, but if not migrating no further questions will be asked).
The second fixed interval is referred to as the ‘second-stage sampling rate’ and is employed to contact a sub-sample of contacts, to which questions relating to travel and tourism are asked.
In an example where the first-stage sampling rate on a shift is 10 and the second-stage sampling rate is 40, three out of four passengers are approached and screened to see if they are migrating, only, whereas the fourth is both screened for migration and additionally interviewed about his or her visit from a Travel and Tourism point of view.
Contacts who are selected for interview and who complete either the whole or majority of the interview are classified as ‘completes’.
Some respondents give very limited information, but enough for them to be included in the weighting – these are classified as ‘minimums’.
Some contacts either can’t be contacted or are contacted and refuse to provide any information and they are referred to as ‘nonresponders’.
The classification of response determines how each contact is treated in the weighting procedure.
The basis of the weighting of IPS survey data is that the total set of respondents interviewed at a port or route is weighted up/calibrated to passenger traffic known to have passed through that port or route in the period in question.
The known passenger traffic information is provided to the IPS team by CAA, Department for Transport, Eurostar, Eurotunnel, BAA and a number of airports themselves.
The weighting approach incorporates a number of stages which take account of all passengers selected for interview.
Weighting is conducted for each port/route and direction of travel combination, employing the same principles at each one.
The stages, listed in order of application, are as follows.
A Design weight is employed, to account for the probability of sampling this passenger using the first-stage sampling rate.
The calculation compares the number of shifts or crossings sampled (at each port/route and direction of travel combination) with the number of shifts or crossings that could have been sampled for that combination in the period.
In addition it takes into account the first-stage sampling rate. For example, in a case where a contact was sampled at a port with the following details:
10 shifts were run in the period
100 shifts could have been run in the period the contact was sampled employing a first-stage sampling rate of 20 (that is, every 20th passenger was selected)
the design weight for this contact would be 200, calculated as (100/10) x 20
As well as port/route and direction, this weight incorporates weekday or weekend, and am, pm or night as weighting strata.
A Nonresponse weight factor is employed to take account of contacts selected for interview but who were subsequently not interviewed, either because it was not possible to contact them or they refused to participate.
The weight is applied at each port/route and direction travel combination and also incorporates weekday versus weekend as weighting strata.
It involves uplifting ‘complete’ and ‘minimums’ cases by a factor calculated as:
the sum of weights applied to all ‘completes’, ‘minimums’ and ‘non response’ records
divided by the sum of ‘completes’ and ‘minimums’ at that port/route and direction of travel combination
A second Design weight is applied to account for the second-phase of the sample design and relates to the subsampling of non-migrants.
The weight for this factor is simply equal to:
the ratio second-stage sample interval: first-stage sample interval for non-migrants, 1 for migrants
A weight factor is applied for discarding minimum respondents.
Minimum interviews are discarded in this step of the weighting, with other cases weighted up to compensate.
The purpose of applying this weight is that it is possible that the profile of minimums might be skewed to certain nationalities or residents of a certain countries (for example driven by language difficulties meaning that only minimal information is provided to the interviewer).
This weighting step works to the same principle as the nonresponse weight. It utilises port/route and direction of travel as weighting strata.
Weighting to sampling frame. Here the population (that is, passenger traffic) or the ports and routes covered by the sampling frame are used to weight the data.
The population excludes interlining passengers (those neither entering nor leaving the UK from this port, that is, simply changing international flights) and out-of-hours traffic (that is, arriving or departing outside the hours covered by the IPS interviewing at that port).
The weight is applied at each port/route and direction of travel combination.
Weighting for frame under coverage. This extends the above population weighting to compensate for not covering certain ports and times of day (out-of-hours traffic) in the survey sample.
The weight utilises port/route and direction of travel as weighting strata and also incorporates region of the world that traffic has come from/gone to.
The weight reflects the fact that flights to and from some parts of the world are more likely than others to arrive or take off at night when no interviewing is conducted at airports.
Weighting for observed imbalance. This step is used to correct an observed imbalance between the number of non-migrants entering and leaving the UK.
These are applied as a series of fixed factors, relating to direction of travel, port/route and country/residence.
It has been noted since spring 2009 that there has been an increase in the proportion of respondents in the IPS overseas travel and tourism sample who are starting their visit compared to the proportion ending their visit.
This proportion of the two types of traveller in the sample defines the estimates of travel and tourism.
There is no clear reason for this trend and ONS has taken steps to calibrate its overseas travel and tourism estimates with external data, notably estimates from surveying conducted at departure gates at main airports in the UK by the Civil Aviation Authority (CAA) and e-borders data.
This work has shown general consistency between the datasets with the result that the factors used in the imbalance weight have been retained. However, this is an area of ongoing research.
A final weight is applied, which combines each of the weighting stages listed above.
Where the responses for key items of interest are missing from the survey data for an individual record the values are imputed.
Imputation is applied to the following items:
Length of stay
Cost of fare (expressed in terms of cost of the single fare for the respondent)
Town of stay
For each of length of stay, cost of fare and spend, a value is calculated for the survey record whichhad the information missing.
The IPS employs a mean-value within class imputation procedure where the missing value is replaced with the average value for records with similar characteristics.
The matching variables used for each of these items are:
|Length of stay:||Country of visit/visiting from|
|Purpose of visit|
|Cost of fare:||Port in UK travelled to/from|
|Overseas port travelled to/from|
|Month of travel|
|Spend:||Country of visit/visiting from|
|Purpose of visit|
Where the respondent has travelled on a package holiday, the cost of the fare is imputed and thendeducted from the total cost of the package, and the residual cost (after removal of a percentage to cover travel agent fees) is assigned to expenditure.
Overseas residents staying in the UK are asked about their total expenditure in the UK.
This information is then imputed across the towns stayed in, proportionate to the length of stay in each one.
It is recognised that people tend to spend more when they stay in London than in other towns in the UK and therefore an uplift index is calculated and applied to the spend allocated to London in cases where the respondent stayed in both London and other towns in the UK.
In cases where an overseas resident hasn’t given details of all the towns in the UK they stayed in, an uplift is applied to towns stayed in by similar records, using the same principles as outlined above for the imputation of stay, fares and spend.