 |
 |
Research - TVS
How to interpret TVS report data
Why survey estimates may differ from the “real” number?
The estimates in this report are based on information obtained from a sample of visitors coming to Tasmania. As with all surveys, the estimates are subject to sampling variability. This means the survey results may vary from the results that would have been produced, if all visitors could have been interviewed.
Therefore, it is often prudent to talk about results in a range, rather than as a single figure. These ranges are called “confidence intervals”. In other words, the true result lies within a range, to which a level of confidence can be given. The above table shows ranges for visitors, nights and expenditure estimates. Statistically, we are 95 percent confident that the actual result lies somewhere within the lower and upper limits of these ranges.
A basic assumption is that the bigger the sample (i.e. the more visitors that were interviewed), the smaller the range can be. Conversely, results based on a small number of interviews are less reliable, meaning the size of the range is bigger to compensate. This means than smaller estimates are less reliable than larger estimates.
How to interpret TVS statistics using confidence intervals – examples:
Visitors example (bottom figure, left hand table). The TVS estimated that 800,000 scheduled air and sea visitors came to Tasmania. The lower range of this result is 762,667 and the upper range is 837,333. Therefore, there are 19 chances in 20 (i.e. we are 95 percent confident) that the true number of visitors to Tasmania was somewhere between 762,667 and 837,333 visitors.
Expenditure example (top figure, right hand table). When a group of visitors returns a result that they spent $1,500,000, we are 95 percent confident that they spent somewhere between $740,000 and $2,260,000.
Non sampling error
Non-sampling errors may occur due to such things as non-response and imperfections in reporting by surveyed visitors. Every effort has been made to keep non-sampling error to a minimum by careful design and testing of TVS questionnaires, efficient interviewing procedures and appropriate sampling methodology.
You will need Adobe's Acrobat Reader to read this document set:
Download
Acrobat Reader
A
PDF version of this explanation is
available for print-out.
For further information on interpreting TVS statistics please contact:
Research Section, Tourism Tasmania
+ 61 3 6230 8141
statistics@tourism.tas.gov.au