Our questionnaire has been carefully constructed to provide both statistical (quantitative) information about the visitor and to search for the subjective (qualitative).
The qualitative structure is based upon the Manning-Haas Hierarchy of Recreational Demand (Manning, 1986). [find references here] This North American recreational model has previously been adopted by Prentice & Light (1994) in their research on visitor perceptions of heritage sites but the present researcher is not aware of it having been used to determine visitor perceptions of a whole destination. The value of the hierarchy, claim Prentice & Light (1994:207 ) is that "it has great potential as a means of defining visitors' satisfaction...". Desired outcomes can be reached by looking at the context in which the visitor experience takes place. The model is based on the premise that the benefits are generated from the experiences had from the settings of the visit and the activities pursued. This model has been adapted in accordance with our study and the structure of our questionnaire can now be read as follows:
Expectations - Activities - Settings - Experience - Satisfaction
A glance at the questionnaire will reveal that this structure is not concrete and that the categories may easily overlap. Integration is a key feature of the model as it allows for information to be compared during analysis. The questions have also been structured to invite participants to build a chronological account of their visit so that they are ready to address each issue spontaneously. In a later paper, Beeho & Prentice (1996) recommend using the model in conjunction with a SWOT analysis. A brief summary of Strengths, Weaknesses, Opportunities and Threats will appear in the conclusion.
The construction phase of the interviews were complemented by declared pilot testing. That is, selected participants ( friends, relatives and colleagues) were informed of the purpose of the research and invited to take part in interviews to help to develop the questions. It was found that the concept of tourist markers was not easily understood by participants and was not being adequately represented. The questions were re-formulated and the wording was simplified.
Longitudinally, Tintagel can be defined as encompassing the area between two deep, natural valleys that extend from the coast; Trebarwith Strand (046862) and Rocky Valley (072894)2. The depth of the area logically includes the surrounding hamlets, main roads and key visitor attractions to the point where urbanisation is subverted by wilderness. A respondent was considered to be eligible so long as they did not reside within this area. All personal interviews were conducted randomly in Fore Street. Structured mail questionnaires were distributed to three key visitor sights in the village; the Visitor Centre, Dragon's Breath gift shop and King Arthur's Halls.
Non controlled observation and unstructured personal interviews were recorded spontaneously and have been integrated into the report. Structured personal interviews (i.e. questionnaires) have been coded and fed into a Statistical Package for the Social Sciences (SPSS)3. This means that some information will be lost but its condensation will allow for generalisations and comparisons to be made
Because our sample is non-probability and does not make use of a sampling frame, Rose and Sullivan (1996) advise that it is unnecessary to apply inferential statistics Statistical analysis will therefore be confined to measures of central tendency and crosstabulation and will be displayed in a series of simple tables. There is much information to be gained even from these statistics and the researcher has attempted to select the most relevant points, given the scope of the report. The reader is invited to make their own inferences by consulting the tables and the raw data in the appendix .
The presentation of the report will be guided by the layout of the questionnaire and will be segregated into quantitative and qualitative analysis.