3. Critique of Questionnaire The introduction, although longer than a couple of sentences, effectively communicates the objective of the survey through polite and inviting language. Having an envelope attached establishes further convenience. The initial identification of the questionnaire is that is appears relatively clear and is divided into sections to however the use of colour could increase the appeal of the questionnaire (Birdie et al 1992) and there is insufficient space to answer open questions with an appropriate response such as question 14, question 24 and question 48. This could lead to a lack of obtainable qualitative information and could inhibit the objective of the survey. With regards to the sequencing of the questionnaire the flow is consistent and the use of instructions to skip to certain questions if answered 'no' avoids respondent's answering irrelevant questions and allows the respondent to easily navigate their way around the questionnaire without much confusion. The wording of questions is overall clear and easy for respondent's to understand, however there is use of vague language, for example in question 2 the use of 'occasionally' and 'regularly' shouldn't be used as these quantifiers don't convey the same meaning to all respondent's. As an improvement the question should use an appropriate time frame, in order to make it possible for respondents to to provide accurate answer, which also makes it easy for data to be analysed. Ideally the questionnaire should begin with 'non-threatening' and interesting questions to increase the likelihood of the respondent completing the questionnaire (Levine and Gordon 1958) and in order to create a basic customer profile the questionnaire should also include personal questions regards age and gender, information that this questionnaire fails to obtain. The questionnaire has a variety of questions where respondent's answer in response categories, in order to measure attendant's opinions that can easily be transferred into data and analysed. Question 25 for example, has a scale of 5 which is balanced scale that allows respondents to select a neutral option, Bradburn, Sudman and Wansink (2004) believe that middle-of-the-road-respondents should not be forced to express opinions.
However, there is use of biased rating scales within the questionnaire, for example question 6, gives the respondent the option to answer 'yes', 'probably' and 'no', rather than a neutral option which will generate invalid data. The design of the questionnaire influences the way the answers are coded to be analysed in SPSS. There are less open questions and most of them occur when the respondent is asked to specify their opinion if none of the categories match their opinion. If enough respondents give similar answers the open answers collected can be changed into closed answers in order for the data to be inputted into SPSS. The questionnaire is made up of majority closed questions which give the respondent a restricted amount of answers which in turn restricts the meaning of the question to match the research objective and allows for the data to be transferred to SPSS with ease. Closed answers generates 3 levels