Mr Hungry's Burger Bar is a national chain of hamburger restaurants with branches in all major cities of the UK and most large towns. The burger bars typically attract three main categories of customer: teenagers looking for a cheap, filling lunch while cruising shopping malls, parents who take young children to the restaurants for a treat or birthday celebration and office workers and others looking for a quick lunch stop during the day or a pre-cinema snack in the evening.

In an attempt to gauge customer attitudes to the restaurant, each customer is asked to complete a form, which uses a five-point Likert scale to assess responses across a range of issues. The form is shown below. At Mr Hungry's we are always looking for ways to improve our service to you, the customer. To help us do this, we would be grateful if you could take a few moments to fill out this form.

Return of the forms tended to be low (less than 10% from most restaurants) but given the very large numbers of customers involved, the firm's customer services department found that it was overwhelmed with forms, with 12,532 useable responses coming in during the first month. A raw tabulation of the forms showed the following pattern.

These figures caused some consternation; on the face of it, it seemed that on most measures the restaurants were not doing well. The trend was for customers to be very dissatisfied or dissatisfied with each element of the restaurant, with only the convenience of location coming out well.

Another problem was that the forms could not be segmented; the analysts were unable to tell which sub-category of the firm's target customers had filled in each form. Since the restaurants remained busy, and indeed business seemed to be increasing rather than decreasing, senior management felt that the research must be flawed in some ways.

After all, the customers were perfectly capable of voting with their feet by going to any one of half a dozen other hamburger outlets; if the restaurants were obviously all right, and the research said the restaurants were all wrong, then the research had to be at fault. The directors of Mr Hungry's told the researchers to look again at the research design and report back.

Questions
1. Plot the figures in the form on a graph, using a separate plot for each attribute. What can you deduce from the shape of the lines?

2. How might the respondents be divided so that the researchers could determine whether responses regarding one attribute of the restaurants relate to responses regarding another attribute?

3. Which of the attributes show significant correlations with each other? What might be the reasons for these correlations?

4. What are the sources of bias in data of these types?

5. How might the firm improve the form in order to obtain better quality data?

6. How might you check whether the variation shown in the data is statistically significant?