This chapter shows a poor but all too common approach to undertaking a sample survey. No initial thought is given to the aims of the survey. As a result, the information/data may not be appropriate. Consequently, one may not be able to do the data analysis he/she would like. The library loan survey Suppose people simply went ahead and only recorded the number of times each book in the sample was borrowed during the year. After conducting the survey they might wonder whether or not there are any differences in the loan rates between various degree courses. If they had not thought about recording the courses that the students borrowing the books were on, people could not do this analysis. The key point is the order in which things should come. One should formulate the aims of his survey and decide what data analysis is needed right at the beginning.