Introduction: Inputting Questionnaire Data versus Other Types of Data

This chapter shows how to create SAS datasets in a number of different ways, and it does this by illustrating how to input the types of data that are often obtained through questionnaire research. Questionnaire research generally involves distributing standardized instruments to a sample of participants, and asking them to respond by circling or checking fixed responses. For example, participants might be asked to indicate the extent to which they agree or disagree with a set of items by selecting a response along a 7-point Likert-type scale where 1 represents “strongly disagree” and 7 represents “strongly agree.”

Because this chapter (and much of the entire book, for that matter) focuses on questionnaire research, some readers might be concerned that it is not useful for analyzing data that are obtained using different methods. This concern is understandable, because the social sciences are so diverse and so many different types of variables are investigated. These variables might be as different as “the number of aggressive acts performed by a child,” “rated preferences for laundry detergents,” or “levels of serotonin in the frontal lobes of chimpanzees.”

However, because of the generality and flexibility of the basic principles of this discussion, you can expect to input virtually any type of data obtained in social science research upon completing this chapter. The same can be said for the remaining chapters of this book; although it emphasizes the analysis of questionnaire data, the concepts taught here can be readily applied to many types of data. This fact should become clear as the mechanics of using SAS are presented.

This book emphasizes the analysis of questionnaire data for two reasons. First, for better or for worse, many social scientists rely on questionnaire data when conducting their research. By focusing on this method, this text provides examples that are meaningful to the single largest subgroup of readers. Second, questionnaire data often create special entry and analysis problems that are not generally encountered with other research methods (e.g., large numbers of variables, “check all that apply” variables). This text addresses some of the most common of these difficulties.