Descriptive Statistics

Abstract

Market research projects involving data become more efficient and effective if a proper workflow is in place. A workflow is a strategy to keep track of, enter, clean, describe, and transform data. These data may have been collected through surveys or may be secondary data. Haphazardly entering, cleaning, and analyzing bits of data is not a good strategy, since it increases one’s likelihood of making mistakes and makes it hard to replicate results. Moreover, without a good workflow of data, it becomes hard to document the research process and cooperate on projects. For example, how can you outsource the data analysis, if you cannot indicate what the data are about or what specific values mean? Finally, a lack of good workflow increases one’s risk of having to duplicate work or even of losing all of your data due to accidents. In Fig. 5.1, we show the steps necessary to create and describe a dataset after the data have been collected.

Learning Objectives

After reading this chapter, you should understand:

– The workflow involved in a market research study.

– Univariate and bivariate descriptive graphs and statistics.

– How to deal with missing values.

– How to transform data (z-transformation, log transformation, creating dummies, aggregating variables).

– How to identify and deal with outliers.

– What a codebook is.

– The basics of using IBM SPSS Statistics.