Introductory Statistics and Analytics: A Resampling Perspective

1

DESIGNING AND CARRYING OUT A STATISTICAL STUDY

In this chapter we study random behavior and how it can fool us, and we learn how to design studies to gain useful and reliable information. After completing this chapter, you should be able to

• use coin flips to replicate random processes and interpret the results of coin-flipping experiments,
• define and understand probability,
• define, intuitively, p-value,
• list the key statistics used in the initial exploration and analysis of data,
• describe the different data formats that you will encounter, including relational database and flat file formats,
• describe the difference between data encountered in traditional statistical research and “big data,”
• explain the use of treatment and control groups in experiments,
• explain the role of randomization in assigning subjects in a study,
• explain the difference between observational studies and experiments.

You may already be familiar with statistics as a method of gathering and reporting data. Sports statistics are a good example of this. For many decades, data have been collected and reported on the performance of both teams and players using standard metrics such as yards via pass completions (quarterbacks in American football), points scored (basketball), and batting average (baseball).

Sports fans, coaches, analysts, and administrators have a rich array of useful statistics at their disposal, more so than most businesses. TV broadcasters can not only tell you when a professional quarterback's ...