Data Analysis for MPA Students

Concise Managerial Statistics, authored by Alan H. Kvanli, Robert J. Pavur, and Kellie B. Keeling, is a comprehensive and intuitive textbook for teaching graduate students in a variety of disciplines the fundamental concepts of data analysis and statistical inference. Through extensive use of illustrations and examples, the authors emphasize the practical application of statistical concepts and analytical tools while giving proper attention to the underlying theoretical frameworks. Providing detailed instructions for all data analysis tools covered in the textbook in Excel, Minitab, and SPSS along with the authors’ own Excel macro add-in file is unique and especially useful aspects of this textbook. Adding the plethora of additional resources for both students and instructors that accompany this text make it a great choice—from among many available—for use in the classroom.

I have personally used this textbook for several years to teach a required data analysis course for graduate students pursuing their master’s degree in public administration. Upon successful completion of this course, I expect my students to be able to (1) identify and propose questions of analysis that are pertinent to contemporary public policy and the broader study of public administration, (2) formulate a step-by-step approach for analyzing public management problems and policy questions, (3) identify the most appropriate methodological techniques for analysis of given research questions and available data, and (4) conduct data analyses using various methodologies covered in the course. Many of my students come from backgrounds without much emphasis on mathematical skills or statistical concepts, if at all. Although these students find the textbook challenging, the learning curve is not insurmountable. I attribute the approachability of this textbook to the authors’ ability to break down complex topics and explain them in everyday language while not ignoring the mathematical and theoretical foundations of the concepts being taught. I very much appreciate this approach and believe it contributes to a successful learning environment for my students; it certainly helps students accomplish the learning objectives I have laid out for the course. Although many of the textbook’s applied examples are topic specific to a business context, there are sufficiently relevant examples and appropriate practice exercises for students of public policy and/or administration. For example, I often create my own examples for use during class lecture, but always assign homework problems directly from the textbook. Practice exercises can be found throughout each chapter (not only at the very end of each) and are always categorized as problems to help students (1) understand the mechanics of the chapter’s section topic, (2) extend the topic and its use by applying new concepts, and (3) use computer data sets for further application of each statistical concept. The progression of difficulty, multitude of practice exercises, data sets and computer application problems, and detailed explanations of simulation exercises all offer ample opportunity for students to practice the tools and techniques covered throughout each chapter, which I believe is an essential component of developing statistical analysis skills.

In Concise Managerial Statistics, chapters 1 through 3 cover the important fundamentals of data collection, presentation, and descriptive statistics. These chapters lay the foundation for the remainder of the textbook by introducing essential vocabulary, providing clear differentiation of data types and levels of measurement, describing various means of graphically displaying data, explaining frequency distributions, and covering the most commonly used measures for summarizing and describing data. The appendices to these chapters are especially useful as the authors walk students through each mouse click required to produce the graphs and statistics in Excel, Minitab, and SPSS. These instructions make it possible for even the first-time user of the software packages to professionally analyze and present data for a variety of purposes. Because our Masters in Public Administration (MPA) is a terminal degree program aimed at preparing students for professional careers in the public and nonprofit sectors, I exclusively use Excel for all statistical applications throughout the course. Although I agree that using Excel for data analysis is not always the most user friendly, and Excel does not have the power of other software for conducting statistics, I firmly believe having expertise in data manipulation, analysis, and presentation in Excel is most beneficial for my MPA students because it is what will be most widely available when they enter the workforce. In addition, feedbacks from several alumni of our MPA program have indicated that federal and state agencies often use the ability to generate pivot tables in Excel as an important metric for gauging expertise in the software and data analysis skills. Because the students entering my class are typically not proficient in Excel, the textbook’s focus on its use and detailed explanations are essential for my students to gain this necessary expertise.

Chapters 4 thorough 6 focus on theoretical concepts of probability and probability distributions. These are the chapters that my students often find most challenging, yet I believe are essential for understanding the application of inferential statistics. In these chapters, the authors make use of contingency tables and Venn diagrams to introduce basic probability definitions and rules, as well as to provide relevant and understandable examples to explain the fundamentals of random variables and several discrete and continuous probably distributions. For this information, the authors’ extensive use of illustrations and summaries of formulas are especially helpful for students.

Chapter 7 introduces the key concepts of sampling and statistical inference. Here, the authors use Excel to show students how the means from multiple samples can be described with the frequency distribution tools students learned in earlier chapters and then translated into a sampling distribution for calculating probabilities. The statistically empowering central limit theorem is also introduced in this chapter. I believe chapter 7 of this textbook is best described as a culmination of the prior six; this is the point each semester when my students begin to see the payoff of several weeks’ worth of struggling through learning unfamiliar concepts that seem vague and disparate. This chapter always seems to refocus and reinvigorate students as they begin to feel equipped to use data analysis to answer research questions that are relevant and interesting.

The remainder of the textbook covers application of statistical tools for inferential data analysis. Chapters 8 and 9 introduce hypothesis testing and apply the concept to means testing a single and then two or more populations. Chapter 10 provides the tools necessary for analyzing categorical data. Chapters 11 and 12 cover correlation and both simple and multiple regression using ordinary least squares. Finally, chapter 13 introduces the basics of time series analysis and forecasting. In each of these chapters, the authors continue to provide appropriate, practical examples of application for each analytical tool covered, to explain each step for using Excel, Minitab, and SPSS for data analysis, to offer formula summaries and illustrations to aid students in learning, and to provide numerous opportunities for students to practice each new skill learned. In addition, the authors are particularly adept at walking students through each step of the process for hypothesis testing—taking time to show students how to state the null and alternative hypotheses, to identify the appropriate test statistic, to calculate the value of the test statistic, to predetermine an alpha level and testing procedure and then use these to make decisions about whether to reject or fail to reject the null hypothesis, and finally to offer a practical conclusion based on the results of hypothesis testing that addresses the initial research question. Focusing on each of these important steps and using repetition to reinforce proper procedure is an aspect of this textbook that I find highly commendable. This approach is one of the many reasons I find the textbook to be valuable for teaching data analysis to MPA students.

Aside from the content of the textbook and useful aspects of content presentation I outlined above, the resources that accompany this book for both students and instructors are some of the best on the market. Particularly unique to this textbook is an Excel macro add-in file that was developed by the authors. As mentioned earlier, I teach the course exclusively in Excel format because it is the most widely available software in the workforce my students are entering into. I believe this approach gives my students more advanced Excel skills than the average MPA graduate, which makes them highly competitive on the job market and well qualified for their employment positions. In this textbook, the KPK Excel macro is a data analysis tool pack that further enhances the Excel expertise of my students due to its two advantages over Excel’s built-in data analysis add-in file. First, the KPK macro file offers some data analysis capabilities that Excel does not have. Second, for the data analysis tools that are replicated, the KPK macro file offers them in a user-friendlier format that is easier for students to use and interpret. The downside of this approach, is that the KPK macro file is not usable in versions of Excel for Mac, which is becoming increasingly popular among my graduate students. However, when Microsoft implemented its significant upgrade in 2007, the textbook authors wrote and issued an updated version of the KPK macro file so that it would remain compatible with Microsoft Excel. After personally e-mailing the textbook authors for the updated version, I was able to share the new file with my students and continue its use for my class.

Student resources offered in addition to the textbook and Excel macro add-in file include a CD that offers extended content of each chapter along with the databases and data files (in Excel, Minitab, and SPSS formats) for all examples and exercises contained in the textbook. Resources for qualified instructors include an instructor’s manual, Powerpoint slides, and test bank questions for each chapter of the textbook. These are perhaps the most comprehensive resources accompanying any textbook I have used.

Overall, I believe this book is an excellent resource for a beginning level graduate course on data analysis and statistics. Although the textbook is geared toward students in a business program, I find it relevant and useful for students in public administration as well. While I sometimes find it necessary to adapt the examples and applications of the book’s concepts toward a more public sector focus, the book’s contents are certainly within the realm of appropriateness for training our future public managers and leaders.