Making Sense of Statistical Analysis: A Researcher’s Handbook for Data Analysis using IBM SPSS starts with introducing research design. Starting a data analysis book with the introduction of a research design seems unusual. However, this book takes this unusual approach. The reason for starting a data analysis book using SPSS with research design is interesting in two ways. Firstly, research design guides the overall research process from conceptualization to report writing. Secondly, when to use which research design in a research study depends on the nature of the research question. Put explicitly, the research question determines the research design of the study. Therefore, to develop a sound and logical understanding of the readers, we have started the book with an introduction to the four types of research designs.
There are many excellent books on data analysis using SPSS. Nevertheless, data analysis using IBM SPSS still remains one of the sources of stress, frustration, and anxiety for beginners. The key reason for this is the non-availability of books that cover the theoretical understanding of statistics, how to conduct statistical analysis, and how to read and report information from the output generated as a result of statistical analysis. This book bridges this gap and blends the theoretical background of statistical analysis with practical guidance on conducting analysis. It adopts a step-by-step approach to reading and reporting information from the output according to the American Psychological Association (APA) format.
Section I familiarizes readers with the design of a study. It is divided into three chapters. Chapter 1 introduces the importance of choosing a research design. Selecting a suitable research design is essential to accomplishing a research report. Chapter 2 addresses the choice of an appropriate scale for data collection. Choosing a suitable scale after the research design selection is essential because the scale used for data collection largely determines the tests performed on the data to answer the questions posed in a research study. Chapter 3 covers the techniques for developing a coding book and outlines the principles necessary for preparing data entry procedures.
Section II provides researchers with a guide to the SPSS environment. Chapter 4 deals with creating an SPSS file, defining variables and data entry, creating a new data file, opening an existing one, and saving a data file. Chapter 5 introduces the techniques of data editing and handling in SPSS. It offers a summary of inserting a new variable, adding a new case, and selecting cases while performing tests on specific cases.
Section III takes a step-by-step approach to guide researchers in preparing files for analysis. Chapter 6 elucidates the researchers with data exploration using graphs and tables. Chapter 7 delves into data screening techniques, such as reversing negative items, calculating the total score of a scale, transforming data, converting continuous variables into categorical variables, and recoding variables into different variables in SPSS.
Section IV is your gateway to understanding descriptive statistics, a fundamental tool in the social sciences and statistics. These statistics are divided into two main categories: measures of central tendency and measures of variability. In Chapter 8, we equip you with knowledge of measures of central tendency, including mode, median, and mean. Chapter 9 then guides you through the application of measures of variability, such as variance, range, and standard deviation.
Section V deals with the reliability and validity of the instrument used for data collection. In the research process, having stated a problem, formulating one or more hypotheses or research questions, and describing a sample, the description of the instrument for data collection is a key factor. Chapter 10 briefly introduces the reliability of scales. Chapter 11 presents the validity of a scale used for data collection in a research study and several types of validity of a scale.
Section VI presents a map of the hypothesis testing logic and selecting an appropriate test for a research project. Chapter 12 discusses whether our research hypothesis should focus on the given population or samples. Why do we formulate our research hypothesis about the population? How do we select a sample to test the hypothesis about the population? Chapter 13 summarizes the techniques for selecting a suitable test for analyzing the collected data in a research project. It presents statistical tests in a very simple, understandable, and comprehensive manner, avoiding the complexity and confusion that creates panic and fear among the general readers in general, and students, in particular, regarding the logic and categorization of tests. Chapter 14 presents a summary of the assumptions of parametric tests.
Section VII illuminates tests of correlation and is comprised of two chapters. Chapter 15 reviews parametric tests of correlation, such as Pearson Product-Moment Correlation Coefficient and Partial correlation. Chapter 16 presents non-parametric tests of correlation, such as Spearman Rank Order Correlation. Section VIII elucidates comparison tests and consists of four chapters. Chapter 17 explains parametric tests of comparing two means. In contrast, chapter 18 deals with non-parametric versions of comparing two means. Chapter 19 explains parametric tests for comparing more than two means, while chapter 20 explicates non-parametric statistical tests for comparing more than two means.
Section IX enlightens on tests of prediction and is divided into three chapters. Chapter 21 reviews simple regression procedures and assumptions. Chapter 22 extends the concept of simple regression to multiple regression procedures, thus explaining the addition of more than one predictors to explain the outcome variable. When we have categorical dependent variables and categorical or continuous independent variables, logistic regression is a suitable tool to analyze data. Chapter 23 succinctly explains the concept and procedure of logistic regression.
The last section of the book comprises five chapters. Chapter 24 enlightens readers about the Chi-Square Goodness of Fit test; when researcher has one categorical variable, which is paired with a value that comes from the population. To elaborate on the analysis of more than one categorical independent and dependent variables, the Chi-Square Test of Independence is an appropriate statistical tool for analyzing data. Chapter 25 illuminates the assumptions, procedures, and reporting output of the Chi-Square Test of Independence. Chapter 26 explicates the use of Fisher’s Exact test when the expected frequency in each cell is less than 80%. McNemar’s test is used to find differences in a categorical dichotomous dependent variable between two related groups. It is similar to paired or repeated samples t-tests, which is also used to study the same group at two points in time. In paired samples t-test, the dependent variable is continuous. However, the data measurement level makes McNemar’s test different from that of paired samples–t-test. Chapter 27 is devoted to elaborating the use of McNemar’s test. Chapter 28 elucidates the use of Cochran’s Q test. Cochran’s Q test is used to determine differences in a dichotomous dependent variable between three or more related groups.
Nasir Mahmood, Ph.D.
Muhammad Yousaf, Ph.D.
This is a concise and comprehensive book that thoroughly explores both the theory and practice regarding using statistics in research. Many excellent books deal with statistical theory, such as Gravetter & Wallnau (2014), Field (2017), and Stevens (1992). Likewise, some good books explain the practical use of SPSS to analyze the data in research projects. For example, (Pallant, 2020); Brace, Kemp & Sneglar, 2012). In a nutshell, some of these texts solely focus on the theory of SPSS, and others pay attention to the use of tests on a given data set. However, using SPSS without sound and logical statistical knowledge is useless and misleading for students and researchers. No book deals with reading the generated output as a result of analysis and reporting it according to APA format. Therefore, there was a need to write a book that is a blend of statistical theory and practical use of SPSS and equip the readers, students, and researchers with the knowledge of how to read relevant information from many confusing tables that are generated as output after analysis. We designed this book to understand the aforementioned three aspects of data analysis. Hence, this book presents statistical analysis as a dynamic process, from choosing a research design to selecting an appropriate scale, developing a codebook, data entry, editing data, scanning data, checking scale reliability, and validity. In addition to these basic introductory techniques, the book categorizes statistical tests into two broad groups. From a pedagogical point of view, this book intends to address the gap in the existing literature. It presents a comprehensive text that deals with theoretical and practical aspects of data analysis. In addition to this, we have added in ‘focus box’ to highlight the salient points for conducting analysis and reading outputs when reporting results.
At the most obvious level, these statistical analysis groups are grouped under descriptive statistics and inferential statistics. At the descriptive statistics level, mode, median, and mean are presented as measures of central tendency, whereas range, variance and standard deviation are treated as measures of variability. At the level of inferential statistics, we have divided statistical tests into four broad categories to make this text direct, logical, and reader-friendly. To be more specific, inferential statistics are covered under the umbrella of the tests of association, tests of difference, tests of prediction, and tests of nominal data. We have followed the aforementioned approach to make understanding of diverse inferential statistics simpler, more straightforward, and understandable for the students and the researcher alike. In this quest to pursue simplicity, clarity, and avoiding ambiguity, we categorized each of the aforementioned inferential tests into further sub-categories.
To achieve this end and make the learning process more practical, the tests of association are categorized into parametric tests of association and non-parametric tests of association. Similarly, the comparison tests are divided into parametric tests of comparing two means and non-parametric tests of comparing two means. Moreover, comparing more than two means is grouped under the parametric tests of comparing more than two means and non-parametric tests of comparing more than two means. This practical categorization will help you understand and apply these tests in real-world scenarios.
Going a step further to ascertain this clarity, we have explained tests of prediction in detail both for continuous and nominal/categorical data. The section on nominal data covers five chapters to provide students and researchers with in-depth knowledge about the tests of nominal data. This is the less explored part of statistics analysis books. As a result, we are confident that this division would make the use of SPSS useful to beginners and students taking advanced statistical analysis courses, as well as to the researchers intending to use SPSS in analyzing the data for their research projects. Aside from this simple and logical division of inferential statistics, we are sure that the direct, simple, clear and reader-friendly style of this book would be beneficial for understanding statistical analysis. We hope this book will be an excellent addition to aid both theoretical and practical understanding of SPSS.