1 Introduction

1.1 Overview

In the world of data analysis, the choice of software can significantly impact the flexibility, efficiency, and depth of your analyses. For many years, SPSS has been a popular tool among social scientists, market researchers, and others who require a straightforward interface for statistical analysis. However, as data analysis becomes more complex and expansive, many analysts are turning to R, an open-source programming language and environment that offers unparalleled flexibility, a vast array of packages, and a growing community of users.

This course is designed to help you transition from SPSS to R, demonstrating that all of the functionality you’re accustomed to in SPSS can be replicated—and often enhanced—in R. By the end of this course, you will be equipped with the knowledge and skills to conduct your analyses in R, whether you’re dealing with basic descriptive statistics, survey data, or more advanced statistical models.

1.2 Why Learn R?

1.2.1 Flexibility and Power

R is a full-fledged programming language, which means it is inherently more flexible than SPSS. While SPSS provides a user-friendly interface with point-and-click options, R allows for a deeper level of customisation and control over your analyses. You can automate tasks, create reproducible reports, and even develop new statistical methods if needed.

Example: In SPSS, you might run a regression using a dialog box and then manually record the output. In R, you can automate this process, run multiple regressions in a loop, and automatically output the results to a report.

1.2.2 Reproducibility

Reproducibility is a cornerstone of modern data analysis. R’s script-based workflow ensures that every step of your analysis is documented, making it easy to reproduce results or adjust your analysis if new data becomes available. This contrasts with SPSS, where much of the analysis is done through a GUI, making it harder to track and reproduce each step without explicitly saving syntax files.

Example: In R, you can save your entire analysis pipeline in a script, which can be rerun with new data or shared with colleagues for replication.

1.2.3 Extensive Community and Package Ecosystem

One of R’s greatest strengths is its vast and active community, which has contributed thousands of packages to CRAN (the Comprehensive R Archive Network). Whether you need advanced statistical techniques, machine learning algorithms, or specialised visualisations, there’s likely an R package available. This is in contrast to SPSS, where functionality is often limited to what is provided out-of-the-box or through costly add-ons.

Example: In R, you might use the ggplot2 package for advanced visualisations, dplyr for data manipulation, or survey for complex survey analysis. In SPSS, creating custom visualisations or handling complex survey designs might require more manual effort or external software.

1.2.4 Cost

R is completely free and open-source, which is a significant advantage over SPSS, especially for organisations with budget constraints. This means you can install R on as many computers as you need, share it with colleagues, and access the latest updates and packages without any cost.

Example: Many organisations are moving to R not only because of its capabilities but also because it reduces software costs significantly.

1.3 Replicating SPSS Functionality in R

If you’re used to SPSS, the idea of switching to a command-line-driven environment might seem daunting. However, you’ll find that every major feature of SPSS can be replicated in R—often with greater flexibility and power.

1.3.1 Data Management

In SPSS, data management tasks such as merging datasets, recoding variables, or selecting cases are performed using a series of dialog boxes or syntax commands. In R, these tasks are handled with functions and packages like dplyr, which offer intuitive syntax for manipulating data.

Example: Recoding variables in SPSS might require several steps in the GUI, while in R, you can accomplish the same with a single line of code using mutate() and case_when().

1.3.2 Descriptive Statistics

Both SPSS and R allow you to calculate descriptive statistics such as means, medians, and standard deviations. In R, you can use basic functions like mean() and sd(), or you can employ packages like psych or skimr for more detailed summaries.

Example: Descriptive statistics in SPSS are often displayed in tables generated by the DESCRIPTIVES command, whereas in R, you can achieve this with a simple script, and even automate it for multiple variables.

1.3.3 Statistical Tests

SPSS is known for its user-friendly interfaces for running statistical tests. In R, all the same tests (t-tests, ANOVA, chi-square, etc.) are available, and you can conduct them using straightforward commands.

Example: Running a t-test in SPSS involves navigating through multiple menus, while in R, the same test can be executed with a simple t.test() function.

1.3.4 Regression Analysis

Regression analysis, a staple of SPSS, is fully supported in R. Whether you’re running simple linear regressions or more complex logistic regressions, R provides the tools you need. The lm() function is used for linear models, while glm() handles generalised linear models, including logistic regression.

Example: In SPSS, you might need to manually specify each option in the regression dialog box. In R, you have the flexibility to customise your models directly in code, making it easy to adjust your analysis as needed.

1.3.5 Data Visualisation

While SPSS provides basic charting capabilities, R, with its ggplot2 package, is unmatched in the realm of data visualisation. You can create everything from simple bar charts to complex multi-faceted plots, with full control over every aspect of the appearance.

Example: A simple bar chart in SPSS might look quite basic, whereas in R, you can use ggplot2 to add layers, colors, themes, and annotations to create publication-quality graphics.

1.4 Transitioning from SPSS to R

Transitioning from SPSS to R might feel like a big leap, but with the right guidance, you’ll soon see how R’s power and flexibility make it a worthwhile switch. Throughout this book, we will replicate common SPSS procedures in R, step by step. You’ll see how to import your data, perform the same analyses you’re used to, and even go beyond what SPSS can offer.

1.4.1 Building Confidence in R

We’ll start with simple tasks, such as descriptive statistics and data manipulation, gradually moving to more advanced topics like regression analysis and survey data handling. By building on your existing knowledge of SPSS, you’ll find that learning R is not as daunting as it might seem.

Example: We’ll compare SPSS and R workflows for common tasks, showing how R can simplify and enhance your analytical processes.

1.4.2 Leveraging R’s Ecosystem

One of the goals of this course is to familiarise you with the rich ecosystem of R packages. We’ll introduce you to some of the most useful packages for data analysis, showing how they can replace or enhance the tools you’re used to in SPSS.

Example: For instance, you’ll see how dplyr can streamline your data manipulation tasks, or how ggplot2 can elevate your data visualisation.

1.5 Conclusion

As you proceed through this course, you’ll gain the skills needed to replicate and enhance the analyses you currently perform in SPSS. Each chapter will build on the previous ones, gradually expanding your knowledge and capabilities in R. By the end of this book, you’ll be able to handle all your data analysis tasks in R with confidence, taking full advantage of its flexibility, power, and vast ecosystem of packages.