Intermediate Statistics with R

Reviewed by Zhuanzhuan Ma, Assistant Professor of Statistics, The University of Texas Rio Grande Valley on 10/21/24

Comprehensiveness rating: 5 see less

The textbook covers a wide range of statistical methods and techniques, including data wrangling, basic hypothesis testing, one-way and two-way ANOVA, permutation tests, chi-square tests, simple and multiple linear regression, as well as bootstrapping​. These are core concepts that any student of intermediate statistics would need to master. The integration of R, specifically the tidyverse and ggplot, makes the book comprehensive in terms of modern statistical software use. It provides detailed explanations and R code for each topic, helping students learn both the statistical method and its implementation​.

Content Accuracy rating: 5

The accuracy of Intermediate Statistics with R by Mark C. Greenwood appears to be high, particularly in its explanations of statistical concepts and its use of R for data analysis. The real-world examples and datasets used in the book help ground the theory in practical applications. These examples are representative of common statistical scenarios, ensuring that the material is not only accurate but also applicable to real-world data analysis. The accuracy of the textbook is strong in both its statistical content and its implementation in R. The author’s continual updates help maintain the relevance and correctness of the material, especially given the evolving nature of the R programming language.

Relevance/Longevity rating: 5

Intermediate Statistics with R is highly relevant due to its alignment with modern statistical practices, its use of widely-adopted software, and its focus on practical, real-world applications. The skills and methods taught in the book are not only critical for academic success but are also directly transferable to professional data analysis roles.

Clarity rating: 5

The clarity of Intermediate Statistics with R is one of its notable strengths. The text is written in a straightforward, accessible style, making it easier for students transitioning from introductory to intermediate-level statistics to follow along and engage with more complex topics. The book includes diagrams, flowcharts, and plots that visually clarify complex concepts, particularly when discussing statistical models or hypothesis testing. These visuals help break down information that might be difficult to grasp from text alone​. Each chapter concludes with a summary of key points and practice problems, which helps reinforce the material. These summaries distill the chapter’s content into its most important takeaways, making it easier for students to review and retain information.

Consistency rating: 5

The book is internally consistent in its use of terminology, structure, and instructional approach. The book follows a predictable structure in each chapter, starting with a theoretical introduction, followed by practical implementation in R, and ending with summaries and practice problems, which creates a cohesive learning experience. The integration of R, particularly through the tidyverse package and ggplot for data visualization, remains uniform throughout, allowing students to apply similar coding logic across different statistical problems. Additionally, the progression of topics is logical, moving from basic to more complex methods, ensuring that concepts build upon one another smoothly. The use of visual aids, R code snippets, and consistent pedagogical tools like summaries and exercises further reinforces this consistency, making the text an effective and coherent resource for intermediate statistics students.

Modularity rating: 5

The book is highly modular, making it easy to divide into smaller reading sections that can be assigned at various points in a course. Each chapter is broken down into distinct sections with clear subheadings, allowing instructors to assign specific topics such as hypothesis testing, ANOVA, or regression independently without requiring students to read entire chapters at once. The text is not overly self-referential, so while there is a logical progression in the order of topics, each section can stand on its own, enabling flexible realignment with different subunits of a course. This modularity ensures that instructors can adapt the material to their specific course structure, assigning sections based on the needs of their students without causing disruption or confusion. Moreover, each section includes self-contained explanations, examples, and R code, making it straightforward for readers to engage with the material in smaller, manageable portions.

Organization/Structure/Flow rating: 5

The book is logical and clear, with topics presented in a structured progression that guides the reader through increasingly complex statistical methods. The book begins by reviewing foundational concepts, such as summary statistics and basic hypothesis testing, before moving into more advanced techniques like ANOVA, regression, and bootstrapping. This step-by-step approach allows students to build on their knowledge incrementally, ensuring they grasp the fundamental ideas before tackling more sophisticated models. Each chapter flows naturally into the next, and within each chapter, concepts are introduced with clear explanations, followed by practical R examples and visualizations. The consistent chapter structure, which includes summaries, key R code, and practice problems, reinforces the logical presentation of topics. This clear and methodical organization makes the text easy to follow, ensuring that readers can engage with the material without confusion.

Interface rating: 5

The book is free from significant issues, providing a smooth and user-friendly experience for readers. The text is well-formatted, with clear navigation and distinct sections, making it easy for readers to find specific topics or chapters. Additionally, hyperlinks to external resources or references function properly, and the layout—whether in print or digital form—remains organized and easy to follow, avoiding distractions or confusion for the reader. This well-executed interface design helps enhance the learning experience by keeping the focus on the content.

Grammatical Errors rating: 5

The text contains no grammatical errors.

Cultural Relevance rating: 5

The real data set in the textbook are inclusive of a variety of backgrounds and disciplinary.

Comments

I will use this book for my statistical methods class.

Reviewed by Claire Seungeun Lee, Associate Professor, University of Massachusetts Lowell on 6/4/24

Comprehensiveness rating: 5 see less

The book is comprehensive, covering a wide range of topics in detail. It provides thorough explanations, examples, and practical applications, making it an invaluable resource for readers seeking an in-depth understanding of the subject matter.

Content Accuracy rating: 5

The book presents accurate information on a variety of statistical methods, ensuring that readers receive reliable and up-to-date knowledge.

Relevance/Longevity rating: 5

Potential necessary updates to the book will be driven by developments and changes in the different versions of the R language. As R evolves, certain functions, packages, and methodologies may be updated or deprecated, necessitating revisions to ensure the content remains current and accurate.

Clarity rating: 5

The concept of each statistical method is clearly outlined and thoroughly explained. This ensures that readers gain a solid understanding of the principles behind each method, facilitating better comprehension and application of the techniques discussed.

Consistency rating: 5

The book maintains a consistent style throughout, employing uniform elements such as R plots, table formats, handwritten methods, and visualizations. This cohesive approach ensures that readers can easily follow and understand the material, regardless of the chapter or section they are reading.

Modularity rating: 5

Each section is a separate unit, but it builds upon the previous one, ensuring a comprehensive understanding of the material for readers with different levels of expertise.

Organization/Structure/Flow rating: 5

This book is well-organized, making it a versatile resource suitable for learners at various stages. It is structured to accommodate basic, intermediate, and advanced levels of study.

Interface rating: 5

The book features a user-friendly interface, enhancing the reading experience.

Grammatical Errors rating: 5

No grammatical errors are found in this book.

Cultural Relevance rating: 3

As the textbook focuses on statistics, cultural relevance is not emphasized. But I don't think there are some biased examples.

Comments

Thank you!

Reviewed by Wei Wei, Professor, Metropolitan State University on 2/23/24

Comprehensiveness rating: 4 see less

The text covered both analysis of variance and regression, the two major areas in statistics, along with some categorical analysis method, Chi-square tests. The author introduced each chapter in detailed approaches. The author even wrote their own R package for some of the analysis. This text did not cover the nonparametric analysis area, which is helpful for analyzing small data. It will also be helpful if the author can offer more exercise questions at the end of each chapter.

Content Accuracy rating: 5

The text was written in very good quality. I did not find any bias on any of the analysis or explanations.

Relevance/Longevity rating: 5

The content is very relavant. The author used most up-to-date R functions and packages. Some of the datasets used in the examples are relatively out-dated, but those data work well to help the readers to understand the materials.

Clarity rating: 5

The text is very clearly written. All the R code and output are provided. The author did very good job of explaining the models and results. The author structured each chapter really well, in terms of model validation, data visualization, statistical analysis, and explanation of the results.

Consistency rating: 5

The entire text is structured consistently. All the formats of the R code and output are consistent throughout the text.

Modularity rating: 4

Adding subtitles to each section in each chapter can better guide the readers through the materials. Overall, the text has good modularity with dividing the complicated materials into smaller sections.

Organization/Structure/Flow rating: 5

The text flows well with the consistent organization for each chapter: introducing the concepts, validating the model , analyzing data with visualizations, and interpreting the results.

Interface rating: 5

There is no significant interface issues throughout the text. Some of the hand-written illustrations/pictures can be in better quality. But those are very minor issues and there are not many of those types of illustrations.

Grammatical Errors rating: 5

Very good quality.

Cultural Relevance rating: 5

All the examples used in this text are free of culturally sensitive issues.

Comments

This is a very well written text. Enjoyed reading it through. One comment: the author used one of their own R package, catstats, that the readers will need to install it from their GitHub repository. Hope the author will keep this package updated so the readers can continue accessing to it.

Reviewed by Eric Smith, Professor, Virginia Tech on 3/6/22

Comprehensiveness rating: 5 see less

The text covers material for a one-semester long class on applied statistics. I teach a course on biological statistics and it covers most of the areas that I teach in my course. Some topics I spend some time on are slightly covered – ANCOVA, randomized blocks, mixed models. Some material on logistic regression might be useful. However, I can easily add more material using my notes and examples. I found the material to be current, accurate and free of errors. There is also a nice set of case students for the students to look at to get a sense of what real data is like. I tend to add two things to my course - first I make a number of "how do I" examples with screenshots to help students navigate R and Rstudio. Second I have students read or try to reproduce analyses in publications so they can get a sense of "good" and "bad" analyses. The case studies at the end of the text more or less do this as well so I do not see this as a problem with the text.

Content Accuracy rating: 5

The writing is good quality and I did not find any errors or coding problems. I like the summary at the end of each chapter about highlights and coding.

Relevance/Longevity rating: 4

The book is up to date and the material is quite relevant and current. It will have to be updated regularly as it is based on R (which changes regularly). The inclusion of tidyverse is both good and bad in that some students struggle with this topic.

Clarity rating: 5

The authors have done a good job at explaining statistical analysis and the tools that are available to evaluate data. It would be nice for the authors to add some comments in the coding to help students recall what is being done although again this something easy for me to add. Comments might, for example, help explain some of the ggplot coding. I have found that coding in R is the greatest issue in my course so have numerous examples for them to follow.

Consistency rating: 5

The text is consistent in terminology and the chapters have a similar structure. Again, I like the summaries at the end of each chapter.

Modularity rating: 5

The text is free of significant interface issues, including navigation problems, distortion of images/charts, and any other display features that may distract or confuse the reader. It is quite easy to find topics and link to specific chapters. There might be some minor issues moving chapters around as the examples use code and some of the explanations are in other chapters. This is a rather minor problem though.

Organization/Structure/Flow rating: 5

I have a slightly different ordering of topics in my course however, it would not be a problem to teach most of the chapters in a different order although there are some obvious chapters that you want to teach in sequence (one-way ANOVA before two-way ANOVA or SLR before MLR). I tend to teach simple regression and correlation before ANOVA so that I can better explain the estimates that are produced in the ANOVA summary. Again, this is not a problem as the SLR chapter is relatively independent of the other chapters and can be taught earlier. Since the general structure of the text is quite sound, I can easily supplement the text with additional examples and material. The case studies section is especially useful.

Interface rating: 5

The authors used bookdown to prepare the text. I believe (once the first edition is done) this makes it easy to update, produces a consistent set of graphics and helps with navigation. To update should also be relatively easy.

Grammatical Errors rating: 5

The book has been through a number of editions so the grammar/spelling etc is good.

Cultural Relevance rating: 5

I did not find any issues.

Comments

I look forward to using this as the main text the next time I teach my class. It is a better text than many of the others that involve a first course in statistics using the R package.

Reviewed by Evidence Matangi, Assistant Professor, Taylor University on 11/15/21

Comprehensiveness rating: 4 see less

R introduction is concise, data sets introduction is clear with objectives formulating the the questions to answer from the data.
Plots for categorical variables would compliment those for quantitative variables. Grammar of graphics is condensed, more would be better for an introductory textbook. Data wrangling needs to talk about the 5 verbs and also briefly talk about piping. The ASA statement on p-values would be a great addition to this textbook. Briefly introduce loops in R.

Content Accuracy rating: 5

There is some missing information on page 97 on second bullet point for SS_Total.
page 154, Discussion section needs to be distinct as you comment on it earlier for reference.

Relevance/Longevity rating: 5

The content is so up-to-date and captivating addressing interesting problems to solve using data.

Clarity rating: 4

There is need to be intentional in introducing the R packages and its associated jargon. Distinguish between random and fixed effects and also highlight other multiple comparison tests. For example, introduce functions such as favstats and their normal associatives for the five number summary.

Consistency rating: 5

The flow from chapter to chapter and within chapters is very consistent, leading to the practice problems. The structure of the practice problems is very consistent throughout the textbook.

Modularity rating: 5

This is a typical set-up for a Statistics textbook. However, there is need to delve deeply on modern statistics topics such as bootstrap resampling and confidence intervals.

Organization/Structure/Flow rating: 5

Great organization of the topics.

Interface rating: 5

the outputs from R and their associated visuals add more value to the theoretical aspects.

Grammatical Errors rating: 5

great communication, it is very clear.

Cultural Relevance rating: 5

Awesome text.

Comments

This is a good textbook for intermediate modern statistics. Thank you for this open resource for equality in access to education.