Conditions of Use
This textbook provides appropriate coverage for an intermediate-level course in applied statistics. The first five chapters address content that would typically be covered in an introductory statistics course for students outside of mathematics,... read more
This textbook provides appropriate coverage for an intermediate-level course in applied statistics. The first five chapters address content that would typically be covered in an introductory statistics course for students outside of mathematics, although these chapters do not provide as much depth as would be needed for that course level. Chapters five through eight are typically addressed in intermediate-level biometry courses and provide appropriate coverage for understanding the basics of the tests involved. Appropriate design of experiments and related topics such as fixed vs. random variables are not considered in sufficient depth to guide readers on these topics. The two final chapters provide an introduction to statistical models and quantitative measures specific to forestry, natural resources management, and ecology. No glossary or index was provided; this would have been helpful for readers seeking to refresh their knowledge of statistical terms or locate specific content in the main text.
Overall the content is consistent with my prior training in statistics and biometry, with one major exception. Chapter 6 (Two-Way ANOVA) presents the analysis of statistically significant interactions in a dramatically different manner from my prior training and does not discuss any history of or justification for the change. More specifically, this chapter dispenses with the "slicing" of statistically significant interactions, wherein one main effect is evaluated by first nesting it within each level of the second main effect and carrying out a one-way ANOVA for each level (see, for example, Damon & Harvey, 1987). The remaining degrees of freedom allow the second main effect to be evaluated overall. Instead, this text recommends ignoring both main effects when the interaction is statistically significant, applying multiple hypothesis testing to all possible combinations of the levels for each main effect. A few typos are present, which is to be expected; fortunately, this only interferes with the understanding of the content itself in a single example scenario in Chapter 1.
The book provides foundational knowledge in conventional descriptive and inferential statistics that is unlikely to change substantially in the future. Quantitative models and measures described in the final chapters appear to be largely canonical and therefore maintain relevance even as research in these areas progresses. However, the book makes extensive use of Minitab and Excel software packages to provide examples. If new versions of the software change substantially, then the walk-through examples found in each chapter and the lab exercises in the appendices would require updates.
A major advantage of this textbook is that it presents the statistical workflow in multiple ways, especially in early chapters. I appreciated the use of visual information to depict the comparison of critical values with test statistics and the presentation of critical value, p-value, and confidence intervals back-to-back in multiple examples. These are all complementary views of the same information, so if a student only understands one of these perspectives, this can then be leveraged to help them better understand the other two. The language is generally accessible, although presentation is often confusing. Only some new vocabulary terms are highlighted and defined immediately upon their first introduction. Some terms are defined upon first introduction but not highlighted. Others are not highlighted and a definition is provided a few pages later. For a reader new to the topic, this would likely be confusing. Direct references to figure and table numbers would also help readers in sections where multiple graphs are presented. Finally, section numbers within chapters are numbered independently of the chapter, which could lead to unnecessary confusion.
In general, topics with similar themes are presented consistently. In addition, each chapter shows at least one numerical example to help readers grasp the mathematical equations presented, then notes later on how Minitab software and MS Excel software can be used to carry out the analysis. If the software is not equipped to carry out the analysis, this is stated directly in the text. As previously noted, the textbook is somewhat inconsistent in drawing the reader's attention towards and defining new vocabulary terms. In addition, the text sometimes presents the order of the "triple view" of statistical inference (critical value, p-value, confidence interval) differently in different sections or chapters.
The modularity of this text is excellent, due in part to the consistency of the author's approach in providing theory, underlying assumptions, numerical and visual examples, and examples of software usage. It would easily be possible to remove chapters for particular audiences, and it would be easy to supplement particular topics with outside material.
The organization, structure, and flow are presented logically and clearly in most cases. I have already described two exceptions to this rule, which also related to the text's clarity (e.g., vocabulary terms) and consistency (e.g., perspectives on statistical inference).
In my experience, many of my students prefer to print their e-texts. They have told me that this allows them to hand-annotate the materials and provides a learning-specific context for focusing without the distractions of social media. Based on that prior feedback, I chose to print the text in grayscale using a double-sided format, with two pages per side while reviewing. The only limitation I came across was the use of color in some figures in the early chapters.
I noticed only one or two grammatical errors.
The only item that may hold explicit cultural relevance in this text is a statement in the first chapter that gender and race are examples of qualitative variables. All other examples relate to nonhuman species and represent scientific or natural resources questions embedded within a profession that has historically been predominantly white and male. It would have been forward-thinking, and more directly relevant to a greater proportion of the students I currently teach, if the author had included examples relevant beyond this perspective (e.g., natural resources questions relevant to indigenous peoples or agricultural questions outside the industrial monoculture model).
This easy-to-read text covers the areas and principles of the subject appropriately and provides appropriately related lab/practical exercises. read more
This easy-to-read text covers the areas and principles of the subject appropriately and provides appropriately related lab/practical exercises.
Content is accurate, error-free and unbiased.
The text content is up-to-date and orchestrated so that future updates will be pretty straightforward to carry out.
The text is well-written with many examples that help to elucidate concepts.
The text is written in a consistent manner.
The book is logically and predictably divided/partitioned by chapters related to topic.
Subject areas of the text are presented in a logical manner.
Helpful display features and lab details.
No noted grammatical errors.
The book is relevant to individuals regardless of background.
I applaud an open-source text on intro to statistics.
Table of Contents
- Chapter 1: Descriptive Statistics and the Normal Distribution
- Chapter 2: Sampling Distributions and Confidence Intervals
- Chapter 3: Hypothesis Testing
- Chapter 4: Inferences about the Differences of Two Populations
- Chapter 5: One-way Analysis of Variance
- Chapter 6: Two-way Analysis of Variance
- Chapter 7: Correlation and Simple Linear Regression
- Chapter 8: Multiple Linear Regression
- Chapter 9: Modeling Growth, Yield, and Site Index
- Chapter 10: Quantitative Measures of Diversity, Site Similarity, and Habitat Suitability
About the Book
Natural Resources Biometrics begins with a review of descriptive statistics, estimation, and hypothesis testing. The following chapters cover one- and two-way analysis of variance (ANOVA), including multiple comparison methods and interaction assessment, with a strong emphasis on application and interpretation. Simple and multiple linear regressions in a natural resource setting are covered in the next chapters, focusing on correlation, model fitting, residual analysis, and confidence and prediction intervals. The final chapters cover growth and yield models, volume and biomass equations, site index curves, competition indices, importance values, and measures of species diversity, association, and community similarity.
About the Contributors
Diane Kiernan completed her Ph.D. in quantitative methods in forest science at SUNY ESF in 2007. She is currently teaching Introduction to Probability and Statistics and Forest Biometrics at SUNY ESF and Advanced Statistics at LeMoyne College in Syracuse, New York. She is employed as a biometrician analyzing long-term re-measurement data for the SUNY ESF forest properties and is involved with additional research projects at SUNY ESF. Diane has authored and co-authored two previous books on statistics currently being used in her classes.