Q: The text covers all areas and ideas of the subject appropriately and provides an effective index and/or glossary
The text covers most of the areas that would normally be included in an introductory course with a few exceptions that I will note later. The index is definitely not effective and I feel that the glossary, while complete, needs revision.
The only major topic that is omitted is experimental design but that is not an important omission unless the course is for science or social science students. There is no section on ethics but very few Statistics texts include such a section. Probability plots are not covered and the chapter on regression makes no reference to residual plots which is highly unusual.
In my opinion the biggest thing this textbook is missing is motivation for studying statistics. Statistics plays a huge part in trying to answer many important questions and this text gives little or no indication of this. The examples and problems generally deal with uninteresting questions predominantly with made up data. Even when the data is real there is rarely any motivation given or apparent reason to analyze it. Here is an example (pages 398-399) from Chapter 9, Hypothesis Testing: Single Mean and Single Proportion which is typical of most of the student generated questions in the chapter.
“NOTE: The following questions were written by past students. They are excellent problems!
18. "Asian Family Reunion" by Chau Nguyen
Every two years it comes around
We all get together from different towns.
In my honest opinion
It's not a typical family reunion
Not forty, or fifty, or sixty,
But how about seventy companions!
The kids would play, scream, and shout
One minute they're happy, another they'll pout.
The teenagers would look, stare, and compare
From how they look to what they wear.
The men would chat about their business
That they make more, but never less.
Money is always their subject
And there's always talk of more new projects.
The women get tired from all of the chats
They head to the kitchen to set out the mats.
Some would sit and some would stand
Eating and talking with plates in their hands.
Then come the games and the songs
And suddenly, everyone gets along!
With all that laughter, it's sad to say
That it always ends in the same old way.
They hug and kiss and say "good-bye"
And then they all begin to cry!
I say that 60 percent shed their tears
But my mom counted 35 people this year.
She said that boys and men will always have their pride,
So we won't ever see them cry.
I myself don't think she's correct,
So could you please try this problem to see if you object?”
I am not sure what hypothesis I am being asked to test here. I would certainly disagree with it being described as an excellent problem. While many of the student generated problems are similar to this one there was one about the endings of Japanese girl’s names (9.16.25 Page 402) that I found quite interesting.
The index clearly had little or no human input. As well as reasonable entries the index includes a host of random words. For example, the index includes 80 references for the word “elementary” and 186 references for the word “statistics”. It also includes references for many words such as “answer”, “box”, “word”, “good” and “two” that should not be in any index.
I would rate the glossary as somewhat effective. The glossary is fairly complete but I believe that many of the entries should be rewritten. It includes some minor errors such as the definition of a geometric distribution “The probability of exactly x failures before the first success is given by the formula: P (X = x)= p (1- p)^(x-1).” In at least one case an entry is given with no definition.
Some of the other definitions are somewhat unclear. For example:
An observation cannot fall into more than one class (category). Being in more than one
category prevents being in a mutually exclusive category.
Standard Normal Distribution
A continuous random variable (RV) X~N (0, 1) .. When X follows the standard normal
distribution, it is often noted as Z~N (0, 1).
Other definitions just don’t match my preferences. For example the definition of correlation includes the so called computational formula which I feel doesn’t belong in any statistics textbook. I also didn’t like the definition of “Random Variable” being given under the heading “Variable”. Doing that accentuates the confusion between a variable in algebra and a random variable in probability.
Comprehensiveness Rating: 3 out of 5
Q: Content is accurate, error-free and unbiased
The content is generally accurate and unbiased, although I am not sure what a biased statistics text would look like. There are some errors such as the previously mentioned definition of the geometric distribution which is not much more than a typo and the occasional more serious error such as the statement: “True random sampling is done with replacement.” on page 20. In my opinion, virtually every graph in the chapter on graphing is done badly but they are not really errors.
Content Accuracy Rating: 4 out of 5
Q: Content is up-to-date, but not in a way that will quickly make the text obsolete within a short period of time. The text is written and/or arranged in such a way that necessary updates will be relatively easy and straightforward to implement
This text is a mix, up-to-date in some ways, quite old fashioned in others.
It makes good use of graphical calculator technology using the calculator to calculate probabilities rather than using antiquated tables although the tables are still included if an instructor prefers to use them. It also uses the graphical calculator in all aspects of statistical analysis. If you are convinced that a graphical calculator is the best technology to use when teaching introductory statistics, this is one of the primary strengths of the text. The fact that it includes no other technology is a weakness. For example, the text gives long detailed instructions for creating frequency tables and histograms from scratch. I do not feel that this section was done well and even done well it should have disappeared 30 years ago.
The text correctly indicates that the normal approximation to the binomial is no longer necessary with the technology that is currently available. However it then uses the same normal approximation when doing inference with proportions. While this is still the norm for introductory classes and should probably be included, it would have been nice to include a justification for using the normal approximation after saying it isn’t necessary.
One of the first sections I look at when I review a text for possible adoption is the section on comparing means using independent samples. The more modern texts use the Welch’s t-test. That is the test used by this text so for me that is a positive. However it follows that section with a long section using the assumption that the variances are known. The variances are never known so the only justification for including such a section is as a lead-in for Welch’s t-test. In that case it should be much shorter and should be included first as was done in the single population chapter. While the text indicates “In practice, we rarely know the population standard deviation.” (I would replace rarely by never) it devotes more space to the case when the variance or variances are known than when they are unknown.
I also check to see if the text differentiates between large and small sample inference for means since there is no reason to do so. This text does not differentiate and it says why which is another plus.
As I have mentioned before, this text gives very few examples of what statistics is being used for. Since few of the examples or problems are topical, it will take them a long time to become dated. I would consider this to be a minus but in the context of this question it might be considered a plus.
The textbook is written in a way that updates and revisions will be straightforward to implement but in my opinion, so many are needed before I would consider adopting this text that it would not be easy.
Relevance Rating: 3 out of 5
Q: The text is written in lucid, accessible prose, and provides adequate context for any jargon/technical terminology used
The text is written very clearly in some places less so in others. It gives a very clear, step by step set of instructions for taking a small simple random sample from an already given sampling frame. However, no mention is made of how difficult it is to create a sampling frame for a large population and no mention is made of how a large simple random sample could be taken from a sampling frame. It also gives relatively clear instructions on how to create a frequency table and histogram including detailed instructions for calculating the number of bars of width 1 required to graph data consisting of the integers 1, 2 ,3 ,4, 5, and 6. (Spoiler: the answer is 6.) It gives a pretty good job of relating decisions using p-values to the concept of rare events.
Other parts are less clear. My guess is that no one in a class of tourism students would get anything from the chapter on analysis of variance. It contains lots of jargon with very little context. For example, this is how the description of the F test starts out:
“To calculate the F ratio, two estimates of the variance are made.
1. Variance between samples: An estimate of σ^2 that is the variance of the sample means
multiplied by n (when there is equal n). If the samples are different sizes, the variance
between samples is weighted to account for the different sample sizes. The variance is also
called variation due to treatment or explained variation.
2. Variance within samples: An estimate of σ^2 that is the average of the sample variances
(also known as a pooled variance). When the sample sizes are different, the variance within
samples is weighted. The variance is also called the variation due to error or unexplained
While most of the text is written clearly, I feel that a general shortcoming throughout this textbook is that it does not provide sufficient context for the techniques it looks at.
Clarity Rating: 3 out of 5
Q: The text is internally consistent in terms of terminology and framework
The text is consistent in terms of terminology and framework.
Consistency Rating: 4 out of 5
Q: The text is easily and readily divisible into smaller reading sections that can be assigned at different points within the course (i.e., enormous blocks of text without subheadings should be avoided). The text should not be overly self-referential, and should be easily reorganized and realigned with various subunits of a course without presenting much disruption to the reader.
The text is easily and readily divisible into smaller reading sections; it is not overly self-referential and should be easily reorganized to the extent that any statistics text could be.
Modularity Rating: 5 out of 5
Q: The topics in the text are presented in a logical, clear fashion
The organization is similar to most old-school intro stats texts and while it is not the same as what I use I am sure that it conforms to the organization that many instructors use. The only really awkward place that I noticed was introducing box-plots before measures of centre or location. It meant that the authors had to define quartiles and medians in that section and then define them again later. It would be easy to move the section on box-plots after the discussion of quartiles and medians.
Organization Rating: 4 out of 5
Q: 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
I was working from the pdf file so I cannot comment on these issues.
Interface Rating: 3 out of 5
Q: The text contains no grammatical errors
I did not notice any grammatical errors.
Grammar Rating: 4 out of 5
Q: The text is not culturally insensitive or offensive in any way. It should make use of examples that are inclusive of a variety of races, ethnicities, and backgrounds
The text is not culturally insensitive or offensive in any way. The names it uses in its examples are inclusive of a variety of ethnicities.
Cultural Relevance Rating: 4 out of 5
Q: Are there any other comments you would like to make about this book, for example, its appropriateness in a Canadian context or specific updates you think need to be made?
The six recommendations of the GAISE (Guidelines for the Assessment and Instruction in Statistics Education) college report prepared for the American Statistical Association are:
1. Emphasize statistical literacy and develop statistical thinking
2. Use real data
3. Stress conceptual understanding, rather than mere knowledge of procedures
4. Foster active learning in the classroom
5. Use technology for developing conceptual understanding and analyzing data
6. Use assessments to improve and evaluate student learning
This textbook does an excellent job on points 4 and 5. There are many group exercises throughout the text. It is a conscious focus of the text and is its primary strength. The textbook is also based on the use of a graphic calculator. While I feel that it is a poor tool for doing statistics, it is a reasonable tool for use in an introductory statistics class. This textbook does an excellent job of integrating it into the curriculum. This is the other strength of the textbook. However, as I mentioned earlier, I feel that ignoring other technologies is a weakness.
The book also is less successful in stressing conceptual understanding rather than mere knowledge of procedures, point 3. For example, in the chapter on sampling it gives brief descriptions of different sampling methods but says nothing about the conditions under which one method is better than another. It lists possible problems in sampling but gives no context. Another example is that it lists the properties of correlation but doesn’t relate them to data and the only formula given is the computational formula which I feel has no pedagogical value what-so-ever.
It uses some real data but I don’t feel that it uses enough. The real data it uses does involve the students in the collection of data, making that data more relevant and fostering active learning, obviously a good thing. However, it does not include much data that was used to answer interesting questions.
I feel that the critical failure of this textbook is that it doesn’t do a good job of teaching statistical thinking. Far too often it emphasizes how to do questions in a textbook rather than how to do statistics. This is a consistent focus throughout the text. Here are a few examples:
These listed learning outcomes all talk about textbook questions:
“By the end of this chapter, the student should be able to:”
“Classify discrete word problems by their distributions.”(Chapter 4 Page 159)
“Classify continuous word problems by their distributions.”(Chapter 7 Page 281)
“Discriminate between problems applying the normal and the student-t distributions.”
(Chapter 8 Page 319)
As it introduces confidence intervals for proportions it does so in the context of a textbook problem:
“How do you know you are dealing with a proportion problem? First, the underlying
distribution is binomial. (There is no mention of a mean or average.)” (Page 331)
In the discussion of using hypothesis testing to make decisions on page 375:
“A systematic way to make a decision of whether to reject or not reject the null
hypothesis is to compare the p-value and a preset or preconceived α (also called
a "significance level"). A preset α is the probability of a Type I error (rejecting
the null hypothesis when the null hypothesis is true). It may or may not be given to
you at the beginning of the problem.”
When working an example of a test for two means:
“Example 10.1: Independent groups
The average amount of time boys and girls ages 7 through 11 spend playing sports
each day is believed to be the same. … Is there a difference in the mean amount of
time boys and girls ages 7 through 11 play sports each day? Test at the 5% level
“The words "the same" tell you Ho has an "=". Since there are no other words to indicate
Ha, then assume "is different." This is a two-tailed test.”
Another example of the lack of statistical thinking is that while the textbook mentions the assumptions for the various procedures, it never indicates how to assess whether they are reasonable for a particular set of data. The only assumption checking it does is again based on textbook questions rather than data.
For example (Page 381):
Statistics students believe that the mean score on the first statistics test is 65.
A statistics instructor thinks the mean score is higher than 65. He samples ten
statistics students and obtains the scores 65; 65; 70; 67; 66; 63; 63; 68; 72; 71.
He performs a hypothesis test using a 5% level of significance. The data are from
a normal distribution.
“Distribution for the test: If you read the problem carefully, you will notice that
there is no population standard deviation given. You are only given n = 10 sample
data values. Notice also that the data come from a normal distribution. This means
that the distribution for the test is a student’s-t.”
Since the data are given for the question, the decision on whether to use a t-test should be based on the data, not artificially given in the statement of the question.
While this textbook does an excellent job of integrating graphical calculators and includes a large number of collaborative exercises it does not come close to matching my needs for a textbook for an introductory statistics course. I feel that the first three recommendations of the GAISE college report are all critical and I do not believe that this textbook adequately addresses any of the three. I personally would not consider adopting it without extensive revision.