I am looking forward to using the 3d edition of Millers Principles of Everyday Behavior
Analysis in my senior Applied Behavior Analysis course this spring. I like the study
questions. They seem to be especially well crafted for this edition and I am going to enjoy
hearing my students echoing them and then using their textual or thematic equivalents to
compose new phrases . (I'm not trying to be a verbal behavior wonk here. It seems to be
just exactly what happens with this type of textbook.) I have some reservations about the
answers to the text questions, reading quiz, and examples being included in the book. I
personally prefer the format introduced by Whaley & Malott (1971)1 and recently favored
by others (e.g., Sulzer-Azaroff and Hall, 2 Martin and Pear, 1992,3 Miltenberger, 1974,4)
in which the students can find the answers only by reading. browsing re-reading, picking,
choosing, or , better yet, asking the instructor or other students. But this has never been a really serious problem.

The greatest improvement that I can see in the 3d edition is that every study and/or
test question requires a tact ("It is [thus]. . . .") as an answer. In earlier editions the
answer key allowed autoclitics ("There is not enough information given here to answer
this question [in my opinion]") to serve the same end. I am uncomfortable with the latter
for the same reason that I am uncomfortable with the use of distracters in the case of
multiple-choice or true-false tests: to say what you know not to be true is to describe
your own history or personal condition rather than the controlling environment
What Keith's book certainly should accomplish, in addition to teaching a heck of a lot of
good behavior analysis, is to prompt and reinforce the use of weekly-quiz formats in
college teaching. As one behavior analyst put it in a recent note to mourn the passing of
Fred Keller, "I am fully aware of the difficulty of starting PSI courses and the likelihood of
experiencing aversive consequences for doing so" ( Sidman, 1996)5. Millers 3d Ed.
should help all of us preserve the vitality of Dr. Keller's invention.

1. Whaley, Donald L., and Malott, Richard W. (1971) Elementary Principles of Behavior
. Englewood Cliffs, Prentice-Hall.
2. Sulzer-Azaroff, B. and Hall, L. J. (1991). Study Guide to Accompany Behavior
Analysis for Lasting Change. Fort Worth: Holt, Rinehart and Winston.
3. Martin, G and Pear, J. (1992). Behavior Modification: What it is and How to Do It (
Fourth Edition). Englewood Cliffs: Prentice Hall.
4. Miltenberger, R. (1997). Behavior Modification: Principles and Procedures.
Belmont: Brooks/Cole.
5. Sidman, M. (1996) Fred S Keller: Rememberings. Journal of the Experimental
Analysis of Behavior, 66, 1-6

Keith Miller's Principles of Everyday Behavior Analysis (PEBA) is an unusual text because
it was constructed using the behavioral principles it attempts to teach. In addition, the
book was experimentally evaluated by the author before it was published. For those
unfamiliar with it, here is a brief description.

PEBA was initially published in 1975. It attempts to teach 68 behavioral concepts,
organized into 25 lessons. Four lessons are review. Lessons are sequenced such that
material from prior lessons is also included in the next lesson. Each lesson involves three
steps. First, the student reads up to five pages of material and then takes a quiz over the
material. Answers to this quiz are provided in the text. Next, the student studies a series of
20 programmed examples which apply the concept to fictional situations and answers
questions about the examples. Answers to these examples are also given. Finally, during
class, the student is given one of a set of 3 quizzes (printed in the text but without
answers) which consist of 11 fill-in-the-blank questions.

Miller's book teaches basic concepts such as reinforcement, data evaluation,
discrimination, shaping, fading, extinction, and more. It describes applications of the term
in everyday situations and teaches how to discriminate one behavioral concept from
another. Perhaps most importantly, the format requires students to be actively involved
with the text. The following describes how three instructors have used the text along with
a brief critique of the 3rd edition. I will describe my own procedure first.
My class meets twice a week for one and one quarter hours. The class buzzes with
conversation over possible answers to quiz questions before class begins. The 11 item quiz
may be given at either the beginning or end of class. This unpredictability ensures
attendance and takes about 15 minutes. Lectures, films, and discussion take up the
remaining class time. Since there are three versions of each quiz, one particular form of
the quiz is chosen either by the instructor or the option is given to a student who has
contributed to class discussion. Students exchange the quiz in their book with one from
another member of the class to prevent writing answers on the quiz form before it is taken.
Quizzes are immediately corrected in class and handed in to be recorded. In order to
encourage optimum initial performance, quizzes taken in class are worth one more point
than those taken in the TA's office. Students who miss more than one question are
required to take a make-up quiz within two weeks. A perfect score on the first make-up
quiz, adds one point to the original test score. Overall grades are determined by a point
system with an A equating to 95%, B 90%, C 81%, D 70% and below that an F. There is
also a final exam which includes questions over the text and lecture material.

Student Performance
Class attendance is high - about 90%. Few students fall behind. Of 1225 quizzes given to
49 students in Fall 1995, 13% were retaken. Concerns over frequent quizzes are offset by
those who delight in knowing quiz questions in advance. Students like the brevity of the
lessons and the pace of the material.

A sampling of grades over a 16 year period showed that 69% of students received A's,
22% B's, 5% C's, and 4% D's, F's, or Incompletes. Students frequently comment that they
believe they have learned more under this format than using a non-programmed text with
fewer exams. They also indicate that they remember the material better when it comes to
taking GREs or exams in related courses.

A survey of responses in 1991 and 1995 showed the following: (n=89) using a 5 point
Likert Scale with 1, easy to understand and 5, difficult to understand, students gave the
text a mean score of 1.66.

In terms of how interesting the text was, with 1, very interesting and 5, boring, the mean
score was 2.4. In terms of how much the students felt they learned with 1, more than
other classes and 5, less than other classes, the mean was 1.67.
In terms of the value of the material learned, with 1, very valuable and 5, little value, the
mean was 1.75.

In terms of how much the students found themselves thinking and talking about the
subject matter of this class compared to other classes, with 1, much more and 5, much
less, the mean score was 1.77. PEBA is a useful book for students beginning to learn
behavior analysis. Its' format models what it attempts to impart to the student. It is a text
that teaches.

References: Miller, L. K. (1996). Principles of everyday behavior analysis (3rd Ed.) Pacific
Grove, CA: Brooks/Cole.Col

Teaching behavior analysis offers a unique opportunity to design a course that explicitly
models the concepts and principles students are expected to master. This presentation
describes such a behavior analysis course taught at an open admission, Historically Black
Land Grant College. Special emphasis is placed on how Miller's Principles of Everyday
Behavior Analysis (PEBA) contributes to that effort. The course is the first required
course for Psychology majors and minors. In addition to teaching basic behavior analysis
concepts, it contributes to computer literacy and technical writing components of the
curriculum.

PEBA is covered using a modified PSI format in which mastery (zero or one errors in 11
questions) of each unit is required for progressing to the next but points are only awarded
for unit mastery achieved on or before scheduled deadlines. Lesson tests begin when the
student presents blank lesson tests. One form selected at random is taken, and graded
immediately. Students are asked to orally defend at least one of the answers to a test,
including any questions answered incorrectly. The interview begins with the student
defining their answer, then identifying the relevant parts of Miller's question that
correspond to the key parts of concept or procedure they provided in their answer.
Where an incorrect answer was provided, this gives the student the opportunity to
"discover" the absence of a key element of the definition. In this case the student is
prompted to provide and define an alternative answer, and find its key parts in the
example. Successfully defended answers are counted as correct.

Pretest-posttest results using alternate forms of Miller's Lesson 25 Review Exam show
mean gains in excess of 70% + 4% (standard error) in each of the last five semesters.
Well-functioning PSI courses are easy to identify. They produce J-shaped or U-shaped
final grade distributions. Uniform or normally distributed final grade frequencies generally
indicate that mastery is not being achieved. J-shaped grade distributions in my course
suggest that students are achieving mastery. Almost all of the F grades were awarded to
students who withdrew from the course but neglected to inform the university they had
done so. Students consistently rate PEBA as one of the most positive features of the
course.

Request full text of this presentation by e-mail:
wmbeneke@mail.coin.missouri.edu