A COMPARISON OF AUDIO ANO WRIITEN
PROMPTING IN THE TRAINING OF PERCEPTUAL
‘ MOTOR SKILLS

DiSsertatIon for the Degree Of Ph. D. '
NIOIIIOAN STATE UNIVERSITY '
GEORGE FRANCIS SARGENT
1975 ,

 

fmlwymm.uwn‘;,
.

.' v‘fyfi,n'~"
’ ‘32. - .-'

M i ”5:“! \ml-

1.
U Jam‘rsm'

This is to certify that the

thesis entitled

A COMPARISON OF AUDIO".AND WRITTEN
PROMPTING IN THE TRAINING OF PERCEPTUAL
' MOTOR SKILLS
l’

 

_ ,_ presented'lby
GEORGE FRANCIS SARGENT

has been accepted towards fulfillment
of the requirements for

Ph.!) degree in Education

 

 

Q

Date_ 2 20 74

 

0-7639

 

 

 

 

II II II II III IIII II II III IIII IIII II IIII IIII IIII

3 1293

 

 

ABSTRACT

A C(MPARISON OF AUDIO AND WRITTEN PROMPT ING
IN THE TRAINING OF PERCEPTUAL IVD'IUR SKILLS

By
George Francis Sargent

The objective of this study was to compare the relative
effectiveness of audio and written prompting given to students as they
were learning perceptual motor tasks. Tasks of interest here were those
a learner can usually master with minimum practice once he has acquired
the necessary knowledge, as opposed to those such as unicycle riding, or
piano sight reading, which require a high ratio of practice to knowledge
fer mastery.

The primary research question was: Does audio prompting make a
worthwhile contribution to student learning of perceptual motor tasks,
or in other words, are written instructions just as good?

The perceptual-motor task chosen for the experiment was to load,
thread, record and playback the recording using an Ampex 5100 Video Tape
Recorder.

The first part of the experimental treatment consisted of the
subject watching an 11 minute training video tape that taught him all he

needed to know about operating the Ampex 5100 video tape recorder.

George Francis Sargent

Immediately after watching the training tape the subject received hands—on
practice. unknown to the subject, he was assigned to one of three treat-
ment groups, audio, written, or control. He then completed three repeti-
tions of loading, threading, recording and playing-back his recording on
the 5100. During the first two repetitions the subjects in the audio and
written treatment groups received prompts. In the first repetition these
prompts were extensive--much like complete instructions or tutoring.
During the second repetition, the prompts were much less frequent and
shorter. There were no prompts during the third repetition for any group.
The audio group received these prompts via a cassette tape recorder that
stopped after each prompt until restarted by the subject with a fbot switch.
The prompts fer the written group were virtually identical to the auditory
ones except they were written on five pages of paper.

During each repetition of the experiment the observer (1) timed it,
and (2) counted the number of times the subject requested assistance from
the observer (which he gave). After the third repetition, the subject was
administered a Likert type attitude test to measure his attitude towards
the prompting he received during this hands-on period.

The dependent variables of the study were (1) the number of times

assistance was provided, (2) time to completion, (3) efficiency (computed

 

 

from.the time variable), and (4) the sUbject's attitude towards the

prompting. The assistance, time, and efficiency dependent variables were

 

 

analyzed via a one-way ANOVA with repeated measures at the alpha equals
0.10 level. The attitude dependent variable was analyzed via a one-way

ANOVA.

George Francis Sargent

The main conclusions of the study fellow.
The subjects were unable, in most cases to effectively learn
how to operate the video tape recorder from.only watching the
11 minute training tape.
Both the audio and.written groups used similar amounts of

.Assistance and Time, and had similar.Attitude measures.

 

In the early stages of learning a skill task, audio instructions

have higher Efficiencies than their written counterparts, i.e.,

 

they are more quickly implemented.

Audio instructions for perceptual-motor tasks are gasigg_to
understand than their written counterparts. This is because
the learner can keep his eyes on the equipment as its operation
is being explained.

The main recommendations of the study fellow.

When developing prompting materials fer use in the training of

perceptual-motor tasks, and your objectives are to:

 

A. minimize outside assistance required by the user, or
B. minimize overall Lung required by the user, or

C. have favorable learner attitude

Then:

1. If the training period is short (less than one hour)--Audio
and written prompting work equally well.

2. If the training period is prolonged (more than one hour,
perhaps done every day)--Audio prompting is likely to be

superior to written prompting.

A.COMPARISON OF AUDIO AND WRITTEN PROMPTING
IN THE TRAINING OF PERCEPTUAL MITOR SKILLS

By

George Francis Sargent

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

College of Education

1975

ACIGNIOWLEDGIENTS

The writer wishes to express his appreciation to several persons,
who not only made this dissertation possible, but made my entire doc—
toral program an enriching experience.

I thank Dr. Allan Abedor for many hours overseeing not only
this dissertation, but my comprehensives and doctoral program as well.

I also acknowledge Dr. Norman Bell, who not only guided this
.study, but provided many an inspiration during my years of graduate
school.

Additionally, I thank Drs. Andy Porter and John Schewitzer, who
provided me the opportunity to gain much insight and practical experi-
ence into the research process.

I also thank Dr. Jim Page for his assistance on my doctoral
conmittee and for providing the equipment used in this research.

And, I thank Dr. Paul Witt, the Grand Old Man of the IDGT
Department, who really cares for all of "his" students.

Now, aclmowledgments go to my family, beginning with my brother,
Jim, who has always encouraged me and set the example with his own
dissertation.

I give special thanks to my wife, Jackie, who is always there
with support when I need her.

Finally, my sincere appreciation goes to both of my parents.

ii

TABLE OF CONTENTS

Page

LIST OF TABLES . . . . . . . . . . . . . . . . . . V

LIST OF FIGURES . . . . . . . . . . . . . . . . . . Vi
Chapter

I. STATEMENT OF PROBLEM . . . . . . . . . . . . . . 1

Overview of Chapter I . .

Statement of the Problem .

Proposed Solution . .

Purpose of this Research. . . .
Research Question and Hypotheses . .
Definition of Important Terms . . .
Overview of Other Chapters in Study

OVC‘U'IUJI-‘H

II. REVIEW OF RELATED LITERATURE . . . . . . . . . . . 10

Overview . . . . . . . . . 10
Audio- Tutorial Teaching System . . . . . . . . 10
Stimulus Presentation . . . . . . . . . . lS
Learner Response to Stimulus . . . . . . . . . 19

Summary . . . . . . . . . . . . . 24

III. RESEARCH DESIGN AND PROCEDURES . . . . . . . . . . . 26

Design. . . . . . . . . . . . . . . 26
Design Over Time . . . . . . . . . . . 26
Experimental Treatments . . . . . . . . . 28
Subjects . . . . . . . . . . . . . . 30
Instrumentation . . . . . . . . . . . . 30
validity Concerns . . . . . . . . . . . . . 35

variable Matrices . . . . . . . . 40
Statistical Model of Analysis . . . . . . . 41
Research Hypotheses . . . . . . . . . . 42

Procedures . . . . . . . . . . . . . . . . 43

iii

Chapter Page
IV.ANALYSISOFRESJLTS..............47

Findings . . . . . . . . . . . . . . 47
Student Perfbrmance . . . . 47
Dependent variable -.Assistance: Hypotheses
Tests 1 and 2 . . . . 47
Dependent variable - Time: Hypotheses
Tests 3 and 4 . . . 50
Dependent variable - Efficiency: Hypotheses
5, 6, and 5a, Sb, and 5c . . . . . 51
Summary of Student Perfbrmance Findings . . . . 56
Student Attitude . . . . . 58
Dependent variable - Attitude: Hypothesis
Test 7 . . . . . . . . . . . . 58

Interpretation of Findings . . . . . . . . . . . 59
Student Perfbrmance . . . . . . . . . . . . 6O
Interdependences . . . . . . . . . . . . 60
Assistance . . . . . . . . . . . . . . 61

Time . . . . . . . . . . . . . . . 62
Efficiency. . . . . . 63

Other Observations and Interpretations . . . . 66
Sumnary of Student Performance Interpretations . . 67
Student Attitude . . . . . . . . . . . . . 68

V. SIMARY,CONCI.USIONSANDRECCMI4ENDATIG\IS . . . . . . . 69

Summary and Conclusions . . . . . . . . . . . . 69
Reconmendations...............7l

REFERENCES . . . . . . . . . . . . . . . . . . . 76
APPENDIX . . . . . . . . . . . . . . . . . . . . 79

iv

LIST OF TABLES

Table Page
1 ANOVA for Assistance . . . . . . . . . . . . . . 48
2 ANOVA for Time . . . . . . . . . . . . . . . . 51
3 ANOVA for Efficiency . . . . . . . . . . . . . . 53
4 Simple Effects ANOVA for Efficiency . . . . . . . . . SS
5 ANOVA for Attitude . . . . . . . . . . . . . . 59

Figure

3.1 Design Over Time .

3.2 Mbde in which the Treatment Groups Received the
TUtoring and Prompting.Aid . . . . . . .

3.3 Perfbrmance variable Matrix

3.4 Attitude variable.Matrix

3.5 Mean Optimal Times in Decimal Minutes for All
Treatment Conditions . . . . . . .

4.1 Variable Matrix of Means for: Assistance

4.2 Interaction of Treatment and Repetition on
.Assistance . . . . . . . . . .

4.3 Variable Wtrix of Means for: Time (In Decimal
Minutes) . . . . . . . . .

4.4 Variable Matrix of Means for: Efficiency

4.5 Interaction of Treatment and Repetition on
Efficiency . .

4.6 Variable Mtrix of Means for: Attitude

LIST OF FIGURES

vi

Page
27

29
40
41

45
48

50

51
52

S4
S8

I . STATWT OF PROBLEM

OVERVIEW OF CHAPTER I

 

The first tepic to be presented is a general STATEWENI‘ OF THE

 

w which states that an improvement in the teaching methods for

perceptual -motor behaviors, resulting in the learners requiring less

human assistance, is desirable. The next topic presents the PROPOSED
SOLUTION of the problem. An Audio Tutor-Prompter is to be introduced
to provide the learner with more careful guidance as he is engaged in
actual hands -‘on practice to learn a perceptual -motor task.

The third topic of Chapter I is the PURPOSE OF THIS RESEARCH,

 

which is to determine the impact of the Audio 'Dutor-Prompter on the
learning of perceptual-motor skills. This is followed by the RESEARCH
QUESTION AND HYPOTHESES and DEFINITIONS OF IMPORTANT TERNS sections.

 

 

Chapter I concludes with an OVERVIEW OF OTHER CHAPTERS IN STUDY section.

 

STATEMENT OF THE PROBLEM

 

SKILLS TRAINING

 

Skills training, as used here, refers to training an individual
to perform a psychomotor task. Singer (1972) has provided a model of the
Psychomotor Domain. This model identifies three categories of psycho-
motor tasks, (1) motor skills, (2) physical tasks, and (3) perceptual-

motor behavior .

The first psychomotor task group, motor skills, refers to
acts of physical coordination having relatively little emphasis upon
the knowledge required to perfbrm.them. Examples of motor skills are:
playing a piano, riding a unicycle, pole vaulting, and ski racing.

These skills, although they require varying amounts of knowledge, may
be identified as motor skills because they all require massive amounts
of practice to master.

The second group of psychomotor tasks, physical tasks, is
characterized by minimal mental involvement. Following an understanding
of the act, the execution of it becomes quite repetitive, with stimulus
discrimination of minor importance. Push-ups and foot races are amongst
the tasks that fall into the physical group.

The third group of tasks, and the one of concern here, is per-
ceptual-motor behavior. This includes those tasks that have a high
degree of the perceptual component present. According to Singer, per-
ceptualdmotor training programs generally concern "relatively non-difficult
movement responses to specific directions; involving the interpretation of
directions and the activation of acceptable responses" (p. 14). Examples
of perceptual-motor behaviors are: using a keypunch machine, disassembling
a carburetor, and operating a computer. In all of the above skills, the
knowledge of the individual steps, and their proper sequencing is of
primary importance in perfbrmdng the task. Practice is useful to develop
one's proficiency.

Hencefbrth, when skills training is referred to in this study, it
should be understood that this does not include all skills of the psycho-
motor domain. Rather, it is restricted to those behaviors, in the third

group above, which Singer has classified as perceptual-motor behaviors.

NEED TO IMPROVE SKILLS TRAINING

 

The teaching of perceptualdmotor skills, as they have been defined
above, includes an important portion of all training activities. A cannon
characteristic of such training is that the student practices the task in
a laboratory setting as he is learning to master it during what will be

called a trial period. This fellows logically from the well known premise

 

that "...students learn by doing" (Anderson and Faust, 1973, p. 200). This
is a circumstance that by nature requires some degree of individual atten-
tion from an instructor, or his sUbstitute, because each student is working
individually on his own project rather than in a group. Thus when a learner
encounters a need, he may be the only person with that particular need in
the laboratory at that time, and so needs individual help.

It would seem clear that an.improvement in the teaching of skill
tasks would be made if either the student's need fer individual human
assistance could be reduced, or if that assistance could be provided in a

more satisfactory manner.

PROPOSED SOLUTION

 

GENERAL APPROACH

 

The preceding section has illustrated a basic need for learners
to have either less need fer human assistance while they are learning to
perform a skill task, or a better way to get that assistance. The general
approach of the proposed solution is to do the former, i.e., reduce the
need fer human assistance.

SPECIFIC APPROACH:
AUDIO TUTOR- PRGIIPTER

 

 

It is proposed that better instruction for skill task learning

can be accomplished by more careful guidance of the behavior of the
learner during the time which he is actually engaged in hands-on practice
of the task that he is being trained to perfbrm.

The more careful guidance of the learner's behavior will be
accomplished via a device herein called an Audio Tutorjgrompter, or
ATP for short. The purpose of the ATP is to provide individual students
with anything along the continuum fran prompting to tutoring via pre-
recorded messages. Tutoring refers to actually delivering the instruc-
tional message. Prompting refers to either abbreviated versions of the
instructional message, or calling for a particular learner response. The
ATP is then a substitute for a person acting as a tutor or a prompter.

The ATP consists of a special cassette tape player and a cassette
tape. Each tape is locally created specifically for the particular skill
task which the student is trying to learn. Being made locally, it can
integrate with the particular laboratory equipment, workbook, text books
and people involved.

During use the ATP tape player provides the learner additional
statements concerning the individual steps, in sequence, of the skill to
be perfbrmed. .After the tape player issues a cue concerning the next step
of the task being perfbrmed, it stops. The learner perfbrms the step, and
restarts the tape player for the next cue. The ATP normally is portable
so the learner can take it with him to wherever he is practicing the skill
which he is learning.

The cues to the learner can be of any degree of complexity, and
may be faded on subsequent tapes or portions of a single tape. For
example, side two may contain a more terse version of the instructions

fbund on side one. This becomes a shift from using the ATP more as a

tutor to using it more as a prompter.

The ATP would appear to be useful in the instruction of those
skills where some individualization is desirable and either is currently
not available or when available, for some reason is not a completely
acceptable solution. It may be too expensive to be cost effective,
available only at restricted.times, or restricted amounts because of
too few people to man the lab, or other such reasons. Thus a learner,
in need of some individual prompting or tutoring, during the practice
period, may receive it from the ATP.

This study will use the ATP to provide assistance to the subjects
as they learn the perceptual-motor (i.e., skill) task of operating a
reel-to-reel video tape recorder. This video tape recorder was chosen
as being representative of a large number of skill tasks where more care-
ful guidance of the learner, through methods of prompting, may result

in a decrease in the amount of human assistance required by the learner.

PURPOSE OF THIS RESEARCH

 

This research investigates the effects on the learning of a
skill task by extending the learner's prompts and tutoring materials
into the actual hands-on practice period. The learner can take the
ATP module with him as he begins to perform the task he is learning, and
use it until he has the operations and their sequence sufficiently
mastered to perfbrm the entire operation without prompts.

Thus the primary purpose of this study is to introduce, and
test the impact of an Audio TUtor-Prompting system to be used when

practicing a perceptual-motor task. To do so, the following research

6

questions and hypotheses are of interest.

RESEARCH QUESTION.AND HYPOTHESES

RESEARCH QUESTION

 

The primary question of interest is: Does audio prompting make
a worthwhile contribution to student learning of skill tasks? This
results in hypotheses concerning both student perfbrmance of the task,

and student attitude towards the learning experience.

RESEARCH HYPOTHESES

 

To determine if audio prompting makes a.worthwhile contribution,
what is worthwhile needs further explanation.

As used here, a worthwhile improvement in teaching methods may
either improve (a) the student '5 performance of the task, or (b) the
student's attitude towards the task. One would hope to improve both.
The research hypotheses fellow:

I. Student Perfbrmance Hypotheses

IA. Dependent variable: Assistance

 

H1: There will be a treatment main effect.

H2: There will be a. treatment by repetition

 

interaction.
B. Dependent variable: Time
H3: There will be a treatment main effect.

H4: There will be a treatment by repetition

 

interaction.

C. Dependent Variable: Efficiency

 

HS: There will be a treatment main effect.

H6: There will be a treatment by repetition

 

interaction.
II. Student Attitude Hypothesis
H7: There will be a treatment main effect using the

Ampex Attitude Scale to measure student attitude

DEFINITIONS OF IMPORTANT TERIIB

 

Definitions for the key terms used in the study follow to provide

a conmon basis for tmderstanding.

1.

Audio Tutor-Prompter - The name chosen to describe the process of

 

prompting a student practicing skill tasks during the practice
period.
Skill Tasks - Those psychorIIotor/cognitive operations that the

 

student is being trained to perform.

Practice Period - That time segment in which the student practices

 

. the skill task to be learned.

Learner - A subject in the experiment. The learners are college
students.

Pm - Abbreviated versions of an instructional message, or calling
for a particular learner response.

_T:u_t_o_1: - To provide one-to-one instruction- Tutoring may be considered
the opposite end of a contintnrm from prompting.

fade - To gradually remove, i.e. , the learner is provided with many

prompts initially, then fewer prompts, and finally no prompts.

8. I§I§9.' short
The fellowing two terms are the independent variables of the
study.
9. Treatment - The experimental variable. There are three treatment
groups:
a) Audio group - Received tutoring and pranpting from the ATP.
b) written group - Received tutoring and prompting from
'written instructions.
c) Control group - Received no tutoring or prompting.

10. Repetition - The repeated.measure independent variable. The subjects

 

perfbrm.the same task three times. Each performance represents
one repetition.
The following four terms are the dependent variables of the
study.
11-.3292 ' The time in decimal minutes required for the subject to
complete one repetition.

12. Efficiency - How effectively the subject uses the time available to

 

perform the task. Computed as the ratio of the subject's time
to an optimal time.

13. Assistance - The number of times the subject is given help so that

 

he is able to complete the task.

14. Attitude - The degree to which the subject favors (or disfavors)
the methods used for cueing in the three experimental groups.
Attitude was measured via the specially developed Likert type
Ampex Attitude scale. (Appendix A)

OVERVIEW OF OTHER CHAPTERS IN STUDY

 

Chapter II provides a REVIEW OF THE LITERATURE pertinent to

 

audio tutoring and prompting, ard learner response. Chapter III details
the RESEARCH DESIGN AND PROCEDURES, and Chapter IV states the ANALYSIS

 

OF RESULTS . Finally , Chapter V contains the SUMVIARY AND RECOWIENDATIONS

 

 

of the study.

II. REVIEW OF RELATED LITERATURE

OVERVIEW

The review of related literature is divided into three main
areas: (1) Audio-Tutorial Teaching System, (2) Stimulus Presentation,
and (3) Learner Response to Stinulus. Each of these areas are discussed

and smmarized, and an overall surmary is given.

DETAILS OF MAJOR AREAS

 

AUDIO- TUTORIAL TEACHING SYSTEM

 

General Features. Professor S. N. Postlethwait (1971) is widely

 

known for his develoment of the Audio-Tutorial System. This began in
1961 due to the diversity of student backgromds in a freshnan botany
course. Students of equal capacities could not perform equally well due
to large differences in their prior Biology training. To help the less
well prepared students, Dr. Postlethwait made a special lecture on tape
each week that the students could use in the language lab at their own
convenience. (Postlethwait, et. al., 1969, 1971) Soon, he began making
the tapes integrate with the text. The student then brought his textbook,
and the tape amplified it for him. The same was done with the lab manual,
and eventually the student was provided with real specimens to relate to

the text, lab manual and tape. It became a full audio-video teaching

10

11

system, relieving the instructor of the major portion of materials pre-
sentation fermally done in lecture.

Dr. Postlethwait has had notable success with his Audio-TUtorial
Approach to teaching, and it is worth further exploration. The following
quote illustrates the basic philosophy of his audio-tutorial teaching
system:

A.fundamental guideline which.must be given prime consideration is

that "learning is an activity done by an individual and not some-

thing done to an individual. The structuring of an educational
system should be done on the basis that the program must involve

the learner. The teacher at best can only create a situation con-

ducive to learning by providing the direction, facilities, and

motivation to the individual learner. Inmediately, it becomes
apparent that the program must allow for individual differences in

interests, capacity, and background. (p. l)

Postlethwait, Novak and Wm (1969, p. 96) list advantages of
the audio-tutorial approach upon which there is general agreement:

In the audio-tutorial approach:

1. Emphasis is placed on student learning rather than on teaching.

2. Students can adapt the study pace to their ability to assimilate
the information. Exposure to difficult subjects are repeated as
often as necessary for any particular student.

3. Better students are not a "captive audience" and can use their
time most effectively. Their interests are not dulled by
unnecessary repetition of infermation already learned but they
are free to choose those activities which are more challenging
and instructive.

4. The student can select a listening time adapted to his diurnal
efficiency peak.

5. Tapes demand the attention of the students. Students are not

distracted by each other.

12

6. Students have more individual attention, if they desire it.

7. Scheduling problems are simplified. The hours of scheduled time
from which the students are relieved under the new system can now
be distributed throughout the week as necessary to adjust to the
student's activities.

8. lbre students can be acconmodated in less laboratory space and
with less staff.

9. Fake-up labs and review sessions can be accomnodated with a
minimum of effort.

10. The student is essentially "tutored" by a senior staff member.

Other advantages listed fer audio-tutorial instruction that are
particularly relevant to the ATP are: l) repetition, 2) concentration,
3) association, and 4) pacing. If these are desirable attributes of the
educational system, the ATP should have similar qualities, at least to
some degree. Consider Postlethwait's fbur factors, repetition, concen-
tration, association, and pacing, as they relate to the ATP concept of
this research:

Repetition. The nature of many learning tasks requires multiple

 

repetition fer their achievement. However, some individuals require more
repetition than others. This leads to the need to individualize materials
presentation. It should be noted that the learner knows best how many
repetitions he needs to learn a specific concept. The ATP will allow each
student to use the tape as many times as he believes is necessary to master
the perceptual motor task.

Concentration. It is often difficult to concentrate in most class-

 

rooms. Students are distracted by one another and other extraneous events

13

which may be happening. This has been substantiated by Price (1969) in
an experiment where one half of a group of feurth grade subjects were
subjected to distractions on an audio tape they were listening to and
also to conlnon aural and visual classroan distractions while listening
to the tape. Price found that these cannon distractions significantly
decreased the subject's listening comprehension. Postlethwait isolates
the students from their surroundings by using earphones to carry the
audio message.

Like Postlethwait's system, the ATP obviously works largely
through the auditory channels. If a student uses earphones, his sensi-
tivity to extraneous distractive sounds is diminished largely, and re-
placed by the content of the ATP program.

Additionally, it should be pointed out that when the student is
practicing some training task, such as keypunching, assembling a rifle,
and so ferth, he may already be using his eyes to read data to be key-
punched, or to identify and fit the next rifle conponent. It is dis-
tracting to shift one's sight from the task at hand to "read the
directions" -— while learning or practicing the task. Instead of making
the visual sensory channel do double duty, as is conmon, the ATP allows
the learner to use both visual and audio channels to pronpt people
through the trial period, the same as a human tutor would do.

Association. When the major objective is to learn about plants

 

(etc.), according to Postlethwait, it makes sense that one should study
where plants (etc.) are available. Diagrams, models, photographs, and
other such devices should be a means to the end so that the students'

attention is directed to the literal object itself.

14

Likewise, when one is learning to perform some skill task con-
cerning a physical object, it is mandatory to have that Object available
to the learner. For example, when learning to keypunch, one needs a
keypunch machine to practice on, although the knowledge of how to key-
punch.may be learned outside the presence of an.actual keypunch machine.
Thus when the ATP is used to p:gmpt_a learner who is learning to perform
a skill task, it is being used in a way which is similar to the current
way of doing things by association.

Hewever, as the ATP is integrated into the teaching process as
a tutor rather than a prompter, it may allow the learner to have more
association with that which is being learned. This is because the ATP
is here being used as a portable information delivery system that can be
integrated with the student's actual "hands on” laboratory experience.
Traditional textbooks are mass produced and therefbre cannot effectively
comment on the local experiences that the student is exposed to. It
should be noted that the above statement refers to the possible tutoring
use of the ATP, while this study is primarily concerned with prompting
uses of the ATP.

Paging, People vary considerably in the amount of subject matter
that can be grasped in a given amount of time. A program of study there-
fbre should provide the student the opportunity to adjust the size of the
unit so that those who can proceed more quickly are allowed to do so.

The ATP, as used in this study, works at a fixed audio delivery
speed. Hewever, the pace from student-to-student does vary. This is
because the ATP is stopped while the student perfbrms the required
prompted-for behaviors, and the recorder-off—time can vary from.one student

to another.

15

Pacing will be further discussed in the Stimulus Presentation
section.
Summagy. The discussion of the literature pertaining to an

Audio-Tutorial Teaching System has reported that:

 

l. The teacher creates a situation conducive to learning by pro-
viding the direction, facilities and.motivation to the individual
learner.

2. Emphasis is placed on student learning rather than on teaching.

3. Students progress through the instructional materials individually,
at their own pace, with the help of media and actual specimens to

amplify the instructional objectives.

STIMULUS PRESENTATION

 

A.discussion of three facets of stimulus presentation, as they

relate to the ATP follows. These are the uses of Branching, Pacing, and

 

Compressed Speech.

 

Branching. Crowder (1959) introduced a new form of programmed
instruction first identified as a "scrambled book" and later as "intrinsic"
programming when machines replaced the books. The scrambled book is not
read in page sequence, rather the learner is given a paragraph or two of
information followed by a multiple-choice question. If he chooses the
correct alternative, he is directed to a page on which he is reminded of
why he is right and then directed to another page having new material and
another test question. Hewever, if he was wrong, he is directed to another
page that tells him he is wrong and gives an explanation of why he is wrong.

Crewder has also developed "teaching machines" that present the material

16

and do the page turning for the learner. Branching, then, is a technique
designed to allow the student who already knows a portion of the subject
matter, or who masters the portion with minimal instruction, to skip ahead,
as well as to provide remedial instruction when necessary.

Branching programs seem to make good sense, and one would expect
them to be superior to the less complex linear programs, because they
appear to provide the information the learner needs when he needs it
without his being bothered by superfluous information. However, the re-
search on branching programs has generally not found any advantage for
branching (Anderson, 1967; weisgerber, 1971). Anderson suggests the
reason that:

...branching procedures usually fail to teach better than fixed-
sequence procedures, because, for one reason, any single item

in a branching program is a fallible basis fer branching decisions,
often yielding false positives such that the student skips over
material he cannot afford to miss. When more elaborate and less
error-prone procedures, such as those that presumably could be
implemented with computer-assisted instruction or by a teacher,
are used to make branching decisions, then there may be an advan-

tage to branching techniques, provided the instruction is com-
petently prepared. (p. 154)

ngipg, Much has been written about the advantages of individual-

izing the rate at which the learner may progress through a set of in-
structional materials. It seems intuitively correct that the individual
learner should be allowed to choose his own pace, however, after several
studies on pacing, Kress (1966) and Cropper (1966) suggest that self-
pacing can be nonadaptive to the needs of the learner. Some learners

need to be speeded up and some need to be slowed down to improve the
effectiveness and efficiency of learning.

Self-adapted work rate does coincide with student ability

17

among upper-classmen in high school and among college students, however.
Consequently, self-paced program administration appears to optimize both
learning effectiveness and efficiency among these students. For them,
self-pacing seems clearly preferable to the fixed-paced strategies.

HOwever, among younger students (sixth and eighth graders were
studied) self— adapted work rates neither coincide well with ability nor
do they lead to learning which can be described as optimally efficient.
Substantial numbers of those students demonstrated a need for remedial
treatment by the non-adaptive work rates. Kress found that some high
ability, high achievers worked too slowly to be efficient, while some low
ability, low achievers worked too fast for their ability and did not
learn effectively.

The maladaptive patterns of work habits observed in this study
may be characteristic of students described as "under-achievers." Both
students who learn less and students who learn more slowly than would be
predicted on the basis of their ability are displaying under-achievement.

Compressed Speech. Since the normal rate of speech is generally

 

less than that of reading, audio methods of prompting the learner seem to
be inherently slower than written ones. However, there are means of in-
creasing the words per minute of recorded passages.

Talking at a faster rate than normal is the simplest method of
compressing speech. While this method does have the advantage of requiring
no special equipment, it has limitations in that it introduces undesired
changes in vocal inflection and fluctuations in rate, and has a relatively
low upper limit.

A second way to compress speech is simply to make the playback of

a recording faster than its recording speed. This is the speed changing

18

method, and results in an increase in frequency (pitch) directly pro-
portional to the increase in speed.

The most acceptable method of compressing speech at the present
time is the sampling method. This method removes bits of words in time
units (20-100 milliseconds, depending on the amount of compression de-
sired) from an original recording to shorten the amount of time it takes
to play the recording. Equipment is now available (e.g., varispeech I)
that can compress a normal cassette tape, over a wide range of speeds
controlled by the user, as the tape is being played. Thus it is now
possible for the user to either slow-down or speed-up a cassette tape
as he listens to it.

If it works, this may be an important means by which auditory
learning can be made more efficient, i.e., faster. Significantly, Orr
(1968) reports that "humans find some degree of accelerated speech in-
telligible and comprehensible. There has been almost complete consistency
in the research on this point." (p. 288) HOwever, he goes on to say that
"there is substantial variability among people in their ability to compre-
hend compressed speech." (p. 289) This suggests that the feature of the
varispeech I which allows the user to choose the amount of compression is
probably more satisfying and productive than providing the users with a
tape of fixed speed, whether compressed or not. Foulke (1969) reports
that an inexpensive device will soon be available that will take the ferm
of an attachment to a standard cassette tape recorder and allow the user
to manage the word rate himself.

Sunmagy. The discussion of the literature pertaining to Stimulus

Presentation has reported that:

 

19

l. Branching programs have generally not been found to be superior
to linear programs. This is probably due to the fallability of
branching on the basis of a single item.

2. Self-pacing is preferable to fixed-pacing fer high school and
college aged learners who are normal achievers, but

3. Selpracing fer grade school learners is quite often maladaptive
to their needs.

4. Maladaptive pacing habits (either too fast or too slow for the
learner's ability) were Observed primarily in those students who
are under-achievers.

5. Most all people can tolerate some degree of speech compression,
although there are substantial variations amongst people.

6. The use of compressed speech is becoming more practical.

LEARNER RESPONSE TO STIMULUS

Anderson and Faust (1973), in discussing the role of learner
response, make the fbllowing statement:

One of the most important principles in this entire book is that
students learn by doing. For purposes of discussion three levels
Of active response may be distinguished. At the first level the
student is required to read, listen, or watch. At the second
level he is required to make a particular covert response. And,
finally, at the third level, he is required to make a particular
overt response. (p. 200)

The sections fellowing discuss Overt vs. Covert Responses,

 

Conditions Under Which Overt Responding Facilitates Learning, and When to
Require Explicit Responses, and are followed by a Summagy.
Overt Vs. Covert Responses. IA student may be asked to make

 

a particular response, such as answering a multiple choice question

20

contained in the text. If the student "thinks" the answer, the response
is covert. But if he makes some publicly observable response (i.e.,
speaks, marks) the response is overt. Anderson and Faust cite research
findings on response mode, much of which was summarized by Gates (1917).
These findings suggest that when students respond, they learn more than
when they do not respond. .Also, overt responding produces slightly better
results than does covert responding. This is motivation fer the great
emphasis on student response in programmed instruction.

Anderson (1967) cites an interesting experiment by van wagenen
and Travers (1963) designed to study whether a peer group and a teacher
enhance the effects of feedback. Using groups of eight feurth, fifth,
or sixth graders, the teacher held up large cards upon which a German
word and two English alternatives appeared. The children were divided

into two groups. One group was the actively responding group and the

 

other observed without overt responses. The actively responding group

 

guessed at the correct English alternative and then heard feedback from
the teacher. .A similar pair of experimental groups were instructed by a
teaching machine instead of a live teacher. The posttest results showed
that those who participated actively with either the live teacher or with
the teaching machines, learned more than the Observers. Items that were
actually practiced were learned best, but the participants were consis-
tently better than the observers even on items that they did not directly
practice.

Travers (1964) went on to confirm that participating students do

better not only on directly-practiced items but also on items for which

21

they are observers than do students who are always observers. Further-
more, the performance of the observing students fell off to a greater
degree on the second and third days than did the performance of partici-
pating students. These data fit nicely with the results of the studies
of overt responding within written lessens, indicating some generality
fer the rule that the requirement to make overt responses has both a
direct effect and an indirect, attentional effect. However, student
response does not always facilitate learning.

Conditions Under Which Overt Responding Facilitates Learning.

 

The requiring of overt responses is a distinct characteristic of pro-
graxnned instruction, but many studies (Alter and Silverman, 1962;
Stolurow and walker, 1962; Tbbias and weiner, 1963) feund that covert
responding students, and students reading programs having the responses
already filled in, perfbrmed as well as overt responding students.
Anderson and Faust explain this unexpected development. That is,
relevant overt response occurring fer the right reasons increases
learning. Hewever, making irrelevant overt responses may actually inter-
fere with learning.

An irrelevant response is one that is not related to the
critical content of the lesson. When students are asked to answer frames
containing questions unrelated to the critical content of the lesson,
Holland and Kemp (1965) have shown that learning is impaired. Addition-
ally, even when the item is related to the critical content of the lesson,
the student must get it right for the right reasons. That is, the item

may have enough cues that the student can answer it correctly without

22

knowing the critical content. When this happens, the expected benefit
of requiring overt student responses does not result.

Additionally, Anderson (1967) reports that an overt constructed
response should be required from a student if he is expected to be able
to emit an unfamiliar, technical term. Hewever, if he will only have to
recognize the term, or if he is already able to make the response, as is
the case with high-frequency words in the language, then a multiple-choice

response or "thinking" the response, or just reading may do as well.

Anderson notes that "why overt responding works best with difficult,
unfamiliar material has never been investigated. One possibility is
that the student does make easy responses to himself, but that covert
responding drops out when the response is more difficult to make...
This possible interpretation is illustrated by the phenomenon of
encountering an odd name in a foreign novel and saying to oneself,

'Oh, that guy,‘ instead of'trying to pronounce the name." (p. 141)

When To Require Explicit Responses. Anderson and Faust cite
clear evidence from Spitzer (1939) that the sooner after instruction a
response is called for, the less fbrgetting that takes place.

Additionally, Margolius and Sheffield (1961) conducted research
that gives support for relatively small modules of instruction between
student responses. Students learning to assemble an automobile distrib-
utor were able to do so faster when learning from brief presentations
fellowed by short periods of practice, than when learning from fewer, but
longer presentations with correspondingly longer periods of practice;

Both Spitzer's and the Margolius and Sheffield findings give

support to Postlethwait's Audio Tutorial System of teaching and the ATP

23

being studied here. Spitzer's findings that students should respond

as soon as possible after learning lends support to the concept of
learning new material right in the laboratory where the students have
opportunity to interact (i.e., respond to) that which they are learning
about. Similarly, the ATP probably should have some tutoring fUnction
as opposed to only prompting.

Margolius' and Sheffield's findings that instructional
modules should be relatively short and followed by student practice,
when learning psychomotor skills, also supports the use of the ATP.

The ATP, when used as a tutor, is designed to give a segment of in-
struction and stop while the student practices the task. When used as

a prompter, the ATP can give extremely short instructions, to be fellowed
by student practice, the ultimate extrapolation of Margolius' findings.
And, of course, the ATP can provide for fading from longer segments of
instruction, to shorter, to merely prompts, to nothing at all.

These studies suggest that the instructional materials used
in this study (1) require overt student responses that are, (2) dependent
on knowledge of the critical content of the lesson, and (3) student
responses irrelevant to the critical content of the lesson are to be
avoided.

Thus the prompting by the ATP is designed to result in correct
student responses when the student understands the critical content of
the lesson. However, the ATP prompting may help teach this critical content.

SurnnaLy. The discussion of the literature concerning the Learner

Response to Stimulus has reported that:

 

l. Learner response is very important. Learners who respond learn

more than those who do not.

24

Relevant overt responses are better than covert responses.

Overt response has a positive effect on learner attention.

Overt responses should be required.when learning unfamiliar,
difficult material.

The size of the module between learner responses should be small.
The learner response should occur as soon as possible after the

instruction to minimize fOrgetting.

SUMMARY

The Review of Related Literature was divided into three main

topical areas: (1) An.Audio-TUtorial Teaching System, (2) Stimulus

Presentation and (3) Learner Response to Stimulus. The summaries of each

of these areas are collated and summarized here:

1.

The teacher creates a situation conducive to learning by pro-
viding the direction, facilities and motivation to the indi-
vidual learner.

The Audio-Totorial System places emphasis on student learning
rather than on teaching.

Students progress through the Audio-Tutorial instructional
materials individually, at their own pace, with the help of
media and actual specimens to amplify the instructional ob-
jectives.

Branching programs have generally not been fbund to be superior
to linear programs.

Self-pacing is preferable to fixed-pacing fOr high school and

college aged learners, who are normal achievers, but must be

10.
ll.
12.

13.

25

used with reservation with grade school learners.

Maladaptive pacing habits were generally observed with under-
achievers.

Mbst people can understand some degree of speech compression,
although there are substantial variations amongst people.

The use of compressed speech is becoming more practical.
Learner response is very important as learners who respond
learn more than those who do not.

Relevant overt responses are better than covert responses.
Overt response has a positive effect on learner attention.
Overt responses should be required when learning unfamiliar,
difficult material.

The size of the module between learner responses should be
small and the learner response shoudd occur as soon as possible

after the instruction.

III. RESEARCH DESIGN AND PROCEDURES
This chapter has two major divisions, the DESIGN of the
study, and the PROCEDURES that were fOllowed to implement that design.

First the design.

DESIGN

 

There are two separate design sections in this chapter. First,

the Design Over Time section illustrates the overview of the procedures,

 

and builds the foundation fer the discussion on the validity of the

research, i.e., validity Concerns. Second, the Design Over variables

 

 

section details the interrelation of the variables used in the study

and their analysis. It provides the fOundation fer the Research Hypoth-

 

eses section.

DESIGN OVER TIME
This study takes the form of a posttest only, control group

design, similar to Campbell and Stanley's (1963, p. 8) design number 6,
with the addition of’a materials dimension and two additional repetitions.
Figure 3.1 shows the relationship between the three treatments, assignment
of subjects to treatments (R's - Random), and the measurements (O's) across

time.

26

27

 

 

 

TTreatment .AsSign Repetition 44_111
Group ment 1 2 3
Audio (X1) R M1x1010203 M2X1010203 M3X101020304

 

Written (X2) R M1X2010203 M2X2010203 M3x201020304

Control (X3) R MSBQIOZOS MSXSO] 0203 M3).(_:,l(’)}020304
* a a:

 

 

 

 

 

 

 

tinee———-e>- *concurrently

 

FIGURE 3.1 -- DESIGN OVER TIME

Legend:

X : Treatment group
R : Random Assignment
M : Tutoring - Prompting Materials used by the treatment
groups
‘M - Complete instructions
M - Prompts only
M - No materials
0: OGServations (i.e., Dependent Measures)
01 - Assistance

O2 - Time
03 - Efficiency
04 - Attitude

A posttest-only design was chosen over a pretest-posttest design
fer three reasons. First, Campbell and Stanley indicate that a posttest
only design generally has higher external validity than a similar pretest-
posttest design when the subjects have been randomly assigned to the
treatments, as Figure 3.1 indicates was done here. Second, the computa-
tion of gains, made possible by the inclusion of pretests, were not
required to test the hypotheses. And finally, the student perfOrmance
variables were measured as an integral part of the treatment and so were

not possible to use as pretests.

28

In considering the DESIGN OVER TIME, the following components

 

of Figure 3.1 will be explained: EXPERIMENTAL TREATMENTS (the X's),
SUBJECTS within the X's, and INSTRUMENTATION (the 0's).

 

EXPERIMENTAL TREATMENTS

 

The subjects went through the treatments on an individual basis,
the time for which varied from approximately 40 to 75 minutes. There
were two distinct components to the experience, the video tape instruc-
tion and hands-on practice.

The first portion of the treatment consisted of the subjects
independently watching an 11-minute video tape that taught them all they
needed to know about operating the Ampex 5100 video tape recorder.

Before the video tape was started, the subject was given the assignment
sheet that explained what would be required of him during the hands—on
portion of the treatment . The experimenter also explained the contents
of the assignment sheet so that whether the subject read it or not he was
given an advance organizer before he saw the video tape.

Irnnediately after watching the tape the subject was taken to
the 5100 for his hands-on practice. Again the assignment (to load, thread,
record and playback the recording) was explained to him. Also several
other relevant details were explained and any questions the subject had
were answered (if they did not jeopardize the study). Unlmown to the
subject, he was assigned to one of three groups. Figure 3.2 illustrates
that one group received additional tutoring and prompts via a cassette
recorder, a second received this aid via print, and the third received no
such aid at all. This extra aid was given to the subject and its use was

explained. The subject was not ready to begin his hands-on practice.

29

 

 

 

 

 

 

 

 

 

T Repetiticn
Treatment f ' 2 I
Group Carplete Instructions Pratpts No Help
x1 M1 M1 M3
x2 M2 M2 M3
X3 M3 M3 M3
Figure 3 . 2

NODE IN WHICH THE TREATMENT GRCIJPS
RECEIVED THE 'I'U'IORING AND PRGVIPTING AID
M1: Audio aids via cassette tape
M2: Written aids (transcribed from M1 cassette tape)

M3: N0 aids

All subjects completed three repetitions of loading and thread-
ing the tape on the machine, recording to 15 on the digital counter
(i.e. , everyone recorded the same length of time), played back their
recording, and rewound the tape completely onto its reel. During each
repetition the observer (1) timed it, and (2) provided assistance to the
subject as he asked for it. The time period began when the first reel to
be loaded touched its spindle, and terminated when the tape being rewound
snapped off of the take-up reel. Assistance was provided only when asked
for, although sometimes the experimenter asked the subject if he would

like assistance when he was in obvious need of it.

30

SUBJECTS ,

The subjects used in the experiment were all Michigan State
university students, and.most were volunteers from a gradhate level media
course. Over 90% of the students enrolled in the course actually partici-
pated in the study, probably because extra credit was offered as an

incentive.

INSTRUMENTATION

 

This section first discusses the measurements (0's) taken, and
second, discusses the properties of the instruments used to make the

measures in greater detail.

w
Four separate dependent variables were measured to determine the

subject's perfbrmance and attitude as fOllows:

PerfOrmance - The test of how proficient the student is at perfbrmdng

 

the task. It has three separate variables.

01 - assistance - The number of thmes the subject asked
the observer fer assistance during a single
repetition. For each request fbr help, a single
problem was corrected, and one point added to
the subject's score.

02 - time - The number of decimal minutes required to
complete a single repetition of the task.

03 - efficiency - How effectively the subject uses the
remaining time, after prompting, to perfbrm the

task. Since the various prompting conditions

31

all require a portion of the overall time, a
subject's time measure (02) is the sun of (1)
time to receive pranpting, and (2) time to

perform the task.

Attitude - The degree to which the subject favors (or disfavors) the
method used for cueing.
O4 - attitude - A single Likert type attitude ques-

tionnaire was administered to each subject upon

the completion of the third repetition. See

Appendix A. The questionnaire measures the

student's attitude toward the non—hunan assistance

he received while learning to Operate the tape

recorder.

INSI'RIMEN’I‘ATION CHARACTERISTICS

Four separate methods were used to measure the subject's per-
formance and attitude , as indicated above . These measured variables
were (I) assistance, (2) time, (3) efficiency, and (4) attitude, and
are discussed below.

_(_)_1 - Assistance. The first performance variable is the number

 

of times a subject asks the observer for assistance. For example, the
learner may be in the process of threading the tape, and not know where
the tape should go next. He may say something like "I'm stuck." At
this point the observer asks him if he wants assistance. If he

answers affirmatively, assistance would be given until the problem

was overcome .

32

Assistance was measured by defining a procedure to follow when
giving the assistance and cmmting the nunber of times it was given.
This procedure was followed for all subjects and was as described
below.

No assistance was given the subject unless he asked for it,
except for the special case of checking the tape threading. Whenever
a subject completed threading a tape, the observer checked to make sure
it was threaded correctly. This was not counted as assistance unless
a threading error was discovered. When threading errors were found,
the subject was told how many errors were found and this was counted
as one unit of assistance. Notice that the errors were not identified,
thus the subject still has to find and correct the error and may require
further assistance to do so.

When assistance was asked for, a single error was explained
until the subject had it corrected. This counted as one unit of
assistance.

0

—2
time interval required for a subject to complete the task, measured

- Time. The second measured performance variable is the

with a decimal minute stop watch. The time interval conmenced the
moment the subject first touched the spindle with a reel to be loaded,
and terminated when the tape being rewound snapped off the take-up reel.
Since both ends of the time interval are clearly defined, and since
time can be directly and precisely measured, the time measure is both

relatively reliable and valid.

33

However, it should be noted that the time variable is a measure
of at least two other confounded variables. The time required is related
to (l) the subject's proficiency at perfbrming the task, and (2) the
pacing inherent in the instructional materials. Clearly, a person pro-
ficient at performing a task can usually perfonm it quickly. This is the
basis for all timed contests. However, the time required can be influenced
by the pacing inherent in the instructional materials. For example, audio
cues may require more time to deliver because speech is generally slower
than reading. Hewever, audio cues may be less distracting than visual
cues (i.e., reading) when a subject's eyes are needed to help him.perform
the task at hand. In this case, the inherently slower audio cues may
result in faster or better student perfbrmances because they are more
compatible with perfbrming the task at hand. This represents the logic
behind using audio cues to help a subject learn a psychomotor task like
operating a video tape recorder.

Since time is affected by the pacing of the instructional
materials, a second time variable, called efficiency (03), was computed
from.it, as described below.

_0_., - Efficiency. The last performance variable, efficiency,

J

 

is an index which indicates how quickly the subject performed the task
when considering that prompting required a portion of his 02 time measure.
Efficiency is computed from the following ratio: 03 - efficiency = 100 x
Optimal time (fer treatment)/02.

O2 is the ordinary time required, measured via stopwatch. It
ignores the treatment group which the subject is in. 03, efficiency, is

computed in recognition that the Audio treatment should result in longer

34

times than the written or Control treatments because speech is normally
slower than reading, which, in turn, obviously requires more time than

not doing anything at all. The time required to listen to the audio tape,
or read the instructions is not normally available to the subject to
perform the experimental task. Therefore, how well a subject used his
available time is not necessarily reflected by the raw time measure, 02.
However, an efficiency index of how well a subject used his available time
may be obtained by dividing an optimal time for the subject's treatment,

by the subject's time. This efficiency index ranges generally from O to 100,
with the higher numerical values being more desirable. High efficiencies
indicate the subject perfbrmed the task quickly, with few errors and little
fumbling.

Efficiency, as used here, has a reliability no greater than the
reliability of O2 - time since efficiency is a function of time. The
determination of the optimal times, which fOrm the numerator of the
efficiency ratio, is what turns the time variable into the efficiency index.
It is also a source of possible error. The deviation of these optimal times

is discussed in the PROCEDURES section of Chapter III. In short, they were

 

derived from "experts" acting as naive subjects. The optimal times are the
times required by the "experts" when they performed the task at a fast pace
using the prompting materials.

Although the validity of the "experts" optimal times cannot readily
be established, it is believed not to be biased toward any one treatment
condition. In other words, they attempted to perfbrm at the same pace for

the audio, written and control group conditions. Therefore, the optimal

times are believed to be reasonably valid, and thus the efficiency variable -

O3 is also believed to be reasonably valid.

‘35

_Q, - Attitude. The attitude scale was a measure of the

‘7

 

subject's perceptions of the adequacy and desirability of the non-

human instructions and prompts given him. It contained Likert items

such as:
SA A. N D SD
1. The aid was not useful --------------------- l 2 3 4 5
2. I should have had more aid ----------------- l 2 3 4 S

The complete scale is given in Appendix A. The O4 variable was the
sum of all seven Likert items, after the negatively stated items had
been reversed. A.HOyt's reliability coefficient value of 0.67 was com-
puted for this scale using the M.S.U. version of the J'ENNRIO-I ANOVA
program.

This concludes the development of the DESIGN OVER.TIMB. The
following section examines possible concerns for the validity of the

design.

VALIDITY CONCERNS

This section discusses possible concerns for first, the internal
validity of the experiment, and then the external validity. Internal
validity asks the question: was the experimental treatment the real
cause of the observed change? (Campbell 8 Stanley, 1963) When internal
validity is violated, the values of the measurements (0's) for the
various groups (i.e., treatments, etc.) are unequally affected.

External validity asks the question: to what populations,
settings, treatment variables, and measurement variables can this effect
be generalized? Violations of external validity, therefore, reduce the

generalizability of the findings.

36

INTERNAL‘VALIDITY

 

The design over time is a form of the posttest only control
group design. This is a design which Campbell and Stanley show to have
excellent internal validity. It, as used in this experiment, sufficiently
controlled fOr the concerns of HISTORY, MATURATION, STATISTICAL REGRESSION,
SELECTION, EXPERIMENTAL mRTALITY and their interactions. No unusual cir-
cumstances arose to warrant further discussion of the above concerns.
However, the design's control over the sources of invalidity arising from
TESTING and INSTRUMENT DECAY is less clear and does warrant further dis-
cussion.

Because there are three repetitions of the treatment, it might be

suspected that the post measures (01, O and 03) from the first repetition,

2’
might affect the second set of posttests. These post measures (01 -
assistance, 02 - time, 03 - efficiency) are not taken by the subjects,
rather they are observations made by the observer of the experiment. There-
fOre, the subject obtains no feedback from the measures, during the exper-
iment, that could be used to alter his perfbrmance during the subsequent
treatment and thus this concern vanishes.

INSTRUMENT DECAY is, however, a threat to the internal validity
not to be taken lightly. The methods fOr measuring time (02), efficiency
(03), and attitude (04) are stable over time since time and efficiency are
objective mechanical measures and attitude is measured via a Likert scale
questionnaire that does not change.

The measurement of assistance (01), however, does have an element

of experimenter judgment in it. This allows fOr both experimenter biases

and changes over time to affect the data. If this were to happen, both

37

the assistance and time variables would be affected because it requires
time to give assistance, or unsolicited assistance could be used to

reduce the time required for the subject to perfbrm the task. To combat
this potential source of bias, a procedure to give and count assistance
was established as discussed earlier in the Instrumentation Characteristics
section. This procedure was followed for all subjects and resulted in a
relatively stable measure of assistance across experimental treatments and
time.

Additionally it should be noted that the subjects will ordinarily
vary in their willingness to ask for help. Some subjects will be keenly in-
terested in correcting their problems themselves, while others will readily
35k for help. This results in longer times and lower assistance scores
fbr those more interested in correcting their problems themselves and
shorter times and higher assistance scores fbr those who readily seek
help. This contributes towards higher variability of both the performance
measures than desired, but this variability is reflecting the true nature
of the variability amongst individuals. Since the subjects were randomly
assigned to treatments, and since the treatments all had approximately
50% males and 50% females to control for sex differences, there is no reason
to expect differences between the treatments due to differences in the sub-

ject's willingness to ask fOr assistance.

EXTERNAL VALIDITY

 

The threats to the external validity, or generalizability are
reasonably well controlled fOr in this design, although several possible
threats to external validity identified by Campbell and Stanley warrant

explanation.

38

The INTERACTION of TESTING and TREATMENT is not a significant
concern. While it is true that there are X's fbllowing 0's, the subject
himself does not take the test (0), rather the observer measures the
subject's perfbrmance. Thus the subject has no feedback from earlier
O's that he can use to modify his behavior in the succeeding X's. This
is the same argument given earlier to eliminate the possible concern for
TESTING, where an earlier 0 might affect the scores on a later 0.

MJLTIPLE-TREATMENT INTERFERENCE seems to be a likely problem,
but, in fact, is not. This is where an earlier X might affect a later X.
Actually the three X's shown for each group in Figure 3.1 really combine
to make a single treatment. The term repetition has been used to describe
one complete cycle of the subject loading, threading, recording and
playing back his recording. .Although this task is perfOrmed three times,
the circumstances change. The first two repetitions represent the learning
period such as one might have in a laboratory. The third repetition repre-
sents the test situation that normally occurs after the learning has sup-
posedly taken place. So the three repetitions combine to represent a single
complete learning sequence. When learning, it is desirable for the first
repetition to affect the second, and the second the third. So multiple
treatment interference is not a problem, rather it is the normal part of
the overall learning experience defined by the three sets of X's.

The last concern fer the external validity is the REACTIVE SETTING
of the experiment. Circumstances did not allow this experiment to be done
so that the subjects were unaware that they were participating in a study.
Consequently, all subjects were equally informed that they were in a study

and how they were being measured. Due to two sets of video tape recorders

39

used, probably most all subjects believed they were in some experimental
group rather than any mundane control group. Therefore, there is no
reason to expect the Hawthorne effect to help the two experimental groups
outdo the control group.

However, one might logically expect the performance of all three
groups to be enhanced due to the reactivity of the experimental setting.
This is not really a problem since the purpose of this study is to obtain
relative, not absolute, measures of the effectiveness of the Audio-Tutor
compared to written instructions, and no instructions. The relative
measures, even when enhanced by the reactive setting, will show the pre-
ferred treatment, unless one treatment is affected by the reactivity'more
than another. But, the argument above has already suggested that all
treatments should be similarly affected because the control treatment
probably seemed like an experimental treatment to the subjects.

It should also be noted that absolute scores on these experimental
tasks are not important. The purpose of the experiment is to show the
applicability of the concept of an Audio Tutor-Prompter. Wherever it may
be employed, it is virtually certain that the task being tutored or prompted
‘will not be that of learning to use the Ampex 5100 video tape recorder.

In conclusion, the design of this study has been shown to be
reasonably sound with respect to violations of either internal or external

validity, thus the data collected is reasonably free from contamination.
DESIGN OVER VARIABLES

Whereas the design over time gave the big picture of the procedures

of the design, the design over variables gives the inter-relationships of

40

the variables used in the study. It leads to the research hypotheses
and together they lead to Chapter IV, the Analysis of Results.

VARIABLBIMATRICES

The dependent variables fall into two groups, perfbrmance
measures and attitude measures. They have slightly different variable
matrices.

Perfbrmance Dependent variables. The independent variables in
the performance variable matrix are treatment group and repetition. The
dependent variables are assistance (01), time (02)’ and efficiency (03).
Each dependent variable is analyzed in turn, resulting in.three separate
analyses. Each cell of the variable matrix contains the scores on eight

subjects, fOur males and fOur females.

 

 

 

 

 

 

 

 

 

 

 

 

Tm REPETITICN
GROUP 1 2 3 *Dependent variables:
S1
5
2 l. O - Assistance
Audio'- Xl ' l
38 2. 02 " T1118
3. O - Efficien
s9 3 Cy
tl'll _ x2 .
516
S17
Cbntrol - X3 '
524
Figure 3.3

PERFORMANCE* VARIABLE MATRIX

41

Attitude Dependent Variable . The independent variable in the

 

attitude variable matrix is treatment group. The dependent variable
is the total score from the Ampex Attitude Scale. The same subjects
are used in the attitude variable matrix as are in the performance

variable matrices .

 

 

AUDIO WRI'I'IFN CCNI‘IOL

 

S1 59 S17

S

 

 

 

 

S3 316 24

*Dependent variable: 04 - Ampex Attitude Scale Sum

 

Figure 3.4
ATTITUDE* VARIABLE MATRIX

 

STATISTICAL mDEL OF ANALYSIS

 

All hypotheses were tested using Analysis of Variance (i.e. ,
ANOVA) at the a = .10 level as reconlnended by Borg and Gall (1971) for
exploratory studies.

The student performance hypotheses were tested in three one-
way ANOVA's with a repeated measure dimension, as illustrated in
Figure 3.3. The student attitude hypothesis was tested via an ordinary

one-way ANOVA as shown in Figure 3.4.

42

The usual ANNA assumptions of (l) independence between and
within groups, (2) equality of variances between groups, and (3) normal
distribution of scores, are reasonably well met. In particular, the
first and most important assumption, independence, is well met because
the subjects were taken through the experiment individually and they
are believed not to have interacted outside of the experiment.

Research has shown that even large violations of the
ANOVA assunptions (z) and (3) above, have little effect on
the ANOVA F-test when the design is balanced and the cell sizes are equal
(Glass, Peckham, and Sanders, 1972, p.273). Since this design is balanced
and has equal cell sizes, and since there is no reason to expect large
violations of assumptions (2) and (3), the ANNA F-tests of this study

are considered to be both normal in their power and true a. level .

RESEARCH HYPOTHESES

The testable hypotheses, stated in the research form, are
given below. They are divided into Student Performance and Student
Attitude hypotheses to be analyzed via their corresponding variable
matrices given in the Design Over Variables section.

I . Student Performance Hypotheses

A. Dependent Variable: Assistance - 01

H1: There will be a treatment main effect.

 

H2: There will be EL treatment by repetition

interact ion .

43

B. Dependent Variable: Time - O

2
H3: There will be a treatment main effect.
H4: There will be :1 treatment by repetition

 

interact ion.

C. Dependent Variable: Efficiency - 03

H5: There will be a treatment main effect.
H6: There will be _a_l_ treatment by repetition

 

interaction.
II. Student Attitude Hypothesis - 04
H7: There will be a treatment main effect using
the Ampex Attitude Scale sun to measure

student attitude.

PRIIIEIIJRES

 

The procedures of the study are the actions that have been

taken to implement the design. Most of the procedures of the experiment

have already been outlined in the Design Over Time section of this chapter.

This procedures section gives additional detail on Constructing the In-

structional Materials, and Determining Efficieng, a dependent variable

 

 

of the study.

CONSTRUCTING THE INSTRIKITIONAL MATERIALS

 

As described in the Design Over Time section, the experimental

treatment was divided into two parts. The first part consisted of watching

a video tape about the use of the Ampex 5100. This video tape was in

44

existence before the time of the experiment and was used almost "as is."
However, the tutoring and prompting materials which the subjects
received during the "hands-on" practice session were specially constructed
for the experiment in the following way. First the operation of the
machine was studied and it became apparent that the sequence of proper
operation was virtually fixed. A first prototype of the tutoring portion
was created on tape and later transcribed to paper. A revised prototype
was created after the author used it on himself. Several revised cassette
tape prototypes were created and tested via the tutorial approach to
formtive evaluation (Abedor, 1972). That is, a single subject used the
materials under the supervision of the author. He observed troublesome
points in the materials and gave assistance where necessary. The materials
were revised on this evidence until the pilot subjects could proceed through
the cassette tape reasonably well. A written transcription of the final
tape was made and a professionally made tape recording was also created.
Thus the final version of the tape and written instructions used
in the experiment were arrived at after several revisions based on the
tutorial approach to formative evaluation. It should be noted that the
written instructions were transcribed from the tape, and thus they were

essentially word-for-word identical to the tape version.

DETERMINING EFFICIEI‘CY

 

Efficiency, as used here, refers to how fast the subject performed
the task compared to the minimun time possible under the circunstances.
The efficiency variable (03) and its properties have been discussed in

the Design Over Time portion earlier in this chapter. The discussion here

45

is primarily to explain how the optimal times for the five different
treatment conditions were established .
To obtain the mininun time for the five conditions (there were

two different parts to the cassette tape, two sets of written instruc-

tions corresponding to the two cassette tape parts , and one control
condition of no help at all), three experts performed the task under all
five conditions. Their times were averaged to compute the times shown

in Figure 3.5.

 

 

 

 

 

 

 

W m L
(mm 1 3 7 1 3 4
.Audio - x1 7.10 ' 3.48 ' 2.21 T;
i
fiwt-itten - x2 4.09 E 2.83 2.21 i
% i E
' Control - x3 2.21 ’ 2.21 i 2.21
i 1 I i

 

 

fi

RITE: 2.21 minutes is the optimal time for all
repetitions of the Control group, and
repetition nunber 3 for both the Audio
and Written groups as none of these cells
received prompting materials.

Figure 3. S

MEAN OPTIMAL TIMES IN DBCIMAL MINUTES
Wm CONDITIGIS

46

The experts each canpleted one or more repetitions of the task
until they were satisfied with their performance. If they made a time
consuning error, that repetition was aborted. Extreme times for each
expert were discarded, and a single average for each expert in all five
treatment conditions was canputed. The optimal times displayed in
Figure 3.5 are the averages over the three experts for each of the
five treatment conditions .

Each subject's efficiency was obtained by dividing the optimal
time fran the corresponding cell in Figure 3.5 by the subject's measured
time (i.e., 02) for that repetition, and nultiplying the result by 100.

IV. ANALYSIS OF RESULTS

This chapter is divided into two major parts. The first
part reports the findings of the study, and the second part discusses
the findings.

‘ FINDIMSS

There are findings in two major areas: student performance,
and student attitude. The student performance findings are reported first,

followed by the findings cancerning student attitude.

STUDENT PERFOIMANCE

The student performance findings stem from three different
dependent variables: assistance, time, and efficiency. Each of these
three dependent variables are discussed in turn below. In each case, the
hypothesis is restated, the suunary data and associated AMA tables are
given, followed by the discussion related to the hypothesis.
DEPENDENT VARIABLE. - ASSISTANCE:
m l and 2

The mean differences in assistance required between the three
groups tested is illustrated in Figure 4.1. The results of the one way,
with repeated measures, ANOVA are shown in Table l.

47

48

 

..r__

 

 

 

 

 

 

 

 

 

 

 

 

 

TREATMENT REPETITTON
GROUP '
l 2 3
dio
1 1.625 .750 .250 0.875
itten
2.250 1.000 .500 1.250
trOl
ii ‘7.375 2.250 1.125 3.583
N’- 8 for all cells
Figure 4.1
VARIABLE.MATRIX.OF1MEANS
FOR: ASSISTANCE
TABLE 1: ANOVA FOR ASSISTANCE.
Source DF 1W5 F
Treatment (T) 2 51.58 6.04*
Subjects: T(S:T) 21 8.55
T x Repetition (TR) 4 17.37 5.66*
RS:T 42 3.07

 

*Significant at o = .10

49

The ANOVA tests of the Assistance Hypotheses (H1 and Hz) were both

significant and are explained below.

figmesis l. The Audio group will require less assistance
t wi e Written group, which in turn will require

less than the Control group.

In this main effect test, the means differed in the hypothesized
directions, and the F ratio of 6.04 (df 2, 21) was significant at the
a s .10 level, lending swport to this hypothesis. A Tukey post hoc
analysis revealed that for pairs of means to be different, they must be
different by 1.76 or more. Therefore it may be concluded that the Audio
group required less assistance than the Control group (3.583-0.87S =
2.708 > 1.76); the Written group required less assistance than the Control
group (3.583-1.250 = 2.333 > 1.76); but the Audio grow and the Written
group required indistinguishable, using the Tukey method, amounts of
assistance (1.250-.875 = .375 i 1.76). .Thus, the data lends support to
a major portion of the hypothesis, but not all of it.

Hypothesis 2. Initial differences, in assistance, amongst

m three treatment grows will tend to disappear over

time (i.e., there will be a treatment by repetition inter-
action).

 

The F ratio of 5.66 (df 4, 42) was significant at the a =- .10
level. Thus the data swported the hypothesis. Figure 4.2 illustrates
the trend towards requiring less and less assistance with additional
repetitions of practice for all groups. It also indicates the inter-
action is an ordinal one (i.e., when graphed, the lines do not cross)
and therefore the interpretations of the main effect hypothesis (H1)
can be generalized across all repetitions rather than being restricted

to a single repetition.

50

 

 

8-
* ’ Control
g 7 V
H
g” 6P \\
\.
L
2* 5 \
as. 4' \
a .Written \
3
5 2b \
\
1 Audio/ "“ ‘~\
1 2 3
REPBTITION
Figure 4.2

INTERACTION OF TREA'IWT AND REPETITION 0N ASSISTANCE

DEPENDENT VARIABLE - TIME:
an

The mean differences in time required between the three grows
was tested as illustrated in Figure 4.3. The results of the one way,
with repeated measures MINA are shown in Table 2. The ANOVA tests of
the time hypotheses (H3 and H4) were not significant as explained below.

{Ethesis 3. The Audio grow will require less time

t w1 t Written group, which in turn will require

less than the Control grow.

In this main effect test, the F ratio of .01 (df 2, 21) was
not significant at any a level. The hypothesis was rejected

l-b'pothesis 4. Initial differences, in time, amongst the

three treatment groups will tend to disappear over time

(i.e. , there will be a treatment by prepetition inter-
action).

 

51

 

 

 

 

 

 

 

 

 

‘TREATMENT REPETITION

GROUP . i

j l 2 ' 3

o

1 13.067 6.840 5.337 8.414

‘tten

2 13.099 6.639 5.130 8.289
Cantroi'
4;: 12.750 6.820 5.695 8.422

 

 

N = 8 fbr all cells
Figure 4.3

VARIABLE MATRIX OF MEANS
FOR: TIME (IN DECIMAL MINUTES)

TABLE 2: ANOVA fOr TIME.

 

 

Sources DF MS F
Treatment (T) 2 0.13 .01
Subjects: T (S:T) 21 15.63
T x Repetition (TR) 4 0.46 .09
RS:T 42 5.30

 

The F ratio fbr this interaction hypothesis was .09 (df 4, 42)
which is not significant at the o - .10 level. Thus the data does not

support the hypothesis and it was rejected.

DEPENDENT VARIABLE - EFFICIENCY:
m; g, andTaLSb, and Sc

The mean differences in efficiency between the three groups

 

 

was tested as illustrated in Figure 4.4. The analysis of this dependent

variable was more complicated than other variables. The normal ANOVA

52

analysis revealed a significant disordinal (i.e. , the lines cross) inter-
action (H6) which complicated interpretation of the main effects (H5)
and motivated a second analysis. First, the results of the hypotheses

tests via the normal ANOVA.

 

TREATMENT REPET ITION

 

 

 

GROJP
1 2 3
fidio
1 58 . 750 53. 062 42 . 962 51. 6
itten

2 34.962 45.425 47.862 42.8
L'CTE-Lftrol
g18.687 39.687 43.125 33.8

N = 8 for all cells
where:

 

 

 

 

 

 

 

Efficiency - Expert's time
Learner‘s time

 

x 100

Figure 4.4

VARIABLE MATRIX OF DEANS
FOR: EFFICIENCY

Main Effects Analysis. The results of the one way, with

 

repeated measures , ANOVA are shown in Table 3. Here the sources of
variation are Split w in the usual way yielding a testable main effect,
with two degrees of freedom, and a testable interaction having four
degrees of freedom, for a total of six degrees of freedom. The ANOVA
tests of the efficiency hypotheses (H5 and H6) directly correspond to

these sources. Both tests were significant.

53

TABLE 3: ANOVA for EFFICIENCY.

 

 

 

Source DF MS F
Treatment (T) 2 1892.16 4.58*
Subjects: (T(S:T) 21 413.17

T x Repetition (TR) 4 888.82 13.96*
RS:T 42 63.67

 

*Significant at o = .10

HEggthesis 5. The Audio group will perfOrm.more efficiently
t w1 t e Written grow, which in turn will perform

more efficiently than the Control group.

In this main effect test, the means differed in the hypothesized
directions, and the F ratio of 4.58 (df 2, 21) was significant at the
o.= .10 level, lending support to the hypothesis. However, it should be
noted that the two-way interaction (H6) was also significant, furthermore,
this interaction appears to be a disordinal one (i.e., when graphed the
lines crossed). This suggests that HS was not fully supported but further
analysis is necessary (H6 and simple effects) to clarify the interpreta-
tion of H5.

Hypothesis 6. Initial differences, in relative efficiency,

amongst the three treatment groups will tend to disappear

over time (i.e., there will be a treatment by repetition
interaction).

 

The F ratio of 13.96 (df 4, 42) was significant at the a =
.10 level. Thus the data supported the hypothesis. The data are graphed
in Figure 4.5, which shows the lines clearly crossing. Therefbre,
although the findings of H5 indicate there is a difference amongst the
three treatments, the best treatment is not best at all three repetition

levels. Consequently, the best treatment must be defined at each

54

 

 

60
Audic?\
ES 50 A -
E "1:72;;
B 40 //
H //
ES 30 Written /
20 ./’
E /
10 Control
0
1 2 3
REPETITION
Figure 4.5

INTERACTION OF TREATMENT AND REPETITION ON EFFICIENCY

repetition level instead of being able to generalize across all repeti-
tions. .A Simple Effects ANOVA.and.TUkey'post-hoc techniques were used
fer this further analysis (Kirk, 1968, p. 263).

Re-Analysis via Simple Effects. The simple effects ANOVA
separately tests for differences amongst the three treatments at
epetitions l, 2 and 3. Thus, two hypotheses (HS and H6), a main effect
and an interaction test, turns into what looks like three one-way ANOVAS.
These differences show up as the three simple effects sources in Table
4, which correspond to Treatment and Treatment x Repetition sources in
Table 3, the original ANOVA table for hypotheses 5 and 6. Since two
hypotheses have been replaced by three, the simple effects a level was
reduced to 2/3 of its original value (from .10 to .067) to control the

overall level of a on the efficiency dependent variable.

55

TABLE 4: SIMPLE EFFECTS ANOVA FOR.EFFICIENCY.

 

 

 

Source DF MS F
Treatment: REP 2 3247.6 l8.02*
Treatment: REP1 2 360.2 2.00
Treatment: REP; 2 62.0 .34
Within Cell (Pooled) 63 180.2

 

*Significant at o = .067

The test results of the three simple effects hypotheses are
explained below. For the sake of better nomenclature they are called
hypotheses 5a, 5b, and 5c. Viewing Figures 4.4 and 4.5 while reading
the results of the tests of 5a, b, and c Will help clarify the dis-
cussion.

Hypothesis 5a. On Repetition #4, the Audio group will

perform more efficiently thanwv—Vill the Written grow,

which in turn will exceed the performance of the
Control group.

 

 

 

In this first of three simple effects tests the means (58.75,
34.962, 18.687) differed in the right directions, and the F ratio of
18.02 (df 2, 63) was obviously significant at the a = .067 level. Thus
the data lends support to the hypothesis. However, since ANOVA only
tests for any difference amongst the three means, a Tukey post—hoe
technique was used to test fer significance on each of the three pairs
(Kirk, 1968, pp. 268-269).

This post hoc requires a ratio to be fermed from the data and
compared to a tabled critical value (q') from the studentized range.
All three pairs (Audio vs. written,.Audio vs. Control, Written vs.

Control) were found to be significantly different at the a = .067 level.

56

Therefore the data fully swports hypothesis 5a; the Audio grow per-
formed more efficiently than the written grow, which in turn were

more efficient than the control group.

Typothesis 5b. On Repetition #2, the Audio grow will
perform more efficiently fimn will the Written grow,
which in turn will exceed the performance of the
Control grow (i.e., same as 5a except now Repetition

#2).
In the second simple effects test the means (53.062, 45.425,
38.687) differed in the desired directions, however the F ratio of 2.00
(df 2, 63) was not significant at the o = .067 level. Hypothesis 5b is

not swported.

Hypothesis 5c. ()1 Repetition #3, the Audio grow will
perform more efficiently than will the Written grow,
which in turn will exceed the performance of the
Centrol grow (i.e., same as 5a and 5b except now
Repetition #3).

 

In the third and last simple effects test the means (42.962,
47.862, 43.125) did not differ in the desired directions; however, the
F ratio of 0.34 is not significant at any a level. Since the test is
not significant, it cannot be determined that the Written group performed
more efficiently than the other two grows; rather the three means are
statistically indistinguishable fran each other. Hypothesis 5c is not

swported.

SLWARY OF STUDENT PERFORMANCE FINDINGS

 

There were three dependent student performance variables:
Assistance, Time, and Efficiency. They each had a main effect on an
interaction hypothesis tested via a one way with repeated measures ANOVA.

Efficiency became a special case as discussed below.

S7

Assistance. Both the main effect (H1) and interaction (H2)

 

hypotheses were significant at the o = .10 level. The treatment by
repetition interaction was a1 ordinal one, and therefore it was not
improper to test the treatments main effect across all three repetitions.
Tukey post hoc tests showed that the Audio and Written grows generally
required less assistance than the Control grow. However, it could not
be shown that the Audio grow required less assistance than the Written
grow.

Th3. Neither the main effect (113) nor the interaction (H4)
hypotheses were significant at the a =- .10 level. Thus the time
required was essentially the sane for all three treatments.

Efficiency. Things got complicated here. Both the main effect

 

(HS) and interaction (”6) hypotheses were significant. However, unlike
the conditions under Assistance above, the interaction was a disordinal
one, as, when graphed in Figure 4.3., the lines crossed. Because a
disordinal interaction makes general interpretations of main effects
improper, a simple effects AMNA was performed on the data.

When changing to simple effects, two hypotheses (HS and H6)
becane three (H5 a’ HSb’ H5c)' To preserve the overall a level for the
efficiency dependent variable, a was reduced by l/3 from .10 to .067.

The results of the simple effects aralysis showed the effi-
ciencies of the three treatments at repetition #1 to be significantly
different at the a = .067 level. A post hoc analysis showed that all
three pairs were statistically different. This means that at repetition
#1, the Audio grow used their time more efficiently than did the
Written grow, which in turn was more efficient than the Control grow.
In other words, hypothesis 5a was fully swported. Hypotheses 5b and 5c

58

were not swported. This nears the efficiencies of the three treat-

ments at repetitions 2 and 3 were statistically indistinguishable.

STUDENT ATTITUDE
The student attitude findings stem from a single dependent
variable measured via the specially written Likert type attitude scale
to measure the subject's reaction to the aid they received during their

hands -on learning experience.

DEPENDENT VARIABLE - ATTITUDE:
HYPOIHESIS nasr T

 

The participant ' s attitude was computed by reversing the scoring
on the negatively stated items of the attitude scale (i.e. , 5 becanes
a l, 4 becomes 2, 3 remains 3) and then totaling to get an overall
measure of attitude.

The mean differences in attitude required between the three
grows was tested as illustrated in Figure 4.6. The results of the one
way ANOVA are shown in Table 5. The test was significant at the .10

level .

 

TREATMENT GRCXJP

 

Audio-XJ : Written - X, Control — X

 

 

 

 

 

29.25 28.63 24.38

 

N = 8 for all cells

Figure 4. 6

VARIABLE MATRIX OF MEANS
FOR: ATTITUDE

59

TABLE 5: ANNA for ATTITUDE.

 

 

Source DF LB ' F
Treatment 2 56 . 292 5 . 02*
Within 21 ll . 202

 

*Significant at on = .10

Hypothesis 7. The Audio grow will have a mare favorable
attitude than will the Written grow, which in turn will
55 nnore favorable than the Control grow.

 

In this main effect test, the means differed in the hypothesized
directions, and the F ratio of 5.02 (df 2, 21) was significant at the
on = .10 level, lending swport to this hypothesis. However, simple
inspection reveals that the means for the written and audio grows are
very similar, in fact, they are not a meaningful distance apart. The
significance comes from the control grow mean which is meaningfully less
than the other two. Since the written and Audio grows are not meaning-
fully different, the data does r_n_ot_ swport the research hypothesis.
Therefore the data analysis has failed to reject the null hypothesis.

 

INTERPRETATION OF FINDINGS

 

The preceding portions of this chapter have presented the
statistical findings of the experiment without discussion or interpreta-
tion. The remaining part of the chapter interprets the meaning of those
findings. The interpretation section is organized in the sanne order as
the earlier findings section. Student performance interpretations come

first.

60

STUDENTPERFOIMANCE

 

INTERDEPENDENCES

 

Before the student performance findings are interpreted , one
should be aware that (l) the main effect and interaction tests in the one
way with repeated ANOVA are not independent of each other, and (2) the
three student performance dependent variables: Assistance, Time, and
Efficiency, are also not independent of each other. These interdepen-
dencies are discussed below, beginning with the relationship between the
main effect and interaction tests.

Non- Independence of Main Effects and Interactions. When testing
the student performance variables, significance was found in both the
main effects and interaction tests for assistance and efficiency while
neither test was significant for time. A short reflection will reveal
that:

Interpretation 1. The main effects and interaction tests

are not independent of each other for any of the student
performance measures . Rather, they are positively related.

 

The logic behind the interpretation follows. The subjects
invariably learn how to operate the Video Tape Recorder no matter what
treatment grow they are in because they are given assistance whenever
they ask for it . Eventually, all grows would become competent at the
task, performing it without any assistance and at a high level of effi-
ciency. If the treatments have an effect, there will be nuch larger
differences anongst their means at repetition #1 than at repetition #3
when all grows are much nnore competent at performing the task. See
Figure 4.2 above, as an exanple of this. The dispersion of the means
at repetition #1 produces a main effect. The convergence at repetition

#3 makes the lines non-parallel and therefore yields a significant

61

interaction when there is a time main effect. Consequently the tests of
the main effects and interactions for the student performance variables
are not independent of each other. Rather , they are positively
related so they tend to be either both significant or both not signi-
ficant. Now, there is one more interdependency that must be considered.
Non-Independence of Student Performance Measures. It also can
be readily shown that:
Inte retat ion 2 . The three student performance dependent

varia les: Assistance, Time and Efficiency, are not
independent of each other , but are interrelated instead.

 

 

To illustrate the inter-relationships, note that Assistance is
a function of understanding . (To understand generally means to perceive
clearly and correctly.) The more understanding one possesses , the less
assistance one requires. Also, the Time variable is related to under-
standing. If the level of understanding is not satisfactory, more Time
may be spent studying the situation until the level of understanding
becomes adequate. 80, both Time and Assistance are related to under-
standing, and therefore do not act independently of each other. Effici-
ency is a mthematical function of Tinne, and so all three dependent
variables, Assistance, Time and Efficiency are interrelated.

Now that the interdependencies amongst the student performances
data have been discussed, proper interpretations can be uncle of the

student performance findings for Assistance, Time and Efficiency.

ASSISTANCE

 

The assistance main effect was significant and its post hoc

follow-w indicates that:

62

Interpretation 3. The subjects did not learn all they needed
tcflonow fran passively watching the video tape demonstration.
Actual hands-on experience was necessary.

 

This is supported by the facts that (l) the control group
required a large amount of assistance from the observer and (Z) the Audio
and Written groups required less help than did the Control group. So the
audio and written instructional materials did help the subjects perfbrm
the task as they reduced the subjects' dependency on outside help, but
they did not completely eliminate it.

Because the differences in the amounts of assistance between
the audio and written groups were not statistically detectable using the
Tokey post hoc method, it can be inferred that:

Interpretation 4. Both the audio and written groups required

similar amounts of Assistance from the observer to be able to
complete their tasks.

 

Therefore it seems clear that both grows achieved similar
levels of understanding, but that does not mean that it was equally easy
for both grows to achieve their similar levels of understanding. In
fact, an argunent will be presented in the EFFICIENCY section suggesting

 

that the audio grow was able to follow the directions more easily than
the written grow.
The next student performance dependent variable to discuss is

Time .

TIME
Since neither the time main effect or interaction were signifi-
cant, it nnust be interpreted that under these experimental conditions:
Interpretation 6. None of the methods of cueing has the

advantage of yielding shorter overall times for the
subjects to complete their tasks.

 

63

However, it should be noted that when learning a skill task,
the time required to complete it is highly dependent won one's under-
standing of the task and the way in which help or assistance is provided.
If a student needs help, and the lab instructor (or substitute) are not
there to help him, the work may cone to a halt. In this experiment, help
was provided immediately which eliminated time delays that might actually
result in the real world when there are differences in the amounts of
assistance required.

Since it takes longer to listen to a tape than it does to read
the sane information, which in turn takes longer than no external instruc-
tion, if all other things were equal, there should be time differences
amongst the three treatmnent grows. But the overall timnes for the three
grows were not significantly different, therefore, some grows must be
using their available timne more efficiently than others. The efficiency

variable was developed to highlight this effect.

EFFICIENCY

 

The graph of efficiency vs. repetition for the three treatment
grows, Figure 4.5, provides some of the mnost revealing information in
the study. This shows the Audio grow performing much more efficiently,
during the first repetition, after which its advantage statistically
disappears in the second repetition, and is absolutely gone in the third
repetition as the lines on the graph actually cross. The convergance of
the lines is expected since as the grows become nnore competent, they
act more nearly alike.

The clear sweriority, in efficiency, of the Audio grow over

the other two at repetition #1 suggests the following interpretation:

64

Interpretation 6. In the early stages of learning a skill
ta§k, audio instructions are nnore quickly carried out
than their written counterparts. That is, the audio grow
had higher efficiency.

 

This does not mean that the overall time is less (as indeed
it was not for this study). Rather it mneans that the timne interval
beginning at the end of the _f_i_r_'_s_t_ hearing (or reading) of the instruc-
tion, and ending when the instruction has been properly carried out,
is shorter for the audio grow than it is for the written grow. In
other words, the audio grow spent less time actually implementing the
instructions, after they received them, than did the written grow.

The audio grow had higher efficiency.

Interpretation 6 follows directly from the analysis of simple
effects at repetition #1 for the efficiency variable. This showed the
efficiency of the audio grow exceeding the written grow, both of which
exceeded the control grow's efficiency. The following discussion
explains how the grow's efficiency is related to its level of under-
standing.

After a subject has listened to a single tape instruction (or
read the sane words from paper), his understanding of that instruction
may range from clearly understood to no comprehension at all. If his
understanding is less than clear, he may benefit from replaying the
instruction, to help him understand more clearly.

If the subject understands the instruction well enough on the
first attempt, he has learned quickly. But if the instruction is not
clear, there is a decision to make. He may choose between forging ahead
to attempt to complete the task quickly, but having less understanding,

or take the time to repeat the instruction for greater understanding .

65

It should be obvious that the subject must understand quickly
to have high efficiency. He also needs a clear understanding of the
instruction or his chances of making a mistake increase and he is likely
to waste time deliberating on his predicament and.may even have to seek
outside assistance. Therefore, for high efficiency the subject must
understand both quickly and clearly. It seems that the audio group
understood more quickly and just as clearly as did the written group,
as they were able to implement their instructions more quickly (Inter-
pretation 6) and with equal understanding as evidenced by the fact that
they required similar amounts of Assistance (Interpretation 4). First,
why they understood more quickly will be explained, and second, why they
did not understand more clearly than the written group will be discussed.

First, the interpretation of the data regarding why the audio
group understood more quickly than the written group is as follows:

Interpretation 7. Audio instructions for skill tasks are
eaSier to understand than written.instructions.

 

This interpretation was substantiated by observation by the
experimenter during the course of the experiment. The Audio group seldom
backed the tape up, while the Written group did seem to have greater
tendency to re-read the instructions. This is attributed to the Audio
group being able to focus their eyes on the part of the video tape
recorder that was being explained, during the explanation. The written
group, on the other hand, obviously had to focus their eyes on the
printed instructions to read the explanation, and then shift their eyes
to the video tape recorder, and quite often turn back to the instruc-
tions again.

Second, the interpretation of why the audio group did not

understand more clearly than the written group is as fellows.

1‘-

66

Differences in these grows' levels of understanding shoudl logically
show w as differences in the dependent variable Assistance required.
This did not happen because:

Interpretation 8. The mastery nature of the skill task

training had the effect of holding knowledge constant
and allowing timne to vary.

 

This is because the subjects had to keep trying until they
successfully completed the task. To do this they needed some minimum
level of knowledge. When they did not understand an instruction, mnany
mnade the trade-off decision (discussed above) by choosing to repeat the
instruction to enable them to continue making progress towards finishing
the entire task. By repeating the instruct ion they brought their level
of understanding w to the same level as the audio grow, without having

to rely on the observer for outside human assistance.

OTHER OBSERVATIONS AND INTERPRETATIONS

 

The audio instructions were given via a cassette tape recorder
that stopped after each instruction until restarted by the subject's foot
switch. This set-w has the following problem:

Interpretation 9. Users of the audio cassette tape system

are slowed down because it is difficult to either preview
the content of the next frame , or review the previous frame .

 

Quite often, the subjects in the audio grow wanted to lonow
what was coming next so they would know if somne essential component was
either omitted from the instructional prompting materials, or if they
might have accidentally mnissed it. In other words, they either (1) did
not want to blindly follow the tape, or (2) did not have faith that the
tape would lead them completely through the entire task. This phenomenon
suggests that some nneans of previewing or reviewing tape instructions

(such as a topic outline) would reduce the timne spent by the student

67

wondering if he had a problem. This should reduce the total timne

required as many subjects hesitated to go forward, knowing that it was

womewhat difficult to back w if the need arose.

The written grow had the opposite problem. They sometimes had

difficulty finding the frame they were last reading after they performed

the current Operat ion on the video tape recorder and looked back to the

instruction sheet. Some mneans of assisting themn in keeping their place

would be helpful .

SIMNARY OF STUDENT PERFORMANCE INTERPRETATIGNS

 

The following student performance interpretations were made as

a result of this study.

1.

The student performance main effects and interact ions are not
independent of each other.

The student performance dependent variables, (1) Assistance,
(2) Time, and (3) Efficiency are not independent of each
other.

The subjects did not learn all they needed to know from
passively watching the video tape demonstration. Actual
hands-on experience was necessary.

Both the audio and written grows required similar amounts
of Assistance from the observer to be able to conplete their
tasks.

None of the mnethods of cueing has the advantage of yielding
shorter overall timnes for the subjects to conplete the tasks.
In the early stages of learning a skill task, audio instruc-
tions are more quickly carried out than their written

counterparts. That is, the audio grow had higher efficiency.

68

7. Audio instructions for skill tasks are easier to understand
than written instructions.

8. The mastery nature of the skill task training had the effect
of holding knowledge constant and allowing time to vary.

9. Users of the audio cassette tape system are slowed down
because it is difficult to either preview the content of

the next frame, or to review the previous franne.
STUDENT ATTITUDE

There were no statistical student attitude findings of conse-
quence. This is perhaps due to either (1) the low reliability of the
attitude scale and/or (2) the inability of the students to have absolute
differences in attitude, when they themselves are not comparing the two
methods against each other.

However , when the audio subjects were asked whether they pre-
ferred the written or audio presentations , they invariably chose the
audio. But since the instrument slowed no meaningful differences in
attitude between the audio and written grows, this observation is not

interpreted as having importance in this study.

V. SUMNARY, CONCLUSIONS AND REQNMENDATIONS

This chapter is divided into two mnajor divisions as follows.

The SUWIARY and CONCLUSIONS section reviews the objective of the study,

 

the procedures used to meet that objective, and the conclusions drawn

directly from the data. The IMPLICATIONS section applies those con-

 

clusions to a rationale for choosing between audio and written prompting

and develops several recommendations.

SUMMRY and CONCLUSIONS

 

OBJECTIVE AND PROCEIXIRES

 

The objective of this study was to compare the relative
effectiveness of audio and written prompting given to students as they
were learning perceptual motor tasks. Tasks of interest here were those
a learner can usually mnaster with minimum practice once he has acquired
the necessary knowledge, as opposed to those such as unicycle riding, or
piano sight reading, which require a high ratio of practice to knowledge
for mnastery.

The primary research question was: Does audio prompting make
a worthwhile contribution to student learning of perceptual mnotor tasks,
or in other words, are written instructions just as good?

The skill task chosen for the experiment was loading, threading,
recording and playback on an Ampex 5100 Video Tape Recorder. It was chosen
because it was relatively complicated and unfamiliar to the population
selected and because it was an exarple of a perceptual-motor task where knowledge

rather than large amounts of practice, was the essential ingredient of success.

69

70

.All three groups, audio, written, and control, watched an
11 minute training video tape after which they completed three repetitions of
loading, threading, recording and playing-back their recordings on the
machine itself. During the first two repetitions the audio and written group
received prompts. In the first repetition these prompts were
extensive-~much like complete instructions or tutoring. During the second
repetition, the prompts were much less frequent and shorter. The audio group
received these prompts via a cassette tape recorder that stopped after each
prompt until restarted by the subject by a foot switch. The prompts for
the written group were virtually identical to the auditory ones except
they were written on five pages of paper.

The primary conclusions of the study fellow.

CONCLUSIONS

 

This section contains the most important findings of the study
as they have been interpreted.

Conclusion 1. The subjects were unable, in most cases, to

 

effectively learn how to operate the video tape recorder from only watching
the 11 minute instructional tape. They needed the hands-on experience to
supplement what they learned from the video tape.

Conclusion 2. The subjects needed additional instruction or

 

prompting during their hands-on experience.

Conclusion 3. Both the audio and written groups required similar

 

amounts of Assistance from the observer to be able to complete their tasks.

 

Conclusion 4. Neither audio prompting nor written prompting has

 

any Time advantage. Both.methods require approximately the same amount

of time from the learner.

71

Conclusion 5. In the early stages of learning a skill task,

 

audio instructions have higher Efficiencies than their written counterparts,

 

that is, they are more quickly implemented. Since people generally read
faster than they speak, written instructions require less time to present,
but appear to be more difficult to understand. Because of this, the peOple
receiving audio instructions are able to implement them more quickly so
that their total time was no longer than the written groupis time.

Conclusion 6. Audio instructions for skill tasks are easier to

 

understand than their written counterparts. This is because the learner
can keep his eyes on the equipment as its operation is being explained.
Because the audio instructions are easier to understand, they are carried
out more quickly than equivalent written instructions.

Conclusion 7. Learner Attitude toward the learning experience

 

fer both the audio and written groups was similar.

Conclusion 8. The audio group's progress was slowed by the

 

difficulty of either previewing or reviewing adjacent frames of the
instructional materials.
New, the application of these conclusions in skill training in-

volving the selection of audio and written prompts is discussed.

RECOMMENDATIONS

 

RATIONALE FOR RECOMMENDATIONS

 

Before any recommendations are made, it should be recalled that
the primary objective of this study was to determine if audio prompting
makes a worthwhile contribution to student learning. Or stated another
way, are audio instructions for skill task learning just as good, or better,

than their written counterparts? To answer such a question one needs some

72

set of terminal objectives on which to compare the contributions of both

kinds of instructions. The terminal objectives chosen directly relate to

three of the four dependent variables in the study and are as follows:
Terminal Objectives. A worthwhile contribution to student learning
of a skill task will result in the student having:

1. a reduction in Assistance required, or

 

2. a reduction in the overall T‘_'_m_e_ required, or
3. an improvement in Attitude toward the learning experience.

The question now becomes, which mode of prompting, audio or
written, best satisfies these three objectives, or is there no difference
between the two modes? Conclusions 3, 4, and 7 of the preceding section
may be used to compare the contributions of audio and written prompting
towards objectives 1, 2 and 3 respectively.

Based on conclusions 3, 4 and 7, which were for periods of time
less than one hour, audio instructions are po_t_ considered either superior
or inferior to written instructions on the basis of the learner's (1) amount
of Assistance required, or (2) overall Tile required, or (3) Attitude
toward the learning experience. Therefore, the data of this study have not
clearly indicated that either mode of pronpting, audio or written, is
preferred.

Because longer term effects of the two prorpting methods may differ
from short term effects, the Efficiency dependent variable is looked to for
further information to clarify the choice between audio and written prompting
methods for perceptual mnotor skill learning.

Higher efficiency, as it has been defined here, is not a desirable
end in itself, and therefore it alone is not a sufficient reason to choose

either mode of prompting. However, if greater efficiency will eventually

73

lead to improvement in the terminal objectives investigated, that is,
Assistance, Time or Attitude, then higher efficiency is a means to the
desired end.

The findings of the study do show that in the early stages of
learning a perceptual motor skill task, audio instructions have higher

Efficiencies than their written counterparts, that is, they are more

 

quickly implemented. Therefore, the data of this study does clearly
indicate that audio prompting is preferred to written prompting for the
purposes of higher efficiency in the early stages of training.

It remains, however, to show that the higher efficiency of audio
prompting may eventually result in greater success on the terminal objec-
tives. Toward this end, conclusion 6 states that audio prompts are easier
to understand than written ones. The audio students generally understand
the prompt easier because their eyes are free to examine the components
being explained.gppipg_the explanation. On the other hand, the written
group must shift their eyes back and ferth between the paper and object
being explained. Therefore it may be reasoned that the audio group will
suffer less fatigue over extended periods of instruction than will the
written group. Less fatigue could be expected to result in better all around
perfbrmances on the terminal objectives concerning.Assistance, Time and

Attitude. So it is reasonable to expect higher Efficiencies for the audio

 

group, to eventually, over time, show up as differences in the other
dependent variables in the study. This is a conjecture which goes beyond
the data at hand and therefbre cannot be a conclusion of the experiment.

It can be a recommendation, however, and is given in the next section.

74

RECOMMENDATIONS FOR SELECTION
AND USE OF FROMPTS

The following recommendations are offered as a result of this

 

 

study.
1. When developing prompting materials fer use in the training
of skill tasks (such as equipment operation), and your ob-
jectives are to:

A. minimize outside assistance required by the user, or

 

B. minimize overall pipe_required by the user, or
C. have favorable learner attitude
Then:
i. If the training period is short (less than one hour)--
Audio and written prompting work equally well. Use
whichever costs the least.

ii. If the training period is prolonged (more than one
hour, perhaps done every day)-<Audio prompting is
likely to be superior to written prompting.

2. If one chooses audio prompting, and it is not inconvenient for
the learner to also see written instructions--each frame on
the written copy should be numbered and the current frame
number should be given at the end of each audio frame. (This
will help the learner preview and review adjacent frames.)

3. If the skill task to be learned is at all complicated-~provide
some kind of prompts, learning aids or human assistance during

the student's hands-on learning time.

RECOMVIENDAT IONS FOR FUTURE RESEARCH

 

The following three recommendations are given as logical extentions

of this research.

75

Further research on the fatigue that results under audio

and written prompting, used for long periods of time, should
be done to further substantiate recommendation 1, part ii above--
that audio prompting is likely to be superior to written
prompting because it is less fatiguing.

Audio prompting is generally delivered more slowly than
equivalent written prompting. A user controlled speech com-
pressor might be introduced to speed the delivery of audio
prompts.

For future research into psychomotor tasks, Analysis of
Covariance should be considered as a means of reducing within
group variability to increase the statistical power. A timed
psycho-motor task given all students before the experiment,

could be used to provide the co-variable.

REFEREMIES

 

Abedor, Allan J. "Development and Empirical Test of a Lbdel for
Formative Evaluation of Self-Instructional Mnlti-Media Learning
Systems." Prepared for: 1972 Annual Meeting of the Association
for Educational Communications and Technology; Research and Theory
Division, April 19, 1972.

Alter, M. and Silverman, R.E. The Response in Programmed Instruction,
Journal of Programmed Instruction, (1962), 1, 55-78.

 

Anderson, Richard C. "Educational Psychology," P.R. Farnsworth, et. al.
(Ed) Annual Review of P§ychology, (1967), 18, 103-164.

 

Anderson, Richard C. and Faust, Gerald W. Educational Psychology The
Scch of Instruction and Learnipg, Dodd, Mead and co. , New York,
1973, pp. 187-230.

 

Borg, Walter R. and Gall, fibredith D. Educational Research: An Intro-
duction. New York: David McKay Company Inc. , 1971.

 

Campbell, Donald T. and Stanley, Julian C. Experimental and Quasi-
Experimental Designs For Research. Chicago: Rand McNally and Co. ,
1963.

 

Crowder, Norman A. "Automatic Tutoring by Means of Intrinsic Programming,"
in Eugene Galanter (Ed), Automatic Teachm: The State of the Art.
New York: John Wiley and Sons, Kc” 1959, pp. 109-16.

 

Foulke, Emerson, ed. Proceedings of the Second Louisville Conference
on Rate and/ or Fregpnenct ControlIed SpeeEh, Louisville ,Tentucky,
October 22-24, 1969, p. 2.—

 

 

Gates, A. I. Recitation as a Factor in Memorizing. Archives of P§ychology,
(1917), 26 (Whole No. 40).

 

Glass, Gene V.; Peckhan, Percy D.; Sanders, James R. "Consequences of
Failure to Meet Assumptions Underlying the Analysis of Variance and
Covariance." Review of Educational Research, Vol. 42, No. 3, (Summer
1972): 237-273.

 

Cropper, George L. and Kress, Gerard C. Jr. A Summary Report, Report 4 ~-
Studies in Televised Instruction, Individualizin Group Instruction.
PittsEurgh: American Insfitute for Research in Be vioral Sciences,
1966.

 

 

Holland, J. G. and Kemp, F. D. A Measure of Programming in Teaching-Machine
haterial, Journal of Educational Psychology, (1965), 56, 264-269.

 

Kimble, Gregory A. and wulff, J. Jepson. "Response Guidance" as a Factor
in the value of Student Participation in Film Instruction, A..A.
Lumsdaine (Ed) Student Response in Programmed Instruction, National
Academy of Sciences - National Researcth6unci , wa§hington, D.C.,
1961, pp. 217-240.

 

Kirk, Roger B. Eycperimental Design: Procedures for the Behavioral Sciences.

 

Belmont, Ca.: Brooks/Cole Publishing Co., 1968, pp. 245-270.
76

77

Klaus, D. J. The Art of Auto-Instructional Programming. Audio-Visual
Communication Review, 1961, 9, l30-l42.--4_74_, 4_8£.

 

Kress, Gerard C. Jr. The Effects of Pacing on Pro amned Learning Under
Several Administrative Conditions. Pittsbur : American Institutes
for Research, 1966.

 

Lumsdaine, Arthur A. Some Conclusions Concerning Student Response and a
Science of Instruction (Ed). Student Regponse in Programmed
Instruction, National Academy oFSciences-qNational Research Council,
Washington, D.C., 1961, pp. 471-500.

 

 

Margolius, Garry J. and Sheffield, Fred D. wtiimmm Methods of Combining
Practice with Filmed Demonstration in Teaching Complex Response
Sequences: Serial Learning of a Mechanical-Assembly Task, A. A.
Lumsdaine (Ed) Student Response in Programmed Instruction, 1961,
pp. 33-54.

 

Merrill, M. David and Boutwell, Richard C. Instructional Development:
Methodology and Research. Fred Kerlinger (Ed) Review of Research
in Education, 1973.

 

 

Merrill, M. David and Tennyson, R. D. Attribute Prompting Variables in
Learning Classroom Concepts. Journal of Educational Psychology,
1972.

 

Orr, David B. "Time Compressed Speech--a Perspective," Journal of
Communication, XVIII (1968), 288.

 

 

Postlethwait, S. N. Teaching Tools and Techniques: An Audio-Tutorial

Approach to Teaching. Robert A. Weisgerber (Ed) Develo ntal
Efforts in Individualized Instruction, F. E. PeacocE RISIisfiers,

Inc., Itasca, 111., pp. 314-320, 1971.

 

Postlethwait, S. N. and Novak, J. and Mnrray Jr., H.T. The Audio-
T‘utorial Approach to Learning. Burgess Publishing Co. ,
Minneapolis, Minn., 1969.

 

Postlethwait, S.N. and Novak, J. and Murray, Jr., H.T. Independent Study
Session: Excerpts from a Week's Unit of Study. Robert A. Weisgerber
(Ed) Developmental Efforts in Individualized Instruction, F. E.
Peacock Pubiishers Inc., Itasca, I11., pp. 325-325, 197I.

 

Price, Jack A. The Influence of Three Listenin -Environment Factors on
Listening Comprehen§ion of Fourfh-Grade St ents. Unpublished Ed.D.
dissertation, Ndfth Texas State University, 1969.

 

 

Spitzer, H. F. Studies in Retention. Journal of Educational Psychology,
(1939), 30, 641-656.

 

Stolurow, L. M. and Walker, C.C. A Comparison of Overt and Covert Response
in Programmed Learning, Journal of Educational Research (1962), 55,
421-429.

 

78

Tobias, 8. and Weiner, M. Effect of Response Mode on Immediate and
Delayed Recall from Programmed Materials, Journal of Programmed
Instruction (1963), 2, 9-13.

 

Travers, R.M.W. et a1. "Learning as a Consequence of the Learner's Task
Involvement-We} Difficult Conditions of Feedback." Journal of
Educational Psychology. Vol. 55, 1964, 167-73.

Urbach, Floyd. "The Psychomotor Domain: General Considerations,"
NSMI (Ed) The Psychomotor Domnain, Gryphon House: Washington, D.C. ,

1972, pp. 7—42.

Van Wagenen, R.K. and Travers, R.M.W. "Learning Under Conditions of
Direct and Vicarious Reinforcemnen ," Journal of Educational
Psychology, Vol. 54, 1963, 356-62.

 

weisgerber, Robert.A. Pegspggtives in Individualized Learning, F. E.
Peacock Publishers, Inc., Itasca, Ill., 1971, pp. 50-51.

APPENDIX A

AMPEX STUDENT ATTITUDE SCALE
AND
ASSIGNMENT SHEET

AMPEX ATTITUDE SCALE

These questions refer to the non-Tnmnan pronpts or other aids yon

received while practicing using the Ampex SlOO‘Yideo Tape Recorder (VTR).

>. o >.o
H H 0) H0)
60 m H one
a o o :4 on c an
o o o 4H m o m
£356 ‘“ 8 .23 M
an 5? 2: c: unc:
SA A. N D SD
1. The aid was not useful. 5 4 3 2 l
2. The aid I received was sufficient. 5 4 3 2 l
3. Learning to operate the VTR was more 5 4 3 Z l
difficult than it needed to be.
4. The aid I received was very good. 5 4 3 2 l
5. I would like to see these kinds of aids 5 4 3 2 l
(checksheet, etc.) used in other modules.
6. The aid slowed me down. 5 4 3 2 l
7. I should have had more aid. 5 4 3 2 l

79

80

ASSIGNNENT SHEET

LOAD, THREAD, RECORD, PLAY & REWIND
Do not daudle, you are being timed.

I. PRACTICE
*A. TOAD the tape.
B. THREAD (proceed to C after threading is Oked.)

at timne of recording. (Hint: Reset the counter, record to 15,

E C. RECORD 15 units, by the counter. Subject matter to be identified
4. rewind to 0.)

D. PLAYBACK the recording until Oked.

 

*E. REWIND tape completely back onto the reel.
F. Remnove tape, and take w reel; turn machine off.

G. Repeat A-E one time.

II. TEST - Do parts A-E above.

 

 

MICHIGAN STATE UNIV. LIBRARIES
TIl\lWWINHIIIillIHHIIWIHIIIMITIHIHHIHIIIIHIHII
31293103514786