0 .

 

 

    

1:12...
W '

 

   

       

. .... U A ..T....

.. ,. m ”....HP..,..
mmub.

- M . m. . . s a!

            

.

IE

.'1

as

us:

 

    

  

.~-I
'.‘

. .

 

1

 

i

‘.

m

u

0

ION?
'e"
{’13

 

 

.‘
. .

 

.€.: 3

PW

..
:o.

I,
55819

    

'o
..l.

i

A.

1,-

. U.
for ., ,e.

'4

01:1.

.
“Ed.-
"' “ |.'l
an“-
.. ',.\- n

0

art

?

This is to certify that the

thesis entitled

An Adaptation of Programmed Instruction for
Selected Telecommunication Audience Analysis
Concepts: Deve10pment and Evaluation

presented by

Erik Linn Fitzpatrick

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Higher Education

 

 

Major professor

0-7639

LIBRARY

Michigan State
University

 

 

ABSTRACT

AN ADAPTATION OF PROGRAMMED INSTRUCTION FOR SELECTED
TELECOMMUNICATION AUDIENCE ANALYSIS CONCEPTS:
DEVELOPMENT AND EVALUATION

By

Erik Linn Fithatrick

The purpose of this study was to provide a viable
aid in the teaching and learning of selected telecommunication
audience analysis fundamental concepts. Toward that end,
the specific purposes of the study were:

1) To develop and evaluate for retention and applica-
tion effectiveness, in relation to traditional classroom
lecture-recitation instruction, a series of short, branching
programmed instruction units dealing with broadcast audience
studies fundamentals.

2) To evaluate the developed programmed units in
regard to their relative teaching efficiency; that is,
learner time expended in grasping the programmed content as!
'compared to learner time consumed in conventional classroom
instruction.

3) To assess student differences in attitudes toward

.3"
(9% Erik Linn Fithatrick
the developed audience analysis programmed instruction units
and conventional classroom lecture instruction.

It was suggested that programmed instruction was a
promising method by which to improve the learning process at
the undergraduate fundamental level and that a systematic
investigation should be made of this instructional procedure.

Forty undergraduate Television and Radio students
enrolled in a Broadcast Management course at Michigan State
University were randomly assigned to one of two instructional
conditions: non-directed branching programmed instruction,
and conventional classroom lecture instruction.

An objective post-test over the content of three
prepared instructional units for both experimental groups and
a Likert scale questionnaire dealing with student attitudes
was administered to each of the subjects. Mean learner time
spent with the programmed instruction treatment was compared
with the known mean time taken for conventional lecture
instruction.

Cognitive retention/application post-test student
scores, total subject attitudinal scores and mean learner
times were analyzed for statistically significant differences
between experimental groups using a portion of the.Finn
Univariate and Multivariate Analysis of Variance, Covariance

and Regression computer program for the post-test-only

control group research design.

Erik Linn Fitzpatrick

The results of this study (all statistically signifi-
cant differences are at the .05 level of confidence) can be
summarized as follows:

a) Students taught via programmed instruction scored
significantly higher than students receiving conventional
classroom lecture instruction on the same objective post-
test which assessed retention and limited application of
telecommunication audience analysis fundamental information.

b) Programmed instruction subjects spent significantly
less time in learning identical material taught by means of
non-directed programmed instruction and traditional classroom
lecture instruction.

c) Students taught selected audience analysis funda-
mentals through programmed instruction possessed significantly
more favorable attitudes toward their mode of instruction than
did students taught by conventional classroom lecture means.

In brief, experimental subjects taught selected tele-
communication audience analysis concepts by programmed
instruction retained significantly more information, consumed
Significantly less time in instruction and reported signifi-
cantly more favorable attitudes toward their mode of
instruction than did students taught via conventional class-

room lecture presentation.

AN ADAPTATION OF PROGRAMMED INSTRUCTION FOR SELECTED
TELECOMMUNICATION AUDIENCE ANALYSIS CONCEPTS:
DEVELOPMENT AND EVALUATION

By

Erik Linn Fithatrick

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Administration-Higher Education

1974

ACKNOWLEDGMENTS

With sincere appreciation I wish to thank Dr. William
Sweetland for his curricular guidance, professional counsel
and constant encouragement. Special thanks are extended to
Dr. John Abel for his unselfish willingness to direct the
experimental research and his perceptive service in the
compilation and editing of the final manuscript. I am also
most grateful to the other members of my Guidance Committee,
Drs. Norman Bell and Richard Featherstone, without whose
respective research interests and faculty efficiency concerns
the present study might never have reached fruition.

Next,] would like to express my indebtedness to Mr.
Robert Carr who assisted in processing the collected data,
and Mrs. Virginia Foster who assisted with the typing and
reproduction of the programmed instructional units. Apprecia-
tion is also extended to Dr. Leroy Olson for his statistical
aid in implementing the item analyses for the cognitive
post-test instrument.

To Deborah, I wish to express my unreserved gratitude
for her seemingly infinite patience, unfailing inspiration

and otherwise indefatigable personal support.

ii

TABLE OF CONTENTS

LIST OF TABLES .

LIST OF FIGURES

Chapter

I.

II.

III.

IV.

THE PROBLEM, RATIONALE, AND RELATED RESEARCH .

Need .
Purpose .
Theory and Related Research
Programmed instruction- -related
Attitude- related . . .
Mass communication instruction- -related .
Hypotheses . . . . . . . . . . .
Overview . . . . . . . . . . . . . . .

EXPERIMENTAL DESIGN AND METHODOLOGY

Subjects

Instructor

Treatments

Procedures

Instruments .

Testable Hypotheses

Design and Statistical Analysis
Summary

ANALYSIS OF RESULTS

Dependent Variable: Cognitive Achievement
Dependent Variable: Time Consumed
Dependent Variable: Attitudes

Summary . . . . . . .

SUMMARY AND DISCUSSION .
Summary

Discussion . .
Suggestions for Further Study

iii

Page

vi

APPENDICES . . . . . . .

A.

HEC'HL‘UUO w

(.4

LIST

PROGRAMMED INSTRUCTION FOR SELECTED AUDIENCE
ANALYSIS CONCEPTS . .

BEHAVIORAL OBJECTIVES FOR AUDIENCE ANALYSIS
INSTRUCTION UNITS . . .
DIRECTIVE TO CONVENTIONAL LECTURE SUBJECTS . .
DIRECTIVE TO PROGRAMMED INSTRUCTION SUBJECTS .
CONVENTIONAL LECTURE ATTITUDINAL INSTRUMENT .
PROGRAMMED INSTRUCTION ATTITUDINAL INSTRUMENT
COGNITIVE TEST FOR AUDIENCE ANALYSIS CONCEPTS

ITEM ANALYSIS DATA FOR COGNITIVE POST-TEST .

CROSS TABULATION OF PAIRED BIPOLAR ATTITUDINAL '

QUESTIONNAIRE ITEMS . . .
TIMES CONSUMED BY CONVENTIONAL LECTURE AND
PROGRAMMED INSTRUCTION SUBJECTS .

OF REFERENCES .

iv

41

4.1

7.9
80

82
86

95
96
98
99

LIST OF TABLES

Table Page
'3.1 Univariate ANOVA - cognitive achievement . . . . . 25
3.2 Cell means and standard deviations - achievement . 25
3.3 Univariate ANOVA - mean and adjusted mean time . . 27
3.4 Cell means and standard deviations - time
consumed . . . . . . . . . . . . . . . . . . . . 28
3.5 Univariate ANOVA - attitude scores . . . . . . . . 29
3.6 Cell means and standard deviations - attitudes . . 29
APPENDIX H ITEM ANALYSIS DATA FOR COGNITIVE POST-TEST . 95

APPENDIX I CROSS TABULATION OF PAIRED BIPOLAR
ATTITUDINAL QUESTIONNAIRE ITEMS . . . . . 96

APPENDIX J TIMES CONSUMED BY CONVENTIONAL LECTURE AND
PROGRAMMED INSTRUCTION SUBJECTS . . . . . 98

n»
“1:".

I can

LIST OF FIGURES

Figure

1. Graphic representation of the post-test-only
control group research design .

2. Fourfold representation of possible subject
attitudes toward instruction . . . . . . .

vi

Page

20

22

CHAPTER I
THE PROBLEM, RATIONALE, AND RELATED RESEARCH

E2921.

Virtually all national higher education commissions
have warned that the years ahead for American colleges and
universities will be characterized by general parsimony in
funding and intense pressures for faculty teaching efficiency.
Academic departments of telecommunication can claim no exemp-
tion from the future's financial realities. As the demand for
ever more faculty productivity intensifies (many faculty
members are currently over-extended in the diverse areas of
teaching, research, administrative internal duties and
Community service activities), telecommunication faculty
members must discover more efficient means than conventional
lecture—recitation by which to convey the fundamentals of
their undergraduate subjects so as to reserve their valuable
"free time" and occasional classroom hours for the kinds of
teaCheI‘~student interactions which are totally inimitable.
Certainly prograImned instruction cannot and should not
5UPP13nt face-to-face "classroom" discussion, but it can
perhaps provide the basic concepts with which potential
telecommunication practitioners and scholars must be equipped,

l

while releasing faculty personnel from the unproductive
drudgery of continually presenting the somewhat mundane,
but nonetheless essential, principles of broadcast studies
to each new succession of undergraduate students.
Traditionally content to subject their students to
a seemingly endless chain of in~classroom "courses," tele-
communication divisions have been remiss in exploring newer
Ways by which students might be taught--a reluctance which in
the future could lead to a loss of student, industry and
general public trust. A mode or modes of instruction must be
developed-~this study suggests branching programmed instruc-
tion--by which mature students of telecommunication audience
analysis can gain a rudimentary understanding of the under—
lying principles within their subject area without physically
attending scheduled "classes" day after day, month after
month. If broadcast faculty members are to approach maximum
Productivity, given available resources, alternatives to
traditional undergraduate classroom lecture-recitation for

instruction in fundamentals must be actively sought.

Purpose

The purpose of this study is to provide a viable aid
in the teaching and learning of telecommunication audience
analysis fundamentals. Toward that end, the specific purposes
0f the study are:

1) To develOp and evaluate for retention and

application effectiveness, in relation to traditional class-
room lecture-recitation instruction, a series of short,
branching programmed instruction units dealing with broad-
cast audience studies fundamentals.

2) To evaluate the develOped programmed units in
regard to their relative teaching efficiency; that is,
learner time expended in grasping the programmed content as
compared to learner time consumed in conventional classroom
instruction.

3) To assess student differences in attitudes toward
the developed audience analysis programmed instruction units

and conventional classroom lecture instruction.

Theory and Related Research

 

This study's components draw from three diverse areas
of knowledge: the extremely broad subject of programmed
instruction, the indefinitive domain of attitudes and the yet
emerging field of mass communication—related instruction.

Reviewed literature relating to these three spheres is cited.

Programmed instruction-related

 

The relatively brief history of programmed instruction
began with the research of Pressey (1926, 1927, 1932).
Initially, he was interested in the scoring of tests and

built a machine that did not move to the next question until

5::

min

9v
VA

au 1
”I

,
6..

..
on.

.f‘,‘
~‘~\

the student correctly answered the question he or she was
currently studying. Thus, the student knew immediately whether
or not he or she was correct; the student received immediate
feedback. Pressey found that students using this machine

and getting feedback actually learned the material on which
they were being tested. But Pressey had no real influence

on educational theory or practice at the time he published

his results. His work was generally ignored by educational
institutions for a number of years.

Programmed instruction did not reach general attention
and influence until the publication of Skinner's article in
1954, "The Science of Learning and the Art of Teaching."
According to Skinner, a child performed learning tasks in
order to escape aversive stimuli. He suggested that princi-
ples of learning theory be employed in a method of teaching.
Later, Skinner (1954, 1958) described a machine in which the
important features were immediate reinforcement for the right
answer and individually self-paced learning. The material
is presented in sequential order and the student responds
overtly to the material. There is an interaction between
the student and the material to be learned. On the basis of
reinforcement theory, the student is rewarded immediately
for each correct response. Such a program usually consists
of small steps or frames, each containing 20 to 30 words.
This is a stimulus-response pattern with immediate feedback

of results. Skinnerian programs are called linear programs

due to their step-by-step presentation. Skinner's ideas were
accepted by many who were interested in improving practical
aspects of education.

Crowder (1959, 1960) developed what is termed branching
programmed instruction. He asserted that people in daily
life are confronted with situations in which they must make
decisions and that in these circumstances they make mistakes.
When a learner makes a mistake, Crowder stated, he or she
should be branched. The term "branched" means that the
material to be learned is explained to the learner in a new
way or the learner is directed to a remedial sub-routine.
Thus, a branching frame usually contains more material than
a linear frame. Answers to the frames determine how the
student is branched through the different parts of the program
and how many frames he or she will study.

Some of the characteristics of programmed learning are
common to all methods. Macdonald-Ross (1969) summarized the
research on programmed instruction in the 19605 and listed
several common characteristics of programmed learning.

First of all, programmed learning requires clearly stated
objectives. They are necessary in order to define precisely
what is going to be taught (what the student is to know at
the completion of instruction) and to enable the develOpment
of ways of measuring whether learning has in fact taken
place (criterion test). The objectives should be behavioral

objectives, written in observable and measurable terms.

Another characteristic is the feedback control made
available in programmed instruction; the immediate checking
of the right answer serves as a reinforcer. Anderson, Kulhavy
and Andre (1971) reported a study of 356 subjects who completed
a computer-based instructional lesson which insured that the
subject responded before he or she received knowledge of the
correct response (KCR). Students who received KCR after
every frame performed better on the criterion test than stu-
dents who were tested on the same criterion without receiving
KCR. It was suggested that students attempting to read the
XOR before responding performed poorly on the criterion test.
An additional common characteristic of programmed
instruction is the fact that it is essentially an individual-
ized method of instruction. Each individual progresses at
his or her own pace in making responses and receiving feedback,
or in the case of branching programs, follows a unique path
through the learning process according to individual responses.
To this point the theory and assumptions underlying
linear and branching programmed instruction have been
presented. Another vital question is: Do students learn
from programmed instruction? Schramm (1964) in his review
of the research on programmed instruction mentioned 36
experiments comparing different types of programs with
conventional classroom instruction. Sixteen of these
experiments were done in colleges, four in secondary schools,

five in primary schools, 10 with adults and one with retarded

children. Of the 36 experiments, 18 showed no significant
differences when the two groups were measured on the same
criterion test. Seventeen showed a significant increase
for classroom students working with programmed instruction,
and only one showed a favorable significant increase for
students taught in a conventional manner. In addition, a
number of studies have shown that programmed material is
more effective or efficient for retention of information,
learner time spent and application of learned material than
are prose-text and conventional modes of instruction (Pressey,
1963; Hough, 1962b).

Roe (1962) reported research conducted with 189
freshmen engineering students. The program consisted of
an introduction to probability concepts written in linear
and branching forms. Significant learning occurred with
both programs, but when considering programs separately,
there were no significant differences between any of the
simple branching methods and the linear program when measured
by the amount of learning that took place. However, it was
determined that a logical sequenced program resulted in more
learning over time than did a random sequenced program.

Conversely, Shull (1969) reported a difference
between branching and linear programmed instruction. One
hundred and twenty students enrolled in an industrial arts
course were assigned to one of three groups: control,

linear and branching. The control group received only a

post-test. The linear group received linear programmed
instruction and the post-test and the branching group
received branching programmed instruction and the post-test.
The findings indicated a significant difference between the
three experimental groups at the .05 level of significance

in favor of the branching group.

Attitude-related

 

An additional issue relevant to this study is the
attitude of a given student exposed to programmed instruction.
Dawes (1972) has described the nebulous nature of

attitudes in the following manner:
In 1935 Gordon Allport observed that "attitudes are
measured more successfully than they are defined."
This statement is still true today (p. 2).
Although the term "attitude" is elusive, Osgood
(1956) stated that attitudes are (l) learned, (2) implicit
and (3) the basis for an evaluative response. Thus, attitudes
are referred to as "tendencies of approach or avoidance"
or as "favorable or unfavorable."
Zimbardo and Ebbesen (1969) summarized the general-
evaluation nature of attitudes:
Attitudes have generally been regarded as either
mental readinesses or implicit predispositions
which exert some general and consistent influence
on a fairly large class of evaluative responses (p. 34).

Katz and Stotland (1959) considered that attitudes have

three major components. The information component forms the

foundation on which the attitude is built. The affective
component involves feelings and it is this segment which
attitude scales attempt to measure. The third component,
the action component, represents the extent to which the
attitude has associated with it any habits of action.
Dealing specifically with attitudes toward programmed
instruction, O'Toole (1964) stated that there is evidence
that attitudes will be a strong factor in the successful
implementation of programmed instruction in a school setting.
Nauman (1962) described the performance reactions of
approximately 40 college students who worked the first third
of the Holland-Skinner psychology program written in 1961.
Seventy-five percent of the subjects indicated that without
a teaching machine they would have learned much less from
the course. The use of self-instructional programs was
favored by 64 percent of the group. Other reactions of
students reported in this study include: (1) about one-
fourth of the students felt that at some point during the
course they were treated like experimental organisms. A
small minority felt that the use of teaching machines
reflected adversely upon their dignity as human beings.
(Z) About two-thirds of the group stated that they thought
the instructor was trying to teach as much as possible
within the limits of basic reality consideration. (3) The
final point was a statement about the possible lack of

opportunities to reflect on the material learned and to

tnqnfifi
quJO'
. ‘l
34
fivhla
1
131

10

consider its implication.

Tobias (1969) assessed the attitudes of students in
educational psychology classes as they pertained to non-
traditional media and conventional instruction. Student
attitudes accounted for a substantial percentage of achievement
variance. Subjects with attitudes favorable to conventional
instruction tended not to learn as well from unfamiliar
media such as programmed instruction.

Scherman (1973) found student sex category to be a
determinant of favorable or unfavorable attitudes to programmed
instruction. Male students expressed generally positive
attitudes to programmed instruction, while females were equally
split between favorable attitudes to programmed instruction

and conventional instruction.

Mass communication instruction-related

There is a dearth of programmed instruction research
(or any other form of instruction research) in the general
area of mass communication course-work content, the reason
for which is not altogether clear. Directing his remarks to
America's educational broadcasters, Eurich (1968) stated that
although educational systems generally lag behind other
societal institutions in the adoption of innovations, higher
education is perhaps the most reluctant of all to change.
Eurich's assertion finds support in the literature.

Three research studies are reported, all of which deal

with programmed instruction of selected journalism rudiments;

330 0
‘43.“

I. Lev

ZICEC

11

two of the three studies are unpublished doctoral disserta-
tions. No research is reported in the area of this study--
broadcast audience analysis and programmed instruction.

Francois (1967) compared programmed instruction in
beginning news writing to conventional instruction in the
subject. He found no significant difference in news writing
ability between the two experimental groups, but he reported
that programmed instruction proved to be more efficient
than conventional instruction, with learner time used as the
measure of efficiency. Boredom with the linear programmed
instruction and a lack of concrete, real-life experiences
were the salient perceived weaknesses.

Prejean (1968) constructed a linear programmed unit
on journalistic style with frames logically grouped and
compared learning gain through programmed and conventional
instruction. She found that programmed self-teaching outside
of the classroom was as effective as conventional lecture
in terms of retention and application, and that programmed
instruction usage provided extra class hours for material
requiring teacher-student clarification.

In measuring the effectiveness of programmed instruction
as a method for teaching basics of elementary photography
(film exposure and flash), Griffith (1969) found that in eight
comparisons of programmed and conventional instruction, three
showed no significant differences and five comparisons showed

a significantly higher level of achievement for students

'3‘
oh»

are

.
.11

12

receiving programmed instruction. The Griffith findings

are consistent with those reported in the general programmed
instruction literature in which either no significant
differences were found between programmed and conventional
instruction in terms of retention and application, or when
differences were found, they were generally in favor of
programmed instruction. Griffith also concluded on the basis
of his comparisons that non-directed programmed instruction
was the most efficient form to utilize.

As aforementioned, programmed instruction research in
this study's area of telecommunication audience analysis is
nonexistent, but Meeske (1972) has alluded to the implementa-
tion of measurable behavioral objectives (the prime consti-
tuents of programmed instruction) in broadcast course-work
instruction and the seemingly universal reluctance of broad-
cast faculty members to innovate:

It is possible to determine the skills a (broadcast)

course is designed to impart, to devise tests to

see if the students can demonstrate those skills,

and to apply the tests. The simple fact is that little
effort has been made to do so (p. 221).

Hypotheses

 

To explore the questions raised by the investigator,
based upon the findings of the literature and general field
observations, the following hypotheses were developed.

1. Students receiving branching programmed instruction

will score significantly higher than students

receiving conventional classroom lecture instruc-
tion on the same objective post-test that assesses

l3
retention and application of broadcast audience
analysis fundamental information.

2. Students taught by means of branching programmed
instruction will consume significantly less time
than students taught via conventional classroom
lecture instruction in learning identical
telecommunication audience analysis fundamental
concepts.

3. Students taught broadcast audience analysis
fundamentals through branching programmed instruc-
tion will possess significantly more favorable
attitudes toward their mode of instruction than

will students taught by traditional classroom
lecture means.

Overview

In Chapter I the problem was introduced, theory and
related research were reviewed and hypotheses were stated.
In Chapter II the subjects, treatments, procedures and
instruments will be described; the research design identified,
and the tested hypotheses, statistical analysis and level of
significance presented. In Chapter III the results of the
data analysis will be reported. Chapter IV will conclude
the study with a discussion of the results and some suggestions

for further research.

CHAPTER II

EXPERIMENTAL DESIGN AND METHODOLOGY

Upon reviewing the related literature, stating the
hypotheses, and describing the purpose(s) of the investigation
as presented in the previous chapter, an experimental study

was planned and subsequently conducted.

Subjects

Students enrolled during the Spring term of 1974 in
Television and Radio 401 (Broadcast Management) at Michigan
State University served as treatment subjects for the study.
Of the total enrolled students available, approximately one-
third were excluded from treatment because of their prior
knowledge gained in Television and Radio 335 (Audience
Studies); that is, any students reporting that they had
experienced the audience studies course-work were excluded
from any treatment for the purposes of the study.

The Television and Radio undergraduate subject
population was preponderantly male-~nearly 80 percent of the
students were males. The studied subjects, with few excep-
tions, fell into an age range of 20 to 23 years. Without

exception, all of the subjects were either Juniors or Seniors

14

15

and all were Television and Radio majors. After accounting
for students excluded from the study due to prior knowledge
and other miscellaneous exemptions (such as the inevitable

few students withdrawing from the management course), the
studied population consisted, quite conveniently, of precisely
40 students, 20 served as programmed instruction subjects and

20 as conventional lecture subjects.

Instructor

 

No formal instructor was anticipated or necessary for
the non-directed programmed instruction component of the
study. The conventional lecture component required an in-
classroom instructor and that instructor was the scheduled
Television and Radio faculty member assigned to the broad-

cast management undergraduate class.

Treatments

 

A short course consisting of three units in selected
telecommunication audience analysis fundamentals was designed
by the investigator. The designed fundamentals course was
then written into two forms: branching programmed instruction
(Apppendix A), and paragraphed text to be used in conventional
lecture. These instructional materials served as the treatment
conditions for the study. The short audience analysis

fundamentals course was built upon conventional, Mager-style

16

behavioral objectives for discriminative forms of behavior

(Appendix B).

Procedures

 

Within the studied Television and Radio Management
class (excepting those students who were automatically
excluded for the previously stated reasons), each individual
student was randomly assigned to one of the two treatment
groups--conventional lecture or programmed instruction.

During the first class meeting of the term, students
to be excluded from the study were identified through self-
report. Approximately two weeks later--allowing ample time
for class student numbers to stabilize due to the customary
"drops" and "adds"--the random assignment of students to
treatment groups was accomplished.

In actuality, students to be "excluded" from the
study results did receive the conventional lecture presenta-
tion, but their cognitive scores and attitude indices were
not calculated for the data analysis purposes of the study.

Students assigned to conventional lecture treatment
from their instructor were required to attend two succeeding
class periods (Appendix C), while subjects assigned to
programmed instruction treatment were gathered at one prede-
termined time in a separate, reserved classroom to receive
the developed non-directed programmed units (Appendix D).

In short, subjects were given an instructional treatment

f
env-

1%.:

“PW.

.
A

17

corresponding to the experimental group to which they had
been randomly assigned. The instruction was in one of two
forms: conventional lecture, or branching non—directed
programmed instruction.

Conventional lecture subjects were admonished to take
clear, reasonably detailed class notes and to review those
notes at their leisure. Programmed instruction students
were advised to complete their individual units at one
study session; that is, without undue interruption, just
as one might receive one-hour lecture-recitation.

Programmed instruction subjects were also asked to review
their completed instructional units at their leisure, and

one important additional requirement was that each programmed
instruction subject indicate on the front page of his or her
instructional unit how much time in minutes was spent in
completing the unit. Absolutely no "help" (instructor
clarification of terms, intended meanings of phrases, etc.)
was afforded the non-directed programmed instruction student,
while conventional lecture subjects were allowed the usual
classroom interactions characteristic of traditional

instruction.

Instruments

 

Following the scheduled lecture and programmed
instruction treatments, the two instructional treatment groups

were assembled at the same hour in their usual classroom to

18

evaluate the audience analysis "course" they had experienced
through a l6-item Likert scale attitudinal questionnaire
(Appendices E, P). Then, a 30-question application and
retention post-test constructed from representative areas

of the instructional treatment units was administered
(Appendix G). Both of the evaluative instruments--the
attitudinal questionnaire and the cognitive post-test--were
developed by the investigator.

Item difficulty for the cognitive achievement post-test
was determined via a standard item analysis using extreme
criterion groups (Appendix H). During develOpment of the
post-test, the scores of tested subjects (Television and Radio
majors who had experienced audience studies course-work) were
selected at the upper 27 percent and lower 27 percent ends
of the rating continuum, as suggested by Borg and Gall (1971),
and a determination made as to whether different proportions
of subjects in each extreme group answered a given test
item correctly.

Attitudinal questions dealt with the clarity and
organization of presented materials, time spent in receiving
instructional presentations, the degree to which instruction
motivated a given student to seek additional information in
the subject area, overall stimulation from the course and
other general affective "feelings" concerning the treatments.
An attempt was made to check the consistency of subjects'

responses to the attitudinal items by cross tabulating

19

individual responses to three pairs of items considered to
be biqpolar in meaning; students reporting negative attitudes
to one item might be expected to express positive attitudes
towartl another item with identical, but positively stated

conterit (Appendix I).

Testable Hypotheses

 

In order to assess the difference between the two
treatments employed as represented by the measures used,
mnl on the basis of the hypotheses developed and described

in Chapter I, the following testable hypotheses were

fornuilated:
1 H0 MP (cog. achiev.) = ML (cog. achiev.)
H1 MP (cog. achiev.) # ML (cog. achiev.)
Key: - null hypothesis

- alternate hypothesis

H
H
M - mean of group
L - lecture

P

- programmed instruction
There will be no difference in post-test mean scores

between the two treatment groups when the dependent variable

is cognitive achievement.

20

MP (time) = ML (time)

H : MP (time) # ML (time)

There will be no difference in minutes consumed for
instruction between the two treatment groups when the

dependent variable is time.

MP (attitudes) = ML (attitudes)

Mp (attitudes) # ML (attitudes)
There will be no difference in mean scores between
the two treatment groups when the dependent variable is

attitudes.

Design and Statistical Analysis

 

The post-test-only control group design, using the
notation of Campbell and Stanley (1966), can be graphically

represented as shown in Figure l.

 

 

.Programmed Instruction R X 01

 

 

 

 

. Conventional Lecture ' R ‘ 02

 

Figure 1. Graphic representation of the post-test-only
control group research design.

Key: R - random assignment of subjects to groups
0.- post-test
X - treatment

21

According to Campbell and Stanley, the design is well
suited to "avoiding an experimentor-introduced pre-test
session, and in avoiding the 'giveaway' repetition of identical
or highly similar unusual content, as in attitude change
studies" (p. 26). The design may be considered as the last
two groups of the Solomon four-group design.

An analysis of variance test for significance of
differences between variances was made to determine whether
tfluzbetween-groups variance was significantly greater than
the within-groups variance.

It was determined prior to undertaking the study that
the .05 level of confidence would be used in interpreting "F"
ratios and "p" values. The study design assumed the two
experimental groupS'UJbe basically alike in such characteristics
Gas aptitude scores, sex distribution, years of television
and radio formal education and major field of study.

The mean learner time spent using programmed instruction
was compared with the known mean time consumed by the conven-
tional lecture treatment group (Appendix J).

As the collected attitudinal data regarding programmed
instruction and traditional lecture resembled that depicted
in Figure 2, the mean group attitudinal scores were analyzed

as any other dependent variable.

22

 

 

 

 

Attitudes Favorable Unfavorable
C.L. ## %%
P.I; ## %%

 

 

 

Figure 2. Fourfold representation of possible subject
attitudes toward instruction.

The five possible alternatives for each question on
the Likert attitudinal instrument were ordered so that a
weight of "one" was given the most negative expressed attitude
and a weight of "five" was ascribed to the most positive
attitude. A total index or "score" of attitudes (which were
assumed to be unidimensional) was calculated and tested for
differences between experimental treatments.

All of the investigated dependent variables (cognitive
achievement, time, and attitudes) were analyzed using a
portion of thelthIUnivariate and Multivariate Analysis of

Variance, Covariance and Regression computer program.

Summary

Undergraduate Television and Radio students enrolled
in a Broadcast Management course at Michigan State University
were randomly assigned to one of two instructional treatments:

1. Non-directed programmed instruction.

2. Conventional classroom lecture.

23

An objective post-test over the content of the three
prepared instructional units for both treatment groups and a
Likert scale questionnaire dealing with student attitudes
was administered to each of the subjects. Mean learner
time spent with the programmed instruction treatment was
compared with the known mean time taken for conventional
lecture instruction.

Cognitive retention/application post—test student
scores, total subject attitudinal scores and mean learner
times were analyzed for statistical significance between
treatment groups using an analysis of variance test for
significance for the post-test-only control group experimental

design.

CHAPTER III

ANALYSIS OF RESULTS

The Control Data 6500 computer system at Michigan State
University was employed in calculating the statistical
analyses. Use of the Michigan State University computing
facilities was made possible through support, in part, from
the National Science Foundation. The level of confidence for

all of the significance testing was .05.

Dependent Variable: Cognitive Achievement

 

A Finn univariate analysis of variance program was
used to test for differences between experimental treatment
(programmed instruction) and control (conventional lecture
instruction). The primary ANOVA components of the test for
cognitive achievement differences are presented in Table 3.1.
Observed cell means and standard deviations comprise Table 3.2.

Inspecting Table 3.1, it appears that there was a
significant difference in performance between the experimental
groups on the 30-question objective post-test for retention
and limited application of fundamental audience analysis

concepts (as measured by mean test scores).

24

25

Table 3.1. Univariate ANOVA - cognitive achievement.

 

 

 

 

Source df MS F P less than
Between 1 140.625 41.634 .0001
Within 38 3.377
Total 39
Ke df - degrees of freedom
MS - mean squares

F - F ratio

P - P value: the probability of obtaining an
F ratio as large or larger than
the calculated F ratio if the
null hypothesis is true.

Realizing a "p" value of less than .0001, the next
step in analyzing the data was to consult the calculated cell

means presented in Table 3.2.

Table 3.2. Cell means and standard deviations - achievement.

 

 

 

Group N Cell Means Standard Deviation
P.I. 20 28.850 .988
C.L. 20 25.100 . 2.403

 

On the basis of the data expressed in Table 3.2, a
conclusion can be drawn that the significant difference
found through ANOVA was in favor of the treatment group--

programmed instruction.

26

It should be noted that a ceiling effect is evident
in the results; few students scored below 25 correct on
the 30-item post-test, and a perusal of Appendix H reveals
that the test was not a particularly good discriminating
instrument. The nature of the test itself (Appendix G)
predicts a simple chance score of 12 and a perfect score of
30. For purposes of least- and most-knowledgeable discrimina-
tion, an ideal overall mean score would be approximately 21.
The actual mean score for the post-test obtained by the two
experimental groups (treatment and control combined)
approached 27 correct.

Nevertheless, the "p" value contained in Table 3.1 is
small, and the sufficiently small standard deviation figures
suggest that the difference between treatment and control
was statistically significant. Table 3.2 reports that the
significant difference is in favor of the programmed instruc-

tion treatment group.

Dependent Variable: Time Consumed

 

The statistical computer analysis of the data for the
dependent variable time spent with instruction proved to be
rather a formality and foregone conclusion. First, there
could be no within-group variance of total time spent for
the conventional lecture control group because all lecture
subjects consumed the same amount of time. Secondly, and

more importantly, the two most deliberate programmed

27

instruction subjects used a total of only 35 minutes apiece
to complete their three short units of instruction and thus
complete the deve10ped "course" for fundamentals. In
comparison, the conventional lecture control group used

94 minutes (close to three times more minutes than the
slowest-paced P.I. subjects) in receiving the audience
analysis course content.

Such a wide discrepancy in time consumed by the two
experimental groups indeed constitutes a statistically
significant difference in favor of the programmed instruction
treatment group. Table 3.3 details the univariate ANOVA

results for group-mean/adjusted-mean times.

Table 3.3. Univariate ANOVA - mean and adjusted mean time.

 

 

 

Source df MS F P less than
Between 1 105,125.000 2122.608 .0001
Within 19 49.526

Total 20

 

A study of the cell means in Table 3.4 points out
that the ANOVA significance testing utilized adjusted mean
time for conventional lecture subjects (due to the fact
that there could be no within-group variance) and actual
or "real" mean time for the programmed instruction group

(within-group variance did exist).

28

The total time figures for the two experimental groups
(Appendix J) explain that the real time taken for completion
of the three units of instruction was 94 minutes for the
conventional lecture control group and a mean of 21.5

minutes for the programmed treatment group.

Table 3.4. Cell means and standard deviations - time consumed.

 

 

 

: Mean Time Adjusted Standard
Group N (minutes) Mean Time Deviation
P.I. 20 21.5 ---- 7.037
C.L. 20 ---- -72.5 7.037

Adjusted mean time for the lecture control group was
arrived at by adding the mean time of the programmed instruc-
tion group (21.5) to -94 minutes. The negative number was
necessary to test the null hypothesis that mean time minus
94 minutes is equal to zero.

It is eminently clear (at first blush and upon ANOVA
statistical analysis) that a significant difference existed
between the two experimental groups when the dependent
variable was time taken to complete three units of audience
analysis instruction. The known uniform time taken by
conventional lecture was 94 minutes vs. 21.5 minutes mean
time consumed by programmed instruction subjects-~a minutes

ratio of approximately four to one.

29

Dependent Variable: Attitudes

 

Once again, a univariate ANOVA computer program was
used to analyze the data and again it is apparent that a
significant difference existed in total attitudinal scores
between the experimental groups. Table 3.5 presents the
results of the ANOVA significance test for differences with

attitudes as the dependent variable.

Table 3.5. Univariate ANOVA - attitude scores.

 

 

 

Source df MS , F P less than
Between 1 366.025 4.154 .0486
Within 38 88.103

Total 39

 

Having established that a significant difference
obtained between groups, the investigator next considered

the cell means contained in Table 3.6.

Table 3.6. Cell means and standard deviations - attitudes.

 

 

Group N Cell Mean Standard Deviation

 

P.I. 20 58.500 7.104
C.L. 20 52.450 11.213

 

30

The analyzed data point to a significant difference
between experimental group attitudinal scores in favor of
the treatment programmed instruction subjects. Due to the
weighting of the attitudinal responses, higher group mean
scores are to be interpreted as more positive attitudes toward
a given mode of instruction; in this case, the programmed
group students expressed significantly more positive
attitudes toward their mode of instruction.

As established in Chapter 11, an attempt was made to
determine how consistent subjects were in their attitudinal
responses (the attitude instrument itself was considered to
be unidimensional). The complete cross tabulations are
encapsuled in Appendix I, but it may be generally stated
that subjects' responses to the attitudinal items were
rather consistent.. If the cognitive achievement instrument
can be characterized as a somewhat poor discriminator, then
the attitudinal instrument proved to be its contrast--it

apparently possessed considerable discriminating power.

Summary

Three investigated dependent variables were analyzed
for differences between experimental treatment and control
groups using a Finn univariate ANOVA computer program. The
dependent variables and the results were as follows:

1. Cognitive Achievement--a significant difference

between groups was found in favor of the programmed

31

instruction treatment group (obtained significantly higher
post-test scores).

2. Time Consumed in Instruction-~a significant
difference between groups was found in favor of programmed
instruction (used significantly less time to complete three
units of course content).

3. Attitudes-~a significant difference between groups
was found in favor of programmed instruction (subjects
expressed significantly more favorable attitudes toward
their mode of instruction).

The level of confidence for all of the significance
testing was .05. A discussion of the above statistical

results will be presented in the following chapter.

CHAPTER IV

SUMMARY AND DISCUSSION

Summary

Developing three branching units of non-directed
programmed instruction for selected telecommunication
audience analysis concepts and testing their relative
effectiveness to conventional classroom lecture instruction
was the general purpose of this study. Toward that overall
end, the more specific purposes of this study were: 1) to
evaluate the programmed units in the light of relative
cognitive achievement of experimental subjects; 2) to
evaluate the programmed units in regard to their relative
teaching efficiency, with time consumed in instruction as
the efficiency measure; and 3) to evaluate expressed student
attitudes toward the programmed units in relation to tradi-
tional classroom lecture instruction. It was suggested that
programmed instruction might be a viable means by which to
improve teaching productivity for' audience analysis funda-
mentals and that a systematic study should be undertaken of
this instructional procedure.

The investigator developed three short programmed

units touching upon rudimentary principles of audience studies

32

33

such as a definition of a rating, an outline of the most
popular audience measurement method and the primary reason
for conducting audience research. The units were subsequently
tested at Michigan State University. The hypotheses advanced
were: 1) students receiving programmed instruction will
achieve significantly greater post-test scores than will
conventional lecture subjects; 2) a significant difference
between groups in favor of programmed instruction will be
found in time consumed by students learning identical
information; and 3) programmed students will possess signifi-
cantly more favorable attitudes toward their mode of instruc-
tion than those taught by conventional means.

Forty undergraduate Television and Radio students
were randomly assigned to one of two instructional conditions:
1) non-directed branching programmed instruction; or 2)
conventional classroom lecture instruction. A post-test for
cognitive achievement and an attitudinal questionnaire were
administered to both of the experimental groups. Programmed
instruction subjects were additionally required to report
the time(s) taken in completing their instructional units.

The three dependent variables of the study--cognitive
achievement, time consumed in instruction, and expressed
attitudes toward mode of instruction--were tested for signifi-
cant differences between treatment (programmed instruction)
and control (conventional lecture) using a Finn univariate

analysis of variance computer program.

34

Discussion

 

The results of the present study (all statistically
significant differences are at the .05 level of confidence)
can be summarized as follows:

a) Students taught via programmed instruction scored
significantly higher than students receiving conventional
classroom lecture instruction on the same objective post-test
which assessed retention and limited application of tele—
communication audience analysis fundamental information.

b) Programmed instruction subjects spent significantly
less time in learning identical material taught by means of
non-directed programmed instructionzunitraditional classroom
lecture instruction.

c) Students taught audience analysis fundamentals
through programmed instruction possessed significantly more
favorable attitudes toward their mode of instruction than did
students taught by conventional classroom lecture means.

In brief, experimental subjects taught by programmed
instruction retained significantly more information, took
significantly less time in instruction, and generally felt
better about their method of instruction than did students
taught by conventional classroom lecture.

In regard to the above statistically significant
differences, the question might logically be posed: "Are
these differences socially as well as statistically

significant?" A reasoned answer may be: "Yes, with some

35

reservations."

First, a conscious effort was made to ensure that
investigator-introduced biases were held to a minimum.
For example, the programmed units were developed by the
investigator, but the conventional lecture presentations
were prepared by the regularly scheduled classroom instructor
(a Television and Radio faculty member), not the investigator.
Also, the investigator attempted to meet with the experimental
subjects as little as possible. But, most importantly, the
programmed instruction units were completely non-directed;
no questions from programmed subjects were fielded and no
instructional help offered. Meanwhile, conventional lecture
subjects were allowed the usual classroom interactions such
as clarification of misunderstood concepts, multiple questions
(although due to the fundamental nature of the information,
few questions were actually posed) and the sorts of other-
wise general human contacts that should have provided a wider
stimulus array and, thus, a learning advantage.

It must be stated, however, that even though efforts
were made to reduce possible biases in the research,
the experimental subjects did vaguely know that they were
participating in "some sort of study" and the programmed
instruction treatment students may have fallen under the hazy
glow of a "halo" effect due to their unusual instructional
condition. Also, conventional lecture subjects were

"required" to attend class (Appendix C) but no attendance

36

was taken (to duplicate a true daily lecture setting).

Hence, a few lecture subjects came to and left from the
classroom at erratic intervals, as they normally did during
their usual Broadcast Management lecture presentations,

but most subjects (perhaps due to simple courtesy) did not
leave the classroom until the completion of the respective
unit lecture periods--some of these students, left to their
own devices, might have actually grasped or "learned" the
audience information in a shorter period of time. Programmed
students were all at least physically presented with the
prepared instructional materials, but it must be recalled
that whether or not programmed instruction subjects actually
chose to give their units anything more than cursory attention
was left entirely to their discretion (to duplicate a true
non-directed programmed setting). It is not known if these
differences in attendance vs. attention produced anything
other than randomly distributed biasing effects, however some
comfort can be found in the knowledge that each instructional
method was employed according to its inherent behavioral
properties.

The literature concerning programmed instruction suggests
that programmed instruction enjoys the potential to be more
effective than other traditional modes of instruction (most
notably, conventional lecture), but even programming's
enthusiasts recognize limitations. None of the most ardent

see programmed instruction replacing the instructor himself;

37

they realize that individualized conceptual learning must
‘ of necessity be so varied that no self-instruction can supplant
the interactions between human beings. Filep (1962)
emphasized twelve years ago that programmed instruction should
become "a part, not the sole teaching method, a measure, not
an end," and that the use of self instruction should not
"absolve the instructor of his responsibilities" (p. 175).

It is essential to note that the programmed units
evaluated in the present study dealt with telecommunication

audience analysis fundamentals for undergraduate students.

 

Had the programmed units incorporated more advanced material
(e.g., the phi1050phical implications of collecting personal
mass media behavior data from supposedly "anonymous"
respondents), the significant differences found in favor of
the programmed instructional treatment might easily have
been swayed in the direction of conventional classroom
lecture.

The results of this study, then, can be said to suggest
promising new directions in the effective, efficient teaching
of telecommunication audience analysis fundamentals. The
present study results do not constitute a statistical diatribe
aimed at traditional lecture-recitation instruction or those
who practice such instruction. Indeed, Gagne (1970) has
observed that probably the finest teaching available recognizes
and uses the intrinsic differences of dissimilar modes of

instruction to full advantage, each instructional method

38

complementing the other:

It may be that the most striking effects of instruc-
tional planning are to be sought in various combina-
tions of media, where each may perform a particular
function best (p. 61).

Suggestions for Further Study

 

The sheer bulk of telecommunication audience analysis
fundamentals available for instruction is considerable.
Perhaps a second and third series of units could deal with
more advanced concepts for undergraduate students. In the
interest of physical dimensions, these succeeding units
could omit the principles detailed in the "beginner's"
units.

The electronic media have become pervasive tools in
American higher education, and the value of programmed
audience analysis units could be enhanced by structuring them
for electronic media such as video tape, or most easily, the
interactive computer. The investigator has written an intro-
ductory APL computer program dealing with audience analysis
problem-solving instruction and the future of similar
applications is promising.

Another variation of programming the audience analysis
units, through either the medium of print or one of the
electronic media, would be to compile the information into
Skinner's linear form explained in Chapter I.

These suggestions obviously place much emphasis on the

39

deve10ped programmed units themselves, but audience research
is to practitioners and scholars of telecommunication as the
periodic table of elements is to the chemist—-all telecommuni-
cation observation, theory and investigation hinges upon a
systematic knowledge of differing audiences.

The recommendations thus far have been in the interest
of the student and instructor. There are studies of equal
importance in research where quantitative data would be of
assistance.

Examination of the data from this experimental study
suggests further statistical investigation. Pre- and post-
tests could be administered to determine actual gain in
learning, but such experimentation should allow considerable
time lapse between the two measures so as to minimize
pre-test information effects. Also, the relationship of
significantly greater cognitive achievement to intelligence
test scores or grade point averages warrants experimental
style, using undergraduate telecommunication majors. One
of the more interesting postulates is the effect, if any, of
programmed instruction on long-term learning retention of
audience analysis information. Since mastery of telecommuni-
cation audience analysis fundamentals is an accepted quality
of professionalism, and since the audience principles receive
constant use, a series of programmed audience analysis units
should provide ideal treatment material for longitudinal

study of long-term retention.

40

Finally, much additional experimentation is required
to validate the present study's audience analysis programmed
units, the cognitive achievement post-test and the Likert
scale attitudinal instrument. Such validation evolves only
through an empirical research and development cycle of testing,
analyzing and revising of programmed instructional units,
cognitive achievement instruments and attitudinal question-

naires--a cycle which demands extraordinary patience and even

greater quantities of time.

APPENDICES

APPENDIX A

APPENDIX A

PROGRAMMED INSTRUCTION FOR SELECTED

AUDIENCE ANALYSIS CONCEPTS

 

Three Non-Directed Branching Units for

Students of Telecommunication

 

Compiled By
Erik L. Fithatrick
April, 1974

...——

Not for distribution-~contains c0pyrighted material
incorporated for limited research purposes under
provision of academic fair use.

41

42

Time Spent Completing
this unit:

 

Your Name:

 

Student Number:

 

UNIT ONE

 

WHO NEEDS BROADCAST "RATINGS? AND WHAT ARE THEY?

Module I. -- The Main Reason for Audience Measurement

Advertising agencies who plan and place broadcast
campaigns must be able to choose the "best" vehicle for
their selling purposes, and that means measuring audiences
to determine not only the numbers of persons reached by
messages but their ages, income levels and so forth.

Other less important reasons for conducting audience
research include station programming decisions, evaluation
of on-air personnel (particularly in radio), and long-range
station policy planning.

Module II. ~- What the Ratings Really Mean

What's a rating? Let's talk in terms of television.
An audience rating is a statistical estimate (not a
guess) of the number of homes viewing a program as a
percent of all homes owning a television set. For
example, a rating of 20 for a network television program
indicates that 20% of U.S. television homes tuned in
the program. All audience ratings are based upon samplings.

Why use sampling? Sampling is clearly the best way.
No one could afford a complete "census" of the television-
viewing population and, even if they could, the "census
of viewing" would have to be conducted far too often in
order to detect changing audience trends. Besides, the
U.S. Census Bureau checks the validity of its data with
a national sam 1e. Without sampling, how could a fire-
cracker manufacturer test the quality of his product and
still have a product left to sell?

Are 1,165 homes enough? The A.C. Nielsen Co. samples
about that number of homes nationally to determine network
TV viewing. Okay, so sampling works (anyone who has had
a blood test can agree), but isn't this stretching things
a bit? No, statisticians state that a sample of 1,000

43

households provides ratings of about the same accuracy
for a nation of over 50 million households as for a city
of only 100,000. (But, it is far more difficult to
select a quality 1,000 home sample for the U.S. than for
a city of 100,000.)

Can sampling measure not only firecrackers but people?
The idea that sampling is no match for man may be ego-
building, but it's incorrect. The failure of the Truman
vs. Dewey political polling is often used to show how
sampling can't deal with peOple, but the major source of
error in political polling stems from predicting and Egg
from sampling. A pollster is trying to pre 1ct what
people will do on election day by what they tell him on
a given day prior to the election; he is measuring future
behavior and that's very risky. What are ygp going to do
next week? You get the idea. Ratings aren't predictions!
They report only what people did after the fact, not
before. Nielsen for example, does not report what people
plan to watch or "expect" to watch on television.

 

How accurate are ratings? As we have seen, ratings are
arrived at by sampling and they are statistical estimates.
This means that although ratings are expressed as numbers
(Johnny Carson pulled a 10 rating), they don't have the
precision we usually associate with a number. Statisticians
refer to the indefinite nature of ratings numbers as
"sampling error." As a general rule, the larger the
sample, the smaller the sampling error. The idea to
remember is that just because a rating of 10 is reported
for a given segment of the "Tonight Show," that does not
mean that exactl 10% of U.S. television homes watched—The
program-~tfie truth is somewhere "around" 10 (it could
easily be 9 or 11). To repeat, ratings points are not
exact numbers; they are estimates.

How should ratings be used? A first rule of carpentry
is: "only use a tool the way it was designed to be used."
It applies equally to the use of ratings. Keep the limi-
tations in mind. Remember that a rating is not a
precise number and that it shouldn't be interpreted
as such. If you are a station manager and the "competi-
tion" is "beating" you by a ratings point or two (say
30 for you and 31 for them), don't get too excited and
fire the entire staff from janitor on up. On the other
hand, if "they" are whipping you by 20 ratings points, you
might try appointing the janitor as manager!

(NOW, TURN THE PAGE AND WORK PROGRESS CHECK 1)

44

PROGRESS CHECK 1

 

The main reason for measuring broadcast audiences is:

(a) because it makes for good reading.

(b) advertisers are interested in the data.

(c) chambers of commerce often require it.

(d) the "seal of good practice" depends upon the results.
A rating of 20 for a network TV program indicates

(a) an advertiser would have to pay 20 dallars to advertise

on it. .
(b) the NAB has evaluated the program's worth with a score
of 20.
(c) of the U.S. homes with television sets, 20% were
watching.

(d) 20 critics thought that the program would ease a
fertilizer shortage.

Sampling has come to receive a bad name. Why?

(a) Most of the problems have been with prediction of
human behavior.

(b) The problems have been in describing past behavior
after the fact.

Ratings companies frequently report what pe0ple will watch
in the future.

(a) true
(b) false

In terms of ratings accuracy, what is the important point
to recall?

(a) ratings-points are precise numbers.

(b) ratings are statistical estimates with ranges of error.

(c) ratings are program evaluations by columnists.

(d) ratings are to be taken at their face number value
without hesitation.

(NOW TURN THE PAGE AND SCORE YOUR PROGRESS CHECK)

01‘th

45

ANSWER KEY: PROGRESS CHECK l

 

(IF YOU MISSED MORE THAN ONE QUESTION, TURN THE
PAGE AND READ THE "VERBOSE EXPLANATION” FOR UNIT
ONE. IF YOU MISSED NONE OR ONLY ONE, YOU ARE READY
TO WORK THROUGH UNIT TWO AT YOUR CONVENIENCE.)

 

H I
.40»;

46

UNIT ONE: VERBOSE EXPLANATION

 

Module 1. -- The Main Reason for Audience Measurement

Every manufacturer of goods or purveyor of services
using broadcasting to any degree, and every advertising
agency concerned with planning and placing broadcast
campaigns stands to benefit from audience research.
Knowing and being able to select the audiences most
likely to purchase a given firm's products (older persons
for denture cleansers, teenagers for acne preparations,
for example) can result in a very real competitive
advantage. Audience measurement supplies this vital
data to advertisers to help them peddle their wares.
Remember, it is not only the big-volume advertiser who
has a stake in efficient use of broadcasting. Indeed,
the smaller advertiser with a limited budget must be
especially aware of the audience(s) he is attempting

to reach.

Module II. -— What the Ratings Really Mean
What's a rating? Let's confine our discussion to
television, although it's the same idea for radio.
In television, an audience rating is a statistical
estimate of the number of homes viewing a program as
a percent of all homes having a television set. A
rating of 20 for a network television program indicates

that 20% of U.S. TV homes tuned in the program.

47

Ratings of this sort aren't confined to television.
Radio, magazines and newspapers all use ratings to report
statistical estimates of the sizes of their audiences.
These ratings are similar to batting averages, with one
important difference: all audience ratings are based
upon samplings.

Because of sampling, statisticians call audience

ratings estimates. This is their way of saying that a

 

rating is subject to a margin of statistical error
because it is based upon information obtained from a
sample and not from the entire population.

Why use sampling? Is there a better way than sampling
to obtain TV ratings? People have suggested that the
best way to get accurate information about television
viewing is to check everybody, so as to avoid the use of
sampling. But a little reflection shows this simply is
not practical.

A complete census of television would be prohibitively
expensive. A census costs many millions of dollars. And
a "census" of television viewers would have to be taken
many times per year-~almost every week--so that sponsors
and broadcasters would know public reaction to their
continuing changes in program offerings.

Sampling is used in television research just as
sampling is used throughout industry, agriculture, medicine

and government (to mention only a few areas of application)

48

as the best way of obtaining information. The Campbell
Soup Company knows that if they keep stirring, one can
of tomato soup tastes just as good and exactly the same
as the rest of the kettle. The quality of wheat, milk
and cotton are all checked by examining small samples.
City water systems are checked daily for impurities by
samples of a few gallons out of millions. For a blood
test, even the least mathematical doctor takes only a
few drops.

But can sampling measure people? Many have implied
that to use sampling we must reduce the richness of our
experience to the simplest common denominator, that
sampling is no match for man. The argument may be ego-
building, but it is incorrect.

The talented writer, Goodman Ace, once quipped:
"Polls are fascinating. They are read by everyone from
the farmer in the field all the way up to Tom Dewey,
President of the U.S."

The famous failure of political polling in the 1948
Presidential election is often cited as an example of
the inability of sampling to c0pe with peOple. But was
the predicted election of Tom Dewey an indictment of
the sampling process? Not at all.

The major source of error in political polling comes

from predictipg and not from sampling. A pollster who

 

 

had questioned every registered voter in the country

49

about his voting intention before an election would be in
substantially no better a position to predict the outcome
of the election than his rival who had used proper
sampling techniques. This is true because some voters
are undecided about how they are going to vote, others
'change their minds after they are asked and still others
don't make it to the polls at all. In a close election--
such as in 1948--these peOple can make the difference.
But TV ratings aren't predictions. They report only
what peOple did after the fact, not before. Nielsen,
.for example, does not report what programs people plan
to watch or expect to watch on television.
Of course the behavior of people is more difficult
to measure than the characteristics of an inanimate
object. ‘A family's choice of TV programs will change
from one week to another. But this doesn't mean that
a family's TV viewing behavior can't be accurately
measured; it means only that TV viewing behavior must

be measured more often.

.How accurate are ratings? How should ratings be
used? Re-read these sections in the initial presentation
of UNIT ONE:‘ THEN, turn this page and worE PROGRESS
CHECK 2.

50

PROGRESS CHECK 2

 

It is essential to measure the commercial broadcast
audience. Why?

(a) Because it's there.
(b) Advertisers want the data.
(c) It builds character.
(d) It isn't essential at all.

The 6:00 p.m. News receives a rating of 50. What does
it mean?

(a) 50 stories were presented.

(b) 50 people work for the station, probably in the
research department.

(c) Of the 100 TV homes in the market, 50 are tuned to
the 6:00 News.

(d) All 50 of the community's leaders have evaluated the
program.

Which statement best characterizes ratings?

(a) Ratings are accurate predictors of future viewing
behavior.
(b) Ratings report only what pe0ple did after the fact.

The first rule of carpentry applies to ratings, too.
What is it?

(a) Always get your fee in advance.

(b) Use a tool only as it was designed to be used.
(c) Use a tool any way you can; get the most from it.
(d) Never work any harder than is necessary.

If you were a rating, how would you describe yourself?
(a) a precise number
(b) beautiful

(c) an estimate
(d) a phalarope

(NOW TURN THE PAGE AND SCORE YOUR PROGRESS CHECK)

h M
o 0

0‘0‘

51

ANSWER KEY: PROGRESS CHECK 2

 

(IF YOU MISSED MORE THAN ONE QUESTION, BETTER
RE-READ THE SECTIONS THAT GAVE YOU DIFFICULTY:
THEN MOVE ON TO UNIT TWO AT YOUR DISCRETION.

IF YOU MISSED NONE OR ONLY ONE, YOU ARE READY TO
WORK THROUGH UNIT TWO AT YOUR CONVENIENCE.) A

52

Time Spent Completing
this Unit:

 

Your Name:

 

Student Number:

 

UNIT TWO

ONE METHOD FOR GATHERING AUDIENCE DATA
AND SOME DIFFICULTIES IN MEASUREMENT

 

 

Module I. -- The Most Popular Measurement Method: The Diary

Diaries are by far the broadcast industry's primary
source of audience information. More commercial time
is bought and sold on the basis of diary-produced
estimates than by any other method. The actual diaries
used, placement and sampling procedures vary between
audience research companies, but the principle is the
same--people are asked to keep a written record of their
listening or viewing behavior.

 

In addition to basic viewing (speaking in terms of
television), diaries usually furnish information about
the viewing audience. Diaries supply detailed instruc-
tions for recording separately, for each day of the week,
time viewed, station, program and audience composition.

Some of the advantages of the diary method: it
provides a record of who is viewing rather than merely
set Operation; adaptable to any size geographic area;
can cover all broadcast hours; and can provide a wide
variety of "other" information very economically. This
"other" information can include the occupation of the
male head of household, which commercial during the week
was considered ”best" by family vote and like topics.

Each week during a survey the sampled households
are changed--the reasons are: 1) the mail delivery of
diaries makes them easy to change from house to house,
neighborhood to neighborhood, and 2) studies have
shown that families will usually record one week's viewing
faithfully and that the viewing record becomes less
satisfactory if a family is retained in the sample for
a longer period of time.

Diary keeping does require an effort on the part of
the families, but studies conducted by the American

53

Research Bureau have found that, when properly approached
and informed as to the purpose of the survey, most families
are quite willing to cooperate. In most cases (contrary

to p0pu1ar belief), no money is given for their efforts,
although token payments have proven useful in chronic
low-c00peration areas (inner city residents are notoriously
reluctant to participate).

Problem areas in audience measurement are discussed
later in Module II, but a few issues are particularly
associated with the Diary. For example, we know that
some diary entries are made days after the actual
viewing has occurred on the basis of hazy memories.
Coding respondent entries to make them digestible for
computers can be an invitation to human and mechanical
error. And, there is the question of possible viewing
differences between those who keep a diary and those
who refuse--they may be different "types" of people.

Module II. -- Problem Areas in Audience Measurement

Differences in data: When two ratings services don't
turn up with the same numbers it is not because of sampling
theor but rather because of sampling ractice. We have
aIreagy seen that sampling is the only IogicaI solution
to today's audience research needs. However, in
implementing sampling theory, differences will occur.

Some of the factors which can contribute to these differences
are: individual corporation measurement techniques

(phone interview, diary, etc.); differing geographical
areas measured; and differing survey dates.

 

Misuses of data: Audience estimates are to be used
only in the environment in which they were obtained.
National ratings cannot be directly applied to local areas;
they were not obtained in any one "local" area but
rather in a series of areas that by themselves mean
nothing statistically. The use of ratings in a vacuum
is another misuse. Ratings alone cannot and should not
carry the entire burden when appraising a station as an
advertising vehicle. . . there are important differences
in types of audiences as well as sizes.

Standard deviation: We repeat that a rating is an
estimate, an approximation. When a program is rated at
10 (remember, that's 10% of the potential viewing homes--
those with TV sets), it means about 10, not exactly 10.
The plus or minus range of the rating will depend upon
standard deviation. The important thing to remember is
that this "range" can be measured mathematically.

54

The non-response factor: Audience measurement firms
are constantly employing new techniques or "gimmicks" to
improve their cooperation rate. But, in the real world,
there will always be those who will not participate in
any form of survey. Once one household refuses, the
carefully preselected sample is no longer really "pure."
The research companies have attempted to identify and
describe these non-respondents, but characterizing them
is very difficult because--what else?--they won't
cooperate.

(NOW TURN THE PAGE AND WORK PROGRESS CHECK l)

SS

PROGRESS CHECK 1

 

By far the most p0pu1ar audience measurement method is

(a) the telephone coincidental
(b) the mechanical device

(c) the personal interview

(d) the diary

Each week during a typical diary survey, the homes are
changed. Why?

(a) The audience research firms run out of money.

(b) Most families will only keep accurate viewing records
for a week.

(c) It is a custom established by the post office to sell
stamps.

(d) Who could watch TV for seven straight days?

Two rating services have reported different data for the
same program.

(a) They probably just trumped up the figures for kicks.

(b) Each one may have used its own individual measurement
technique.

(c) No one trusts them anyway, so why worry?

(d) The trouble is that liars can figure, and these firms
can, too.

Ratings alone should decide which station's commercial
time is purchased.

(a) True--they're fair and objective.
(b) False--they can't tell us enough.

What is probably the worst part of the non-response factor?

(a) We cannot tell if these persons' behaviors might be
different.

(b) Such peOple slam doors and break too many noses.

(c) There is no place for a recluse in modern America.

(d) It bothers those characters in academia, but no one
else.

(NOW TURN THE PAGE AND SCORE YOUR PROGRESS CHECK)

O‘U‘O‘Qa

56

ANSWER KEY: PROGRESS CHECK 1

 

(IF YOU MISSED MORE THAN ONE QUESTION, TURN THE
PAGE AND READ THE "VERBOSE EXPLANATION" FOR UNIT
TWO. IF YOU MISSED NONE OR ONLY ONE, YOU ARE
READY To WORK THROUGH UNIT THREE AT YOUR
CONVENIENCE.)

57

UNIT TWO: VERBOSE EXPLANATION

 

Module I. 7' The Most Popular Measurement Method: The Diary
We have seen that diaries are the primary source of
audience information for practically every audience
measurement firm. More television time is bought and
sold on the basis of estimates produced by the diary
than by 33y other method. The principle is always the
same—-a sample of peOple are asked to keep a written
record of a particular media activity, such as television
viewing or radio listening.
The chief advantage of the diary method is that it can

be designed to obtain a wide variety of information on

 

the sample, in addition to specific media activity. Also
of importance is that the diary can be adaptable to

any size geographical area; it covers all broadcast
hours; it is economical; and, in the case of television
and radio, it can report individual activity, not merely

set Operation (as with mechanical devices).

 

Diaries furnish detailed instructions for recording
separately, for each day of the week, time viewed,
station, program and audience composition. In addition,
the diary gathers information on the age, sex and number
of school years completed by each member of the household;
whether or not the lady of the house works outside the
home more than 35 hours a week; occupation of male head

of household and type of employment; number of TV

58

sets being used in the home and so on.

Although this requires no small effort on the part of
diary keepers, research firms have found that, when
properly approached and instructed as to the purpose of
the survey, families are usually quite willing to cooperate.
In most cases, no direct compensation is given for their
efforts, although a token payment may be made in low-
cooperation areas, such as inner city dwellings.

Each sample used for diary surveys is completely
changed and composed of new and different families
every week during a survey and from survey to survey.

Most families will keep one week's record of their viewing
and, after that, the record becomes less satisfactory

if the family is retained in the sample for a longer
period.

To review briefly the advantages of the diary method--
it provides a record of individuals viewing rather than
merely set Operation; it is adaptable to any size area;
it can cover all broadcast hours; and it can provide a
wide variety of information economically. But these
advantages hrnatheir meaninig if the original source-~the
diary itself--is lacking in validity or is inaccurate in
any way. We know, for example, that some diary entries
may be made on the basis of hearsay or the "guesstimate"
of the diary keeper. As with any written questionnaire,

the transcribing of information from the point of its

59

inception to the final report is subject to human and
mechanical error. And, there is the question of possible
differences in viewing habits between those who keep a
diary and those who do not.

Of course, many of the errors do tend to at least
partially cancel out in the end results. A number of
tests have been conducted which compare the results of
a diary study with the results of other techniques of
audience measurement, such as personal and telephone
interviews conducted in the same areas at the same time.
As a general conclusion, it can be stated that for
practical business decisions, the diary is the most
valuable technique in providing audience estimates for
broadcast industry use.

To revisit a point: the diary is Ehg most p0pu1ar
measurement device currently used in broadcast audience
research. It is cheap, reasonably reliable and quite

versatile.

Module II. -- Problem areas in Audience Measurement
Differences in data: Differences in data between
one research firm and another do not occur because of
sampling theory but rather because of sampling practice.
There is undeniable evidence all around us that sampling
is, indeed, the only solution to the audience research

needs of today. However, it is in implementing the

 

60

sampling theory that differences in data will occur.
Some of the factors which can contribute to these

differences are: individual corporation measurement

 

techniques (phone, personal interview, diary, etc.);
differing survey dates; and differing geographical areas
measured.

Misuses of data: There are obviously cases when
estimates are misused within the broadcasting and
advertising industries--these further compound the problems
of audience research. For instance, difficulties arise
when audience estimates are used out of the environment
in which they were obtained, such as applying a local
rating nationally or applying it to another local area.
Local ratings apply only to the area in which they were
obtained, and at the same time national ratings cannot
be directly applied to various local areas. Before any
survey begins, research companies have a clear understand-
ing of the area to be measured and, when that area is so
defined for the user, it is the only area for which the
estimates are applicable. The use of ratings in a vacuum
is another instance of misuse. Ratings alone cannot and
should not carry the entire burden when appraising a
station as an advertising medium. Some of the best
television campaigns, in terms of sale of merchandise,
have been the lowest-rated in a market. Probably the

best examples of such programs are the daytime women's

61

shows. The secret here seems to be that the products
are all extremely well-matched to reach and sell a big
part of a comparatively small, loyal audience of house-
wives. Today's successful broadcaster recognizes the
importance of audience pypgg as well as audience £133.
He realizes that the lowest cost per prospect is not
necessarily found zumnug the higher rated programs.
Also, station "image" has become important not only

as a measure of responsibility to the viewing public,
but also as a saleable value to prospective clients.

Standard deviation: To repeat, a rating is an
approximation. When a program is rated at 10 (remember,
that's 10% of the potential viewing homes), that means
gpppp 10, not exactly 10. The plus or minus range around
the rating which makes up this "about" depends upon the
standard deviation. The important thing is that this
range in a "good" sample can be measured mathematically.
All audience estimates, no matter who produces them,
have a standard deviation, as does any measurement based
upon sampling.

The non-response factor: Ideally, each home selected
Should participate in the survey in order to create, or
maintain, a pygg probability or "good" sample. But,
practically speaking, there are obviously some peOple
who will never participate in surveys. Once one person

refuses, the sample is no longer really pure. The

62

important consideration, however, is what effect a non-
cooperator has on the overall validity of the resulting
estimates. The American Research Bureau's studies of
cooperating families vs. non-cooperating homes have
stated that these households are essentially the same,
at least in terms of the gmgppp of time spent viewing

television.

(NOW TURN THE PAGE AND WORK PROGRESS CHECK 2)

63

PROGRESS CHECK 2

 

The absolute most popular audience measurement method is

(a) personal interview

(b) mechanical device

(c) diary

(d) telephone coincidental

The same program has been rated by two services and their
data disagree.

(a) One may have used personal interview while one used
diary.

(b) No one knows why these things happen.

(c) The figures are probably just fabricated anyway.

(d) What, me worry?

Should ratings alone decide which station's commercial
time is purchased?

(a) Sure, they're black and white and what could be
fairer.

(b) No way, Often audience types may be more important
than size.

The homes are changed each week during a typical diary
survey. Why?

(a) No one could stand to watch TV for over seven days.

(b) It is a custom established by the post office to
hustle stamps.

(c) Most families will only keep accurate viewing records
for a week.

(d) The rating firms need time to duck out of town.

When we say that a rating of 8 doesn't mean exactly 8,
what do we mean?

(a) It could easily be a bit more or less than 8.
(b) We are just being evasive.

(c) Ratings have no meanings to advertisers.

(d) So, what's this "we" stuff?

(NOW TURN THE PAGE AND SCORE YOUR PROGRESS CHECK.)

64

ANSWER KEY: PROGRESS CHECK 2

 

(IF YOU MISSED MORE THAN ONE QUESTION, BETTER
RE-READ THE SECTIONS THAT GAVE YOU DIFFICULTY:
THEN MOVE ON TO UNIT THREE AT YOUR DISCRETION.
IF YOU MISSED NONE OR ONLY ONE, YOU ARE READY
TO WORK THROUGH UNIT THREE AT YOUR CONVENIENCE.)

65

Time Spent Completing
This Unit: pp

.1

Your Name:

 

Student Number:

 

UNIT THREE

 

SOME TELEVISION VIEWER BEHAVIOR PATTERNS AND
MEASUREMENT SELF-REGULATION

 

 

Module 1. -- A Few Essential Facts and Figures

First of all, virtually every American household
owns at least one television set--about 96%, to get
Specific. Forty percent have two sets or more.

Of those households owning a TV set, the majority
have at least one color TV set lurking under their
roofs—-about 55% of those households.

TV usage tends to be higher in homes with color
sets-~during the weekend ball games, the plus margin for
color-equipped homes is 21%.

The TV audience builds in size throughout the broad-
cast day, and TV viewing reaches its peak (understandably)
in mid-evening (between 8 and 9).

Audience composition changes by season of the year--
viewing in general is highest in the Winter, but there
are also shifts in the composition of the audience
itself. For example, teens watch more late night TV in
the Summer than they do in the Winter.

, TV households average a whopping 49 hours of TV
usage per week-~an average of 7 hours every day.

Women view almost 30 hours of TV weekly—-4 hours more
than men. Of course, most of the extra female viewing
can be attributed to housewives pulling up the average.

Walk into an average American home, and over half of
its evening time will be devoted to television-~and it
doesn't make muCh difference as to income levels; the
affluent like their tube-watching, too.

As might be expected, different types of programs vary
in audience appeal--young adults watch more feature

 

66

films and general drama, but older adults prefer variety
Shows and westerns.

Sunday night still attracts the biggest adult TV
audience-~Friday continues as the night with the least
amount of viewing.

Module II. -- The BRC: Referee for Audience Research

The Broadcast Rating Council (BRC), located in New
York City, oversees the activities of the various rating
services (A. C. Nielsen, ARB, and a host of others) to
ensure that they meet certain standards of audience
research and integrity. To those research firms meeting
its standards, the BRC grants a check-mark of approval.
It's very much like the National Association of Broad-
casters' "Seal of Good Practice"--strictly voluntary
compliance on the part of participating research organiza-
tions. To repeat, the Broadcast Rating Council has no
regulatory power at all. If a research firm should Hippen
to continually violate the BRC standards of performance,
about the most drastic action that the BRC could take
would be to send a guy down to the firm's offices to
scrape their "seal of approval" off the front door with
a razor blade.

Nevertheless, the fact remains that most reputable
research firms engaged in broadcast audience measurement
do adhere to the BRC'S minimum standards for ratings
YEporting and methodology. Just for good measure-~no
pun intended--the BRC conducts periodic audits of the
various research firms. During the course of these audits,
the BRC examines a firm's compliance with the following
sorts of standards:

 

ETHICAL AND OPERATIONAL STANDARDS

 

The anonymity of all interviewers, research
supervisors and field servicemen is to be protected.

If a respondent (such as a family keeping a diary)
has been assured that his anonymity will be protected,
his name, address or other information is not to be
disclosed.

Any sample chosen for an audience study must reasonably
represent that population which it was chosen to measure
(don't choose homes with no males present if you're
trying to measure viewing in "typical" homes across the
country).

67

Appropriate quality control steps must be taken in
regard to data editing, keypunching, printing and other
Operations which might affect the final study results.

The field work of interviewers should be verified
by the use of spot checks to certify that interviews
were indeed completed.

All field personnel shall be thoroughly trained in
their work. They Should know the duties of their
positions, adhere to interview instruction, and avoid
attempts to bias information obtained from respondents.

DISCLOSURE STANDARDS

 

Each ratings report shall mention any omissions, errors
or biases known to the rating service which could influence
the report findings.

If the rating organization has deviated from its
standard Operating procedures, the rating publication
must report the deviation(s).

The geographic areas surveyed, and the reasons for
their selection, should be clearly defined in each
ratings report.

Each rating service Should indicate the normal
sample return for each survey (for example, how many
managed to mail back their diaries) and, if the return
is below normal, the ratings publication must explain the
fact in a prominent position.

If a station has engaged in special, non-regular
promotional activity during a survey period in order to
build an inordinately large audience (called "hypoing" the
ratings), this activity must be published along with the
rating report.

If a rating service has knowledge of possible rating
distorting influences such as pre-emptions, station
failures, etc., the rating report must disclose that
such conditions occurred during the survey period.

(NOW TURN THE PAGE AND WORK PROGRESS CHECK 1)

68

PROGRESS CHECK l

 

Of all American homes, how many own at least one television
set?

(a) 47%
(b) 68%
(c) 32%
(d) 96%

The night of the week which attracts the largest adult
TV audience is

(a) Wednesday
(b) Sunday
(c) Monday
(d) Thursday

Most reputable audience research firms are accredited by
the BRC.

(a) true
(b) false

Any sample chosen for an audience study must reasonably
represent

(a) anything the rating firm can get away with.

(b) the number of persons who purchased television sets
last fiscal year.

(c) that population which it was chosen to measure.

(d) only married couples.

Special, non-regular station promotion during a survey
period is called

(a) doing what comes naturally.

(b) the name of the game in broadcasting.
(c) hypoing the ratings.

(d) insuring the sampled audience.

(NOW TURN THE PAGE AND SCORE YOUR PROGRESS CHECK)

N
o

«504

69

ANSWER KEY: PROGRESS CHECK 1

 

(IF YOU MISSED MORE THAN ONE QUESTION, TURN THE
PAGE AND READ THE "VERBOSE EXPLANATION" FOR UNIT
THREE. IF YOU MISSED NONE OR ONLY ONE, YOU HAVE
COMPLETED YOUR ASSIGNED PROGRAMMED UNITS. DON'T
FORGET TO REVIEW ALL THREE OF YOUR UNITS AT WHAT
YOU CONSIDER TO BE THE APPROPRIATE TIME.)

70

UNIT THREE: VERBOSE EXPLANATION

 

Module I. -- Some of the Necessary Facts and Figures

Ninety-six percent of U.S. households own a TV set--
Television ownership continues to inch up closer to
the point where almost every household owns a TV set.
Only 9% of the households owned TV sets in 1950,
compared to 87% in 1960 and 9§% today.

Majority of households own color sets--Estimated color
set penetration is about 55% of all TV households. The
influx of small solid-state sets has contributed to this
growth figure.

TV usage higher in color set households: PeOple
who own color TV sets tend to watch more television than
those owning only a black and white set. The plus
margin among color-equipped homes ranges from 9% during
late night to 21% weekend daytime.

TV viewing reaches peak in mid-evening: The television
audience builds in size throughout the day as housewives,
school age children and then the working force get into
the action. Between 8 and 9 p.m., estimates Show that
two-thirds of all Eastern and Central time zone TV
households have their sets on. Although viewing follows
the same overall pattern in both of these time zones,
it is interesting to note the lunchtime rise and the
higher mid to late night usage in the Central zone.

Audience composition changes by season: It's no

71

surprise that viewing, in general, is at its highest in
the Winter, but there are some interesting changes in

the composition of the audience by seasons of the year.
For example, non—adults watch more morning and late night
television in the Summer than they do in the Winter.
Overall, women make up the largest share of the audience,
except during Summer mornings and early evening in the
Winter, when they are out—viewed by non-adults. (The
term "non-adults" includes teens as well as children. It
is the teens who account for most late night non-adult
viewers.) The only daypart when total viewing goes down
in the Winter is late night, despite a substantial
increase in TV usage by men.

TV households average 49 hours of TV per week: According
to Nielsen estimates, television households average almost
49 hours of TV usage a week, or almost 7 hours a day.
Highest viewing levels, averaging over 62 hours a week,
were registered among households with five or more
peOple. Also well above average were $10,000+ households
with non-adults.

Women view TV over 4 hours more than men: During
December, women viewed nearly 30 hours of television a
week, over four more hours than men. The group with the
lowest rate were men in households with $15,000+ income,
but even they viewed over 21 hours a week.

Over half of evening time devoted to TV: Household TV

72

usage avearages 216 minutes out of the total 420 minutes
available each evening. (6 p.m.-1 a.m.)

The rate is highest in households with income $10,000-
$l4.999--232 minutes an evening. Households with incomes
$15,000+ view television 213 minutes per evening-~just
three minutes below the composite average.

Types of programs vary in audience appeal: Each
category of TV programming varies in its appeal to
different segments of the audience. Younger adults
watch more feature films and general drama. Older adults
prefer variety shows and westerns. Situation comedies,
along with westerns, also attract the largest number of
non-adults.

Sunday night attracts biggest adult audience: Sunday
night is by far the most popular TV night of the
week among adults. Tuesday has replaced Monday as the
second heaviest viewed night. Friday still continues

as the night with the least amount of viewing.

Module 11. -- The Broadcast Rating Council and Standards
The Broadcast Rating Council Inc. believes that
adherence to a set of minimum standards is necessary to
meet the basic objective of valid, reliable and effective
broadcast audience measurement. Acceptance of these
minimum standards by a rating service is one of the

conditions of accreditation by the BRC. Most reputable

73

audience research organizations 19 comply with the BRC'S
standards of methodology and reporting, but bear in mind
that it is a voluntary compliance -~the BRC has no power
to punish; it can only withdraw its accreditation which,
in the audience research game, can mean a lot to image-
conscious research firms.

During its periodic audits of approved rating
companies, the BRC checks for compliance with minimum

standards such as the following:

ETHICAL AND OPERATIONAL STANDARDS

 

The anonymity of all interviewers, supervisors and
servicemen Should be preserved. The audit firm, however,
would have the right to check with these and any other
apprOpriate persons as part of the auditing process.

If a respondent has been led to believe, directly or
indirectly that he is participating in an audience measure-
ment survey and that his anonymity will be protected,
his name, address or such other identifying information
shall not be made known to anyone outside the rating
service organization, with the following exceptions:

(l) The audit firm of the Broadcast Rating Council,
Inc., in the performance of an audit, or such
disclosures as required in a hearing before the
Broadcast Rating Council, Inc.

(2) The broadcast rating service at its discretion may
permit other reputable research organizations to

reinterview respondents in the conduct of
special research studies.

74

The sample design for each rating report should be SO
constructed to represent, to a reasonable degree, the
universe being measured (households, individuals, sets,
etc.). Where significant deviations are considered, by
the rating service, to be desirable and/or unavoidable,
such deviations will be described clearly in each rating
report.

Appropriate quality control measures Shall be
maintained with respect to all internal and external
Operations which may influence the final results.

Quality control Shall be applied, but not necessarily
limited, to data collection, editing, collating, tabulating
and printing.

All field personnel (including supervisors) Shall be
thoroughly trained in their work. Such training Should
provide assurance that: they know the responsibilities
of their positions; they understand all instructions
governing their work, that they will deviate from such
instructions only when apparently justified by unusual
conditions and that such deviations will be reported in
writing; they recognize and will avoid any act which might
tend to prejudge, condition, misrepresent or bias the
information obtained from respondents.

The field work of each rating service should be
verified by spot checks or other procedures apprOpriate

to the techniques used to verify or inspect the work

75

of interviewers, supervisors and other field personnel.

DISCLOSURE STANDARDS

 

Each report shall mention all omissions, errors and
biases known to the rating service which may exert an
effect on the findings shown in the report.

Each rating report should point out known deviations,
from standard operating procedures of the rating service,
which could also affect the reported results.

Geographic areas surveyed should be clearly defined
in each rating report. In each case, the criteria and/or
source used in the selection of the survey area Shall
be given. (Thus, if the area surveyed is the Metro area
as defined by the U.S. Census, it should be so recorded
in the report.)

Each rating service shall indicate the normal sample
return for each survey, and shall indicate in a prominent
position when the return is below normal but not below
the minimum required for issuance of a report.

If a rating service has established that any station
has employed special non-regular promotional techniques
that may distort or "hypo" ratings, then said rating
service will publish in the apprOpriate report a notice
of this effect.

If a rating service has knowledge of apparent rating
distorting influences such as unusual weather, catastrophes,

political or social events, pre-emptions such as world

76

series, elections, congressional hearings, station failures,
etc., the rating service will indicate in its reports the

existence of such conditions during the survey period.

(NOW TURN THE PAGE AND WORK PROGRESS CHECK 2)

77

’PROGRESS CHECK 2

 

On the average, who watches more TV: a black 8 white or
color set family?

(a) black 8 white-~it's cheaper to operate.
(b) color-~it's more fun to watch.

In general, TV viewing is at its highest in

(a) Summer.
(b) Fall.

(c) Winter.
(d) Spring.

It is mandatory that research firms be accredited by the
BRC.

(a) true
(b) false

What action can the BRC take against those who violate
its standards?

(a) fine them heavily and impose S-year prison sentences.
(b) withdraw their seal of approval check-mark.

(c) kick then square in the pants.

(d) help them beat the system.

If a station is detected "hypoing" its ratings, what
happens?

(a) It is reported in the rating publication for that
survey period.

(b) The station receives the NAB medal of honor.

(c) It is overlooked as the way of the world.

(d) NO one cares, so nothing happens at all.

(NOW TURN THE PAGE AND SCORE YOUR PROGRESS CHECK.)

b M
o o

78

ANSWER KEY: PROGRESS CHECK 2

(IF YOU MISSED MORE THAN ONE QUESTION, BETTER
RE-READ THE SECTIONS THAT GAVE YOU DIFFICULTY.
YOU HAVE NOW COMPLETED YOUR ASSIGNED PROGRAMMED
UNITS. DON'T FORGET TO REVIEW ALL THREE OF YOUR
UNITS AT WHAT YOU CONSIDER TO BE THE APPROPRIATE

TIME.)

APPEND I X B

APPENDIX B

BEHAVIORAL OBJECTIVES FOR AUDIENCE
ANALYSIS INSTRUCTION UNITS

Primary Objective: Upon completion of the branching programmed
instruction materials, given a fixed alter—
native cognitive test of achievement, the
student will identify correct responses
to questions relating to representative
areas of the instructional treatment units
(minimum acceptable level of performance
not established, pending further revision).

 

Sub-Objectives: The student will select from among alter-
native responses the chief purpose of
conducting broadcast audience analysis.

The student will discriminate between the
nature of broadcast rating figures and the
precision associated with whole numbers.

The student will distinguish the advantages
and disadvantages characteristic of diary
audience measurement methodology.

The student will select appropriate
general problem areas in broadcast audience
analysis.

The student will isolate essential facts
and figures pertaining to American tele—
vision viewer behavior patterns.

The student will indicate ethical, opera-
tional and disclosure standards necessary
to achieve valid, reliable and Objective

broadcast audience measurement.

79

APPENDIX C

 

8**8*

APPENDIX C

DIRECTIVE TO CONVENTIONAL LECTURE SUBJECTS

YOU HAVE BEEN RANDOMLY SELECTED to receive a few units
of Broadcast Audience Studies concepts via short
classroom lectures from your instructor.

IT IS ESSENTIAL (AND MANDATORY) that you attend the
class meetings of:

Wednesday -- April 17
and
Friday -- April 19
YOU WILL EARN COURSE CREDIT for your participation,

so be on time and please don't "cut" either of the
above class meetings.

THANKS for your cooperation; your help is greatly
appreciated!

80

 

APPENDIX D

 

***8*

APPENDIX D

DIRECTIVE TO PROGRAMMED INSTRUCTION SUBJECTS

YOU HAVE BEEN RANDOMLY SELECTED to receive a few units
of Broadcast Audience Studies concepts via 3 short
programmed instruction packages.

IT IS ESSENTIAL (AND MANDATORY) that you attend the
class meeting of:

Wednesday -- April 17

Friday -- April 19

YOU WILL EARN COURSE CREDIT for your participation,
so be at class promptly on Wednesday, April 17.
You will receive further instructions at that time.

THANKS for your cooperation; your help is greatly
appreciated!

81

 

APPENDIX E

 

C.

L.

APPENDIX E

CONVENTIONAL LECTURE ATTITUDINAL INSTRUMENT

Affective Questionnaire for Audience Analysis Concepts

Name:

 

Student Number:

 

Sex:

 

The classroom lecture units dealing with Audience Analysis
concepts were too long.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

The classroom lectures held my attention very well; I
was not easily distracted from the concepts being presented.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

I found that I was generally turned off by the lecture
presentation of the audience information.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

82

 

 

83

If I had a friend who was looking for some worthwhile
instruction, I would suggest the presented Audience
Analysis units.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

The repetition within the lectures was very objectionable.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

 

Presenting the Audience Analysis concepts via classroom
lecture proved to be a very efficient use of my time.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

The lecture units tried to present too much material for
the average undergraduate student.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

I probably would not recommend the presented Audience
Analysis units to other T-R students.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

I was stimulated by the audience concepts presented to
the point that I would like to take other audience-related
courses and/or expand my knowledge by self reading.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

10.

11.

12.

13.

14.

84

I find the subject of broadcast Audience Studies very

interesting.

1) strongly
2) agree

3) neutral
4) disagree
5) strongly

agree

disagree

I frequently found myself thinking about the material
presented, after listening to the classroom lectures.

1) strongly
2) agree

3) neutral
4) disagree
5) strongly

The lecture
ppabsorbing
college.

1) strongly
2) agree

3) neutral
4) disagree
5) strongly

agree

disagree
units presented qualify as some of the most

and otherwise dull experiences I've had in

agree

disagree

Overall, taking everything into consideration, I would
rate the content organization of the presented audience
units as excellent.

1) strongly agree

2) agree

3) neutral
4) disagree
5) strongly

disagree

The redundant nature of the lecture units probably served
as a learning aid for me.

1) strongly agree

2) agree
3) neutral
4) disagree

5) strongly disagree

 

15.

16.

85

If the presented Audience Analysis concepts were to be
expanded into an elective "course," I would like to
take the course.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

I frequently found my mind wandering as I sat listening
to the classroom lectures.

I) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

 

APPENDIX E

 

.1.

APPENDIX E

PROGRAMMED INSTRUCTION ATTITUDINAL INSTRUMENT

Affective Questionnaire for Audience Analysis Concepts

Name:

 

 

Student Number: g3

 

Sex:

 

The programmed instruction units dealing with Audience
Analysis concepts were too long.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

The programmed instruction units held my attention very
well; I was not easily distracted from the concepts being
presented.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

I found that I was generally turned off by the programmed
presentation of the audience information.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

86

87

If I had a friend who was looking for some worthwhile
instruction, I would suggest the presented Audience
Analysis units.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

The repetition within the programmed units was very
objectionable.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

 

Presenting the Audience Analysis concepts via programmed
instruction proved to be a very efficient use of my time.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

The programmed units tried to present too much material
for the average undergraduate student.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

I probably would not recommend the presented Audience
Analysis units to other T-R students.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

10.

11.

12.

13.

88

I was stimulated by the audience concepts presented to
the point that I would like to take other audience-related
courses and/or expand my knowledge by self reading.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

I find the subject of Broadcast Audience Studies very
interesting.

1) strongly agree

2) agree

3) neutral

4) disagree w
5) strongly disagree

I frequently found myself thinking about the material
presented, after reading the programmed instruction units.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

The programmed instruction units presented qualify as some
of the most ppabsorbing and otherwise dull experiences
I've had in college.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

Overall, taking everything into consideration, I would
rate the content organization of the presented audience
units as excellent.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

14.

15.

16.

89

The redundant nature of the programmed units probably
served as a learning aid for me.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

If the presented Audience Analysis concepts were to be
expanded into an elective ”course,” I would like to take
the course.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

I frequently found my mind wandering as I sat reading
the programmed instruction units.

1) strongly agree

2) agree

3) neutral

4) disagree

5) strongly disagree

 

 

APPENDIX C

APPENDIX C

COGNITIVE POST-TEST FOR AUDIENCE CONCEPTS

Cognitive Test for Audience Analysis Concepts

 

Name:

 

Student Number:

 

Sex:

 

Television viewing levels tend to change by season of
the year. When is viewing generally the highest?

1) Spring
2) Fall
3) Winter

Are most American families willing to COOperate in audience
surveys?

1) yes
2) no

All things considered, is there a better way than sampling
to gather audience information?

1) yes
2) no

Ratings reports speculate as to what people will probably
be watching in the near future.

1) true
2) false

During an average evening in an average American household,
about how much of the night will be devoted to television?

1) 70%
2) 50%
3) 30%

90

 

 

"“; t. u _'_‘.

10.

11.

12.

91

All audience research firms operating in the U.S. must be
accredited by the Broadcast Rating Council.

1) true
2) false

Any special, non-regular station promotion during a
survey period is called:

1) insuring the sampled broadcast audience
2) adversarial ratings methodology
3) hypoing the broadcast ratings

About how many American homes own at least one television
set?

1) 96%
2) 87%
3) 78%

A national television rating for a particular segment
of "Bonanza" is 25. What does that mean?

1) an audit by the Broadcast Rating Council will result
in an independent variable of 25 Dollars per second of
advertising

2) of the households in the U.S. owning television sets,
about 25% were watching the program

3) the United National Church Services have given the
program their wholesomeness evaluation number of 25

Ratings alone should probably decide which station's
commercial time is to be purchased.

1) true
2) false

About how many households are sampled nationally by the
A. C. Nielsen Co. to determine network television
viewing?

1) 1,000 homes
2) 1,000,000 homes
3) 100,000 homes

You are a station manager and your local competition is
beating you in the latest audience report by one full
rating point. Should you be unusually worried?

1) yes
2) no

 

13.

14.

15.

16.

17.

18.

19.

92

On the average, which household watches more television:
a black 8 white or color set home?

1) black 6 white
2) color

National ratings cannot be directly applied to local areas.

1) true
2) false

When two ratings services do not report the same figures
for identical time periods, what is probably the difficulty?

1) sampling theory
2) sampling initiation
3) sampling practice

The most important reason for measuring the broadcast
audience is:

1) it is required by the NAB Board of Station Practices
for representation

2) broadcast advertisers will want to use audience data
for selling purposes

3) local citizen pressure groups can now demand the
quantitative information

You are explaining the accuracy of ratings to a friend.
What is the most important point to stress?

1) ratings are program evaluations by leading TV columnists
2) ratings are statistical estimates with ranges of error
3) ratings are precise numbers that may be interpreted

as such

During a survey period a station is caught conducting
special, non-regular promotion. What happens?

1) the station receives a fine from the Broadcast Rating

Council

2) the act is reported in the ratings publication for that
period

3) the promotion department of the offending station is
indicted

The "plus" or "minus" interpretation of a given audience
estimate will depend most directly upon:

1) the quality of research personnel involved
2) the general climate for broadcast evaluation
3) the reported standard deviation

 

20.

21.

22.

23.

24.

25.

26.

27.

93
On the average, who watches more television per week:
a typical male or female?

1) male
2) female

American households average how many hours of television
usage per day.

1) 11 hours
2) 9 hours
3) 7 hours

The major source of error in political polling stems from:

1) predicting future behavior
2) describing past behavior
3) acquiring unbiased respondents

What exactly is a "rating?”

1) an evaluation of a given program by a panel of BRC
broadcasters

2) the published number of available advertiser dollars

3) a percent of all homes owning a television set

One audience measurement method is by far the most popular.
Which is it?

1) the diary
2) the personal interview
3) the telephone coincidental

There is a distinct possibility of different program
preferences between those who participate in audience
surveys and those who do not.

1) true
2) false

Most reputable audience research firms adhere to the
Broadcast Rating Council's minimum reporting and methodology
standards.

1) true
2) false

One night of the week attracts the least amount of
television viewing. Which is it?

1) Wednesday
2) Friday
3) Monday

28.

29.

30.

94

A sample chosen for an audience study must reasonably
represent:

I) married couples owning one or more receiving sets in
working order .

2) that population which it was chosen to measure

3) homes purchasing receiving sets in the last survey
period.

Your rich Uncle is a broadcast rating. What word would
best characterize him?

1) a phalarope
2) a precise number
3) an estimate

Each week during a typical local-market survey period
the sampled households are changed. Why?

1) families keep accurate viewing records for a week

2) postal departments have encouraged the switching
techniques

3) field teams can run short of research funds in local
markets

 

 

APPENDIX H

APPENDIX H

ITEM ANALYSIS DATA FOR COGNITIVE POST-TEST

Number of subjects - 40 Number of test items - 30

RAW SCORE DISTRIBUTIONS

 

 

 

Raw Cumulative Percentile Standard
Score Frequengy Frequency Rank Score
30 6 6 92 61.5
29 10 16 72 57.7
28 5 21 53 53.9
27 3 24 43 50.1
26 6 30 32 46.2
25 2 32 22 42.4
24 3 35 16 38.6
23 Z 37 9 34.8
22 1 38 6 31.0
21 2 4O 2 27.2

Mean - 26.97

Standard Deviation - 2.62

Variance - 6.90

Standard Score has Mean of 50 and Standard Deviation of 10

(Perfect score = 30; Chance score = 12; Ideal discriminating
score = 21)

A

Mean Item Difficulty - 10. Difficulty is the percentage of
the total group marking a wrong answer.

Mean Item Discrimination - 21. Discrimination is the difference
between the percentage of the upper 27% group
marking the right answer and the percentage of the
lower 27% group marking the right answer.

Kuder-Richardson Reliability #20 — .6635. K-R Reliability
measures the internal consistency of the test
through analysis of the individual test items.

95

 

APPENDIX I

APPENDIX I

CROSS TABULATION OF PAIRED BIPOLAR

ATTITUDINAL QUESTIONNAIRE ITEMS

The following cross tabulation tables can be

perceived as one might interpret a simple correlation matrix--

 

in this case, the more nearly subjects' numerical responses

tend to approach the drawn diagonal lines, the more "consistent"
their responses to bipolar attitudinal items can be said to be.
For example, subjects providing a response of "2" (agree) to

the positively-stated item #2 could be expected to provide a

 

response of "4" (disagree) to the negatively-stated item #16

 

which possesses essentially the same attitudinal content.

Attitudinal Item #2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\O
H
\
T \ 5 4 3 2 l
e
\.
31 1 3
H
r32 )21
.5
.U 3 l 1 l l
3
«441 4181
:3
<5 MK
Totals 2 6 7 20 5 (40 subjects)

96

97

Attitudinal Item #4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\
4 (40 subjects)

 

 

 

 

 

 

oo

* 5 4 3 2 l

E \. ‘

Q)

r: \c

E 11\2 l

Lg 124\1ppp

\.

.2 51161

p

4.)

< l 1 2\

Totals 2 9 17 8
Attitudinal Itemfi_

t,

r-i

* 54321

5\

p 1\

H

3;; 11>-52

£2

lg 4\31

fl \\

i3 65 \fi

<< \

\
Totals 2 22 13 3 0

 

 

 

 

 

 

(40 subjects

APPENDIX J

APPENDIX J

TIMES CONSUMED BY CONVENTIONAL LECTURE
AND PROGRAMMED INSTRUCTION SUBJECTS

Conventional Lecture Times

 

Time Spent Presenting Unit ONE: 25 minutes
" " " Unit TWO: 34 minutes

" " " Unit THREE: 35 minutes
Total: 94 minutes

Programmed Instruction Times (Minutes)

 

 

Unit Unit Unit Total

Subject # ONE TWO THREE ATime
l. 4 5.5 5.5 15
2. 7 S 5 l7
3. 7 8 7 22
4. 4.5 3.5 4 12
5. l3 9 13 35
6. S 5.5 7.5 18
7. 5 6 6 17
8. 4 S 5 l4
9. 4.5 S 4.5 14
10. 9 6 7 22
11. 12 10 10 32
12. 10 10 15 35
13. 8 10 7 25
14. S S 7 17
15. 8 9 6 23
16. 6 7 10 23

17. 6.5 5.5 5.5 17.5
18. 8 6 5 19
19. 9 ll 12 32
20. 6 7 7 20

Average Total Time for Group: 21.5

98

LIST OF REFERENCES

LIST OF REFERENCES

Anderson, R.; Kulhavy, R. W.; and Andre, T. "Feedback
Procedures in Programmed Instruction." Journal of
Educational Psychology, Vol. 62 (1971), pp. 148456.

 

 

Borg, W. R., and Call, M. D. Educational Research: An
Introduction. New York: David McKayFCo., I971.

 

 

Campbell, D. T., and Stanley, J. C. Experimental and Quasi-
Experimental Designs for Research. CHICago: ‘Rand—
McNally, 1966.

 

 

Crowder, N. A. "Automatic Tutoring by Means of Intrinsic
Progamming." Automatic Teaching. Edited by
E. Galanter. New York: John Wiley 6 Sons, 1959.

 

"Automatic Tutoring by Intrinsic Programming."
Teaching_Machines Programmed Learning: A Source
Book. Edited by A. A. Lumsdaine andiR. Glaser.
Washington, D. C.: National Education Association,
1960.

 

Dawes, R. M. Fundamentals of Attitude Measurement. New
York: John WiIey 8 Sens, I972.

 

Eurich, A. C. "Man and Media in Higher Education."
Educational Broadcasting Review, Vol. 2 (1968),
pp. 3-9.

 

Filep, R. "The Potential Impact of Programmed Instruction
on the Curriculum and Counseling." Prospectives in
Programming. Edited by J. B. Granger. New York:
The Macmillan Co., 1962.

 

 

Francois, W. B. "Designing, Developing and Testing
Programmed Instruction for Beginning News Writers."
Unpublished Ph.D. dissertation. Dissertation
Abstracts, Vol. 28 (1968), 5002a.

 

 

Gagne, R. M. "Educational Media and Individualized Instruc-
tion." Educational Broadcasting Review, Vol. 4
(1970), pp. 49162.

 

Griffith, J. L. "An Evaluation of Programmed Instruction in
Journalism." Journalism Quarterly, Vol. 46 (1969),
pp. 613-17. TI

 

99

100

Hough, J. B. "Research Vindication for Teaching Machines."
Phi Delta Kappan, Vol. 42 (1962b), pp. 240-42.

 

Katz, D., and Stotland, E. "A Preliminary Statement to a
Theory of Attitudinal Structure and Change."
Psychology: A Study of a Science. Edited by S. Koch.
New York: McGraw Hill, 1959.

 

Macdonald-Ross, M. "Programmed Learning: A Decade of Deve-
lopment." International Journal of Man-Machine
Studies, Vol. 1 (1969), pp. 73-100.

Meeske, M. D. "Teaching Radio-Television in a Department of
Communication." Educational Broadcasting Review,
Vol. 6 (1972), pp. 219-223.

Nauman, T. F. "A Laboratory Experience in Programmed
Learning ikn' Students in Educational Psychology."
Journal of Programmed Instruction, Vol. 1 (1962),
pp.’9-181

 

Osgood, C. E.; Suci, G. J.; and Tannenbaum, P. H. The
Measurement of Meaning. Urbana: University of
IIlinois Press, 1957.

 

O'Toole, J. F. "Teachers' and Principals' Attitudes Towards
Programmed Instruction in the Secondary School."
AV Communication Review, Vol. 12 (1964), pp. 429-433.

Prejean, B. G. "Programmed Instruction in Journalism:
An Experimental Study." Unpublished Ph.D. dissertation.
Dissertation Abstracts, Vol. 29 (1969), 3569a.

 

Pressey, S. L. "A Simple Apparatus Which Gives Tests and
Scores-~and Teaches." School Sociology, Vol. 23 (1926),
pp. 373-6.

 

. "A Machine for Automatic Teaching of Drill Material."
School Sociology, Vol. 25 (1927), pp. 549-52.

 

"A Third and Fourth Contribution Towards the
Coming 'lndustrial Revolution' in Education."
School Sociology, Vol. 36 (1932), pp. 668-72.

 

"A Puncture of the Huge 'Programming Boom'?" Edited
by W. Scramm. The Research on Programmed Instruction,
Vol. 35 (1964), p. 87.

 

 

Roe, A. "A Comparison of Branching Methods for Programmed
Learning." Journal of Educational_Research, Vol. 55
(1962), pp. 407-16?

 

101

Scherman, A. "Free-choice, Final Performance and Attitudes
Toward Different Types of Programmed Instruction."
Journal of Education, Vol. 155 (1973), pp. 56-63.

 

Shull, H. I. "Programmed Instruction: A Comparison of
Learning and Retention of Information Learned
Through the Use of Small Step (Linear) Programmed
Instruction and Large Step (Branching) Programmed
Instruction." Unpublished Ph.D. dissertation.
Dissertation Abstracts International, Vol. 30
(1970), p. 5266a.

 

Skinner, B. F. "The Science of Learning and the Art of
Teaching." Harvard Educational Review, Vol. 24
(1954), pp. 86-97.

 

. "Teaching Machines." Science, Vol. 128 (1958),
pp. 969-77.

Tobias, 8. "The Effect of Attitudes to Programmed Instruc-
tion and Other Media on Achievement from Programmed
Materials." AV Communication Review, Vol. 17
(1969), pp. 299-306.

 

Zimbardo, P., and Ebbesen, E. B. Influencing Attitudes

and Changing Behavior. Menlo Park: Addison-Wesley,
I969.

 

 

ml
A“.
RI!
B"
UT

"
V”
”All
S”
R“
E"
V”
N”
U"
T”

u

H

"I
I"
H"

3 1293 03056 5125