ZREMTIVE EEFEETIVENESS 0F DISCOWERY AND
EXPOSI'EQRY METHODS OF TEACHING CONCEPTS
'EHROUGH THE SINGLE- CONCEPT FILM

Thesis for the Degree of Ed. D.
MICHIGAN STATE UNKVERSITY
CASTELLE G. GENTRY
1 9 6 S

 

—: ‘ __.

Wests

nmungfxxnxngmgw «14%;?

 

This is to certify that the

thesis entitled
Relative Effectiveness of Discovery and Ex-

pository Methods of Teaching Concepts Through the
Single-Concept Film.

presented by

Castelle G. Gentry

has been accepted towards fulfillment
of the requirements for

__L‘-_D___ degree in __Educa t ion

Mm

Major professor

Date August 2L 1965

0-169

. pa-

f-._1.,

F. IIN‘DING av '
7 HOME 8: SUNS'
‘ 300K BINDERY W

LIBRARY BINDEF
l "sunny. meme

 

 

 

 

ABSTRACT

RELATIVE EFFECTIVENESS OF DISCOVERY AND
EXPOSITORY METHODS OF TEACHING CONCEPTS

THROUGH THE SINGLE—CONCEPT FILM
by Castelle G. Gentry

The abstract consists of three parts: summary,

conclusions, and recommendations.
Summary

The broad purpose of the study was to provide
guidelines for programing and use of single-concept films.
The specific purpose was to analyze the teaching effects
of the single—concept film, programed to teach science
concepts through two contrasted teaching modes: exposi-
tion and discovery.

The concept "Adaptation" was the subject of a

four minute film. One film version had the concept

Castelle G. Gentry

instances ordered from simple to complex, while the second
presented the concept instances in a random sequence. The
film was shown in four treatment modes: 1) ordered se-
quence of concept instances using discovery narration.

2) ordered sequence of concept instances using expository
narration, 3) random sequence of concept instances using
discovery narration, and 4) random sequence of concept
instances using expository narration.

The sample consisted of 280 junior high school
students selected from eighth grade general science classes.
They were tested on three independent variables: sequence
of concept instances, intelligence level, and teaching
method. And they responded to three criterion tasks: 1)
generation of new instances of the concept, 2) applica-
tion of the concept to problem situations, and 3) recog-
nition of new instances of the concept. These tasks were
given immediately after the treatment, and again three
Weeks later. i

The statistical hypotheses were:

1. There is no differential effect for the interaction

of intelligence level and teaching method, on

2

Castelle G. Gentry

immediate criterion performances, When science
concepts are taught through the single—concept

film.

2. Method does not have an effect on the retention
of science concepts, as taught by the single-
concept film, when measured by delayed criterion

tasks.

Two analytic procedures were employed: three,
three-way univariate factorial analyses of variance,

and product-moment correlations.

Conclusions

The conclusions are presented in two parts:
variable control and the hypotheses.

Variables which may require greater control or
modification are: 1) effect of practice, 2) number of
film showings per student, 3) relative shortness of the

film, 4) number of instances of the concept, 5) concept

3

Castelle G. Gentry

prerequisites, and 6) the novelty effect of the concept
instances.

No significant differences were found for the
hypotheses. However, a trend contrary to the first hy-
pothesis was noted. It indicated that bright students
learned best when taught by single-concept films pro-
gramed for the expository method, while less bright
learned best when the programing followed the discovery
method. This trend reoccurred during three regroupings
of the data. In one of the regroupings (i.e., subjects
separated by sex) significant differences were found
favoring the trend.

A finding not directly relevant to this research,
but of importance, was that the three criterion tasks

were independent measures of concept formation.

Recommendations

Replication of this study under greater controlled

conditions was recommended. Changes suggested included:

4

10.

Castelle G. Gentry

Students practice discovery method prior to

treatment administration.

Film viewed by each student more than once.
Vary running time of film.

Vary number of instances in film.

Use commonplace instances of the concept to re-

duce novelty effect.

Determine sub-concepts necessary for attaining

concepts.
Increase categorization of intelligence.
Differentially assign students by sex.
Test concepts from other subject areas.

Explore trend of bright students learning best
through expository programing of single-concept
films, contrasted with less bright learning best
through the discovery programing of single-concept

films.

Castelle G. Gentry

It was proposed that continued research in the
development of guidelines for the programing and use of
the single-concept film could be aided by these recom-

mendations.

RELATIVE EFFECTIVENESS OF DISCOVERY AND EXPOSITORY
METHODS OF TEACHING CONCEPTS THROUGH THE
SINGLE-CONCEPT FILM

3

BY

|
\t"

Castelle G. Gentry

A THESIS

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

DOCTOR OF EDUCATION

College of Education

1965

'tv,

ACKNOWLEDGEMENTS

Grateful acknowledgement is made to the following
people:

To Dr. Charles Blackman, my major professor, for
his advice, assistance, and patience.

To Dr. William Stellwagen for his valuable help
in the early development of this thesis.

To Dr. Charles Schuller for his timely advice and
assistance.

To Dr. David Krathwohl for his assistance in the
development of the design and the statistical procedures
of this study.

To Dr. Jean LePere for her assistance in the de-
velopment of the measurement instruments. -

To the students and faculty of Pattengill Junior
High School of Lansing for their invaluable cooperation.

To my wife Nancy, and our children Linda, Karen,
Melanie, and Jennifer, whose confidence and sacrifice

made this endeavor possible.

ii

TABLE OF

ACKNOWLEDGEMENTS. . . . .
LIST OF TABLES. . . . . .
LIST OF FIGURES . . . . .

LIST OF APPENDICES. . . .

Chapter

I. INTRODUCTION OF THE
TERMINOLOGY . . .

The Problem . .

PROBLEM

CONTENTS

Purpose of the study. .
Hypotheses in broad research form
Limitations of this study . . . .

Theory related to

Definition of Terms . .

Overview of the Study

II. REVIEW OF THE LITERATURE. .

Summary . . . .

III. THE EXPERIMENTAL DESIGN .

iii

RELEVANT

this study. . .

Page

ii

ix

xi

ONU'IU'Ib

l6

l7

19

56

58

Table of Contents--continued.

Chapter

The Population. . . .
The Sample. . . . . .

Instrumentation . . .

Experimental Design and Analysis. . .

Summary . . . . . . .

IV. ANALYSIS OF RESULTS . . .

Statement of Statistical Hypotheses .

Probability Statement
Analysis of Variance.
Interpretation. . . .

Summary . . . . . . .

V. SUMMARY, CONCLUSIONS, AND

APPENDIX.

Summary . . . . . . .

Conclusions . . . . .

Recommendations . . .

BIBLIOGRAPHY. . . . . . . . . .

iv

Page
58
58
59
64
73
76
76
77
77
82

116
121
121
127

130

132

182

LIST OF TABLES

Table Page

2.1 SUMMARY OF SWENSON'S DISCOVERY LEARNING EXPERI-
MENT . O C O O C O O O O O C O O O O C O O C 36

3.1 PRODUCT MOMENT CORRELATIONS BETWEEN JUDGES
SCORES FOR TASK I O O O O O O O C C O C C O O 64

3.2 PRODUCT MOMENT CORRELATIONS BETWEEN JUDGES
SCORES FOR TASK II . . . . . . . . . . . . . 64

4.1 ANALYSIS OF THE IMMEDIATE CRITERION PERFORMANCE
WITH TWO LEVELS OF INTELLIGENCE: GENERATION
OF NEW INSTANCES OF THE CONCEPT. . . . . . . 79

4.2 ANALYSIS OF THE IMMEDIATE CRITERION PERFORMANCE
WITH TWO LEVELS OF INTELLIGENCE: APPLICATION
OF THE CONCEPT TO PROBLEM SITUATIONS . . . . 80

4.3 ANALYSIS OF THE IMMEDIATE PERFORMANCE WITH TWO
LEVELS OF INTELLIGENCE: RECOGNITION OF NEW
INSTANCES OF THE CONCEPT . . . . . . . . . . 81

4.4 ANALYSIS OF THE DELAYED CRITERION PERFORMANCE
WITH TWO LEVELS OF INTELLIGENCE: GENERATION
OF NEW INSTANCES OF THE CONCEPT. . . . . . . 83

4.5 ANALYSIS OF THE DELAYED CRITERION PERFORMANCE
WITH TWO LEVELS OF INTELLIGENCE: APPLICATION
OF THE CONCEPT TO PROBLEM SITUATIONS . . . . 83

4.6 ANALYSIS OF THE DELAYED CRITERION PERFORMANCE
WITH TWO LEVELS OF INTELLIGENCE: RECOGNITION
OF NEW INSTANCES OF THE CONCEPT. . . . . . . 84

V

List of Tables--continued.
Table Page

4.7 ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION
PERFORMANCE WITH THREE LEVELS OF INTELLI-
GENCE: GENERATION OF NEW INSTANCES OF
THE CONCEPT. . . . . . . . . . . . . . . . . 86

4.8 ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION
PERFORMANCE WITH THREE LEVELS OF INTELLI-
GENCE: APPLICATION OF THE CONCEPT TO PROB-
LEM SITUATIONS . . . . . . . . . . . . . . . 86

4.9 ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION
PERFORMANCE WITH THREE LEVELS OF INTELLI-
GENCE: RECOGNITION OF NEW INSTANCES OF
THE CONCEPT. . . . . . . . . . . . . . . . . 87

4.10 ANALYSIS OF VARIANCE OF DELAYED CRITERION PER-
FORMANCE WITH THREE LEVELS OF INTELLIGENCE:
GENERATION OF NEW INSTANCES OF THE CONCEPT . 89

4.11 ANALYSIS OF VARIANCE OF DELAYED CRITERION PER-
FORMANCE WITH THREE LEVELS OF INTELLIGENCE:
APPLICATION OF THE CONCEPT TO PROBLEM SITU-

ATIONS . . . . . . . . . . . . . . . . . . . 90

4.12 ANALYSIS OF VARIANCE WITH THE DELAYED CRITERION
PERFORMANCE WITH THREE LEVELS OF INTELLIGENCE:
RECOGNITION OF NEW INSTANCES OF THE CONCEPT. 90

4.13 ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION
PERFORMANCE WITH MIDDLE INTELLIGENCE LEVEL
EXCLUDED: GENERATION OF NEW INSTANCES OF
THE CONCEPT. . . . . . . . . . . ..-. . . . 91

4.14 ANALYSIS OF'VARIANCE OF IMMEDIATE CRITERION
PERFORMANCE WITH.MIDDLE INTELLIGENCE LEVEL
IflKCIJZDEH): JKPFULICVMFICHV CH? TTHE CKMNCIHWP 1K)

FTNQBIJWW’EEITTUHPICHWS . . . . . . . . . . . . . 532

.115.AAD&LITSIE? CH7 VVUQIYUVCHE CH” IfiflWEIXLAflHE CHZITHMRIINN
PEERIWQRTLAARZE'FKIQVY AIIEUILEI ITRPEHILIIIEBHIE JJEVHEL

iii

List of Tables--continued.

Table

EXCLUDED: RECOGNITION OF NEW INSTANCES OF
HE CONCEPT.

4.16 ANALYSIS OF VARIANCE OF THE DELAYED CRITERION

PERFORMANCE WITH MIDDLE INTELLIGENCE LEVEL

EXCLUDED: GENERATION OF NEW INSTANCES OF
THE CONCEPT.

4.17 ANALYSIS OF VARIANCE OF THE DELAYED CRITERION
PERFORMANCE WITH MIDDLE INTELLIGENCE LEVEL

EXCLUDED: APPLICATION OF CONCEPT TO PROBLEM
SITUATIONS .

4.18 ANALYSIS OF VARIANCE OF THE DELAYED CRITERION
WITH MIDDLE INTELLIGENCE LEVEL EXCLUDED:
RECOGNITION OF NEW INSTANCES OF THE CONCEPT.

4.19 ANALYSIS OF VARIANCE FOR THE IMMEDIATE CRITER-
ION PERFOMRANCES FOR MALES:

GENERATION OF
NEW INSTANCES OF THE CONCEPT

4.20 ANALYSIS OF VARIANCE FOR THE IMMEDIATE CRITER-
ION PERFORMANCE FOR MALES: APPLICATION OF
THE’CKMWCEPT’HKD.PRCEHJNH SITTUVPIONS.

4.21 ANALYSIS OF VARIANCE FOR THE IMMEDIATE CRITER-

ION PERFORMANCES FOR MALES: RECOGNITION
OF NEW INSTANCES OF THE CONCEPT.

4.22 ANALYSIS OF VARIANCE FOR THE DELAYED CRITERION

PERFORMANCE FOR MALES: GENERATION OF NEW
IAHWFAAKHRS CH? TENS CIHWZEPT? . .

4.23 ANALYSIS OF VARIANCE FOR THE DELAYED CRITERION

PERFORMANCE FOR.MALES: APPLICATION OF THE
(CCMWZETHP 1Y3 PTNQBIJMW SEFTLUNPIIMNS.

‘24 .AADTLIQSIE? CH? ILAEKLADWCEI.FIH2 ITIE IlEIJXYEH) CEKITTHRICflfi
.PEHZEYZRALAARZE’JFCUQ fiflAIdES:

FUECIXRNSIPICHS CH? BEEN
.ZNERRALWZEM3 CU? TYLE'thmWCEJNP

trini

Page

. 92

93

93

. 94

. 1130

. 1130

..-

List of Tables-~continued.

Table Page
4.25 PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS:
BETWEEN IMMEDIATE CRITERION TASKS I. II. AND
III ACROSS ENTIRE SAMPLE . . . . . . . . . . . 105

4.26 ROTATED FACTOR LOADINGS FOR ORTHOGONAL FACTOR
ANALYSIS OF THE MAJOR CORRELATION MATRIX . . . 106
4.27 MINOR PEARSON PRODUCT-MOMENT CORRELATION MATRIX
OF IMMEDIATE CRITERION TASKS I.

II. AND III
FOR LOW INTELLIGENCE

. . . . . . 108

4.28 MINOR PEARSON PRODUCT-MOMENT COREELATION MATRIX
OF IMMEDIATE CRITERION TASKS I.
FOR MIDDLE INTELLIGENCE.

II. AND III
. . . . . . . 108

4.29 MINOR PEARSON PRODUCT-MOMENT CORRELATION MATRIX
OF IMMEDIATE CRITERION TASKS I.
FOR HIGH INTELLIGENCE.

II. AND III
. . . . . . . . 108

4.30 ROTATED FACTOR LOADINGS FOR ORTHOGONAL FACTOR

ANALYSIS OF THE MAJOR CORRELATION MATRIX OF
LOW'IC213flflfl1LATE POST-TASKS.

. . . . . . . . . 109
4.31 ROTATED FACTOR LOADINGS FOR ORTHOGONAL FACTOR
ANALYSIS OF THE MAJOR CORRELATION MATRIX OF

MIDDLE IQ ON IMMEDIATE POST-TASKS. . . . . . . 109
4.32 .ROTATED FACTOR LOADINGS FOR ORTHOGONAL FACTOR
ANALYSIS OF THE MAJOR CORRELATION MATRIX OF

.HIGH IQ ON IMMEDIATE POST-TASKS. . - . . . . . 110

viii

AH-

LIST OF FIGURES

Figure Page

4.1 A plot of immediate Post-task I means for the
intelligence-method interaction . . . . . 81

4.2 A plot of immediate Post-task II means for the
intelligence-method interaction . . . . . . 81

4.3 A plot of immediate Post—task III means for
the intelligence-method interaction . . . 81
4.4

A plot of the means for delayed Post—tasks I.
II, and III of the method variable. . . . . 81
A plot of immediate Post-task I means of the
intelligence—method interaction .

4.5
. . . . . 88

A plot of immediate Post—task II means of the
intelligence-method interaction .

4.6

. . . . . 88
4.7 A plot of immediate Post-task III means of the
intelligence—method interaction . . . . . . 89
4.8 A plot of immediate Post-task I means of the
intelligence-method interaction . . . . . . 96
4'9

A plot of immediate Post-task II means of the
intelligence-method interaction

0 O O

. . . 96
-10 A plot of immediate Post-task III means of the
intelligence-method interaction

. . . 96
l A plot of delayed Post—task I means of the in-

telligence—method interaction . . .

. . . . 101.
i}(

 

 

 

.us

 

List of

Figure
4.12 A
4.13 A

Figures-—continued.

Page
plot of delayed Post-task II means of the
intelligence-method interaction . . . . . . 101
plot of delayed Post-task III means of the
intelligence-method interaction . . . . . . lOl

LIST OF APPENDICES

Appendix Page
A. PRETESTS USED IN A PILOT STUDY TO DETERMINE
THE STUDENTS' KNOWLEDGE OF THE CONCEPT
"ADAPTATION" o o o o o o o o o o o o o o o o 0 l3 3
B. SCRIPTS FOR THE DISCOVERY AND EXPOSITORY
142

NARRATION OF FILM . . . . . . . . . . . . . .

C. THE TEST MANUALS WITH TESTS. AND DIRECTIONS
FOR SCORING TESTS . . . . . . . . . . . . . . 151

D. MEANS AND STANDARD DEVIATIONS OF FACTOR LEVELS
FOR TEACHING METHOD. AND FOR THE INTELLIGENCE-

METHOD INTERACTION. . . . . . . . . . . . . . 174

xi

CHAPTER I

INTRODUCTION OF THE PROBLEM

AND RELEVANT TERMINOLOGY

The focus of this study is directed toward an
instructional medium called the single-concept film.
The showing time of this film is usually of short du-
ration; approximately two to five minutes. As its
title suggests, it is designed to teach one concept.
although it is understood that subconcepts may also
be taught in the process. At present. the single-
concept film is identified with the cartridge-loaded
projectors. That is. the film is packaged in a perma-
nent plastic container which is slid into the projec-
tor without the bother of threading; a boon in itself,
most teachers would agree. The present and predicted
use of this medium is for small groups and for indi-
vidual use. The Sponsors of the medium are convinced
that its relatively low price will allow its extensive
adoption throughout the educational field.

1

The single concept film is not new. Bray,
in 1926. produced some six hundred of them. Percy
Smith.2 in England. produced a number during the thir-
ties. as did the Commission on Human Relations of the
Progressive Education Association.3 All of these at-
tempts were short—lived. The present movement support-
ing the single-concept film is making its entrance on
the coattails of a technological innovation. It comes
as a rider on the extraordinary utility of the new
eight millimeter cartridge—loading projectors. An
unfortunate effect is that the technological aspects
are so striking that little attention has been given
to the principles underlying the development and use
of the "single-concept film." Its creation and use

have been largely the product of the combined intuitions

 

1Robert W. Wagner, "The Educational Film in
Transition." Audiovisual Instruction. IX. No. 174
(March. 1964).

2Ibid.. p. 172.

3Charles F. Hoban. Jr., and E. B. van Ormer.
Instructional Film Research 1918—1950 (Rapid Mass
Learning). A Report on the Pennsylvania State College
Report Jointly Sponsored by the Department of the Army
and the Department of the Navy (Port Washington, N. Y.:
U. S. Navy Special Devices Center. 1951).

of the film maker and the subject-matter expert. And
it is suspected that their intuitions, particularly
the film maker's, have been structured by certain mun—
dane expectations:

One factor that tends to work towards the
formalization of the nontheatrical motion
picture is that the production of such films
has become big business. More films are
being made than ever before. and more money
is being spent on mass production and dis-
tribution. The producer Who finds himself
in a highly competitive market with the
knowledge that he must sell at least 500
prints to recover his production costs and
show a profit. realistically tries to design
his product for the widest possible audience
within the limits of the stated objectives
of the film. He may envision its use at
several grade levels in the public school.
speculate on its adaptability with adult
groups, and have in mind the possibility
that it might be used on television as well.
He is tempted to select currently popular
themes. to limit experimentation with new
and untried formats. and to emulate or re-
work ideas and treatments found successful
in the past.4

 

4Robert W. Wagner. "The Formula Film," A. V.
Communication Review. III. No. l (1955). p. 54.

The Problem

Little or no effort has been expended toward
isolating the unique teaching properties of the single-
concept film. nor to deveIOp guidelines for its pro—
graming. Further. there has been no concerted effort
to deve10p guidelines for the use of single-concept
films.

As an important beginning, this study will
examine the single-concept film in two contrasted
teaching modes. It is hOped that such research will
have a number of effects: one. it will provide film
makers with guidelines for programing their single-
concept films; two. it will further encourage the use
of a valuable educative tool; and three. observations
from this study might point the direction for further

research in programing and use of the single—concept

film.

Purpose of the study. This study is designed
to analyze the teaching effects of the single—concept

film when programed to teach a science concept through

the two contrasted teaching modes of expository and

discovery.

Hypotheses in broad research form. In teach-
ing science concepts via the medium of the single—concept
film. a differential relationship exists between intelli—
gence level and teaching method. Further. there is a
relationship between the teaching method employed and

the amount of material retained.

Limitations of this study.
1. It will deal only with one concept in one subject

area. and will therefore not be generalizable to

other subject areas.

2. Since the teacher is completely removed from the

study. it will make no statement as to the effect

of teacher interaction.
3- It deals only with one age group in one school.
4- The study will utilize the teaching method called

5 .
These hypotheses are restated in a testable
form in Chapter III.

”Guided Discovery." It will not concern itself
with any of the other varieties of the Discovery
Method. except in the historical development of

the rationale for guided discovery.

5. Within the proposed limits. the study will allow
conclusions concerning the effect of teaching method.
sequence of concept instances. and intelligence

level. on learning of a concept through the medium

of the single-concept film.

Theory related to this study. Three bodies
of theory undergird this study; the theory of concept
formation. the rationale for the discovery method of
teaching. and the use of film as a teaching medium.
By concept formation. is meant the process of discov-
ering and identifying detail. relation, or principle
through which a series or collection of instances can
be classified. The general consensus of learning the—
orists is that concept formation plays a major role

in learning.6 In fact. some psychologists contend

6

D. W. Johnson, The Psycholoqy of Thought and
EEQQEQQE (New York: Harper and Brothers. 1955).

 

kit

«NU

that learning and concept formation are synonymous.
The two most clearly stated theories of con—
cept formation depend upon the distinction between
abstraction and generalization.8 In the composite
Photograph Theory, a model first described by Hull,9
the primary emphasis is on abstraction. Here the fea-
tures common to a class of objects or events summate
their impressions on the observer. who thus gradually
acquires a picture in which the common features stand
out strongly while the variable characteristics are
washed out. The Active Search Theory emphasizes gen-
eralization. Here the concept originates as a hypo-
thesis, after which the conceptualizer tests the con—

cept by trying the hypothesis on fresh members of the

class.

7David P. Ausubel. "Some Psychological Aspects
of the Structure of Knowledge." Education and the Struc-
ture of Knowledge. ed. Stanley Elam (Chicago: Rand Mc-

Nally and Company. 1964). p. 230.

8W. Edward Vinacke. "The Investigation of Con-
cept Formation," Psychological Bulletin. XLVIII (Janu—

ary. 1951), p. 6.

9 . .

C. L. Hull. "Quantitative Aspects of the Evo-
lution of Concepts." Psychological Monograph, XXVIII
(1920). p. 4.

At present there is strong support for the
combination of these two theories, since experiments
show that abstraction and generalization are often

difficult to distinguish in the actual behavior of

the organism:

Under some conditions. the individual may,

in fact. be essentially a passive recipient
of sensory impressions which gradually sum-
mate into the concept. Under other condi-
tions it may be that an individual proceeds
by establishing an hypothesis and then de-
liberately checking it against the instances.
More than likely, they are mutually comple-
mentary approaches to a situation and occur
as interrelated processes.

The classical paradigm of concept formation
developed by Hullll under the influence of Herbart
consisted of three steps which. with variations. still
forms the essential paradigm for research on concept
learning:

1. The subject is presented with a number of
experiences. either simultaneously or in
succession. each of which contains a certain

characteristic common to others.

 

10Vinacke. op. cit.. p. 7.

llHull, op. cit., p. 4.

2. The subject is brought to compare deliberately
the various situations presented with a view

to discovering similarities and differences

among them.

3. Lastly. the significant element common to all
the experiences is deliberately sought out,

found. and formulated in language.

With only slight variations this study has
followed Hull's classical Model for concept formation.

As stated in the hypotheses. this study is
primarily concerned with the relative effectiveness
of teaching methods as applied to the single—concept
film. The two methods under study are discovery and
expository.

The expository method presents (to the student)
a concept with a detailed explanation of its applica-
tion and the identification of its instances. This
method stands deep in pedagogical tradition. but per-
haps some comment is appropriate for the purposes of
this study. Possible bad effects of the expository

method are propounded by Bruner:

10

Insofar as possible. a method of instruction
should have the objective of leading the child
to discover for himself. Telling children and
then testing them on what they have been told
inevitably has the effect of producing bench—
bound learners whose motivation for learning

is likely to be extrinsic to the task at hand——
pleasing the teacher. getting into college.
artificially maintaining self esteem.

Othersl3 feel that this view of expository
teaching is exaggerated. They freely admit that the
above description fits many learning situations in
our schools and colleges. but they question that it
is inherent in the expository method itself. They
maintain that it can be supplemented but not replaced.
"because it is the only feasible and efficient method
of transmitting large bodies of information."14 Later
in an extended discussion of Guided Discovery it will
be noted that there is a measure of agreement between

the proponents of the expository method and the position

12J. S. Bruner. "After Dewey What." Saturday
Re‘vieiw (June 17, 1961). p. 77.

13David P. Ausubel. The Psychology of Meaning-

iEl_§§é£fllflg (New York: Grune and Stratton, 1963).
P- 160.

1-

)

U)

f).

11

held in this study. The question becomes one of which
method is supplementing the other.

The discovery method needs to be defined. be-
cause it has several variations. and the claims for
each do not necessarily overlap. The definition of
discovery adopted for this study is: the teaching
method in which the student is presented instances
of objects or events through which run a common re-
lationship or a common element, and is asked to "dis-
cover" the common relationship or element. This var-
iation of the discovery method is more structured than
some of the "pure" types, which maintain that the stu-
dent should be responsible for such things as: sequenc-
ing. rate of intake of information. and the accumulation
Of the data necessary to arrive at the solution to a
particular problem.

The approach of the present study tends to

agree with Suchman:

 

15M. C. Wittrock. The Learning by Discovery

HYBQEEEEIfi- A Review sponsored by Stanford University
With support from the U. S. Office of Education.

12

The teacher must see to it that the child's
efforts at inquiry are rewarded by success,
that the child is able to obtain the infor-
mation that he needs and that he does dis-
cover new concepts on his own. The teacher
can help the child by posing problems that
are reasonably structured and will lead to
exciting new discoveries. The teacher can
also coach him in the techniques of data
collection and organization that will lend
power and control to his searching.

Claims supporting this method include: "prac-
tice in discovering for oneself teaches one to acquire
information in a way that makes the information more

. . . . "l7 . .
readily Viable in problem solVing. learning by dis—
covery produces knowledge which transfers to new situ-

. 18 . . . .
ations. learning by discovery increases retention

of knowledge.19 and learning by discovery serves to

 

16Ibid.. p. 4.

17J. S. Bruner. "The Act of Discovery.“ Harvard
Educational Review. XXXI (1961), p. 27.

186. L. Anderson, "Quantitative Thinking as De-
veloped under Connectionist and Field Theories of Learn-
ing." Learning Theory in School Situations (Minneapolis:
University of Minnesota Press. 1949). pp. 40-73.

19G. N. Haslerud and Shirley Meyers. "The Trans-
fer Value of Given and Individually Derived Principles."
figurnal of Educational Psycholqu. XLIX (1958). pp. 293-
298.

p-1

w ‘v

(T‘

‘ r

.0.

 

"v
V.

I).

b.

O .

in

[(7

13

2
motivate the student. 0

This research will be concerned with two of the
above claims: 1) learning by discovery produces know—
ledge which transfers to new situations, and 2) learn-
ing by discovery increases retention of knowledge.

The relationship and implications that exposi—
tory and discovery methods draw from the combined the—
ories of concept formation need to be identified. The
desired event is for a student to attain a working
understanding of a concept. That is to say, to attain
a concept allowing him to go beyond a set of observed
critical properties. exhibited by an object or event,
to the class identity of the object or event in ques-
tion. The science concept tested in this study is
"Adaptation,” or the idea that organisms change in
order to better c0pe with their environment. The tea—
cher's aim is for his students to be able to identify

the instances of this concept, to be able to generate

208. Y. Kersh. "The Motivating Effect of Learn-

ing by Directed Discovery." Journal of Educational Psy-
shslssx. L111 (1962), pp. 65-71.

an
I Q

14

new instances from their experiences gained prior to
attaining the concept, and to apply the concept to
problem situations. By the previous definitions of
concept formation, the student may accomplish this by
either generalization or abstraction, or through some
combination of these two processes. By generalization,
then. they may be led to "discover" the hypothesis
used as a combining model for the various instances

of the concept. or they may be given the hypothesis
and Shown how the instances of the concept concur with
the model. The combination of abstraction and gener-
alization lends itself to both of the two teaching
methods.

How does the use of film for teaching a concept
fit into the general scheme of concept formation theory?
Considering the two parts of the combined theory, it
can be assumed that presenting a number of instances
that are exemplars of the concept give the learner his
best opportunity to recognize the common elements or
common relationships that determine the concept. To

quote Gagne's design for concept learning:

U)

15

Present a suitable variety of stimuli to rep-

resent the class. each stimulus having a con—

nection with a common response. Verify by

presenting a novel stimulus that is also a

member of the class.21

The abstracted common elements or relationships

can be transformed by the student into a hypothesis or
generalization. that is “an instant having these char—
acteristics can be expected to behave in a certain way."
The student can then test his hypothesis through the
simulated instances provided by the film. As was
stated previously, exactly how generalization and ab-
straction are combined by the student in arriving at
the desired concept is not clear. Nonetheless, it
seems reasonable that the conditions necessary for

generalization and abstraction can be established

through the film medium.

 

1Robert M. Gagne, The Conditions of Learning
(New York: Holt. Rinehard. and Winston. Inc.. 1965),
p. 254.

2.

16

Definition of Terms

Criterion Performances consist of three tasks:

a) the generation of new instances of the concept
taught. b) the application of the concept taught
to an appropriate problem situation, and c) the
identification of instances of the concept taught,
among a collection of instances from several con-

cepts.

Concept: an abstraction consisting of some ele-
ments or relationships common to particular events
and/or things. Possession of a concept is inferred

from behavior (success on the criterion performances).

Discovery Method: a teaching method whereby a stu-

 

dent is presented instances of objects or events,
through which run common relationships or common

elements and is asked to discover the common rela-

»tionships or elements.

Expositorprethod: a teaching method whereby a

student is presented a concept with a detailed

l7

explanation of its application and the identifica-

tion of its instances.

5. Single-concept Film: an instructional motion pic-
ture of short duration that is designed to teach
one concept. although supporting concepts may also

be taught as part of the process.

Overview of the Study

The general plan of this study is as follows:
Chapter Two will consist of a review of the literature.
and its implications for this study. Studies that will
be reviewed in depth are: teaching methods (emphasis
on discovery), use of film in teaching concepts. and
those studies relating teaching method to intelligence.
The related areas section of the review of the litera-
ture consist of: concept formation studies, transfer
of training studies, and retention of learning.

Chapter Three lays out the design of the study.
Here data are provided on the sample and the instrumen-

tation. the statistical hypotheses are stated. the plan

U)

r—v.

18

of the study or the experimental design is presented,
and the statistical model used to test the hypotheses.
with an explanation for its selection included.

The fourth chapter consists of an analysis of
the results. Here the hypotheses are restated, and
the relevant data for each of the hypotheses is pre-
sented. Also, with each hypotheses the probability
statement and the statement of rejection or failure
to reject are attached. After this presentation of
fact. the chapter concludes with an interpretation
of the results.

The last chapter will cover the major summary,

the conclusions, and implications of the study.

CHAPTER II
REVIEW OF THE LITERATURE

In this chapter. the related literature is
surveyed. and its implications for this study are dis-
cussed. The areas reviewed include: 1) the use of
the film medium in teaching, 2) teaching methods with
emphasis on the contrast between the expository and
directed discovery methods, 3) the relationship of
teaching method to intelligence, 4) concept formation,
5) transfer of training, and 6) retention of learning.
The major part of this review will be concentrated in
the first three areas.

Researchers have accumulated masses of data
supporting the value of film in the learning process.
The most comprehensive compilation of this research

evidence is found in a government technical report

19

20

prepared by Hoban and van Ormer.l Since 1950 the amount
of research being done on educational films has increased
steadily. In nearly all of the educational films, expo-
sition is the method employed for presenting the film
content. Rarely is this method contrasted with another.
There are innumerable examples of film—teaching con—
trasted with other methods such as film versus lecture,
film versus filmstrip, film versus tape recording, and
so on. In some of the research, two versions of the
same film were contrasted but. on those occasions when
separate versions of the film were compared, they dealt
with such factors as sound versus silent, inserted ques-
tions versus no questions. one showing versus multiple
showings, color versus black and white, amount of narra-
tion, audio readability. density of information, and
rate of presentation, to name a few. While these fac—

tors are important, they do not deal with the focus

1Charles F. Hoban, Jr., and E. B. van Ormer.
Instructional Film Research 1918-1950 (Rapid Mass Learn-
lflfl). A Report on the Pennsylvania State College Project
jOintly sponsored by the Department of Army and the De-
Partment of the Navy (Port Washington, N. Y.: U. S.
Navy Special Devices Center. 1951).

 

.
A k

u!

21

of this study, which is the comparison of teaching
methods in film.

A second characteristic of this study makes
it especially difficult to find similar studies. The
teacher variable has been deliberately removed from
this study. The compendium of Hoban and van Ormer.
and succeeding research have, in the main, included
the teacher in their studies. In fact they have shown
clearly that for optimum use of a film the teacher is
essential. But the teacher variable is extremely dif-
ficult to control for. and even more difficult to gen-
eralize from.3 In order to get an accurate picture of
the effect of the medium. the teacher has been removed
from the study, while fully realizing that the most
successful use of any educational film depends upon
the application of the principles of film usage devel—

oped through research.

2Ibid.

3Donald M. Medley, and Harold E. Mitzel, "Meas-
uring Classroom Behavior by Systematic Observation.“
Handbook of Research on Teaching. ed. N. L. Gage (Chi-
cago: Rand McNally & Co., 1963), II, Part II. pp. 247-
48.

22

Of the work reviewed. open-ended films, like
those of Suchman.4 come closest to the instructional
programing notion. But he. like many others. has tied
his research so tightly to the teacher that it is very
difficult to separate out the effect of the film. And.
he does not compare his discovery films with comparable
versions using other teaching methods.

Brenner, Walter. and Kurtz5 did a study some—
what analogous to this study. They developed six ver-
sions of two different films: The Care and Use of Hand
Epglg (a training film), and Snakes (an information
film). In the two versions relevant to this research.

one had questions inserted, making it somewhat similar

to the discovery method, and statements were inserted

 

4J. Richard Suchman, The Elementary School Train-
ipqurogram in Scientific Inquiry. A Project Sponsored by

U. S. Department of Health. Education and Welfare. Office
of Education. and by the Research Board of the University
of Illinois, Title VII Project, No. 216: National Defense

Education Act of 1958, June 1962.

5H. R. Brenner. J. S. Walter. and A. K. Kurtz.
"The Effects of Inserted Questions and Statements on
Film Learning," Progress Repprt No.glg, State College
(Pennsylvania: Pennsylvania State College. Instruc-
tional Film Research Program).

I I
bu!

F.

23

in the same spot in the second version in an expository
manner. The results were not conclusive. The effects
of inserting statements and questions in the two films
varied. They also discovered that when the original
film without insertions was shown twice, it proved to
be about as effective as the experimental film. In
contrast with this study. their criterion specified
rote memorization of facts in the film, whereas in

this study students are asked to form a concept.

Some other studies have dealt with "dramati—
zation versus exposition” of the content.6 A few of
them require students to draw conclusions from the dra-
matic presentations in a manner comparable to the dis—
covery method. But the majority of them use an exposi-
tory approach in both versions.

Along the same line, Hovland, Lumsdaine, and
Sheffield7 contrasted two versions of a documentary

film dealing with the difficult jOb in Japan after

6Hoban, and van Ormer. pp;_gip.. pp. 15-16
(flap. 8) o

7c. I. Hovland, A. A. Lumsdaine, and F. D. Shef-
field. Experimenpg on Mass Communication (Princeton. N.J.:
Princeton University Press. 1949), pp. 130—141.

(I)

I)

24

V-E Day. The events in one of the versions were inter-
preted by a commentator, while the events in the second
version were presented by professional actors. with mu-
sical accompaniment. The dramatic presentation had a
consistent but slight and unreliable advantage in meas-
ures of audience interest and acceptance of authenticity.
In contrast with this study. their research is concerned
with attitudinal change rather than the attainment of
concepts.

In his similation experiments. Kersh8 uses film
in an open-ended manner as did Suchman. And as with
Suchman. the effectiveness of his method is tied closely
to teacher interaction. As was pointed out. the curtail-
ment of the teacher variable severely limits the number
of relevant film studies. Because of this limitation.
literature on the discovery method is primarily studied

outside of the film context.

 

Bert Y. Kersh. Classroom Simulation: A New
Dimension in Teacher Education, The Final Report, Title
VII. Project No. 886: National Defense Education Act of
1958. June 1963, pp. 38-44.

25

This paper assumes the reader is familiar with
the expository method and its various forms. All var-
ieties are similar in that the learner plays a passive
role While the instructor gives complete explanations
of the desired information. The learner is periodically
tested as to his assimilation of the instructor's message.
In contrast with the expository method, the discovery
method. while dating back to the work of Judd in 1908,9
has not been used extensively. An educator who very
early determined some of the working principles for
this method was Maria Montessori, who believed "it is
necessary for the pupil to perfect himself through his
own efforts.”10

As Ausubel has pointed out. the progressive

education movement also had some responsibility for the

development of the discovery method of teaching:

9C. H. Judd. "The Relation of Special Training
to Special Intelligence.” Education Review. XXVI (1908).
pp. 28—42 0

0Maria Montessori. The Montessori Method (London:
William Heineman, 1912). pp- 171‘172-

26

The Progressive Education movement obvi-
ously furnished several major strands in the
design of the discovery method. One aspect
of this movement was a growing dissatisfaction
with the empty formalism of much educational
content in the latter part of the nineteenth
century and the early part of the twentieth
century: with stultifying drill and catechism-
like methods of teaching: with the curriculum's
lack of relatedness to the everyday experience
of the child, his physical world, and social
environment; and with pupils' rote veraliza-
tion and memorization of ideas for which they
had no adequate reference in experience. Over—
statement of the realities underlying this dis-
satisfaction constituted the basis of the later
mystique that gll_verbal learning is little more
than glib verbalism and parrot-like recitation.
This led, in turn, to the exaggerated emphasis
that progressivists placed on relating the cur-
riculum to the physical and social environment
of the child: on direct, immediate. and con-
crete experience as a prerequisite for meaning-
ful understanding; on active learning and inquiry;
and on incidental learning in natural, uncontrived
situations.11

A major controversy among advocates of discovery
methods has to do with theories of its true value. On
one hand there is support for it in its ability to teach
information. while the other side insists that its value

lies in practicing the method itself. The three major

 

11D. P. Ausubel. The Psychology of Meaningful

Verbal Learning (New York: Grune & Stratton, 1963),
p. 139.

27

claims supported by the former are: the discovery
method motivates the students to learn, learning by
discovery assures greater transfer than does the ex-
pository method. and last. the relatively unstructured
discovery method allows a student to fit information
into his unique cognitive structure in the manner most
natural to him, and according to his particular rate
of reception.12 The proponents of the second view in-
sist that information derived through the practice of
the discovery method is secondary, that it is only
through the practice of this method that the student
truly prepares himself for a world that pays dividends
to those who are able to solve their problems rather
than having the answers given to them.13 Wittrock
summarizes these two positions thus:

.-. . learning by discovery is both an end

and a means for some educational theorists.

By discovery a student is supposed to learn
regularities and concepts within a discipline.
But more importantly he is supposed to learn
how to solve problems. to go beyond the data.

12J. S. Bruner. "The Act of Discovery." Harvard
EQEQational Review. XXXI (1961). p. 32.

13Ibid., p. 27.

 

'C,

28

to behave as a junior scientist. He is sup-
posed to become motivated and enthusiastic
about the discipline. He is to know personal
satisfaction because he has selected his own
sequence of problems and. through active re-
Sponses of his own. has succeeded at these
problems.

A second division among proponents of discov-
ery methods is between those who support "independent
discovery" where the learner is almost entirely on his
own and where the teacher serves only as a resource
person, and those who support "guided discovery" in
which the student is led to the ”discovery." A state-
ment by Craig helps to clarify these views:

Many have advocated relatively independent
problem solving in the belief that learning
situations should be similar to anticipated
transfer situations. This point of view

rests on the assumption that future discovery
of principles will be through independent
prOblem-solving, hence, more like pupil self—
discovery than directed discovery. A differ-
ent view is that problem-solving and discovery
are never independent except in the sense that
no one is physically present to prompt the
learner. Principles previously learned in

an area serve to direct discovery. Out—of

14M. C. Wittrock, The Learning by Discovery Hy-
BQEhéfiig. A Review prepared for the Conference on Learn-
ing by Discovery. New York City, January 28-29, 1965.
SPonsored by Stanford University with support from the
U- 8- Office of Education. 1964, p. 10.

29

school discovery is not independent but di-
rected by the knowledge gained under the di-
rection of previous teachers. The more di-
rection of this kind available to the learner.
the more effective his discovery of new re-
lations. The cumulative effect of greater
learning through directed discovery over
months or years may offset the effect of
any lack of similarity between learning

and transfer situations and proves to be 15
the best preparation for new discoveries.

A more middle—of-the—road view is expressed
by Stanley:

It is evident that the young human being
must receive considerable instruction but
also that he should be eternally vigilant

in making additional observations. His

life is a complicated blending of instruc-
tion and discovery. Many facts will be
handed to him outright. At the same time.
during every day of his life he will be en-
gaged, almost unknowingly, in inductive rea-
soning. the process of bringing together a
number of experiences and abstracting from
them some common factor. The issue becomes,
then, not instruction versus discovery.
since both are essential, but a considera-
tion of the relative importance to be accorded

each in the educative process.

 

15R. C. Craig, "Directed Versus Independent Dis-
covery of Established Relations," Journal of Educational
PSVcholoqy. XLVII (1965). p. 233.

16J. C. Stanley, "The Role of Instruction, Dis-
covery. and Revision in Early Learning." Elementary
School Journal, XLIX (1949), p. 457.

 

 

 

1

Yb

CV

30

The major criticism of the method is that it
is too time consuming. If, say its critics. everyone
had to rediscover each fact or idea for himself, prog—
ress in acquiring knowledge would stumble to a halt.

A second criticism is that the discovery method in-
creases the possibility that the student will learn
false information. Critics use a statement by Thorn-
dike in support of this contention:

The attainment of active vs. passive learn—
ing at the cost of practice in error may
often be a bad bargain . . . . The almost
universal tolerance of imperfect learning

in the early treatment of a tOpic, leaving
it to be improved by the gradual elimination
of errors in the later treatments, is prob—
ably unsound and certainly is risky.17

Another reviewer of "learning by discovery"
concludes:

The issue of expository teaching versus inde-
pendent discovery in the learning, retention,
and transfer of principles is still very much
in doubt because of the non-comparability of
the various studies, serious deficiencies in
research design, and the failure to hold con-
stant or take into account rote-meaningful,

l7E. L.'Thorndike, The Psychology of Wants,

Interests, and Attitudes (New York: Appleton-Century.
1935). p. 147.

 

31

inductive-deductive, verbalization, ability
level, cognitive maturity, subject-matter
sophistication. and motivational variables.
In general the research findings support
Thorndike's well-known conclusion that "re-
fusal to supply information on the ground
that the learner will be more profited by
discovering the facts himself, runs the risk
not only of excessive time-cost but also the
strengthening of wrong habits." Providing
guidance to the learner in the form of verbal
explanation of the underlying principles al-
most invariably facilitates learning and re-
tention and sometimes transfer as well. Self—
discovery methods and the furnishing of com-
pletely explicit rules on the other hand.

are relatively less effective.

The most efficacious type of guidance (guided
discovery) is actually a variant of expository
teaching that is very similar to Socratic ques-
tioning. It demands the learner's active par—
ticipation and requires him to formulate his
own generalizations and integrate his knowledge
in response to carefully programed leading
questions: and it is obviously much more highly
structured than most discovery methods, with
the possible exception of the UICSM [University
of Illinois Committee on School Mathematics].
Further research is necessary to determine
whether guided discovery is superior to simple
didactic exposition in terms of relative effec—
tiveness for the time-cost involved when such
factors as cognitive maturity, subject-matter
sophistication, and verbal ability are varied.18

 

18Ausubel. op. cit., pp. 171-172.

Ar
»L

(.1.

32

The following studies are arranged in a chron-
ological order so that a clearer picture of the devel—
opment of the discovery method may be perceived. This
procedure closely follows the excellent review of M. C.
Wittrock, although the emphasis is somewhat different
since the purposes of the study do not demand such a
comprehensive review.

In Judd's19 early (1908) empirical study of a
discovery method. he tested the ability of two groups
to hit a target placed under water. One group was told
the principle of refraction and allowed time to practice
throwing darts at the submerged target. The second
group was not given an explanation of refraction and
spent the entire time throwing darts at the submerged
object. Judd tested the group by changing the depth
of the water. The group given the explanation of re-
fraction was able to hit the target more often than
the second group. The purpose in referring to his re-

search is to point out the value of his empirically
l9 .
Judd. 2.r>_-_c_11:_-

 

(9‘

C)

In

33

asking the question. "How does independent learning
contrast with expository learning?"

The influence of Thorndike slowed experimenta—
tion with the discovery method. The following dictum
of his tended to discourage research to the contrary:

The teacher and the learner must know the

characteristics of a good performance in

order that practice may be apprOpriately

arranged. Errors must be diagnosed so that

they will not be repeated. When there is

lack of clarity about what is being taught

or learned, practice may be strengthening

the wrong connections as well as the right

ones. At the same time. needed connections

may be weakened by disuse.20
In 1949 three researchers, Swenson, Anderson and Stacey,
set out independently to show that Thorndike's experi-
ments on the memorization of nonsense syllables were
too primitive a base from which to generalize to human,
complex. verbal learning. It was believed by these re-
searchers that studies using "meaningful" materials

might produce results disagreeing with those of Thorn-

dike.

 

OErnest R. Hilgard. Theories of Learning (New
York: Appleton-Century—Crofts. Inc.. 1956). p. 23.

34

Swenson21 randomly assigned. by levels. 332
second graders into three treatment groups: A gener-
alization Group which was treated by a variation of
the discovery method, a Drill Group which followed
procedures agreeable with Thorndike's connectionism,
and a Drill-Plus Group judged as a prototype of methods
currently used in the schools. Students in the Gener-
alization Group were encouraged by the teachers to
"discover" relationships among number combinations
of addition problems (a form of reward). Children
in the Drill Group were taught mainly by group drill.
A wrong answer was immediately corrected (a form of
punishment). In the Drill—Plus Group concrete objects
were used to present number combinations, either through
pictures or by actual handling of the objects. This
was followed by the same drill procedure used with the
Drill Group.

During the 20 weeks that the treatments were

applied, three sets of facts were taught: an original

 

21Esther J. Swenson, G. L. Anderson, and G. L.
Stacey. Leagning Theory in School Situations (Minneapo-
lis. University of Minneapolis Press, 1949), pp. 9-39.

 

In
(I)

(II
(T)

'u‘i

0*

f)

35

set of facts, an interpolated set of facts. and a final
set of facts. Students were tested over all of these
facts five different times during the instruction. At
the completion of the instruction. three transfer tests
were given. The first consisted of 100 facts on sub-
traction, the second was over 100 addition problems
slightly different from those given during instruction,
and the last consisted of addition problems with an in-
creased number of digits. The following table summar—
izes the results.

The best statement that can be made about her
findings in reference to the contrast between the dis-
covery and expository methods is that: students given
the set to generalize and who are reinforced for devel-
oping and verbalizing their generalizations do better
than students who are discouraged from verbalizing gen-
eralizations. Unfortunately, the effect of the discov-
ery method cannot be separated out from the effects

of reward and punishment used in the treatments.

36

TABLE 2.1

SUMMARY OF SWENSON'S DISCOVERY LEARNING EXPERIMENT

 

 

 

Generaliza- Drill Drill—Plus
Criteria tion Group Group Group
Original Facts Best Middle Poorest
Interpolated
Facts (no significant difference among
the groups)
Final Facts (no significant difference among

the groups)

Transfer (with
Adjustment for
Mental Age)

1.

Addition
Facts Best (no significant differ-

ence between Drill and
Drill-Plus Groups)

Transfer

a. Subtrac-

tion Best Middle Poorest
b. Digit

Number (Generalization and Drill Groups were
significantly better than the Drill-
Plus Group)

 

. 22
In a similar research prOJect, Anderson also

contrasted Thorndike's connectionism with active discovery.

¥

22Ibido I Pp. 41-73 .

37

His carefully trained teachers taught arithmetic to
389 elementary students. His design consisted of two
treatments. The Drill Treatment incorporated the major
tenets of Connectionism: students were taught discrete
elements. their questions were answered concerning the
arithmetic prdblems, and they were drilled on the dis-
crete elements. In the Active Discovery Treatment.
which tended to align itself with field theory. organ-
ization was around the following elements: students
were encouraged to discover organized patterns among
the arithmetic problems, direct questions on the prob-
lems were not responded to. and the students were given
practice at using generalizations.

The performance criteria were computational
skills. problem solving. understanding of social con—
cepts in arithmetic and vocabulary, and of mathematical
thinking. The significant differences occurring for
the computational skills indicated that the discovery
method (meaning method) was best for students scoring
high on the ability test. while the drill method seemed

best for those scoring low on the ability test. On

38

tests of mathematical thinking, the discovery method
again proved superior(.01 level of significance) to
the drill method for high ability students. while the
drill method was superior (again at the .01 level)
for the low ability students. In the case of the
social concepts and vocabulary tests there were no
significant differences. For teaching a few specific
associations it was found that drill was more effec-
tive with both the high and low ability students. A
final conclusion was that, if transfer of information
to new situations is the learning objective, the dis-
covery or meaning method is superior to drill. Since
the teacher variable was not well controlled and the
time involved was for a period exceeding six months,
care must be taken in interpreting the results.
Gertrude Hendrix23 compared three treatments
using both high school and college students. In the

first treatment. a generalization (the sum of the first

3 .

Gertrude Hendrix. "A New Clue to Transfer of
Training." Elementary School Journal, XLVIII (1947),
Pp. 197-208.

39

p odd numbers is.p square.) was stated, illustrated.
and then applied to new problems. In the second treat-
ment, called the Unverbalized Awareness Method, the
students were asked to find the sum of the first two
odd numbers. then the first three, then the first four.
and so on until they showed by a change in behavior
(eg. increased rate of solving problems) that they had
discovered the generalization stated for Group I. The
third treatment, called the Conscious Generalization
Procedure. was the same as for Group II, with the addi-
tional task of stating verbally the rule they had dis-
covered.

On a transfer test given two weeks later there
was no statistical difference. But. the results indi-
cated a trend favoring the Unverbalized Awareness Method.
The small number of students used in her study further
limits the conclusions to be drawn from it. A major
value of the study is that the effect of discovery is

measured through transfer of learning.

 

«C

ta

.9».

EV

HU-

40

Like Swenson and Anderson. Stacey24 argued that.
provided the stimuli are meaningful (rather than non-
sensical as in Thorndike's studies). the discovery method
can aid students in learning. He particularly pointed
out the lack of guidance received by Thorndike's sub-
jects. He deliberately used amount of guidance as the
major variable in his study. Items consisting of five
words were presented to his subjects. The student's
task was to determine which one of the five words did
not belong with the other four because of some common
element possessed by the four and not by the other.
Group A was told that a choice was correct or incorrect
but not why. Group B was given the same information as
Group A and also informed that there was a reason for
the item not belonging and allowed to work three sample
problems before beginning the learning task. Group C
was given the correct answer with each item. Group D
was given the correct answer with each item and was

allowed to work three sample prdblems prior to beginning

24Swenson, Anderson. and Stacey, 0p. cit., pp.
74-103.

 

(*
--.

’9‘-

41

the learning task. Group E's treatment was the same
as Group D's except they were also told why the answer
was correct.

He used the same instrument for both pretest
and post—test. There were no significant differences
among the five treatments. On a measure requiring
correct reasons for the responses on the pretest and
the post-test there was a strong trend favoring methods
A and B (Stacey's discovery method). He interprets
this to mean strong support for discovery, and that
errors made through discovery are not as damaging as
are those made through the expository method. In Sta—
cey's study a shift of emphasis may be noted. from a
contrast of Connectionism and Field Theory to a con-
trast between the discovery method and the reception
or expository method.

Craig,25 using the same learning task as did

Stacey. attempted to prove that guidance could produce

25R. C. Craig. "The Transfer Value of Guided

Learning," Bureau of Publications. Teacher's College
Columbia University, New York. 1953.

42

a positive effect on discovery. Of his four treatment
groups. Group A was given a minimum of guidance. Group
B was given the items grouped according to common prin—
ciples. Group C received a blank space between the sets
of items which were grouped by a common principle and
informed that all items in a group were organized into
a common principle, and Group D was given all of the
information provided by the other groups plus a short
statement of each principle. Craig found that in both
initial learning and on transfer tasks. the groups with
the most clues had a superior ability to discover.

In a follow—up study. Craig26 hypothesized that
directed discovery would not only produce superior re-
sults on initial learning but on transfer and retention
as well. This time he used two treatment groups: a
No-help Group who were told only that test items grouped
together could be solved by a common principle. and a
Directed Group which was given the same information as

the first group except the principle was stated above

26Craig. op. cit., pp. 223-234.

43

each group of four items. Of the three retention tests
given. there was no significant difference on the tests
given 3 and 17 days after the treatment. but there was
a significant difference in favor of the Directed Group
on the test given 31 days later. He found no signifi-
cant differences on transfer to new principles. How-
ever. on the test immediately following the treatment.
the Directed Group was again superior to the No-help
Group.

Studies carried out by Corman27 and Kittell28
tended to support Craig's conclusion that by giving
rules transfer could be increased.

Haslerud and Meyers29 got results that contra-

dicted those of Craig. Corman, and Kittell. Their two

 

27B. R. Corman, "The Effect of Varying Amounts

and Kinds of Information as Guidance in Problem Solving,"

Wanna. LXXI (1957). 1313- 1-21-

28J. E. Kittell. "An Experimental Study of the

Effect of External Direction During Learning on Transfer

and Retention of Principles." Journal of Educational Psy-
cholpgy. XLVIII (1957), pp. 391-405.

29G. N. Haslerud, and Shirley Meyers. "The Trans-
fer Value of Given and Individually Derived Principles,"

Journal of Educational Psychology. XLIX (1958), pp. 293-

298.

 

44

treatments consisted of a No-direction Group. and a
Specific Directions Group. The subjects were asked
to solve 10 enciphering problems. While there was no
statistical difference between the two group's initial
learning. the Specific Directions Group had the largest
mean performance on the test of transfer to new sent-
ences. The main reason that the researchers favored
the No-direction Group treatment was because the per-
centage of loss on gain scores was less for this group.
In his review of this study Wittrock suggests that:

Since each subject was given practice on

each of the two experimental conditions,

one might expect that the practice at dis-

covering would generalize from one set of

items to the other and that the treatments

would then be contaminated.30

Elementary and junior high school students (180)

in Baltimore were divided into six, ten, and fourteen
year old age groups. Half of the students were of av-
erage intelligence and the other half were of high in-

telligence according to IQ scores from the Wechsler

Intelligence Scale for Children. A concept attainment

 

3OWittrock, 0p. cit., p. 41.

 

(1’

0‘
.A

SC

Ci

Cl.

45

task was presented under general and explicit instruc-
tions. Data were analyzed using t—tests, F—ratio, mean
scores, standard deviation, variance. and rank order
correlations. The results were analyzed for errors.
number of students who achieved the criterion of suc-
cess, and number of students who verbalized the concept
correctly. Results indicated that, under nonspecific
instructions. superior intelligence is associated with
more effective concept attainment; under explicit in-
structions students of average intelligence improved
while those of superior intelligence remained unchanged,
and performance improved throughout the age range
studied.31

In a study of intelligence levels. Klausmeier

and Loughlin32 obtained results which have meaning for

 

318. F. Osler and S. R. Weiss, ”Studies in Con-

cept Attainment: III. Effect of Instructions at Two
Levels of Intelligence," Jpprnal of Experimental Psy-
cholpgy. LXIII (1962), pp. 528-533.

32H. J. Klausmeier and L. J. Loughlin, "Behavior
Problem Solving Among Children of Low. Average, and High
Intelligence,” Jougpgl of Educational Psychology. LII
(1961), pp. 148-152.

n
O.—

f)

”D

46

the choice of instructional method for different levels
of intelligence. Sixty fifth grade boys and sixty fifth
grade girls acted as subjects. They were divided into
three intelligence levels: 1) 20 boys and 20 girls with
intelligence scores of 55-80, 2) 20 boys and 20 girls
with scores of 90—110, and 3) 20 girls and 20 boys with
scores of 120-146. Each child. after a period of pre-
training, was assigned an arithmetic problem that was
graded in difficulty for his or her intelligence level.
Close observation was made of the child's method of
solving his particular problem. Results indicated that
application and verification of information are corre-
lated with intelligence. Since verification and appli-
cation skills are necessary for the student learning
through the discovery method. it suggests that the
brighter student would be more successful in using the
discovery method than the less bright. The same study
noted that below average subjects do not persist at

the task when they find their first solution to be in—
Correct. while above average students do persist. Since

use of the discovery method requires a great deal of

47

persistence, the implication is again that bright stu-
dents would be more successful with the discovery method
than the less bright.

The research of Kersh directs our attention
away from "meaningfulness" of the learning task as the
intervening variable responsible for the effect of the
discovery method. In his early work33 he randomly as-
signed sixty subjects to six treatments. The first 20
were called the No-help Group and were required to dis-
cover the desired rules without aid. The second 20
formed the Direct—reference Group and were given help
in the form of perceptual aids (symbol patterns). The
third group was called the Rule-given Group and was
told the rules directly and given practice in applying
them. Each of these three groups were further broken

p
into two sub-groups on the basis of the problems-to-be—

solved being presented either in Hindu-Arabic form, or

in the more nearly iconic form.

33Bert Y. Kersh. "The Adequacy of Meaning as an

Explanation for the Superiority of Learning by Independ-
ent Discovery," Journal of Educational PsyChology, XLIX
(1958). pp. 282—292.

([3
(IX

48

Teaching two rules. the Odd-numbers rule, and
the Constant-difference rule was the task set for the
six treatments. The students were taught individually.
Several of them were unable to discover the intended
rules within the 60—90 minutes time period scheduled.
Immediately following the learning period. the subjects
were given 20 problems to solve. From four to six weeks
later the students were retested. Kersh was not pri-
marily interested whether the student got the correct
answer. but rather in the method used to solve the prob-
lem. The No-help Group produced more acceptable methods
on the test of transfer to new examples of rules than
did the others. The Directed Group did best on the first
test. On the retest the No-help or Discovery Group was
superior although not with high reliability. Kersh in-
terprets this last fact as meaning that as a result of
their experience during the learning period. the No-help
Group were motivated to continue learning after the
learning period whereas subjects in the other treatment

groups were not. In Kersh's words:

49

Presumably. the motivating power is of the
type that lies in acquired interest or ego
involvement in a task, and develops to the
extent that the individual relies on his
own cognitive capacities in learning.34

And finally:

The results oftfifis experiment suggest that
when the learner is forced to rely on his
own cognitive capacities. it is more likely
that he will become motivated to continue
the learning process or to continue prac-
ticing the task after the learning period.
Consequently, the learning becomes more
permanent and is more effectively trans-
ferred than when the learner is not moti-
vated.

In a follow-up study. Kersh tested the hypothe-
sis that guided discovery motivates the student to prac-
tice more and thus to remember and transfer more than
if he were taught directly. His data supported the hy-
pothesis.36

Gagne and Brown37 assigned 33 ninth and tenth

 

34Ibid., p. 291.

35Ibid., p. 292.

36Bert Y. Kersh. "The Motivating Effect of Learn-
ing by Directed Discovery." Journal of Educational Psy-
chology. L111 (1962). PP. 65-71.

37R. M. Gagne and L. T. Brown. "Some Factors in
the Programing of Conceptual Learning." Journal of Ex-

perimental Psychology. LXII (1961), pp. 313—321.

50

graders to three treatments designed to teach principles
pertaining to a number series. The first treatment was
called the Discovery Program which asked the subject

for a rule for the number series. and although hints
were given. the rules were not stated. The second
treatment was labeled the Guided Discovery Program and
proceeded very much as did the Discovery Program except
the questions were more specific and items presented ex-
amples of relationships within the number series. The
last treatment was called the Rule and Example Program
which provided the subject with the correct principle
(formula) which the student was allowed to copy before
working several examples.

The students went through the same program on
two different days. They were then measured in three
ways: on the time required to solve the problems, num-
ber of clues necessary to solve the problems. and a
weighted time score combining these. All subjects were
required to derive the formula for the sum of.p terms

in a number series.

 

(37

51

The results favored the Guided Discovery Group.
with the Discovery Group second and the Rule and Example
Group last. All measures were statistically significant
below the .01 level.

Bruner38 offers an explanation for the superior-
ity of the discovery method in one of his studies. He
presented 30 paired-associates to three groups of twelve—
year-old children. One group was told only to remember
the pairs, and that they would be asked to repeat them
at a later time. The second group was told to remember
them by using a word or idea to tie the associates to-
gether (e4; Chair-forest: "Chairs are made from trees
in the forest). Bruner calls the words or ideas tying
associates together "mediators." The third group was
asked to use the second group's mediators for remember-
ing their paired-associates. Bruner found that the
children who developed their own mediators did best.
In fact, after going through the thirty pairs once.

they recovered up to 95% of the second words when

38Br‘uner. OE- Cit- 3 pp. 31-32.

52

presented with the first word of the pair, while the
other groups reached a maximum of less than 50%. The
value of the discovery method seems to lie in its flex-
ibility of information handling. Bruner concludes:

. . . in general, material that is organized

in terms of a person's own interests and cog-

nitive structures is material that has the
best chance of being accessible in memory.

39
Gagne and Brown also gave as an explanation for the su-
periority of self-discovery of a principle over being
taught the principle. that the former procedure allows
the principle to fit better into the individual's ex—
isting verbal system.

There is considerable research evidence support-
ing the hypothesis that redundancy of relevant informa-

tion facilitates performance on concept learning. The

data from studies by Bourne and Haygood4l suggest that

 

39Ibid. p. 32.

40Gagne and Brown. op. cit., pp. 319-320.

41L. E. Bourne. and R. C. Haygood. "Supplement-
ary Report: Effect of Redundant Relevant Information
Upon the Identification of Concepts," Journal of Exper-
imental Psychology, LXI (1961). pp. 259-260.

 

“C

.9

'a

"(1

In

53

simplification in audio visual presentations. while
desirable. should not be done at the expense of needed
redundant relevant information. The treatment condi-
tions of this study have adhered to the cautions of
Bourne and Haygood.
42 . 43

Work by Smoke. Hovland and Weiss. and

44 - '
Bruner. have shown rather concluSively the superior
teaching value of positive instances of concepts over

. . 45

negative instances. Some later work by Huttenlocker
indicated that if positive instances are sequenced
properly with negative instances of a concept the re-

sults are superior to the use of all positive instances.

 

42 . .
K. L. Smoke, "Negative Instances in Concept

Formation," Journal of Experimental Psychology. XVI
(1933). pp. 583-588.

43C. I. Hovland. and W. Weiss, "Transmission
of Information Concerning Concepts Through Positive

 

and Negative Instances," Journal of Experimental P815
shelgsx. XLV (1953). pp. 175—182.
44

J. S. Bruner. J. J. Goodnow. and A. Austin.
A Study of Thinking (New York: Wiley. 1957). pp. 119-
122.

45J. HuttenlOCRer. "Some Effects of Negative
Instances on the Formation of Simple Concepts." sy-

stelssissl.§spert§. XI (1962). pp. 35-42.

 

.(2

(n

54

His use of extremely simplified concepts raises the
question of whether his combinations would still hold
for a concept as complex as "adaptation.” The decision
was made to accept earlier findings treating more com-
plex concepts which support the use of positive in-
stances only.
46 47
Huttenlocker and Bruner, among other re-
searchers, take issue with Vinacke's statement about
the sequential order of instances of a concept:
The order of the instances (of a concept)
does not particularly matter if an equal
amount of time is allowed to each.
Studies of sequencing in programed learning

have tended to agree with Vinacke. For example. Roe,

Case. and Roe49 reported no difference in the learning

 

46Ibid.

47Bruner. Goodnow. and Austin, 0p. cit., pp.
96—103.
48 . n . .
W. E. Vinacke. The Investigation of Concept
Formation," Psychological Bulletin, XLVIII (1951), p. 22.
49 .
K. V. Roe. H. W. Case. and A. Roe, ’Scrambled
Versus Ordered Sequence in Autoinstructional Programs,"

Journal of Educational Psychology. L111 (1962), pp. 101-
104.

 

old

'(1

*‘I

55

of programed materials presented in the original se-
quence as compared with the same program with the in-
stances arranged in a random sequence.

Because of the complexity of the problem. this
study attempted only to control for the effect of se-
quencing by: l) keeping the time for presentation of
each instant of the concept as nearly equal as was
possible: and 2) providing both an ordered and a
random treatment of sequences.

A question adjunctive to this study was the
degree to which a student attains a concept. A major
finding by Hull50 was that the ability to state a con—
cept requires greater abstracting ability than recog-
nizing instances of the concept. In order to distin-
guish between the effects of teaching method on differ—
ent conceptual tasks and the degree of attainment of
the concept. three criterion variables were used in
this study. They were, in decreasing order of diffi-

culty: the application of the newly learned concept

 

50C. L. Hull, "Quantitative Aspects of the Evo-

lution of Concepts." Psychological Monograph, XXVIII
(1920). p. 85.

56

to a prdblem situation: generation of new instances
of the concept; and recognition of a new instance of

the concept.

Summary

The review of the literature relevant to this
study has illustrated the diversity of Specific inter-
ests among the researchers.

Research on the usage of film in teaching is
plentiful but for the most part not appropriate to this
study. Its inappropriateness is due primarily to two
factors: few researchers contrast alternate versions
of a film using two distinct instructional methods,
and fewer still have the effect of the teacher variable
completely removed from the treatment as has been at-
tempted here.

A great deal of evidence has been provided in

support of the Directed Discovery Method of teaching.

along with its historical development. The major lines

 

.
A}
0'.

"fl

57

of divergence of the proponents of the discovery approach
have been presented. One group maintains that the advan-
tage of discovery lies in learning content. While the
second insists that its major value lies in the actual
practice of the method. This study limits its concern

to the method's application in teaching content: spec-
ifically the teaching of science concepts through the
single-concept film.

Of special interest was the evidence garnered
from among the several studies in support of the moti-
vational. transference, and retention effects of the
discovery method.

Other studies were selected for their support
of the treatment of the science concept taught by the
single-concept film. Some of the elements considered
were: redundancy of information. intelligence level,
positive versus negative instances of the concept, se-
quencing of the concept, instances, and the degree of

attainment of the concept.

CHAPTER III

THE EXPERIMENTAL DESIGN

The Population

The population consisted of students in nine,
eighth grade General Science II classes of Pattengill
Junior High School, in Lansing, Midhigan. This urban
junior high school draws its students from families
whose incomes vary from high to low, and Whose members

include all races.

The Sample

Of the population, 288 students were available
to take part in the experiment. Because of illness and
transferring to other schools, 4 students were lost from
the experiment. In order to satisfy design and statisti-
cal requirements, 4 other students were randomly selected

58

 

LL;

59

out of the experimental treatments, leaving a total of
280. Of this number, 146 were boys, and 134 were girls.
Ages varied from 13 years 2 months to 15 years 3 months.
The variance in IQ was great; ranging from a low of 51

to a high of 138.

Instrumentation

This section describes the content. the medium,
the apparatus. and the testing materials used in the

experiment.

Content- The concept "Adaptation” was selected
on the basis of teacher judgment and a pretest. The
General Science II teachers at Pattengill were given a
list of concepts that pertained to the area of study in
the time period for which the treatment was planned.
There was unanimous agreement among them that "Adapta-
tion" was one of the concepts that was not held by their
students. A pretest was devised to test their conviction

(Appendix A). A random selection of 30 students was used

(3
AA

r‘f

60.

in the pretest. The results substantiated the teachers'
claims. An agreement was reached with those teachers
whose classes were participating in the experiment. not

to introduce the concept prior to the experiment.

Medium. The concept "Adaptation" was the sub—
ject of two, four minute films. The films consisted of
ten instances of the concept: these instances were ex—
cerpted from existing film. Both biological and beha—
vioral instances of adaptation were included in the two
films. The same instances appeared in both films. but
the order of presentation of the instances varied. In
one of the films the instances were randomly assigned
their position, while in the second film they were or-
dered from easy to difficult as determined by a pilot
study using other eighth grade general science students.

Each of these films was used to teach both by
the discovery method and by the expository method. This
was accomplished through the provision of two sound
tracks for each film. When the film was utilized in

the expository mode. the sound track stated the concept

61

and gave careful explanations of the instances of the
concept presented by the film. When using this same
film for the discovery method. the second sound track
directed the student to find the relationship or common
element among the instances of the concept as presented
by the film. Where necessary, the sound tracks also
clarified instances of the concept in order to prevent
misinterpretation. Scripts for the Discovery and Ex-
pository narrations are in Appendix B. Support for

the use of narration in the control of visual stimuli
(motion pictures) may be found in research studies by

McClusky.l Einbecker.2 Jaspen.3 and McGuire.4

 

1S. D. McClusky and H. Y. McClusky. "Comparison

of motion pictures. slides. stereographs and demonstra—
tion as a means of teaching how to make a reed mat and
a pasteboard box," Visual Education, ed. F. N. Freeman
(Chicago: University of Chicago Press. 1924), pp. 310—
334.

2W. F. Einbecker. "Comparison of Verbal Accom—
paniment to Films." School Review; XLI (1933), pp. 185-
192.

3N. Jaspen, Effects on Training of Experimental
Film Variables, Stpdyilll Progress Report No. 14-15-16.
State College, Pennsylvania: Pennsylvania State College,
Instructional Film ResearCh Program, 1950.

4W. J. McGuire, "Slow Motion, Added Narration
and Distributed Showing as Factors Influencing Teaching

 

62

Apparatus. The apparatus selected to show the
two films to the treatment groups was the Technicolor
cartridge-loaded projector with the Audio-Sell sound
attachment. The device synchronized the narrated tape

with the film.

Criterion Performance. Three tasks were de-
vised to determine the relative effects of the treat—
ments. In their order of increasing difficulty they
were: 1) To identify instance instances of the concept
from a collection of examples of several concepts. 2)
to generate new instances of the concept taught in the
film, and 3) to apply the newly learned concept to
prdblem situations. A copy of these tests, along with
the rating scales deve10ped to guide scorers, may be
found in Appendix C. These tasks fitted roughly into

Bloom's taxonomic scheme5 of intellectual ability and

 

Effectiveness of a Training Film," Visual Communication,

ed., John Ball and Francis C. Byrnes (Washington, D. C.:

The Department of Audiovisual Instruction of the National
Education Association, 1960), p. 96.

5B. S. Bloom (ed.), Taxonomy of Educational Ob-
jectives (New York: David McKay Company, Inc.. 1956).

63

skills. The identification of new instances of the
concept demanded of the student a level of comprehen—
sion.
such that the individual knows what is being
communicated and can make use of the material
or idea being communicated without necessarily
relating it to other material or seeing its
fullest implications.6
The second task. that of generation of new in-
stances of the concept, was a translation problem where
the student was required to paraphrase or render the
communication from one form to another. The most dif-
ficult task. using the newly learned concept to solve
a problem, fitted Bloom's Application Category.7
The first two tasks were evaluated by three
judges using a.five point rating scale. Inter-judge
reliability was investigated by means of correlations
(Tables 3.1 and 3.2). The high correlations between

the judges' scores indicated an extremely high inter-

judge reliability. It was noted that the correlations

6;p;g.. p. 204.

71bid.. p. 120.

64

were highest for Task I. The third task was treated
as a true or false test. Its inter-item reliability
was evaluated by the Kuder-Richardson Formula 20 at

.7013.

TABLE 3.1 TABLE 3.2

PRODUCT-MOMENT CORRELATIONS PRODUCT-MOMENT CORRELATIONS

 

 

 

BETWEEN JUDGES SCORES FOR BETWEEN JUDGES SCORES FOR
TASK I TASK II

Judge Judge Judge Judge Judge Judge Judge Judge
A B C A B C

A .917 .966 A .799 .871

B .938 .836

C

 

Experimental Design and Analysis

Hypotheses. The following hypotheses were in-
vestigated in order to determine the effectiveness of
the discovery method of teaching. as opposed to the ex—

pository method, when applied to the medium of single-

concept films:

65

Directional Hypotheses One. The nature of the

unique relationship (e.g. interaction) between

intelligence and method is of the following

form:

a.

All immediate criterion performances of
students above an IQ of 100 will be greater
if they are taught science concepts through
the discovery method, than if they are
taught by the expository method, regardless
of the sequence of the concept instances.

Symbolically: H : M > M

la 1 2
Legend: M1 = Discovery method group mean;
M2 = Expository method group mean.

All immediate criterion performances of stu-
dents below or equal to an IQ of 100 will be
greater if they are taught science concepts
through the expository method, than if they
are taught through the discovery method. re-
gardless of the sequence of the concept in—

stances.

66

S 1' : :
ymbo ically Hlb M2 > Ml

Legend: M1
M2

Discovery method group mean:

Expository method group mean.

Directional Hypothesis Two. The retention of
science concepts, as measured by delayed criter-
ion performances, will be greater for all stu-
dents taught by the discovery method than for
those taught by the expository method, regard-
less of the sequence of the concept instances.

Symbolically: H - M > M

2' l 2
Legend: M1 = Discovery method group mean;
M2 = Expository method group mean.

Statistical Hypothesis One. In the relationship
between intelligence and method (e.g. the unique
component called interaction) there will be no
difference among the cell means for the immedi—
ate criterion other than that which could be ac-
counted for independently by intelligence and
method. regardless of the sequence of concept

instances.

67

Symbolically: H

3
II
{Z

01 l 2

Legend: M Discovery method group mean:

1
M2

Expository method group mean.

4. Statistical Hypothesis Two. No differences
will be found in the retention of science con-
cepts when presented through the single—concept
film as measured by delayed criterion perform-
ances, between students taught by the discovery
method. and students taught by the expository
method. regardless of the sequence of the con-
cept instances.

Symbolically: H02: M1 = M2

Legend: M1 = Discovery method group mean:

M2 = Expository method group mean.

Treatments and Assignment of Students. The
median of the IQ scores (100) was used as a point for
seParating the students into two levels. These two
groups responded to four treatment conditions:

1- The ordered sequence of the concepts was viewed.

and the discovery tape was heard.

68

2. The random sequence of the concepts was viewed,

and the discovery tape was heard.

3. The ordered sequence of the concepts was viewed.

and the expository tape was heard.

4. The random sequence of the concepts was viewed.

and the expository tape was heard.

The procedure for assigning subjects to the
four treatment conditions was as follows: within the
two IQ levels the subjects were ranked by their IQ
scores. these ranks were divided into quartets, and
then the members of each quartet were randomly assigned
to the treatments. No student was allowed to partici-
pate in more than one treatment. Each treatment was
administered in a separate room. Directions to each

student, within limits, were the same.

Analysis of Data. The statistical procedure
involved a three, three-way univariate factorial analy-
ses of variance with an inspection of the data (means

and standard deviations of the factor levels). For

{u

«1

69

eadh analysis the main effects were: 1) sequence of
concept instances. 2) intelligence level, and 3)
teaching method. The dependent variables were: 1)
generation of new instances of the concept, 2) appli-
cation of the concept to problem situations, and 3)
recognition of new instances of the concept. The in-
terpretation underlying analysis of variance--normal
distribution of errors, equal error variance (homogen—
eity), and independence of error portions entering into
the respective observations--follow Hays.8 He reported
that for a relatively large number of observations per
cell (above 25) it has been both theoretically and em-
Pirically determined that the requirement for normal
distribution of errors may be violated without serious
consequences. Similarly. when an equal number of ob-
servations were maintained in each cell the assumption
of homogeneity of variance could be violated without
causing serious consequences. In deference to these

two assumptions, 35 Observations were reported for each

 

8W. L. Hays, Statistics for Psychologists (New
York: Holt, Rinehart, and Winston. Inc.. 1963), pp.
396! 4080

70

cell. In satisfying the third assumption--independence
--the usual restrictions were placed on the selection
and use of the subjects: they were randomly assigned
to treatments, each student was used in only one treat-
ment, and safeguards were provided to prevent interac-
tion among students during the treatment period.

A number of Pearson product-moment correlations
were also run to determine if the three criterion tasks
were independent. and to establish inter-judge reliabil-
ity. Again. the question of underlying assumptions was
based upon the authority of Hays, who noted that as long
as one was not predicting beyond the sample. the assump-
tion of linearity could be violated without serious con-
sequence. In fact, the usual consequence of violating
this assumption was an underestimate of the true relation-
ship. In concluding he stated:

So long as there are N distinct cases, each
having two numberical scores. X and Y, then

the desgpiptive statistics of correlation
and regression may be used.9

 

91bido I pp. 509-510 a

71

An Orthogonal Factor Analysis of the above db-
tained correlation matrix was run in order to determine
dependence or independence of the criterion variables.
The factors were rotated by means of the verimax rota—
tion, and the rotation was halted or st0pped according

to the Kiel-Rigley criterion.

Chronological Procedure of the Experiment.
Three weeks prior to the administration of the treatments,
the subjects were given an intelligence test.10 Results
from this test, as was mentioned earlier, were used to
assign the students to the four treatment conditions.

Two weeks before administration of the treatments,
a pilot study was run using a group of students considered
comparable to the subjects in the major study. The pilot
study was designed to accomplish four things: to ascer-
tain that the students did not possess the proposed con-
cept. to determine the ordered sequence of the concept

instances, to train administrators in the use of the

 

10Otis Quick-Scoring Mental Ability Tests. Beta

Test: Form cm.

72

apparatus and in handling their respective groups. and
to train judges in the use of the scales developed for
evaluation of the test data. The results from this pi—
lot study indicate that the four aims were satisfactor-
ily met.

On April 22, the films were shown to the four
treatment groups. Immediately after viewing the film
the students were required to take the three criterion
tests. Treatment and testing conditions. except for
the three independent variables being tested, were held
as near constant as circumstances would allow.

Approximately three weeks later the students
were given the delayed post-test. which was the same
as the one given for the immediate post-test. The pur-
pose in re-administering this test was to determine re-
tention effects.

In the period between the post-test and the de-
layed post-test and immediately following the delayed
post-test the judges scored the tests. The data they

provided were placed on IBM cards in a form commensurate

73

with the requirements of the two computer programs:

FACREPll and FANOD.12 used to analyze the data.

Summary

The sample consisted of 280 junior high school
students from nine. eighth grade General Science II
classes from Pattengill Junior High School in Lansing,
Midhigan.

The plan of this study was to teach the concept
"Adaptation" through the use of a single-concept film.
This single-concept film was shown to the subjects in
four modes: 1) with the instances of the concept in
an ordered sequence using the discovery narration, 2)
with the instances of the concept in a random sequence

using the discovery narration. 3) with the concept

 

11D. F. Kiel. A. L. Kenworthy. W. L. Ruble. Use

of Analysis of Variance Routines on the CDC 3600 (East
Lansing: Michigan State University. 1963). p. 22.

12J. J. DeJonge and F4 M. Sim, Factor Analysis
Programs: Fanod 3 and Fanim 3 (East Lansing: Michigan
State University. 1964). pp. 1-7.

74

instances in an ordered sequence using the expository
narration. and 4) with the concept instances in a ran-
dom sequence also using the expository narration.

Subjects were randomly assigned to the treat-
ments. after an initial separation on the basis of high
and low intelligence using an IQ of 100 as a dividing
point.

Three criterion tasks were devised to determine
the degree of concept attainment: 1) the generation of
new instances of the concept taught by the film, 2) the
application of the new concept to problem situations,
and 3) the recognition of new instances of the concept
taught in the single-concept film.

Judges were used to evaluate student responses
on the first two tasks. They were guided by two rating
scales devised for that purpose.

The data were analyzed through two models: a
three. three-way univariate factorial analysis of vari-
ance, and a Pearson product-moment correlation. The
first model was used to determine the independence of

the criterion tasks and inter-judge reliability.

75

Kuder-Richardson Formula Twenty was used to establish
inter-item reliability in task three.

The chronological order of events in the ex-
periment was summarized through the description of four
administrations: 1) intelligence tests, 2) treatments,

3) immediate post-tests, and 4) delayed post-tests.

CHAPTER IV

ANALYSIS OF RESULTS

The first section of this paper presents the
findings of the analyses related to each of the hypo-
theses. Included are: relevant data from the analy-
ses of variance and the product-moment correlations.
the probability statement, and the statement of accep-
tance or rejection of the hypotheses. The second sec—

tion deals with the interpretation of the data.

Statement of Statistical Hypotheses

Statistical Hypotheses 1. In the relationship
between intelligence and method (e.g. the unique compon~
ent called interaction) there will be no difference
among the cell means for the immediate criterion per-
formances other than that which could be accounted for
independently by the main effects, intelligence and

76

77

method. regardless of the sequence of the concept in-

stances.

Statistipgl Hypothepes 2. When students are
taught concepts by means of the single-concept film,
there is no difference in means on the delayed criter-
ion performances between those taught through the dis—
covery method. and those taught by the expository method,
regardless of the sequence of concept instances in the

film. or of intelligence level.

Probability Statement

For the analysis of variance, the level of sig-
nificance for rejecting the statistical hypotheses was

set at the five percent level of confidence.

Analysis of Variance

On the basis of the five percent level of con-

fidence. no significant differences were found for the

78

interaction between intelligence level and method for
Hypothesis One nor were there any significant differ-
ences for retention due to competing methods as pro—
posed by Hypothesis Two. Tables 4.1, 4.2, and 4.3
summarize the respective analyses of the immediate
criterion performances: 1) generation of new ins-
tances of the concept, 2) application of the new
concept to the problem situations, and 3) recogni—
tion of new instances of the concept from a collection
of examples of several concepts.

The element of the data in Tables 4.1, 4.2, and
4.3 particularly pertinent to Hypothesis One is the in-
teraction of intelligence with method (AC). Although
no significant difference was noted, plotting the means
(figs. 4.1. 4.2, and 4.3) revealed a trend indicating
an interaction between intelligence and method. This
trend was consistent throughout all analyses. However.

its direction was opposite to what had been predicted

by the 7Directiona1. Hypothesis. The trend

79

indicated that bright students (with IQ's above 100)
learned best through the expository method. while the
below median students (with IQ's equal to or below 100)
learned best through the discovery method of programing
single-concept films. A fuller explanation of this is

found in the interpretation section of this chapter.

TABLE 4.1l

ANALYSIS OF THE IMMEDIATE CRITERION PERFORMANCE WITH
TWO LEVELS OF INTELLIGENCE: GENERATION OF NEW
INSTANCES OF THE CONCEPT

fl *-
-__ —

Degrees of Mean F Sta-

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 272.06 49.06 Reject
B. Sequence 1 4.63 .83
C. Method 1 4.13 .74

AB. 1 .01 .00
AC. 1 4.63 .83
BC. Interactions 1 8.23 1.48
ABC. 1 .70 .13
Error 272 5.55
* . . _
Reject if F > F 95(1, 272) — 3.89.

 

1For this table and for all ensuing analyses of
variance tables. the Sums of Squares are not included
since to do so would be superfluous. If the reader needs
this information it can be easily derived by multiplying
the Degrees of Freedom by the respective Mean Square.

:3,—

rn

80

TABLE 4.2

ANALYSIS OF THE IMMEDIATE CRITERION PERFORMANCE WITH
TWO LEVELS OF INTELLIGENCE: APPLICATION OF THE
CONCEPT TO PROBLEM SITUATIONS

—
_-

Degrees of Mean F Sta-

 

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 504.91 490.32 Reject
B. Sequence 1 13.73 2.46
C. Method 1 1.16 .21

AB. 1 12.01 2.15
AC. Interactions l 5.16 .92
BC. 1 9.66 1.73

ABC. 1 4.63 .83

Error 272 5.59
* ' ' = o .
Reject if F > F.95(1, 272) 3 89

ANALYSIS OF THE IMMEDIATE CRITERION PERFORMANCE WITH
TWO LEVELS OF INTELLIGENCE: RECOGNITION OF NEW
INSTANCES OF THE CONCEPT

 

 

 

Degrees Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 669.60 64.57 Reject
B. Sequence 1 2.23 .22
C. Method 1 2.60 .25

AB. 1 10.80 1.04
AC. Interactions 1 20.09 1.94
BC. 1 .60 .06

ABC. 1 5.43 .52

Error 272 10.37
*Reject if F > F = 3.89.

.95(1, 272)

 

6
5
4 (high IQ)
3
2
*——___________*
1
(low IQ)

0

Discovery Expository

Figure 4.l--A plot of imme-
diate Post-task I means for
the Intelligence—method in-
teraction.

 

30
25 W'M
(high IO)
20
Amy
15 ‘\\\\\\\\\*
10
(low IQ)
5
0 . L
Discovery Expository

Figure 4.3-—A plot of imme-

81

diate Post-taSk III means for

the intelligence-method in-
teraction.

 

12
10
8 i/(high IQ)
6 \
4
2 (low IQ)
0 . 1
Discovery Expository

Figure 4.2--A plot of imme-
diate Post-task II means for
the intelligence-method in—
teraction.

 

 

16 **~—-—~_____*
15
14 (Post-task III)
13
12
ll
10
9
8
7 #7 —*
6 (Post-task II)
5
4
3
2
l (Post-task I)
0 li
Discovery Expository

Figure 4.4--A plot of the
means for delayed Post-tasks
I, II. and III of the method
variable.

82

Hypothesis Two (directional) predicted that all
students. regardless of intelligence level or sequence
of the concept instances. would retain more, as measured
by the delayed criterion performances.2 when taught through
the discovery method. than when taught through the exposi-
tory method. No significant differences were found between
students taught through the discovery method when contrasted
with those taught by the expository method. Further, a plot
of the means (refer to Figure 4.4) revealed no consistent
trends. Tables 4.4, 4.5, and 4.6 summarize the results of
the analyses of variance on the delayed criterion perform—
ances. A more complete listing of the means, along with

the standard deviations. may be found in Appendix D.

Interpretation

Several factors suggest explanations for the re-

sults of these analyses. Among them are lack of precision

2The delayed criterion performances are the same
as the immediate criterion performances. The former were
given 3 weeks after administering the treatment.

TABLE 4.4

83

ANALYSIS OF THE DELAYED CRITERION PERFORMANCE WITH TWO

LEVELS OF INTELLIGENCE:

INSTANCES OF THE CONCEPT

r

—_‘
‘—

J

GENERATION OF NEW

 

 

Degrees of Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 317.16 52.97 Reject
B. Sequence 1 13.73 2.29
C. Method 1 16.51 2.76

AB. 1 1.16 .19
AC. Interactions 1 18.51 3.09
BC. 1 4.63 .77
ABC. 1 1.43 .24
Error 272 5.99
* ' ' = 3.89.
Reject if F > F.95(l, 272)
TABLE 4.5

ANALYSIS OF THE DELAYED CRITERION PERFORMANCE WITH
TWO LEVELS OF INTELLIGENCE: APPLICATION OF THE
CONCEPT TO PROBLEM SITUATIONS

 

 

Degrees of Mean F Sta-

Sources of Variance Freedom Square tistic Decision*
A. Intelligence 1 455.18 90.09 Reject
B. Sequence 1 5.43 1.07
C. Method 1 .60 .12

AB. 1 .43 .09
AC- Interactions 1 .03 .01
BC. 1 .18 .03

ABC. 1 1.03 .20

Error 272 5.05
*Reject if F > F = 3.89.

N
\l
N
V

.95(l.

 

k

tn

84

TABLE 4.6

ANALYSIS OF THE DELAYED CRITERION PERFORMANCE WITH TWO
LEVELS OF INTELLIGENCE: RECOGNITION OF NEW
INSTANCES OF THE CONCEPT

 

 

Degrees of Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 1064.70 80.48 Reject
B. Sequence 1 3.66 .28
C. Method 1 .51 .04

AB. . l 2.06 .16
AC. Interactions l 32.91 2.55
BC. 1 1.16 .09
ABC. 1 17.50 1.36
Error 272 12.91
*Reject if F > F = 3.89.

.95(1, 272)

 

due to the grossness of the intelligence levels used. the

possibility that the criterion tasks are not functioning

as expected. and imperfect experimental conditions.

Precision. The factors listed above would tend

to increase the estimate of the within variance by add-
ing to the truly systematic and instrumental based errors.

The within variance is, in essence, an estimate of the

sampling error which is a function of individual variance

in each cell p0pulation and sample size only. The above

factors. acting as sources of error. contaminate the

85

estimate of within variance, and the resulting increase
of the estimate of within variance decreases the size of
the F ration thereby producing F values smaller than the
true F values. Because of this condition there will be
situations in which the hypotheses should be rejected
but will not since the F value is reduced. That is, the
power of the analysis of variance is reduced.

In order to increase precision the data were re-
grouped in three ways: 1) the students were divided into
three intelligence levels, 2) the middle IQ group was
removed and the analysis run for the extreme groups. and
3) the subjects were divided on the basis of sex and the
data re—analyzed on this basis.

In the first regrouping of the data, the three
intelligence levels were: high (107-138), middle (93—106),
and low (63-92). This rearrangement resulted in the loss
of four subjects who were randomly selected out ixiorder
to maintain equal numbers of subjects per cell. The new
N equaled 276, and the n per cell equaled 23. The analy-
ses of variance for this regrouping is summarized in

Tables 4.7, 4.8, and 4.9.

86

TABLE 4.7

ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION PERFORMANCE
WITH THREE LEVELS OF INTELLIGENCE: GENERATION OF NEW

INSTANCES OF THE CONCEPT

 

 

Degrees of Mean F Sta-

 

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 2 178.22 34.51 Reject**
B. Sequence 1 2.84 .55
C. Method 1 3.71 .72

AB. 2 18.54 3.59
AC. Interaction 2 1.10 .21
BC. 1 7.67 1.48
ABC. 2 3.24 .63

Error 264 5.16

* ' ' = 3.89.

Reject if F > F.95(l, 264)
** ' ' = 3.04.

Reject if F > F 95(2' 264)

TABLE 4.8

ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION PERFORMANCE
WITH THREE LEVELS OF INTELLIGENCE: APPLICATION OF THE

CONCEPT TO PROBLEM SITUATIONS

 

W
Degrees of Mean F Sta-

 

Source of Variance Freedom Square tistic Decision*

A. Intelligence 2 331.05 65.17 Reject**
-B- Sequence 1 10.57 2.08

C. .Method 1 1.45 .29

AB. 2 2.82 .55

AC. Interactions 2 2.46 .48

BC. 1 11.36 2.24
ABC. 2 4.70 .93

Error 264 5.08
* . . = 3.89.
Reject if F > F.95(1. 264)
= 3.04.

** o -
Reject if F > F 95(2’ 264)

87

TABLE 4.9

ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION PERFORMANCE
WITH THREE LEVELS OF INTELLIGENCE: RECOGNITION OF NEW
INSTANCES OF THE CONCEPT

 

Degrees of Mean F Sta-

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 2 444.79 42.80 Reject**
B. Sequence 1 2.45 .24
C. Method 1 .06 .01

AB. 2 10.07 .97
AC. Interactions 2 3.53 .34
BC. 1 .36 .03
ABC. 2 7.35 .71
Error 264 10.39
* . . = 3.89.
Reject if F > F 95(1' 264)
** ' ' = 3.04.
Reject if F > F 95(2' 264)

 

It may be noted that there were no significant
differences found for the interactions between intelli-
gence level and teaching method as proposed by Direc-
tional Hypothesis One. and when intelligence was consid-
ered as a three level factor instead of two, the F values
decreased indicating that there was no interaction and
grossness of the categorization could not be blamed for
Wiping out the interaction between intelligence and method.

Plotting of the means for these analyses (Figures 4.5,

88

4.6, and 4.7) produced the same trend found in the first
analyses. That is, bright students learn best through
the expository method when taught by single-concept films,
While slow students do best when taught by the discovery
method. A complete listing of the means and standard de-
viations for the three regroupings of the data may be
found in Appendix D.

The delayed criterion performances were also re-
grouped into three IQ levels and an analysis of variance
run on them to determine retention effects. Tables 4.10.

4.11. and 4.12 summarize the results.

6 12
5 10
4 8 (High IQ)

3 ‘/(H:gh IQ) 6 \
2 4

 

 

(Low IQ)
0 . _ 0
Discovery Expository Discovery Expository
Figure 4.5--A plot of imme- ’ Figure 4.6--A plot of imme-
diate Post-task I means of diate Post-task II means of
the intelligence-method in— the intelligence method in-

.teraction. teraction.

89

 

30
25
20 *l_._—————-‘**
(High IQ)
15 *____~_~_‘-~*
10 (Low IQ)
5
0 .
Discovery Expository

Figure 4.7--A plot of immediate Post-task III means of
the intelligence-method interaction.

TABLE 4.10
ANALYSIS OF VARIANCE OF DELAYED CRITERION PERFORMANCE

WITH THREE LEVELS OF INTELLIGENCE: GENERATION OF
NEW INSTANCES OF THE CONCEPT

Degrees of Mean F Sta-

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 2 198.62 34.68 Reject
B. Sequence 1 10.18 1.78
C. Method 1 17.25 3.01

AB- 2 8.87 1.55
AC. Interactions 2 10.60 1.85
BC. 1 4.44 .77
ABC. 2 .33 .06
Error 264 5.73
* ' . - _
Reject if F > F 95(1’ 264) — 3.89.
**Reject if F > F = 3.04. .-

.95(2, 264)

 

90

TABLE 4.11

ANALYSIS OF VARIANCE OF DELAYED CRITERION PERFORMANCE

WITH THREE LEVELS OF INTELLIGENCE:

THE CONCEPT TO PROBLEM SITUATIONS

APPLICATION OF

 

 

 

Degrees of Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 2 318.55 71.06 Reject**
B. Sequence 1 5.80 1.29
C. Method 1 .93 .21

AB. 2 .72 .16
AC. Interactions 2 .79 .18
BC. 1 .06 .0]—
ABC. 2 .13 .03

Error 264 4.48

* ' ' > = 3.89.

Reject if F F 95(1’ 264)
** . . = .04.

Reject if F > F 95(2' 264) 3

TABLE 4.12

ANALYSIS OF VARIANCE OF THE DELAYED CRITERION PERFORMANCE

WITH THREE LEVELS OF INTELLIGENCE:
INSTANCES OF THE CONCEPT

RECOGNITION OF NEW

 

 

 

Degrees of Mean F Sta- . '
Source of Variance Freedom Square tistic DeClSlon*
A. Intelligence 2 612.71 48.77 Reject**
B. Sequence 1 1.60 .13
C. Method 1 .44 .03
AB. 2 15.96 1.27
AC. Interaction 2 2.14 .17
BC. 1 1.05 .08
ABC. 2 9.80 .78
Error 264 12.56
* ' ' = 3.89.
Reject if F > F.95(l, 264)
= 3.04.

**Reject if F > F

.95(2. 264)

91

In the analysis of the delayed criterion per-
formances there was no significant evidence to support
the directional hypothesis that method affects retention.
although it was noted that precision increased substan-
tially in Task I and moderately So in Task II (not sig-
nificantly so). Again, there was no consistent trend
for Hypothesis Two.

The second regrouping was similar to the one
above except that the middle intelligence level (93-106)
The results of this analysis are summarized

was excluded.

in Tables 4.13. 4.14, 4.15, 4.16, 4.17. and 4.18.

TABLE 4.13

ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION PERFORMANCE
WITH MIDDLE INTELLIGENCE LEVEL EXCLUDED: GENERATION OF
NEW INSTANCES OF THE CONCEPT

 

 

Degrees of Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 356.17 74.27 Reject
B. Sequence 1 .54 .11
C. Method 1 2.17 .45

AB. 1 23.67 4.94 Reject
AC. Interactions l 2.17 .45
BC. 1 11.50 2.40
ABC. 1 2.63 .55
Error 176 4.80
= 3.91.

*Reject if F > F

.95(1.

176)

92

TABLE 4.14

ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION PERFORMANCE
WITH MIDDLE INTELLIGENCE LEVEL EXCLUDED: APPLICATION OF
THE CONCEPT TO PROBLEM SITUATIONS

_ _
—

Degrees of Mean F Sta—

 

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 661.96 125.31 Reject
B. Sequence 1 5.22 .99
C. Method 1 2.40 .45

AB. 1 5.22 .99
AC. Interactions l 3.96 .75
BC. 1 4.57 .87

ABC. 1 8.27 1.56

Error 176 5.28
* ‘ I = .91.
Reject if F > F.95 (l, 176) 3
TABLE 4.15

ANALYSIS OF VARIANCE OF IMMEDIATE CRITERION PERFORMANCE
WITH MIDDLE INTELLIGENCE LEVEL EXCLUDED: RECOGNITION OF
NEW INSTANCES OF THE CONCEPT

#—
—__—

 

 

 

Degrees of Mean F Sta—

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 873.92 93.72 Reject
B. Sequence 1 1.57 .17
C. Method 1 .66 .07
AB. 1 .92 .10
AC. Interactions 1 5.92 .63
BC. 1 5.92 .63
ABC. 1 3.40 .36
Error 176 9.32
*Reject if F > F 95(1' 176) = 3.91.

 

93
TABLE 4.16
ANALYSIS OF VARIANCE OF THE DELAYED CRITERION PERFORMANCE

WITH MIDDLE INTELLIGENCE LEVEL EXCLUDED: GENERATION OF
NEW INSTANCES OF THE CONCEPT

 

 

 

Degrees of Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 396.20 77.98 Reject
B. Sequence 1 .35 .07
C. Method 1 17.04 3.35

AB. 1 5.57 1.09
AC. Interactions 1 19.57 3.85
BC. 1 1.76 .35

ABC. 1 .20 .04

Error 176 5.08
* ' ' = 3.91.
Reject if F > F.95(1, 176)
TABLE 4.17

ANALYSIS OF VARIANCE OF THE DELAYED CRITERION PERFORMANCE
WITH MIDDLE INTELLIGENCE LEVEL EXCLUDED:

APPLICATION OF

CONCEPT TO PROBLEM SITUATIONS

 

Degrees of Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 635.67 134.39 Reject
B. Sequence 1 7.04 1.49
C. Method 1 .02 .00

AB. 1 .02 .00
AC. Interaction l .35 .07
BC. 1 .20 .04

ABC. 1 .09 .02

Error 176 4.73
- 3.91.

*Reject if F > F

095(11

94

TABLE 4.18

ANALYSES OF VARIANCE OF THE DELAYED CRITERION PERFORMANCE
WITH MIDDLE INTELLIGENCE LEVEL EXCLUDED: RECOGNITION OF
NEW INSTANCES OF THE CONCEPT

 

Degrees of Mean F Sta-
Sources of Variance Freedom Square tistic Decision*

 

A. Intelligence 1 1180.20 103.43 Reject
B. Sequence 1 .78 .07
C. Method 1 .20 .02
AB. 1 20.89 1.83
AC. Interaction l 1.39 .12
BC. 1 9.59 .84
ABC. 1 4.26 .37
Error 176 11.41

*Based on a confidence of F equal to or greater than 3.91.

 

The only factor on the immediate Post-task prov—
ing significant, other than the expected intelligence
variable, was the intelligence—sequence interaction (AB).
This was only true for one of the tasks, that of genera—
tion of new instances of the concept (Table 4.13). A look
at the means for this analysis shows that the strength of
the interaction comes from the bright students. They did
Significantly better when the instances of the concept
were arranged in a random sequence. While the hypotheses

were not directly concerned with the variable of sequence

95

of instances of the concept, it was interesting to specu-
late on its causes.

Although the intelligence-method interaction (AC)
was not significant for the immediate Post—tasks, an in-
spection of the means (Figures 4.8, 4.9, 4.10) revealed
the same trend previously experienced: bright students
learn best by the expository method, while slow students
learn best through the discovery method.

On the delayed Post-tasks there was no signifi-
cant evidence indicating that method had an effect on
retention.

The third attempt at increasing precision was
by means of regrouping the data on the basis of sex.
This approach was selected because of the possibility
of systematic variance due to sex differences affecting
the error variance. Tables 4.19, 4.20, and 4.21 sum-

marize the results for males on the immediate criterion

performances.

96

6

5

4

3 */(h':9h IQ)
2

l *\\\\\\\\£::w IQ)
0 . L

 

Discovery Expository
Figure 4.8--A plot of imme—
diate Post-task I means of

the intelligence-method
interaction.

30
25

20

12
10
8 #I/
(high IQ)
6
4 \.
2 (low IQ)
O l L

 

Discovery Expository

Figure 4.9--A plot of imme-
diate Post-task II means of
the intelligence-method
interaction.

(high IQ)

l5. *F“\-I-N*

 

10 (low IQ)

5

0 . .
Discovery Expository

Figure 4.10--A plot of immediate Post—task III means of
the intelligence—method interaction.

97

TABLE 4.19

ANALYSES OF VARIANCE FOR THE IMMEDIATE CRITERION PERFORM-
ANCE FOR MALES: GENERATION OF NEW INSTANCES OF THE CONCEPT

Degrees of (Mean F Sta-

 

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 171.11 30.16 Reject
B. Sequence 1 35.11 6.19 Reject
C. Method 1 25.31 4.46 Reject

AB. 1 4.51 .80
AC. Interactions l 15.31 2.70
BC. 1 1.01 .18
ABC. 1 1.51 .27
Error 72 5.67

*R' ' > = . .

eject If F F 95(1’ 72) 3 98
TABLE 4.20

ANALYSES OF VARIANCE FOR THE IMMEDIATER CRITERION PERFORM-
ANCE FOR MALES: APPLICATION OF THE CONCEPT TO PROBLEM
SITUATIONS

Degrees of Mean F Sta-

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 127.51 21.64 Reject
B. Sequence 1 12.01 2.04
C. Method 1 1.01 .17
AB. 1 6.61 1.12
AC. Interaction l 23.11 3.92
BC. 1 .01 .00
ABC. 1 2.81 .48
Error 72 5.89
*Reject if F > F = 3.98.

.95(1, 72)

98

TABLE 4.21
ANALYSES OF VARIANCE FOR THE IMMEDIATE CRITERION PERFORM-

ANCE FOR MALES: RECOGNITION OF NEW INSTANCES OF THE
CONCEPT

Degrees of Mean F Sta-

 

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 277.51 31.04 Reject
B. Sequence 1 1.01 .11
C. Method 1 6.61 .74
AB. 1 4.51 .50
AC. Interaction 1 46.51 5.20 Reject
BC. 1 .01 .00
ABC. 1 10.51 1.18
Error 72 8.94
*Reject If F > F 95(1’ 72) — 3.98.

 

Of the immediate Post-tasks, only in post-task
three was there a significant difference obtained for
the interaction of intelligence and method (Table 4.21,
factor AC). However, it may be worth noting that in the
other two tasks the F statistic was much closer to sig-
nificance than in previous analyses.

The results of a similar analysis on the delayed
post—task scores produced highly significant differences
at the five percent level of confidence again for the

boys, but not for the girls. Tables 4.22 and 4.24 show

99

that method made a significant difference in the amount
boys retained for generation of new instances of the
concept and for recognizing new instances of the con-
cept. A look at the means (Figures 4.11, 4.12, and 4.13)
for these tasks reinforced the trend of bright boys doing
best through the expository programing of single-concept
films, while the less bright students did best When taught
through the discovery method. As the evidence in Table
4.23 indicated, the differences were not significant for
the second criterion task. Nor did the means support the

trend mentioned above.

TABLE 4.22

ANALYSES OF VARIANCE FOR THE DELAYED CRITERION PERFORMANCE
FOR MALES: GENERATION OF NEW INSTANCES OF THE CONCEPT

k

 

Degrees of Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 183.01 39.87 Reject
B. Sequence 1 43.51 9.48 Reject
C. Method 1 9.11 1.99

AB. 1 17.11 3.73
AC. Interactions 1 46.51 10.13 Reject
BC. 1 .31 .07
ABC. 1 .31 .07
Error 72 4.59
*Rejected if F > F = 3.98.

.95(1, 72)

100

TABLE 4.23

ANALYSES OF VARIANCE FOR THE DELAYED CRITERION PERFORMANCE

 

 

 

 

 

FOR MALES: APPLICATION OF THE CONCEPT TO PROBLEM
SITUATIONS
Degrees of Mean F Sta—

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 127.51 19.04 Reject
B. Sequence 1 6.61 .99
C. Method 1 4.51 .67

AB. 1 .31 .05
AC. Interactions 1 1.51 .23
BC. 1 .01 .00
ABC. 1 2.81 .42
Error 72 6.70
= 3.98.

* ' '
Rejected 1f F > F.95(1, 72)

 

TABLE 4.24

ANALYSES OF VARIANCE FOR THE DELAYED CRITERION PERFORMANCE

 

 

FOR MALES: RECOGNITION OF NEW INSTANCES OF THE
CONCEPT
Degrees of Mean F Sta-

Source of Variance Freedom Square tistic Decision*
A. Intelligence 1 418.61 31.18 Reject
B. Sequence 1 9.11 .68
C. .Method 1 1.01 .08

AB. 1 5.51 .41
AC. Interaction 1 94.61 7.05 Reject
BC. 1 9.11 .68
ABC. 1 21.01 1.57
Error 72 13.42
*Rejected if F > F = 3.98.

.95(1,

72)

101

 

6
5
.‘ll*
4 (high IQ)
3
2
1 *\(low IQ)
*
O . .
Discovery Expository

Figure 4.11--A plot of de-
layed Post-task I means of
the intelligence-method
interaction.

30
25

2O

 

 

12
10

8 #7 4%

4

(low IQ)
2
O I n
Discovery Expository

Figure 4.12—~A plot of de-
layed Post-task II means of
the intelligence-method
interaction.

*_,,Lifl~*r“*(high IQ)

15

10

*‘\\\~\“‘*(low IQ)

 

Discovery

Expository

Figure 4.13--A plot of delayed Post-task III means of
the intelligence-method interaction.

102

The possibility that sex differences were con—
tributing systematic variance to the estimate of the ran-
dom variance (within) has been supported by this analysis.
Male-female differences are in evidence. When the analy-
ses were run for females, the results washed out complete-
ly, but for males the F ratios immreased substantially
for the intelligence-method interaction on both immediate
and delayed performance tasks, although Task II of the
delayed tasks (Table 4.23 remained low. This increase
in F values indicated that, for males on immediate tasks
and for two of the three delayed tasks, the intelligence-
method interaction does exist, and is consistent with
the trends noted earlier. The small F values noted
earlier must have been due to the inclusion of the fe—
males. These findings suggest that method should be dif—
ferentially assigned to boys and girls. For girls it
doesn't seem to matter which method is used, but for
boys the choice of method appears to be critical.

This increase in F values resulting from the re—
moval of the systematic sex difference from the within

variance estimate caused the writer to believe that the

103

intelligence-method interaction was a true interaction
for males and would be very much in evidence in a prop—
erly controlled experiment treating males and females
as experimental variables.

A possible explanation for this differentiation
between males on the two IQ levels might have been due
to their attitude toward the teaching methods. Bright
students have been consistently rewarded by the exposi-
tory method. Thus, in moving from a procedure which had
provided past successes to one whose merits were unknown,
may have produced anxieties that reduced their learning
effectiveness. The less bright student has, relatively
speaking, been punished by the expository procedure. On
the other hand there was no reason for the discovery
method to be threatening to him. In fact, in his need
to gain some of the extrinsic rewards enjoyed by his
brighter counterpart, he may have been openly receptive
to the discovery method.

Pushing conjecture a little farther, a possible
reason for the trend washing out in the analysis for the

girls may have been due to the relative maturation of

104

verbal abilities. The girls, particularly at that age
level, had a more effective command of verbal processes
than did boys, and this facility may have hidden any
small effect attributable to an intelligence-method

interaction.

Criterion Tasks. There are implications in the

 

analyses of variance that the three criterion tasks:
generation of new instances of the concept, application
of the concept to problem situations, and recognition
of new instances of the concept may be measuring differ-
ent abilities. In Tables 4.22 and 4.24 there is a sig-
nificant interaction between intelligence and teaching
method, but in Table 4.23 this difference has washed
out despite the assumption that all three tasks were
measuring the same thing--concept attainment. To test
the possibility that these tasks might in fact be meas-
uring different things, correlations were run between
them. The findings are presented in Table 4.25.

It is noted that the correlations between the
tasks are small, substantiating the hypothesis that,

for the most part, these are different tasks requiring

105

different abilities for their completion. Such a con-
tention is supported when the squared values of the cor-
relation coefficient are noted. These values range from
.19 to .29, indicating that the amount of common variance
in the tasks is small (e.g. 19%, 19.4%, 29.5%) for task
correlations I and II, II and III, I and III respectively.
Further, one might think of the correlations as test—
retest reliability coefficients, and if one insists that
the tasks measure the same thing the criticism then be-
comes one of low reliability. The data matrix was fac-
tor analyzed in order to obtain a picture of the source
of these correlations and to determine, if possible, the
underlying structure of the instruments. The results of

this analysis are presented in Table 4.26.

TABLE 4.25

PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENTS:
BETWEEN IMMEDIATE CRITERION TASKS I, II. AND III
ACROSS ENTIRE SAMPLE

 

Task I Task II Task III
Task 1 .432 .440
Task II .504

Task III

 

106

TABLE 4.26

ROTATED FACTOR LOADINGS FOR ORTHOGONAL FACTOR ANALYSIS
OF THE MAJOR CORRELATION MATRIX

 

 

 

Variables Factor 1 FactOr 2 Factor 3
l .9499 .2187 .1174
2 .9308 .2244 .1292
Postwtask I 3 .9604 .2123 .1188
4 .9668 .2193 .1277
5 .9630 .2256 .1078
6 .2037 .9120 .1249
7 .2139 .8880 .1547
Post-task II 8 .2163 .9283 .1129
9 .2248 .9631 .1406
10 .2057 .9576 .1344
Post—task III 11 .2763 .3384 .8993

 

The relative independence of the three tasks is
again in evidence when one notes the loadings on the
three resulting factors. Factor 1 might be called a
Task I factor, since all measures loading on this factor
involve Task I. Task II and Task III measures have very
low loadings on factor 1. Similarly, measures loading
on factor 2 involve TaSk II, and measures loading on
factor 3 involve Task III. The existing structure.

therefore, closely resembles a simple structure case

107

(all dependent tasks).3 It was suspected, then, that
the existing correlations might be due to intelligence
and not truly be a function of the tasks. Knowing that
the correlations increase or decrease due to increased
number of IQ levels, the same analysis was run for three
separate intelligence groupings (high, middle, and low).
Such an analysis should indicate whether or not the cor-
relation in evidence is due simply to IQ difference in
people (e.g. concomitant factor) or due to a true rela—
tionship in the measure. If the former were true, the
correlation would decrease when one analyzes within IQ
groupings as contrasted with across IQ groupings. If
the latter were true, the correlation would increase or
stay the same. 'In each case the correlations (Tables
4.27, 4.28, and 4.29) dropped and the factor structure
tended even more toward simple structure. This tendency
was very obvious in the case of low intelligence (Table
4.30), although less so for middle and high intelligence
(Tables 4.31 and 4.32 respectively). The same results were
found for an analysis of the delayed post-tasks.

3Lee J. Cronbach, Essentialsfiof Psychological

{Testing (New York: Harper & Brothers, 1960). Pp. 255-
256.

108
TABLE 4.27
MINOR PEARSON PRODUCT-MOMENT CORRELATION MATRIX OF

IMMEDIATE CRITERION TASKS I, II, AND III FOR
LOW INTELLIGENCE

 

 

 

 

Task I Task II Task III
Task I .0652 .1812
Task II .0838
Task III
TABLE 4.28

MINOR PEARSON PRODUCT-MOMENT CORRELATION MATRIX OF
IMMEDIATE CRITERION TASKS I, II, AND III FOR
MIDDLE INTELLIGENCE

 

 

Task I Task II Task III
Task I .3807 .3264
Task II .4529
Task III
TABLE 4.29

MINOR PEARSON PRODUCT-MOMENT CORRELATION MATRIX OF

IMMEDIATE CRITERION TASKS I, II. AND III FOR
HIGH INTELLIGENCE

 

 

Task I Task II Task III
Task I .2309 .3089
Task II .3771

Task III

109

TABLE 4.30

ROTATED FACTOR LOADINGS FOR ORTHOGONAL FACTOR ANALYSIS

OF THE MAJOR CORRELATION MATRIX LOW IQ
IMMEDIATE POST-TASKS

 

 

 

 

Variables Factor 1 Factor 2 Factor 3
l .9564 .0421 .0443
2 .8414 .0873 .1109

Post-task I 3 .9699 .0475 .0340
4 .9946 .0622 .0656
5 .9817 .0521 .0459
6 .0100 .9511 .0865
7 .0389 .8749 .0588

Post-task II 8 .0205 .9494 .0955
9 .0180 .9983 .0183
10 .0103 .9843 .0063

Post-task III 11 .1284 .0844 .9857

TABLE 4.31

ROTATED FACTOR LOADINGS FOR ORTHOGONAL FACTOR ANALYSIS

OF THE MAJOR CORRELATION MATRIX OF MIDDLE IQ ON
IMMEDIATE POST-TASKS

 

 

Variables Factor 1 Factor 2 Factor 3
1 .9501 .2248 .0995
2 .9484 .1297 .0771
Post-task I 3 .9722 .1550 .0944
4 .9795 .1763 .0931
5 .9702 .1902 .0864
6 .1763 .8846 .0718
7 .1088 .8869 .0999
Post-task II 8 .2051 .8806 .1597
9 .1797 .9743 .1232
10 .1875 .9506 .1484
Post-task III‘ '11 .1958 .2924 .9354

 

110

TABLE 4.32

ROTATED FACTOR LOADINGS FOR ORTHOGONAL FACTOR ANALYSIS
OF THE MAJOR CORRELATION MATRIX OF HIGH IQ ON
IMMEDIATE POST-TASKS

 

 

Variables Factor 1 Factor 2 Factor 3
1 .9694 .1191 .0801

2 .9537 .1244 .0998

4 .9886 .1092 .0981

5 .9847 .1312 .0598

6 .1262 .8840 .0556

7 .1428 .8832 .1168

Post-task II 8 .1127 .9398 .1231
9 .1399 .9820 .1120

10 .0936 .9751 .1064

Post-task III 11 .2208 .2622 .9388

 

Hence, it is not unreasonable to assume that the

measures are independent, and while this fact has no di-

rect bearing upon the purposes of this study (other than
to increase precision) the development of three indepen-
dent tests of concept formation represents an original

and important research finding, and should be pursued

extensively in another context.

A second question that comes to mind concerns the

validity of testing outcomes of a treatment depending on

111

visual stimuli with a purely verbal device. Schalock
and his associates suggest that the results of a meas-
urement would be more meaningful if a filmic treatment
were also tested with film. There are apparent diffi-
culties in this approach for the criterion task of gen-
erating new instances of the concept, but certainly
Tasks I and II of this study could be presented through

film.

Treatment Conditions. The question arises,
were there aspects of the treatment conditions that
might account for the unexpected results? Some aspects
which seem to bear consideration are: running time of
film, number of times film was viewed, the number of
instances of the concept, the arrangement of the in-
stances, validity of the concept, and variables not
controlled for.

The film has a running time of approximately

four minutes. Is this too short an exposure for teaching

 

4H. D. Schalock, et a1., Motion Pictures as Test

§timuli: An Application of New Media to the Prediction

9:,Comp1ex Behavior, A Final Report, Title VII, Project
No. 971, Dec. 1964.

112

this rather complex concept? In conjunction with this
question, would it have been better to have shown the
film to each subject more than once? Hoban and van
Ormer5 cite evidence that there is a tremendous learn-
ing advantage accrued from repetitive showings of a
film. Researchers have measured increases in compre-
hension and retention for extra showings of a film that
range as high as 35% for one extra showing, 42.4% for
two extra showings, and 43.5% for three extra showings.
The researcher of this study was not unaware of these
facts at the initiation of the research; the decision
to show the film only once was made to reduce the number
of contaminating variables. In retrospect it might have
been wiser to have set up a control group where the ef-
fect of multiple showings could have been observed.

The four minute treatment film, in both the ran-
dom and ordered sequence versions, was made up of ten

instances of the concept adaptation. The question in

 

5Charles F. Hoban, Jr., and E. B. van Ormer. In:
structional Film Research 1918-1950. (Rapid Mass Learn-
ing). A Report of the Pennsylvania State Collge Project
jointly sponsored by the Department of the Army and the
Department of the Navy (Port Washington, N. Y.: U. S.
Navy Special Devices Center, 1951), pp. 34-35 (chap. 8).

113

this case asks about the density of information. Perhaps
an analysis of the visual and auditory information in the
film would reveal that the "seven bits" limitation6 has
been exceeded. Travers,7 in his support of the theory
that the human receiver is capable only of receiving in-
formation through one sense modality at a time, states
that when information is presented to two senses at the
same time, as is the case in the sound motion picture,

the receiver is constantly switching back and forth be—
tween the visual and aural channels. He cites research
under his sponsorship which indicated that "where there

is an auditory and a visual input of information of rela-
tively high information density (in relation to what the
processing system can handle), then the switching process

itself may interfere with the taking in of information."8

 

6George A. Miller, "The Magical Number 7, plus
or minus 2: Some limits on our capacity for processing
information, Psychological Review, LXIII (1956), pp. 81—
97.

7 . .
Robert M. Travers, ”The Transm1531on of Infor—
mation to Human Receivers," AV Communication Review,

81bid., p. 376.

114

Such evidence defends the thought that the present re-
search design might have been improved by the control
of another variable where the number of instances of
the concept in the film would be varied While keeping
the showing time constant.

A factor worthy of consideration is the concept
itself. Adaptation was chosen for a number of reasons,
some of them mentioned earlier: it was a concept norm-
ally taught in the time period in which the students
were available for treatment, teacher judgment and a
pretest indicated that the students did not possess it,
and there was a wealth of film present containing in-
stances of the concept. In selecting the instances of
the concept there was an attempt to get novel examples.
It is wondered if perhaps the unusualness of these in—
stances might not have detracted from the teaching objec-
tive, particularly among the discovery treatments. An—
other question about the concept might be in reference
to its prerequisites. Did the individual student have
in his repertoire the necessary subconcepts to allow him

to arrive at the concept of Adaptation?

115

Last, it is possible that the effect of practice
may be a variable confounding the results of this study.
Nearly all of the formal learning of the subjects has
been of an expository nature. They expect to receive
verbal statements and demonstrations of concepts, prin-
ciples, and processes. They do not have a comparable
set for digging out unknowns from an array of data.
Students are used to being told what they are going
to see, what they see, and what they have seen. Their
experience in viewing films has conditioned them to ex—
pect that points of major importance will be repeated
several times, thus allowing them to ignore much of a
film's content. In a film, such as the experimental
film used in this study, each of the segments contribute
to the learning of the concept, and ignoring large por-
tions of it could logically distort the learning effect.
It seems reasonable that a preliminary training period
in discovery for the subjects could have affected the

outcomes of this study.

116

Summary

Chapter Four consisted of two major parts. The
first section presented the findings of the analyses for
each of the hypotheses. These findings included: rele-
vant data from the analyses of variance and correlations,
the probability statement, and statements of acceptance
or rejection of the hypotheses. The second section
dealt with the interpretation of the data.

At the .05 level of confidence no significant
differences were found for the interaction between in-
telligence level and teaching method in support of the
first hypothesis, nor was there a significant difference
between methods in support of the second hypothesis.
Plotting the means revealed a trend contrary to Direc-
tional Hypothesis One of this study. The trend indicated
that bright students (IQ > 100) learned best through the
expository method of programing of single-concept films,
while the less bright (IQ Z 100) learned best through
the discovery method of programing.

In the interpretation section, several factors

were examined and data were regrouped in an attempt to

117

discover the causes underlying the results of the analy—
ses. Various aspects of precision, independence of cri-

terion tasks. and experimental conditions were examined.

Precision. In order to increase precision, the

 

data were regrouped on intelligence in three ways.

First, the students were regrouped on three in—
telligence levels in order to gain precision on this
factor and the data were re-analyzed. Again, no sig—
nificant differences were found, but the F values did
rise appreciably. Plotting of the means revealed the
same trend seen in the original analyses; bright students
learned best from the single—concept film when it was
programed with the expository method, while the less
bright learned best through the discovery method of
programing. A similar analysis of the delayed criter-
ion tasks failed to produce significant evidence of
method affecting retention..

Second, an analysis was run where the middle
group of the three intelligence groups was excluded in
order to observe the effect of the treatments on the

extreme groups. Except for a significant difference

118

between intelligence and sequence of concept instances
(order vs. random), the results were the same as for
the three intelligence groupings above. The means in-
dicated that bright children do significantly better
When the instances of the single-concept film are
arranged in a random sequence.

Third, the subjects were regrouped on the basis
of their sex because of the possibility of systematic
variance, due to sex differences, affecting the error
variance. On the immediate Post-task III (recognition
of new instances of the concept) significant differences
were found for the intelligence-method interaction, for
males only. And in the other two immediate post-tasks.
for the same interaction, the F values were close to
significance. An analysis of the delayed post-tasks
produced comparable results. The trends from previous
analyses were supported there; that is, bright boys
learned best through the expository method, while the
less bright learned best through the discovery method
of programing single-concept films. The conclusion

drawn from these analyses is that the possibility that

119

sex differences were contributing systematic variance
to the estimate of the within variance has been supported.
The findings indicate that method should be differentially

assigned to boys on the basis of their intelligence level.

Criterion Tasks. Correlations were run between
the three criterion tasks in order to test the possibil-
ity that they were not measuring the same thing. The
small correlations between the tasks supported the pos-
sibility that the tasks were indeed measuring different
things. A factor analysis of the data matrix gave further
support to this contention.

Similar correlations and factor analyses were run
separately for the three intelligence levels (high, middle,
and low). In each case the correlations dropped and the
factor analyses tended toward simple structure indicating
that the correlation results were due to IQ differences
in the subjects. It was noted that, while the discovery
of this fact resulted from an attempt to increase pre-
cision, the development of three independent tests for
the formation of a concept is a valuable research finding

and should be pursued extensively in another context.

120

Treatment Conditions. Other factors considered
as possible causes of the results were: limited running
time of the single-concept film, number of repetitive
instances in the film, the question of prerequisite con-

cepts. and the effects of practice.

CHAPTER V
SUMMARY. CONCLUSIONS. AND RECOMMENDATIONS
Summary

The summary is composed of three parts: the
problem, the design and procedure of the study, and the

analysis.

The Problem. This study was designed to analyze
the teaching effects of the single-concept film when pro-
gramed to teach science concepts through two contrasted

teaching modes: expository and discovery.

The design and Procedure of the Study. The sample
consisted of 280 junior high school students selected from
nine, eighth grade General Science II classes from Patten-
gill Junior High School in Lansing, Michigan.

The plan of this study was to teach the concept
”Adaptation" by means of the single-concept film. This
single-concept film was shown to the subjects in four

121

122

modes: 1) with the instances of the concept in an ordered
sequence using the discovery narration, 2) with the in-
stances of the concept in a random sequence using the
discovery narration, 3) with the concept instances in

an ordered sequence using the expository narration, and

4) with the concept instances in a random sequence also
using the expository narration.

Subjects were randomly assigned to treatments
after an initial separation on the basis of high and low
intelligence, using an IQ of 100 as a dividing point.

Three criterion tasks were devised to determine
the degree of concept attainment: l) the generation of
new instances of the concept taught by the film, 2) the
application of the new concept to problem situations,
and 3) the recognition of new instances of the concept
taught in the single-concept film.

Three judges independently evaluated student re-
sponses on the first two tasks. They were guided by two
rating scales devised for that purpose. The third task
was objective in nature (true-false) and was machine scored.

Two procedures of analysis were employed: three.

three-way univariate factorial analyses of variance, and

123

Pearson product—moment correlations. The analyses of
variance were used to test the hypotheses under consid-
eration. Correlation coefficients were calculated in
order to determine the independence, or non-independence
of the criterion tasks and inter-judge reliability.
Kuder—Richardson formula 20 was used to establish

inter-item reliability in Task III.

Analysis. The two statistical (null) hypotheses
were:
1. There is no differential effect for the interac-
tion of intelligence level and teaching method
when science concepts are taught through the

single-concept film, regardless of the sequence

of the concept instances.

2. Method does not have an effect on the retention
of science concepts taught by the single-concept
film as measured by the delayed criterion tasks,

regardless of IQ level, and sequence of concept

instances.

The directional research hypotheses were:

124

1. There is a differential effect for the interac—
tion of intelligence level and teaching method
when teaching science concepts through the
single-concept film regardless of the sequence
of concept instances. These interactions take
two forms: bright students (IQ > 100) perform
best on the criterion tasks when the concept is
taught through the discovery method, and less
bright students (IQ 5 100) perform best on the
criterion tasks when the concept is learned

through the expository method.

2. The retention of science concepts as measured by
delayed criterion tasks is greater for all students
taught through single-concept films using discovery
method programing, than for those taught by single-
concept films using expository programing, regard-
less of the IQ level and the sequence of concept

instances.

There were no significant differences found for

either of the hypotheses. An inspection of the means

125

indicated a consistent trend contrary to the first (di-
rectional) hypothesis. The trend indicated that bright
students learn best when taught through the expository
method, and the less bright students learn best through
the discovery method.

Operating under the assumption that there should
in fact be significant main and interaction effects, a
reason for the non-significant F values was sought. If
in fact there exist such causes, their effect would have
to be one of increasing the within sum of squares, there—
by decreasing the power of precision of the F statistic.
Two possible sources of non-random error were considered.
First, it was thought possible that the categories of the
intelligence factor were too grossly defined. Second, the
investigator was concerned with the possibility that sex
differences might exist and thereby be contributing syste-
matic differences which were thrown into the error term.
Following this line of thinking the data were regrouped
in three ways: 1) the analyses were re-run with three
intelligence levels (high, middle, and low), 2) the an-

alyses were re-run with the middle group excluded, and 3)

126

the analyses were re-run with subjects divided on the
basis of sex with separate runs for males and females.

The first two data regroupings did not obtain
significant differences for the treatments, but in sev-
eral cases the F value increased appreciably and the
trend noted in the original analyses was repeated. The
third regrouping, on the basis of sex, proved more fruit-
ful. While there was no evidence indicating that assign—
ment of method was important for girls, it produced sig-
nificant differences for the boys. An inspection of the
means indicated that this differential effect was in the
same direction as the trend mentioned earlier; bright
boys learn best through the expository method, and the
less bright learn best through the discovery method.
These findings support the investigator's contention
that sex differences do exist and have had a suppress-
ing effect on the F values.

Pearson product—moment correlations were run be-
tween the criterion tasks in order to test the possibility
that the tasks were not measuring the same thing. The
correlations (over total N) between the tasks were low,

giving further support to the idea that the tasks were

127

measuring different abilities. A Factor analysis of
the measures supported these findings. The same analysis
run within IQ groupings indicated an even stronger tend-

ency toward simple structure.

Conclusions

The conclusions of this study. within those limi-
tations enumerated in Chapter I, are presented in two

parts: variable control, and the hypotheses.

Variable Control. In several.cases a clear ra—
tionale was made about control procedures for variables
affecting the study. However, in some cases study results
raised a question about how this control might be modified
to improve the study. In other instances variables were
not seen as crucial to results and were largely uncon-
trolled, but outcomes of the study indicated a more cru—
cial role for a number of them. Variables that fall rea—
sonably into these two categories were: 1) effect of
practice, 2) number of film showings per student, 3)

relative shortness of the film. 4) number of instances

128

of the concept of the film, 5) concept prerequisites.
and 6) the novelty effect of the concept instances.

It is possible that the students lack of train-
ing in learning by the discovery method affected the
outcomes of this study.

There is evidence that showing the film twice
increases retention and understanding of the film content
as high as 35 percent. Such an increase might have also
affected this analysis.

The experimental film used in this study is ap-
proximately four minutes long. There is a question as
to whether this is enough exposure to insure learning of
a concept as complex as Adaptation.

A large number of instances of the concept were
included in the relatively short four minute film. It
is possible that examples with more extensive development
of the few would have had a more desirable effect.

There is also the need to answer the question
of what concepts must be present in the student’s reper-
toire in order for him to grasp the concept being taught.

Novel examples of the concept were chosen in the

hopes of emphasizing the distinguishing elements of the

129

concept. Perhaps, the very novelty of the instances
tended to distract the student from the instructional
objective.

Finally, it is not clear just what abilities
the criterion instruments are measuring. A surface,
content inspection of the tasks leads one to believe
that the tasks are measuring the same ability. But
statistical evidence indicated that the three tasks
are independent measures, suggesting that more than
one ability is involved in concept formation. This
important outcome should be researched thoroughly in
a context where the major emphasis is on the formation

of concepts.

The Hypotheses. It must be concluded from the
original analyses of this study that there is no differ—
ential effect of intelligence and teaching method result-
ing from teaching science concepts through the single-
concept film, nor is the retention of science concepts
as measured by delayed criterion tasks, when taught
through the single-concept film, significantly affected

by the differential assignment of the two methods.

130

Post hoc consideration of the data gives support
to a trend, indicated by the original analyses, that
bright students learn best through the expository method.
while the less bright learn best by the discovery method.
Statistically significant evidence of this trend, for

boys, existed for one of the post-tasks.

The conclusions and recommendations that follow
are intended to aid those persons continuing research in
the development of guidelines for the programing and use

of the single-concept film.

Recommendations

It is suggested that in replicating this study
an attempt should be made to control more factors.

Changes recommended include the following:

1. Students should be given practice in the discov—

ery method prior to administering the treatment.

2. The film should be shown to each student more
than once, under controlled conditions to de—

termine the effect of multiple showings.

 

 

10.

131

The running time of the film should be varied.

The number of instances of the film should be

varied.

The concept should be taught with more common-
place examples in order to reduce the novelty

effect that might be present.

Study is needed to determine the subconcepts

required by the student prior to his attainment

of the concept.

Intelligence should be divided into a larger

number of categories.

The effect of differential assignment on the basis

of sex should be studied.

Concepts in other areas should be taught to in—

crease generalizability of the findings.

The trend of bright students learning best through
the expository method of programing single—concept
films, and slow students learning best by the dis-

covery method needs to be explored in depth.

132

Finally, in considering these conclusions and
recommendations in the light of the broad purposes out-
lined in Chapter I of the study, it would seem that this
work can be regarded as a first step toward developing
guidelines for the programing and use of single—concept

films. While this study has not provided immediate sup—

 

port for the use of the single-concept film in the field.
it has indicated a number of promising directions for

further research.

APPENDIX

APPENDIX A

PRETESTS USED IN A PILOT STUDY TO DETERMINE THE
STUDENTS' KNOWLEDGE OF THE CONCEPT "ADAPTATION”

PRETEST I

Below are a number of statements, if you agree with a
statement encircle "A" on your answer sheet. If you
disagree, encircle "D" on your answer sheet.

EXAMPLE: 0.1 All horses are black.
On your Answer Sheets: 0.1 A @

l. The bodies of many animals will change if their
environment changes.

2- The plants and animals that we see are basically
unchanged over the thousands of years.

3- There are plants that catch insects and use their
bodies for nourishment.

4. The ancestors of birds. have always had wings.

5. Men, living millions of years in the future, may
100k a great deal different from men of today.

6. One of the reasons for breathing is to remove use-
less gases from the body.

7- The main purpose of the circulatory system is to
keep the body warm.

133

134

8. Some animals are capable of growing two headS.
9. All plants produce seeds.

10. If examples were here before us, it would be easy

to recognize the ancestor of the robin that lived
a million years ago.

11. The offspring of animals who have been subjected
to x-rays may be quite different from those off-

spring of animals who have not been treated with
x-rays.

12. The blood in humans is exactly the same for all
individuals.

13. It is possibly true that at one time the bodies
of men were completely covered with hair.

14. In the future, there may be human diseases that we
have never had in the past.

15. Dinosaurs can no longer be found on earth because
they could not change fast enough to match the
changes in their environment.

PRETEST II

Below are a number of examples arranged in groups of three.
The examples in each of these groups are expressing one
main idea or concept. It is your task to determine what
concept best describes each of these groups of three. The
same concept may be applied to more than one group if you
are convinced that the examples in the groups refer to

the same thing. Study the following three examples care-
fully:

135

a. Sugar placed in water seems to disappear.

Qiffiggigg 0.1 b. The girl's perfume could be smelled all
' the way across the room.

c. The red dye slowly spread to all parts
of the container.

a. The ameoba divided into two smaller
ameobas.

Reproduction 0.2 b. The uniting of the sperm and the egg
cell formed a seed.

c. The small piece of potato grew into
a new potato plant.

a. The eggs laid by the ant were fertil-
ized.

Reproduction 0.3 b. The willow branch stuck into the soil
of the riverbank began to grow.

c. A shoot and root began to grow from
the covered end of the rose bush branch.

Write the best answer you can think of for the following
suggestions, and place your answer on the answer sheet.
Please do not write on this test.

a. Flounders have developed the unusual ability
of changing their color to match their sur-

roundings.

l. b. Bats of today, unlike their ancestors of a
few thousand years ago, can find their way
in the dark by sending out sound waves ahead

of them.

c. Certain ants have learned to keep another in—
sect type, the aphid, a prisoner. The ants per—
iodically remove a sweet liquid from the aphids.

136

There are some animals in which bleeding is
difficult to stop.

When the 02 content in swamp water drops to
a certain level, the fish in the swamp die.

Cancer occurs in all organisms.

Whenever Bill eats eggs. his temperature rises.

During the flowering time of golden rod, Jane
sneezes constantly.

It is difficult to make a skin graft, from
one person, grow on another.

A crayfish is able to lock his claw into place
when grasping an enemy, and then deliberately

breaks off his own claw. He escapes while the
enemy tries to break loose the locked claw.

The Snyr bird has over many generations stopped
building nests. Instead, they lay their eggs
in the Albatross‘ nest, who raises the baby
Snyr birds as her own.

When the climate changed to a colder one, some
of the bears survived because they were able
to slow their bodily processes down and sleep
through the winter.

The tadpole develops legs and loses his tail
and gills.

The butterfly that breaks out of the pupa is
nothing like the caterpillar which looked it-

self in.

Mosquito larva live under water. but the flying
adult mosquito would drown if he were placed
under water.

137

'Mushrooms get their nourishment from dead

and living plants.
Tapeworms live in the intestines of animals.

Athletes foot is caused by a tiny fungus
which gets its nourishment from animals.

When leaving bright sunlight and going into
a darkened room, it is difficult to see but
after a few minutes full vision is restored.

At the end of fall, the brown coat of the
snowshoe rabbit begins turning white. It

remains white throughout the winter.

Man is able to be heard much farther than

his voice normally carries, by using a tele—

phone.

Robins who have never seen other robins' nests
will build their nests in the same way.

Certain species of pigeon, when let out in un-
known territory, miles from their home, are
able to fly directly to their home.

Each species of spider spins a particular web
pattern. '

Bees, while drawing nectar from flowers, carry
pollen from one flower to another.

A brisk wind can fill the air with pollen.

Corn tassels, which contain pollen. grow above
the ears of corn. The pollen falls from the
tassels and some of the pollen lands on the

corn silk below.

138

a. Termites and certain tiny protozoa have
learned to live together. The protozoans
which live in the termites' stomach digest
the wood that the termite eats. The termite
cannot digest the wood by himself.

10. b. The feet of young boys are very tender when
they first begin going barefoot in the spring.
but in a short time a thick protecting callus
grows over the bottoms of their feet.

c. The skunk of today is somewhat different from
the skunk of several thousand years ago. He
is different because he possesses a gland
that his ancestors did not have. This gland
produces a very disagreeable odor that serves
to discourage animals that would attack the
skunk.

 

PRETEST III

Most, but not all, of the following examples can be grouped
under five headings or concepts: Instinct, Allergy, Meta-

morphosis, Adaptation, and Paracitism. Read each item and

mark each on your answer sheet.

Allergy a EXAMPLE: 0.1 Sugar placed in water
Adaptation b seems to disappear.
Instinct c

Metamorphosis d

Paracitism e On Answer Sheet:

None of these f

0.1 a‘b c d e(:)

l. Flounders have developed the unusual ability of changing
their color to match their surroundings.

2. There are some animals in which bleeding is difficult
to stop.

10.

11.

12.

13.

14.

15.

139

Whenever Bill eats eggs, his temperature rises.

A crayfish is able to lock his claw into place when
grasping an enemy, and then deliberately breaks off
his own claw. He escapes while the enemy tries to
break loose from the locked claw.

The tadpole develops legs and loses his tail and gills.

Mushrooms get their nourishment from dead and living
plants.

When leaving bright sunlight and going into a darkened
room. it is difficult to see but after a few minutes
full vision is restored.

Rdbins who have never seen other robins' nests will
build their nests in the same way.

Termites and certain tiny protozoa have learned to
live together. The protozoans which live in the ter-
mites' stomach digest the wood that the termite eats.
The termite cannot digest the wood by himself.

The skunk of today is somewhat different from the skunk
of several thousand years ago. He is different because
he possesses a gland that his ancestors did not have.
This gland produces a very disagreeable odor that serves
to discourage animals that would attack the skunk.

Corn tassels, which contain pollen, grow above the ears
of corn. The pollen falls from the tassels and some of
the pollen lands on the corn silk below.

Each species of spider spins a particular web pattern.

Man is able to be heard much farther than his voice
normally carries, by using a telephone.

Athletes foot is caused by a tiny fungus Which gets
its nourishment from animals.

 

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

140
Mosquito larva live under water, but the flying adult
mosquito would drown if he were placed under water.
When the climate changed to a colder one, some of the
bears survived because they were able to slow their

bodily processes down and sleep through the winter.

It is difficult to get a skin graft taken from one _
person to grow on another. '1

Cancer occurs in all organisms.

 

Certain ants have learned to keep another insect i
type, the aphid, a prisoner. The ants periodically
remove a sweet liquid from the aphids.

Bats of today, unlike their ancestors of a few thous-
and years ago, can find their way in the dark by send-
ing out sound waves ahead of them.

When the 0 content in swamp water drops to a certain
level. the fish in the swamp die.

During the flowering time of golden rod, Jane sneezes
constantly.

The Snyr bird has, over many generations, stopped
building nests, instead they lay their eggs in the
Albatross' nest, who raises the baby snyr birds as
her own.

The butterfly that breaks out of the pupa is nothing
like the caterpillar which locked itself in.

Tapeworms live in the intestines of animals.

When leaving bright sunlight and going into a dark-
ened room, it is difficult to see but after a few
minutes full vision is restored.

Certain species of pigeon, When let out in unknown
territory, miles from their home, are able to fly
directly to their home.

141

29. A brisk wind can fill the air with pollen.

30. The feet of young boys are very tender When they

first begin going barefoot in the spring, but in
a short time a thick protecting callus grows over
the bottoms of their feet.

 

APPENDIX B

SCRIPTS FOR THE DISCOVERY AND EXPOSITORY
NARRATION OF THE FILM

Script I - Discovery

The ten examples in this film are connected by
an idea that is common to all of them. It is your task
to discover the idea that connects all of these examples.
Occasionally, I will direct your attention to facts about

the examples in this film. Listen and watch carefully.

1.

Scientists believe that at one time baby deer did
not have White spots on their coats. Furthermore, by the
time this fawn is one year old, all of his spots will have
disappeared. Would it be an advantage to him if he could

keep his spots?

142

143

Would it make a difference to this giraffe if his
neck and tongue were even longer? If a giraffe depends
upon the foliage of trees for their food, what do you sup-
pose has happened to giraffes whose necks and tongues were

much shorter?

The ancestors of the fish you are about to see
were trapped in this cave thousands of years ago. Unlike
his ancestors, this fish has neither pigment to protect
him from the sun, nor eyes with which to see, but he does
have something they did not. He has tiny growths on his
head with which he can sense vibrations made by other ani—
mals moving in the water. Would he be better off, in his

cave environment, if he were exactly like his ancestors?

The nitrogen in the soil, in which the Venus Fly

Trap plant grows, was used up by similar plants long ago.

But the Venus Fly Trap plant is able to get the nitrogen

144

it needs from the bodies of the insects it captures.
What do you suppose happened to the relatives of this

plant who did not have the ability to catch insects?
5.

Sometime in the past, the water animals that the
Archer Fish killed for food died out. By chance, some
of the Archer Fish had a groove in their upper jaw that
enabled them to fire tiny pellets of water at insects
above the surface of their pond. What do you think hap—

pened to those Archer Fish who were hatched without this

groove?

The small fish attached to the shark is called a
Remora. One of his upper fins has a suction disc on it.
By attaching this disc to the shark he gets free trans-
portation, and protection from his enemies. He also
shares the shark's food. Fossil remains of the Remora's

ancestors show that they lacked this suction disc. Which

of the two were better off?

145

The peculiar looking yellow fish is called the
Angler Fish. Fossils show that over several years the
Angler has developed a fleshy growth on his head that 4

greatly resembles a fishing worm. He wiggles it entic-

“

ingly at each passing fish. Sometimes their curiosity

gets the best of them.

The Spade-Foot Toad depends upon insects for its
food, and during dry periods these insects die out. To
carry him over this dry spell, nature has developed in
the toad the ability to slow down all of his bodily pro-
cesses. As a result he uses up very little energy. He
simply digs into the protecting soil and goes to sleep

until the rain and the insects return.

The Hermit Crab has one serious weakness, his
soft abdomen, or stomach, is unprotected. To correct

this weakness he pushes his soft abdomen into an empty

146

snail shell and, turtle—fashion, wears it for protection.
When he outgrows this shell, he will search out another,

larger shell.
10.

Sometime in the past, the area in which this air
breathing spider lived was invaded by enemies too strong
to fight off. Some of these spiders, called Diving Spiders.
escaped by building nests among plants below the water's
surface. In order to breathe they use their hairy hind

legs to carry air bubbles down to their nests.

Conclusion

Considering these ten examples you have just
seen, what is the common idea that connects them all to-

gether?
Script II - Expository

l- The following examples illustrate the concept ”adap—

tation." Adaptation is the idea that living things

 

147

change to fit their surroundings. A rabbit‘s coat
changing from brown in summer, to white in winter is
an obvious change; but others, like the changes in

the coat of a fawn are not so obvious. Scientists
tell us that there was a time when baby deer were not
born with these camouflaging spots. It was by a lucky
chance thousands of years ago, that a fawn was born
with this characteristic. And since a spotted coat
increased the fawn's chances of escaping detection

by his enemies, he survived and reproduced, while

others without spots did not.

Since the giraffe depends upon speed to escape his
enemies, he tends to live in flat areas that have
little ground cover. His food comes mainly from the
foliage of trees. As this breed has gradually devel-
oped a longer neck and tongue, he has become more

successful in obtaining food.

Fish that are trapped in a completely dark cave have
no need for pigment to protect them from the sun, or
eyes with which to see. This blind cave fish has made

a very interesting adjustment. Instead of eyes, he

148

has developed little growths on his head which enable
him to detect vibrations in the water. By these vi-
brations he can determine if it is an enemy or food

that is approaching.

Plants as well as animals have unusual ways of better
fitting their surroundings. The venus fly trap plant
that you see here has perfume glands which attract in-
sects. The insects trigger the trap by touching tiny
hairs. The plant absorbs the insect in order to get
nitrogen, an element not readily found in the soil

where this plant is grown.

The Archer fish has developed a most unusual weapon
to bring insects within his reach. Making use of a
groove that has developed in his upper jaw, he fires
drops of water at insects above the surface of his

pond. He is deadly at any distance up to twelve feet.

Some animals increase their chances of survival by
forming partnerships with other animals. The Remora
fish has a small suction disk on one of its fins which

it uses to attach itself to a shark. It not only gets

149

free transportation and protection from its enemies,
but when the shark makes a kill it also shares the
food. It is not clear what benefit the shark derives

from this relationship.

The yellow angler fish has, over several thousands
of years, developed a fleshy growth on his head that
closely resembles a large fishing worm. He wiggles
this growth enticingly at each passing fish. Like

any good fisherman, he gets his share.

In a period of drought, the insects that the spadefoot
toad feeds on tend to disappear. The toad, to carry
himself over this dry period, buries himself in the
soil and goes to sleep. His bodily processes slow
down so that he uses very little energy. When the
rains come a week or a month later, he digs out and

resumes his normal activities.

There are a number of fish that regard the soft body
of the hermit crab as good eating. But he has de-
veloped a clever way of escaping them. He forces

his soft abdomen or stomach into an empty snail shell,

 

150

and turtle-like, uses it for protection. When he
outgrows his present shell, he searches out a larger

one.

10. Some time in the past, the area in Which this air

breathing spider survived was invaded by enemies too

 

strong to fight off. Some of these spiders, called

 

diving spiders, escaped by building nests among plants
below the water's surface. In order to breathe they
use their hairy hind legs to carry air bubbles down

to their nests.

Conclusion

Notice that in each of the ten examples that you
have seen, living things have changed in order to better

fit their surroundings.

APPENDIX C

THE TEST MANUAL WITH TESTS. AND
DIRECTIONS FOR SOCRING TESTS

MANUAL OF DIRECTIONS
Treatment for Single-Concept Film Study

Verbal instructions to be given by the Tester to
the subjects will be printed in capital type. Directions
for the administrater will be given in lower case letters.

Pull blinds. Set up projection equipment.

HELLO. MY NAME IS . YOU ARE TAKING PART IN A NEW
WAY OF LEARNING CONCEPTS. I AM GOING TO SHOW YOU A FILM
CONTAINING TEN EXAMPLES OF SOME COMMON AND UNUSUAL PLANTS
AND ANIMALS. ALL OF THE TEN EXAMPLES IN THE FILM ARE
CONNECTED BY ONE MAIN IDEA OR CONCEPT. IT IS YOUR TASK
TO LEARN THAT IDEA FROM THE FILM. AS YOU VIEW THE FILM
WATCH AND LISTEN CAREFULLY SO THAT YOU WILL UNDERSTAND
THE MAIN IDEA THAT IS COMMON TO ALL OF THE TEN EXAMPLES

THAT YOU SEE IN THE FILM»

REMOVE ALL OBJECTS FROM YOUR DESK.

Make sure that all objects have been removed from the
students' desks.

IT IS VERY IMPORTANT THAT YOU TALK WITH NO ONE EXCEPTING
ME DURING THE REMAINDER OF THIS EXPERIMENT. THE FILM IS
ONLY FIVE MINUTES LONG. AS SOON AS YOU ARE FINISHED VIEW-
ING IT. I WILL GIVE YOU A SERIES OF QUESTIONS TO ANSWER
IN ORDER TO DETERMINE HOW WELL YOU HAVE LEARNED THE CON-
CEPT THE FILM IS ATTEMPTING TO TEACH. LET ME REMIND YOU
OF THE IMPORTANCE OF NOT TALKING TO ANY OF YOUR FELLOW
STUDENTS UNTIL AFTER THE TESTING HAS BEEN COMPLETED.

151

152

Turn off the lights. Show the film. Stay by
the projector so that you can control the film speed so
that it will match the audio.

The instructions for giving the test follow:

Post-tasks for Single—concept Film Study F4

 

Verbal instructions given by the Tester to the
subjects will be printed in capitals. Directions to the f
Tester will be given in lower case letters. i

Immediately after showing the film; turn on the
lights and open the curtains.

WE ARE NOW GOING TO GIVE YOU SOME TESTS THAT MEASURE YOUR
ABILITY TO USE THE INFORMATION GIVEN IN THE FILM. PLEASE
CLEAR YOUR DESK OF ALL OBJECTS.

Allow time necessary.

THIS TEST BOOKLET WILL NOW BE PASSED TO YOU. Hold booklet
up. DO NOT OPEN THE BOOKLET UNTIL I DIRECT YOU TO. PART
OF THE TEST IS TO SEE IF YOU CAN FOLLOW DIRECTIONS. AS
SOON AS YOU RECEIVE THE TEST BOOKLET, PRINT YOUR NAME, THE
DATE , AND YOUR HOME ROOM NUMBER AT THE TOP OF

THE TEST BOOKLET IN THE SPACE PROVIDED.

 

Have the test booklets passed, or pass them your-
self. It is very important that none of the copies are
carried out by the subjects, since the test will be given
again at a later date. Allow students time to write their

name, the date, and their home room number.

IS THERE ANYONE WHO HAS NOT FINISHED WRITING THEIR NAME.
THE DATE, AND THEIR HOME ROOM NUMBER? Give help if needed.

TURN YOUR TEST BOOKLET TO PAGE ONE. READ THE DIRECTIONS
SILENTLY WHILE I READ THEM ALOUD. Read the directions for
Post-task I. ARE THERE ANY QUESTIONS ABOUT THE DIRECTIONS?

153

Answer only those questions which help Clarify
the directions.

NUMBER EACH OF YOUR EXAMPLES. IF YOU HAVE DIFFICULTY
SPELLING A WORD DO THE BEST YOU CAN. IF YOU HAVE NEED

FOR MORE SPACE FOR YOUR EXAMPLES WRITE THEM ON THE BACK
OF THE PAGE. YOU WILL HAVE TEN MINUES TO COMPLETE THIS
SECTION. BEGIN. F

Allow ten minutes.

 

STOP. TURN TO PAGE TWO AND READ THE DIRECTIONS SILENTLY w
WHILE I READ THEM ALOUD. E

Read the directions for Post-task II to the studenus.
ARE THERE ANY QUESTIONS ABOUT THE DIRECTIONS FOR THIS SECTION?

Allow time for questions. Answer only those que—
tions which help clarify the directions.

YOU HAVE 15 MINUTES TO<ZOMPLETE THIS SECTION. BEGIN.
Allow 15 minutes to complete Post-task II.

STOP. TURN TO PAGE FOUR. READ THE DIRECTIONS FOR POST-
TASK III SILENTLY WHILE I READ THEM ALOUD.

Read the directions aloud to the students.

ARE THERE ANY QUESTIONS ABOUT THE DIRECTIONS?

Answer only those questions which help clarify
the directions.

YOU WILL HAVE TEN MINUTES TO COMPLETE THIS SECTION. BEGIN.
Allow ten minutes to complete this section.

STOP. CLOSE YOUR BOOKLET. LAY DOWN YOUR PENCIL.

154

Pick up all of the test booklets. Make certain
that you get one from each student. Discourage others
from picking up the test booklets. As you pick up the
booklets make sure the student has placed his name on
the front of the booklet.

I I ‘5'. - -
II ‘ .‘
'. .51-li-

 

NAME

155

DATE HOMEROOM

DO NOT TURN THIS PAGE
OR WRITE ON THIS BOOKLET

UNTIL DIRECTED TO DO SO

 

156

POST-TASK I

The ten examples in the film that you saw were
connected by an idea that is common to all of them. To
Show how clearly you understand this idea, list as many
examples as you can think of which were not shown in the
film.

After each new example that you list, briefly
explain why you think it illustrates the main idea of
the film. You will have ten minutes to complete this
section.

157

DO NOT TURN THIS PAGE

UNTIL DIRECTED TO DO SO

158

POST-TASK II

The ten examples that you saw in the film were

Connected by an idea that is common to all of them. See
if you can use this idea to solve the following problems.
You have fifteen minutes to complete this section.

1.

The Aberdeen angus is a beef steer that is usually
born without horns, but recently an offspring from
pure Aberdeen angus parents was born with horns.

a. Explain to the best of your ability why this
happened.

b. Is it important to this breed of animal that
some of the offspring are different from their
parents? Explain your answer.

We are told by archeologists (scientists who study
fossils) that the camel has Changed greatly over
the centuries. For example, at one time he did
not have the hump in which camels now store water.

a. Explain to the best of your ability why this
happened.

b. Is it important to this breed of animal that
this should happen. Explain your answer.

After we stand in a cold shower for a few minutes,
the temperature of the water seems to be much
warmer than when we first got under the Shower.
Similarly. after our hands have been in hot dish
water for a short time, the water does not seem
to be as hot as it was when we first placed our
hands in it.

a. Explain to the best of your ability why this
happened.

b. Is it important that this should happen?
Explain your answer.

159

DO NOT TURN THIS PAGE

UNTIL DIRECTED TO DO SO

160

POST-TASK III

The ten examples that you saw in the film are con-
nected by an idea that is common to all of them. Read the
following statements carefully, and encircle the numbers
next to the statements that you feel best indicate the
idea that was common to the ten examples you saw in the
film. Take ten minutes to finish this section.

1. All plants and animals require water.

2. Termites and protozoa (tiny one-celled animals) have
formed an unusual partnership. The protozoa live in
the termite's stomach and breaks down food swallowed
by the termite. Otherwise, the termite would die.

3. Flounders (a breed of fish) have developed the ability
of changing the color of their bodies to match their
surroundings.

4. The spider spins a sticky web in which he catches
edible insects.

5. Flies are the only insects which have red blood. All
others have either clear or green blood.

6. There are some animals in which bleeding is very diffi-
cult to stop. Human beings, who have this condition,
are called hemophiliacs.

7. An octopus has the same number of legs as does a spider.

8. The puffer fish, unlike his early ancestors, can swell
up so large that attacking fish cannot swallow him.

9- Most male birds are brightly colored. Despite this
fact, they are usually very difficult to see when they
are sitting perfectly still.

10. The circulation of the blood was discovered by an
English scientist named Priestly.

ll.

12.

13.

14.

15.

16.

17.

18.

19.

20.

161

Crabs have developed an unusual way of defending
themselves. They lock one of their claws onto an
enemy and then deliberately break off the Claw.
While the enemy struggles to free himself. the
crab escapes.

The green katydid, over thousands of years. has come
to closely resemble the leaves of the plants upon
which it feeds.

Man, like dogs, may catch the disease called rabies.

One of the major differences between male and female
mosquitos is their eating habits. The males mouth
part is made for chewing, while the female has a
sharp mouth which is good for puncturing the skins
of animals.

Some birds, when let out in an unfamiliar area, can
fly directly to their nests. Some of the ancestors
of these birds did not have this ability.

When the oxygen content in swamp water drops below
a certain level, the fish living in the swamp die.

Increasing age tends to reduce the effectiveness of
most plants and animals. However, there are a select
few who, if it were not for predators, accidents and
disease, would live forever.

When traveling in a forest, we can return to a de-
sired place by learning to use the stars or the sun
to guide us.

Although both bats and birds have wings with which
to fly, they are not related.

The vice-roy butterfly has come to look a great deal
like the beautiful monarch butterfly. Birds have a
strong dislike for the monarch's taste.

162

21. It is difficult to make a skin graft taken from one
person grow on another person.

22. The feet of young boys are very tender when they first
begin going barefoot in the spring, but in a short
time a thick protecting callus grows over the bottoms
of their feet.

23. The Snyr bird has, over many generations, stopped
-building nests. Instead they lay their eggs in the
Albatross' nest, who raises the baby Snyr birds as
her own.

24. Athletes foot is caused by a tiny fungus Which gets
its nourishment from animals.

25. Corn tassels, which contain pollen. grow above the
ears of corn. The pollen falls from the tassels
and some of the pollen lands on the corn Silk below.

 

 

163

EVALUATION OF POST-TASKS

Post-Task I - Generation of new instances of the concept.

A.

Criteria for judging
1. Not a repeat of the filmed examples.

2. Answer must indicate that organism has Changed
in order to improve its conditions.

Score will be the total number of valid instances
listed by the student.

Judging will be carried out by three science teachers.
1. Judges will score Post-Task I independently.

2. The three independent scores will be averaged

to get final score.
\

Post-Task II - Application of the concept to a problem.

A.

B.

The following scale will be used to evaluate the
solutions to the three problems:

- definitely do not have the concept
- doubtful if subject has concept
subject may have the concept

— subject probably has the concept

- definitely has the concept

QUONI-‘O
I

A student's score on the three problems in Post—
Task II will be determined by the following steps:

1. Each of the three judges will score the three
problems according to the scale in (A). Each
will take an average of the score assigned to
each problem.

 

I.

A.

2.

164

The three average scores determined by the
three judges will in turn be averaged to give
the student's final score for Post-task II.

Post-Task III - Recognition of instances of the concept.

The task will be treated as a "True or False"

test. Each correct answer will be treated as +1.

No correction for guessing. Score is number right.

Evaluation of Post-task I

Criteria for judging examples in Post-task I.

a.

The answer must indicate that the organism has

changed to better fit his surroundings.

The answer should not be a repeat of the filmed

examples which follow:

(
(

l)
2)

3)

.4)

5)

The development of spots by fawns

The development of long neck and tongue

by giraffe

Blind cave fish that has lost functional
use of eyes, but has developed growths on
its head which allow it to sense vibrations
made by other animals moving in the water.
The development of trap by Venus Fly Trap
plant in order to get needed nitrogen.
Archer fish has developed a groove in upper
jaw which allows him to fire pellets of
water at insects above the pond.

 

( 6)

( 7)

( 8)

( 9)

(10)

165

Remora fish attaches itself to a Shark by

a suction disc it has developed in order

to get food, protection, and transportation.
Angler fish has deve10ped a fleshy growth
on its head that resembles a fishing worm.
It uses this growth to capture food.
Spade-foot Toad buries itself in the soil
when food is scarce. Its bodily processes
are slowed down so that it does not use
much energy.

The Hermit Crab protects his soft abdomen
by pushing it into a snail shell, and wear-
ing it turtle-fashion.

A Diving Spider that lives under water by
carrying bubbles of air down to its under
water nest.

The score will be the total number of valid instances

listed by the student.

Judging will be carried out by three science teachers.

a. The
b. The

get

Scale

btuhabao
I

three judges will score Post-task I independently.
three independent scores will be averaged to

the final score.

Evaluation of Post-task II

- definitely do not have concept

- doubtful if subject has concept
subject may have concept

- subject probably has the concept
- definitely has the concept

166

**(In scoring, judges are to look at the two parts
under each problem and assign only one score for
each problem. The score should be in favor of
the best answer of the two parts. Example: for
prOblem one, (la) is scored 3, (lb) is scored l--
the final score for problem one would be 3.)

II. Model answers for the five categories of the scale:

0 - definitely do not have concept

(la)

(lb)

(2a)

When the Aberdeen angus is born it is not
grown. As it grows their horns develop.

This could be caused by the mother getting
hurt.

The mother might have mated with a steer
with horns.

No. it just isn't necessary.

Yes, because then they would look like
their parents and you couldn't tell them
apart.

Because they don't come from the same ani-

mal or flower and so they are different.

He may have mated with an animal that did
have a hump and they've been this way ever
since.

Camels have always had humps.

Because maybe there are other kinds of camels.

 

(2b)

(3a)

167

Yes, because they are made for the desert.
and that is why they have a hump on their
back.

Yes so we can see the difference between
them.

Yes, because it wouldn't seem right somehow.

Because the cold water runs so long it begins
to get hotter.

Yes because drinking warm water would be
terrible.

Yes, if it didn't when you were in cold
water you would catch cold.

1 - doubtful if subject has concept

(1a)

(lb)

Because the other steer probably has some
hormones or something mixed in the blood-
streams of the body to cause such functions.

Because they say that all things are not
alike. so the one with horns is different.

It would have to depend on the mother. if
it is born with horns.

No, because you wouldn't wont the offspring
to look just like you.

Yes, because if everything in this world
was the same it would look pretty dull.

Yes. because then you can tell which is
which. Like he is the baby or this one
is older.

A ms-

 

168

(2a) I was told that two different camel breeds
met and as the result the baby was born
with a hump.

Over the years scientists have experimented
with animals.

No. because if it was caused by the weather
then other animals living in the desert
would have humps.

(2b) Yes. because pe0ple can use them when they
go across the desert.

No, this is nature.

Yes. because other camels would not like
him without his hump.

(33) Your body gets numb.
Your blood gets to the same temperature.
Because of the heat being transferred from

the body to the water.

(3b) If it didn't happen there wouldn't be any
dishes washed. and we wouldn't take any
baths.

Because you wouldn't want it to stay hot
all the time.

Yes because Lf you was to keep your hands
in hot water you could possibly burn them.

2 - subject may have concept

(la) It is just nature's way of making them
different.

 

(lb)

(2a)

(2b)

169

Because his ancestors once had them.

I think it is to protect him, because his
mother Aberdeen may be in an area that is
dangerous.

Because this would decide whether the plant
or animal would be better or worse. Or how
it would effect its whole way of life. E

Because if everything was the same we
wouldn't have different kinds of plants or
animals, and we wouldn't have the different
thoughts about these different things.

 

Yes. because they can't be the same way as
the parents are all the time.

Because probably, the camel needs to drink
much more water now than he did before.

This might have happened because the animal
lived in a hot climate where little water
could be found.

The camel didn't drink so much before. so
no hump. Now that the camel drinks a large
amount of water he gets his hump.

Yes, because the camel is used mostly in the
desert. and there is little water in the
desert so it is good that they have humps
to store water.

Yes because he would die of thirst without
water.

Yes because it makes it easier for camels
to travel long distances.

(3a)

(3b)

170

The hot water makes the blood in your hands
warmer, so you don't notice the heat.

Because our body temperature gets used u)
the new temperature.

Your hand will build a resistance to the
temperature.

Yes. for if it didn't we always would have
to drink all water the same temperature.
And to drink warm water would be awful.

 

Yes. because you could die of too sudden
a change.

Yes, because it wouldn't be as hot to your
hands after a while.

3 —_§gbject probably has the concept

(la)

(lb)

In every race, no matter if plant or animal
or person. there is a need for something
different in life.

Because they Change from year to year.

It happened so that they can protect them-
selves.

Yes, because maybe they can improve.

Yes, so that there will be new kinds.

Some plants and animals need to be different

from their parents because there might be a
reason why the parents can't get along.

171

(2a) Just as horses used to be very small,
camels have inherited or grown a hump.

Because he needed this hump to survive.

When the camel didn't have the hump he
couldn't live in as dry places as he can
today.

(3a) When you are in an environment for so long.
your body temperature tends to equalize
with the environment.

 

Your body gets accustomed to the water so
you can't tell if it is hot or cold.

When you put hands into hot water, your
hands are colder than the water so you let
off some cold molecules that cool the water
a little and you absorb some hot molecules
so your hands and the water have a closer
temperature.

(3b) Yes, so you could be comfortable and could
work or shower easier.
Yes. because otherwise you could only handle
those things that have the same temperature
as you do.
Yes, to Show us how hot, so we will not be
burned.

4 - definitely has concept

(1a) This is prObably caused by the Change that
takes place in all animals.

 

(lb)

(2a)

(2b)

172

Nature keeps producing animals a little bit
different than their parents so that the
animals will improve so they can live better.

This happened in order to improve the breed.

They may need this different thing for pro-

tection or for the place they live in. They

may need it so they can adjust to their sur-
rounding.

Yes, because sometime the children will
need something else that their parents
didn't need.

Yes, because with the Changing conditions
of the world. all animals and plants have
to Change or else become extinct.

It had to adjust to the desert.

Maybe there was a baby camel born that by
some sort of accident had a hump and it
never went away.

In order to survive the changes in the cli-
mate When it turned dryer, the camel race
had to find some way to carry more water
in order to survive.

Yes. every camel should be able to change
in order to survive in his environment.

Yes. because if they did not change the
camel might become extinct.

Yes. because this is one way that nature
can provide animals with things that will
make life easier.

 

 

173

(3a) Because your body usually stays at a steady
temperature. But when the temperature
changes outside it makes your nerves react.
They send a message to your mind. After a
time it feels normal.

Because the skin becomes conditioned to
that temperature.

The body is able to regulate its tempera-
ture to fit the type of environment it is in.

 

(3b) Yes. humans have to adjust to different con-
ditions too.

Yes, because if we couldn't adjust, we
couldn't live.

Yes. because if it didn't. you could not
live under unusual temperatures or differ-
ent climates.

 

 

APPENDIX D

MEANS AND STANDARD DEVIATIONS OF FACTOR LEVELS FOR
TEACHING METHOD. AND FOR THE INTELLIGENCE-METHOD

 

 

 

 

 

INTERACTION
Legend:

High IQ = 1

Middle IQ = 3

Low IQ = 2

Discovery = 4

Expository = 5

I.--Origina1 Analyses of Variance - Immediate Post—tasks
IQ, Method Mean Standard Deviation

Post- 1 4 2.6000 2.8812
task 1 5 3.1000 2.8343
I 2 4 .8857 1.6378
2 5 .8714 1.7439
4 1.7428 2.4884
5 1.9857 2.5976
Post- 1 4 7.7428 2.0689
task 1 5 7.8857 2.5283
II 2 4 5.3285 2.4889
2 5 4.9285 2.3975
4 6.5357 2.5822
5 6.4071 2.8685
Post— 1 4 17.9571 3.5033
task 1 5 18.3000 3.2631
III 2 4 15.4000 2.9012
2 5 14.6714 3.1333
4 16.6785 3.4521
5 16.4857 3.6707

174

175

II.--Original Analyses of Variance - Delayed Post-tasks

 

.IQ Method Mean Standard Deviation

Post- 1 4 2.6000 3.0187
task 1 5 3.6000 2.6778
I 2 4 .9857 1.8454
2 5 .9571 2.0531

4 1.7928 2.6211

5 2.2785 2.7223

Post- 1 4 7.3000 2.2024
task 1 5 7.4142 2.3559
II 2 4 4.8428 2.2691
2 5 4.7714 2.1139

4 6.1285 2.6412

5 6.0357 2.4972

Post- 1 4 17.6857 3.5896
task 1 5 18.2857 3.4479
III 2 4 14.4714 3.8437
2 5 13.7000 3.4192

4 16.0785 4.0412

5 15.9928 4.1230

III.--Data Regrouped on Three IQ Levels - Immediate Post-tasks

.19 Method Mean Standard Deviation

Post- 1 4 3.0434 3.0980
task 1 5 3.4782 2.8498
I 2 4 .4982 1.0052
2 5 .4782 .9366

3 4 1.6739 2.2015

3 5 1.9347 2.6533

4 1.7318 2.4865

5 1.9637 2.6037

 

 

IQ

Post- 1
task 1
II 2
2

3

3

Post- 1
task 1
III 2
2

3

3

IV.--Data Regrouped on Three IQ Levels - Delayed Post-tasks

IQ

Post— 1
task 1
I 2
2

3

3

Post- 1
task 1
II 2
2

3

3

Method

mubtfi.b(fi.b(fi.b

UlubU'l-bU'lkUl-p

Method

mbmbmpmp

mbmbmbm-b

176

Mean

8.3043
8.3695
4.8043
4.2826
6.4782
6.5000
6.5289
6.3840

17.7173
17.9565
13.7173
13.2391
16.0869
16.2391
15.8405
15.8115

Mean

2.9130
4.1739

.6304

.5869
1.8043
2.0869
1.7826
2.2826

7.8260
7.9347
4.1956
4.1304
6.3260
6.0217
6.1304
6.0144

Standard Deviation

2.1173
2.5065
2.2471
2.3157
2.0948
2.4845
2.5749
2.8803

3.4426
3.0980
2.5962
2.9225
3.3053
3.7605
3.5231
3.7988

Standard Deviation

2.9046
2.7511
1.4508
1.3916
2.7778
2.5632
2.6214
2.7312

1.9241
2.3036
2.1563
2.2371
1.9328
2.0164
2.6576
2.5083

 

1.9
Post- 1
task 1
III 2

2
3
3

Method

UIubUlibmobU‘Inh

177

Mean

18.1521
18.3913
13.2608
13.1521
16.7826
16.4130
16.0652
15.9855

Standard Deviation

3.5525
3.4671
3.2278
3.2177
3.7351
3.9416
4.0529
4.1416

V.-—Data Regrouped with Middle IQ Level Excluded - Immediate

 

 

Post-tasks
‘IQ Method Mean Standard Deviation

Post- 1 4 3.0434 3.0980
task 1 5 3.4782 2.8498
I 2 4 .4982 1.0052
2 5 .4783 .9366

4 1.7608 2.6285

5 1.9782 2.5932

Post- 1 4 8.3043 2.1173
task 1 5 8.3695 2.5065
II 2 4 4.8043 2.2471
2 5 4.2826 2.3157

4 6.5543 2.7946

5 6.3260 3.1591

Post- 1 4 17.7173 3.4426
task 1 5 17.9565 3.0980
III 2 4 13.7173 2.5962
2 5 13.2391 2.9225

4 15.7173 3.6383

5 15.5978 3.8203

178

VI.--Data Regrouped with Middle IQ Group Exluded - De-
: layed Post-tasks

 

IQ_ Method Mean Standard Deviation

Post— 1 4 2.9130 2.9046
task 1 5 4.1739 2.7511
I 2 4 .6304 1.4508
2 5 .5869 1.3916

4 1.7717 2.5553

5 2.3804 2.8200

Post- 1 4 7.8260 1.9241
task 1 5 7.9347 2.3036
II 2 4 4.1956 2.1563
2 5 4.1304 2.2371

4 6.0326 2.9592

5 6.0108 2.7315

Post— 1 4 18.1521 3.5525
task 1 5 18.3913 3.4671
III 2 4 13.2608 3.2278
2 5 13.1521 3.2177

4 15.7065 4.1761

5 15.7717 4.2429

VII.--Data Regrouped with Subjects Separated by Sex —
Immediate Post-tasks (Boys)

.IQ Method Mean Standard Deviations

Post- 1 4 2.5500 3.2682
task 1 5 4.5500 3.2682
I 2 4 .7500 1.1641
2 5 .5000 1.0000

4 1.5250 2.6016

5 2.6500 3.0930

Post- 1 4 7.2500 2.1974
task 1 5 8.1000 2.2918
II 2 4 5.8000 2.8946
2 5 4.5000 2.2360

4 6.5250 2.6407

5 6.3000 2.8840

 

 

“IQ Method Mean Standard Deviation
Post- 1 4 16.1000 3.3857
task 1 5 18.2000 2.5874
III 2 4 13.9000 3.3857
2 5 12.9500 2.2589
4 15.0000 3.5228
5 15.5750 3.5797

179

 

Vlll.-—Data Regrouped with Subjects Separated by Sex -
Delayed Post-tasks (Boys)

 

.IQ Method Mean Standard Deviation

Post- 1 4 2.5000 3.0174
task 1 5 4.7000 2.8302
I 2 4 1.0000 1.8064
2 5 .1500 .4893

4 1.7500 2.5695

5 2.4250 3.0541

Post— 1 4 7.1500 2.7003
task 1 5 6.9500 2.5438
II 2 4 4.9000 2.6734
2 5 4.1500 2.2307

4 6.0250 2.8866

5 5.5500 2.7544

Post- 1 4 17.0500 4.5011
task 1 5 19.0000 2.8654
III 2 4 14.6500 3.7735
2 5 12.2500 3.2424

4 15.8500 4.2760

5 15.6250 4.5611

 

IX.-—Data Regrouped with Subjects Separated by Sex-

180

Immediate Post-tasks (Girls)

 

 

.IQ Method Mean Standard Deviation

Post- 1 4 3.1000 2.7510
task 1 5 2.9500 2.2820
I 2 4 .6500 1.3484
2 5 1.2500 2.7120

4 1.8750 2.4722

5 2.1000 2.6194

Post- 1 4 8.5500 2.4381
task 1 5 7.7500 2.9177
II 2 4 5.2000 2.3078
2 5 4.7500 2.6532

4 6.8750 2.8928

5 6.2500 3.1439

Post- 1 4 17.4500 4.1609
task 1 5 18.5000 2.7047
III 2 4 14.8500 2.5188
2 5 13.9000 3.0932

4 16.1500 3.6412

5 16.2000 3.6947

x.--Data Regrouped with Subjects Separated by Sex -
Delayed Post-tasks (Girls)

 

‘IQ Method Mean Standard Deviation

Post- 1 4 3.8000 3.8607
task 1 5 3.8000 2.2384
I 2 4 1.1000 1.9166
2 5 1.4000 2.5422

4 2.4500 3.3046

5 2.6000 2.6583

Post- 1 4 8.4000 2.3033
task 1 5 7.6000 2.4793
II 2 4 4.1000 2.2454
2 5 4.9000 1.8035

4 6.2500 3.1276

5 6.2500 2.5394

181

 

_IQ Method Mean Standard Deviation
Post- 1 4 18.4000 3.3633
task 1 5 19.3500 2.7198
III 2 4 14.1000 3.5377
2 5 14.1000 2.9540
4 16.2500 4.0430
5 16.7250 3.8629

 

BIBLIOGRAPHY

 

 

BIBLIOGRAPHY

Anderson, G. L. "Quantitative Thinking as Developed
Under Connectionist and Field Theories of Learn-
ing," Learninngheory in SchOol Situations, Min—
neapolis: University of Minnesota Press, 1949.
pp. 40-73.

Ausubel, D. P. The Psychology of Meaningful Verbal
Learning. New York: Grune & Stratton, 1963.

. "Some Psychological Aspects of the Structure
of Knowledge," Education and the Structure of
Knowledge, ed.. Stanley Elam, Chicago: Rand
McNally and Company, 1964.

Bloom. B. S., ed. Taxonomy of Educational Objectives.
New York: David McKay Company, Inc.. 1956.

Bourne, L. E. and Haygood, R. C. ”Supplementary Report:
Effect of Redundant Relevant Information Upon the
Identification of Concepts," Journal of Egperi-
mental Psychology, LXI, 1961, pp. 259-260.

Brenner, H. R.; Walter, J. S.; and Kurtz, A. K. ”The
Effects of Inserted Questions and Statements on
Film Learning," Progress Report No. 10, State
College, Pennsylvania: Pennsylvania State College,
Instructional Film Research Program.

Bruner, J. S. “The Act of Discovery.” Harvard Educational
Review, XXXI, 1961. pp. 21-32.

Bruner, J. S.; Goodnow. J. J.; and Austin, A. A Study of
Thinking. New York: Wiley, 1957.

183

184

Corman, B. R. "The Effect of Varying Amounts and Kinds
of Information as Guidance in Problem Solving."
Psychological Monographs, LXXI. 1957, pp. 1-21.

Craig, R. C. "Directed Versus Independent Discovery of
Established Relations." Journal of Educational
Psychology. XLVII. 1956. pp. 223—234.

"The Transfer Value of Guided Learning,"
Bureau of Publications, Teacher's College.
Columbia University, New York. 1953.

 

Cronbach, Lee J. Essentials of Psychological Testing.
New York: Harper and Brothers, 1960.

DeJonge, J. J. and Sim, F. M. Factor Analysis Programs:
Fanod 3 and Fanim 3. East Lansing: Michigan State
University, 1964.

Einbecker. W. F. "Comparison of Verbal Accompaniment to
Films." School Review. XLI. 1933, pp. 185-192.

Gagne, Robert M. The Conditions of Learning. New York:
Holt, Rinehart, and Winston, Inc., 1965.

Gagne, R. M. and Brown, L. T. "Some Factors in the Pro—
graming of Conceptual Learning." Journal of Exper-
imental Psychology. LXII. 1961. pp. 313-321.

Haslerud, G. N. and Meyers, Shirley. "The Transfer Value
of Given and Individually Derived Principles,"
Journal of Educational Psychology, XLIX, 1958,
pp. 293-298.

Hays, W. L. Statistics for Peychologists. New York:
Holt. Rinehart and Winston. Inc.. 1963.

Hendrix. Gertrude. "A New Clue to Transfer of Training,”
Elementary School Journal, XLVIII. 1947. pp. 197-
208.

 

 

185

Hilgard. Ernest R. Theories of Learning. New York:
Appleton-Century-Crofts, Inc., 1956.

HOban, Charles F., Jr., and van Ormer. E. B. Instruc-
tional Film Research 1918-1950 (Rapid Mass Learn-
ing). Report on the Pennsylvania State College
Project Jointly Sponsored by the Department of
the Army and the Department of the Navy. Port
Washington, N. Y.: U. S. Navy Special Devices
Center, 1951.

Hovland, C. I.; Lumsdaine, A. A.; and Sheffield, F. D.
Experiments on Mass Communication. Princeton.
New Jersey: Princeton University Press, 1949.

Hovland, C. I., and Weiss, W. "Transmission of Infor-
mation Concerning Concepts Through Positive and
Negative Instances," Journal of Experimental Psy-
chology. XLV, 1953, pp. 175-182.

Hull, C. L. "Quantitative Aspects of the Evolution of

Concepts." Psychological Monograph, XXVIII, 1920,
p. 85.

Huttenlocker. J. "Some Effects of Negative Instances on
the Formation of Simple Concepts," Psychological
Reports. XI. 1962. pp. 35—42.

Jaspen, N. Effects on Training of Experimental Film
VariablesL_Study II. Progress Report No. 14-15—16,
State College, Pennsylvania State College, Instruc-
tional Film Research Program, 1950.

Johnson, D. W. The Psychology of Thought and Judgment.
New York: Harper and Brothers. 1955.

Judd, C. H. "The Relation of Special Training to Special

Intelligence," Education Review, XXVI, 1908, pp.
28-42.

186

Kersh, Bert Y. "The Adequacy of Meaning as an Explana-
tion for the Superiority of Learning by Indepen-
dent Discovery," Journal of Educational Psychology,
XLIX, 1958, pp. 282-292.

. Classroom Simulation: A New Dimension in
Teacher Education. The Final Report, Title VII.
Project No. 886; National Defense Education Act
of 1958, June 1963, pp. 38-44.

 

"The Motivating Effect of Learning by Directed
Discovery," Journal of Educational Psychology.
L111. 1962. pp. 65-71.

 

Kiel, D. F.; Kenworthy, A. L.; and Ruble. W. L. Use of
Analysis of Variance Routines on the CDC 3600.
East Lansing: Michigan State University. 1963.

Kittell. J. E. "An Experimental Study of the Effect of
External Direction During Learning on Transfer
and Retention of Principles," Journal of Educa-
tional Psychology, XLVIII, 1957, pp. 391-405.

Klausmeier, H. J., and Loughlin, L. J. "Behavior Problem
Solving Among Children of Low, Average, and High
Intelligence," Journal of Educational Peychology,
LII (1961), pp. 148-152.

McClusky, S. D. and H. Y. "Comparison of motion pictures,
slides, stereographs and a demonstration as a means
of teaching how to make a reed mat and a pasteboard
box," Visual Education, ed. F. N. Freeman, Chicago:
University of Chicago Press, 1924, pp. 310-334.

McGuire, W. J. "Slow Motion, Added Narration and Distrib-
uted Showing as Factors Influencing Teaching Effec-
tiveness of a Training Film," Visual Communication.
eds. John Ball and Frances C. Byrnes, Washington,
D. C.: The Department of Audiovisual Instruction
Of the National Education Association, 1960, p. 96.

 

 

187

Medley. Donald M. and Mitzel, Harold E. "Measuring Class-
room Behavior by Systematic Observation," Handbook
of Research on Teaching, ed. N. L. Gage, Chicago:
Rand McNally & Co., 1963, II, Part II, pp. 247-248.

Miller. George A. "The Magical Number 7, Plus or Minus 2:
Some Limits on our Capacity for Processing Informa-
tion,” Psychological Review, LXIII (1956L pp. 81-97.

Montessori. Maria. The Montessori Method. London: William
Heinemann. 1912, pp. 171-172.

Osler, S. F. and Weiss, S. R. "Studies in Concept Attain—
ment: III. Effect of Instructions at Two Levels
of Intelligence,” Journal of Experimental Psychol-
pgy, LXIII, 1962, pp. 528-533.

Roe. K. V.; Case, H. W.; and Roe, A. "Scrambled Versus
Ordered Sequence in Autoinstructional Programs,"
.gpprnal of Educational Behavior PsycholoQY. LIII,
1962, pp. 101-104.

Schalock, H. D. et. a1., Motion Pictures as Test Stimuli:
An Application of New Media to the Prediction of
Complex Behavior. A Final Report. Title VIII.
Project No. 971, Dec. 1964.

Smoke, K. L. "Negative Instances in Concept Formation,"
Journal ofggxperimental Psychology, XVI, 1933.
pp. 583-588.

Stanley, J. C. "The Role of Instruction. Discovery, and
Revision in Early Learning," Elementary School
Journal. XLIX, 1949, pp. 455-458.

Suchman. Richard J. The Elementary School Training Pro-
gram in Scientific Inquiry. A Project sponsored
by U. S. Department of Health, Education and Wel—
fare, Office of Education, and by the Research
Board of the University of Illinois, Title VII
Project, No. 216: National Defense Education Act
of 1958. June'1962.

188

Swenson, Esther J.; Anderson, G. L.; and Stacey, C. L.
Learning Theory in School Situations. Minneapo-
lis, University of Minneapolis Press. 1949.

Thorndike, E. L. The Psychology of Wants. Interests.
and Attitudes. New York: Appleton—Century.
1935. .

Travers. Robert M. "The Transmission of Information to
Human Receivers," AV Communication Review. XII.
NO. 41 1964! pp. 373-3850

Vinacke. W. Edward. "The Investigation of Concept For—
mation," Psychological Bulletin, XLVIII, January,

1951, pp. 1-31.

Wagner, Robert W. "The Educational Film in Transition."
Audiovisual Instruction. IX, No. 174. March. 1964.

"The Formula Film" A.V. Communication Review.
III, No. l, 1955. p. 54.

Wittrock, M. C. The Learning by Discovery Hypothesis.
A Review prepared for the Conference on Learning
by Discovery, New York City, January 28-29, 1965;
Sponsored by Stanford University with support from
the U. S. Office of Education, 1964, p. 10.

 

HICHIGRN STATE UN
llWWII!WWI
31293102

IV. LIBRQR IES

MIUIIHIIIIW lllWll

099904