Citation
Differential effects of pictorial and written presentation on the acquisition of scientific concepts by Indians taught in Bengali and in English

Material Information

Title:
Differential effects of pictorial and written presentation on the acquisition of scientific concepts by Indians taught in Bengali and in English
Creator:
Roy, Protima Kumir, 1944-
Publication Date:
Language:
English
Physical Description:
xiv, 194 leaves. : illus. ; 28 cm.

Subjects

Subjects / Keywords:
Science -- Study and teaching (Secondary) ( lcsh )
Curriculum and Instruction thesis Ph. D
Dissertations, Academic -- Curriculum and Instruction -- UF
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis -- University of Florida.
Bibliography:
Bibliography: leaves 186-192.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Protima Roy.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
This item is presumed in the public domain according to the terms of the Retrospective Dissertation Scanning (RDS) policy, which may be viewed at http://ufdc.ufl.edu/AA00007596/00001. The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
022775041 ( ALEPH )
14076172 ( OCLC )

Downloads

This item has the following downloads:


Full Text
DIFFERENTIAL EFFECTS OF PICTORIAL AND WRITTEN
PRESENTATION ON THE ACQUISITION OF
SCIENTIFIC CONCEPTS BY INDIANS
TAUGHT IN BENGALI AND IN ENGLISH
By
PROTIA ROY
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THlE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1974




Dedicated to my grandmother,
BROJOBALA KUMAR-a wonderful concept




ACKNOWLEDGMENTS
The writer wishes to express her deepest gratitude to Professor John J. Koran, Jr., Chairman of the supervisory committee, for his time, effort, encouragement and continuous guidance throughout the present study. The author is also thankful to all her committee members, Professor Vynce A. Hines, Dr. Joseph J. Shea, Dr. James D. Casteel, for their generous help and suggestions.
Special thanks are also extended to Dr. Mary Lou Koran for her help in interpreting the aptitude treatment interactions data and to Professor John M. Newell for his generous advice.
The investigation owes a special debt of gratitude to Dr. James E. McLean, for his assistance in computer programming, to Dr. Arun Mazumdar and Dr. Shantanu Maitra for their assistance and technical help in preparation of the data for analysis.
The author also expresses her appreciation to her husband, Professor Rabindranath Roy, for his interest and encouragement throughout this study.
Finally, the author wishes to express her sincere appreciation to the logistical support of the administration of the school system in Calcutta which participated in this study.
iii




TABLE OF CONTENTS
Page
ACKN~OWLEDGMENTS..............................ti
LIST OF TABLES.................................. vii
LIST OF FIGURES.................................. xt
ABSTRACT...................................... xii
CHAPTER
I THE PROBLEM........................... 1
Definition of the Problem...................1I
Background to the Problem............... 1
Definitions........................... 3
Types of Concepts .. .. .... ..... ....5
Distinguishing Characteristics of
Concept Formation ....... .. .... ...7
Research on Concept Formation. .. .. .. ...8
Effects of Different Variables that Influence
Concept Formation. .. .... ...... ...8
Effects of Positive and Negative Examples ...8
Effects of Redundant Relevant Information on
Concept Acquisition. .. .. .... ......10
The Effects of the Order and Sequence of
the Examplars .. .. ...... .......11
Deductive and Inductive Instruction .. .. .....11
Nature of Critical and Noncritical
Attribut es .. ... ...... ...... ..12
Type of Concept Variables .. .. .. ........12
Response Variable. .. .. .... ...... ..13
Theory and Research on Pictorial
Presentation. .. .. ..... ...... ..13
Theory and Research on Aptitude Treatment
Interactions .. .. ........ ......20
Theory and Research on Language and
Cognition. .. ... ..... ...... ...23
Language and Its Influence on Concept
Learning. .. ... ...... ...... ..24
Cultural Influence. .. .. ...... ......26
Bilingual Children. .. .. ..... ........27
Rationale for This Study .. .. ..... ......29
Experimental Questions .. .. ...... ...36
iv




TABLE OF CONTENTS (Continued)
CHA PTER Page
11 EXPERLMNENTAL DESIGN................... 37
The Design........................... 37
Independent Variables .. .. ..... .....37
Ability Measures .. .. .. ...... .....39
Letter Sets Test (I-11. .. ..... .. .....40
Vocabulary Test (V-1) and (V-2). .. .....40 First and Last Names Test (Ma-3). .. .....40 Auditory Number Span Test (Ms-i) ......41 Dependent Variables .. .. ... ...... ..41
Methods. .. .. ...... ..... ..... ..41
Subjects .. ... ..... ..... ......41
Treatment Materials. .. .. .... ......43
Procedures. .. .. .............44
The Instrument and Its Reliability .. .. .....45 Data Collection Procedures. .. .. .......46
III RESULTS .. .. ... ..... ...... ......48
Purpose of Study and Data Collected. .. ... ..48
Independent and Dependent Measures. .. ....48
Intercorrelations Between Different
Measures .. .... ...... ......57
Instructional Treatment Main Effects. .. ....62
Acquisition of Concepts .. .. ... ...... ..63
Total Achievement on Post- and RetentionTest. .. .. .... ..... ...... ..63
Performance on Concept I. .. .. ..... ..70
Performance on Concept R .. .. .. ......70
Performance on Concept III. .. .. ......81
Effects of Post-Test and Retention-Test
Time .. .. ...... ... ........86
Aptitude X Treatment Interactions .. .. .. ..90
IV DISCUSSION AND IMPLICATIONS .. .. .... ..104
Summary of Data and Interpretation .. .. ....104 Instructional Treatment 'Main Effects. .. .. ...107
Acquisition of Concepts .. .. ...... ..107
Differential Effects of Written and
Pictorial Presentation .. .. ..... ..108
Implications for the Educational Practice . 117 Limitations .. .. .. ...... ..... ..119
APPENDICES. .. .. ..... ..... ..... ...... ..121
V




TABLE OF CONTENTS (Continued)
APPENDIX Page
A TREATMENT MATERIALS FOR THE BENGALI
MEDIUM SCHOOLS....................... .122
B POST-TEST AND RETENTION -TEST FOR THE
BENGALI MEDIUM SCHOOLS................ 140
C TREATMENT MATERIALS FOR THE ENGLISH
MEDIUM SCHOOLS....................... 155
D POST-TEST AND RETENTION-TEST FOR THE
ENGLISH MEDIUM SCHOOLS................ 173
LIST OF REFERENCES............................. 186
BIOGRAPHICAL SKETCH............................ 193
vi




LIST OF TABLES
TABLE Page
1 EXPERIMENTAL DESIGN. ............. 38
2 APTITUDE MEASURES ................ 42
3 INDEPENDENT AND DEPENDENT MEASURES.. 47
4 MEANS AND STANDARD DEVIATIONS OF THE
THREE GROUPS (BENGALI MEDIUM SCHOOLS) 49
5 MEANS AND STANDARD DEVIATIONS OF THE
THREE GROUPS (ENGLISH MEDIUM SCHOOLS) 51
6 MEANS AND STANDARD DEVIATIONS OF
DEPENDENT VARIABLES (BENGALI MEDIUM
SCHOOLS) ...................... 52
7 MEANS AND STANDARD DEVIATIONS OF
DEPENDENT VARIABLES (ENGLISH MEDIUM
SCHOOLS) ...................... 53
8 MEANS AND STANDARD DEVIATIONS OF
APTITUDE MEASURES (BENGALI MEDIUM
SCHOOLS) ..... ........................ 55
9 MEANS AND STANDARD DEVIATIONS OF
APTITUDE MEASURES (ENGLISH MEDIUM
SCHOOLS) ...................... 56
10 INTERCORRELATION MEASURES OF THE
APTITUDE TESTS (BENGALI MEDIUM SCHOOLS) 58
11 INTERCORRELATION MEASURES OF THE APTITUDE TESTS (ENGLISH MEDIUM SCHOOLS). . 59
12 INTERCORRELATION MEASURES OF THE
DEPENDENT VARIABLES (BENGALI MEDIUM
SCHOOLS) ..... ........................ 60
13 INTERCORRELATION MEASURES OF THE
DEPENDENT VARIABLES (ENGLISH MEDIUM
SCHOOLS) ..... ........................ 61
vii




LIST OF TABLES (Continued)
TABLE Page
14 ANALYSIS OF VARIANCE FOR CRITERION MEASURES TOTAL ACHIEVEMENT (BENGALI
MEDIUM SCHOOLS) ................. 64
15 TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (BENGALI MEDIUM
SCHOOLS) ...................... 65
16 TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (BENGALI MEDIUM
SCHOOLS) ...................... 66
17 ANALYSIS OF VARIANCE FOR CRITERION MEASURES TOTAL ACHIEVEMENT (ENGLISH
MEDIUM SCHOOLS) ................. 68
18 TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (ENGLISH MEDIUM
SCHOOLS) ...................... 69
19 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT I (BENGALI MEDIUM
SCHOOLS) ...................... 71
20 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT I (BENGALI MEDIUM SCHOOLS) .... 72
21 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT I (ENGLISH MEDIUM
SCHOOLS) ...................... 73
22 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT I (ENGLISH MEDIUM SCHOOLS) .... 74
23 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT II (BENGALI MEDIUM
SCHOOLS) ...................... 76
viii




LIST OF TABLES (Continued)
TABLE Page
24 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT II (BENGALI MEDIUM SCHOOLS) .... 77
25 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT II (ENGLISH MEDIUM
SCHOOLS) ....................... 78
26 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT II (ENGLISH MEDIUM SCHOOLS)..... 79
27 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT II (ENGLISH MEDIUM SCHOOLS) .... 80
28 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT III (BENGALI MEDIUM
SCHOOLS) ....................... 82
29 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT III (BENGALI MEDIUM SCHOOLS) .... 83
30 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT II (ENGLISH MEDIUM
SCHOOLS) ....................... 84
31 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT III (ENGLISH MEDIUM SCHOOLS) .... 85
32 ANALYSIS OF VARIANCE FOR CRITERION MEASURES TIME USED FOR POST- AND RETENTION-TEST (BENGALI MEDIUM SCHOOLS)..... 87
33 TUKEY'S TEST FOR PAIRWISE COMPARISONS TIME USED FOR POST- AND RETENTION-TEST
(BENGALI MEDIUM SCHOOLS) ........... 88
34 ANALYSIS OF VARIANCE FOR CRITERION MEASURES TIME USED FOR POST- AND
RETENTION-TEST (ENGLISH MEDIUM SCHOOLS). 89
ix




LIST OF TABLES (Continued)
TABLE Page
35 TUKEY'S TEST FOR PAIRWISE COMPARISONS TIME USED FOR POST- AND RETENTION-TEST
(ENGLISH MEDIUM SCHOOLS) ........... 91




LIST OF FIGURES
FIGURE Page
1 Regression Analysis for Letter Sets Test (Part
U[) on Post-Test Time Used (Bengali medium
school SO). .. .... ...... ..... ......92
2 Regression Analysis for First and Last Names
Test on Post-Test Time (Bengali medium
school Ss). .. .... ...... ..... ......94
3 Regression Analysis for First and Last Names
Test (Part HI) on Post-Test Time Used (Bengali
medium school Ss).. .. ... ...... ..... ..95
4 Regression Analysis for First and Last Name
Test on Time Used for Retention-Test (Bengali
medium school Ss).. .. ... ...... ..... ..96
5 Regression Analysis for Vocabulary Test (V-2),
Part H., on Concept I (English medium school Ss)O 97
6 Regression Analysis for First and Last Names
Test on Concept I (English medium school Ss). . 99
7 Regression Analysis for Auditory Number Span
Test on Concept I (English medium school Ss) .. 100
8 Regression Analysis for Vocabulary Test (V-2),
Part I on Concept III (English medium school Ss). . 101
xi




Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
DIFFERENTIAL EFFECTS OF PICTORIAL AND WRITTEN
PRESENTATION ON THE ACQUISITION OF
SCIENTIFIC CONCEPTS BY INDIANS
TAUGHT IN BENGALI AND IN ENGLISH
By
Protima Roy
December, 1974
Chairman: Dr. John J. Koran, Jr. Major Department: Curriculum and Instruction
The purpose of this study was to compare the relative effectiveness of pictorial presentation and written passages on the acquisition of three class ificational concepts in biology by Indians taught in English and in Bengali. This study should provide further information regarding whether the acquisition of concepts depends on the cultural background, language and specific characteristics of the student.
Independent measures include: written passages about three concepts (geotropism, phyllotaxy, feather), single-line pictorial diagrams with appropriate labeling of the same concepts and nine aptitude tests, measuring verbal abilities, inductive reasoning and memory.
Dependent measures were the immediate post-test (corresponding to the treatment materials) measures for the acquisition of the
xii




three concepts, delayed retention-test (equivalent to post-test) measures and time used.
One hundred and two students were selected from the Bengali medium schools and also from the English medium schools. The subjects were randomly assigned to two treatments and a control group. A split-plot nested design with repeated measures and equal cell size (34) was used to evaluate the instructional treatment main effects. The same procedures were followed for both types of school subjects.
Subjects receiving treatments showed significantly greater
behavioral change than those in the control. Although it was anticipated that there would be a differential effect of written and pictorial treatments, this was not found except in one case, that is for Concept III in the Bengali medium schools. In the latter case, the written group was significantly different than the pictorial group. This finding is basically consistent with a pilot study conducted by the author using American subjects.
Comparing the means it was evident that subjects of the two different types of schools showed a similar performance on the criterion measures.
F test for homogeneity of regression was employed to study
aptitude x treatment interactions. Analyses of interactions disclosed that scores on Letter Sets Test. First and Last Names Test, Vocabulary Test, and Auditory Number Span Test interacted significantly vith the instructional treatments.
xiii




These findings suggest that subjects in different treatment
groups responded differently to the criterion measures and consequently could be differentially assigned to instructional method according to these characteristics; thus achieving individualization.
xiv




CHAPTER I
THE PROBLEM
Definition of the Problem
The main purpose of this investigation was to compare the
relative effectiveness of pictorial representation and written passages on the acquisition of three class ificational concepts of biology by Indians taught in English and in Bangali. Another major purpose was to find out whether acquisition of the concepts was dependent on specific learner characteristics. This study will also give some information regarding whether the cultural background of the learner is a factor in concept learning. and the extent to which languages (Bengali and English) affect concept acquisition in science. Background to the Problem
Recent curricula in science rely increasingly on teaching
scientific processes and concepts rather than facts alone. A number of advantages are claimed for emphasizing conceptual knowledge. Since scientific knowledge is increasing and changing so rapidly. it is not possible for scientists themselves to keep abreast of changing "facts" in different disciplines of the sciences. Science students share the problem of learning and remembering specific scientific facts in all the sciences they are exposed to. Emphasizing scientific concepts
I




2
is one way to equip the learner with the skills to process factual knowledge in a rapidly changing world.
The values of concepts have been emphasized by leading
psychologists and educators. Vygotsky (1962) points out that the word that stands for a concept does not refer to a single object but to a group or a class of objects. Every word is therefore already a generalization. The acquisition of a concept therefore permits the learner to classify subsequent knowledge. to generalize beyond the knowledge given. and to reduce redundancy in the environment (Bruner. 1961). A concept is a type of summarizing "system" which provides the learner with the potential to deal with large amounts of new knowledge (Voelker. 1969). Being comprehensive in nature, concepts are useful to the learner in gaining some grasp of a larger field of knowledge than he has personally experienced. They permit the knower to interpret and assimilate new information into old systems through the modification of existing concepts (Pella. 19 66).
Concept learning not only permits more efficient information
processing but also results in a conservation of memory. Remembering a concept requires recalling the critical attributes that describe many dimensions of the concept. Once the concept is recalled, critical attributes can be reconstructed- In contrast. factual knowledge can be thought of as discrete, specific, disconnected knowledge which has been memorized in unrelated parts and is consequently easy to partially or completely forget. "Knowing a class of items reduces the necessity




3
of knowing each element of the class; accordingly. fewer concepts than facts need to be remembered" (J. J. Koran, 1971.. p. 408).
The domain of concept learning occupies a unique position
between elementary behavioral processes --such as stimulus -response learning, chaining, verbal association, and discrimination learning-and more complex behavioral processes like rule learning. problem solving. and creativity (Gagng. 1973). On the one hand, verbal association, chanining. and multiple discriminations are necessary to build concepts; on the other hand, these concepts combine to form principles and to provide the raw materials for problem solving and, subsequently, divergent thinking.
Definitions
A wide range of definitions of concept learning have been given by many authorities. In general, one can identify common elements in the following definitions.
Dictionary definitions try to distinguish the boundaries of a concept by indicating its genus, that is, what it has in common with other things and experiences, and its differentia. that is. the extent to which it differs from these things (Glaser, 1968). When the dimensions of a definition are overlooked, misinterpreted, or mistaught, then an error in the use of the concept occurs (Carroll, 1964).
A concept is a category which a person uses to classify the
stimuli he perceives (McDonald, 1965). In human beings, this category




4
is often a verbal one; that is, a word describes the name of a class of objects or events. such as "vertebrates" as opposed to "invertebrates" (J.J. Koran, 1971).
Concepts are properties of organizing experience --more
specifically they are the abstracted, and usually cognitively structured. classes of "mental" experience learned by organisms in the course of their life histories (Carroll. 1964).
A concept can be defined as "the common element shared by an array of objects" or "the relationship between the constituents or parts of a process" (George, 1962, p. 260).
A concept may be thought of as an artifact extracted by verification of the contexts or sentences in which it occurs (Bronowski and Bellugi, 1970).
"A concept is a generalized and abstract symbol; it is the sum total of all our knowledge of a particular class of objects. . In short, a concept is a condensation of experience"t MVaud. 1960, pp. 7576).
"The basic concepts are essentially high-level abstractions expressed in verbal cues and labels" (Taba. 1965, p. 465).
"A concept. . is something about an idea expressed in the words of our language" (Platt, 1963. p. 21).
All of the preceding definitions allude to the critical role that concepts play in school learning. At the same time, they suggest why one might expect the efficiency of concept learning to vary with




5
culture and language and to be sensitive to individual cognitive characteristics.
Types of Concepts
Psychologists, scientists, and science educators present at
least two related methods of classifying concepts. In Pella's taxonomy (1966), three types of concepts are elaborated: classificational, correlational, and theoretical.
Classificational. This type of concept is concerned with the
classification of facts. It facilitates the description of phenomena. An example of this type of concept is "insect. An insect is an animal with six legs and three major body divisions. The facts that contribute to characterizing an insect are placed together and used to distinguish them from spiders, crustaceans, and other arthropods.
Although there are many different kinds of concepts, classificational concepts probably represent the foundation, or knowledge base, on which more complex concepts and conceptual schemes can be built. Recognition of them by educators may be lacking because in common practice, they are taught through memorization (J. J. Koran. 1971), and, consequently, educators think of them as facts rather than concepts.
Correlational Concept. This type of concept is concerned with the correlation of facts. It facilitates prediction. An example of this type of concept is "force. A force is a push or pull that tends to change the motion of a body. Here force is the concept and requires other




6
concepts (push. pull. motion, change) to clarify it.
Theoretical Concept. This type of concept is concerned with the explanation of phenomena. An example of this type of concept is an "atom. An atom is the smallest particle of an element possible and consists of electrons, protons, neutrons, and other particles.
There are certain similarities, as well as distinctions, among these three types of concepts. Class ificational and correlational concepts are concerned with "abstractions from a field of direct experience, and they are "descriptions of human experience. whereas a theoretical concept is an "abstraction of a created idea,." and "explanation of human experiences" (Pella, 1966, p. 32).
Bruner, a psychologist, and his associates use a different
taxonomy with three classes of concepts: conjunctive, disjunctive, and relational. Each is related to a different mode of coupling attributes (Martorella, 1972).
Conjunctive Concepts. These can be defined as concepts with a set of common attributes that describe the concept. An example of which is a "square. a four-sided plane figure with four right angles.
Disjunctive Concepts. These are defined as concepts with either one or the other of two or more sets of attributes describing the concept, an example of which is a "cell. A cellis a structural and functional unit of living organism. A structure with a nucleus, cytoplasm, and cell wall is a cell (plant cell), or a structure without a cell wall




7
is also a cell (animal cell), or a structure without a nucleus is also a cell (human red blood cell).
Relational Concepts. Here a relationship between two or more attributes describes the concept, an example of which is "waste, the relationship between the given state of an object or class and the value that one places on it.
Conjunctive and disjunctive concepts provide a distinction usually made by psychologists to facilitate study under laboratory conditions. They may be thought of as being analogous to classificational, correlational, and theoretical concepts in school learning. Distinguishing Characteristics, of Concept Formation
Conceptual behavior involves two basic characteristic processes: generalization within classes and discrimination between classes (Mechner. 1965). Generalization and discrimination also form the basis for evaluating concept formation. Conceptual behavior also incorporates direct perception or perception through words and symbols (Glaser, 1968). For the most part. concepts probably occur hierarchically, with more complex concepts subsuming simpler ones (Gagne", 1973).
Several steps. models, and schemes for concept formation are given by authorities in the field (DeCecco, 1968; Gagng, 1973; Taba, 1966; Ausubel, 1968; J. J. Koran, 1971).
Briefly. one can identify the following steps in concept formation:




8
1) The essential characteristics of the concept must be
emphasized; 2) Both positive and negative instances of the concept should be made available; 3) The correct language for the concept and its characteristics should be established; 4) Proper sequencing of materials should be provided so that a student who is attempting to formulate one concept has had the prerequisite materials; 5) The generation of concepts should be encouraged, guided. and reinforced; and 6) Situations should be provided in which the concept can be generalized and discrimination between concepts can be made (J. J. Koran. 1971).
Research on Concept Formation
A large number of experimental investigations have been conducted in the field of concept formation using different variables. Clark (1971) points out that over the last 30 years there have been well over 250 experimental studies conducted in the area of concept attainment. Many of these studies need replication in order to identify and assess variables which are relevant to the development of concepts in school students. The following review emphasizes some major lines of research which directly relate to the problem of this study.
Effects of Different Variables that Influence Concept Formation Effects of Positive and Negative Examples
Opinion differs among practitioners and researchers whether to use positive instances or negative instances alone or both together in




9
concept instruction. A large number of research findings in this area are briefly summarized below:
Bourne (1966) found that subjects learn more efficiently from positive than from negative examples. Subjects show an inability or unwillingness to use information efficiently based on instances of what a concept is not (Bruner, Goodnow and Austin, 1956). Problems defined by positive examples are more easily solved than those defmed by negative examples; mixed sequences are of intermediate difficulty (Hovland and Weiss. 1953). Even at a time when the possible informational content of positive and negative instances is equaled, there is still an advantage in presenting positive examples (Hovland and Weiss. 1953).
Huttenlocher (1962) points out that negative instances in a mixed series of positive and negative instances can result in efficient concept formation. He proposes that subjects must be taught to process this type of information since it is not characteristically used in society. Frelbergs and Tulving (1961) show that, while practice time was longer for learning to use negative instances than for positive ones, subjects gradually used negative instances almost as well as positive. Information obtained from negative instances does not seem to be trusted (Bruner et al. 1956). although one can logically determine the minimum number of positive and negative examples required to define a concept correctly (Hovland, 1952).
Increased difference between positive and negative instances




10
facilitates concept acquisition. The degree of difference between positive and negative instances of a concept may be a more significant variable in concept formation than the form of these instances (Clark. 1971). The form of an example apparently determines the number of noncritical attributes that will be displayed or evoked by it; as this number decreases, ease of concept attainment increases. Hence, examples in verbal form, because they evoke fewer noncritical characteristics, increase the ease of concept attainment more than examples presented in visual form. e.g.. drawings, photographs (Clark. 1971). However Heidbreder (1949) disagrees with this finding. These findings are particularly relevant to this study, since the independent variables were presented in visual and verbal form.
A display of positive and negative examples in the same form
accelerates concept attainment to a greater extent than a display which allows form variation among instances. When examples are in verbal form. the degree to which positive examples evoke appropriate critical properties, and negative instances evoke necessary noncritical properties increases. ease of concept attainment increases (Clark, 1971). Effects of Redundant Relevant Information on Concept Acquisition
Bourne (1966) found that redundancy of relevant information accelerates concept learning, whereas redundancy of irrelevant information interferes with acquisition of a concept.
Irrelevant dimensions which interfere with concept formation have not received as much attention as relevant dimensions. Most




11
often it is the learner's task to find some kind of rule, or principle of internal structure. concerning the relevant dimensions that defne a concept.
The Effects of the Order and Sequence of the Examplars
The sequence variable has considerable influence on the
acquisition of concepts. However, these have not been investigated extensively. Hovland and Weiss (1953) studied sequence by presenting mixed orders of positive and negative examples and reported no effects. Detambel and Stolurow (1956) conducted an investigation concerned with change in the sequence of presentation of relevant and irrelevant dimensions from trial to trial. They concluded that a nonsequential presentation showed a striking superiority in contributing to the efficiency of concept learning. Anderson (unpublished paper) followed up this investigation using a longer series of transitions than adjacent trials and confirmed only part of the findings of the above study. He mentions that his study "extends the previous result by showing that the effective variable is not merely the number of relevant stimulus dimensions in the concept learning task, but the number of these that changes from trial to trial" (reviewed by Glaser, 1968. p. 46).
Deductive and Inductive Instruction
Almost all studies of concept learning utilize inductive behavior. That is. the instances are presented and the rule must be induced




12
(Glaser. 1968). On the other hand it appears that. in school learning, a concept often is taught by a deductive method. of presenting a rule and having students identify examples. Where time is not a factor or where the task is suitable, instruction may well be inductive. However, for some tasks, and with time economy an issue. deductive sequences appear useful.
Nature of Critical and Noncritical Attributes
When the difference between critical and noncritical attributes increases, east of concept attainment also increases. When critical attributes become more prominent and noncritical attributes become less prominent. ease of concept attaim-nent increases. When a critical attribute shows increasingly less variation from instance to instance, ease of concept attainment usually increases (Clark. 1971). Type of Concept Variables
Decrease in the number of relevant attributes (movement from a complex to a simple concept) generally increases ease of concept attainment. However, this finding is not supported by Laughlin (1966). Laughlin and Jorden (1967), Laughlin and Doherty (1967), as reviewed by Clark (1971).
These findings have direct relevance for school learning in that strategies for concept attainment must be selected that minimize distracting variables and maximize defining characteristics.




13
Conjunctive concepts are easiest to attain (Ciborowski and Cole. 1972) while disjunctive concepts appear more difficult.
'Ihe type of concept introduced may be a more significant variable in concept formation than the method of presentation (Clark, 1971).
Again these findings are particularly relevant for experimental studies on concept formation in school subjects. In order to test the efficacy of different instructional materials, it is essential to select concepts that lend themselves to experimental studies in school setting. Response Variable
Ease of concept formation and transfer to new concepts is
presented along a continuum from high to low when students 1) verbalize critical characteristics only. 2) verbalize critical and noncritical properties, 3) do not verbalize, 4) verbalize noncritical properties only (Clark. 1971).
When the student is able to verbalize both critical attributes and the concept name and to verbalize the concept name throughout the concept task, the ease of concept attainment and degree of retention increase (Clark, 1971).
When the student is allowed to physically manipulate instances, ease of concept attainent increases more than when he is not allowed to do so (Clark. 1971).
Theory and Research on Pictorial Presentation
An instructional medium as defined by DeCecco (1968) is




14
only a means of transmitting instruction and not the substance of that instruction.
At present, a large number of media are available for instruction. This introduces the practical problem of teachers making an appropriate selection among media at a time which is useful in meeting specific instructional objectives.
The following criteria could be used by teachers for the adequate selection of media: [11 "information about available media; [21 analy sis and design of an instructional system; [ 3] knowledge of the research findings on the use of the media" (DeCecco. 1968, p. 529).
Gestalt psychologists have proposed that the brain has built-in
systems for organizing visual information. In contrast to this proposal it was argued that many principles of visual organization reside not in the brain, but in stimuli themselves (Gibson, 1950. 1966). Modern scientific views of how one looks at pictures (Neisser. 1967) generally agree that seeing is basically a constructive experience. That is, if one could slow down the process of picture looking and study it in detail, one would discover that the mind puts together, in a unified way, selected information brought in by the eyes. The first part of the pictorial comprehension process is an analytical one--one has to locate and select the critical features of what is being observed.
Visual illustrations are becoming increasingly widespread as a means of instruction. However, comparatively few attempts have,




15
subsequently. been made to determine the relative effectiveness of various types of visual illustration. Supposedly, not all types of visual illustration are equally efficient in accelerating the learning of different types of educational tasks.
Holliday (1973) suggests that it is more likely that specific types of pictures facilitate the learning of specific types of objectives for certain students with certain kinds of characteristics. However, definite relationships have not yet been established. As a result, instructional designers have had to depend largely upon intuition in trying to match instructional systems with learner types. One result of these complex relationships and problems is that many investigators examining the contribution of pictures to verbal materials have concluded that the type of visuals investigated did not significantly improve criterion achievement scores.
Bruner et al. (1956) and Travers et al. (1964) arrived at the conclusion that learners do not need a wealth of stimuli in order to recognize the attributes of an object or situation which place it in a particular category. It is suggested by Travers et al. (1964) that there is no guarantee that useful information will be retained if a person is confronted with stimuli similar to those emitted by the real environment.
Travers et al. (1964) adds that the realistic presentation of much content provides unnecessary detail. and that the real objective of visual education is not so much to bring the student into close touch with reality,




16
but to help students become more effective in dealing with reality. They contend that this can be done effectively with symbols. Broadbent (1958. 1965) describes one reason why reduction of learning occurs as cues increase. According to him. it is caused by a filtering process in the central nervous system which prevents many realistic stimuli from receiving active reception in the brain. This assumption is supported by Jacobson (1950. 1951). who said that the brain is capable of utilizing only minute proportions of information perceived. This point of view is supported by Travers et al. (1964, p. 520): "the amount of information which is utilized by the higher centers is vastly less than the informational capacity of the channels involved. "
A study conducted by Attneave (1954, pp. 185 -86) to test the hypothesis that a function of the perceptual machinery is to reduce redundant stimulation and to encode incoming information so that only essentials travel through the nervous system to the brain. To support this contention he mentions that "lines bordering objects supply the essence of the information to be conveyed, explaining the effective ness of cartoons and stick drawings as conveyors of information.
If this is true, the proposal of Travers et al. (1964) that visual information is stored in the nervous system in some form isomorphic with line drawings. thereby permitting the individual to remember and reproduce such information with greater facility than more realistic information, gains more credibility. One could interpret this as




17
meaning that these visuals closely representing line drawings and containing the essence of information to be transmitted would be more effective and more efficient in facilitating learning than would detailed illustrations. which have to be initially coded by the central nervous system before being transmitted. This proposition is relevant for this investigation because single-line pictorial treatment materials were used in order to facilitate concept acquisition in school subjects. Travers et al. (1964) explain that inputs of information when received by the subjects are coded and most of the original stimuli initially presented to the senses not only never enters the perceptual system, but is not remembered by the system.
Both theory and research seem to be contradictory to the assumption of "realism theories" that presenting a subject with a large number of stimuli which approximate "reality" is not necessarily the most effective way to accelerate learning. Because these excessive stimuli may interfere with the transmission of information, and some of the stimuli may not be perceived, researchers should make some effort to find those specific characteristics of visuals that will enhance particular types of learning (Dwyer, 1967).
The most comprehensive study of visual illustration as an adjunct to instruction in science education has been conducted by Dwyer and reviewed by Holliday (1973). Dwyer studied the relative effectiveness of eight visual types in relationship to numerous instructional and other categorical variables. During the last five years. he reported




at least thirty-two separate studies using similar treatments, experimental designs, statistical analyses. and evaluative criteria. In his investigations, Dwyer commonly used as illustrations cross sections of the human heart including the following visual types: 1) simple-line drawings. 2) detailed shaded drawings. 3) photographs of a heart model. and 4) photographs of an actual heart. Black-andwhite and colored pictures were also used.
Dwyer drew the following conclusions from his investigations: 1) no sex differences were observed. 2) color had a differential effect on achievement. 3) treatment differences were not observed on delayed retention tests. 4) student choices for particular types of visuals were not reliable indicators of relative successfulness as measured by the criterion tests. 5) the use of verbal questions related to visuals did not seem to be helpful. 6) differing size images of visuals used in the investigations did not appear to enhance learning differentially. 7) small amounts of realistic detail. usually via single -line drawings. formed a more effective visual aid when the subjects received grouppaced instruction. via slides and television. After studying the differential effects of visual types on high school students, Dwyer arrived at the conclusion that verbal presentation is most effective in terms of "effectiveness, economy and/or simplicity of production" (Dwyer, 1972, p. 24).
Without any doubt, Dwyer has made a broad spectrum of contributions in the area of visual illustrations. However, some replication




19
is needed for practical application in classroom instruction. In this regard. the following research must be considered: Vernon (1950, 1951, 1953, 1954) studied the effectiveness of graphs and pictures as adjuncts to verbal lessons and concluded (1953, 1954) that pictures generally contribute little to general comprehension of verbal material. Burdick (1960) arrived at the conclusion that pictures used to complement the verbal instruction do not facilitate reading comprehension. These two investigations suggest that perhaps pictures do not facilitate learning. In a similar study, both a graph and a map proved to be a significantly more effective visual advanced organizer than a verbal expository treatment (Weisberg, 1970. as reviewed by Holliday, 1973).
The optimal size of a picture has been studied separately by Moore and Sasse (1971), and Dwyer (1970). and it depends on the objectives. Larger images inhibited student achievements for certain objectives (Dwyer. 1970). Dwyer suggests that the inhibiting or limiting factor in the instruction was the amount of time the subject was permitted to view and grasp relevant cues. However, any relationship between specific size of a picture and specific objective is not yet established. A study conducted by Chan, Travers, Van Mondfrans (1965) supports the hypothesis that color is a complex characteristic and is not necessarily a positive attribute for a visual.
Research findings (Bruner and Potter. 1964; Dwyer, 1967)
indicate that pictorial cues stimulate a rich flood of mental associations.




20
Under certain conditions it may be difficult for a student to consolidate selected associations into a single. clear concept. thereby necessitating that the teacher choose visual aids with extreme care (FarnhamDiggory, 1972).
. "Pictures should be introduced into a curriculum for the purpose of expanding and enriching associations, rather than constricting them. . Pictorial comprehension involves the arousal and consolidation of personal ideas that is both its power and its danger" (FarnhamDiggory, 1972,.p. 432).
There are many more areas in visual illustration where further investigation needs to be done. Investigation regarding the relationship between different aptitudes and the picture types used in science instructional materials should provide information applicable to individualized instruction (Holliday, 1973) and should explain many of the nonsignificant or conflicting results previously reviewed. Theory and Research on Aptitude Treatment Interactions
It has been known for a long time that individuals differ from each other with regard to mental abilities as well as in physique and traits of personality. However. only recently have educators devised a methodology of research to correspond with their concern for this phenomena. Instructional methods effective for training large numbers of individuals possessing different ability patterns have become a major concern of present-day educators. One way of achieving the




21
goal of individualizing instruction, as proposed by Fuller (1969). Hereford (1971). and Merrill (1968), is to vary educational objectives. pacing. and sequencing according to student characteristics. Another appropriate approach, as proposed by Cronbach (1967), concerns varying instructional methods for different subjects to reach the same educational goals. As researchers point out, a subject learns more easily from one method than another, and this best method differs from subject to subject. Such differences between treatments are related to learner characteristics. As a result, a larger number of subjects may be expected to achieve the instructional objectives when instruction is differentiated for different types of subjects (Cronbach and Snow, 1969, reviewed by M. L. Koran, 1972). Cronbach (1965) developed a theoretical framework to deal with the nature of different aptitudes and differing instructional treatments. He named these studies aptitude treatment interaction studies, or more commonly ATI studies.
Generally speaking, an interaction is present when one treatment is significantly better for one type of subject, while an alternative treatment is significantly better for a different type of subject. Particularly, the presence of ATI depends on nonparallel regression slopes of aptitude on performance for different instructional methods (M.L. Koran. 1972). "The general objective of ATI research is to match specific instructional methods or materials to selected learner




22
characteristics (M.L. Koran, 1972, p. 136).
Two different types of interactions may be obtained. either
ordinal or disordinal in nature (Cronbach and Snow, 1969). An interaction in which the regression slopes intersect within the range of the aptitudes being considered is known as a disordinal interaction. An ordinal interaction is present when regression functions have different slopes, but one is superior to the other through the range of aptitudes being considered.
In ATI research. aptitude is defined as any characteristic of the learner that functions selectively with respect to learning. Or, in other words, it facilitates or interferes with his learning from some designated instructional method (Cronbach and Snow, 1969).
Large numbers of student characteristics might be considered in adapting instruction to individual differences. But, there are only a few studies which have investigated the possible relationships between perceptual aptitude tests and different kinds of pictorial and nonpictorial instructional treatments. Nevertheless.* significant interactions between certain perceptual aptitudes and two instructional treatments (a film-mediated and written model), concerned with modeling procedures in the acquisition of a teaching skill, have been observed (McDonald and M. L. Koran, 1969).
In general, much more research is needed to identify the kinds
of abilities and processes necessary for different tasks. It is necessary to find out whether students with particular characteristics are able




23
to master particular tasks and under what conditions. This kind of relationship would provide us with the information for individualizing instruction by aptitude factors other than the differential rate of human learning.
Further investigations involving the relationship between different aptitudes and visual illustration in science instructional material should shed some light on appropriate uses of visuals for particular student characteristics and educational outcomes.
Theory and Research on Language and Cognition
The relationship between language and thinking is intrinsic
to the nature of intellectual growth. The child's ability to manipulate words is a reliable indicator of his thinking capacity. If his vocabulary is limited, his capacity for elaborating perceptions into concepts-the general process known as thinking--is blocked (King and Kerber, 1968).
Language is a dynamic entity--a critically significant way of handling information. One can represent the world by actions and images only to a limited extent. In order to represent it flexibly and powerfully, one must have symbols for it, and language supplies these symbols (Farnham-Diggory, 1972).
The universal nature of language development has also been emphasized by Farnham-Diggory (1972). It seems that all children begin to speak in about the same way at about the same time.




24
regardless of nationality. The characteristic features of this universal developmental system may give some important clues to a basic human learning process.
Language functions as a symbol system and as a process of
thinking, without which human education is inconceivable. The growth and education of language capacities is almost another way of representing the growth and education of the mind (Farnham-Diggory, 1972). Language and Its Influence on Concept Learning
The influences of language on concept formation constitute an important aspect of school learning. Words represent concepts that may have been learned preverbally or with the help of verbal labels. Concepts have been defined by some psychologists as meaningful words which label classes of otherwise dissimilar stimuli (Archer, 1964). Most studies designated as concept learning by psychologists have been limited to laboratory situations involving direct, nonverbal dimensions (Glaser, 1968).
There is evidence which suggest that the ability to use words is an important factor in the rate of concept acquisition (Glaser, 1968). Dietze (1955) reports that children learn to categorize nonconventional forms faster when distinctive names are given to them. Verbalization, when compared with nonverbalization, enhances concept learning in four-year-olds, but makes insignificant difference for seven-year-olds because they tend to verbalize the solution. However




25
Jenson (1966) found that, when instruction is given to seven-year-olds to verbalize the irrelevant dimension, it delays the learning of a reversal shift. The correlation between verbalization of a rule and correct responding is not clear. Green (1955) reports, from a study with college students, that verbalization covaried with correct responses but was not necessary to making the correct response. Kendler and Kendler (1962) suggest that, in studies with children, verbalizations are not always a guarantee that choice behavior will be appropriate (reviewed by Glaser, 1968).
The discussion of language in the study of concept formation and its relationship to bilingual learning of science concepts brings up the question of the definition of stimulus dimensions involved and emphasizes mediational processes. Evidently there is a difference between children and adults in performing solution shifts in concept problems, which probably depends on prior verbal habits. The nature of the verbal repertoire which exists prior to concept learning determines the way the concept is learned and, consequently, the way in which it can be taught (Glaser. 1968). Experiments in this area can be designated generally as examples of the way in which a higher-order conceptual system, once taught to subjects. may make it possible for them to do a kind of thinking they could not otherwise do (FarnhamDiggory. 1972).
Benjamin Lee Whorf, linguist and anthropologist, developed a hypothesis known as the linguistic relativity hypothesis. Whorf




26
derived this complex hypothesis after extensive studies of the Hopi and Shawnee languages. He proposed that it was not just the words and values of the cultures that were different; what the peoples perceived was itself different (Keele. 1973). He concluded that the thought content of these peoples differed very much from that of English-speaking Americans.
To paraphrase Wharf'Is hypothesis, the world of sensation is really a continuum. However, in dealing with large numbers of sensory events, people learn to categorize the world. Stimuli which are close together on a continuum evoke in the memory the same categorical representation. The human's limited processing mechanism is then able to deal with categories rather than a multitude of sensory events. People in different cultures categorize the world of sensations in different ways. and the categories are reflected in the languages of the cultures. The categorizations, in turn. determine what is perceived (Keele. 1973). If so, instructional methods incorporating pictures and/or prose necessarily have to be designed to correspond with cultural peculiarities.
Cultural Influence
Sapir (1960). from a broad study of languages, did not arrive
at the conclusion that cultural (or racial) distinctions of any sort could reliably be attributed to language. There are plenty of instances of similar cultures (e. g..* among Indian tribes) having different languages.




27
Again, there are examples of dissimilar cultures having similar languages (e. g.,* among English-speaking people). It is also Sapir's contention that any language is constructed in such a way that it can respond to the intentions of any speaker, and that any language system can be translated into another. He also mentions that some language forms might be used in a language long before the ideas they contained were consciously recognized by the speakers of the language. For example, perhaps the concept of causation may be expressed quite unconsciously in the languages of primitive societies that have not yet developed formal, scientific concepts of causality (Farnham -Diggory. 1972). The implications of the above alternative interpretation of the language -culture relationship for this study are dubious.
Bilingual Children
Children who learn two languages more or less at the same time have been able to capitalize on their own natural language -learning powers. Two basic questions regarding classroom relevance can be put forward:
1) Do two languages improve one's thinking ability?
2) Should one teach the second language the way it is naturally
learned by bilinguals, that is, by simply putting the child into a second-language environment and letting him learn
according to his ability?
It is suggested by Farnham-Diggory (1972) that the answers to these questions should be in the affirmative, with some qualification. A general notion has been that knowing two languages may set up a




28
kind of interference or linguistic "static, which would have an inhibitory effect on learning ability. On the basis of recent investigation, Peal and Lambert (1962) disagree with earlier evidence supporting the interference hypothesis.
Kolers' (1963) investigations, which are more technical, suggest reasons for the above findings--for example, that separate verbal memories may exist for each language. Peal and Lambert concluded that bilingual children (approximately ten years old) were significantly superior to monolingual children on many different measurements of reasoning ability.
Intellectually. the bilingual's experience with two language
systems seems to leave him with a mental flexibility, a superiority in concept learning. and a more diversified set of mental abilities, in the sense that the patterns of abilities developed by bilinguals are more heterogeneous. It is not possible to state, from the present investigation, whether the more intelligent child became bilingual, or whether bilingualism aided intellectual development, but there is no doubt of the fact that he is superior intellectually.
However, one could propose that, because of superior intelligence, bilingual children are also further ahead in school than the monolinguals. Superior achievement of these children in school seems to be dependent on a verbal facility (Peal and Lambert, 1962, p. 20).
These findings have a special significance for this study since




29
all the experimental subjects were exposed to three languages. Hence the findings of this investigation may shed some light on whether experience with two or more languages aids in the acquisition of scientific concepts.
Rationale for This Study
India is a diversified country. This diversification is reflected in her cultures, languages, and in many other ways. The country is not united through a common language, although English was, and still is. used to overcome this condition to some extent. Histo-ically speaking, the English language came to India with the British and has Dow become a factor in the language controversy. After independence, many Indian educators, scientists, and common people felt strongly that English should be abandoned and Indian languages should be adopted in the educational system and other official, business, and administrative systems of the country.
Hindi was declared to be the national language of India just before independence (Chatterjee, 1954). but it has not yet been completely implemented throughout the nation. According to the census taken in 1961. there is a total of 1.,652 mother tongues in India. which have been grouped into languages /dialects. There are 14 major languages (which includes 380 mother tongues) specified in Schedule VIII of the constitution (India, 1973). It was mentioned by DasGupta (1970) that language diversity has always been a characteristic feature of India.




30
This diversity creates a tremendous communication problem between the average people of different states and complicates educational processes. The Indian government has been, and still is, trying to solve these problems.
The educational aystem in India is bilingual to some extent. The regional language of the state of West Bengal is Bengali. Previously, the study of three languages, that is, Bengali, English, and Sanskrit, was compuilsory in this state. Recently, in most schools, Sanskrit. classical language, has been replaced by Hindi in order to implement the thrme-language formulas proposed in the National Policy Resolution of 1968 (India. 1971-2). In West Bengal, the medium of instruction in most primary and secondary schools is Bengali. However, there is another set of schools where the medium of instruction is exclusively English from the elementary level. The student population in these schools is more heterogenous in that a broader spectrum of students from different backgrounds attend these schools. Most of these students speak in English at school but converse in a mother tongue at home. There is still another type of school which falls in between the above mentioned categories. where the medium of instruction at the elementary level is Bengali and of the secondary level is English.
The situation is quite flexible at the University level in Calcutta, the capital of West Bengal. Lectures are given either in English or in Bengali or in a mixture of English and Bengali. The latter is the most commonly used technique. It is also permissible for students to answer questions in the mother tongue in all subjects except English.




31
However, it is also accepted that a working knowledge of English will be an asset for all students and a reasonable proficiency in the language will be necessary for students who will go for higher education. Hence, English should be allowed to continue, placing, however, greater emphasis on the mother tongue (Chatterjee, 1954). "Virtually the right of the Indian boy or girl to be instructed in his or her mother tongue, or in whatever language he or she chooses in its place, has been admitted both by the general consensus of opinion and by legis lation" (Chatterjee, 1954, p. 6).
At the Poona University linguistic conference held in 1953,
it was unanimously recognized that the education of a child must begin in the mother tongue. Also, instruction and evaluation at all levels up to high school should be in the mother tongue. This view had already been accepted by the University of Calcutta (Chatterjee, 1954).
The regional languages are already in use as media of education at primary and secondary levels. Many educators are now emphasizing education entirely in the mother tongue at colleges and universities. Chatterjee (1954) mentions that this will inevitably lead to an undesirable condition, that is, the formation of linguistic states.
There w ill be a great damage in the intellectual domain if
English is abandoned at the high school stage, which leads to the college. The replacement of English with regional languages will not only isolate people of one region of India from another, but it will also isolate India




32
from the rest of the world. English is one of the most important European languages or the "world language. and, since it is already being used in India, it is relatively easier for Indians to learn it than any other foreign language. Hence, Chatterjee (1954) proposed that English should be retained side by side with the mother tongue in our university education.
At the Poona linguistic conference, held in 1953. the following
recommendations were made regarding the use of international scientific symbols and expressions:
1) "All technical terms should be drawn as far as possible
from Sanskrit sources.
2) "All international symbols should be retained.
3) "International scientific terms and expressions should be
retained if Indian equivalents cannot be framed" (Chatterjee.
1954. p. 14).
There are also a large number of reactions adverse to retaining
English. Some of them seem to be quite legitimate and cannot be ignored as such. It was suggested by Krishna (1974) that there is a psychological disadvantage in using English as a medium of instruction. For example. when children are taught to memorize numbers in English. the objective of learning to count becomes secondary to the misplaced concern of learning to count in English. "From the very beginning, knowledge of arithmetic or mathematics becomes subsidiary to a knowledge of English. A student has to spend a large amount of time in mastering a




33
foreign language instead of mastering the subject content (Krishna. 1974. p. 20) even when the students are not interested in literature or linguistics (Mathai, 1974). When students are instructed in any science subject in English, they usually become more conscious about the Language than the subject matter content. On the contrary, if they are to be taught in a mother tongue, consciousness of medium would play an insignificant part and thus permit them to devote more attention to the subject matter. Krishna (1974, p. 21) mentions, "in considering the causes of our scientific backwardness, a major factor that would emerge is the lack of involvement of young minds in science because it is taught in a foreign language that is not fully understood. Many Indian educators complain about the absence of good texts in our own language. Krishna (1974) refuted this criticism by suggesting that this is a question of time, and in due course we will be able to develop good texts in our own languages.
In addition to this language conflict. there are many other problems, drawbacks, and difficulties in teaching science in India. Age-old traditional curricula, testing, and evaluative procedures still exist at most places. Science is taught mostly by using traditional approaches. The major emphasis is on theoretical aspects rather than on practical implications of science. The inquiry or discovery approach is mostly unknown to students. Science teaching is not related to the environment.
The present status of science in India has been pointed out by Vaidya (1971).




34
1) Science instruction has been and is still oral in character
and includes only very few demonstrations. Students are not involved in doing practical work in the elementary schools. After this, students are to follow rigidly a prescribed list of experiments. These experiments are mostly in the nature of verifying knowledge. or working according to set rules. 2) Science instructions is based strictly on prescribed textbooks. which are usually not of high quality. 3) Differentiated and sequential curricula are lacking for various categories of students. 4) Methodologies of science instruction are traditional and nonstimulating. 5) Teacher training programs are also inadequate, and "there is hardly any supervision and availability of expert advice in science teaching by those who are really competent, 6) There has been very little research conducted in the field of science education. There is a "complete blackout of research in science education" (Vaidya, 1971..p. 30).
Since independence, efforts have been made to improve science
instruction in India. Government of India has undertaken many programs to upgrade our scientific knowledge. To improve science instruction in India, the following three integrated steps are essential (Vaidya. 1971).
1) Development of a curriculum which will include modern
concepts and understandings of the subject-fields and a
rigorous, analytic study of fundamentals.
2) Preparation of new textbooks. teacher's guides/manuals.
and other instructional kits and apparatus in order to satisfy




35
the demands of the new curriculum.
3) Establishment of a new teacher training program in order
to equip the teachers with skills required for the new
curriculum.
A great deal of research is needed to be done in order to improve science instruction in India. To develop a conceptually oriented new curriculum, according to the need for the society, some empirical evidence is essential in order to find out the type of specific concepts to be included for a particular group of students with particular language -cultural backgrounds.
Similarly, evidence is needed regarding how students learn from different media and what are the essential factors that facilitate concept acquisition and the acquisition of other types of technical and nontechnical materials. The latter area, in addition to the medium of instruction, poses a major problem in science teaching in India. The general notion that students should be taught in the mother tongue from the beginning seems to be very reasonable. It is generally assumed that students will be able to think, process information and express themselves much better in their mother tongue than in any foreign language. However. before someone makes an attempt to replace English, either by Hindi or by any other regional Indian language, some basic investigation is essential in order to find out whether the particular language has the capacity to express clearly and efficiently all the ideas and thoughts of science and technology. Decision making




36
should be based on empirical evidence.
An attempt has been made in this investigation to find out the relative effectiveness of English and Bengali instruction and pictorial or written presentation in acquisition of three scientific concepts by Indian students with different language and cultural backgrounds. Experimental Questions
Based upon the previously reviewed research and theory, the following questions were asked:
1) What effect do written passages have upon acquisition of
a scientific concept?
2) What effect do written passages have on retention of a
scientific concept?
3) What effect does a single-line pictorial presentation have
on acquisition of the same scientific concept?
4) What effect do pictorial presentations have on retention
of a scientific concept?
5) What differential effect does language (English or Bengali)
have on acquisition and retention of a scientific concept?
6) What effect do individual differences have on the acquisition
and retention of scientific concepts?




CHAPTER II
EXPERIMENTAL DESIGN
The Design
A post-test only control group design (Campbell and Stanley, 1971). as described in Table 1, was used with four experimental and two control groups. This design efficiently controls for pre -test sensitization, but it does not provide information about entering behavior. With a large sample, as in this study, one can assume that the entering behavior of subjects is randomly distributed over groups and hence subjects are equivalently distributed over treatments. This design was selected because it permits an adequate evaluation of the relative effects of an independent variable upon a dependent variable with control for major threats to internal validity. Aptitude testing preceded the administration of each treatment. Independent Variables
Written Passages. Written passages about three classificational concepts in biology were described in a small booklet. The concepts used were: geotropisms, phyllotaxy. and feathers. This booklet constitutes one of the independent variables.
37




TABLE 1
EXPERIMENTAL DESIGN
Aptitude Tests Treatment Post-Test Retention-Test
Letter Sets
Vocabulary R X 1 Written passages in 0 107
Bengali17
First and Last R X2 Single-line pictorial 0 2 08
Names presentation labeled
in Bengali
Auditory Number R None Placebo (control) 03 0 9
Letter Sets
Vocabulary R X3 Written passages in 04 010
English
First and Last R X4 Single-line pictorial 0 5 0 11
Names presentation labeled
In l1nth
Auditory Number R None Placebo (control 0 6 012
Span
CA




39
Pictorial Presentation. Single-line drawings of the same three biological concepts with appropriate labeling were included in another booklet. This booklet constituted the other independent variable.
Aptitude Measures. The following aptitude measures were used for this investigation.
1) Letter Sets Test UI-1)
2) Vocabulary Test (V-i) and (V-2)
3) First and Last Names Test ('Ma-3) 4) Auditory Number Span Test (M-%s-i) Ability Measures
The selection of ability measures used for this study was made from the Kit of Reference Tests for Cognitive Factors (French, Ekstrom and Price, 1963). The abilities assessed by the selected measures were thought to influence learning from the two modes of information types presented in the treatments.
The Kit of Reference Tests for Cognitive Factors consists of groups of tests representing frequently obtained factors in the area of cognitive ability. These tests are not standardized batteries of tests, and the manual does not provide measures of reliability, validity norms, or other information commonly included in a test manual; for this reason, the tests are recommended for experimental use only. A brief description of the ability tests used in this study and the factors they represent follows.




40
Letter Sets Test (1-1)
A test of divergent production of symbolic classifications (DSC) is thought to be concerned with the ability to group items of symbolic information in different ways. In this test students are supposed to indicate different common properties that sets of letter combinations may have in common (Meeker, 1969).
Divergent Production: Generation of information from
given information, where the emphasis is upon variety and quality of output from the same source. Likely to involve
what has been called transfer. This operation is most clearly
involved in aptitudes of creative potential. (Meeker, 1969, p. 20). Vocabulary Test W-1) and (V-2)
These tests measure the student's ability to understand the
English language. Individual differences are perhaps most obviously seen in the size of comprehensive vocabularies.
First and Last Names Test (Ma-3)
The first and last names test is a memory test. Memory can be defined as "retention or storage. with some degree of availability, of information in the same form it was committed to storage and in response to the same cues in connection with which it was learned" (Meeker, 1969, p. 16). Memory is a well known intellectual abilitywell known in the sense that it is one of the oldest. Memory is also universally and historically recognized as a primary mental operation (Meeker, 1969).




41
This test requires students to remember what kind of last name goes with first name and to recall perfectly for immediate reproduction. Auditory Number Span Test (Ms -1)
Memory for symbolic system (MSS) concerns the ability to
remember the order of symbolic information (Meeker. 1969). This test involves immediate recall of series of numbers after only one auditory presentation of the series. Table 2 identifies each test and the time required for its administration. Dependent Variables
Acquisition of Concepts. A post-test was developed corresponding to the content materials and was used as a criterion measurement. It contained thirty items; of these, five items dealt with Concept I (geotropism). seven items with Concept II (phyllotaxy). and eighteen with Concept M (feathers).
Retention of Concepts. An equivalent retention test to the posttests was also prepared from the same set of instructional objectives.
Methods
Subject
A total of 302 Indian students were selected from five different high schools in Calcutta to participate in this study. These schools were selected on the basis of availability, cooperation, and spoken language. The subjects were all science students. They had just




TABLE 2
APTITUDE MEASURESa
Number of
Test Total Time Parts Given
Kit of Reference Tests
for Cognitive Factors
Letter Sets (I-I) 14 minutes 2
Vocabulary (V-1) 8 minutes 2
Vocabulary (V-2) 8 minutes 2
First and Last Names (Ma-3) 10 minutes 2
Auditory Number Span 4 minutes (approx.) 1
Practice Time 1 minute
aAll tests were administered during one setting.




43
finished ninth grade and had been promoted to tenth grade. A total of 169 students were selected from Bengali medium schools. and 133 students were selected from English medium schools. However. due to differential experimental mortality (i. e. different students were absent at different testing times, etc. ) and the statistical procedures used. only 204 students who took all the tests constituted the final sample for this investigation. A total number of 102 students was used from the Bengali medium schools, and the same number from English medium schools.
Treatment Materials
Written passages about three biological concepts were included in a small booklet. These concepts were: geotropisms. phyllotaxy. and feathers. The textual information was selected from standard biology books. The concept of feather was described more elaborately than phyllotaxy and geotropism. Critical attributes of each of the concepts were highlighted. The pictorial presentation of the same concepts was included in another booklet. This included single-line black-and-white simple drawings. The critical attributes of each of the concepts were labeled. These treatment materials were written and labeled in-English for a pilot study conducted by the author at the P. K. Yonge Laboratory School of the University of Florida. The same materials were later translated in Bengali by the experimenter, with the help of some authorities in the field, for the Bengali medium schools. Hence, the




44
study wbich followed the pilot sought to replicate findings from the pilot. done with American students, on. Indian students taught in Bengali and English.
Procedures
Aptitude tests were given to all students seven days prior to the treatments. Directions for the administration of each test were followed according to the test manual. Different parts of this test and the directions were included in a small booklet. The test was administered during one sitting. For the First and Last Names Test, English names were replaced by Bengali names, keeping the same number of letters. For example. instead of the first name Edward, Kamala was used. The same rule was followed for other names.
In the first part of this experiment, all students in each class room of Bengali medium schools were assigned randomly, using a table of random numbers, to one of three groups: written group, pictorial group, control group. All students in the control group were moved to one side of the classroom, then treatments were administered to the treatment groups. Treatment time, forty minutes, was kept constant for all groups. After the treatments all students were given the post-test, with constant testing time. Students were given a time period of twenty minutes to finish the test. After the post-test, ten minutes was given to the students to write information about themselves on an information sheet. The retention-test was administered seven days after the treatment, again with constant testing time. Twenty




45
minutes was given to complete the test. The same procedures were followed for the English medium schools, with the exception of an unavoidable time gap of four days between the post-test and retentiontest.
The Instrument and Its Reliability
The post-test and retention-tests were prepared to correspond with content material and were included separately in small booklets. These tests were prepared from the same set of objectives. Each test consisted of thirty items. Five items were selected from the concept geotropism. seven from phyllotaxy, and eighteen from feathers. Two types of items were chosen: fill in the blanks and multiple choice. Subsequently, the tests were evaluated and found to be equivalent. A point bi-serial correlation was employed to evaluate the reliability of each item. The range of point bi-serials varied from -0. 014 to 0. 799 for the post-test in Bengali medium schools and for the retentiontest, from 0. 101 to 0. 681. The mean difficulty for the post-test in Bengali medium schools was 0. 427, and for the retention-test, 0. 416. Difficulty, here, is defined as the percentage of subjects failing to answer the item correctly. The Cronbach alpha of the post-test used in Bengali medium schools was 0. 89, and of the retention-test, 0. 87.
The point bi-serial for the post-test in English medium schools ranged from 0. 193 to 0. 745, and for the retention-test from 0. 228 to
0. 692. The mean difficulty of the post-test was 0.418, and of the




46
retention-test. 0. 389. The reliability coefficient for the post-test ts 0. 88, and for the retention-test. 0. 88. Data Collection Procedures
Experimental subjects recorded their responses in three
separate booklets. The first booklet contained aptitude measures. the second one post-test, and the third one retention measures. Appropriate student information data sheets were attached with corresponding booklets. These tests were bandscored. and the data were transferred to a data summary sheet. Points were given for correct responses; no penalty was given for incorrect responses. The necessary data were then transferred to an IBM data card for statistical analysis. Key punching was done by the experimenter and by trained technicians. Copies of different booklets including treatment materials and the instruments are included in Appendices A, B, C and D. These manuals were included in the original form as they were used for the experiment.
Table 3 summarizes all of the independent and dependent variables in this study.




TABLE 3
INDEPENDENT AND DEPENDENT MEASURES Independent Measures Dependent Measures
Written Passages Post-test measures includes
Pictorial Presentation Relevant question from written
passages and pictorial presentation which include:
Letter Sets Test (1-1). Part 1 Concept I (geotropism)
Letter Sets Test (I-1). Part 2 Concept 11 (phyllotaxy)
Vocabulary Test (V-l). Part 1 Concept III (feather)
Vocabulary Test (V-l). Part 2 Total achievement
Vocabulary Test (V-2), Part 1 Time taken for post-test
Vocabulary Test (V-2). Part 2 Retention-test measures include:
First and Last Names Test (Ma-3). Part 1 Concept I (geotropism)
First and Last Names Test (Ma-3), Part 2 Concept II (phyllotaxy)
Auditory Number Span Test (Ms-1). Part 1 Concept III (feather)
Total achievement
Time taken for retention-test




CHAPTER III
RESULTS
Purpose of Study and Data Collected
The major purposes of this investigation were:
1) To assess the relative effects of pictorial and written
presentation upon the acquisition of three biological concepts.
2) To find out if there is any relationship between student
characteristics and different modes of instructional content.
This chapter is a description of statistical procedures and the results obtained in this study. The results from the instructional treatment main effects will be discussed first and then aptitude treatment interactions.
Independent and Dependent Measures
Independent measures included written passages, pictorial presentations, and nine aptitude measures. Dependent measures were the immediate post-test measures for the acquisition of three scientific biological concepts, delayed retention measures and time used.
Means and standard deviations of both independent and dependent variables were computed. Table 4 shows the means and standard deviations of three groups of subjects from Bengali medium schools. It is evident from this table that the means of Ss in the written group
48




49
TABLE 4
MEANS AND STANDARD DEVIATIONS OF THE THREE GROUPS (BENGALI MEDIUM SCHOOLS) Standard Number of
Treatment Groups Mean Deviation Measurements
Post-Test Measures
Written Group 22.29 3.86 34
Pictorial Group 20.38 2.74 34
Control Group 8.94 2.74 34
Retention-Test Measures
Written Group 22.06 3.12 34
Pictorial Group 19.85 2.98 34
Control Group 10.32 2.91 34




50
(22. 29) on the post-test is not much different from Ss in the pictorial group (20. 38) on the post-test. However. both of them are much higher than the Ss in the control group (8.94) on the post-test. This trend is also found on the retention-test. Table 5 represents the means and standard deviations for three treatment groups obtained from Ss from the English medium scilools. Here also. the means of Ss in written treatment (22.15) on post-test and of Ss in pictorial treatment (20.85) are very close to each other, whereas both of them are much higher than the mean for control subjects (9.32) on post-test. A similar trend is also seen for the retention-test From these data one may conclude that subjects from both types of schools responded in a very similar manner to the different treatments. Table 6 represents means and standard deviations of dependent variables obtained from Ss in Bengali medium schools: 2. 81 for Concept I on post-test,
4. 73 for Concept 11, 9. 67 for Concept M, 17.21 for total achievement, and 14. 67 for time used for post-test. A similar trend was also observed for the retention test. with the exception of time taken, which is 7. 45. Evidently, students took more time on the post-test than on the retention-test. Table 7 represents means and standard deviations of dependent variables obtained from Ss in the English medium schools and these are: 2. 78 for Concept I on post-test,
4. 77 for Concept 11, 9.91 for Concept M, 17.47 for total achievement, 14.49 for time used on post-test. Again, a similar trend is also




51
TABLE 5
MEANS AND STANDARD DEVIATIONS OF THE THREE GROUPS
(ENGLISH MEDIUM SCHOOLS)
Standard Number of
Treatment Groups Mean Deviation Measurements
Post-Test Measures
Written Group 22.15 3.04 34
Pictorial Group 20.85 3.03 34
Control Group 9.32 3.59 34
Retention-Test Measures
Written Group 22.80 3.18 34
Pictorial Group 21.59 3.36 34
Control Group 10.03 3.76 34




TABLE 6
MEANS AND STANDARD DEVIATIONS OF DEPENDENT VARIABLES (BENGALI MEDIUM SCHOOLS)
Standard Number of
Performance Measures Mean Deviation Measurements
Post-Test
Concept I (geotropism) 2.81 1.14 102
Concept II (phyllotaxy) 4.73 2.16 102
Concept III (feather) 9.67 4.15 102
Total Achievement 17.21 6.67 102
Time Taken 14.67 3.72 102
Retention-Test
Concept I (geotrapism) 2.83 1.01 102
Concept I (phyllotaxy) 4. 92 2.01 102
Concept III (feather) 9.66 3.74 102
Total Achievement 17.41 5.89 102
C"
Time Taken 7.45 3.19 102 W




TABLE 7
MEANS AND STANDARD DEVIATIONS OF DEPENDENT VARIABLES (ENGLISH MEDIUM SCHOOLS)
Standard Number of
Performance Measures Mean Deviation Measurements
Post -Test
Concept I (geotropism) 2.78 1.30 102
Concept II (phyllotaxy) 4.77 2.30 102
Concept III (feather) 9.91 3.67 102
Total Achievement 17.47 6.61 102
Time Taken 14. 49 3.43 102
Retention-Test
Concept I (geotropism) 3.03 1.36 102
Concept II (phyllotaxy) 5.1 8 2.12 102
Concept III (feather) 10. 12 3.80 102
Total Achievement 18.32 6.53 102
CA
Time Taken 7.52 3.19 102W




54
exhibited by the retention-test. In general. the means of the retentiontest are slightly higher than the post-test means, with one exception; that is. the testing time (7. 52) is much lower than the post-test time. Comparison of Tables 6 and 7 shows that the means for the dependent variables for Ss in the Bengali medium schools (6) are very close to the corresponding means (7) of dependent variables for the Ss in the English medium schools, i. e., the corresponding values are not much different Table 8 shows the means and standard deviations of aptitude measures obtained from Ss in the Bengali medium school.
The mean scores on these tests were: Auditory Number Span Test, 15. 77; First and Last Names Test, Part H, 9. 34, Part 1, 7. 63; Letter Sets Test. Part 1. 7.07, Part 11, 6. 61; Vocabulary Test (V-2), Part 1, S. 81; Vocabulary Test (V-1). Part 1, Part 111, 2. 53; Vocabulary Test (V-2). Part H. 2. 10. Table 9 shows the means and standard deviations of aptitude measures obtained from Ss in the English medium schools. The range of means from highest to lowest are: Auditory Number Span Test, 14. 85; First and Last Names Test, Part 1. 11. 41, Part 11. 10. 91; Letter Sets Test, Part 1, 8. 35; Vocabulary Test (V-2). Part 1, 8. 33, Letter Sets Test. Part 11, 8.20; Vocabulary Test (V-1), Part 1, 7. 69; Part 11. 7.13. Vocabulary Test (V-2). Part H, 6.23. A comparison of Tables
8 and 9 shows that in general. the means of aptitude measures for Ss in the English medium schools were slightly higher than for Ss in Bengali medium schools. Specifically. the means of different vocabulary tests for English medium school Ss were higher than for Bengali medium schools.




TABLE 8
MEANS AND STANDARD DEVIATIONS OF A PTITUDE MEASURES (BENGALI MEDIUM SCHOOLS)
Standard Number of
Performance Measures Mean Deviation Measurements
Letter Sets Test, Part I 7. 07 2. 90 102
Letter Sets Test, Part 11 6. 61 2. 66 102
Vocabulary Test (V-i). Part I 2. 53 1.80 102
Vocabulary Test WV-1). Part 11 2. 53 1.66 102
Vocabulary Test (V-2). Part 1 3. 81 2. 28 102
Vocabulary Test (V-2), Part 11 2.10 1.66 102
First and Last Names Test, Part I 7. 63 3. 91 102
First and Last Names Test, Part 11 9. 34 3. 70 102
Auditory Number Span Test 15. 77 4.34 102




TABLE 9
MEANS AND STANDARD DEVIATIONS OF APTITUDE MEASURES (ENGLISH MEDIUM SCHOOLS)
Standard Number of
Performance Measures Mean Deviation Measurements
Letter Sets Test. Part I 8. 35 2. 63 102
Letter Sets Test. Part 11 8. 20 2. 39 102
Vocabulary Test (V-i). Part I 7. 69 3. 50 102
Vocabulary Test (V-i). Part 11 7.13 3.02 102
Vocabulary Test (V-2), Part I 8.33 2.79 102
Vocabulary Test (V-2), Part 11 6. 23 2. 93 102
First and Last Names Test. Part 1 11.41 3.47 102
First and Last Names Test, Part 11 10. 91 3.28 102
Auditory Number Span Test 14. 85 4. 45 102
W,




57
Intercorrelations Between Different Measures
Intercorrelations of both independent and dependent measures were computed for the total sample in Bengali and in English medium schools. Table 10 summarizes the intercorrelations of aptitude measures for Ss in Bengali medium schools and Table 11, for Ss in the English medium schools. Evidently, the intercorrelations between the first part and the second part of the same test are relatively higher. This is appropriate because they measure the same mental abilities. In other cases. intercorrelations among the different aptitudes tests are relatively low, again because they measure different mental constructs. Table 10 shows relatively higher intercorrelation between Letter Sets Test, Part L with Vocabulary Test (V-2) (0. 40) and First and Last Names Test. Part 11 (0.45); between Vocabulary (V-2), Part I, with First and Last Names Test, Part 1 (0.45) and Part H1 (0.48). Table 11 also indicates that one part of the test is relatively highly correlated with the other part of the same test. In addition, it also shows that the First and Last Names Test, Part II. is relatively highly correlated (0.41) with the Auditory Number Span Test. Tables 12 and 13 show intercorrelations among the dependent variable measures obtained from Ss of the Bengali and English medium schools respectively. Intercorrelations among dependent variables both in post- and retentiontests show relatively higher correlations than among aptitude measures. Most of the dependent variables displayed higher intercorrelation indices with one another and with themselves. The comparisons of




TABLE 10
INTERCORRELATION MEASURES OF THE APTITUDE TESTS (BENGALI MEDIUM SCHOOLS)
Number of
MeasurePerformance Measures 1 2 3 4 5 6 7 8 9 ments
1. Letter Sets Test. Part i 1.00 0.53 0.32 0.19 0.40 0.16 0.37 0.45 0.13 102
2. Letter Sets Test. Part II 1.00 0.34 0.24 0.27 0.21 0.35 0.30 0.08 102
3. Vocabulary (V-), Part i 1.00 0.54 0.51 0.54 0.32 0.36 0.18 102
4. Vocabulary (V-I), Part II 1.00 0.47 0.51 0.21 0.20 -0.03 102
5. Vocabulary (V-2), Part I 1.00 0.43 0.45 0.48 0.21 102
6. Vocabulary (V-2), Part II 1.00 0.26 0.22 0.19 102
7. First and Last Names Test,
Part I 1.00 0.76 0.16 102
8. First and Last Names Test,
Part H 1.00 0.22 102
9. Auditory Number Span Test 1.00 102




TABLE 11
INTERCORRELATION MEASURES OF THE APTITUDE TESTS (ENGLISH MEDIUM SCHOOLS)
Number of
Measure
Performance Measures 1 2 3 4 5 6 7 8 9 ments
1. Letter Sets Test. Part I 1.00 0.62 0.30 0.20 0.34 0.25 0.17 0.16 0.09 102
2. Letter Sets Test. Part II 1.00 0.27 0.23 0.27 0.24 0.10 0.07 0.04 102
3. Vocabulary (V-i). Part i 1.00 0.72 0.59 0.64 0.22 0.06 0.06 102
4. Vocabulary (Vi), Part II 1.00 0.64 0.59 0.15 0.15 0.02 102
5. Vocabulary (V-2). Part I 1.00 0.70 0.12 0.19 0.10 102
6. Vocabulary (V-2), Part II 1.00 0.29 0.16 0.20 102
7. First and Last Names Test,
Part I 1.00 0.61 0.53 102
8. First and Last Names Test,
Part II 1.00 0.41 102
9. Auditory Number Span Test 1.00 102




TABLE 12
INTERCORRELATION MEASURES OF THE DEPENDENT VARIABLES (BENGALI MEDIUM SCHOOLS)
Number of
Measure-,
Performance Measures 1 2 3 4 5 6 7 8 9 10 ments
Post-Test
1. Concept I (geotropism) 1.00 0.52 0.59 0.70 0.31 0.64 0.41 0.57 0. 61 -0.42 102
2. Concept II (phyllotaxy) 1.00 0.74 0.87 0.38 0.46 0. 71 0.64 0. 73 -0.23 102
3. Concept III (feather) 1.00 0.96 0.30 0.61 0. 66 0.87 0. 88 -0.33 102
4. Total Achievement 1.00 0.36 0.64 0. 71 0.84 0.89 -0.35 102
5. Time Taken 1.00 0.14 0.17 0.29 0.26 -0.04 102
Retention-Test
6. Concept I geotropismm) 1.00 0. 49 0. 61 0. 73 -0. 31 102
7. Concept II (phyIlotdxy) 1.00 0.60 0.81 -0.33 102
8. Concept III (feather) 1.00 0.95 -0.29 102
9. Total Achievement 1.00 -0.35 102
10. Time Taken 1.00 102




TABLE 13
INTERCORRELATION MEASURES OF THE DEPENDENT VARIABLES (ENGLISH MEDIUM SCHOOLS)
Number of
MeasurePerformance Measures 1 2 3 4 5 6 7 8 9 10 ments
Post-Test
1. Concept I (geotropism) 1.00 0.67 0.69 0.81 -0.42 0. 84 0.60 0.66 0.76 -0. 51 102
2. Concept II (phyllotagcy) 1.00 0.74 0.89 -0.20 0.71 0.87 0.67 0.82 -0.41 102
3. Concept III (feather) 1.00 0.95 -0.21 0.66 0.69 0.89 0.88 -0.44 102
4. Total Achievement 1.00 -0.27 0.78 0.81 0.86 0.92 -0.49 102
5. Time Taken 1.00 -0.42 -0.19 -0.20 -0.26 0.39 102
Retention -Test
6. Concept I (geotropism) 1.00 0.68 0.69 0.83 -0.47 102
7. Concept II (phyllotaxy) 1.00 0.67 0.86 -0.36 102
8. Concept III (feather) 1.00 0.94 -0.36 102
9. Total Achievement 1.00 -0.48 102
10. Time Taken 1.00 102




62
Tables 12 and 13 reveal that the correlation coefficient indices for total time taken in post-test for Ss in Bengali medium schools are low positive, whereas in the English medium schools these are low and negative. That is, in the latter case, time is inversely related with the different performance measures.
Instructional Treatment Main Effects
The following hypotheses represent the major concern of this investigation:
1) Subjects receiving treatments will show significantly
greater behavioral change, in terms of the acquisition
and retention of concepts, than control.
2) There will be a significant differential treatment effect
between written presentation and pictorial presentation.
3) There will be a differential relationship between criterion
measures and aptitudes of subjects relative to the
treatment obtained.
This study will also yield data regarding language and cultural effect.
In the tables. groups were named according to the treatment
they received. Group 1 received written presentation, hence is known as written group; group 2 received pictorial presentation, hence is known as pictorial group; and group 3 received no treatment, hence is known as control group. Analysis of variance was performed using




63
both post- and retention-test measures. to determine if there were significant treatment effects. Bengali medium school Ss and English medium school Ss were studied separately. A split plot, nested statistical design (BMD 08V), with repeated measures analysis of variance with equal cell size (34) was employed. Separate analyses were done using: total achievement scores (a cumulative score of Concept 1, Concept U., and Concept III); scores on Concept 1; scores on Concept II; scores on Concept 1II, and the time used for both the post- and retention -tests. Whenever an overall significant F ratio was obtained. Tukey's HSD test, which performs all pairwise comparisons, was employed to determine the pair of means which was significantly different from the others.
Acquisition of Concepts
Total Achievement on Post- and Retention-Test
Acquisition of the concepts in this study was measured by an immediate post-test and delayed retention-test. These test scores were analyzed to determine instructional treatment main effects.
Table 14 reports the results obtained for analysis of variance
using three groups (34 Ss in each group) and the total achievement on two tests for Ss in Bengali medium schools. Significant group differences (F =212.74, P< .01) were found, and significant interactions (F = 3. 40. P < .05) between groups and achievement were also obtained. Tukey's pairwise comparison test was performed using both groups (Table 15) and interactions (Table 16). It is evident




TABLE 14
ANALYSIS OF VARIANCE FOR CRITERION MEASURES TOTAL ACHIEVEMENT (BENGALI MEDIUM SCHOOLS) Source Error Term df MS F
Group S(G) 2 3.077.40 212.74**
Total Achievement S x T" (G) 1 2.16 0.48
S(G), 99 14.47
G x T" S x T" (G) 2 18.01 3.40*
S x T" (G) 99 4.51
Note--Abbreviations: G Groups; S = Subjects; T" Total Achievement G= Written T" = On Post-Test
G2= Pictorial T" = On Retention-Test
G3 = Control
*P < .05.
**P < 01.




65
TABLE 15
TUKEY'S TEST FOR PAIRWISE COMPARISONS
TOTAL ACHIEVEMENT (BENGALI MEDIUM SCHOOLS) Differences Among Means G3 2
G3 = 9.63 -- 10.49* 12.54*
G2 = 20.12 -- 2.06
G1 = 22.18 -*P< .01.




TABLE 16
TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (BENGALI MEDIUM SCHOOLS) Differences Among Means
xi X2 34 56
3i=8.94 (G x Tj) -- 1.38 10. 91* 11.44 13.12** 13.35**
R2 =10. 32 (G3 x T") -- 9. 53',"* 10. 068** 11. 74'** 11. 97**
X319.85 (G2 x T") --.53 2.21** 2.44**
X4 20.38 (G2 x Ti') -1. 68* 1. 91*
X5 =22. 06 (Gl x T") --.2
X6 22. 29 (Gl x T")
:,P < .05.
**P < 01.




67
from Table 15 that groups 1 and 2 are greater than group 3. Or. in other words. the written group and the picture group performed better than the control group. To rephrase the above statement, there was a significant difference between the treatment groups and the control group. There is no evidence to support the contention that the written and pictorial groups are significantly different. Table 16 depicts the results obtained from Tukey's test performed on the means of the interactions. From this table it can be concluded that interactions X5 (G, x T") and X6 (Gl x Tj') are superior to the rest. Or. in other words, interactions between the written group and the total achievement on retention-test (X 5) and post-test (X 6) are significantly better than other interactions. Similar analysis of variance results on total achievement for Ss in English medium schools are represented in Table 17. It is evident from this table that the groups were significantly different (F = 169. 55, P < 01) and the total achieve ments were also significantly different (F = 8. 78, P < 01). A Tukey's test was performed on group means, and results are shown in Table 18. Evidently, group 1 and group 2 are significantly superior (P < 01) to group 3. and there is no significant difference between group 1 and group 2. It is also evident from comparing means that the mean for total achievement on the retention-test (18. 34) is higher than the mean for total achievement on the post-test (17. 56).
After using total achievement scores, i. e., scores on Concept I, Concept HI and Concept III, for analysis of variance, each concept




68
TABLE 17
ANALYSIS OF VARIANCE FOR CRITERION MEASURES TOTAL ACHIEVEMENT (ENGLISH MEDIUM SCHOOLS) Source Error Term df MS F
Groups S(G) 2 3,373.16 169.55*
Total Achievement S x T" (G) 1 30. 59 8.78*
S(G) 99 19.90
G x T" S x T" (G) 2 1.27 0.36
S x T" (G) 99 3.49
*P< .01.




69
TABLE 18
TUKEY'S TEST FOR PAIRWISE COMPARISONS
TOTAL ACHIEVEMENT (ENGLISH MEDIUM SCHOOLS) Differences Among Means G3 G2 G1
G3 = 9.87 -- 11.40* 12.87*
G2 = 21.26 -- 1.47
G1 = 22.74 -*P< .01.




70
was evaluated separately with respect to the three treatments. The purpose was to find out if the treatments differed significantly from each other with respect to the acquisition of a particular concept. Performance on Concept 1
Table 19 represents an analysis of variance obtained from using the three treatment groups (34 Ss in each group) and performance on the post-test and retention-test on Concept I for the Bengali medium school Ss. Table 19 shows that the groups are significantly different (F = 43.91. P< .01). A Tukey's test for pairwise comparison between meins was employed. and the results are represented in Table 20. This table shows that group 1 and 2 are significantly different (P < 01) from group 3. and there is no significant difference between group 1 and 2. Table 21 shows the same analysis performed on the English medium school Ss. The results obtained from analysis of variance also show a significant difference (F = 65. 00, P < 01) between groups. A significant difference (F = 10. 86, P < .05) also exists between tests. Table 22 shows Tukey's test comparing the means of different groups: group 1 and group 2 are significantly greater (P < 01) than group 3. and no significant difference exists between group 1 and group 2. When comparing the means of post-test and retention-tests for groups 1 and 2, the retention-test (3.03) is higher than the post-test (2. 78). Performance on Concept 11
The results obtained from the analysis of variance using
performance on Concept II. with three groups (34 Ss in each group)




71
TABLE 19
ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT I
(BENGALI MEDIUM SCHOOLS)
Source Error Term df MS F
Group S(G) 2 45.28 43.91*
Test S x T (G) 1 0.0120 0.05
S(G) 99 1.03
GxT SxT(G) 2 0.12 0.28
S x T(G) 99 0.43
Note--Abbreviation: T = Test.
*P< .01.




72
TABLE 20
TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT I
(BENGALI MEDrUM SCHOOLS) Differences Among Means G3 G2 G1
G3 = 1. 91 -- 1.16* 1.57*
G2 = 3.07 -- 0.41
G1 = 3.49 -*P< .01.




73
TABLE 21
ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT I
(ENGLISH MEDIUM SCHOOLS)
Source Error Term df MS F
Group S(G) 2 94. 73 65. 00*
Test S x T (G) 1 3.06 10. 86*
S(G) 99 1.46
GxT S x T (G) 2 0.25 0.90
S x T (G) 99 0.28
*P< .01.




74
TABLE 22
TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT I
(ENGLISH MEDIUM SCHOOLS)
Differences Among Means G3 G2 G1
G3 = 1.54 -- 2.03* 2.06*
G2 = 3.57 -- 0.03
G1 = 3.60 -*P< .01.




75
and two tests for the Bengali medium school Ss are shown in Table 23. This table shows that the groups are significantly different (F 98. 66. P < 01). Table 24 shows the results obtained from a Tukey's test performed on the group means: group 1 and group 2 are significantly better (P < 01) than group 3, and no significant difference is obtained between group 1 and group 2. Table 25 shows a similar analysis for Concept 11 using the English medium school Ss: the groups differ significantly (F = 123. 56, P < 01). A significant difference (F= 13. 09. P< 01) also existed between tests. A significant interaction (F =3.12, P < 05) also existed between groups and tests. Table 26 shows the results obtained from a Tukey's test: group 1 and group 2 are significantly superior (P < 01) to group 3. Table 2 6 does not indicate the presence of a significant difference between group 1 and group 2. Comparing means of post-test and retentior-tests, it is evident that means of the retention-test (5. 18) are higher than means for the post-test (4. 77). Table 27 indicates the results obtained from pairwise comparisons of means for interaction: the interactions between group 1 and post-test. group 1 and retention-test, group 2 and post-test. group 2 and retention-test. are significantly higher (P < 01) than between group 3 and post-test, and group 3 and retentiontest. In other words, interactions between written and pictorial groups with two tests were significantly higher than the control group with two tests.




TABLE 23
ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT II
(BENGALI MEDIUM SCHOOLS)
Source Error Term df MS F
Group S(G) 2 251.93 98.66*
Test S x T (G) 1 1.96 1.54
S(G) 99 2.55
GxT SxT(G) 2 2.36 1.85
S x T (G) 99 1.28
*P< .01.




77
TABLE 24
TUKEY'S TEST FOR PAIRWSE COMPARISONS CONCEPT II
(BENGALI MEDIUM SCHOOLS) Differences Among Means G3 G2 G,
d3 =2.60 --3.41* 3.25
G2 = 6.01 -- 0.16
Gi = 5.85 -*P <. 01.




78
TABLE 25
ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT II
(ENGLISH MEDIUM SCHOOLS)
Source Error Term df MS F
Groups S(G) 2 332.84 123.56**
Tests SxT (G) 1 8.24 13.09**
S(G) 99 2.69
GxT SxT(G) 2 1.96 3.22*
SxT (G) 99 0.63
*P < .05.
**P < 01.




79
TABLE 26
TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT II
(ENGLISH MEDIUM SCHOOLS)
Differences Among Means G3 G2 G1
G3 =2.43 -- 3.68* 3.97*
G2= 6. 10 -- 0.29
G1 = 6.40 -*P < .01.




TABLE 27
TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT II
(ENGLISH MEDIUM SCHOOLS)
Differences Among Means X1 2 3 4 5 6
1 =2.03 (G3 x T1) -- 0.79 3.97* 4.18* 4.26 4.47*
X2 = 2.82 (G3 x T2) -- 3.18* 3.38* 3.47* 3.68*
X3 = 6.00 (G1 x T1) -- 0.21 0.29 0.50
X4= 6.21 (G1 x T2) -- 0.09 0.29
X5 = 6.29 (G2 x T1) X6 6.50 (G2 x T2)
*P< .01.




81
Performance on Concept M
Table 28 represents the analysis of variance data obtained for Concept III using three groups, (34 Ss in each group), and two tests for Ss in Bengali medium schools. This table demonstrates a significant difference (F = 148. 40, P < 01) between the groups. A Turkey's test was employed for comparing the group means. Table 29 shows the results obtained from this analysis: group 1 and group 2 are significantly superior (P < 01) to group 3. and group 1 is significantly greater (P < 05) than group 2. An exactly similar analysis was performed using Ss from English medium schools. Table 30 shows the results: there is a significant difference (F = 97. 86, P < 01) between groups. A Tukey's test was performed to find out the locus of difference here. Table 31 summarizes the results obtained from this test: group I and group 2 are significantly better (P < 01) than group 3. and no significant difference exists between group I and group 2.
The data obtained from an analysis of variance. using performance on total achievement, Concept I, Concept If and Concept M for both types of schools, can be summarized as follows:
1) Written presentation (treatment 1) and pictorial presentation
(treatment 2) produced greater behavioral change, measured
in terms of positive change in scores on the criterion
measures, than the control (treatment 3).
2) There is no apparent significant difference between
written treatment and pictorial treatment. This pattern




82
TABLE 28
ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT III
(BENGALI MEDIUM SCHOOLS)
Source Error Term df MS F
Group S(G) 2 1,108.72 148.40*
Test S x T (G) 1 0.0049 0.0023
S(G) 99 7.47
GxT SxT (G) 2 5.59 2.64
S x T (G) 99 2 11
*P < 01.




83
TABLE 29
TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT MI
(BENGALI MEDIUM SCHOOLS) Differences Among Means G3 G2 G1
3 = 5.12 -- 5.91** 7.72*
G2= 11.03 -- 1.81
1 = 12. 84 -*P< .05.
**P< .01.




84
TABLE 30
ANALYSIS OF VARIANCE FOR CRITERK)N MEASURES CONCEPT II
(ENGLISH MEDIUM SCHOOLS)
Source Error Term df MS F
Groups S(G) 2 940.78 97.86*
Tests S x T (G) 1 0.83 0.44
S(G) 99 9.61
GxT SxT(G) 2 0.064 0.03
S x T (G) 99 1.87
*P< .01.




85
TABLE 31
TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT III
(ENGLISH MEDIUM SCHOOLS)
Differences Among Means
3 G2G1
G3 = 5.90 -- 5.40* 7.13*
2 = 11.29 -- 1.74
1 = 13.03 -*P< .01.




86
was consistent for Ss from both types of schools. However,
there is one exception to this pattern: the case of Concept
M In the Bengali medium schools. Here. the written
group was significantly better (P < .05) than the pictorial
groupA similar set of analyses of variance were done with the time students required on the post-test and retention-test. It was intended that these analyses give some information regarding the time and treatment relationship.
Effects of Post-Test and Retention-Test Time
Table 32 demonstrates the data obtained from an analysis of variance using three groups (34 Ss in each group) and the post-test and retention-test time used Ss in Bengali medium schools. Table 32 demonstrates a significant difference (F = 234.05, P< .01) between test times. It also indicates significant interaction (F 15. 76, P < 01) between groups and times. It is obvious from comparing the two means that post-test time (14. 67) was larger than retention -test time (7. 72). A comparison between interaction means was performed using Tukey's test. Table 33 summarizes the data, showing that an interaction, between the written group for post-test time and the pictorial group for post-test time. was significant (P < 01). A similar analysis was performed using the Ss in English medium schools. The results are represented in Table 34: a significant difference (F = 17.23.




Full Text

PAGE 1

DIFFERENTIAL EFFECTS OF PICTORIAL AND WRITTEN PRESENTATION ON THE ACQUISITION OF SCIENTIFIC CONCEPTS BY INDIANS TAUGHT IN BENGALI AND I N ENGLISH By PROTIMA ROY A DISSERTATION PRESENTED TO THE GRADUATE COU N CIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1974

PAGE 2

Dedicated to my grandmother, BROJOBALA KUMAR-a wonderful concept

PAGE 3

ACKNOWLEDGMENTS The writer wishes to express her deepest gratitude to Professor John J. Koran, Jr Chairman of the supervisory committee, for bis time, effort, encouragement and continuous guidance throughout the present study. The author is also thankful to all her committee members, Professor Vynce A. Hines, Dr. Joseph J. Shea, Dr. James D. Casteel, for their generous help and suggestions. Special thanks are also extended to Dr. Mary Lou Koran for her help in interpreting the aptitude treatment interactions data and to Professor John M. Newell for his generous advice. The investigation owes a special debt of gratitude to Dr. James E. McLean, for his assistance in computer programming, to Dr. Arun Mazumdar and Dr. Shantanu Maitra for their assistance and technical help in preparation of the data for analysis. The author also expresses her appreciation to her husband. Professor Rabindranath Roy. for his interest and encouragement throughout this study. Finally, the author wishes to express her sincere appreciation to the logistical support of the administration of the school system in Calcutta which participated in this study. iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDG M ENTS LIST OF TABLES LIST OF FIGURES ABSTRACT Page iii vii xi xii CHAPTER I THE PROBLEM . 1 Definition of the Problem. 1 Background to the Problem 1 Definitions . 3 Types or Concepts . 5 Distinguishing Characteristics of Concept Format i on . 7 Research on Concept Formation. 8 Effects of Different V ariables that Influence Concept Formation . . 8 Effects of Positive and Negative Examples 8 Effects of Redundant Relevant Information on Concept Acquisition. . . 10 The Effects of the Order and Sequence of the Examplars . . 11 Deductive and Inductive Instruction. 11 Nature of Critical and Noncritical Attributes . 12 Type of Concept Variables. . 12 Response Variable, . 13 Theory and Research on Pictorial Presentation . . . 13 Theory and Research on Aptitude Treatment Interactions . . 20 Theory and Research on Language and Cognition ....... 23 Language and Its Influence on Concept Learning 24 Cultural Influence 26 Bilingual Children 27 Rationale for This Study 29 Experimental Questions 36 iv

PAGE 5

CHAPTER II III IV TABLF OF CONTENTS (Continued) FXPEROlENTAL DESIGN 1be Design ....... Independent Variables Ability Measures ... Letter Sets Test (I-H Vocabulary Test (V-1) and (V-2) First and Last Names Test (Ma-3) Auditory Number Span Test (Ms-1) Dependent Variables Methods ........ Subjects ..... Treatment Materials Procedures . The Instrument and Its Reliability. Data Collection Procedures RESULTS ........ ... Page 37 37 37 39 40 40 40 41 41 41 41 43 44 45 46 48 Purpose of Study and Data Collected 48 Independent and Dependent Measures 48 Intercorrelations Between Different Measures . . . 57 Instructional Treatment Main Effects 62 Acquisition of Concepts . . 63 Total Achievernent on Postand RetentionTest. . . 63 Performance on Concept I 70 Performance on Concept II 70 Performance on Concept III 81 Effects of Post-Test and Retention-Test Time . . . 86 Aptitude X Treatment Interactions. 90 DISCUSSION AND IMPLICATIONS .. Summary of Data and Interpretation lnstnictional Treatment ;\lain Effects Acquisition of Concepts ..... Dilferential Effects of Written and .104 .104 .107 .107 Pictorial Presentation . 108 Implications for the Educational Practice. .117 Limitations 119 APPENDICES .. .121 V

PAGE 6

TABLE OF CONTENTS (Continued) APPENDIX Page A TREATMENT MATERIALS FOR THE BENGALI MEDIUM SCHOOLS. 122 B POST-TEST AND RETENTION-TEST FOR THE BENGALI MEDIUM SCHOOLS 140 C TREATMENT MATERIALS FOR THE ENGLISH MEDIUM SCHOOLS 155 D POST-TEST AND RETENTION-TEST FOR THE ENGLISH MEDIUM SCHOOLS 1 73 LIST OF REFERENCES BIOGRAPHICAL SKETCH. vi 186 193

PAGE 7

LIST OF TABLES TABLE Page 1 EXPERIMENTAL DESIGN. 38 2 APTITUDE MEASURES 42 3 INDEPENDENT AND DEPENDENT : MEASURES 47 4 MEANS AND STANDARD DEVIATIONS OF THE THREE GROUPS (BENGALI MEDIUM SCHOOLS) 49 5 MEANS AND STANDARD DEVIATIONS OF THE THREE GROUPS (ENGLISH MEDIUM SCHOOLS) 51 6 MEANS AND STANDARD DEVIATIONS OF DEPENDENT VARIABLES (BENGALI MEDIUM SCHOOLS) 52 7 MEANS AND STANDARD DEVIATIONS OF DEPENDENT VARIABLES (ENGLISH ::\IEDIUM SCHOOLS) 53 8 MEANS AND STANDARD DE VIA TIO!\'S OF APTITUDE MEASURES (BENGALI MEDIUM SCHOOLS) 55 9 MEANS AND STANDARD DEVIATIONS OF APTITUDE MEASURES (ENGLISH l\U:DIUl\l SCHOOLS) 56 10 INTERCORRELA TION MEASURES OF THE APTITUDE TESTS (BENGALI MEDIUM SCHOOLS) 58 11 INTERCORRELATION MEASURES OF THE' APTITUDE TESTS (ENGLISH MEDIUM SCHOOLS) 59 12 INTERCORRELA TION MEASURES OF THE DEPE NDEN T VARIABLES (BENGALI ::\IEDIUM SCHOOLS) 60 13 INTERC ORR ELA TION MEASURES OF THE DEPEND E~ T VARIABLES (ENGLISH MEDIUM SCHOOLS ) 61 vii

PAGE 8

TABLE 14 LIST OF TABLES (Continued) ANALYSIS OF VARIANCE FOR CRITERION MEASURES TOTAL ACHIEVEMENT (BENGALI MEDIUM SCHOOLS) 15 TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (BENGALI MEDIUM Page 64 SCHCX)LS) 65 16 TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (BENGALI MEDIUM SCHCX)LS) 66 17 ANALYSIS OF VARIANCE FOR CRITERION MEASURES TOTAL ACHIEVEMENT (ENGLISH MEDIUM SCHOOLS). . 68 18 TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (ENGLISH MEDIUM SCHCX)LS) 69 19 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT I (BENGALI MEDIUM SCHCX)LS) . 71 20 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT I (BENGALI MEDIUM SCHOOLS) 72 21 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT I (ENGLISH MEDIUM SCHOOLS) . 73 22 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT I (ENGLISH :ME DIUM SCHOOLS) 74 23 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT II (BENGALI MEDIUM SCHCX)LS) 76 viii

PAGE 9

LlST OF TA BL ES ( Con ti nued ) TABLE Page 24 TUKEY'S TEST F O R P AIRW:ISE CO MPA R ISO::-."S CONCEPT II (BE N GA LI MEDIUM SCHOOLS) 77 25 ANALYSIS OF V A RIANCE FOR CRITERION MEASUR E S CO N C E P T II (ENG L ISH MEDIUM SCHOOLS) 78 26 TUKEY'S T E ST FOR PAIRWISE CO MPAR1SO~'S CONCEPT II (E N G LISH MEDIUM S CHOOLS) 79 27 TIJKEY'S TE ST F O R PAIRWISE C O MPARISOl\'S CONCEPT II ( EN GL ISH 1IEDIUM SCHOOLS) 80 28 ANALYSIS OF V A RIANCE FOR CRITERION MEASURES CO N CEP T ill (BENGALI MEDIUM SCHOOLS) 82 29 TUKEY'S TEST FO R PAIR W ISE C O MPAR1SO~ CONCEPT III (B EN GA LI ~lE D IUM S CHOOLS) 83 30 ANALYSIS OF V ARIA NCE FOR C RI TERION MEASURES CO NCEPT ill (ENGLISH MEDiliM SCHOOLS) 84 3 1 TUKEY 1 S TEST FOR P AIRWISE CO MPAR IS0 1'.'S CONCEPT III (E N GL ISH MEDIUM S C HOOLS) 85 32 ANAL Y SIS OF V AR IANCE FOR CRI TERION MEASURES T IME' USED FOR POST-AND RE'TEN TION-TEST (B ENGALI MEDIUM S CHOOLS) 87 3 3 TUKEY'S T E ST FO R P AIRWISE C O M PAR1SOl\'S TIME USED FOR POST A N D RETE N TIO N TEST (BENGALI ME DIU M SC HOOLS) . .. 88 34 ANALYSIS OF VARIAN C E' FOR CR I TERION MEASUR ES TI ME USED FOR POSTA N D RETE N TIO N TE S T (ENGLISH MEDIU M S C H OO LS) 89 ix

PAGE 10

LJST OF TABLES (Cont inued) TABLE Page 35 TUKEY'S TEST FOR PAIBWISE COMPARISO!\'S TIME USED FOR POST AND RETENT ION-TEST (FNGLJSH MEDIUM SCHOOLS ) 91 X

PAGE 11

FIGURE 1 LJST OF FIGURES Regression Analysis for Letter Sets Test (Part II) on Post-Test Time Used (Bengali medium school Ss) 2 Regression Analysis for First and Last Names Test on Post-Test Time (Bengali medium Page 92 school Ss) . . 94 3 Regression Analysis for First and Last Names Test (Part II) on Post-Test Time Used (Bengali medium school Ss). . . 95 4 Regression Analysis for First and Last Name Test on Time Used for Retention-Test (Bengali medium school Ss). . . 96 5 Regression Analysis for Vocabulary Test (V-2). Part II. on Concept I (English medium school Ss) 97 6 Regression Analysis for First and Last Names Test on Concept I (English medium school Ss) 99 7 Regression Analysis for Auditory Number Span Test on Concept I (English medium school Ss) 100 8 Regression Analysis for Vocabulary Test (V-2). Part I on Concept III (English medium school Ss) ... 101 xi

PAGE 12

Abstract of Dissertation Presented to th e Graduate Council of the University of Florida in Par tial F ulfillment of the Requirements for the Degree of Doc to r of Philosophy DIFFERENTIAL EFFECTS OF PICTO R L.\L AND WRITTEN PRFSENTA TION ON THE AC QCJSI TI ON OF SCIENTIFIC CO N CEPTS BY L'-D IA N S TAUGHT IN BENGALI AND IN E:'.\ GLISH By Protima Roy December, 1974 Chairman: Dr. John J. Koran, Jr. Major Department: Curriculum and Instruction The purpose of this study was to compare the relative effective ness of pictorial presentation and written passages on the acquisition of three classificational concepts in biology b y Indians taught in English and in Bengali. This study should provide further information regarding whether the acquisition of concepts depends on the cultural background, language and specific characteristics of the student. Independent measures include: written passages about three concepts (geotropism, phyllotaxy, feather), s ing le-line pi.c:torial diagrams with appropriate labeling of the same concepts and nine aptitude tests. measuring verbal abilities, inductive reasoning and memory. Dependent measures were the immediate post-test (corresponding to the treatment ma t erials) measures for the acquisition of the xii

PAGE 13

three concepts. delayed retention-test (equivalent to post-test) measures and time used. One hundred and two students were selected from the Bengali medium schools and also from the English medium schools. The subjects were randomly assigned to two treatments and a control group. A split-plot nested design with repeated measures and equal cell size (34) was used to evaluate the instructional treatment main effects. The same procedures were followed for both types of school subjects. Subjects receiving treatments showed significantly greater behavioral change than those in the control. Although it was anticipated that there would be a differential effect of written and pictorial treat ments. this was not found except in one case. that is for Concept III in the Bengali medium schools. In the latter case. the written group was significantly different than the pictorial group. This finding is basically consistent with a pilot study conducted by the author using American subjects. Comparing the means it was evident that subjects of the two different types of schools showed a similar performance on the criterion measures. F test for homogeneity of regression was employed to study aptitude x treatment interactions. Analyses of interactions disclosed that scores on Letter Sets Test. First and Last Names Test. Vocabulary Test. and Auditory Number Span Test interacted significantly with the instructional treatments. xiii

PAGE 14

These findings suggest that subjects in different treatment groups responded differently to the criterion measures and consequently could be differentially assigned to instructional method according to these characteristics; thus achieving individualization. xiv

PAGE 15

CHAPTER I THE PROBLEM .Definition of the Problem The main purpose of this investigation was to compare the relative effectiveness of pictorial representation and written passages on the acquisition of three classificational concepts of biology by Indians taught in English and in Bangali. Another major purpose was to find out whether acquisition of the concepts was dependent on specific learner characteristics. This study will also give some information regarding whether the cultural background of the learner is a factor in concept learning, and the extent to which languages (Bengali and English) affect concept acquisition in science. Background to the Problem Recent curricula in science rely increasingly on teaching scientific processes and concepts rather than facts alone. A number of advantages are claimed for emphasizing conceptual knowledge. Since scientific knowledge is increasing and changing so rapidly, it is not possible for scientists themselves to keep abreast of changing "facts" in different disciplines of the sciences. Science students share the problem of learning and remembering specific scientific facts in all the sciences they are exposed to. Emphasizing scientific concepts 1

PAGE 16

is one way to equip the learner with the skills to process factual knowledge in a rapidly changing world. 2 The values of concepts have been emphasized by leading psychologists and educators. Vygotsky (1962) points out that the word that stands for a concept does not refer to a single object but to a group or a class of objects. Every word is therefore already a generalization. The acquisition of a concept therefore permits the learner to classify subsequent knowledge. to generalize beyond the knowledge given. and to reduce redundancy in the environment (Bruner. 1961 ). A concept is a type of summarizing "system" which provides the learner with the potential to deal with large amounts of new knowledge (Voelker. 1969) Being comprehensive in nature. concepts are useful to the learner in gaining some grasp of a larger field of knowledge than be has personally experienced. They permit the knower to interpret and assimilate new information into old systems through the modification of existing concepts (Pella. 1966). Concept learning not onl y permits more efficient information processing but also results in a conservation of memory. Remembering a concept requires recalling the critical attributes that describe many dimensions of the concept. Once the concept is recalled, critical attributes can be reconstructed. In contrast. factual knowledge can be 1hougbt of as discrete, specific. disconnected knowledge which has been memorized in unrelated parts and is consequently easy to partially or completely forget. ''Knowing a class of items reduces the necessity

PAGE 17

of knowing each element of the class; accordingly. fewer concepts than facts need to be remembered" (J.J. Koran, 1971, p. 408). 3 The domain of concept learning occupies a unique position between elementary behavioral processes--such as stimulus-response learning. chaining, verbal association. and discrimination learningand more complex behavioral processes like rule learning. problem solving. and creativity (Gagne. 1973). On the one hand, verbal association. chanining. and multiple discriminations are necessary to build concepts; on the other hand. these concepts combine to form principles and to provide the raw materials for problem solving and, subsequently. divergent thinking. Dermitions A wide range of definitions of concept learning have been given by many authorities. In general. one can identify common elements in the following definitions. Dictionary definitions try to distinguish the boundaries of a concept by indicating its genus, that is, what it has in common with other things and experiences, and its differentia, that is, the extent to which it differs from these things (Glaser. 1968). When the dimensions of a definition are overlooked, misinterpreted, or mis taught. then an error in the use of the concept occurs (Carroll, 1964). A concept is a category which a person uses to classify the stimuli he perceives (McDonald, 1965). In human beings. this category

PAGE 18

4 is often a verbal one; that is, a word describes the name of a class of objects or events. such as "vertebrates" as opposed to "invertebrates" (J. J. Koran. 1971). Concepts are properties of organizing experience --more specifically they are the abstracted, and usually cognitively structured. classes of "mental" experience learned by organisms in the course of their life histories (Carroll. 1964). A concept can be defined as "the common element shared by an array of objects" or "the relationship between the constituents or parts of a process" (George. 1962, p. 260). A concept may be thought of as an artifact extracted by verification of the contexts or sentences in which it occurs (Bronowski and Bellugi. 1970). "A concept is a generalized and abstract symbol; it is the sum total of all our knowledge of a particular class of objects. In short, a concept is a condensation of experience" (Viaud. 1960, pp. 7576). "The basic concepts are essentially high-level abstractions expressed in verbal cues and labels" (Taba. 1965, p. 465). "A concept is something about an idea expressed in the words of our language" (Platt. 1963, p. 21). All of the preceding definitions allude to the critical role that concepts play in school learning. At the same time. they suggest why one might expect the efficiency of concept learning to vary with

PAGE 19

culture and language and to be sensitive to individual cognitive characteristics. Types or Concepts 5 Psychologists. scientists. and science educators present at least two related methods of classifying concepts. In Pella's taxonomy (1966). three types of concepts are elaborated: classificational. correlational. and theoretical. Classificational. This type of concept is concerned with the classification of facts. It facilitates the description of phenomena. An example of this type of concept is "insect. An insect is an animal with six legs and three major body divisions. The facts that contribute to characterizing an insect are placed together and used to distinguish them rrom spiders. crustaceans. and other arthropods. Although there are many different kinds of concepts. classifica tional concepts probably represent the foundation, or knowledge base. on which more complex concepts and conceptual schemes can be built. Recognition of them by educators may be lacking because in common practice. they are taught through memorization (J. J. Koran, 1971). and. consequently, educators think of them as facts rather than concepts. Correlational Concept. This type of concept is concerned with the correlation of facts. It facilitates prediction. An example of this type or concept is "force. A force is a push or pull that tends to change the motion of a body. Here force is the concept and requires other

PAGE 20

concepts (push. pull, motion, change) to clarify it, Theoretical Concept. This type of concept is concerned with the explanation of phenomena. An example of this type of concept 8 is an "atom." An atom is the smallest particle of an element possible and consists of electrons, protons, neutrons. and other particles. There are certain similarities, as well as distinctions, among these three types of concepts. Classificational and correlational concepts are concerned with "abstractions from a field of direct experience." and they are "descriptions of human experience." whereas a theoretical concept is an "abstraction of a created idea." and "ex planation of human experiences" (Pella. 1966, p. 32). Bruner. a psychologist. and his associates use a different taxonomy with three classes of concepts: conjunctive, disjunctive. and relational. Each is related to a
PAGE 21

7 is also a cell (animal cell). or a structure without a nucleus is also a cell (human red blood cell). Relational Concepts. Here a relationship between two or more attributes describes the concept. an example of which is "waste," the relationship between the given state of an object or class and the value that one places on it. Conjunctive and disjunctive concepts provide a distinction usually made by psychologists to facilitate study under laboratory conditions. They may be thought of as being analogous to classificational, correlational, and theoretical concepts in school learning. Distinguishing Characteristics of Concept Formation Conceptual behavior involves two basic characteristic processes: generalization within classes and dis crimination between classes (Mechner, 1965). Generalization and discrimination also form the basis for evaluating concept formation. Conceptual behavior also incorporates direct perception or perception through words and symbols (Glaser. 1968). For the most part, concepts probably occur hier archically. with more complex concepts subsuming simpler ones (Gagne. 1973). Several steps. models, and schemes for concept formation are given by authorities in the field (DeCecco, 1968; Gagne. 1973; Taha. 1966: Ausubel, 1968; J. J. Koran. 1971). Briefly. one can identify the following steps in concept formation:

PAGE 22

8 1) The essential characteristics of the concept must be emphasized; 2) Both positive and negative instances of the concept should be made available; 3) The correct language for the concept and its characteristics should be established; 4) Proper sequencing or materials should be provided so that a student who is attempting to formulate one concept has had the prerequisite materials; 5) The generation of concepts should be encouraged, guided, and reinforced; and 6) Situations should be provided in which the concept can be generalized and discrimination between concepts can be made (J. J. Koran, 1971). Research on Concept Formation A large number of experimental investigations have been con ducted in the field of concept formation using different variables. Clark (1971) points out that over the last 30 years there have been well over 2 50 experimental studies conducted in the area of concept attainment. Many of these studies need replication in order to identify and assess variables which are relevant to the development of concepts in school students. The following review emphasizes some major lines or research which directly relate to the problem of this study. Effects of Different Variables that Influence Concept Formation Effects of Positive and Negative Examples Opinion differs among practitioners and researchers whether to use positive instances or negative instances alone or both together in

PAGE 23

concept instruction. A large number of research findings in this area are briefly summarized below: 9 Bourne (1966) found that subjects learn more efficiently from positive than from negative examples. Subjects show an inability or unwillingness to use information eff i ciently based on instances of what a concept is not (Bruner. Goodnow and Austin. 1956). Problems defined by positive examples are more easily solved than those defined by negative examples; mixed sequences are of intermediate diff i culty (Hovland and Weiss. 1953). Even at a time when the possib l e informa tional content of positive and negative instances is equaled, there is still an advantage in presenting positive examples (Hovland and Weiss. 1953). Huttenlocher (1962) points out that negative instances in a mixed series of positive and negative instances can result in efficient concept formation. He proposes that subjects must be taught to process this type of information since it is not characteristically used in society. Freibergs and Tulving (1961) show that, while practice time was longer for learning to use negative instances than for positive ones. subjects gradually used negative instances almost as well as positive. Information obtained from negative instances does not seem to be trusted (Bruner et al.. 1956). although one can logically determine the minimum number of positive and neg ative examples required to deflne a concept correctly (Hovland. 1952 ). Increased difference between pos i tiv e and negative instances

PAGE 24

10 facilitates concept acquisition. The degree of difference between positive and negative instances of a concept may be a more significant variable in concept formation than the form of these instances (Clark. 1971). The form of an example apparently determines the number of non critical attributes that will be displayed or evoked by it: as this number decreases, ease of concept attainment increases. Hence. e%3Dlples in verbal form. because they evoke fewer noncritical cbaracteristics, increase the ease of concept attainment more than ~pies presented in visual form. e.g drawings, photographs (Clark. 1971). However Heidbreder (1949) disagrees with this finding. "Ibese f"mdings are particularly relevant to this study, since the independent variables were presented in visual and verbal form. A display of positive and negative examples in the same form accelerates concept attainment to a greater extent than a display which allows form variation among instances. When examples are in verbal form. the degree to which positive examples evoke appropriate critical properties, and negative instances evoke necessary noncritical properties increases, ease of concept attainment increases (Clark, 1971). Effects of Redundant Relevant Information on Concept Acquisition Bourne (1966) found that redundancy of relevant information accelerates concept learning, whereas redundancy of irrelevant information interferes with acquisition of a concept. Irrelevant dimensions which interfere with concept formation have not received as much attention as relevant dimensions. Most

PAGE 25

11 often it is the learner's task to find some kin d of rule. or prin<:iple of internal structure. concerning the relevan t dimensions that define a concept. The Effects of the Order and Sequence o f th e E xamplars Tbe sequence variable has considerab l e influence on the acquisition of concepts. However. these have not been investigated extensively. Hovland and Weiss (1953) studied sequence by presenting mixed orders of positive and negative examp l es and reported no effects. Detambel and Stolurow (1956) condncted an investigation concerned with change in the sequence of presentation of relevant and irrelevant dimensions from trial to tria L They concluded that a nonsequential presentation showed a striking superiority in contributing to the efficiency of concept learning. Anderson (unpublished paper) followed up this investigation using a longer series of transitions than adjacent trials and confirmed only part of the findings of the above study. He mentions that his study "extends th e previous result by showing that the effective variable is not merely the number of relevant stimulus dimensions in the concep t learning task. but the number of these that changes from trial to trial" (reviewed by Glaser. 1968. p. 46). Deductive and Inductive Instruction Almost all studies of concept learning utilize inductive behav-ior. That is. the instances are presented and the rule must be induced

PAGE 26

12 (Glaser. 1968). On the other hand it appears that. in school learning, a concept often is taught by a deductive method, of presenting a rule and having students identify examples. \vbere time is not a factor or where the task is suitable, instruction may well be inductive However, for some tasks, and with time economy an issue, deductive sequences appear useful. Nature of Critical and Noncritical Attributes When the difference between critical and noncritical attributes increases. east of concept attainment also increases. When critical attributes become more prominent and noncritical attributes become less prominent, ease of concept attainment increases. When a critical attribute shows increasingly less variation from instance to instance, ease of concept attainment usually increases (Clark. 1971 ). Type of Concept Variables Decrease in the nwnber of relevant attributes (movement from a complex to a simple concept) generally increases ease of concept attainment. However, this finding is not supported by Laughlin (1966), Laughlin and Jorden (1967), Laughlin and Doherty (1967), as reviewed by Clark (1971). These findings have direct relevance for school learning in that strategies for concept attainment must be selected that minimize distracting variables and maximize defining characteristics.

PAGE 27

13 Conjunctive concepts are easiest to attain (Ciborowski and Cole, 1972) while disjunctive concepts appear more difficult. 'Ihe type of concept introduced may be a more significant variable in concept formation than the method of presentation (Clark. 1971). Again these findings are particularly relevant for experimental studies on concept formation in school subjects. In order to test the efficacy of different instructional materials, it is essential to select concepts that lend themselves to experimental studies in school setting. Response Variable Fase of concept formation and transfer to new concepts is presented along a continuum from high to low when students 1) verbalize critical characteristics only, 2) verbalize critical and noncritical properties, 3) do not verbalize, 4) verbalize noncritical properties only (Clark, 1971 ). When the student is able to verbalize both critical attributes and the concept name and to verbalize the concept name throughout the concept task, the ease of concept attainment and degree of retention increase (Clark. 1971). When the student is allowed to physically manipulate instances. ease of concept attainment increases more than when he is not allowed to do so (Clark, 1971). Theory and Research on Pictorial Presentation An instructional medium as defined by Dececco (1968) is

PAGE 28

only a means of transmitting instruction and not the substance of that instruction. 14 At present. a large number of media are available for instruction. This introduces the practical problem of teachers making an appropriate selection among media at a time which is useful in meeting specific instructional objectives. The following criteria could be used by teachers for the adequate selection of media: [ 1] "information about available media; [ 2] analysis and design of an instructional system; [ 3] knowledge of the research findings on the use of the media" (DeCecco. 1968, p. 529). Gestalt psychologists have proposed that the brain has built-in systems for organizing visual information. In contrast to this proposal it was argued that many principles of visual organization reside not in the brain. but in stimuli themselves (Gibson. 1950. 1966). Modern scientific views of how one looks at pictures (Neisser. 1967) generally agree that seeing is basically a constructive experience. That is. if one could slow down the process of pie ture looking and study it in detail, one would discover that the mind puts together. in a unified way. selected information brought in by the eyes. The first part of the pictorial comprehension process is an analytical one--one has to locate and select the critical features of what is being observed. Visual illustrations are becoming increasingly wi despread as a means of instruction. However. comparatively few attempts have,

PAGE 29

subsequently, been made to determine the relative effectiveness of various types of visual illustration. Supposedly. not all types or visual illustration are equally efficient in accelerating the learning of different types of educational tasks. 15 Holliday (1973) suggests that it is more likely that specific types of pictures facilitate the learning of specific types of objectives for certain students with certain kinds of characteristics. However, definite relationships have not yet been established. As a result. instructional designers have had to depend largely upon intuition in trying to match instructional systems with learner types. One result of these complex relationships and problems is that many investigators examining the contribution of pictures to verbal materials have con cluded that the type of visuals investigated did not significantly improve criterion achievement scores. Bruner et al. (1956) and Travers et al. (1964) arrived at the conclusion that learners do not need a wealth of stimuli in order to recognize the attributes of an object or situation which place it in a particular category. It is suggested by Travers et al. (1964) that there is no guarantee that useful information will be retained if a person is confronted with stimuli similar to those emitted by the real environment. Travers et al. (1964) adds that the realistic presentation of much content provides unnecessary detail. and that the real objective or visual education is not so much to bring the student into close touch with reality,

PAGE 30

16 but to help students become more effective in dealing w i th real ity They contend that this can be do n e effectively with symbols. Broadbent (1958, 1965) describes one reason why reduction of learn in g occurs as cues increase. According to him. i t is caused by a filter in g process in the central nervous system wh i c h prevents many rea li s ti c s timu li from receiving active reception in the brain. This assump ti o n is supported by Jacobson (1950. 1 9 51). who said that the bra in is capable of utilizing only minute propor ti ons of information perce iv ed. Th i s point of view is supported by Tra v ers et al. (1964. p. 520) : th e amount of information which is u til ized by the higher cente rs is vastly less than the informational capacity of the channels involved. A study conducted by Attnea v e (1954, pp. 185-86) to tes t th e hypothesis that a function of the perceptual machinery is to reduce redundant stimulation and to encode incoming informat i on so that only essentials travel through the nervous system to the brain. T o support this contention he mentions that 'lines bordering ob j ects suppl y the essence of the information to be conveyed." explaining the e ff ective ness of cartoons and stick dra win gs as conveyors of informa ti on. If this is true. the propos al of Travers et al. (1964) tha t visu al information is stored in the nervous system in some form iso m orphic with line drawings. thereby per mi tting the individual to re m e m ber and reproduce such information with greater facility than more r ea lis tic information, gains more cred ibili ty. One could interpret this as

PAGE 31

17 meaning that these visuals closely representing line drawings and containing the essence of information to be transmitted would be more effective and more efficient in facilitating learning than would detailed illustrations, which have to be initially coded by the central nervous system before being transmitted. This proposition is relevant for this investigation because single-line pictorial treatment materials were used in order to facilitate concept acquisition in school subjects. Travers et al. (1964) explain that inputs of information when received by the subjects are coded and most of the original stimuli initially presented to the senses not only never enters the perceptual system, but is not remembered by the system. Both theory and research seem to be contradictory to the assumption of "realism theories" that presenting a subject with a large number of stimuli which approximate "reality" is not necessarily the most effective way to accelerate learning. Because these excessive stimuli may interfere with the transmission of information, and some of the stimuli may not be perceived, researchers should make some effort to find those specific characteristics of visuals that will enhance particular types of learning (Dwyer, 1967). The most comprehensive study of visual illustration as an adjunct to instruction in science education has been conducted by Dwyer and reviewed by Holliday (1973). Dwyer studied the relative effectiveness of eight visual types in relationship to numerous instructional and other categorical variables. During the last five years, he reported

PAGE 32

18 at least thirty-two separate studies using similar treatments, experi mental designs, statistical analyses. and evaluative criteria. In his investigations. Dwyer commonly us ed as illustrations cross sections of the human heart including the following visual types: 1) simple-line drawings, 2) detailed sha ded drawings, 3) photographs of a heart model. and 4) photographs o f an actual heart. Black-and white and colored pictures were also used. Dwyer drew the following conclm i ons from his investigations: 1) no sex differences were observed, 2) color had a differential effect on achievement. 3) treatment differe n ces were not observed on delayed retention tests. 4) student choices for particular types of visuals were not reliable indicators of relative successfulness as measured by the criterion tests. 5) the use of verbal questions related to visuals did not seem to be helpful, 6) differing size images of visuals used in the investigations did not appear to enhance learning differentially, 7) small amounts of realistic detail. u.:,-ually via single-line drawings. formed a more effective visual aid when the subjects received group paced instruction, via slides and television. After studying the differential effects of visual types on hi gh school students, Dwyer arrived at the conclusion that verbal presentation is most effective in terms of "effectiveness, economy and / or simplicity of production" (Dwyer, 1972. p. 24). Without any doubt. Dwyer has made a broad spectrum of contri butions in the area of visual illustrations. However, some replication

PAGE 33

19 is needed for practical application in classroom instruction. In this regard, the following research must be considered: Vernon (1950, 1951, 1953, 1954) studied the effectiveness of graphs and pictures as adjuncts to verbal lessons and concluded (1953, 1954) that pictures generally contribute little to general comprehension of verbal material. Burdick (1960) arrived at the conclusion that pictures used to complement the verbal instruction do not facilitate reading compre hension. These two investigations suggest that perhaps pictures do not facilitate learning. In a similar study. both a graph and a map proved to be a significantly more effective visual advanced organizer than a verbal expository treatment (Weisberg. 1970. as reviewed by Holliday, 1973). The optimal size of a picture has been studied separately by Moore and Sasse (1971), and Dwyer (1970). and it depends on the objectives. Larger images inhibited student achievements for certain objectives (Dwyer, 1970). Dwyer suggests that the inhibiting or limiting factor in the instruction was the amount of time the subject was permitted to view and grasp relevant cues. However. any relationship between specific size of a picture and specific objective is not yet established. A study conducted by Chan, Travers, Van Mondfrans (1965) supports the hypothesis that color is a complex characteristic and is not necessarily a positive attribute for a visual. Research findings (Bruner and Potter. 1964; Dwyer, 1967) indicate that pictorial cues stimulate a rich flood of mental associations.

PAGE 34

20 Under certain conditions it may be difficult for a student to consolidate selected associations into a single, clear concept, thereby necessitating that the teacher choose visual aids with extreme care (Farnham Diggory, 1972). "Pictures should be introduced into a curriculum for the purpose of expanding and enriching associations, rather than constricting them. Pictorial comprehension involves the arousal and consolida tion of personal ideas that is both its power and its danger" (Farnham Diggory, 1972, p. 432). There are many more areas in visual illustration where further investigation needs to be done. Investigation regarding the relationship between different aptitudes and the picture types used in science instructional materials should provide information applicable to individualized instruction (Holliday, 1973) and should explain many of the nonsignificant or conflicting results previously reviewed. Theory and Research on Aptitude Treatment Interactions It has been known for a long time that individuals differ from each other with regard to mental abilities as well as in physique and traits of personality. However, only recently have educators devised a methodology of research to correspond with their concern for this phenomena. Instructional methods effective for training large numbers of individuals possessing different ability patterns have become a major concern of present-day educators. One way of achieving the

PAGE 35

21 goal or individualizing instruction, as proposed by Fuller (1969), Hereford (1971), and Merrill (1968), is to vary educational objectives, pacing, and sequencing according to student characteristics. Another appropriate approach, as proposed by Cronbach (1967), concerns varying instructional methods for different subjects to reach the same educational goals. As researchers point out, a subject learns more easily from one method than another, and this best method differs from subject to subject. Such differences between treatments are related to learner characteristics. As a result, a larger number of subjects may be expected to achieve the instructional objectives when instruction is differentiated for different types of subjects (C ronbach and Snow, 1969, reviewed by M.L. Koran. 1972). Cronbach (1965) developed a theoretical framework to deal with the nature of different aptitudes and differing instructional treatments. He named these studies aptitude treatment interaction studies, or more commonly A TI studies. Generally speaking, an interaction is present when one treatment is significantly better for one type of subject, while an alternative treatment is significantly better for a different type of subject. Particularly, the presence or ATI depends on nonparallel regression slopes of aptitude on performance for different instructional methods (M. L. Koran, 1972). "The general objective of ATI research is to match specific instructional methods or materials to selected learner

PAGE 36

22 characteristics" (M. L. Koran. 1972. p. 136). Two different types of interactions may be obtained. either ordinal or disordinal in nature (Cronbach and Snow. 1969). An inter action in which the regression slopes intersect within the range of the aptitudes being considered is known as a disordinal interaction. An ordinal interaction is present when regression functions have different slopes. but one is superior to the other through the range of aptitudes being considered. In ATI research. aptitude is defined as any characteristic of the learner that functions selectively with respect to learn ing. Or. in other words. it facilitates or interferes with his learning f rom some designated instructional method (Cronbach and Snow. 1969). Large numbers of student characteristics might be considered in adapting instruction to individual differences But. there are only a few studies which have investigated the possible rela tionshi ps between perceptual aptitude tests and different kinds of pictorial and nonpictorial instructional treatments. Nevertheless. significant interactions between certain perceptual aptitudes and two instructional treatments (a film-mediated and written model). concerned with mo deling procedures in the acquisition of a teaching skill. have been ob served (McDonald and M. L. Koran. 1969). In general. much more research is needed to identify the kinds of abilities and processes necessary for different tasks. It is necessary to find out whether students with particular characteristics are able

PAGE 37

23 to master particular tasks and under what conditions. This kind of relationship would provide us with the information for individualizing instruction by aptitude factors other than the differential rate of human learning. Further investigations involving the relationship between different aptitudes and visual illustration in science instructional material should shed some light on appropriate uses of visuals for particular student characteristics and educational outcomes. Theory and Research on Language and Cognition The relationship between language and thinking is intrinsic to the nature of intellectual growth. The child's ability to manipulate words is a reliable indicator of his thinking capacity. If his vocabulary is limited. his capacity for elaborating perceptions into concepts-the general process known as thinking--is blocked (King and Kerber. 1968). Language is a dynamic entity--a critically significant way of handling information. One can represent the world by actions and images only to a limited extent. In order to represent it flexibly and powerfully, one must have symbols for it, and language supplies these symbols (Farnharn-Diggory, 1972). The universal nature of language development bas also been emphasized by Farnham-Diggory (1972). It seems that all children begin to speak in about the same way at about the same time,

PAGE 38

regardless of nationality. The characteristic features of this universal developmental system may give some important clues to a basic human learning process. Language functions as a symbol system and as a process of thinking. without which human education is inconceivable. The growth and education of language capacities is almost another way of represent ing the growth and education of the mind (Farnham-Diggory, 1972). Language and Its Influence on Concept Learning The influences of language on concept formation constitute an important aspect of school learning. Words represent concepts that may have been learned preverbally or with the help of verbal labels. Concepts have been defined by some psychologists as meaningful words which label classes of otherwise dissimilar stimuli (Archer, 1964). Most studies designated as concept learning by psychologists have been limited to laboratory situations involving direct, nonverbal dimensions (Glaser, 1968). There is evidence which suggest that the ability to use words is an important factor in the rate of concept acquisition (Glaser, 1968). Dietze (1955) reports that children learn to categorize nonconventional forms faster when distinctive names are given to them. Verbalization, when compared with nonverbalization, enhances concept learning in four-year-olds, but makes insignificant difference for seven-year-olds because they tend to verbalize the solution. However

PAGE 39

25 Jenson (1966) found that, when instruction is given to seven-year-olds to verbalize the irrelevant dimension, it delays the learning of a reversal shift. The correlation between verbalization of a rule and correct responding is not clear. Green (1955) reports, from a study with college students, that verbalization covaried with correct responses but was not necessary to making the correct response. Kendler and Kendler (1962) suggest that, in studies with children, verbalizations are not always a guarantee that choice behavior will be appropriate (reviewed by Glaser, 1968). The discussion of language in the study of concept formation and its relationship to bilingual learning of science concepts brings up the question of the definition of stimulus dimensions involved and emphasizes mediational processes. Evidently there is a difference between children and adults in performing solution shifts in concept problems, which probably depends on prior verbal habits. The nature of the verbal repertoire which exists prior to concept learning determines the way the concept is learned and, consequently, the way in which it can be ~aught (Glaser, 1968). Experiments in this area can be designated generally as examples of the way in which a higher-order conceptual system, once taught to subjects, may make it possible for them to do a kind of thinking they could not otherwise do (Farnham Diggory, 1972). Benjamin Lee Whorf, linguist and anthropologist, developed a hypothesis known as the linguistic relativity hypothesis. Whorf

PAGE 40

26 derived this complex hypothesis after extensive studies of the Hopi and Shawnee languages. He proposed that it was not just the words and values of the cultures that were different; what the peoples per ceived was itself different (Keele, 1973). He concluded that the thought content of these peoples differed very much from that of English-speaking Americans. To paraphrase Whorf's hypothesis, the world of sensation is really a continuum. However, in dealing with large numbers of sensory events, people learn to categorize the world. Stimuli which are close together on a continuum evoke in the memory the same categorical representation. The human's limited processing mechanism is then able to deal with categories rather than a multitude of sensory events. People in different cultures categorize the world of sensations in different ways, and the categories are reflected in the languages of the cultures. The categorizations, in turn, determine what is perceived (Keele, 1973). If so, instructional methods incorporating pictures and/or prose necessarily have to be designed to correspond with cultural peculiarities. Cultural Influence Sapir (1960), from a broad study of languages, did not arrive at the conclusion that cultural (or racial) distinctions of any sort could reliably be attributed to language. There are plenty of instances of similar cultures (e.g., among Indian tribes) having different languages.

PAGE 41

27 Again. there are examples of dissimilar cultures having similar languages (e.g among English-speaking people). It is also Sapir's contention that any language is constructed in such a way that it can respond to the intentions of any speaker, and that any language system can be translated into another. He also mentions that some language forms might be used in a language long before the ideas they contained were consciously recognized by the speakers of the language. For example. perhaps the concept of causation may be expressed quite unconsciously in the languages of primitive societies that have not yet developed formal, scientific concepts of causality (F arnham-Diggory. 1972). The implications of the above alternative interpretation of the language-culture relationship for this study are dubious. Bilingual Children Children who learn two languages more or less at the same time have been able to capitalize on their own natural language-learning powers. Two basic questions regarding classroom relevance can be put forward: 1) Do two languages improve one's thinking ability? 2) Should one teach the second language the way it is naturally learned by bilinguals, that is. by simpl y putting the child into a second-language environment and letting him learn according to his ability ? It is suggested by Farnham -Diggory (1972) that the answers to these questions should be in the affirmative, with some qualification. A general notion has been that knowing two languages may set up a

PAGE 42

28 kind of interference or linguistic "static," which would have an inhibitory effect on learning ability. On the basis of recent investiga tion. Peal and Lambert (1962) disagree with earlier evidence supporting the interference hypothesis. Kole rs' (1963) investigations, which are more technical, suggest reasons for the above findings--for example, that separate verbal memories may exist for each language. Peal and Lambert conchlded that bilingual children (approximately ten years old) were significantly superior to monolingual children on many different measurements of reasoning ability. Intellectually, the bilingual 's experience with two language systems seems to leave him with a mental flexibility, a superiority in concept learning, and a more diversified set of mental abilities, in the sense that the patterns of abilities developed by bilinguals are more heterogeneous. It is not possible to state, from the present investiga tion, whether the more intelligent child became bilingual, or whether bilingualism aided intellectual development, but there is no doubt of the fact that he is superior intellectually. However, one could propose that, because of superior intelligence, bilingual children are also further ahead in school than the mono linguals. Superior achievement of these children in school seems to be dependent on a verbal facility (Peal and Lambert, 1962, p. 20). These findings have a special significance for this study since

PAGE 43

29 all the experimental subjects were exposed to three languages Hence the findings of this investigation may shed some light on whether experience with two or more languages aids in the acquisit i on of scientific concepts. Rationale for This Study India is a diversified country. This diversification is reflected in her cultures. languages. and in many other ways. The country is not united through a common language. although English was. and still is. used to overcome this condition to some extent. Histo-ically speaking. the English language came to India with the Bri tish and has now become a factor in the language controversy. After independence. many Indian educators. scientists. and common people felt strongly that :English should be abandoned and Indian languages should be adopted in the educational system and other official. business. and administrative systems of the country. Hindi was declared to be the national language of Ind i a just before independence (Chatterjee, 1954). but it has not yet been completely im plemented throughou.t the nation. According to the census taken in 1961. there is a total of 1. 652 mother tongues in India. which have been grouped into languages/dialects There are 14 major languages (which includes 380 mother tongues) specified in Schedule VIII of the constitution (India. 1973). It was mentioned by DasGupta (1970) that language diversity has always been a characteristic feature of India.

PAGE 44

30 This diversity creates a tremendous communication problem between the average people of different states and complicates educational processes. The Indian government bas been. and still is, trying to solve these problems. The ew.catiooal aystem in India is bilingual to some extent. The regional language of the state of West Bengal is Bengali. Previously. the study of three languages. that is. Bengali. English. and Sanskrit. was compulsory in this state. Recently. in most schools, Sanskrit, a classical language. has been replaced by Hindi in order to implement the three-language formulas proposed in the National Policy Resolutioo of 1968 (India. 1971-2). lo West Bengal. the medium of instruction in most primary and secondary schools is Bengali. However. there is another set of schools where the medium of instruction is ex clusively English from the elementary level. The student population in these schools is more heterogenous in that a broader spectrum of students from different backgrounds attend these schools. Most of these students speak in English at school but converse in a mother tongue at home There is still another type of school which falls in between the above mentioned categories. where the medium of instruction at the elementary level is Bengali and of the secondary level is English. The situation is quite flexible at the University level in Calcutta. the capital of West Bengal. Lectures are given either in English or in Bengali or in a mixture of English and Bengali. The latter is the most commonly used technique. It is also permissible for students to answer questions in the mother tongue in all subjects except English.

PAGE 45

31 However. it is also accepted that a working knowledge of English will be an asset for all students and a reasonable proficiency in the language will be necessary for students who will go for higher education. Hence. English should be allowed to continue, placing, however, greater emphasis on the mother tongue (Chatterjee, 1954). "Virtually the right of the Indian boy or girl to be instructed in his or her mother tongue. or in whatever language he or she chooses in its place, has been admitted both by the general consensus of opinion and by legis lation" (Chatterjee, 1954, p. 6). At the Poona University linguistic conference held in 1953, it was unanimously recognized that the education of a child must begin in the mother tongue. Also, instruction and evaluation at all levels up to high school should be in the mother tongue. This view had already been accepted by the University of Calcutta (Chatterjee, 1954). The regional languages are already in use as media of education at primary and secondary levels. Many educators are now emphasizing education entirely in the mother tongue at colleges and universities. Chatterjee (1954) mentions that this will inevitably lead to an undesirable condition, that is, the formation of linguistic states. There will be a great damage in the intellectual domain if English is abandoned at the high school stage. which leads to the college. The replacement of English with regional languages will not only isolate people of one region of India from another, but it will also isolate India

PAGE 46

32 from the rest or the world. English is one or the most important European languages or the "world language, 11 and, since it is already being used in India. it is relatively easier for Indians to learn it than any other foreign language. Hence. Chatterjee (1954) proposed that English should be retained side by side with the mother tongue in our university education. At the Poona linguistic conference. held in 1953, the following recommendations were made regarding the use of international scientific symbols and expressions: 1) "All technical terms should be drawn as far as possible from Sanskrit sources. 2) "All international symbols should be retained. 3) "International scientific terms and expressions should be retained if Indian equivalents cannot be framed" (Chatterjee. 1954 .. p. 14). There are also a large number of reactions adverse to retaining English. Some of them seem to be quite legitimate and cannot be ignored as such. It was suggested by Krishna (1974) that there is a psychological disadvantage in using English as a medium of instruction. For example, when children are taught to memorize numbers in English, the objective of learning to count becomes secondary to the misplaced concern of learning to count in English. "From the very beginning. knowledge of arithmetic or mathematics becomes subsidiary to a knowledge of English. A student has to spend a large amount of time in mastering a

PAGE 47

33 foreign language instead of mastering the subject content (Krishna. 1974. p. 20) even when the students are not interested in literature or linguistics (Mathai. 1974). When students are instructed in any science subject in English. they usually become more conscious about the language than the subject matter content. On the contrary. if they are to be taught in a mother tongue. consciousness of medium would play an insignificant part and thus permit them to devote more attention to the subject matter. Krishna (1974. p. 21) mentions. "in considering the causes of our scientific backwardness. a major factor that would emerge is the lack of involvement of young minds in science because it is taught in a foreign language that is not fully understood." Many Indian educators complain about the absence of good texts in our own language. Krishna (1974) refuted this criticism by suggesting that this is a question of time. and in due course we will be able to develop good texts in our own languages. In addition to this language conflict. there are many other problems. drawbacks. and difficulties in teaching science in india. Age-old traditional curricula. testing, and evaluative procedures still exist at most places. Science is taught mostly by using traditional approaches. The major emphasis is on theoretical aspects rather than on practical implications of science. The inquiry or discovery approach is mostly unknown to students. Science teaching is not related to the environment. The present status of science in India has been pointed out by Vaidya (1971).

PAGE 48

34 1) Science instruction has been and is still oral in character and includes only very few demonstrations. Students are not involved in doing practical work in the elementary schools. After this, students are to follow rigidly a prescribed list or experiments. These experiments are mostly in the nature of verifying knowledge. or working according to set rules. 2) Science instruc t ion is based strictly on prescribed textbooks, which are usually not of high quality. 3) Differ entiated and sequential curricula are lacking for various categories of students. 4) Methodologies of s cience instruction are traditional and nonstimulating. 5) Teacher training programs are also inadequate, and "there is hardly any supervision and availability of expert advice in science teaching by those who are really competent, 6) There has been very little research conducted in the fi eld of science education. There is a "complete blackout of research in science education" (Vaidya, 1971, p. 30). Since independence. efforts have bee n made to improve science instruction in India. Government of India has undertaken many programs to upgrade our scientific knowledge. To improve science instruction in India, the following three integrated steps are essential (Vaidya, 1971 ). 1) Development of a curriculum which will include modern concepts and understandings o f the subject-fields and a rigorous, analytic study of fundamentals. 2) Preparation of new textbooks, t eacher's guides/manuals, and other instructional kits and apparatus in order to satisfy

PAGE 49

35 the demands of the new curriculum. 3) Establishment of a new teacher training program in order to equip the teachers with skills required for the new curriculum. A great deal of research is needed to be done in order to improve science instruction in India. To develop a conceptually oriented new curriculum, according to the need for the society. some empirical evidence is essential in order to find out the type of specific concepts to be included for a particular group of students with particular language -cultural backgrounds. Similarly, evidence is needed regarding how students learn from different media and what are the essential factors that facilitate concept acquisition and the acquisition of other types of technical and non technical materials. The latter area, in addition to the med i um of instruction, poses a major problem in science teaching in India. The general notion that students should be taught in the mother tongue from the beginning seems to be very reasonable. It is generally assumed that students will be able to think, process information and express themselves much better in their mother tongue than in any foreign language. However, before someone makes an attempt to replace English. either by Hindi or by any other regional Indian language, some basic investigation is essential in order to find out whether the particular language has the capacity to e x press clearl y and eff i ciently all the ideas and thoughts of science and te chn ology. Decision making

PAGE 50

36 should be based on empirical evidence. An attempt bas been made in this investigation to find out the relative effectiveness of English and Bengali instruction and pictorial or written presentation in acquisition of three scientific concepts by Indian students with different language and cultural backgrounds. Experimental Questions Based upon the previously reviewed research and theory. the following questions were asked: 1) What effect do written passages have upo n acquisition of a scientific concept? 2) What effect do written passages have on retention of a scientific concept? 3) What effect does a single-line pictorial presentation have on acquisition of the same scientific concep t? 4) What effect do pictorial presentations ha ve on retention of a scientific concept? 5) What differential effect does language {Fnglish or Bengali) have on acquisition and retention of a sc ientifi c concept? 6) What effect do individual differences ha v e on the acquisition and retention of scientific concepts?

PAGE 51

CHAPI'ER II EXPERIMENTAL DESIGN The Design A post-test only control group design (Campbell and Stanley. 1971), as described in Table 1. was used with four experimental and two control groups. This design efficiently controls for pre-test sensitization. but it does not provide information about entering behavior. With a large sample. as in this study. one can assume that the entering behavior of subjects is randomly distributed over groups and hence subjects are equivalently distributed over treatments. This design was selected because it permits an adequate evaluation of the relative effects of an independent variable upon a dependent variable with control for major threats to internal validity. Aptitude testing preceded the administration of each treatment, Independent Variables Written Passages. Written passages about three classificational concepts in biology were described in a small booklet. The concepts used were: geotropisms. phyllotaxy. and feathers. This booklet constitutes one of the independent variables. 37

PAGE 52

TABLE 1 EXPERIMENTAL DESIGN Aptitude Tests Treatment Post-Test Retention-Test Letter Sets Vocabulary R Xl Written passages in 01 07 Bengali First and Last R X2 Single-line pictorial 02 Oa Names presentation labeled in Bengali Audltory Number R None Placebo (control) 03 09 Span Letter Sets Vocabulary R X3 Written passages in 04 0 10 English First and Last R X4 S Ingle -line pictorial 05 011 Names pr esenta tion labeled In E' ngll H h Auditory Number R None Placebo (control 08 0 12 Span w 00

PAGE 53

39 Pictorial Presentation. Single-line drawings of the same three biological concepts with appropriate labeling were included in another booklet. This booklet constituted the other independent variable. Aptitude Measures. The following aptitude measures were used !or this investigation. 1) Letter Sets Test (I-1) 2) Vocabulary Test (V-1) and (V-2) 3) First and Last Names Test (:Ma-3) 4) Auditory Number Span Test C\ts-1) Ability Measures The selection of ability measures used for this study was made from the Kit of Reference Tests for Cognitive Factors (French. Ekstrom and Price, 1963). The abilities assessed by the selected measures were thought to influence learning from the two modes of information types presented ii? the treatments. The Kit of Reference Tests for Cognitive Factors consists of groups of tests representing frequently obtained factors in the area of cognitive ability. These tests are not standardized batteries of tests. and the manual does not provide measures of reliability. validity norms. or other information commonly included in a test manual; for this reason. the tests are recommended for experimental use only. A brief description of the ability tests used in this study and the factors they represent follows.

PAGE 54

40 Letter Sets Test (I-1) A test of divergent production of symbolic classifications (a3C) is thought to be concerned with the ability to group items of symbolic information in different ways. In this test students are supposed to indicate different common properties that sets of letter combinations may have in common (Meeker, 1969). Divergent Production: Generation of information from given information, where the emphasis is upon variety and quality of output from the same source. Likely to involve what has been called transfer. This operation is most clearly involved in aptitudes of creative potential. (Meeker, 1969, p. 20). Vocabulary Test (V-1) and (V-2) These tests measure the student's ability to understand the English language. Individual differences are perhaps most obviously seen in the size of comprehensive vocabularies. First and Last Names Test (Ma-3) The first and last names test is a memory test. Memory can be defined as "retention or storage. with some degree of availability. of information in the same form it was committed to storage and in response to the same cues in connection with which it was learned" (Meeker. 1969, p 16). Memory is a well known intellectual ability-well known in the sense that it is one of the oldest. Memory is also universally and historically recognized as a primary mental operation (Meeker, 1969).

PAGE 55

41 This test requires students to remember what kind of last name goes with first name and to recall perfectly for immediate reproduction. Auditory Number Span Test (Ms-1) Memory for symbolic system (MSS) concerns the ability to remember the order of symbolic information (Meeker, 1969). This test involves immediate recall of series of numbers after only one auditory presentation of the series. Table 2 identifies each test and the time required for its administration. Dependent Variables Acquisition of Concepts. A post-test was developed correspond ing to the content materials and was used as a criterion measurement. It contained thirty items; of these, five items dealt with Concept I (geotropism), seven items with Concept II (phyllotaxy), and eighteen with Concept ill (feathers). Retention of Concepts. An equivalent retention test to the post tests was also prepared from the same set of instructional objectives. Methods Subjects A total of 302 Indian students were selected from five different high schools in Calcutta to participate in this study. These schools were selected on the basis of availability, cooperation, and spoken language. The subjects were all science students. They had just

PAGE 56

Test Kit of Reference Tests for Cognitive Factors Letter Sets ( 1-1) Vocabulary (V-1) Vocabulary (V-2) First and Last Names (Ma-3) Auditory Number Span Practice Time TABLE 2 APTITUDE MEASURFSa Total Time 14 minutes 8 minutes 8 minutes 10 minutes 4 minutes (approx,) 1 minute aAll tests were administered during one setting. Number of Parts Given 2 2 2 2 1

PAGE 57

43 finished ninth grade and bad been promoted to tenth grade. A total of 169 students were selected from Bengali medium schools. and 133 students were selected from English medium schools However. due to differential experimental mortality (i.e different students were absent at different testing times, etc.) and the statistical procedures used. only 204 students who took all the tests const ituted the final sample for this investigation. A total number of 102 students was used from the Bengali medium schools. and the same number from English medium schools. Treatment Materials Written passages about three biological concepts were in cluded in a small booklet. These concepts were: geotropisms. phy ll otaxy. and feathers. The textual information was selected from standard biology books. The concept of feather was described more elabora tely than phyllotaxy and geotropism. Critical attributes of each of the concepts were highlighted. The pictorial presentation of the same concepts was included in another booklet. This included single-line black-and-white simple drawings. The critical attributes of each of the concepts were labeled. These treatment materials were written and labeled in..English for a pilot study conchcted by the author at the P. K. Yonge Laboratory School of the U niversity of Florida. The same materials wer e later translated in Ben gali by the experimenter, with the help of some authorities i n th e field. for the Bengali medium schools. Hence. the

PAGE 58

study which followed the pilot sought to replicate findings from the pilot. done with American students, on Indian students taught in Bengali and English. Procedures 44 Aptitude tests were given to all students seven days prior to the treatments. Directions for the administration of each test were followed according to the test manual. Different parts of this test and the directions were included in a small booklet. The test was administered during one sitting. For the First and Last Names Test, English names were replaced by Bengali names, keeping the same number of letters. For example, instead of the first name Edward, Kamala was used. The same rule was followed for other names. In the first part of this experiment, all students in each class room of Bengali medium schools were assigned randomly, using a table of random numbers. to one of three groups: written group, pictorial group, control group. All students in the control group were moved to one side of the classroom, then treatments were administered to the treatment groups. Treatment time, forty minutes, was kept constant for all groups. After the treatments all students were given the post-test. with constant testing time. Students were given a time period of twenty minutes to finish the test. After the post-test, ten minutes was given to the students to write information about themselves on an information sheet. The retention-test was administered seven days after the treatment, again with constant testing time. Twenty

PAGE 59

45 minutes was given to complete the test. The same procedures were followed for the English medium schools, with the exception of an unavoidable time gap of four days between the post-test and retention test. The Instrument and Its Reliability The post-test and retention-tests were prepared to correspood with content material and were included separately in small booklets. These tests were prepared from the same set of objectives. Each test consisted of thirty items. Five items were selected from the concept geotropism. seven from phyllotaxy. and eighteen from feathers. Two types of items were chosen: fill in the blanks and multiple choice. Subsequently. the tests were evaluated and found to be equivalent. A point bi-serial correlation was employed to evaluate the reliability of each item. The range of point bi-serials varied from -0. 014 to 0. 799 for the post-test in Bengali medium schools and for the retentiontest. from 0. 101 to 0. 681. The mean difficulty for the post-test in Bengali medium schools was 0.427. and for the retention-test. 0.416. Difficulty. here. is defined as the percentage of subjects failing to answer the item correctly. The Cronbach alpha of the post-test used in Bengali medium schools was 0. 89, and of the retention-test. 0. 87. The point bi-serial for the post-test in English medium schools ranged from 0.193 to 0. 745. and for the retention-test from 0. 228 to O. 692. The mean difficulty of the post-test was 0. 418, and of the

PAGE 60

retention-test. 0. 389. The reliability coefficient for the post-test is O 88. and for the retention-test; O. 88. Data Collection Procedures Experimental subjects recorded their responses in three separate booklets. The first booklet contained aptitude measures, 46 the second one post-test, and the third one retention measures. Appropriate student information data sheets were attached with corresponding booklets. These tests were bandscored, and the data were transferred to a data summary sheet. Points were given for correct responses; no penalty was given for incorrect responses. The necessary data were then transferred to an IBM data card for statistical analysis. Key punching was done by the experimenter and by trained technicians. Copies of different booklets including treatment mater i als and the instruments are included in Appendices A, B. C and D. These manuals were included in the original form as they were used for the experiment. Table 3 summarizes all of the independent and dependent variables in this study.

PAGE 61

TABLES INDEPENDENT AND DEPENDENT MEASURES Independent Measures Wrltten Passages Pictorial Presentation Letter Sets Test (I-1), Letter Sets Test (I-1), Vocabulary Test (V-1). Vocabulary Test (V-1), Vocabulary Test (V-2), Vocabulary Test (V-2). Part 1 Part 2 Part 1 Part 2 Part 1 Part 2 First and Last Names Test (Ma-S), Part 1 First and Last Names Test (Ma-3). Part 2 Auditory Number Span Test (Me -1 ). Part 1 Dependent Measures Post-teat meaeuree Lnclude1 Relevant question from written passages and pictorial presen tation which include: Concept I (geotropiem) Concept II (phyllotaxy) Concept III (feather) Total achievement Time taken for post-test Retention-test measures include: Concept I (geotropism) Concept II (phyllotaxy) Concept III (feather) Total achievement Time taken for r e tention-teat

PAGE 62

CHAPTER ill RESULTS Purpose of Study and Data Collected The major purposes of this investigation were: 1) To assess the relative effects of pictorial and written presentation upon the acquisition of three biological concepts. 2) To find out if there is any relationship between student characteristics and different modes of instructional content This chapter is a description of statistical procedures and the results obtained in this study. The results from the instructional treatment main effects will be discussed first and then aptitude treatment interactions. Independent and Dependent Measures Independent measures included written passages, pictorial presentations, and nine aptitude measures. Dependent measures were the immediate post-test measures for the acquisition of three scientific biological concepts, delayed retention measures and time used. Means and standard deviations of both independent and dependent variables were computed Table 4 shows the means and standard deviations of three groups of subjects from Bengali medium schools. It is evident from this table that the means o f Ss in the written group 48

PAGE 63

49 TABLE 4 MEANS AND STANDARD DEVIATIO N S OF THE THREE GROUPS (BENGALI MEDIUM SCHOOLS) Treatment Gr ou ps Post-Test M easures Written Group Pictorial Group Control Group Retention-Test M easures Written Group Pictorial Group Control Group Mean 22.29 20.38 8.94 22.06 19.85 10.32 Standard Deviation 3. 86 2.74 2.74 3.11 2.98 2.91 Number of M easurements 34 34 34 34 34 34

PAGE 64

50 (22. 29) on the post-test is n ot mu ch different f rom Ss in the pictor i al group (20. 38) on the pos ttes t. However, bo th o f them are much higher than the Ss in the control group ( 8 94) on the post-test. This trend is also found on the retention-test. Table 5 represents the means and standard deviations for three treatmen t groups obtained from Ss from the English med i um scqools. Here also, the means of Ss in written treatment (22.15) on post-test and of Ss in pictorial treatment (20. 85) are very close to each other. whereas both o f them are much higher than the mean for control subjects (9. 32 ) on post-test. A similar trend is also seen for the retention-test. From these data one may conclude that sub j ects from both types of schools responded in a very similar manner to the different treatments. Table 6 represents means and standard deviations of dependent variables obtained from Ss in Bengali medium schools: 2. 81 for Concept I on post-test. 4. 73 for Concept II. 9. 67 for Concept m. 17. 21 for total achievement. and 14. 67 for time used for post-test. A similar trend was also observed for the retention test. with the excep ti on of time taken, which is 7. 45. Evidently. students took more time on the post-test than on the retent i on-test Table 7 represents means and standard deviations of dependent variables obtained f rom Ss in the English medium schools and these are: 2 78 for Concept I on post-test, 4. 77 for Concept II. 9. 91 for Concept m, 17. 47 for total achievement, 14. 49 for time used on post-test. Again, a similar trend is also

PAGE 65

51 TABLE 5 MEANS AND STANDARD DEVIATIONS OF THE THREE GROUPS (ENGLISH MEDIUM SCHOOLS) Treatment Groups Post-Test Measures Written Group Pictorial Group Control Group Retention-Test Measures Written Group Pictorial Group Control Group Mean 22.15 20.85 9.32 22.80 21. 59 10.03 Standard Deviation 3.04 3.03 3. 59 3.18 3. 36 3.76 Number of Measurements 34 34 34 34 34 34

PAGE 66

TABLE 6 MEANS AND STANDARD DEVIATIONS OF DEPENDENT VARIABLES (BENGALI MEDIUM SCHOOLS) Performance Measures Post-Test Concept I (geotroplsm) Concept II (phyllotaxy) Concept III (feather) Total Achievement Time Taken Retention-Test Concept I (geotraplsm) Concept II (phyllotaxy) Concept III ( feather) Total Achievement Time Taken Mean 2. 81 4.73 9, 67 17,21 14. 67 2. 83 4,92 9. 66 17. 41 7. 45 Standard Deviation 1. 14 2.16 4. 15 6.67 3,72 1. 01 2,01 3,74 5. 89 3.19 Number of Measurements 102 102 102 102 102 102 102 102 102 102 CII w

PAGE 67

TABLE 7 MEANS AND STANDARD DEVIATIONS OF DEPENDENT VARIABLES (ENGLISH MEDIUM SCHOOLS) Performance Measures Post-Test Concept I (geotropism) Concept II (phyllotaxy) Concept Ill (feather) Total Achievement Time Taken Retention -Test Concept I (geotropism) Concept II (phyllotaxy) Concept III (feather) Total Achievement Time Taken Mean 2,78 4,77 9,91 17,47 14.49 3,03 5.t8 10.12 18. 32 7. 52 Standard Deviation 1. 30 2,30 3,67 6,61 3,43 1. 36 2,12 3.80 6. 53 3, 19 Number or Measurements 102 102 102 102 102 102 102 102 102 102 UI w

PAGE 68

exhibited by the retention-test. In general, the means of the retention test are slightly higher than the post-test means, with one exception; that is, the testing time (7. 52) is much lower than the post-test time. Comparison of Tables 6 and 7 shows that the means for the dependent variables for Ss in the Bengali medium schools (6) are very close to the corresponding means (7) of dependent variables for the Ss in the English medium schools; i.e the corresponding values are not much different. Table 8 shows the means and standard deviations of aptitude measures obtained from Ss in the Bengali medium school. The mean scores on these tests were: Auditory Number Span Test, 15. 77; First and Last Names Test. Part II, 9. 34, Part I, 7. 63; Letter Sets Test. Part I. 7. 07. Part II, 6. 61; Vocabulary Test (V-2). Part I, 3. 81; Vocabulary Test (V-1). Part I, Part II. 2. 53; Vocabulary Test (V-2). Part II. 2.10. Table 9 shows the means and standard deviations of aptitude measures obtained from Ss in the English medium schools. The range of means from highest to lowest are: Auditory Number Span Test, 14. 85; First and Last Names Test, Part I, 11. 41, Part II, 10. 91; Letter Sets Test. Part I, 8. 35; Vocabulary Test (V-2). Part I, 8. 33; Letter Sets Test. Part II, 8. 20; Vocabulary Test (V-1), Part I, 7. 69; Part II, 7.13, Vocabulary Test (V-2). Part II. 6.23. A comparison of Tables 8 and 9 shows that in general, the means of aptitude measures f or Ss in the English medium schools were slightly higher than for Ss in Bengali medium schools. Specifically, the means of different vocabulary tests for English medium school Ss were higher than for Bengali medium schools.

PAGE 69

TABLE 8 MEANS AND STANDARD DEVIATIONS OF APl'ITUDE MEASURES (BENGALI MEDIUM SCHOOLS) Standard Number of Performance Measures Mean Deviation Measurements Letter Sets Test. Part I 7. 07 2.90 102 Letter Sets Test. Part II 6. 61 2.66 102 Vocabulary Test (V-1), Part I 2. 53 1. 80 102 Vocabulary Test (V-1 ). Part II 2. 53 1. 66 102 Vocabulary Test (V-2). Part I 3. 81 2. 28 102 Vocabulary Test (V-2). Part II 2.10 1. 66 102 First and Last Names Test, Part I 7. 63 3.91 102 First and Last Names Test. Part II 9.34 3.70 102 Auditory Number Span Test 15. 77 4.34 102 UI UI

PAGE 70

TABLE 9 MEANS AND STANDARD DEVIATIONS OF APTITUDE MEASURES (ENGLISH MEDIUM SCHOOLS) Standard Number or Performance Measures M e an Deviation Measurements Letter Sets Test, Part I 8,35 2,63 102 Letter Sets Test, Part II 0. 20 2 39 102 Vocabulary Test (V-1), Part I 7. 69 3.50 102 Vocabulary Test (V~l), Part II 7,13 3.02 102 Vocabulary Test (V-2), Part I 8,33 2,79 102 Vocabulary Test (V-2), Part II 6 23 2,93 102 F i rst and Last Names Test, Part I 11 41 3.47 102 F i rst and Last Names Test, Part II 10. 91 3, 28 102 Auditory Number Span Test 14. 85 4,45 102

PAGE 71

57 Intercorrelations Be tw een Different Measures Intercorrelations of both independent and dependent measures were computed for the total sample in Bengali and in English medium schools. Table 10 summarizes the intercorrelations of aptitude measures for Ss in Bengali medium schools and Table 11. for Ss in the English medium schools. Evidently. the intercorrelations between the first part and the second part of the same test are relatively higher. This is appropriate because they measure the same mental abilities. In other cases. intercorrelations among the different aptitudes tests are relatively low. again because they measure different mental constructs. Table 10 shows relatively higher intercorrelation between Letter Sets Test. Part I. with Vocabulary Test (V-2) (0. 40) and F i rst and Last Names Test. Part II (0. 45); between Vocabulary (V-2). Part I. with Fil,"St and Last Names Test. Part I (0. 45) and Part II (0. 48). Table 11 also indicates that one part of the test is relatively highly correlated with the other part of the same test. In addition. it also shows that the First and Last Names Test. Part IT. is relatively highly correlated (0. 41) with the Auditory Number Span Test, Tables 12 and 13 show intercorrelations among the dependent variable measures obtained from Ss of the Bengali and English medium schools respectively. Inter correlations among dependent variables both in postand retention tests show relatively higher correlations than among aptitude measures. Most of the dependent variables displayed higher intercorrelation indices with one another and with themselves. The comparisons of

PAGE 72

TABLE 10 INTERCORRELATION MEASURES OF THE APTITUDE TESTS (BENGALI MEDIUM SCHOOLS) Performance Measures 1 2 3 4 5 6 7 8 1. Letter Sets Test, Part I 1.00 0,53 0,32 0,19 0.40 o. 16 o. 37 0,45 2, Letter Sets Test, Part II 1.00 0.34 0.24 o. 27 0.21 0,35 0,30 3. Vocabulary (V-1 ), Part I 1.00 0.54 o. 51 0,54 0.32 0,36 4. Vocabulary (V-1 ), Part II 1.00 0.47 o. 51 o. 21 o. 20 5, Vocabulary (V -2), Part I 1.00 0.43 0,45 0.48 6. Vocabulary (V-2), Part II 1.00 0,26 0.22 7. First and Last Names Test, Part I 1. 00 0,76 8. First and Last Names Test, Part II 1,00 9. Auditory Number Span Test Number or Measure9 ments o. 13 102 0.08 102 o. 18 102 -0.03 102 0,21 102 o. 19 102 0.16 102 0,22 102 1.00 102 UI 00

PAGE 73

TABLE 11 INTERCORRELATION MEASURE'S OF THE APTITUDE TESTS (ENGLISH MJ!'DIUM SCHOOLS) Performance Measures 1 2 3 4 5 6 7 8 1, Letter Sets Test, Part I 1.00 0,62 0,30 0,20 0,34 o. 211 0,17 0, 16 2. Letter Seta Teet, Part II 1.00 0,27 o. 23 0,27 0,24 o. 10 0,07 3, Vocabulary (V-1), Part I 1.00 0,72 0,59 0,64 0,22 0, 06 4. Vocabulary (V;..1 ), Part II 1.00 0.64 0.59 0.15 o. 15 5. Vocabulary (V-2), Part I 1.00 0,70 0,12 0.19 6, Vocabulary (V-2), Part II 1.00 0,29 0.16 7. First and Last Names Test, Part I 1.00 o. 61 8. First and Last Names Test, Part II 1.00 9. Auditory Number Span Teat Number of Measure9 ments 0,09 102 0,04 102 0,06 102 0.02 102 o. 10 102 0.20 102 0.53 102 0,41 102 1,00 102

PAGE 74

TABLE 12 INTERCORRELATION MEASURES OF THE DEPENDENT VARIABLES (BENGALI MEDIUM SCHOOLS) Number of Measure. Performance Measures 1 2 3 4 5 6 7 8 9 10 ments Post-Test 1. Concept I (geotropism) 1.00 0.52 o. 59 0.70 o. 31 0.64 o. 41 0.57 o. 61 -0.42 102 2. Concept II (phyllotaxy) 1.00 0.74 0,87 o. 38 0,46 0.71 0,64 0.73 -o. 23 102 3. Concept III (feather) 1.00 0.96 0.30 0.61 0,66 0,87 0,88 -0.33 102 4. Total Achievement 1.00 0.36 0.64 o. 71 0.84 0.89 -o. 35 102 5, Time Taken 1. 00 o. 14 o. 17 o. 29 0,26 -0.04 102 Retention-Test 6, Concept I (gcotroplsm) 1.00 0,49 0,61 0,73 -0,31 102 7, Concept II (phyllotdxy) 1.00 0,60 o. 81 -0.33 102 8. Concept III (feather) 1.00 0.95 -o. 29 102 9. Total Achievement 1.00 -0.35 102 Cl) 10. Time Taken 1.00 102 0

PAGE 75

TABLE 13 INTERCORRELATION MEASURES OF THE DEPENDENT VARIABLES (ENGLISH MEDIUM SCHOOLS) Number of MeasurePerformance Measures 1 2 3 4 5 6 7 8 9 10 ments Post-Test 1. Concept I (geotroplsm) 1.00 o. 67 0,69 o. 81 -0.42 0.84 0.60 0.66 0,76 -0.51 102 2. Concept II (phyllota,cy) 1.00 0,74 0,89 -0.20 0,71 0,87 0,67 o. 82 -0,41 102 3. Concept III (feather) 1.00 o. 95 -0.21 0,66 o. 69 0,89 0,88 -0.44 102 4. Total Achievement 1.00 -0.27 o. 78 o. 81 0.86 0,92 -0.49 102 5. Tlme Taken 1.00 -0.42 -0.19 -0.20 -0.26 o. 39 102 Retention-Test 6. Concept I (geotroplsm) 1,00 0.68 0,69 0,83 -0.47 102 7. Concept II (phyllotaxy) 1.00 0.67 0.86 -0.36 102 8. Concept III (feather) 1.00 0,94 -o. 36 102 0, o. Total Achievement 1, 00 -0.48 102 ... 10. Time Taken 1.00 102

PAGE 76

Tables 12 and 13 reveal that the correlation coefficient indices for total time taken in post-test for Ss in Bengali medium schools are low positive. whereas in the English medium schools these are low and negative. That is, in the latter case. time is inversely related with the different performance measures. Instructional Treatment Main Effects The following hypotheses represent the major concern or this investigation: 1) Subjects receiving tre atm ents will show s i gnificantly greater behavioral change. in terms of the acquisition and retention of concepts. than control. 2) There will be a significant differential treatment effect between written presentation and pictorial presentation. 62 3) There will be a differential relationship between criterion measures and aptitudes of subjects relative to the treatment obtained. This study will also yield data regarding language and cultural effect. In the tables. groups were named according to the treatment they received. Group 1 received written presentation. hence is known as written group; group 2 received pictorial presentation, hence is known as pictorial group; and group 3 received no treatment. hence is known as control group. Analysis of variance was performed using

PAGE 77

63 both postand retention-test measures. to determine if there were significant treatment effects. Bengali medium school Ss and English medium school Ss were studied separately. A split plot. nested statistical design (BMD 08V). with repeated measures analysis of variance with equal cell size (34) was employed. Separate analyses were done using: total achievement scores (a cumulative score of Concept I. Concept II. and Concept III); scores on Concept I; scores on Concept II; scores on Concept III. and the time used for both the postand retention-tests. Whenever an overall significant F ratio was obtained. Tukey's HSD test. which performs all pairwise compari sons. was employed to determine the pair of means which was signifi cantly different from the others. Acquisition of Concepts Total Achievement on Postand Retention-Test Acquisition of the concepts in this study was measured by an immediate post-test and delayed retention-test. These test scores were analyzed to determine instructional treatment main effects. Table 14 reports the results obtained for analysis of variance using three groups (34 Ss in each group) and the total achievement on two tests for Ss in Bengali medium schools. Significant group differences (F = 212. 7 4, P < 01) were found, and significant inter actions (F = 3. 40. P < 05) between groups and achievement were also obtained. Tukey's pairwise comparison test was performed using both groups (Table 15) and interactions (Table 16). It is evident

PAGE 78

Source Group Total Achievement S(G) G X T" s X T" (G) TABLE 14 ANALYSIS OF VARIANCE FOR CRITERION MEASURES TOTAL ACHIEVEMENT (BENGALI MEDIUM SCHOOLS) Error Term S(G) s X T (G) S x T" (G) df 2 1 99 2 99 MS 3,077,40 2,16 14, 47 18, 01 4, 51 Note-Abbreviations: G Groups; S = Subjects; T" Total Achievement P < 05, **P < 01. G 1 = Written G 2 = Pictorial G 3 = Control T 1 = On Post-Test T 2 = On Retention-Test F 212,74 0,48 3,40

PAGE 79

TABLE 15 TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (BENGALI MEDIUM SCHOOLS) Differences Among Means a 3 = 9. 63 (=20.12 o 1 = 22.1s *P < 01. 10. 49* 65 12.54* 2.06

PAGE 80

Differences Among Means X1 = 8. 94 (G3 x Ti.'> X2 = 10. 32 (G3 X T2'> X3 = 19. 85 (G2 X T2> X4 = 20. 38 (G2 x Ti.'> X5 = 22. 06 (Gl X T2) x 6 = 22. 29 (G 1 x Ti.'> '-' P< 05. ~'* P < 01. TABLE 16 TUKFY 1 S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (BENGALI MEDIUM SCHOOLS) x\ X2 X3 X4 X5 1.38 10. 91 ,,,:, 11.44 13.12 '~ 9. 53 :o:< 10. 05 : ,* 11. 74 '' 53 2.21 1. 6 3 ,, X6 13.35 11.97 ** 2.44 1. 91 .24

PAGE 81

67 from Table 15 that groups 1 and 2 are greater than group 3. Or. in other words. the written group and the picture group performed better than the control group. To rephrase the above statement, there was a significant difference between the treatment groups and the control group. There is no evidence to support the contention that the written and pictorial groups are significantly different. Table 16 depicts the results obtained from Tukey's test performed on the means of the interactions. From this table it can be concluded that interactions x 5 (G 1 x T2) and x 6 (G 1 x Ti') are superior to the rest. Or, in other words, interactions between the written group and the total achievement on retention-test (X 5 ) and post-test (X 6 ) are significantly better than other interactions. Similar analysis of variance results on total achievement for Ss in English medium schools are represented in Table 17. It is evident from this table that the groups were significantly different (F = 169. 55, P < 01) and the total achieve ments were also significantly different (F = 8. 78, P < 01 ). A Tukey's test was performed on group means, and results are shown in Table 18. Evidently, group 1 and group 2 are significantly superior (P < 01) to group 3, and there is no significant difference between group 1 and group 2. It is also evident from comparing means that the mean for total achievement on the retention-test (18. 34) is higher than the mean for total achievement on the post-test (17. 56). After using total achievement scores. i.e scores on Concept I. Concept II and Concept m. for analysis of variance, each concept

PAGE 82

68 TABLE 17 ANALYSIS OF VARIANCE FOR CRITERION MEASL""RE S TOTAL ACHIEVE MENT (ENGLISH ME DI UM SC HOOLS) Source Error Term Groups S(G) Total Achievement s X T" (G) S(G) G x T" S x T" (G) S x T" (G) *P< .01. df 2 1 99 2 99 MS 3.373.16 30.59 19.90 1.27 3.49 F 169.55 8.78 0.36

PAGE 83

TABLE 18 TUKEY'S TEST FOR PAIRWISE COMPARISONS TOTAL ACHIEVEMENT (ENGLISH MEDIUM SCHOOLS) Differences Among Means G3 = 9. a1 G2 = 21. 26 a 1 = 22. 74 *P < 01. 11. 40* 69 12.87* 1.47

PAGE 84

'10 was evaluated separately with respect to the three treatments. The purpose was to find out if the treatments differed significantly from each other with respect to the acquisition of a par ti cular concept. Performance on Concept I Table 19 represents an analysis of variance obtained from using the three treatment groups (34 Ss in each group) and performance on the post-test and retention-test on Concept I for the Bengali medium school Ss. Table 19 shows that the groups are sign ifi cantly different (F = 43. 91, P< 01). A Tukey's test for pairwise comparison between means was employed. and the results are represented in Table 20. Tb.is table shows that group 1 and 2 are significantly different (P < 01) from group 3, and there is no significant difference between group 1 and 2. Table 21 shows the same analysis performed on the English medium school Ss. The results obtained from analysis of variance also show a significant difference (F = 65. oo. P< Ol) between groups. A significant difference (F = 10. 86, P < 05) also exists between tests. Table 22 shows Tukey's test comparing the means of different groups: group 1 and group 2 are significantly greater ( P < 01) than group 3, and no significant difference exists between group 1 and group 2. \\lien com paring the means of post-test and retention-tests for groups 1 and 2. the retention-test (3. 03) is higher than the post-test (2. 78). Performance on Concept II The results obtained from the analysis of variance using performance on Concept II. with three groups (34 Ss in each group)

PAGE 85

TABLE 19 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT I (BENGALI MEDIUM SCHOOLS) Source Error Term df MS Group S(G) 2 45.28 Test S x T (G) 1 0.0120 S(G) 99 1.03 GxT S x T (G) 2 0.12 S x T(G) 99 0.43 Note--Abbreviation: T = Test. *P < .01. n F 43. 91 0.05 0.28

PAGE 86

TABLE 20 TUKEY'S TEST FOR PAffiWISE COMPARISOXS CONCEPT I (BENGALI MEDfUM SCHOOLS) Differences Among Means G3 = 1. 91 c; 2 = 3. 01 c; 1 = 3. 49 *P < 01. 1.16* 72 1.57* o. 41

PAGE 87

TABLE 21 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT I (ENGLISH MEDIUM SCHOOLS) Soarce Error Term elf MS Group S(G) 2 94. 73 Test S x T (G) 1 3.06 S(G) 99 1.46 G:11:T S x T (G) 2 0.25 S x T (G) 99 o. 28 *P<. 01. 73 F 65.00* 10.86* 0.90

PAGE 88

TABLE 22 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT I (ENGLISH MEDIUM SCHOOLS) Differences Among M eans G3 = 1. 54 'a2 = 3. 57 'a1 = 3. so *P< 01. 2.03* 2.06 0.03

PAGE 89

75 and two tests for the Bengali medium school Ss are shown in Table 23. This table shows that the groups are significantly different (F 98. 66. P < 01 ). Table 24 shows the results obtained from a Tukey's test performed on the group means: group 1 and group 2 are significantly better (P< .01) than group 3. and no significant difference is obtained between group 1 and group 2. Table 25 shows a similar analysis for Concept II using the English medium school Ss: the groups differ significantly (F = 123. 56. P < 01). A significant difference (F = 13. 09. P < 0l) also existed between tests. A significant interaction (F = 3.12. P < 05) also existed between groups and tests. Table 26 shows the results obtained from a Tukey's test: group 1 and group 2 are significantly superior (P< .01) to group 3. Table 26 does not indicate the presence of a significant difference between group 1 and group 2. Comparing means of post-test and retentioP.-tests. it is evident that means of the retention-test (5 18) are higher than means for the post-test (4. 77). Table 27 indicates the results obtained from pairwise comparisons of means for interaction: the interactions between group 1 and post-test. group 1 and retention-test, group 2 and post-test. group 2 and retention-test. are significantly higher (P < 01) than between group 3 and post-test. and group 3 and retention test. In other words. interactions between written and pictorial groups with two tests were significantly higher than the control group with two tests.

PAGE 90

TABLE 23 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT II (BENGALI MEDIUM SCHOOLS) Source Error Term df MS Group S(G) 2 251.93 Test S x T (G) 1 1. 96 S(G) 99 2.55 GxT S x T (G) 2 2.36 S x T (G) 99 1. 28 *P < 01. F 98.66 1.54 1.85

PAGE 91

TABLE 24 TO.KEY'S TEST FOR PAIR WISE COMPARISO N S CONCEPT II (BENGALI MEDfl~l SCHOOLS) Differences Among ~l eans 5 3 = 2. 60 5 2 = 6. 01 o 1 = 5. s5 *P < 01. 3.41 77 0.16

PAGE 92

TABLE 25 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT II (ENGLISH MEDIUM SCHOOLS) Source Error Term df MS Groups S(G) 2 332.84 Tests S x T (G) 1 8. 24 S(G) 99 2. 69 GxT S x T (G) 2 1.96 S x T (G) 99 0.63 *P <. 05. **P < 01. 78 F 123.56** 13. 09** 3.12*

PAGE 93

TABLE 26 TUKEY'S TEST FOR PAffiWlSE C0 :\1 PARISONS CONCEPT II (ENGLISH MEDIUM SCHOOLS) Differences Among Means G3 = 2. 43 02 = 6.10 'o1 = 6. 4o *P<. 01. 3.68 79 3. 97* 0.29

PAGE 94

Differences Among Means X1 = 2. 03 (G3 x T1) 'x2 = 2. 82 (G 3 x T 2 ) X3 = 6. 00 (Gl x Tl) X4 = 6. 21 (G 1 x T 2 > X5 = 6.29 (G2 x Tl) X5 6. 50 (G2 x T2) P < 01. X1 TABLE 27 TUKFY 'S TEST FOR PAIRWISE COMPARISONS CONCEPT II (ENGLISH MEDIUM SCHOOL.S) X2 X3 X4 x5 0.79 3.97 4.18* 4. 26* 3.18* 3.38* 3.47* 0.21 o. 29 0.09 x6 4. 47* 3.68* o. 50 0.29 CII 0

PAGE 95

81 Performance on Concept III Table 28 represents the analysis of variance data obtained for Concept musing three groups, (34 Ss in each group), and two tests for Ss in Bengali medium schools. This table demonstrates a significant difference (F = 148. 40, P < 01) between the groups. A Tukey's test was employed for comparing the group means. Table 29 shows the results obtained from this analysis: group 1 and group 2 are significantly superior (P < 01) to group 3, and group 1 is significantly greater (P < 05) than group 2. An exactly similar analysis was per formed using Ss from English medium schools. Table 30 shows the results: there is a significant difference (F = 97. 86, P < 01) between groups. A Tukey's test was performed to find out the locus of difference here. Table 31 summarizes the results obtained from this test: group 1 and group 2 are significantly better (P < 01) than group 3, and no significant difference exists between group 1 and group 2. The data obtained from an analysis of variance, using performance on total achievement, Concept I, Concept II and Concept III for both types of schools, can be summarized as follows: 1) Written presentation (treatment 1) and pictorial presentation (treatment 2) produced greater behavioral change, measured in terms of positive change in scores on the criterion measures, than the control (treatment 3). 2) There is no apparent significant difference between written treatment and pictorial treatment. This pattern

PAGE 96

TABLE 28 ANALYSIS OF VARIANCE FOR CRITERION MEASURES CONCEPT III (BENGALI MEDIUM SCHOOLS) Source Error Term df MS Group S(G) 2 1. 108. 72 Test S x T (G) 1 0.0049 S(G) 99 7.47 GxT S x T (G) 2 5. 59 S x T (G) 99 ~11 *P<. 01. 82 F 148. 40* 0.0023 2.64

PAGE 97

TABLE 29 TUKEY'S TEST FOR PAffiWISE COMPARISONS CONCEPT III (BENGALI MEDIUM SCHOOLS) Differences Among Means a 3 = 5.12 = 11.03 a 1 = 12. 84 *P< 05. **P< .01. G2 5.91** 83 7. 72** 1. 81*

PAGE 98

TABLE 30 ANALYSIS OF VARIANCE FOR CRITERION ME AS U RES CONCEPT III (ENGLISH MEDIUM SCHOOLS) Source Error Term df MS Groups S(G) 2 940.78 Tests S x T (G) 1 0. 83 S(G) 99 9. 61 GxT S x T (G) 2 0.064 S x T (G) 99 1.87 *P< .01. F 97.86* o. 44 0.03

PAGE 99

TABLE 31 TUKEY'S TEST FOR PAIRWISE COMPARISONS CONCEPT III (ENGLISH MEDIUM SCHOOLS) Differences Among Means a 3 = 5. 9 o a 2 = 11.29 a 1 =13.o3 *P < .01. 5. 40* 85 7.13* 1.74

PAGE 100

86 was consistent for Ss from both types of schools. However. there is one exception to this pattern: the case of Concept min the Bengali medium schools. Here. the written group was significantly better ( P < 05) than the pictorial group. A similar set of analyses of variance were done with the time students required on the post-test and retention-test. It was intended that these analyses give some information regarding the time and treat ment relationship. Effects of Post-Test and Retention-Test Time Table 32 demonstrates the data obtained from an analysis of variance using three groups (34 Ss in each group) and the post-test and retention-test time used Ss in Bengali medium schools. Table 32 demonstrates a significant difference (F = 234. 05. P < 01) between test times. It also indicates significant interaction (F = 15. 76, P < 01) between groups and times. It is obvious from comparing the two means that post-test time (14. 67) was larger than retention-test time (7. 72). A comparison between interaction means was performed using Tukey's test. Table 33 summarizes the data. showing that an interaction. between the written group for post-test time and the pictorial group for post-test time. was significant (P < 01 ). A similar analysis was performed using the Ss in English medium schools. The results are represented in Table 34: a significant difference (F = 17. 23,

PAGE 101

87 TABLE 32 ANALYSIS OF VARIA N CE FOR CRITERIO N MEASURES TIME USED FOR POSTA N D RETENTIO N -TEST (BENGALI MEDIU M SCHOOLS) Source Group Time S(G) GxT' s X T' (G) *P < 01. Error Term S(G) s X T' (G) s X T' (G) elf 2 1 99 2 99 MS 19.07 2,464.12 12.74 165.97 10.53 Abbreviations: Ti = (1) Time used for post-test~ T 2 = (2) Time used for retention -test. F 1.50 234.05 15. 76

PAGE 102

Differences Among Means X1 = 6,97 (G2 X T2> X2 = 7,21 (Gl X T2> x 3 = a. 97 (G 3 x 'fd> X 4 = 12.47 (G2 x Ti> X 5 = 14. 97 (Gl x Ti) X5 = 16. 56 (G2 x Ti) P < 01. TABLE 33 TUKEY 'S TEST FOR PAIRWISE COMPARISONS TIME USED FOR POSTAND RETENTION-TEST (BENGALI MEDIUM SCHOOLS) X4 X5 5,50 8,00* 1. 76 5. 26 7. 76* 3 50 6. 00* 2 50 X5 9. 59* 9,35* 7. 59* 4,09 1,59 011 011

PAGE 103

Source Groups Time S(G) GxT' TABLE 34 ANALYSIS OF VARIANCE FOR CRITERIO N MEASURES TIME USED FOR POSTAND RETENTION-TEST (ENGLISH MEDIUM SCHOOLS) Error Term df MS 89 F S(G) 2 s X T' (G) 1 205.30 2,457.18 17.23 356.40 99 11. 91 s X T' (G) 2 19.07 2.77 s X T' (G) 99 6. 89 *P<. 01.

PAGE 104

90 P < 01) between groups and between times (F = 356. 40, P < 01 ). Group means were compared and the results are presented in Table 35: group 3 exceeds (P < 01) group 1 and group 2. Or, in other words, the control groups used more time than either written or pictorial groups. Comparisons between the means of post-test time (14. 57) and retention-test time (7. 63) show that the post-test time was greater than the retention-test time; that is, students used more time for the post-test than the retention-test. Aptitude X Treatment Interactions The major purpose of these analyses was to find out if there were relationships between student aptitudes and the treatments Aptitude X treatment interactions were studied by comparing regression slopes obtained, for each aptitude-criterion pair for the three different treatments, using an F test for homogeneity of regression. Out of 180 regression analyses performed, 8 were significant at 0. 05 level. These results are represented in eight graphs. These graphs, Figures 1 through 4, were obtained from the measures on Ss in Bengali medium schools, and Figures 5 through 8 were obtained from the measures on Ss in English medium schools. Figure 1 compares the three regression slopes correspond in g to the three treatments, with the aptitude measured by Letter Sets Tests (the divergent production of symbolic classification) and the criterion measure, post-test time used. Both the written (w) and pictorial (p) treatments

PAGE 105

TABLE 35 TUKEY'S TEST FOR PAIRWISE COMPARISONS TIME USED FOR POSTAKD RETE N TION-TEST (ENGLISH MEDIUM SCHOOLS) Differences Among :\leans a 1 = 10. 03 a 2 =10.1s a 3 = 13.10 *P < 01. 91 2.94*

PAGE 106

20 F = 3,17 (P< ,05) 16 C p E-< tll .I E-< 12 I E-< ci:: 0 rx. 8 Cl .I Ul :::> J .... E-< 4 0-----------------------.. 0 3 6 9 12 15 LETTER SETS TEST, Part II Figure 1: Regress i on Analysis for Letter Sets Test ( Part II) on Post-Test Time Used (Bengali m ed i um school Sa) Abbreviations: w = written treatment, p = pictorial treatment, c = control treatment co N

PAGE 107

93 show a negative relationship (negative slopes) whereas the control (c) shows a positive relationship (positive slope). The written group started slightly higher than the pictorial group and decreased more than the pictorial group. There is an interaction between control group and the written and pictorial groups. This is evident by the inter section of planes for two treatments and control. There is also an interaction between the written and pictorial treatment groups. Figure 2 represents regression analysis for the First and Last Names Test, Part I (a memory test), on post-test time. Control group shows a positive slope, written group shows a parallel slope. and the pictorial group shows a negative slope. There is also an interaction present between the pictorial treatment group with the written and control treatment groups. Figure 3 represents the regression analysis for the First and Last Names Test, Part II (memory test). on post-test time used. Control group shows a positive slope. whereas the pictorial treatment and written treatment groups show negative slopes. Pictorial group shows a steeper slope than written. There is also an interaction present between pictorial group with written and control groups. Figure 4 shows regression analysis for First and Last Names T est. Part II (memory test). on retention-test time used. Control group shows a relatively steeper positive slope. whereas written and pictorial groups show negative slopes. These two slopes are almost parallel with respect to each other. Figure 5 depicts the regression analysis

PAGE 108

20 F = 3, 60 (P < 05) 16 w C E-p Cl.I 12 I f.; u, 0:: 8 Q t/l ;::> ;:-;E 4 E-< 0 0 3 6 9 12 15 FIRST AND LAST NAMES TEST, Part I Figure 2: Regression Analysis for First and Last Names Test on Post-Test Time (Bengali medium school Ss)

PAGE 109

20 F II. 02 (P < 01) 16 w e,.. C l'/l p 12 I e,.. rn ex: 0 r.. 8 0 .I Ul ;:) .I .... ,.,: E-< 4 o._ ___ __. ____ ..._ ____ ,_ ________ _. 0 3 6 9 12 15 FIRST AND LAST NAMES TEST, Part II Figure 3: Regression Analysis for First and Last Names Test (Part II) on Post-Test Time Used (Bengali medium school Ss) co CII

PAGE 110

20 F = 4. 50 (P < 05) f'4 16 Ul I z 9 E-< 1 2 z C E-< ll:. ll:. 0 8 rz. 0 w Vl :::> p .... 4 E-< o ____ _... ____ .._ ____ .__ ___ ....., ____ 0 3 6 9 12 15 FIRST AND LAST NAME TEST, Part II Figure 4 ;: Regression Analysis for First and Last Name Test on Time Used for Retention-Test (Bengall medium school Ss)

PAGE 111

p 5 F 3, 32 (P < 05) 4 w ... 3 t r..l u z 0 u 2 C 1 0 -------------------------------' 0 3 8 9 12 18 VOCABULARY TEST (V-2), Part 11 Figure 51 Regreaalon An a lysis for Vocabulary Teet (V -2 ), Part II, on Concept I ( Englleh medium school Se)

PAGE 112

98 for Vocabulary Test (V-2). Part II (measures knowledge or English). on Concept I. Here the written group and the control group show a parallel slope. This figure shows a steeper positive slope for the pictorial treatment group. Figure 6 is the graphic representation of the regression analysis for the First and Last Name Test (memory test), on Concept I. It is seen that the written and the control groups show negative slopes. The control treatment group has a greater negative slope than the written group, whereas the pictorial group shows a steeper positive slope. There are also interactions between the pictorial and the written groups and the pictorial and the control groups. Figure 7 shows regression analysis for the Auditory Number Span Test (memory test) on Concept I. Even though both the written and the control groups show negative slopes, the control group shows a steeper decrease than the written group. The pictorial group has a steeper positive slope. Figure 8 depicts the regression analysis for the Vocabulary Test (V-2). Part I (measures knowledge of English), on Concept III. Here the written group shows a slight negative slope. while both the pictorial and the control groups show positive slopes. There is also an interaction present between the written and the pictorial groups. In summary, the aptitude X treatment interaction analysis shows that, for Ss in Bengali medium schools: 1) Both the written and pictorial treatment groups are negatively related to Letter Sets Test, Part II (divergent

PAGE 113

5 Fa 7.46 (P< .01) 1 0 '---....ii..... __ __., _____________ .. 0 3 6 9 12 111 FIRST AND LAST NAMES TEST, Part I Figure 6: Regression Analysis for First and Last Names Test on Concept I (English medium school Sa) co co

PAGE 114

4 ... fi: 3 .I u u 2 p F 5,46 (P< ,01) 1-----------------------------.. 0 4 8 12 16 20 AUDITORY NUMBER SPAN TEST Figure 7: Regression Analysis for Auditory Number Span Test on Concept I (English medium school Ss) 24 ... 0 0

PAGE 115

101 F = 3. 78 (P < 05 ) 18 16 14 p 12 C 10 ... ... ... t 8 u z 0 u 6 4 2 0 0 3 6 9 1 2 1 5 18 VOCABULAR Y TEST ( V -2), P a rt I Figure 8 : Reg r e ssi o n Ana l ysis for V o ca bu l a ry Test (V-2), Pa r t I on Concept III ( Englis h mediu m school S s )

PAGE 116

102 production of symbolic classification). The written group shows more negative slope than the pictorial group whereas the control group shows a positive relationship with the aptitude mentioned above. 2) The pictorial group shows an inverse relationship with the First and Last Names Test, Part I (memory test). The written group shows no relationship, and the control group shows a positive relationship. 3) The pictorial group shows a negative relationship with the First and Last Names Test, Part II (memory test). The written group a::..:;o shows a slight negative relationship with the above-mentioned aptitude test. whereas the control group shows a positive relationship. 4) Both the written and pictorial groups show an inverse relationship with the First and Last Names Test. Part II (memory test). whereas the control group shows a positive sl:ope. From Ss in the English medium schools: 5) The pictorial group shows a positive relationship with the Vocabulary Test (V-2). Part II (measures knowledge of English), whereas the written and control groups show no relationship. 6) The pictorial group shows a positive relationship with the First and Last Names Test. Part I (memory test).

PAGE 117

103 The written and the control groups show negative relationships. However. the control group shows relatively more negative slope. 7) The pictorial group shows a high positive relationship with the Auditory Number Span Test (memory test). whereas the written and the control groups show negative relationship~ However. the control group shows a more negative slope. 8) The pictorial group and the control group show positive relationships with the Vocabulary Test (V-2). Part I (which measures knowledge of English). but the written group shows a slight negative slope. In general. these data further substantiate the fact that the control group is different from both treatment groups with respect to corresponding aptitude measures. In addition. it also reveals differential effects of written and pictorial treatments with respect to the aptitude measures.

PAGE 118

CHAPn'.R IV DISCUSSIO~ A~D I:\IPLICATIONS Summary of Data and Interpretation This study sought to examine the relative effectiveness of written and pictorial presentations on the acquisition of three bio logical concepts by students ta ught in Bengali and in English. The scores of subjects on the post-test and retention-test measures were taken as ev i dence of acquisition of thes e concepts. Comparing the data of means in Table 4 with Table 5 and in Table 6 and Table 7. it can be conc luded in general. that Ss from Bengali med iu m and English medium schools show a similar per formance on criterion measures. 'W hil e it was expected that students speaking na tiv e Bengali and taugh t in Bengali would perform differently than those tanght in English. due to linguistic difference, this was not substantiated. From the data on m eans of treatment groups using Ss in Bengali medium schools (Table 4) and Ss in English medium schools (Table 5), one can concl ude that the treatment groups showed a greater behavior change than the control groups, but the difference between the treatment groups was not significant. Table 6 shows the means and standard deviations of dependent variables obtained from Ss in Bengali medium schools. and Table 7 represents Ss in English medium schools As mentioned in Chapter III, a similar trend was 104

PAGE 119

105 observed for different dependent variables (achievement in Concept I. Concept II, Concept III. total achievement, time used) in the post-test and the retention test, with the exception of testing time. The finding of a similarity in achievement scores for corresponding measures is consistent with the Ss in both types of schools. Students from both types of schools used more time during the post-test than the retention test. This finding can be attributed to the fact that during the posttest Ss were introduced for the first time to the criterion measures. In addition, multiple choice .types of objective items are not commonly used in most schools in India. Since Ss were not familiar with the criterion measures in general and specifically with multiple choice types of items. they took more time to understand each item and to respond. Students used less time for the retention-test. which is not unusual since they were already familiar with the criterion measures. Another interpretation might be that some interference due to the time gap (between the postand retention-tests) resulted in forgetting. or it might be that the immediate recall of different items during post test was easier than the delayed retention-test. Hence, students used less time because they forgot some items. However, it does not appear that the latter is the most plausible explanation. It is evident from comparing the individual mean scores of each dependent variable (Tables 4, 5, 6. 7) on the post-test with corresponding scores on the retention-test that they are very close for Ss in Bengali medium schools (Tables 4, 6). It is even slightly higher for Ss from the

PAGE 120

106 English medium schools on the retention-test (Tables 5, 7). This gain on the retention-test may be due to the fact that Ss acquired the relevant information either from consulting each other or other teachers or from consulting books during the time gap. It may be inferred from data on the Ss in Bengali medium schools (Tables 4, 6) and Ss in English medium schools (Tables 5. 7) that student performances do not differ much between these two types of schools with respect to the treatments. This finding leads one to the conclusion that. if students are taught from the elementary level with a foreign language. the student can process information and acquire concepts as well as a comparable group of students who ar e taught in the mother tongue from the very beginning. One s i, gnificant departure from this generalization will be discussed shortly. Comparison of Table a 8 and 9 shows that the means of each individual aptitude measure for Ss in English medium schools is slightly higher than for Ss in Bengali medium schools except on the Auditory Number Span Test. Specifically. the means on the vocabulary tests for the English medium school Ss are much higher than for the Bengali medium school Ss. One way to interpret this is that the booklets for administration of the aptitude measures were in English and, since Ss from English medium schools were taught from the elementary level in English, these Ss had a richer English vocabulary than equivalent Ss who were taught in Bengali from the very beginning and learned English as a second language. What is surprising is that this advantage in vocabulary

PAGE 121

107 is not reflected in their average performance on the criterion measures. One would expect, however, to be able to shed light on this finding with the aptitude treatment interaction data. Examination of Tables 10 and 11 shows that the intercorrelations of the two parts of the same aptitude measure are relatively higher. This higher intercorrelation is appropriate because these two parts measure the same mental abilities. In other cases. intercorrelations between different aptitude measures are relatively low. thus sub stantiating that the different aptitude tests used measure different mental abilities and are independent. It is evident from Tables 12 and 13 that the intercorrelations between dependent variables are, in general, higher than between independent variables. maybe because they measure similar cognitive abilities. Again, the correlation coefficient between total achievement and Concept I, Concept II, Concept m. is higher, because these concepts as a group constitute the total achievement score. Instructional Treatment Main Effects Acquisition of Concepts The first hypothesis tested was: 1) Subjects receiving treatments will exhibit greater behavioral change in terms of acquisition of concepts than Ss in the control group. Support for this hypothesis was derived from analysis of variance

PAGE 122

108 and subsequent Tukey's tests comparing group differences. There were significant group differences exhibited on total achievement. Concept I. Concept II. Concept III, and time used. The results obtained from both types of Ss, Bengali and English. are quite consistent in this respect. Hence, it ~an be concluded that Ss who received either treatments show':d greater acquisition of concepts than Ss in the control group. or. in other words, the treatment made a difference between groups. This is further substantiated by results obtained from regression analysis. Differential Effects of Written and Pictorial Presentation The second and third hypotheses tested were: 2} There will be a differential effect of written and pictorial treatments. 3) Students with particular characteristics (or abilities) will perform better in one treatment than another. Support for the second hypothesis is dependent upon significant between-group differences in post-test and retention-test measures. Appropriate statistical analyses involved significant group differences followed by comparisons of group means. Analyses using the total achievement for both types of Ss failed to support this hypothesis. However. each concept was evaluated separately with the three treatment groups in order to determine wh ether groups differed from each other with respect to a particular concept. No significant group

PAGE 123

109 differences were observed using Concept I and Concept II for both types of schools. However. for Concept ill, Ss in Bengali medium schools, written group Ss performed significantly be tte r ( P < 05) than Ss pictorial group. However. in the English me di um schools ttie same Concept III showed no significant difference between the written and pictorial groups. One interpretation of this result could be related to the way the concepts were described; that is, Concept ill was analyzed Ex Post Facto and appeared t o be described more elaborately with a greater variety of positive and negative examples. Since written texts are the major source o f inf ormation used in most schools in India. students are more fam il iar with pro cessing information from written textual material than p i ctorial; it is likely that pictorial information involved more co m p l ex information processing and placed a larger burden on cognitive processes. Thus processing involves decoding symbolic information t o verbal and then translating it to a familiar viable form. As Gagne (197 3 ) bas pointed out, concepts probably occur hierarchically. with m ore complex concepts subsuming simpler ones. These particular groups of students may already have been equipped with the prerequis it e kn owledge; hence, when they were exposed to the treatment. i t w as easier for them to acquire than for the other groups. Further research would be necessary to determine. conclusively. cause and e ffect relationships. The superiority of the wri tt en group for Co n ce pt ill can be attributed to the fact that perhaps this group alre ady bad a general

PAGE 124

110 knowledge (Ausubel, 1968) about the specific concept; as a consequence, when more specific detail about the concept was introduced, it was easier ror the written group to acquire the concept. This finding is quite consistent with a pilot study conducted by the experimenter using twenty-nine American eighth-grade science students. These students were randomly assigned to one or three groups: the written group consisted of ten Ss, the pictorial group consisted of ten Ss, and the control group consisted of nine Ss. A pre-test post-test control group des ign (Campbell and Stanley, 1971) with retention-test was employed to find out instructional treatment main effects. The same three concepts as in the present study and the same criterion measures, with the exception of an additional equivalent pre-test (to evaluate entering behavior since groups are v ery small), were used to assess instructional treatment main effects for the pilot The reliability coefficient for the instrument was O. 78. and the mean difficulty was 0. 572. The pre-test was given to all twenty-nine Ss. After that, the control group was removed from the room, and two treatments were administered. All twenty-nine Ss took the post-test aft er the treatment. Treatment time (forty minutes) and testing time (twenty minutes) were kept constant. A reten ti on-test was adm inist ered after five days. A one-way analysis of variance was used to tes t th e main effects. It was found that student performance on the pre-test was n ot

PAGE 125

111 significantly different. In other words. the groups were equivalent before treatment. The post-test scores showed significant differences (P < 01) between the two treatments and the control. Analysis of retention-test scores also showed a significant difference (P < 01) between the treatments and the control. It is evident from these results that the treatment made a difference between groups, since Ss in each exhibited greater behavioral change than those in the control group. The results obtained from the pilot study support these findings. On the basis of these data, one could speculate that science has a universal language which permits communication, on the average, to be similarly effective across culture. regardless of the nature of the moade of presentation. However, differences in the effectiveness of one mode over the other appears to become apparent as one looks at highand low -ability students. Ciborowski and Cole report similar data from a cross-cultural study of conjunctive and disjunctive concepts. They used American and Liberian subjects differing in age and educational background. They concluded that "Overall, the performance of both cultural groups was strikingly similar; evidence of cultural differences centered on measures of the way stimuli were classified into subclasses and the relation between learning and verbalization" (Ciborowski and Co l e. 1972. p. 774). Another study conducted by Allen and Shannon (19TI). on students at the University of Papau and New Guinea, using concept selection strategies among students from a non-Western culture.

PAGE 126

112 reported that subjects did not differ much in the two different cultures. Some interactions were also obtained from the analysis of variance: 1) G x T" (Groups x total achievement) (Tables 14, 16) A significant interaction ( P < 05) was obtained between groups and total achievement. This interaction can be explained as. depending upon task. some groups achieved more than others. Or. in other words. this additional effect was due to combined influence of groups and achievement. An inspection of Table 16 shows that the written group of Bengali medium school Ss showed a superiority in total achievement in both postand retention-test scores to the other groups. This finding also supports the previous finding that the written group performed significantly better for Concept ill in Bengali medium schools, that is probably due to the fact that Concept III is also a part of total achievement. 2. G x T (Group x test) (Tables 25, 27) A significant interaction (P < 05) was obtained between groups and tests for Concept II for Ss of the English medium schools. In general. it can be concluded that. depending upon the nature of the concept. some groups gained more on one concept than others. Table 27 reveals that both treatment groups showed superiority on both the postand retention-tests to the control group. This finding also supports the previous finding on main effects that. in general. both

PAGE 127

treatment groups were significantly superior to the control group. Another set of similar analyses was performed to investigate the relationship between groups and times. Or. in other words, whether the treatment groups differed with respect to the time they used for postand retention-test measures. Table 32 depicts the analysis of variance data obtained using postand retention-test 113 times for Ss in the Bengali medium schools: the times are significantly different. with a significant interaction between groups and times. A comparison of the mean time of the post-test (14. 67 minutes) and retention-test (7. 72 minutes) times indicates that Ss used more time for post-test than for retention-test, perhaps due to practice effect. In general, the interaction between groups with t esting time can be explained as depending on concept tasks; some groups used more time than others on certain of the concepts. This finding provides additional credibility to the contention that concepts may have been hierarchically organized and also that, as they became more complex the processing burden increased. Table 33 points out that the interaction between written group with post-test time and pictorial group with post-test time is superior to the others. Similar analysis using Ss from the English medium schools (Table 34) shows significant differences between groups and between times. One can conclude from Table 35 that the control group was superior to the other two groups; that is, they took more time than either treatment group for the criterion measure. Again. comparing

PAGE 128

114 mean times on post-test (14. 57 minutes) and retention-test (7.63 minutes). it is evident that Ss used more time for the post-test than the retention-test. The results discussed so far are based on group means for the analysis since the purpose here was to find out if there were differences on the average Many statistical analyses use the mean score. which represents a typical score for the whole group. They give some evidence regarding which group is superior or inferior to the o~er and to what extent. However. since the mean is a single score representing the whole group. it ignores the variability of scores. or individual differences of subjects and their resulting performances. and hence. sometimes fails to represent these differences adequately. In order to explore further the differential effects of the treatments on subjects with different characteristics. the aptitude x treatment interaction were employed. The presence of an aptitude x interaction suggests that the treatments were more or less effective for Ss with different aptitude patterns and that. using this data. instructional treatments could be adapted to the individual differences of Ss. However. instructional treatments can be adapted only for those cases where alternative treatments leading to the same terminal objective have different effects on different Ss. and only if the regression of criterion scores on aptitude scores of one treatment intersect the regression slope for the alternative treatment (Cronbach. 1965).

PAGE 129

115 The positive slope of the control group (Figure 1) suggests that it is positively related with the Letter Sets Tests. It also points out that students with higher aptitudes in Letter Sets Tests, a measure or inductive reasoning. used more time for the criterion measures. It appears that Ss high in this aptitude used it and thus took more time to C-md out the answer for an item to which they had not been exposed previously; whereas both the written and pictorial groups show negative slopes. hence inverse relationship with the same aptitude measures. The written group started slightly higher than the pictorial group and decreased more than the pictorial group. From Figure 1, one can conclude that the pictorial and the written treatments would be better for students who score lower on Letter Sets Tests. This finding further suggests that control Ss without a treatment but with higher aptitudes and more time could not perform as well as either written group Ss or the pictorial group Ss. These findings are consistent with the previous findings on main effects. The control group also shows a positive slope (Figure 2), hence a positive relationship with the First and Last Names Test (a measure of memory); whereas the pictorial group shows a negative slope, and the written group shows a parallel slope. These can be interpreted as meaning that the control group Ss with higher aptitude measures used more time to consider the answers in the criterion measures. Pictorial treatment will be better for Ss who score low on this aptitude measure. The written treatment shows a parallel or zero

PAGE 130

116 slope. which means that there is no relationship between the First and Last Names Test and post-test time for this treatment group. Figure 3 also shows a positive slope for the control treatment; that is. Ss in this group with higher aptitudes in First and Last Names Test used more time on post-test. whereas both written and pictorial groups show a negative relationship. Hence, students with low scores in First and Last Names Test will benefit from written and pictorial treatment. Figure 4 also shows that the control group is positively related with the First and Last Names Test. whereas both the written and the pictorial groups showed negative relationship. One can interpret this as written and pictorial treatment would be better for students who score low in First and Last Names Test Again, written and pictorial treatment slopes are parallel, meaning that Ss scoring the same on the aptitude tests also perform the same way on the treatments. In Figure s. both the written and the control groups show parallel slopes or zero slopes; that is, there is no relationship between Vocab11lary Test and Concept I for these two groups. On the contrary. the pictorial group shows a steeper positive slope, hence a positive relationship with the Vocabulary Test. This means that pictorial treatment is better for those Ss who score higher in Vocabulary Test (V-2). Perhaps these students use this ability for complex information processing. translating pictorial symbols into verbal and then interpre tin g them Figure 5 shows an ob v ious

PAGE 131

117 difference. between the two treatment groups with respect to the Vocabulary Test that was not disclosed by the main effects in Ss from the English medium schools. The pictorial group also shows a steeper positive slope (Figure 6). whereas both written and control groups show negative slopes. control group being more negative. Here. students with higher scores on the First and Last Names Test. which is a memory test. will do better with the pictorial presentation. This aptitude probably helps in the recall of symbols which require decoding. It also shows that written treatment is better for those students with lower scores in First and Last Names Test. From Figure 7 it can be concluded that Ss with higher scores on the Auditory Number Span Test will do better with pictorial treatment. while Ss with lower scores on this test adapt better to written treatments. It is evident from Figure 8 that pictorial treat ment will be better for Ss who score higher in Vocabulary Test. whereas written treatment will be effective for the student who scores low in Vocabulary Test. Here the control treatment also shows a positive relationship with the aptitude test. It may be that students in this treatment group used this skill to explore correct response on the criterion measure. Figure 8 also suggests that there is a differential effect of pictorial and written treatments with respect to particular aptitude measures (Vocabulary Test V-2). Implications for the Educatio n a l Practice The results of this inves ti gation support the initial hypothesis

PAGE 132

118 of significant treatment effect. The control group used more time and even when aptitude x treatment interactions were taken into account. subjects in the control group could not exceed the treatment groups. These data illuminate further the importance of instructional treatments. This study did not demonstrate the differential effects of the two treatments except in one case (Concept III in the Bengali medium schools). In the latter case. the written group was significantly different from the pictorial group. lf no significant differential treatment effects had existed (between the written and the pictorial group). and/or if the written treatment was significantly different. recommendations may be made to use written textual materials as an effective means of instruction. since written materials are easily available as tl'ell as economical. However, further investigation is needed on this issue before one can make such a generalization. Sunilarity of results between the present study and the pilot suggest that, most likely, science {biological) has a universal language whic h allows communication, on the average, across culture. regardless of mode of representation. 'nie comparison of average achievement scores on the criterioo measures of Ss in Bengali medium schools with Ss in the English medium schools reveals that they are very close to each other. This result allows us to extend our generalization that, if, on the average. students are taught from the ,ery beginning using a particular language.

PAGE 133

119 they can think. understand, and process information in the same way as in the mother tongue. Regardless of this finding, this particular topic requires more elaborate investigation using students with different mother tongues and language backgrounds. If the above mentioned results could be replicated and the above generalization made more confidently. it would help a great deal to solve a part of the language problem in India. The results obtained from the aptitude x treatment interactions could provide a general guideline toward individualization of instruction by student characteristics as well as rate of learning. Both of these methods are almost nonexistent in India. Limitations Although the results obtained from this investigation satisfied the major purposes of this study. the collected data introduced some questions regarding methodological and other technical issues. For one. the experimental design should also include a group that will receive both pictorial and written treatments, in order to provide data regarding additional or interfering effects. if any. In addition, design steps and sampling methods could be adjusted to attempt to increase the generalizability of the data. To evaluate the relative merit of each concept. one should try to make each concept equivalent with respect to complexity and number of items on the criterion measure.

PAGE 134

120 Multivariate analyses would ha v e b een another statistical approach which wo ul d allow evaluat i on of all the variables together in the same exper im ental design. A b roade r spectrum of different aptitude tests. concepts from differe n t dis c i plines of science, and a larger representa tiv e sample would a lso hav e aided in generalization. Finally. additional research could be c onm cted using monolingual students and students from different cultic:ral backgrounds i n the same experimental format.

PAGE 135

APPENDICES

PAGE 136

APPENDIX A TREATMENT MATERIALS FOR THE BENGALI MEDIUM SCHOOLS

PAGE 137

ms ,~ ~'.iC~lf ~11 f.1/1;1!" 1 m~r-~B ~f5"! .!!t~1~t ~If W I ~~lf~f.;i ~1(1 1!t-! l
PAGE 138

5IJ '>'1"i ( Growth m u vem,nt ) ~~ii fH f.l(rfi'.! O ~If~ ii;IJ.ffi (Stimulu") C11 fir~ m 'l!f1~1"5
PAGE 139

JRt5C1 1'1..n"I ,irq \.. 125

PAGE 140

126 -..~.._l "IT<'lI-? '1T"llil 'Sl-0 ,e (l"fl f 'l I ~ll (I~ ~'1il llH:f'l~ <111:"IC("( Jfo -i (rachis) i~ flA llllJ>!i!"l'tT-! ~:EM >llt"l_-1 (~l! Spo11gt1 of air filled <' ell s) 'iTTI 1_.( 1 (rachit1) ~'!I '1Tr-f ~~ltoll f.l~~ (ISol (Yane) I: i "IT"TT"fTf"l ~ nfo a:Bf.l ~C"l '! ~f"l ..!IC4 m7-!CllC"lj is"!il DTf"f11 1 o!fC~~ m~i5C"l"il (barbul,) '51~1Sl11 e .;)csilfuiHf>IClf':'!ll ( barbicd:1) <1"1 1 n 1 >l f il~~ll'P\~f"'I 1111 ~ClTC1f ~~llCI! >ltllllJ ~il I ( Down fo.iher .. ) : 9f f "!'l\~f.--i l]__"l~ ~>Ill HCll"il fit:~ ha 11h! 9f~"!l ~'I I ~lill 1i_~i! (!ft~ .!!l!lol "!!(:"l~'l ~>I~ 91._9<1~~ "11ftlil I ~ Clfm ~"111~ Jl:"il'll'C'l .!!'1-1 'llHll!"il~ (barbs) ,a t "fl"!4 (Filoplwnros) : ~i" ~Til 9!T"IC11 (Pin feath .. r$f I ~~T"il1 "'J'I (~ 5 ?C"lil l.C-!1 "11"1~ I l~ICH ~-6~'8 (Shaft) ,"I '3~ I TITCTI llCIIJ ~'(-1 if."1-,'8 11 Clf..~"! (Rt-miges) U I W ~,a C"IC!f >IC9] >I~ '1Tcf1"11 H I "1TT'iil ClfITT 11 -rt ITT
PAGE 141

127

PAGE 142

128 ,Jr',r". y79'/St5{(T'ROPISM5) 1f ":a7 b;/~ ( THE G;ROWTH MOVEMENT

PAGE 143

J/T!,}, r;!) P7'!1Jlf(:/! 'v7~7?f !fl~r ~l2grc12 129 'vljf~-\Jili;pf c;,_ (POSITIV f \JfJJ,<1>$ G.EoTROPJSJf) ( NEC,RTIVE C.EOTl?.OPISM) /"""'-. / r-~t0; 4> :iJ 1] 3 (GEOTROP!S!15)

PAGE 144

130

PAGE 145

,,.._,_,_,_:::.,___91{~( INTER f{O'])E) ~--PJ;J (NODE) ~" 07::J-c?J ~Tr;~?} -1 iii) \fl 1 )t;T 131

PAGE 146

132

PAGE 147

133 ~>P?'!;'iJ r-.... W:J:ilvWW"' ,,..,. 5'/. \JTl?u ( llLTEl
PAGE 148

134

PAGE 149

.q)~iJ (CONTOUR) ?fJ,;:pp ___,;J
PAGE 150

136

PAGE 151

137 ~(~/!(;/ / <1"11-_(BAR]) :~ :~ ?7~7?3H5{(13!lRB UL ES) fJ~ (QUILL) 076~ 9 f/rq


PAGE 152

138 /""'~7 a:/('6't ("REMI<:E5)

PAGE 153

139 W~:/1 "l,7JI~ ---1-lii/ 47N (3{5"'n'd/) w f f !fit "---.., -.!>7:fl ( ...J;/n) 'Vfio/ ffi/)J f t ~?A' _;J;'l/ W,;JJ "/?1)

PAGE 154

APPENDIX B Fa3T-TEST AND RETE N TIO N -TEST FOR THE BENGALI MEDIU M SCHOOLS

PAGE 155

141 'll'i.2t' ~m "J'Tnl"l~f-, ~., ~11 I C1I 'l~ffi~M ~'11111 m~ "'J~~ .!ITTT~~. ~1~1 11~61 :\~"' f,rf., ~11 fu~ C6<61 ~fw1 1 .. ,~ ......... ..... ... ........... 'lllf~ (111tl ........ ~----.. -.tfi,c ........ .... ............... Rllli1Ct'II OfTll .. ................................... ........... ....... .. ...... .. f~ f-1,.~"'111 ............ ........................................ ~d ~~51 "1Tf "fTt'f ~,, fi1fr~ ,n..........................

PAGE 156

142 ...... ..................... .. .. ... ... ..... .. ... .... ........... '1111 ,ns, ~1f1uni r.i'f q~ m c~ J iE"l>J l.!l<1: ~~>J ~, olll

PAGE 157

' f.r.rf.ef~ 6Tfif6 ~f=e,11 'llC'II }lf6 6~"-! ( Direction movement) ,._ Jftt~}l:twi ( Photosy,\lhesis ) ,.-. ~f"5 ( Phototropism ) '(. "'lf~4~~fo ( Geotropism ) 11,~'I{ ( Germination ) 'II(. ll>lltf.14~f~ ( Chemotropism ) ). ~Ifill ,,_ 9flft~ !mfJl tTI1tff > ) '() ').~~,e 11) 8 t 'II() ~0 t 143

PAGE 158

l!tl .. ru 1(. ~" 11. ,._ f111 Rm .. '(. 11. { Growth hormone ) ..ret1~ 9ft'ltfl ~'tAA -!ad l~" l!j'lf : ( Mammals ) 1(. ~ffl ( Reptiles ) 'I~"I (Birds) ~~H ( Amph i bians) (Quill) t(. ~Mnff ~-,~ ( Superior umbilicus ) 11. ~m ( lnfrrior umbilicus ) 11. tv:il ( Y4n) 144

PAGE 159

/ ~-~(Barb) ~CW-I' (Vane) ,r. ( Bub ulo) ~~C'f'Pl ( Bar bie-els ) ~{ Rcm i? ) ~~M ( Ptc!) lae) ,r Cid ... ( Rccr-ric~ ) ~cmr1 ( Racha) ~. t'tfl ( Quill ) ~~J (Shaft) ,r ( Remi tcs ) ~CN!ttJ!11. ( R ectrices ) 145

PAGE 160

146 J'IJ~ i -l~T"I ~.( U :,,,,..,,.. ___ ,()),---tt----(2)--h) (~) _. -(o)

PAGE 161

)r I 14'1 I --~. 151 "lffe.vtt~ f='l>t -I 'l(IJ-'1~ "fPj ,~~1 fl I 11>1-! 61-611 ,~C"l ~~o! '.51';i1!' ~ ~TI ~~fml

PAGE 162

148 ,...'i'211f -1'cR ,-in1-..~~ "1._l'l n 1 Ji~ffi~~ ~ll c'Pidf!JC ('P,~Q cwm 1f~Sl II~~ f.l'i'"I ~H fu~ ,~
PAGE 163

149 .. ........... .. ..... ......... ....... .. ~sf~ m fl"ftl! ,~ fl I m ~m, c,t;tc-n ~t"l11 ffl m ltf'fC:"5 -rt 9fITTi m "!l!Rlt~ ~l fiit'' "IITT1 I Ci!T'lltl -l"H f.i~ ~. ~:Sf."f ~l ~l11l1 ~
PAGE 164

ilUI .!f'!l'!:f.~ "'J.---it----(2). -(~) --(,e) 160

PAGE 165

151 I -m,(f ,!lift.! 1 ~cu Clf~ lltt mtrrfl -l"til~M n:~ I ~Ci11'1_111i (Filoplumet) ~I "'' --CtJm"fl (rachis } 9ttwfclf'I!
PAGE 166

f.rrf'ff~ rn~ '!{{'fl ffl "Ti~ fl ..!~ m1 TI I I~~:~. 6""R ( Growth movement) 1f. ~m 5ifol" ( Stimulus movement ) ,r. 5"1-1 ( Direction movement ) ~,m;~:t"..A ( Photosynthesis ) ~( Phototrop is m ) 1f. ~fs ( Geotrop i~ m ) 11( GerminaJion) ~'alll~f.<4~fs ( Cbemotropism ) l& 1 l"" ,nm~ 1IT1R f~ m .?tt~~~1 ). :t lfi'I ,. 'flfli g. t. ~) ) l.!I~~ O '), ,!!~~ ) 8 t ,. ) ~. -0 t 152

PAGE 167

)~I -.. 'IIITT 9fim'Jt'I ,r. 11. ~1 R'Jt'I ... 'ff! -.if1,PJ. ( Superior umbilicus ) ,r. ~-IPl'f~ ~irfi1'! <1c,1 ,1 : / -..m( BarL) ~Ce-I (Vane) ,r. ffl~IS!'\'l ( Barbules ) 1(. ~tt~~"!'l ( Barbi c elR )

PAGE 168

-f ~'l ( Rcmige) ~~fot ( Ptcrylal.' ) ,r. Ct~C'{'l ( RcctTim ) ~'1JW'I ( Racliie ) >.. I 9fTT1ff C~ 'ef.{fir~ lfllm-!1 9tlottfl '1' : ,f. (Quill) ~'ft1F _it ( Shaft ) ,r ( Remigcs ) ~(flltl ( Rectrices ) ... .. -r ,~ r.,~ 1~ 'lll-I,m fflll arM (~11 11;91--.1 fl I l>t -e~ ,m "!~ ~".iYl "TUl (.l~ fml

PAGE 169

APPENDIX C TREATMENT MATERIALS FOR THE ENGLISH MEDIUM SCHOOLS

PAGE 170

15S General Directions The materials in this booklet consist of a series of short passages. Study each passage carefully. Try to remember it. At the conclusion of all the passages, you will be asked to answer a set of questions to find out how much you remembered from each passage. PLEASE 00 NOT TURN BACK TO THE PASSAGE AFTER YOU HA VE FINISHED READING IT.

PAGE 171

TROPISM: GEOTROPISM A tropism is a growth movement whose direction is determined by the direction from which the stimulus strikes the p l ant. If the plant grows in the direction from which the stimulus or ig inates the tropism is referred to as positive. Growth in the oppos i te direction constitutes a negative tropism W hen the plant responds to the stimulus of gravity it is known as a geotropism. Stems are negatively geotropic while roots are positiv e l y geotropic. The value to the plant o f these responses seems quite c l ear. R oots growing downwards and/or away from light are more likely to find soil, water, and minerals. Stems growing upwards or towards light will be able to expose their leaves so that photosynthesis can occur. PHYLLOTAXY: It is rather interesting that leaves are not haphazardly placed on the stem but grow predictably at definite and regular intervals. The 15 7 points of leaf attachment on the stem is called a nod e The gap between two nodes is known as the internode. The arrangement of leaves on the stem is known as "phyllotaxy a t erm meaning leaf order. Leaves may be opposite in which two leaves occur at a node, usually on opposite sides of the stem as in maples, carnations and buckeye. Leaves may be whorled in their arrangement in which there are three or more leaves at a node, more or less equall y spaced around the node. This arrangement occurs in catalpa and in some species of lilies. Leaves are most commonly alternate, or spiral in their arrangement; that is, only one leaf occurs at a node, as in elms, lindens and apples. In the simplest type of spiral or a lte rna t e phyllotaxy, every third leaf is directly over the first o ne o f the cycle thus forming two vertical rows around the stem, as in elm and in corn and in many other grasses. Phyllotaxy deals only with the arrangement of leaves as determined by their points of or i gin on the stem. The arrangement of th e l e a v es is greatly influenced by their position, with respect to l~t. the stem, or the plant as a whole. Leaves are usually found in posi ti ons that enable them to receive a maximum amount of light so they can produce food. FEATHER: The possession of feathers is a distinguishing mark of the bird.

PAGE 172

FUNCTION: Feathers serve two major function in birds: 1. As a body covering they form an effective insulation, aiding in the mainte nance of the high body temperature wh i ch is -as characteristic of birds as of mammals. 151l 2. Bird ~tis made possible through the development of large feathers which form the wing surface and tail rudder Flight itself is not unique among anima l s --most in s ects fl y and one family of mammals, the bat, has developed t r ue fl ight However, a feather is extremely li ght and structurall y s t rong, m uch more versatile than the stre t ched skin on which a bat supports itself in flight and far more readily repaired or replaced when damaged. TYPES: There are three types of feathers which are called: 1. the down feather 2. the ccmtour feather 3. the filoplume DESCRIPTIONS: CONTOUR FRATHERS: These are the large feathers which cover the body surface and include as well the feathers o f wing and tail. The main axis of the feather is call e d the shaft. The basal portion of the shaft is known as quill, which is a hollow cylinder with its cavity more or less fille d by mesodermal tissue. At the base of the feather is a small opening into this cavity, an inf e rior umbilicu s and a superior sumbilicus is present at the distal (other end) of the quill. The remaind e r of the s ha ft is referred to as the ra c his. U nlike the quill, the rachis is solid rather than a hollow structure, its interior filled with a sponge o f air-filled cells. On either side of t he rachis lies the exposed and e x pan d ed portion of the feather, known as the~ If the vane is examined under a microscope it w ill be seen to be composed of barbs, w h ich extend out laterally from the rachis, and barbules, which in turn extend out laterally from the barbs. Thus, adjacent rows o f barbules overlap one another. The tip and underside of each barbule possesses small filaments called barb ice ls which further assist in holding overlapping barbules together. DOWN FEATHERS: Down feathers are basically similar, but simpler in build. These make an entire body covering for the chick and underlie the contour feathers over much of the bod y of the adult, forming the main insulation. A basal quill re g ion is present in these small feathers as in the contour type. Distally, howe v er, beyond the quill termina ti on, there is no rachi s but in stea d, a s pr a y o f slender simple branches. B arbs and barbules are present.

PAGE 173

159 FILOPLUMES: Filoplumes, "pinfeathers," are still simpler in appearance. These are small hair-lik e feathers with a slender shaft continuing out from the body; they may terminate in a tiny tu ft of barbs. Feathers are not, of course, uniformly distributed over the bird body Contour feathers occur in definite feather tracts, pterylae. Most prominent is a long row of large feathers along the back o f the "forearm" and "hand," the remiges, which form the wing surface; another group of large feathers, the rectrices, forms the tail. Over a number of body areas no contour feathers arise, but the arrange ment of the pterylae is such that the entire body is completely and smoothly enclosed.

PAGE 174

160 General Directions The materials in this booklet consist of a series of d i agrams with appropriate labeling. Study each diagram carefully and try to remember it. At the end of all the diagrams~ you will be asked to answer a set of questions to find out how much you remembered from each diagram. PLEASE DO N OT T U R N BACK TO TH E DIAGRA M S AFTER YOU HAVE LO OKE D AT THE M

PAGE 175

TROPISNS-THE GROWTH MOVEMENT 161

PAGE 176

. ROOT 1'LANT PLACED ON SI])E 162 PO SI IVE N&A JV E GEOTROP!S/1 CEOTR.0PJS!l 4-DAYS LATER ~EOTROPI S JVJS

PAGE 177

LEAF RRRf)NG:EMENTS ON THE STEM 163

PAGE 178

,-..;.__INTERNO.DE NO])E --~-PART OF A STEM or A FLOWERl N"G. -PL ANT 164

PAGE 179

PHYLLOTAXYTHE LEAF ARRI\N"CEMENTS ON THE 5TEM 165

PAGE 180

186 I A.ALTERNJ1TE :B.OPPOSITE C.WHOR"LED DIFFERENT TYPES OF -PHYLLOTAXY

PAGE 181

VARIOUS TYPES OF FEATHERS AND THEI'R DIFFERENT PARTS 167

PAGE 182

VA 188 SHAFT(THE NAIN AXlS) ~~--SUPERIOR UMBILI cus \,,1-------QU I LL w::.-----1NFER10R UMBILICUS CONTOUR FEATHER AND IT~ DIFFERENT --PARTS

PAGE 183

A CONTOUR FEATHER 169 ])IRC.l
PAGE 184

110 ~~BARB t -BARBU LES qUILL DOWN FEATHER ,ir--SHRFT ( SLENDER) FI LO PLUME FEA TH ER~

PAGE 185

17 1 A FL)'ING :BIRD

PAGE 186

172 Pi FL-YING. 13AT(AMAMMAL) A FLYINC BUTTERFL~ (AN INSECT)

PAGE 187

APPENDIX D FOST-TEST AND RETENTION -TEST FOR THE ENGLISH MEDIUM SCHOOLS

PAGE 188

174 Student Information Sheet Please complete the following statements. The information you will provide will be important, so please be accurate. Name: ___________________ Year in school: _______________ Sex: ____ Age: _______________ Date: --------------------School: -------------------Teacher: ------------------Record the time it takes to read the material __________ GO ON TO THE NEXT PAGE.

PAGE 189

175 Name ____________ Record the time you start this test. _______ Directions for Answering Test Items Answer the following test items as well as you can. If you cannot remember the answer, it is all right to guess. Your score will be based upon how many items you get right with NO penalty for wrong answers. GO ON TO THE NEXT PAGE.

PAGE 190

176 POST-TEST Direction: For each of the following statements select from among the four responses that one which i s the correct answer. M ark the answer of your choice. 1. A tropism is: A. Growth movement in plants B. Stimulus movement C. Direction movement D. Photosynthesis 2. The plant responds to the stimulus of gravity. Th i s is known as: A. Phototropism B. Geotropism C. Germination D. Chemotropism 3. The reason roots grow downward is that they are likely to find: (1) Soil (2) Water (3) Minerals (4) Air (5) Light A. 1 and 2 B. 1, 2 and 3 c. 4 and 5 D. 2. 3 and 5 4. The arrangement of leaves on the stem is known as: A. Cyclic B. Phyllotaxy C. Inflorescences D. Venation 5. The actual arrangement of the leaves is greatly influenced by the position with respect to: A. Water

PAGE 191

B. Gravity C. Grow th hormone D. Light 6. Presence of feathers is a distinguishing feature of: A. Mammals B. Reptiles C. Birds D. Amphibians 7. The basal portion of the shaft of the feather is known as: A. Quill B. Superior umbilicus C. Inferior umbilicus D. Vane 8. On either side of the rachis lies the exposed and expanded portion of the feather which is known as: A. l3~rb B. Vane C. Barbules D. Barbicels 9. Large feathers at the wing are arranged in a definite tract which is known as: A. Remiges B. Pterylae C. Rectrices D. Rachis 10. A group of large feathers arranged in a definite tract in the tail of a bird is known as: A. Quill B. Shaft C. Remiges D. Rectrices 17'1

PAGE 192

Directions: Write the word or phrase that correctly completes the statement in the space pz:ovided. 11. Label the numbered parts in the diagram below. 178 12. Label the following diagrams according to the leaf arrangement on the stem. l') ____ (.J) ___ C..,l ___

PAGE 193

179 13. The diagram below is an example o f __________ SOYBEAX SEEDLINGS ROOT SOIL 14. The two major functions of th e f eather in birds are ______ and -----15. There are mainly three types of feathers which are known as 16. The main axis of the feather iS called ________ 17. Filoplumes are __________ feathers. 18. _______ extend out laterally from the rachis 19. Names of the two animals which do not have wings made up of feathers. but still can fly are _____ and _______ 20. Down feathers consist of a spray of ----Record the time: _____ Close your booklet. Please wait quietly. When asked. please turn in all th e mat erials you were given. THANK YOU FOR YOUR HELP.

PAGE 194

180 Student Information Sheet Please complete the following statements. The information you will provide will be important, so please be accurate. Name: ___________________ Year in school: ----------------Sex: ____ Age: _______________ Date=-----------------'----Scbool: ---------~---------Teacher: ___________________ Record the time it takes to read the material ____________ GO ON TO THE NEXT PAGE.

PAGE 195

181 Name -------------Record the time when you start this test ______ Directions for Answering Test Items Answer the following test items as well as you can. If you cannot remember the answer, it is all righ t t o guess. Your score wili"be based upon how many items you ge t r i ght with N O penalty for wrong answers. GO ON TO THE NEXT PAGE

PAGE 196

182 RETENTION-TEST Directions: Write the word or phrase that correctly completes the statement in the space provided. 1. Label the numbered parts in the diagram below. 2. L abel the following diagrams according to the leaf arrange ment on the stem. (!) _____ (.J) ___ lJ) ___

PAGE 197

183 3. The diagram below is an example of __________ SOYBEAN SEEDLINGS ROOT 4. The two major functions of the feather in birds are ____ and _____ __ 5. There are mainly three types of feathers which are known as and -----------6. The main axis of the feather is called -------7. Filoplumes are ____________ feathers. 8. _______ extend out laterall y from the rachis. 9. Names of the two other animals which do not have wings made up or feathers, but still can fly are ______ and _____ 10. Down feathers consist of a spray of ___________

PAGE 198

184 Directions: For each of the following statements select from among the four responses that one which is the correct answer. Mark the answer of your choice. 11. A tropism is: A. Growth movement in plants B. Stimulus movement C. Direction movement D. Photosynthesis 12. The plant responds to the stimulus of gravity. This is known as: A. Phototropism B. Geotropism C. Germination D. Chemotropism 13. The reason roots grow downward is that they are likely to find: (1) Soil (2) Water (3) Minerals (4) Air (5) Light A. land 2 B. l. 2 and 3 c. 4 and 5 D. 2, 3 and 5 14. The arrangement of leaves on the stem is known as: A. Cyclic B. Phyllotaxy C. Inflorescences D. Venation 15. The actual arrangement of the leaves is greatly influenced by the position with respect to: A. Water B. Gravity C. Growth hormone D. Light

PAGE 199

16. Presence of feathers is a distinguishing feature of: A. Mammals B. Reptiles C. Birds D. Amphibians 17. The basal portion of the shaft of the feather is known as: A. Quill B. Superior umbilicus C. Inferior umbilicus D. Vane 18. On either side of the rachis lies the exposed and e x panded portion of the feather which is known as: A. Barb B. Vane C. Barbules D. Barb ice ls 19. Large feathers at the wing are arranged in a definite tract which is known as: A. Remiges B. Pterylae C. Rec t rices D. Rachis 185 20. A group of large feathers arranged in a definite tract in the ta il of a bird is known as: A. Quill B. Shaft C. Remiges D. Rectrices Record the time: ____ Close your booklet. Please wait quietly. When asked. please turn in all materials you were given. THANK YOU FOR YOUR HELP.

PAGE 200

LIST OF REFERENCES Allen. A. L. and Shannon. A.G. Concept Selection Stra t e gi es of New Guinea Students. The Journal of Ex perimen ta l Edu ca ti on. 1971. 1-4. Archer. E. J. On Verbalizations and Concepts. In M elton. A. (Ed.). Categories of Human Learn i ng. N ew York: Academ i c Press. 1964. Attneave. F. Some Informational Aspects of Visual Percept i on. Psycholog i ca l Review, 1954. 183-193. Ausubel. D. P. Educational Psychology: A Cognitive Vi ew New York: Holt Rinehart and W inston, Inc 1 9 6 8. Bourne. L. E Jr. Human Conceptual Behavior Boston: Allyn and Bacon. 1966. Broadbent. D.E. Perception and Communication. New York: Pergamon Press. 1958. ____ Information Processing in the Nervous S y stem. Sc i ence. Massachusetts: American Association for the Advancement of Science, !965, 3695, 457-462. Bronowski, J. and Bellugi, U. Language Name and Concept. Science. 1970, 148. 673. Bruner, J.S. The Act of Discovery. Harvard Educational Re vi ew. 1961. 31, 21-23. ____ and Potter. M. C. Interference in Visual Recognition. Science. 1 9 64, 144, 424-25. ------' Goodnow. J. J and Austin, G.A. A Study of Thinking. New York: John Wiley and Sons, 1956. Burdick. J. G. A Study of Cross-Section Drawings U sed as Technical illustrations in High School Science Te x tbooks. S y racuse, N Y.: Syracuse t:niversity. 1959, Abstract: Dissertation A b stracts, 20: 2707-2708. No. 7 (1960). 186

PAGE 201

187 Campbell. D. T. and Stanley, J.C. Experimental and Quasi Experimental Designs for Research. Chicago: Rand McNally and Company. 1971. Carroll. J.B. Words. Meanings and Concepts. Harvard Educational Review. 1964. 34, 178-202. Chan. A Travers. R. M. W and Van Mondfrans. A. P. The Effect of Colored Embellishments of a Visual Array on a Simultaneously Presented Audio Array. Audio Visual Communication Review. 1965. !!, 159-164. Chatterjee. S. K. The Language Problem in Indian Education. Bombay: International Book House Ltd., 1954. Ciborowski. T. and Cole. M. A Cross-Cultural Study of Conjunctive and Disjunctive Concept Learning. Child Development, 1972, 43. 774-789. Clark, C. D. Teaching Concepts in the Classroom: A Set of Teaching Prescriptions Derived from Experimental Research. Journal of Educational Psychology, 1971. 62, 253-278. Cronbach, L. J. Individual Differences in Learning Ability as a Function of Instructional Variables. Basic and Applied Research Proposal. Submitted to the U.S. Commission of Education. August. 1965. --,,.---=How Can Instruction be Adapted to Individual Differences? In Gagne. R. M. (Ed.) Learning and Individual Differences. Columbus, Ohio: Charles E. Merrill Publishing Company. 1967. __ -,--_and Snow. R. E. Individual Differences in Learning Ability as a Function of Instructional Variables. Final Report, U.S.O.E Contract No. OEC-4-6-061269-1217, School of Education. Stanford University. 1969. DasGupta. J. Language Conflict and National Development Group Politics and National Language P olicy in India. Berkeley, Los Angeles. and London: University of California Press, 1970. DeCecco. J.P. The Psychology of Learning and Instruction: Educa tional Psychology. Englewood Cliffs, N.J.: Prentice-Hall. Inc 1968.

PAGE 202

188 Detambel, M. H and Stolurow. L. M. Stimulus Sequence and Concept Learning. Journal of E xperimental P s ychology, 1 9 56, 51, 34-40. Dietze, D. The Facilitating Effect of Words on Discriminat i on and Generalization. Journal of Experimental Psychology, 1 9 55, 50, 255-260. Dwyer, F. M Jr. Adapting Visual Illustrations for Effective Learning. Harvard Educational Review, 1967, 37, 250-263 ____ The Effect of Image Size on Visual Learning. The Journal of Experimental Education, 1970, 39, 36-41. ____ .. A Guide for Improving Visualized Instruction. Learning Services, State College, Pennsylvania, 1972. Farnham-Diggory, S. Cognitive Processes in Education: A Psvchological Preparation for Teaching and Curriculum D eve l o p ment Z\ ew York, Evanston, San Francisco, London: Harper and R ow, Publishers, 1972. Freibergs. V. and Tulving, E. The Effect of Practice on l:t ilization of Information from Positive and Negative Instances in Concept Identification. Canadian Journal of Psychology. 1 9 61, 1 5, 101-106. French, J. W Ekstrom, R. B and Price, L.A. Kit of Re f erence Tests for Cognitive Factors. Princeton, New Jersey: Educational Testing Service, 1963. Fuller, F. F. Concerns of Teachers: A Developmental Conceptualization. American Educational Research Journal, 1969, ~. 207-226. Gagne, R. M. The Conditions of Learning. San Francisco: Holt, Rinehart and Winston, Inc., 1973. George, F. H. Cognition London: Methuen and Company, Ltd., 1962. Gibson, J. J. The Perception of the Visual World. Boston: Houghton Mifflin, 1950. ____ The Senses Considered as Perceptual System. Boston: Houghton Mifflin, 1966.

PAGE 203

189 Glaser, R. Concept-Learning and Concept Teaching. Learning Research and Development Center, U niversi ty o f Pittsburgh, 1968. Preprint of the chapter to appear in Gagne, R. (Ed ), Research Approaches to School-Subject Learning, to be published by F. E. Peacock Publishers, Itasca, Illinois. Green, E. J. Concept Formation: A Problem in Human Operant Conditioning. Journal of Experimental Psychology, 1955, 49, 175-180. Heidbreder, E. The Attainment of Concepts: VII. Conceptual Achievements Dur i ng Card Sorting. Journal of Psychology, 1949, 27, 3-39. Hereford, C. F. Instructional Variables Related to Educational Goals. Paper presented at the meeting of the American Educational Research Association, New York, 1971. Holliday, W. G. Critical Analysis of Pictorial Research Related to Science Education. Science Education, 1973, 57, 201-214. Hovland, C. I. A "Communication Analysis" of Concept Learning. Psychological Review, 1952, 59, 461-472. ---,-and Weiss, W. Transmission of Information Concerning Concepts Through Positive and Negative Instances. Journal of Experimental Psychology, 1953, 45, 175-182. Huttenlocher, J. Some Effects of Negative Instances on the Formation of Simple Concepts. Psychological Reports, 1 9 62, !.!_, 35-42. lndia--A Reference Annual. Publication Division, Ministry of Information and Broadcasting, Government of Ind i a; 1971-1972, 1972-1973. Jacobson, H. The Informational Capacity of the Human Ear. Science, 1950, 112, 143-144. The Informational Capacity of the Human Eye. Science, 1951, !,!!, 292-293. Jensen, A. R. Verbal Mediation and Educational Potential. Psychology in the Schools, 1966, ~. 99 109.

PAGE 204

190 Keele, S. W. Attention and Human Performance. Pacific Palisades. California: Goodyear Publishing Company, Inc., 1973. Kendler. H. H. and Kendler T. S. Vertical and Horizontal Processes in Problem Solving. Psychological Review, 1962, ~. 1-16. King, E.W. and Kerber, A. The Sociology of Early Childhood Edu cation. American Book Company, 1968. Kolers. P.A. Interlingual Word Associations. Journal of Verbal Learning and Verbal Behavior, 1963, ~. 291-300. Koran. J.J. Concepts and Concept-Formation in the Teaching of Biology. The American Biology Teacher, 1971, 33, 405-408. Koran. M. L. Varying Instructional Methods to Fit Trainee Characteris tics. Audio Visual Communication Review, 1972, 20, 135-146. ____ Snow, R. E and !\lcDonald, F. J. Teacher Aptitude aoo Observational Learning of a Teaching Skill. Journal of Edu cational Psychology, 1971. 62, 219-228. Krishana, R. G. The ABC of Backwardness. The Illustrated Weeklv of India, 1974, 95, 20-22. Laughlin, P.R. Selection Strategies in Concept Attainment as a Function of Number of Relevant Problem Attributes. Journal of Experimental Psychology, 1966, 71, 773--776. ---- and Doherty, M.A. Discussion Versus Memory in Cooperative Group Concept Attainment. Journal of Educational Psychology. 1967, 58, 123-128. --~~ and Jordan. R. M. Selection Strategies in Conjunctive, Disjunctive, Biconditional Concept Attainment. Journal of Experimental Psychology, 1967, 75, 188-193. Martorella, P.H with Jensen, R. S., Kean, J.M. and Voelker, A. M. Concept Learning Designs for Instruction. Intext Educational Publishers (Co ll ege Division of Intext): Scranton, San Francisco, Toronto, London; 1972. Mathai, S. English for Unity. The Illustrated Weekly of India, 1974. 95. 12-15.

PAGE 205

M~Donald. F. J. Educational P s ychology Wadsworth Publishing Company, Belmont, California, 1 9 65. 191 ---,-and Koran, M. L. The Effect of Individual D i fferences on Observational Learning in the Acquisitfon of a T e aching Skill, Final Report, Washington, D.C U.S.O.E March, 1969. Mechner, F. Science Education and Behavioral Technolog y In Glaser, R. (Ed.) Teaching !\ lach i nes and Programm e d L e arning II: Data and Directions. W ashington, D.C.: N E .A 1 9 65. Meeker, M. N. The Structure of Intellect: Its Interpre tati on and Uses. Columbus, Ohio: Charles E. M errill Publish i ng Company. A Bell and Howell Company, 1969. Merrill, M. D. Teachers: Technologists or Technic i ans. The Journal of Teacher Education, 1 9 68, ~. 325-330 Moore, D. M. and Sasse, E. B. Effect of Size and Type of Still Projected Pictures on Immediate Recall of Content. Audio Visual Communication Re v iew, 1 9 71, ~. 437-450 Neisser, U. Cognitive Psychology. New York: Appleton, 1967. Peal, E. and Lambert, W. E. The Relation of Bilingualism to Intelligence. Psychological l\lonographs, Whole No: 546, 1 9 62. Pella, M. O. Concept Learning in Science. Science Teacher, 1966, 33, 31-34. Platt, M. M. Concepts and the Curriculum. Social Education, 1963, 27, 21. Sapir, E. The Grammarian and his Language. In Mead, M. and Benzel, R. L. (Eds.). The Golden Age of American Anthropology. New York: Braziller, 1 9 60. Taba, H. Techniques of In-Service Training. Social Education, 1965, 29, 465. ____ Teaching Strategies and Cognitive Functioning in Elementary School Children. Cooperat i ve Research Project N o 2404 (Washington, D. C.: U.S. Office of Education, 1966). Travers, R.M.W., McCormick, M C., Van Mondfr ans A.P and Williams, F E. Research and Theory Rel a ted t o Au d iovi s ual Information Transmi s sion. U tah: Bureau of Educ atio n al Research, Universit :,r of U tah. 1964.

PAGE 206

192 Vaidya, N. The Impact Science Teaching. Oxford and IBM Publishing Company, New Delhi, Bo.(llbay. Calcutta, 1971. Vernon, M. D The Visual Presentation of Factual Data. British Journal or Psychology, 1950, 20, 174-185. ____ Learning and Understanding. Quarterly Journal of Experi mental Psychology. 1951, l 19-23. ____ The Value or Pictorial Illustrations. British Journal or Educational Psychology, 1953, 23, 171-179. ____ .. The Instruction of Children by Pictorial Illustrations. British Journal of Educational Psvchology. 1954, 24, 171-179. Viaud, G. Intelligence: Its Evolution and Form. New York: Harper and Row, Publishers. 1960. Voelker, A. M. Concept Teaching and Nature of the Scientific Enterprise. School Science and :\Yathematics. 1969, 69, 3-8. Vygotsky, L. S. Thought and Language. John Wiley and Sons, Inc. and MIT Press. New York, 1962. Weisberg. S. The Use or Visual Advance Organizers for Learning Earth Science Concepts. Journal of Research in Science Teaching, 1970, '!.. .. 161-165.

PAGE 207

BICXiRAPHICAL SKETCH Protima Roy was born on July 22, 1944, lived with her parents, grandmother and other members of her family in Calcutta, India. She attended schools in Calcutta. Upon her graduation from high school, she went to Presidency College under the University of Calcutta. She also attended music school in Calcutta at the same time. At Presidency College her major was zoology with two minors. botany and physiology. In the year of 1965 she received her bachelor of science degree with honors in zoology. She earned her master's in the year of 1968 in zoology and comparative anatomy. During her master's work, she became interested in doing research in cell biology. She conducted her research on tissue transplantation under the S1J.pervtsion of Professor Sivatosh Mookerjee, Chairman of the Department of Zoology. Presidency College and prepared her master's dissertation in that subject area. She was engaged in more advanced research in cell biology until she came to the United States of America with her husband in January. 1969. She started doing research on diabetics with Professor of Bio chemistry. Jorge L. Padron. Chairman of the Department of Chemistry, Drury College, Springfield, '.\Iissouri. During her stay at Drury College, she took advanced courses in physics, chemistry, mathematics, and elementary courses in German language. 193

PAGE 208

19f She joined the University of Florida in the fall of 1971. She took various courses in different disciplines of sciences. She started her doctorate in the fall of 1972 in science education with Dr. John J. Koran. Jr. She has presented three papers in different professional meetings (excluding her dissertation). Her working and research activities include: research graduate in the cell-biology laboratory. research assistant at biochemistry laboratory. producing behavioral objectives in the biological sciences for the State of Florida Project, working in competency based dental education and science teacher education in the Dental Medical School. graduate research assistant in the college of education. conducting microteaching laboratory.

PAGE 209

I certify that I have read th is study and that in my opinion it conforms to acceptable st anda rds o f scholarly presentation and is fully adequate. in scope a n d quality as a dissertation for the degree of Doctor of Philosophy. Kora fe sor of I certify that I have re a d this s dy and that in m y opinion it conforms to acceptable sta nca:rds o f scl'folarly presenta fi on and is fully adequate. in scope and ~i ty. as a dissertation for the degree of Doctor of Philosophy. Professor of Education I certify that I have read this study and that in my opinion it conforms to acceptable stan da.--ds of scholarly presentation and is fully adequate. in scope and quali ty. as a dissertation for the degree of Doctor of Philosophy. = Associate Professor of Education I certify that I have re a d this study and that in my opinion it conforms to acceptable stan dards of scholarly presentation and is fully adequate, in scope and q uality as a dissertation for the degree of Doctor of Philosophy. J? (1,,.dz/ I) es D. Casteel ~i=e Pro f essor of Education J

PAGE 210

This dissertation was submitted to the Graduate Faculty of the College of Education and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December, 1974 13 J.Ji, ducat ion