7..
....
.

vx. :91;

. r. a . . I ‘ . ,. . V ‘ . . ,
._... F... . . , . , . . V T. 7...... ’

 

 

.;...‘1.a»

. ‘fiw V;.. w-..—

 

: éw. ‘
s.» Wavy

.A.

 

 

 

'\.

-\Q. ~

3 1293 10015 9247
_~__fi_ __ r f.

P ‘ .
n3 _ ,
. '\
' er".
I - wt
1, ’4", av )1.
4

b

_ .DING I... ~

 

- ".1" an: H"

D
’ 1
‘ ?
J—v .‘0--
. ,.. x,

 

 

7 I .1 V I I g
‘3
;_
Z

 

 

In? --*" 1' -
‘33?

-:‘..

_-| N.
':.

 

   

ABSTRACT

THE EFFECT OF A PROGRAMMED COURSE OF INSTRUCTION
ON THE DEVELOPMENT OF INFORMATION-PROCESSING
COMPETENCE AND DECISION-MAKING STYLES

BY

Nancy Garrison Harries

The purpose of this study was to examine the effects
of a programmed course of instruction on information-
processing complexity and decision-making style, in order to
determine the implications for organization of information
in gradations of conceptual complexity as a training envi-
ronment for complex decision making. The specific research
objective was to identify the relationship between the exper-
ience of working through a programmed sequence of textile
instruction and (a) level of complexity used in information
processing, as defined by Schroder, et al. (1967); and
(b) decision-rule complexity and (c) time orientation,
defined as decision style elements in the.model conceptual-
ized by Bustrillos (1963).

The conceptual framework of this study was based on
EEZE individuals typically process information and make deci-
sions, rather than on the content of the information or
decisions. A distinction was made between what a person

thinks and how he thinks. The structure of human decision

 

Nancy Garrison Harries

making was examined in regard to the number and connected—
ness of integrating rules used for organizing information
for the purpose of decision making.

The three criterion variables——information-processing
complexity, decision-rule complexity, and decision time
orientation--were examined in relation to structure of
thinking. The variables were measured before and after
subjects worked through the programmed sequence of textile
instruction.

The population selected for the study consisted of
150 Michigan State University undergraduate students enrolled
in an introductory textile class (HED 171), which was taught
using the programmed text treatment designated for the study.
The students were divided into two groups to test for any
effects of the pretest on posttest results. Group I, the
experimental sample, consisted of two-thirds of the class;
these subjects received measures designed to test the vari-
ables at the beginning and conclusion of the textile class.
Group II was composed of the remaining one-third of the
class, which served as a placebo sample that received only
posttest measures of the variables. Personality measures
unrelated to the criterion variables were given to Group II,
instead of the regular pretest measures. Out of the experi—
mental group, 50 subjects were randomly selected for purposes
(Df data analysis, and 25 subjects were randomly selected from
the placebo group. No sensitizing effects were found from

lyretest to posttest results; thus, the pretest and posttest

Nancy Garrison Harries

measures of the 50 students in the experimental sample were

used for hypothesis

testing in the data analysis.

The first three hypotheses tested predicted that

there would be gains in the three criterion variables after

subjects worked through the programmed treatment:

Hypothesis I:

Hypothesis II:

Hypothesis III:

The level of complexity of textile infor—
mation processing will be increased after
working through the textile program.

After the experience of working through
the textile program, the decision rule of
decision—making style will be more
complex.

After the experience of working through
the textile program, time reference of
decision-making style will be more
future oriented.

The remaining four hypotheses predicted relationships

among the variables:

Hypothesis IV:

Hypothesis V:

Hypothesis VI:

Hypothesis VII:

Those who Operate at higher levels of
conceptual complexity are likely to use
more complex decision rules.

An increase in complexity of information
processing over time will be accompanied
by an increase in complexity of decision
rule.

Those who Operate at higher levels of
conceptual complexity are likely to be
more future oriented.

An increase in complexity of information
processing over time will be accompanied
by an increasing orientation toward
future time.

A multivariate analysis of gain score variance indi-

catfifli that there were over-all significant gains from pretests

to Exasttests on the measured variables. An examination of

the linivariate F tests for each variable indicated that the

Nancy Garrison Harries

gain scores for information—processing complexity and time
orientation contributed to the over—all multivariate signif—
icance level (p<.0026), while the gain score for rule com—
plexity did not contribute to significant gains.

No significant correlations were found among pretest
and posttest scores of information processing, time refer-
ence, or decision rule; nor among gains scores of these
variables.

From these findings, there is evidence that the
categorical organization of information in gradations of
increasing complexity contributes to increased levels of
information—processing complexity and orientation toward
future time. These results have implications for the design
of organizational learning environments which can help indi—

'viduals process information more efficiently for complex

decision making .

‘ THE EFFECT OF A PROGRAMMED COURSE OF INSTRUCTION
1‘ ‘ ON THE DEVELOPMENT OF INFORMATION—PROCESSING

COMPETENCE AND DECISION-MAKING STYLES
BY

Nancy Garrison Harries

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Family Ecology

1972

Dedicated to the
HED l 7 l STUDENTS
whose cooperation in taking numerous

measures made the study possible.

ii

ACKNOWLEDGMENTS

To Thomas Harries, friend, colleague, and life com—
panion, who nurtures and makes substantial contributions to
my career, including introducing me to information—processing
theory——a basis for this study; who provided intellectual
stimulation and shared in the conceptual development and
frustrations of this research as he simultaneously completed
his own dissertation; and who most importantly demonstrates
understanding, encouragement, and support of all of my
activities.

To Margaret Jacobson Bubolz, Frances Magrabi, Beatrice
Paolucci, and Everett Rogers, my committee members, for their
cooperative input and useful critique of this research. In
addition, for their special contributions as major professors:
To Dr. Magrabi, who initially chaired the research during
conceptual formulation and development prior to leaving
Michigan State University and continued to provide major
guidance and consultation throughout the study; and to
Dr. Paolucci, who later became major professor and provided
the encouragement and impetus for expediting the research
through to completion.

To the organizations that contributed financial support
toward my doctoral studies: To the American Home Economics
Association for awarding the Effie I. Raitt Fellowship for
1971-72, which made possible reduction of teaching responsi-
bilities so as to complete the dissertation in a shorter
time period; and to the College of Human Ecology for awarding
the Thelma Porter Fellowship the second half of the 1970—71
school year, which permitted full—time pursuit of course work

for a term.

To all of my colleagues and friends who supported
this endeavor; in particular:

To Marlene Wamhoff for her hours of time in supervision
of the data collection, with the aid of Sara Butler and
Elizabeth Dolan.

To Jo Lynn Cunningham and Verda Scheifley, research and
statistical consultants with the Institute for Family and
Child Sciences, for their willingness to go the "extra mile"
in offering time and advice despite their own demanding
schedules.

To Cathe Axinn and Stella Garrison for their long hours
and dedication to coding the measures.

To Mary Gephart Donnell and Robert Rice for their adminis—
trative support and visible c00peration in making possible
pursuit of my doctorate in conjunction with a professional
faculty role.

To Sue Cooley, friend as well as expert typist and editor,
for her ability to meet deadlines under pressure, cheerfully
and competently.

To Virginia Beauchamp for her ever-present encouragement
and willingness to listen and offer assistance on a moment's
notice.

To Stella and Lester Garrison, both college educators
themselves, for creating a family environment which set the
stage for a lifetime of ongoing growth and development; and

flor their tangible support throughout my doctoral program.

iv

LIST OF
LIST OF

LIST OF

Chapter

I.

II.

III.

IV.

TABLE OF CONTENTS

TABLES. . . . . . . . . . .
FIGURES . . . . . . . . . .

APPENDICES. . . . . . . . .

INTRODUCTION. . . . . . . .

Statement of the Problem.
Conceptual Framework. . .
Explication of Conceptual

Framework:

The Textile Instructional System.

Summary . . . . . . . . .

REVIEW OF LITERATURE. . . .

Effect of Cognitive Structure

Information—Processing Complexity
Related to Elements of Decision

Summary . . . . . . . . .
METHODOLOGY . . . . . . . .

Treatment:
Assumptions . . . . . . .
Research Design . . . .
Selection of the Sample .
Description of the Sample

Instrumentation: Development
Data Collection . . . . .
Coding Procedures . . . .
Data Analysis . . . . . .
RESULTS AND DISCUSSION. . .

0

Test for Effect of Pretest.

Gain Scores of Information Processing,
Time Orientation, Decision Rule

The Textile System.

0

mr4

16
20

22

24

43
58

60

 

Chapter Page

Correlations Between Information—
Processing Complexity and Decision—
Rule Complexity . . . . . . . . . . . . . . 109

Correlations Between Information—
Processing Complexity and Decision

Time Orientation. . . . . . . . . . . . . . 112

Summary of Results. . . . . . . . . . . . . . 116

V. CONCLUSIONS AND IMPLICATIONS. . . . . . . . . . 118
Effects of the Programmed

Cognitive Structure . . . . . . . . 118

Relationships Among the Variables . . . . . . 120

Limitations of the Study. . . . . . . . . . 122

Implications of the Study . . . . . . . . . . 123

Summary . . . . . . . . . . . . . 132

BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . . . 134

APPENDICES. . . . . . . . . . . . . . . . . . . . . . . 144

vi

ll.
12.
l3.

14.

15.

LIST OF TABLES

Chi Square Test of Nominal Descriptive Data. .

Multivariate Analysis of Continuous

Descriptive Data . . . . . . . . . . . . . .
Age and Year in College. . . . . . . . . . . .
Marital Status and Class in School . . . . .

College Major and Average GPA by Major
Average GPA by Class in School . . . . . . . .
Student Occupation . . . . . . . . . . . . . .
Student Credit—Hour Load . . . . . . . . .
Parent's Occupation. . . . . . . . . . . .
Size of Sample's High School Graduating Class.
Summary of Statistical Treatment of Data . .
Multivariate Analysis of Posttest Scores . . .
Multivariate Analysis of Gain Scores . . . . .

Cell Means and Standard Deviations of
Gain Scores. . . . . . . . . . . . . . . . .

Cell Means and Standard Deviations of
Pretest and Posttest Scores. . . . . . . . .

vii

Page

69

69
70
71
72
72
73
74
75
75
94
97

98

99

100

Figure

1.

LIST OF FIGURES

Hypothesized Relationships . . . . . . .

Relationship Between Environment and
Behavioral Complexity. . . . . . . . .

The TOTE Unit. . . . . . . . . . . . . .
Design for Data Collection . . . . . . .

Scoring Schemata for Criterion Variables

viii

Page

22

28
40
66

101

LIST OF APPENDICES

 

Appendix Page
A. SAMPLE PRETEST AND POSTTEST . . . . . . . . . . 145
B. PARAGRAPH COMPLETION TEST (PCT) SCORING . . . . 157
C. CODING MANUAL: DECISION TIME ORIENTATION . . . 161
D. CODING MANUAL: DECISION RULE . . . . . . . . . 166

 

ix

CHAPTER I

INTRODUCTION

Statement of the Problem

 

Information can be organized in gradations of concep-
tual complexity to train individuals for complex decision
making. The benefits of a simple to increasingly complex
categorical organization of information are examined in this
study, as to effect of the organization on level of complex-
ity in processing information and decision making.

The cognitive structure used to organize the infor-
mation was a programmed sequence of textile instruction,
which had been developed and tested during the past three
years in undergraduate courses at Michigan State University.

Specifically, the research objective was to identify the

 

effect of the experience of working through the programmed
cognitive structure upon: (a) the student's level of complex-
ity in processing information, and (b) the student's decision—
making style, which was defined through the elements —-
decision rule and time orientation. Decision rule, time
orientation, and level of information-processing complexity
were measured before and after students worked through the

sequence of programmed textile instruction.

Research in this area is needed if learning environ-
ments are to be developed which contribute to an individual's
ability to process increased amounts of information at higher
levels of conceptual complexity. Competence in processing
information and making decisions in complex environments is
becoming essential in societies where masses of information
are bombarding peOple daily.

Although high or low levels of complexity of informa-
tion processing are inherently neither good nor bad, high—
level conceptual functioning is more adaptive because it
allows the individual to c0pe with greater environmental com-
plexity, to generate and utilize conflict in his thinking,
and to withstand more effectively the negative effects of
stress (Schroder, 1971). In complex environments, higher
levels of complexity allow for more efficient processing of
information in decision making than do lower levels of com-
plexity. At lower levels of conceptual complexity the indi-
vidual typically uses very little information in forming
concepts, and has difficulty in developing alternative con-

ceptions of events.

Conceptual Framework

 

The conceptual framework for this study draws upon
three areas of research: information-processing theory,
decision theory, and related instructional theories.
InfOrmation-processing theory is important to the study

because it emphasizes the structure rather than the content

 

of human thinking. Decision theory provides a basis for
examining the process of decision making, which relates to
how choices are made in the context of observable behavior.
Last, the instructional theories of Piaget, Bruner, and
Skinner establish a rationale for use of the textile pro-
grammed instructional system as the cognitive structure.
Each area is discussed in turn, as to how it contributes

to the conceptual formulation of the current study.

Information Processing

 

The information—processing approach to cognition and
personality, which focuses on the structure of human thought
rather than on its content (Schroder, Driver, and Streufert,
1967; Schroder, 1971), was the conceptual basis for this
study. Conceptual complexity theory, or conceptual systems
theory, as this approach has been called, focuses on how
individuals typically process information rather than on the
outcomes of such processing. The emphasis is on the number
and connectedness of conceptual or integrating rules used for
organizing information in thinking, valuing, judging, and
decision making. A sharp distinction is made between what a
person thinks and h9w_he thinks. Thus, two classes of infor—
mation are basic to this information—processing model:

(1) content variables, which provide information about acqui-

 

sition, direction, and magnitude of what a person thinks; and

(2) structural variables, which provide a means for measuring

 

the way a person combines information perceived from the

outside world, as well as internally generated information.
Information processing in a given situation involves the
perception and the subsequent organization of various kinds
of information.

The approach allows for the systematic relation of
individual differences in information-processing ability to
a wide variety of personality and task parameters. The com-
plexity of the conceptual processes used in thinking and
making judgments about a given set of stimuli is defined in

terms of two sets of variables: (1) the degree of differen—

 

tiation, and (2) the level of integrative complexity. Differ—

 

entiation refers to the number of items of information used

in a given information-processing situation, while integrative
complexity is operationally defined as the number of differ-
ent concepts which the individual forms on the basis of
derived informational elements.

Integrative Complexity.——A central notion to the model

 

is that different levels of integrative complexity indicate
different ways of processing information. The level of infor—

mation processing is viewed as ranging from a low integration

 

index or concrete processing structure, to a high integration

 

index or abstract structure. The level of complexity of
information processing is measured in reference to a given

range of stimuli, based upon how many different dimensions

 

(a dimension is defined as a unique arrangement of stimuli)
are perceived related to the stimuli, and to what extent

rules are used in integrating the dimensions.

The number of dimensions is not necessarily related
to the integrative complexity of the information-processing
structures; but the greater the number of dimensions, the
more likely is the development of integratively complex
rules. Research indicates that the relationship between the
number of dimensions of information presented and complexity
of information processing is curvilinear (Streufert and
Schroder, 1965). A low integration index is similar to a
hierarchical form of integration, in which fixed rules are
used; and the relationship between the dimensions (however
many there may be) is relatively static. High—integration—
index structures have more connections between rules, and
there are more schemata available for forming new hierarchies,
which are generated as alternate rule structures.

Behavioral Characteristics.—-Examples of behavioral

 

patterns generated by low integration levels of information
processing include: categorical, black—white thinking;
minimizing of conflict; habitual avoidance of ambiguity; a
high degree of conformity; anchoring of behavior in external
conditions; standardization of judgments in a novel situa—
tion; and increased susceptibility to the anxiety—producing
effects of such stress variables as environmental complexity,
frustration, or rapid change (Schroder, et al., 1967, pp. 43-
106; Schroder, 1971). At higher levels of complexity, how—
ever, the individual moves away from this simplistic and
rigid type of functioning. He uses more information in

forming concepts and also develops the capacity to

conceptualize events in alternative ways, to think in multi-
conceptual terms. Some of the behavioral implications asso—
ciated with high—level integration include the utilization
of alternative interactive processes and the ability to cope
with situation change over time.

Yet, individuals operating at low levels of concep—
tual functioning can, in certain situations, make decisions
more quickly and efficiently than those who function at such
high levels of complexity that many alternatives are perceived
and decision closure is difficult. Generally, however,
an ability to operate at high levels of conceptual function—
ing has been recognized as an important stage in personality
development (Piaget, 1932; Kohlberg, 1964). Thus, an objec-
tive of instruction is to increase the level of complexity /
of information processing.

Successful performance in most complex task environ-
ments requires: (a) sufficient performance of skills and
knowledge (content), (b) a near—optimum level of interest or
motivation, (c) adequate competence in interpersonal rela-
tions (since most complex task environments require group
activity for decision-making purposes), and (d) the capacity
to engage in complex information processing (Schroder, et al.,
1967, p. 13). Research efforts should focus on developing
training environments that foster these competencies; and the
focus of the present research is on the development of (a)

and (d) above.

Decision-Making Style

 

To determine how decision-making style relates to
information-processing levels, an adaptation of the Bustrillos
model (1963) was selected from the decision literature as the
conceptual framework for this study. Bustrillos viewed deci-
sion making as a conflict—resolution process, composed of
three components: movement, relationship, and discrimina-
tion. Bustrillos recognized that these components were not
directly observable, but she proposed that they be identi—
fied through study and analysis of verbalized behavior.

Bustrillos defined decision-making style as the man-
ner in which individuals pattern decision behavior. The
pattern was seen as the decision maker's creative product of
attempts to recognize demands from the environment and blend
them with his own needs and desires. Style might vary from
situation to situation, but patterns of behavior were devel-
oped which were distinctly one's own (Bustrillos, 1963, p. 5).

Three behavioral or style elements were conceptual-
ized in the Bustrillos model as indicative of an individual's
decision-making process. These interdependent elements were
called mode, time reference, and decision-making rule; and
each style element was described as manifesting all three
process components.

Mode, which was defined as the distinctive way one
develops ideas in a decision situation, was not included for
study as a decision—style element; the mode component was

explored through the information—processing approach developed

by Schroder, et al. (1967). According to Bustrillos, mode

was the expressive component of decision-making style; it
represented how ideas were developed, analyzed, classified,
and then related to the decision-making problem. The central
feature of mode did not focus on the content of ideas, but

on the structure, which is the thesis of the Schroder model.
The latter two elements of style, time reference and decision-
making rule, were selected from the Bustrillos model for this
study.

Decision Time Orientation.——Time reference was

 

selected as a variable to relate to information processing,
since no decision is independent of time. One must be able
to perceive events in a time relationship, as time gives

meaning and continuity to events. Past, present, and future

 

time orientations were specified, with future time references
considered most relevant to conceptual complexity of infor—
mation processing. Decisions made with thought to future
orientation involve increasingly complex levels of integra-
tion and abstract reasoning. To make projections from present
or past actions to future consequences, additional dimensions
and rules would need to be considered.

Decision Rule.--The other element of decision style

 

selected for study was the decision rule, which is the method
by which alternative courses of action are evaluated. Rule,
thus, comes after a set of alternatives has been perceived.
By using a decision—making rule, an individual is able to

differentiate alternatives and then arrive at a discriminatory

point or decision. This element of style is comparable to
the evaluation and choice phases designated in the normative,
sequential model of decision process. Different labels have
been used to describe various decision rules; a discussion
of rules considered for this study follows, concluding with
the classification decided upon for the rule variable.

Three decision rules were identified by Bustrillos:
preference ranking, objective elimination, and immediate

closure. Preference ranking involves the ordinal ranking of

 

perceived alternatives. Alternatives are evaluated and placed
in order from best to worst, according to a subjectively

defined criterion. This rule is essentially a lexicographic

 

ordering approach, in which the decision maker ranks or
evaluates alternatives with respect to values placed on an
alternative's most important attribute. If more than one
alternative exhibits the same value for the most important
attribute, the tie is broken by looking at the second most
important attribute, and so on until there are either no more
ties or no more attributes. All perceived alternatives are
thus ranked according to the decision maker's preference along
several dimensions. The lexicographic ordering label was
selected for this study as being more description than pref—
erence ranking.

Bustrillos defined objective elimination as a situa-

 

tion in which no one best alternative is consistently chosen;
the "best" depends on the conditions of the environment.

Alternatives suggested as desirable are qualified, with

10

phrases such as: "under these conditions is best";
"but if then ." Unless forced by the situation,
closure is not readily made; the decision maker does not
identify personally with the possible alternatives, but rather
remains detached. An objective factual approach is used, with
little thought given to subjective value judgments. Objec—
tive elimination involves a recognition of the objective
demands of the situation and analysis of the alternatives
which are perceived to apply. However, if more than one
alternative is feasible, based on the situation, the decision
maker may have difficulty in making a final choice.

Simon's satisficing model (1957) is similar to objec-

 

tive elimination, in that it suggests that the decision maker
will choose the first alternative which exceeds a set of
minimal criteria. Satisficing was chosen instead of objec-
tive elimination as a rule classification in the current
study. Satisficing has a widespread and precise meaning and
is used more frequently in the decision literature. Alterna-
tives are evaluated according to the satisficing model by
comparing their attribute values with a set of goals or
standards for these attributes; and any alternative, usually
the first one discovered, which meets or exceeds all of these
standards is chosen. Following a satisficing model, an indi—
vidual uses fewer dimensions and perhaps operates at a lower
level of complexity in processing information than with the
preference or lexicographic ordering model. In the latter

model, all perceived alternatives are ranked according to

11

preference along several dimensions, which requires integra-
tion and differentiation of information.

In the last rule specified by Bustrillos, immediate

 

closure, only one action is focused upon. The one perceived
alternative is immediately selected without explicitly going
through ranking or elimination. No other alternatives are
mentioned at the conscious level, and the decision process
is quickly executed, with little analysis occurring before
choice. This rule would no doubt be associated with low
levels of conceptual functioning.

In this research, the following rules were evaluated
according to complexity of information processing required:

(1) immediate closure, (2) satisficing, (3) lexicographic

 

 

 

ordering, and (4) a combination of satisficing and lexico—
graphic ordering, called SATISLEX by Russ (1971a). The use
of immediate closure was considered the simplest, least com-
plex decision rule, with each rule increasing in complexity
from (1) to (4).

The use of lexicographic ordering was, therefore,
considered a more complex decision rule than the use of satis-
ficing alone. The use of both rules, i.e., when alternatives
which fail to meet certain goals or standards are eliminated
from further consideration and those that remain are ranked
lexicographically, was considered more complex than the use

of either rule alone.

12

Related Instructional Theories
The development of the programmed cognitive structure,
used as the vehicle for this study, is grounded in several

learning theories which are related to information processing.

 

Piaget's Development Theory.-—Piaget is recognized as
a leading investigator of concept formation in children;
his approach is a genetic one. He attempted to distin-
guish levels or stages of development in the evolution of
thought and to show how each stage reveals a progressive
sequence from simpler to more complex levels of organization
(Piaget, 1950, 1953). He believed that the individual's
neural structure matures at a given rate, and that at cer-
tain stages of maturity there are "wired—in" responses,
which naturally occur in an information—processing context.

Given an individual's history of processing informa-
tion within the context of his neural structure, adaptation
continually occurs through a series of phases from birth to
maturity. The individual constructs his "cognitive map"
through interaction with experiences. Piaget maintained that
thought activities may be analyzed in terms of groups or
systems of such operations, which appear as actions (Thompson,
1959, p. 91).

Bruner began where Piaget left Off--with mature
adults who have mastered all the basic Operations and group-
ings necessary for thought. Therefore, Bruner's work has
particular bearing on the development of cognitive structures

for learning environments at the college level.

13

Bruner‘s Concept Attainment Theory.--Bruner and his
associates have been concerned with research in the area of
cognitive growth——the way in which human beings learn to
represent their experience and the way in which they are
affected by their cultural environment (Bruner, Goodnow, and
Austin, 1956). However, Bruner researched a narrower area
than Piaget; Bruner concentrated on categorizing in concep—
tual thinking, while Piaget covered the entire field of
conceptual activity from forming of the notion of concrete
permanent Objects beyond the range of the subject's percep-
tual field to the concepts of space, time, and number (Piaget,
1952, 1953).

Bruner, et a1. succeeded in showing the type of

behavior which actually goes on in conceptual activity

(Thompson, 1959, p. 67). They defined a concept as a cate-
gory, and distinguished two categories: (1) identity classes,

in which a variety of different stimuli are identified as the
same object or event; and (2) equivalence classes, in which
different items are treated as equivalent for some purpose

or other (Bruner, et al., 1956, pp. 2—4). Their work further
dealt with the meapg peOple use for categorizing, which

Bruner called strategies. Bruner's strategies are somewhat

 

comparable to Piaget's concept of groupings, which he defines
as systems fulfilling certain conditions of reversibility,
composition, etc. (Piaget, 1950).

Bruner defined strategies as purposive patterns of

behavior used in moving toward some goal. Information was

14

viewed as having to be acquired, retained, and used in a
series of moves and countermoves with the goal of solving

the problem of defining attributes of the concept at issue.
Four ideal heuristic strategies were worked out from the
experiments, which consisted of the most logical and econom-
ical ways of tackling the specific problem of the experiments,
namely: successive scanning, simultaneous scanning, con—
servative focusing, and focus gambling (Bruner, 1956, pp. 83-
90).

Put in a decision frame of reference, the steps
involved in attaining a concept are successive decisions,
earlier ones of which affect the degrees of freedom possible
for later decisions. Decision strategies were not seen as
fixed; rather, they altered with the nature of the concept
being sought, with pressures of the situation, perceived con—
sequences, and so forth (Bruner, et al., 1956, pp. 54—55).

Bruner presumed that strategies are learned, and in
later writings (1964, 1966) he took the View that human
development is shaped by a series of advances in the use of
the mind. He stated that growth depends upon the mastery of
techniques and cannot be understood without reference to such
mastery. He saw the techniques as mainly skills transmitted
with varying efficiency and success by the culture, e.g.,
language. Three systems of processing information are iden—
tified, by which human beings construct models of their world:
through action, through imagery, and through language. These

systems allow human beings to represent in a manageable way

15

the recurrent features of the complex environments surround—
ing them (Bruner, 1964, p. 1).
The manner in which these systems of information

processing are integrated is another concern expressed by

 

Bruner, which supports Schroder's view of integrative con-
ceptual complexity. The means through which acts are organ—
ized into higher—order ensembles, making possible the use of
larger and larger units of information for the solution of
complex problems, is an important aspect of maturity (Bruner,
1964, p. 2).

Skinner's S-R Theory.——The development of programmed

 

instruction as a prominent instructional tool has its theoret-
ical roots in that portion of the behaviorist psychology
school of thought developed by B. F. Skinner (Skinner, 1953).
Skinner conceived of the human being as a "black box" in
which external stimuli are input and observable behavior is
output. Skinner was not concerned with hypothesizing as to
what is occurring within the black box. Rather, one only
need Observe input-output relationships, in order to make
probability statements about the likelihood of the occur-
rence Of a specified behavior in the presence of a given
stimulus array.

A consequence of inputs can therefore be structured
through a program of instruction so as to increase the pre-
dictability of behavior occurrence after passing through the
stimulus array called the "program." The development of

programmed instruction, however, has almost exclusively been

16

used for content purposes. Programs are designed to make
content output (i.e., correct use Of terms, naming, and
identification of concepts) rather than to develop information

processing ability. (See T. Harries, 1972, Chapter I.) There

 

is little question that programmed instruction is an effec-
tive tool for developing content—oriented behaviors in stu-
dents. However, the questions at issue in this study are

concerned with what effect the structure of the program itself

 

(rather than simply its content) contributes to the behaviors
related to information processing, decision rule, and time
orientation.

Explication of Conceptual Framework:
The Textile Instructional System

 

 

Explication means making visible the elements of the
conceptual structure in observable behavior. The textile
instructional system was designed to effect change on student
information—processing complexity and decision—making beha—
viors, in a manner that behavior could be Observed and
analyzed by appropriate tests. (The design and development
of the instructional system are discussed in Chapter III,
Methodology.)

A programmed text was selected as the cognitive struc—
ture for information transmission, rather than a conventional
text, because Of the difference in nature of structures.
Programmed materials are highly structured, with fixed degrees
Of freedom set by the program. The student is led down a

prescribed path, with observable behavioral outcomes traceable

17

to the program. Further, the textile program is revised
as it is developed, based upon student performance and
feedback, so that its results are predictable.

Convergent thinking, rather than divergent thinking,
is believed to be fostered by programmed instruction, causing
the student to focus on the point at issue. Uncertainty
is reduced, in that the student knows what information he is
to master. The concreteness Of the program structure is more
visible than the structure of the typical textbook. Differ-
ential student interaction with the program can be looked at
in relation to the information-processing level of the stu-
dent. The effect Of differences in individual information—
processing complexity can be related to the program structure.

By defining the constraints within which the student
must operate, alternatives for decision making are initially
narrowly prescribed. Basic information of low complexity is
presented first, and complexity is increased as the student‘K
is able to integrate additional information. Information
is organized in gradations of increasing complexity, in which
experiential referents are initially paired with textile
data. As student mastery of textile concepts increases,
experimental referents are phased out and replaced by an
increasing hierarchy of textile symbols and referents. Such
a categorical organization of information is intended to
provide a framework for later use, allowing new inputs of
information to be processed and decisions made, within a much

broader framework than the program permits.

18

Bruner's work, indicating that classification reduces
the complexity of the environment and makes information easier
to deal with, supports the advantages of a structured envi—
ronment, expecially for low-level conceptual processors
(Bruner, et al., 1956). By providing all students with a
prescribed structure, which leads them down the path toward
greater complexity, increased predictability of successful
attainment of the information by a majority of students is
possible.

Using a conventional text, which is less structured
than a programmed text, high—level conceptual processors
would likely be able to integrate and master the information;
but there is no planned means for seeing that concepts are
systematically attained. Particularly, low—level conceptual
processors are not as likely to succeed following unstructured
environments, because an environment with fewer degrees of
freedom (thus less complexity and low uncertainty) is required
to Optimize their level of information processing (Schroder,
1971, p. 267).

High—level conceptual processors should initially be
able to integrate the textile information at higher levels
of complexity than low-level conceptual processors, and per-
haps demonstrate greater increases in their abilities to
integrate information than the low—level processors, as the
textile environment increases in complexity. By the end of
the instructional sequence, complexity is purposefully

increased in what is intended to be a systematic manner of

19

developing a framework which allows for multidimensional,
multiconnected combinatory rule structures of processing
textile information.

Of research interest in the effect of a highly struc-
tured information environment on altering the processing
levels of individuals who operate at varying levels of con—
ceptual complexity. Past testing of the textile system has
demonstrated that performance on textile exams increases for
all students after working through the textile program; (see
T. Harries, 1972, for a description and results of the study)"-
but what happens to information-processing and decision-
making abilities?

Accepting the proposition that individuals disposed
to process information at a given level will function opti-
mally at a particular level of environmental complexity
(Schroder, 1971, p. 267), the concern in designing training
environments becomes how to structure information so that
individuals can categorize and integrate information in a
manner that reduces environmental complexity to their Opti—
mal level of coping ability.

In closing, the point should be made that a high
level of information processing does not necessarily define
the "best" performance for problem solving. What is "good"
performance depends upon the criterion which one applies.
When the situation is more or less static, and when the
appropriate interpretation and strategy for achieving a par-

ticular goal are known, then single rule or low levels of

20

information processing can achieve a high measure of "success."
‘When the situation is changing, when the task calls for the
definition of goals and means, and when alternative perspec-
tives are meaningful, then higher levels of information pro-

cessing are more effective (Schroder, 1971, p. 273).

Summary

A discussion of information-processing theory and
decision theory was undertaken for the development of the
conceptual framework to explain the focus of this research
on information structure rather than content. Attention has
typically been given in the past to what_was to be learned,
to the content of decisions. However, this framework takes
into account hpw individuals differentiate and integrate
information at certain levels of complexity, and the influ—'
ence that this processing of information has upon selection
among alternatives for the decision choice.

The instructional theories which bear upon how con-
cepts are learned are important to the development and selec—
tion Of a cognitive structure to use in presenting informa—
tion. The programmed textile instructional system was
selected as the vehicle of information transmission to study
because of the highly structured nature of programmed instruc—
tion. The categorical organization of information, programmed
with a prescribed logic which initially allows few degrees
of freedom for subjects, transmits content efficiently; but

what effect does information structure have on how human

 

21

beings process the content and make decisions? What deter-
mines effective techniques which aid individuals in coping
with the masses of information surrounding them? Can indi—

viduals be provided with a cognitive structure for processing

 

diversified information input in order to transform it into
useful behavioral output? How can learning environments
(cognitive structures) be designed which "teach" the process—
ing of information at increased levels Of complexity, as well
as teaching information content?

The intent of this study is to look at the relation-
ships between a programmed cognitive structure and
(a) information—processing complexity, (b) decision rule
complexity, and (c) time orientation, in order to establish

their relevance to the above questions.

CHAPTER II

REVIEW OF LITERATURE

The literature review is organized according to the
hypotheses set forth for the study. The diagram below
depicts the relationships examined among the three criterion
variables: (1) information-processing complexity,

(2) decision-rule complexity, and (3) decision time orienta—
tion, and how they are viewed in regard to experience with

the programmed cognitive structure.

Information—Processing Complexity

   

 

’8
Programmed II 5+4 :~§ Decision—Rule
Cognitive Structure <: > >4J >2 Complexity
0 on“
o O m
:11 32>
.9

 

 

Decision Time Orientation

HYPOTHESIZED RELATIONSHIPS: Diagram Of how the
three variables are examined in relation to each
other and to the programmed cognitive structure.

Figure l.--Hypothesized relationships.

22

23

The first three hypotheses predict increases in each
of the variables after working through the textile program.
The remaining four hypotheses suggest relationships between
information-processing complexity and the elements of deci-
sion style (time and rule). The literature is reviewed
according to the hypothesis grouping described above: the
first section contains support for Hypotheses I, II, and III,
under the heading "Effect of Cognitive Structure"; the second
section, entitled "Information—Processing Compexity as
Related to Decision Style Elements," contains support for
the variable relationships specified in Hypotheses IV through
VII.

In the first section, literature in the areas of
information processing and cognitive development is reviewed,
as the research relates to the effects of information organi-
zation on human behavior. The human being is viewed as a
cybernetic system, which transforms inputs Of information
into outputs of behavior, specifically decision—making behav-
ior. Each Of the following hypotheses is discussed in rela-
tionship to information organization or cognitive structure.

Hypothesis I: The level of complexity Of textile
information processing will be increased
after working through the textile
program.

Hypothesis II: After the experience of working through
the textile program, the decision rule

of decision-making style will be
more complex.

24

Hypothesis III: After the experience of working through
the textile program, time reference of
decision-making style will be more
future oriented.

Effect of Cognitive Structure

Information—Processing
Complexity

 

 

Processing Models.——The information model proposed
by Kuhn (1963) identifies human behavior as the product of a
controlled or cybernetic system, "in which the concept—
perception subsystem is the detector, the motive subsystem
is the selector, and the motor and behavioral subsystem is
the effector, with feedback to the detector system." The
motive system is seen as dealing with preference functions
and the concept system with opportunity functions (Kuhn,
1963, p. viii).

Information input is viewed in two phases. First,
an individual must devise his means of coding or classifying
information, which can be considered comparable to what psy—
chologists refer to as concept learning, or what philosophers
refer to as induction. The second step in processing infor-

mation inputs involves perception, which could be identified\,

 

with deductive processes (Kuhn, 1963, pp. 57-58).

Concepts are stored information accumulated over a
period of time, and must be identified by the individual as
representing certain things in his environment (Bruner,
et al., 1956). Perceptions are the information about par—

ticular things currently happening; and how information is

25

perceived relates not only to the stored concepts, but to
stimuli currently received. One of the most basic communi-
cation concepts is that perception is within an individual.
What he perceives is true to him, regardless Of what someone
else sees as reality. Thus, interest in perception lies in
individual differences observable in the classification and
interpretation of information.

Information alone, therefore, does not determine
behavior, since behavior is the result of selectively process—
ing some information inputs and not others. Human behavior 1
is typically viewed as selective or adaptive behavior, mean—
ing that the individual selectively perceives which informa—
tion to process. There are many possible responses to any
given set of stimuli, and adaptive behavior consists of choos—
ing the relatively satisfactory responses in preference to
the relatively unsatisfactory ones (Kuhn, 1963, p. 55).

Because there is so much information available for
processing, individuals must devise their means of filtering
out certain stimuli. "Information—input overload" is one
of the prevalent problems of current society, interfering
With the performance of most human beings. Methods of deal—
ing with information overload discussed in communication
literature include: omission, queueing (piling up unattended
items for later attention), approximation, multiple channels,
chunking (dealing with blocks of items), filtering, and

escape (Platt and Miller, 1969, p. 295).

26

Platt and Miller suggested that queueing is used too
much, while approximation not enough. Further, multiple
channels might be effectively used to give greater attention
to nonroutine decisions and let routine ones by handled
automatically by other designated individuals. Chunking and
filtering likewise are cited as being useful, and individ—
uals need to be made aware of such means of handling over—
load. Training programs, i.e., educational packages, must
be developed to help individuals deal with increasingly com-
plex information inputs, or man will be overloaded beyond
his capacity to process information effectively when called
upon in varying problem-solving situations. Such consequen-

ces are discussed in Alvin Toffler's book, Future Shock

 

(1970).

The experiments of Bruner's research group (1956,
1964, 1966) indicated that there is much variation in how
individuals process information and adapt their learning
strategies to particular problem—solving situations. Strate—
gies were also found to vary with problem difficulty.

Other research groups have also found that the means
used in processing information and solving problems vary
greatly from individual to individual, and with the situation.
A large body of research, much of which has been conducted by
G. A. Miller, et al. at Harvard, indicates that amounts of
information processed, and the complexity at which it is
processed, vary considerably, but have a finite limit

(Miller, et al., 1956, 1960). Yet, at any given level of

27

processing abilities, individuals can be trained to process
greater amounts of information at higher levels of complexity.

Although training environments can be designed which
will increase an individual's level of complexity of informa-
tion processing, individuals disposed to process information
at a given level tend to function optimally at a particular
level of environmental complexity (Schroder, 1971, p. 267).
A family of inverted "U" curve hypotheses has been generated
and researched by Schroder, et a1. (1967, 1971), which sup—
ports the predisposition toward differential levels of
processing: Information processing by people in general
reaches a maximum level of structural complexity at some
Optimal level of environmental complexity. Increasing or
decreasing environmental complexity from the Optimal point
lowers the conceptual level.

The curve representing a higher-level conceptual
functioning should reach its peak of integrative complex—
ity level in more complex environments than the curve rep—
resenting lower—level conceptual functioning. As diverse
input increases, the more complex structure curve begins to
rise earlier than the low—level complexity structure curve;
and the low conceptual curve reaches its optimum and begins
to decline while the higher conceptual curve is still rising.
The high-level conceptual curve would fall more slowly than
the low-level curve. The levels at which individuals process
information would therefore be indicative of the degrees of

environmental complexity which could be adequately handled.

 

 

m
c
-.-1
U)
m
8 High
0
H
m
6 High Conceptual Level*
"-4
4.)
m
E
H
O
m
C.
H Low
m Conceptual
o Level*
r-i
m
>
m
T Low
Low High

Environmental Complexity
Information Load and Diversity

*Level Of conceptual structure is a mediating linkage between
environmental factors and information—processing character—
istics.

Source: Schroder, 1971, p. 264.

Figure 2.——Relationship between environment and
behavioral complexity.

Overly simple environments, which fail to present
sufficiently diverse units of information, fail to stimulate
the processes of integration; therefore, a low level of con—
ceptual processing is sufficient for coping with simple envi-
ronments. Overly complex environments, which provide
excessively diverse units of information, reduce the genera—
tion of integratively complex rules for processing information
and also reduce the levels of differentiation and integration

used.

29

Research in educational development congruent with
that Of Schroder, et al. (1967, 1971, 1972) has also examined
the match between the schemata within the organism, and
characteristics of the environment as the conceptual founda-
tions for understanding development. Relevant research in
these areas includes: Piaget (1932), J. McV. Hunt (196l)--
forerunners in the area--Grant, et al. (1963), D- E. Hunt
(1966, 1968), Kelman (1961), Kohlberg (1964), Palmer (1968),
and Turiel (1966).

Organizational Learning.-—Schroder (1971, 1972) inves—

 

tigated the development Of "organizational learning" through
his model of information processing. The combining or organ-
izing process is perhaps one of the most persistent and
important problems in psychology. Although studied in the
context of Gestalt psychology, person perception, and clinical
judgment, organizational processes have been largely neglected
in learning and development studies (Schroder, 1971, p. 250).
While differentiation and the learning of dimensional
properties can be understood in the framework of traditional
learning principles, the development of conceptual struc—
tures for organizing information depends upon the nature and
extent of "organizational learning." New categories or dimen-
sions are learned via conditioning, the selective operation
Of rewards and punishments, and through observation and lan-
guage. How already acquired information is organized is a
product of the type of environment provided for organizational

learning (Schroder, 1971, p. 260).

30

The integrative complexity of conceptual structures
used in human information processing is the index for defin-
ing organizational processes proposed by Schroder, et al.
(1967). Schroder's integrative complexity refers to precise
organization prOperties, i.e., the number and connectiveness
of combinatory rules, which can be measured with the Paragraph
Completion Test (Schroder, et al., 1967; Phares and Schroder,
1969).

One type of environment commonly used for organiza-
tional learning is characterized by: (l) training-agent
determination of combinatory rule and the goal to be reached
using the dimensions learned, (2) agent determination Of
criteria for evaluating child's responses when using the
rule, and (3) agent administration of punishments and rewards
depending on the degree to which the child's responses match
the standards of the agent. This unilateral environment
rests on the principles of agent determination of combinatory
rule and agent "control of behavior" (Schroder, 1971, p. 260).
The child learns a more or less stable dispositional tendency
to look externally for a fixed rule for combining information;
he learns to avoid ambiguity and to think by the application
and modification of a single perspective.

On the other hand, another type of training for organ-
izational learning is characterized by the development Of an
environment (1) in which the child generates combinatory
rules and goals which are relevant to the dimensions or cate-

gories previously acquired; (2) which feeds back information

31

relevant to the perspectives, strategies, and goals gen-
erated; and (3) which is intrinsically interesting and
encourages the generation of different uses of the same
information. This interdependent environment rests on the
principles of the self-generation of combinatory rules and

of environmental control (Schroder, 1971, p. 260). Of course,
there are many gradations between these two extreme environ-
ments for learning organizational properties.

Experimental studies of organizational learning which
indicate that organizational training environments can influ—
ence conceptual complexity include: Schroder and Rotter
(1952), "Rigidity as Learned Behavior"; Gardiner (1971),
"Complexity Training and Prejudice Reduction"; and Krohne
and Schroder (1971), "Anxiety Defense and Complex Information
Processing."

The less an individual is exposed to organizational
training in an inductive interdependent environment, the
more conceptual organization becomes arrested at a lower
stage of develOpment. Given an Optimal inderdependent envi—
ronment, organizational development proceeds through univer—
sal stages in any stimulus domain—-from categorical
single—dimensional valuing and thinking, through multi-
dimensional single rule structure; multidimensional multirule
(unconnected) structure; to multidimensional, multiconnected
combinatory rule structure. More unilateral, overpowering,
or stressful environments lead to arrestation at some lower

stage of develOpment. Such long—term training effects in a

32

particular area, what Schroder called a domain, result in the
development of a more or less stable dispositional information-
processing structure in that domain (Schroder, 1971, pp. 262-
263). However, extreme caution should be followed in gener-
alizing from level of complexity of information processing

from one stimulus domain to another. The degree of generali—
zation will depend upon the similarity of the learning condi-
tions across domains (Schroder, 1971, p. 255).

The previous discussion supports Hypothesis I: The
level of complexity of textile information processing will
increase after working through the textile program. The
ability to integrate and process textile information (one
stimulus domain) at higher levels of conceptual complexity
should result after exposure to a cognitive structure consist—
ing of a specially designed training environment composed of
textile stimuli. The textile instruction system was based
on organizational learning techniques, in that concepts
embedded in a simple environment were presented first, with
the same concepts later presented in progressively more com—
plex environments to stimulate integrative processes. The
flexibility of integration involved in information processing
would be expected to increase as the environment became richer
and presented more diverse information, until an optimal
level of functioning was reached (Streufert and Castore,

1968; Suedfeld and Streufert, 1966). If the complexity of

the environment were increased beyond this optimal point for

33

individuals, their level of complexity of information pro—

cessing would tend to decrease.

Decision Rule Complexity

 

Decision rule complexity should also increase after
working through the textile program, as hypothesized, based
on Schroder et al.‘s information-processing research concern-
ing decision making in complex environments.

A simulated war game was used as a complex problem-
solving situation in several studies to test these "U" curve
hypotheses in relationship to decision making (Brooks, 1962;
Driver, 1962; Karlins, et al., 1965, 1967; Lawrence, 1962;
Streufert, et al., 1965a; Suedfeld, et al., 1966). Each
team in the game was given the task of making decisions
regarding an invasion, and teams were told they were playing
against another team. However, all the functions of the
Opposing team were programmed in order to achieve experi—
mental control over such factors as complexity Of feedback
and success. Team members could arrive at hypotheses about
the enemy and develop strategies on the basis of feedback.
One Objective index Of level of information processing was
defined as the number of different connections made between
different kinds of feedback and the number of integrations
or perspectives generated from a given set of Observations
in arriving at decisions.

In order to vary the complexity of the environmental

input, each team was exposed to seven differing load

 

34

conditions. The "U" curve relationship between environmental
load and diversity and level of information processing was
supported. In informationally restricted environments, the
integrative processes remained unstimulated. Conceptual
properties increased in complexity until reaching an Optimal
point, and further increases in environmental load beyond

this point led to inability to process information effectively
(Schroder, 1971).

Using a similar experimental situation, the same
relationship was demonstrated to hold between variations in
team success and failure and level of information processing
when environmental complexity was held constant (Streufert
and Streufert, 1967). This finding demonstrates the signifi—
cance of motivation and affective factors in the environment
on conceptual structure. In this experiment two groups Of
40 subjects, differing in conceptual level, were selected
on the basis Of performance on Schroder's Paragraph Comple—
tion Test. Individuals in the high conceptual level group
used connected multiple rules, while the low conceptual
groups were judged to use single or unconnected single rules
in responding to the paragraph completion stems in the
domain of interpersonal conflict and uncertainty. The
results again showed that differences in dispositional con—
ceptual level may be expressed as a family of "U" curves
which define the function relating information load and level
Of information processing in performance. Although not

expected, the optimal environmental complexity was the same

35

for both dispositional groups; but in keeping with expecta-
tions, differences in information processing between groups
were at a maximum in the central ranges of environmental
complexity and decreased as the extremes were approached
(Schroder, 1971, p. 265).

The results of the experiment do not necessarily
imply that dispositional organizational properties general-
ize from the interpersonal domain (Paragraph Completion Test)
to problem solving in a complex strategy game. In setting up
the teams and in structuring the situation as a team game,
the level of information processing in decision making was
a direct function of such interpersonal factors as (1) team
members' openness to another‘s perspectives; (2) the number
of perspectives generated; (3) the tendency to study, under—
stand, and use others' perspectives; and (4) the tendency to
support others' perspectives (Schroder, 1971, p. 266). Inde-
pendent studies supported the contention that such tenden—
cies are significantly higher in high dispositional conceptual
level teams (Schroder, et al., 1964; Lee, 1968). Another
experiment (Suedfeld and Hagen, 1966) showed that high
conceptual level subjects were better at solving complex
verbal problems, but not at solving simple ones.

In related studies using different experimental set-
tings, it has been shown that high dispositional conceptual
level teams, as compared to low conceptual level teams,

(1) discover more information about competing teams in inter-

group, international, and business simulations (Tuckman, 1964;

36

Driver, 1962); (2) exhibit higher levels of information pro-
cessing under stressful competitive conditions in a simulated
international simulation situation (Driver, 1963); and

(3) engage in more complex information processing when pre-
sented with conflicting attitudinal information in a stimulus—
social deprivation situation (Streufert, et al., 1965b;
Suedfeld, 1963). Sieber and Lanzetta (1964) also noted a
tendency for more "abstract" subjects to entertain more
conflicting hypotheses in a problem—solving task. Abstract
persons were found to be more sensitized to the conflicting
cues inherent in complex and uncertain stimuli, and were
therefore more highly motivated than concrete persons to
resolve uncertainty.

In another related set of studies, Driver (1962)
investigated the effect Of involvement on information-
processing performance resulting from attitudes developed in
an international simulation situation. He found that the
amount of different and conflicting information integrated
in decision making and the level of information processing
were highest at a moderate level of involvement. As involve-
ment became either increasingly intense, or decreased from
the Optimal point, the level of information processing
decreased. Evidently, too little involvement fails to stim-
Ulate organization processes and reduces the conceptual
Processes for structuring stimuli. TOO much involvement in

an issue can reduce the connectedness between and the number

 

37
\

of combinatory rules for organizing stimuli (Schroder, 1971,
p. 268).

Of the studies in other fields which have investi—
gated the possibilities for developing proficiency in problem
solving, one home management investigation was relevant to
the current research. Peterson (1963) investigated a ninth
grade home economics course which was designed around
problem-solving situations. The pupils appeared to gain in
ability to recognize important aspects of problem situations,
and Peterson proposed that repeated experiences in problem—
solving learning experiences would assist students in develop—
ing skills of thinking and problem solving. Since programmed
instruction capitalizes on the value of repetition in learn—
ing, the textile program would presumably aid students in
problem solving. Again, there is the consideration of how
much repetition is "Optimal" for individuals with varying
levels of conceptual complexity; a high—level conceptual
thinker could possibly become bored.

Although individuals can no doubt benefit from being
taught to use decision procedures which lead to satisfac—
tory results, a crucial aspect to improved decision making

1 is the quality, availability, and use of information (Magrabi
and Paolucci, 1970, p. 19). The significance of information
points to the importance of determining the means by which
information can be organized to allow for efficiency in com—

plex decision making.

 

l'IIII:::_____________________________________________________________—_f

38

Brim, et al. found that general values and orienta-
tions toward life, together with the cultural background of
the respondents, seemed to account for more variability in
decision making than the more traditional personality traits
(Brim, 1962, p. 234). These findings suggest that environ—
ment (information input) does influence decision making, and
training environments can be designed to teach individuals
to cope with increased conceptual complexity. This research
also suggests that people differ in the kinds of problems
they are good at solving (Brim, 1962, p. 19), which no doubt
relates to the Optimal degree of environmental complexity
at which an individual can operate, i.e., the "U" curve
hypothesis, even with organizational training.

Thus, there is support in the literature for
Hypothesis II: After the experience of working through the
textile program, the decision rule of decision—making style

will become more complex.

Decision Time Orientation

 

Information—processing theory emphasizes the impor—
tance of the structure of information--in McLuhan's terms,
"the medium is the message" (1964). The textile program is
structured with a future orientation in a manner that stu-
dents continually are required to project themselves into
future time. For example, the Objectives are stated from
the stance Of what the student will be able to do after

completing the program; the hypothetical decision situations

39

require future considerations about textile serviceability
and end use, and specific frames in the program frequently
ask: if , then .

As stated at the beginning of this chapter, man is
viewed as a controlled behavioral system with inputs of
information and outputs of behavior. When information is
input in a future frame, behavioral output likewise follows.
The view of human behavior as a cybernetic system supports
this theory.

In a cybernetic system, human behavior is considered
the product of a controlled feedback system. A controlled /
system implies the presence of a governor (decision maker), 7
and the human being governs or selects responses to informa-
tion input from among perceived Opportunities. When human
beings are perceiving, selecting, and controlling, which
includes conscious management and decision—making processes,
the effects of selected responses are fed back into the
system so as to maintain a level of equilibrium. A controlled
human system produces a particular quality of equilibrium,
not just an equilibrium. The complexity of inputs and out-
puts is a functional component between the system and its
environment, and an adaptive human system is capable of
changing goals as it acquires new information about its envi-
ronment. (This process of coping with changing stimuli and
still maintaining balance is known in systems terms as eqpi—
finality, rather than as a homeostatic type Of equilibrium.)

The cybernetic or controlled system, thus, is a primary tool

40

for analyzing goal—directed behavior, which includes manage—

ment and conscious decision making (Kuhn, 1963, p. 43).
Miller, Galanter, and Pribram (1960) conceptualized

another cybernetic model based on the postulate that human

beings organize their behavior according to plaps. Planning

implies consideration of the future. Their notion of a

plan is a TOTE model, which represents the simplest type Of

feedback loop: test-Operate—test—exit (Miller, et al.,

 

 

 

 

 

 

 

 

 

 

 

 

 

1960, p. 26).
Congruity
-——-9'- TEST >— EXIT
Incongruity
OPERATE
Figure 3.-—The TOTE unit.
The plan sequence includes: test for a discrepancy
between what appears to be and what should be. If no dis-

Crepancy exists, then exit; if there is discrepancy, operate
to reduce it; test again, and so on. This TOTE unit is based
onWhat has been called the "cybernetic hypothesis," namely
tjurt the fundamental building block of the nervous system is

the feedback 100p (Miller, et al., 1960, p. 26). Flows in

thfii TOTE unit include energy, information, and control, which

 

41

can be related to the View of human behavior as the product
of the controlled system presented previously.

Miller, et al. (1960, p. 168) indicated that all
people are "cognitive gamblers" to some extent, generally
more than is realized. Heuristic plans, rather than system-
atic or algorithmic plans, are typically followed by indi—
viduals in processing information and adapting to environ-
mental stimuli. When a systematic plan is possible, it is
sure to work but many take too long, cost too much, or are
dull and inefficient (Miller, et al., 1960, p. 160). Thus,
searches for information or a set of alternatives for prob—
lem solving are usually heuristic, which may fail to produce
the most Optimum results, but are quicker and less costly
than algorithmic plans.

Simon‘s satisficing model, in which the first accept—
able alternative is followed, would be comparable to follow-
ing an heuristic plan. Simon stated in one of his reviews
of problem solving that "human problem solving, from the
most blundering to the most insightful, involves nothing
more than varying mixtures of trial and error and selectivity"
(Simon, 1969, p. 97). He further indicated that selectivity
derives from various rules Of thumb. He equated selectivity
With some kind of feedback of information from the environ-
ment. Two kinds of selectivity were cited: (1) trial and
error or heuristic means, where various paths are tried out,

the consequences noted, and this information used to guide

42

further search; and (2) that based on previous experience,
where the problem to be solved is similar to one solved
before.

By viewing the human being as a cybernetic system
that processes selectively perceived inputs of information,
which result in certain types of behavioral outputs, one
could hypothesize that information inputs are related to
environmental complexity. Cognitive growth and the manner
in which concepts are attained and information processed
for future use are related to information organization.

Although no studies were found which examine ways
of increasing orientation toward future time, information—
processing theory and cybernetic models would indicate that
if the information structure is organized in future terms,
behavior can be altered to focus on planning and looking
toward the future.

Past research which has examined time orientation
has supported the value of increasing one's orientation
txxnard the future. Results of the Elgidaily study (1971) of
tiJne orientation and activity patterns indicated that indi-
Vixiuals who were more future time oriented performed resource-
CCulserving activities more frequently than those who were
PIKesent (Dr past time oriented; and that younger, more highly
eéhacated homemakers tended to be more future time oriented
t1hanthose who were older and less educated (Elgidaily,

1971, pp. 164—65) .

43

Therefore, the need to design organizational learn—
ing environments to increase future time orientation is indi—
cated, and information systematically presented which requires
that students consider future consequences would support
Hypothesis III: After the experience of working through the
textile program, time reference of decision-making style

will be more future oriented.

Information Processing Complexity as Related
to Elements of Decision Style

 

 

This section contains literature about the decision
process and how it relates to cognitive and environmental
complexity. The discussion is presented in two parts:
First, the relationship between information processing and
rule complexity is included in support of Hypotheses IV and
V; second the relationship between information—processing
complexity and time orientation is reviewed in support of
Hypotheses VI and VII.

Ipformation-Processing and
Decision-Rule Complexity

 

 

Hypotheses IV and V concern the relationship between
information-processing complexity and decision-rule complex-
ity:

Hypothesis IV: Those who operate at higher levels Of
conceptual complexity are likely to use
more complex decision rules.

Hypothesis V: An increase in complexity of information
processing over time will be accompanied
by an increase in complexity of dec151on
rule.

44

Several approaches have been utilized in studying
choice and attempting to relate specific decision situations
to a broader framework applicable to recurring decision
problems. Choice can be studied relative to the decision
context, to the characteristics of the decision maker, or
to the cognitive processes used in organizing information
for problem solving (Magrabi and Paolucci, 1970). Emphasis
in this section of the review will be on the last approach,
with decision—making behavior viewed in terms Of information-
processing abilities.

Decision making has been described from many differ-
ent perspectives in various frameworks, and ambiguities exist
in distinguishing decision process from product. Decision
making has typically been treated as a sequential process,
in which there are clearly distinguishable phases or steps
which may overlap and develOp simultaneously or at different
rates. The following delineation of phases in the decision
process is representative of the many listings found in the
literature:

(1) Recognize and identify the problem or issue.

(2) Search for information in the problem field (delin-
eate alternatives).

(3) Analyze and sort out the various alternatives per-
ceived (weigh alternatives).

(4) Make choice and execute action.

(5) Justify or legitimize choice has been included as a

final stage by some authors.

45

Research in the area of decision making has typically
focused on aspects other than the process, except in the field
of home management. In that discipline, however, behavioral
theory has received attention, which has yielded research
about the decision process that may be compared with
information—processing research.

Schomaker (1961) investigated financial decision
making, using a rational, sequential model with farm families.
She found the steps in decision making were not necessarily
sequential, in that the phases were Often combined or elimi-
nated. The indication that there is variance in how indi-
viduals progress through decision phases using rules of
varying complexity, supports the previous body of literature
about the differences in how individuals process information.

The process of decision was not conceptualized by
Bustrillos (1963) in the traditional manner as a sequential
process. Rather than considering phases in the decision
process, three elements-—mode, time, and rule——were
defined as influencing the process. The Bustrillos model,
explained previously, was chosen as the framework for this
research, since decision-making behavior was viewed as the
Creative product of an individual's attempt to recognize
demands from the environment and blend them with his own
needs and desires. The relationship between environmental
factors and information-processing levels is of importance
in this current research, and a decision model that could

be related to complexity was required.

 

46

Findings from the Bustrillos study indicated that
the elements of style were identifiable in the decision
protocols, and the styles varied much more than was predicted.
The styles varied with individuals and with the situation;
however, the facutal mode, present time orientation, and
preference ranking rule recurred at least twice, in responses
of more than half of the sample consisting of 16 Mexican
homemakers. Results further suggest that the nature of the
decision problem might affect the use of elements and there-
fore, the resulting style (Bustrillos, 1963, p. 122).

Rivenes (1964) replicated the Bustrillos study, using
written reports of 36 students enrolled in a college decision-
making course. Students were assigned to reconstruct recent
decision situations by writing a paper as an outside class
assignment. Factual mode and present time reference were
the two elements reported most frequently, as with the
Bustrillos study. However, the decision-making rule did not
appear frequently enough to be analyzed by Rivenes.

Another study was conducted by Hogan (1965) using
this behavioral approach to decision style. Results from
this study are not entirely consistent with the other find-
ings, perhaps due to cultural differences between the samples.
HOgan Obtained data on decision-making style from 42 home—
makers in two socioeconomic levels and found that decision
Style of the lower socioeconomic group most frequently con—
SiSted of action-suggestive mode, past time reference, and

Preference ranking rule. Homemakers in the upper socioeconomic

 

47

group typically used Objective elimination as the most fre-
quent decision rule with a factual mode and past time refer-
ence (Hogan, 1965, p. 68). The Bustrillos study indicated
that homemakers in lower socioeconomic levels most frequently
used decision-making styles which were factual, present time
oriented, and based on preference ranking.

Researchers have Often preferred to focus on specific
phases or steps of the decision process, e.g., the evaluation
stage, rather than attempting a holistic approach to studying
the entire process. In such process studies, attention has
been devoted to studying the behavior associated with various
phases of decision making in order to reduce the complexity
of the unit of analysis. The advantage of such an approach
would perhaps be supported by the findings of Bruner, et al.,
which suggest that information is more manageable when
analyzed or classified in smaller units. However, at some
point in time, there is benefit in examining the various
phases together, to determine the complexity of information-
processing activities involved in decision making.

One phase of the decision process Often studied is
the way in which individuals claim to evaluate alternatives.

Eyaluation Process (EP) models, which are based on the assump—

 

tion that an individual's preference among alternatives will
be some function of his preference for the various attributes
0f each alternative and how important these attributes are

tO him, have frequently been used for such purposes. EP

mOdels are numerous in the normative literature (MacCrimmon,

 

 

48

1968), and they occur also in the descriptive literature
sometimes, typically in simulation models.

EP models are useful in analyzing multiple-attribute
decision making, and the major EP models which have been
generated to date are based on one or more of three notions
about the way evaluations are made: additive weighting,
lexicographic ordering, or satisficing (Russ, 1971).
Bustrillos developed her decision rules, which she considered
as only one of the three decision-style elements, from these
latter approaches. Yet, the largest portion of the research
in the area of evaluation of multiple—attribute alternatives
has focused on some form of the additive model, which requires
that the decision maker choose the alternative which has the
best score on some weighted additive evaluation function.

Recent research, however (Russ, 1971a), indicated that
individuals tend to use satisficing, lexicographic ordering,
or a combination of the models, rather than additive weight-
ing in evaluating consumer products. Furthermore, "it was
clear that subjects were more comfortable and less likely to
be irritated when required to provide ranking rather than
ratings" (Russ, September, 1971b, p. 5), which are required
With the additive weighting model. Although Russ was pri-
marily concerned with predicting consumer preferences, as
are most marketing researchers, his findings, as well as
those of other researchers in the consumer decision—making
area (e.g., Alexis, Haines, and Simon, 1968), support the

Selection of satisficing, lexicographic ordering, and a

49

combination of the two, SATISLEX, as decision rules to iden-
tify in the current research.

Magrabi (1960) studied cognitive aspects Of the deci—
ssion process using concepts from economic literature. Uni-
xrersity students were asked to make hypothetical choices
(zoncerning the purchase of a second family car, and responses
xvere coded according to the decision rules: satisficing and
lexicographic ordering. Ninety—five per cent of the respon—
dents made choices which were consistent with either satis-
ficing or lexicographic ordering, and 14 per cent used the
Inost complex lexicographic rule.

Other examples Of behavioral studies related to
1phases of the decision process include: Bross's (1953)
:reference to decision "systems" of action; Simon's (1960)
"decision—making activities"; and Brim‘s (1962) studies on
'the evaluation and strategy phases of the decision process
515 related to personality characteristics.

Bross studied phases of the decision process as
"systems," and he described the deliberating step or weighing
(Df alternatives as the "value system." This system includes
€¥Ja1uation of alternatives in terms of the desirability
Eflssociated with the outcomes. The final choice phase of the
decision process involves the "decision criterion," which
Serves to integrate knowledge and permit the selection Of
aPprOpriate action (Bross, 1953).

Simon (1960) described and studied the behavior in

the first phase of the decision process as "intelligence

 

50

activity," which includes the searching of the environment
for conditions that define the problem. His "design activity"
characterizes the decision maker's action in the determina-
tion of alternatives. This designing results in the inven—
tion, develOpment, and analysis of possible alternatives.
Simon believed, however, that decision makers cannot

cope with the numerous alternatives in the environment in
terms Of "optimal rationality." An individual faces an
impossibly complex problem if he is completely rational,
carefully weighing and evaluating all possible alternatives.
The necessity of ignoring enough elements to allow a decision
to be made is known as "bounded rationality." As Simon
stated (1957):

The capacity Of the human mind for formulating and

solving complex problems is very small compared with

the size of the problems whose solution is required

for objectively rational behavior in the real world

-—or even for a reasonable approximation to such objec-

tive rationality. (p. 198)
ESimon believed that the environment possesSes properties
tflnat simplify a human's choice mechanisms by filtering out
(Martain stimuli. Put in Schroder's model, a concrete pro—
<Xassor would tend to filter out more alternatives from his
Enrvironment than would an abstract processor. Other writers
have also expressed this point of View that it is rarely
Possible to become aware of all alternatives relevant to the

decision to be made (Tannebaum, 1958; Shackle, 1961; Wilson

and Alexis, 1962).

51

Problem solving has Often been described as a search
through a vast maze of possibilities, a maze that describes
‘the environment; and successful problem solving involves
:searching the maze selectively and reducing it to manage-
eible proportions (Simon, 1969, p. 26). Further, the more
(iifficult and novel the problem, the greater the chance of
aan increasing amount of trial and error required to find a
solution (Simon, 1969, p. 95).

A large part of the literature of decision rationality
is not relevant to understanding processing since it concerns
classification of decision as to its degree or type of
rationality. However, Halliday‘s (1964) study of the ration-

ality of the decision procedure is of relevance. She defined

 

:rational decision making as "a reasoned, information-using
‘May of determining positive guides to action" (Halliday,
.1964, p. 9). This concept Of rationality applies to the
(flecision process; not to the "goodness" of the choice among
éilternatives, but whether the choice was determined through

a reasoning, weighing process based upon utilized information.
Esusentially, this approach to rationality relates to the con-
Cmnptual level of information processing and how environmental
cc>mp1exity is handled. The findings of Halliday's study
indicated that those homemakers who evidenced a more rational
Procedure in their decision making tended to perceive them—
selves as being able to exercise control over their envi-
rOnments, while those individuals using a less rational

approach perceived themselves as being more subject to chance

52

or fate. The importance attached to the situation, however,
was the most significant influence on the amount of reason—
ing, weighing, and information using done by the respondents
in thinking through the hypothetical decisions presented to
them.

Although high—level conceptual processors could be
expected to exhibit a greater amount of reasoning, weighing,
and information-using than low-level processors, the perceived
importance of the situation may be a determining factor in
how much complex processing is done about a given decision.
Since the hypothetical decision situations in the current
research study are not "life—changing" or irreversible,
complex processors may not be operating at their full con—
ceptual complexity level.

I Yet, the potential of high-level conceptual processors
to differentiate and integrate increased amounts of infor-
Ination would accompany the use of more complex decision
rules which involve evaluation, elimination, and ordering
of alternatives. An increase over time in ability to process
information at higher levels of conceptual complexity would
likewise support an accompanying increase in complexity of
the decision rule.

The research reviewed in this section on p93 decisions
are made indicates that individuals use decision rules of
Varying complexity, based on how information input is selec-

tively perceived and processed.

53

Information-Processing
Complexity and Decision
Time Orientation

 

 

Hypotheses VI and VII concern the relationship between
information-processing complexity and decision time orienta-
tion:

Hypothesis VI: Those who operate at higher levels of
conceptual complexity are likely to be
more future oriented.

Hypothesis VII: An increase in complexity of information
processing over time will be accompanied
by an increasing orientation toward
future time.

Research studies using the Bustrillos model of deci-
sion style elements have revealed several significant corre—
lations between mode (the element relating to hep ideas are
developed) and time reference, which support Hypothesis VI.
Rivenes (1964) found significant correlations frequently,
between factual mode and present time reference; and between
action-suggestion mode and future time reference. In fact,
all of the studies which have included the Bustrillos mode
Of decision style (e.g., Bustrillos, 1963; Velasco, 1964;
HOgan, 1965) have supported the relationship between factual
mode and present time reference; however, only the Rivenes
Study had enough future time references to establish any
intercorrelations with future time.

The findings of the Elgidaily study (1971) that
Younger, more highly educated individuals used more future

time references than Older, less-educated subjects, provides

an explanation of why only the Rivenes study contained

 

54

significant references to future time. Subjects in the
Rivenes study were from a college study population, while
subjects from the other studies were homemakers of varying
ages, many with limited education. I
Nonetheless, from these studies it can be hypothe-
sized that those who operate at higher levels of conceptual
complexity are likely to be more future oriented. The
recurring relationship between factual mode (which indicates
that ideas are stated conclusively with no explicit or verbal
relationship between ideas or actions, nor any consequences
given) and present time indicates that individuals who think
in the present are not processing information at high levels
of conceptual complexity. On the other hand, the action-
suggested mode, which was found to be associated with future
time, indicates that individuals are adaptive and capable Of
increased conceptual complexity in changing environments.

Hypothesis VII concerning increases over time in

 

conmdexity of information processing and future time orienta-
tzion is also supported by the Elgidaily study mentioned above.
Silice higher educational levels are associated with orienta-
tixan toward future time, organizational training environments
MIhich can increase level of processing complexity (Schroder,
1971) should be accompanied by increased orientation toward
f'thure time references.

Further, Kuhn stated (1963, pp. 254-55) that decision
making, of which time orientation is an element, is not dif—

ferent from cognitive development and information processing,

55

except that it is an extension of psychological processes

into unfamiliar and more complex environments. He maintained
that complexity in decision making excludes any stimulus sit—
uation to which a reliable response has already been learned.
Once an activity has been learned, the identification and
selection of response are not complex in the sense involved

in decision making. Simple decisions that are made auto-
matically do not require consideration of future consequences,
as do complex decisions.

Neisser's distinction between two types of human
thinking (1963, p. 2): (l) rational, controlled, and routine,
as opposed to (2) irrational, uncontrolled, and creative,
can also be tied to greater projections into future time as
complexity of decision making and processing increases. The
simpler the decision—making situation, the more routinized
the process tends to be, making possible a single sequential
process for rational thinking; whereas in complex and uncer-
tain environments, there are typically multiple paralled
Ixrocesses interacting with each other, allowing for creative
tlhinking and future projections. As environments increase
ir1 complexity and there is the need for increased levels of
CCuiceptual complexity in processing information, planning
VWith consideration of future consequences would likely
increase .

Planning becomes more apparent and critical as human
beings organize their behavior in complex situations. Miller,

SELEiL'S (1960) TOTE unit, introduced earlier in the literature

56

review, considers the benefit of developing Plans in complex
problem-solving situations. They maintained that individuals
must have ways to generate alternatives, develop Plans; and
then to Operate on them, test and evaluate (Miller, et al.,
1960, p. 169). As did Simon, Miller felt that the more com-
plex the problem, the more likely a heuristic (rather than
algorithmic) plan will be used that allows for selective
trial and error. The task of an adaptive organism is to
find the difference between the state of affairs and the
desired state of affairs, and then to find the correlating
process that will erase the difference (Simon, 1969, p. 99).

Problem solving requires a continual translation
between the state and process description of the same com-
plex reality. Planning is a way of integrating information
and decisions on a time dimension-—the more complex the
information, the more complex the decision process, with
attention given to how the future will be affected. There
is a growing body of research indicating that human problem
SCIlving is basically a form Of means-ends analysis that aims
at; discovering a process description of the path that leads
t£> a desired goal (Simon, 1969, p. 112). To the extent that
EKDal direction is necessarily future oriented, there is a
lcDgical relationship between future time orientation and
goal-directed behavior .

The condition of any goal-seeking system is that it
iS connected to the outside environment through two kinds of

Channels: the sensory or afferent channels, through which

57

the system receives information about the environment, and the
motor or efferent channels, through which it acts on the envi-
ronment. Individuals must have means of sorting memory infor-
mation, both sensory information and information about actions
(motor information), concerning states of the world. This
ability to attain goals depends on building up associations,
which may be simple or very complex, between particular
changes in states of the world (future projection) and spe-
cific actions that will bring these changes about (Simon,
1969, p. 66).

In order to study and analyze further how human beings
strive to obtain goals in complex environments, computer
problem-solving programs have been developed. Perhaps some
of the most significant computer simulations in recent years
on the processes underlying goal—directed thought are the
General Problem Solver (GPS) developed by Newell, Shaw, and
Simon (1958, 1960), Newell and Simon (1961); the Geometry-
Theorem-Proving machine developed by Gelernter, et al. (1960);
and Argus, an information-processing system, formulated by
Reitman, Grove, and Shoup (1964).

The GPS system was the forerunner of these types of
Programs, which were designed to model some of the main fea—
tures of human problem solving. GPS is a system that searches
selectively through an environment, possibly quite large and
comPlex, in order to discover and assemble sequences of
actions that will lead it from a given situation to a desired

Solution. To represent the relation between the sensory and

~58

motor worlds, the GPS is conceived as moving through a large
maze (Simon, 1969, pp. 66—67).

Simon maintained that man, ". . . viewed as a behaving
system, is quite simple. The apparent complexity of his
behavior over time is largely a reflection of the complexity‘
of the environment in which he finds himself" (Simon, 1969,

p. 52). He continued further, that most of the complex struc-
tures found in the world are enormously redundant, and that
this redundancy can be used to simplify environmental comp-
plexity.

Until complexity is simplified, decision making is
not routine. As the environment calls for increased concep-
tual complexity in processing information for complex decision
making, increased orientation toward future time follows.
Future consequences are inherent considerations when nonroutine

plans and decisions are made.

Summary
A review of the literature in the areas of information

Ixrocessing, cognitive development, and decision process and
EHTvironmental complexity indicates that there is support
ifior'the hypothesized relationships set forth in this study.
Vflfien man is viewed as a cybernetic system that transforms
jJfiputs of information into behavioral outputs, the cognitive
EStructure Of the information influences how that information

18 processed.

59

Therefore, the programmed cognitive structure used
as the information—transmission vehicle in this research
would be expected to alter subject behavior in the three
variables of interest: (1) information-processing complexity,
(2) decision-rule complexity, and (3) decision time orien—
tation. An increase in complexity of information processing
and decision rule and an increase toward future time orien-
tation are predicted after working through the programmed
cognitive structure, based on past research which indicates
that the categorical organization of information in increas—
ing gradations of complexity (i.e., organizational learning)
contributes to the efficiency with which information is
jprocessed and decisions made.

Relationships among the variables are likewise sup-
}Oorted by the literature. Information—processing complexity
is hypothesized to be related to decision-rule complexity
annd time orientation, since past research supports that the
(manoeptual complexity used in processing informatiOn contrib—

Luxes to complexity evident in the decision process.

CHAPTER III

METHODOLOGY

College students enrolled in an introductory textile
course at Michigan State University were tested at the begin-
ning and conclusion of an academic quarter to determine any
changes in (a) information—processing complexity, (b) decision-
rule complexity, and (c) decision time orientation. Treat-
ment for this descriptive study consisted of a textile pro-
grammed cognitive structure, which was used in conjunction
with two multimedia mass lectures per week, and one small-
group recitation section.

The develOpment of the textile instructional system
is explained in this chapter, followed by a description of
the research design and sample for the present study, and
‘the procedures used for examining decision style and

iJiformation—processing levels.

Treatment: The Textile System

Background

 

The design of a systems approach to information
transfer was initiated in 1968 at Michigan State University
as the foundation for development of an introductory textile
Course. This method of instruction is described as a systems

60

61

approach because of the carefully organized and tested
presentation of information, and because more than one means
of develOping appropriate student behaviors was used. Three
phases of the system have been identified to effect learning
behavior: (1) Phase I: the programmed text, which has been
develOped and tested during the past three years, is the only
completed phase and is currently in press with a national
publisher (Harries and Harries, 1973); (2) Phase II: a

series of film lopps, which is planned to provide single-

 

concept illustrations for certain basic textile principles
and processes, is scheduled for production in 1973; and

(3) Phase III: an instructional game or simulation, which

 

is intended for instructors to use as a teaching tool, is
currently under development.

These communication techniques selected to convey
the appropriate textile information were developed in coop-
eration with agencies of the Michigan State University
Instructional Improvement Service. A grant of $7,000 was
Obtained from the Educational Development Program at Michigan
State University for the design and development of Phase I
Of the system, which is the programmed cognitive structure

used as the treatment in the current research.*

*A discussion of the rationale behind selecting the
bighly structured programmed method for transmitting the
information-rich textile content is presented in Chapter I,
with the conceptual framework for the study.

62

Evaluation of Programmed
Cognitive Structure

 

 

To determine the effectiveness of the sequence of
programmed instruction as a learning tool, two groups of
students were compared over several quarters: (1) classes
studying textiles via multimedia mass lectures, recitation
small-group discussions, using a traditional text; and
(2) classes studying textiles via the experimental method
using the programmed cognitive structure, in conjunction
with the multimedia lectures and recitation small-group
discussions. Rigorously developed final exams, designed to
measure to what extent behavioral Objectives of the course
were met, were used as a criterion of student performance.
Difficulty and discrimination scores from several terms of
question usage and develOpment were taken into account in
selecting the 50 identical questions that students in both
groups answered.

An analysis of covariance was the statistical proce-
dure followed, based upon: (a) grade—point average (covari-
ant) of students prior to their enrollment in the textile
Course, and (b) performance of students on the final exam in
the course. For the purpose of the study, a student's exam
Score was recorded as the number of correct responses recorded
Out of the 50 identical questions on both tests. The hypoth-
eSis that the "Performance of the Experimental Group will be

greater than the Performance of the Traditional Group" was

63

supported beyond the .02 level of significance (T. Harries,
1972).

Since the significantly increased performance of
the experimental group was attributable to the programmed
cognitive structure, the program is the treatment of interest
in this study. The multimedia mass lectures and recitation
sessions were structured essentially the same for both groups,
and are being used as a part of the textile system only until
Phases II and III, the film loops and game, are ready for
classroom use.

Since fall quarter, 1969, 75 per cent of the 1500+
students using the programmed materials have earned a 2.0 or
better, with more than 40 per cent regularly scoring above
a 3.0, when tested against the established performance meas-
ures. Less than 10 per cent of the students using the pro-
gram have made a grade below 1.5. This performance record
<>f students using the systems approach to textile instruction
VVith the programmed cognitive structure is in comparison to
the "normal" curve performance record of students who have
taken the course via the lecture/traditional textbook method.

Improved performance on exams was not the only bene-
fit; student evaluations demonstrated that over 80 per cent
0f the class were approving of the system, with less than 5
Per cent overtly negative. Student ratings of the course
increased significantly over ratings taken when a conven—
tional text was used. A detailed study has been conducted

Of student attitudes and personality factors (locus of

64

perceived control and intrinsic/extrinsic motivation), as
related to information-processing levels and to this par-

ticular method of instruction (T. Harries, 1972).

Assumptions

 

Assuming that the textile programmed cognitive struc-
ture increases the number of correct decisions on exams in
an introductory textile course, based on past experience with
the program, the manner in which decision styles and
information-processing levels are altered is the research
question of interest. A review of past behavioral studies
in the area of decision making (Magrabi and Paolucci, 1970)
and information processing (Schroder, et al., 1967, 1971)
supports the feasibility of the current study, in that the
effect of information organization can be analyzed and eval-
uated. Therefore, the following assumptions were accepted
and underly the design of the study:
1. The programmed sequence of textile information
improves student performance on textile exams.
2. Valid and reliable measures exist to determine
levels of complexity of information—processing and

decision-making styles.

Research Design

 

All students enrolled in an introductory textile
course were exposed to the same treatment, a programmed cog-
nitive structure, since past research has indicated that the

textile system of instruction is a more effective means of

65

transmitting information content than the traditional method
of instruction.

Pretests and posttests were designed, using both
textile and nontextile stimuli, to measure the three criter—
ion variables Of interest: (1) informationjprocessing
complexity, from the Schroder, et a1. information-processing

model; (2) decision time reference and (3) decision-rule

 

complexity, elements from the Bustrillos model Of decision
style. The measures were differentially administered to two

groups Of students:

Group I, which contained two-thirds (90 students) of the
150-member class in the study, received both
pretest and posttest measures.

Group II, which consisted of one-third (60 students) of
the class,received posttests only, to deter-
mine if the pretest measures had a sensitiz-
ing effect on posttest results.

The two groups were assigned according to recitation
sections;* three recitation sections were assigned to Group I,
and the remaining two recitation sections to Group II. The
Group II students,who were designated to take only the post-
test measures for evaluating decision style and information-
processing level, received a placebo pretest consisting of
personality measures which were unrelated to the measures

used to examine the criterion variables. The design for the

data collection is represented in Figure 4.

*Recitation sessions are the smaller subdivisions (30—
35 students) of the large class, which meet once weekly to
allow for discussion of content presented in the two mass
lectures per week and of the programmed cognitive structure
the students have worked that week.

66

.GOHuomHHOO sump mew How ccflmwcll.v musmflm

 

 

omam
ezmzeamme ommoaqm . Aaonucooc
Ham HH QDOHU
0mm
. AmHmEmm
mm Hmpcmeflnmdxmv
mm
a a
m H SOHO

 

 

 

T“. N
e o
x u
1 i
I. e
I x
a 1
In

TL

9

masm
coemflomo

 

 

I N
e O
X U
i i
I. e
I x
e 1

T-

I

e
wocmummmm

mEHB

 

 

mamum coflmflomo

BmMBBmOm

eIInxei
etilxanuon

 

wuflxmamfioo
mcflmmmoonm
[coflumsnomcH

 

 

 

 

 

I N I N
e o a O
x u x u
1 1 1 1
I. e I. e
I X I X
e 1 e 1
T I.

I I

e e

wasm mocmuwwmm
:OHmHomc oEHB

mawpm coflmflomo
EmmemMm

 

atiaxei
eIrqxequoN

 

huflxmameou
mcflmmmooum
ICOHumEH0mcH

 

 

67

Selection of the Sample

 

Students enrolled in an introductory textile course
at Michigan State University (HED 171) were selected as the
population for this study, since successful performance in
the course requires that decisions be made based on informa—
tion input of increasing complexity. The subject matter in
the course progresses in complexity from simple, basic ter-
Ininology presented first, to the consideration of various
component factors which comprise textile products, through to
the final level where integration of components is required
in making textile decisions for maximum serviceability in
specific end uses.

Further basis for selecting this particular group of
subjects was the programmed cognitive structure systematic-
ally developed and tested over the past several years (as
«explained previously), which has the following characteris—
‘tics: (1) highly structured environment; (2) rich in infor-
'mation content; and (3) information organized in gradations
of increasing complexity, in which experiential referents are
paired with textile data initially, with experiential refer-
ents phased out and replaced by an increasing hierarchy of
textile symbols and referents. A carefully designed system
Of iiiformation organization, which has potential for organi—
zaticnial or process learning, was required to determine
Possijale educational practices that influence how individuals

think '.

68

Lastly, this sample was deemed appropriate for study
laecause the 150 students in the course provided adequate
saubjects for sampling and analysis. Of the 90 Group I stu-
ékents receiving pretest and posttest measures, a random
sample of 50 was drawn for data analysis; and of the 60
<3roup II students receiving only the posttest measures, 25
“Kare randomly selected for analysis. A description of demo—

«graphic characteristics of the sample follows.

Description of the Sample
Demographic data were collected from all subjects to
(establish the constituency of the textile class. (See
Puppendix A for "Background Information" cover sheet used

vvith the pretest measures.)

(Ramparison of Groups I and II

To determine how comparable the samples were from
(Eroups I and II, chi squares were run on the nominal descrip-
txrve data, and a multivariate analysis of variance on the
Ccuitinuous descriptive data. The results in Table l and
Talile 2 indicate that the two groups were similar on all
descriptive characteristics .

Since the two groups of subjects were comparable with
re8pectLtO demographic data, and because Group I was used for
hypothesis testing, frequency counts of demographic charac-
teriStics are presented only for Group I. Group II, composed
Of subjects who took only posttest measures Of the criterion

variables, was included in the research design to determine

69

if pretests were sensitizing posttest results; these subjects
were not suitable for the hypothesis testing, which required

the use of gain scores for part of the analysis.

TABLE l.—-Chi square test of nominal descriptive data.

 

 

 

 

 

 

Chi Square C at .01
Variable Value Significance DF

Class in School 2.178 11.345 3

Major 9.121 18.475 7

Marital Status 1.562 9.210 2

Student's Occupation 2.114 15.086 5

Father's Occupation 14.675 20.090 8

Mother's Occupation 8.558 20.090 8

Has Completed FE 331 .309 6.635 1

(Decision-Making Course)

TABLE 2.--Mu1tivariate analysis of continuous descriptive data.
F—ratio for Multivariate Test of Equality of Mean
Vectors = 1.4672; D.F. = 6 and 65.0000; P<0.2035

Variable Univariate F P<

Grade-Point Average 0.204 0.653

Age 0.527 0.470

Hours Employed/Week 0.103 0.749

NO. in High School

Graduating Class 1.252 0.267

Credit-Hour Load 5.637 0.020

Hours Sleep in Past 24 0.022 0.882

Degrees of Freedom for Hypothesis = 1

Degrees of Freedom for Error = 70

7O

Demographic Character-
Istics of Group I

 

 

The demographic data of the random sample Of 50 from
Group I are included to provide the reader with the general
characteristics Of subjects used for hypothesis testing;
however, there is no design over subjects in this particular
study, and the data are not analyzed according to these
descriptive characteristics.

Class in School, Age, Sex, Marital Status.--Of the
50 students randomly selected for Group I, a majority of
them were the typical types of students enrolled in lower-
level, undergraduate courses in the Department of Human
Environment and Design. Freshmen and sophomores accounted
for 88 per cent of the sample, with 42 per cent from the
freshman class and 26 per cent from the sophomore class.
Juniors and seniors accounted for 24 and 8 per cent, respec—
tively. Most of the sample fell in the normal age range for
college students: 56 per cent were 17-19 years old, 34 per
cent were 20-21, and 10 per cent were 22 and over. Table 3

.illustrates the distribution by age and class in school.

TABLE 3.~-Age and year in college.

\

 

 

 

Age

Year in 23 or

(Rollege 17 18 19 20 21 22 more Total
Freshman 2 15 3 .. .. .. l 21

Sophomore .. .. 8 3 l 1 .. 13

Junior .. .. .. 10 .. 1 1 12

Senior .. .. .. 1 2 . l 4

 

Totals 2 15 11 14 3 2 3 50

71

The sample was primarily female, with only one male.
Ninety per cent were unmarried, with the single male in the
unmarried category, and 5 per cent were married females.

Juniors and senior accounted for four of the five married

students. (See Table 4.)

TABLE 4.--Marital status and class in school.

 

Marital Status

 

 

 

Class in School Married Single Total
Freshman 1 20 21
SOphomore .. 13 13
Junior 1 2 10 12
Senior 2 2 4

Totals 5 45 50

 

Major and Grade-Point Average.--Majors represented
in the College of Human Ecology are itemized separately,
while "no preference" and "other majors" are the remaining
grouped categories. Seventy—four per cent of the students

ill the sample are majors in the college, and the other 26 per

oeult have named majors outside the college or indicated no

prxaference." In Table 5 the percentages by major are pre-
Senlted, with the average grade-point average of subject by
miijor. Table 6 shows the grade-point average by class

1J1 school.

TABLE 5.--College major and average GPA by major.

72

 

 

 

 

 

 

 

Number of Per Cent Average GPA
Major Students of Sample by Major

Retailing Clothing

and Textiles 14 28 2.66
General Clothing

and Textiles 5 10 2.89
Interior Design 9 18 2.70
General Home Economics

(Human Ecology) 3 6 3.08
Community Services 4 8 2.78
Home Economics Ed. 2 4 2.90
No Preference 6 12 3.20
Other Majors

(Outside College) 7 14 3.30

Totals 50 100

TABLE 6.--Average GPA by class in school.

Class in Number of Per Cent

School Students of Sample Average GPA

Freshman 21 42 2.93

Sophomore 13 26 2.89

Junior 12 24 2.74

Senior 4 8 3.12

Totals 50 100

73

Student Occupation.-—The number of credit hours a

 

subject was enrolled in and hours of employment per week were
obtained to gain insight into how time was allocated, and
possibly to determine if fatigue or "overload" were a factor
in completing measures. The largest proportion of the sample,
64 per cent, was in the full-time student, unemployed category.
A breakdown of student occupations and average credits

enrolled in is presented in Table 7.

TABLE 7.-—Student occupation.

 

 

 

Number of Per Cent Average Credit-

Student Occupation Students of Sample Hour Load
Full-time Student

Unemployed 31 62 15.4
Full-time Student

Employed On Campus 10 20 15.7
Full-time Student

Employed Off Campus 5 10 14.8
Part-time Student

Unemployed l 2 11.0
Part-time Student

Employed Off Campus 1 2 g 10.0
Part-time Student

Housewife 2 4 7.5

Totals 50 100

 

Only four students were enrolled for less than 12
credits, the minimum load for regular or full-time student

status. The largest number of students was enrolled for

74

14 credits, with 15 and 16 credits accounting for a sizeable
proportion, too. The credit loads are represented in

Table 8.

TABLE 8.--Student credit-hour load.

 

 

 

 

Student Credit Number Of Per Cent
Hour Load Students of Sample
7—11 Hours 4 8

12-13 Hours 2 4
14 Hours 14 28
15 Hours 10 20
16 Hours 10 20
17 Hours 6 12
18-20 Hours 4 8

Totals 50 100

Environmental Factors.--The occupation of the parents

 

and size of the subject's high school graduating class are
reported herein to provide the reader with information about
the environment in which the subject was raised, and to pro-
vide a basis for comparison with other university student

populations. (See Tables 9 and 10.)

Instrumentation: Development Of Measures
Measures for determining decision—making style were
adapted from the Bustrillos hypothetical decision situation

techniques (1963) and related follow-up research. Schroder,

75

TABLE 9.--Parent's occupation.

 

Father's Occupation Mother's Occupation

 

 

Number of Per Cent Number Of Per Cent

 

 

Occupations Subjects of Sample Subjects of Sample
Professional 3 6 1 2
Public Education/

Service 2 4

Agriculture 5 10

Housewife 0 31 62

Skilled Worker 11 22 4

Semi-skilled Worker 1 l

Unskilled Laborer 0 1

Clerical 2 5 10

Business—-Sa1es 8 16 3 6

Business——

Management 15 30 0 0
Totals 47* 94* 50 100

 

*Three fathers deceased.

TABLE 10.--Size of sample's high school graduating class.

 

 

Number in Number<of Per Cent
Graduating Class Subjects of Sample
1- 99 5 10
100-299 12 24
300-499 14 28
500—599 11 22
600-999 - 6 12
1000+
No Data 1

 

Totals 50 100

 

76

et al.'s Paragraph Completion Test (1967) was used to deter-
mine level of integration of information processing. Both
procedures were modified to include opportunity for textile
decision making and for textile information processing, since
past research has indicated that decision making tends to be
situation—specific and information—processing levels are
related to subject-matter domains.

In addition to the textile measures used, nontextile
decision situations and paragraph completions were included
to determine if any increase in information processing or
decision-making complexity carried over into cases which were
not specifically covered in the programmed cognitive struc—
ture. Further, nontextile stimuli have been used in the
above-mentioned studies, and were a more direct replication
of past research.

The measures used in the study were revised and
decided upon, based on the extent to which they:

1. Elicited responses which reveal or display character—
istics intended to be measured.

2. Yielded a level of the characteristic which does not
vary or fluctuate across responses as a result of:
a. Order of presentation of stimuli (measures are to

be administered randomly).

b. Age of respondent within range of 17 to 30.
c. Mood or immediate past experiences of respondent:

fatigue.

77

3. Elicited responses which discriminate among subjects.
4. Could be adequately responded to in less than 50

minutes per test session.

Pilot Study to Test Measures

 

Fall quarter, 1971, preliminary versions of all meas—
ures were administered to determine the extent to which the
above criteria were met. Decision style and information-
processing measures were administered at the same time and
under the same conditions to two groups of students: (1) to
48 students who had never had any college—level textile
courses (to represent the population in the pretest situation
of the study), and (2) to 50 students just completing HED 171
(to represent the population in the posttest situation).

Paragraph Completion Test Pilot.--The Paragraph Com-

 

pletion Test (PCT) developed by Schroder, et al. (1967) has
been successfully replicated in numerous studies using high
school and college student populations, and found to be a
reliable measure of level of integration of information
processing in the interpersonal uncertainty domain. Schroder
reported that scorers familiar with the theoretical variables_
can be trained in about four days of concentrated work to
reach satisfactory reliabilities of .80 - .95 (Schroder,
et al., 1967, p. 190).

Six of Schroder's paragraph completion stems were
exactly duplicated to represent nontextile stimuli, and six

textile stems very similar to Schroder's stimuli were

78

developed. Each subject was asked to respond to six stems
at a given test situation (three textile and three nontextile
stimuli).

The stems were slightly modified, based on coding
results; samples of the finalized pre and post PCT used in
the study are included in Appendix A.

Decision Measures Pilot.--Six hypothetical decision

 

situations were developed for pretesting, three nontextile
situations and three textile situations. The situations used
in previous decision-making studies (Bustrillos, 1963;
Halliday, 1964; Velasco, 1964; and Hogan, 1965) were evalu-
ated and adapted to suit the college student pOpulation

under study, composed primarily of nonmarried female, under-
graduate students.

Variations in the previously used hypothetical situ-
ations were required, since subjects in related past studies
were homemakers rather than college students. Further, low-
incomes, limited formal education, and ethnic considerations
(Bustrillos' pOpulation was composed of Mexican homemakers
and Velasco's of Filipinos) were factors considered in revis—
ing former hypothetical decision situations, which were not
appropriate to the current population. Probe questions were
also revised and developed to generate responses which would
be relevant enough to generate codable responses from the
college student population.

Each subject participating in the pilot study

responded to two decision situations: one involving textile

79

information, and the other, nontextile information. Six
situations were tested in all, three textile and three non—
textile problems, with the intent of selecting the four
situations (two in each subject category) which elicited the
most-codable responses. The four hypothetical decision
situations selected as yielding the most—codable results are
included in Appendix A, in the format used in the test situ-

ation.

Data Collection

 

The revised Pilot Study measures, included in Appen-
dix A, were administered during winter quarter, 1972, to
students enrolled in an undergraduate textile course (HED 171)
at Michigan State University. The pretest and posttest data

collection periods are described below.

Pretest Situation

 

During the first recitation period of the quarter,
subjects responded to a designated set of measures. Students
in the three recitation sections designated as Group I
received paragraph completion stems designed to measure
information-processing complexity, and hypothetical decision
problems designed to determine decision rule and time refer-
ence. Students in the two recitation sections designated
as Group II, established as a type of control, received
placebo measures relating to personality factors.

The recitation periods, in preference to the mass

lecture section, were chosen for administering the pretests

80

and placebo measures, since the PCT required 120-second
interval timing and the personality measures did not. To
prevent any Hawthorne or sensitizing effects, the subjects
were not to know that different measures were administered.

The entire 50-minute recitation period was devoted
to the pretest data collection. The development and history
of the textile system of instruction were briefly explained
to the class, and students were told that they would be par-
ticipating in a study winter term to evaluate further the
programmed sequence of instruction. They were encouraged to
answer all measures as thoroughly as possible. However, they
were assured that there were no right or wrong answers, and
that grades would not be assigned to the responses. Further,
it was emphasized that the grade they received in the course
would not be influenced by their performance on the measures.

Before timing for the test began, the students were
asked to use all the time allowed for completing each phase
of the measure, and they were asked to remain in their
chairs until the papers were collected at the end of the
period. The measures were then distributed, and five minutes
were allotted for students to read the instructions. Ques-
tions were called for, but none was asked. The subjects
Vfiare then requested to fill out the "Background Information
Sheet," which was the first page of the pretest.

All subjects who took the decision style and infor-
mation processing pretest measures responded first to the

paragraph completion stems, and second to the decision

81

situations. However, ordering within the stems series and of
the decision situations was systematically varied so that no
one stem nor decision situation consistently occurred first,
and each arrangement occurred an equal number of times.

Each subject responded to six paragraph completion
stems, three containing textile stimuli and three with non-
textile stimuli. One half of the sample received the "Y"
series, and the other half received the "X" series (subjects
received the Opposite series on the posttest); there were no
stems duplicated from one series to the other, and order of
the stems was rotated. For example, the stems were sequenced
according to the following format: X-l, X-2, X-3, X-4, X-S,
X—6; then X—2, X-3, X-4, X-5, X—6, X-l, and so on.

Each subject responded to Part Two Of the test, the
Decision Style Measure, by answering the four probe questions
for both a textile and a nontextile hypothetical situation.
One—half of the subjects received an A-l textile situation,
the other half the A-2 textile situation; likewise, half of
the students received the B—1 and the remainder the B—2 non-
textile situation. (Again, situations were reversed for all
subjects on their posttest.) "A" and "B" situations were
combined and ordered in all possible ways; for example, an
equal number of subjects received the following ordered com-
binations: A-l, B-l; A-l, B—2; A—2, B-l; A-2, B-2; B-l, A—l;

B-l, A-2; B-2, A-l; B-2, A—2.

82

Posttest Situation

 

The last class meeting winter quarter of the mass
lecture period was selected for administering the posttest
measures. Since all subjects were receiving similar measures
that required the same timing, the mass lecture period was
used rather than the five small-group recitations, in the
interest Of keeping the situation similar for the group.

Students were directed to sit by recitation sections
at the posttest session, and were told the reason was to make
distribution of tests efficient. All students received a test
package with their student numbers on the cover sheet. This
procedure was necessary to insure that no subject received a
repeat decision situation or paragraph completion stem. For
example, if a subject received an "X" PCT series on the pre—
test, he received "Y" series on the posttest. If he received
an A-l, B—2 combination and ordering on the pretest, a B-l,
A-2 arrangement was administered on the posttest. The stu-
dents who received the personality measures for the pretest
were systematically assigned an equal number of situation

combinations and processing series.

Make—up Measures

 

All students enrolled in the course were required to
take all measures. For those who were absent when the class
took the measures, make—up sessions were established. However,

subjects who took the make-up measures were not included in

83

the population when the random sample was drawn for data

analysis.

Coding Procedures

 

Information-Processing Coding

Scoring Manual.——A structural scoring manual

 

(Schroder, et al., 1967, 1969), developed and validated for
coding responses to the Paragraph Completion Test (PCT),
was available to use in determining level of integration of
information processing. Since the manual is based on the

scoring of processing structure in the uncertainty inter-

 

personal domain, rather than content, there were no manual
revisions necessary for scoring the textile stems added to
the original PCT for this study. See Appendix B for a rep-
resentative sample of the PCT Scoring Manual.

Coders.--There was no need to train and familiarize
coders with the theoretical framework and structural scoring
procedures, because experienced coders trained by the Schroder
Group at the University of Southern Illinois agreed to score
the paragraph completion responses.

Scoring.--Each subject was asked to respond to six
paragraph completion stems on the pretest, and a different
six on the posttest. All twelve of the paragraph completions
were scored on a one to seven scale, representing level Of
integrating of information processing. The lower the score,
the lower the level of complexity of information processing.

Each of the seven designated levels is explained in detail in

84

the Scoring Manual (Schroder, et al., 1967, pp. 186—196;
Phares and Schroder, 1969).

Three stems in each set Of six paragraph completions
were textile stems, and three were nontextile stems, as can
be Observed from the sample protocol in Appendix A. Follow-
ing Schroder's procedures for determining the final score
representing level of information processing, the top two
values given each set of stems were summed. Thus, to determine
level of textile information processing on the pretest, the
top two out of three textile scores were summed and the top
two out of three nontextile scores were summed.

Schroder recommends taking the top two scores in a
set of paragraph completion stems, due to the repetitive and
somewhat boring nature of the Paragraph Completion Test
(Schroder, et al., 1967, p. 195). Further, he now recommends
adding scores rather than averaging, to broaden the range of
scores (Phares and Schroder, 1969).

Each subject, therefore, ended up with four summed
scores representing the level of integration of:

Pretest-—Textile Information Processing
Pretest--Nontextile Information Processing
Posttest--Textile Information Processing

Posttest—-Nontextile Information Processing

Decision—Style Coding
The Coding Manuals developed for scoring time orien—

tation and decision rule are found in Appendices C and D. The

85

procedures used for selecting and training decision-style
coders, and for developing the coding manual are described
in this section.

Coders.--Two coders were selected for this study,
with the understanding that they would code both decision
time and rule variables for the pretests and posttests. They
were told that several days of training were required for
each variable, and that the same individuals needed to stay
with the project and code all data.

The coders were further chosen based on varying
backgrounds and experience, since one purpose of this study
was to develop a coding manual which could be used to allow:

1. Any researcher trained in use of coding manual pro-
cedures to classify responses in the same way.

2. Resulting classifications which display the char-
acteristics intended to be measured.

3. Resulting classifications which are independent of
situations or content used as stimuli.

4. Resulting classifications which adequately discrimi—
nate between levels of characteristic.
Descriptions of the two coders further illustrate the

"test" of the manual for use by varied researchers:

1. Coder One was 65 years old and just retired from a

 

long and varied teaching career, including experience
with numerous subjects from grade school through col-

lege. She had two Masters degrees and several years

86

of post—graduate work. She was from the South and
had a conservative orientation to life.

2. Coder Two was 22 years Old and a recent college grad-

 

uate, with an undergraduate multidisciplinary social

science degree. She had not yet been able to Obtain

regular employment, and had little work experience,
with no teaching experience. She was from the North
and had liberal views.

A general orientation to introduce the coders to the
study involved their reading Chapter I of the present thesis,
and discussing the research with this writer to clarify and
examine further the coders' roles. Both coders were extremely
competent, dedicated to the project, and quick to internalize
what was expected of them.

An intercoder reliability of 90 per cent was set as a
goal to reach before beginning to code the sample data, and
the goal was reached after three days of intensive training
On each variable. Although a high intercoder reliability was
obtained, there continued to be some recurring difficulties
based on the backgrounds of the coders; the problems specific
to COding each variable are discussed in the following two

sections,

Time-Orientation Scoring. After reviewing manuals

 

which had been used in past studies and reading descriptions
0f prOblems encountered, a preliminary coding manual delin-
eating eight categories of time orientation was developed to

use in training coders. Former manuals typically included

87

provision for categorization of only past, present, or future
time orientations. From working with the pilot study data
and reading research reports of problems in not having cate-
gories which allowed a combination of those three categories,
the following scale was initially proposed to allow for finer
discrimination in time coding: (1) Distant Past, (2) Recent
Past, (3) Past/Present, (4) Present, (5) Present/Future,

(6) Near Future, and (7) Distant Future. Criteria for coding
in each category are explained in detail in the Time—
Orientation Coding Manual, included in Appendix C.

Additions were made to the manual to describe further
the various time categories,based on coder training and their
work with the pilot data. An eighth category was deemed
necessary, and was developed to include those time references
to future occurrences based on past experiences. The coders
quickly progressed to the degree that they could agree on a
particular time orientation, pp they used the same unit of
analysis.

The next phase of training was, therefore, devoted to
determining the "smallest unit of analysis which had a time
referent," as was specified by the manual. A word, phrase,
clause, or even a sentence was sometimes found to be the
smallest analyzable unit, and a final description of how the
coders arrived at their breakdowns is included in the manual
instructions, included in Appendix C.

A coding sheet was devised after the pilot data were

coded, to provide a uniform format for coding and to allow

88

the averaging of time scores separately for the textile and
nontextile situations. A sample of the coding sheet is also
included in Appendix C, immediately following the Time-
Orientation Coding Manual.

With the revised coding manual and the coding sheets,
the coders then scored 20 additional protocols, this time from
the actual population but not the sample selected for analysis.
Discrepancies in coding were discussed, and agreement was so
close that scoring of time orientation for the 50 pretest
papers was begun on the fourth day of work.

Time coding of the pretest protocols went smoothly,
except for some differences encountered throughout, about
how past experience was evaluated in determining final score.
Coder One had a somewhat different view of past experience
references than did Coder Two, no doubt due to variance in
age. They did reconcile differences, though neither was com—
pletely convinced of the other's stand on the relationship
between past experiences and present or future decisions.
Further, Coder One was somewhat troubled by giving "low" time
orientations to individuals she thought were making "good"
use of the past experiences in making decisions. Her years
of experience as a teacher continued to show through, in that
she associated a low score with low grading. Repeatedly, it
was emphasized that the students were ppp_being graded or even
scored on the "goOdness" of content of the protocols, but
rather on the process used. The end result Of pretest scor—

ing, however, still yielded high intercoder reliability between

89

the coders: the average difference of time scores for the
textile situations was .261, and nontextile situation scores
averaged a .276 discrepancy.*

Upon completion of pretest coding of time orientation,
training for coding rule was initiated. The pretest and post-
test protocols were.scored for decision rule before return-
ing to the scoring of time orientation for the posttest. To
check that the coders were still together before beginning
posttest time scoring, three protocols were scored by each
coder, and discussed among the coders and the researcher. The
coders continued to be within the same range as on the pretest
coding; thus, posttest coding was begun and proceeded very
smoothly, even more so than coding of the preteSt data. At
the conclusion of posttest coding, the average difference
between coders on the textile protocols was .247, and the
nontextile situations were scored to within .217; both average
differences, therefore, totalled less on posttest coding than
on pretest coding. The total average discrepancy between the
two coders for all of the time coding was .244.

To check further the reliability of coding, later
time scores were compared with earlier ones, and pretest and
posttest scores were compared under masked identification.

At no time were the respondents identified or grouped accord—

ing to any characteristic.

 

*Formula for computing average time scores:

total of all time scores for a decision situation

 

number of clauses or phrases coded in that situation

9O

Decision-Rule Scoring.—-A preliminary coding guide

 

was developed for classifying a subject's decision rule, again
building upon coding manuals which had been used in the past.
As was done with the time variable, additional categories
beyond those used in former manuals were incorporated into

the rule coding manual. The following categories were speci—
fied for rule coding at the beginning of coder training:

(1) Immediate Closure, (2) Satisficing, (3) Lexicographic
Ordering, and (4) SATISLEX. Criteria for each category are
included in the rule section of the Coding Manual, included

in Appendix D.

After the coders mastered the different types of
decision rules and completed the first round of practice cod—
ing, they suggested separating the lexicographic ordering rule
into two categories to differentiate between an ordering
involving many alternatives and one involving few alternatives.
Both coders had difficulty in giving a score of "three"--
both to a subject who had used some ordering words, but had
not generated alternatives, and to a subject who had gener-
ated many alternatives, comparing and ranking them, too.

Thus, an additional category was added, changing the scale to:
(3) Lexicographic Ordering Of Few Alternatives, (4)Lexico—
graphic Ordering of Many Alternatives, and (5) SATISLEX, which
remained the most complex rule.

The unit of analysis for rule coding was the entire
protocol, with question four, "If you were making this decision,

how would you choose among your generated alternatives?"

91

developed to focus specifically on the rule used in the
decision process. However, if the subject did not indicate
the "how" under question four, the coders studied responses
to the preceding questions to uncover any evidence of rule in
how the subject generated alternatives, compared or ranked
them, and selected among them. In only two protocols were
there so few evidences of rule that the subject received a
zero, indicating an unclassified, noncodable response.

A Rule Coding Sheet was developed to provide a uni-
form format for coding decision rule, which allowed space
for an explanation of why a certain rule was designated. A
copy of this sheet is included in Appendix D, immediately
after the Rule Coding Manual.

During three days of intensive training sessions for
coding rule, the coders scored all of the protocols available
except the actual sample. A small group of protocols was
scored by each coder and then the protocols exchanged. Dis-
crepancies were then examined and discussed between the
coders and with the researcher, and differences reconciled.
As a result of discussion and clarification, additions were
made in the manual, further delineating criteria for the
different rules.

A problem occurred of coders viewing decision making
from different perspectives, as was the case with time coding.
However, there were increased problems in resolving rule
scoring than time scoring. With time scoring, the two scores

could be slightly different, and averaged for the data

92

analySis. However, the rule scores were not averageable,
and agreement had to be reached.

Scoring difficulties appeared to revolve around the
concept of "grading" a subject. Coder One continued to feel
that a "good" job of satisficing deserved a higher score than
a "bad" or "weak" job of lexicographic ordering. Again, the
concern with decision process in this study, rather than
content, was stressed. Due to this problem, there were some
discrepancies occurring between a score of two versus three.
The differences were generally reconciled through discussions,
and the discrepancies lessened as coding progressed.

Another problem area was with the SATISLEX rule.
Coder Two had difficulty in perceiving a protocol which rated
a five, although she would see the reasoning after discus~
sions. She was more likely to score the protocol a three or
four, picking up the ranking, but not the elimination of cer—
tain alternatives before ranking. On the other hand, Coder
One saw some fives as a result of a subject's clear statement
of preference for one alternative over the others; she felt
that in some instances (depending on wording) low preference
alternatives were, in fact, eliminated, though the subject
did not always clearly state an elimination.

When agreement was consistently reached in nine out
of ten protocols, coding of the sample data began. Pretest
and posttest protocols were mixed to increase coding relia-

bility, and later scores were compared with earlier ones to

93

insure that scores continued to be consistent. As with time,
respondent identity was masked.

The coders scored the sample data independently, and
then discussed the instances in which discrepancies occurred.
In a majority of situations, they reached agreement; in the

remaining cases, the researcher made a judgment.

Data Analysis

 

After scoring was completed, data from the measures
and demographic data on each subject were transferred to
data-processing cards for use in statistical analyses.) A
summary chart of statistical treatment is presented in
Table 11.

Multivariate procedures were selected for hypothesis
testing in this study, since such an analysis allows exami-
nation Of multiple variables which have potential relation-
ships. Advantages of multivariate methods are that they
address general questions of relationship and discrimination,
and that they consider several measures simultaneously and
can, therefore, ask a somewhat broader question than univar-
iate conterparts. Further, if a series of univariate tests
were run, an inflated alpha level would result. The multi-
variate approach is reported to be profitable when the theo-
retical constructs benefit from being broken down into their
components, and when the interactions among dependent as well

as independent variables are of interest (McCall, 1970).

94

TABLE 11.--Summary of statistical treatment of data.

 

 

 

 

 

 

 

Results
Purpose of Analysis Sample Data Statistical Test Reported
Sample Description
Comparispn of charaitir- Chi square*,** p. 69
13mlis 3.t:xpir1mgn a Groups Demographic Univariate analysis
sa p e l p ace 0 groug I and II Data of variancei p. 69
giggiigziigloiam 1e Demographic Frequency count,
p p Group I Data percent, mean*,** pp. 70-75
Test for Effect
of Pretest
Check to see if pretest Posttest scores Multivariate
measures influenced Groups of information analysis of
posttest results I and II processing, time variance of p. 97
orientation, and posttest
decision rule Of scoresi
both experimental
and placebo
samples
Hypothesis Testing
Hypotheses I, II, III Group I Gain scores from pretests Multivariate
to posttests on informa- analysis Of p. 98
tion processing, time gain scoresi
orientation, and decision
rule
Hypotheses IV, VI Group I Pretest and posttest
scores Of information Pearson 110
processing, time orienta— Product-Moment ppil3 ’
tion, and decision rule Correlationi
Hypotheses V, VII Group I Gain scores from pretests
to posttests on informa-
tion processing, time Pearson
orientation, and decision Product—Moment pp. 112,

rule Correlationi 116

 

*Leighton A. Price and James L. Peterson, A New CISSR Library System, A Pre-
liminary Report on the Overlay II System, Technical Report 70-2 (East Lansing, Mich:
Michigan State University, 1970); A File Building Routine for the New 6500 CISSR
Library System, A Preliminary Report, Technical Report 70-3 (East Lansing, Mich:
Michigan State University, 1970); Control Cards for Programs on the New 6500 CISSR
Library System, A Preliminary Report, Technical Report 70-4 (East Lansing, Mich:
Michigan State University, 1970).

**Alan M. Lesgold, Bob Zerby, and Glenn Foster, Analysis of Contingency Tables,
ACT II, Technical Report No. 14.1, Computer Institute for Social Science Research
(East Lansing, Mich: Michigan State University, 1969).

TDavid J. Wright, Jeremy D. Finn's Multivariance--Univariate and Multivariate
Analysis of Variance and Covariance: A FORTRAN IV Program, Occasional Paper NO. 8,
Office Of Research Consultation, College of Education (East Lansing, Mich: Michigan
State University, 1970).

95

The results of the data analysis Of sample character-
istics were presented earlier in this chapter under Descrip-
tion of the Sample"; the remaining statistical results used

to answer the hypothesized relationships are reported in the

next chapter.

CHAPTER IV

RESULTS AND DISCUSSION

Time orientation and decision rule, two elements of

 

 

decision style, and information-processing level Of college

 

students enrolled in an introductory textile course at Michi—
gan State University were measured at the beginning and con-
clusion of an academic quarter. A multivariate analysis of
variance procedure was used to test Hypotheses I, II, and
III, to determine if there were significant gains at the end
of the quarter in (l) information—processing complexity,
(2) decision-rule complexity, and (3) orientation toward
future time. A correlational analysis was used to test
Hypotheses IV, V, VI, and VII, to determine if there were
significant relationships among the variables.

The results are reported and discussed in this chapter
by hypothesis, under three sections: (1) Test for Effect
of Pretest; (2) Gain Scores of Information Processing, Time
Orientation, and Decision Rule; and (3) Correlations Among

Information Processing, Time Orientation, and Decision Rule.

Test for Effect of Pretest

 

The class pOpulation of 150 students was divided into
two groups to determine if pretest measures were influencing

96

97

posttest results. A multivariate analysis of variance was
run on the posttest scores of the 50 randomly selected stu-
dents from Group I, the experimental sample, and the 25
randomly selected students from Group II, the sample receiv-
ing placebo pretests.

The null hypothesis that: Posttest results of the
experimental sample will be comparable to posttest results
of the placebo sample was accepted (p<0.3266). The results,
depicted in Table 12, indicate that the pretest did not sensi-

tize subjects' responses on the posttest.

TABLE 12.--Mu1tivariate analysis of posttest scores.

F-ratio for Multivariate Test of Equality of

 

 

 

Mean Vectors = 1.1815; D.F. = 6 and 68.0; P<0.3266
Variable Univariate F* P<
Textile Processing 2.1054 0.1511
Nontextile Processing 1.1853 0.2799
Textile Time 0.5233 0.4718
Dknrtextile Time 4.2152 0.0437
Textile Rule 0.4081 0.5250
Nontextile Rule 0.0560 0.8137
Degree”; of Freedom for Hypothesis 1 N = 50

Degreess<3f Freedom for Error = 73

’WPhe univariate tests have inflated alpha values, but
do shcwv contributors to over—all multivariate analysis
probability level .

98

Gain Scores of Information Processing,
Time Orientation, Decision Rule

 

The results of the multivariate analysis of gain
scores indicate that there were over—all gains on posttest
scores over pretest scores, with a reported probability of
less than 0.0026. Thus, the general null hypothesis that
gain scores were not significantly different from zero was
initially rejected.

TO determine which variables were contributing to
the over-all significance of gain scores, the univariate
tests were examined, realizing that they had inflated alpha
levels. Table 13 shows that the over-all significance in
gain scores was attributable to the information—processing
and time-orientation variables, whereas the rule variables

did not contribute significantly to over-all gain levels.

TABLE l3.——Multivatiate analysis of gain scores.

F-ratio for Multivariate Test of Equality of

 

 

 

Mean Vectors = 4.0583; D.F. = 6 and 44.0; P<0.0026

Variable Univariate F P<
Textile Processing 6.4936 0.0141
Nontextile Processing 5.6538 0.0214
Textile Time 5.9833 0.0181
Nontextile Time 12.1054 0.0011
Textile Rule 0.2258 0.6368
Nontextile Rule 0.0479 0.8278
Degrees of Freedom for Hypothesis = 1 N = 50

Degrees of Freedom for Error = 49

99

The cell means of the gain scores further demonstrate
the more significant gains in level of information processing

and time orientation than do the gains Of complexity in

 

 

 

decision rule. The means represent the gains of pretest
scores subtracted from posttest scores. See Table 14.
TABLE 14.--Ce11 means and standard deviations of gain scores.
Variable Cell Mean Standard Deviation
Textile Processing 0.56 1.55
Nontextile Processing 0.60 1.78
Textile Time 0.17 0.49
Nontextile Time 0.19 0.38
Textile Rule 0.10 1.49
Nontextile Rule —0.04 1.29
N = 50

The above mean table illustrates the gains in each
of the variables; however, relative position on the variable
scales is not shown. Thus, the cell means for the raw pre-
test and posttest scores are presented in Table 15 to indi-
cate where subjects are in regard to the variable values.
The scoring schemata for each variable is shown in Figure 5
to make the pretest and posttest mean scores meaningful.
(See Chapter III for additional information on scoring

procedures, and Appendices B, C, and D for scoring manuals.)

100

TABLE 15.-—Cell means and standard deviations of pretest and
posttest scores.

 

 

 

 

 

Standard

Variable Cell Mean Deviation
Pretest Textile Processing 4.82 1.32
Posttest Textile Processing 5.38 1.18
Pretest Nontextile Processing 4.10 1.37
Posttest Nontextile Processing 4.70 1.18
Pretest Textile Time 5.03 0.40
Posttest Textile Time 5.20 0.37
Pretest Nontextile Time 4.73 0.30
Posttest Nontextile Time 4.92 0.37
Pretest Textile Rule 3.12 1.06
Posttest Textile Rule 3.22 1.09
Pretest Nontextile Rule 3.34 1.04
Posttest Nontextile Rule 3.30 0.89

N = 50

Each of the variables is discussed according to the
hypothesis represented, in the remaining sections under
gain scores.

Information Processing:
Hypothesis I

 

The null hypothesis--The level of complexity of

textile information processing will remain the same after

101

mmHQMHHm> wasp HDOM
Orb How OOOHOOOH
wms GOAuMpGOHHO
mEHp mo ODHM>
pmnoom Hanson one

mmaflmenm> OHDH snow
one How copuoomn
mp3 wbflxmamsoo

mass mo msHm>
OOHOOm Hmsuom one

muasmmn ummpumom

pew bmmuwnm How
.mcflmmmoond maflpxmu
Icon can mafluxmb MOM
pmpnooon was mmuoom
mourn mo uso OBu

now may mo Sun was

 

pomflnsm mom
pmcuoomm msHm>

.mmHQMHHm> sofiumbflno HOw mmemnom mcfluoomll.m musmfim

OHSpSM pcmmmum
m IIIII h lllll m lllll m IIIII v IIIII mullllm
mcflumpmo mcHOHw onSmOHO
xmqueam .mooexma -meppm .pmssH
m llllllll v IIIIIIII m IIIIIIII m uuuuuuuu H
case 30s
50H: wampmnopoe hampmnmpoe
h llllll Ollilslm llllll v IIIIII m IIIIII ml:

 

mssflpsOO msflnoom

....... 0

[line

coflpmucmfluo
mafia
ICOHmHOOO

mass mocmp
IH>O Oz wpflmeQEOU
masm
Icoflmeomo

prxOHQEOO
@cflmmmooum
ICOHmeHOMGH

30H

peppepm>

working through the textile

of the research hypothesis:

102

program--was rejected in favor

The level of complexity of

textile information processing will be increased after work-

ing through the textile program.

The univariate F tests of information processing

contributed to the over-all
F test for gain scores. As
information-processing gain
(F = 6.4936, df = 1, 49,
information-processing gain

Increased levels of

working through the textile

p<.

significance of the multivariate
was expected, the textile

was of greater significance
0141) than the nontextile

(F = 5.6538, df = l, 49, p<.0214).
information processing after

program support previous research

relating to the influence of information organization and

organizational learning as reported in the Review of Lit—

erature chapter.

A categorical organization of information

Of increasing complexity, such as the textile programmed

text, does appear to contribute to increased levels of com—

plexity of information processing.

Since the programmed cognitive structure is directed

toward the transmission of textile information,

a greater

increase in level of textile information-processing complex-

ity than nontextile information-processing complexity was

anticipated,

evidence that there is some

as the gain score results indicate.

Yet, the

transfer of level of information-

processing complexity to nontextile areas indicates that

training in one content area can increase processing com-

plexity in other areas.

The nontextile measures in this

103

study are, however, still related to consumer decision making
and do not deal with a totally different domain of informa-
tion processing.

The pretest means of textile information processing
(4.82) and nontextile information processing (4.10) indi-
cate that subjects, on the average, were operating at low to
moderately low levels of complexity in their information
processing upon entering the textile course. Conversations
with experienced scorers of the Paragraph Completion Test
(from Southern Illinois University, Edwardsville) revealed
that summed scores of four to five were about the national
average for college students on this measure. The posttest
means of 5.38 for textile processing and 4.70 for nontextile
processing indicated an increased level of complexity of
information processing, though still at the moderately low

end of the information-processing continuum.

Decision Rule: Hypothesis II

The null hypothesis—-After the experience of working
through the textile program, the decision rule of decision-
making style will not change-—was accepted. Although the
over-all multivariate analysis of gain scores was signifi-
cant (p<.0026), an examination of the univariate F tests
clearly showed that the rule variables did not exhibit
significant gains (textile rule: F = .2258, df = l, 49,
p<.6368; nontextile rule: F = .0479, df = l, 49, p<.8278).

Therefore, the research hypothesis--After the experience of

104

working through the textile program, the decision rule of
decision—making style will be more complex-~was rejected.
The mean gain scores were so slight (textile mean = .10 and
nontextile mean = —.O4) that chance alone accounted for the
difference scores from pretests to posttests.

There are several possible explanations for the non-
significant results relating to decision rule: (1) The
measures used to classify rule were not sufficiently discrim—
inating or valid; (2) the measures were acceptable, as
Assumption 2 would support, but increase in rule complexity
did not occur after the relatively brief experience of work-
ing through the textile program. In light of the results,
this writer takes the position that the measures did not
prove adequate to elicit subject responses which reflected
the total decision—making process used in making the final
choice. Therefore, Assumption 2—-Valid and reliable measures
exist to determine levels of complexity of information—
processing and decision—making styles--needs re-examination.

Past researchers have had difficulty in coding
decision rule, and both the Bustrillos (1963) and Rivenes
(1964) studies lacked evidence of the rule element to the
extent that the data could not be thoroughly analyzed. Hogan
(1965), however, was able to code a majority of responses
according to rule, and the scores in the current study were
codable in all but two situations as to rule. Although the
coders for the present study were able to score responses

for most of the variables, the rule variable was by far the

105

most difficult to determine and reach agreement on. The
different dimensions along the rule scale probably did not
discriminate discretely enough to reflect slight gains which
might have occurred in decision-rule complexity.

Perhaps the major reason that decision rule is so
hard to code relates to the problem of getting subjects to
verbalize all of the mental processes and reasoning used in
arriving at a decision. Verbal responses about decision
process are hard enough to secure, and written responses are
perhaps even more abbreviated by respondents than oral
responses. Further, the decision situations in this study
were hypothetical and subjects did not always show evidence
of seriously trying to complete the protocols as thoroughly
as requested. Thus, in real life situations, the subjects
may actually use more complex decision rules than were indi-
cated on the written, hypothetical decision protocols.

Pretest and posttest mean and frequency scores of
the decision—rule variable indicate that subjects were, on
the average, following a simple, lexicographic ordering rule,
which is of moderate complexity on the rule scale (textile
rule Y: pretest = 3.12, posttest = 3.22; nontextile rule X:
pretest = 3.34, posttest = 3.30). Past research does not
really allow for comparison as to how these results relate
to decision-rule usage of comparable subjects. However,
since the subjects in this study were near the national
average of level of information processing, which is at a

moderately low level of complexity, the moderate complexity

106

predominating with decision rule was perhaps reasonable.

The complexity of information processing, based on (1) degree
of differentiation (which refers to the number of items of
information used in a given information—processing situation)
and (2) integration of information, conceptually would seem
to contribute to increased complexity in weighing and eval-
uating alternatives,allowing for increased complexity in

the decision rule.

The conceptual relationship between decision rule and
information—processing complexity seemingly broke down when
the level of complexity of information processing was found
to increase significantly after working through the textile
program, while rule complexity did not increase. Since the
Paragraph Completion Test used to measure processing com-
plexity has been tested and validated to a much greater
extent than the decision measures, an even stronger case
might be made for the lack of validity or discrimination of
the decision measure, which showed no gains in rule complex-
ity.

Another unanticipated result of the analysis of rule
complexity scores was the slightly higher rule score recorded
for the nontextile situations than the textile situations.
This result can perhaps be attributed to the fact that the
students appeared to be more familiar with the nontextile
decision situations involving automobile and apartment selec—
tions than with the textile selections required in the tex-

tile decision situations.

 

107

Time Orientation:
Hypothesis III

 

 

The null hypothesis--After the experience of working
through the textile program, time reference of decision-
making Style will not change--was rejected in favor of the
research hypothesis: After the experience of working through

the textile program, time reference of decision-making style

will be more future oriented.

 

The univariate F tests on the time variables indi—
cated that there were significant gains toward future orien—
tation, which contributed to the over-all multivariate sig-
nificance of gain scores (F = 4.0583, df = 6, 44.0, p<.0026).
Considering that the univariate alpha levels are inaccurate
and inflated, they are still useful in showing the greater
gain in nontextile time reference (F = 12.1054, df = 1, 49,
p<.0011) than in textile time reference (F = 5.9833, df = l,
49, p<.018l).

Actually, a greater increase toward future time ref-
erence was anticipated in the textile area than the nontex-
tile area, though the univariate results above and the gain
score time reference means (textile = .1705 and nontextile =
.1861) showed just the opposite. Perhaps the relatively
short-term time constraint associated with studying textile
decisions in a college course for one term affected the
smaller gains in future time references of textile decisions

rather than nontextile decisions.

108

However, an examination of the pretest and posttest
raw mean scores indicated that subjects were initially more
future oriented in the textile areas than the nontextile
areas. With career interests related in many cases to tex-
tile areas, it would appear that textile time orientation
is more future oriented than nontextile time references.
Both pretest and posttest mean scores in the textile area
fall within the present/future range on the time—orientation
continuum (pretest textile time mean = 5.0326 and posttest
textile time mean = 5.2031), while the pretest and posttest
mean scores in the nontextile area were slightly lower on
the time scale (pretest nontextile time mean = 4.7306 and
posttest nontextile time mean = 4.920).

In the current study a tendency toward more future
time orientations was found than in past research on the
time element of decision-making style. Both Bustrillos
(1963) and Rivenes (1964) found that evidences of the present
dimension dominated the future and the past. However,
Bustrillos in her study of Mexican homemakers found that the
past dimension ranked second and future third, whereas
Rivenes with a population of college students found that
future was second and past third.

Hogan (1965) achieved still different results in her
comparative study of decision-making styles of two socio-
economic groups of homemakers, in the finding that both
groups used the past dimension of time reference most fre—

quently. A more recent study of time orientation of selected

109

rural homemakers (Elgidaily, 1971) showed that present time
references predominated the conversations of the homemaker
subjects.

Further comparisons were made in the Elgidaily study
to determine characteristics of individuals identified with
the various time orientations, since there was some evidence
in the literature of a relationship between time orientation
and age (Fink, 1953). Elgidaily (1971, pp. 142-145) found
some evidence that younger, more highly educated subjects
were more future oriented; and older, less-educated subjects
were more past oriented. Although not all past research
indicates these age/time/educational-level relationships,
the results of the current study of college students pro-
vide further support that younger, more-educated subjects
tend to be future oriented. The tendency for younger people
to be more educated today than in the past might also account
for the tendency toward future orientations of this popula—
tion of young college students.

Correlations Between Information—Processing
Complexity and Decision—Rule Complexity

 

 

No significant correlations were found between
information—processing complexity and decision—rule complex-
ity. Thus the following null hypotheses were accepted:

Null Hypothesis IV: There will be no relationship between

level of conceptual complexity and
decision-rule complexity.

 

 

110

Null Hypothesis V: An increase in complexity of information
processing over time will not necessarily
be accompanied by an increase in complex—
ity of decision rule.

 

Information-Processing Complex—
ity and Decision-Rule Complex—
ity: Hypothesis IV

 

 

 

Hypothesis IV--Those who operate at higher levels
of conceptual complexity are likely to use more complex deci—
sion rules--was rejected because of the following correlation
values:

Pretest Textile Rule and Pretest

Textile Processing = .0885
Pretest Nontextile Rule and

Pretest Nontextile Processing = -.l953
Posttest Textile Rule and

Posttest Textile Processing = -.0187
Posttest Nontextile Rule and

Posttest Nontextile Processing = -.0876

The correlation value must have been greater than .3541 to

be significant at the .01 level, which is the level necessary
to represent a meaningful correlation with this many vari—
ables; obviously, none of the variables of interest was close
to being significantly related.

The lack of relationship between decision—rule com-
plexity and information-processing complexity may be due to
the previously explained difficulties with the rule measure.
However, findings which provided some support for the valid-
ity of the rule measure were the significant correlations at
the .01 level between the pretest textile decision rule and

the pretest nontextile decision rule (.3864), and between

111

posttest textile and posttest nontextile decision rules
(.4359).

The significant gain scores for processing complex-
ity, but nonsignificant gains for rule complexity, further
indicate the lack of a directly observable relationship or
sensitivity of the variables to the same treatment. Since,
conceptually, the manner in which individuals process infor-
mation would appear to be related to how that information is
used in making decisions, other potential relationships might
exist that did not surface through the examination of the
subjects as a group. A design across subjects, with indi—
viduals grouped by the type of decision rule they used, would
provide insight into how complexity of information processing
and decision rule were related. There could be low and high
information processors within each rule group; for example,
there may be concrete and abstract processors who exhibit
satisficing behavior. Then as processing complexity was
increased, so might the ability to integrate and differen-
tiate information in a manner which would "improve" a satis—
ficing decision style, but not necessarily change it to an
ordering rule. Of course, determination of the validity of
the decision-rule measure would be a first step prior to
examining how decision-rule categories might relate to
processing complexity; further, means would need to be deter-

mined as to how to evaluate an improvement in the use of a

 

given decision rule.

 

112

Additional research is necessary to determine if, and
how, information-processing complexity might be related to
decision-rule complexity.

Gains in Information—Processing
Complexity and Decision-Rule
Complexity:- Hypothesis V

Hypothesis V—-An increase in complexity of informa-
tion processing over time will be accompanied by an increase
in complexity of decision rule-—was rejected because of the

following correlation values:

Gains in Textile Rule and

 

Textile Processing = -.1218
Gains in Nontextile Rule
and Nontextile Processing = -.2282

Neither of these correlation values was close to the
.01 level of significance of >.354l.

Since there were significant gains in level of infor-
mation processing, but no significant gains in rule complex-
ity, it is logical that there was no correlation between the
processing and rule gain scores. As has been discussed pre-
viously, the rule measure may not have been sensitive enough
to determine rule complexity and/or increases in complexity;
or perhaps the programmed treatment did not contribute to
training in the use of more complex decision rules.

Correlations Between Information-Processing
Complexity and Decision Time Orientation

 

 

No significant correlations were found between

information-processing complexity and decision time

113

orientation. The following null hypotheses were therefore

 

 

 

accepted:

Nyll Hypothesis VI: There will be no relationship between
level of conceptual complexity and
decision time orientation.

Null Hypothesis VII: An increase in complexity of informa-

 

tion processing over time will not
necessarily be accompanied by an
increasing orientation toward future
time.

Information-Processing Complex-
ity and Decision Time Orienta—
tion: Hypothesis VI

 

 

 

Hypothesis VI--Those who operate at higher levels of

 

conceptual complexity are likely to be more future oriented—-
was rejected because of the following correlation coeffi-
cients (value had to be >.354l to be significant at the .01
level):

Pretest Textile Time and

Pretest Textile Processing = -.1218
Pretest Nontextile Time and

Pretest Nontextile Processing = -.0531
Posttest Textile Time and

Posttest Textile Processing = .0521
Posttest Nontextile Time and

Posttest Nontextile Processing = —.0950

Decision time orientation was not significantly
related to information processing, based on these correlation
values. The only significant correlations at the .01 level
relating to decision time orientation were the correlations
between pretest and posttest nontextile time orientation

(.3758) and between posttest textile and nontextile time

114

(.4350). There appeared to be no particular problems in
coding time orientation, and there were some significant
correlations between different time variables; thus, the
time measure does not appear to present the same measuring
difficulties as did decision rule.

Given that both the time orientation and information-
processing measures appear to be valid indicators of the
variables, the conclusion is that time orientation and
processing complexity are not, in fact, directly correlated.

Since past research (Schroder, et al., 1967, 1971) has indi—

 

cated few relationships between age and educational levels
(intelligence) and information-processing complexity, and
there are some indications that future time orientation may
be related to younger, more highly educated individuals
(Elgidaily, 1971), perhaps time and processing are not
directly related or influenced by the same environmental
treatments.

A future time orientation was hypothesized to accom-
pany higher levels of conceptual complexity, because looking
toward the future does require consideration of uncertain
and disparate types of information, which must be integrated
and projected in order to make decisions about the future.
The ability to use and combine alternative rule structures,
which is a characteristic of high levels of information
processing, would seem to be fostered by a future time
orientation. One of the behavioral implications associated

with high—level integration of information processing includes

115

the utilization of different interactive processing and the
ability to cope with situation change over time—-which implies
coping with future considerations.

Since these findings, nonetheless, did not support a
direct relationship between high levels of information pro—
cessing and future time orientation, perhaps individuals
differentiate and integrate information based on the number
of stimuli at a given time, regardless of the time reference.
For instance, an individual who Operates at high levels of
complexity of information processing may be able to differ-
entiate and integrate numerous present stimuli, if a present
decision is at issue, or numerous past stimuli to use in
making a future decision. Different educational experiences
may be needed to develOp complex information—processing
skills and orientations toward future time.

In many decision-making situations in complex envi—
ronments, a high level of complexity of information process-
ing is considered desirable, and a future time perspective;
thus, additional research is needed to determine how these
characteristics can be fostered, and if there are common
dimensions to both.

Gains in Information—Processing
and Time Reference: Hypothesis VII

 

 

Hypothesis VII——An increase in complexity of infor—
mation processing over time will be accompanied by an
increasing orientation toward future time--was rejected

because of the following correlation values:

 

116

Gains in Textile Time and

Textile Processing = .0407
Gains in Nontextile Time
and Nontextile Processing = -.0172

Time orientation was not significantly related to
information processing, based on these correlation values,
This was not as expected, especially since both time and
processing showed significant gain scores. Thus, the subjects
who increased in complexity of information processing were
not the same ones who became more future oriented after
working through the textile program.

Perhaps in the relatively short time constraint of
one academic quarter, increases in all variables were too
much to anticipate. The treatment may well be effective,
as the correlation results indicate, in increasing one vari-
able at a time. Apparently the processing structures of some
subjects were affected by the programmed means of information
categorization, while the programmed treatment altered time
reference in other subjects.

Additional research is needed to determine what
accompanies and causes gains in information-processing
complexity and gains in future time orientation, since
increases in each are believed to be desirable in information-

rich environments.

Summary of Results

 

Significant gain scores were found with two of three

variables, measured before and after subjects worked through

117

a programmed sequence of textile instruction. The categorical
organization of information appeared to contribute to sig—
nificant increases in level of complexity of information
processing, and in orientation toward future time reference;
but there were no significant increases found in decision-

rule complexity after subjects worked through the programmed

cognitive structure .

Further, there were no significant correlations among
any of the three variables--information—processing complexity,
decision-rule complexity, and decision time orientation--on

pretest and posttest scores, nor among gain scores of the

variables.

CHAPTER V

CONCLUSIONS AND IMPLICATIONS

The results of this study indicate that the organi-
zation of information in gradations of conceptual complexity
can increase level of complexity of information processing
and orientation toward future time. The three criterion
variables of interest--information-processing complexity,
decision—rule complexity, and decision time orientation—-
were measured before and after college students worked
through a programmed course of textile instruction.

Conclusions from the findings are presented in the
first two sections of this chapter under (1) Effects of the
Programmed Cognitive Structure and (2) Relationships Among
the Variables, followed by (3) a discussion of limitations
of the study. The final section (4) of this thesis includes
consideration of educational research implications which

point to areas for further study and application.

Effects of the Programmed Cognitive Structure

 

Information Processing

 

There were significant gain scores in level of com-
plexity of information processing, measured before and after
subjects worked through the programmed sequence of textile

118

119

instruction. The gains in processing level were significant
in both textile and nontextile areas; however, as expected,
the gains in complexity of textile information processing
were greater than those recorded for nontextile information
processing.

Although no cause/effect conclusion can be drawn,
there appears to be evidence that the categorical organiza-
tion of information presented in increasing degrees of inte-
gration and differentiation (i.e., the programmed cognitive
structure) does increase the level of complexity at which
information is processed, especially in the content area at
issue.

Decision Rule of
Decision Style

 

 

There were no significant gains in decision-rule
complexity measurable from pretest to posttest. Rather than
concluding that information organization does not affect the
process of decision making, additional refinement of the
rule measure is recommended. There were problems in coding
the decision rule, which appeared to stem from difficulties
in getting subjects to verbalize their total thought pro-
cesses used in making the final choice. An initial assump—
tion that valid and reliable measures exist to determine
levels of complexity of information processing and decision-
making styles requires re-examination in regard to decision—

rule measures. The rule measure possibly did not discriminate

120

discretely enough between the decision rules specified, to
reflect any gains in rule complexity which might have
occurred.

Time Orientation of
Decision Style

 

 

Significant gains toward future-oriented time refer-
ences were found in both the textile and nontextile areas
after subjects completed the textile program. These results
suggest that the modification of time orientation toward the
future can be achieved by the organization and presentation
of information in increasing degrees of complexity, in a
manner to encourage subjects to consider future consequences

of given textile decisions.

Relationships Among the Variables

 

No significant correlations were found among pretest
and posttest scores of information-processing complexity,
decision-rule complexity, and decision time reference; nor
among gain scores of any of the variables.

Information Processing
and Decision Rule

 

 

There was no evidence that individuals who operate
at higher levels of conceptual complexity are likely to use
more complex decision rules. The rule measure may not have
reflected the true level of complexity used in finalizing
choice, due to the problems presented previously of deter—

mining the actual thought processes used by the decision

121

maker. Individuals perhaps were operating at higher levels
of complexity than they recorded in their written responses
to hypothetical decision problems. Thus, this writer does
not conclude from these results that no relationship exists
between complexity of information processing and complexity
of decision rule.

Even considering the difficulties observed with the
rule measure, there may be relationships not tapped with the
method of data analysis used. A design over subjects in
which individuals were grouped according to the type of
decision rule recorded, might show how different levels of
processing relate to specific levels of decision—rule
complexity.

Information Processing
and Time Orientation

 

 

The correlations did not indicate, either, that
individuals who operate at higher levels of conceptual com-
plexity are likely to be more future oriented. Since the
measures used to determine decision time reference and
information-processing levels were believed to represent
accurate scores, the conclusion was reached that processing
complexity was not directly correlated with orientation
toward future time. Apparently, there are high and low
processors within each time reference; i.e., individuals
who operate at high levels of information—processing complex-

ity integrate and differentiate information in regard to

122

past, present, or future time reference, based upon their

time orientation.

Gains in Information Processing
and Decision Rule

 

 

With no significant gains in rule complexity, it
followed that there were no significant correlations between
gains in information-processing and rule complexity. The
hypothesized relationship that an increase over time in
information processing complexity should be accompanied by
an increase in decision-rule complexity was not rejected.
However, since there is some evidence that the rule measure
was not sensitive enough to pick up slight gains in rule
complexity, further study is needed.

Gains in Information Processing
and Time Orientation

 

 

Although there were significant gains in processing
scores and in orientation toward future time, the programmed
treatment did not produce gains in both variables in the
same subjects. The treatment apparently influenced indi-
viduals differently; a tentative conclusion from these find-
ings is that over a relatively short time period, significant
gains in complexity can be expected in only one variable for

a given subject.

Limitations of the Study

 

The following factors limited the research findings,

and should be considered when interpreting results:

123

l. The sample was selected from a class of 150 textile
students at Michigan State University, which is not repre-

sentative of the general population, nor necessarily even of

college students. Further, the subjects were primarily
female.
2. Gains scores are not always considered a reliable

measure, and the relatively short time constraint of one
academic quarter from pretest to posttest measurement was a

limiting factor.

Implications of the Study

 

The two major implications from the findings of this
study are that information can be organized to produce:
1. Increases in the level of complexity of information
processing.
2. A more future-oriented time reference.
Implications of these findings are discussed in

this section according to educational applications and

 

indications for future research.

 

Educational Applications

 

The results are particularly applicable to the
design and organization of information in educational train-
ing environments. Providing means for individuals to cope
more effectively with information—rich environments is
critical to decision making in an industrialized society.

The past legacy of formal education has emphasized the acqui-

sition of information content, rather than information

124

processing, or how to use the information. The static model
of education (memorize, describe, and analyze given struc-
tures),which is a part of this legacy, is appropriate only
when man's environment changes very slowly and has a low
level of symbolic referents, such as in primitive or nontech-
nological societies. The twentieth century human being
cannot exist compatibly with the high—information external
environment unless his neural structure (internal map) is
specifically trained in the techniques of information acqui-
sition analysis. This research indicates that a programmed
structure of information presentation can provide structure
for processing information at increased levels of complexity.
Although content is important and must accompany
training in information processing, a structure must be pro—
vided so that individuals might selectively process a variety
of types of content for given purposes. Information content
changes so rapidly, and there is so much of it, that learn~
ing volumes of content seems neither practical nor possible.
The ability to process information at high levels of concep-
tual complexity has been demonstrated to be an asset in
complex environments which demand complex decision making
(Schroder, et al., 1967, 1971). There is too much "informa—
tion overload" now. Thus, training in the "how" of informa—
tion integration and differentiation in increasingly complex
environments is going to become increasingly critical.
Specific examples of educational applications are

now considered. First, the way these results can be applied

125

to classroom instruction is discussed, followed by implica-

 

tions for general education in the areas of management and

 

consumer decision making.

 

Classroom Instruction.-—Although an increase in level

 

of information—processing complexity resulted after subjects
worked through the programmed treatment, the implication
should not be that such a structured type of information
organization is the answer. In fact, programming is highly
structured and leads toward convergent thinking, which can
create rigid, concrete behavior in students. When program-
ming is the type of information organization used, therefore,
it should allow as much flexibility as possible. Perhaps
a less structured type of organization would contribute to
even greater increases in level of complexity of information
processing. Though the gains in processing were statistically
significant, the subjects, on the average, were still pro-
cessing information at the moderately low end of Schroder's
continuum (1967, 1971).

The categorical organization of programmed instruc-

tion, and the highly structured presentation of information

 

from simple to increasingly complex concepts are probably
the advantageous by—products of programming. They contrib-
uted to subjects' ability to integrate information better
(process at higher levels of conceptual complexity) and to
consider future consequences after working through the tex-
tile program. However, to allow for increased flexibility

of student interaction with instruction, programmed

126

instruction might be used in conjunction with other methods
of transmitting information.

Simulation and gaming techniques are one promising
means of allowing individuals to input diverse types of
information into an instructional system, and come up with
creative ways of using it. Also, the student-centered
instructional systems allow individuals to define their own
structure for dealing with information.

Therefore, the design of total instruction systems
is called for, perhaps including programming techniques as
the delivery system for certain types of information content.
The current emphasis on process curriculums in educational
literature (e.g., Berman, 1968; Schroder, et al., 1972)
illustrates the greater awareness of need for training in
information processing. Research such as the present study
provides insight into organizational learning environments
which give students content, as well as a structure for
processing information and thinking in terms of the future.

General Education.-—A main concept of management is

 

based on the fact that individuals can imagine alternatives
and consequences, and project abstractly into the future.
Presenting information in a framework which aids individuals
in differentiating and integrating alternatives, and in
evaluating them as to future consequences,would help families
manage their lives. This research suggests that programmed

units of instruction might be developed for adult, continuing

127

education programs designed to train individuals for complex
decision making.

For years, literature on decision making has sug—
gested that attention be given to the decision—making

process, typically viewed as a sequential one composed of

steps or phases. As early as 1910, Dewey was discussing the
"steps of reflection" (1910, p. 12), which were necessary
in formulating plans about future decisions. Yet, there has

been little progress made in how to help individuals organ—
ize information into a framework for decision making.

Other writers have also considered the internal or
cognitive map of individuals in terms of how information
is perceived. For example, the internal plan has been
likened to an image (Boulding, 1956), which can be cast in
a time frame, or a processing structure for handling infor-
mation. The nature of the individual's image exercises
constraint upon the decision—making process, and broadening
the perspective of the internal map is a task of learning
environments. Until recently, past or present content-
oriented time frames functioned adequately in coping with
environmental stimuli. However, in the present high-
information, high—flux environment, such orientations are
becoming increasingly dysfunctional.

The current research indicates that training envi—
ronments can be designed in specific content areas to influ-
ence behavior; for example, information can be structured

to aid consumers in decision making. The textile program

128

was designed to help students select from among textile
alternatives, and implications are promising for designing
programs in a variety of content domains for general edu-
cation.

Training is particularly needed in evaluating con-
sumer information, as is indicated by the current consumer
movement which has developed out of frustrations and dis-
satisfactions occurring in the consumer arena. Technology
has increased the alternatives available for consumption to

such an extent that a structure for processing the con—

 

stantly changing information inputs is needed. Granted, the
wide variety of options from which the consumer may choose
are readily attainable by a limited number of people due to
economic factors; yet those individuals with limited resources
are the ones to which training environments should be directed
which can help consumers evaluate information.

Individuals benefit from a structure for processing
diverse units of information, and this research indicates
that programmed techniques do provide a structural framework

which can be used in transmitting information to consumers.

Future Research

 

Follow—up research which gives insight into how
information can be structured to foster complex decision

making can be explored relative to (a) cognitive processes

 

used in decision making, (b) the decision situation or

context, or (c) characteristics of the decision maker (as

 

129

suggested by Magrabi and Paolucci, 1970). Implications for
research in each of these areas are suggested in the remainder
of this chapter.

Cognitive Decision Process.-—Research studies need

 

to get at what triggers choice—-i.e., information organized
in a particular fashion? a series of environmental factors?
a combination of both?

In the present study, decision rule was intended to
get at what triggers choice. However, additional study is
required of behavioral aspects of decision making, to refine
the means of evaluating an individual's decision rule. After
more discriminating measures are developed, research designs
other than the one used in this study should be selected.

A design that groups subjects according to the type of rule
they used in a particular decision situation might provide

new insights into potential relationships between how con-

ceptual complexity relates to decision—making processes.

Further research should also explore the effect of
different training environments on the decision process.
Although this study indicated that programming techniques
aid in developing processing skills, there are other delivery
systems which might be as effective, or even more effective
in developing cognitive processes for decision making. Par-
ticularly, delivery systems which are less structured than
programming may be very beneficial, used in conjunction
with programmed instruction. A study of total instructional

systems which train individuals in complex decision making

130

should be designed and systematically evaluated as to the
effectiveness of component parts of the instructional system.
Systems which build in adaptive behavior are particularly
desirable; thus, the structure should allow for flexibility.
Also, the influence of information organization
during the developmental stages is an area ripe for research.
The type of structure required for preschool organizational
learning environments through graduate education learning
environments would give additional insights into how adap—
tive processing behaviors are developed. Specifically,
further research is needed to determine how choices are
made, under varying conditions. How do individuals react
to organizational learning environments designed to increase
the ability to cope with many alternatives available in

complex decision making? The processes used in decision

 

making have, in the past, been found to be situation—
specific; therefore, research must follow which allows
examination of the decision situation.

Decision Situation or Context.——The current research

 

included decision situations which could be classified as

routine, sequential, repetitive types of consumer decisions;

 

 

as compared to central, unique, irrevocable types of deci-
sions which occur frequently in life and perhaps require the
most complex processing of all. Critical, life/death types
of decisions which are irrevocable may really be the kinds
of decisions in which the highest levels of conceptual

complexity operate. These decision situations need

131

researching as to how information can be organized to aid
processing; yet, these areas are very difficult to explore,
except hypothetically.

However, future research should include a wide array
of managerial decisions of varying consequences. Potential
relationships may be uncovered between certain decision situ—
ations and levels of information-processing complexity, which
have implications for designing training environments for
use in family crisis intervention, as well as for training
in the routine types of day—to-day decisions (which are the
types of decisions attended to in the textile program).

Decision Maker.—-Lastly, research is needed concern-
ing individual personality characteristics and how they
influence information processing and decision making. Do
individuals with particular personality characteristics
exhibit particular decision-making styles? Do they respond
more favorably to certain types of organizational training
environments?

Individuals could be categorized according to per-
sonality variables, such as locus of perceived control (e.g.,
internal versus external control), intrinsic versus extrinsic
motivation, and then researched for common decision—style
characteristics which surfaced with certain personality
variables. Also, consideration of the decision maker's envi-
ronment is open for research. How might family environment,

parental influences, living conditions, rural versus urban

132

surroundings, influence the way in which information is
processed for complex decision making?

The interaction of the organism with his environment
is perhaps the most fruitful approach for future research.
This approach requires an ecological thrust, in which the
interface between the organism and the situation is examined.
For instance, do genetic characteristics of the individual
override ethnic or environmental factors of the situation in
which the individual makes decisions?

Actually, information—processing and decision-making
styles cannot be fully explored by examination of the decision
process, the decision situation, or the decision maker in
isolation. Rather, all factors should be considered in the
context of an ecological system. This type of integrative
research is no doubt difficult to pursue, requiring special
research techniques and statistical tools, which need to be

developed through future research endeavors.

Summary

There is evidence that structured learning environ—
ments can raise the level of complexity of information
processing and contribute to an increased orientation toward
future time. Based on the findings of this study, a pro-
grammed course of instruction, designed with the information
presented in gradations of increasing conceptual complexity,
has the potential to increase levels of conceptual complexity

and future time orientation.

133

These findings can be applied to the design of edu-
cation training environments in order to improve individuals'
abilities to integrate and differentiate large amounts of
diverse information for complex decision making. Additional
research which takes into account the decision maker in his
environment is needed to investigate thoroughly the cognitive
process a decision maker uses in varying situations. An
ecological research approach which brings together the organ—
ism and his environment is the type of research ultimately

to strive toward.

 

BIBLIOGRAPHY

134

 

BIBLIOGRAPHY

Adams, E. W., and Fagot, R. "A Model of Riskless Choice."
Behavioral Science, IV (1959), 1-10.

 

Alexis, M.; Haines, G. H., Jr.; and Simon, L. "Consumer
Information Processing: The Case of Women's
Clothing." Marketing and the New Science of Planning.

 

Edited by R. L. King. Chicago: American Marketing
Association, 1968.

Berlyne, D. E. Structure and Direction in Thinking. New
York: John Wiley and Sons, Inc., 1965.

 

Berman, L. New Priorities in the Curriculum. Columbus,
Ohio: Charles E. Merrill Publishing Company, 1968.

 

Bloom, B. S. Stability and Change in Human Characteristics.
New York: John Wiley and Sons, Inc., 1964.

Boulding, K. The Image. Ann Arbor, Michigan: University
of Michigan Press, 1956.

 

Brim, O. G., Jr.; Glass, D. C.; Laven, D. E.; and Goodman, N.
Personality and Decision Processes. California:
Stanford University Press, 1962.

 

Brooks, S. A. "Group Composition and Satisfaction in Group
Function: The Contribution of Abstractness of
Personality Structure and Ascendance." Unpublished
Ph.D. dissertation, Princeton University, 1962.

Bross, I. D. F. Design for Decisions. New York: MacMillan,
1953.

 

Bruner, J. S. "The Course of Cognitive Growth." American
Psychologist, XIX (1964), 1-15.

 

; Goodnow, J. J.; and Austin, G. A. A Study of
Thinking. New York: John Wiley and Sons, 1956.

 

 

Bruner, J. S.; Olver, R. R.; and Greenfield, P. M.; et al.
Studies in Cognitive Growth. New York: John Wiley
and Sons, 1966.

 

135

136

Buckley, W. "Society As a Complex Adaptive System." Modern
Systems Research for the Behavioral Scientist.
Edited by Walter Buckley. Chicago: Aldine Publish-
ing Co., 1968.

 

Bustrillos, N. R. "Decision—Making Styles of Selected
Mexican Homemakers." Unpublished Ph.D. dissertation,
Michigan State University, 1963.

Dewey, J. How We Think. Boston, Mass.: Heath, 1910.

 

Driver, M. J. "Conceptual Structure and Group Processes in
an Inter—Nation Simulation. Part 1: The Perception
of Simulated Nations.” ETS Research Bulletin (1962),
RB 62-15.

 

"Conceptual Structure and Group Processes in an
Inter-Nation Simulation: The Performance of Simu—
lated Nations." Unpublished manuscription, Purdue
University, 1963.

 

Dunham, J. L., and Bunderson, C. V. "Effect of Decision-
Rule Instruction Upon the Relationship of Cognitive
Abilities to Performance in Multiple-Category Concept
Problems." Journal of Educational Psychology, XL, 2
(1969), 121-125.

 

Elgidaily, D. A. "Time Orientation and Activity Pattern of
Selected Homemakers." Unpublished Ph.D. dissertation,
Michigan State University, 1971.

Fink, H. H. "The Relationship of Time Perspective to Age,
Institutionalization and Activity." Unpublished
Ph.D. dissertation, Michigan State University, 1953.

Gardiner, G. S. "Complexity Training and Prejudice Reduc—
tion." Unpublished Ph.D. dissertation, Princeton
University, 1971.

. "Some Correlates of Cognitive Complexity." Unpub-
lished Master's thesis, University of Alberta, 1968.

 

Gelernter, H. "Realization of a Geometry-Theorem-Proving
Machine." Proceedings of the International Confer-
ence on Information Processing (ICIP). Paris:
UNESCO, 1960.

 

 

Grant, M. 0.; Warren, M.; and Turner, J. K. "Community
Treatment Project: An Evaluation of Community
Treatment for Delinquents." CTP Research Report
No. 3. Sacramento: California Youth Authority,
1963.

137

Halliday, J. R. "Relationships Among Certain Character-
istics of a Decision Event: Decision Procedure,
Decision Context, and Decisionmaker." Unpublished

Ph.D. dissertation, Michigan State University, 1964.

Harries, N. G., and Harries, T. E. Textiles: A Systems
Approach to Serviceability. New York: McGraw-Hill
Book Company, in press, 1973.

 

 

Harries, T. E. "Human Information Processing, Locus of
Perceived Control, and Intrinsic—Extrinsic Motivation
as Related to Success and Attitude in a Programmed
Course of Instruction." Unpublished Ph.D. disserta-
tion, Michigan State University, 1972.

Harvey, O. J.; Hunt, D. E.; and Schroder, H. M. Conceptual
Systems and Personality Organization. New York:
John Wiley and Sons, Inc., 1961.

 

 

Hogan, M. J. "Decision—Making Styles of Two Socio-economic
Groups of Homemakers." Unpublished Master's thesis,
Michigan State University, 1965.

Hunt, D. E. "Matching Models and Moral Training." Paper
presented at Conference on Moral Education, Ontario
Institute for Studies in Education, June, 1968.

 

 

. "A Model for Analyzing the Training of Training
Agents." Merrill-Palmer Quarterly, XII (1966), 137-
156.

Hunt, J. McV. Intelligence and Experience. New York:
Ronald Press, 1961.

 

Karlins, M. "Conceptual Complexity and Remote-Associative
Proficiency as Creativity Variables in a Complex
Problem-Solving Task." Journal of Personality and
Social Psychology, VI (1967), 264—278.

 

 

, and Lamm, H. "Information Search as a Function of
Conceptual Structure in a Complex Problem—Solving
Task." Journal of Personality and Social Psychology,
V (1967), 456—459.

 

 

Karlins, M.; Schroder, H. M.; and Streufert, S. "A Fixed
Input Program for a Tactical War Game." Unpublished
Manuscript, American Documentation Institute,
Auxiliary Publications Project, Document No. 8621,
1965.

Kelman, H. C. "Processes of Opinion Change." Public
Opinion Quarterly, XXV (1961), 57-79.

 

138

King, R. H. "A Study of the Problem of Building a Model to
Simulate the Cognitive Processes of a Shopper in a
Super-Market." Consumer Behavior: Learning Models
of Purchasing. Edited by George H. Haines, Jr.

New York: Free Press, 1969.

 

 

Kohlberg, L. "Development of Moral Character and Moral
Ideology." Review of Child Development Research.
Edited by J. Hoffman and L. Hoffman. Vol. I. New
York: Russell Sage Foundation, 1964.

 

Kornreich, L. B. "Strategy Selection and Information
Processing in Human Discrimination Learning."
Journal of Educational Psychology, LIX, 6 (1968),

 

438-448.
Krohne, H. W., and Schroder, H. M. "Anxiety and Complex
Information Processing." Unpublished research

paper, Princeton, New Jersey, 1971.

Kuhn, Alfred. The Study of Society: A Unified Approach.
Homewood, Illinois: The Dorsey Press, Inc., 1963.

 

Lawrence, E. A. "An Investigation of Some Relationships
Between Personality Structure and Group Functioning."
Unpublished Ph.D. thesis, Princeton University, 1962.

Lee, R. E. III. "Dispositional and Induced Information
Processing Structures." Technical Report No. 18,
Princeton University, Office of Naval Research,
1968.

Lesgold, A. M.; Zerby, B.; and Foster, G. Analysis of
Contingency Tables, ACT II, Technical Report No.
14.1. Computer Institute for Social Science Research.
East Lansing, Mich: Michigan State University, '
1969.

MacCrimmon, K. R. "Decision Making Among Multiple—
Attribute Alternatives: A Survey and Consolidated
Approach. Rand Corporation Memorandum, RM-4823-ARPA,
December, 1968.

McCall, R. W. "Addendum. The Use of Multivariate Procedures
in Developmental Psychology." Carmichael's Manual of
Child Psychology. Edited by Paul H. Mussen. Vol. I.
New York: John Wiley & Sons, 1970.

 

McLuhan, M. Understanding Media: The Extensions of Man.
New York: McGraw-Hill Book Company, 1964.

Magrabi, F. M. "Am Empirical Test for Decision Rule." The
Journal of Consumer Affairs, III, 1 (Summer, 1960),
41-51.

 

139

, and Paolucci, B. Decisionmaking Styles of Family
Managers. Published with the approval of the Director
of the Michigan Agricultural Experiment Station as
Journal Paper No. 5199. Revision of a paper read at
the Annual Meeting of the American Psychological
Association at Miami, Florida, September, 1970.

 

 

 

Miller, D. W., and Starr, M. K. The Structure of Human
Decisions. Englewood Cliffs, New Jersey: Prentice-
Hall, Inc., 1967.

 

 

 

Miller, G. A. "The Magical Number Seven, Plus or Minus Two:
Some Limits on Our Capacity for Processing Informa-
tion." The Psychological Review, LXIII (1956), 81-
97.

; Galanter, E.; and Pribram, K. H. Plans and the
Structure of Behavior. New York: Holt, Rinehart,
adn Winston, Inc., 1960.

 

 

 

Neisser, U. "The Multiplicity of Thought." British Journal
of Psychology, LIV (1963), 1—14.

Newell, A.; Shaw, J. C.; and Simon, H. A. "Elements of a
Theory of Human Problem Solving." Psychological

 

Review, LXV, 3 (1958), 151-166.

. "Report on a General Problem-Solving Program."
Proceedings of the International Conference on
Information Processing (ICIP). Paris: UNESCO
House, 1960.

 

 

 

Newell, A., and Simon, H. A. "Computer Simulation of Human
Thinking." Science, CXXXIV (December 22, 1961),
2011-2017.

Palmer, T. B. "An Overview of Matching in the Community
Treatment Project." Paper presented at the Meeting

of the Western Psychological Association, San Diego,
March, 1968.

Peterson, B. "Problem Solving in Home Economics. Journal
of Home Economics, LV (1963), 179—181.

 

Phares, J. O., and Schroder, H. M. "Structural Scoring
Manual for Paragraph Completion Test." Unpublished
paper, Princeton University, Princeton University
Press, 1969.

Piaget, J. The Child's Conception of Number. New York:
Humanities Press, 1952.

 

140

Logic and Psychology. Manchester, England:
Manchester University Press, 1953.

The Moral Judgment of the Child. New York: The
Free Press, 1932.

. The Psychology of Intelligence. London:
Routledge and Kegan Paul, 1950.

Platt, J., and Miller, J. G. "Handling Information Over—
load." Ekistics, XXVIII (1969), 295-296.

Price, L. A., and Peterson, J. L. Control Cards for Programs
on the New 6500 CISSR Library System, A Preliminary
Report, Technical Report 70-4. East Lansing, Mich.:
Michigan State University, 1970.

. A File Building Routine for the New 6500 CISSR
Library System, A Preliminary Report, Technical
Report 70-3. East Lansing, Mich.: Michigan State
University, 1970.

 

. A New CISSR Library System, A Preliminary Report
on the Overlay II System, Technical Report 70-2.

East Lansing, Mich.: Michigan State University,
1970.

Reitman, W. R.; Grove, R. B.; and Shoup, R. G. "Argus: An
Information-Processing Model of Thinking." Behavioral

 

Science, IX (1964), 270—281.

Rivenes, D. N. "Study of Decision—Making Style." Unpublished
Master's thesis, Pennsylvania State University, 1964.

Russ, F. A. "Consumer Evaluation of Alternative Product
Models." Unpublished Ph.D. dissertation, Carnegie-
Mellon University, 1971a.

. "Evaluation Process Models and the Prediction of

Preference." Paper presented at the Association for
Consumer Research Annual Meeting, College Park, Md.,
September 1, 1971b.

Schneider, G. A., and Giambra, L. M. "Performance in Concept
Identification as a Function of Cognitive Complexity."
Journal of Personality and Social Psychology, XIX
(1971), 261—273.

 

Schomaker, P. G2 "Financial Decision—Making as Reported by
100 Farm Families in Michigan." Unpublished Ph.D.
dissertation, Michigan State University, 1961.

141

Schroder, H. M. "Conceptual Complexity and Personality
Organization." Personality Theory and Information
Processing. Edited by H. M. Schroder and P. Suedfeld.
New York: Ronald Press, 1971.

 

 

 

 

; Driver, M. J.; and Streufert, S. Human Information
Processing. New York: Holt, Rinehart, and Winston,
1967.

Schroder, H. M.; Karlins, M.; and Phares, J. O. Education
for Freedom. New York: John Wiley and Sons, in
press, 1972.

 

 

Schroder, H. M., and Rotter, J. B. "Rigidity as Learned
Behavior." Journal of Experimental Psychology,
XLIV, 3 (1952), 141—150.

 

Schroder, H. M.; Streufert, S.; and Weeden, D. "The Effect
of Structural Abstractness in Interpersonal Stimuli
on the Leadership Role." Technical Report No. 3,
Princeton University, Office of Naval Research,
1964.

Shackle, G. L. D. Decision, Order, and Time in Human
Affairs. Cambridge, England: Cambridge University
Press, 1961.

 

Sieber, J. E., and Lanzetta, J. T. "Conflict and Conceptual
Structure as Determinants of Decision-Making Behavior."
Journal of Personality, XXXII (1964), 622—641.

 

Simon, H. A. Models of Man. London: John Wiley and Sons,
1957.

 

. The New Science of Management Decision. New York:
Harper, 1960.

 

 

. The Sciences of the Artificial. Cambridge, Mass.:
The M. I. T. Press, 1969.

 

 

Skinner, B. F. Science and Human Behavior. New York: The
MacMillan Company, 1953.

 

Smith, R. D. "Heuristic Simulation of Psychological Decision
Processes." Journal of Applied Psychology, LII, 4
(1968), 325—330.

 

Stephenson, W. The Study of Behavior: Q—Technique and Its
Methodology. Chicago and London: University of
Chicago Press, 1953.

 

142

Streufert, S. "Increasing Failure and Response Rate in
Complex Decision-Making." Journal of Experimental
Social Psychology, V (1969), 310-323.

 

., and Castore, C. H. "Effects of Increasing Suc—
cess and Failure on Perceived Information Quality.
Psychonomic Science, XI (1968), 63—64.

 

 

Streufert, S.; Clardy, M. A.; Driver, M. J.; Karlins, M.;
Schroder, H. M.; and Suedfeld, P. "A Tactical
Game for the Analysis of Complex Decision—Making
in Individuals and Groups." Psychological Reports,
XVII (1965a), 723-729.

 

Streufert, S., and Schroder, H. M. "Conceptual Structure,
Environmental Complexity and Task Performance."
Journal of Experimental Research in Personality,
I (1965), 132—137.

 

Streufert, S., and Streufert, S. C. ”The Effects of Concep—
tual Structure, Failure and Success on Attribution
of Causality and Interpersonal Attitudes." Technical
Report No. 7, Rutgers: The State University, Office
of Naval Research, 1967.

Streufert, S.; Suedfeld, P.; and Driver, M. J. "Conceptual
Structure, Information Search, and Information
Untilization." Journal of Personality and Social
Psychology, II (1965b), 736—740.

 

 

Suedfeld, P. "Conceptual and Environmental Complexity as
Factors in Attitude Change." Technical Report No.
14, Princeton University, Office of Naval Research,
1963.

., and Hagen, R. L. "Measurement of Information
Complexity: I. Conceptual Structure and Information
Pattern as Factors in Information Processing."
Journal of Personality and Social Psychology, IV
(1966), 233—236.

 

 

Suedfeld, P., and Streufert, S. "Information Search as a
Function of Conceptual and Environmental Complexity."
Psychonomic Science, IV (1966), 351-353.

 

 

 

Tannenbaum, R. "Managerial Decision-Making." Selected
Readings in Management. Edited by Fremont A. Shull.
Homewood, Illinois: R. D. Irwins, 1958.

Thomson, R. The Psychology of Thinking. Baltimore:
Penguin Books, Inc., 1959.

 

143

Toffler, A. Future Shock. New York: Random House, 1970.

 

 

Tuckman, B. W. "Personality Structure, Group Composition and
Group Functioning." Sociometry, XXVII (1964), 269—
487.

Turiel, E. "An Experimental Test of the Sequentiality of

Developmental Stages in the Child's Moral Judgments."
Journal of Personality and Social Psychology, III
(1966), 611-618.

 

Velasco, E. B. "Decision-Making in the Use of Family Money:
Case Studies of Five Families." Unpublished Master's
thesis, University of the Phillippines, 1964.

Wason, P. C., and Johnson-Laird, P. N., eds. Thinking and
Reasoning. Baltimore: Penguin Books, Inc., 1968.

 

 

Wilson, C. E., and Alexis, M. "Basic Framework for Decision.I
Journal Acad. Management, V (1962), 150-165.

 

Wright, D. J. Jeremy D. Finn's Multivariance-—Univariate and
Multivariate Analysis of Variance and Covariance:
A FORTRAN IV Program, Occasional Paper No. 8, Office
of Research Consultation, College of Education.
East Lansing, Mich.: Michigan State University,
1970.

APPENDICES

144

APPENDIX A

SAMPLE PRETEST AND POSTTEST

145

SAMPLE PRETEST

BACKGROUND INFORMATION

 

 

 

 

Student Number Class in School
Major GPA
Age Marital Status Number Children

 

 

Number of hours employed per week

 

Your occupation

 

Father's occupation

 

Mother's occupation

 

Number of students in your high school graduating class

Number of credits you are enrolled for this term

 

Check the courses you have previously taken:

HED
(or

HED
(or

FCS
FE

Number of hours sleep you have had in the last 24 hours

146

171
170)

143
140)

261
331

INSTRUCTIONS TO PARTICIPANTS

Part One: PARAGRAPH COMPLETION

 

On the following pages you will be asked to complete
certain Sentences and write a short paragraph.

On each page you will find the beginning of a sentence,
and your task is to complete it.

For example: I like....

When you are given the signal turn to Page 1. Complete
the sentence given and write at least three additional
sentences. You will be given 120 seconds. After 100
seconds I will say "Finish your sentence," and at 120.
seconds I will ask you to turn to the next page. Make
sure you complete your last sentences. There are 4
sentence completions in all.

Write your sentences as quickly but as clearly as possible.

Do not turn this page until you are given the signal.

147

148

When others criticize me it usually means....*

Rules....

When I am in doubt about a garment purchase....

When someone disagrees with my textile selections....

Confusion....

When a fabric purchase proves inappropriate....

 

*In the actual test situation, each sentence stem is on
a separate page.

II.

149

INSTRUCTIONS TO PARTICIPANTS

Part Two: PROBLEM SITUATIONS

 

 

On the following pages you will be asked to respond to
questions about two problem situations.

A brief description of each situation will be presented
and your task is to answer these questions about both
situations:

1) What factors would you consider in this situation
before recommending specific alternatives?

 

2) What recommendations would you make?

 

3) What other alternatives might you recommend?

 

4) If you were making this decision, how would you choose

 

among your generated alternatives?

When you are given the signal, turn the page. You will
have approximately 15 minutes to respond to the four
questions for each situation.

Please think through your answers and write responses

as clearly and fully as possible. Use the back of the
pages if you need additional space.

Do not turn this page until you are given the signal.

150

A-l. The Smith Family (husband, wife, son age 9, daughter
age 13) is considering how to redecorate their large
family room, perhaps re—upholstering a couch and lounge
chair, and selecting new carpeting and draperies. They
want to keep down costs and yet have an attractive room
that will hold up under constant use.

1) What factors would you consider in this situation before
recommending specific alternatives?

2) What recommendations would you make?

 

3) What other alternatives might you recommend?

 

4) If you were making this decision, how would you choose
among your generated alternatives?

 

151

B-l. The Smith Family (husband, wife, son age 9, daughter
age 13) is considering the purchase of an automobile.
They would like an all—purpose family car for routine
use and for travel, at moderate cost.

1) What factors would you consider in this situation before
recommending specific alternatives?

2) What recommendations would you make?

3) What other alternatives might you recommend?

4) If you were making this decision, how would you choose
among your generated alternatives?

 

SAMPLE POSTTEST

INSTRUCTIONS TO PARTICIPANTS

Part One: PARAGRAPH COMPLETION

 

 

On the following pages you will be asked to complete
certain sentences and write a short paragraph.

On each page you will find the beginning of a sentence,
and your task is to complete it.

For example: I like....

When you are given the signal turn to Page 1. Complete
the sentence given and write at least three additional
sentences. You will be given 120 seconds. After 100
seconds I will say "Finish your sentence," and at 120
seconds I will ask you to turn to the next page. Make
sure you complete your last sentences. There are 4
sentence completions in all.

 

 

Write your sentences as quickly but as clearly as possible.

Do not turn this page until you are given the signal.

152

153

When I am criticized about a textile purchase....*

When I am confused about which carpet to recommend....

When I am in doubt....

Parents... .....

When someone disagrees with me....

Nonserviceable textile selections....

 

*In the actual test situation, each sentence stem is on
a separate page.

II.

154

INSTRUCTIONS TO PARTICIPANTS

Part TWO: PROBLEM SITUATIONS

 

 

On the following pages you will be asked to respond to
questions about two problem situations.

A brief description of each situation will be presented
and your task is to answer these questions about both
situations:

1) What factors would you consider in this situation
before recommending specific alternatives?

 

2) What recommendations would you make?

 

3) What other alternatives might you recommend?

 

4) If you were making this decision, how would you
choose among your generated alternatives?

 

When you are given the signal, turn the page. You will
have approximately 15 minutes to respond to the four
questions for each situation.

Please think through your answers and write responses

as clearly and fully as possible. Use the back of the
pages if you need additional space.

Do not turn this page until you are given the signal.

155

A-2. You are assigned as a designer—consultant to make recom-
mendations on how to redecorate dormitory lounges.
Carpeting, furniture upholstery, draperies, etc. are to
be considered. University personnel want to keep costs
down and yet have an attractive room that will hold up
under constant use.

1) What factors would you consider in this situation before
recommending specific alternatives?

 

2) What recommendations would you make?

3) What other alternatives might you recommend?

 

4) If you were making this decision, how would you choose
among your generated alternatives?

 

156

B-2. A student couple (age 20 and 22) is seeking an apart-
ment to rent for the college school year. They are on

a tight budget, but want an apartment that is attrac—
tive and comfortable.

 

1) What factors would you consider in this situation before
recommending specific alternatives?
2) What recommendations would you make?

 

3) What other alternatives might you recommend?

4) If you were making this decision, how would you choose
among your generated alternatives?

 

APPENDIX B

PARAGRAPH COMPLETION TEST (PCT) SCORING

157

PARAGRAPH COMPLETION TEST (PCT) SCORING

For a discussion of the general background and theoret-
ical assumptions underlying the PCT and its scoring, the
reader is referred to Schroder, Driver, and Streufert (1967,
pp. 185-204) and Schroder (1971, pp. 240—273) in Personality
Theory and Information Processing, ed. by Schroder and Suedfeld.

The scOring manual includes a set of general Operations
for inferring the level of conceptual structure that gen—
erated the response. Responses are scored along a seven-
point scale, which represents a continuum from.low (score of
l) to high (score of 7) levels of integrative complexity.

Representative samples of the scoring of actual sub—
jects' responses are reproduced, with an analysis of the
response in terms of its structural prOperties, to indicate
how the scoring procedures operate. These examples were
taken from the Structural Scoring Manual (Phares and Schroder,
1969), developed to help train scorers.

 

 

Scoring of Responses to Stem: Rules...

a) Score of 1, indicating low complexity:

 

. . . are very boring but completely necessary. Many
rules and laws I don't enjoy, but I always obey them.
Rules are not made to be broken.‘

Scorer l - This is a 1. He only sees rules in one way.
Scorer 2 — I scored it 1. There is only one perspective.
Scorer 3 - I think this is a 2 because he qualified the

fact that he never breaks rules by saying that
they are a pain in the neck.

 

Scorer 4 - I scored it a 1. He closes down completely.
Discussion:
This is a score 1 response. The S gives no indication

of viewing rules in any way other than that of obedience.
The mere fact that he does not always enjoy the imposition
of rules is not a sufficient qualification to raise the
response to a 2."

b) Score of 6, indicating high complexity:

 

"'. . . are formed and amended through trial and error.
They weren't made to be broken but to change with changing

times and societies. Should revolve around one main
theme--moderation——and should be flexibly applied to sit-
uations.‘

158

159

Scorer l - It's a higher response all right, and I would
have scored it higher than a 4 except that the
S doesn't talk about the purposiveness of
rules. So, I gave it a 4.

Scorer 2 — This is really a sophisticated one—dimensional
judgment--they're right if they're moderate;
if they aren't, they must be changed which is
just another way Of saying 'break them.‘
Because the S uses the word, 'flexibly,‘ he has
Opened up a bit and so I scored it 2.

Scorer 3 - I disagree. It's definitely above a 2 except
that I'm just not sure how much above. A 4
or a 5 maybe?

Discussion:

 

Although it might be argued that 'moderation' in this
response gives indication of the purpose of rules (see
Scorer 1), there are enough other referents to establish
a higher score: a very high degree of conditionality; a
sense Of dynamism, change; a view that, rather than being
externally imposed standards, rules are formed and modi—
fied (and re-formulated) by an interaction between need
and circumstance. Since these processes are articulated
a bit more clearly in this response than in the earlier
response (see p. 4, Score 5), a score of 6, rather than
5, is given."

Scoring of Responses to Stem: When I Am in Doubt...

 

a) Score of 3, indicating medium-low complexity:

 

"'. . . I think carefully about the situation. I search

for various alternatives to resolve the doubt. If the

latter persists, I am somewhat disconcerted but I press

on.‘

Scorer l - This is a 2 score because the S sees the possi—
bility of more than one alternative.

Scorer 2 — Scored it 3——there is more than one alternative
stated.

Scorer 3 — I scored it 3 because there is evidence of

toleration Of the uncertainty as well as
alternatives expressed.

Scorer 4 - I was going to score it 3 because the S
expresses different perSpectives but then he
admitted that he was ‘disconcerted‘ and I felt
that would bring it back down to a 2.

b

v

160 -

Discussion:

There is a clear expression of different alternatives in
this response which is sufficient to score it 3. The
expression Of concern over an uncertain situation, per se,
is not a referent for lowering a score (see Scorer 4).
Only if such an expression were accompanied by verbal
evidence that the S was also "closing down" or warding
off the conflict would the scorer consider giving a lower
score."

Score of 4, indicating mid—point on complexity continuum:

. . I try to analyze the situation to the best of my
ability. I study the solutions to the problem and pick
a few Of the best solutions after comparing all of them.
Then I act on them in the best way I know how.‘

 

Scorer 1 - I scored it 3 because the S is seeing differ—
ent solutions.
Scorer 2 — I scored it 3 for the same reasons I stated

earlier; the S sees various alternatives and
chooses among them.

Scorer 3 - I think it is a 4. The S sees different per-
spectives and is doing some uniting Of them
when he starts talking about 'comparing.‘

Discussion:

As each of the scorers has already pointed out, there is
good evidence in this response of the S's ability to see
more than one alternative. He also indicates that he may
be engaging in a process Of combining these alternatives
('comparing all of them'). This, then, permits a score
of 4."

APPENDIX C

CODING MANUAL: DECISION TIME ORIENTATION

 

161

CODING MANUAL: DECISION TIME ORIENTATION

 

Decision time orientation is based on how ideas are perceived
in a time relationship. Time reference is identified by con—
text of statement as it relates to a time base. Grammatical
connotations, e.g., the relationship of nouns to verbs, and
adjectives to nouns, help to determine the context of the
statement.

 

Unit of analysis is the smallest component in a sentence
that has a time reference, whether it be a word, phrase,
clause, or sentence.

 

Numerical rating scale from zero to eight is designated, with
lower numbers representing greater orientation toward the
past and higher numbers representing greater orientation
toward the future.

 

Rating Scale

 

TimeO--Unclassified. NO distinguishable reference to time.

 

Timel-—Distant Past. References to experiences or events
which occurred years in the past; overall life experi—
ences. Age of respondent provides insight as to how
long ago experience might have occurred.

 

Examples: I experienced....as a child.
My mother, etc. used to....
Based upon my childhood, my past, etc.....
Previously....
Already, I have learned....

Time2--Immediate Past. References to experiences or events
which occurred within the immediate past, i.e., the
last year or two. Include those references to past
which only indicate past time, without referring to
events which took place a specified long time ago.

 

Examples: From experience....
It was my....(job, duty, etc.)
I have found....
I did....previously

162

163

Time --Past/Present. Combination of past and present refer-
ences and/or experiences. Present references based
on past experiences. References to life style, atti-
tudes, preferences which are presently held, but were
developed through past experiences.

 

Examples: From past experience I can now....
Considering tastes and preferences, I
recommend....

I value, I prefer....

Time4--Present. Indication of present behavior based on that
which is on—going but not habitual; statements which
denote on-going or present conditions. Feelings at
the present time. (Consider when recommendations are
carried out, not when they are made, since any recom—
mendation is made at present time.)

 

Examples: ....should be or do.
....are considered.
....are or is.
I think she should.... e
I like....
Right now the demands are....
Under these circumstances....
My friends presently....

 

 

 

They need....

Time5-—Present/Future. Combination Of present and future
references and/or experiences.
Examples: They wish, desire....

She, he, does want....
They plan....
....can do, can be.

Time -—Near Future. Predictive statements or explicit
expectation of fulfillment Of a "near" future state
of affairs, i.e., next month, next year references.
Statement as to what might happen in the immediate
future.

Examples: If she waits awhile....
The next time she....
Tomorrow, or next month....

By next year....

The job will probably be....

....would choose, would want, would like,
would need, etc.

....could do.

Perhaps, might, maybe, probably, etc....

Time

Time

7

8

164

-—Distant Future. Predictive statements about long-

 

range consequences: references to years in advance,
where one expects to be in distant future. References

to over a lifetime. When no near future time con-
straint is referred to, and implication is distant,
consider reference into distant future.

Examples: I expect....to happen someday.
....may result eventually.
I aSpire to....
....would anticipate.
....expect to last.
Over a long period....
In the long run....
I would select....for long-range durability.

—-Future/Past. Referents to future decision based on

 

past experiences. Indication of profiting from past-—
evaluating. Being critical of past and making future
projections.

Examples: From continued experience....
I would p1an....based on....(some past
referent).

Each codable unit receives one of the above numbers,

which are averaged by adding up the values and dividing by
the total "n" for a given situation. The following score
sheet is suggested.

165

 

mafia waflbxoucoz coflbmpcmflno
ommuo>¢ WWHWmepOB mmmno><

 

 

 

A
V

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

oEHB OHprOB

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mmmum><

Aw

Am

Am

AH

mamuOB mmmunm HO .mmsmau..ouoz >3 mmuoom oEHB HOQEDZ
coflumoso GOHmeSO
moonm

 

pmmgm acawoo

ZOHBfiBZMHmO MEHB ZOHmHUWQ

Im

coaumsuflm
OHprmusoz

I<

COHDOSDHm
OHHume

# humansm

APPENDI

CODING MANUAL:

166

X D

DECISION RULE

 

 

CODING MANUAL: DECISION RULE

Decision rule represents the methods by which alternative
courses Of action are evaluated, and how one alternative is
decided upon.

Unit of analysis is a sentence or group of sentences indi—
cating the method used in comparing alternatives. Evidence
Of rule may appear throughout protocol.

Numerical rating scale from zero to five is designated, with
lower numbers representing less complex rules than higher
numbers.

 

Rating Scale

Rule0-—Unclassified. No evidence of selection among alter-
natives, nor indication that one acceptable alterna—
tive is chosen. NO closure evident.

 

 

Rulel--Immediate Closure. One alternative or set of alter—
natives is focused upon and immediately grasped. The
process of decision is quick, and any analysis of
choices comes after the decision is made. The limits
allowable in making the decision are perceived as
fixed; no evidence of weighting or evaluating. 1

Examples: From the beginning, I thought or believed....
I can't think Of any other choice....
All the time I knew....would be best.
Clearly, the recommendation would be....

Rule2—-Satisficing. The advantages and disadvantages of
one alternative at a time are perceived and evaluated,
EEEording to a minimum set of goals and expectations;
the alternative is either accepted or rejected. The
first alternative which meets or exceeds the minimum
criteria is accepted. Often there is inferred
rationalization——the decision maker indicates that
the chosen alternative is not necessarily the best or
an optimum, but it is "good enough." There is an
indication of two categories-~acceptable and not
acceptable--and the first acceptable alternative is
selected.

167

168

Examples: I chose....because....
....a1ternative must be chosen because....
There would be some satisfaction with....
Due to these factors....I have decided....
Since....is not acceptable, I have elimi—
nated it and chosen....

I know it's not the best, but....

....is good enough.

I would compromise and choose....

....is just as good.

Rule --Lexicographic Ordering of Few Alternatives. Two or
three alternatives are ranked, on the basis of per-
ceived values or attributes or characteristics under
consideration. A11 perceived factors are weighed.
Closure is reached when the best alternative is chosen
from other alternatives. If two or more alternatives
are equivalent with respect to the most important
attribute, then they are ranked on the basis of the
second most important attribute. The key is some
indication of ordering of alternatives--weighing and
comparison must be evident.

 

Examples: My first recommendation would be....
My last recommendation would be....
....would be a far better choice than....
After considering....alternatives, I would
evaluate....
....is preferable (better, more desirable)
than....

Rule --Lexicographic Ordering of Many Alternatives. Same
factors considered as with Rule , except four or more
alternatives are indicated and then ranked.

Examples: This (fourth) undesirable alternative
would be last because....
....has more advantages than....

Rule —-SATISLEX. Alternatives are ranked on the basis of
values or attributes which are most important; then
next most important, etc., as with lexicographic
ordering. Then unacceptable alternatives are elimi-
nated, first on the basis of the most important char-
acteristics or attributes. After unsatisfactory
alternatives are eliminated, the first acceptable one
is selected (as with satisficing). Or, a satisficing
rule may be applied first, in which unsatisfactory
alternatives are eliminated immediately, with a weigh-
ing and evaluating of the remaining acceptable alter—
natives.

169

Examples: After considering....alternatives, I would
reject....and accept....
Since....wou1d be unacceptable, I would
rank....as follows.

Each decision situation receives a single rule rank, with
justification cited from the protocol for the value decided
upon. See rule coding sheet, which follows.

DECISION RULE

Coding Sheet

Subject #

 

Evidence of Rule in Protocol

 

 

Textile
Situation
A—

Rule

 

Nontextile
Situation
B—

Rule

 

 

 

 

 

“m1

CHIGAN STATE UNIV

IHIWIHIIIWIIWI

 

r513