IIIIIIIIII' II”;

«I I IE5“. Fl»!

retards?

CHI

lllglllll‘ggllllljlfl‘l‘.llllll

 

 

   

 

 

I.
is

Ill-[O I. o
. I t-‘II-I] P4 '-""9‘”
"'""'";_,‘ fi tug—ah";

t... o
"" a' 9'1 zit,

 

 

 

This is to certify that the
dissertation entitled

The Relationship Between Student Characteristics
And The Effective Use Of A Computer-Assisted
Career Guidance System

presented by

Alan J Farber

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Counseling Psychology

 

 

Wig/4M

ajor professor

Date July 22, 1986

 

MS U is an Affirmative Auction/Equal Opportunity Institution 0-12771

 

 

 

IV1ESI_J RETURNING MATERIALS:

Place in book drop to
LJBRARJES remove this checkout from
“ YOUY‘ Y‘ECOY‘d. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

”2.0 M?

h >4 M

$28
553061996

 

 

 

THE RELATIONSHIP BETWEEN STUDENT CHARACTERISTICS
AND THE EFFECTIVE USE OF A COMPUTER-ASSISTED
CAREER GUIDANCE SYSTEM

BY

Alan J Farber

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

School of Health Education, Counseling Psychology
and Human Performance

1986

"D

W >93

ABSTRACT

THE RELATIONSHIP BETWEEN STUDENT CHARACTERISTICS
AND THE EFFECTIVE USE OF A COMPUTER-ASSISTED
CAREER GUIDANCE SYSTEM

BY
Alan J Farber

Some researchers have suggested that certain
individuals are more likely than others to benefit from the
use of computer-assisted career guidance systems (CACGS).
The present study sought to examine the influence of
personal and demographic traits on the outcome of CACGS use.
The CACGS under investigation was the System of Interactive
Guidance and Information (SIGI). It was predicted that task
oriented participants, in comparison to people oriented
participants, would benefit more from the use of SIGI. It
was further predicted that a relationship would exist among
the three dependent variables under investigation. College
students were classified according to their Holland's
typologies (people or task orientation), sex, and previous
experience with CACGS. The outcome of SIGI use was measured
by three dependent variables: evaluations of SIGI (as
measured by an investigator-created rating scale), time
spent on the system, and post-SIGI levels of career
decisiveness (as measured by the Career Decision Scale).

Participants were 69 university undergraduate men and
women. Prior to using SIGI, the students completed a SIGI

User Information Form, Career Decision Scale, and Vocational

Preference Inventory. Upon completion of SIGI, they
completed the Career Decision Scale, and the SIGI Evaluation
Questionnaire. Regression analysis did not reveal any
statistically significant relationship between the three
independent and dependent variables. However, significant
relationships were found between two of the three pairings
of dependent variables: SIGI evaluations and post-SIGI
career decisiveness; and SIGI evaluations and time spent on

SIGI.

ACKNOWLEDGMENTS

I would like to acknowledge the help and support of my
dissertation committee, Drs. Linda Forrest, Gordon Williams,
and Casimer Heilman.‘ Special thanks to my dissertation
chairperson, Dr. Richard Johnson, for his scholarly advice
and genuine concern. The excitement I feel about career
counseling and research was largely fostered by Drs. Johnson
and Forrest. I admire their scholarship, and appreciate
their guidance.

I appreciate the cooperation and assistance of the
professional and support staff at the Counseling Center,
Brody Hall, Wonders Hall, and Learning Resource Center. My
deepest gratitude goes to Cheryl and Sam Baer for assisting
in the procurement of data, and Douglas Rosenthal for his
assistance in the analysis of data. I am especially
appreciative of the 69 students who graciously volunteered
to participate in this study.

This work is dedicated to Mary Theresa Gawrys - my

friend, wife, and heart's desire. Thanks Gaws.

ii

TABLE OF CONTENTS

List Of Tables OOOOCOOCOOOOOOOCOOCO00.0.00...0...... iv

CHAPTER I. INTRODUCTION 0.0.0.0...OOOOOOOIOOOOOOOOO 1.6
Introduction to the Problem
Importance of the Study

CHAPTER II. REVIEW OF THE LITERATURE ............... 7-57
Computer-Assisted Career Guidance Systems
System of Interactive Guidance and Information
Computer-Assisted Career Guidance Systems Research
Attribute-Treatment Interaction Research
Holland's Theory
Research Hypotheses
Exploratory Issues

CHAPTER III. METHODOLOGY ........................... 58-93
Research Design
Participants
Setting
Procedures
Independent Variables
Dependent Variables
SIGI Evaluation Questionnaire
Career Decision Scale
Time Spent on SIGI
Additional Forms and Instrumentation
Vocational Preference Inventory
SIGI User Information Form
Consent Form
Primary Statistical Analysis Procedures
Secondary Statistical Analysis Procedures

CHAPTER IV. RESULTS OOOOOOOOOOOOOOOOOOOOOOOO00...... 94-105
Primary Statistical Results
Secondary Statistical Results

CHAPTER v. DISCUSSION OOOOOOIOOOOOOOOOO0.0.0.0000... 106-131
Procedural Issues
Design Issues

Measurement Issues
Suggestions for Further Research

APPENDICES OO...O..00....0...OOOOOOOOOOOOOOOOOOOOOOOO 132-140

REFERENCES 0.0...0....O..0.000000000000000000000000.0 141-156

iii

LIST OF TABLES

Table Page

1 Means and Standard Deviations:
Three Dependent Variables by Type
(Tests Of Hypotheses 1-3) OOOOOOOOOOOOOOOOO 100

2 Stepwise Multiple Regressions:
Three Dependent Variables ................. 101

3 Means and Standard Deviations: Three
Dependent Variables by Contact With
a Career Counselor ........................ 102

4 Chi-Square Test of the Relationship
Between Orientation and Contact With a
career counselor O0....OOOOOOOOOCOOOOCOO... 103

5 Mean Scores on Three Dependent Variables
When Artistic Types are Included
and EXCIUded .0...OOOOOOOOOOOOIOOOOOO0.0... 104

6 Demographics of Participants,

Non-Completers, and Comparison Group
SIGI users OOOOOOOOOOCCOOOOCOOOOOOIOO00.... 105

iv

CHAPTER I

INTRODUCTION

Introduction to the Problem

 

Empirical evidence and common sense alike suggest that
vocational clients' personal characteristics may influence
their reaction to, and the effectiveness of, particular
treatment interventions. Three subject areas have received
considerable attention in the psychology and vocational
counseling literature: (a) Attribute-treatment interaction
research directed at determining whether the effects of
different therapeutic methods are influenced by the
cognitive or personality characteristics of the client, (b)
Studies using Holland's Occupational Themes to categorize
clients according to stated vocational preferences, and (c)
Outcome studies examining the effectiveness and impact of
computer assisted career guidance systems. The present
attribute-treatment interaction study represents an attempt
to incorporate Holland's Occupational Themes in assessing
the effectiveness of, and user reactions to, a computer
assisted career guidance system.

Computers have been used to assist career counselors in
providing up-to-date career information and guidance for
approximately 20 years. Research on such computer assisted

career guidance systems (CACGS) has provided evidence that

2
the systems are generally well received by counselors and
users alike. Numerous studies have compared the
effectiveness of “traditional" (non-computer assisted)
vocational counseling to the use of computer-based systems
(English, 1974; Glaize & Myrick, 1984; Grisham, 1978; Maola
& Kane, 1976). Other outcome investigations have measured
the comparative effectiveness of two computer assisted
programs (Kapes, Borman, & Kimberly, 1985). Because the
studies have involved a variety of computerized systems,
traditional counseling approaches, and criterion measures,
findings have been equivocal. However, in general, results
of such studies indicate that the computerized systems are
favorably evaluated by users and counselors, provide useful
information, increase the users' vocational decidedness, and
encourage appropriate vocational planning and decision
making.

Far less is known about the differential effectiveness
of CACGS among various user populations. Only one study
(Melhus, Hershenson, & Vermillion, 1973) has been
specifically devoted to assessing the "best fit" between
different kinds of users (according to vocational counseling
readiness) and different types of treatment (a computerized
career information delivery system versus "traditional”
vocational counseling). The findings of this study strongly
suggest that the efficacy of CACGS may differ according to
users' personal characteristics.

Several anecdotal reports have appeared in the

3

literature describing individuals who are likely users,
non-users, beneficiaries, and non-beneficiaries of CACGS
(Haring-Hidore, I984; Walz, 1984). The goal of the present
study is to provide further empirical support for the notion
that users' personal traits differentially influence the
effectiveness of CACGS. The CACGS under investigation is
the System of Interactive Guidance and Information (SIGI), a
widely used and researched program designed to provide
career guidance and occupational information. Specifically,
it is predicted that SIGI users' degree of “task
orientation” and "people orientation" will differentially
influence their evaluations of SIGI, time spent on SIGI, and
post-SIGI career decidedness. Task oriented individuals
tend to gravitate toward, and function most effectively
among, things rather than people. The reverse is true of
people oriented individuals. Because SIGI is a
pre-programmed, systematic, and largely non-social activity,
task oriented users would be expected to score higher than
people oriented users on the three aforementioned dependent
measures. Orientation toward tasks or people will be
determined through the use of the Vocational Preference
Inventory, a personality measure and interest inventory
designed to categorize people according to Holland's six
Occupational Themes.

In addition, it is predicted that a positive
correlation will exist among the three criterion measures

under investigation. For example, a user spending an above

4
average amount of time on SIGI would be expected to
favorably evaluate the system, and demonstrate a high level
of career decidedness following its use. SIGI users' degree
of career decidedness will be assessed by the Career
Decision Scale. An investigator-created questionnaire will
be employed to determine user ratings and time spent on

SIGI.

Importance of the Study

It is, or should be, the goal of every career counselor
to match treatment approaches to clients according to their
personal traits and needs. The popularity of career
development theories and vocational maturity measures attest
to researchers' and practitioners' recognition of individual
client differences in terms of vocational identity,
develOpment and treatment needs. Unfortunately, however,
vocational counseling and guidance practices often minimize
or ignore the unique client-specific traits that may be
influencing the efficacy of the treatments in use. The two
major reasons for such a de-emphasis on attribute-treatment
interactions in career counseling appear to be: (a)
Insufficient empirical evidence supporting the need to
contour treatments according to client traits, and (b)
Insufficient funds, resources, and personnel to provide the
needed individually client-tailored interventions.

The present study addresses both of the aforementioned

5
issues. First, it represents an attempt to demonstrate that
various client personal and demographic traits influence the
outcome of CACGS use. Second, it involves a potentially
helpful, seemingly nonthreatening, enjoyable, and
cost-efficient career guidance procedure that may serve to
alleviate some of the burden of understaffed and overworked
career guidance programs, while providing a valuable service
to clients.

Despite the relative dearth of empirical evidence to
support their use, CACGS have been overwhelmingly embraced
by the career guidance community. Nonetheless, warnings
have sounded in the professional literature against the
indiscriminant use of CACGS (Cairo, 1983; Wrenn, 1970).

Some fear that CACGS's favorable user ratings,
cost-effectiveness, and minimal need for human intervention
may result in their misuse, much in the fashion of the
common misuse of interest inventories with inappropriate
clientele or by unqualified personnel. Furthermore, global
generalizations have been made concerning the merits and
outcomes of CACGS despite questionable research practices
and vast differences between the various systems' formats,
contents, purposes, and intended users.

The present study represents one in a recent and
rapidly increasing attempt to accurately and systematically
evaluate this relatively new career counseling and guidance
procedure. The use of Holland's typologies, undoubtedly the

most well accepted and empirically sound vocational

6
classification system, provides further credence and support
for the present attribute-treatment interaction study.
Since Holland's typologies are universally recognized and
employed by career counselors, the findings of the present
study may have direct applicability and relevance to
everyday career guidance procedures. For example,
counselors may chose to offer SIGI alone to Realistic and
Investigative students, while offering it in conjunction
with group discussions with Social and Enterprising
students. Finally, such a study may serve to stimulate
further attribute-treatment interaction CACGS research
involving other computerized systems and/or other

independent and dependent variables.

CHAPTER II

REVIEW OF THE LITERATURE

Computer Assisted Career Guidance Systems

Computer assisted career guidance systems (CACGS) are
currently one of the most widely used applications of
computer technology in the counseling field (Sampson, 1983).
Authors in the area of career counseling have predicted
computer-based approaches to be one of the priorities of the
19708 and beyond (Campbell, 1974). In 1984, Katz estimated
that approximately 1,000 post-secondary institutions were
using the best-known computer—based guidance and information
systems. Originally run solely on mainframes, most programs
are now available in microcomputer versions, making them far
more accessible and affordable to career counselors and
guidance personnel. Although no current estimate of the
number of microcomputers in career guidance exists, their
use is expanding rapidly. For example, according to Lambert
(1984), few if any microcomputers were in use in Wisconsin
in 1980-81. In 1981-82, about 25 sites used microcomputers,
and in 1982-83 over 125 sites had career information
available on microcomputers. Their proliferation is
primarily due to the computer's ability to provide ready
access to extensive amounts of updated information in a
convenient and relatively inexpensive fashion.

7

8

CACGS are programs in which a user communicates
directly, by means of a typewriter or visual display screen
terminal, with the computer's prestored text and files to
receive vocational and educational information designed to
assist with personal career decision making (Rayman,
Bryson, & Bowlsbey, 1978). Computer assisted guidance is
currently available to a variety of users in a number of
settings, including secondary schools, two- and four-year
colleges, technical schools, state employment offices,
public libraries, correctional institutions, and private
career counseling agencies.

Sampson and Pyle (1983) note that the use of
computerized systems has increased as a result of the
realization that there are aspects of the helping process
that can be more effectively handled by a computer than a
counselor. Growth of CACGS continues due to the increasing
demands on inadequate numbers of career counselors to have
relevant, accurate, and up-to-date information available to
increasing numbers of students (Kapes, Borman, & Kimberly,
1985). Computers have been recommended as a means of
freeing the counselor from the routine tasks of information
dissemination (Minor, 1970) and ”search-retrieval"
activities (Cooley, 1964). White (1981), in pointing out
that the computer is not competition for the counselor but
rather an aid to the completion of mundane tasks, notes
that ”people will matter more not less" (p. 54). The

Computer-based program affords the counselor the greater

9
opportunity to focus on the more affective and
interpersonal aspects of career choice (Harris, 1972;
Jacobson & Grabowski, 1981; Lackner & Feller, 1984). It
has been observed that counseling psychologists and
counselors-in-training tend to dislike and/or actively
avoid career counseling (Pinkney & Jacobs, 1985). If the
computer proves capable of handling the more bromidic
tasks, career counseling may be viewed more favorably by
professionals in the field.

In recent years technological advancements have allowed
CACGS to provide guidance and teaching services previously
performed by counselors. The capability of CACGS, and the
point at which human intervention becomes necessary, has
been the subject of much debate in the career counseling
literature. Some studies have suggested that CACGS alone
are as effective as, or more effective than the counselor
at providing occupational information (Jones, 1976; Maola &
Kane, 1976). However, the consensus is that CACGS are best
used to supplement and enhance, rather than replace, the
interpersonal career counseling process (Maze 8 Perlmutter,
1983). Harris (1971) notes that guidance programs should
strive to achieve an optimum blend of human and machine to
avoid the overreliance upon either counselors or computers.
Similarly, Wrenn (1970) expresses fear for the dangers of
both nonuse and overuse of computers.

Johnson (1983) has identified circumstances in which

counselor intervention is essential when individuals use a

10
computer-based guidance system: (a) When the computer
program raises anxieties, (b) When the user has difficulties
or concerns regarding computer technology, (c) When the user
regards the computer as an authority, and (d) When
debriefing the user after computer use to correct
misunderstandings.

Harris-Bowlsbey (1983) has listed the specific
capabilities that computerized systems offer in meeting
career guidance needs as: (a) Arousing career awareness of
the need to plan, (b) Providing a broad awareness of
alternatives without the screening of Options based on
non-relevant client characteristics, (c) Teaching a process
of decision making, (d) Providing up-to-date and easily
accessible banks of information, and (e) Providing
systematic treatment with individualization.

Although the systems vary widely in terms of purpose,
audience, and content, all appear to be directed toward
persons in the exploratory life stage (Myers, 1978). Myers
claims that all CACGS address the following three realms of
user awareness: (a) The realm of self in which users are
encouraged to consider a variety of data about their own
characteristics, (b) The realm of opportunity in which users
are permitted to manipulate information about courses,
majors, schools, jobs and positions, and (c) The realm of
prediction in which users are guided in projecting their
selves, as they now see them, into some selected

Opportunity. Accordingly, he claims that the three

11
questions to be answered by the user are, respectively: "Who
am 1?," "What is there that I might do in the future?,” and
"What are my chances of doing that thing satisfyingly and
satisfactorily?" (p. 53).

Myers claims that CACGS facilitate career development
by (a) teaching that decision making is a rational process,
(b) providing information about occupational opportunities,
(c) increasing self-awareness; and (d) providing and
encouraging exploratory activities.

The content and format of the numerous systems are
varied, but commonly include one or more of the following
components (Maze and Cummings, 1982): personal review and
assessment (some using vocational inventories such as the
Strong-Campbell Interest Inventory), decision making
(identifying options, gathering information, assessing
desirability of options, assessing risks, selecting goals),
planning (assessing needed skills, identifying training
opportunities, identifying resources for job openings,
outlining plans), and implementation.

Several means of classifying CACGS have been proposed.
Rayman and Bowlsbey (1977) have used a three generation
classification system of computer systems. The first
generation consists of batch information storage and
retrieval systems. In this type of system, the user
completes a questionnaire on which the desired
characteristics of a school or occupation are indicated.

The questionnaire, along with many others, is sent to a

12
central location to be "batch processed." A print-out is
then returned to the user, listing occupations or schools
that possess the combination of characteristics desired by
the student. Maze and Cummings (1982) contend that such
batch-process instruments do not qualify as CACGS because
they fail to allow clients to directly interact with the
computer and to change responses after observing their
impact on the occupational selection process. However,
Clyde (1979) reports that such a batch-process system,
despite its inability to directly communicate with users and
provide immediate feedback, ”assists individuals in the
decision-making process by using a computer to search large
data files and identify options with predetermined
characteristics" (p. 8).

Second generation systems are described by Rayman and
Bowlsbey (1977) as on-line information storage and retrieval
systems (examples include Educational and Career Exploration
System, Guidance Information System, and Oregon Career
Information System). Such systems allow the user to be in
direct communication (on-line) with the computer by means of
a phone line or cable and terminal devices, and offer
user-computer interaction based on both the users' responses
and the computer's stored files. Users have the opportunity
to alter responses and note the subsequent effect on the
computer-supplied information.

Third generation systems are on-line information

systems that deliver guidance content beyond career

13
information (examples include DISCOVER and the System of
Interactive Guidance and Information). In addition to
providing the aforementioned on—line services, these CACGS
add content and functions traditionally undertaken in
one-to-one or group career counseling. For example, Rayman
and Bowlsbey (1977) claim that a systematic career guidance
system such as DISCOVER should include the following
components: (a) Selfeinformation, including values,
interests, and competencies; (b) The systematic exploration
of occupations; (c) Teaching of decision making; (d)
Relationship of self-information to occupational
alternatives; and (e) Informational assistance with
implementation of choice. In more recent years, both the
second and third generation systems have been developed for
use on microcomputers. Updated information, rather than
becoming available on-line, is periodically provided by the
CACGS developers via newly distributed diskettes.

Clyde (1979) has developed a continuum of leading
on-line systems, ranging from those that solely provide
information access and retrieval (Rayman and Bowlsbey's
second generation), to those that emphasize guidance and
teach a decison making or valuing process (third
generation). Similarly, the report of the National
‘Vocational Guidance Association's Commission on
(Zomputer-Assisted Systems (Harris, 1971) distinguishes
loetween indirect inquiry systems (first generation) and

<3irect inquiry systems (second and third generation).

14

Penn (1981) classified the direct inquiry systems into
two broad categories. Occupational Information Systems
store and retrieve data as well as provide educational and
occupational information. Career Guidance Systems, in
addition to offering these capabilities, teach a career
decision making or valuing process. Maze and Cummings
(1982) note, however, that all CACGS contain both guidance
and information functions.

The late 19603 saw the simultaneous development of
several computerized guidance systems throughout the United
States and Canada. The number of existing CACGS is
difficult to estimate, as obsolete and/or unwieldy systems
disappear from use, and new systems are developed and
introduced. In 1974, Harris reported that approximately 30
systems became operational since 1966. A decade later, Herr
and Cramer (1983) cited the same number. Several of the
major systems will be briefly described, followed by a more
detailed description of the System of Interactive Guidance
and Information.

The Education and Career Exploration System (ECES) was
developed in 1966 with funds from IBM, and has undergone
several subsequent revisions. ECES consists of six
Subsystems (Awareness, Search, Explore, Experience, Plan,
iand Placement) which allow students to consider their own
educationally and vocationally relevant characteristics,
Obtain career related information, learn decision-making and

employability skills, and receive career placement

15
assistance. This system is still in use in the Genesee
Intermediate School District in Flint, Michigan.

The Computerized Vocational Information System (CVIS),
also developed in the late 19603, was designed to provide
students with a means for exploring information about
themselves, and their educational and occupational interests
and Opportunities (Wilhelm, 1978). It provides occupational
information based on Roe's (1956) Occupational
Classification System. Among the content of CVIS is
information about occupations, apprenticeships, local job
Opportunities, technical and specialized schools, two and
four-year college programs, and military opportunities. The
system is also capable of handling school registration and
schedule changes.

Career Information System (CIS) was developed by
McKinlay in 1969 at the University of Oregon under a
Department of Labor grant. The purpose of the system is to
provide accurate, locally-relevant career information to
individuals who make personal career decisions (McKinlay &
Franklin, 1975). Unlike other systems described herein, CIS
emphasizes local information and is not marketed directly to
schools and agencies on a nationwide basis. Instead,
National CIS was developed to coordinate the development of
systems on a local site-by-site basis.

The Information System for Vocational Guidance (ISVD),
<ieveloped by Tiedeman and O'Hara in the late 19603,

attempted to simulate a dialogue between a career counselor

16
and a client. However, the system proved to be quite
unwieldy and the concept too advanced for the available
technology. In 1971, several of the ISVD collaborators
designed an inexpensive information storage and retrieval
system called Guidance Information System (618). According
to the GIS Resource Kit (1981), the intention of the system
is to "close the communication gap between the facts and
those who need to know them for effective decision-making"
(p. 47). Maze and Cummings (1982) report that due to its
inexpensiveness, flexibility, and simplicity Of use, GIS is
currently used at more sites and runs on more makes of
computers than any other computer-based guidance system.
GIS is typically used for only 5-10 minute periods by
students in need Of specific career-related information on a
national or local level (e.g., vocational and technical
programs, financial aid, apprenticeships).

The Computerized Heuristic Occupational Information and
Career Exploration System (CHOICES) was developed in Canada
in 1976 by the Department of Manpower and Immigration, and
is currently in use in both Canada and at least 20 states in
the United States (Pinder & Fitzgerald, 1984). CHOICES
possesses both an information and guidance function, having
been classified by Jacobson and Grabowski (1982) as an
<3nrline information system, and by Clyde (1979) as falling
141 the middle of the information-guidance continuum. The
System includes a 26-page guidebook that must be completed

Prior to beginning the first of four access routes. Its

17
first component (Explore) provides career guidance, whereas
the subsequent three (Specific, Compare, and Related)
provide occupational information.

DISCOVER was developed by Harris-Bowlsbey in 1976
through the sponsorship Of IBM. It is an eclectic system
combining Holland's Self-Directed Search, American College
Testing Program's ”World Of Work Map," Katz's work values,
and Harris-Bowlsbey's concept of systematic guidance.
DISCOVER is a highly comprehensive system composed of 21
modules which contain didactic material, several
self-assessment inventories, and extensive information
files. Versions have been designed for both the secondary
school and college/adult level. DISCOVER II, released in
1982, is an abbreviated version designed for use by re-entry
students and older adults.

In 1976, the National Occupational Information
Coordinating Committee (NOICC) was created to facilitate
interagency cooperation among the Education and Training
Administration, the Bureau of Labor Statistics, the 0.8.
Office of Education, and the National Center for Educational
Statistics. Similar committees have been created on a
state-by-state basis, and in 1980 fourteen states were given
grants to develop Career Information Delivery Systems
(CIDS). Since that time, dozens of states and regions have
adopted one of the major CACGS to form localized systems.
For example, as of March, 1982, 17 localized systems have

adopted CIS as their base system, 18 use GIS, four use

18
CHOICES, and two use DISCOVER. In addition, the Departments
of Education in several states have developed or financed
computerized systems which focus upon statewide and/or
regional occupational information (e.g., the Colorado Career
Information System, and the Michigan Occupational
Information System).

In recent years a number of software packages have been
introduced for use by high school and college students,
students' parents, re-entry students, adults, and
counselors. These microcomputer packages have been
classified by Bellotto (1983) as falling into one of six
groupings: self-assessment, self-assessment and career
information, educational information, administrative
systems, and test preparation. In addition to DISCOVER II
and System of Interactive Guidance and Information (SIGI),
the self-assessment and career information programs include
Career Directions, Coordinated Occupational Information
Network Career Search, ENCORE, Micro-Quest, Micro-Skills:
Eureka Skills Inventory Self-Scoring System, Holland's
Occupations Finder, Systems Design Associates' Computerized
Career Assessment and Planning Progam, and Peterson's Career
Planning Service.

Because of differences in the various programs' format,
content, and audiences, and problems associated with the
research that has been undertaken with CACGS (see following
section on CACGS Research), it is difficult to make

definitive statements concerning the various systems'

19
effectiveness and impact on users . However, tentative
conclusions have been drawn which suggest the following
(Cairo, 1983; Clyde, 1979; Harris, 1974; Maze & Cummings,
1982; Sampson, 1984):

1. Users acquire relevant information about
educational and occupational alternatives.

2. CACGS promote a greater awareness of the need for
planning.

3. CACGS increase users' concern with vocational
choice, amount Of exploratory behavior, and time devoted to
thought and conversation about career planning.

4. Users Often increase the appropriateness of their
occupational preferences, making them more congruent with
personal attributes.

5. Users report that CACGS are understandable, helpful
in facilitating career development, and enjoyable to use.

6. Users generally report that they prefer to Obtain
career-related information from a computer than from other
traditional sources.

7. Career counselors and parents of student users
generally welcome CACGS as valuable additions to career

guidance services.

System of Interactive Guidance and Information

The present study utilized SIGI as it was made

available to students in four locations at Michigan State

20
University in East Lansing, Michigan during the Spring,-
Summer, and Fall quarters of.1985. SIGI was developed in
1967 by Martin Katz, a senior researcher for the Educational
Testing Service. Since that time it has been field tested
throughout the United States and refined with the aid of
funding from the Carnegie Foundation, National Science
Foundation, and Kellogg Foundation. SIGI was designed for
use with two-year and four-year college students and is in
use in over 150 institutions.

According to Maze and Cummings (1982), SIGI is
"probably the most theoretically pure computer-assisted
career guidance system" (p. 79). In a series of
publications that preceded the introduction of SIGI, Katz
(1963, 1966) expounded upon his ”model of guidance for
career decision-making," a departure from the traditional
trait-and-factor theory Of occupational choice. Katz
contends that a humanistic process requires that individual
values and manpower needs be assessed by the individual in
the process of selecting an occupation. In subsequent
articles, Katz (1969, 1974) presents SIGI as the working
representation of this model.

According to Katz (1981), the main purposes of SIGI are
"to increase students' freedom of choice, to develop their
understanding of the elements involved in choice, and to
improve their competence in the process of making informed
and rational career decisions“ (p. 3). The two major

premises upon which SIGI Operate are: (a) The individual

21
must clarify and examine his or her values as they relate to
occupational choice, and (b) Having recognized the role
values play, rational and structured decision making can
occur.

Whereas conventional vocational counseling focuses on
the domains Of Interests and Aptitudes, Katz advocates the
inclusion of Values as a third component. He has identified
ten ”Occupational Values” that are designed to help students
determine those occupational rewards and satisfactions that
are most important to them. They consist Of High Income,
Prestige, Independence, Helping Others, Security, Variety,
Leadership, Leisure, Field of Interest, and Early Entry.
Clarification of these values proportedly facilitates
increased self and occupational knowledge, and improved
decision making skills.

After logging on, and following a brief introduction,
the user progresses through each of six subsystems. The
VALUES subsystem has the student examine the ten
occupational values and weigh the importance Of each. The
LOCATE subsystem has the student supply specifications for
five values at a time and receiving a list of occupations
that meet the specifications. The COMPARE subsystem allows
the student to ask questions and receive specific
information about occupations of interest. The PREDICTION
subsystem has the student supply probabilities of getting
various grades in key courses of programs that prepare for

specific occupations. The PLANNING subsystem identifies the

22
steps to be taken to prepare for an occupation. The
STRATEGY subsystem evaluates occupations in terms of the
rewards they Offer and the risks of trying to enter them.
After proceeding through the six subsystems, the user is
free to return to any system as he or she sees fit.

SIGI is essentially a didactic system which provides
information about the client (VALUES and PREDICTION
subsystems), about occupations and preparation for entry
(COMPARE and PLANNING subsystems), and about the
relationship between clients and occupations (LOCATE and
STRATEGY subsystems). As is the case in the proposed study,
the PREDICTION subsystem is Often excluded from use by
schools' guidance departments and counseling centers.

‘SIGI contains write-ups of 198 occupations covering
approximately 300 titles. Because it is designed for use
with college students, this number is smaller than most
other CACGS geared for use with high school rather than
college students. (In comparison, GIS Offers 875 primary
occupational listings with reference to approximately 2,500
related jobs). However, a great amount of detail is
available for each occupation in the form of 28 topics such
as job description, starting and average salaries, and
percentage of women within the field. In contrast to some
other CACGS, SIGI does not provide local labor market
information, job search assistance, or military information.
Also, unlike other systems, SIGI does not require interest

or personality inventories, or the assessment of abilities.

23

CACGS Research

 

In Holland, Magoon, and Spokane's (1981) review of
career interventions, research, and theory, only a single
reference is made to CACGS despite their widespread use for
over a decade. In contrast, two journals have more recently
devoted entire issues to the role of computers in

counseling: The Counseling Psychologist, Vol. 11, 1983; and

 

Journal of Counseling and Development, Vol. 63, 1984. This

 

reflects the burgeoning interest in, and implementation of
CACGS among counselors, educators, and researchers.
Nonetheless, after almost 20 years of use "little research
has been conducted on the effects and impact Of computers on
career counseling" (Pyle, 1984, p. 143). Furthermore, those
studies that have been undertaken provide little support for
their effectiveness in helping people plan their careers or
make better career decisions. Relatedly, Maze and Cummings
(1982) note that it remains unclear which services computers
can perform and at what point a counselor is needed. Thus,
it would seem essential, as suggested by Sampson (1977),
that counselors actively engage in well designed research
projects to assess the value of CACGS in the process of
career counseling. A review of CACGS research follows.
Numerous studies (Chapman, Katz, Norris, & Pears, 1977:
Nickerson & Davis, 1983: Wolff, 1976) have used investigator
or publisher created questionnaires to assess users' and/or

counselors' attitudes toward particular computerized

24
systems. In a typical study, Savin (1978) found that 73% of
245 high school students liked using DISCOVER and 56%
recommended it to their friends. Ryan, Drummond and Shannon
(1980) list the following self-reports among GIS users
ranging in age from 10-20: they learned a great deal about
occupations (75%); they reported the system helped them
confirm career plans they already had (41%): they reported
GIS helped them clarify educational plans they needed to
make (40%): they learned how to explore occupations (37%);
and they learned more about their interests as they relate
to career planning (45%). Although great differences among
computer programs and client populations exist, in all such
descriptive studies the majority of users have reported
satisfaction and generally positive overall impressions
(Clyde, 1979). The majority of extant CACGS studies appear
to be solely or primarily concerned with user ratings of the
systems. Cairo (1983), among others, warns that if CACGS
are to represent a significant part of the delivery of
counseling services, "evaluations of their impact must go
beyond determining their popularity” (p. 55).

A second common dependent variable in CACGS outcome
research involves the assessment of the users' level of
career maturity or career development before and after
exposure to one or more Of the computer-based systems. Such
studies (Bowlsbey, Rayman & Bryson, 1976; Devine, 1975:
Drake, 1978: Traver, 1977) have relied upon such instruments

as the Assessment of Career Development (ACT, 1974), Career

25
Development Inventory (Super, Thompson, Lindeman, Jordaan, &
Myers, 1981), and Career Maturity Inventory (Crites, 1973)
to measure the participants' level or stage of career
develOpment.

Kapes et a1. (1985) performed a comparative evaluation
Of SIGI and DISCOVER. Effectiveness was measured with pre-
and post- measurements on four instruments designed to
assess various aspects of career development: Career
Development Inventory, Career Decision Scale (Osipow,
Carney, Winer, Yanico & Koschier, 1976), Survey of Career
DevelOpment (Garis, 1982), and the Self-Assessment Of
Confidence and Progress in Educational/Career Planning
(Garis, 1982), an investigator-created evaluation inventory.
Although findings were complicated by simultaneous student
involvement in a career planning course, the results
generally indicated moderate increases in career development
indices following the use of both systems, with no
differences found between the two systems. As is typical of
CACGS studies, predominantly positive evaluations were
reported for both SIGI and DISCOVER.

Pyle and Stripling (1976) compared the adjusted Career
Maturity Inventory Attitude Scale posttest scores of a
control and SIGI treatment group consisting of community
college students. It was found that the SIGI users'
posttest scores were significantly different from those of
the control group, indicating greater levels of career

maturity. Using the same computer-based system and

26
dependent measure, Devine (1975) found no significant
posttest differences between treatment and control subjects
drawn from a community college population. It is noteworthy
that unlike Devine, Pyle and Stripling included group
orientation sessions prior to the use of each of the five
subsystems, and debriefing sessions following the completion
of each subsystem. This would suggest the possibility that
the human intervention positively influenced the students'
career maturity scores.

Cochran, Hoffman, Strand, and Warren (1977) used
Harren's Vocational Decision-Making Checklist (Harren, 1966)
to assess changes in decision-making stages of undergraduate
SIGI users. Following three 1-hour interactions with SIGI,
the treatment group, in comparison to the control group,
revealed significant positive changes on pre-post measures
of decision making stage related to choice of college major.
No significant changes were found in pre-post
decision-making stage measures related to vocational choice
or to locus of control of reinforcement, as measured by the
Locus of Control Scale.

Numerous studies have compared the outcomes of
computer-based systems with more “traditional” career
counseling approaches (English, 1974; Grisham, 1978: James &
Smith, 1972: Ross, 1971; Sampson & Stripling, 1979).
Reardon, Bonnell and Huddleston (1982) compared student
satisfaction ratings for CHOICES and the Self-Directed

Search (Holland, 1979), a paper-and-pencil career planning

27
guide. Seventy-five undergraduates who voluntarily
completed both treatments indicated a strong preference for
the computer-based system, although both programs received
overall positive ratings.

Glaize and Myrick (1984) compared the effect of three
career guidance interventions and a control group experience
on 120 high school students. The interventions were a
Vocational Exploration Group (VEG) only (Daane, 1972),
DISCOVER only, VEG and DISCOVER, and a control group. The
dependent measures were (a) the Career Maturity Inventory
Attitude Scale, (b) the Career Decision Scale, (c) a count
of the number of visits made to the school's career resource
center during a two week period following the interventions,
and (d) an investigator-created measure of attitudes
concerning self, school, and the world of work. The Career
Maturity Inventory and Career Decision Scale scores improved
significantly more for each Of the interventions when
compared to the control group, but no inter-treatment
differences were revealed. No significant differences were
found among the four conditions on the latter two criterion
measures.

Similarly, Garis (1982) compared the Career Development
Inventory gain scores Of a DISCOVER-only, Individual
Counseling-only, Individual Counseling and DISCOVER, and a
wait-list control group. The results, which typify career
counseling outcome research in general (Fretz, 1981),

revealed that all treatment groups scored significantly

28
higher on the questionnaire than did the control group, with
no significant differences among treatment groups. In
contrast, Maola and Kane's (1976) high school student CVIS
users achieved significantly greater gains on Assessment of
Career Development scores than did both non-CACGS career
counselees and control group members.

Risser and Tulley (1977) randomly assigned college
students to one of two groups: SIGI plus traditional
testing-oriented counseling, and traditional career
counseling and testing. The results, based on counselor
ratings, showed that the group provided with the use of SIGI
scored significantly higher on various indices of career
development.

In a study involving non-high school graduates, Penn
(1981) examined the differential effects of the use of
DISCOVER alone or in conjunction with one Of three other
career exploration strategies. The three additional
interventions were (a) the presentation of ”lifestyle"
films, (b) an interview procedure entitled “Psychosocial
Visit Exploration," and (c) ”structured traditional career
counseling" which included administration and interpretation
of the Career Assessment Inventory and the use of the
Occupational Outlook Handbook and the Dictionary of
Occupational Titles. DISCOVER was offered alone and in
conjunction with the first and second strategies.
Participants in the third strategy were considered a control

grOup. Using the Assessment Of Career Development and the

29
Assessment Of Career Decision Making (Harren, in press) the
results showed no differences between the treatments for the
following vocationally-related dependent variables: career
planning knowledge, decision making skills, career decision
making style, career planning involvement, and knowledge
regarding occupational information. However, the author
discovered several age related differences. For example,
DISCOVER was more effective in providing career planning
knowledge to older clients (22-24 years Old) when used in
conjunction with a psychosocial career exploration strategy
than was the traditional career counseling program.

Several studies, including Glaize and Myrick (1984) and
Kapes et a1. (1985) described above, have used measures of
career decidedness in addition to indices of career
development to determine the effectiveness of CACGS. Using
the Career Decision Scale and the Assessment of Career
Decision Making (Occupational Decision Making Subscale only)
as pre- and posttest measures with undergraduate and
graduate volunteers, Pinder and Fitzgerald (1984) found
significant differences between a CHOICE and control group's
adjusted posttest indecision scores on both scales.

Such studies that compare the effectiveness Of career
interventions such as "traditional counseling," ”group
counseling,“ or “individual counseling" have been criticized
for their failure to describe the nature of the
interventions under investigation (Fretz, 1981). For

example "traditional career counseling” may or may not

30
include the use Of interest and aptitude inventories, the
teaching of decision making skills, a discussion of personal
values, etc. The approach may be highly eclectic, didactic,
client-centered, cognitive-behavioral, structured or
unstructured. Similarly, although the internal structure Of
programs such as SIGI are standardized, investigators Often
fail to consider the potential contributory impact of such
factors as orientation and debriefing sessions, the absence
or presence of counselors or support personnel, the required
or recommended use of paper-and-pencil CACGS-related
exercises or support material, and the amount of time spent
on the system.

Many additional weaknesses of CACGS research have been
cited in the literature (Cairo, 1983: Clyde, 1979; Garis,
1982; Snipes & McDaniels, 1981), including:

1. Broad conclusions have been reached on the basis of
user Opinions measured by investigator-created
questionnaires which lack proven validity, are reactive,
and/or subjective.

2. The lack of congruence between criterion measures
and the Objectives Of the system under evaluation.

3. Overreliance upon user evaluations and measures of
vocational maturity.

4. A tendency to ignore the prerequisite skills needed
by CACGS users.

5. The absence Of subject randomization.

6. A paucity Of longitudinal studies.

31

7. Failure to include behavioral measures such as time
spent on the system, time spent on particular components of
the system, or subsequent visits to career resource
libraries.

An additional commonly cited shortcoming of career
counseling and CACGS research that is particularly germane
to the present study involves the failure to take into
consideration client variables that may influence the
effectiveness of the intervention. A study by Melhus,
Hershenson, and Vermillion (1973) is apparently only one of
two CACGS investigations devoted to matching treatment
methods with client variables. The investigators rated high
school students on "readiness for counseling,” based on
standardized educational and "personal coping” ability
tests. High readiness and low readiness high school
SOphomores were randomly assigned to either a computer-based
system (CVIS), or a traditional individual counseling
treatment. Effectiveness of the interventions was measured
by progress toward crystallization of vocational goals using
the Occupational Plans Questionnaire (Hershenson, 1967).
Results indicated that high readiness students progressed
equally well with both treatments. The low-readiness
students made progress in both groups, but showed greater
progress in the traditional individual counseling group.

The authors suggest that counselors should spend a greater

Proportion of their time with low readiness students.

In the second such CACGS study, Marin (1984) examined

32
the differential effectiveness of a computer-only (DISCOVER
II) and computer-plus-counselor career guidance
intervention. The interaction between these procedures and
subject decision making styles (rational, intuitive,
dependent) was also examined. Using pre- and posttest data
from the Assessment Of Career Decision Making (Occupational
Subscale), Career Decision Scale, and Occupational
Alternatives Questions, the investigator concluded that: (a)
the computer-plus-counselor intervention was more effective
than the computer-only intervention, and (b) individual
decision making styles did not affect progress by
interacting with the intervention method.

In a relevant study not specifically involved with
CACGS, Coovert and Goldstein (1980) found that a group of
undergraduates with internal locus of control scores had
more favorable attitudes toward computers than external
locus of control scorers. Regarding client-CACGS
interaction factors, it has been reported that deaf students
struggle with the vocabulary on SIGI (Slusher and White,
1978), and the same program may be deficient in areas of
concern to the adult learner (Adams, 1978; Diamond, 1975).

Lacking empirical evidence to support their
contentions, several authors have raised thought-provoking
questions about the advisability and appropriateness of
CACGS use with various other user populations. For example,
Haring-Hidore (1984) claims that female students, minority

Students, and low achievers are likely nonusers of CACGS.

33
Walz (1984) suggests that clients with linear cognitive
styles may find the computer less appealing than those with
inferential thinking and learning styles. Walz does not
explain the rationale for this supposition, the reverse of
which would seem more plausible. He also contends that
experience with computer games may influence one's response
to CACGS. Such claims suggest the importance of undertaking
research aimed at determining the "goodness of fit" between
CACGS and various client populations. The following section
cites examples of, and describes in more detail the
rationale behind such a client-treatment matching approach

in career counseling and CACGS research.

Attribute-Treatment Interaction

Educational and psychological experimentation has
traditionally focused on the differential influence of one
or more independent variables such as teaching methods or
therapeutic interventions, on one or more dependent
variables. In recent years a line of experimentation has
expanded upon this model, giving equal consideration to a
second form of independent variable, learner or client
characteristics. This line of study was originally called
aptitude-treatment interaction research because it dealt
With the varying aptitudes Of students (Borg & Gall, 1983).
It has more recently been labelled attribute-treatment

interaction research because its focus centers upon a wider

34

range of learner and client characteristics. Such studies
generally consist of two types of independent variables:
the first a treatment or teaching method; the second a
student or client characteristic such as a personality
dimension, demographic classification, or level of academic
achievement. Attribute-treatment interaction research thus
gives equal consideration to both the client and treatment
characteristics that may influence counseling outcomes.

In 1964, Colby warned against the "patient uniformity
assumption” (p. 350), wherein individual traits of research
subjects are ignored or overlooked by psychotherapy
investigators. Hunt (1977) states that "the better we are
able to answer the question 'What approach produces what
effects with what kind of client?'... the more we will
understand about the counseling-client process" (p. 12).
Hunt has outlined a differential treatment model for use
with drug abusers that exemplifies this attribute-treatment
interaction approach. His model adopts the formula provided
by Lewin (1935), B a f (P, E), or Behavior is a function of
the Person and the Environment. It states that in terms of
outcomes (B) the purpose of this differential treatment
model is to coordinate client characteristics (P) with
treatment approaches (E). Or otherwise stated, E : P > B, a
Specific counseling approach used with a specific person
leads to a particular result. In this 'BPE' drug abuse
Program, the treatment assignment, rather than being based

SOIely on the type of drug being abused or pattern of abuse,

35
takes into consideration client characteristics (individual
differences) representative Of some system Of personality
organization, developmental stage, or other psychological
consideration.

Cronbach and Snow (1977) reviewed all of the extant
attribute-treatment interaction studies in educational
instruction and concluded that students with high-average as
compared to low-average intelligence levels have been shown
to benefit significantly more from less structured
curricula, and from roughly structured computer assisted
instructions. Students with a high need for achievement and
low amount of defensiveness benefited most from teaching
conditions that provided more freedom and challenge, and
treatments that required the learner to supply a good deal
of the structure and task specification. Defensive persons
were found to do better when instructions placed less demand
on them for structuring their learning.

Numerous psychotherapy outcome studies have found that
such client characteristics as level of anxiety (Brown,
1969), degree of dependence (Gilbreath, 1967), locus of
control (Abramowitz, Abramowitz, & Roback, 1974),
intelligence (Rogers, 1954), level of achievement (Schaefer,
1976), self-confidence (Wachowiak, 1973) and sex (Thoresen,
Krumboltz, & Varenhorst, 1967) differentially influence the
results of specific interventions. In his review of
research on client variables in psychotherapy, Garfield

(1971) noted that clients' degrees of defensiveness, level

36
of motivation, expectations from therapy, degrees of
discomfort, and socioeconomic status have been shown to
influence treatment outcome.

The therapist-patient matching model is exemplified in
the pioneering work Of Whitehorn and Betz (1960) who
differentiated therapists whose schizophrenic patients
improved most from therapists whose schizophrenic patients
improved least. These researchers, labelling the former
therapists ”As," and the latter, ”Bs," stimulated numerous
subsequent investigations into this A-B interaction
hypothesis. Although the results of numerous studies are
highly equivocal, such research has suggested that A-type
therapists, as derived from scores on the Strong Vocational
Interest Blank, may work better with schizophrenics, and
B-type therapists with neurotics.

In 1968, Hebert investigated the effects of similarity
and dissimilarity between selected personality needs of
vocational clients and counselors on the counseling
relationship. Nineteen counselors and 108 clients were
administered the Edwards Personal Preference Schedule (EPPS)
(prior to counseling, and the Communication Rating Scale at
the conclusion of the last (second) interview. Results
indicated the following: (a) significant interaction

between client sex, counselor sex, and degree Of similarity
011 the EPPS, (b) male clients gave significantly higher
1rantings to male counselors than did female clients, (c) male

clients tended to agree with their counselors, regardless of

37
counselor's sex, on the rating of the relationship, to a
greater extent than did female clients. Such a study
suggests that client and counselor traits may interact to
influence the therapeutic outcome.

Butcher and Hebert (1985) investigated the influence of
locus of control similarity between counselor and client in
a matched case study procedure. Two university clients were
selected for participation based on matching by sex (male),
and Rotter IE Scale scores (one highly similar to the
counselor, one highly dissimilar). Through time-series IE
self-recordings, and Counselor Comfort, Counseling Climate,
Client Satisfaction, and Counseling Evaluation Inventory
ratings by the clients, counselors, and an expert rater,
levels of counseling effectiveness were determined. In
general, it was found that the counselor-client intervention
matched according to IE scale scores was significantly more
effective than the client-counselor mismatched intervention.
Their findings led the authors to conclude that they had
provided valuable additional evidence which suggests that
"client and therapist variables account for the largest
percentage of variability in clinical outcomes" (p. 108).

The majority of attribute-treatment interaction studies
have matched clients and therapists, rather than clients and
treatment methods. The distinction between ”therapist” and
"treatment method" is an important one for several reasons.
First, therapists' treatment approaches are Often eclectic

and/or non-aligned with any single therapeutic orientation.

38
Secondly, the actual therapeutic procedure employed by a
therapist often bears little or no resemblance to his/her
stated therapeutic orientation (Argyris, 1976). For
example, a series Of studies by Fiedler (1950, 1951)
established that those who adhere to a particular
therapeutic theory can not be regarded as interchangeable
units. Finally, therapist-client matching studies often
focus upon the former's personal traits, therapeutic
expectations, counseling experience, or demographic
characteristics, while ignoring or minimizing treatment
method.

Client-treatment method interaction studies have
appeared with regularity in the personal/social
psychotherapy literature. For example, Abramowitz,
Abramowitz, Roback, and Jackson (1974), and Best (1975)
examined the effects of tailoring treatment procedures to
clients' internal-external beliefs (as measured by Rotter's
1966 Locus of Control Scale) so as to facilitate abstinence
from smoking. Abramowitz et al. used directive group
therapy for externals, and nondirective therapy for
internals. Best exposed internals to stimulus satiation
aversion procedures, whereas externals were encouraged to
conduct situational analyses of environmental cues that
influence their smoking. Client-treatment interaction
effects were significant in both studies.

Meichenbaum, Gilmore, and Fedoravicius (1971) examined

the comparative efficiency of desensitization,

39
rational-emotive therapy, and the two combined in group
therapy with speech-anxious clients. The researchers found
that the two treatment modalities were equivalent, but
superior to control procedures. Post hoc analysis revealed
a client-treatment interaction when participants were
subdivided according to the specificity or generality of
their anxiety. Specifically, desensitization yielded better
results with the subgroup with anxiety specific to speech
situations, whereas the rational-emotive therapy yielded
better results with the group experiencing generalized
interpersonal anxiety.

Similarly, DiLoretO (1971) examined the comparative
effectiveness of three treatment modalities on introverted
and extraverted undergraduates suffering from interpersonal
and general anxiety. While participants in all three
treatment groups showed more improvement than controls,
systematic desensitization appeared equally effective with
introverts and extraverts, whereas rational-emotive therapy
favored introverts, and client-centered therapy favored
extraverts.

There have apparently been far fewer empirical
examinations of the interaction between clients and
treatment methods in the career counseling literature. It
is not uncommon to see a comparative career counseling study
involving the ambiguous label "traditional vocational
counseling." Such descriptive vagueness serves little value

to the research consumer interested in the comparable worth

40
of counseling practices. In addition, it has been suggested
that the failure of career counseling research to find
differential effects among various treatment approaches may
be due to the investigators' failure to include clients'
individual differences as a major independent variable
(Kivlighan, Hageseth, Tipton, & McGovern, 1981). Fretz
(1981) has suggested that career counseling studies which
find client variables to have a significant effect on
outcomes may be offering a type of treatment that is
interactive with a given attribute, whereas those studies
without effects may be providing treatments that are not
interactive with client characteristics. Numerous authors
(Krumboltz, 1966: Super & Hall, 1978; Takai & Holland, 1979)
have noted the need for career counseling research that
takes into consideration how individual differences interact
with treatments.

Rubinton (1980), taking into account both client and
treatment characteristics, predicted correctly that rational
decision makers would do best with a "rational" career
intervention, whereas intuitive decision makers would do
best with a more “intuitive” approach. Like many such
career counseling outcome studies, she relied upon pre-post
measures Of career maturity and certainty Of vocational
choice as the determinants of intervention effectiveness.

Johnson, Smither, and Holland (1981) evaluated two
college career development seminars “to determine what kind

of vocational interventions were helping what kind Of

41
people" (p. 180). Indices of vocational decidedness and
identity were derived using the Vocational Decision-Making
Difficulty Scale (Holland & Holland, 1977) and My Vocational
Situation (Holland, Daiger, 8 Power, 1980). The criterion
variables consisted of treatment ratings for numerous
seminar experiences such as the Vocational Exploration and
Insight Kit, the Self-Directed Search, the Occupational
Handbook, and talking to employed people about their jobs
and careers. Although an overall main effect Of all
treatments on raising students' level of vocational
decidedness and identity was found, no interactions between
characteristics of students and treatment ratings were
revealed. These results were echoed by Henkels, Spokane,
and Hoffman (1981) who found that vocational identity had no
significant relationship to students' preferences for
receiving interest inventory results by profile only, by
audiotape, or by direct contact with a counselor.

Warner and Jepsen (1979) designed an experiment to
study the interaction effects of conceptual level and group
counseling format on adolescents' career decision-making
behavior. It was predicted that students low on conceptual
complexity and interpersonal maturity (low CL) would profit
from a highly structured treatment environment, and that
high CL students would benefit more from low structure. NO
significant interaction or main effects were Obtained for
attitudes toward career decisions and planning activity. It

was found that group format was related to students ”choice

42
basis complexity" (the differentiation of factors considered
in making a career decision) for the high CL group. The
interaction, however, was in the opposite direction Of that
predicted, which the authors attribute to flaws in the
research design.

Power, Holland, Daiger, and Takai (1979) examined the
effects of the Self-Directed Search as related to 493 high
school students' sex, expectation for treatment, vocational
identity, and vocational decision-making difficulties.
Outcome measures were the number Of occupations being
considered at posttest, the variety of Options, and the score
on a revised Treatment Evaluation Scale (Zener & Schnuelle,
1976). Results revealed that students generally evaluated
the Self-Directed Search favorably, regardless Of their level
Of treatment expectation. Also, students with low identity
scores and high levels Of decision-making difficulties tended
to have the lowest Self-Directed Search ratings. Those who
were high on identity and low on decision-making difficulties
tended to rate the treatment more positively.

Sampson (1984) contends that the interaction between
various user characteristics and CACGS' effectiveness is an
area in need of investigation. In his review of CACGS
outcome research, Cairo (1983) states:

In most of the evaluations... there is the implicit

assumption that all users have the same needs. Thus all

users are lumped together into one experimental group
and compared with nonusers on all the same criterion
variables... There is also a question regarding what

type of user is most likely to benefit from using a

computer-assisted system... but overall this issue
remains relatively unexplored. (p.58)

43

As discussed in the CACGS Research section, the studies
by Marin (1984) and Melhus et al. (1973) represent the only
two CACGS studies concerned with the interaction between
client and treatment characteristics. Berzins (1977),
referring to psychotherapy process and outcome studies in
general, criticized the relatively few extant
client-treatment matching studies on the grounds that they
Often fail to demonstrate true differences among treatment
groups, and often fail to account for possible interaction
effects between counselors' personalities and the treatment
methods used. The present study attempts to avoid the first
shortcoming by employing Holland's General Occupational
themes to differentiate among SIGI users. The second
shortcoming is not relevant due to the absence of counselor
intervention.

Interventions such as SIGI differ markedly from the
treatments discussed in the aforementioned client-treatment
method interaction studies in their nonreliance upon
therapists. As previously noted, therapeutic procedures
Often differ from the practitioner's stated orientation.
However, a CACGS is a generally standardized procedure
unaffected by such potentially intervening factors as
therapists' expectations, personal traits, theoretical
orientation, age, sex, race, experience, etc. It is true
that the manner in which a CACGS is Offered to a clent may
influence its effectiveness (e.g., with a pre-treatment

orientation session or post-treatment debriefing: in

44
conjunction with a discussion group or career planning class;
used partially or in its entirety; made available in a
library, career center, or college placement Office, etc.).
Nonetheless, the intervention itself varies little among
users, and provides a generally well defined ”treatment
method” for investigation. The following section consists of
a description Of Holland's theory and research incorporating

his constructs.

Holland's Theory

 

Holland's (1966) theory of vocational typologies appears
to be an outgrowth Of Lewin's (1936) dictum that "behavior is
a function of personality and environment" (p. 57). To
Holland, the choice Of a vocation is, in part, an expression
of an individual's personality. Vocational interests are
viewed as a product of one's life history, rather than as an
isolated entity.

Holland contends that people can be characterized by
their resemblance to one or more of six personality types. A
'type' represents a cluster Of personal attributes which may
be used to measure the person (Walsh, 1973). According to
Holland, people generally possess characteristics Of all six
types (paraphrased below) with one or two predominating
orientations.

Artistic people have a high need for original and

individualistic expression. They tend to be asocial,

45
intraceptive, emotional and more conventionally feminine than
masculine. These peOple dislike structure, and prize
aesthetics.

Conventional people prefer structure and order in verbal,
numerical, and behavioral activities. They tend to be
concerned with rules and regulations, and identify strongly
with power and status.

Enterprising individuals seek situations where being
popular, verbally assertive, gregarious, and taking
leadership roles are important. They strive to acquire power
and status, while avoiding analytic and more individually
oriented endeavors.

Investigative people tend to have creative, analytic, and
intellectual interests, and prefer to think through rather
than act out problems. They often report themselves as being
scholarly, precise, inventive, confident, and lacking
leadership qualities.

Realistic individuals are characterized by aggressive
behavior, mechanical skill, practical mindedness, and
conventional masculinity. They prefer to act out problems,
to avoid interpersonal tasks, and to seek concrete rather
than abstract problem situations.

Social people enjoy understanding and helping others.
They prefer settings where they can help, train or inform
others. Social individuals usually value socially and
ethically oriented issues and activities, while avoiding

mechanical, highly ordered, or scientific ones.

46

A related assumption is that the environments in which
people work may be characterized by their resemblance to one
or more environments. Thus, for each personality type there
is a logically related environment. For example, the
Enterprising environment is characterized by tasks requiring
verbal responses meant to influence people.

Holland claims that the degree of person-environment
congruence influences such outcomes as vocational stability
and achievement, personal stability, creative performance,
and personal development. The theory contends that
Conventional types search out and experience more
satisfaction within Conventional environments, and so forth
among the other five classifications. The present study
attempts to incorporate Holland's theory of congruence with
the attribute-treatment interaction model. This approach
tests Holland's contention that “the classifiction can be
used to form groups and subgroups to interpret the results
of experiments or surveys or to simplify the
individual-differences treatment problem" (1985, p. 147).

It is predicted that the degree of congruence between
personality type and treatment will influence the
effectiveness of the computerized career guidance
intervention.

For many years researchers and practitioners have
attempted to apply Holland's theory for the purpose of
matching people and jobs. More recently, researchers have

attempted to employ this trait-factor approach for the

47
purpose Of matching clients and treatments. According to
Holland (1966), "The formulations for the types imply that
people of different types have different reactions to
psychotherapy.... The formulations also imply that different
kinds Of treatment are suitable for different types of
people" (pp. 92-93). Although CACGS do not represent a form
of psychotherapy per se, it is reasonable to assume that
Holland's contentions are equally applicable to the career
guidance provided by computerized systems.

Studies using Holland's six personality types to
measure client-counselor and client-treatment interactions
(Jones, 1969; MacDonald, 1984; Whitney and Whittlesey, 1972)
have had mixed results. Cox and Thoreson (1977)
investigated the effect of congruent Holland personality
environments between clients seeking career counseling and
career counselors. One hundred and forty-four college
students selected among three career counselors scripted to
represent one Holland personality type. Results indicated
no clear-cut selection of similar versus dissimilar
counselors, but a predominant selection of the Holland
Social-type counselor.

In contrast to the Cox and Thoreson investigation,
Kivlighan et al.'s (1981) study supports the notion of
client-treatment matching according to Holland's typologies.
Forty undergraduate vocational counseling clients were
categorized as either more people oriented (Enterprising and

Social types) or more task oriented (Investigative and

48
Realistic types) according to their Vocational Preference
Inventory (VPI) scores. The participants were then assigned
to one of two groups; one emphasizing vocational counseling
through numerous interactions with participants (peOple
oriented), and the other emphasizing individual problem
solving (task oriented). Both groups met twice. Analyses
of covariance on subject responses to the Career Maturity
Inventory indicated that participants matched with
treatments similar to the participants' orientations
demonstrated greater career maturity, engaged in more career
information seeking behavior, and enjoyed the group sessions
more.

With the exception of the present investigation, and a
CACGS study currently being undertaken by Kirschner
(personal communication, December 18, 1985), the Kivlighan
et a1. investigation apparently represents the only study
employing this people and task orientation classification
derived from Holland's General Occupational Themes. In
addition, no accounts of attribute-treatment interaction
studies involving both CACGS and Holland's typologies have
appeared in the literature to date. However, Bruch (1978),
in discussing the matching of male clients and career
interventions according to Holland's types, states:

The use of direct counseling approaches employing simple
instructions, examples and guided practice seems
desirable for R, C, and IR type clients because of their
preference for structured activities and action rather
than symbolic-introspective activities.... The use of
reading materials and “apparatus” such as programmed

learning or computer assisted materials... may appeal to
this client's desire for structure, manipulation of

49

materials, and more impersonal approaches to problem

solving.... Programs such as the System of Interactive

Guidance Information (sic)... are illustrations Of

approaches suited to individuals with these personality

types. (PP. 30-31)

The present study represents a modified test of Bruch's
contention that certain personality types, as conceptualized
by Holland, will be more amenable to, and benefit more from
a CACGS intervention than others. Based on the six
aforementioned personality types, clients have been
categorized as either peOple oriented or task oriented. As
in the Kivlighan et al. study, people oriented participants
are those whose highest Vocational Preference Inventory
scores are either Social or Enterprising. Task oriented
participants are those who score highest on either the
Investigative, Realistic, or Conventional scales. Kivlighan
et al. did not include the Conventional type in their task
oriented classification. However, that study did not
involve CACGS, and the inclusion of the Conventional type
among those who would seemingly be amenable to CACGS use,
would appear warranted.

Bruch, quoted above, proposed that R, IR and C types
would be likely CACGS users. According to Holland (1985),
Conventional types prefer activities that entail the
ordered, systematic manipulation of data. They are disposed
toward the use of "data processing machines” in the
attainment of economic and organizational goals. According

to Campbell and Hansen (1981) Conventional individuals are

effective at well-structured tasks, and avoid ambiguous

50
situations and problems involving interpersonal
relationships. They tend to describe themselves as
efficient, orderly, and practical. All of these
characteristics seem congruent with the tasks involved in
SIGI.

Osipow (1973) and Walsh (1973) have noted that
Investigative and Realistic individuals prefer an
environment in which there is little contact with others.
Holland (1973), in noting the traits shared by Investigative
and Realistic individuals, included 'unsociability, (and) an
orientation toward things rather than people” (p. 21). In
contrast, he reported that Of the six types, the Social and
Enterprising persons rely most on other people for their
vocational information. Vierstein (1972) found that the
Social and Enterprising individuals were the most involved
with peOple. Similarly, Osipow (1973) and Walsh (1973) have
emphasized the interpersonally oriented learning environment
of the Social and Enterprising types. It is noteworthy
that of the fifteen possible two code pairings, SE and RI
are the second and third most highly correlated: .54 and
.46, respectively (Holland, Whitney, Cole, & Richards,
1969). Thus it is logical that a great many similarities
between S and E's, and I and R's have been cited in the
literature.

Holland (1985) claimed that people with different
personality patterns respond to teaching methods according

to the formulations for the types. Relatedly, in the

51
present study it is hypothesized that SIGI will be more
effective with task oriented (I,C,R) than people oriented
(S,E) individuals for the following reasons:

1. SIGI involves the clarification of personal values,
and instruction in the decision-making process. Both
involve individual, rather than socially interactive,
problem solving and self-exploratory processes. Such
procedures are more likely to be favorably responded to by
the Investigative and Realistic individuals, who rely on
self-examination in their vocational decision making
(Osipow, 1973; Walsh, 1973).

2. SIGI is undertaken in the absence of other persons,
entails the use of a computer keyboard, monitor and
printouts, and involves the printed rather than spoken word.
The Social and Enterprising individuals will likely be less
amenable to such a mechanistic and nonsocial approach.

3. SIGI's VALUES and STRATEGY subsystems involve
quantified weightings of values and ratings of occupations,
respectively. Such an approach is more likely to appeal to
the structured and analytical disposition of the
Conventional, Investigative and Realistic individuals.

4. There is evidence to suggest that Conventional,
Investigative and Realistic persons are more likely than
Social and Enterprising individuals to have an affinity for
computer use. For example, the following occupations and
matching code types are used in the Strong-Campbell Interest

Inventory: Computer programmer (IRC): and Computer Operator

52
(IC and C). The code type for each occupation was derived
either by testing people in that occupation to determine
their highest codes, or by asking people of known code types
which occupations they liked.

A review of existing studies employing Holland's
General Occupational Themes, and the VPI in particular,
reveals no consensus regarding the manner in which
individuals should be classified, nor the manner in which
data should be analyzed. Holland (1985) claims that the
"flexibility of the classification makes it possible to cope
with both representative and unrepresentative populations"
(p. 147). He further recommends the expansion or
contraction of classifications in order to deal with skewed
distributions of data. He cites this as a necessity due to
the skewed distribution of types among various groups. For
example, Nafziger, Holland, Helms, and McPartland (1972) had
to analyze their data in such a manner to accomodate the
following skewed distribution of young people: R=82%, I=5%,
A=2%, S=2%, E=6%, and C=4%.

Because of the nature of the hypotheses under
investigation in the present study, and the profiles of the
participants, the classifications derived from the VPI were
categorized as follows: All participants with high point
codes of C, I, and R were labeled 'task oriented,” those
‘with S and E primary codes are "people oriented.“
Participants whose primary codes were A, were categorized as

task.or peOple oriented, according to their secondary codes.

53
Six remaining participants who could not be classified
according to this procedure were excluded from analysis.
Ten of the people oriented participants whose primary and
secondary codes were 8 or E, in either order, were
subclassified and separately analyzed as "Pure SE's.” The
breakdown of participants according to their Holland codes
appears in Table 2 of the Results section.

It bears mentioning that a construct bearing
resemblance to this task and people orientation
classification has received considerable attention in the
industriallorganizational psychology literature. In his
attempt to determine optimal leadership characteristics,
Fiedler (1964) described two styles of interaction: a task
(or production) orientation, and a relationship (or
employee) orientation. The former is said to primarily
satisfy ”the leader's need to gain satisfaction from
performing the task," whereas the latter “is primarily
oriented toward attaining a position of prominence and
toward achieving good interpersonal relations“ (Fiedler,
1967, p. 13). Numerous studies have provided evidence
suggesting that the two styles are differentially effective
according to the nature of the organizational structure and
climate.

In 1961, Bass created the Orientation Inventory, a 27
item self-report instrument that generates three scores:
self-orientation, interaction-orientation, and

task-orientation. Like Kivlighan et al.'s people and task

54
orientations, Bass's personality types relate to the
respondents' preference for and concern with others versus
self, and social versus asocial activities. Because the
Orientation Inventory possesses poor and unproven
psychometric properties (Bouchard, 1978), has generated
relatively little scientific attention, and is not commonly
used among career counselors and researchers, it was deemed
inappropriate for use in the present study. Instead,
Holland's VPI has been used to determine the participants'
orientation toward people or tasks. A description and
rationale for the use of the VPI appears in the Methodology

section.

Research Hypotheses

 

A consideration of the literature relevant to CACGS,
attribute-treatment interaction studies, and Holland's
typologies led to the following hypotheses. Hypotheses 1,
2, and 3 represent the major hypotheses to be tested in the
present study. Hypotheses 4, 5, and 6 serve to determine
the relationship between the dependent variables under
investigation.

Hypothesis 1: Task oriented participants will

 

evaluate SIGI more favorably than people oriented
participants.

Hypothesis 2: Task oriented participants will spend

 

more time on SIGI than people oriented participants.

55
Hypothesis 3: Task oriented participants will be more
career decisive following the use of SIGI than people
oriented participants.
Hypothesis 4: A positive relationship will exist
between time spent on SIGI and SIGI evaluations.

Hypothesis 5: A positive relationship will exist

 

between time spent on SIGI and career decisiveness following
SIGI use.

Hypothesis 6: A positive relationship will exist

 

between SIGI evaluations and career decisiveness following

SIGI use.

Exploratory Issues

 

In addition to the above six central hypotheses, a
number of additional issues have been examined. They are as
follows:

1. The people oriented participants (N=44) were
subclassified as ”Pure SE's” (N=10) and “People-minus Pure
SE's' (N=34). The former subgroup consists Of those people
oriented participants whose primary and secondary
occupational codes are Social and Enterprising, in either
order. The latter group are those people oriented
participants whose primary code is Social or Enterprising,
and whose secondary code is neither Social nor Enterprising;
or stated otherwise, peOple oriented participants who are

not "Pure SE's." It is expected that the "Pure SE's,” will

56
exemplify this orientation toward people. In comparison to
the peOple and task oriented participants, they are expected
to respond least favorably to SIGI, as measured by the three
dependent variables under investigation.
2. To determine if the participants' level of career
decisiveness changed after exposure to SIGI, pre- and
post-SIGI CDS scores were compared.
3. A test was performed to determine the internal
consistency of the investigator-created SIGI Evaluation
Questionnaire.
4. A further attempt was made to determine the relationship
between the two levels of independent variable (task and
people orientation), two additional independent variables
(sex and previous use of CACGS), and the three dependent
variables (time spent on SIGI, evaluation of SIGI, and
career decisiveness following the use of SIGI). The purpose
of this analysis is to shed light on the impact of
additional independent variables on the use Of SIGI.
5. An attempt was made to determine if differences exist
between those participants who were and were not
concurrently seeing a career counselor. These participants'
mean scores on the three dependent variables were compared.
In addition, a test was performed to determine if any
relationship exists between these two categories of
participants and the two orientation (task and people)
classifications.

6. A non-statistical comparison was made between the 69

57
research participants, 35 students who did not complete the
required forms, and a comparison group of 639 Michigan State
University SIGI users, on eight variables (sex, age, race,
college class, previous CACGS use, orientation, concurrent
career counselor contact, and pre-SIGI CDS scores). This
was to determine if (a) noncompletion represented a
confounding variable in the present study, and (b) the
present study's findings can be generalized to university
SIGI users in general.
7. Ten of the participants were Artistic types whose
secondary codes were used to determine task or people
orientation. The task and people orientation groups'
dependent variable mean scores were compared when the
Artistic codes are both included and excluded from analysis.
This comparison was performed to determine the potential
confounding nature of this heretofore untested procedure.
The small number of Artistic types (10) made statistical

analysis inadvisable.

CHAPTER III

METHODOLOGY

Research Design

The present study is a "correlational design" (Campbell
& Stanley, 1966). According to Borg and Gall (1983) the
first three hypotheses are ”confirmatory" in that they are
directed at providing empirical support for theoretically
based hypotheses. The additional three hypotheses, which
are aimed at studying the relationship between heretofore
uninvestigated variables, are ”exploratory" in nature.

The central (dichotomous) independent variable is
orientation (task or people). Two secondary (dichotomous)
independent variables are sex (female or male), and previous
use of SIGI (yes or no). The three dependent variables are
evaluation of SIGI (as measured by the SIGI Evaluation
Questionnaire), time spent using SIGI, and career

decisiveness (as measured by the Career Decision Scale).

Participants

Sixty-nine undergraduate Michigan State University students

participated in the present study during the 1985, Spring,

Summer and Fall quarters. Forty-nine (71%) were female, 20

(29%) were male. The students ranged in age from 17 to 38.

(Excluding the single 38 year-old, the range was 17 to 23).
58

59
The mean age was 19 years, 6 months. The distribution of
participants by college class was follows: 27 freshmen
(39%), 28 sophomores (41%), 7 juniors (10%), and 5 seniors
(7%). Two students failed to report their age and class
standing. Because of the preponderance of underclasspersons
and Caucasians, the data were not separately analyzed by
age, class standing, nor race. (See Table 6 for a full
demographic breakdown).

Initial efforts at obtaining treatment and control
volunteers for the present study proved unsuccessful,
resulting in thirteen participants, only two of whom
completed the necessary forms to qualify for inclusion in
the study. The explanation for this investigator's
inability tp procure participants, despite extensive
advertisement and recruitment efforts, seems rather
apparent. Students at Michigan State University, at the
time the study occurred, had free access to eight SIGI
terminals across campus. There was no personal advantage to
their participation in the study, nor any reason for those
randomly placed in the control group condition to agree to
delay their use of SIGI for several weeks.

Subsequent participants consisted of undergraduates who
either independently requested the use of SIGI, or were
referred to SIGI by career counselors, residence hall
personnel, or academic advisors. Although referral sources
were not formally ascertained, a random inquiry Of

participants suggested that the great majority

60
were self-referred, having heard of SIGI from friends or
classmates at CSU.

It is noteworthy that 35 students who completed the
Consent Form failed to provide ample data to warrant
inclusion in the study. Of these participants, 33 completed
the SIGI User Information Form, 30 completed the VPI, and 26
completed the pre-SIGI CDS. Several of these students
failed to complete one or more Of these inventories in their
entirety. Three of these students completed the post-SIGI
CD8, and only two completed the SIGI Evaluation
Questionnaire, which provided information regarding two of
the study's three dependent variables (time usage and SIGI
ratings). Several plausible explanations for such
noncompletion are possible. It is likely that many students
indicated to the receptionists that they intended to
reschedule fr SIGI and then failed to do so: or rescheduled
for a subsequent appointment and then did not return. It is
also likely that many students left the ite without checking
in with the receptionist as instructed prior to SIGI se.
Also, the receptionists may have failed to provide the
post-SIGI materials to students who indicated tha they had
completed the program. Finally, it is known that some
students received the post-SIGI forms, but failed to provide
any or all of the necessary information.

Because of the potential confounding effects of such
noncompletion, a non-statistical comparison has been made Of

the 69 participants and the 35 non-completers (see Table 6).

61
No obvious discrepancies appear to exist between the
completers and noncompleters on the eight demographic
categories for which information was available. The
potential confounding effect that this incomplete data may
have had on the study's findings is examined in the
Discussion section. Table 6 also includes the sex, race and
class standing of the approximately 639 students who used
SIGI at the Counseling Center, Brody Hall, and Learning
Resource Center during the Fall, 1985 and Winter, 1986
quarters at Michigan State University. Wonders Hall did not
Offer SIGI during these terms. Once again, a
non-statistical comparison would ppear to demonstrate that
this study's participants accurately represent Michigan Stae
University SIGI users in general.

To further determine the generalizability of the
present study, an attempt was made to compare the
participants with those in similar studies involving CACGS
in general, and SIGI in particular. Unfortunately, most
authors fail to provide demographic information regarding
their studies' participants. As in the present study,
virtually all Of the investigations cited in the CACGS
Research section involved volunteer CACGS users, rather than
students paid or required to participate. The few authors
who provide the gender distribution of their undergraduate
participants, reveal a female-male ratio similar to that of
the present study (e.g., Coovert & Goldstein, 1980: 62%

female: Reardon et al., 1982: 77% female; Kapes et al.,

se
of
0n

8k:

Stu

62
1985: 69% female). Because individual differences are
rarely considered in CACGS studies, breakdowns according to
class standing, race, age and other demographic variables
are unavailable. The fact that a large percentage of
participants in the present study are underclasspersons
(82%) would certainly seem typical of undergraduate career
clients in general. Given these facts, it is reasonable to
assume that the present study's participants are typical of

CACGS users at other universities.

Setting

Participants used SIGI at one of four locations at
Michigan State University during the 1985, Spring, Summer
and Fall semesters. The sites were the University
Counseling Center (Spring and Summer, N=17); two of its
satellite Offices: Wonders Hall (Spring, N=11), and Brody
Hall (Spring, N=4); and the Learning Resources Center
(Spring and Fall, N=37). The Counseling Center offices
provide a full range Of personal and vocational counseling
services. The Learning Resources Center, in addition to
offering SIGI, provides workshops and individual counseling
on a variety of academically related areas such as study
skills, time management, test taking, and word processing.
Each of the locations had one SIGI station available for
student use by appointment.

At the time data were being collected, no formal

63
written or oral pre-SIGI orientation procedures existed at
any of the sites. NO effort was made to account for all
SIGI users' preparation for, or familiarity with, the system
prior to its use. Those students referred by career
counselors, residence hall personnel, or academic advisors
likely received cursory descriptions of the program. It is
unlikely that any such brief and informal preparation would
have had significant influences on the outcomes under
investigation in the present study.

At each location, use of SIGI occurred in relative
isolation, with nearby technical assistance available.
Although the physical settings, noise levels, and degrees of
privacy differed somewhat in the four locations, each
appeared quite adequate for the successful implementation of

the system.

Procedures

 

The manner in which students used SIGI and completed
the study's inventories was largely the same at all four
sites. The components and format Of the SIGI program itself
was identical at each location. Any administrative,
procedural, or environmental dissimilarities that may have
existed do not appear to warrant separate analyses or
exclusion from the study. Students scheduled, either in
person or over the phone, an initial two-hour appointment to

use SIGI. Upon arrival they completed the pre-SIGI

64
materials (Consent Form, SIGI User Information Form, Career
Decision Scale, and Vocational Preference Inventory; see
Appendices A-D, respectively). They then proceeded to use
SIGI independently and at their own pace for all or any
portion of their allotted two hour period.

Upon completion, or when they elected to discontinue
use, students returned to the receptionist to indicate that
their use Of SIGI was complete, or that they wished to
reschedule for an additional one hour appointment. In the
former instance, students completed the post-SIGI materials
(Career Decision Scale, and SIGI Evaluation Questionnaire;
see Appendices C and E, re3pectively), ending their
participation in the study. In the latter case, students
returned on one or more occasions to continue their use Of
SIGI. Upon completion of the system, or when they decided
to terminate its use, these students also completed the
post-SIGI materials, ending their participation in the

study.

Independent Variables

As discussed in the previous chapter, participants'
task or people orientation represents the central
independent variable under investigation in the present
study. However, as is always the case in correlational
research, additional extraneous variables which may account

for the variance in the dependent variable have to be taken

65
into consideration. The potential influence of several such
variables have been cited in the literature; for example,
participants' intelligence, grade point average,
socioeconomic status, cognitive strategies, sex, attitude
toward computers, experience with computer programming or
game-playing, etc.

In the present study, participants' sex and prior CACGS
use have been included as the second and third independent
variables under investigation. The reasons for selecting
these particular predictor variables, and excluding others,
follows:

1. The interaction of treatment strategy and sex is a
common research theme among counseling and career-related
outcome studies. Although there is little evidence to
suggest that males and females differ significantly in terms
of intellectual-cognitive functions (Maccoby & Jacklin,
1974), the sexes are undoubtedly socialized in a manner that
would differentially influence their attitude toward and
comfort with computers. Haring-Hidore (1984) claims that
for women, "the social environment generally has not
encouraged exploration into the mechanical and technological
world that is commonly experienced by men" (p. 140).

2. Common sense suggests that participants who had
previously used CACGS are likely to approach SIGI with a
positive attitude regarding CACGS in general. Conversely,
it would seem logical that those who previously disliked or

did not benefit from their use of CACGS would be less likely

66
to voluntarily use SIGI than those who had previously valued
and/or benefitted from the experience.

3. Other potential demographic predictor variables did
not "naturally occur” in sufficient number to allow for
their examination. For example, only 5 of the 69
participants in this study identified themselves as minority
group members. Also, many Of the participants were first
quarter freshmen who had no college grade point averages to
report. The relatively few grades that comprise the second
and third quarter freshmen GPA's are of doubtful value for
determining a student's academic progress. In addition,
because so many participants were undecided about their
major, that variable could not be examined. In contrast,
males and females, as well as previous CACGS users and
nonusers, both participated in sufficient number to allow
for investigation.

4. Orientation (task or people), sex and previous
CACGS use are all easily and accurately measured. These
variables would likely be perceived as relevant to career
counselors, in contrast to the more esoteric and/or
unrelated psychological or demographic independent variables
often encountered in the literature (e.g., locus of control,
birth order, self-esteem, etc.).

5. The number of participants in the present study
largely dictated that no more than three predictor variables

be included in the multiple regression formula.

67

Dependent Variables

 

The three dependent variables under investigation in
the present study are users' evaluation of SIGI, time spent
on SIGI, and post-SIGI career decisiveness. The
recommendations Of numerous authors concerning career
counseling research criteria influenced the selection of
dependent variables and criterion measures in this study.
Some Of those suggestions involved the inclusion of multiple
criteria (Fretz, 1981; Oliver, 1979), behavioral and
nonreactive measures (Cooper, 1976: Webb, Campbell,
Schwartz, & Sechrest, 1966), instruments with adequate
psychometric properties (Jenson, Coles, & Nestor, 1955:
Zytowski & Betz, 1972), instruments which are consistent
with the systems' objectives (Cairo, 1983), and criteria
which are potentially valuable yet previously neglected
(Cairo, 1983). A description Of the three dependent

variables and the procedures for their measurement follows.

SIGI Evaluation Questionnaire

As earlier discussed (see section on CACGS Research)
numerous descriptive studies have reported favorable user
reactions to SIGI (e.g., Chapman et al., 1977; Nickerson &
Davis, 1983; Sampson & Stripling, 1979) and other CACGS
(e.g., Reardon et al., 1982: Ryan et al., 1980). -The nature

of this particular criterion measure is such that it

68
precludes the use of control group data. This problem can
be minimized by comparing the evaluations Of treatment
subgroups, or two or more treatment groups, on the basis of
multiple independent variables. However, as Cairo (1983)
points out, most CACGS outcome studies involve a single,
undifferentiated treatment group. What has resulted is an
abundance of descriptive accounts Of generally favorable
user reactions.

In past investigations, user reactions have been
procured through the use Of questionnaires constructed by
the investigators or the CACGS publishers (e.g., DISCOVER
Foundation's DISCOVER Student Questionnaire, and Educational
Testing Service's SIGI Evaluation Questionnaire: Form E).
Apparently no efforts have been made to establish the
reliability or validity Of such measures. In the present
study posttest scores on the Career Decision Scale and time
spent on the system will both serve as tests Of concurrent
validity of the SIGI Evaluation Questionnaire in use. It is
predicted that evaluation scores will be related to scores
on the Career Decision Scale; that is, favorable ratings are
related positively to decidedness and negatively to
undecidedness. It is further predicted that favorable
ratings will be positively related to task orientation and
negatively related to people orientation.

The SIGI Evaluation Questionnaire developed for use in
the present study consists of 11 Likert-style items related

to the system's components and the user's overall

69

impressions of the system's usefulness. Ten of the 11 items
were taken verbatim from the Educational Testing Service's
SIGI Evaluation Questionnaire: Form E. Revisions were made
according to the recommendations of several Michigan State
University career counselors familiar with SIGI. Those
revisions include the addition of the item "How likely would
you be to recommend SIGI to a friend?” Three items from
Form E were excluded because they dealt with subject matter
absent from Michigan State University's version of SIGI.

The four-point Likert-style method of response has been
substituted for the ABCDF scaling method used in ETS's
version because of the former's similarity to the Career
Decision Scale's scaling procedure, and its frequent use in
attitude scale construction. It is doubtful that the
five-point ABCDF scale represents a valid summated rating
procedure. Given today's ”grade inflation" and the negative
attitude of many students toward a grade of ”C," such a
"neutral" item may represent a source of ”response bias”
(Nunnelly, 1978). In fact, the use of this ABCDF scale by
ETS and CACGS researchers may help to explain the
inordinately favorable SIGI user reactions in virtually all
studies reported in the literature. The instrument is
scored by tallying the responses to each of the eleven
items. The scores can range from 11 (very favorable rating)
to 44 (very unfavorable rating).

A test of internal consistency was performed to assess

the instrument's reliability (see Secondary Statistical

70
Results). To summarize, in the present study the evaluation
form in use was subjected to several procedures, and
constructed in such a manner, to provide evidence of its
validity and reliability. They include (a) using items that
have previously been employed in numerous SIGI studies, (b)
eliminating inappropriate items, (c) revising the scaling
method in an attempt to provide consistency between measures
and improve its psychometric properties, (d) constructing
the questionnaire with the input of career counselors
experienced in the use Of SIGI and similar rating scales,
(e) correlating the ratings with other relevant variables,
and (f) subjecting the instrument to a test of internal
consistency.

As previously discussed, numerous CACGS studies have
elicited user reactions, reporting generally favorable
ratings. However, there has apparently been little or no
attempts to (a) distinguish between evaluators according to
relevant personality or demographic traits, (b) determine
the relationship between user ratings and other variables,
or (c) provide evidence of the rating scale's validity and
reliability. The present study has been designed to address

all three of these seemingly important factors.

Career Decision Scale

 

A great deal of the concern of vocational development

theorists relates to the process of career decision making,

71
and the problem of career indecision, or its counterpart -
career decidedness. Harren (1979) and Van Matre and Cooper
(1984) discussed the importance of assessing career
indecision when counseling college students. Carney,
Savitz, and Weiskott (1979) and Harman (1973) reported that
undecided students seeking help with vocational decisions
comprise a majority of the clientele Of most university
counseling centers. Numerous studies have compared decided
and undecided students, revealing that undecided students
are more anxious (Kimes & Troth, 1974: Walsh & Lewis, 1972),
more dependent (Ashby, Wall, & Osipow, 1966), more dogmatic
(Maier & Herman, 1974), have lower self-esteem (Barrett &
Tinsley, 1977: Resnick, Fauble, & Osipow, 1970), and have
lower high school and college grade point averages
(Lunneborg, 1975) than decided students. These findings
notwithstanding, Osipow (1983) claims that such students
have revealed "few systematic reliable personality or
ability factors involved in whether an individual is decided
or undecided about a career" (p. 283).

It is generally recognized that it is not enough to
merely supply vocationally indecisive clients with
information about occupational alternatives and assume they
will use it to make wise decisions (Mendonca & Siess, 1976).
Barak, Carney, and Archibald (1975) found that little
relationship existed between the amount of vocational
information seeking participants engaged in, and their level

of career decidedness. This would support SIGI develOper

72
Martin Katz's contention that a computerized career guidance
system must teach a decision making process as well as
supply vocational information to its users.

The Career Decision Scale (CDS; Osipow, Carney, Winer
et al., 1976) has been employed in the present study as a
meaure of career decidedness. The CD8 was designed to
identify barriers that prevent individuals from making
educational and career decisions. It is based on the
rationale that a finite number of relatively discrete
circumstances are responsible for problems individuals have
in reaching appropriate closure in implementing career
decisions (Osipow, 1980).

The instrument consists of 16 Likert-style items which
are commonly reported explanations for career indecision.
Two additional items indicate the level of certainty of a
career and school major and, if included, are scored
negatively with the other items. Thus, the first two items
measure decidedness, and items 3 through 18 measure
indecision. A numerical value for each item can be assigned
to each response Of 1, 2, 3, or 4. An index Of indecision
ranging from 16 (1x 16) to 72 (4x 16) can then be produced
for each student (Taylor, 1982: Utz, 1983). The higher the
score, the greater the level of career indecision. A 19th
item is an unscored item that offers the respondent the
Opportunity to clarify previous responses.

Some researchers have included the first two items in

the generation of a global indecision score (MacDonald,

73
1984; Niece & Bradley, 1979). However, the majority Of the
research directed at demonstrating the instrument's
reliability and validity appears to have used only items
3-18 in the scoring procedure. Since the CDS developers
(Osipow, Carney, Winer et al., 1976) do not describe a
method for scoring, the more commonly used procedure of
including only items 3-18 has been employed in the present
study.

The reported test-retest correlations of CD8 items
range from .70 to .90 for the entire scale (Osipow, Carney,
Winer et al., 1976). Osipow, Carney, and Barak (1976) found
that test-retest correlations for the sum of items 3-18 were
.90 and .82 for two samples Of college students over a two
week period. Slaney, Palko-Nonemaker, and Alexander (1981)
reported an overall test-retest correlation of .70 over a
six week period with 857 college students.

Numerous studies have established its validity as an
instrument to measure the reduction of career indecision
(Osipow, 1980: Sutera, 1977: Taylor, 1979). For example, in
two CACGS outcome studies employing the CDS, Glaize and
Myrick (1984) and Pinder and Fitzgerald (1984) found that
college students using the computerized system CHOICES and
DISCOVER, respectively, had significantly greater career
decidedness gain scores than did control group members.
Also, as previously described, Marin (1984) and Kapes et al.
(1985) used the CDS to measure decreases in career

indecision following participants' CACGS use.

74

In addition, numerous career intervention outcome
studies not involving CACGS have employed the CDS as a
dependent measure (Carney, 1977; Cooper, 1986; Neice &
Bradley, 1979; Rubenstein, 1978; Taylor & Betz, 1983: Utz,
1983). Slaney, Palko-Nonemaker, and Alexandr (1981) found
that the total score Of the CD3 and all of the individual
items discriminated between vocationally undecided and
decided undergraduates. Similarly, Limberg (1980) found
that students seeking assistance at a Career Center or
enrolling in a career planning class showed a higher level
of indecision than a group not pursuing career assistance.
In addition, several studies (Barak & Friedkes, 1981; Neice
& Bradley, 1979: Slaney, 1978) found no differences by sex,
indicating that the CDS does not reflect gender. Therefore,
there appears to exist ample empirical evidence
demonstrating the CDS's reliability, validity, and
sensitivity to career decidedness changes due to career
counseling interventions.

In his review of CACGS-related research, Cairo (1983)
noted that the outcomes which were measured by various
career-related instruments were often inconsistent with the
objectives, content and format of the system being
evaluated. The CD5 appears to have substantial reliability
and acceptable validity for the purpose of measuring levels
of career decidedness. Career decidedness, "the ability to
clarify one's career goals” (Glaize & Myrick, 1984), is

generally considered by vocational theorists to be a

75
learnable skill. SIGI, as earlier discussed, is designed to
clarify values as they related to occupational choice, and
teach a decision making procedure. Thus, the CDS would
appear excellently suited to the purpose of evaluating
SIGI's effectiveness.

A particular advantage of the CDS is its brevity.
Because the participants in the present study were regularly
scheduled SIGI users who had to arrange their SIGI use
around their class schedules, it would have been unrealistic
and unreasonable to require the completion of overly
time-consuming inventories before and after the use of SIGI.
This is especially important because of the additional
required completion of the Consent Form, VPI, SIGI User
Information Form, and SIGI Evaluation Questionnaire.

In at least one similar study (Kapes et al., 1985) the
length and expense of one of the criterion measures (Career
Development Inventory) necessitated that only 40% of the
participants complete the instrument. As the authors
reported, ”the CDI... is an expensive instrument and takes
considerable time to administer, therefore it is probably
not a good instrument to use on further research with
computer-based guidance systems..." (pp. 11-12). Brevity is
particularly germane as one of the predominant complaints
registered in reaction to CACGS use is the amount of time
required for their use.

An additional advantage of the CDS over the more

commonly used career "development" or "maturity" instruments

76
(e.g., Career Development Inventory, Career Maturity
Inventory, and Assessment of Career Development) relates to
the apprOpriateness of using vocational maturity as a
criterion variable in such research. In most CACGS
investigations vocational maturity has been measured just
before and immediately after participants used a particular
system. But, as Cairo (1983) noted, "since vocational
maturity is, by definition, a developmental process, to
expect resounding increases over short periods of time even
with the use of computer-based systems is probably
unreasonable" (pp. 57-58). Conversely, increased
”decidedness" may reasonably be expected as a result of the
acquisition of decision making skills and career/educational
information over a relatively brief period of time.
Therefore, the CDS appears better suited to discerning such
changes in decidedness than the more traditional "career
development" instruments.

There exist several additional instruments which, like
the CDS, purport to measure career decidedness. The
Vocational Decision Scale (Jones & Chenery, 1980) was
designed to measure three dimensions of this particular
construct: degree of decidedness, comfort with decidedness,
and reasons for decidedness. However, the instrument has
apparently generated little or no scientific attention since
its development and initial testing. Similarly, Holland,
Daiger, and Power's (1980) My Vocational Situation has

apparently received considerably more use as a clinical

77
diagnostic and screening device, than as an experimentally
valid measure of career decidedness. Nonetheless, pending
more extensive use in outcome research, My Vocational
Situation would appear to be potentially well suited to the
assessment of CACGS effectiveness.

The Assessment Of Career Decision Making (Harren,
1976), previously called the Vocational Decision Making
Checklist bears considerable similarity to the CDS. Like
the CDS, it has been used to measure the outcomes of various
career counseling interventions (Cochran et al., 1977:
Pinder & Fitzgerald, 1984; Smith & Evans, 1973), and appears
to possess adequate reliability and validity. However, only
one portion Of this 56-item inventory measures a construct
comparable to the CDS's ”vocational decidedness;" the
20-item Occupational Decision Making Scale's measure of
”vocational commitment." Studies by Carney (1977), and
Osipow and Schweikert (1979) found significant correlations
between overall scores on the ACDM and CDS. These two
instruments appear in the literature with approximately
equal frequency, possess adequate validity and reliability,
and appear to be measuring a similar construct. No distinct
advantages Of either instrument are apparent.

Finally, the Vocational Decision Making Difficulty
Scale (Holland & Holland, 1977), a 13-item scale devised to
assess the number of reasons given by persons for vocational
indecision, has been utilized in career counseling outcome

studies (Johnson, Smither, & Holland, 1981; Slaney, 1980).

78
A study by Rogers (1980) found convergent validity between
the Vocational Decision Making Difficulty Scale and the CDS,
which were significantly positively correlated. However,
the CDS appears to have generated more empirical scrutiny
and support, and thus was deemed preferable for use in the

present investigation.

Time Spent on SIGI

 

According to Cairo (1983) "one of the most glaring
omissions from the literature is an examination of how the
amount of time spent using a system affects the user... this
is both an obvious and valuable research question" (p. 58).
In the only published CACGS study to include time spent on
the program as a dependent variable, Myers, Lindeman,
Thompson, and Patrick (1975) found a positive correlation
between use-time and gain scores related to planning
orientation and knowledge, and use of resources for
occupational exploration. However, it is noteworthy that
heavy users also tended to have higher pretest scores than
light users on the aforementioned measures Of career
maturity.

Some computerized career information systems, such as
GIS, offer the first-time user the opportunity to bypass the
initial guidance component. For this reason users who are
only interested in the acquisition of career-related

information may only spend from 5-10 minutes on the system.

79

However, first-time users Of SIGI are required to complete
the initial Values and Locate subsystems, which necessitates
a greater investment of user time and energy. Individuals
who spend less than two hours on SIGI have only accessed a
fraction of the system's material. It would appear logical
to assume that the extended-time user (two hours or more)
will derive greater benefits from SIGI than the brief user.

In their unpublished descriptive study, Nickerson and
Davis (1983) sought explanations from SIGI users who had not
completed all of the program's subsystems. According to the
authors, 57% of the responses "could be interpreted very
positively, i.e., needed information was found, jobs were
obtained, assignments completed, other resources were used,
or decisions concerning a major or career were made" (p. 7).
Furthermore, 45% of the noncompleters reported their
intention of returning to finish the program at a later
date. Such findings, although tentative, lend doubt to the
assumption that noncompletion or brief use of a CACGS
indicates program ineffectiveness and/or client
dissatisfaction. One purpose of the present study is to
shed light on this issue involving CACGS usage time.

Two questions (see Appendix E) will be asked of each
participant upon completion of SIGI to determine the amount
of time spent on the system. Unfortunately, the logistics
of SIGI use in the four settings rendered the recording of

student usage time by receptionists impossible.

80

Additional Forms and Instrumentation

Two additional instruments, the Vocational Preference
Inventory, and the SIGI Users Information Form were used to
Obtain information concerning participants' personality and
demographic characteristics. In addition, a Consent Form
was used to briefly describe the study and obtain the
students' written consent to participate. Descriptions and

the rationale for these instruments' use follows.
Vocational Preference Inventory

The VPI was developed by Holland in 1978 as both a
vocational interest measure and personality inventory. The
rationale for the develOpment of the VPI was based on the
notion that preferences for specific occupations are
expressions of personality. This brief, self-administered
inventory requires that respondents indicate their attitudes
toward 160 occupations. Eleven scores are generated, six of
which (Realistic, Investigative, Artistic, Social,
Enterprising, and Conventional) comprise the previously
described Holland's typologies. Inventory users blacken a
"Yes” circle by those occupations that interest or appeal to
them: a 'No' circle by those occupations they dislike or
find uninteresting: or make no marks when they are undecided
about an occupation. The VPI is scored by tallying the

number of occupations marked "Yes,” ranging from a low of 0

81
to a high of 14. Thus, for each of the six Holland's
typologies, a score of 0 to 14 is generated.

The test-retest reliability coefficients, ranging from
.62 to .98 with college seniors over six weeks, and .61 to
.93 with college freshmen over one year (Holland, 1978),
suggest that the instrument has moderate to high
reliability. Likewise, internal consistency scores ranging
from .69 to .85 with employed adults, and .76 to .89 with
college freshmen, demonstrate the scales to have moderate to
high homogeneity of content.

The VPI has been subjected to numerous tests of
construct validity (Cole & Hanson, 1971; Edwards & Whitney,
1972; Lacey, 1971). Similarly, a large number Of cOncurrent
validity studies (Dunnette, Kirchner, & DeGidiO, 1958;
Gaffey, 1972; Holland, 1960) have demonstrated that "the VPI
scores differentiate between men and women; between
salesmen, advertising and scientific personnel; between
normal, psychiatric, TB patients and psychopaths; and
between psychotic and non-psychotic patients" (Walsh, 1972,
p. 168). Walsh further concluded that ”the results of the
validity studies lend support to the meaning attributed to
the scales and to the rationale underlying the development
of the inventory" (1972, p. 168). Hundreds of subsequent
studies using and examining the VPI, and the continued
popularity of the instrument suggest that Walsh's generally
positive assessment is shared within the career counseling

and research community today.

82

There are several instruments available by which
experimental participants' General Occupational Theme can be
assessed: the VPI, Strong-Campbell Interest Inventory,
Self-Directed Search, and Personal Survey. The rationale
for using the VPI in the present study, rather than one of
these other instruments, follows:

1. According to Walsh (1972), "In terms Of application,
Holland's theory and the VPI are easily translated into
research terms" (p. 168). Conversely, the Personal Survey,
which fails to receive mention by Holland in his 1973 book,
has apparently received little scientific attention since
its development in 1963. In fact, in career counseling
outcome studies where the identification of the
participants' RIASEC theme is required (e.g., Cox &
Thoreson, 1977; Guthrie & Herman, 1982; Holland 3 Whitney,
1968; MacDonald, 1984; Whitney & Whittlesey, 1972), the VPI
appears to be the overwhelmingly favored instrument.

2. The VPI is a nonreactive measure whose
administration alone is unlikely to hinder or enhance one's
career decidedness, or influence any career related
intervention method. Such is not the case with the
Self-Directed Search, which is a self-administered
"vocational counseling experience" (Crites, 1978, p. 1608).

3. The VPI is brief, easily scored, and inexpensive.
In contrast, the Strong-Campbell Interest Inventory,
requires rather expensive and time consuming computer

scoring, and consists of approximately twice the number of

83
items as the VPI.

4. The VPI possesses very acceptable psychometric
properties (see above). Unlike the SCII, separate female
and male scales do not exist. This undoubtedly more
accurately reflects student interests, and the accessibility
of occupations across genders in the 19803. The VPI scales
are equal in length and have "useful homogeneity, ranging
from .76 to .85" (Holland, 1966). Such is not the case with
the Self-Directed Search, a shortcoming cited by Crites
(1978).

Other less frequently practiced procedures exist for
ascertaining people's typological codes. For example,
Grandy and Stahmann (1974), and Schneider, DeWinne, and
Overton (1980) recorded participants' occupational choices
and subsequently classified them according to Holland's
(1974) Occupational Finder codes. However, such a procedure
is infrequently used and seems to Offer no advantage over
the more rigorously researched VPI procedure. Therefore, to
summarize, the VPI's favorable psychometric features,
gender-neutral occupational titles, brevity, simple
administration, low cost, and popularity within the career
counseling and research community make it the preferred
procedure for determining participants' General Occupational
Themes.

Congruence, consistency, and differentiation are three
concepts related to Holland's theory that warrant brief

consideration herein. Congruence is said to exist when

84
people actually choose environment types compatible with
their own typologies. Consistency refers to the
similarities of personality types to other personality types
within Holland's schema. A person is said to be well
differentiated when his/her profile closely resembles a
single type and shows little resemblance to other types. A
poorly differentiated or undifferentiated profile is
characterized by about equal numbers of the six types.
These concepts, common to many studies involving Holland's
typologies, are excluded from the present investigation,
despite Holland's (1983) assertion that the three concepts
“are needed to apply the theory to practical problems" (p.
37). The reasons for this exclusion are as follows:

1. Congruence is a match between a typology and a
vocational environment, not a personality trait per se. The
participants in the present study are predominantly
pre-employed underclassmen with either undeclared or only
tentatively declared majors. Therefore, in the present
study, the common practice Of matching according to college
major or present occupational status is not relevant or even
possible.

2. Holland (1973) and Bruch (1978) have stressed the
importance of including all three criteria in
client-counselor matching research based on Holland's
theory. However, the present study involves the far less
frequently addressed issue Of client-treatment interaction.

The relevance of, and need to consider these three variables

85
in such attribute-treatment interaction studies remains
unsubstantiated.

3. Few studies have provided evidence for the validity
Of the concepts Of consistency and differentiation
(Wigington, 1983). The few studies focusing on differential
treatment outcomes according to clients' Holland's
typologies (e.g., MacDonald, 1984; Whitney & Whittlesey,
1972) generally fail to support the necessity of including
these dimensions. In fact, the concepts of consistency and
differentiation are absent from Osipow's (1983) most recent
review of Holland's theory, even though an earlier edition
(1973) depicted the constructs as important to Holland's
theory.

4. The only extant attribute-treatment interaction
study employing Holland's typology (Kivlighan et al., 1981)
did not take into consideration these three dimensions. As
previously reported, this study's results generally
supported the conclusiOn that consonance between client
personality type and type of treatment enhance treatment
effectiveness. The only criterion for classification within
one of the six Holland's typologies was participants' single
high point code.

5. Like the Kivlighan et al. (1981) study, the majority
Of investigations employing the VPI to classify participants
appear to include the primary code only (e.g., Schneider,
DeWinne, & Overton, 1980; Whitney & Whittlesey, 1972). Two

additional factors may further explain the sole use of the

86
dominant VPI score in determining participants' types.
First, a paucity of functional definitions of the three
dimensions exists. For example, studies that have relied
upon ”well differentiated” profiles all differ in their
criteria for differentiation (e.g., two, three, or four raw
score points above the next highest type). Second, the
inclusion of one or more of the dimensions drastically
reduces the usable participant population. For example,
MacDonald (1984), who included only participants who
qualified on the three dimensions according to criteria he
had established, found that 75% of a community college
student sample, 70% of a graduate student sample, and 68% of
a mixed sample failed to qualify for inclusion in the study.
Experimenters are unlikely to eliminate approximately 70% of
their sample in the absence Of compelling empirically-based
evidence for doing so.

As discussed in the section on Holland's theory,
participants with high point codes of C, I, or R are
classified as "task oriented;' those with S or E primary
codes are ”peOple oriented." Ten participants with Artistic
'primary codes were categorized as task or people oriented
according to their secondary codes. Four participants with
Artistic codes and no scores (zeroes) on the remaining five
types were excluded from the study.

Unlike the other five types, the Artistic individual
does not appear to be easily classifiable as task nor people

oriented. Like task oriented individuals, Artistic types

87
prefer to work with objects or machines, than people
(Osipow, 1983). However, like people oriented individuals,
Artistic types dislike structure, and have a great need for
individual expression (Campbell & Hansen, 1981). Thus, the
person scoring high on the Artistic code would appear to
possess similarities and dissimilarities with both the task
and people oriented individual.

An examination of the Strong-Campbell Interest
Inventory's occupational code types reveals a great
disparity between Artistic types; largely a function of
their secondary and tertiary codes. For example, a drafting
technician, who relies heavily on well-ordered, mathematical
activities is an ”AR” (task oriented) type. In contrast, an
"AB” (people oriented) type, such as Public Relations
Director or Advertising Executive would rely more on his or
her interpersonal and creative skills in the workplace.
Thus, Artistic types can be either task or people oriented,
according to their secondary and terciary codes.

Relatedly, the Artistic classification appears to
comprise a more diverse group Of people and work
environments than the other five types. For example,
lawyers, interior decorators, and music teachers would
appear to have more dissimilarities than similarities.
However, the classifications gain credence when their
secondary codes are taken into consideration (I, E, and A,
respectively) (Campbell 8 Hansen, 1981). Thus, it would

seem reasonable to classify Artistic participants in the

88
present study as either task or people oriented according to
their secondary code. Admittedly however, there is no
precedent for such a practice, as the only other study
employing this ”orientation” schema (Kivligan et al., 1981)
did not include Artistic types. The possible confounding
effects of such a procedure are examined in the Discussion
section.

In addition, a separate analysis will be performed for
"Pure SE's," ten participants whose primary and secondary
codes are Social and Enterprising, in either order. The
”Pure SE's” are expected to epitomize the people oriented
classification, and any differences between their scores and
those of the larger sample of SE's (people oriented) and the
ICR's (task oriented) bears examination. Unfortunately, no
such "Pure ICR" type was derived due to the small number of

task oriented participants.

SIGI User Information Form

A SIGI User Information Form was developed for use in
the present study (see Appendix B). The lZ-item form was
designed to gather information of potential value to such a
correlational study. Some of the information was
statistically analyzed to determine its possible confounding
effects on the findings. For example, data concerning SIGI
users' sex (item #3), and previous experience with CACGS

(item #9) were included as independent variables in the

89
multiple regression equation (see Statistical Analysis
Procedures). Other potential intervening variables such as
age (item #2), grade point average (item #7), race (item
#4), college class (item #6), and current career counseling
status (item #12) were excluded due to a lack of group
variability, or insufficient numbers to warrant analysis.
No hypotheses were generated, nor analyses performed, for
items 5, 9, 10, and 11, which were included because of their
potential value for subsequent examination. No attempts
were made to check the validity of these 14 self-report

responses.

Consent Form

 

All participants were required to sign a standard
Consent Form (see Appendix A) prior to participation in the
study. The form served to explain the procedures for
completing the pre- and post-SIGI materials, and outline
both the participants' role and rights within the study. In
addition, general results were offered to any students who

requested them. No such requests were made.

PrimarypStatistical Analysis Procedures

Statistical analyses were performed at Colorado State

University on a Control Data Corporation Cyber mainframe

using the Statistical Package for the Social Sciences (SPSS;

90
Nie, Hull, Jenkins, Steinbrenner, 8 Bent, 1975). In order
to test Hypotheses 1, 2, and 3 (see Research Hypotheses
section) three separate analyses of variance were performed.
Hypotheses 1 and 2 were tested using Analyses of Variance
(SPSS: ANOVA); Hypothesis 3 was tested with an Analyses of
Covariance (SPSS: ANCOVA) to control for initial differences
in career decisiveness scores. Participants were
categorized as either people oriented (N=44) or task
oriented (N=18).

To determine the relationship between the three
dependent variables, Hypotheses 4, 5, and 6 were tested
through the calculation Of separate correlation
coefficients. Hypothesis 4 was tested using a Pearson
Product Moment Correlation Coefficient (SPSS: PEARSON CORR).
Hypotheses 5 and 6 were tested using Partial Correlation
Coefficents (SPSS: PARTIAL CORR) in order to statistically
control for the effects of the pre-SIGI levels of career
decisiveness. The correlations are, respectively: (a) Time
spent on SIGI and SIGI evaluations; (b) Time spent on SIGI
and post-SIGI career decisiveness; and (c) SIGI evaluations

and post-SIGI career decisiveness.
SecondaryiStatistical Analysis Procedures
As discussed in the Exploratory Issues section, a

number of additional tests will be performed. They are as

follows:

91
1. In addition to the aforementioned people and task
orientation classifications, the people oriented
participants were subclassified as either "Pure SE" (N=10:
those participants whose primary and secondary codes were
Social and Enterprising, in either order; or ”People-minus
Pure SE” (N=34: those people oriented participants who were
not Pure SE's). In addition to comparing the task and
people oriented participants' mean scores on the three
dependent variables (tests of Hypotheses 1-3), the following
mean differences between group scores were calculated (using
SPSS: ANOVA when comparing means for time and evaluation;
and ANCOVA when comparing means for post-SIGI career
decisiveness): (a) ”Pure SE” and task oriented
participants; (b) "Pure SE" and "People-minus Pure SE"
participants; and (c) ”People-minus Pure SE" and task
oriented participants.
2. A T-test for Dependent Samples (SPSS: T-TEST PAIRS) was
performed to measure mean differences between pre- and post—
career decisiveness scores.
3. A Coefficient Alpha test (SPSS: RELIABILITY) was
performed to determine the internal consistency of the
investigator-created SIGI Evaluation Questionnaire.
4. In a further attempt to determine the relationship
between the two levels of independent variable (task and
people orientation) and the three dependent variables (time
spent on SIGI, evaluation of SIGI, and career decisiveness

following the use of SIGI) three separate Stepwise Multiple

92
Regression procedures (SPSS: REGRESSION) were performed.
The third dependent variable, career decisiveness following
the use Of SIGI, has been determined through participants'
post-SIGI Career Decision Scale (CDS) scores. Pre-SIGI CDS
scores have been included in the multiple regression formula
as a covariate to control for the effects of initial
differences. In an attempt to control for the effects of
extraneous variables, two additional independent variables,
sex and previous use of CACGS have been included in the
multiple regression equations. Thus, three separate
stepwise multiple regression equations were performed, one
for each dependent variable. In each equation, the three
dichotomous independent variables are: (a) orientation
(task or people), (b) sex (female or male), and (c) previous
use of CACGS (yes or no).
5. In order to determine if differences existed between
those participants who were and were not concurrently seeing
a career counselor, Analyses of Variance (SPSS: ANOVA and
ANCOVA) were performed to compare mean scores for these two
groups on each Of the three dependent variables. Although
no interaction between orientation (task or people) and
concurrent contact with a career counselor (yes or no) was
predicted, a chi-square test was performed to determine if a
systematic relationship exists between these two variables.
6. A table was produced to non-statistically compare this
study's participants, 35 non-completers of necessary forms,

and a comparison group Of MSU SIGI users, on eight

93
variables.
7. A table was produced to non-statistically compare the
dependent variable mean scores of task and people oriented
participants when “Artistic" individuals were included and
excluded from analysis (see Table 6). Because of the small
number of Artistic types whose secondary codes were employed
(four were classified as task oriented, and six as people
oriented), a statistical analysis of mean differences would

have been meaningless.

CHAPTER IV

RESULTS

In this chapter, the statistical results of the
experiment will be presented. First, the central hypotheses
(1-3) will be addressed. As previously discussed,
participants (n=62) were classified as either task or people
oriented, according to their VPI primary codes (or secondary
codes in the case Of Artistic types). The three dependent
variables were SIGI evaluations, time spent on SIGI, and
career decisiveness. Hypotheses 4-6 posit that the three
pairs of dependent variables are positively correlated to
one another.

Hypothesis 1: Task oriented participants will
evaluate SIGI more favorably than people oriented
participants. An Analysis of Variance revealed that the
mean scores of task oriented participants (23.89) were not
significantly different than the mean scores of people
oriented participants (24.20). Hypothesis 1 was not
confirmed (see Table 1).

Hypothesis 2: Task oriented participants will spend
more time on SIGI than people oriented participants. An
Analysis of Variance revealed that, on the average, task
oriented participants spent more time on SIGI (2.57 hours)
than people oriented participants (2.08 hours). However,
these differences did not reach statistical significance.

Hypothesis 2 was not confirmed (see Table 1).

94

95

Hypothesis 3: Task oriented participants will be more

 

career decisive following the use of SIGI than people
oriented participants. After being adjusted for pre-SIGI
levels Of career decisiveness, an Analysis of Covariance
revealed that task oriented participants' mean post-SIGI
career decisiveness scores (35.45) were not significantly
different than people oriented participants' mean career
decisiveness scores (35.75). Hypothesis 3 was not confirmed
(see Table 1).

Hypothesis 4: A positive relationship exists between

 

time spent on SIGI and SIGI evaluations. This hypothesis
was tested using a Pearson Product Moment Correlation
Coefficient. The correlation between these two dependent
variables was significant at'p < .05 (r = .-20). Because
evaluations were reversed scored (i.e., low evaluations
imply high ratings), the negative correlation indicates a
positive relationship between ratings and time, as

predicted. Hypothesis 4 was confirmed.

Hypothesis 5: A positive relationship exists between

 

time spent on SIGI and career decisiveness following SIGI
use. This hypothesis was tested using a Partial Correlation
Analysis. Controlling for pre-test differences in career
decisiveness, the analysis resulted in a non-significant
correlation (r = -.10). Although not statistically
significant, this correlation was in the predicted
direction. Hypothesis 5 was not confirmed.

Hypothesis 6: A positive relationship exists between

 

96
SIGI evaluations and career decisiveness following SIGI use.
This hypothesis was tested using a Partial Correlation
Analysis. Controlling for pre-test differences in career
decisiveness, a correlation significant at p < .01 (r = .28)
was found. In this case, both lower evaluation scores
(i.e., higher ratings) and lower CDS scores, (i.e.,
increased decisiveness) were predicted. This accounts for
the positive correlation. Hypothesis 6 was confirmed.
Thus, two of the three pairings of dependent variables were,
as hypothesized, correlated to a statistically significant

extent.

Secondary Statistical Analyses

The results Of the seven previously discussed
”exploratory issues,” are as follows:
1. As described in the Methodology section, people oriented
participants were subcategorized as "Pure SE's' and
"PeOple-minus Pure SE's.” Three additional pairings of
participants were created: (a) "Pure SE” and task oriented
participants; (b) "Pure SE" and "People-minus Pure SE"
participants; and (c) "People-minus Pure SE" and task
oriented participants. Analyses of Variance revealed no
significant differences among these groups' mean scores on
the three dependent variables (see Table 1).
2. Pre- and post-SIGI career decisiveness scores were tested

using a T-test for Two Dependent Samples. As expected, the

97
mean difference (pre = 38.97; post = 35.90) was
statistically significant (p < .001).
3. The SIGI Evaluation Questionnaire demonstrated high
internal consistency at .88, as calculated using Coefficient
Alpha.
4. Three separate Stepwise Multiple Regression analyses were
performed on the three dependent variables. Due to missing
data, seven participants were eliminated from the multiple
regression equations. In addition to orientation, sex and
previous use of a CACGS were included as two additional
dichotomous independent variables. Pre-SIGI career
decisiveness scores were used as a covariate when the
dependent variable was post-SIGI career decisiveness. The
covariate was used to statistically control for initial
differences on career decisiveness scores. As depicted in
Table 2, none of the independent variables accounted for
significant variance in any of the three criterion
variables.
5. In an attempt to determine if clients differed according
to whether or not they were concurrently seeing a career
counselor, mean scores for these two groups were compared
for each of the three dependent variables. An Analysis of
Variance revealed that the mean evaluation scores for the
two groups did not differ to a statistically significant
extent. Non-significant differences were also found between
the two groups' mean time scores. However, an Analysis of

Covariance revealed that, after adjustment for pre-SIGI

98
career decisiveness, students who were not concurrently
seeing a career counselor were more indecisive following the
use of SIGI, than those who were in contact with a
counselor. The mean scores and standard deviations for each
dependent variable appear in Table 3. The significance of
these results is covered in the Discussion section.

In addition, a Chi-Square test was performed to
determine if participants who were and were not concurrently
seeing career counselors were categorized as either task or
people oriented to a greater extent than would be expected
by chance. As Table 3 indicates, the Chi-Square (.03, df=1)
did not reach statistical significance. Thus, no systematic
relationship was found to exist between orientation and
concurrent counselor contact.

6. A non-statistical comparison was done between
participants, non-completers, and a comparison SIGI user
group, on eight separate variables (sex, race, class,
previous CACGS use, orientation, concurrent contact with a
career counselor, pre-SIGI career decisiveness, and age).
The only potentially meaningful difference that appears to
exist, in this author's opinion, is the somewhat smaller
percentage of females (60%) in the comparison group as
compared to the participant (71%) and non-completer (74%)
groups (see Table 6). The potential significance of such a
finding is discussed in the subsequent chapter.

7. A non-statistical comparison was done between the

dependent variable mean scores Of the task and people

99
oriented participants, according to whether or not Artistic
secondary codes were included or excluded from the analysis

(see Table 5). No Obvious differences between means appear

to exist.

100

Table 1

Means and Standard Deviations:
Three Dependent Variables by Type
(Tests of Hypotheses 1-3)

 

 

Dependent
Variable Orientation N Mean Stand. Dev.
a
SIGI
Evaluations
People 44 24.22 5.90
PeOple-minus Pure SE 34 24.20 6.00
Pure SE 10 24.30 5.77
Task 18 23.89 5.52
b
Time Spent
on SIGI
People 44 2.25 .88
People-minus Pure SE 34 2.08 .86
Pure SE 10 2.64 .91
Task 18 2.57 1.23
c
Post-SIGI
CDS Scores d
People 44 36.69 6.95
36.75 adj.
PeOple-minus Pure SE 34 36.77 7.63
35.75 adj.
Pure SE 10 36.70 4.64
36.77 adj.
Task 18 34.50 7.83
35.45 adj.

 

a

ANOVA revealed non-significant

b

ANOVA revealed non-significant

C

differences among means.

differences among means.

ANCOVA revealed non-significant differences among means.

d

Observed means and means adjusted for pre-CDS scores.

Note: The total N262. ”People-minus Pure SE” and "Pure SE”

are subgroups of the "People"

(N=44) classification.

101

Table 2

Stepwise Multiple Regressions:
Three Dependent Variables

 

Dependent
Variable

SIGI
Evaluation

(N862)

Step

Independent
Variable

Sex
Use
Orient

Multiple
Correlation (R)

a
R Square
Increment

Simple
Correlation (r)

Time Spent
on SIGI

(N862)

Orient.
Sex
Use

Post-SIGI
Decisiveness

(N-62)

fiUNH

Pre-CDS
Sex
Use
Orient.

.668
.675
.680
.681

.446
.009
.008
.001

.668
.284
-.072
-.l45

 

Addition of each of the three independent variables to the three equations
failed to account for significant variance in the respective dependent

variables.

Pre-SIGI career decisiveness scores were used as a covariate in this

analysis.

102
Table 3

Means and Standard Deviations:
Three Dependent Variables By Contact With A Career Counselor

 

 

 

fiependent COunselor Standard
Variable Contact N Mean Deviation
a
SIGI
Evaluations
Yes 18 22.94 6.45
No 44 24.20 5.28
b
Time Spent
on SIGI
Yes 18 2.58 .91
No 44 2.25 1.06
c
Post-SIGI
CDS Scores d
Yes 18 34.56 adj. 8.77
No 43 36.47 adj. 7.51
a
ANOVA revealed non-significant differences among means.
b

ANOVA revealed non-significant differences among means.

0

ANCOVA revealed significant differences among means (p<.05)
d -
Means adjusted for pre-CDS scores.

103
Table 4.

Chi-Square Test of the Relationship Between Orientation
and Contact With a Career Counselor

 

 

Task-Oriented People-Oriented
Seeing A Observed = 11 Observed = 24
Counselor Expected = 10.2 Expected = 24.8
Not Seeing Observed = 7 Observed = 20
A Counselor Expected = 7.8 Expected = 19.2

 

Chi-square = .03; df=1; .03 < 3.84 (p<.05)

Results indicate that there is no relationship between
orientation and contact with a career counselor

104

Table 5

Mean Scores on Three Dependent Variables
When Artistic Types are Included and Excluded

 

a
Evaluation Time Post-SIGI CDS

 

Task-oriented including

Artistic types (N=18) 23.89 2.57 35.45
Task-oriented excluding

Artistic types (N214) 25.50 2.90 35.21
People-oriented including 24.22 2.25 36.75

Artistic types (N=44)

People-oriented excluding 22.70 2.35 35.11
Artistic types (N338)

 

a
Means adjusted for pre-SIGI CDS scores

Note: Of the ten Artistic types whose secondary codes were used to
determine task or people orientation, four were classified as
task oriented and six were classified as people oriented.

105

Table 6

Demographics of Participants, Non-Completers,

and Comparison Group SIGI Users

a

 

 

 

 

 

Participants

Sex

Female 49 (71%)

Male 20 (29%)
Race

Non-Minorities 64 (93%)

Minorities S ( 7%)
Class

Freshmen 27 (40%)

Sophomores 28 (42%)

Juniors 7 (10%)

Seniors 5 ( 7%)

Graduate 0 ( 0%)
Previpgs CACGS Use

Yes 12 (19%)

No 52 (81%)
Orientation

Task 18 (29%)

People 44 (71%)
Presently Seeing a Career Counselor

Yes 15 (27%)

No 22 (67%)
Pre-SIGI Mean CDS Score

39.0 (N=62)
Mean Age
19-6 (N=6l)

a

Non-Completers

 

26 (74%)
9 (26%)

28 (85%)
S (15%)

18 (53%)
11 (320)
3 ( 9%)
2 ( 68)
0 ( 08)

8 (248)
26 (76%)

8 (27%)
22 (73%)

11 (338)
22 (67%)

40.2 (N=26)

19.3 (N=33)

a

b
m

382 (60%)
257 (40%)
C
425 (90%)
48 (10%)
253 (41%)
246 (40%)
72 (118)
37 (128)
12 ( 2%)

Not Available
Not Available
Not Available

Not Available

Not Available

This data encompasses student SIGI users during the Spring, Summer,
and Fall, 1985 quarters at the Counseling Center, Brody Hall,
Wonders Hall, and Learning Resource Center.

b

This data encompasses student SIGI users during the Fall, 1985 and
Winter, 1986 quarters at the Counseling Center, Brody Hall, and
Fifteen of the same Learning Resource

Learning Resource Center.

Center SIGI users during the Fall 1985 quarter are included in the
Participant/Non-Completer and Comparison group data.

C

This information was not available from Learning Resource Center.

CHAPTER V

DISCUSSION

This investigation represented an initial examination
of the relationship that exists between individual student
traits and the outcome of CACGS use. The central hypotheses
received little statistical support. It was predicted that
task oriented, in comparison to people oriented
participants, would (a) spend more time on SIGI, (b)
evaluate the program more positively, and (c) demonstrate
greater degrees Of career decisiveness. Mean differences
among groups occurred in the predicted directions, but not
to statistically significant extents.

The secondary hypotheses, which posited a positive
correlation between the three dependent variables, received
more, albeit modest, statistical support. It was found that
statistically significant positive relationships existed
between (a) time spent on SIGI and SIGI evaluations, and (b)
post-SIGI career decisiveness and SIGI evaluations. No such
statistically significant positive correlation was found to
exist between time spent on SIGI and post-SIGI career
decisiveness.

Negative findings nothwithstanding, the results of this
investigation may have career counseling significance and
suggest directions for subsequent research in this
relatively new area of investigation. The following

106

107
discussion focuses on the hypotheses investigated in this
study, addresses the theoretical and methodological
strengths and weaknesses Of this investigation, and
concludes with comments on the implications of these
findings and recommendations for future research.

The generally negative results of this investigation
have undoubtedly been influenced by methodological issues
related to procedure, design, and measurement. In addition,
the major theoretical assumptions contained within this
study bear re-examination, in an attempt to determine the
role they may have played in the generally negative
findings. In this author's opinion, the two factors that
may have primarily contributed to the non-confirmatory
findings are (a) the manner in which participants were
categorized as either task or people oriented, and (b) the
premise that such individuals would differentially respond
to a computerized guidance system. These two issues will be
discussed first, followed by a number of additional
potentially confounding factors.

As discussed in the section on Holland's Theory, there
appears to be no consensus among researchers concerning the
manner in which ”types” are determined based on VPI results.
The Kivlighan et al. (1981) study, upon which the present
study was loosely based, relied solely on primary codes
(i.e., the highest raw score among the six types). The
results, as previously discussed, support the notion of

matching vocational treatment approaches and client

108 r
personality types. Such a classification by high-point code
alone appears to be the favored procedure by researchers.
However, other coding procedures have appeared in the
literature. For example, Cox and Thoreson (1977) defined a
VPI ”single high-point code" as any code 10 percentile
points or 4 raw score points higher than any other code.
These authors excluded from participation any subjects whose
profile did not meet these criteria.

A review of the literature by this author revealed no
compelling argument for or against any particular method of
classification. Holland claims that his early validation
studies "are characterized by the use of multiple dependent
variables and multiple methods for the definition of types"
(1985, p. 59); many of which employed the same high-point
code classification used in the present study. In fact,
those studies that take into consideration Holland's notion
Of "differentiation” generally disregard the proximity of
secondary and tertiary codes.

Differentiation is generally operationally defined as
the arithmetic difference between the highest and lowest
scores on the six occupational scales of the VPI or
Self-Directed Search (although other formulas have been
used) (Lowe, 1981). So, for example, the two following
profiles would result in identical differentiation scores of
12: R213, I212, A212, S212, E212, C21; and R213, I21, A21,
S21, E21, C21. The scores of 12 suggest equally high levels

of differentiation despite the Obvious quantitative

109

differences in the two profiles. Thus, it would seem that
high-point codes, in one context or another, are the favored
method of determining Holland's typologies. Admittedly,
however, this does not speak to the validity (only the
popularity) of such a method.

In clinical use, a VPI interpreter would certainly take
a client's entire profile into consideration, rather than
simply focusing on the primary code. In the above two
RIASEC profiles, the two "Realistic” individuals differ
drastically in their expressions of occupational interests.
However, in the scoring system employed in the present
study, both would be categorized as ”task oriented." This
method also ignores the individual's total number of ”Yes,"
"No,“ or ”Indifferent” (no marks) responses. For example,
the following two individuals would be classified as "task
orientedz' R214, I210, A211, S28, E29, C212; and R21, I20,
A20, S20, E20, C20. The first individual found 64 of the 84
occupations interesting or appealing, the second individual
only 1 of 84. Although he originally proposed that people
resembled one type largely to the exclusion of others,
Holland later suggested that, whereas one type usually
predominates, second and third types Often contribute to the
individual's overall make-up (Weinrach, 1984).

Several explanations can be Offered for why the
majority of researchers, this author included, use only the
primary code in career-related research. First, there

appears to be no empirical evidence to suggest against its

110
use. Second, it results in a higher percentage of usable
participants than other more discriminating methods. Third,
it is extremely easy to determine one's primary code.
Fourth, it appears to be the favored method of
classification in previous investigations. Admittedly,
these four explanations are hardly sufficient justification
for the use of primary codes alone.

Fifth, and perhaps most importantly, is the fact that
no trait-factor approach can ever hope to take into account
all possible variables and contingencies. For example,
whereas strong arguments could be made for including
differentiation and/or consistency in the typological
classification, one could also recommend the inclusion of
any number of equally credible independent variables (e.g.,
number or percentage of ”Yes” responses - which purports to
provide clues into the nature of one's indecisiveness,
congruence with college major, sex, age, grade point
average, etc.). 1

It could be argued that excluding Holland's other ”key
concepts" is an example of pragmatism superseding scientific
rigor. Conversely, however, little empirical support
appears to exist for the use of differentiation and
consistency in career counseling outcome research (see
Holland's theory section). The relative simplicity and
"elegance” of Holland's theory has undoubtedly contributed
to its acceptance in applied and research settings.

However, those very qualities may lend themselves to

111
simplistic and/or questionable applications in counseling
and research efforts.

Kazdin (1980) claims that the "primary task of research
design and the goal of a given experiment" is ”minimizing
ambiguity” (p. 4). It would certainly be desirable for the
career counselor or researcher to be able to
(non-ambiguously) categorize people, with 100% accuracy and
confidence, as 'task' or "people" oriented. But even if
this unlikely task was achieved in the present study, it
does not address the influence and interaction of
potentially confounding variables. As Kazdin also states,
"The better an experient is designed, the fewer the
alternative plausible explanations that can be advanced to
account for the findings" (p. 3). Thus, perhaps in an
attempt to ”minimize ambiguity" by dichotomously classifying
people as task or peOple oriented based on their responses
to a brief interest inventory, this author may have been
guilty of (a) failing to design the experiment so as to
minimize confounding explanations, and (b) oversimplifying
complex human characteristics. However, a review of the
extant research reveals that these shortcomings are in no
way unique to the present study.

This leads to the second major proposed explanation for
noneconfirmatory findings - that the central premise of the
study is flawed. That is, given that task and people
oriented individuals do in fact exist (and were correctly

identified in the present studY), their reaction to SIGI may

112
have been incorrectly predicted. Because SIGI is
structured, mechanized, and does not involve human
interaction, it was hypothesized that the program would not
appeal to, and would be less effective with, people oriented
users. However, several possible flaws in this reasoning
can be identified.

First, unlike other ”inanimate objects" (e.g., books,
appliances, tools, mechanical devices, etc.), later
generation CACGS are interactive. They differentially
respond to user input by providing individualized feedback.
The user knows that his or her interaction is uniquely
geared to his or her responses. The output on screen and
paper is “personalized.” It is not uncommon to hear college
students complaining of the "depersonalization" of college
and university life. SIGI, although non-human, is
individually geared - a uniquely individualized guidance
approach in an increasingly 'depersonalized' environment.

At the risk of belaboring this point, it is also
noteworthy that SIGI is infinitely ”patient.” If, as some
SIGI users have reported, they feel rushed while using SIGI,
this is a function of time constraints imposed by humans
rather than the program itself. In addition, SIGI was
designed to be 'user friendly,“ containing instructions and
procedures that are easily followed. Perhaps the people
oriented individual is more predisposed toward these
positive "human" qualities (interactivity, individualistic

attention, patience, and "user friendliness") than to the

113
simple presence of another human being.

Secondly, the characteristics of task and people
oriented individuals may have implications that confounded
the findings of the present study. For example, Holland
(1985) reported that high vocational aspirations are
positively associated with the types in the following order:
Enterprising, Social, Artistic, Investigative, Conventional,
and Realistic. Thus, people oriented individuals' higher
vocational aspirations may have influenced their use and
reactions to SIGI to a greater extent than their social
orientation. Future studies should attempt to "partial out"
the potential influence of vocational aspirations in its
participants.

Finally, the activities encompassed in SIGI may hold
more appeal to people oriented than task oriented
individuals. Recall that SIGI requires a considerable
amount of self-exploration and values clarification.

Whether involved with a human or a computer, the former
group may react more favorably to such activities than the
latter group. Thus, the self-examining activities involved
in SIGI, rather than their mode Of presentation, may
differentially influence user reactions.

In this author's opinion, these two factors (the method
of assigning types, and the major assumptions concerning
these individuals' reaction to CACGS) underlie the generally
negative findings. The following issues may have influenced

the study's outcome as well.

114

Procedural Issues

Several procedural difficulties arose that were either
unanticipated or seemingly insoluable. As previously
discussed, 35 participants did not complete the required
materials, rendering their data unusable. This is
unfortunate because, in addition to the loss of potentially
valuable data, it is possible that such "noncompleters"
possessed personality characteristics of potential
importance to this study. For example, noncompleters may
have been dissatisfied with SIGI. Their failure to follow
instructions may have reflected this dissatisfaction.
Another possible explanation is that such people who fail to
follow instructions may be poor candidates for CACGS use,
which involve rather methodical and uniform procedures.

An attempt was made to determine if differences existed
between completers and noncompleters regarding eight
variables (sex, race, age, college class, previous CACGS
use, orientation, pre-SIGI career decisiveness, and
concurrent contact with a career counselor). A
non-statistical comparison (see Table 6) reveals no apparent
differences between completers and noncompleters among these
variables. This list is not comprehensive ofcourse (it does
not include any of the three dependent variables), and does
not suggest that other differences may not exist.

In addition, participants were questioned to determine

if they were simultaneously receiving career counseling

115
elsewhere (see SIGI User Information Form, Appendix B,
Question #12). Of those who responded, 73% indicated that
they were not currently seeing a counselor about
career-related issues. Of the 27% who were seeing
counselors at that time, approximately 80% indicated that
they had seen the counselor only once. Therefore, it would
appear that the great majority Of SIGI users in the present
study had received little or no career counseling
immediately prior to, or during, their use of SIGI. NO
relationship was found to exist between contact with a
counselor and orientation (See Table 4).

To further determine if exposure to a career counselor
may have influenced the results of the study, participants'
mean scores on the three dependent variables were compared
on the basis of whether or not they were concurrently seeing
a career counselor (see Table 3). The two groups did not
differ in terms of time or SIGI evaluation scores. However,
it was found that participants who were not seeing
counselors were more indecisive following the use of SIGI
than those simultaneously in contact with counselors. This
finding supports the notion that a CACGS is more effective
in conjunction with career counseling, than as a
“stand-alone" intervention. It also suggests that greater
efforts should have been made to control for this seemingly
meaningful independent variable. Realistically however,
career counseling and information come in many forms and

from many sources (e.g., from academic advisors, friends,

116
parents, library resources, etc.). In a large university
setting, it is unlikely that any screening procedure could
assure that a CACGS (or any career counseling intervention)
represents the participants' sole source of career guidance
or information. The same would be true of any control group
members that are included in any such research design.

On a more positive note, participants in the present
study may not have represented the oft-criticized ”volunteer
personality” common to many counseling outcome studies.
These students were not motivated by externally imposed
conditions such as college credit, money, course
requirements, etc. Instead, they independently and
voluntarily sought the treatment, a fact that increases the
study's generalizability to other college career counselees.

The fact that 71% of the participants were female can
be viewed in two ways. It is known that female college
students are more likely than their male counterparts to
volunteer for psychologically-oriented research (Rosenthal 8
Rosnow, 1975). However, as previously discussed, it has
been found that female college students pursue career
counseling and use CACGS in greater percentages than male
college students. Thus, the female majority in the present
study likely portrays a typical CACGS user group, rather
than necessarily representing a ”volunteer personality"
artifact.

Nonetheless, it is interesting that the comparison MSU

SIGI user group consisted of a smaller percentage Of females

117
(60%) than this study's participant (71%) and non-completer
(74%) groups. Females completed and failed to complete the
forms in similar percentages (71% vs. 74%, respectively).
The same is true for males (29% vs. 26%, respectively).
Thus, there is little basis for the argument that, among
those who voluntarily sought the use of SIGI, females were
more amenable to participation in the study than males. No
explanation for this discrepancy is readily apparent,
although it is likely that a larger sample size would have
resulted in female-male percentages closer to those existing
in the larger comparison group.

The fact that the participants agreed to use SIGI
distinguishes them from the general "career client”
population. Such CACGS volunteers are more likely than a
randomly assigned group of individuals to have high
expectations prior to SIGI use. This may partially account
for the generally favorable ratings in this and virtually
all other studies, and the general increase in decidedness
scores. Some of the participants, aware that their
responses were being recorded, may have tried to please the
investigator by responding in a manner they deemed
desirable. Thus, in this, as in many such investigations,
the findings may have been confounded by the researcher's
inability to determine if the results were due to the
treatment or to the effect of demand characteristics.
Largely because of this, a behavioral dependent variable

(time spent on SIGI) was included in the study.

118

Design Issues

Although causal inferences can not be drawn from
correlational research, such a method can lead to the
discovery of possible causal relationships which may then
become the subject of experimental investigation (Wood,
1983). The oft-cited major shortcoming of the one-group
pretest-posttest design is the investigator's inability to
evaluate whether or not the treatment is related to
performance on the dependent measure.

Two of the present study's three criterion variables
(time spent on SIGI, and SIGI evaluations) required the use
of SIGI, and therefore were not subject to control group
comparisons. Only post-SIGI level of career decisiveness
could have been compared to control members' Career Decision
Scale (CDS) scores over an equivalent period Of time.
Although no control group existed in the present study,
numerous CACGS investigations have demonstrated significant
differences between CACGS users' and control group members'
post-CACGS level of career decidedness and/or career
maturity (Chapman et al., 1977; Myers et al., 1975; Pinder 8
Fitzgerald, 1984; Pyle 8 Stripling, 1976). There also
exists substantial empirical evidence demonstrating the
CDS's sensitivity to career counseling interventions
(Osipow, 1980). Therefore, although the existence of a
control group would have been preferable, there remains

substantial evidence to suggest that pre-post changes in the

119
Career Decision Scale scores were largely a result of the
intervention rather than other ”sources of invalidity”
(Stanley 8 Campbell, 1966).

Previous studies employing control groups would suggest
that the act of completing the CDS prior to a career
counseling intervention does not result in a "practice
effect" which may contaminate the responses to the
post-intervention CDS. Likewise, the nature of the CDS
scoring procedure guards against the occurrence of
"instrumentation” as an additional ”uncontrolled rival
hypothesis.“ In addition, the short-term nature of this
intervention eliminates the potential influence of
"maturation” on pre-post scores. Therefore, it is safe to
assume that these three potential confounding variables did
not influence nor account for this study's generally
negative findings.

Campbell and Stanley (1966) point out that statistical
regression is a particular problem in one-group
pretest-posttest designs when the participants are selected
for their extremity. NO such selection procedure occurred
in the present study. A large variance existed among the
participants' pre-treatment levels of career decisiveness.
This is not to suggest, however, that a regression effect
could not have partially accounted for some pre-post change

scores in the present study.

120

Measurement Issues

The rationale for the classification of Artistic
participants according to their secondary codes appears in
the Methodology section. This procedure, to the best
knowledge of this author, has no precedent in the
literature. A non-statistical comparison was done between
task and people oriented participants' dependent variable
scores when the Artistic types' secondary codes were
included and excluded. No differences appear to exist among
the groups' mean scores (see Table 5). Nonetheless, because
of the unknown validity of such a procedure, its elimination
could only serve to enhance the tenability of such a
classification procedure.

Similarly, the reasons for excluding differentiation
and consistency in the present study appear in the
Methodology section. Because of the reasons listed, no
attempt was made to distinguish participants according to
their profiles' degrees of differentiation or consistency.
It is noteworthy that any such procedure would have resulted
in the elimination of the majority of participants,
rendering the results virtually meaningless. For example,
only 18% (8 of 44) of the people oriented participants'
profiles would qualify as "well differentiated" when a
highest-minus-lowest cut-Off score Of 5 is used (a
"moderate" level of differentiation, at best). Although a

comparison could be made between poorly, moderately, and

121
highly differentiated individuals, there simply is not
enough empirical support to justify such a procedure.

Although consistency was also excluded from the
analyses, the classification of participants as "Pure SE's"
(primary and secondary code of Social and Enterprising, in
either order) demonstrates a recognition of the potential
importance of code consistency within participants'
profiles. As is discussed in the following section (see
Suggestions for Future Research), a larger sample size would
have allowed for the inclusion of differentiation and
consistency in the analyses.

The CDS is a reactive measure, and thus subject to the
potential shortcomings inherent to such instruments. The
very act of completing such a scale may influence an
individual's attitude toward his/her career decisiveness,
identity, exploratory behaviors, etc. However, its proven
sensitivity to changes resulting from career counseling
interventions, as well as its acceptable psychometric
properties, support its use as a dependent measure in such a
study.

The amount of time spent on SIGI served as a less
reactive measure of clients' attitude toward and use of the
system. It is unlikely that a participant would have
invested an extra hour or two on the system merely to please
the experimenter. It is possible, but seemingly unlikely,
that some participants may have purposely over- or

under-reported their usage time. Nonetheless, the manner in

122
which usage time was ascertained was less than ideal. Some
people failed to indicate the amount of time Spent on the
system. In addition, there exists the possibility of
inaccurate reporting Of time. Schedules were checked on a
random basis to determine if students were recording more
time than was available to them. No such discrepancies were
discovered. However, this did not guard against students
reporting more time than they had actually used. For
example, a student with a two hour appointment may only have
used the system for one hour, but recorded two. There is no
strong reason to believe students may have done so, although
the aforementioned demand characteristics might help to
explain its possible occurrence. By relying on self-report,
this intended nonreactive measure took on the potential for
reactivity in some cases. Although the receptionists were
requested to record time of usage, this practice
discontinued almost immediately upon onset of the study.

The intention of examining the relationship between the
reactive measures (decisiveness scores and SIGI ratings) and
time spent on SIGI, was to better determine the credibility
of all three dependent variables. The findings, which
revealed significant positive relationships between time and
ratings, and time and degree of decisiveness, support the
notion that meaning can be derived from an examination of
the amount of time spent on SIGI. However, as is the case
with all the correlational findings reported herein, no

causal inferences can be drawn. For example, the positive

123
relationship between time and ratings permits several
interpretations: (a) To reduce cognitive dissonance, a
person who spends a lot of time on SIGI may report having
enjoyed and profitted from it, regardless of his/her true
attitude and benefit; (b) Students who spend a lot of time
on tasks may also tend to perceive those activities in a
more favorable light; (c) Students who rush through the
system may be reflecting a harried, frenetic disposition,
and a related tendency toward displeasure and
dissatisfaction toward things and events in general; (d)
Students who spend more time on SIGI are those who need it
more, and as such have a greater personal stake in its use
and evaluation. These and other possible interpretations
acknowledge the tentativeness with which all correlational
findings must be treated.

Another shortcoming related to the present study's
examination Of time relates to the differential use of that
time. For example, some psychotherapy clients make
excellent use of long-term therapy, whereas others
maladaptively prolong treatment. An assumption of the
present study was that SIGI is used optimally when used in
its entirety and/or in an unhurried fashion. However, this
is not to suggest that the longer the use, the greater the
benefit. And relatedly, this ignores the manner of use.
One four hour user may divide the time evenly among the five
subsystems, whereas another might skip quickly to a later

subsystem. Neither pattern is necessarily desirable nor

124
undesirable, but rather a function of the user's attitudes
and/or needs. For example, one hour on the STRATEGY section
alone may be all a highly vocationally decisive individual
requires. For a less decisive person, such a practice would
undoubtedly be premature and insufficient.

Ideally, the time spent on each of the system's five
subsections should have been separately examined to
determine any significant patterns of use or preference.

For example, it is reasonable to hypothesize that certain
types Of individuals spend more time on certain subsystems
than others. For example, Social and Enterprising types may
invest greater time and energy on the Values subsystem than
Realistic and Investigative types.

The two criticisms most Often cited about the use of
CACGS evaluation devices as dependent variables in such
studies are their dubious or unknown psychometric qualities,
and their ultimate meaning. The first criticism was
addressed in the present study. In an attempt to provide
evidence of content validity, the investigator-created SIGI
Evaluation Questionnaire was written with items relevant to
the system's content, and supplied by the Educational
Testing Service and MSU's career counselors. Slight yet
statistically significant positive correlations were found
with two other related dependent variables, thus supporting
its concurrent validity. Also, a test of internal validity
which resulted in a Coefficient Alpha of .88, demonstrated

acceptable reliability.

125

The second criticism was only partially addressed.
Correlations such as those examined in the present study
provide added meaning to user ratings. CACGS rating tools
such as this may in fact be reliable, and may be measuring
that which they purport to measure. But there is some
question among authors in the field regarding the meaning
and importance of ratings. Psychotherapists have long known
that it is not sufficient for clients to merely like their
therapists and treatment. Such could also be said regarding
teaching and the provision of medical services. One could
enjoy an intervention and not benefit from it, or dislike
yet profit from it. Why then do CACGS studies, the present
one included, stress the importance of user ratings? There
are several possible explanations:

1. Positive ratings, although not sufficient, are
likely necessary to CACGS's survival. If CACGS are found to
be effective in achieving their particular goals (e.g.,
decreasing indecisiveness, supplying up-to-date information,
clarifying career-related values, etc.) but are somehow
aversive to users, they would likely disappear from use. If
enjoyable 32g effective, CACGS will undoubtedly proliferate.
Because it is more easily demonstrated, researchers have
likely focused on the former rather than the latter.

2. Many CACGS researchers seem unclear about the
purpose of CACGS and therefore do not know what to measure
in order to determine their effectiveness. Ratings are

easily obtained and analyzed. Furthermore, practitioners

126
more than likely need some justification for the initial
substantial expenditure involved in CACGS purchases and
implementation.

3. Evaluation studies generally Obtain positive
ratings. The initial CACGS studies were undertaken or
supported by CACGS developers who had an obvious vested
interest in results demonstrating favorable user reactions.

4. College students are generally reluctant to seek
career services (Brooks 8 Hagler, 1984). Although there are
undoubtedly many reasons for this, positively rated services
such as CACGS can only serve to decrease that historic
reluctance.

5. It is Often reported that counselors dislike
engaging in career counseling activities. They may be more
receptive to career counseling if a procedure exists that
clients clearly enjoy, and which simultaneously decreases
the counselors' active participation in career-related
activities.

Although all of the above may have influenced the
inclusion of SIGI evaluations in the present study, the
nature Of that examination differed from previous
investigations in several ways. First, attempts were made
to substantiate the psychometric qualities of the rating
instrument. Second, the measure was compared to related
dependent measures. Third, the interaction between ratings
and raters' personal traits was examined. The failure to

discover such interactions between ratings and user

127

characteristics may reflect the absence of any such existing
relationship. The same could be said about the present
study's failure to demonstrate a relationship between user
traits and the two other dependent vriables: post-SIGI
career decisiveness and time spent on the system. Before
drawing such conclusions however, many of the deficiencies
of the present study require correction. Such potential

improvements appear in the following section.

Suggestions for Future Research

 

It has been suggested herein that the two primary
reasons for the generally negative findings in the present
study may have been (a) the manner in which SIGI users were
classified according to Holland's occupational types, and
(b) possible flaws in the central attribute-treatment
interaction premise. In order to conclude that the latter
is true, the former issue, as well as other reported
shortcomings, require correction.

Studies such as the present one which involve the
subclassification of participants according to various
psychological and demographic traits, require a larger
sample size than was achieved in the present investigation.
A larger sample size allows the inclusion of more
independent variables in the multiple regression equation,
and increases the power to detect significant effects.

In addition, a larger N would allow the investigator to

128
be more discriminating when classifying individuals
according to Holland's types. For example, the task and
people orientation schema could be replaced by the six
Holland types, a more familiar research procedure. Much
could be learned by separately Observing the correlations
between all six types and the numerous proposed independent
and dependent variables. This would also serve to eliminate
the practice of classifying Artistic individuals according
to their secondary occupational codes. A larger sample size
would also allow for the elimination, or separate analysis,
of participants with poorly differentiated profiles. Thus,
for example, one could compare "Pure E's" with "Pure 1'3"
with Pure C's,” etc.

As discussed in the Procedures section, the present
study was originally designed as a pretest-posttest control
group design, but control participants proved unobtainable.
Optimally, attribute-treatment interaction studies should
involve two or more treatment groups, in addition to a
control group. It would be desirable to have one SIGI group
and one less structured, more interpersonal career
counseling treatment group. Participants could be
classified as peOple or task oriented, with half of each
assigned to one of the two treatment groups. SIGI would be
considered the task-oriented intervention, the group
treatment representing the people-oriented intervention.
Control group members' pre- and post-treatment CDS scores

would be compared to the treatment members' scores. As in

129
the aforementioned Melhus et a1. (1973) study, it would be
hypothesized that people in attribute-consistent treatments
would benefit more than those in attribute-inconsistent
treatments.

The interaction between CACGS and numerous other
relevant personal attributes should be examined. For
example, pre-CACGS expectations and attitudes undoubtedly
influence students' use of, and reaction to the programs.
Also, students should be classified according to reasons for
use; for example, too few or too many career Options, need
for career information, desire to confirm present choice,
etc. It may be discovered, for example, that CACGS are
particularly helpful for students needing to limit their
choices, but less helpful for those wanting to expand their
options.

A largely overlooked area of research involves the
impact of pre-CACGS orientation procedures and/or post-CACGS
debriefing sessions. Both procedures have an intuitive
appeal, but little is known about their value or influence
upon the use of CACGS. Sharf (1984), in a similar manner,
compared user reactions to the same program (Exploring
Careers) with and without the inclusion of occupational
information. He found no significant difference between the
two groups' ratings. If in this case it is found that
orientations and/or debriefings fail to enhance the value of
the systems, their practice could be altered or eliminated.

Conversely, if they are shown to facilitate CACGS use, their

130
inclusion would be warranted. Surprisingly, many of the
existing CACGS studies have paid little or no attention to
the potential influence Of such orientation and debriefing
sessions that were included in the treatment.

The present study demonstrated the advisability Of
using the systems' internal programming to obtain data that
is presently Obtained obtrusively and inefficiently by
paper-and-pencil means (e.g., time Spent on the system,
subsystems used, number and type of occupational titles and
descriptions requested, interest inventory results, etc.).
Having the questionnaires appear on the screen rather than
on separate forms would be a highly efficient means of
procuring information. The data could then be automatically
compiled internally.

The examination Of the process of SIGI use, rather than
solely the outcome, has important ramifications for
subsequent research. Whereas previous research has focused
on how people react to CACGS, subsequent studies should
examine how people use the systems. Data from the present
study, although not reported herein, are available regarding
differential reactions to SIGI's five subsystems. The first
eight items in the SIGI Evaluation Questionnaire relate to
one of the five subsystems, and thus the interaction of
demographic and personal traits is subject to examination.
Unfortunately, the time spent on the five subsystems is
unavailable in the present study.

As is often suggested in the literature, the design of

131

the present study would have been strengthened by the
inclusion of behavioral, nonreactive and longitudinal
outcome measures, such as visits to a career information
center, use Of career resources, number of college major
changes, post-CACGS grade point average, level of
satisfaction with college, etc.

To summarize, it is recommended that future research
(a) include larger numbers of participants, (b) alter the
manner in which participants are classified according to
Holland's typologies, (c) employ two or more treatment
groups in addition to a control group, (d) include
additional relevant independent variables, (e) study the
potential influence of orientation and/or debriefing
procedures, (f) employ the computer's internal programming
for the procurement of data, (9) focus upon the process as
well as the outcome of CACGS use, and (h) examine more

behavioral and longitudinal variables.

APPENDICES

APPENDIX A

CONSENT FORM

132
Appendix A
Dear SIGI User:

We are very interested in evaluating the effectiveness Of
SIGI and learning more about the students who use the
system. The information which you and hundreds of others
supply us with will better enable us to provide quality
career services to MSU students. Please take a few minutes
to complete this Consent Form and the three additional
questionnaires. When you are done with SIGI, you will be
asked to complete two more brief questionnaires. Your use
of SIGI will not be altered in any way, and you will not be
contacted after you complete the questionnaires. Thank you
for your participation.

CONSENT FORM

I understand that the study being conducted by Alan Farber,
in conjunction with the MSU Service Learning Center, is for
the purpose of examining personal, vocational and
demographic characteristics of SIGI users. I also
understand that participating in this study will not result
in direct benefits to me, and I can discontinue my
participation at any time.

I understand that the information I provide will be kept
strictly confidential. General results will be reported,
but none of these will identify individual participants'
results. I understand that only Mr. Farber will have access
to the original data. I also understand that I will upon
request receive a report of this study's general results,
within the restrictions of confidentiality outlined above.

 

 

Signature Name

 

 

Student Number Date

For additional information contact:
Alan Farber, Ph.D. candidate
Counseling Psychology

4th floor Erickson Hall, Desk 13

APPENDIX B

SIGI USER INFORMATION FORM

133

Appendix B

SIB) USER (”FORMATION

   

 

The (oilowi'ng questions are designed to provide demographic iniormation about student 3101 users. Where a
niimber is applied. you need only circle the appropriate umber. Other items need to be completed by you in
the space provided. Thanks.

i. What is you- name?
2. Mat is yoir age (in years and months)?

3. mat is your sex? (1) Female (2) Male

‘

. that is your race? (1) Asian American (2) Black (3) Caucasian (4) Hispanic
(3) Native American (“Other (specity)

S. that is you- college mabr?

4. that is you- college classification?
(i) Freshman (2) Sophomore (3) Junior (4) Senior (5) Other (specify)

7. that is your overall grade point average (GPA)?
8. Have you ever used 5101 or another computer-ind career guidance system? (i) Yes (2) Mo

9. How much experience do you have with computers? (e.g.o Mamming. word processing. spread sheets. etc.)
(i) A lot (2) Some (3) Very little (4) None

(0. How much time have you spent playing video games?
(i) A lot (2) Some (3) Very little (4) None

ii. How did you hear about 310)?
(i) Friend (2) Career Counselor (3) Teacher/Class
(4) Resident Advise (3) Academic Advisor (6) Advertisement

(2. mich o4 the (ensuing is true o4 you. regarding you- seeing a coinseior about career-related issues?
(i) I am cirrently seeing one on an ongoing basis.
(2) i am not currently seeing one. but have in the past.
mleeneverseenonemutplanondou-igsointheneartutire.
(4))havenever ummunddon'tpiantodosointheme.
B) OM (Sanity)

APPENDIX C

CAREER DECISION SCALE

134

Appendix C

CAREER DECISION

GCopyrlght by Samuel H. Osipow, Clarke G. Carney. Jane Winer. Barbara Yanico. and
Maryanne Koschier 1 976 (3rd revision) all rights reserved

Marathon Consiltlng 8. has:
F.0.onfl1'
comm Ohio 432“ 201‘

Your Name Today's date __Your date of birth

This questionnaire contains some statements that people commonly make about their
educational and occupational plans. Some of the statements may apply to you; others may
not. Please read through them and indicate how closely each item describes you in your
thinking about a career or an educational choice by circling the appropriate number on the
answer sheet.

If you are excited about going to work and feel no hesitation about it you would
circle "4" as it is circled in the example on the next page to indicate the
description was exactly the way you feel. lithe item is very close. but not exactly
the way you feel—for example, you're generally excited about going to work after
you graduate. but are experiencing some minor concerns about it—you would
circle the number ”'13 You would circle "2" if the item described you in some
ways. but in general it was more unlike than like your feelings; for example, if
you were generally more concerned than excited about work after graduation.
Finally, you would circle ”1 " it the item did not describe your feelings at all; that
is. you were experiencing a great deal of concern and no excitement about
graduation and work.

135

An example is given below:

Sample Answer
Sample Self-Description Item

. _ Exactly Very much Only slightly Not at all
I am excned about graduatmg like me like me like me like me
and going to work.

4 3 2 1

If you change your answer, please be sure that all previous marks are completely erased.
Please give only one response to each item and respond to all items.

CIRCLE ANSWER

Like Me Not Like Me

1. l have decided on a career and feel comfortable with it. I also 4 3 2 1
know how to go about implementing my choice.

2. l have decided on a major and feel comfortable with it. I also know 4 3 2 1
how to go about implementing my choice.

3. If I had the skills or the opportunity. I know I would be a _ 4 3 2 1
but this choice is really not possible for me. I haven't given much
consideration to any other alternatives. however.

4. Several careers have equal appeal to me. I'm having a difficult 4 3 2 1
time deciding among them.

5. I know I will have to go to work eventually. but none of the careers 4 3 2 1
I know about appeal to me.

136

REMEMBER — 4 is exactly like me. 3 is very much like me. 2 is only slightly like me,
and 1 is not at all like me.

6. I'd like to be a _ but I'd be going against the 4
wishes of someone who is important to me if I did so. Because of
this, it's difficult for me to make a career decision right now. I
hepe I can find a way to please them and myself.

7. Until now, I haven't given much thought to choosing a career. I 4
feel lost when I think about it because I haven't had many
experiences in making decisions on my own and I don't have
enough information to make a career decision right now.

8. I feel discouraged because everything about choosing a career 4
seems so "ify" and uncertain; I feel discouraged, so much so that
I'd like to put off making a decision for the time being.

9. I thought I knew what I wanted for a career, but recently I found 4
out that it wouldn't be possible for me to pursue it. Now I've got to
start looking for other possible careers.

10. I want to be absolutely certain that my career choice is the "right" 4
one, but none of the careers I know about seem ideal for me.

11. Having to make a career decision bothers me. I'd like to make a 4
decision quickly and get it over with. I wish I could take a test that
would tell me what kind of career I should pursue.

12. I know what I'd like to major in, but I don't know what careers it 4
can lead to that would satisfy me.

137

REMEMBER — 4 is exactly like me, 3 is very much like me. 2 is only slightly like me,
and 1 is not at all like me.

13. I can't make a career choice right now because I don't know what 4
my abilities are.

14. I don't know what my interests are. A few things ”turn me on" 4
but I'm not certain that they are related in any way to my career
possibilities.

15. So many things interest me and I know I have the ability to do 4

well regardless of what career I choose. It's hard for me to find
just one thing that I would want as a career.

 

16. I have decided on a career. but I'm not certain how to go about 4
implementing my choice. What do I need to do to become a
anyway?
17. I need more information about what different occupations are like 4

before I can make a career decison.

18. I think I know what to major in, but feel I need some additional 4
support for it as a choice for myself.

APPENDIX D

VOCATIONAL PREFERENCE INVENTORY

Reproduced by special permission of the publisher,
Psychological Assessment Resources, Inc., Odessa, FL 33556,
from the Vocational Preference Inventory by Dr. John L.
Holland, Ph.D., Copyright 1978.

138

Appendix D

THE VOCATIONAL PREFERENCE INVENTORY
Developed by john L. Holland, Ph.D.

This is an inventory of your feelings and attitudes about many kinds of work. Fill out your
answer sheet by following the directions given below:

1. Show on your answer sheet the occupations which interest or appeal to you by blackening

Y for "Yes."

2. Show the occupations which you dislike or find uninteresting by blackening N for "No."
3. Make no marks when you are undecided about an occupation.

 

Criminologist

Private Investigator

. Restaurant Worker
Detective

Photoengraver '
Truck Gardener

Physical Education Teacher
. Humorist

Photon-eh"

. Diplomat

fl
°P°N9SPPUPP

. Airplane Mechanic
. Met ’

13. Sociologist

14. Bookkeeper

15. Speculator

16. Poet

17. Deep Sea Diver
18. Stock Clerk

19. Dramatic Coach

20. Lawyer

Fish and Wildlife Specialist
Biologist

High School Teacher
Business Teacher

Buyer

Symphony Conductor

. Wrecker (Building)

. Veterinarian

.. Elementary School Teacher
. Physician '

ands-a
Ne—I

SSENRESERE

U
us

. Auto Mechanic

32. Astronomer

33. juvenile Delinquency Expert

34. Budget Reviewer

35. Advertising Executive

36. Musician

37. Prizefighter

33. Post Office Clerk

39. Experimental Laboratory Engineer
40. Bartender

77
78
79
80

3§¥QRP 3$$§$$$$$$ $$$§$$¥$$3

. Carpenter
. Medical Laboratory Technician

. Speech Therapist

. Certified Public Accountant

. Manufacturer’s Representative
. Author

. Firefighter

. Airline Ticket Agent

. Entertainer

. Novelist

. Power Shovel Operator
Anthropolosm
Marriage Counselor
Credit Investigator
Television Producer
Commercial Artist

Wild Animal Trainer
Administrative Assistant
Physical Therapist
Cashier

Surveyor

Zoologist

School Principal

Court Stenographer
Hotel Manager
Free-Lance Writer

Stunt Man/Stunt Woman (Movies)
Route Salesperson

Professional Athlete
Flight Attendant

Construction inspector
Chemist

Playground Director
Bank Teller

Business Executive
Musical Arranger

. jockey

. Interior Decanter

. Airplane Pilot

. Banker

 

(Continued)

THE VOCATIONAL PREFERENCE INVENTORY (ContlnuOdI

 

81. Radio Operator
82. Independent Research Scientist
83. Clinical Psychologist
84. Tax Expert
85. Restaurant Manager
. journalist
. Motorcycle Driver
Department Store Manager
. Referee (Sporting Events)
Mail Carrier

Filling Station Worker
Writer of Scientific Articles
Sodal Sa'ence Teacher
inventory Controller
Master of Ceremonies
Portrait Artist

Blaster (Dynamiter)

. Police Officer

English Teacher

. U.N. Official

§sssraxsss 8838:

101. Tree Surgeon

102. Editor of a Scientific journal
103. Director of Welfare Agency
104. lBM Equipment Operator
105. Salesperson

106. Concert Singer

107. F.B.l. Agent

103. Probation Agent

109. Astronaut

110. College Professor

111. Long Distance Bus Driver
112. Geologist

113. Youth Camp Director
114. Financial

115. Real Estate Salesperson
116. Composer

117. Mountain Climber

118. Cook/Chef

119. Stage Director

120. Ticket Agent

121. Locomotive Engineer
122. Botantst

12.3. Personal Counselor
124. Cost Estimator

125. Publicity Director

126. SculptorlSculptress
127. Explorer

128. Nursery School Teacher
129. Quality Control Expert
130. judge

131. Machinist

132. Sdentific Research Worker
13. Psychiatric Case Worker
134. Payroll Clerk

135. Sports Promoter

136. Playwright

137. Test Pilot

138. Computer Programmer
139. Clothing Designer

140. Truck Driver

141. Electridan

142. Physicist

143. Vocational Counselor
144. Bank Examiner

145. Sales Manager

146. Cartoonist

147. Racing Car Driver
148. Forester

149. Social Worker

150. Sales Clerk

151. Funeral Director

152. Mind Reader

153. Architect

154. Shipping 4: Receiving Clerk
155. Criminal Psychologist

156. Insurance Clerk

157. Barber

158. Bill Collector

159. Ward Attendant

160. MasseurIMasseuse

 

ms II. Minuteman” hmu-m.~uam~~meu~neam

CONSULTING PSYCHOLOGISTS mass

3 arcane-Ammalemcwemlastaos

APPENDIX E

SIGI EVALUATION QUESTIONNAIRE

140

Appendix E

? 51m EUHLUHIIDH cuesnunnnme

How usetul was SIGI in accomplishing the iollowrng’ (circle i-é. per item:

   

”Extremely Useful 28Very Useful 3850mewhat Useful 4-Not at all Useiul

Helping you decide which occupation to prepare tor. i 2 3 4
Helping you become more aware at you‘ values. 1 2 3 4
Showing you the relationship between values and career decisions. 1 2 3 4
Helping you tind out which occupations might tit your values. 1 2 3 4
Helping you get information about occinations. ‘ l 2 3 4
Helping you plan a program appropriate tor an occtnation you i 2 3 4

are considering.

Helping you learn how to make career decisions. 1

O J
.U
.

 

How interesting was 5161 to you?

_ Extremely Very Somewhat Not at all

How clear was 5161 in proViding intonation?

_ Extremely Very _ Somewhat Not at all

How likely would you be to recommend 8101 to a friend?

_ Extremely _ Very __ Somewhat __ Not at all
Overall. how would you rate 5161?
_ Excellent _ Good Fair _ Poor

How many total hours did you spend on 5101? (please round o" to me nearest quarter hour):

Over how many sessions did you use SIGI? _l 2 3 ___4 or more

Are you cu-rently seeing a cotnselor about career issues? __ Yes No

it yes. how many times have you seen him/her since you began using 5101?

LI ST 01“ REFERENCES

LIST OF REFERENCES

Abramowitz, C.V., Abramowitz, S.I., Roback, H.B. & Jackson,
C. (1974). Differential effectiveness of directive and
nondirective group therapies as a function of client
internal-external control. Journal of Consulting and
Clinical Psychology: 42, 849-853.

 

Abramowitz, C., Abramowitz, S. & Roback, H. (1974).
Differential effectiveness of directive and nondirective
group therapies as a function of client internal-external
control. Journal of Consulting and Clinical Psychology,
42L 849-853.

 

ACT (American College Testing Program). (1974). Assessment
of career development handbook user's guide and report of
research. Boston, MA: Houghton Mifflin.

 

 

Adams, C. (1978). Women's educational equity act: The SIGI
component. Dallas, TX: Eastfield College Counseling
Department. . ‘

 

 

Argyris, C. (1976). Increasing leadership effectiveness. NY:
Wiley.

 

Ashby, J.D., Wall, H.W. & Osipow, S.H. (1966). Vocational
certainty and indecision in college freshmen. Personnel
and Guidance Journal, 44, 1037-1041.

 

 

Barak, A., Carney, C.G. & Archibald, D. (1975). The
relationship between vocational information seeking and
educational and vocational decidedness. Journal of
Vocational Behavior, 7, 149-159.

 

Barak, A. & Friedkes, R. (1981). The mediating effects of
career indecision subtypes on career-counseling
effectiveness. Journal of Vocational Behavior, 20,
120-128.

Barrett, T.C. & Tinsley, H.B. (1977). Vocational
self-concept crystallization and vocational indecision.
Journal of Counseling Psychology, 24, 301-307.

Bass, B.M. (1961). Manual for the Orientation Inventory.
Palo Alto, CA: Consulting Psychologist's Press.

141

142

Bellotto, B. (1983). Guidance and counseling directory of
microcomputer software. San Jose, CA: Santa Clara County
Office of Education.

Berzins, J.I. (1977). Therapist-patient matching. In A.S.
Gurman, & A.M Razin (Eds.), Effectivegpsychotherapy: a
handbook of research. Oxford: Pergamon Press.

 

Best, J.A. (1975). Smoking modification procedures tailored
to subject characteristics. Behavior Therapy, 3;

 

Borg, W.R. & Gall, M.D. (1983). Educational research: An
introduction. New York: Longman.

 

 

Bouchard, T. (1978). Review of The Orientation Inventory. In
0. Buros (Ed.) The eighth mental measurements yearbook
(Vol. 2). NJ: Gryphon Press.

 

Bowlsbey, J., Rayman, J. & Bryson, D. (1976). DISCOVER field
trial report, Westminster. MD: DISCOVER Foundation, Inc..

 

 

Brooks, L., & Haigler, J. (1984). Contract career
counseling: An option for some help seekers. The
Vocational Guidance Quarterly, 33, 178-182.

 

Brown, R. (1969). Effects of structured and unstructured
group counseling with high- and low-anxious college
underachievers. Journal of Counseling Psychology! 16,
209-214.

Bruch, M. (1978). Holland's typology applied to
client-counselor interaction: Implications for counseling
men. The CounselingAPsychologist, .ZL 26-32.

 

Butcher, E. & Hebert, D.J. (1985). Locus of control of
similarity and counselor effectiveness: A matched case
study. Journal of Counseling and Develgpment, 64,
103-108.

Cairo, P. (1983). Evaluating the effects of
computer-assisted counseling systems: A selective review.
The Counseling_Psycholggist, 11, 55-59.

Campbell, R.E. (1974). Career guidance practices
transcending the present. Vocational Guidance Quarterly,
22, 292-300.

Campbell, D.T. & Stanley, J.C. (1963). Experimental and
Quasi-Experimental Designs for Research. Chicago, IL:
Rand McNally.

 

Carney, C.G. (1977). Changes in career decision scales
compared. Unpublished data, Ohio State University.

 

143

Carney, C.G., Savitz, C.J. & Weiskott, G.N. (1979).
Students' evaluations of a university counseling center
and the intentions to use-its programs. Journal of
Counseling Psychologyr 26, 242-249.

Chapman, W., Katz, M.R., Norris, L. & Pears, L. (1977).
SIGI: field test and evaluation of a computer-based
system of interactive guidance and information (2
volumes). Princeton, NJ: Educational Testing Service.

Clyde, J. (1979). Computerized Career Information and
Guidance Systems. Columbus, OH: NCRVE.

Cochran, D., Hoffman, 5., Strand, K. & Warren, P. (1977).
Effects of client/computer interaction on career
decision-making processes. Journal of Counseling
Psychology, 24, 308-312.

 

Colby, K.M. (1964). Psychotherapeutic processes. Annual
Review of Psychology, 15, 347-370.

Cole, N.S. & Hanson, G.R. (1971). An analysis of the
structure of vocational interests. Journal of Counseling
Psychology, 18, 478-487.

 

 

Cooley, W.W. (1964). Research frontier: Current research on
the career development of scientists. Journal of
Counseling Psychology: 11, 88-93.

 

Cooper, J.F. (1976). Comparative impact of the SCII and the
vocational card sort on career salience and career
exploration of women. Journal of CounselingyPsychology,
23L 348-352.

 

C00per, S.E. (1986). The effects of group and individual
vocational counseling on career indecision and personal
indecisiveness. Journal of College Student Personnel,
27, 39-42.

 

Coovert, M.D. & Goldstein, M. (1980). Locus of control as a
predictor of users' attitude toward computers.
Psychological Reports, 47, 1167-1173.

Cox, J. & Thoreson, R. (1977). Client-counselor matching: a
test of the Holland model. Journal of Counseling

 

Crites, J.O. (1973). Career Maturity Inventory: Theory and
research handbook. Monterey, CA, CTB/McGraw-Hill.

Crites, J.O. (1978). Review of the Self-Directed Search. In
0. Buros (Ed.) The eighth mental measurements yearbook
(Vol. 2). NJ: Gryphon Press, p. 1608.

144

Cronbach, L.J. 8 Snow, R.E. (1977). Aptitudes and
instructional materials. New York: Wiley.

 

 

Daane, C.J. (1972). Vocational Exploration Group kit: Theory
and research. Tempe, AR: Studies for Urban Man.

 

 

Devine, H. (1975). The effects of a computer-based career
counseling program on the vocational maturity of
community college students. Dissertation Abstracts
International, 7865-A.

 

 

Diamond, E.E. (Ed.) (1975). Issues of sex bias and sex
fairness in career interest measurement. Washington, DC:
Department of Health, Education and Welfare, National
Institute of Education, U.S. Government Printing Press,
Spring.

 

DiLoreto, A.O. (1971). Comparative psychotherapy: An
experimental analysis. Chicago: Aldine-Atherton.

 

 

Drake, J.W. (1978). The effects of a computer-assisted
career guidance support system upon the vocational
maturity of high school sophomores. Dissertation
Abstracts International, 5947A.

 

Dunnette, M.D., Kirchner, W.R. & DeGidio, J. (1958).
Relations among scores in Edward's Personal Preference
Schedule, California Psychological Inventory, and Strong
Vocational Interest Blank for an individual sample.
Journal of Applied Psychology, 43L 178-181.

Edwards, K.J. & Whitney, D.R. (1972). A structural analysis
of Holland's personality types using factorial and
configural analysis. Journal of Counseling Psyghology,
lggL 136-145.

English, T. (1974). A comparison of the effects of two
methods of disseminating occupational information on the
vocational maturity of senior high school students.
Dissertation Abstracts International, 35L 1976A.

Fiedler, F.A. (1950). A comparison of therapeutic
relationships in psychoanalytic, non-directive, and
Adlerian therapy. Journal of Consulting Psychology, 14,
436-445.

Fiedler, F.A. (1951). Factor analyses of psychoanalytic,
non-directive, and Adlerian therapeutic relationships.
Journal of Consulting Psychology, 15, 32-38.

Fiedler, F.A. (1967). A theory of leadership effectiveness.
New York: McGraw-Hill.

145

Fretz, B. (1981). Evaluating the effectiveness of career
interventions. Journal of Counseling Psychologyr 28,
77-90.

Gaffey, R.L. (1972). An investigation of the concurrent
validity of Holland's theory. Dissertation Abstracts
International, 33, 54903.

 

 

Garfield, S.L. (1971). Research on client variables in
psychotherapy. In S. Garfield & A. Bergin (Eds.) Handbook
ofgpsychotherapy and behavior change. New York: John
Wiley & Sons.

Garis, J. (1982). The integration of the DISCOVER
computer-based guidance system in a college counseling
center - its effect uppn career planning. University
Park, Pa.:The Pennsylvania State University.

 

Gilbreath, S. (1967). Group counseling, dependence, and
college male underachievement. Journal of Counseling
Psychology, 14, 449-453.

 

GIS Resource Kit. (1981). Avon Park, CT: Time Share
Corporation.

 

Glaize, D. & Myrick, R. (1984). Interpersonal groups or
computers? A study of career maturity and career
decidedness. The Vocational Guidance Quarterl , 32,
168-176.

 

Grandy, T.G. & Stahmann, R.E. (1974). Family influence on
college students' vocational choice: predicting Holland's
personality types. Journal of College Student Personnel,
15, 404-409.'

 

Grisham, D.W. (1978). The effects of a computer-based career
guidance program upon the career planning and decision
making skills of economically disadvantaged youth.
Dissertation Abstracts International, 32, 3373A.
(University Microfilms No. 7822903).

Gurman, A.S. & Razin, A.M. (1977). Effective psychotherapy:
a handbook of research. Oxford: Pergamon Press.

Guthrie, W.R. & Herman, A. (1982). Vocational maturity and
its relationship to Holland's theory of vocational
choice. Journal of Vocational Behavior, 21: 196-205.

Haring-Hidore, M. (1984). In pursuit of students who do not
use computers for career guidance. Journal of Counseling
and Development, 63, 139-140.

 

Harman, R.L. (1973). Students who lack vocational identity.
Vocational Guidance Quarterly, 21, 169-173.

146

Harren, V.A. (1966). The vocational decision-making process
among college males. Journal of Counseling Psychology,
13, 271-277.

Harren, V.A. (1976). Preliminary manual for interpretation
of the assessment of career decision making (Form B) .
Unpublished manuscript. Southern Illinois University.

Harren, V.A. (in press). Assessment of Career Decision
Making. Los Angeles: Western Psychological Services.

 

Harris, J.E. (1971). Toward guidelines for computer
involvement in guidance. (Commission Report).
Washington, DC: National Vocational Guidance Association.

 

Harris, J.E. (1972). Analysis of the effects of a
computer-based vocational information system on selected
aspects of vocational planning. Dissertation Abstracts
International, 3;, 4089A-4090A.

 

Harris, J.E. (1974). The computer: Guidance tool of the
future. Journal of Counseling Psychology, 21, 331-339.

Harris-Bowlsbey, J. (1983). A historical perspective. In
C.S. Johnson (Ed.), Microcomputers and the school
counselor. Alexandria, VA: American School Counselor
Association.

 

Hebert, D.J. (1968). The counseling relationship as a
function of client-counselor personality need and sex
similarity. Dissertation Abstracts International, 23,
4479A-4480A.

 

Henkels, M.T., Spokane, A.R. & Hoffman, M.A. (1981).
Vocational identity, personality, and preferred mode of
interest inventory feedback. Measurement and Evaluation
in Guidance,r $l4.712‘715r

 

Herr, E. & Cramer, S. (1983). Career guidance and counseling
through the lifeSpan. Boston: Little-Brown.

Hershenson, D.B. (1967). Sense of identity, occupational
fit, and enculturation in adolescence. Journal of
Counseling Psychology: 14, 319-324.

Holland, J.L. (1960). The relation of the Vocational
Preference Inventory to the Sixteen Personality Factor
Questionnaire. Journal of Applied Psychology, 44,
291-296.

Holland, J.L. (1962). Some explanations of a theory of
vocational choice: I. one- and two-year longitudinal
studies. Psychological Monographs, ‘16 (26, Whole No.
545).

147

Holland, J.L. (1966). Psychology of Vocational Choice.
Waltham, MA: Blaisdell.

Holland, J.L. (1973). Making vocational choices: A theory,of
careers. Englewood Cliffs, N.J.: Prentice-Hall.

 

Holland, J.L. (1974). Occupations Finder. Palo Alto, CA:
Consulting Psychologists Press.

 

Holland, J.L. (1978). Manual for the Vocational Preference
Inventory. Palo Alto, CA: Consulting Psychologists Press.

 

Holland, J.L. (1979). Professional manual for the
Self-Directed Search (2nd ed.). Palo Alto, CA: Consulting
Psychologists Press.

 

 

Holland, J.L., Daiger, D.C. 8 Power, P.G. (1980). My
Vocational Situation. Palo Alto, CA: Consulting
Psychologists Press.

 

Holland, J.L. 8 Holland, J.E. (1977). Vocational indecision:
More evidence and speculation. Journal of Counseling
Psychology; 24, 404-414.

 

Holland, J.L., Magoon, T.M. 8 Spokane, A.R. (1981).
Counseling psychology: Career interventions, research
theory. Annual Review of Psychology, 32, 279-305.

Holland, J.L. 8 Whitney, D.R. (1968). Changes in the
vocational plans of college students: Orderly or random?
ACT Research Report No. 25, Iowa City, IA: The American
Coliege Testing Program.

 

Holland, J.L., Whitney, D.R., Cole, N.S. 8 Richards, J.M.
(1969). An empirical occupational classification derived
from a theory of personality and intended for practice
and research. ACT Research Report No. 29, Iowa City, IA:
The American College Testing Program.

Hunt, D.E. (1977). A paradigm fo developing and analyzing
differential treatment programs. In C.S. Davis 8 M.S.
Schmidt (Eds.), Differential treatment of drug and
alcohol abusers. Palm Springs, CA: ETC Publication.

Jacobson, M.C. 8 Grabowski, B.T. (1982). Computerized
systems of career information and,guidance: A
state-of-the-art. Evanston, IL: Northwestern University.
(ERIC Document Reproduction Service No. ED 210 552)

 

James, E.M. 8 Smith, J.C. (1972). Traditional versus
computer-based vocational guidance and counseling
systems: Implications for disadvantaged youth. East
Lansing, MI: Michigan State University. (ERIC Documents
Reproduction Service No. ED 089 150).

148

Jenson, B.T., Coles, G. 8 Nestor, B. (1955). The criterion
problem in guidance research. Journal of Counseling
Psychology, 2, 58-61.

 

Johnson, C.S. (Ed.). (1983). Microcomputers and the school
counselor. Alexandria, VA: American School Counselor
Association.

 

 

Johnson, J.A., Smither, R. 8 Holland, J.L. (1981).
Evaluating vocational interventions: A tale of two career
development seminars. Journal of Counseling Psychology ,
38, 180-183.

 

Jones, H. (1969). The relationship of counselor-client
personality to counseling process and outcome.
Dissertation Abstracts International, 29, 2962A.

 

Jones, S.N. (1976). Computer-assisted vocational counseling
in two Wyoming high schools. Dissertation Abstracts
International, 37, 8A.

 

 

Jones, L.K. 8 Chenery, M.F. (1980). Multiple subtypes among
vocationally undecided college students: A model and
assessment instrument. Journal of Counseling Psychology, .
21, 469-477.

 

Kapes, J.T., Borman, C.A. 8 Kimberly, R.M. (1985).
Evaluation of microcomputer based career guidance systems
with college students: SIGI and Discover. Paper presented
at the annual meeting of the American Educational
Research Association, Chicago.

 

 

Katz, M.R. (1963). Decisions and values: A rationale for
secondary school,guidance. New York: College Entrance
Examination Board.

 

 

Katz, M.R. (1966). A model of guidance for career
decision-making. Vocational Guidance Quarterly, No. 15,
2-10.

 

Katz, M.R. (1969). Can computers make guidance decisions for
students? College Board Review, no. 72, 13-17.

 

Katz, M.R. (1974). Career decision-making: A computer-based
system of interactive guidance and information (SIGI). In
Measurement for self-understanding_and personal
develOpment, Proceedings of the 1973 Invitational
Conference. Princeton, NJ: Educational Testing Service,
pp. 43-69.

 

 

Katz, M.R. (1981). SIGI information bulletin. Princeton, NJ:
Educational Testing Service.

 

149

Kazdin, A.E. (1980). Research design in clinical psychology.
New York: Harper 8 Row.

Kimes, H.G. 8 Troth, W.A. (1974)._Relationship of trait
anxiety to career decisiveness. Journal of Counseling
Psychology: 21, 277-280.

Kivlighan, D., Hageseth, J., Tipton, R. 8 McGovern, T.
(1981). Effects of matching treatment approaches and
personality types in group vocational counseling. Journal
of Counseling Psychology, 28, 315-320.

Krumboltz, J. (Ed.) (1966). Revolution in counseling.
Boston: Houghton Mifflin.

Lacey, D.W. (1971). Holland's vocational models: A study of
work groups and need satisfaction. Journal of Vocational
Behavior, 1, 105-123.

 

Lackner, K.G. 8 Feller, R. (1984). The information age: Will
guidance, counseling, and placement personnel respond?
Awareness: The Journal of the Colorado Association for
Counseling,and Development, g2, 34-38.

Lambert, R.H. (1984). Using microcomputers for career
guidance information. In R. Lambert 8 J. Rodenstein
(Eds.) Microcomputers: Applications in career counseling
and career edugation. Madison, WI: Vocational Studies
Center, pp. 225-254.

 

Lewin, K. (1935). A dynamic theory of,personality. New York:
McGraw-Hill.

Lewin, K. (1936). Principles of topological psychology. New
York: Harper 8 Row.

Limberg, D.F. (1980). A comparison of various student groups
on career indecision. Unpublished master's thesis, East
Tennessee State University.

Lunneborg, P.W. (1975). Interest differentiation in high
school and vocational indecision in college. Journal of
Vocational Behavior, 1, 297-303.

Maccoby, E.E. 8 Jacklin, C.N. (1974). The psychology of sex
differences. Stanford: Stanford University Press.

 

MacDonald, D.J. (1984). A test of client-counselor matching
according to Holland's theory, Unpublished doctoral
dissertation, Michigan State University.

150

Maier, D. 8 Herman, A. (1974). The relationship of
vocational decidedness and satisfaction with dogmatism
and self-esteem. Journal of Vocational Behavior, 5,

 

Maola, J. 8 Kane, G. (1976). Comparison of computer-based
versus counselor-based occupational information systems
with disadvantaged vocational students. Journal of
Counseligg Psychology, 2;, 163-165.

 

 

Marin, P. (1984). The differential effectiveness of
computer-based career counseling intervention and
decision making style on progress in career decision
status. Unpublished doctoral dissertation, University of
-¥——¥——

Michigan, Ann Arbor.

 

 

 

Maze, M. 8 Cummings, R. (1982). How to select a computer
assisted career guidance system. Madison, WI: Wisconsin
Vocational Studies Center.

 

Maze, M. 8 Perlmutter, D. (1983). Managing a computerized
guidance program. In C.S. Johnson (Ed.), Microcomputers
and the school counselor. Alexandria, VA: American School
Counselor Association.

 

 

McKinlay, B. 8 Franklin, P.L. (1975). Education components
for a career information system. Eugene, OR: University
of Oregon. (ERIC Document Reproduction Service No. ED 129
980)

 

Meichenbaum, D.H., Gilmore, J.B. 8 Fedoravicius, A. (1971).
Group insight versus group desensitization in treating
speech anxiety. Journal of Consulting and Clinical
Psychology: 36, 410-421.

 

Melhus, G., Hershenson, D. 8 Vermillion, M. (1973). Computer
assisted versus traditional vocational counseling with
high and low readiness clients. Journal of Vocational
Behavior, 3, 137-144.

Mendonca, J.D. 8 Siess, T.F. (1976). Counseling for
indecisiveness: problem-solving and anxiety-management
training. Journal of Counseling Psychology, 23, 339-347.

Minor, F.J. (1970). An experimental computer-based
educational and career exploration system. In D.B. Super
(Ed.), Computer-assisted counseling. New York: Teachers
College Press, Columbia University.

Myers, R.A. (1978). Exploration with the computer. The
Counseling Psychologist, 1, 51-55.

 

151

Myers, R., Lindeman, R., Thompson, A. 8 Patrick, T. (1975).
Effects of educational and career exploration system on
vocational maturity. Journal of Vocational Behavior, g,
245-254.

Neice, D. 8 Bradley, R. (1979). Relationship of age, sex,
and educational groups to career decisiveness. Journal of
Vocational Behavior, 14, 271-278.

 

 

Nickerson, N. 8 Davis, D. (1983). System of Interactive
Guidance and Information: 1980-1982. University Center,
MI.

 

 

Nie, N.H., Hull, C.H., Jenkins, J.G., Steinbrenner, K. 8
Bent, D.H. Statistical Package for the Social Sciences.
(2nd ed.).New York: McGraw-Hill.

Nunnally, J.C. (1978). Psychometric theory. New York: McGraw
Hill.

 

Oliver, L.W. (1979). Outcome measurement in career
counseling research. Journal of Counseling Psychology,

 

Osipow, S.H. (1973). Theories of career development. (2nd
ed.). New York: Appleton Century-Crofts.

Osipow, S.H. (1980). Manual for the Career Decision Scale
(2nd ed.). Columbus, OH: Marathon Consuiting and Press.

Osipow, S.H. (1983). Theories of career development (3rd
ed.). Englewood Cliffs, NJ: Prentice-Hall, Inc.

 

Osipow, S.H., Carney, C. 8 Barak, A. (1976). A scale of
educational-vocational decidedness: a typological
approach. Journal of Vocational Behavior, 2, 233-243.

Osipow, S.H., Carney, C., Winer, J., Yanico, B. 8 Koschier,
M. (1976). The career decision scale (3rd rev.).
Columbus, OH: Ohio State University Press.

Osipow, S.H. 8 Schweikert, D. (1979). The Career Decision
Scale: A test of concurrent validity. Unpublished
manuscfipt, Ohio State University.

 

Penn, P.D. (1981). Differential effects of
vocationally-related behaviors of a computer-based career
guidance system in conjunction with innovative career
exploration strategies. Qissertation Abstracts
International, 43,_5070A-5071.

Pinder, F. 8 Fitzgerald, P. (1984). The effectiveness of a
computerized guidance system in promoting career decision
making. Journal of Vocational Behavior, 24, 123-131.

 

 

152

Pinkney, J.W. 8 Jacobs, D. (1985). New counselors and
personal interest in the task of career counseling.
Journal of Counseling Psychology, 32, 454-457.

Power, P., Holland, J., Daiger, D. 8 Takai, R. (1979). The
relation of student characteristics to the influence of
the SDS. Measurement and Evaluation in Guidance, 12,

Pyle, K. (1984). Career counseling and computers: Where is
the creativity? Journal of Counseling,and Development,
63, 141-144.

Pyle, K. 8 Stripling, R. (1976). The counselor, the
computer, and career development. The Vocational Guidance
Quarterly, 25, 71-75.

Rayman, J. 8 Bowlsbey, J. (1977). DISCOVER: A model for a
systematic career guidance program. Vocational Guidance

 

Rayman, J., Bryson, D. 8 Bowlsbey, J. (1978). The field
trial of DISCOVER: A new computerized interactive
guidance system. The Vocational GuidanceQuarterly,
June, 349-360.

Reardon, R., Bonnell, R. 8 Huddleston, M. (1982).
Self-directed career exploration: A comparison of CHOICES
and the self-directed search. Journal of Vocational
Behavior, 22, 22-30.

 

Resnick, H., Fauble, M.L. 8 Osipow, S.H. (1970). Vocational
crystallization and self-esteen in college students.
Journal of Counseling Psychology, 17, 465-467.

Risser, J. 8 Tulley, J. (1977). SIGI,projgct research
summary report. Pasadena, CA: Pasadena City College.

 

Roe, A. (1956). The psychology of occupations. New York:
John Wiley 8 Sons.

Rogers, L. (1954). Comparisons of two kinds of test
interpretation interview. Journal of Counseling
Psychology, ‘1, 224-231.

Rogers, W.B., Jr. (1980). An investigation of career
indecision theory as it applies to male college students.
Unpublished doctoral dissertation. North Carolina State
University, Raleigh.

Rosenthal, R. 8 Rosnow, R.L. (1975). The volunteer subject.
New York: John Wiley 8 Sons.

153

Ross, L.L. (1971). The effectiveness of two systems for
delivering occupational information: A comparative
analysis. Unpublished master's thesis, University of
Oregon, Eugene.

 

Rotter, J.B. (1966). Generalized expectancies for internal
versus external control of reinforcement. Psychological
Monggraphs, 80, (1, Whole No. 609).

 

Rubinstein, M.R. (1978). Integrative interpretation of
vocational interest inventory results. Journal of
Counseling Psychology, 25, 306-309.

 

Rubinton, M. (1980). Integrative interpretation of
vocational interest inventory results. Journal of
Counseling Psychology, 25, 306-309.

Ryan, C., Drummond, R. 8 Shannon, M. (1980). Guidance
Information System: An analysis of impact on school
counseling. The School Counselor, Nov., 93-97.

Sampson, J.P., Jr. (1977). Counselor intervention with
computer-assisted career guidance. Dissertation Abstracts
International, 38, 3297A.

 

Sampson, J.P., Jr. (1983). Computer-assisted testing and
assessment: Current status and implications for the
future. Measurement and Evaluation in Guidance, 15,

Sampson, J.P., Jr. (1984). The use of computers. In H.D.
Burck 8 R.C. Reardon (Eds.), Career development
interventions. Springfield, IL: Charles C. Thomas.

 

Sampson, J.P., Jr. 8 Pyle, K.R. (1983). Ethical issues
involved with the use of computer-assisted counseling,
testing and guidance systems. Personnel and Guidance
Journal, 61, 283-287.

Sampson, J.P., Jr. 8 Stripling, R. (1979). Strategies for
counselor intervention with a computer-assisted career
guidance system. The Vocational Guidance Quarterly,
March, 230-238.

Savin, G. (1978). The DISCOVER program in Highland, Indiana.
Unpublished manuscript, School Town of Highland, IN.

Schaefer, B. (1976). Holland's SDS: Is its effectiveness
contingent upon selected variables? Journal of Vocational
Behavior, s, 113-123.

 

154

Schneider, L.J., DeWinne, R.F. 8 Overton, T.D. (1980).
Influence of congruity between parental personality types
on offspring's personality development. Journal of
Counseling Psychology, 27, 40-43.

 

 

Slaney, R.B. (1978). Expressed and inventoried vocational
interests: a comparison of instruments. Journal of
Counseling Psychology, 25, 520-529.

Slaney, R.B. (1980). Expressed vocational choice and
vocational indecision. Journal of Counseling Psychology,
27, 122-129.

Slaney, R.B., Palko-Nonemaker, D. 8 Alexander, R. (1981). An
investigation of two measures of career indecision. ‘
Journal of Vocational Behavior, 18, 92-103.

 

Slusher, N. 8 White, K. (1978). Apreliminaryjevaluation of
the System for Interactive Guidance and Information with
NTID students .National Institute for the Deaf,
unpublished report, Spring.

 

 

 

Smith, D.R. 8 Evans, J.R. (1973). Comparison of experimental
group guidance and individual counseling as facilitators
of vocational development. Journal of Counseling
Psychology, 20, 202-208.

 

Snipes, J.K. 8 McDaniels, C. (1981). Theoretical foundations
for career information delivery systems. The Vocational
Guidance Quarterly, June, 307-314.

 

Super, D.B. 8 Hall, D.T. (1978). Career development:
Exploration and planning. Annual review of,psychology,
29, 33-37. -

Super, D.E., Thompson, A.S., Lindeman, R.H., Jordaan, J.P. 8
Myers, R.W. (1981). The career development inventory.
Palo Alto, CA: Consulting PsychoIOgists Press.

Sutera, J. (1977). An exploratory inquiry of a residence
hall career planning program. Unpublished Master's
thesis. Ohio State University.

Sweetland, R. 8 Keyser, D. (Eds.) (1983). Tests (first
edition). Kansas City, MO: Test Corporation of America.

Takai, R. 8 Holland, J. (1979). Comparison of the Vocational
Card Sort, the SDS, and the Vocational Exploration and
Insight Kit. Vocational Guidance Quarterly, 27, 312-319.

Taylor, K. (1979). Th5 relationships amongjvocational
indecision: fear of success and locus of control as
moderated by sex and ability. Unpublished doctoral
dissertation, Ohio State University, Columbus.

 

155

Taylor, K.M. (1982). An investigation of vocational
indecision in college students: correlates and
moderators. Journal of Vocational Behavior, 21, 318-329.

 

Taylor, K.M. 8 Betz, N.E. (1983). Applications of
Self-Efficacy Theory to the understanding and treatment
of career indecision. Journal of Vocational Behavior,
22, 63-81.

 

Thoreson, C., Krumboltz, J. 8 Varenhorst, B. (1967). Sex of
counselors and models: effect on client career
exploration. Journal of Counseling Psychology, 14,
503-508.

Traver, G.J. (1977). Effects of a computerized career
information system upon job and self knowledge of
selected prison inmates in Tennessee. Dissertation
Abstracts International, 38, 7160A.

 

 

Utz, P.W. (1983). A comparison of three groups of
vocationally indecisive students. Journal of Counseling
Psychology, 30, 262-266.

 

Van Matre, G.L. 8 Cooper, S.E. (1984). Concurrent evaluation
of career indecision and indecisiveness. Personnel and
Guidance Journal, 62, 637-639.

Vierstein, M.C. (1972). The extension of Holland's
Occupational Classification to all occupations in the
Dictionary of Occupational Titles. Journal of Vocational
Behavior, 2, 107-121.

 

 

Wachowiak, D. (1973). Personality correlates of vocational
counseling. JOurnal of Counseling Psychology, 20,
567-568.

Walsh, W. (1972). Vocational Preference Inventory. Journal
of Educational Measurement, 2, 167-169.

 

Walsh, W. (1973). Theories ofjperson-environment
interaction: Implications for the college student. Iowa
City, Iowa: The American College Testing Program.

Walsh, W.B. 8 Lewis, R.C. (1972). Consistent, inconsistent
and undecided career preferences and personality. Journal
of Vocational Behavior, 2, 309-316.

 

Walz, G.R. (1984). Role of the counselor with computers.
Journal of Counseling and Development, 63, 135-138.

Warner, S.G. 8 Jepsen, D.A. (1979). Differential effects of
conceptual level and group counseling format on
adolescent career decision-making processes. Journal of
Counseling Psychology, 2g, 497-503.

 

156

Webb, E.J., Campbell, D.T., Schwartz, R.D. 8 Sechrest, L.
(1966). Unobtrusive measures: Nonreactive research in the
social sciences. ChiCago: Rand McNally.

 

Weinrach, S.G. (1985). Determinants of vocational choice:
Holland's theory. In D. Brown 8 L. Brooks (Eds.) Career
choice and development. San Francisco: Jossey-Bass, pp.
61-93 0

 

White, P. (1981). The impact of computers on career
education and guidance provision. In V. Wheelock (Ed.)
Careers in an information society, education for the
future. Huddersfield: Greenwood Graphics Ltd., pp. 52-64.

Whitehorn, J.C. 8 Betz, B.J. (1960). Further studies of the
doctor as a crucial variable in the outcome of treatment
with schizophrenic patients. American Journal of
Psychiatry, 117, 215-223.

 

 

Whitney, D. 8 Whittlesey, R. (1972). Two hypotheses about
Holland's personality types and counseling outcomes.
Journal of Counseling Psychology, 19, 323-327.

 

Wigington, J.H. (1983). The applicability of Holland's
typology to clients. Journal of Vocational Behavior, 23,
286-293.

Wilhelm, S.J. (1978). Computer-based guidance systems: Are
they worth it? Journal of College Placement, Winter,
65-69 0

Wolff, W.W. (1976). The COCIS annual report. Boulder, CO:
University of Colorado. (ERIC Document Reproduction
Service No. ED 127 518)

Wood, G. (1981). Fundamentals of Psychological Research.
Boston, Little, Brown and Co.

Wrenn, C.G. (1970). The dangers within. Personnel and
Guidance Journal, 49, 183-184.

 

Zener, T.B. 8 Schnuelle, L. (1976). Effects of the
Self-Directed Search on high school students. Journal of
Counseling Psychology, 23, 353-359.

Zytowski, D.G. 8 Betz, E.L. (1972). Measurement in
counseling research: a review. Counseling Psychologist,
2, 72-86.

"Illlllfllllilfllllllllr