,
ail»

i~i

ini’f’ﬁm u

Hi}

- K “I.“
a“...

: “173).".
{ii

in.

 

vac-In.
izrg‘u. men-2s;
.. .. u, .1

...
I

   

‘2'

, .
Jaw

   

3.
L

.. . ".‘.':.z.’:.‘::.“:7,,,
f“; L '3‘“ '

..
a ..

.., .Z.’ .
.mua‘

. .
’l‘l‘

 

r.
...v
. ,4, 22..
.....,'... . " ;.'
.x A . . ' ,~ . . ‘yv.r";n- .
. H.‘

«4.
f
v
«a

 

V \“
.‘. . "urnrr‘a,A.
...4:"a.' .' may. w .._.
,
l

 

 

“5..
’fﬁ': " "
u”.

 

 

..-
.5...
~. .
T‘ ~ ~—
1 . ’ -v:.'-.I~"'.r
-,_ ._ - 1., .. “.5: . r
, . , .4: ..,, _..
g ..-.;-,, -. .... "H.131- {up .- 3
.1 ”WWW“, ,,
I .v . ,,. ,,_. .0 ,

>.~:4
~——.~wo.-u-"

 

. .
.nn.-. . ,
.-.a..’....".........y’:f~
vJ'N-va-anzhop .4... 4
. 1

   

.. u. .. "rm-nil...“-
f' L- ~ 'm“"
:,v...’. N...
.3. . .. -.....- 4.
“.72.... u.

.

E
.‘
;
z
t
l‘

"‘4‘. 01“!"

v- '0

 

,- nova-u
- » v 4 .53.?-
.

‘

 

no- u- 7L~qo v
am». mum-:9» {-
. n... v wityl‘. a (O
CN-Ohulﬁyn . ”4‘ l<-:4;V".V'
4 ,... Hr

 

4....
«v... u
~:,-.-— V.mt'.'iv~

.nuw. u um~am -.

n . n

'1
v

   

. ."

4 a“. ..- n"

.4», .. _ .. --... 4... .54..--”
"‘3‘.”

£2.
3“;
.

.r
{$5
‘5

i
89:
rﬁr'

 

3 a

$2. .54;
‘ 51‘, 49:2133‘ 72:4.

'7‘- “I"

4:: -x»
x
.,,.',.‘,.‘3 .
v,-_ -.r<_>-.:';é’-4.
. - ,.
. . 4...,“ . 4,3
r1-;?o="

~22-

53.-
k. 5. r .4
- "3" ..... ,..
,..,-.....,,..'
Cir-- 'ufs-rr-sr. '
I

.9» a;-

  
 
 
     

"w

., K;‘.-'27.":.', a“... .-

J;..,,,.,.:. I"”47”? ~,.. :4. ”2.22:...
n:

zwv
"-3. .n;l
.un— -»{~

. .r .
..
. 1 .4; ”gm”..-
w- ._. ... . ,. "m...
., ,.. w5'v:-;,;é,~"-- 7
n 0 »-.:

v pom-v ~
ov- - rr‘” pwnu-xuu-_-~uum
n..ﬂ~r,o.—raurur‘ “run-r».- .«n

.-..~......~o—v -
v 41w.

 

a... .ww-m
_... o.

   

uo'ui DJ!
4.... ~..
3:..."
..

 

"VI. .
o. a...
01!! 4-Way
' ”*3” noon-I'm

"Mm

no.»

wruwl

a 4r~rm
I'I '—

awe—new.

rmpﬁzrauaw:
U: "a.” yam-no
mtmv': .ﬂ_ .
WM..&"“"7..:*.
, W 'F‘QI'U‘W %?~C:O 0:31
.....r. . . ..
.. ﬁrm 1':
:U' 0"» I” V
WW“

   

 

_’...._......_ ..
”.5“; >

 

h
w»...
' ”-4352.
.2.

"-0-- cumin-vi..-
v u..- -

 

l’rlEﬁ'l‘

IHIIHHIIII'll!llllllllllllllllllllllllllllllllllllllllllll

300901 9369

This is to certify that the
dissertation entitled

Organizational empowerment of volunteer
citizen monitors: A field experiment

presented by

Marion Terenzio

has been accepted towards fulﬁllment
of the requirements for

 

 

Ph.D. mgmem Psychology

Ww/‘ﬂ/ieba,

Major professor

Date ////(/7’

MSU is an Afﬁrmative Action/Equal Opportunity Institution 042771

 

LIBRARY

Michigan State

University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ll

 

 

 

 

 

 

 

 

 

 

 

 

l

 

MSU lo An Affinnetivo ActionlEquel Opportunity institution

omens-9.1

ORGANIZATIONAL EMPOWERMENT OF
VOLUNTEER CITIZEN MONITORS:
A FIELD EXPERIMENT

By

Marion Ann Terenzio

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1991

A / /’\
:P)‘. b.) \x/

6 8‘” *3”.

ABSTRACT
ORGANIZATIONAL EMPOWERMENT OF
VOLUNTEER CITIZEN MONITORS:

A FIELD EXPERIMENT
By

Marion Ann Terenzio

One purpose of the present study was to provide a conceptual framework derived
from theories of empowerment and organizational behavior, with particular emphasis
on volunteer satisfaction, commitment, self-efﬁcacy and participation within an
advocacy agency. Another purpose of the present study was to empirically test an
empowerment intervention that was collaboratively implemented by volunteers and
the investigator to address low volunteer perceived self-efﬁcacy. The study
participants were volunteers who monitored the quality of life of adults with
developmental disabilities who were living in group homes, After three years of
monitoring, the volunteers felt ineffective in their ability to make changes in the
group homes. They, in collaboration with the investigator, developed and
empowerment intervention that addressed their concerns. The intervention strategies
which represented the independent variable in the present study, involved role
clariﬁcation and access to appropriate information. A repeated measures
experimental control group design was used to test the impact of the empowerment
intervention on volunteer self-efﬁcacy, satisfaction, commitment and program
participation trends. Volunteers ﬁlled out a questionnaire prior to and three months

after the intervention implementation. Archival data was used to ascertain the

volunteers' level of participation. Regression analyses were performed to test the
mediational model between and the volunteers ' level of participation and their levels
of perceived self-efficacy, satisfaction and commitment. Results indicated that the
volunteers in the experimental group signiﬁcantly increased in their levels of program
self-efficacy, satisfaction and participation, while the volunteers in the control group
displayed a significant decrease in these variables. The intervention did not
significantly increase the volunteers ' sense of program commitment, nor their overall
perceptions of general self-efﬁcacy. Results from the regression analyses did not
support a mediational model. This study demonstrated the need to further examine
the relationship between volunteers and agencies, particularly in terms of meeting
volunteer expectations and agency goals. This study also provided an example of

how to broaden the application of the construct of empowerment.

I dedicate this dissertation to my first mentor, Dr. Patricia O'Connor, who
continously afﬁrmed me as a competent student and valued colleague, and to my

parents for their support during those many trying times.

iv

ACKNOWLEDGEMENTS

I wish to publicly thank Dr. Thomas Reischl for his patience and guidance in
this dissertation process, and to acknowledge Dr. Paul F reddolino and Dr. Daniel
Ilgen for their expert advice and words of encouragement. I am most grateful to
Mrs. Sandy Orne-Adams and the staff and volunteers at Michigan Association for
Retarded Citizens for allowing me the opportunity to be involved in their program
and for their integrity in acting upon their convictions. Finally, I would like to
particularly thank Dr. William Davidson, 11 for his conﬁdence in me and his words

of wisdom throughout my tenure at MSU.

Thank you, all!

TABLE OF CONTENTS

Contents WI
Introduction ............................... . ................... 1
Theoretical Perspective of the Study ................................ 4
Literature Review .............................................. 6
Volunteerism ............................................ 6
Organizational Commitment ................................ 10
Organizational Satisfaction .................................. 15
Contingencies and Feedback ................................ 18
Model of Feedback Processes . . . . . . . . . . . . . ............. 19

Goal Setting and Feedback ............................ 21

Self-Efficacy ....................................... 26
Organization-Based Self-Esteem ........................ 27
Empowerment ........................................... 29
Present Study ................................................. 33
Model of an Empowerment Intervention ....................... 33
Setting for the Intervention ................................. 36
Intervention Goal ........................................ 38

mm mm

Intervention Rationale ..................................... 39
Intervention Procedures ............... _ .................... 44
Intervention Overview ............................... 44

Step one ..................................... 44

Step two ..................................... 45

Step three .................................... 46

Step four .................................... 46

Step ﬁve ..................................... 47

Summary ............................................... 47
Hypotheses .................................................. 48
Hypothesis One .......................................... 49
HypothesisTwo ............. 49
Hypothesis Three ......................................... 49
Hypothesis Four ......................................... SO
Hypothesis Five .......................................... 50
Hypothesis Six ........................................... 51
Methods .................................................... 51
Sampling ............................................... 51
Procedures ............................................. 55
Data Sources ............................................ 58
Manipulation Check ....................................... 59

vii

mums 2mm

Participants ............................................. 61
Measures ........................... i .................... 63
Self-Efﬁcacy Measures ................................ 63
The Generalized Self-Efficacy Scale
(GSE) ....................................... 63
Program Self-Efﬁcacy Scale ....................... 63
The Organization-Based Self-Esteem Scale
(OBSE) ..................................... 64
Satisfaction Measures ................................ 65

The Job Diagnostic Survey (IDS)

General Satisfaction Sub-scale ..................... 65
The Frustration Index ........................... 67

The Organization Commitment Questionnaire .
(OCQ) ........................................... 67
The Intervention/Process Measures ...................... 68
Results ..................................................... 69
The Relationship Among the Dependent Variables ............... 71
The Intervention Impact ................................... 74
Self-Efficacy ....................................... 74
General Satisfaction ................................. 77
The Participation (Dependent) Variables .................. 79
Program Commitment ................................ 82

viii

ntn W

Summary .................................................... 82
Discussion ................................ _ ................... 84
Program Self-Efﬁcacy ..................................... 84
Satisfaction ............................................. 88
The Dependent (Participation) Variables ....................... 91
Organization-Based Self-Esteem .............................. 93
General Self-Efﬁcacy ...................................... 94
Organizational Commitment ................................ 96
The Process of Empowerment ............................... 98
Application of the O ' Connor Model of Competence ............. 104
Methodological Considerations .............................. 105
Conclusion ....................................... _ ........... 108
References .................................................. 120

LIST OF TABLES

labia.

Table 1: Correlation matrix of the

variables in the present study ....................

Table 2: Correlation coefﬁcients between

the mediating and dependent variables .............

Table 3: Means (and SDs) of self-efﬁcacy measures

and E coeﬁicients for Group X Time design .........

Table 4: Means (and SDs) of satisfaction measures

and _E coefﬁcients for Group X Time design .........

Table 5:1[coefﬁcients of four items from

the Frustration Index ..........................

Table 6: Means (and SDs) of the dependent variables

and _E coefﬁcients for Group X Time design .........

Appendix A, Table 1: it coefﬁcients of major categorical

variables of the study at pre-test ..................

Appendix A, Table 2: Results of a test for homogeneity

of conditions between the groups at pre-test ........

Appendix A, Table 3: Results of a test for differences
between Wayne County and the remaining the counties

Appendix A, Table 4: ”f coefﬁcients for engagement

in the intervention strategies by each volunteer group . .

Buiﬂumlm

................. 70

................. 73

................. 76

................. 78

................. 80

................. 81

................ 110

................ 113

Iables Mailer

Appendix B, Table 1: Items of the Generalized

Self-Efﬁcacy Scale ............................................ 114
Appendix B, Table 2: Program Self-Efﬁcacy Scale items ................ 115
Appendix B, Table 3: Items of the Organization-Based

Self-Esteem Scale ............................................. 116
Appendix B, Table 4: Items of the " General Satisfaction"

sub-scale of the Job Diagnostic Survey ............................. 117
Appendix B, Table 5: Items of the Frustration Index ................... 118

Appendix B, Table 6: Items of the Organization
Commitment Questionnaire ..................................... 119

LIST OF FIGURES

Figures Pagzﬂumhcr
Figure 1: O ' Connor ' s sources of competence model ................... 35

xii

INTRODUCTION

In times of ﬁnancial constraints and budgetary cuts, more and more service
oriented agencies have to rely on volunteers to perform tasks once done by paid
employees. However, individuals who in the past may have been able to devote time
to voluntary endeavors now need to seek employment in order to maintain their own
ﬁnancial security. As the potential volunteer pool shrinks, and the demand for
volunteers increases, an acute shortage of human resources has resulted, which has
forced organizations to pay more attention to the issues of volunteer recruitment,
participation and retention (Gamm & Kassab, 1983). The competition for volunteers
has been exacerbated by the fact that individuals have become more selective in their
choice of agencies to which to volunteer their time (Gidron, 1987). One example of
this is that individuals have begun to expect agencies to provide them with
opportunities for personal grth and development (Houghland, Turner, &
Hendricks, 1988).

As the role of the volunteers become more critical in agencies, researchers of
volunteerism have begun to focus on the relationship between volunteers and
agencies, particularly in terms of the ability of agencies to fulﬁll volunteers’ personal

expectations. This focus directs us to take into account the organizational behavior

2

of volunteers, such as their perception of self-efficacy, satisfaction and commitment,
and level of participation (Gidron, 1987).

Although the notion of organization and volunteer interaction is relatively new
to the realm of volunteer research, this concept has received some attention. Several
researchers (Dailey, 1986; Gam & Kassab, 1983; & llsley and Niemi 1981) have
borrowed from models explicated in the organizational and employee behavior
research literature. One tenet that these researchers have supported is that
organizations that rely on volunteers should closely examine their internal structures,
in particular their volunteer management practices. One purpose of the present
study was to apply organizational models that depict the interactional relationship
between volunteer characteristics and organizational structures in a ﬁeld setting.

Through his research with volunteers and supervisors of an agency, Dailey
(1986) demonstrated that the application of organizational behavior models to
explain volunteer behavior within an organizational context is plausible. His process
model of volunteer commitment, which was tested in his study, delineated the
interactions between the various organizational factors and the individual volunteers,
and the possible enhancement of active participation and future retention of
volunteers. Borrowing from Porter and Iawler (1968), Dailey emphasized that
volunteer perceived satisfaction with outcomes, which is a function of the values that
volunteers place on those outcomes, is a key element for successful volunteer
recruitment and retention.

Another purpose of the present study was to test the effects of an

3

empowerment intervention that was developed and implemented by the volunteers
in collaboration with the investigator. The theoretical premise which framed the
intervention was partly derived from expectancy theories of organizational
satisfaction, commitment and participation, and partly from theories of
empowerment. The intervention was developed to enhance the volunteers’ perceived
self-efﬁcacy in terms of their involvement with the program, and their general
satisfaction and commitment, (mediating variables of the study), and their level of
participation (dependent variable).

The theory of employee commitment, ﬁrst explicated by Porter and Lawler
(1968), and later adapted to the volunteer sector by Dailey ( 1986), suggested that
complex interactions exist among many factors related to employees or volunteers
and the organization, and that the outcomes of the interactions can be manipulated
and inﬂuenced through many diverse mechanisms. One such mechanism that seemed
to appropriately address the issue of enhanced perceived self-efﬁcacy and increased
satisfaction and commitment among volunteers was the process of empowerment.
Empowerment, as explicated in the present study, was deﬁned as the process through
which volunteers clariﬁed their roles and gained access to appropriate information.
The anticipated end result was the enhancement of perceived self-efﬁcacy,
satisfaction and commitment, and an increased level of participation.

Conger and Kanungo (1988) provided an empowerment model that
accommodated the factors and issues identiﬁed in this study. The model stipulates

that change should occur in the organizational setting, and not within the individuals.

The application of this model to the present study seemed appropriate because; (a)
it has speciﬁc application within an organizational context, (b) it is based on the
notion of interactions between the individual and the organization, and (c) it is a
process through which individuals manipulate the envirOnment in order to increase
self-efﬁcacy.

In summary, the major purpose of this study was to test the effects of an
empowerment intervention that was collaboratively implemented by a group of
citizen advocate volunteers and the investigator. The purpose of the intervention
was to increase the volunteers’ sense of self-efﬁcacy in their advocacy tasks and
enhance their sense of satisfaction with and commitment to that program. It was also
hoped that pertinent information as to the interactive relationship between volunteers

and the organization in which they participate would be obtained.

Theoretical Perspective of the Study

The theoretical perspective of this study explicates a reciprocal interaction
process between volunteers and the agency, which determines the levels of self-
efficacy, satisfaction and commitment on the part of the volunteers, and their level
of participation in the agency (Dailey, 1985 ; Rothman, Erlich & Teresa, 1981; Porter
& Lawler, 1968). According to this view, individuals join a particular agency as

volunteers with expectations as to valued future outcomes or rewards. The level of

5

satisfaction and commitment that these volunteers experience is contingent upon the
agency’s ability to fulﬁll their expectations. Fulﬁllment of volunteer expectations will
most likely enhance volunteer participation and intent to remain.

The nature of the reciprocal interaction between the volunteers and the
agency is dynamic and nonrecursive. The overall description of the process is
dependent upon the initial point of examination. For example, if the process is
initially viewed from the point of the volunteers’ perceptions of self-efﬁcacy and
satisfaction toward their roles, tasks and overall involvement with the agency, then
these perceptions are seen as inﬂuencing their level of participation and performance
(Bandura, 1982; Conger & Kanungo, 1987; Mento, Steel, & Karren, 1987, Locke,
Iatham & Erez, 1988; Locke, Lee & Bobko, 1984). As Bandura (1982) has asserted,
individuals who do not feel that they can accomplish a task will not attempt that task.
Decreased attempts at tasks will impact on the level of participation. The level of
participation, which has been partly determined by the volunteers’ perceptions, can
eventually affect the potential fulﬁllment of the volunteers’ expectations. Thus, a
decrease in volunteer participation can result in the obtainment of fewer goals. The
realization of fewer goals can adversely impact on the volunteers’ levels of
satisfaction with and commitment to the program. These lower levels of satisfaction
and related experiences can then determine the future levels of volunteer perceived
self-efﬁcacy and expectations, which affect future levels of volunteer participation
(Porter & Lawler, 1968; Locke, Lee & Bobko, 1984). This process is further reﬁned

and explicated in terms of its relationship to the empowerment strategy employed in

the present study.

The following literature review will discuss various theoretical perspectives and
research studies which have influenced the development of the above noted
theoretical orientation of the present study and the effects of an empowerment
intervention on volunteer perceived self-efﬁcacy and participation. The two
fundamental constructs which provide the framework for the review and the rationale
for the present study are volunteerism and empowerment. The deﬁnition of the
dynamic interaction between the individual and the organization for the present study
will be explicated through a discussion of self-efﬁcacy, organizational satisfaction and
commitment, and feedback and goal setting constructs and concepts. The review will

begin with a discussion of volunteerism.

Literature Review

In rim

N 0 single deﬁnition of volunteerism exists. However, several attributes of
volunteerism seem to emerge in the literature (Jenner, 1982; Schorr, 1970). These
attributes are: (a) volunteers actively participate; (b) volunteers join a particular
organization with expectations as to outcomes of their efforts, and potential

satisfaction of their needs; (c) volunteers do not receive ﬁnancial renumeration for

7

their involvement; (d) and the function of volunteerism is dependent upon the social,
political and moral orientation of an individual volunteer within a particular social
structure.

One attribute that has been shown to be of particular import is that the roles
of volunteers are inﬂuenced by the environmental context. Both Rothman, Erlich
and Teresa (1981) and Gidron (1987) found this to be the case in terms of the
fulﬁllment of volunteer expectations for different groups of volunteers. According
to these researchers, individuals choose to volunteer based on their expectations that
agencies will provide certain beneﬁts. If there is congruence between the volunteers’
expectations and the agency’s ability to provide the expected beneﬁts, the level of
volunteer participation increases. Thus, Rothman, et al. (1981) and Gidron ( 1987)
both concluded that, to foster voluntary participation, an organization must provide
appropriate beneﬁts. Such beneﬁts may be either instrumental, or expressive.
Instrumental beneﬁts include concrete results such as policy reform or increased
allotment of resources; where-as expressive beneﬁts include psychosocial results, such
as increased satisfaction, or worthwhile social interactions.

Anderson and Moore (1977) corroborated the ﬁndings of Rothman, et al.
(1981) and Gidron (1987), and concluded that if voluntary agencies have knowledge
as to why people volunteer, these agencies can gain an understanding as to what
types of expectations volunteers have. Such knowledge could then be used by the
agencies to develop programs that address potential and current volunteer

expectations. Anderson and Moore obtained a comprehensive list of voluntary

8

agencies in Canada, from which they randomly selected 189 service oriented
voluntary agencies. The investigators asked the directors of each agency to randomly
distribute the questionnaires (with self-addressed stamped envelopes) to their
volunteers. The sampling procedure yielded 1062 responses, which represented a 37%
response rate. The results supported the conclusion that most individuals volunteer
for humanitarian reasons, and that the motivation to continue to volunteer stems
from expectations of self-fulﬁllment and growth.

In their review of the literature, Ilsley and Niemi (1981) found that most
researchers assert that volunteer participation issues should be addressed through the
assessment of the interactive relationship between the organization and the volunteer.
Ilsley and Niemi compiled a list of organizational factors gleaned from their review,
that represent potential mechanisms which can inﬂuence an organization’s ability to
fulﬁll volunteer expectations. The factors included; (a) the mission of the
organization, (b) the goals and objectives that have been developed to obtain the
mission, and (c) the ﬁnancial support provided for the volunteer program. Although
researchers call for a need to examine the issue of volunteer participation and
retention within the context of the organization, very few have proposed a conceptual
model that explicates the interaction between the individual characteristics of the
volunteer and organizational factors. Dailey (1986) was successful in adapting an
organizational commitment model to a volunteer population. The basic tenet of
Dailey’s model of volunteer commitment, which was borrowed from models of

employee organizational commitment (Porter & Lawler, 1968; Porter, Mowday,

9

Steers, & Boulian, 1974). The basic tenet of the model is that there is an interaction
between the individual characteristics of the volunteer, and the organizational factors
which inﬂuences the volunteer’s sense of satisfaction and level of participation with
the organization.

Dailey (1986) collected data from a convenience sample of volunteers and
their supervisors who were afﬁliated with a charitable organization. The study
examined the hypothesis that volunteers’ satisfaction with their role and involvement
with an agency are correlated with their commitment to that agency. Dailey’s results
demonstrated that general volunteer satisfaction was predicted by three task
dimensions; task signiﬁcance, skill variety, and task identity. Task signiﬁcance was
found to be the strongest predictor of satisfaction. Personal characteristics did not
signiﬁcantly predict commitment, but satisfaction did.

Dailey’s study provides support to further examine volunteer behavior in terms
of employee organizational paradigms. Although Dailey’s results are not unique to
the volunteer sector, several implications can be drawn, the ﬁrst is that satisfaction
with volunteer tasks plays an important role in volunteer commitment. The second
implication is that the volunteers’ personal characteristics alone are not strong
predictors of volunteer commitment. Dailey’s results support the need for
researchers to examine organizational characteristics such as task identity, variety and
signiﬁcance, and uses of feedback, along with volunteer perceptions of organizational
structures to better understand volunteer participation.

In summary, the trend in volunteerism research is to examine the issue of

10

increased volunteer participation and eventual increased volunteer retention rates,
as the public and private service sectors become more reliant on volunteer resources.
In order to assess the critical factors involved with increased participation,
researchers need to search for appropriate conceptual models that capture the
pertinent dynamic relationship between volunteers and their agencies. It has been
asserted by Ilsley and N iemi (1981) and Dailey (1986) that conceptual models should
include an interaction between the expectations of volunteers and the agency’s ability

to meet such expectations.

0 "MC .11

As noted earlier, volunteers have expectations as to their future satisfaction
as well as to their sense of self-efﬁcacy with their participation within an agency’s
program. In turn, enhanced satisfaction and fulﬁlled expectations of positive
outcomes associated with their participation can enhance individuals’ levels of
commitment to the agency or program. Research on organizational commitment can
provide some important theoretical links and guidelines for the understanding of the
relationship between volunteers and agencies and potential strategies to strengthen
that relationship.

The literature on organizational commitment is varied, however Naylor,

Pritchard and Ilgen (1980) have provided an initial broad conceptual explication of

11

organizational commitment in relation to organizational motivation. These authors
deﬁned organizational motivation as a process which involves allocating personal
resources to certain activities. The process of allocation, which they labelled as
commitment, can be described in terms of time and energy expended to cause the
maximum gain in an individual’s anticipated affective level or level of satisfaction.
The strength behind the resource allocation process is dependent upon the
individual’s affective level, his or her cognitive appraisal of past experiences and
related prediction of future outcomes.

This deﬁnition seems to indicate that an individual’s perception of future
outcomes can be a determinant of commitment. However, it is not that simple.
Many factors, including but not limited to outcomes that are valued by the person,
can play a role in the individual’s expectation of future outcomes and rewards, as
well as choice of behavior used to obtain personal and organizational goals. This
assertion was ﬁrst put forth by Porter and Lawler (1968) who identiﬁed some of the
factors that could inﬂuence outcomes. Such factors included self-esteem, affective
and cognitive appraisals of the individuals as to their expectations, and the work
environment itself.

Commitment reﬂects variations in the level of motivation through the arousal
of factors outside and within an individual’s control. Thus, the assessment of only
effort expended by an individual and the end products obtained through a particular
act of resource allocation precludes the identiﬁcation of other factors that play a

signiﬁcant role in the determination of the relationship between the motivation and

12

commitment process and outcomes.

The discussion of commitment as provided by Naylor, et al. ( 1980) focuses on
a general resource allocation process. Porter, Steers, Mowday, and Boulian (1974)
and Porter (1979) provided a deﬁnition of commitment in relation to a speciﬁc
context, namely the organization. They conceptualized organizational commitment
as an individual’s desire to remain with an organization and his ore her ability to
identify with the organizational goals. According to the Porter et al. (1974)
deﬁnition, three characteristics comprise organizational commitment. They are: (a)
a strong belief in the acceptance of the organization’s goals and values, (b) a
willingness to exert considerable effort on behalf of the organization and, (c) a strong
desire to maintain membership in the organization. Mowday, Porter and Steers
(1982) further recognized the existence of exchange mechanisms that operate in the
establishment of organizational commitment. These mechanisms are represented by
(a) the extent to which expectations of the individual are being met by the
organization, (b) the extent to which the individual feels important, and (c) the
individual’s social involvement with the organization.

Steers (1977) and Morris and Sherman (1981) provided empirical evidence
that supports the above deﬁnitions of organizational commitment. Steers’ model of
commitment, although it merely deﬁnes clusters of variables associated with
commitment, represents a two fold process; the ﬁrst identiﬁes antecedents of
organizational commitment, and the second identiﬁes commitment outcomes. The

three major antecedents of commitment are: personal characteristics, job

13

characteristics, and work experiences. The personal characteristics include age,
education and central life interest; where as the job characteristics include, job
challenge, opportunities for social interaction, and feedback related to performance.
Work experiences include, organizational dependability and trust, group attitudes
toward the organization, and perceptions of personal investment and importance to
the organization. The outcome components of commitment include the intent and
desire to remain, attendance, retention and job performance.

Steers collected data from a randomly selected sample of employees in two
separate organizations, a hospital and an independent research laboratory. Measures
of self-reports were administered to the employees and their respective supervisors,
and behavioral measures recorded the outcome variables. Steers’ ﬁndings
demonstrated that personal characteristics, job characteristics and work experiences
are important antecedents of commitment.

Morris and Sherman (1981) performed a study which tested the
generalizability of the existence of the antecedent variables of organizational
commitment as identiﬁed in the Steer’s model, with one alteration. They added
another characteristic in the personal characteristics category, the sense of
competence. The results that Morris and Sherman obtained through a regression
analysis using data from three different agencies that service developmentally
disabled individuals were similar to the ﬁndings of Steers. In addition, a sense of
competence was also found to be a highly signiﬁcant predictor of organizational

commitment. Morris and Sherman concluded that Steer’s model provides a grounded

14

theoretical framework of organizational commitment which can be utilized across a
heterogenous group of organizations.

Further empirical evidence which supports the Outcome component of the
Steers model comes from a study designed by Koch and Steers (1978) that examined
the predictive roles of job commitment and satisfaction. The results indicated that
organizational commitment played a signiﬁcant predictive role in employee turnover.
Data for the study was collected from a random sample of entry-level employees in
three public sector agencies. Zero order correlations between satisfaction and
turnover were not signiﬁcant. In addition, when satisfaction was partialled out, the
relationship between commitment and turnover did not signiﬁcantly change. When
commitment was partialled out, the relationship between satisfaction and turnover
signiﬁcantly increased. Thus, commitment was demonstrated to be a stronger
predictor of outcome, in this case turnover, than satisfaction.

It has been empirically demonstrated that the Steer’s model of organizational
commitment can be applied to a diverse group of individuals in organizations. Thus,
it can be assumed that the model can be applied to volunteers and their
organizations in determining the antecedent factors and outcomes of volunteer
organizational commitment, and its relationship to self-efﬁcacy.

To summarize, four points of interest concerning commitment have emerged
from the research that may have implications for volunteer participation and
~ retention. The points are: (a) both a sense of competence and role conﬂict were

found to be signiﬁcant predictors of commitment, (b) commitment is signiﬁcantly

15

related to the desire to remain, (c) commitment represents a dynamic interaction
between the individual and the organization, and (d) satisfaction is a strong predictor
of commitment.

As stated earlier, satisfaction maybe a strong predictor of organizational
commitment, which may have important implications for the understanding of
volunteer behavior and ways to enhance the volunteer experience. The following
discussion will provide some insight as to the role of satisfaction in an organizational

context.

r'in ifin

A review of the literature on organizational satisfaction reveals that its role
in determining employee expected outcomes and future performance levels and its
relationship with organizational factors are not directly linked. There is evidence
that mediating factors, such as supervisors, co-workers, pay, career progress and work
related tasks can inﬂuence the various types of satisfaction relationships and
interactions with other variables (Shore & Martin 1989, Van De Ven & Ferry, 1980).
The implications from such literature for the present study is that manipulation of
the mediating variables may enhance volunteer satisfaction, and perhaps increased
length of afﬁliation with an agency.

Although various theories of satisfaction exist, Hackman and Lawler’s (1971)

16

model and measurement of the construct seemed to address the most salient factors
related to the volunteer experience in terms of organizational satisfaction. The
Hackman and Lawler model of satisfaction, which was based on the tenets of
expectancy theory as delineated by both Vroom (1964) and Porter and Lawler (1968),
represents an attempt to identify the complex relationships and pertinent factors
involved. The major tenets of the model that have received empirical support were;
(a) an individual believes that a valued outcome can be reached through the
engagement of a particular behavior and (b) the value of the outcomes is partly
determined by an individual’s ability to satisfy his or her needs directly and indirectly.
According to Hackman and Lawler, the work environment should provide the
necessary conditions that will allow for valued outcomes to be reached. In addition
to environmental work factors, individual characteristics such a sense of personal
responsibility and higher ordered need strengths could also affect an individual’s level
of satisfaction with his or her job. The organizational characteristics that could
impact on satisfaction included the speciﬁc task that the individual performed in
terms of its identity and the opportunity for variety. Task identity refers to the
opportunity afforded to the employee to gain a sense of the entirety of the task.
Task variety refers to the degree to which an employee is allowed to utilize a variety
of skills and talents within a particular task, as well as perform a variety of tasks.
Although Hackman and Lawler (1971) delineated two separate task
dimensions (entirety and variety), they maintained that it is not the mere existence

of the environmental aspects that determine the level of satisfaction, but how these

17

aspects are experienced and perceived by the individual. Hackman and Oldham
(1975) further articulated the sense of satisfaction as an intrinsic motivation that is
derived from the tasks themselves. The basic tenets of their theory is that high levels
of satisfaction can be obtained when three psychological states, which include: a
sense of meaningfulness with the job, a sense of responsibility of the outcomes
produced and an awareness of the work outcomes are present for the employee.
These psychological states are primarily caused by the actual parameters of the tasks
of the job. For example, a sense of job meaningfulness seems to be enhanced by the
signiﬁcance of the task, the variety of skill that the task demands and the capability
the task affords the individual to perform it in its entirety. Other task parameters
include feedback from the performance of the task itself, feedback from other
individuals, autonomy and the amount of interaction employees have with each other
in order to perform the tasks.

Although, Hackman and Oldham have claimed that the task itself provides the
sense of satisfaction felt by an individual, personal and environmental factors may act
as mediators between satisfaction and outcomes (Naylor, Pritchard & Ilgen, 1980;
Morris & Sherman, 1981; Mowday, Porter & Steers, 1982). Well articulated
examples of the role of mediating factors on outcomes have been supplied by
research in the areas of feedback processes and goal setting behavior. Below is a
brief overview of the research ﬁndings and explication of feedback and goal setting

and their applicability to the volunteer sector.

18

Ilgen, Fisher and Taylor (1979) provided a comprehensive deﬁnition of
feedback and identiﬁed three essential components: a message, someone to send that
message (the source), and someone to receive it (the recipient). The actual feedback
component of this process is the message or information which speciﬁcally pertains
to the recipient. Although the source is not considered part of the feedback content,
assessment of the impact of the feedback message can only be ascertained through
the identiﬁcation of the source.

According to Ilgen, et. a1 (1979) three distinct types of sources of feedback
have emerged from the research ﬁndings. They are: (a) other individuals who are
both internal and external to the particular organizational setting, (b) the task
associated with the overall performance orientation, and (c) the individual. The
inﬂuence that each source has on the feedback process is dependent upon factors
such as, the credibility of the source, the power the source has over the rewards, and
the trustworthiness of the source.

Although the feedback message, or information may be generated from
external sources, it is the individual receiver who interprets the information, through
his or her own psychological and cognitive perspectives. The individual then interacts
with the feedback stimulus, which eventually determines the utilization of the

feedback information in future behavior.

19
M l f F ackPr

Ilgen, et al. (1979) presented a conceptual model of the feedback process, and
discussed speciﬁc interactions that may occur betWeen the three feedback
components mentioned above. The model outlined four stages that an individual
may engage in when he or she receives feedback information. The execution of the
stages, and consequent decisions and behaviors are contingent upon the nature of the
source, the feedback stimulus interaction, the recipient’s personal cognitive and
affective styles, and environmental factors. The four stages are: (a) perceived
feedback, (b) acceptance of feedback, (c) desire to respond to the feedback, and (d)
the intended response or goals. The ﬁnal segment of the model is the response that
follows from the employment of this four stage process.

As noted earlier, the source of the feedback can play an inﬂuential role in the
outcome of the feedback process. Some evidence points to the fact that the most
relied upon sources of feedback are those that are perceived to be close to the
recipient, with the individual recipient being the closest. Greller and Herold (1975)
provide empirical evidence that individuals do rely on their own perception of
performance efﬁcacy more often than the other sources of feedback.

The purpose of the Greller and Herold study was to assess the importance of
different feedback sources on job information. A convenience sample of 64 evening
division students at an urban community college participated in the study.

The questionnaires, which tapped two types of information, sources of information

and issues, were administered to the students. Issues were broken down into two

20

categories: the role of the source in providing information concerning job
requirements (referent), and the extent to which the source provided information
about how well a person is doing on the job (feedback). Five separate sources were
identiﬁed: self, task, supervisor, co-workers and organization.

Results from a factorial repeated measures AN OVA design, with issue by
source, indicated that job attributes, such as friendly atmosphere, types of rewards,
and type of tasks performed, had an impact on the relative importance of sources.
Individuals relied more on sources that were closer to them for feedback issues
rather than referent issues, regardless of the job attributes. Internal sources (self and
task) became more relevant as sources of referent information when certain job
characteristics, such as autonomy, task identity, enjoyable atmosphere, were present.

It has been demonstrated that the source of feedback can affect the validity
of the message given. In addition to the type of source that gives feedback, the
frequency of information or the message given in the feedback process can partly
affect the outcome response of the recipient. In general, the more frequently positive
feedback is given, the more accurate the recipient will be in his or her interpretation
of the message, the more competent he or she will feel, and the more likely the
appropriate subsequent behavior change will occur. This relationship, however, is
not always simple. It seems that the type of feedback, positive or negative, interacts
with its frequency and the recipient’s cognitive appraisal and can produce situations
in which the feedback can be detrimental, and may even inhibit future performance

(Ivancevich & McMahon, 1982).

21

Although it is evident that the source of information and the frequency of
giving such information, along with other related factors, can inﬂuence the recipient’s
acceptance of the message, the individual characteristiCs of the recipient can also
impact on the outcome of the feedback process. For example, the intentions of the
recipient in terms of his or her response to the feedback can dictate the
consequences of the entire feedback process. If recipients perceive that they have
control over setting the goals which will determine the type and level of their
participation, then their expectations of their future performance and rewards will
be enhanced. Goal setting theories, as ﬁrst postulated by Locke (1968), and later
elaborated on by Steers & Porter (1974), and Ivancevich and McMahon (1982)

address the salient issues involved in recipients’ intentions and the feedback process.

W

In their review of the relevant research on the role of task-goal attributes,
Steers & Porter (1974) and Latham (1983) could not provide a universal conclusion
regarding the effects of employee participation in goal setting and their performance.
Both sets of researchers concluded that contextual and personal factors should be
taken into account in determining performance implications of participation in goal
setting procedures. The main point here is that the relationship between
participation in goal setting and increased performance is more a function of
mediating factors that impact on both rather than a direct linkage between the two

activities.

22

Feedback is one such mediating factor that seems to play an inﬂuential role
in goal setting and its impact on performance is the feedback that is given about
performance. Feedback seems to be most effective when coupled with goal setting,
but the type of feedback is dependent upon many factors. Speciﬁc feedback given
to speciﬁc goals was found to be the most effective enhancement of motivation and
performance (Locke, 1968; Latham, Mitchell, & Dossett, 1978; Ivancevich &
McMahon, 1982). However, few studies have been done that speciﬁcally look at the
role of feedback in goal setting behavior.

Ivancevich and McMahon (1982) examined the motivational impact of goal
setting and performance feedback on performance, intrinsic and extrinsic satisfaction
and organizational commitment. They hypothesized that goal setting combined with
feedback would enhance performance, satisfaction and commitment more than
feedback alone. The study participants were randomly selected from six separate
engineering organizations, and randomly assigned to six different conditions. In the
ﬁrst three conditions, the groups were involved in goal setting. They varied on the
following dimensions: (a) received extrinsic supervisory feedback only, (b) received
extrinsic supervisory feedback and positive evaluative praise, and (c) received self-
generated feedback only. The three groups in the second condition were not
involved in goal setting, and varied on the following dimensions: (d) received general
feedback from their supervisors on their performance, (e) received feedback from
their co-workers, and (f) received only annual evaluations on performance. Data was

collected over three quarterly periods.

23

A repeated measures AN OVA was used to test the relationship between the
conditions and the performance outcomes. Both a signiﬁcant main effect of time
period and an interaction effect by time and control costs were obtained. A follow-
up repeated measures ANOVA was then used to determine the location of the
signiﬁcant differences, followed by a post hoc contrast coding comparison that further
examined the main and unique effects of each condition.

The general results of the contrast coding were that goal setting was found to
have a stronger effect on performance than non-goal setting, and that self-generated
feedback had a more profound effect on performance than external feedback. In
terms of speciﬁc outcomes, goal setting conditions coupled with self-generated
feedback had a higher impact on intrinsic satisfaction than the non-goal setting, and
external feedback goal setting conditions. In terms of commitment, both goal setting
conditions had the same impact. However, the conditions with feedback had a
greater impact on commitment than the non-feedback conditions, and the self
generated feedback was more signiﬁcant than the externally generated feedback.
These results support the results found by Greller and Herold (1982) cited above.
Internal feedback mechanisms of information seem to be the most effective source.

Further support for the above ﬁndings of Ivancevich and McMahon (1982)
comes from the meta-analytic study of goal setting behavior and performance
conducted by Mento, Steel and Karren (1987). Although, they found that feedback
as a moderator variable between goal difﬁculty and performance could not be

substantiated, a signiﬁcant zero order correlation was found between feedback and

24
goal speciﬁcity / difﬁculty-performance in the multiple regression analysis.

Mento, et al. were able to provide substantial support for the two major
premises that Locke postulated in 1968. They demonstrated that hard goals lead to
a higher level of performance, and that speciﬁc hard goals lead to higher
performance than do general goals. Seventy studies that examined goal difﬁculty,
and forty nine studies that examined speciﬁcity/ difﬁculty in relation to performance
were assessed across ﬁve moderator variables. These variables were: the experiment
setting (laboratory versus ﬁeld), study type (experiment versus correlation), feedback,
level of education, and incentives. The overall conclusion of the meta-analysis is that
none of the ﬁve variables were able to moderate the relationship between goal
difficulty and performance, and that only the setting of the study (laboratory versus
ﬁeld) was found to signiﬁcantly moderate the relationship between goal
difﬁculty/ speciﬁcity and performance.

Based on these ﬁndings and other empirical results, Mento, et al. ( 1987)
contended that feedback is desirable for the setting of speciﬁc goals that are
perceived as difﬁcult to obtain in order to enhance effective performance. In
general, they supported the goal setting paradigm.

The provision of a self-generated feedback mechanism for the present study
was included in the intervention or independent variable. Goal setting strategies
were the ﬁrst intervention tasks performed by the volunteers. The speciﬁc targets
for intended change were not identiﬁed until the goals for both the volunteer

program and the intervention were in place. It was assumed by the researcher in this

25

study that the effectiveness of the intervention strategies were going to be partly
inﬂuenced by the level of commitment that volunteers felt toward the goals that they
set.

Locke, Latham and Erez (1988) provided empirical support for the notion that
a relationship exists between performance and goal commitment. Through their
review of the research they concluded that it is important to examine the major
determinants of goal commitment in order to better identify the relationship between
goal commitment and performance. Locke, et. al also identiﬁed three major
categories of goal commitment determinants which include, external factors,
interactional factors and internal factors.

These researchers found that the reason why the studies that they reviewed
did not produce similar results between the salient factors involved in goal
commitment and performance was methodological. Most studies used different
design and measurement techniques that made it virtually impossible to compare or
contrast results in a meaningful fashion. Future studies that use a within rather than
a between subjects design to better ascertain the existence and inﬂuence of
commitment differences are needed.

Although, Locke, et al. (1988) provided a brief, but comprehensive discussion
about each category of determinant factors of goal commitment, they contended that
the one internal determinant that shows strong promise as to its inﬂuential role in
goal commitment and consequent performance is self-efﬁcacy. It has already been

noted that Sherman and Morris (1981) found self-competence to be an important

26

mediator between commitment and outcome behaviors. As suggested by Bandura
(1986) low self-efﬁcacy can override the beneﬁcial impacts of incentives, because the
initial motivation for people to perform is not present. Thus Locke, et al. (1988)
suggested that rewards may play a more minor role in performance expectation than

originally thought when self-efﬁcacy is taken into account.

Self-Efﬁcacy

Bandura (1977 ) deﬁned self-efﬁcacy as individuals’ expectations of their ability
at mastering behaviors, skills or tasks. Without this belief or expectation, individuals
will not attempt to perform particular behaviors or tasks. Wood and Bandura (1989)
further articulated the nature of the construct in that obtainment of a personal goal
can enhance an individuals’ expectations for future rewards, and provide the
standards against which individuals measure their capabilities. Self-efﬁcacy enables
an individual to pursue performance goals, as set by external sources, while utilizing
self-directed behavior and developing personal incentives. In light of this deﬁnition,
the examination of the construct of self-efﬁcacy within a volunteer population seem
appropriate, because volunteers are identiﬁed as group that strives to obtain personal
goals. Although the notion of the fulﬁllment of personal incentives is inherent in a
volunteer population, the ﬁndings of Morris and Sherman (1981), Locke, Latham and
Erez (1988), and Locke, Frederick, Lee and Bobko (1984) for employee populations
empirically support the notion that an individual’s sense of competence can be a

signiﬁcant predictor or determinant of organizational and goal commitment and can

27

play an inﬂuential role in enhanced performance.

Researchers have suggested that other internal factors may also play mediating
roles between goal commitment and performance, which may have speciﬁc relevance
for volunteers. Pierce, Gardner, Cummings and Dunham (1989) contended that
there exists a strong relationship between self-esteem, deﬁned as a belief in the
ability to fulﬁll needs through a particular role within an organization, and self-
efﬁcacy in terms of an individual’s ability to fulﬁll a role within an organization. The
implications for the present study is that volunteer perceived efficacy and self-esteem,
as they relate to role identity within a particular agency, may be good predictors of
commitment and thus produce higher levels of participation and eventual increases
in volunteer retention. This implication can be well served in the model presented

by Steers ( 1977) and Morris and Sherman (1981).

gamma-BMW

Pierce, Gardner, Cummings and Dunham (1989) proposed that self-esteem is
a function of the interaction of the individual and the organization to which he or she
is afﬁliated. The rationale behind this tenet is that most of the psychological
constructs that are studied in organizational perspectives are functionally related to
the role of the employee within an organization. Factors that impact on a person’s
motivation, performance and desire to remain with an organization are deeply
embedded in the organizational context, be it the characteristics of the organization

itself, or the interaction between personal and external characteristics.

28
Pierce et al. (1989) deﬁned organization-based self-esteem (OBSE) as "the

degree to which organizational members believe that they can satisfy their needs by
participating in roles within the context of an organization" (p. 625). Those
individuals who possess a high organization-based self-esteem (OBSE) perceive
themselves as competent members who have been able to satisfy their needs through
their member roles. Thus, it can be inferred from the above deﬁnition that OBSE
is an index of a sense of value and worthwhileness that individuals have of
themselves in terms of their roles and organizational afﬁliation.

Pierce et al. (1989) contended that individuals who possess high OBSE also
possess self-efficacy, and feel conﬁdent in their ability to perform a task. Such
expectations, as noted earlier by Bandura (1977, 1982), Steers (1977), Morris and
Sherman (1981), and Dailey (1985) can lead to enhanced organizational commitment
and participation.

Although Pierce et al. (1989) labeled an individual’s sense of organizational
worthwhileness as self-esteem, it should be noted that it is the explication of a global
construct within a speciﬁc context that is being proposed. Justiﬁcation for this
hypothesis comes from other researchers who have demonstrated the application of
global psychological constructs to the organizational context. Just as other
researchers have claimed, the contention of Pierce et al. was that global self-esteem
measures do not adequately capture the essence of the construct within
organizational settings.

The Pierce et al. (1989) model of organization-based self-esteem identiﬁes

29

antecedents and consequences of the self-esteem construct. The antecedents are
primarily organizational characteristics that impact upon the individuals’ sense of
worthwhileness and value. The sense of value that individuals develop in terms of
their afﬁliation with the organization will result in consequences such as their sense
of intrinsic motivation, organizational satisfaction and commitment, participation and
performance.

Now that the important constructs and variables of the study have been
identiﬁed and discussed, an overall framework in which to explain the relationship
among these variables needs to be articulated. This framework should take into
account the dynamic interaction between individuals and their environment, the
impact of personal and contextual characteristics on outcome behaviors, and the
feedback processes and roles necessary for access to appropriate information. The
concept of empowerment as deﬁned by Rappaport ( 1984) can provide such a

framework.

Emmwerment

Rappaport (1984) provided a deﬁnition of empowerment that conceptually
links this construct with the identiﬁed tasks noted above, and volunteer commitment,
perceived self-efficacy and participation. It states that:

Empowerment is viewed as a process: the mechanism by which people,

30

organizations, and communities gain mastery over their lives. However, the

content of the process is of inﬁnite variety and as the process plays itself out

among different people and settings the end products will be variable and
even inconsistent with one another. The inconsistency is in the ends rather

than in the process; yet the form of the process will also vary (p. 3)

This deﬁnition depicts the construct as a dynamic process that can only be activated
by those individuals who design, develop and implement the components of that
process, and who have identiﬁed not only the goals of the process, but what aspects
of their lives should be and have become to be under their speciﬁc control. Mastery
over one’s life, as stated above, can only be identiﬁed and gained by those people
who are experiencing the empowerment process.

Kieffer’s (1984) deﬁnition of empowerment provides further conceptual
support for the use of empowerment as a mechanism to enhance volunteer perceived
self-efﬁcacy, which can inﬂuence satisfaction and commitment. Kieffer identiﬁes the
construct of empowerment as a transition through which individuals develop and
enhance personal skills and resources needed to inﬂuence social policy. Although,
Rappaport (1984) did not mention social policy in his deﬁnition of empowerment,
Kieffer demonstrated what Rappaport meant when he said that the process of
empowerment can be utilized in very diverse settings. Kieffer applied Rappaport’s
general deﬁnition of empowerment to grass roots movements as they attempted to
gain mastery over instituting social policy change.

Conger and Kanungo (1988) proposed a four stage process model of

31

empowerment that captured the above stated deﬁnitions within an organizational
context. The model suggests that a state of psychological powerlessness can exist
within the employees of an organization, which can ' be altered by managerial
strategies and techniques. Such strategies target contextual factors or organizational
structures that have a detrimental impact on employees’ sense of self-efﬁcacy, and
thus increase their sense of powerlessness. Contextual factors may include, poor
communication networks, low incentives, arbitrary reward allocations, lack of role
clarity, unrealistic goals, low task variety, limited participation in programs or
meetings, and lack of training and technical support.

Once the empowerment strategies which have been developed to alter the
organizational structures that have been deemed detrimental to employees are in
place, employees should be able to receive information concerning their perception
of self-efﬁcacy. After such information is received, employees can then adopt
behavior that not only makes them feel efﬁcacious, but leads to the fulﬁllment of the
organization’s goals as well.

Conger and Kanungo (1988) not only provided a conceptual link between
empowerment and perceived individual mastery, but also explicated empowerment
as a process through which individuals accomplish personal and organizational goals.
Conger and Kanungo (1988) stated that, "Empowerment is a process of enhancing
feelings of self-efﬁcacy among organizational members through the identiﬁcation of
conditions that foster powerlessness and through their removal by both formal and

organizational practices and informal techniques of providing efﬁcacy information”

32
(p. 474).

This deﬁnition provides the conceptual link between the tenets of volunteer
participation and retention as stated by Ilsley and Niemi (1981), Anderson and
Moore (1977), Dailey (1986), and Rothman, Erlich and Teresa (1981), and
organizational constructs that pertain to commitment and satisfaction (Porter &
Lawler, 1968, Hackman & Lawler, 1971; Hackman & Oldham, 1975; Naylor,
Pritchard & Ilgen, 1980). Voluntary organizations must be cognizant of and address
the contextual factors that impact on volunteers and their participatory role within
that agency.

Two factors have been identiﬁed by both Rappaport and Conger and Kanungo
which are central to the empowerment process. The factors are the roles that each
key player has access to and the information necessary to carry out those roles. In
terms of the present study, this implies that through the empowerment intervention
the volunteers secure opportunities for access to roles that they have identiﬁed as
necessary for enhancing their self-efﬁcacy and increased satisfaction, and information
and support that will enhance those roles.

Although, Rappaport did not speciﬁcally identify particular settings that seem
to be the most conducive for the emergence of empowerment, other researchers
have. Zimmerman and Rappaport (1988) proposed that settings which involve
citizen participation are appropriate contexts in which to examine the process of
empowerment, because individuals have the opportunity to experience self-efﬁcacy

within a social context. Zimmerman and Rappaport deﬁned citizen participation as,

"an organized activity in which an individual participates without pay in order to

achieve some stated goal” (p. 726).

Present Study

Medel ef en Empowerment Interventien

As noted earlier, Conger and Kanungo’s (1988) model of empowerment
identiﬁes strategies which can be utilized in organizational settings. However, they
did not adequately identify who would determine the strategies to be used or how to
set such strategies into motion. O’Connor ( 1989) proposed an intervention model
of competence that provides guidelines as to who should implement empowerment
strategies, and under what circumstances change is warranted. ‘ Even though the
model addresses the issue of competence, its application to an empowerment
intervention can be justiﬁed.

The O’Connor model is based on two assumptions; the ﬁrst being that
competence does not hold to one speciﬁc deﬁnition. The second assumption asserts
that the values upon which competence criteria are based are subject to change. The
implication here is that competence is perceiver bound, which is similar to the
explication of empowerment (Rappaport, 1984). Figure 1 depicts the model.

The O’Connor model (1989) is two dimensional, with each dimension

33

34

containing two levels, internal and external. The ﬁrst dimension is the source from
which standards are developed and competency is determined for a particular group
or organization. The standards can be derived from an‘ internal or external source
as it relates to that particular group or organization. The second dimension is the
assessment, be it internal or external to the particular group or organization, of the
level of competence obtained or acquired by that group or organization. Thus, cell
one in Figure 1 represents standards of competence which are developed by a
particular group for itself (internal), and the assessments that are performed by that
same group in regards to the achievement of the competencies set forth by that
group (internal).

O’Connor proposed that when standards and assessments are congruent within
or across the cells, there is no need for intervention. It is when discrepancies exist
either. between sources of standards, or between assessors of achievement, or
between the sources and assessors that intervention is warranted. Such incongruities
can be played out in a myriad of ways within and across the cells. This takes into
account the relationships between individuals and their environments, and among
individuals themselves. Thus, according to O’Connor, the change agent has to be
aware of the origin of both dimensions (standards and achievement), the
incongruities that have prevailed, and the know how to infuse him or herself into the
dynamics of the situation.

O’Connor’s (1989) model represents the conceptual framework upon which

the present empowerment strategy was designed. After listening to all the

35

Sources of Competence

Standards
Assessment
Internal External
Internal internal only internal judges
external
CHANGE AGENT
External external judges external only
internal
Figure 1
’ nn r’ e n l

N ete. From paper presented at the Second Biennial Division 27 Conference,
Community Action and Research, East Lansing, Michigan. "A competency model for
community interventions: A theoretical proposal for change agents" P. O’Connor,

1989, p. 14.

36

stakeholders involved in the monitor program, and carefully reviewing the issues
presented, this researcher, as the potential change agent, decided to frame the
situation as deﬁned by the volunteers as sources of standards and achievement (cell
one of the model). Viewing the situation in this manner directed this researcher to
collaboratively develop an intervention strategy with the volunteers that addressed
their concerns. If this researcher chose to assess the situation in terms of other
sources of standards and achievements in the monitor program, a different
intervention strategy would have emerged. The volunteers detected an incongruence
between their standards, their ability as citizen monitors to initiate positive quality
of life changes for individuals with developmental disabilities who reside in group
homes, and the actual situation in which the volunteers felt that they have not made
a difference (internal sources and internal assessors). This situation represents cell
one of O’Connor’s model. The intervention strategy in this study addressed the

perceived incongruities of the volunteers.

S'Erll .

In 1983, the Michigan Association for Retarded Citizens (ARC/ M) received
a grant from the Department of Mental Health to implement a citizen, advocacy
monitor program. The program’s mission was to monitor the level of quality of life

in the group homes and provide direct feedback to the group home providers, and

37

to the Department of Mental Health as to the extent that individuals who resided in
the homes were experiencing a quality of life. The primary advocacy activity that the
volunteers performed were group home visits, which entailed ﬁlling out assessment
forms as to the presence of normalization principles, and providing suggestions to the
group home staff to enhance the quality of life.

The Michigan Association for Retarded Citizens (ARC/ M) is a collective
advocacy organization primarily comprised of local chapters which are located
throughout the state, with its headquarters in Lansing. Although there is a state level
ARC/ M ofﬁce, it is within the individual local ARC chapters that the advocacy
efforts take place, and to which monitors are speciﬁcally afﬁliated. The overseeing
supervisors of the monitor program are state level staff who work closely with the
volunteer liaisons, known as local coordinators, within the local chapters. The local
coordinators are the immediate volunteer monitor supervisors for the speciﬁc local
chapters.

Three years after the implementation of the program, the volunteer monitors
had not been consistently monitoring the group homes, because they felt frustrated
about their roles as monitors. Thus, their level of participation dropped, as seen
through the infrequent visits to the group homes, and attendance at monitor
meetings. Their primary complaint was that they felt ineffective in their role as
monitors. In collaboration with the investigator, the volunteers, with support from
their state level ARC supervisors, decided to revamp the monitor program in an

effort to enhance their perceived self-efﬁcacy as advocates. The following

38

intervention has been developed in response to this need.

The premise behind this intervention was that changes in the monitor program
procedures and increased access to other sources of information and agencies
involved with group home operations could enhance the roles of the monitors as
effective advocates. It was also assumed that increased feelings of self-efﬁcacy would
increase levels of monitor satisfaction and level of participation. The goal of this
intervention was for the monitors to utilize an empowerment process through which
they would gain a sense of self-efﬁcacy in their activities and increase their level of
satisfaction with and commitment to the monitor program. For the purpose of the
present study, an empowerment process indicates that the volunteers identiﬁed the
problems or incongruities between their perceived standards of advocacy
participation and their level of achievement, and developed the strategies to address
the incongruities.

The practical application of the empowerment process was operationalized
through two different global strategies. The goal of the ﬁrst strategy was to increase
and enhance the opportunity for inﬂuence on the part of the monitors in terms of
access to other types of channels of communication and feedback within the group

home network, including local and state level regulatory agencies. The second

39

strategy involved the development and mobilization of support systems within
ARC/ Michigan and within the local community for each local chapter. Both
strategies targeted the contextual factors and not the volunteers. The empowerment
process, as deﬁned by the two primary strategies, served as the independent variable

in the present study, and was under the control of the monitors and the investigator.

Inrv 'n

Research has demonstrated that the mission of the organization, the goals and
objectives that are utilized to realize that mission, the amount of ﬁnancial support
and the ability of the organization to fulﬁll volunteer expectations are important
inﬂuential factors that impact upon a volunteer’s decision to remain (llsley & Niemi,
1981; Gidron, 1987; Ivancevich & McMahon, 1982). In the case of the ARC/M
monitor program, some of these factors were lacking. For example, the volunteers
were aware of the program mission: to improve the quality of life of the adults who
live in group homes. But, the volunteers perceived the procedures to be
inappropriate to realize the mission and the program objectives to be vague. The
volunteers’ primary role was to visit the group homes and provide various types of
feedback to different agencies which were directly or indirectly involved with the
operation of group homes. During the visits the volunteers performed two tasks; one

was ﬁlling out the assessment form which addressed the homes’ responsiveness to

40

quality of life related behaviors, atmosphere and issues. The second task involved
the volunteers providing suggestions to the group home staff as to how they could
enhance the quality of life in the homes.

The volunteers felt frustrated with both tasks for various reasons. The
volunteers found the forms to be too rigid and the information contained on the
forms to be incomplete and inappropriate for dissemination to agencies involved with
group home operations. Although the group home staff had direct contact with the
adults in the homes, the volunteers felt that the staff were in no position to
implement the suggested changes in the homes due to their status in the group home
organization.

The articulation of these frustrations led the volunteers, in collaboration with
the investigator, to target those organizational and program factors that the
volunteers perceived to be barriers to their monitor role and tasks. The targeted
factors were primarily identiﬁed as channels of communication and feedback
processes. After identiﬁcation of the perceived barriers, the group developed the
empowerment intervention strategies which particularly focused on the
communication and feedback channels between the volunteer monitors and the
ARC/ M, group home providers, state level regulatory agencies and local community
agencies involved with group home operation, as well as general local community
systems. An example of a feedback barrier as perceived by the monitors was that
they needed to be publicly endorsed by ARC/ M as credible sources of feedback for

the Department of Mental Health and those agencies who owned and operated

41

group homes. ARC/ M was responsive to this concern, and implemented the
proposed changes that the monitors made. Three of the proposed changes included:
the formation of a task force to revise the assessment forms and review the training
program, the formation of a task force which dealt with sexuality issues and their
relationship to quality of life, and a ARC/ M sponsored working conference in which
the monitors took an active role in presenting, discussing issues and problems, and
formulating recommendations.

Although, the main focus of the interVention was for monitors to enhance
their effectiveness through their own initiatives to make changes in the monitor
program, the investigator had hypothesized a mediational effect would result as well.
Through direct volunteer manipulation of targeted program and organizational
factors, the volunteers’ level of satisfaction with and commitment to the program
would increase, which might eventually impact on the volunteers’ future expectation
of obtained future valued outcomes and performance (Naylor, et al., 1980; Porter &
Lawler, 1968; Hackman & Lawler, 1971).

The second global empowerment strategy involved accessing support from
various ARC and local community groups, which would lead to enhanced and
broadened feedback systems for the monitors to utilize. One particular support
strategy involved the formation of the local coordinator dyads. Two local
coordinators were paired with the purpose of providing them a forum in which they
could more effectively problem-solve among and with each other, rather than always

having to turn to the state ofﬁce for help. External, local community sources of

42

support were mobilized with the purpose of seeking more communal understanding
and involvement in the monitor program. Such involvement might lead to new
potential resources for future use.

The basic underlying framework of the study which links the empowerment
intervention with organizational commitment, satisfaction, self-efﬁcacy, goal setting
and feedback processes is that the empowerment intervention was treated as a two
fold process through which volunteers identiﬁed goals, roles, tasks and rewards, and
set up the mechanisms which dictated the integration and interaction of these
identiﬁed factors. The empowerment intervention process was a representation of
two feedback loop systems; the ﬁrst one treating the volunteers as recipients of
information that they deemed necessary to possess in order to be more effective and
credible sources of information in the second feedback loop system. This initial
feedback loop process included activities that provided the monitors with information
as to their effectiveness. Such activities included receiving input from the ARC/ M
staff, the group home providers and revamping the training program and assessment
forms used on visits. The second feedback loop in the empowerment intervention
included activities such as the development of other types of monitor strategies than
just group home visits and meeting attendance through which the volunteers were
able to provide more accurate information to a broader base of people and agencies.
Such activities in this second loop included writing articles in the newsletter and
holding meetings with group home providers, local community mental health boards

and the Department of Mental Health.

43

This researcher assumed that if the monitors felt that they were receiving as
well as providing accurate and appropriate information then they would feel more
efﬁcacious and satisﬁed with their roles as monitors. As stated earlier by Bandura
(1977) individuals will feel more efﬁcacious if they have positive expectations as to
their ability to perform the necessary tasks. Since the volunteers identiﬁed and
modiﬁed the barriers to their monitoring performance, then they should begin to feel
effective.

The second basic assumption made by the researcher was that the volunteers
would have an enhanced sense of satisfaction because they valued the outcome of
their efforts; namely to increase their effectiveness to improve the quality of life in
the homes. The volunteers’ satisfaction was a function of their ability to obtain their
desired mission, goals and outcomes. This satisfaction was partially realized by the
type of tasks the volunteers added to their monitoring efforts- The intervention
provided the context in terms of more tasks and better follow through of tasks as well
as improved channels of communication through which the volunteers set their
valued goals and developed a mechanism to realize those goals.

If the volunteers felt efﬁcacious and satisﬁed with the outcomes of their
performance, they would probably intend to remain with the program. Thus, the
intervention was also seen as a vehicle to increase the volunteers’ desire to stay.

This, in turn, would increase ARC/ M’s volunteer retention rate.

Interventien Preeedtges

Interventien merview

As stated earlier, this intervention was a collaborative empowerment project
designed by volunteers to increase their effectiveness as citizen advocates. One
interesting aspect of this intervention was that, although the goals were general, the
operationalization of them took place in each local chapter. The initial starting point
entailed all monitors to review the overall mission of the program as a collective, and
then to move the application of the mission to the local level.

Prior to the implementation of the intervention strategies, the investigator met
with the local coordinators and monitors to discuss the purpose of the intervention,
and to develop the strategies that speciﬁcally addressed the needs of each local
chapter. It was made clear to the monitors that they would identify the problems and
develop the intervention strategies. For the sake of cohesion and continuity for the
collective nature of the monitor program, the intervention was implemented through
the same ﬁve step process for all local chapters. The speciﬁc local chapter issues,

concerns and strategies were addressed in each of the general ﬁve steps.

Step_Qne. The ﬁrst step in the intervention process was the identiﬁcation and
setting of the goals for the monitor program by monitors in each local chapter. This
procedure was the crucial link between the stated mission of the monitor program

and successful accomplishment of the mission as deﬁned by each local chapter.

45

The investigator compiled a list of goals for the monitor program from various
sources of the ARC organization. The monitors in each local chapter reviewed the
list of goals and identiﬁed those that represented the collective monitoring efforts of
all local chapters of ARC/ M, and goals that particularly pertained to their speciﬁc
local chapter program. The monitors included other goals not listed that they felt
were important to the mission of the program, both on an overall and local level.
Once the program goals were established for each local chapter the monitors

prioritized them and developed strategies accordingly.

We. The second step, which was guided by the identiﬁed goals, entailed
the implementation of the opportunities for inﬂuence strategy, which expanded
certain identiﬁed processes and targets of feedback, and curtailed others. The
expansion and curtailment strategies were different for each local chapter, depending
on the list of goals that they identiﬁed and prioritized.

In addition to deve10ping speciﬁc strategies, all local chapters addressed the
following strategies; (a) involvement in the publishing of the monitoring newsletter,
(b) involvement with the preparation of the quarterly report that is sent to the
Department of Mental Health, and (c) the development and implementation of
presentations to state level regulatory agencies concerning the goals, objectives and

progress of the program.

46
W. This step was a further reﬁnement of the opportunities for

inﬂuence strategy and represented activities that expanded the primary role; that of
group home visits. The monitors hoped that these activities would enhance their
assessment efforts and provide them with more monitoring mechanisms. Such
activities included: the establishment of a People First Group, which is a self help
advocacy group for people with developmental disabilities, the training of individuals
who reside in the group homes to become citizen monitors, and the establishment
of a sexuath task force which dealt with the relationship between problems and

issues of sexuality and quality of life.

Stepjeur. This step entailed the development and mobilization of within
agency support for each local chapter. The formation of the local coordinator dyad
involved the pairing of two local coordinators into a problem-solving, information-
sharing, support unit. The main goal of the dyad process was to provide a peer
support system for the local coordinators. The dyad structure would also help link
the local chapters with each other and allay the sense of isolation and frustration felt
by the monitors. It was hypothesized that this type of structure would allow for
better ﬂow of information between the local chapters which were in close proximity
to each other and enhance collaborative efforts in targeting potential providers,
agencies and future monitors. The inclusion of dyad feedback mechanism would also
help create a comprehensive feedback information system for the collective efforts

of ARC/ M.

47
Step ﬁve. The last step involved the mobilization of external support systems

for each local chapter. Particular outside agencies within the local communities were
targeted by each local chapter as providers of either potential resource or support
for the monitor program. This was accomplished through various channels including;
dissemination of monitoring ﬁndings and information, invitations of key individuals
in the community to monitor meetings, invitations of monitors and coordinators to
speak at various community agencies, increased involvement with the local

community mental health boards and other policy making agencies.

Summary

, In summary, this intervention represented a collaboratively developed
empowerment process that targeted organizational and program factors that were
perceived to act as barriers to the monitoring efforts of volunteer citizen advocates.
The process was a two pronged strategy: enhancement of the opportunities for
inﬂuence, and the mobilization of support systems. The goal of the intervention was
to enhance the perceived self-efﬁcacy of the monitors, as well as increase their level
of satisfaction, commitment and participation. The monitors set the standard for

efﬁcacy and provided the means to achieve it.

Hypotheses

Two statistical strategies were used to test the hypotheses in the present study.
The ﬁrst contained two types of analyses, the ﬁrst being cOrrelations which were used
to examine the relationships among self-efﬁcacy, satisfaction and commitment
(mediating dependent variables) and volunteerism as deﬁned by group home visits
and attendance at meetings (primary dependent variables). Only pre-test data were
used in the these analyses in order to assess the relationships between and among the
mediating and dependent variables without the effects of the intervention. The
second set of analyses in the ﬁrst statistical strategies were regression equations
which tested the existence of a mediational model. Post-test data were used, since
the effects of the independent variable needed to be included. Hypotheses One and
Two address the relationship between and among the mediator and dependent
variables.

The second statistical strategy was a repeated measures analysis which was use
to assess the impact of the intervention on the trend of the primary and mediating
dependent variables within the experimental group. Embedded in the experimental
study were manipulations of structural and contextual factors (independent variable;
empowerment intervention) identiﬁed by the monitors as either barriers to their
monitoring tasks or necessary to effectively perform their tasks. Change in the
dependent variables over time in the experimental group would demonstrate that

empowerment strategies did affect volunteerism. Hypotheses Three through Six

48

49

represent the second assessment strategy.

ﬁmthesisme

Organizational satisfaction, commitment and perceived self-efﬁcacy (general
[GSE] and program self-efﬁcacy) should be positively related to the two primary

dependent variables; number of monitor visits and attendance at monitor meetings.

There should be positive relationships between perceived self-eﬁicacy,

organizational commitment and satisfaction and volunteer retention rates.

Hypothesis Three

Members of the experimental group should feel more self-efﬁcacious in their
advocacy efforts than members of the control group. The construct of self-efﬁcacy,
was deﬁned by three independent measures. The ﬁrst was a four item scale which

assessed the individual’s perception of self-efﬁcacy in terms of their roles and tasks

50

in the monitor program. The second measure was the general self-efﬁcacy scale
(GSE) which assessed the individual’s general perception of self-efﬁcacy with no
reference to a particular situation, and the third measure-was the organization-based
self-esteem questionnaire (OBSE) which measured the individual’s feelings of worth

and accomplishment with the program.

Hyppthesis Fem

Members of the experimental group should feel more satisﬁed with the
monitoring program than members of the control group. Two measures were used
to assess satisfaction with the program: the job diagnostic survey (JDS) sub-scale
which assessed the individual’s general satisfaction with the, program, and a
frustration index which assessed the level of frustration individuals had with aspects

of the program.

Hymthesis Five

Members of the experimental group should visit the group homes more often
and attend more monitor meetings than members of the control group. These two

variables represented the original monitoring activities and participation variables

that were subject to change due to the intervention strategy.

Hyppthesis Six

Members of the experimental group should have a stronger sense of
commitment to the program than members of the control group. Commitment was
deﬁned as the desire to remain with an organization. The organizational
commitment questionnaire (OCQ) was used to measure the volunteers’ level of

commitment to the program.

Methods
5mm

Fifteen local ARC/ M chapters (n= 125 volunteers) represented the total
sample for this study. Although sixteen local chapters were involved in the monitor
program, one local chapter declined to participate in the study. This researcher did
not view this as a threat to the validity of the study, because that particular local
chapter was operating under a different set of goals and procedures than the rest of

the local chapters in the monitor program. That local chapter had a monitor

51

52

program in place prior to the implementation of the present monitor program. The
ﬁfteen local chapters who participated in the study were divided into pairs prior to
random assignment and data collection. The pairing was a result of the formation
of the local coordinator dyads. After the dyads were formed, local chapter pairs
were randomly assigned to one of the two experimental groups. Since there were
only three local chapters in the Upper Penninsula, and their geographical location
precluded them from being able to interact with other local chapters, this researcher
formed those three local chapters into a triad. A separate random assignment
process was used for those four Detroit-area local chapters in Wayne County (n =
34 monitors; experimental group n = 18; control group n = 16). One pair of Wayne
county local chapters was randomly assigned to the experimental group and the other
pair was assigned to the control group. The primary reason for this procedure was
that Wayne County was not representative of the rest of the counties in terms of the
number of group homes, the types of overseer boards that dictate policies and group
home procedures, and the location of the city of Detroit, which is more urbanized
than any of the other cities in Michigan. The end result was that there were seven
local chapter dyad/ triad pairings, with eight local chapters assigned to the control
group, and seven assigned to the experimental group.

Random assignment according to local chapters was a preferred sampling
procedure, because it best protected against possible contamination of the
intervention between the experimental and control groups. Random assignment of

individuals would not have adequately ensured that monitors in the control group

53

would not discuss or engage in the intervention procedures with monitors in the
experimental group prior to the post-test. One weakness with the random assignment
by local chapter rather than individual volunteers was that range of variability would
be limited.

All 125 volunteers in the ﬁfteen local chapters (experimental group = 66,
control group = 59) ﬁlled out the pre-test questionnaire, which represented 100%
of the total volunteer population in the monitor program who were eligible to
participate in this study. At post-test, only 109 volunteers (experimental group = 55,
control group = 54) ﬁlled out the questionnaire, which represented 87% of the pre-
test sample and total volunteer population in the monitor program. Of the 16
volunteers who did not ﬁll out a post-test questionnaire, eight volunteers
(experimental group = 5, control group = 3) terminated the program. This
represented a 6% attrition rate for the total sample, a 9% rate for the experimental
group and a 5% rate for the control group. Program termination was deﬁned by the
local coordinators and this researcher prior to the intervention implementation. This
researcher attempted to collect data on those who terminated the program, but was
unsuccessful. Several volunteers had left the area and others did not want to ﬁll out
the questionnaire.

The other 8 volunteers (experimental group = 4, control group = 4) who did
not ﬁll out the post-test questionnaire were still considered active in the monitor
program, but for various reasons, did not respond to this researcher’s efforts to

collect the post-test data. Two major reasons for why these volunteers did not ﬁll

54

out the questionnaire were; they were on vacation and they did not want to ﬁll out
another questionnaire.

To summarize: at pre-test, 100% of the total Volunteer population of the
monitor program (15 local chapters, 11 = 125 volunteers) who agreed to participate
in this study ﬁlled out a questionannire. At post-test, 109 volunteers ﬁlled out the
post-test questionnaire, which represented 87% of the total population of volunteers
who agreed to participate and who ﬁlled out questionnaire at the pre-test. Of the
16 volunteers who did not ﬁll out a questionnaire, 8 terminated the program. This
researcher could not obtain post-test data on any of these 16 individuals

As stated earlier, the local coordinators and this researcher deﬁned program
termination prior to the intervention implementation. The process used by the local
coordinators to determine program termination involved a categorization of monitors
into three different groups: those who actively monitored and went to meetings, those
who had expressed interest in staying with the monitor program, although they had
not monitored within the last month, and those who wanted to keep their names on
the list, but could not be actively involved for several months. A monitor’s name was
dropped from the list if he or she speciﬁcally requested to leave, or did not stay in
contact with the local coordinator for a period of several months. This deﬁnition
seemed adequate, since the local coordinators communicate, either by phone or in
person, with their monitors at least once a month. Thus, the local coordinators could

assess on a month be month basis who terminated and who remained.

55

All study participants in both groups (n = 125) ﬁlled out questionnaires prior
to the intervention implementation (pre-test) which assessed their perceptions of the
program and their level of participation three months prior to the intervention.
Three months after the implementation of the intervention with the experimental
group (post-test), 109 volunteers ﬁlled out another questionnaire which assessed their
perceptions, levels of participation and engagement in the intervention strategies
during that three month time period.

The questionnaires were disseminated through two forums; at the local
monitor meetings and by mail. Since both procedures were used for both groups at
both data collection time periods, procedural bias could be ruled out as a potential
threat to validity. The investigator collected the pre-test data during her initial visit
with all of the local chapters in the study. Time was set aside at the beginning of
each meeting to allow the monitors to ﬁll out the questionnaires. After the
questionnaires were ﬁlled out (which were coded for identiﬁcation and matching at
post-test), the general intervention plans were discussed for the experimental group.
Discussion in the meetings for local chapters in the control group focused on general
monitor program issues, concerns and progress. The local coordinators disseminated
questionnaires to those monitors who were not present at the meetings. A two week
grace period for return was given for those monitors who received their questionnaire

in the mail. If the investigator did not receive a questionnaire within the ﬁrst week

56

of the respondent receiving the form, she called the monitor and mailed another
questionnaire in the self-addressed stamped envelope. All 125 questionnaires were
received within the two week deadline. A grace period of longer than two weeks
would have jeopardized the intervention, and made it harder to control for inﬂuences
due to the passing of time. As each pre-test questionnaire was returned, the
investigator recorded the post-test date and a two week grace period for that
particular respondent (three months from the intervention implementation).
Although, the same data collection procedure was followed for both groups at both
time periods, no intervention plans were discussed with the local coordinators and
monitors in the control group at the meetings. They knew that they would be
addressing their perceived problems within the near future.

The procedures that the investigator used in an attempt to obtain the data
from those 8 individuals who terminated the program involved a) a written request
and questionnaire sent to the individual from the local coordinator, b) a verbal
request from the local coordinator to the individual via the telephone and c) a
written request from the investigator enclosed with a second questionnaire and a self-
addressed stamped envelope with a follow-up phone call from the investigator.
These procedures are slightly different than the procedures used for active volunteer
of the monitor program, because those who terminated the program did not attend
the meetings at which the questionnaires were disseminated.

The primary justiﬁcation for the three month interval between the pre-test and

post-test was that the monitors were anxious to see change in their program.

57

However, their anxiety was directed and tempered by very speciﬁc perceptions of the
problems they were encountering. Thus, the monitors were not easily swayed by just
any change, or the presence of someone who promised to help make change in the
program. This was demonstrated by the fact that the volunteers in the experimental
group signiﬁcantly decreased in their level of frustration with only those aspects of
the monitor program that were addressed by the intervention. The volunteers in the
control group did not signiﬁcantly increase in their level of frustration. (Refer to
Table 5 in the results section of this report.) The empowerment intervention
procedure helped guarantee that the appropriate problems and frustrations as
perceived by the monitors were addressed.

A less salient, but related reason for the three month measurement interval
was that the volunteers were randomly assigned to either an experimental or control
group, which meant that the control group did not immediately receive the
intervention. As stated above, the volunteers were anxious to implement the changes
that would allow them to have access to information and feedback sources that would
enhance their ability to make a difference in the group homes. In order for the
investigator to best assure that the intervention strategies would not spill over into
the control group, perform an adequate assessment of the effectiveness of the
intervention, and be sensitive and responsive to the needs of all of the monitors, an
evaluation period that was long enough to assess change in the experimental group
and yet not cause more frustration for the volunteers in the control group had to be

selected. Thus, a three month time interval was decided upon by the investigator and

58

the ARC state level supervisors of the monitor program, who were extremely

reluctant to agree to a longer time period.

Although, the local chapters were the units that were randomly assigned, the
volunteers themselves were the unit of analysis for testing the hypotheses. The two
primary sources of data used to test the hypotheses were the self-report
questionnaires ﬁlled out by the participants in both groups, and archival data that
were contained in both the local chapter and state ofﬁce ﬁles.

Archival data was collected from the ﬁles of the state and local chapter ofﬁces
to obtain an objective assessment of the volunteers' attendance at meetings and
rates of monitoring visits, intervention activities performed and termination from the
program. Since it was hypothesized that the intervention would help increase the two
major monitoring activities of each local, namely the monitor visits and attendance
at local monitor meetings, it was important to have a data source that was reliable
and not based on the participants' recollection and perceptions. The archival data
information came from local chapter meeting minutes, ofﬁce records on current and
terminated volunteers, date books of general monitoring and intervention activities,
training sessions, and the number of monitor assessment forms on ﬁle at the ARC/ M
state ofﬁce. This data provided information on the following variables; the number

of group home visits that each monitor made, the number of local monitor meetings

59

that monitors attended, the number of monitors recruited, as well as the number who
remained with or terminated the program, the number of group homes that were
monitored, and the total number of visits each 10ml chapter made within a three
month time interval prior to and a three month interval after the intervention
implementation. The archival data also provided the number of volunteers who
participated in the intervention activities prior to and after the manipulation of the
independent variable (the intervention). The use of archival data allowed for

conﬁrmation, since the data was not supplied by monitor self-reports.

Manipu' lgtien Cheek

To assess the success of the random assignment process, a series of between-
group comparisons were used to detect differences between the two groups on 15
pre-test variables. Only one signiﬁcant difference was found; volunteers in the
experimental group had more education than the volunteers in the control group.
Appendix A, Tables 1 and 2 contain the results of the between group comparisons
on the 15 variables.

In addition, no signiﬁcant differences were found between the four chapters
located in Wayne County and the other chapters on the dependent variables of the
study. Refer to Appendix A, Table 3 for the results. Since no differences were

found between Wayne County local chapters and other ARC local chapters, or

60

among the four local chapters themselves, geographical location was not used as a
covariate in subsequent analyses.

A manipulation check was performed to: a) assess the existence of
contamination of the independent variable in both experimental groups at pre-test,
b) to assess the existence of contamination of the independent variable in the control
group at post-test, and c) to determine that the manipulation of the independent
variable took place in the experimental group. At pre-test, all monitors were asked
if they had engaged in any activity associated with the proposed intervention
strategies. The results of the manipulation check at pre-test showed no signiﬁcant
differences between the two groups in terms of their engagement in activities related
to the intervention. A total of three monitors, two from the control and one from
the experimental group, engaged in two intervention-related activities: recruitment
of new monitors and starting a People First group. Thus, out of a total of 125
participants, only three individuals reported engaging in one intervention related
activity. This did not seem to pose a contamination threat for further assessment of
the effectiveness of the intervention.

In order to detect a spill-over effect of the intervention strategy activities into
the control group, during the post-test monitors were asked if they speciﬁcally
engaged in any of the intervention strategies. The inclusion of these responses
served two purposes; the ﬁrst was to assess if the control group engaged in any of the
empowerment strategies (spill-over or contamination check). The second purpose

was to ascertain if the manipulation of the independent variable occurred in the

61

experimental group. Since the major goal of this study was to test the effectiveness
of an empowerment strategy, it was essential to ascertain if the manipulation of the
independent variable took place.

Results of the chi-square analysis at post-test, contained in Appendix A, Table
4, revealed that there was a signiﬁcant difference in the level of engagement on
seven of the 10 intervention activities between the two groups. The two groups did
not differ signiﬁcantly in their engagement of three of the intervention strategies.
Only one individual from the experimental group, and no one from the control group
reported that they attended other local chapter meetings. No one from either group
reported being involved in writing an addendum to the quarterly report, and the
same number of participants in each group reported being involved with writing an
article for the newsletter. The number of control participants (18) who reported
engaging in an intervention strategy represents 2% of the potential number of control
group participants who had the opportunity to engage in at least one intervention
strategy. This minimal percent does not indicate that there was a signiﬁcant spill-

over of the independent variable into the control group.

Manta

Participants in the study included 125 monitors (experimental group = 66,

control group = 59) at pre-test, which represented 100% of the total volunteer

62

monitor population across all 15 local chapters. At post-test 109 monitors responded
to the questionnaire (experimental group = 55, control group = 54), which
represented 87% of the 125 monitors who responded at the pre-test.

Since the random assignment process was somewhat successful in producing
two homogeneous groups prior to the intervention, the following description pertains
to the entire sample population. The mean age of the volunteer monitors was 52
years old, 81% were women and the mean length of volunteer service to the program
was 20 months. The majority of the volunteers were parents (65%) of individuals
with developmental disabilities. A total of 49% of the volunteers had adult children
with developmental disabilities living in a group home.

The reasons for volunteering related to parental concerns about the future
welfare of their children (46%). This is also borne out in the expectations that the
monitors had of the program. The majority of the volunteers (58%) expected that
the program would allow them to improve the quality of life in the group homes.
The major frustration, as indicated by 54% of the monitors was the assessment form
that they used on visits. little access to the group homes was found to be frustrating
by 42% of the volunteers, and 30% found the group home staff frustrating in that
they would not cooperate to improve the conditions of the group homes.

Data gleaned from the ofﬁce ﬁles revealed that the average number of visits
for all monitors within the three month time period prior to the intervention
implementation was 8.1. The average number of local meetings that all monitors

attended within that three month time period was 1.5.

63

W
Self-Efﬁgey Meﬁpres
n r iz lf-Efﬁ . The Generalized Self-Efﬁcacy

(GSE) Scale (Tipton & Worthington, 1984), measures general mm which is
deﬁned as an individual ' s relatively enduring set of perceptions in regards to his or
her ability to generally cope with broad range of situations. Although the reliability
and validity data on this measure are limited, there is some evidence that the
measure signiﬁcantly correlates with the Goal Attainment Scale which measures the
extent to which and individual successfully achieves a personal goal (I; = .33, p <.05).
There is some speculation that the test should be administered in conjunction with
measures that assess the speciﬁc nature of the task at hand (Tipton & Worthington,
1984). This occurred in the present study.

The index of internal consistency for the GSE was low for this sample (alpha

= .60). Appendix B, Table 1 contains the 7 items of the GSE.

WM. Since it has been suggested in the literature that '
self-efﬁcacy should be assessed in relation to a particular task, four items were
developed for this study to tap the monitors' perception of their self-efﬁcacy in
relation the monitor program. Two items addressed the monitors' perception of

their individual efforts as effective monitors, and two items addressed their

 

64

perception of the efﬁcacy of the program as a whole to make a difference in the
group homes. The index of internal consistency for the four items of the "Program
Self-Efficacy Scale " was moderate (alpha: .71). The items are found in Appendix

B, Table 2.

W- The measure of
WW (OBSE) has demonstrated excellent psychometric

properties as indicated by reported alphas of .86 to .91 in seven studies performed
by Pierce, Gardner, Cummings and Dunham (1989). These researchers also reported
a test-retest reliability coefﬁcient of .75 before they controlled for perceived
organizational change. The correlation increased to .87 when perceived change was
taken into account. Two versions of the OBSE, which differed on a semantic
differential dimension, were given to the same group of subjects to assess convergent
validity. The obtained correlation was .69 (Pierce, Gardner, Cummings and Dunham,
1989). Discriminant validity was assessed by a factor analysis which was performed
on the OBSE and the Rosenberg and Beehr global measures of self-esteem with
measures of organizational commitment, intrinsic motivation, and job and
organizational satisfaction to assess discriminant validity. A two factor solution
emerged with the three self-esteem measures loading on one factor and the
organizational and job measures loading on the second factor. The strength of the
predictive validity of the OBSE for level of organizational satisfaction was found to

be signiﬁcant with a correlation of .59.

65
The index for internal consistency for the 9 item, 5 point Organizational-Based

Self-Esteem Scale (OBSE) was excellent for the present sample (alpha= .97).

Appendix B, Table 3 contains the items of the OBSE scale.

'f inM s

-n- u Dunno" L.- -n-r.. A's. -101!!! . JD . The Job
Diagnostic Survey (JDS), (Hackman & Lawler, 1971; Hackman and Oldham, 1975)
was utilized in this study to measure the genetalsatisfaetion level of individuals with
their voluntary activities with one modiﬁcation. The word " job " in the original
survey was substituted with the phrase " monitoring experience" for the present
sample.

This widely used measure has demonstrated good psychometric properties,
with reported internal consistency of the scale items ranging from .82 (Dailey, 1986)
to .56, and the discriminant validity obtained between the JDS general satisfaction
sub-scale and the other sub-scales of the JDS ranged from .12 to .28 (Hackman &
Oldham, 1975; Oldham & Hackman, 1980). The substantive validity of the JDS
indicates that the variables measured do relate to one another, and to external
criterion variables, as postulated by Hackman ' s and Oldham ' s (1975) theory
(Hackman & Iawler, 1971).

The index of internal consistency for the JDS general satisfaction sub-scale

items (14 items) for this sample, was good (alpha=.84). After reviewmg the content

66

of the items in this sub-scale for this sample, it was decided to label the general
factor as "general satisfaction with the program". The items of the JDS general
satisfaction sub-scale are contained in Appendix B, Table 4.

Although the JDS general satisfaction sub-scale was used in the present study
it should be noted that the survey has received some criticism. Roberts and Glick
(1981) caution the use of the JDS for the following methodological reasons: the
survey does not adequately specify the measurement and analysis of the job
characteristics, the issue of common method variance never gets addressed by
researchers who have utilized the survey, and the uni-dimensional index is not valid.
Roberts and Glick (1981) also discussed the inability of the job characteristics model
to address individuals who are low on higher-order growth need strength. These
issues were addressed in the present study through the following actions: a) other
measures, including archival data, of volunteer experience with the monitor program
were used, b) the JDS index was not computed, and c) another scale which assessed
the volunteers ' level of frustration was developed as another indicator of satisfaction
with the program. Research has demonstrated that volunteers, in general, display
high levels of higher-order grth need strength (Anderson & Moore, 1977; Gidron,
1987; Houghland, Turner, & Hendricks, 1988). Thus, this variable was treated as a

constant in the present study.

67
W. The investigator developed a nine item frustration

index, which contains the major frustrations with the program that the monitors
identiﬁed prior to implementation of the study. The index of internal consistency for
this nine item index was moderate (alpha=.67). The frustration items are found in

Appendix B, Table 5.

Th ' i n mmi m n

The Organizational Commitment Questionnaire (OCQ) which was developed
by Mowday, Steers & Porter, 1979 was used to measure prganizaﬁenaleem’tment
in the present study. The psychometric properties of the OCQ were examined by
Mowday, et. al., who administered the questionnaire to 2563 employees in nine
different organizations, ranging from auto companies to hospitals, banks and
universities. Internal consistency coefﬁcients ranged from .82 to.93, with the median
alpha correlation being .90. Test-retest stability coefﬁcients (with 3 to 4 month
intervals) ranged from .56 to .75. A factor analysis performed on six of the nine
employee samples resulted in a one factor solution: that of commitment. Martin and
Shore (1989) obtained an alpha coefﬁcient of .91 on a sample of professional staff
at a large hospital, and an alpha coefﬁcient of .89 on a sample of clerical staff at the
a large bank.

Discriminant validity coefﬁcients obtained between the OCQ and measures
of job involvement ranged from .30 to .56 across four different samples. The

common variance shared by organizational commitment and the other measures was

68

primarily around 24%, and did not exceed 50%.
The index of internal consistency for the 12 item OCQ for the present sample
was good (alpha= .84). Appendix B, Table 6 containsthe items included in the OCQ

scale used for this study.

Th In erv n i n Pr M r 5

Ten different strategies comprised the empowerment intervention (the
independent variable)for this study. The strategies included: a) presentations about
the monitor program to local community groups and agencies, b) presentations about
the effectiveness of the monitor program to state level agencies, c) articles for the
state-wide ARC newsletter pertaining to monitor issues, d) the implementation of
People First Groups, e) attendance at other local chapters' monitor meetings, f)
workshops on normalization and the monitor process for group home providers,
g)attendance at monitor training sessions, h) training adults with developmental
disabilities to monitor, i) the compiling of a manual which contained ideas and
solutions to problems encountered on monitoring visits, j) and the implementation
of the local coordinator dyads. At both pre-test and post-test, all monitors were
asked to indicate whether or not they engaged in any of the 10 strategies. As noted
earlier, the manipulation check revealed that contamination at pre-test for both
groups and the spill-over effect in the control group were minimal, and did not pose
a threat to the study. In addition, it was evident that the manipulation of the

independent variable did occur in the experimental group.

Results

A signiﬁcant difference was found between the experimental and control
groups at pre-test on the level of education, as repOrted above. As shown in the
correlation matrix contained in Table 1, only one variable, general satisfaction, was
signiﬁcantly correlated with education. It seems safe to assume that single signiﬁcant
result was due to chance. Random assignment should have ensured that the groups
are not signiﬁcantly different, and that no further analyses which control for the
between-group difference in the level of education are needed. However, since a
signiﬁcant correlation was found between general satisfaction, identiﬁed as an
important mediational variable in the present study, and level of education, the
relationship between the level of education and the dependent variables in this study
was further assessed, using MANCOVAs and ANCOVAs with education as a
covariate. The lack of signiﬁcant results from the MANCOVA and AN COVA
analyses demonstrated that level of education did not signiﬁcantly impact on either
of the dependent variables. Therefore, the covariance analyses are not reported
here; only the results from repeated measures MANOVAs and AN OVAs are

reported below.

69

70

 

 

.520 H

b u E,

30-8w" 38

 

3 30055. graham—9.9 HS
E $sz mmrmhmagg

@ mmzmwt. m>dmm>352

3 gamadoz rm<mr

9 8335.735

3 mmrmbmammz

3 259. 3028 swam

@ yam—maze >3mzc>zom

3 rm<mr 0m mocgdoz

*bAbM :bAbcH

ANV
.8
r8

Auv
.mm..
.oH

”rec

Asv
as
.8
-8.
HS

AMV
.AH..

.OH

.obaa
JOM
P8

Aav
.8
.3
.8.

a;
.8.

r8

3

ON:

.8
.3...
38
.3...
.5
Poo

@
.5

.w .. _._

Jom
.mw... e
.Nw
.3... _._

PS

3

:3...
.3

.5
bu
who

71
Rli hi

The ﬁrst two hypotheses addressed the existence of potential relationships
between self-efﬁcacy, satisfaction and commitment (the mediating variables) and
participation, as deﬁned by group home visits and attendance at meetings (the
dependent variables). Pre-test data were used to obtain the correlations reported in
Table 1, because these data do not reﬂect the intervention impact. Post-test data
were used to test the mediational model, because these data reﬂected the impact of
the independent variable.

The ﬁrst hypothesis, that organizational satisfaction, commitment and
perceived self-efﬁcacy (mediating variables) should be signiﬁcantly related to the two
primary dependent variables: number of monitor visits and attendance at monitor
meetings was not supported by these post-test data. A set of three regression
equations (a total of 18 regression equations) was performed on each mediator
variable (program self-efﬁcacy, commitment and satisfaction) and dependent variable
(group home visits and attendance at monitor meetings) to test for mediational
effects (Baron & Kenny, 1986). In order to establish the existence of a mediational
model, all three equations for each mediator variable and each dependent variable
set had to show that the speciﬁed independent variable had an effect on the speciﬁed
dependent variable in the equation. In none of the sets of analyses did all three
conditions hold. Thus, the researcher was not able to statistically establish a

mediational model for program self-efﬁcacy, satisfaction or commitment on either

72

group home visits or attendance at meetings. Table 2 contains the zero order
correlations between the two sets of mediating and dependent variables. The
correlation between general satisfaction and group home visits was .27 (p < .01) and
between general satisfaction and attendance was .32 (p < .001). Only two other
signiﬁcant correlations were found between the mediating variables and the primary
dependent variables. Program self-efﬁcacy signiﬁcantly correlated with group home
visits (t = .32; p < .001) and commitment signiﬁcantly correlated with attendance at
meetings Q = .34; p < .001).

The second hypothesis, that there should be positive relationships between
perceived self-efﬁcacy, organizational commitment and satisfaction and volunteer
retention rates was difﬁcult to test because of the limit of range on the retention
variables: only 9 volunteers terminated their involvement, which represents a 7%
reduction of the 125 study respondents. The retention rate for the experimental
group was 91% (n=60) and for the control group was 95% (n=56).

A between group analysis on the retention variables revealed no statistically
signiﬁcant difference between group assignment and those who terminated or
remained with the program. A chi-square analysis revealed no signiﬁcant difference
between those volunteers who terminated and those who remained in terms of their
level of education or if they knew someone with a developmental disability. Post-test
data were used for both types of analyses, since the pre-test represented 100% of the

volunteer population.

73

 

 

 

Table 2
o -l._ in o-fﬁ .- --n h m-cUi v .u g-.-n.- v..._..-l-
Mediating variables Visits Attendance n
OBSE .15 .21 53
Program self-efﬁcacy .32" .19 54
Self-esteem .03 .08 54
General satisfaction .27' .32" 54
Commitment .22“ .33" 54
Level of frustration -.01 -.05 . 54

 

'p<.01;"p<.001

74
W

The analyses performed to test Hypotheses Three through Six were repeated
measures MANOVAs and ANOVAs. When statistical signiﬁcance was found in the
group by time effect in a MANOVA analysis, a repeated measures AN OVA was
performed for each dependent variable to assess the signiﬁcance of the trend
between time pre-test and post-test. When a signiﬁcant group by time effect was
found by the repeated measures ANOVA, a Scheffe’ post hoc test was performed to
learn if the differences in the means over time were statistically signiﬁcant. The
Scheffe post hoc test represents a conservative approach to examining the difference
in the means between pre-test and post-test within each group (Scheffef, 1959;
Shavelson, 1988).

The means in the tables which present the results of the repeated measures
analyses show an interesting trend in which the experimental group increases and the
control group decreases. The ScheffeI post hoc conﬁrmed these trends to be

statistically signiﬁcant.

lf-Efﬁ
The third hypothesis, that members of the experimental group should feel
more self-efﬁcacious in their advocacy efforts than members of the control group,

was supported by the present study. A repeated measures MANOVA was performed

on general self-efﬁcacy (GSE), program self-efﬁcacy and organization-based self-

75

esteem(OBSE) to assess the extent to which the experimental and control groups
change over time. F ollow-up repeated measures ANOVA’s were performed to
examine the trends.

The results of the repeated measures MANOVA showed that there was a
signiﬁcant group by time effect (Hotellingsl‘ =.95, est. 13 [3,104] = 33.0, p < .001).
The results of the follow-up repeated measures AN OVAs, reported in Table 3,
indicated signiﬁcant group by time effects for all self-efﬁcacy measures, with one
exception, general self-efﬁcacy (GSE).

In addition, Scheffe’post hoc comparison tests were performed on the program
self-efﬁcacy and OBSE measures for each group. In each case, members of the
experimental group signiﬁcantly increased in their perception of program self-efﬁcacy
(t [52] = 6.66, p < .01); and their mean score on the OBSE, (t [52] = 6.66, p < .01)
while the members of the control group signiﬁcantly decreased in program self-
efficacy, (t [52] = 4.66, p < .01); and their mean score on the OBSE (t [52] = 3.33,
p < .01).

Although there was a theoretical premise that OBSE would be highly related
to feelings of self-efﬁcacy, this sample did not support this position (Wood &
Bandura, 1989; Pierce, J., Gardner, D., Cummings, L., & Dunham, R., 1989)). The
correlation between OBSE and the two self-efﬁcacy measures were: GSE, 1 = .14;

and program self-efﬁcacy, L = .21.

ﬂ ._‘___...._. .._=. “MFA-cm_;_ -_ M

76

.920 u

 

 

 

Z: O .m 45.4mm 3 3-..- -70 5.. m. 9% .0: m: 20. Xda. .. m5
>00000803 580
$50388. Z0850 50-809 H3033" Ono—6 H.502 .580 H.509 9.96 X .550 mum—.00"
08:0 m K gv R EV um 30 M So m So
wnomawB 038$. .uuﬁhod Sumﬁhod... 3.33.53...
030050Em— ma Fuchtv @8wa
833. m0 P828 @833
60:03— 0388. .83.:3 .ooﬁhod Quarsd
96018039. mm 950qu @8008
02:3— ma 936%: uhuhmov
mem Homﬁhoov 839.53 Nubwﬁhoav.
9601803»— 3 H.338 PcmAQmV
833— mg H.803 H9368

 

.bAbu :bAbS

77
W911

The fourth hypothesis, that members of the experimental group should feel
more satisﬁed with the monitor program than members of the control group, was
supported. A repeated measures MANOVA was performed to assess the extent to
which the groups’ levels of satisfaction changed over time. The results indicated that
a signiﬁcant group by time effect was found for the two satisfaction variables
(Hotellings I = .645, F [2,102] = 32.92, p < .001). The results of the repeated
measures AN OVAs, which are found in Table 4, indicated that there was a
signiﬁcant group by time effect for both general satisfaction and levels of frustration
between pre-test and post-test.

The Scheffe post hoc comparison test for general satisfaction revealed that
there was a signiﬁcant increase in the experimental group’s level of general
satisfaction with the program between pre-test and post-test (t [50] = 6.66, p < .01),
while there was a signiﬁcant decrease in the control group’s level of general
satisfaction (t [51] = 3.33, p < .01). The Scheffe post hoc comparison for the level
of frustration revealed that the experimental group signiﬁcantly decreased in the level
of frustration that they felt between pre-test and post-test (t [53] = 7.33, p < .01).
However, no signiﬁcant change in the level of frustration was found for the control
group between pre-test and post-test.

To further assess the impact of the intervention on the experimental group in
terms of the frustrations felt, four of the nine ﬁ'ustration items, two of which were

addressed by the intervention, and two of which were not, were examined by

78

ﬂ

H020 0

.. .m. Tami m3. ,,_..-.X..DB. .. m.

 

 

 

g.§ .iw. U 0% ”wmmﬁmo: 5...- -
360000803 .580
mmammmamo: K0850 30-8% womi0ﬂ 08% mag» .580 mm0£ 03% x .380 mm09
08% h. K ﬁbv K BE .m Ea .m ac m Ga
00502: mmammwamg 0.03.88 .omCLouv 8.33.88.
00601802»: mN 9.328 m.m2.m8
83:: mm 902.03 MmmAch
mEQOaoB Hmaﬁpoﬁ 3.83.8.3... tuuﬁpod.
96018053: mm N833 Hoﬁbav
02:3— 3 553.3 meﬁbv

 

bAbS

79

chi-square analyses. The two frustrations which were addressed by the intervention
were, the assessment forms used during the visits and access to the group homes.
The two frustration items which were not addressed by the intervention were, the
interactions between the monitors and the group home staff and the monthly group
home visits. The results of the chi-square analyses, which are reported in Table 5,
revealed signiﬁcant differences between the two groups on their level of frustration
with the assessment forms that they use during group home visits, and their limited
access to the group homes at post-test. More members in the experimental group
showed less frustrations on both variables than the members of the control group.

Thus, the intervention had an impact on reducing the frustration levels for the forms

and group home access.

P i ' i n D n n ' l

The fifth hypothesis, that the members in the experimental group should visit
the group homes more often and should attend more monitor meetings than the
members in the control group, was supported.

A repeated measures MANOVA yielded a significant group by time effect for
group home visits and attendance at local meetings (Hotellings I = 1.49, E [2,105]
= 78.04, p < .001). Since a signiﬁcant difference was found, separate repeated
measures ANOVA ' s were performed for each dependent variable, the results of
which are reported in Table 6. In both cases, a signiﬁcant group by time effect was

found.

80

 

 

 

Table 5
ffi i n f f r i fr m h F
N =1 rim = r l = 4
Assessment Time
Pre-test Post-test

Frustration Variable

group .2 (%) .1; (%)
Forms

experimental 35 (63) 4 ( 6)

control 23 (42) 31 (58)

xi." NS 29.16”

Staff interactions

experimental 16 (30) 16 (31)

control 17 (29) 18 (33)

2;- NS NS

Group home access

experimental 20 (38) 9 (15)

control 26 (47) 27 (51)

2‘: NS 1216*

Monthly visits

experimental 6 (11) 4 ( 7)

control 5 (8) 7 (13)

21;" NS NS

 

*p<.001

81

H.020 a

.m r. 2.898 <.. ...._. ...~U o.mmm.8 TH. a... X128. .. m.

 

 

 

z. w n:. U -
388808 4.80
5&8 mn 288800 50.80» $8308 088 mm09 .380 mmonﬂ 9.88 X .380 mm09
08% m K. 69 K 85 m as .m 30 m :5
Saw NbcCLod 3.83.85... HNuMwGLoqv.
08018088 mm NmNANEV uhmﬁme
888— mg NKCMS quﬁbuv
5.208800 mbaﬁhoqv 3.33.86... 5.33.85...
08018088.— mm Hmmﬁwv Nuwbawv
883— mg TESS #0er

 

bAbB

The Scheffe post hoc test indicated that there was a signiﬁcant increase in the
number of group home visits and meetings attended by members of the experimental
group between pre-test and post-test (1 [S3] = 5.29, p < .01; 1 [53] = 5.79, p < .01,
respectively). While there was a signiﬁcant decrease in the control group's home
visits, (1 [52] = 2.77, p < 01.), there was no signiﬁcant change in their attendance

behavior.

Program Commitment

The sixth hypothesis, that the members of the experimental group should have
a stronger sense of commitment to the program than the members of the control
group was not supported by the present sample. The repeated measures AN OVA
yielded no signiﬁcant trends between per-test and post-test on organizational

commitment. Since no signiﬁcant trends were found, no further analyses were

performed.

Summary

As predicted, the volunteers in the experimental group did increase in their
level of perceived program self-efﬁcacy and general satisfaction with the program as
well as in their level of program participation. This observation was consistent with

the study ' 5 primary theoretical perspective on empowerment. The ability of the

82

83

volunteers to directly manipulate those factors that they deemed to be barriers to
effective monitoring can increase volunteers' perceptions of program self-efﬁcacy
and general satisfaction and levels of participation.

In addition to verifying the gains in the treatment group,this study noted a
downward trend in the control group. Volunteers displayed decreases in feelings of
program self-efﬁcacy, general satisfaction with the program and level of participation
between pre-test and post-test assessments. Although this trend was not speciﬁcally
addressed in the hypotheses, the theoretical perspectives underlying this study would
indicate that it should have been anticipated. Research studies have demonstrated
the upward spiral and have suggested that the inverse would be likely to occur
(Wood & Bandura, 1989, Morris & Sherman, 1981). The development of the
hypotheses in this study was in the context of examining the positive effect of an
empowerment strategy, and the potential negative changes in the control group were
not adequately addressed. This observation provides some evidence for speculating
that the intervention had an even stronger impact on the volunteers in the
experimental group, since the groups were not signiﬁcantly different in their levels
of program self-efﬁcacy, general satisfaction or program participation prior to the
intervention implementation. It might be assumed that not only did the intervention
increase perceptions and participation, but it also countered a downward trend.

Two results were not consistent with the hypotheses of this study. The levels
of general self-efﬁcacy and commitment remained stable for the volunteers in the

experimental group. These ﬁndings are further explored in the following section.

Discussion

The data provide empirical support for some of the hypotheses of this study.
Also observed have been trends consistent with the theoretical perspectives
explicated above. However, other interesting results emerged which did not conﬁrm
others of the hypotheses presented. The discussion will ﬁrst focus on those
hypothesized relationships conﬁrmed by the data. These include program self-
efﬁcacy, general satisfaction and the two outcome participation variables: group
home visits and meeting attendance. The discussion will then switch to those
hypotheses not conﬁrmed by the data, which include organization-based self-esteem,

general self-efﬁcacy and commitment.

As demonstrated by the results, the volunteers in the experimental group
signiﬁcantly increased their perception of program self-efﬁcacy, while the control
group volunteers signiﬁcantly decreased. As predicted, the observed trend in the
experimental group may have been due to the volunteers' ability to exercise control
over the factors that they perceived to be inﬂuencing the effectiveness of the monitor
program. The volunteers identiﬁed and manipulated factors which they perceived

to be barriers to their efforts. to improve the quality of life in the group homes. Such

84

85

factors included the lack of clear program goals, the lack of a comprehensive monitor
program plan and tasks, and the unresponsiveness of the agencies within the group
home system to the volunteers' efforts and suggestions. 'It seems this involvement
may have enhanced their expectations that they could improve the quality of life in
the group homes. Support for this explanation is found in several research
endeavors, including goal setting (Mento, Steel & Karren, 1987; Locke, Latham &
Erez, 1988), self-efﬁcacy (Wood & Bandura, 1989), organizational satisfaction (Shore
& Martin, 1989), and organizational commitment (Steers, 1977; Morris & Sherman,
1981). The common theme among these studies, which was also observed here, is
that positive perceptions of situations may be enhanced through direct involvement
in organizational activities, especially if such involvement leads to the fulﬁllment of
expectations, which in turn may enhance positive outcomes, such as increased
participation.

An inverse trend could also be explained in the same manner. Negative
perceptions could be reinforced through the inability to fulﬁll expectations which
eventually may lead to decreased participation outcomes. Such may have been the
case for the volunteers in the control group who signiﬁcantly decreased in their
perception of program self-efﬁcacy.

This general downward trend has also been observed by Bandura (1982), who
asserted that individuals who do not feel that they can accomplish a task will not
continue to attempt that task. Decreased attempts at tasks will impact on the level

of participation, which has been partly determined by individuals' perceptions of

86

their ability to affect the situation. Lower levels of participation result in the
obtainment of fewer goals which can adversely impact on satisfaction, which then
determines future levels of perceived self-efﬁcacy and future expectations. The fact
that the volunteers in the control group not only decreased in their level of perceived
program self-efﬁcacy but also in their level of satisfaction with the program and their
level of monitoring activity (participation), is consistent with Bandura' s hypothesis.

The trend observed in the experimental group in terms of program self-
efﬁcacy corroborates two other theoretical perspectives; the ﬁrst is that the
obtainment of personal goals enhances a sense of self-efﬁcacy (Wood & Bandura,
1989). As noted in the results, 79% of the volunteers became involved with the
monitor program to fulﬁll their personal goal, namely to improve the quality of life
in the group homes. Through the intervention strategies, the volunteers in the
experimental group directly revised aspects of the monitor program that had
precluded them from realizing their personal goal. This empowerment intervention
demonstrates Wood and Bandura ' s assertion that the obtainment of personal goals
enhances individuals ' expectation of future rewards. Such expectations then lead to
the initiation of self-directed behavior to pursue those goals. In the present case, the
empowerment intervention represented the volunteers' self-directed behavior, and
their personal goal was the improvement of the quality of life in the group homes.

The second theoretical perspective supported by this study places importance
on the role of volunteers within an agency and their ability to fulﬁll their

expectations of personal rewards (Houghland, Turner & Hendricks, 1988; Gamm &

87
Kassab, 1983; Gidron, 1987). Since volunteers choose to afﬁliate with an agency

based on their expectations of future obtainment of personal rewards, their
perception of self-efﬁcacy will be determined by the fulﬁllment of their expectations
(Gidron, 1987). At pre-test 50% of the volunteers in the experimental group and
56% of the volunteers in the control group reported that their personal expectations
had been met by the monitor program. The correlation between the volunteers'
perceived program self-efﬁcacy and their feelings of fulﬁlled expectations was a
signiﬁcant .37 (p < .05), consistent with the ﬁndings of Houghland, Turner and
Hendricks (1988) and Wood and Bandura (1989). The present empowerment
intervention was the vehicle that the volunteers used to fulﬁll their personal
expectations which enhanced their effectiveness as monitors. At post-test, 76% of
the volunteers in the experimental group reported and 59% of the volunteers in the
control group reported having their expectations met, yielding a signiﬁcant 26%
increase in the experimental group and a slight decrease in the control group.

The above ﬁndings have important implications for volunteers and agencies
with which they afﬁliate. The central focus in terms of volunteer self-efﬁcacy seems
to be the interaction between fulﬁllment of volunteer expectations and articulation
of agency mission and goals. As hypothesized by Conger and Kanungo (1989),
identiﬁcation of personal needs and expectations could help agency managers clarify
and predict the goodness of ﬁt between their goals and agency operation and the
level of employee satisfaction, performance and turnover. 'Ihe empowerment

intervention represented the identiﬁcation process which impacted on the interaction

88

between volunteer needs and expectations and agency management and mission.

A signiﬁcant increase was observed in the level of satisfaction with the
monitor program for the volunteers in the experimental group three months after
they implemented the intervention. There are two salient issues here. The ﬁrst one
is that the volunteers in the experimental group had the opportunity to take on more
tasks, as well as to follow through with the existing monitor tasks to completion. This
is consistent with Hackman and Lawler (1971) who postulated that satisfaction
increases when individuals have the opportunity to gain a sense of the entirety of a
task as well as to be able to perform a variety of tasks.

The second issue is that both task entirety and variety for the volunteers in
this study were speciﬁcally related to the goals and expectations articulated by the
volunteers, and thus the inclusion of those in the empowerment intervention
strategies led to the increase in general satisfaction. Ivancevich & McMahon (1982)
stressed the importance of recognizing goal and feedback speciﬁcity.

The intervention strategies represented the integration of the two issues
related to the tasks noted above. Changes in the tasks involved the expansion of
existing group home visit tasks (entirety) and inclusion of tasks not directly related

to group home visits (variety). Both types of tasks involved feedback processes that

89
were speciﬁcally related to the volunteers' personal goals; to improve the quality of

life in the group homes. Since both types of changes occurred, as demonstrated by
the post-test manipulation check, it may be assumed that increased satisfaction was
a result of increased task entirety and variety, coupled with feedback and goal
speciﬁcity, an expansion of Hackman and Lawler's (1971) theory.

The Frustration Index provides further evidence of the volunteers' increase
in their general satisfaction on a different dimension. The index listed particular
mediating factors that the volunteers identiﬁed as having a strong negative inﬂuence
on their ability to be effective monitors. As stated by Shore and Martin (1989) and
Van De Ven and Perry (1980), mediating factors such as co-workers, supervisors, and
so forth, can inﬂuence an individual ' 5 sense of satisfaction with an organization.
Since the present intervention was a direct manipulation of mediating factors that the
volunteers perceived to be barriers to the monitor program, it could be assumed that
the decrease in the Frustration Index would substantiate the increased sense of
satisfaction. The correlation between the frustration index and the JDS general
satisfaction sub-scale was signiﬁcant (r= -.27, p < .05).

The Frustration index also served a second purpose. It provided an
opportunity to assess the volunteers ' feelings about speciﬁc elements of the monitor
program that the volunteers identiﬁed to be problematic and that they were able to
directly change through the intervention. This intervention process refers back to the
issue of task speciﬁcity discussed above. As noted in the results section of this study,

at post-test, the volunteers in the experimental group only felt less frustrated with

90

those mediating factors (the assessment forms and access to the group homes) that
were addressed by the intervention strategies. Their level of frustration did not
signiﬁcantly change for the aspects of the program that they chose not to change.
The results further show that the volunteers in the control group did not
signiﬁcantly change in their level of frustration with speciﬁc aspects of the program.
Although, their level of satisfaction signiﬁcantly decreased, as indicated by the JDS
general satisfaction sub-scale, their level of frustration did not. One might speculate
that once the frustrating factors have been identiﬁed, the level of frustration with
those factors might not change, but the adverse impact that such factors have on
individuals ' level of satisfaction continues. Thus, the volunteers ' level of frustration
would not change, but their general level of satisfaction would. Measurement
constraints may partly contribute to this observation, because the Frustration Index
comprised dichotomous variables. It might have been more useful for the volunteers
to indicate their levels of frustration on a likert scale. This would have made it
possible to assess the relationship between the level of frustration with the mediating
factor and the type of intervention strategy or task that volunteers used to address

that factor without the restriction of range imposed by the current measuring system.

91

The volunteers in the experimental group did signiﬁcantly increase their group
home visits and meeting attendance between pre-test and post-test. Although, the
intervention strategies did not directly address aspects of either participation variable,
the volunteers' outcome behaviors did increase.

As stated by several researchers (Conger & Kanungo, 1989; Morris &
Sherman, 1981, Hackman & Lawler, 1971) mediating factors such as individuals'
perceptions, the mission goals of an organization, incentives, co-workers, supervisors,
can inﬂuence performance outcome. The strategies employed by the volunteers of
the present study addressed mediating factors that were identiﬁed as barriers which
blocked the volunteers from achieving their goals. Through the manipulation of
these mediating factors, the participation level of the volunteers in the experimental
group increased.

The number of visits to the group homes by the volunteers in the control
group signiﬁcantly decreased, while their level of meeting attendance did not change.
Although, the reason for the decrease in monthly visits may be due to a demoralizing
factor, which random assignment cannot take into account, the reason for decline is
consistent with the theory discussed in the previous sections of this paper. The
decline was probably due to a continued feeling of ineffectiveness by the volunteers,
who were aware that they had to wait three months before they could implement the

intervention. Although they knew that they would be involved in the intervention,

92
they were still dealing with the present situation, which was adversely impacting on

their ability to participate.

Two processes seemed to have occurred simultaneously in this study. If we
can assume that the volunteers were displaying a downward trend in terms of their
overall participation with the monitor program, then it might be assumed that all
volunteers would continue to decrease in their levels of program participation,
because of their lack of conﬁdence in their ability to make changes (Wood &
Bandura, 1989). Then we might further assume that the present intervention
countered the downward trend for the volunteers in the experimental group, while
the volunteers in the control group continued to display the original downward trend.

Although there is evidence to support the above assumptions, more data
collection time periods would have provided further empirical support. For example,
there should have been at least two pre-test time periods which would have detected
a overall trend prior to the intervention. In addition, a follow-up assessment should
have been done after the volunteers from the control group implemented the
intervention. This would have provided further evidence to detect if the intervention
countered the downward trend as hypothesized.

As stated earlier, models and theories derived from research on employee
behavior have been adequately applied to the volunteers in the present study.
However, it seems that the notion of personal expectations and goals play a more
prominent role in inﬂuencing the volunteers' level of participation in the present

study than for employees in general.

93

Further examination of the interactions between the personal characteristics
of the volunteers and the goals and management processes of the organization is
needed. Conger and Kanungo (1989) provided a model of empowerment which takes
such interactions into account for employees, while llsley and Niemi (1981) found
this process to be true for volunteers. The enhancement of outcome variables is
through the manipulation of organizational and contextual factors which can impact

on personal characteristics, perceptions and expectations.

Although there was a signiﬁcant increase in the level of organization-based
self-esteem for the volunteers in the experimental group, surprisingly, the results of
this study did not provide support for the inclusion of the OBSE as a related measure
of self-efﬁcacy. The correlation between OBSE and program self-efﬁcacy was
nonsigniﬁcant (r = .21). Perhaps self-esteem in terms of organizational afﬁliation was
not as relevant for the volunteers in this study in terms of their perceived self-
efﬁcacy, since the volunteers had a speciﬁc goal in mind; that of improved quality of
life in the group homes. Thus, rather than having a need to feel good about their
afﬁliation with ARC, the volunteers may have had a stronger need to feel good about
the quality of life in the homes.

This position is somewhat substantiated by the fact that 63% of the volunteers

94
chose to afﬁliate with ARC in order to improve the quality of life in the group

homes. In addition, 92% of the volunteers did not ﬁnd their afﬁliation with ARC to
be frustrating which would indicate that their level of organization-based self-esteem
would be high. The correlation between the volunteers' level of OBSE and their
satisfaction with ARC was a signiﬁcant .29 (p < .05).

Although the level of organization-based self-esteem for the volunteers in the
experimental group changed signiﬁcantly between pre-test and post-test and
signiﬁcant correlations were obtained between OBSE and satisfaction and
commitment, this measure did not act as an indicator of self-efﬁcacy in this sample.
Further research may be needed to ascertain the meaning of organization-based self-

esteem in terms of volunteers.

mm

The ﬁrst of the hypotheses not supported by the present study to be discussed
here is the hypothesis regarding general self-efﬁcacy. Although, the intervention had
a positive impact on the volunteers' perception of program self-efﬁcacy, the
intervention did not signiﬁcantly impact on the volunteers' sense of general self-
efﬁcacy. In addition, the results of the study demonstrated that the volunteers'
general feelings of self-efﬁcacy was not a signiﬁcant factor in increasing volunteer

participation, or inﬂuential in determining the volunteers' sense of accomplishing

95

their monitoring goals.

Several explanations may account for the present ﬁndings. One reason may
be that a measure of general self-efﬁcacy was too broad, and thus was not robust
enough to have an impact on a person's sense of accomplishment with a speciﬁc
task. As stated earlier, task speciﬁcity may be an important factor in enhancing
positive perceptions and outcomes. In addition, there may be mediating factors such
as commitment to a goal, or satisfaction of personal expectations that are more
inﬂuential in determining an individual ' 5 overall feeling of self-efﬁcacy. It might be
more plausible to suggest that the volunteers, the majority of whom were parents
(65 %), were determined to make the monitor program work, because they were
committed to a speciﬁc goal; the welfare of their children. This explanation seems
valid in light of the previous ﬁndings and assertions noted above, which speak to the
speciﬁcity of the problems and tasks.

It seems that a consistent pattern has emerged in the results, which is that
speciﬁc attitudes and beliefs can alter and be altered by actions that impact on
speciﬁc problems either directly or through mediating structures. It would be logical
to suspect that measures of global constructs would not provide insight into the
relationships between individuals and their organizational contexts. The fact that the
general commitment levels of the volunteers were not altered by the intervention
provides further evidence of this position.

Perhaps it may have been inappropriate to measure the volunteers' general

feeling of self-efﬁcacy, because the intervention strategy targeted structures that had

96

speciﬁc impacts on the program and not individuals. Thus, the measurement of
internal global psychological constructs may be inconsistent with the theoretical
guidelines of this intervention and study, in which the. independent variable was
manipulated by the study participants who identiﬁed the relevant targets for change
in the monitor program. Thus, the results of the present study seem to demonstrate
that those constructs which are more global and internally based did not change
between pre-and post-test, where as those constructs that were program speciﬁc did
signiﬁcantly change between pre-test post-test.

Two other possible explanations for the present results involve methodological
considerations. The ﬁrst explanation may be that three months may not have been
enough time to detect a change in the global constructs. The second explanation for
the lack of change in general self-efﬁcacy may be measurement error. As indicated
in the correlation matrix in Table 1, the measure of general self-efﬁcacy did not
correlate well with any of the other measures in the study, and the internal

consistency coefﬁcient was only moderate.

0"1L‘mmi

There are two possible explanations why the level of commitment felt by the
volunteers in the experimental group did not signiﬁcantly change. The ﬁrst reason

may be that the volunteers' commitment to the program was the motivating factor

97
for them to not abandon the program, but instead attempt to change it. This speaks

to the possibility that commitment may determine the dynamic interaction between
the individual and the organization. The interaction in this case is the congruence
between the volunteers' personal expectations of rewards and the monitor program
mission: to improve the quality of life in the group homes. Thus, it might have been
inappropriate to measure the success of the intervention in terms of its inﬂuence on
the level of volunteer commitment. As stated by Mowday, Porter and Steers (1982),
three aspects comprise organizational commitment. They are (a) a strong belief in
organizational goals, (b) a willingness to exert effort and, (c) a desire to remain. All
three characteristics were present at pre-test as demonstrated by volunteers'
responses to the OCQ items. The fact that only 7% of the volunteers terminated the
program between pre-test and post-test demonstrates that, in light of the ﬁ'ustrations
that the volunteers felt about the monitor program, they were willing to remain.
The second reason for the lack of a signiﬁcant change in the level of
commitment felt by the volunteers may be that perhaps the volunteers had a greater
sense of commitment to their children than to the program. This is somewhat
supported by the fact that the majority of the volunteers joined the program with the
expectation that they would improve the quality of life in the group homes. As stated
earlier, 79% of the volunteers indicated that their main reason for choosing to
volunteer was to improve the quality of life for their children. The monitor program
was the only formal mechanism that these individuals had to inﬂuence the conditions

of the group homes. Thus, the decision to afﬁliate with the program and later

98

revamp it, may have been the result of the commitment the volunteers felt toward
their children and not toward the program.

The major implication here is that the level of commitment may have been
a strong motivating factor in inﬂuencing the volunteers' desire to change the
program rather than abandon it. The fact that the level of commitment remained
stable for both groups of volunteers suggests that this might be the case. In addition,
the attrition rate for both groups is quite low. Only 7% of the total volunteer

population terminated the program between pre-test and post-test.

The Prmess of Emoowet'otent

Although, the data in the present study demonstrate that the empowerment
intervention did occur and had a signiﬁcant impact on the volunteers' perceived
program self-efﬁcacy, general satisfaction and level of participation, it is possible that
factors other than the intervention strategies inﬂuenced the outcomes observed in the
experimental group. One such factor may be the attention that the researcher gave
to the volunteers. In order to rule out an attention factor, the study design would
have called for a third group of volunteers who did not meet with the researcher on
a regular basis, but who did implement the intervention strategies on their own. If
signiﬁcant increases in the dependent variables were observed in this third group,

attention could be ruled out as a possible cause. However, there were not enough

99

participants in this study to form a third group. Groups with small numbers of
subjects lowers the power or the ability to detect differences.

It is also possible to suggest that part of the empowerment process does
include paying attention to a group. Although, it is difﬁcult to assess how much the
attention from the researcher affected the outcome behaviors of the volunteers, there
is empirical evidence that shows that the volunteers in the experimental group did
not signiﬁcantly decrease their levels of frustration with all aspects of the program.
If one assumes that attention could inﬂuence behavior, then one would expect to see
signiﬁcant changes in the variables across the board. This was not the case in the
present study.

The deﬁnition of empowerment used in this study implies that empowerment
is more than just involvement, which primarily entails engagement in some type of
behavior regardless of the relationship of that behavior to anything else.
Empowerment, as deﬁned in the present study indicates the use of behavior that is
directly related to role clariﬁcation and access to information deemed necessary by
the study participants. In order to articulate this deﬁnition of empowerment, the
study design would have called for a third group of volunteers who did not engage
in the empowerment strategies or other behaviors that were directly related to goal
attainment. If no signiﬁcant increases in the dependent variables were observed in
this third group, then we could assume that the empowerment strategies in this study
were not indicators of involvement. However, there were not enough participants in

this study to form a third group.

100

This empirical study has contributed to the explication of two unique concepts
that have not been fully articulated in published articles; the ﬁrst is that of
empowerment as a feedback process, which was treated as the independent variable
and manipulated by the study participants. The second contribution, discussed in the
next section, was the application of the O'Connor model of competence (1989).

This study did not empirically measure empowerment as a dependent or
outcome variable. The empowerment process was the independent variable which
was directly manipulated by the volunteers who had speciﬁc expectations as to the
outcome; namely to be more effective in their monitoring activities. The main point
is that the explication of empowerment as an outcome; namely that individuals
acquire power, was not the focus of the present study. Rather, this study provided
an example of empowerment as a feedback process through which volunteers
clariﬁed their monitoring roles and obtained information that they deemed necessary
for them to effectively make changes in the program.

Access to information and clariﬁcation of roles are the two key elements of
an empowerment process as identiﬁed by Rappaport (1984). Both aspects, that of
access to information and role clarity, were incorporated into the opportunities for
inﬂuence strategy of the present intervention.

Deﬁning the empowerment process in terms of information accessibility is
further supported by the research ﬁndings on feedback and performance (Ilgen,
Fisher & Taylor, 1979; Greller & Harold, 1975). The notion that the volunteers

should have control of and direct access to information that they deemed necessary

101

to improve the monitor program was in part, borrowed from the work of Greller and
Harold (1975) who maintained that the closer the source of information is to an
individual, the more reliable the individual will ﬁnd the information. In terms of the
present intervention, the volunteers were the source of the information for identifying
and rectifying problems.

The second key element of an empowerment process as deﬁned by Rappaport,
that or role clarity, has been articulated by the research ﬁndings of Ivancevich and
McMahon (1982) who maintained that if individuals perceive that they have control
over the setting of program goals, their expectation for future outcomes is more
positive. Not only did the volunteers in the present study identify the program and
personal goals, but they also reformulated their roles as monitors in terms of those
goals.

The impact of combination of information access and role clarity on outcome
behaviors (which represents the major strategies of the present intervention) has
been examined by Mento, Steel and Karren (1987) who found that speciﬁc feedback
and speciﬁc goals most effectively enhanced individuals' motivation and
performance. The fact that the volunteers utilized self-generated feedback processes
to develop and implement the present intervention, and were successful in increasing
their perceived self-efﬁcacy, satisfaction and their own level of participation, namely
group home visits and meeting attendance, is strong evidence for the inﬂuential role
of such feedback.

Since these processes represented the independent variable in the present

102

study, and a manipulation check veriﬁed that such activities did occur in the
experimental group and signiﬁcant changes were observed in both the mediating and
dependent variables, it can be assumed that the feedback processes positively
inﬂuenced the outcomes of the study. Such results lend support to the notion that
speciﬁc goals and feedback are the most effective means to enhance positive
perceptions and performance (Mento, Steel & Karren, 1987). This has one
important implication in terms of the present study. It offers a possible explanation
for the change in the volunteers' perceptions and behavior within a three month
period. Although, the volunteers were eager to change the monitor program, they
were not swayed by the prospect of any type of change. They had a speciﬁc agenda,
and they were seeking speciﬁc ways to address that agenda. As articulated earlier,
the intervention strategies were designed to address very speciﬁc issues through the
manipulation of speciﬁed mediating factors that were considered to act as barriers
to the volunteers ' abilities to be effective. Changes in the perceptions of volunteers
were linked to speciﬁc aspects, such as program self-efﬁcacy versus general self-
efﬁcacy.

An example of the role of speciﬁcity will further clarify this important
assertion. Prior to the intervention, the volunteers identiﬁed several problems and
developed strategies that speciﬁcally alleviated those problems. For example, at pre-
test, 63% of the volunteers found the assessment forms to be frustrating to use for
a variety of reasons. At post-test, only 7% of the volunteers in the experimental

group found the forms to be frustrating, where as 69% of the volunteers in the

103

control group were still frustrated with the forms. The dramatic jump of over 57%
of the volunteers in the experimental group who were no longer frustrated with the
forms was due to the fact that the volunteers revised the forms and the process of
reporting the information. The revision of the forms partly enhanced the
volunteers ' perception of self-efﬁcacy and level of satisfaction. This change, coupled
with the manipulation of other mediating factors, lead to the signiﬁcant increase in
the dependent variables.

This intervention provided an innovative method to identify and manipulate
factors that can inﬂuence the relationship between goal commitment and
performance. As Latham (1982) and Locke, Latham and Erez (1988) concluded,
internal and external factors and their interactions can strongly inﬂuence the role of
goal setting in performance outcome. By utilizing the internal sources of standards
(goals) and achievements, namely the volunteers, as developers and implementors of
the intervention, salient factors that could impact on the ability of the volunteers to
enhance the quality of life in the group homes have been identiﬁed and incorporated
into the monitoring process.

Since the results of past studies that have examined the relationship between
goal setting and performance have been inconclusive, perhaps the tenets of goal
commitment theory may be best served through studies such as this one, which is
longitudinal and places the manipulation of the inﬂuential factors, as identiﬁed by
the participants in their control. This study demonstrates that empowerment and

feedback processes compliment each other and appropriately operationalize models

104

of competency building and goal setting.

The task of identifying the key players in the monitor program system and
developing strategies for intervention with those key players was the primary
responsibility of the volunteers. The investigator chose to collaboratively design the
intervention with the volunteers, because they represented internal sources of
feedback, standards and assessment of achievement as articulated by O ' Connor

(1989).

Ali'nf 'nnerf n

The use of the O'Connor (1989) model of competence in the present
intervention demonstrates that strategies can be developed which identify potential
conﬂict between the various constituents of an organization, and impel change agents
to seriously consider that the group who is implementing an empowerment
intervention is a formidable one. Thus, in the present case the volunteers were
chosen to be the internal source of standards and assessment of achievement,
because they were responsible for the execution of the program, and thus the most
appropriate group to identify the barriers as well as the solutions. The end result
was that 15 different local chapters accomplished the general goals of ARC/ M while
setting their own goals, strategies and measures of achievement.

The theoretical tenets of the O'Connor model are very similar to the tenets

105

of goal setting and performance outcome models, which basically state that the
relationship between goal setting and performance is mediated by both personal and
situational factors (Locke, 1986; Ivancevich & McMahon, 1982). When internal
sources of standards and achievement identify the problems, as well as the factors
and mechanisms needed to rectify the problem, the processes and outcomes are
closely linked to such sources. In this case, the processes had been identiﬁed as
access to information and role clarity, which are components of goal setting and
feedback processes. These two processes were operationalized through the
opportunities for inﬂuence and social support strategies of the intervention.

The O ' Connor model provides a useful framework for future interventionists
and groups desiring change to identify the sources, targets and strategies involved in
change. Future research in the area of empowerment and intervention should

involve the application of this model or its derivative.

The measurement model used in this study treated the mediational variables
of self-efﬁcacy, general satisfaction, organization-based self-esteem and commitment
as separate constructs. Since this measurement model has received empirical
support, that these variables represent separate constructs, in the organization

behavior research, it seemed valid to utilize this measurement model here. However,

106

as demonstrated in Table 1 in the results section, signiﬁcant correlations were
obtained between several of the mediational variables, which may call into question
the validity of the measurement model used. Justiﬁcation for using the present
measurement model comes from the fact that very few measurement models of
psychological constructs have been developed and tested with a volunteer population.
Thus, it seems appropriate to use existing measurement models that have been
developed and supported by research on populations with similar characteristics as
the population presently being tested. The apparent lack of differentiation of the
mediational variables may be due to a variety of factors including, but not limited to
the type of population being tested. Until further research using this measurement
model on the volunteer population is done, it is imprudent to deem this
measurement model to be invalid.

A question may be raised as to whether volunteers in the experimental group
were particularly receptive to change. Since there was random assignment, and if the
outcome of the experimental group was the product of an inclination or a simple
receptivity to change, it should have been similarly evident in the control group as
well. That the control group not only did not increase, but displayed signiﬁcant
decreases (which is the opposite of the experimental group) in the mediating and
dependent variables is evidence of the experimental group's responsiveness to
treatment and not just a receptivity to change.

Another issue related to design is that random assignment cannot take into

account the possible demoralization factor in the control group, as evidenced by their

107

downward trend in most of the major variables of the study. Although this trend may
be the result of a sense of demoralization, this ﬁnding is consistent with the
theoretical orientation of this study.

Longitudinal designs are more appropriate than cross-sectional designs for
assessing trends, as evidenced by the present study. The pre-test post-test design of
this study allowed for the assessment of signiﬁcant increases and decreases in the
mediating and dependent variables over time for both experimental groups.
Although, signiﬁcant trends were detected, it may be more conclusive to measure
changes over a longer time period. For example, several pre-test and follow-up
measures before and after the implementation of the intervention with both groups
of volunteers would have provided more evidence to detect a trend in volunteer
perceptions and behavior as well as the impact of the intervention over time.

The fact that one local chapter chose to not participate in the study may have
caused a sampling bias problem. However, this was not viewed as a substantial
threat, because the supervisors of the monitor program in that local chapter were
operating under a different set of goals, principles and procedures than the rest of
the local chapters involved in the study. For example, this local chapter was involved
in monitoring group homes prior to the initiation of the present ARC program.
Thus, the volunteers in that local chapter did not have the same expectations or
problems as the participants in this study.

This study relied heavily on self-report measures, which typically represent a

possible threat to validity. To compensate for this extensive use, selected behavioral

data were collected and established the validity of those measures.

In any ﬁeld experimental design there must be concern about spill-over effects
from the experimental to the control group. In this study control group members
were carefully assessed at post-test to identify their involvement in any of the
empowerment strategies. Only 2% of possible activities were engaged in by the

volunteers in the control group, indicating there was not signiﬁcant spill-over.
Conclusion

The major contribution of this study is the articulation and application of a
theory of empowerment as a process. The results of this study suggest that it is
possible to articulate a general paradigm that describes empowerment as a process
to guide the development of intervention strategies rather than the more common
consideration of empowerment as an outcome. In this instance, citizen volunteers
articulated their own needs which led to the development of particular intervention
strategies that addressed those needs. This process became ongoing as volunteers
felt more efﬁcacious and satisﬁed and thus increased their involvement. These
enhanced levels of perceptions and participation by the individuals led to the
increased effectiveness of the overall monitor program.

It is not necessary for researchers to identify those factors or their interactions
that can inﬂuence changes in outcome behavior, because the most effective

identiﬁcation of such factors should come from the volunteers themselves. It is the

108

109

responsibility of the researcher to provide a model through which to implement
interventions that enhance outcomes speciﬁed by study participants, as demonstrated
in the present study.

Future research in this area should explore the interactions between the
volunteer and agency with the purpose of enhancing the ability of volunteers to
realize important personal goals and to increase the likelihood of agencies
accomplishing necessary social goals. In conjunction with this needed research
endeavor, it is time to move empowerment to a more articulated level through which
models of social change interventions can be developed and implemented.

The major theoretical contribution of this study is the attempt to advance the
theory of empowerment and to articulate an intervention that manipulated structures
and processes rather than individuals. Such manipulation forces one to rethink the

nature of the context around them, which is the initial step to change.

APPENDICES

APPENDIX A

1 10
APPENDIX A

Table 1

Otrooefﬁg'eng of major eategorieal vao'aoles of the study at ore-test

 

 

 

Group
Experimental(r_i=66) Control(o=59)
. 1v

Vanable g(%) n(%) E
Know someone with a disability 6508) 56(95) --
Relationship to a person with a disability

parent 40(60) 40(89) —

friend 12(18) 804) —

relative 8( 12) 3( S) ..
Reasons for volunteering

improve the quality of life 41(62) 36(61) --
Frustration with assessment form 38(58) 32(55) 5.60“
Frustration with group home staff 20(30) 17(28) --
Frustration with group home access 25(38) 28(48) --
Frustration with group home visits 7(10) 5( 9) —

 

‘£<.05;"£<.001

 

111

o-r--

 

 

 

APPENDIX A
Table 2
R - , o.ftes fr hmeni of oooni' .- --n b no.
Group
Variable Experimental Control
.9. M62) 2 M62) 150“)
General self-efﬁcacy 64 3.10( .77) 59 3.20( .86) .46(1,123)
Program self-efﬁcacy 66 3.30( .44) 59 3.15( .43) 2.47(1,123)
General satisfaction 66 5.50( .82) 59 5.60( .85) .90(1,123)
Level of frustrations 66 2.20( 1.80) 59 1.70( 1.60) 2.08(1,123)
Commitment 65 5.50( .86) 59 5.50( .73) .06(1, 122)
Self-esteem 64 1.80( .76) 58 2.00( .85) 1.49(1,120)
Age (years) 64 5200(1350) 59 5400(1300) 1.15(1,123)
Length of program
service (months) 64 20.00(12.60) 58 21.(I)(B.00) .64(1,1ﬂ))
# of monitor meetings
attended within three
months 65 1.70( 1.70) 59 1.40( 1.10) 1.75(1,122)
# of monitor vists
within one month 65 2.70( 1.90) 59 2.50( 2.50) 2.11(1,122)
Level of education 55 2.50( 1.40) 54 1.90( 1.30) 7.48(1,107)*

 

"' g<.05

112

APPENDIX A

 

 

 

Group

Dependent variable Wayne Other Counties

0 M62) 0 M62) £011)
General
satisfaction 34 5.70( .76) 91 5.50( .85) 2540,123)
level of
frustration 34 2.20( 1.60) 91 2.00( 1.70) .47(1, 123)
Program ‘
self-efﬁcacy 34 3.10( .38) 91 3.00( .36) 1.7l(1,123)
General
self-efﬁcacy 34 3.00( .70) 91 3.10( .84) 530,123)
Organizational-
based self-esteem 33 1.70( .58) 89 2.00( .86) 3.700,120)
Organization
commitment 34 5.60( .73) 90 5.40( .81) 1330,122)
Age(years) 34 53000230) 91 520004.00) 010,123)
Group home visits
(per month) 34 3.10( 2.10) 90 2.60( 1.90) 1.43(1,122)
Attendance at
meetings
(per three months) 34 1.70( 1.00) 91 1.50( 1.20) 330,122)
Length of program

involvement (months)

34 240007.00) 88 1900(1100)

4,700,120)

113

 

 

 

APPENDIX A
Table 4
a.
0‘- O'fﬁ 'n for .n.1..ﬂe' i i‘ '1-“ 1' ‘u' o f«_ o 0- “ u 0.
W
Group
Experimentalm = 55) Control(n = 54)
Strategy 2(%) M70) 4L,"
Presentation in
local community 16(29) 4( 7) 8.47""
Presentation for
state agencies 15(27) 5( 9) 5.84"
Newsletter 2( 3) 2( 3) .00
People ﬁrst grp 7( 13) 0( 0) 5.37“
Attended other
local chapters 3( 6) 0( 0) 1.33
Workshop for
providers 21(38) 6( 3) 16.73“"
Attended training 8(10) 4( 2) 5.7“
Trained DD to
self-monitor 5( 9) 1( 2) 2.71
Compiled a manual 7(13) 0( 0) 5.37"
Local coordinator
dyads 7(13) 0( 0) 8.75‘

 

‘p<.05;"o<.01;"‘o<.001

APPENDIX B

114

 

APPENDIX B
Table 1
I f n r iz 1f-
1. I am a very determined person.
2. Sometimes things just don’t seem worth the effort.

3. I would rather not try something that I’m not good at.

4. I can succeed in most any endeavor to which I set my mind.
5. Nothing is impossible if I really put my mind on it.

6. When I have difﬁculty getting what I want, I just try harder.

7. I have more will power than most people.

Based on a 7 point scale with 1 representing completely agree and 7 representing

completely disagree.

1 15
APPENDIX B

Table 2

MW

 

1. What is your overall impression of the citizen monitoring program as a whole
in terms of having an impact on group homes?

Based on a 5 point scale (1 represents very bad impression and 5 represents very

good impression).

2. My efforts as a volunteer monitor have made a difference in the group homes

that I monitor.

Based on a 5 point scale (1 represents strongly disagree and 7 represents strongly

agree).

3. Do you feel that the collective efforts of the state-wide volunteer monitoring

program can make a difference? (yes/ no).

4. Do you feel that you as a single volunteer eat; make a difference in the group

homes that you monitor? (yes/no)

116

 

APPENDIX B
Table 3
I f r ' i n- If- m
1. I count around here 2. I am taken seriously
3. I am important 4. I am trusted
5. There is faith in me 6. I can make a difference
7. I am valuable 8. I am helpful

9. I am cooperative

Based on a 5 point scale with 1 representing strongly agree and 5 representing

strongly disagree.

1 17
APPENDIX B

Table 4
I II n ' ' I!

 

1. Generally speaking, I am very satisﬁed with the monitoring program.

2. I feel a great sense of personal satisfaction when I monitor.

3. I frequently think of quitting the monitoring program.

4. My own feelings generally new affected much one way of the other by how well do my monitoring.

Based on a 7 point scale, with 1 representing strongly disagree and 7 representing strongy agree.

 

1. The amount of personal growth and development I get through monitoring.
2. The staff I talk to in the group homes.

3. The degree of respect and fair treatment I receive from my ARC supervisors.
4. ,Thedegreeofrespectandfairtreatmentlrecehrefromthegrouphomestaff.
5. The feeling of worthwhile accomplishment I get from monitoring.

6. The amount of support and guidance I receive from my ARC supervisors.

7. The amount of independent thought and action I am have in my monitoring activities.
8. The chance to help the residents of the group homes.

9. The amount of challenge in monitoring.

10. The overall quality of the supervision I receive on my volunteer work.

11. The training I received to become a monitor.

Based on a 7 point scale with 1 representing extremely dissatisﬁed and 7 representing extremely satisﬁed.

1 18
APPENDIX B

Table 5

I f ininx

 

We also want to know what aspects of your monitoring experience are
frustrating to you, which will help us decide how to make changes in the program.
Please place a check mark by those aspects of the monitoring experience that have

been frustrating to you. You may mark as many as you feel are appropriate.

1. _ The monitoring forms you use

2. _ Working only with the group home staff

3. _ Having little access to others/ agencies who can make change
happen in the group homes

4. _ Monthly visits to the group homes

S. __ The feedback you give to the group home staff

6. __ The feedback you receive from ARC/ M

7. _ The lack of contact with the overall state monitoring efforts

8. __ The local monitor meetings

9. _ The communication channels between you and the rest of
ARC/ M

 

10. Other (Please Specify)

1 19
APPENDIX B

Table 6

I fhr'inmmimn

 

1. I am willing to put in a great deal of effort beyond that normally expected in order to help this
program be successful.

2. I talk up the ARC local chapter to my friends as a great organization to volunteer for.

3. I feel very little loyalty to the local ARC.

4. I ﬁnd that my values and ARC ' s values are very similar.

5. I could just as well be volunteering for a different organization as long as the type of work was
similar.

6. This local ARC really inspires the very best in me.

7. It would take very little change in my present circumstances to cause me to leave ARC.
8. I am extremely glad that I chose ARC to volunteer for over other organizations.
9. There ' s not too much to be gained by sticking with ARC indeﬁnitely.

_10. I really care about the fate of ARC monitoring program.

_11. For me this is the best of all possible organizations for which to volunteer.

12. Deciding to volunteer for ARC was a deﬁnite mistake on my part.

The items are based on a 7 point rating scale with 1 representing strongly disagree and 7 representing

strongly agree.

LIST OF REFERENCES

LIST OF REFERENCES

Anderson, J., Moore, L. (1977). The motivation to volunteer. Jour_nal of Voluntary
Aetion Researeh, 1(3), 120-125.

Bandura, A. (1977). Self-efﬁcacy: Toward a unifying theory of behavior change.
Psyehological Review, 84, 191-215.

Bandura, A. (1982). Self-efﬁcacy mechanism in human agencymu
Psyehologistﬂ, 122-147.

Bandura, A. (1986). Social foundations of thought and action: W
theoty, Englewood Cliffs, New Jersey: Prentice Hall.

Conger, J., & Kanungo, R. (1988). The empowerment process: Integrating theoryd
practice. Agdemy of Management Review, _1_3(3), 471-482.

Cook. T4 & Campbell, D- (1979) WWW
for ﬁeld settings. Chicago: Rand McNally.

Dailey, R. (1986). Understanding organizational commitment for volunteers:

Empirical and managerial implications. MalofloluutaﬂAetjouﬁeoeatoh,
15(1), 19-31.

Gamm, L., & Kassab, C. (1983). Productivity assessment of volunteer programs in

not-for-proﬁt human services organizations. Joumal of Voluntat'y Agiou
Researeh, 1_2(3), 23-38.

Gidron, B. (1987). Integration of vounteer workers into formal human service

agencies: An organizational theory perspective. WM
5.9190922 11(2), 191-205-

Greller, M., & Herold, D. (1975). Sources of feedback: A preliminary investigation.
Qrgam’gtional Behavior and Humau Performanee, 13, 244-256.

Hackman, J ., & Lawler, E. (1971). Employee reactions to job characteristics. Journal
P h l , 55, 259-286.

120

 

. - _ . - No.
4). New Haven, CI‘: Department of Administrative Sciences, Yale University.

 

Hackman, J ., & Oldham, G. (1975 ). Development of the job diagnostic survey.
Journal of Applied Psyehology, 60, 159-170.

Houghland, J. Turner, I-I., & Hendricks, J. (1988). Rewards and impacts of

participation in a gerontology extension program. lout'paLoLYoluptagLAetton
Research, _Z(34) 19-35

Huck, S., Cormier, W., & Bounds, W. (1974). Reaﬂingstatistiosauiteseareh. New
York: Harper & Row, Publishers.

Ilgen, D., Fisher, C, & Taylor, (1979). Consequences of individual feedback on
behavior 1n organizations. WW ﬂ, 349-371.

llsley, P.. & Niemi, J. (1981). W New York:
McGraw-Hill.

Ivancevich, J ., & McMahon, J. (1977). Task-goal attributes, higher order need
strength and performance. WM 20(4), 552-563.

Ivancevich, J ., & McMahon, J. (1982). Effects of goal setting, external feedback, and
self-generated feedback on outcome variables: A ﬁeld experiment. AoademLof

Managementloumal, 25(2), 359-372.

Jenner, J. (1982). Participation, leadership and the role of volunteerism among

selected women volunteers. immune, 11(4), 27-38.

Kieffer, C. (1984). Citizen empowerment: A developmental perspective. Motion
in Humau Serviees, 3, 9-36.

Koch, J, & Steers, R. (1978). Job attachment, satisfaction, and turnover among
public sector employees. W1. 12, 119-128.

Latham, G. (1983). The role of goal setting in human resource management. In K.
Rowland, & G. Ferris (Eds. ), t- -._ ._ . __ ..
_auagerueut, vol. 1 (pp. 169-199). Connecticut: Jai Press Inc.

 

Latham, G., Mitchell, T., & Dossett, D. (1978). Importance of participative goal
setting and anticipated rewards on goal difﬁculty and job performance. Jout‘nal
1 li P h 1 03(2), 163-171.

122

Locke, E., Latham, G., & Erez, M. (1988). The determinants of goal commitment.
AcademnofManaasmenLBm .13 (1), 23-39-

Locke, E., Frederick, E., Lee, C., & Bobko, P. (1984). Effect of self-efﬁcacy, goals,

and task strategies on task performance. 101-WWW 62, 241-
251.

Mento, A., Steel, R., & Karren, R. (1987). A meta-analytic study of the effects of

goal setting in task performance, 1966-1984. WWW
Deet'eion Proeesses, 39, 52-83.

Morris, J ., & Sherman, J. (1981). Generalizability of an organizational commitment
model. Aoademy of Mauagement Journal 24(3), 512-526.

Mowday, R4 Porter L & Steers R (1982) Wallis

psyehology of eommitment, aoeeuteeiem auo tutuover. New York: Academic
Press.

Mowday, R, Steers, R., & Porter, L. (1979). The measurement of organizational
commitment WM 11. 224-247

Naylor, J., Pritchard, R4 8:11ng D- (1980). AW.

New York: Academic Press.

O’Connor P (1989 June) W

W. Paper presented at the Biennial Division
27 Conference, Community Action and Research, East Lansing, Michigan.

Oldham, G., & Hackman, J. (1980). Work design in the organizational context.
WW. 2 247-278

Pierce, J., Gardner, D., Cummings, L., & Dunham, R. (1989). Organization- based
self-esteem: Construct deﬁnition, measurement, and validation. Academy of
Managememlmal, 2(3). 622-648.

Porter, L. (1979). The measurement of organizational commitment. ,[oruuaipf
logatimialﬁehan'or, 1_4, 224-247.

Porter, L. & Lawler, E. (1968). W. Homewood,
Illinois: Dossey Press.

Porter, L., Steers, R., Mowday, R., & Boulian, P. (1974). Organizational
commitment, job satisfaction and turnover among psychiatric technicians. loumal

01002119010120.0023, 22, 603-609.

123

Rappaport, J. (1981). In praise of paradox: A social policy of empowerment over
prevention WWW 2(1) 1-25

Rappaport, J (1984) W
M1. New York: Haworth Press.

Rappaport, J. (1987). Terms of empowerment/ exemplars of prevention: Toward a

theory of community psychology Amen’canhuunalnfgommimitxlsycholm
_s(2), 121 147.

Roberts, K., & Glick, W. (1981). The job characteristics approach to task design: A
critical review MW 66 (2) 193-217

Rothman, J., Erlich, J -, & Teresa, J- (1981). W
mom. Beverly Hills: Sage.

Schorr, A. (1970). The tasks of volunteerism in the next decade. W, 51(8),
425-434.

Shavelson, R- (1988) WWW Boston: Allyn

and Bacon, Inc.

Scheffe', H. (1959) A_naJysis_of_var_ia.uoe, New York: Wiley Press.

Shore, L., & Martin, H. L. (1989). Job satisfaction and orgnizational commitment in
relation to work performance and turnover intentions. W, 1_(7),
625-638.

Steers, R. (1977). Antecedents and outcomes of organizational commitment.

Administrative Soienee magerly, 22, 46-56.

Steers, R., & Braunstein, D. (1976). A behaviorally-based measure of manifest needs
in work settings. Journal of Vooatt'onal Behaviot, 2, 251-266.

Steers, R., & Porter, L. (1974). Task-goal attributes. Psyehologiealﬂuﬂetiu, _8_1, 343-
452.

Tipton, R., & Worthington, E. (1984). The measurement of generalized self- efﬁcacy:
A study of construct validity. Willi, 518, 545-548.

Van de ven. A, & Ferry, P- (1980)Me_a§rmng_and__assessing_organizations.
New York:Wiley.

Vroom, V. (1964) W New York: Wiley-

124

Wood, R., & Bandura, A. (1989). Social cognitive theory of organizational
management. Academy of Management Review,_1_4(3), 361-384.

Zimmerman, M., & Rappaport, J. (1988). Citizen participation, perceived control,

and psychological empowerment. Am ri rn ' Ps h l
_1_6(5), 725-750. ’

 

"llllllillllllllf