 

.94?"

 

:L.

.9? . .1, .ﬁ
. {an

.iﬁnu us:
:7 “gin
, 3.x VI
again.
A. . .

2......” .5:
.huaeaﬁ La
.rﬁnuhwynurr .

2.11.11... \ z 7
1 1 t
"a.“ K???
Zlvuqmém .1.
«i . h
:3." .,
.uuv , misﬁt): .
«23.. «A... .
15.1%.. ,

:a :

“xi

.. i
111.0 K 3
£9.55?

., VI -" ‘h|’lf.
.2 15:15:»?
nil."

 

.. .
z s:

,2ng

 

. 1

nupnmwun.“
.

"a .ll
n».

 

QOOl

 

LIBRARY
Mlchlgan State
University

..___J

 

,m'

 

This is to certify that the

dissertation entitled

NONPROFIT VALUES IN MULTI-SECTOR EMPLOYMENT

presented by

Celeste Sturdevant Reed

has been accepted towards fulﬁllment
of the requirements for

Ph.D. . Social Science
* * degree in

 

with Concentration
in Human Resources and Industrial Relations

WJJ 0% Wang

Major professorL

April 17, 2001
Date _

MS U i: an Afﬁrmative Action/Equal Opportunity Institution 0- 12771

 

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8/01 c-JCIRCIDateDue.p65-p.15

NONPROFIT VALUES 1N MULTI-SECTOR EMPLOYMENT
By

Celeste Sturdevant Reed

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
School of Labor and Industrial Relations

2001

 

ABSTRACT
NONPROFIT VALUES IN MULTI-SECTOR EMPLOYMENT
By

Celeste Sturdevant Reed

It is widely recognized in both the popular and the scholarly literature that
employees’ values contribute to a variety of work-related attitudes, actions and outcomes.
This study extends the investigation of the relationship between individuals’ values and
aspects of their work setting by comparing individuals’ values and work practices across
three sectors, nonproﬁt, public and for-proﬁt. This research was guided by a descriptive
study that proposed sector, industry and organizational inﬂuences in addition to employee
values as explanatory variables accounting for nonproﬁt organizations’ practices (Young,
1984)

The challenge in cross-sector research is to collect enough data in each sector to
test key relationships. The solution used here was to identify a set of individuals who
were then asked to provide self and work-related data, rather than identifying
organizations from whose employees’ self and organizational practices data could be
collected. The advantages and disadvantages of such an approach are discussed.

Three streams of literature were used to develop a model that integrated work and
individual context variables, individual values and work practices as predictors of
individuals’ organizational commitment, job involvement, intent to leave their
Organization and tenure. The contributions made by nonproﬁt literature, institutional

literature and organizational behavior literature are identiﬁed.

 

Mixed results were found. In contrast to hypotheses, sector did not predict the
work values altruism or autonomy, the availability of professional rewards or the use of
professional judgments in making client service decisions. Similarly, ﬁnancial rewards
were most available in the public sector and least available in the for-proﬁt sector.
Generally the interactions between work values and the work practices that mirrored
those values were not significant predictors of organizational commitment, job
involvement or intent to leave. The availability of professional rewards and the
opportunity to use professional judgments were associated with organizational
commitment while reliance on legal requirements and the lack of professional rewards
predicted intent to leave. The impact of the work value altruism was predicted: it was
positively associated with organizational commitment and job involvement and
negatively associated with intent to leave. The ﬁndings are more fully discussed and

recommendations for further research and practice are offered.

Cepyright by
CELESTE STURDEVANT REED
2001

DEDICATION
In loving memory of Phil
And
To my family and friends

who were continuously supportive over this long journey
(“Aren’t you done yet?”)

ACKNOWLEDGMENTS

My heartfelt thanks to the continuing members of my Committee: Mike Moore,
Ed Montemayor, Kevin Ford and John Hollenbeck. To say that this has been a long
process is an understatement. Individually you have stepped up at critical moments to
keep this dissertation alive long enough for me to complete it. Your collective willingness
to provide support, encouragement, and plan for yet another meeting made success

possible.

vi

 

 

 

 

Git?
Nita.

flit?
LHER

CH;
ME“.

 

TABLE OF CONTENTS

CHAPTER 1
INTRODUCTION .......................................................................................................... 1
Importance of the Topic .......................................................................................... 1
Introduction ................................................................................... 1
An Extension ............................................................................................... 2
The Research Approach .......................................................................................... 5
Sector .......................................................................................................... 5
Industry ................... ‘ .................................................................................... 5
Type of Organization .................................................................................. 6
Type of Person ............................................................................................ 6
Key Assumptions and Limitations .......................................................................... 7
Contributions .......................................................................................................... 8
Outline of this Dissertation ................................................................................... 10
CHAPTER 2
LITERATURE REVIEW ................................................................................................... 11
Relationship A: Individual Values and Work Practices Predict Dependent ‘
Variables ................................................................................................... 12
The Relationship between People and Work Settings .............................. 12
Individual Values ...................................................................................... 17
Work Structures ........................................................................................ 19
Summary ................................................................................................... 20
Relationship B: Person Context Predicting Individual Values ............................. 21
Sector Inﬂuences on Work Values ............................................................ 21
Relationship C: Organization Context Inﬂuences on Work Practices .................. 31
Industry ..................................................................................................... 31
Work Practices .......................... 39
Summary ........................................................................................................ 42
CHAPTER 3
METHOD ........................................................................................................ 46
Study Sample ........................................................................................................ 46
Measures ........................................................................................................ 54
Relationship A .......................................................................................... 54
Relationship B ........................................................................................... 61
Relationship C ........................................................................................... 62
Data Analysis Procedures ..................................................................................... 63
CHAPTER 4
RESULTS ........................................................................................................ 65
Introduction ........................................................................................................ 65

vii

Relationship C - Work Context Predicts Work Practices ..................................... 65

Sector Inﬂuences on Reward Practices ..................................................... 66
Sector Inﬂuences on Client Service Decision Practices ........................... 68
Industry Inﬂuences on Work Practices ..................................................... 69
Relationship B - Person Context Predicts Individual Values ............................... 70
Relationship A - Individual Values and Work Practices as Predictors of the
Dependent Variables ................................................................................. 74
Values and Work Practices as Predictors of Organizational
Commitment ................................................................................. 75
Job Involvement ........................................................................................ 79
Intent to Leave ........................................................................................... 79
Tenure ........................................................................................................ 80
Testing the Model ................................................................................................. 81
Work Context and Work Practices ........................................................... 81
Individual Context and Work Values ....................................................... 85
Summary ........................................................................................................ 89
Hypothesis Testing .................................................................................... 89
Model Testing ........................................................................................... 94
CHAPTER 5
DISCUSSION ........................................................................................................ 98
Introduction ........................................................................................................ 98
Summary of Findings ............................................................................................ 98
Implications for Research and Theory ................................................................ 101
Organizational Behavior Theory ............................................................. 101
Nonproﬁt Theory .................................................................................... 105
Institutional Theory ................................................................................. 109
Implications for Practice ..................................................................................... 111
Suggestions for Future Research ......................................................................... 113
APPENDD( A
INSTRUMENTS ...................................................................................................... 119
APPENDIX B
SCALES AND MEASURES .......................................................................................... 137
APPENDD( C
FACTOR ANALYSIS RESULTS FOR UNUSED SCALES ........................................ 143
APPENDDC D
NULL RESULTS ...................................................................................................... 145
BIBLIOGRAPHY ...................................................................................................... 148

viii

 

TABL

Cms-

TABU

Rapx

 

TABL
lkus

Sail

TABLE

TABLE
RSPOZ;

TABLE

Rams

TABLE

Uﬁna

TABLI

Rﬁgfeg

TABLI

a ‘73»
RPS; t3

TABLI
Mean l

TABL]

R53";
‘3," ‘5

TABL]
Rfﬂm

DIV

TABu
kgn:

LIST OF TABLES

TABLE 1

Cross-Sector Comparisons of Employee Values .............................................................. 24
TABLE 2

Respondents’ Characteristics ............................................................................................. 49
TABLE 3

Means, Standard Deviations, Correlations and Scale Reliabilities for the

Study Variables ................................................................................................................. 51
TABLE 4

Results of the Factor Analysis of Client Service Decision Making Practice Items .......... 59
TABLE 5

Reported Frequency of Agency Rewards .......................................................................... 60
TABLE 6

Results of Factor Analysis for Work Reward Items ......................................................... 61
TABLE 7

Distribution of the Respondents by Sector and Industry ................................................... 62
TABLE 8

Regression Results Predicting the Availability of Financial Rewards From Sector ......... 66
TABLE 9

Regression Results Predicting the Availability of Professional Rewards From Sector ....67
TABLE 10

Mean Values for Altruism, Autonomy and Economic Rewards by Sector ...................... 71
TABLE 11

Regression Results Predicting the Work Value Autonomy from Sector .......................... 72
TABLE 12

Regression Results Predicting the Work Value Economic Rewards from Sector ............ 72
TABLE 13

Regression Results Predicting Sector from Work Value .................................................. 73

ix

 

TABL
Regs-:5
lam.

TABL
Cont.

TAB"
31’s?"
ELL:

5

bet“ 9;

Raw.

    

TABLE

Ram-

\vaB.
V

Values.

TABLE
R5355

Tait-es

 

TABLE
Rt‘gess

Work P

TABLE
Regen

Comer

TABLE

3 an.
Rims:

TABLE
Rﬁgress
Comer

TABLE
Reg-Tess

COHICXI

TABLE
R‘C‘EJESS
COntex:

TABLE

RE'WQ»-

a”):

TABLE 14
Regression Results for Organizational Commitment Predicted from the Interactions

between Work Values and Work Practices ....................................................................... 76
TABLE 15
Correlations Among the Regression Variables ................................................................. 77
TABLE 16

Regression Results for Organizational Commitment Predicted from the Interactions
between Work Values and Work Practices (Excluding Altruism X Professional

Rewards ............................................................................................................................. 78
TABLE 17

Regression Results for Job Involvement Predicted from the Interactions between Work
Values and Work Practices ................................................................................................ 79
TABLE 18

Regression Results for Intent to Leave Predicted ﬁ'om the Interactions between Work
Values and Work Practices ................................................................................................ 80
TABLE 19

Regression Results Predicting Organizational Commitment from Work Context through
Work Practices ................................................................................................................... 82
TABLE 20

Regression Results Predicting Intent to Leave from Work Practices and Work

Context .............................................................................................................................. 83
TABLE 21

Regression Results Predicting Tenure from Work Practices and Work Context ............. 84
TABLE 22

Regression Results Predicting Organizational Commitment from Work Values and Work
Context ............................................................................................................................... 86
TABLE 23

Regression Results Predicting Job Involvement from Work Values and Work

Context ............................................................................................................................... 87
TABLE 24

Regression Results Predicting Intent to Leave from Work Values and Work

Context ............................................................................................................................... 88
TABLE 25

Regression Results Predicting Tenure from Work Values and Work Context .................. 89

 

 

TEXT.

 

BLE
oT

:I‘JYQI
MUIL,

 

 

TABLE 26
Predicted vs. Actual Sector-Work Practice Relationships ................................................. 91

TABLE 27
Predicted vs. Actual Context-Work Values Relationships 93

TABLE 28
Direction of the Fit Interactions Predicting Organizational
Commitment, Job Involvement, Intent to Leave and Tenure ............................................. 93

TABLE 29
Direction of Work Practice and Context Variables Predicting Organizational
Commitment, Job Involvement, Intent to Leave and Tenure ........................................ p ..... 95

TABLE 30
Direction of the Work Values and Individual Context Variables Predicting
Organizational Commitment, Job Involvement, Intent to Leave and Tenure .................... 97

TABLE 31
Correlations Among the Main Effects and Their Interactions ......................................... 104

TABLE 32
Differences in the Use of Government Policies by Industry ............................................ 110

xi

LIST OF FIGURES

FIGURE 1

The Relationships Among the Dissertation Variables ........................................................ 4
FIGURE 2

Relationship A: Predicting the Dependent Variables from Work Practices and

Individual Values ............................................................................................................... 13
FIGURE 3

Relationship B: Person Context Predicting Individual Values ......................................... 22
FIGURE 4

Relationship C: Work Context Predicting Work Practices ................................................ 32
FIGURE 5

Sources for Model Development ....................................................................................... 43

xii

CHAPTER 1
INTRODUCTION

This chapter provides an introduction to and overview of the dissertation study.
Five topics are discussed: the importance of a more complete understanding of nonproﬁt
employees’ values; the research approach; the key assumptions and limitations inherent in
this research; the contributions of this study; and an outline of the subsequent chapters in

the dissertation.

Importance of the Topic

lamina

It is widely recognized in both the popular and the scholarly literature that
employees’ values contribute to a variety of work-related attitudes, actions and outcomes.
Because values are stable dispositional traits, there has been continued interest in linking
these affective evaluations to motives and performance (Dawis, 1991). Work values can
inﬂuence such diverse factors as an individual’s choice of employer (Judge & Bretz,
1992), organizational adaptation (Chatman, 1989) or employees’ job satisfaction
(Meglino, Ravlin & Adkins, 1989).

While there is evidence of the applicability of general values—such as
achievement, helping and concern for others, fairness and honesty—to many
organizations (Cornelius, et al., 1985), nonetheless researchers persist in their attempts to
identify meaningful organization-related differences that can be attached to individuals’
Speciﬁc values. Knowledge of organizational values as signaled by compensation systems

has been shown to inﬂuence job seekers’ choices (Judge & Bretz, 1992). In one of the

 

 

 

 

  
   

mos: ;
found
on m.
diffs:

ll.t.._;.:

 

 

om"

v: i ,
Kin.
5‘.

llai’ni

most comprehensive studies of organization-person value congruence, Chatman (1991)
found that organizational values proﬁles could be generated for accounting ﬁrms based
on managers’ descriptions and that these unique proﬁles could be associated with
differences in ﬁrms’ practices (socialization activities) as well as employees’ attitudes
(i.e., job satisfaction) and actions over time (i.e., turnover).

The pursuit of some level of ﬁt between individuals’ values and their work
settings is hardly trivial. Where there is values congruence between employees and their
organizations, studies have shown payoff for organizations in improved job performance,
stronger organizational commitment, and reduced turnover (Meglino et al., 1989).
Further, it has been suggested that better alignment between people and their
organizations, as opposed to people and speciﬁc jobs, is necessary to create the workforce

ﬂexibility needed by employers (Bowen, 1991).

An Extension

This study extends investigation of the relationship between individuals’ values
and aspects of their work settings. The studies cited above focused exclusively on
employment experiences in a single sector. This study compares individuals’ values and
organizations’ practices across three sectors—the public, not-for-proﬁt and for-proﬁt
sectors. This research is guided by a descriptive study that proposed sector, industry, and
organizational inﬂuences as well as employee values as explanatory variables accounting
for nonproﬁt organizations’ practices (Young, 1984). Augmenting document reviews

With interviews of nonproﬁt leaders, Young developed theories about the nonproﬁt sector

that accounted for the relationships among sector, industry, organization and person

 

attributes described. Chief among the explanatory factors was the role that a speciﬁc
value orientation had on sector, industry, organization and individual practices. No
subsequent study has been found that attempts to measure the inﬂuence of the variables
he described.

The inﬂuence of employees’ values is discussed extensively in the nonproﬁt
literature (Drucker, 1974; O’Connell, 1988; O’Neill & Young, 1988) but rarely tested. As
will be seen in Chapter 2, even those researchers who describe sectoral value differences
do so without explicitly testing those values. The dearth of information about the
relationship between employee values and work practices is remarkable, given the
assumptions made that nonproﬁt employees are best able to act on their values in
nonproﬁt organizations (Witty, 1989). In this study I have tested nonproﬁt assumptions
by developing a model of interaction among key variables. The model itself, shown in
Figure 1, is derived from relationships among variables tested primarily in the for-proﬁt
sector.

The relationship among variables in Figure 1 is as follows. Two organization
context variables, sector and industry, inﬂuence an organization’s work practices.
Similarly, two person context variables, sector and position, inﬂuence an individual’s
work values. An individual’s work values interact with the work practices to inﬂuence a
set of individual variables, speciﬁcally one’s job involvement, organizational
commitment, tenure in the organization, and intent to remain with that organizations.

These relationships are described in more detail in Chapter 2.

 

 

J1... v Hire—ZVFJLLTIT...
EEC-93%. Savaa‘UovA- 0.9???an sﬁrv-.V

 

 

outﬂow 625 downer—0mg

Baggage. ”N u 8525
58982 .235 waxy—Atom mm H 330m #96365

“5:80

 

 

 

m

 

$2.3
3:339:

gasped own—080m
$8883.. BEBE mm u 825»

 

 

 

 

055%
Suﬁsm 3 E33

EQBEEoU Ecoumﬁswwco . 838$ Allll i850

«880335 93. x53 0 #83

"833:; Bog—omen <

 

 

 

 

 

 

 

 

 

 

.28: 3:02 63:0? BED
.oorcom 333m mm H .0536:—

E8952
.oznzm JmoELom mm u 38%

core—om 25:0 .Ecommmomem .omanoom
mm .1. 3.3.50“ o_na__a><

aim Beameoésoz 626

.600 .moﬁsmuomozom

mm H 9.33% .53qu cerium «5:0

 

Figure 1. The Relationships Among the Dissertation Variables

 

 

 

 

 

 

 

Battle.
0T0:

chillz

The Research Approach

The research approach used here was to collect data using a large cross-sectional
exploratory study. The challenge was to collect enough data across three sectors to test
key relationships in the model. In order to do so, I have used the approach of identifying
individuals and following them back to their organization rather than identifying
organizations and moving down to the employees. The beneﬁts and liabilities of such an
approach are described at the end of this chapter.

Based on Young’s descriptive study of nonproﬁt organizations, a series of
deﬁnitions and a priori decisions were made to place the following boundaries around the

key variables.

m
As used in this study, sector conforms to the traditional three economic Sectors:
public, nonproﬁt, and for-proﬁt. Some nonproﬁt theoreticians (Smith, 1991) have
suggested that subdivisions could create a more homogeneous sector, for example, by
separating out member beneﬁt organizations. The three-sector model is appropriate for
differences being tested here. I have excluded the member beneﬁts nonproﬁt

organizations from this study.

Indusm

Young (1984) accounted for differences be found in reward structures and
practices by looking at the important relationships nonproﬁt organizations had with

organizations in other sectors. Three industries commonly found in the nonproﬁt sector

have bee,”

indusmes

   

\Tlsrns;

ms. Bart.

CalleT {65.

 

31530 {mi

mirror" (\‘

 

ageﬁfles. j

mmmkkrﬂ“

I." .L}
in OMLT‘S

0?"
I B‘s-\i'ﬂ‘

IL!

 

have been selected, child welfare, family services and mental health. These three
industries provide a contrast among the other-sector relationships of their organizations
(Thompson, 1967). Family service agencies are traditional nonproﬁts shaped most by the
needs and preferences in their communities; clients and donors, not other-sector
organizations, provide the most inﬂuence on their services (Alperin, 1993). The nonproﬁt
organizations being shaped most by their public sector partners are child welfare agencies
(Kramer, 1994) and mental health agencies are increasingly being shaped by market

forces (Winegar, 1993).

Type of Organization

One of the challenges in comparing and contrasting nonproﬁt organizations is to
select a subset of organizations that are sufﬁciently alike to allow for comparison. It has
been suggested that “the differences among not-for—proﬁt organizations—Tor example, a
cancer research institute versus a Harlem street academy versus a cemetery association——
are so great that their common negative characteristic of not seeking a proﬁt becomes
minor” (Newman & Wallender, 1978, p. 26). This study focuses on human service
agencies, those local agencies that provide services aimed at meeting the social needs of a

community (McCauley & Hughes, 1991).

Type of Person
This was a study of employee-employer relationships, so it excluded volunteers
and others without a regular employment attachment. The respondents were all members

of the Michigan Chapter of the National Association of Social Workers. They represent a

 

 

. W
sngle t. .

l

111005; 1T1.

 

me ‘61. T.
outcomes

pram,

‘ “.
1"':|:m,~:.\
““‘\IIL‘

OUTCOTTTC'S

din "JV C(

 

neCCSSaQ‘ l

dim sem

single group with similar training and professional socialization who have chosen to work

in different sectors and/or industries.

Key Assumptions and Limitations

A number of key assumptions are being tested in this dissertation. First, the study
model includes relationships that have been tested in part, but never within an entire
model. For example, work values have been shown to be associated with organizational
outcomes, but speciﬁc contextual variables such as sector, industry or position have not
previously been combined as explanatory variables. A key assumption is that sector
inﬂuences can be traced through organizational practices and individual values to speciﬁc
outcomes. Second, this study represents the ﬁrst-time use of a work values inventory to
directly collect work values information from individuals in the nonproﬁt sector.
Standardized approaches have been used in the for-proﬁt sector. However, these
approaches did not include the theoretically necessary items nor were they usable in a
cross-sectional research. An instrument with theoretically appropriate items was selected
and tested in this study. The second assumption made is that distinctions among
individuals’ values can be made using this measuring approach. Third, these employing
organizations—human service organizations—represent an alternative form not usually
studied. In order to study organizational practices within these organizations, it was
necessary to develop a scale to assess their focal operational practice, that of making
client service decisions. The items in the scale are based on a review of the literature, but
are untested as a measurement device. A third key assumption made is that such practices

can be measured and compared across organizational types. Finally, a fourth assumption

 

made is 2{

Tamil;

med. Tr

 

data to a
Orgsﬁil’.
there is .
unlikely
Organs;
ditTercr:

locale Ie

 

B16 CllC
Because.
irate nu
sellings
ofhuma
health 5:

Chmge.

The W

made is that focusing on employees in a single profession would not draw the sample too
narrowly with respect to important individual attributes.

The limitations of the study are several, and parallel the method and model being
used. First, this is a self-report study. There was no feasible way to collect independent
data to verify the responses or to compare individuals’ responses with desired
organizational outcomes. The latter is a more onerous omission than the former since
there is some evidence that the mean self-report values for variables in this study are
unlikely to be inﬂated (Crampton & Wagner, 1994). Second, this could not be a study of
organizations since the mailing list was composed of individuals, most likely to be in
different organizations. Therefore it was only possible to compare work settings, not to
build proﬁles of speciﬁc organizations. Third, while every attempt has been made to
locate tested scales and measures, for some variables of interest that was not possible.
The client service decision factor scale, for example, had to be developed for this study.
Because of some unique aspects of human service organizations, setting-speciﬁc scales
have not been subjected to the same level of extensive use that scales appropriate for all
settings (organizational commitment, for example) have experienced. Finally, the world
of human service organizations in Michigan is particularly volatile. Especially the mental
health system, but also the child welfare system is undergoing constant and progressive

change. Some of the ﬁndings here may be temporal, or at least evolutionary, in nature.

Contributions
This study contributes to current research on organizational behavior and also to

the knowledge about nonproﬁt employees and their work experiences. There are several

 

COTLIEY...

BOLTON I:

BL‘BPTC :‘l'
as liilE»
nonproi
research

lBFO‘Am

 

been a d;

character

 

speciﬁc areas of contribution to the organizational behavior literature. First, this research
collects a number of variables about the employment relationship within an interactionist
perspective, linking attributes of employees with attributes of their work settings (Taylor
& Giannantonio, 1993). Second, the use of a work values variable adds to the growing
literature that acknowledges this as the preferred method of measuring individual and
organizational attributes (Chatman, 1989). Finally, the collection and analysis of
contextual, organizational and individual variables provides additional support for the
notion of congruence among these factors.

Speciﬁc contributions are also made to the nonproﬁt literature. For the ﬁrst time,
nonproﬁt employees’ values have been measured with a speciﬁc work values scale, and,
as will be seen in Chapters 4 and 5, the empirical evidence challenges the predicted
nonproﬁt sector-work values relationship. This study also responds to the call for more
research on organizational factors and employment practices in the nonproﬁt sector
(Brown, 1986). In addition, not since Young’s (1984) ground-breaking work has there
been a description of the interplay among sector, industry, work setting and individual
characteristics. Much of the descriptive literature that focuses on individuals’ values does
not adequately reﬂect the current challenges facing industries in the nonproﬁt sector.
Finally, practitioners may beneﬁt from the “view from the ﬁeld” of how client service
decisions are viewed. The balancing of context, organization, staff and client issues

should be informative.

t t
3.1 1Dr I -
LITUH.‘

soliy's . ,

olthe IT.
Elinor:
outcome.
Organiza'
to each r;

fillet;

 

 

Coltude .

Outline of this Dissertation

This dissertation contains ﬁve chapters. Chapter 1 has described the rationale
underlying the research, provided an overview of the model being tested, outlined the
study’s limitations, and suggested the potential contributions.

Chapter 2 presents the model in detail, providing the literature that supports each
of the model components. The twenty-one variables are organized into three
relationships, those organizational practices and individual values that predict individual
outcomes, the person context variables that predict individual values, and the
organization context variables that predict organization practices. The hypotheses related
to each relationship are presented at the conclusion of each section of the literature
review.

Chapter 3 discusses the method of investigation. This includes a description of the
sample, the data collection procedure, the operationalization of variables, the measures
and the method of data analysis for each hypothesis.

Chapter 4 provides the results of the data analysis by component for each of the
hypotheses. Descriptive statistics have also been included.

Chapter 5 contains a summary of the ﬁndings. Alternative explanations are
presented as are implications for practice. Finally, suggestions for future research

conclude this chapter.

10

V
9 ’3'".
NL‘\\'\\.

' >3 T ,) 7‘
reprise .

certain .‘

tenure. S

 

 

 

 

COBlpOn
model. i
indent
addition
moo. id;
Practices

Control

CHAPTER 2
LITERATURE REVIEW

Chapter 1 introduced a model that has been used to examine the relationships
between employed individuals and their organizations. Figure 1 illustrated that variables
representing peoples’ values and contexts, when combined with speciﬁc work practices in
certain contexts, are expected to have an impact on individuals’ work attitudes and
tenure. Several streams of literature have been used to build this model:

0 The organization behavior literature on individual-organization relationships;

0 The nonproﬁt literature on individual values and organizational practices;

0 The professional-organization relationship literature;

0 Organizational theory for the impact of context and environment on organizations
and their practices; and

o The institutional literature related to the three sectors and three industries that are
highlighted in this study.

In this chapter, the literature with accompanying hypotheses is presented for each
component of the theoretical model. The review begins with the ﬁnal component of the
model, the relationship between individual values and work practices that predict target
individual attitudes and behaviors. That interaction is supplemented with the two
additional components, those contextual variables that are predicted to inﬂuence
individuals’ values and those contextual variables that are predicted to inﬂuence work
practices. The chapter concludes with a summary of the three key authors whose work

contributed to the model’s development and a review of the hypotheses tested.

11

 

 

no: it.

ones or.

marble:

in ,-
“lh riTSl

 

piﬁﬁilii’l

Vales a

concept
0136?] de-
K71 Siljfj

Biff CQm

Relationship A:
Individual Values and Work Practices Predict Dependent Variables

Figure 2 suggests that individual values interact with work practices to create the
individual outcomes of organizational commitment, job satisfaction, intent to remain with
one’s organization and tenure in one’s organization.

In this section literature is presented to support this component of the model, the
variables selected and the method used to operationalize the model. In so doing, literature
will ﬁrst be presented on the nature of the interaction and the potential positive outcomes
of that interaction. Then a discussion of each of the sets of independent variables,
individual values and work practices, will be presented. Additional literature will be
presented in each of the following two sections where work practices and individual

values are operationalized as dependent variables in those segments of the model.
The Relationshgr between Peopleﬁ and Work Settings

Person-Organization Fit

The focus of this study is to better understand the relationship between speciﬁc
people and their perceived work settings. The model proposes a series of antecedents that
result in desired individual outcomes. The interaction modeled in Figure 2 can be
conceptualized as the construct person-organization ﬁt. Person-organization ﬁt is most
often deﬁned as the compatibility between individuals and organizations (Kristoff, 1996).
Kristoff reﬁnes this deﬁnition by suggesting that “person-organization ﬁt is deﬁned by
the compatibility between people and organizations that occurs when: (a) at least one
entity provides what the other needs, or (b) they share similar fundamental characteristics,

or (c) both” (pp 4-5). The attributes that have been investigated include

12

3330M omﬁocoom
558:2 .Emmsbz mm H 82.5

 

 

83:;
126365

 

 

 

 

955,—.
595% 8 E85
EoEoZgE n8.

EoBMEEoU Ecoumﬁuawco

”mo_nmtw> “newcomoa

 

 

 

 

moouoﬁm
x83

 

 

.0238 Bozo £8882“ £8858
mm n 3333— 03.555.

asp—m Ecommmomoaﬁoc Home

.38 .mBEBBEa

mm M 9:53.. .8331 «3:8 «5:0

Figure 2. Relationship A: Predicting the Dependent Variables from Work Practices and

Individual Values

13

01233.1(.
5'

person:

oppor.

 

motel. ':

[~J
I

E; J
r1

 

 

 

(I?

Operatio

measly:-

organization/person characteristics, including values, goals, culture (organization) or
personality (person); organization/person supplies or demands including resources,
opportunities and sets of knowledge, skills and abilities. Fit has been operationalized as a
match between individuals and organizations in four general ways:
1. Identifying the congruence between individual and organization values (a
supplementary perspective);
2. Matching the goal congruence between individuals and organization leaders (a
supplementary perspective);
3. Deﬁning ﬁt as the match between individual preferences and organizational
systems or structures (a needs-supplies perspective); and
4. The match between the characteristics of individual personality and organizational
climate (interpretable as either a supplementary or complementary perspective).
There have been differences in the results observed based on the
Operationalization of person-organization ﬁt. Researchers who have focused on the
measurement of individual personality or preferences and organizational structures have
found mixed results. Bretz and his colleagues (1989) did not ﬁnd a discernable
relationship between individuals’ internal needs states and a set of individually and
organizationally oriented system characteristics. However, when ﬁt has been
conceptualized as individual and organizational goal congruence (Schneider, 1987),
greater goal congruence, both within work group and across supervisory level, has been
associated with desirable work attitudes (Vancouver & Schmitt, 1991).
The most frequently used and strongly endorsed operationalization of ﬁt is one of

individual and organizational work value congruence (Boxx, Odom & Dunn, 1991;

14

 

.5132;

   
   

ls l: p:
and it.
LITTER;
aspects
iL‘ClL'Cc‘
are pro.
"based or.

Ni,» .‘ ‘
“Its-BL“: ll
~ .

 

‘EhV

Giselle

“liars:

Chatrnan, 1989, 1991; Judge & Bretz, 1992; Posner, 1992). Proponents suggest that this
is the preferred method for assessing congruence because values represent fundamental
and enduring characteristics of both people and organizations (Chatman, 1989). The
importance of individual-organizational values congruence has been demonstrated in all
aspects of the employment relationship (Taylor & Giannontonio, 1993). Researchers have
documented the strength of individuals’ value-based decision making. Where individuals
are provided enough information about an organization’s values, they will choose jobs
based on those values (Cable & Judge, 1994). There is also evidence that individuals will
change jobs in order to achieve a better ﬁt with their values (Chao, O’Leary-Kelly, Wolf,
Klein & Garner, 1994). In the most comprehensive of the employment relationship
studies, Chatrnan (1991) demonstrated that similarity of employee-organization work
values could explain selection choices, the ease of employees’ socialization into an
organization, and retention. Moreover, she was able to rule out possible alternative
explanations for the employee-organization compatibility found, i.e., that individuals with
speciﬁc value proﬁles ﬁt better in any organization, or conversely, that organizations with
particular work value proﬁles accommodated all types of individuals’ values better. It
was the alignment of both sets of values that predicted individual satisfaction, intent to
leave, and actual departure.

It is possible to measure the extent of ﬁt without collecting comparable individual
and organizational data. Direct measures of ﬁt, taken by asking people whether they
believe a good ﬁt exists, have been shown to produce results similar to those found with

comparable comparison data. Higher organizational commitment was reported by

15

mom.

who re:

L
a.

a ,:
5'55”};
mu.»‘

.1

19911. :E

 

l'lt’ll ed- a

TCCflll It

Bill“ ldu

“113‘“ 3:).
it“; VI

individuals who rated their values to be compatible with their organizations than by those

who reported less congruence (Chao, et al., 1994; Posner & Schmidt, 1996).

Consequences of Fit

Prior research has conﬁrmed both affective and behavioral consequences of ﬁt.
Studies done with such diverse populations as school personnel (Vancouver & Schmitt,
1991), three levels of staff in a production organization (Meglino et al., 1989) and
professional accountants (Chatman, 1991) found that the level of goal or value
congruence predicted job satisfaction and organizational commitment. Job involvement
has not been studied as a product of organizational ﬁt. However, it has frequently been
viewed as the product of work values (Kunungo, 1982; Lodahl & Kejner, 1965) and
recent research found that job involvement is stable job attitude (Steel & Rentsch, 1997).

Both intent to leave and actual departure have also been predicted by alignment of
individuals’ and organizations’ values (Chatman, 1991). In a study of professional
engineers, Chao and her colleagues (Chao, et al., 1994) found that people who never
changed jobs in the ﬁve years of the study had a signiﬁcantly better ﬁt between their own
and their organization’s values than those people who changed jobs within their
organizations as well as those who changed organizations. Among job changers, better ﬁt
(higher scores) was found after individuals made job changes.

The challenge is this study was to operationalize theoretically supportable
characteristics of ﬁt given the limitations of this study. Finding commensurate ways to
measure target organization and person attributes was not possible because no

independent organization data were collected. Similarly, verifying individuals’ agreement

16

lllOTGfT

 

 

- ' l.-
Alert:

 

 

”'3 “Q ~.
Bur “k ‘1‘

in order to ascribe an attribute to the organization (James, 1982) was not possible because
these respondents were anonymous individuals across many organizations. In the
discussions that follow, I outline the approach used here to identify necessary parallel

and/or related individual work values and work setting characteristics.

Individual Values

I followed the strong recommendation of prior researchers and chose work values
as the key individual variable for assessing ﬁt. Work values have been deﬁned as the end
state people desire and feel they ought to be able to realize through working (Nord, Brief,
Atieh, & Doherty, 1988). The adoption of a work values approach is particularly
appropriate for this study, where I am trying to capture the unique aspects of nonproﬁt
sector employees’ values. The nonproﬁt literature suggests that the attributes upon which
nonproﬁt employees are most like each other—and most different ﬁ'om employees
attracted to the other sectors—is in their work values (Witty, 1989).

A variety of approaches have been used to assess individual and organizational
work values. Some researchers investigating ﬁt have used the organization’s own set of
values (Posner, 1992). Others have used sets of values drawn from universal values such
as the Comparative Emphasis Scale used by Ravlin & Meglino (1987) or the
Organizational Culture Proﬁle used by O’Reilly and colleagues (O’Reilly, Chatrnan, &
Caldwell, 1991). Neither of these approaches were deemed appropriate for this study.
First, given the diversity and anonymity of organizations, no single set of organizational
values could be offered. More importantly, although their assumptions have not been

empirically tested, nonproﬁt authors persist in asserting that values are different in this

17

senor :1

 

roe?

 

 

 

exzss; to.
founders
commit:

3'1 YounE

IS 31ko
5€CIOI I'a'
WSiilw \
Scineshc

all“! boll

 

sector than in the other sectors, so other formulations of universal values did not seem
appropriate.

Nonproﬁt authors agree that the nonproﬁt sector is unique because the
organizational raison d’etre—its mission, which incorporates it accountability to the
public trust and lack of proﬁt motive—is the key factor that sets these organizations apart
(Hodgkin, 1991; J eavons, 1992; Ott, 1991). The mission or values across the three sectors
differ. In the for-proﬁt sector the objective is to make a proﬁt (J eavons, 1992).
Organizations in the public sector are mandated to carry out programs deﬁned by elected
ofﬁcials on behalf of their constituents. In the nonproﬁt sector, no such targeted objective
exists; their work revolves around the philosophical, moral or religious values of their
founders and supporters. The general values ascribed to the nonproﬁt sector include a
commitment beyond one’s self (O’Connell, 1988), independence and autonomy (O’Neill
& Young, 1988).

Commitment beyond oneself, alternatively named as altruism or service to others,
is thought to be a unique nonproﬁt value. In contrast, autonomy is recognized as a cross-
sector value. Professionals, who are the target respondents in this study, are known for the
positive value they place on autonomy (Bartol, 1979; Hall, 1985; Kerr, Von Glinow &
Schriesheim, 1972).l Autonomy has been deﬁned “a perceived right to make decisions
about both the means and goals associated with one’s work” (Bartol, 1979, p.817). The

salience of autonomy for professionals is supported by international research cited above

_*

|The list of multidimensional professional values includes: professional autonomy, commitment to the
Profession, identiﬁcation with the profession, professional ethics, and belief in collegial maintenance of
Standards (Bartol, 1979; Kerr, et al., 1977).

18

(MOW ‘

and prt

was a s-
organiz-
organiz.
lidl\'ld-.
lndlx :d..
01.9,“; , ..

4
\KHL“
~

$130me
m Ores-".2.

55mm

SIT—iclure‘s

 

 

 

‘Rlnes. 1
”LO-“:J’dniz
{Ward St
0120mm '
191516 t- .
Emilie: -.

3550Claicd

 

(MOW International Research Team, 1989). Across countries and occupations, managers

and professionals reported preferences for autonomy.

Wcﬂt Structures

In Kristoff’s (1996) review of the person-organization ﬁt literature, work values
was a supplementary ﬁt variable, used to assess ﬁt from both individuals and their
organizations. Work structures were deﬁned as one of the resource/opportunity variables
organizations supplied in response to the resource/opportunities needed or demanded by
individuals (a complementary perspective). I have mixed these two approaches, matching
individual values with the work structures (resource/opportunities) available in
organizations to assess ﬁt.

This approach, linking individual values and organizational structures, has been
recommended or used by other researchers. Schneider (1987, 1995) theorized that people
in organizations develop the structures that ﬁt their goals and values. Although this
assertion has not been empirically tested, it is widely recognized that organizational
structures, particularly organizational reward systems, can signal organizational values
(Rynes, 1987). Researchers investigating person-organization ﬁt have described an array
of organizational structures in order to assess congruence. These structures include:
reward system characteristics (Bretz, Ash, & Dreher, 1989), pay, promotional
opportunities, type of work (Judge & Bretz, 1992), and total compensation systems
(Cable & Judge, 1994). While personality traits were not associated with the appropriate
attributes of organizational structures (Bretz, et al., 1989), individuals’ work values were

associated with structures that reﬂected those values when sufﬁcient information was

19

provided to make a value comparison (Cable & Judge, 1994). Sufﬁcient agreement
existed in the literature for reward practices to be used as one set of work practices that
were investigated in this study.

A second set of organizational structures was developed to ﬁt these speciﬁc
organizations. Human service organizations (HSOs) as a class are distinguished by two
key characteristics: these organizations work directly with and on people and they are
mandated to protect and to promote the welfare of the people they serve (Hasenfeld,
1983). Organizations differ in the methods use to work with their clients but all are
similar in that judgments and evaluation are made that inﬂuence clients’ fates. These
judgments are codiﬁed into organizational polices and professional practices. In HSOs,
the primary work practice concern is the management of competing pressures, either for
professional service to clients or for the administrative requirements of the organization
(Blau & Scott, 1962). For this study, a client services practice scale was developed to
describe the agency, professional, client and community factors that represented the array
of professional judgments and industry pressures that should operate in human service
organizations. The items in the scale were drawn ﬁ'om the institutional literature
presented later in this chapter. Individuals’ work values, in particular their preferences for
service to others and autonomy, were expected to interact with work practices that

reinforced or constrained those values.

Summm
This discussion of the terminal relationship in the model can be summarized by

the following hypothesis.

20

HI:

 

roger"
‘iLLSAIuy

lamb};

specific

.6 ‘5

 

H1: The ﬁt between an individual’s work values and work practices that reﬂect those
values positively predicts one’s organizational commitment, job involvement,
intent to remain with the organization and tenure.

In the next two sections, the two sets of antecedent factors promoting ﬁt are
presented: the person context variables predicting individual values and the work context
variables predicting work practices.

Relationship B:
Person Context Predicting Individual Values

Figure 3 identiﬁes the key variables for the prediction that context produces
speciﬁc individual values.

If values are enduring personal attributes, one might question how contextual
variables could inﬂuence an individual’s values. Recall that the investigation here is on
values that one believes can be realized on the job. There is some evidence that values
can be shaped by work experiences. In her study of accountants, Chatman (1991) found
that new recruits’ value preferences did become more closely aligned with their
organization’s values through the socialization process, and this improved alignment, in
turn, increased their satisfaction. In this section, literature is presented to support the
hypothesized inﬂuences that sector and position are expected to exert on the target values

of service to others, economic rewards and autonomy. First, a brief description of sectoral

differences is presented.

Sector Inﬂuences on Work Values
Studies of nonproﬁt managers have shown that they were committed to making

things better for the people they served, and they didn’t believe they would be able to

21

 

0328 82%
doﬁEonsmEoummeEwa mm H .3525
5895: 6:959 .ﬁoaéom mm H heuuom

mucmBoM egocoom
SEQ—83‘ .Emez mm H 825»

 

 

“5:80
Buses

 

 

.m

 

82m>
56.565

 

 

 

Figure 3. Relationship B: Person Context Predicting Individual Values

22

pursue these values in the corporate world (Witty, 1989). Young (1984) argued that
ﬁnancial rewards are less necessary in organizations whose employees value the
contribution they make to society more than they value pay. The following section
summarizes the research on the differential sector inﬂuences—for proﬁt, nonproﬁt and
public—on the three selected work values: altruism, economic rewards and autonomy. A
substantial amount of research has compared preferences for service to others with

economic rewards. Those studies are presented below.

Altruism and Economic Rewards

The literature comparing these two values is presented in Table 1. This table
provides highlights of each author’s ﬁndings related to altruism and economic rewards by
sector.
Khoj asteh (1993) surveyed the management hierarchy of twenty-ﬁve organizations (seven
public and 18 private industries) to explore the differences in motivating public and
private sector managers. The 362 managers who responded were both male and female
and ranged in education from high school diploma to some work beyond masters.
Respondents completed a questionnaire based on Herzberg’s Motivation Hygiene Theory
that asked them to rate twelve intrinsic and extrinsic rewards. He found a signiﬁcant
difference (p = .03) in the mean scores and rankings for pay, with private sector managers
ranking it most important and public sector managers ranking pay 7th in importance.
Public managers ranked achievement most important; for private sector managers

achievement 3rd most important (p = .05). He concluded that pay had a high motivating

23

V

Table 1

Cross-Sector Comparisons of Employee Values

 

 

 

 

Authors by Sector
Values For Proﬁt Nonproﬁt Public
Altruism
Rawls, et al., Lower in willingness to Higher in willingness to
1975 work for others work for others
Weisbrod, Comparison group Had a strong preference for
1983 public interest work over
usual corporate work. This
preference is signiﬁcantly
different from for-proﬁt
counterparts (Godderis,
1988)
Wittmer, Lower ranking (4) on Highest ranking (1)
1991 “being helpful to others” placed on “being helpful
but no difference on to others” [Hybrid
“community service” organizations (health and
schools) no different than
public]

 

Economic rewards

 

Khoj asteh,
1993

Preston,
1989

Rawls et al.,
1975

Weaver &
Franz, 1992

Pay is most important
(ranked #1)

Wages were higher for
white collar worker but not
for clerical workers

Higher economic values

No difference in preference
for high income.
Employees had highest
preference for promotion.
Entrepreneurs low and no

Pay is less important
(ranked #7)

Wages were lower for
white collar workers but
not for clerical workers

Lower economic values

No difference in
preference for high
income. Low preference
for promotion and no
difference from

 

difference from public entrepreneurs
Table l (cont'd).
Weisbrod, Served as the comparison Aware that pay is presently
1983 group and permanently lower.
This choice is signiﬁcantly
different than their for-
proﬁt counterparts
(Godderis, 1988)
Wittmer, Highest ranking (1) for Lower (3rd) ranking for
1991 “higher pay” “higher pay” [No
difference between hybrid

and public organizations]

 

24

 

ample;
parser;
to art:

Values .

studies, 1
“inroti

the Pa} d

 

the lawn
for Public
bemfen ‘1
that pUDilt

Daujg .

591m; 0"}..—

 

afr.‘ '9‘ ,.

5 ”£45.“?

 

potential for private sector managers, but had a low motivating potential for public
managers.

In the earliest study that compared the values of nonproﬁt with for-proﬁt
employees, Rawls, Ullrich, & Nelson (1975) used a battery of tests that measured
personality traits and life values.2 In their study population of graduate students planning
to enter either the nonproﬁt or the for-proﬁt ﬁelds, both personality differences and
values differences were associated with job choice and attitudinal differences They found
no differences between the two groups based on their demographic proﬁles, problem
solving ability, intelligence, or creativity. The two groups did differ in personality traits
and values. People planning to enter the nonproﬁt ﬁeld were signiﬁcantly higher in their
scores for comfortable life, ambition, willingness to pardon others, working for the
welfare of others and signiﬁcantly lower in their economic values and need for security.

Cross sector economic value differences were inferred in two wage comparison
studies. Weisbrod (1983) compared the wage structures of attorneys in for-proﬁt and
nonproﬁt public interest law ﬁrms. He found that public interest lawyers were aware of
the pay differences, knew that these differences were permanent (i.e., would not recover
the income later in their career), and made the choices because of their strong preference
for public interest work. He estimated approximately a $4,000 annual salary difference
between lawyers in the two sectors. Godderis (1988) reanalyzed these data and conﬁrmed
that public interest lawyers were systematically different from their for-proﬁt counterparts

in their employment choices. Using a similar approach, Preston (1989) compared white

2Dawis (1991) deﬁnes two categories of values measures; those that are broad in scope and apply to the
totality of human life and those that are work-related or vocational. In the Rawls, et al. (1975) study, values
were measured with a broad, totality of life inventory, the Rokeach Value Survey (Rokeach, 1973).

25

collar and clerical nonproﬁt with for-proﬁt employees in her economic study of pay
differentials across sectors. She found that managers and professionals with similar
characteristics were foregoing about 30% of their wages in the nonproﬁt sector; this was
not similarly true for clerical workers. She interpreted this difference as a “wage
donation” made in exchange for contributing to worthy causes since the wage differential
persisted when human capital, industry, occupation and job ﬂexibility factors were
controlled.

Weaver and Franz (1992) reanalyzed data from the 1972—1990 cumulative ﬁles of
the General Social Surveys administered by the by National Opinion Research Center at
the University of Chicago in order to compare work-related attitudes among
entrepreneurs, public, and for-proﬁt employees. In the survey, respondents had been
asked to select the “one thing on this list you would most prefer in a job.” They were
offered a list of items including high income, no danger of being ﬁred, and gives a feeling
of accomplishment. All types of employees expressed a preference for jobs with high
income. Private sector employees had the strongest preferences for promotion, and public
sector employees were most likely to prefer job security.

Wittmer (1991) deﬁned his work as a study of reward preferences or work-related
values among private, public and hybrid employees. The term “hybrid” referred to
organizations whose functions were not strictly performed in either the public or the
private sectors, and were composed of individuals working in health and education.
Public and private sector organizations in his study were selected to represent those that
only occurred in one or the other sector. Respondents in this study ranked eight rewards,

including “higher pay than you have now” and “doing work that is helpful to other

26

 

 

mien;
others \
profess

Shem é

people” according to “how important they are to you.” The most disparity occurred in the
rankings for higher pay and being helpful to others. Employees in the private sector
ranked higher pay ﬁrst and the employees in the public and hybrid sectors ranked being
helpful to others ﬁrst. They also reversed their rankings of the opposite reward; i.e.,
private sector employees ranked being helpful as fourth and public employees ranked
higher pay as third most important.

Two additional studies of occupation/profession differences offer supporting
evidence for this discussion of cross-sector differences. A contrasting pay/service to
others work values pattern was found in a study of students anticipating careers in helping
professions when compared with students anticipating careers in other professions (Ben-
Shem & Avi-Itzhak, 1991). Students preparing for careers in the helping professions such
as nursing, library science and special education valued altruism (service to others) higher
than economic rewards. The pattern was reversed for students entering other professions
such as economics/accounting, engineering, industrial graphics. Second, in an
international, multi-occupational study, researchers found that managers and
professionals reported work preferences patterns that de-emphasized pay and emphasized
the organization’s products, services and tasks more than lower level employees (MOW
International Research Team, 1987). Respondents came from one of eight counties and
ten occupations. While neither of these are sector-related studies per se, each has sector
parallels. That is, the helping professions tend to be more commonly practiced in the
nonproﬁt and public sectors and the other professions practiced in the for-proﬁt sector. In
the international study, managers and professionals came from occupations that crossed

the public and for-proﬁt sectors.

27

 

10310“ 1

I

|
ﬂnpkg
public ;
Sulaiu
"public

dimensi

 

nonprof

ﬁNpIO}:

199: )‘ 1
Plfi‘erer

Sudtrk

public 2

T115

The results from these studies are quite consistent although none of the authors
compared employees in all three sectors. I made the following linkages in order to extend
my hypotheses to the missing groups. The ﬁrst relates to the emphasis that individuals
give to their value of service to others. To develop a rank order for all three groups, the
following logic is offered. The study of attorneys working in the nonproﬁt sector
(Godderis, 1988; Weisbrod, 1984) echoes sentiments ascribed generally to public sector
employees: that a “normative foundation for public employment is a desire to serve the
public interest” (Perry & Wise, 1990, p. 368). Further, in a study of professional values,
social workers and public administrators were no different in the value they placed on
“public interest” while business administrators were signiﬁcantly lower on this value
dimension (N albandian & Edwards, 1983).3 Therefore I have proposed that both
nonproﬁt and public employees will place a higher value on serving others than for-proﬁt
employees.

Again, when comparing preferences for pay, no author compared all three sectors.
Although the results across public and for-proﬁt sectors are mixed (Weaver & Franz,
1992), the preponderance of evidence suggests that for-proﬁt employees have a stronger
preference for pay than public employees (Khojasteh, 1993; Wittmer, 1991). In the single
study that compared the for-proﬁt and nonproﬁt sectors (Rawls, et al., 1975), the
nonproﬁt employees had a lower preference for pay. To extend these ﬁndings to
differences between the public and nonproﬁt sectors, I suggest that serving a general

public good (in the public sector) is less compelling than meeting speciﬁc need (nonproﬁt

g

3The Galloway Edwards Professional Values Scale is an 18-item rank order scale patterned aﬁer
Rokeach’s (1973) Instrumental Values Scale. It includes factors such as compromise, effectiveness,

28

sect-0'
la 1
m

110W

htp
' OT."

 

 

 

sector), so that employees in the former are less likely to trade wages for service than the
latter are. Therefore, I have predicted a lower value for economic rewards among
nonproﬁt employees than the other two sectors.

When the cross-sector evidence cited above is accumulated, the cross sector

hypotheses for the values service to others and economic rewards are as follows.
H2: Sector predicts the level of individuals’ work values:

H2a: Employees in the nonproﬁt sector rate altruism more positively (higher)
than those in the public sector, and both higher than those in the for-proﬁt
sector.

H2b: Employees in the for-proﬁt sector rate economic rewards more positively
(higher) than those in the public and both higher than those in the

nonproﬁt sector.

Autonomy

Among these few studies on work values, there is nonetheless a great deal of
variability in the characteristics of the respondents—j obs, education, occupation and
profession—that could contribute to their value differences. In this study, one of the ways
that I limited the range of unmeasured contributing factors was by targeting a group of
individuals in a single profession who have nonetheless chosen to work in different
sectors. Across all sectors, one of the attitudinal dimensions that has been routinely

associated with professionals is a positive value for autonomy.

 

 

management, professionalism, and public interest. Other than “public service,” none of the items are
cOtl'lparable to values being tested in this study.

29

53(11):;
nonpr:
Lis set
q‘Jiiii}

1934.7

 

n .
Pl"; .‘r‘f‘

’ “It:
051:2 if ,«g
‘ .a-A

Given that all professionals are likely to desire autonomous practice, guidance for
sectoral differences comes from discussions about the nonproﬁt sector. It has been
suggested that for those professionals employed in heteronomous organizations, the
nonproﬁt sector is seen as more hospitable for professionals because it is assumed that in
this setting professionals are more likely to be able to exercise their autonomy, pursue
quality service, and operate with the least amount of organizational bureaucracy (Majone,
1984).4 Professionals are thought to gravitate to nonproﬁt organizations because it is in
those settings that the presmned conﬂict between the goals and values inherent in one’s
profession come into the least conﬂict with the bureaucratic goals and values of
employing organizations.

H2c: Employees in the nonproﬁt sector rate autonomy more positively (higher)

than those in the for-proﬁt and public sectors.

Position

Position is predicted to affect one’s value for autonomy; it is not predicted to have
an impact on service to others or economic rewards. In a study of accounting
professionals, level in the organization was been found to be positively associated with
bureaucratic orientation and negatively associated with professional orientation (Sorenson
& Sorenson, 1974). In a cross-sector study of social workers in public and private

nonproﬁt health and social welfare organizations, neither geographic location (Illinois

 

 

. 4Hall (1975) presents the setting for professional work as one of the sources of intra- and
mtel‘professional variation. His three settings are: independent practice (the professional alone), the
autonomous professional organization (i.e., the law ﬁrm), and the heteronomous professional organization.
In the latter, the professional employee is subordinated to an administrative framework in which rules and
routine supervision control many aspects of the profession’s tasks. Social welfare organizations of the type
being studied here are offered as one such example.

30

burg-

i
198':
relation

 

 

IDthI:

Infra;

Organ

and Hawaii) nor public or private nonproﬁt auspice was predictive of respondents’
bureaucratic orientation, but level of job responsibility was (Wilson, Voth & Hudson,
1980)
H2d: Employees in direct service (subordinate) positions rate autonomy more
positively (higher) than those in supervisory positions.

Relationship C:
Organization Context Inﬂuences on Work Practices

Figure 4 provides a graphic illustration for the ﬁrst component of the model—the
relationships between organization context and work practices.

Three summaries are presented to elaborate these relationships: an overview of the
inﬂuence that sector is expected to play on the practices of organizatons, an overview of
the characteristics of the industries, and a discussion of the work practices that were
studied.

No separate discussion of sector is presented here; it is integrated into both the
industry and work practices discussions below. Young (1983) has suggested that the
inﬂuences of these nonproﬁt values can be detected in at the sector, industry, and

organization levels.

Industry

The pressure arising ﬁ'om external expectations of altruism has been seen as an
important factor, not only shaping the mission of nonproﬁt organizations but also their
Practices (Hodgkin, 1991; Young 1983). In this study, respondents employed within the
nonproﬁt sector work in industries that experience substantively different pressures from

customers or clientS, suppliers of required resources, competitors and regulatory groups

31

 

30:85
x83

 

 

 

.0338 S8893 dungeon

mm H «9.933— o_au=a><

mExm Raommmomoagoc “some

.68 $235233

mm n 9.33.. .5386 3E8 23:0

0

 

i850
x83

 

 

 

£35 .852 63:0? BED
33.8w began mm H .0526:—

anao: .333 amp—98m mm H Stem

Figure 4. Relationship C: Work Context Predicting Work Practices

32

 

 

 

indrsm‘

 

mil-'03;

' l

n)

(Thompson, 1967; Scott, 1987). The relationships among those organizations constituting
an industry were predicted to produce a set of organizational structures that were similar
among organizations within that industry and different from organizations in other
industries. Because of the differences in environmental pressures across industries,
variations were predicted in the ways in which employees within the nonproﬁt sector
perceive that they are able to act.

In this study, three industries were selected for study: family services, child
welfare services, and mental health services. These industries have been selected for two
reasons. First, these industries reﬂect the predominant practice specialties for social
workers, the target respondents in this study. Approximately 60% of the National
Association of Social Work’s members practice in these three areas.5 Second, these
industries have been selected as archetypes of nonproﬁt agency interactions with key
environmental actors. The industries differ as follows:

0 To date, family service agencies have maintained a diversiﬁed funding and
regulatory base so that such unitary demands administrative pressures are not seen
as the deciding factor in their client service decisions. However, family service
agencies represent a traditional nonproﬁt form that has experienced the most
substantial change in the type of services rendered and clients served.

0 Nonproﬁt child welfare services have increasingly entered into agreements with
state public welfare agencies as governmental services have been privatized and

downsized, resulting in extensive governmental oversight; and

 

 

5Medical and health care is the third largest practice speciality at 12.2%; all other practice specialities
reported are 5% or less (Ginsberg, 1992, Table 9-22, p. 201).

33

 

The iii

Smut -
resour;
3513K 1;

10cm;

 

0 Mental health services are being subjected to increasing for-proﬁt—like market
pressures as employers view cost containment for these services in the same
manner that they have viewed physical health care costs.

The industries are brieﬂy described below.

F_amilv Services

Family service agencies represent an historic means through which voluntary
commitment to “inﬂuence society and its institutions to encourage, protect, and promote
health family life” has been enacted.6 Family service agencies are descended from the
Charity Organization Societies of the 18008 (Erickson, 1987) and many agencies have a
similarly long history of service in their communities. The programs of voluntary family
service agencies have always been shaped by the needs of communities and by the
resources that were available to meet those needs (Erickson, 1987). Over the years, family
agencies have provided remedial services, been advocates for the poor, and moved from
focusing on individuals’ problems to an emphasis on preventing those problems.

The type and number of programs provided by family service agencies changed
substantially during the 19808 (Alperin, 1992, 1993). A recent study found that family
service agencies may provide any of ﬁfteen types of programs and that between 1980 and
1990, on average agencies expanded from providing services in three sites to ﬁve sites
and from providing services in ﬁve different areas to eight different areas (Alperin,
1993).7 The clients of family service agencies became more diverse and also presented

‘

6Family Service of America, 1990; quoted in Alperin, 1992, p. 33.
7The ﬁﬁeen areas of service are: adoption, adult day care, alcoholism, child care resource/referral, child
day Care, credit counseling, domestic violence, drug abuse, eldercare/supportive services to aging, family

34

111016

hem.

WWPS

sufr

 

 

$035.5

“Tu-mu
Ll‘brh:

nmhn

more serious problems for assistance. Overall demand for services has also increased with
the most grth occurring in the area of services for business and industry.8

Family services agencies appear to have experienced the greatest environmental
press based on changes in their clients’ needs. However, client needs are the traditional
purpose for nonproﬁt organizations. Therefore, when asked to describe agency practices,
staff in family service agencies will report that:

H4a: Staff knowledge and skills and client-related factors play the greatest role

in service delivery decisions.

Family service agencies can be perceived as being true to their mission even
though their employees struggle to manage the challenges presented by a new array of
clients and client problems. Using client-related factors to make service decisions should
reinforce, not alter the previously predicted high level of autonomy that they employees

experience in carrying out work with their clients.

Child Welfare

A system of public and nonproﬁt child welfare services has been developed to
carry out community preferences for protecting children and supporting their families
(Liedennan, 1995). The phrase “child welfare” is used to describe a cluster of public
agency services that may include child protective services, out-of-home care (foster care,
kinship care, and residential services), and adoption services. With changing public

perceptions about the need not only to protect children from harm but also to support

 

life education, family/individual counseling, homemaker service, home health care, services to business and
industry, teenage pregnancy/parenting (Alperin, 1993).

35

families as the best source of child care, additional services have been developed. A
transition to child-centered and family-focused practice has given rise to an array of
family preservation and family support services designed to strengthen parents’ abilities
to care for their children and reduce the likelihood of out-of-home placement. While
public agencies retain the responsibility to determine risk to children and make decisions
about the best locus for their care, private non-proﬁt agencies have increasingly been the
partners in the delivery of many of the child- and family-supportive services.

Among the social services, it is in the area of child welfare that the greatest
impact of privatization of public services has occurred. While states vary in the extent to
which they use purchase of service (POS) arrangements to deliver social services,
national studies have shown that virtually all states use contracts to deliver child welfare
and adoption services (Kettner & Martin, 1994). The largest volume of POS contracts are
with individuals—supporting the provision of foster care, for example——although
organizations receive the greatest amount of ﬁnancial support and provide the largest
volume of service (Gronbjerg, Chen, & Stagner, 1995).

The public sector’s extensive reliance on nonproﬁt providers has altered the
traditional role of nonproﬁt organizations from one of an alternative, supplement or
complement to government services, to a role of substitute for government services
(Kramer, 1994). Concerns abound for the negative impacts of contracting on the
character, goals, and roles of nonproﬁt organizations in the delivery of personal social

services. For some, government funds are seen as inherently controlling, contaminating

8That speciﬁc change can be linked directly to contracts between FSA and the Xerox Corporation and
also between FSA the General Motors Corporation, with FSA providing employees’ services through
members agencies across the country (Erickson, 1987).

36

them

with

in cc:

"0116 .

is ac:

[radii

the mission of the nonproﬁt organization and undermining the essential contribution of
voluntarism to a pluralist society. The limited research suggests that decisions to engage
in contractual relationships are much more likely to arise from practical concerns such as
“one agency has something that the other needs or wants, and can provide it at a price that
is acceptable” (Kramer, 1994, p. 40). However, it is clear that POS contracts do bring a
set of funder expectations, ranging from a minimum expectation of staying within budget
to an expected level of performance for achieving speciﬁc results with clients (Kettner &
Martin, 1994), as well as the expected set of regulator concerns. Therefore, employees in
nonproﬁt child welfare agencies will report that:

H4b: contractual requirements play the greatest role in client service delivery

decisions.

The rationale for this predictions is as follows. POS contracts change the
traditional relationship between the agency and the client. Nonproﬁt agencies are
committed to being responsive to clients who best ﬁt the agencies mission; they are not
committed to serving all clients. POS contracts emphasize contractual (rather than client)
priorities or equity in service delivery in ways that substantially alter the nonproﬁt

agency’s triage patterns and even their right to reﬁrse clients service (Smith, 1989).

Mental Health Services

 

“Mental Health services are activities that promote mental well-being and
alleviate mental disorders” (Lin, 1995, p. 1705). Treatment care for chronically mentally
ill people was originally hospital-based but today many individuals live and are served in

Communities through a network of community mental health agencies. Mental health

37

anph;
Cones"
rice: .'
com:
from
mamas.

from

3CCCSS

1ﬁ16r35t

mm
10°00:
”39011:.
iguana
pml'idc'
hateii:

 

 

“1'6 an:
“1051”1

1“

C

services are ﬁnanced through a combination of private insurance, state general funds,
Medicaid, and Medicare.

The mental health movement was originally shaped by core values that
emphasized access to treatment and continuity of care (Clark, Dorwart, & Epstein, 1994).
Concern has arisen over the extent to which the incentives of the health care markets
aﬁ‘ect nonproﬁt agencies’ fulﬁllment of their public expectations. In a national survey of
community mental health agencies, Clark, et a1. (1994) found that increased competition
from private practitioners prompted both public and private-nonproﬁt agencies to use the
management practice of more aggressive fee collection services. Increased competition
from health maintenance organizations and private psychiatrists also appeared to reduce
access to treatment for ﬁnancially disadvantaged consumers (Clark & Dorwart, 1992).

Some of the most recent and dramatic changes have occurred because of increased
interest in insurance-related cost containment. This is as true for publicly funded
insurances, i.e. Medicaid, as it is for private employer-paid insurance. While less than
10% of the community mental health centers in the Clark & Dorwart (1992) study
reported above were private for-proﬁt organizations, the managed mental health care ﬁeld
is “a new and rapidly growing industry employing thousands of clinicians and network
providers across the country” (Winegar, 1993, p. 171). Large employers, in particular,
have initiated managed mental health care programs as part of their interest in controlling
overall health care costs. For-proﬁt managed care organizations have developed level-of-
care and preferred-practices standards and sought afﬁliations with any type of providers
Who share their philosophies and are willing to deliver services within their guidelines.

Although not all aspects of the managed care approach are antithetical to professional

38

 

ramp

1
com;-

(lime

the 511.1.

 

‘ 3
B-
(.1
—rﬁ—__

 

nonproﬁt providers, the beneﬁcial aspects such as the use of quality indicators (a client
complaint/satisfaction system, written treatment plans, and staff training and development
programs) are always combined with guidelines for providing care, thus impinging on
professionals’ decision-making autonomy. Therefore it is predicted that the general sense
of market pressure will result in nonproﬁt mental health employees reporting that:

H4c: F under or insurer requirements play the greatest role in client service

delivery decisions.

Work Practices

It is in the area of organization work practices that the greatest inﬂuence of
sectoral values are expected to be exerted. Not only are industry forces at play, but
additional organization inﬂuences are brought to bear. For some nonproﬁt organizations,
their board members’ expectations that the agency will operate in a charitable manner is
only one of several distinct and sometimes contradictory constituency expectations
(Kanter & Summers, 1987). Many types of work practices could be examined. Because
the study focused on three work values that characterize professions in nonproﬁt
organizations—service to others, economic rewards, and autonomy—the exploration of
work practices was also limited to those that provided corroborating organizational
evidence for these individual differences. Two work practices, those involving client

service decisions and those involving organizational rewards, were examined.

Client Service Decision_s
I am suggesting that there will be consistency between the broad sectoral values

and the opportunities for acting on those values in human service organizations (11805)

39

 

 

 

3hr:

that:

 

anau
nonp
moﬁi

dent.

 

mod:
hair

pmhf

C0115;

 

within each of the sectors. One way in which those sectoral values can be observed is to

examine how human service organizations (HSOs) balance professional judgments with

administrative needs in delivering services to clients (Blau & Scott, 1962). I predicted
that:

H: In for-proﬁt HSOs, administrative requirements are emphasized more than
professional service to clients, and these administrative requirements are
determined primarily by the associated ﬁnancial implications.

A: In public HSOs, administrative requirement are emphasized more than
professional service to clients, and the administrative requirements
emphasize uniform service for publicly sanctioned problems.

B: In nonproﬁt HSOs, professional service to clients will be emphasized
more than administrative requirements.

In HSOs, professionals may ﬁnd themselves with much, or little, latitude to act as
an autonomous practitioner. Although both Young (1984) and Maj one (1984) assert that
nonproﬁt employees enjoy more work autonomy than individuals in the public and for-
proﬁt sectors, no studies were found that tested these assertions. Thus, this prediction
derives from the assumption that making a decision based on one’s own professional
judgment (the nonproﬁt model; Maj one, 1984) affords one more autonomy than making
decisions based on policies designed to affect all clients of a group similarly (the public
model; Kramer, 1994), and both of these approaches provide more individual autonomy
than making decisions based on the cost of the service to be delivered or the cost of the
problem to be solved (the for-proﬁt model; Winegar, 1993). Therefore, I predicted

consistency between the sector and work practices with respect to autonomy.

40

 

some:
apps:
h gr:
used
sector
organi
mp0?

no: on

 

 

otlerir

\‘olun:

Consist
10 [hem

1116 Cm"

 

A: People working in nonproﬁt HSOs will report more work autonomy than
those in the public sector, and both will report more than those working in

the private sector.

Organizational Rewards

Young (1983, 1984) has provided the most comprehensive evidence linking broad
societal values with organizational rewards in nonproﬁt organizations. He interviewed
approximately 40 directors of nonproﬁt associations to identify reward types and patterns.
In general, be found that performance-based and incentive rewards were unlikely to be
used in nonproﬁt organizations. He attributed these patterns to the general purpose of the
sector and, in particular, external expectations of altruism (Atkinson, 1989). Nonproﬁt
organizations are assumed to signal unique values of altruism, philanthropy, social
responsibility, equity, and fairness (McCauley & Hughes, 1991). Young suggested that
not only would a nonproﬁt organization’s image as a charity be incompatible with
offering ﬁnancial incentives to staff, but the presence of donors and other community
volunteers would reinforce the expectation that any excess funds be channeled toward
services for clients and away from staff salaries or incentives. These predictions will be
tested using a list of varied rewards (Kerr, 1988) that have been previously pilot tested
with the staff of a large nonproﬁt human service organization (Reed, 1992).

While rewards are expected to be differentially distributed across sectors, the
Consistency between an individual’s work values and the organizational rewards. available
t0 them is not expected to be sector-related. There are several alternative explanations for

the consistency between an individual’s reward preferences and an organization’s

41

 

COST

 

and?

 

complement of rewards. Whether employees are attracted to values signaled by the sector,
as Young (1984) suggests, or participate in the design of reward systems they value
(Schneider, 1987), or come to value the rewards that are available to them (Loscocco,
1989), a high degree of consistency is expected.

The following hypotheses predict the relationships between organization context
and both sets of work practices:
H3: Sector predicts the factors emphasized in organizations’ practices:

H3a: In the for-proﬁt sector, the most extensive package of ﬁnancial rewards are
available to employees and the cost factors are rated as being the most
inﬂuential (highest) in client service decisions.

H3b: In the public sector, policies and rules are rated as the most inﬂuential
factor (highest) in client service decisions.

H3c. In the nonproﬁt sector, the most extensive package of client/professional
rewards is available and client/professional factors are rated as most

inﬂuential (highest) in client service decisions.

Summary
This chapter summarized the literature used to build the model and predict the
relationship among the variables in that model. The model itself was developed using the
Work of three authors: Jennifer Chatrnan, Benjamin Schneider and Dennis Young. The
fOCus of their contributions is identiﬁed in Figure 5.
Young’s (1984) work not only provided the initial impetus for this study, but also

Supplied the documentation on the inﬂuence that sector has on organizations’ work

42

 

  

 

    

 

 

 

“5:80

3:365

 

 

m

111)

 

 

mos—5w
RawEeE

 

 

 

waso> £569

 

 

 

 

258.30 detach

 

 

225,—.
235% 8 ESE
“soggoo Ecoummmemwho
EoEoZoZ: pea
”83mg; “coccomoa

 

 

 

‘||"'ll
.I|IIIII all...
\ I,
\

Axuoanaom Ema Fem we

I \

 

 

 

 

8388A A 38280
#83 U #83

 

 

 

 

 

 

 

 

Figure 5. Sources for Model Development
43

 

 

pare:
port
org;
lithe.

Olga:
V

\‘alu

practices and individuals’ values. His assertions form the basis for the predictions tested
in the hypotheses about sector and work values. The relationship between individuals’
values and organizations’ structures is drawn from Schneider’s (1983a & b, 1987)
theories about how people are drawn to organizations and, in turn, how they affect the
work practices that are developed in organizations. His predictions, in part, suggested that
this study’s respondents would be able to provide a reasonable representation of client
services decision making structures in their organizations. Finally, Chatrnan’s (1991)
extensive research on accountants and accounting ﬁrms provided guidance not only for
the model’s development but also lent support to the approach of investigation these
patterns within a single occupational group. She veriﬁed the extent to which individuals’
work values could interact with work practices to promote the individual outcomes of
organizational commitment and intent to remain with the organization. In addition,
although accountants have similar education and training and professional accounting
organizations have similar objectives, nonetheless differences were found in the work
values patterns of both individuals and organizations and similarities were found between
employing organizations with their employed individuals. I have used these three, and
other literature as cited, to build the model that was tested with the following hypotheses.
H 1: The ﬁt between an individual’s work values and work practices that reﬂect those
values positively predicts one’s organizational commitment, job involvement,

intent to remain with the organization and tenure.

44

H2: Sector predicts the level of individuals’ work values:

H2a:

H2b:

Employees in the nonproﬁt sector rate altruism more positively (higher)
than those in the public sector, and both higher than those in the for-proﬁt
sector.

Employees in the for-proﬁt sector rate economic rewards more positively
(higher) than those in the public and both higher than those in the

nonproﬁt sector.

H3: Sector predicts the factors emphasized in organizations’ practices:

H3a:

H3b:

H3c.

In the for-proﬁt sector, the most extensive package of ﬁnancial rewards is
available to employees and the cost factors are rated as being the most
inﬂuential (highest) in client service decisions.

In the public sector, policies and rules are rated as the most inﬂuential
factor (highest) in client service decisions.

In the nonproﬁt sector, the most extensive package of client/professional
rewards is available and client/professional factors are rated as most

inﬂuential (highest) in client service decisions.

In the next chapter the methods and instruments used to test the model components are

presented.

45

CHAPTER 3
METHOD

In this chapter, the details for this exploratory survey are provided. Information on
the subjects of the study, the measures to be used and the analytic techniques will be

presented. Each of the scales described can be found in the Appendices as noted below.

Study Sample

Participants in the study were drawn from the membership of the Michigan
Chapter of the National Association of Social Workers (MI-NASW or the Association).
As members of MI-NASW, all respondents had (or were working toward) a social work
degree at one of three levels: the baccalaureate, master’s or doctoral level. Members’
employment status varied widely; they could be students, retired, unemployed or
employed in a wide variety of settings. For this study, the sample as drawn was a subset
of MI-NASW members who worked in the following industries: child/family welfare,
health, mental health, and occupational social work/EAP programs. The selected work
settings were business/industry, health, managed care, mental health, residential facility,
and social service agencies. The sample represented all Association members in these
industries/settings from six counties: Genessee, Ingharn, Kalamazoo, Kent, Ottawa and
Washtenaw counties. The mailing list purchased from the Association had addresses for
2,237 members who ﬁt the industry and setting speciﬁcations in these counties.

The counties were selected to minimize differences in the rate of nonproﬁt social
service employment. Research on nonproﬁt organizations in Michigan indexed counties

to a state average for distribution of employment in nonproﬁt social services (Wilson,

46

1991). The counties selected for this study were either at (near) the state average (Kent,
Ottawa and Washtenaw Counties) or in the next higher tier (Genessee, Ingham,
Kalamazoo counties; Wilson, 1991 , Map 22, p. 41).9 Counties with the highest level of
nonproﬁt employment in the state were excluded. These were primarily rural counties
(Alpena, Dickinson, Ionia, Iron, Lapeer, Menominee) whose inclusion would add few
additional respondents while greatly increasing the amount of service system difference
that could have created unmeasured variation in respondents’ working conditions.

The questionnaire plus a stamped self-addressed return envelope was mailed to all
eligible MI-NASW members in the six selected counties.10 In the cover letter, respondents
were offered the opportunity to submit their name for a lottery; the prize was a one-year
basic membership in NASW. Approximately ten days after the instrument was mailed, a
generic follow-up postcard was sent, thanking individuals for their response if they had
already completed the questionnaire or urging them to complete and return the
instrument. The cover letter, instrument and follow-up postcard text are located in
Appendix A.

Although the Chapter ofﬁce provided a means to target the mailing list,
nonetheless individuals appeared on the mailing labels who were not eligible to
participate. For this study, individuals could be ineligible because of their employment
status (retired, in private practice, student) or because they worked in an industry not

being studied (higher education, school social work, aging services, etc.). In order to

 

9Wayne County also has nonproﬁt social service employment at the state-wide rate. However, the
inclusion of this county would have added up to 1,514 additional NASW members to the study. That level
of increase was not feasible.

10The mailing list included faculty at schools of social work. Based on my personal knowledge of their
employment, I removed individuals who were not in agency-based practice.

47

discover the actual size of the survey population, individuals were asked to return only
the cover letter identifying the primary reason for not responding. The reasons given
were: engaged solely in private practice, retired too long, out of the work force, and
enrolled as a student. These responses were collected from comments written on the
cover letter, sent via email or offered over the telephone. This study is based on 694
responses from a reduced total population of 1,942; the response rate was 35.7%.

Table 2 shows the demographic characteristics of all respondents and the
characteristics of the respondents who were analyzed in this study.

As can be seen in Table 2, respondents were employed in industries other than
child welfare, family services or mental health. Because I had speciﬁc industry-related
hypotheses, it was necessary to select only those employed in the targeted industries. The
results reported in this study are for the 438 respondents employed in child welfare,
family services or mental health. Mental health is a composite of those who originally
identiﬁed their work in mental health, occupational social work/employee assistance
programs and substance abuse, since those three industries routinely address mental
health concerns. The modal respondent is a woman with an MSW who has worked in her
agency almost 8 years and is currently in a direct service position.

The means, standard deviations, and correlations for the independent and
dependent variables are shown in Table 3. Where applicable, the reliabilies for the scales

are shown on the diagonal.

48

'W—

Table 2

Respondents’ Characteristics

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Selected Indusz
All Respondents Respondents
Variable n = 694 (n = 438)
Gender
Female 75.4% 71.2%
Male 2 1 .0% 24.4%
No Response 3.6% 4.3%
Race
Caucasian 89.0% 90.2%
African American 14.1% 2.7%
Latina/Latino, Asian American, Native American, Arab > 1.0% > 1.0%
American
No Response 3.6% 4.3%
EducationH
BA/BSW 16% 15.5%
Other BA or BS 38.5% 40.0%
Some Graduate Work 4.0% 4.3%
MSW 91.9% 91.8%
Other MA, Other Degree, Other Professional Degree 2 6.1% 2 6.2%
PhD in Social Work, Other PhD 2 0.7% 2 0.5%
Licenses or Certiﬁcation
Certiﬁed Social Worker in MI 75.5% 76.3%
ACSWl2 54.3% 55.0%
Advanced Diplomate 9.1% 8.7%
Certiﬁed Substance Abuse Counselor 10.5% 13.2%
Certiﬁed Marriage Counselor 5.5% 7.5%
Other Licenses or Certiﬁcation 22.7% 21.9%
Pattern of Employment
Full-Time Employment in One Agency 56.9% 56.8%
Less than Full-Time Employment in One Agency 14.7% 12.6%
Full-Time Plus Part-Time Private Practice or 10.0% 11.2%
Consultation
Full-Time Plus Part-Time in Another Agency, Less than 2 6.6% 2 6.6%
Full-Time Enrployment in Each of Two Agencies,
Recently Retired From This Agency, and Other
Sector of Aggncy
Nonproﬁt 51.4% 50.7%
Public 39.0% 38.6%
Private For-Proﬁt 8.6% 9.8%
Industry (Field of Practice)
Child Welfare 10.8% 17.1%
Family Services 10.4% 16.4%
Mental Health 34.9% 55.3% (+ 3.9 + 7.3
= 66.5%)

 

llRespondents could check as many as were true for them for Education and also Licenses or

Certiﬁcation.

”The Academy of Certiﬁed Social Work (ACSW) and the Advanced Diplomate are awarded by the
National Association of Social Workers to masters prepared social workers with additional credentials.

49

Table 2 (cont’d)

 

 

 

Occupational Social Work/EAP 2.4% (3.9%)
Substance Abuse 4.6% (7.3%)
Health 17.0% --
Other 17.3% --
No Response 2.6% --
Position
Direct Service 63.3% 65.1%
Direct Service & Supervision 14.1% 14.1%
Management, Administration or Supervision 15.8% 15.6%
Other 5.2% 3.7%
No Response 1.3% 0.2%
Years in This Agency (All Jobs) Mean = 7.6 yrs; Mean = 7.8; Range:
Range: .08 yrs - 36 .08 - 34 yrs; SD =
yrs; SD = 7.3 yrs 7.5

 

50

Table 3a
Means, Standard Deviations, Correlations and Scale Reliabilities for the Study

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables
Variable N x SD. 1 2 3

1. Position - Direct 438 .6507 .4773
Service+

2. Position - 437 .1625 .3693 -.600"'*
Administration
Mai

3. Private Nonproﬁt 438 .0098 .2979 .097* -. 123‘
Sector+

4. Pubic Sector+ 437 .1259 .3321 .134" -.111"' -.007

5. Private For-Proﬁt 437 .0043 .2042 .015 -.033 -.081
Sector+

6. Family Services 438 .1644 .3710 .002 .038 .036
Industry

7. Child Welfare 437 .1716 .3775 -.086 .21 1" -.008
Industry+

8. Mental Health 438 .6644 .4727 .067 -.198** -.022
Industry+

9. Work Values (WV) - 435 14.0736 1.3334 .040 -.020 .052
Altruism

10. WV - Economic 425 12.1106 2.1212 .034 -.063 -.l79*"'
Rewards

1 1. WV — Autonomy 432 12.4190 2.4358 .085 .075 -.094

12. Client Services - 430 21.9023 5.4614 -.072 .138" .018
Professional

13. CS — Financial 420 45.1000 7.8849 .040 -.080 -.021
Practices

14. CS - Legal 437 7.7803 2.0436 -.025 .031 -.068
Mandates

15. Rewards — 438 7.8627 4.3388 -.042 .029 .103“
Professional

16. Rewards — Financial 438 9.9338 3.8529 -.090 .100* -.106"'

17. Organizational 426 51.9977 1 1.2278 -.157*"‘ .180" -.097*
Commitment

18. Job Involvement 428 26.4813 6.1448 -. 128 .149" -.093

19. Intent to Leave 402 7.7914 7.4660 .092 -.104* -.024

20. Tenure 437 6.0050 2.5178 -.296** .248" -.091

 

 

 

 

 

 

 

+ = Partial r’s for dummy variables

"' = Signiﬁcant at .05
** = Signiﬁcant at .01

51

 

Table 3b

Means, Standard Deviations, Correlations and Scale Reliabilities for the Study

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables
‘var 4 5 6 7 1 8 9 10 11
1.
2.
3.
4.
5. 4081
6. .036 4004
7. ”008 4008 ”2029*
8. .022 .009 a624** 4639**
9. .052 4046 .059 .058 4092 .81
10. a179** 4033 4089 4003 .073 4072 .76
11. a094 .051 4044 .041 .001 .132** .233** .67
12. .018 4023 .038 .069 4085 .069 .047 .064
13. 4021 .002 4019 4015 .027 .135** .184* .103*
14. a068 4038 .002 .009 a010 .022 .215* .093
15. .103* 4061 .030 4018 4010 4044 4148* 4026
16. 4106* .015 4115* .097* .012 4075 4100* 4015
17. 4097* 4181** .038 .036 2060 .107* 4018 4024
18. 4093 a066 .109* .052 4127** .193** .038 .076
19. 4024 .022 4079 4004 .064 4077 .248* .091
20. 4091 .143* 4098* .063 .026 4077 4121* 4003

 

 

 

 

 

* = Signiﬁcant at .05
*"' = Signiﬁcant at .01

52

 

Table 3c

Means, Standard Deviations, Correlations and Scale Reliabilities for the Study

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variables
vs: 12 13 14 15 l 16 17 18 19
r
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12. .87
13. .228“ .76
14. .140** .366** .52
15. .098‘ 4058 4139** .76
16. 4021 4104 4025 .578** .76
17. .233*3 4068 .083 .137** .072 .84
18. .076 4075 5038 .050 .054 .460** .84
19. 4240** .126‘ .168** 4204** a138'* 4383** 4195* .81
20. .015 4012 .007 .003 .202** .149** .102* 4060

 

 

 

 

 

 

 

 

 

* = Signiﬁcant at .05
'1" = Signiﬁcant at .01

53

 

Measures
In Chapter 2 variables were introduced that were used to test the proposed study
hypotheses. One or more measures is associated with each of these variables. The
measures are presented below in the order of their previous discussion:
1. Variables associated with Relationship A, Individual Values and Work Practices
Predicting Dependent Variables;
2. Variables associated with Relationship B, Person Context Predicting Individual
Values; and
3. Variables associated with Relationship C, Organization Context Predicting
Organization Practices.
Each of the scales discussed can be found in Appendix B as noted in the Table of

Contents.

Relationship A

Four dependent variables and three sets of independent variables were associated

in this component of the model. The dependent variables are presented ﬁrst.

Dependent Va_rgrbles

The four dependent variables were: organizational commitment, job involvement,
intent to leave and tenure.

Organizational commitment. Organizational commitment is deﬁned as a strong
belief in and acceptance of the organization’s goals and values; a willingness to exert

considerable effort on behalf of the organization; and a strong desire to maintain

54

membership in the organization (Porter, Steers, Mowday, & Boulian, 1974). Two
approaches to measuring organizational commitment have been used, a single factor
version developed by Porter & Smith (1970; Porter et a1. 1974) and a multiple factor
approach. In this study it seemed prudent to use a scale that differentially measured
affective, continuance and normative commitment, since popular wisdom suggests that
many people stay with public organizations because their tenure beneﬁts outweigh any
undesirable work experiences they might have. I used Meyer and Allen’s (1991) three-
component model of commitment. This sixteen item instrument is scored on a 7-point
scale (1 = strongly disagree; 7 = strongly agree). In their study (Meyer, Allen, & Smith,
1993) of registered nurses, they report reliabilities as follows: Affective commitment (01 =
.82); continuance commitment (or = .74); normative commitment (01 = .83). I used a ﬁve-
item Likert scale to conform with the other scales in my instrument and found the
following: Affective commitment (or = .82); continuance commitment (01 = .76);
normative commitment (01 = .82); and a total scale reliability of .84. The scale as a whole
was used for these analyses. See Appendix B for the scale.

Job involvement. Job involvement is deﬁned as a descriptive positive belief about
a speciﬁc job (Kunungo, 1982). Kunungo’s ten-item Job Involvement scale was used in
this study. In his construct validity study the internal consistency coefﬁcient was .87 and
the test-retest coefﬁcient was .85. In this research the internal consistency coefﬁcient was
.84; see Appendix B for the scale.

Intent to leave. This was measured using two of Chatrnan’s (1991) items. She

reported only that the response was a Likert-type scale and did not report the scale

55

reliability. The two items used as a scale for Intent to Leave (01 = .81) asked if people had
“thought seriously about another job” and “prefer(ed) another, more ideal job.”

Tenure in the agency. This was measured with a single item that question that
asked how long people had worked for the agency in any job. The mean was 7.6 years,

the range from .08 to 36 years, and the standard deviation was 7.4 years.

Independent Variables

Subscales from three scales were used as independent variables. Each measured a
set of individual work values, client service decision factors, or available rewards. Each
of these sets is described below.

Individual work values. In this study, selected subscales from Super’s (1977)
Work Values Inventory (WVI) were used to measure Altruism (service to others),
Economic Rewards, and Independence (autonomy). The items forming these scales can
be found in Appendix B.

Super (1970) reported test-retest reliability coefﬁcients for this form ranging from
.74 to .88 with a median of .83. In the Ben-Shem & Avi-Itzhak (1991) study the following
coefﬁcient alpha reliabilities were reported: Altruism .84; Independence .67; and
Economic Return .76. In this study, the following were found: Altruism .81;
Independence (autonomy) .54; and Economic Return .83. The total scale coefﬁcient alpha
was .83. The levels for the Altruism and Economic Return subscales approach or exceed
the standard suggested by Nunally (1978) that instruments used in basic research have
modest reliabilities of .70 or higher. The difﬁculty that arises from using an unreliable

scale, such as that for Autonomy, is that it introduces measurement error into the

56

analyses. A ﬁnding of non-support for a hypothesis could be the result of true lack of

relationship between the variables or it could be the result of lack of stability in the

measure itself.

Work practices. The work setting practices investigated were selected to mirror
the individual values presumed to be related to sectoral differences. The instrument
measured perceptions of general client service practices and available rewards. Each of
these scales is discussed in turn, beginning with the Client Services Decision Making
Practices scale.

0 Client services decision making practices. Inﬂuences ﬁ'om the agency, the staff, the
ﬁrnders, the volunteers, and the clients themselves all may make a contribution to an
agency’s client service decisions. A list was developed that drew broadly on the
general factors mentioned by Young (1983) as well as the speciﬁc industry-related
factors that are thought to be particularly relevant to mission- or client-oriented
service in nonproﬁt organizations. The list was organized as a rating scale that asked
respondents to identify the type of inﬂuence the item has on client service decisions,
i.e., from “a major inﬂuence” to “little or no inﬂuence” and included the option that
“this is not applicable to our clients or services.” The scale can be found in Appendix
B.

Recall that the hypotheses predicted inﬂuences of ﬁnancial practices,
professional attributes, and legal/mandate practices on client service decision-making.
Principal components analysis with varimax rotation was used in an exploratory
fashion on the twenty-three client services decision-making practices listed. The

factor analysis resulted in these three expected factors and two others that could be

57

interpreted but were not relevant to these hypotheses. The summary ﬁve-factor
solution explained approximately 57% of the variance. The results of the factors
analyses and coefﬁcient alphas for the three hypothesized scales are given below in
Table 4. The other two scales, while of practical importance for client service
delivery, are not of theoretical interest here. The factor analysis results for the unused
agency attributes and client characteristics scales can be found in Appendix C.

As was true for the Work Values subscales, these three are not similarly
reliable. The two-item legal requirements subscale has a reliability of .52, less than
that desirable for this exploration. However, it was theoretically viable and has been
used in the analyses.

Work rewards. I queried the presence (or absence) of highly valued rewards.
Organizations can offer a wide variety of rewards. The list used here was based on
Kerr’s (1988) delineation of forty-three rewards that he organized into ﬁve groupings:
ﬁnancial rewards, prestige rewards, job content rewards, combination ﬁnancial-
prestige-job content rewards, and interpersonal rewards (see Appendix B for the scale
items). This reward list was presented to the administrative team of a large urban
multi-purpose child welfare agency. They conﬁrmed the relevance of the list for their
organization and also added a group of agency speciﬁc rewards to the list (Reed,
1992). The rewards were matched as closely as possible to the items in the selected
Work Values Inventory scales. Respondents were asked if these rewards are available
in their agency, and if present, asked if the rewards are available to them. Table 5

shows that most rewards were available.

58

Table 4

Results of Factor Analysis of Client Service Decision Making Practice Items

 

 

 

 

 

 

 

 

 

Factor Loadings Communality
Items 1 2 5 Estimates

Factor 1: Professional Attributes (01 = .87)
Sl3.Staff motivation and participation .905 .008 -.000 .840
Sl4.Staff cooperation inside the agency .873 .005 .000 .803
SlZ.Staff knowledge and skills .833 .006 .000 .743
Sl6.Staff professional judgment .728 -.007 .005 .622
$1 1.Staff commitment to the agency

mission .710 .119 -.000 .603
SlS.Staff collaboration with others outside

the agency .415 .137 .361 .373
Factor 2: Financial Practices ((1 = .76)
S4. Funder rules regarding costs of services .115 .746 .208 .613
S8. The predictability of external funds .010 .673 .266 .542
S3. Funder rules regarding uniformity of

services to clients .151 .666 .346 .599
SlO.Board member and/or donor

preferences -.002 .641 -.1 1 8 .522
S9. The cost of alternative methods of

service -.000 .617 .217 .487
Sl8.The client’s ability to pay (or

insurance coverage) -.002 .504 -.221 .459
822.The availability of volunteers -.007 .434 -.293 .439
Factor 5: Legal Requirements (01 = .52)
82. Legal or accreditation requirements -.002 .216 .700 .558
S 1. Government policies or mandates .000 .174 .646 .496
Percent of Common Variance 17.5% 13.1% 7.1%

 

59

Table 5
Reported Frequency of Agency Rewards

 

 

 

%
Reporting
Available Reward
96.1 Doing work you believe in
93.6 Important duties and responsibilities
93.4 Positive responses from clients
91.7 Compatible group of co-workers
91.5 Freedom concerning working hours
88.3 Increases in salary
86.9 Pay levels like people with similar jobs in other agencies

86.5 Agency paid life insurance
86.1 Agency paid health insurance

85.3 Agency paid retirement

85.3 Inﬂuence in setting goals and making decisions
84.9 Job security

83.2 Freedom concerning job duties

81.5 Formal commendations and rewards

 

73.8 Longevity pay
73.0 Reassigning staff instead of layoffs

 

66.7 Bonuses for performing quality work

 

56.7 Bonuses for meeting budgets and schedules

 

An exploratory factor analysis was done with these items as well (see Table
6). Four interpretable scales were identiﬁed, two of which were used to test
hypotheses. The summary four-factor solution that explained approximately 54.5% of
the variance.

The reward item “freedom concerning job duties” was not a professional
reward, as expected, but loaded onto an Other Rewards subscale that included such
diverse items as paid expenses to present conference papers and job security. The

factor solutions for the other two scales can be found in Appendix C.

60

Table 6
Results of Factor Analysis for Work Reward Items

 

Factor Loadirygs Communality

 

 

 

 

 

 

Items 1 2 Estimates

Factor 1: Pay and Other Financial Rewards; (,1: .76
R10.Agency paid retirement beneﬁts .759 .1 12 .637
R8. Agency paid life insurance .749 .151 .602
R9. Agency paid health insurance .746 .213 .633
R4. Longevity pay .621 -.005 .458
R1. Increases in salary .465 .308 .388
R5. Pay levels like people with similar jobs in

other agencies .451 .288 .415
Factor 2: Professional Rewards; a = .76
R11.Doing work you believe in .273 .719 .648
R17.Positive responses from clients .128 .714 .544
R12.Important duties and responsibilities .270 .681 .612
R16.A compatible group of co-workers .168 .673 .546
R13.Inﬂuence in setting goals and making

decisions . 140 .624 .583
R14.Freedom concerning working hours -.197 .469 .491
Percent of common variance 16.4 16.0

 

Relationship B

The two individual context variables were sector and position in the organization.
Sector was measured with a single item that asked people to check whether they worked
in the public, nonproﬁt or for-proﬁt sector.

Position in the agency was measured with a single item that asked respondents to
select the “one description that was most like your position” from ﬁve descriptions plus
an option to write in an alternative. The distribution of these responses was presented in

Table 2.

61

Relationship C

Sector and industry were assessed with single items that asked respondents to
check the one best answer. Sector was measured with a single item that asked people to
check one of three choices.

Among the options that were offered for industry were: child welfare, family
services, health, mental health, occupational social work/BAP, substance abuse and other.
The industry list was an expanded version of that provided by the Michigan Chapter-
NASW to sort their membership. The distribution of the three primary industries—child

welfare, family services and mental health—by sector is shown in Table 7.

 

 

 

 

 

 

Table 7
Distribution of the Respondents by Sector and Industry
Sector
Industry Nonproﬁt Public F or-Proﬁt Total

Child Welfare (N) 45 29 1 75
% within industry 60.0% 38.7% 1.3% 100%
% within sector 20.3% 17.2% 2.3% 17.3%
% of total 10.4% 6.7% .2% 17.3%
Family Services (N) 52 19 l 72
% within industry 72.2% 26.4% 1.4% 100%
% within sector 23.4% 11.2% 2.3% 16.6%
% of total 12.0% 4.4% .2% 16.6%
Mental Health and Related Services

(N) 125 121 41 287
% within industry 43.6% 42.2% 14.3% 100%
% within sector 56.3% 71.6% 95.3% 66.1%
% of total 28.6% 27.9% 9.4% 66.1%
Total (N) 222 169 43 434
% within industry 51 .2% 38.9% 9.9% 100%
% within sector 100% 100% 100% 100%
% of total 51.2% 38.9% 9.9% 100%

 

62

As can be seen in Table 7, among these respondents child welfare and family
service organizations are ahnost exclusively found in the nonproﬁt and public sectors;

only mental health agencies are reported to occur in all three sectors.

Data Analysis Procedures

The model was tested in four steps: one for each component and then the model as
a whole. Multiple regression analysis was used. This has been the analytic approach of
choice among most researchers doing multivariate analysis of organizational ﬁt (Chao, et.
al, 1994; Chatrnan, 1991; Cable & Judge, 1994; Meglino etal., 1989). Hierarchical
regression analysis has been the method of choice when performing exploratory analysis
of models as is the case here (James & Brett, 1984).

When multiple regression analysis is used, the existence of substantial correlation

among the independent variables creates a situation referred to as multicollinearity. For
this exploratory study, where the goal is to investigate and interpret the individual
contributions of these highly related factors, the aspect of multicollinearity that was
potentially the most troubling is the fact that it is difﬁcult for these independent variables
to make unique contributions to the dependent variables (Cohen & Cohen, 1983).
J udicious use of hierarchical regression has been suggested in these instances. In these
data there is substantial correlation (r = .578) between professional rewards and pay and
other ﬁnancial rewards and also among the client service decision factors (correlations of
.448 - .484). The treatment of these relationships is discussed in Chapter 4 Results.

In all regression equations where sector, industry and position were predictors,

dummy effects coding was used. Nonproﬁt was the omitted sector value, family service

63

was the omitted industry value, and direct service was the omitted position value. Where
dummy effects coding is used for predictor variables, the constant has speciﬁc meaning
(Cohen & Cohen, 1983). It represents the reference group and is the sample mean for that
group.

The speciﬁc analytic approaches for each component of the model are as follows.
All predicted relationships for A — Work Context Predicts Work Practices and B — Person
Context Predicts Individual Values were tested using regression analysis. In several
instances, the results were elaborated using t-tests.

To explore Relationship C — Individual Values and Work Practices as Predictors
of Dependent Variables, four regression analyses were done, one each predicting
organizational commitment, job involvement, intent to leave and tenure. Fit was
operationalized as the interaction term between the individual work value subscale and
the corresponding work practice.

The entire model was tested with two sets of hierarchical regression equations. A
series of hierarchical regression analyses were carried out with either work practices or
work values entered as the ﬁrst step predicting the dependent variables. Work or person
context variables were entered as the second step in the appropriate equation. This
resulted in eight hierarchical regression analyses. In Chapter 4, the results of these

analyses are presented.

64

CHAPTER 4
RESULTS

Introduction

In this chapter, the results of the data analyses are presented. Again, the model is
used to organize the discussion. The hypothesis testing presented here begins with those
pertaining to Relationship C and ends with those pertaining to Relationship A. This is
done to build the multiple regression equations that test the entire model. First those
results related to the effect of context on work structures are presented; next the results
related to the effects of context on work values are presented; and ﬁnally, the results
related to the relationship between work values and work structures and their impact on

work outcomes. The chapter concludes with a summary of results.

Relationship C:
Work Context Predicts Work Practices

The general hypothesis was that sector would predict the factors emphasized in
organizations’ practices, and two practices were speciﬁed: organizational reward systems
and client service decision systems. The three hypotheses are presented and discussed
together.

H 1 a: In the for-proﬁt sector, the most extensive package of ﬁnancial rewards are
available to employees and cost factors are rated as being the most inﬂuential

(used most often) in client service decisions.

H 1 b: In the public sector, policies and rules are rated as the most inﬂuential factor in

client service decisions.

65

ch: In the nonproﬁt sector, the most extensive package of client/professional rewards
are available and client/professional factors are rated as the most inﬂuential in
client service decisions.
Recall from Chapter 3 that the factor analyses of both the agency reward items
and the client service decision factors yielded interpretable factors. The reward
components of the hypothese are presented ﬁrst, followed by the client service decision

factor components of the hypotheses.

Sector Inﬂuences on Reward Practices

Financial Rewards

Hypothesis Hla states that the most extensive package of ﬁnancial rewards are
available to employees in the for-proﬁt sector. In the last chapter, the frequency table for
available rewards suggested that the rated rewards were generally available. The multiple
regression equation tested the extent to which there were differences in the availability of
rewards based on the sector of employment. Table 8 shows the results for the regression

equation predicting availability of ﬁnancial rewards by sector.

 

 

Table 8
Regression Results Predicting the Availability of Financial Rewards ﬁom Sector
Predictor B Beta Sig.

Constant 8.028 .000
Public sector -1 .373 -.105 .029
Private for-proﬁt sector .130 .006 .n.s.

R .106

R2 .011

~R2 .007

 

66

 

113

17?

11011

“'61

SCCI

 

The regression equation was not signiﬁcant (F = 2.449, sig. = .088). No difference
was found in ﬁnancial reward practices across the three sectors; therefore, Hypothesis 1a
was not supported. In the balance of this chapter, when regression equations are not
signiﬁcant, only the F value and signiﬁcance level will be reported in the text; the

equations can be found in Appendix D as noted.

Professional Rewards

Six items that represented professional (“doing work you believe in”) or affiliation
(“a compatible group of co-workers”) rewards formed the professional rewards scale.
Hypothesis 1c predicts that professional rewards will be more available in the nonproﬁt
sector. Because this is the omitted sector, predictions relating to the sector can only be
tested by entering the other sectors in the regression equation. Table 9 shows the

regression equation predicting professional rewards from sector.

 

 

Table 9
Regression Results Predicting the Availability of Professional Rewards from Sector
Predictor B Beta Sig.

Constant 9.840 .000
Private for-proﬁt sector 1.142 .098 .040
Public sector - .998 -.053 n.s.

R .115

R2 .013

~112 .009

 

This equation approaches signiﬁcance (F = 2.927, sig. = .055). Employees in the
nonproﬁt sector, represented by the constant, reported that these professional rewards
were available to them. This was true at a greater extent for employees in the public

sector, but the lesser availability (represented by the negative value) in the for-proﬁt

67

01‘

d1»

CIT.

15’s
'55

+.
117.1;

1113;

111113

 

 

sector was not signiﬁcant. While most people agreed that these professional rewards were
available to them, a review of the data revealed that employees in the for-proﬁt sector
were twice as likely as employees in the nonproﬁt sector to report that three of these
rewards were not available in their organizations. Those less available rewards were
doing work you believe in, important duties and responsibilities, and a compatible group
of co-workers (for-proﬁt > nonproﬁt = 12% > 5%, 16% > 7%, and 25% > 12%,
respectively).

These regression results suggest that client/professional rewards are available in
the nonproﬁt sector, but this set of rewards was reported to be most available to

employees in the public sector. Therefore ch was not supported.

Sector Inﬂuences on Client Service Decision Practices

The factor analysis of the client service decision factors scale resulted in ﬁve
interpretable factors, three of which were used for these analyses: professional attributes,
ﬁnancial practices and legal requirements. These three scales relate well to the predictions
made for client service decision-making factors. Hypothesis 1a predicted that ﬁnancial
factors are incorporated into client services decisions by employees in the for-proﬁt
sector. That hypothesis was not supported (F = .001, sig. = .974). Hypothesis 1b predicted
that employees in the public sector would use legal requirements to make their client
services decisions. That hypothesis was not supported (F = 2.025, sig. = .155). Finally,
based on the nonproﬁt literature, it was predicted that employees in that sector would be

most likely to use professional factors to make their client service decisions. That

68

 

:1
It.
.

Sl.’

If?

the-

 

 

 

hypothesis was not supported (F = .172, sig. = .842). See Appendix D for complete
regression results.

In summary, sector was not an inﬂuential factor predicting organizational
structures, either in the form of organizational reward practices or client service decision-
making practices. Among the predictions made, only the hypothesis that professional
rewards would be available in the non-proﬁt sector was supported, and in this instance

these rewards were reported to be more available by employees in the public sector.

Industry Inﬂuences on Work Practices
Within the nonproﬁt sector only, industry inﬂuences are predicted on work
practices. Recall that the rationale for these predictions is that nonproﬁt organizations that
interact extensively with organizations in the other two sectors will of necessity take on
attributes of these sectors. The hypotheses are as follows:
H2a: In family services, client/professional factors are given the most emphasis in
making client services decisions.
H2b: In child welfare agencies, policies and rules factors are given the most emphasis in
making client services decisions.
H2c: In mental health agencies, cost factors are given the most emphasis in making
client services decisions.
Since none of the similar sector hypotheses were supported, it was unlikely that
the within-nonproﬁt sector predictions would be supported. And that was the case. None

of these hypotheses were supported.

69

Nonproﬁt family services employees did not report the greatest use of
professional factors in making client services decisions (F = .172, sig. = .842;
Appendix D).

Nonproﬁt child welfare employees did not report more emphasis on legal
requirements for making client service decisions (F = 1.369, sig. = .243;
Appendix D).

Nonproﬁt mental health employees did not report more emphasis on ﬁnancial
factors in their client services decisions (F = 1.460, sig. = .228; Appendix D).

Overall, neither sector nor industry within the nonproﬁt sector were predictors of

these organizational structures.

Relationship B:
Person Context Predicts Individual Values

The general hypothesis was that sector predicts individuals’ work values. Three

related hypotheses were developed. The three hypotheses are presented and discussed

together.

H3a:

H3b:

H3c:

Employees in the nonproﬁt sector rate altruism more positively (higher) than
those in the public sector, and both higher than those in the for-proﬁt sector.
Employees in the for-proﬁt sector rate economic rewards more positively (higher)
than those in the public, and both higher than those in the nonproﬁt sector.
Employees in the nonproﬁt sector rate autonomy higher than those in either the

public or the for-proﬁt sectors.

Table 10 gives the means for the altruism, economic rewards and autonomy subscales by

SCCtOl’.

70

V2;
SC:
01 -

1
all:

 

 

Mean Values for Altruism, Autonomy and Economic

Table 10

 

 

 

 

 

 

Rewards by Sector
Nonproﬁt Public For-Proﬁt
Sector Sector Sector
Altruism Subscale
Mean 13.95 14.14 14.33
N 220 169 42
Standard Deviation 1.38 1.33 1.05
(F = 1.89, n.s.)
Economic Rewards Subscale
Mean 11.88 12.31 12.54
N 217 163 41
Standard Deviation 2.17 2.10 1.85
(F = 2.87, n.s. [.058])
Autonomy Subscale
Mean 12.32 12.36 13.10
N 219 169 40
Standard Deviation 1.98 3.03 1.78
(F = 1.79, n.s.)

 

Although the values are higher for public and for-proﬁt sector employees than for

None of the hypotheses as speciﬁed were supported.

those employed in the nonproﬁt sector, the differences in values were not signiﬁcant.

The model indicates that sector will predict work values. Although the analysis

71

above suggests that no relationship will be found, a series of multiple regression analyses
were done on this component of the model. Each value set was entered as the dependent

variable in a regression equation predicted by sector. Working in the private for-proﬁt
sector did not predict one’s value for altruism (F = .923, n.s.), autonomy (F = 1.125, n.s.)
or economic rewards (F = .469, n.s.). See Appendix D for these equations. Further,
altruistic values were not predicted by sector for any of these employees (F = 1.156, n.s.;

Appendix D). However, employment in both the nonproﬁt and public sectors were

18$

00-
‘l

 

Sta?

 

predictors of these respondents’ values for autonomy and economic rewards. Those

results are shown below in Tables 11 and 12.

Table 11

Regression Results Predicting the Work Value Autonomy from Sector

 

 

 

Predictor B Beta Sig.
Constant 12.507 .000
Public sector -.692 -.094 .051

R .094
R2 .009
~R2 .007
F = 3.827, p 5 .05
Table 12

Regression Results Predicting the Work Value Economic Rewards from Sector

 

 

Predictor B Beta Sig.
Constant 12.255 .000
Public sector -1.159 -. 179 .000

R . 179
R2 .032
~12.2 .030

 

F=14.021,p5.000

In both of these instances, while sector does predict these respondents’ values,

only a small amount of the variance is explained in either equation.

Alternatively, the model may be inappropriately speciﬁed. While it is reasonable

to believe that the context of organizations will predict the practices of those

organizations, such a rationale may not equally apply to the direction of the relationship

between an individual’s work values and their context. Work values are thought to be

stable, enduring characteristics of individuals. Ifthat is the case, then the values of

72

 

 

P:
pl L

.I:
'4)

:1
[/7

C077

 

individuals should predict their choice of sector. This alternate model was tested with

regression analysis. The results are shown in Table 13.

 

 

Table 13
Regression Results Predicting Sector from Work Values
Predictor B Beta Sig.

Constant .438 n.s.
Altruism .0045 .092 n.s.
Autonomy .0010 .030 n.s.
Economic Rewards .0032 .102 .043

R .143

R2 .020

~R’ .013

 

F = 2.883, p g .05

The value placed on economic rewards did predict sector choice. However, only

about 1% of the variance was explained by work values. In this instance, neither of these

empirical tests strongly suggest the preferred direction for the model variables. Therefore

the hypotheses were tested as originally speciﬁed.

An additional hypothesis regarding the inﬂuence of position on values was

presented:

H3d: Employees in direct service positions rate autonomy more highly than those in

supervisory positions.

The regression equation testing this hypothesis was not signiﬁcant (F = 2.423,

n.s.; Appendix D). In this instance, the low reliability of the autonomy subscale may have

contributed to a lack of support.

73

Relationship A:
Individual Values and Work Practices as Predictors of the Dependent Variables

The hypothesis for this set of relationships is as follows:

H4: The ﬁt between an individual’s work values and the work practices that reﬂect
those values positively predicts one’s organizational commitment, job
involvement, intent to remain with the organization and tenure.

This hypothesis was tested with a series of regression equations. Fit between
individual work values and work practices was operationalized as the interaction term
between matching pairs of independent variables.

The same set of interactions between the work values and work practices variables
were used to predict each of the dependent variables. The set of interactions between
work values and reward practices were:

0 The interaction between the work value altruism and professional rewards;

o The interaction between the work value autonomy and professional rewards; and

o The interaction between the work value economic rewards and the ﬁnancial
rewards.

Both altruism and autonomy were associated with professional rewards, since that
scale included aspects of service to others as well as components of independent practice.
The set of interactions between work values and client service decision factors were:

0 The interaction between the work value altruism and the ﬁnancial client service
decision factors;
0 The interaction between the work value altruism and the legal client service

decision practices;

74

5C.“

561-

o The interaction between the work value autonomy and professional client service
decision practices; and

o The interaction between autonomy and the ﬁnancial client service decision
practices.

Recall that the professional client service decision practices are based on staff-
related attributes, as is the work value for autonomy. The work value altruism is paired
with both ﬁnancial and legal practices, because these practices would be restrict
employees’ desire to provide service to others. Similarly, one’s preference for making an
autonomous decision on behalf of their clients would be restricted by their organization’s

emphasis on considering ﬁnancial matters in making those decisions.

Values and Wopkjﬂctices a_s Predictors of Orgapizational Commitment

The results of the regression analysis for the interaction between work values and
work practices predicting organizational commitment are shown in Table 14.

The data indicate that organizational commitment can be predicted from the
interaction between the work value Autonomy and professional client service decision
practices as well as from the interaction between the work value Altruism and legal client
service decision practices.

An examination of the correlation table for this equation was done because three
sets of interactions share a common component. Table 15 shows the correlations among

these variables.

75

Table 14
Regression Results for Organizational Commitment Predicted from the Interactions
between Work Values and Work Practices

 

 

Predictor B Beta Sig.
Constant 39.532 .000
WV Altruism X Professional .0029 .138 n.s.

Rewards
WV Autonomy X -.00049 -.023 n.s.

Professional Rewards
WV Economic Rewards X .00107 .049 n.s.

Financial Rewards
WV Altruism X CS .00089 .107 n.s.
Financial Practices
WV Altruism X CS Legal .0049 .137 .013
Practices
WV Autonomy X CS .0040 .304 .000
Professional Practices
WV Autonomy X CS —.0025 -.320 n.s.
Financial Practices
R .346
R2 .120
~R2 .104

 

F = 7.506, sig. = .000

Two sets of the variables are highly related. These are the interaction between:
0 Altruism and professional rewards with autonomy and professional rewards (r =
.904), and
o Altruism and ﬁnancial practices with autonomy and ﬁnancial practices (r = .760).
This high degree of shared variance prompted a reconsideration of the interactions
used to predict the dependent variables. Based on the item content of the professional
reward scales, the interaction between autonomy and professional rewards was retained
since these two included more comparable items than the altruism and professional

rewards scale did. No such clear rationale was appropriate to choose between the two

76

25. H a:
8.
ma. H a

II
a
it

 

arame.

aaammm.

81.08..

35..

§**C_N.

to:

owe:

86.52am
BUSES x b55224 .w

 

aaavom.

«eawom.

NNor

iaquN.

amZ.

naawaﬁ

x >anoS< . b

 

1:...va

moor

435..

over

*aov_.

:83 a 86:52 .6

 

ammo:

moo.

one.

08.

mouuoﬁm
3985 x 88.52 .m

 

suing.

renown.

Loo.

x 3333* owe—oncom— .v

 

aaavoa.

3&2.

x >80ao§< .m

 

tit-T

weaken

35683.5 x 5.6.152 .N

 

“noggoo
8:88.835 .2

moouoﬂm anﬁmouoi

80:an

 

merged 36:85

6,6836% Enoawowoi

 

 

 

 

 

 

m

 

v

 

m

 

N

 

 

 

 

mug—nut”; nommmohwom wcoﬁ< noun—280
m _ 2an

77

work values for an interaction with ﬁnancial practices, so both interactions were retained.
Table 16 shows the resulting equation.
Table 16

Regression Results for Organizational Commitment Predicted from the Interactions
between Work Values and Work Practices (Excluding Altruism x Profession

 

 

Rewards)

Predictor B Beta Sig.
Constant 39.704 .000
WV Autonomy x .0026 .122 .05

Professional Rewards
WV Economic Rewards x .00095 .044 n.s.
Financial Rewards
WV Altruism x CS Legal .00498 .140 .01
Practices
WV Altruism x CS .00152 .181 .05
Financial Practices
WV Autonomy x CS .00406 .306 .000
Professional Practices
WV Autonomy x CS -.00322 -.411 .000
Financial Practices
R .346
R2 .119
~R2 .106

 

F = 8.764, sig. = .000

In this second analysis, all interactions except one are signiﬁcant. The number of
independent variables has been reduced by one, however there is no difference in the
amount of variance explained by the two equations.

The hypothesis that the ﬁt between these three work values and the ﬁve work

practices that reﬂect those values predict organizational commitment was supported.

78

Job Involvement
The same set of six interactions were regressed on job involvement. The results
are shown in Table 17 below.
Table 17

Regression Results for Job Involvement Predicted from the Interactions between
Work Values and Work Practices

 

 

Predictor B Beta Sig.
Constant 22. 144 .000
WV Autonomy x Professional .000644 .055 n.s.

Rewards
WV Economic Rewards x .000514 .043 n.s.
Financial Rewards
WV Altruism x CS Legal .000372 .019 n.s.
Practices
WV Altruism x CS Financial .000868 .189 .05
Practices
WV Autonomy x CS .00110 .154 .01
Professional Practices
WV Autonomy x CS Financial ‘ -.00105 -.247 .01
Practices
R .181
R2 .033
~R2 .018

 

F = 2.212; p 5 .05

The hypothesis that job involvement is predicted from the ﬁt between three of
these work value — work practice interactions is supported. Note that again the autonomy

x ﬁnancial practices interaction is negative.

Intent to Leave
The six interactions were regressed on intent to leave. The results are shown in

Table 18 below.

79

Table 18
Regression Results for Intent to Leave Predicted from the Interactions between Work

 

 

Values and Work Practices
Predictor B Beta Sig.
Constant 6.763 .000
WV Autonomy x Professional -.000875 -.184 .01
Rewards
WV Economic Rewards x .000079 .016 n.s.
Financial Rewards
WV Altruism x CS Legal .00124 .153 .005
Practices
WV Altruism x CS Financial -.00059 -.317 .000
Practices
WV Autonomy x CS -.00103 -.352 .000
Professional Practices
WV Autonomy x CS .00097 .554 .000
Financial Practices
R .409
R2 .167
~112 .153

 

F = 12.172; p 5 .000

The hypothesis that intent to leave is predicted ﬁ'om ﬁt is supported. Not
surprisingly, the ﬁt between autonomy and professional rewards and also autonomy and
professional client service decision-making practices are negatively related to intent to

leave.

Tenure
When the six interactions were regressed on the variable years on the job, the
equation was not signiﬁcant (F = 1.327, n.s.; Appendix D). The hypothesis that work

value — work practice ﬁt predicts years on the job was not supported.

80

Testing the Model

The concluding steps in the analyses were to test the two paths of the full model
by including context variables as the second level in hierarchical regression equations. In
each instance, if this was a completely mediated model, when context was entered as the
second level, the R2 values should fall to zero (James & Brett, 1984). If, however, context
is a moderator, as suggested by several nonproﬁt authors (c.f., Young, 1984), then
different levels of the dependent variables should be found. The context variable applied
to both paths is sector, operationalized as public or for-proﬁt. A second work context
variable, industry, was not entered in the equations; the industry hypotheses were related
solely to the nonproﬁt sector and were found to have no effects. Position was the second
context variable added to the work values path. The work practices (Relationship A) and

individual work values (Relationship B) segments are presented separately.

Work Contexgnd Work Practices

Organizational Commitment

The regression results for organizational commitment predicted from work
context, through work practices, are shown in Table 19. In each model, the work
practices—reward practices and client service decision-making practices—were entered
ﬁrst, followed by sector. Separate equations were run for public and nonproﬁt sectors
although the results are reported together in the table.

The results in Table 19 indicate that the availability of desired professional

rewards, as well as all the client services decision-making practices predict organizational

81

commitment for these respondents. Further, all three sectors of employment are
signiﬁcantly related to organizational commitment. Recall that when using dummy coded
variables, the negative sign refers to a lesser amount than the omitted variable, not a
negative relationship. Therefore, working in the public sector contributes less to one’s
level of organizational commitment than working in the nonproﬁt sector, and working in
the for-proﬁt sector contributes the least to one’s organizational commitment. Sector is
clearly a moderating variable, since it contributes different levels of commitment but is
not completely contained in work practices, as would be the case if work practices were

mediating variables.

 

 

 

Table 19
Regression Results Predicting Organizational Commitment from Work Context
through Work Practices
Predictor Beta R R2 F AR2 F for AR2
Step 1 — Work Practices
Professional Rewards .161 **
Financial Rewards n.s.
CS Professional 226*“
Practices
CS Financial Practices .073"
CS Legal Practices .117* .300 .090 7.862***
Constant (Nonproﬁt [***]
Sector) [B]
--.S_tt=n_2_-.-_.1?9.bli<.=-.8_<?9.t9! ........ 2.1.2.6311“.---_-.3_2§-_-.._1.9_5.---_.7.-.7..5.5_i".‘!‘i".---.-9.!.5_ ....... 9.9.6.231-
Step 2 — Private For- -.164*** .342 .117 8.743*** .027 12.055***
Proﬁt Sector
* p 5 .05
** p 5 .01
*** p 5 .001

82

Job Involvement

The same analyses were done to predict job involvement from work context
transmitted through work practices. None of the equations predicting job involvement
from work practices and work context were signiﬁcant (public sector full equation, F =
1.846, sig. = .089; private for-proﬁt sector, F = 1.361, sig. = .229). For these respondents,
there was no relationship between their work practices, work context and job involvement

(see Appendix D).

Intent to Leave
Work practices and work context were entered in two steps into equations

predicting intent to leave. The results of these equations are summarized in Table 20.

 

 

 

 

Table 20
Regression Results Predicting Intent to Leave from Work Practices and Work Context
F for
Predictor Beta R R2 F AR2 AR2
Step 1 — Work Practices
Professional Rewards -1.964*
Financial Rewards n.s.
CS Professional - .275***
Practices
CS Financial Practices .026*
CS Legal Practices .154" .379 .143 12.565***
Step 2 - Constant [n.s.]
(Nonproﬁt Sector)
[13]
_thliq-$99.tp.r.--_---_------_------_-----.I1..§-.-----.3.7..9 ..... 114.4 .......... 19:49! ..... 199.9 ........ 2924-.
Private For-proﬁt n.s. .379 .144 10.474 .000 .159
Sector
* p 5 .05
** p _<_ .01
*"'* p 5 .001

83

 

The results in Table 20 indicate that sector does not predict one’s intent to leave.

However, the emphasis on using ﬁnancial and legal information to make client service

decisions does predict intent to leave, as does the lack of desired professional rewards.

Tenure
The ﬁnal set of analyses were done to determine the relationship of work practices

 

and work context on respondents’ length of time in their organizations. The summary of

these analyses are presented in Table 21.

 

 

 

Table 21
Regression Results Predicting Tenure from Work Practices and Work Context
F for
Predictor Beta R R2 F AR2 AR2
Step 1 — Work Practices
Professional Rewards -.178**
Financial Rewards 300*"
CS Professional Practices n.s.
CS Financial Practices n.s.
CS Legal Practices n.s. .243 .059 5.134***
Constant (Nonproﬁt [**]
Sector) [B]
"-3192. :.1.’.1.1b.1.i_c_ _Ssstqr. ................. n.9, ...... 1257.---.991 ..... 4 :4191‘1'3‘._--:99.2. .......... 13 :§:-.
.136" .278 .077 5.709*** .018 8.136”

Step 2 — Private For-
proﬁt Sector
* p 5 .05
*"' p 5 .01
*** p 5 .001

 

The data in Table 21 indicate that the availability of rewards, but not the way
client service decisions are made, predicts longevity in organizations. The availability of

ﬁnancial rewards positively predicts tenure and the lack of professional rewards

84

 

 

negatively predicts tenure. Work context impact is limited to the for-proﬁt sector; for
employees in both the public and the nonproﬁt sectors this context variable is not

signiﬁcant.

Individual Context and Work Values

Organizational Commitment

Independent variables were entered in three steps to test the predictive impact of
individual context and work values on organizational commitment. In the ﬁrst step, the
three focal work values were entered: altruism, autonomy and economic rewards. In the

second step, the dummy variable for position was entered, and in the third step the
dummy variables for context were entered. Direct service was the omitted position
variable and nonproﬁt sector was the omitted context variable. Position was treated as the
more proximate context variable and sector as the more distal context variable for
individuals. The results for these analyses are shown in Table 22.
The data suggest that work values in and of themselves do not predict
organizational commitment. The introduction of context, in the form of position at step
two changes the results: both direct service (the constant) and also administrative
positions predict organizational commitment for these respondents. When the sector
variable is added, again the constant, now representing both direct service and nonproﬁt
sector and also the for-proﬁt sector variable predict organizational commitment. In the
case of the nonproﬁt sector, this is a positive relationship while for those in the for-proﬁt

sector, this is a negative relationship.

85

Table 22
Regression Results Predicting Organizational Commitment from Work Values and

 

 

 

 

 

Work Context
F for
Predictor Beta R R2 F AR2 AR2
Step 1 — Work Values
Altruism .1 19*
Autonomy n.s.
Economic Rewards n.s. .118 .014 1.927
Constant (Direct Service ***
Position) [B]
Step 2 — Position/ .194 .227 .052 5.531*** .038 16.130***
Administrative
Step 3 — Constant (Non- ***
proﬁt Sector) [B]
--.S.t§p.§:.P.vhli.C.-.S§9.tgr. ................ I} a--__-.2_4_2--.--.Q.5.8_---.5.-9.3.5.i"_‘f."i---.-99.7. .......... 2.9.4.7.-.
Step 3 — Private For- -.177*** .288 .083 7.317*** .031 13.768***
Proﬁt Sector
* p 5 .05
** p _<_ .01
*** p 5 .001

Job Involvement

The same set of variables, in the same three steps, were regressed on job
involvement. The results are shown in Table 23.

The data show that the work value altruism is a signiﬁcant predictor for job
involvement. However, the equation is improved by the addition of the position context
variable. Both direct service and administrative positions contribute to the variance
explained in the expanded equation at Step 2. The addition of sector, while a minimal

improvement, is signiﬁcant only for the nonproﬁt sector (the constant).

86

 

Table 23
Regression Results Predicting Job Involvement ﬁ'om Work Values and Work Context

 

 

 

 

 

Predictor Beta R R2 F AR2 F for AR2
Step 1 — Work Values
Altruism .201 ***
Autonomy n.s.
Economic Rewards n.s. .207 .043 6098*“
Constant (Direct Service **
Position) [B]
Step 2 - Position/ Administrative .164*** .264 .070 7.640*** .027 11.784***
Step 3 — Constant (Non-Proﬁt **
Sector) [B]
- -8161)- 3. .—. P 99119 .8599}: ....................... 1.1.15.2 - - :.2.7.5. _ - ”0.7.5- - - .6.:6.6..2.* *3“- - - :O.06. ............ 9.3:- -
Stg) 3 - Private For-Proﬁt Sector n.s. .269 .072 6348*" .003 n.s.
* p _<_ .05
** p 5 .01
*** p g .001

Intent to Leave

The same set of predictors were regressed on intent to leave. The results of these
equations are shown in Table 24.

Only the work value related to preference for economic rewards is a signiﬁcant
predictor on these respondents’ intent to leave their jobs; neither the work values related
to altruism or autonomy have a signiﬁcant impact on intent to leave. The addition of the
context variables adds nothing to the variance explained by the equations; neither the

added position nor sector variables have an impact on intent to leave.

87

Table 24
Regression Results Predicting Intent to Leave from Work Values and Work Context

 

 

 

 

 

F for
Predictor Beta R R2 F AR2 AR2
Step 1 — Work Values
Altruism n.s.
Autonomy n.s.
Economic Rewards 229*" .259 .067 9302*"
Constant (Direct Service [**]

Position)

Step 2 — Position/ n.s. .275 .076 7.886"‘M .008 n.s.

Administrative
Step 3 — Constant (Non-Profit [**]

Sector)
"$35323.:_1?}1b.1.i9.§29t9.r..-----_----_-___-_-__n-.S.-._---.2_7.5.----.9_7.§.-_.6.;2.9..6.".‘i".'!‘----999 ....... 11.-.8..-
Step 3 - Private For-Proﬁt n.s. .275 .076 6293*" .000 n.s.

Sector

* p 5 .05

** p 5 .01
*** p _<_ .001

Tenure

Finally, the same set of variables were regressed on the number of years that
respondents reported being employed in all positions by their organizations. The results
of these analyses are shown in Table 25.

The data show that preferences for economic rewards, and to a lesser extent for
altruism, are negatively related to tenure in one’s organization. Holding an administrative
position, however, is related to the number of years in one organization. Surprisingly, for-

proﬁt and nonproﬁt, but not public sector employment, are related to tenure.

88

 

 

Table 25
Regression Results Predicting Tenure from Work Values and Work Context

 

 

 

 

 

 

 

Predictor Beta R R7 F AR2 F for A]?
Step 1 — Work Values
Altruism -.096*
Autonomy n.s.
Economic Rewards -.141** .161 .026 3.682M
Constant (Direct Service [***]
Position)
Step 2 — Position/ 248*“ .295 .087 9904*“ .061 27.859***
Administrative
Step 3 — Constant (Non- [***]
Proﬁt Sector)
Step 3 — Public Sector n.s. .306 .094 8554*" .006 n.s.
Step 3 — Private For- .152*** .332 .110 10.230*** .023 10.614***
Proﬁt Sector
* p g .05
** p 5 .01
*** p 5 .001

Summary
Two sets of analyses were done. The ﬁrst were analyses to test the study

hypotheses. The second were analyses to test the model. The results are presented in that

order.

Hypothesis Testing

The hypothesis testing produced mixed results. The following is a discussion

 

organized by hypothesis, augmented by several summary tables.

89

 

Work Practice Hypotheses

H1 3: In the for—proﬁt sector, the most extensive package of ﬁnancial rewards are

available to employees and cost factors are rated as being most oﬁen used
in client service decisions.

This hypothesis was not supported. First, there was no differential package of
ﬁnancial rewards among the sectors; the ﬁnancial rewards scale included a standard
package of pay, beneﬁt and ﬁnancial rewards. Second, the equation predicting availability
of ﬁnancial rewards from sector was not signiﬁcant. Finally, sector was not a predictor of

employees’ reported use of ﬁnancial factors in making client service decisions.

Hlb: In the public sector, policies and rules are rated as being most often used in

client service decisions.

This hypothesis was not supported. The equation predicting the use of legal
requirements for client service decisions was not signiﬁcant.

ch: In the nonproﬁt sector, the most extensive package of professional rewards

are available and professional factors are rated as the most often used in
client service decisions.
This hypothesis was partially supported. Analyses indicated that professional
rewards were available in the nonproﬁt sector, but these rewards were also reported to be

more predicted by public sector employment. Finally, the multiple regression equation

predicting use of professional attributes to make client services decisions was not

signiﬁcant.

These results are shown in Table 26.

90

 

Table 26
Predicted vs. Actual Sector — Work Practice Relationships

 

 

 

Sector
Work Practices Nonproﬁt Public For-Proﬁt

Client Service Decision-Making Practices

Professional Attributes 1/n.s. x/n.s. x/n.s.
Financial Considerations x/n.s. x/n.s. 1/n.s.
Legal Mandates x/n.s. l/n.s. x/n.s.
Reward Practices

Professional Rewards 1/2 x/ l x/n.s.
Financial Rewards x/n.s. x/n.s. l/n.s.

 

x = No predicted order

n.s. = Not signiﬁcant

H2a: In family services, professional practices are used most often to make
client service decisions.

H2b: In child welfare agencies, policies and rules are used most often to make
client service decisions.

H2c: In mental health agencies, ﬁnancial practices are used most often to make
client service decisions.

None of these hypotheses were supported. The regression equations predicting

professional practices, policies and rules, and ﬁnancial practices from industry were not

signiﬁcant.

Work Value Hypotheses
H3a: Employees in the nonproﬁt sector rate altruism more positively (higher) than

those in the public sector, and both higher than those in the private for-proﬁt

sector.

91

This hypothesis was not supported. There were no differences across sectors in the

mean value for the altruism scale. In addition, the regression equation predicting altruism

from sector was not signiﬁcant.

H3b: Employees in the for-proﬁt sector rate the work value for economic

rewards more positively than those in the public sector, and both higher

than those in the nonproﬁt sector.

This was not supported. In fact, employees in the nonproﬁt sector had the highest
economic rewards values, followed by those in the public sector. The regression equation

predicting economic work value from nonproﬁt sector was not signiﬁcant.

H3c: Employees in the nonproﬁt sector rate autonomy higher than those in

either the public or the for-proﬁt sectors.
This was supported. Nonproﬁt sector afﬁliation was a stronger predictor of

autonomy than public sector. The equivalent for-proﬁt sector equation was not

signiﬁcant.

H3d: Employees in direct service positions rate autonomy more highly than

those in supervisory or administrative positions.

This hypothesis was not supported. However, the unreliability of the autonomy

scale may have contributed to the null results.
These results are summarized below in Table 27.
H4:

The ﬁt between an individual’s work values and the corresponding work practices

predicts one’s organizational commitment, job involvement, tenure and intent to

remain with the organization.

92

Table 27
Predicted vs. Actual Context — Work Values Relationships

 

 

 

Sector
Work Value Nonproﬁt Public F or-Proﬁt Position
Altruism 1/ 1 2/1 3/1 -—
Autonomy 1/1 2/2 2/n.s. direct service > adm/n.s.
Economic 3/1 2/2 1/n.s. ——

Rewards

 

This hypothesis was tested with a series of regression equations in which the
independent variables were the interactions between work values and work practices. The
results are shown in Table 28.

Table 28

Direction of the Fit Interactions Predicting Organizational Commitment, Job
Involvement, Intent to Leave and Tenure

 

 

Intent
Organizational Job to
Predictors Commitment Involvement Leave Tenure

WV Autonomy x Professional +/+ +/0 -/- +/0
Rewards

WV Economic Rewards x +/0 x/O -/0 +/0
Financial Rewards

WV Altruism x CS Financial -/+ -/0 +/- -/0
Practices

WV Altruism x CS Legal -/+ -/+ +/+ -/0
Practices

WV Autonomy x CS +/+ +/+ -/- +/0
Professional Practices

WV Autonomy x CS Financial -/- -/- +/+ -/0
Practices

 

The seven interactions that were initially tested included

93

o Altruism x professional rewards

o Autonomy x professional rewards

0 Economic reward values x ﬁnancial rewards

o Altruism x the use of ﬁnancial practices to make client service decisions

0 Altruism x the use of legal mandate to make client service decisions

0 Autonomy x the use of professional attributes to make client service decisions
0 Autonomy x the use of ﬁnancial practices to make client service decisions.

The altruism x professional rewards and autonomy x professional rewards
variables overlapped to such a great extent (r = .904) that the altruism interaction was
dropped from subsequent regression equations.

Organizational commitment, job involvement and intent to leave were all
predicted by a set of interactions between work values and work practices. The equation
predicting tenure from the series of interactions was not signiﬁcant. The following table
shows the contribution of the interactions to each of the outcome variables. The judgment
of “better ﬁt” is drawn from the hypotheses presented earlier and is shown ﬁrst in each
cell below.

Among the interaction terms, only the variable for economic work values x

ﬁnancial reward practices was never a signiﬁcant predictor of the dependent variables.

Model Testing
Following the hypothesis testing, the entire model was tested. Two sets of
hierarchical regression equations were analyzed, one for the work practices component of

the model (Relationships C to A) and one for the work values component of the model

94

(Relationship B to A). In these equations, the independent variables were entered and not

the interactions.

Worl_( PracticeL and Context Variables

Table 29 summarizes the predictive contributions of the independent variables to
the four dependent variables. In sector, only those sectors found to be signiﬁcant

predictors are identiﬁed. Recall that industry was not tested because it had no explanatory

value.

Table 29
Direction of the Work Practice and Context Variables Predicting Organizational
Commitment, Job Involvement, Intent to Leave and Tenure

 

 

Organizational Job Intent to

Predictors Commitment Involvement Leave Tenure

Professional + 0 - -
Rewards

Financial 0 0 0 -
Rewards

CS Professional + 0 - 0
Practices

CS Financial + 0 + 0
Practices

CS Legal + 0 + 0
Practices

Sector + (Nonproﬁt > 0 0 + (For-proﬁt >

Public > Nonproﬁt)
For-proﬁ9

 

As the table shows, job involvement was not predicted by any structural or
contextual variables. The emphasis on using ﬁnancial or legal practices to make client

service decisions was positively related to one’s intent to leave the organization, coupled

95

with the lack of professional rewards and inability to use professional practices in making

client service decisions.

Four unexpected ﬁndings are highlighted in the table. First, the availability of

ﬁnancial rewards makes such limited contribution to these outcome variables. Second,
job involvement is predicted by no structural or contextual variables. Third, tenure is
predicted by nonproﬁt and for-proﬁt employment but not public; however, the mean
value for years in the organization is highest for the public sector. Finally, the use of all
practices for client service decision making contributes to organizational commitment; for
this group it was expected that professional practices would be positively related but that

ﬁnancial and legal practices would be negatively related. These results will be discussed

more fully in Chapter 5.

Work Values and Context Va_rgbles

The work values component of the model testing provides an overview of the
value differences predicted to be most reﬂective of nonproﬁt sector employment. Table
30 summarizes the predictive contributions of the work values and individual context
variables. Again, only those values for the context variables of position and sector that

were signiﬁcant predictors are reported.

Both conﬁrmatory and unexpected results are highlighted in the table. First, the

organizational commitment predictors are most similar to assertions made by nonproﬁt
authors: altruism, providing direct services and working in the nonproﬁt sector (also the
for-proﬁt sector) are signiﬁcant predictors. Similarly, the intent to leave relationships

reinforce comments that have been made by nonproﬁt managers who recognize that their

96

low salaries contribute to turnover. The most surprising ﬁnding highlighted in the table is
that the value placed on autonomy makes no contribution. Because of the low reliability

of the scale, however, it is difﬁcult to interpret the extent to which this is a supportable or

non-supportable ﬁnding.

Table 30
Direction of the Work Values and Individual Context Variables Predicting Organizational
Commitment, Job Involvement, Intent to Leave and Tenure

 

 

Organizational Intent to
Predictors Commitment Job Involvement Leave Tenure
Altruism + + 0 -
Autonomy 0 0 0 0
Economic 0 0 + -
Rewards
Position +/ l + + +
(Direct Service) (Administrative > (Direct (Administrative >
Direct Service) Service) Direct Service)
Sector + 0 + +
(Nonproﬁt > (Nonproﬁt) (Nonproﬁt >
For-proﬁt) For-Pro ﬁt)

 

In Chapter 5, the study ﬁndings are reviewed and discussed more extensively. The
implications of those ﬁndings for contributions to theory and applications to practice are

presented as are implications for ﬁrture research.

97

CHAPTER 5
DISCUSSION

Introduction
This chapter presents a discussion of the results of the dissertation research. This
discussion contains an overall summary of the ﬁndings presented in Chapter 4 compared

to the assertions presented in Chapters 1 and 2, implications for research and theory,

implications for practice and suggestions for future research.

Summary of Findings

This was an exploratory study, designed to identify and test the linkages between
context, work structures and/or work values as predictors of organizational commitment,
job involvement, intent to leave and tenure using reports from a group of professional
social workers. In so doing, I was attempting to verify three statements made by nonproﬁt
authors about both sectoral work practices and work values. First, nonproﬁt authors assert
that there are detectable value differences across the three sectors. The second statement
was that these values differences are manifested as sector-speciﬁc differences in
individual work values and work practices. A third key assertion was that the nonproﬁt
sector, more than the public or for-proﬁt sectors, provides the most hospitable
employment for professionals. Organizational behavior research, particularly that
describing the relationship and impact of work values, was used to develop a plausible
model to test the nonproﬁt authors’ statements.

Several previously normed instruments as well as specially developed scales were

used to test these assertions. The Work Values Inventory (Super, 1977) was used to test

98

the work values differences assertion. Because these individuals worked in organizations
that have not been the subject of similar empirical research, two sets of work practice
scales were developed. The study hypotheses were tested using multiple regression and
the full model was explored using hierarchical regression. The speciﬁc ﬁndings of this
study indicate the following:

This group of respondents were homogeneous with respect to their very high level
of the work value altruism (14.7 on a 15 point scale). Sector did predict the work value of
economic returns (for-proﬁt sector > public sector > nonproﬁt sector, p g .058).
Autonomy was not predicted by sector.

The factor analyses of the two sets of work practice items—reward practices and
client service decision-making practices——resulted in interpretable scales with mixed
reliabilities. Those scales were used to identify a set of reward practices and client service
decision-making practices that were used in other analyses.

Sector did not predict either the availability of professional reward practices or the
use of professional judgments to make client service decisions. Contrary to predictions,
ﬁnancial rewards were reported as more available in the nonproﬁt sector than in the for-
proﬁt. As predicted, the use of legal requirements to make client service decisions was
most prevalent in the public sector. Finally, there were no industry-based differences in
reward or client service decision-making work practices within the nonproﬁt sector.

Generally, the interactions between work values and the work practices that
mirrored those values were not signiﬁcant predictors of the dependent variables. The

interaction between the value respondents placed on economic returns and the availability

99

of ﬁnancial rewards was the sole interaction to contribute to the dependent variables. It
had an impact on tenure.

In testing the plausibility of the model using work structure and work context
variables, sector played no part. Other independent variables did predict the dependent
variables. Professional rewards and the use of professional judgments to make client
service decisions were associated with organizational commitment. Reliance on legal
requirements to make client service decisions and the lack of professional rewards
predicted intent to leave. Rewards only—an emphasis on ﬁnancial rewards and a lack of
professional rewards—predicted tenure. Finally, those independent variables had no
relationship to job involvement.

In the second test of the model, context, and particularly one’s position, was a
much better predictor of the dependant variables. In all instances, holding an
administrative position was more inﬂuential than holding a direct service position. The
work value altruism generally inﬂuenced the dependent variables in a predictable fashion:
more emphasis on altruism was positively associated with organizational commitment
and job involvement and negatively associated with intent to leave. It was also negatively
associated with tenure; a rationale for that ﬁnding is offered below. A value for economic
returns positively predicted intent to leave and negatively predicted tenure; autonomy had
no predictive impact.

No model-testing relationship successfully approximated the nonproﬁt sector’s
predictions since nonproﬁt sector was never the dominant sector.

The work practices/context path predicting intent to leave was most like the

professional work setting predictions. That is, the use of ﬁnancial practices and legal

100

practices to make client services decisions positively predicted intent to leave while the
availability of professional rewards was negatively associated with intent to leave.
Sector could predict both work values and work practices, but did not play a

signiﬁcant role in the prediction of affective outcomes; it did play a role in explaining the

behavioral outcome, tenure.

In the next section, these results are presented and discussed in one of three

subsections based on their relevance for organizational behavior, nonproﬁt or institutional

theory.

Implications for Research and Theory
Three streams of theory—drawn from organizational behavior, nonproﬁt and
institutional literature—were used to develop the hypotheses and model tested in this

study. The discussion addresses the ﬁndings and questions raised for each of these

streams.

Organizational Behﬂor Theory

The organizational behavior literature was used for three main purposes. These

were to establish:

0 The use of work values as a legitimate individual variable in explorations of

person-organization ﬁt (c.f., Chatrnan, 1989).

The rationale for ﬁt between individual and organizational attributes (Relationship

A in the model; c.f., Kristoff, 1996).

0 The predicted individual attitudinal and behavioral outcomes based on the ﬁt

between an individual and their organization (Meglino, et. al, 1989).

101

 

 

It must be noted that this study lacked the ﬁdelity with many of the cited
organizational behavior studies because it was not an exploration of practices within
speciﬁc organizations but rather a study of work practices across a class of organizations.
In the past, cross-organization studies have generally been limited to a comparison of
members’ perceived attitudes (c.f., Posner & Schmidt, 1996), although sometimes these
studies have been augmented with data on individuals’ behaviors (c.f., Chao, et. a1, 1994).
In this study, data were not available for comparisons between individuals’ and their
employing organizations’ work values nor speciﬁc individuals’ characteristics and their
employing organizations’ practices. As a consequence, the model used in this study
represented a mixed conceptualization of ﬁt with indirect cross level measurements
(Kristoff, 1996). This approach was not seen as a barrier to collecting and analyzing the
data because the work practices studied—reward practices and client service decision
making practices—were not speciﬁc to nor divisible by individuals (Roberts, Hulin &
Rousseau, 1978) and therefore did not rely on achieving some level of individual
agreement for veriﬁcation (James, 1982). Theoretically, the reward practices and client
service decision-making practices could have been veriﬁed by examining organization
policies or records.

The main effects predictably inﬂuenced the dependent variables: organization
commitment, job involvement, intent to leave and tenure. However, the ﬁt between work
values and work practices predicted none of individuals’ attitudinal outcomes as
hypothesized. Only the interaction between individuals’ value for economic returns and

ﬁnancial reward practices predicted any outcome, and that was for the behavioral

102

 

outcome tenure. Two explanations—one statistical and one conceptual—are offered for
the dearth of successful ﬁt predictions.

Two statistical artifacts may explain the failure of the interactions to make
signiﬁcant contributions to the outcomes. First, the work values scale for autonomy was
unreliable. Therefore, this may have accounted for the fact that the variable made no
signiﬁcant contribution to any of the multiple regression equations. Second, it appears

that multicollinearity was problem in these equations. Table 31 shows the correlations

among the main effects and their interactions.

The interactions overlap substantially with the work practices they include, from a
high correlation of .906 between professional rewards and the autonomy x professional
rewards interaction to a low of .817 between the use of professional judgments in client
service decisions and the interaction between autonomy and the use of professional
judgments. Clearly there is no new information that these interactions can add.

This table suggests that not only does the inclusion of key interactions NOT
improve the prediction of affective and behavioral outcomes, in all instances the
interactions appear to suppress main effects. Only in the instance of tenure is there an
interaction effect and it occurs at the expense of four main effects.

The lack of statistical success in using these interactions prompted a
reconsideration of the appropriate way to model ﬁt in these equations. As was noted
above, in other work values research, supplementary ﬁt is operationalized as the
interaction between an individual’s work values and those of the employing organization.
That information was not available here. Studies of complementary ﬁt, where personality

data is used in conjunction with work practices, have modeled ﬁt differently. With mixed

103

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50. V Q 3.2..

8. v a r.

8. v a ..
ﬂ neeoom. §§§w_v. «cicNm. 2:95. on; _. 9!..wa mgr mEmBoM _ncoﬁmomen— X cancer—9:" .w
ooor mmo. .vmo. «ooh—w. «gem—c. see—2. mace—avg. _acgmmomohm X buoy—0:34 .b
xii—mm. oceemv. aoor go. cannon. mENBDM _ﬂocaﬁm X mEEoM EEO—50m .0
23.3%. _No... 20.- «CO—.1 mcagom _EocuEm .m
owao. 0N9- nemgv mEaBom _mcoﬁmouohm .v
V8. 53. muCOE—a—n —ﬂ=0mwwOHO.—n~ I Domtom wCOSU .M
noemmm. echo—522‘ >3 .N
mac—502m ”VEGA—00m Ev OS—ﬂ> v10; A
m _ A .e .m .3. .m H .~ # .2

 

 

$852225 :23. 98 38am :3: 05 meoE< mecca—2.80

3 2an

104

results, they have used personality traits to predict work practice preferences in job search
research (c.f., Bretz, et al., 1989 and Cable & Judge, 1994). The equivalent here would be
to use individuals’ work values to predict work practices. It must be noted that these are
job incumbents, whose work practices are less likely to solely reﬂect their individual
preferences than is the case when making job search decisions. The alternate speciﬁcation
of the relationships between work values and work practices was not the focus of this

research, but it an area for future exploration.

Nonproﬁt Theory

The nonproﬁt literature was used to specify the values, work practices, and the
nature of relationship between them that were used in this study. In particular, nonproﬁt
authors suggested that sector values differences were striking, especially with regard to
the relative emphasis on altruism versus economic returns (c.f., Young, 1984). Further,
they suggested that these differences could be clearly seen in the values held by
individuals working in that sector (Witty,1989) and that these values would be reﬂected
in the structures developed by nonproﬁt organizations. As was demonstrated in the
literature review, nonproﬁt authors may have consistently asserted these value
differences, but it remained the task of others to actually test for value differences. Varied
methods were used to conﬁrm value differences; none were directly tested with a work
values scale. The selection and use of a work values scale was a task undertaken, with
moderate success, in this study. Based on the nonproﬁt literature, scales that identiﬁed
levels of altruism, autonomy, and economic returns were used (Super, 1970). In this

population, the items deﬁned as altruism and economic returns formed reliable scales (or

105

= .81 and .83, respectively) and the autonomy items did not (or = .54). A set of items
drawn from earlier work with a large nonproﬁt child welfare agency was used to deﬁne
the possible range of reward practices that might be used. Those items were analyzed to
develop sub-scales for categories of rewards that organizations might be using, and the
subscales conformed to categories discussed in the nonproﬁt literature: pay/other
ﬁnancial rewards, professional rewards, bonuses and other miscellaneous rewards.

In this study I achieved mixed results with regard to key nonproﬁt assertions about
work values. Contrary to predictions, there were no sector-related differences in the value
altruism. The sector-related predictions about the value for Economic Returns was
supported. In this sample, those people working in the for-proﬁt sectors valued economic
returns more than those in the public sector; and both more than individuals working in
the nonproﬁt sector. Finally, a preference for the work value autonomy approached
signiﬁcant, but was opposite the hypothesis. In this sample, people working in the for-
proﬁt sector valued altruism more than those in the nonproﬁt sector.

Different explanations are offered for each of these work-values-related results.
First, in this population altruism was a strongly held value. As noted previously, there
was almost no variance in the population; respondents averaged 14 points on a 15-point
scale. As a result, there was no variance that could be used to test the assertions.

This research provided empirical support for accumulation of data that has been
assembled about the differences in the value placed on economic returns across the three
sectors. One aspect of the sample that was not tested (and about which there were no
hypotheses), the inter-play between individuals’ values for job security and economic

returns. While every attempt was made to identify a population in which a three-sector

106

 

 

study would be possible, there appear to have been substantial limitations in this
population. F or-proﬁt employment was not evenly distributed in this population; most
respondents working in that sector held jobs in mental health or substance abuse agencies.
Given realities in Michigan, it is unlikely that most of the individuals currently working

in the for-proﬁt sector voluntarily chose that sector. Those agencies have been most
affected by the privatization of formerly public or nonproﬁt organizations. It is reasonable
to assume that these employees remained with their jobs as those jobs made the transition
from one sector to another, not that they voluntarily sought employment in that sector.
Nonetheless, the results are as hypothesized.

Finally, the contradictory ﬁndings with regard to values for autonomy—which
merely approached signiﬁcance—may be the result of scale unreliability. That could
cause ﬁndings that were opposite the predictions or this could actually be a null
hypothesis. However, the true outcome cannot be known.

Mixed results were also found for relationships between sector and work
practices. In this sample, sector did not predict professional practices, neither the
availability of professional rewards nor the use of professional judgments to make client
service decisions. This is consistent with the ﬁnding above that sector did not predict the
work value autonomy. However, this result could also be an artifact of this all-
professional population. Sector did predict the use of legal requirements/rules as
hypothesized, with those in the public sector placing a greater emphasis on the use of
rules to make client service decisions than those employed in the nonproﬁt sector. There
was support for the use of ﬁnancial factors in client service decision making. As

predicted, those employed in the for-proﬁt sector placed a greater emphasis on ﬁnancial

107

 

 

factors than those in the nonproﬁt sector. However, those working in the public sector
placed the greatest emphasis on ﬁnancial factors; there was no hypothesis about the
relative emphasis placed on ﬁnancial factors in the public sector.

Model testing was undertaken to test relationships that might be inferred from the
nonproﬁt literature. If sector (a context variable) has an impact on work structures and
work values, then it is reasonable to expect the inﬂuences of sector to inﬂuence
individuals’ affective and behavioral outcomes. That supposition was not borne out. In
the hierarchical regression equations, sector made no contribution over and above that
made by work values or practices.

While the sector-related inﬂuences were not borne out, the proﬁle of the
committed professional worker was. A separate proﬁle of the professional employee was
related to intent to leave one’s organization. Although professionalism was seen as a
hallmark of nonproﬁt employment (Majone, 1984), in fact this study conﬁrmed it as a
universal (non-sectoral) attribute. The opportunity to act using one’s professional skills
and be rewarded for them contributed to these respondents’ plans to stay with their
organizations; the de-emphasis of these same factors contributed to intent to leave. Intent
to leave one’s employment was negatively predicted from the use of professional
attributes to make client service decisions and also the availability of professional
rewards and positively predicted by the use of ﬁnancial and legal practices to make client
service decisions.

Organization (Thompson, 1967) and nonproﬁt theory (Young, 1984) were
combined to develop a rationale for ﬁnding industry differences. The predicted

differences and ﬁndings are discussed in the next section.

108

 

 

_ILs/titutional TheoLv

Young (1984) suggested that differences found in nonproﬁt organizations’
practices could be accounted for by examining the organization set within which they
functioned. In this study, distinctions were made among agencies that were identiﬁed as
providing family service, child welfare or mental health services. Each of these industries
was predicted to have a different set of constraints which would in turn be reﬂected in
different work practices and different employee values. None of the industry predictions
were supported: no individual work value or work practice differences were found within
the nonproﬁt sector among these three industries. While these results were consistent

with the lack of sector differences, an explanation will assist in considering future

research efforts.

If the descriptions of these three industries do suggest that some differences might
reasonably be found, then it is useful to speculate on the factors that account for these null
results. An initial explanation might lie with the measuring devices used here. It has
already been suggested that the scales used to deﬁne reward practices might have
obscured sector differences. The same may be true for the client service decision-making
scales. Recall that the Legal Practices scale had low reliability (or = .52). A further
investigation into the two items revealed that while no differences were discemable in the

use of legal/accreditation requirements (S2) there were differences in the application of

government polices. Table 32 shows the responses.

109

 

 

Table 32

 

 

 

Differences in the use of Government Policies by Industpy
Reported use of government policies or Industry
mandates in making client service decisions Child Family Mental
Welfare Service Health
No or limited inﬂuence 2% 12% 19%
Some inﬂuence 18% 31% 26%
Great deal or major inﬂuence 80% 57% 55%

 

Percents rounded to nearest whole number.

x2 = .05
Respondents in the mental health industry were most likely to report that
government policies had little inﬂuence on their practices and those in the child welfare
industry, in support of the hypothesis, were most likely to report the use of government
policies in making client services decisions. These results suggest that important industry
differences may have been obscured by the production of scales that combined items that
operated in different ways. It would have been rewarding to ﬁnd that similar differences
occurred in the use of cost factors; that is, that respondents in the mental health industry
reported either that funder rules regarding the costs for services (S3) or the cost of

alternative methods or services (S9) more ﬁequently inﬂuenced their client service

decisions. However, neither item’s predictions were supported (x2 = .267 and .532,

respectively).
In this study, a plausible model for associating context, work values and work

practices with individuals’ affective and behavior work outcomes was developed and
tested. The ﬁndings were least consistent with those from prior organization behavior
research. Work values (measured here as altruism, economic returns and autonomy) could

not be associated with work practices (measured here as sets of reward practices and sets

110

 

 

of client service decision making practices) to predict organizational commitment, job
involvement, intent to leave and tenure.

Modest success was achieved using sectoral (nonproﬁt). While the inﬂuences of
sector could not be traced through the model, sector did predict the components: work
values and work practices. Institutional (industry) theory predictions were not supported.
As has been discussed above, some of the null ﬁndings could be the result of scale
unreliability. However, the “nonproﬁt” value altruism was a consistent predictor of all
dependent variables. And, in the case of job involvement, was one of only two factors
(position was the other) that made a contribution.

A review of the ﬁndings suggests that lack of empirical research related to
“nonproﬁt” work values has lead to potentially erroneous conclusions. When one
examines the results from this study, it becomes apparent that these respondents are not
trading their value for altruism with their value for economic returns toward the same
organizational outcomes. Rather, these are values that often operate in concert to predict
the affective and behavioral outcomes. The results in this study suggest that these
respondents are not trading preferences for pay with those for service to others, they have

both values but are applying them separately to enact affective and behavioral outcomes.

Implications for Practice
This study included a single type of professional from a relatively narrow range of
organizations. Therefore the implications that can be drawn from this study are
necessarily limited to similar individuals in similar organizations. One of the clearest

implications from these data is that altruism and economic returns are not opposite ends

111

 

 

of a single value continuum. Rather, the clear ﬁnding is that people like these have
similarly high regard for both values, and that the gains from each are different. Take, for
example, the organization that experiences high staff turnover. Turnover not only has
dollar costs associated, but for human service organizations has consequences for the
successful delivery of interpersonal services; consumers must connect and reconnect with
staff in order to achieve their goals.

Human service organizations are likely to recruit based on the opportunity to do
something important—or, appeal to job seekers’ altruism. Strategically this is important,
because altruism is a consistent predictor of the dependent variables. However, altruism is
one of several variables that inﬂuence intent to leave or tenure. Here, intent to leave is
based on the availability of ﬁnancial rewards and one’s value for economic returns.
Further, these data suggest that employees’ intent to leave is potentially forestalled by the
availability of professional rewards and the opportunity to make client service decisions
based on professional, not legal or ﬁnancial factors. No predictions can be made
regarding when trade-off between professional and ﬁnancial rewards might occur, but
these data suggest that some balance could be achieved between providing professional
rewards and ﬁnancial rewards with the beneﬁt of longer employment.

Work values do play a role in employees’ affective outcomes. Employees come to
organizations with values, but those values can be shaped. A realistic match between
available economic returns and one’s value for economic returns could again potentially
forestall one’s intention to leave. Since many of these respondents felt that they had pay
comparable to others in similar jobs, one of the activities that these organizations might

sponsor are periodic salary surveys so that their employees have an accurate sense of the

112

 

going rate for their work. (This presumes that the organization will be able to maintain
comparability.) Altruism is a signiﬁcant contributor to both organizational commitment
and job involvement. So the tactic of describing the contribution one would be able to
make in a job interview would have the beneﬁt of attracting people who want to provide

that service to others, with the eventual outcome of increased organizational commitment
and job involvement.

Suggestions for Future Research

This line of inquiry could be strengthened by additional research that would
illuminate both organization behavior and nonproﬁt concerns. These include: improved
deﬁnitions of context variables, better work values measurement (both the tool and
matching individuals’ and organizations’ work values) and clariﬁcation of what role, if
any, respondents’ professional afﬁliation has for their organizational afﬁliation.

A key tenet of nonproﬁt literature is that there are discemable differences among
the three sectors. One of the dilemmas of pursuing cross-sector research is generating a
pool of potential respondents that are similar in important ways but different in their
context. The approach used here of following individuals with similar educational
characteristics and work situations was only partly successful in generating these context
differences. It may be that for these human service organizations, context deﬁned as
sector is not a distinguishing characteristic, but other contextual factors are. In this study,
it was not possible to pursue any of a number of organization-based characteristics that
might have shown differences in work practices or individuals’ work values. Among the

organizational contextual characteristics that might be discemable and different are

113

number and source of funding streams, board characteristics, strategy or purpose, the
presence of absence of the founder at the head of the organization, and/or role in the
community (c.f., Drucker, 1990; Hodgkin, 1991). Further organization-based research
into the most important context variables could help to focus the nonproﬁt sector
discussion on those variables that do make a difference.

The investigation of context can also be improved at the person level. In addition
to position, which was an individual level contextual variable, employment status may be
a ﬁ'uitﬁrl variable to include in ﬁrture research. In this study, respondents were advised to
answer questions using their full-time employer or one of their multiple employers.
Mixed employment status (more than one employer and/or less than full-time with one
employer) was a characteristic of more than one third of this study’s respondents. This
status should be more proximal than sector for individual employees and therefore likely
to have more of an effect on their work values, organizational commitment and job
involvement.

There are two general approaches that improved investigation of work values
could take: the development of better tools and improved collection of matching person
and organization data. The ﬁrst improvement has to do with the device used to investigate
individuals’ work values. Super’s Work Values Inventory (1970, 1977) was used because
it had items that speciﬁcally reﬂected the identiﬁed unique characteristics of nonproﬁt
organizations (and also could be used here without matching data from others in the
organization). Since the nonproﬁt authors are so convinced that value differences exist, it
might be a worthwhile endeavor to develop a sector-speciﬁc instrument. Alternatively,

the use of a general work values tool like the Organizational Culture Proﬁle (OCP;

114

 

O’Reilly and colleagues, 1991) could provide important insight into their real or imagined
unique characteristics. In this study, the three focal values were not similarly predicted by
sector; with another tool that might or might not be true.

The second improvement would be the better speciﬁcation of ﬁt. One way
measurement of ﬁt could be improved is through the collection of matching organization
work value information. Collection of matching organization work values data could be
accomplished by direct assessment (“my values are like my organization’s values”). Data
collected in this manner could be paired with individual data using a tool such the Work
Values Inventory to determine comparability between individuals’ and organizations’
values. Indirect assessment will require the development of matching organization
proﬁles from multiple people in the same organization—a different strategy than that
used in this study. Alternatively, the speciﬁcation of ﬁt could be altered to conform more
closely to that used by others pursuing complementary ﬁt approaches.

Finally, a potentially ﬁuitful avenue of investigation, particularly for populations
similar to this one, is the examination of the role that professional afﬁliation might play
in the person-organization ﬁt relationship and/or the prediction of desirable person and
organization outcomes. Summaries of prior organizational behavior research have shown
that loyalty to both union and employer can contribute to organizational commitment
(Reed, Young & McHugh, 1994). It may be that professional commitment also adds to
organizational commitment.

This study tested a model that proved plausible for linking contextual variables to
individuals’ work values and also to their settings’ work practices as predictors of

individual and organizational outcomes. This study described a set of work practices that

115

 

were used across human service organizations in three industries. While the independent
variables operated in predictable fashion, the operationalization of ﬁt between work
values and work practices did not. Suggestions have been made to improve the
speciﬁcation of ﬁt. The data supported nonproﬁt authors’ assertions that those who work
in the nonproﬁt sector value economic returns less than their public or for-proﬁt sector
counterparts. Further, in this population, the “nonproﬁt” value altruism was similarly and
highly valued across all sectors. However, in contrast to the assertions by nonproﬁt
authors, one of the important discoveries was the ﬁnding that these individuals’ values for
altruism and economic returns do not represent trade-offs, but rather are similarly valued

and can be used in concert to predict affective and behavioral outcomes.

116

 

APPENDICES

117

APPENDIX A

INSTRUMENTS

118

.1 3'.

 

lll

APPENDIX A1
COVER LETTER

February 1999
Dear Social Work Professional,

You received this survey because you are a member of the Michigan Chapter of the
National Association of Social Workers. This survey is being sent to members in ﬁve counties in
order to find out more about professional social workers who work in health and human
services organizations. The summary results from this study will be available to Michigan
Chapter members.

Every attempt has been made to include only those people with agency-based
employment. Two employment situations may create questions:

1. If you have changed jobs or retired in the last six months, PLEASE COMPLETE THE
SURVEY based on the job you recently left. Your comments about your past employment
are very important.

2. If you work only in private practice, please DO NOT complete the survey. Please do
return this cover letter with the information checked here:
I only do private practice. Unfortunately, only people who are eligible to complete
the survey can enter the lottery described below.

There are four sections in this survey that include questions about:

- background information on your agency;

- your attitudes toward your work;

- various agency practices;

- you and your profession.
It will take approximately 30 minutes to complete this survey. Please return it in the enclosed
envelope.

Your responses are anonymous. No answers can be associated with any other
information about you or your agency. All answers will be reported in summary form. Examples of
summary reports may include information about types of agencies such as “mental health” or
sectors such as “nonproﬁt.” The zipcode has been placed on your questionnaire in order to
summarize data by geographic area, such as “Grand Rapids and the surrounding cities."

You indicate your voluntary agreement to participate by completing and returning this
questionnaire.

Please return the survey by:

If you have any questions, please contact me:
Celeste Sturdevant Reed, M.S.W.
Michigan State University
Institute for Children, Youth, 8. Families
27 Kellogg Center
E. Lansing, MI 48824
51 7/353-661 7
csreed@pilot.msu.edu

Thank you for reviewing this information.

119

 

 

 

If you would like to be entered in a lottery to win a one year basic membership in the National
Association of Social Workers, please detach this section and return it to me. You may either
mail this with your survey or send it in a separate envelope to:

Celeste Sturdevant Reed

305 E. Middle Street

Williamston, Ml 48895

Name

 

Address

 

Telephone

 

(Needed to conﬁrm arrangements with you)

120

 

 

APPENDIX A2
FOLLOW-UP POSTCARD

Dear Michigan Chapter NASW Member,

Approximately ten days ago I mailed you a survey about professional social workers who work in
human services agencies.

If the survey has been completed and returned to me, please accept my sincere thanks. If you
have not returned the survey, please complete and return it in the envelope that was provided. In
order to accurately reﬂect the views of social workers about their situations, it is extremely
important that your answers be included in the results.

If by some chance you did not receive the survey, it has been misplaced. or you have additional
questions, please call me at 517/353-6617 and I will be happy to respond.

Sincerely

Celeste Sturdevant Reed

121

Appendix A3
SURVEY INSTRUMENT

 

SECTION 1
BACKGROUND INFORMATION

 

The ﬁrst three items ask for basic background information about you.

Bl.

82.

B3.

Please indicate the education that you have. Check all that apply.

BA or BS in ﬁeld other than social work

BA in social work or BSW

graduate course work not culminating in a degree
MSW

other MA

other professional degree (nursing, law, medicine, etc.)
PhD in social work
other PhD

other; please describe

 

Please indicate the licenses or certiﬁcates that you hold. Check all that apply.

certiﬁed social worker in Michigan

ACSW

Advanced Diplomate

Licensed social worker (state other than Michigan)
certiﬁed substance abuse counselor

Limited license psychologist

Certiﬁed Marriage and Family Counselor

Other; please name

 

 

Please describe your current pattern of employment.
Check the one answer that best characterizes your situation.

Recently retired from agency being described.

Less than full-time employment in one agency. How many hours/week?

F ull-time employment in one agency.

Less than full-time employment in each of two different settings.

Full-time employment in one agency plus part-time employment in another agency.
Full-time employment in one agency plus part-time private practice or consultation.
Other. Please describe:

 

 

122

 

 

This group of questions asks for basic information about the agency that you work in. If you work in more
than one agency, select one agency to serve as the reference point for your answers. Do not report on any
private practice that you may also perform

B4. Please check the one sector that applies to your agency
Private not-for-proﬁt

Public
Private for-proﬁt

BS. Please check the one ﬁeld of practice that best represents your agency’s work:

Child welfare
Family Services
Health

Mental Health
Occupational Social Work/BAP
Substance Abuse
Other (please identify)

 

 

B6. Please check the one type of work setting in which you are employed:

_ Business/Industry __ Mental Health; which type of setting?
__ Health; which type of setting? _ inpatient

inpatient __ outpatient

outpatient _ Social Service agency

Managed care
Residential facility (group home, etc.)

Other; please identify

 

B7. Which of the following client populations are served by your agency? Check all that apply.

 

 

_ children _ families

_ adolescents __ citizens in general
_ young adults _ neighborhood groups
_ adults _ other organizations
_ seniors Other:

special populations (migrants, the unemployed, women, men, people of color, etc.)

88. Please check the one description that is most like your position:

Direct service to clients (either intervention or advocacy)

Combined direct service and supervisory position

Supervision only

Management or administration

Supportive services such as consultation, training, fund-raising, or personnel
management.

Other (Please describe)

lllll

 

123

ll

B9. Adding together all of the jobs that you have held with this agency, how long have you worked
here?

years; or months if less than one year.

B 10. If you have your way, three years from now will you be working with this agency?

Yes

No

Not applicable (i.e., already retired or recently taken another job). If this is the case,
skip questions B11 and B12 and GO TO THE NEXT SECTION BELOW.

 

 

 

To a

Circle the number of the response that best describes your great Very Not
situation. extent little at all
Bl 1. To what extent have you thought seriously about

. . . . . 5 4 3 2 l
changmg agencres srnce begmmng to work here?
BIZ. To what extent would you prefer another more ideal
. . 5 4 3 2 1
job than you now work 1n?

 

 

 

 

 

 

 

 

 

 

SECTION 2
YOUR A'ITITUDES ABOUT YOUR AGENCY, WORK AND CLIENTS

 

In this section there are no right or wrong answers. I am interested in understanding your speciﬁc situation.

Organization Goals and Values

 

 

 

 

 

 

For the following statements, circle the
number of the response that best Neither
describes your relationship with your Strongly Agree nor Strongly
agency. Agree Agree Disagree Disagree Disagree
G1. The goals of my agency are also 5 4 3 2 1
my goals.
G2. I believe that I ﬁt 1n well 1n my 5 4 3 2 1
agency.
G3. I do not always believe in the

5 4 3 2 1
values set by my agency.
G4. I would be a good example of an
employee who represents my agency’s 5 4 3 2 1
values.
G5. I support the goals that are set by 5 4 3 2 1
my agency.

 

 

 

 

 

 

 

 

124

Work Values

The statements below represent values which people consider important in their work. These are
satisfactions which people often seek in their jobs or as a result of their jobs. They are not all considered
equally important; some are very important to some people but of little importance to others. Read each
statement carefully and circle one answer that indicates how important it is for you.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

How important is work in Very Moderately Of Little
which you... Important Important Important Importance Unimportant
W1. help others 5 4 3 2 1
W2. ...can get a raise 5 4 3 2 1
W3. ...have freedom in 5 4 3 2 1
your own area.
W4. ...gain prestige in 5 4 3 2 1
your ﬁeld.
W5. know yourjob will 5 4 3 2 1
last.
W6. ...get the feeling of
having done a good day’s 5 4 3 2 1
work.
W7. ...know by the results
when you’ve done a good 5 4 3 2 1
job.
W8: ...are sure of always 5 4 3 2 l
havmg a job.
W9. ...make your own 5 _ 4 3 2 1
decrsrons.
W10. ...have pay increases
that keep up with the cost 5 4 3 2 l
of living.
W11. ...know that others
consider your work 5 4 3 2 1
important.
W12. ...feel you have
helped another person. 5 4 3 2 1
W13. ...add to the well-
being of other people. 5 4 3 2 1
W14. are looked up to 5 4 3 2 1
by others.
W15. ...are paid enough to
live right. 5 4 3 2 1
W16. ...are your own boss. 5 4 3 2 1
W17. ...are sure of another
job in the agency if your 5 4 3 2 1
present job ends.
W18. ...see the results of
your efforts. 5 4 3 2 1
l 25

Agency Relationships

The following statements reﬂect ways that people can feel about their organizations. You may or may not
have these same feelings about working in your agency. Remember, there are no right or wrong answers.

 

 

 

 

 

 

Neither

Please circle the one response that Strongly Agree nor Strongly
best identiﬁes your feelings. Agree Agree Disagree Disagree Disagree
A1. The agency has a great deal of

. 5 4 3 2 1
personal meaning for me.
A2. One of the few negative
consequences of leaving this agency 5 4 3 2 1
would be the scarcity of available
alternatives.
A3. I would feel guilty if I left my 5 4 3 2 1
agency now.
A4. I do not feel like part of the 5 4 3 2 1

family” at my organization.

A5. If I had not already put so much of
myself into this agency, I might 5 4 3 2 1
consider working elsewhere.

 

 

 

 

 

 

 

 

A6. I owe a great deal to my agency. 5 4 3 2 1
A7. I do not feel “emotionally 5 4 3 2 1
attached” to my agency.
A8. I feel that I have too few options to

. . . 5 4 3 2 1
consrder leavmg this agency.
A9. This agency deserves my loyalty. 5 4 3 2 1
A10. I do not feel a strong sense of
‘6 ‘ ’9 5 4 3 2 1

belongmg to my agency.

Al 1. Too much of my life would be
disrupted if I decided to leave my 5 4 3 2 1
organization now.

 

A12. Even if it were to my advantage,
I do not feel it would be right to leave 5 4 3 2 1
my agency now.

 

 

 

A13. I really feel as if this agency’s

5 4 3 2 1
problems are my own.
A14. It would be very hard for me to
leave my agency now, even if I wanted 5 4 3 2 1
to.
A15. I do not feel any obligation to 5 4 3 2 1

remain with my current employer.

 

A16. I would be very happy to spend 5 4 3 2 1
the rest of my career with this agency.

 

 

 

 

 

 

 

 

126

 

 

 

 

 

 

 

 

 

 

 

Neither

Please circle the one response that Strongly Agree nor Strongly
best identiﬁes your feelings. Agree Agree Disagree Disagree Disagree
A17. Right now, staying with my
agency is a matter of necessity as much 5 4 3 2 1
as desire.
A18. I would not leave my agency
right now because I have a sense of 5 4 3 2 1
obligation to the people in it.

Job Aspects

People may work in the same type of agency but still have very different jobs that they do. Below are a
number of statements each of which you may agree or disagree with depending on your own personal

evaluation of your job.

 

 

 

 

 

 

 

 

 

 

 

 

 

job most of the time.

 

 

 

 

Circle the one answer that best Strongly Neither Agree Strongly
reﬂects your current job situation. Agree Agree nor Disagree Disagree Disagree
J l. The most important things that
happen to me involve my present 5 4 3 2 1
job.
J2. To me, my job is only a small
part of who I am. 5 4 3 2 1
J3. I am very much involved 5 4 3 2 I
personally in my Job.
J4. I live, eat and breathe my job. 5 4 3 2 1
J5. Most of my interests are 5 4 3 2 1
centered around my Job.
J6. I have very strong ties with my
present job that would be hard to 5 4 3 2 1
break.
J7. IJsually I feel detached from 5 4 3 2 1
my Job.
J8. Most of my personal life goals

. . 5 4 3 2 l
are Job-oriented.
J9. I consider my job to be very

. 5 4 3 2 1

central to my exrstence.
J10. I like to be absorbed in my 5 4 3 2 1

 

127

 

 

 

 

SECTION 3

AGENCY PRACTICES

 

Many factors are taken into consideration by people delivering services to clients. Please rate the extent to
which each of these factors has an influence on services delivered to your agency’s clients.

Client Service Practices

 

Source of
inﬂuence on
client service
decisions

Has the
major
inﬂuence on
service
decisions

Has a great
deal of
inﬂuence on
decisions

Has some
inﬂuence on
service
decisions

Has limited
inﬂuence on
service
decisions

Has little or
no inﬂuence
on service
decisions

This is not
applicable to
our clients or

services

 

Sl . Government
policies or
mandates

4

NA

 

82. Legal or
accreditation
requirements

NA

 

83. F under rules
regarding
unifomiity of
service to clients

NA

 

S4. F under rules
regarding costs
for services.

NA

 

S5. Agency
mission or

purpose

NA

 

S6. Agency
policies and rules

NA

 

S7. Agency public
imaje

NA

 

S8. The
predictability of
external funds

NA

 

S9. The cost of
alternative
methods or
services

NA

 

810. Board
member and/or
donor preferences

NA

 

SI 1. Staff
commitment to
the agency
mission.

NA

 

S12. Staff
knowledge and
skills

NA

 

S 1 3. Staff
motivation and
articipation

NA

 

S 14. Staff
cooperation inside
the agency

 

 

 

 

 

 

 

NA

 

128

 

 

 

Source of
inﬂuence on
client service
decisions

Has the
major
inﬂuence on
service
decisions

Has a great
deal of
inﬂuence on
decisions

l-Ias some
inﬂuence on
service
decisions

I-Ias limited
inﬂuence on
service
decisions

Has little or
no inﬂuence
on service
decisions

This is not
applicable to
our clients or

services

 

S15. Staff
collaboration with
others outside the

_agency

NA

 

816. Staff
professional

Meme“

NA

 

Sl7. Severity of
client’s problem

NA

 

S18. Client’s
ability to pay (or
insurance
coverage)

NA

 

Sl9. Client’s
motivation and
participation

NA

 

820. Client’s level
of family or social

support

NA

 

821. Local
community values
or sentiments

NA

 

822. Availability
of volunteers

NA

 

823. Availability
of and/or access
to other
community
services

 

 

 

 

 

 

 

 

NA

 

821. What three factors have the most positive impact on client services?

 

S22. What three factors have the most negative impact on client services?

 

129

 

 

 

Staff Rewards

Agencies can provide a variety of rewards to staff. In the following list, please check off ( ‘1 ) each of the
rewards that are available in your agency and also check off ( ‘l ) rewards that are available to you.

 

Available in your Available to
Rewards agency you

 

R1. Increases in salary

 

R2. Bonuses for performing quality work.

 

R3. Bonuses for performance related to budgets and
schedules.

 

R4. Longevity pay.

 

 

R5. Pay levels that are comparable to people with similar
jobs in other agencies.

 

R6. Job security.

 

R7. Policy or practice of reassigning staff rather than laying
people off.

 

R8. Agency paid life insurance.

 

R9. Agency paid health insurance.

 

R10. Agency paid retirement benefits.

 

R11. Doing work you believe in.

 

R12. Important duties and responsibilities.

 

R13. Inﬂuence in setting goals and making decisions.

 

R14. Freedom concerning working hours.

 

R15. Freedom concerning job duties.

 

R16. A compatible group of co-workers.

 

R17. Positive responses from clients.

 

R18. Formal cormnendations and rewards.

 

R19. Expenses paid to present papers at conferences.

 

 

R20. Membership in high status task forces.

 

 

 

 

130

 

General Management Practices

In order to be a cost-effective agency, a number of different management practices may be used in
agencies. This list includes some of the more common practices that agencies use.

For each of the items listed below, please indicate whether the management practice is being used now in
your agency. If it is in use, please identify the general length of time that the practice has been in use. A
“traditional” practice is one that has been in operation for a number of years; a “recent” practice would
have been adopted in the last year or two.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Is this technique

practiced in your If you answered “yes,” is this a

agency? traditional or a recent practice?
MANAGEMENT PRACTICE NO YES 9 Traditional Recent
G1. Financial incentives to employees N Y T R
62. Efforts are made to collect fees. N Y T R
G3. Billable services are emphasized. N Y T R
G4. Reductions have been made in
administrative costs. N Y T R
G5. Fees are charged for missed
appointments. N Y T R
G6. Clients are required to pay for N Y T R
some servrces in advance.
G7. A collection agency is used for N Y T R
past due fees.
G8. There are separate facilities or
waiting rooms for clients with N Y T R
insurance.

13 1

Monitoring Practices

You, yourself, the agency and outside funders may all be engaged in monitoring the quality and quantity of
services to clients.

 

Not
Very Very at Don’t
Much Somewhat Little All know

 

Ml . In evaluating your agency’s performance, to
what extent do others outside of the agency pay
attention to the practices of other agencies like
yours?

 

M2. In evaluating your agency’s performance, to
what extent do you pay attention to the practices of 5 4 3 2 1
other agencies like yours?

 

 

M3. How much are your agency’s operations
regulated by govermnent agencies?

 

 

 

 

 

 

 

 

For each of the following factors, check off all of the instances that are true in your agency. For example,
both the agency and the outside funders may be monitoring whether or not the agency “stays within its
budget”, but only you may be concerned about “speciﬁc results with clients.”

 

This is a concern This is a factor that This is a factor that
Factors that represent the quality of yours that you the agency routinely outside funder(s)
or quantity of services to clients. monitor monitors routinely monitor
M4. Staying within the budget

M5. Ensuring eligibility of the
clients

 

 

 

M6. Providing the quantity
(number, amount, duration) of
services as planned or contracted

M7. Providing the quality of
services as planned or contracted

 

 

M8. Achieving specific results
with clients.

M9. (Something else?)

 

 

 

 

 

 

 

132

Funding Sources

Agencies receive ﬁnancial support for programs from many sources. For each of the sources of funds listed
below, circle one answer representing the contribution of that source.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A major
source of A modest Not a source I don’t
funds source of funds of funds know

F 1. Consumer/client fees for service 2 1 0 DK
F2. United Way funding and/or other 2 1 0 DK
federated charitable giving

F3. Tax dollars (directly allocated) 2 1 0 DK
F4. Private Insurance reimbursements 2 1 0 DK
F5. Medicaid reimbursements 2 1 0 DK
F6. Government purchase of service 2 1 0 DK
contracts

F7. Employee Assistance Program 2 l 0 DK
contracts

F8. Private sector contracts (excluding

E AP) 2 1 0 DK
F9. County grants/contracts 2 l 0 DK
F 10. Municipal grants/contracts 2 1 0 DK
F11. Private foundation grants 2 1 0 DK
F12. Corporate donations and/or 2 1 0 DK
contributions

D13. Individual donations and/or 2 l 0 DK
contributions

F14. Sales of unrelated products, 2 1 0 DK
goods or services

F15. Something else?

2 1 0 DK
F16. Something else?
2 1 0 DK

 

 

 

 

 

 

 

F17. In the last ﬁve years, what source(s) of funds have increased the most?

 

F18. In the last ﬁve years, what source(s) of funds have decreased the most?

 

133

 

SECTION 4
YOU AND YOUR PROFESSION

 

Professional Relationships

The following statements describe how people feel about their profession. These may or may not be true
for you. Please circle the one responses that best reﬂects your feelings.

 

 

 

 

 

 

 

 

Neither
Agree
Strongly nor Strongly
Agree Agree Disagree Disagree Disagree

P1. I feel very loyal to the social work 5 4 3 2 1
profession.
P2. I am proud to tell others that I am a social 5 4 3 2 1
worker.
P3. I talk up social work as a great ﬁeld. 5 4 3 2 1
P4. I am glad I chose social work over other 5 4 3 2 1
professions I was considering.
P5. For me social work is the best profession to 5 4 3 2 1
be in.
P6. I really care about the fate of the social 5 4 3 2 1

work profession.

 

P7. Being a social worker inspires the best in
. 5 4 3 2 1
my 1013 performance.

 

P8. I am willing to put in a great deal of effort
beyond that normally expected in order to help 5 4 3 2 1
the social work profession be successful.

 

P9. I ﬁnd that my values and the social work 5 4 3 2 1
profession’s values are very similar.

 

P 10. Deciding to work in the social work
. . 5 4 3 2 1
professron was a rmstake.

 

 

 

 

 

 

 

 

OVER " OVER "' OVER

134

 

Professional Participation

Please indicate the regularity with which you participate in the following types of professional activities.
Regularity is based on the frequency of the offering. Attending an annual meeting every year is “regular;”
attending a monthly meeting once a year is “irregular”.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Please circle one answer. Regularly Irregularly Not at All
V1. Local social work professional meetings. 2 l 0
V2. Local practice-related professional meetings 2 1 0
(child welfare, mental health, etc.)

V3. Local advocacy related meetings (child abuse 2 l 0
and neglect, fair housing, etc.)

V4. State-wide social work professional meetings 2 1 0
V5. State-wide practice-related professional meetings 2 1 0
V6. State-wide advocacy-related meetings 2 l 0
V7. National social work professional meetings 2 1 0
(NASW Annual Meeting)

V8. National practice-related professional meetings 2 1 0
(Family Services of America, Child Welfare League)

V9. National advocacy meetings (Children’s Defense 2 l 0
Fund, National Alliance for the Mentally Ill)

 

V10. How many subscriptions to professional journals—other than Social Work—do you have?
(number)

 

Thank you for your time and responses. This ends the questionnaire.
OOOOOGOOGOGOGOO

If you would be willing to give the following information, I would appreciate it.

Gender Race/Ethnicity

_ Female __ Caucasian __ Latina/Latino_ Native American

_ Male _ African American _ Asian American _ Arab American
_OOOOOOOOOOOOOGO

Please return your questionnaire in the enclosed envelope to me:
Celeste Sturdevant Reed
305 E. Middle Street
Williamston, MI 48895

Remember to send in your lottery slip to be included in the pool to win a national NASW basic
membership.

135

 

 

APPENDIX B

SCALES AND MEASURES

 

136

APPENDIX B1
ORGANIZATIONAL COMMITMENT

Commitment to Occupation from Meyer, Allen, & Smith (1993)
7 point scale: 1 = strongly agree to 7 = strongly disagree [“nursing” is replaced with
“social work”]

Affective Commitment [alpha = .82]

99:593"?

Co
1

2
3.
4

5

6

No
1

PP!”

5"

[Social work] is important to my self image.

I regret having entered the [social work] profession. (R)
I am proud to be in the [social work] profession.

I dislike being a [social worker]. (R)

I do not identify with the [social work] profession. (R)
I am enthusiastic about [social work].

ntinuance Commitment [alpha = .74]
. I have put too much into the [social work] profession to consider changing now.
. Changing professions now would be difﬁcult for me to do.

Too much of my life would be disrupted if I were to change my profession.

. It would be costly for me to change my profession now.
. There are no pressures to keep me from changing profession (R)
. Changing professions now would require considerable personal sacriﬁce.

rrnative Commitment [alpha = .83]
. I believe people who have been trained in a profession have a responsibility to stay in

that profession for a reasonable period of time.

I do not feel any obligation to remain in the [social work] profession. (R)

I feel a responsibility to the [social work] profession to continue in it.

Even if it were to my advantage, I do not feel that it would be right to leave [social
work] now.

I would feel guilty if I left [social work].

I am in [social work] because of a sense of loyalty to it.

R = reverse coded

137

APPENDIX B2
JOB INVOLVEMENT

Kunungo (1982) Job-Involvement Scale
Questionnaire Format

8.

9.

. The most important things that happen to me involve my present job.

To me, my job is only a small part of who I am.

I am very much involved personally in my job.

I live, eat, and breathe my job.

Most of my interests are centered around my job.

I have very strong ties with my present job which would be difﬁcult to break.
Usually I feel detached from my job.

Most of my personal life goals are job-oriented.

I consider my job to be very central to my existence.

10. I like to be absorbed in my job most of the time.

138

 

APPENDIX B3
INTENT TO LEAVE

Chatman (1991) Intent to Leave
1. To what extent would you prefer another more ideal job than the one you now work
in?

2. To what extent have you thought seriously about changing organizations since
beginning to work here?

3. How long do you intend to remain with this organization?

4. If you have your own way, will you be working for this organizations three years
from now?

139

 

APPENDIX B4
WORK VALUES

Super (1977) Work Values Inventory.

The statements below represent values which people consider important in their work.
These are satisfactions which people often seek in their jobs or as a result of their jobs.
They are mot all considered equally important; some are very important to some people
but of little importance to others. Read each statement carefully and indicate how
important it is or would be for you.

5 means “YE! Important”

4 means “Important”

3 means “Moderately Important”
2 means “of Little Importance”

1 means “Unimportant”

Work in which you...

[Achievement]

[Security]

[Prestige]

[Independence]

[Economic Returns]

[Altruism]

get the feeling of having done a good day’s work
know by the results when you have done a good job
see the results of your efforts

know your job will last

are sure of always having a job

are sure of another job I the company if your present job ends
gain prestige in your ﬁeld

know that others consider your work important

are looked up to by others

have freedom in your own area

make your own decisions

are your own boss

can get a raise

are paid enough to live right

have pay increases that keep up with the cost of living
help others

feel that you have helped another person

add to the well-being of other people

140

APPENDIX B5
CLIENT SERVICE FACTORS

Many things have an impact on the services that are delivered to clients—including agency policies, staff
experience, client problems, etc. The factors that have an impact can improve or impede services to clients,
or sometimes do both.

For the list below, please identify the type of impact that each item has on the clients you serve. Please
select only one answer for each item. In the case of multiple inﬂuences, select the response that is true for

the most clients.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Generally has a Generally has a
positive or negative or Generally has little
beneﬁcial impact on detrimental impact on or no impact on
Source of inﬂuence on client client service client service client service
service decisions decisions decisions decisions
Agency mission or purpose + - 0
Agency rules and + _ 0
requirements
Agency public image + - 0
The predictability of external + _ 0
funds
Legal or accreditation + - 0
requirements
Board member and/or donor + _ 0
preferences
Funder(s) requirements + - 0
Staff commitment to the + _ 0
agency mrssron.
Staff knowledge and skills + - 0
Staff motivation and + _ 0
particrpation
Staff cooperation each other + - 0
Staff collaboration with + _ 0
others
Staff professional judgment + - 0
Severity of client’s problem + - 0
Client’s ability to pay or + _ 0
insurance coverage
Client’s motivation and + _ 0
particrpatron
Client’s level of family or + - 0
socral support
Local community sentiments + - 0
Availability of volunteers + - 0
Availability of and/or access + _ 0
to other community servrces

 

 

 

 

141

 

 

 

APPENDIX C

FACTOR ANALYSIS RESULTS FROM UNUSED SCALES

142

 

Table 33
Client Service Decision-Making Practices: Agency Attributes

 

 

 

 

 

 

 

 

 

 

 

Communality
Items Factor Loadings Estimates
S6. Agency policies and rules .774 .677
S7. Agency public image .714 .586
S5. Agency mission or purpose .669 .612
Coefﬁcient alpha = .71
Table 34
Client Service Decision-MakingPractices: Client Attributes
Communality
Items Factor Loadings Estimates
820. Client’s level of family or social support .688 .552
S23. Availability of and/or access to other community
services .688 .531
S 19. Client’s motivation and participation .667 .527
S17. Severity of client’s problem .493 .477
S21. Local community values and sentiments .464 .442
Coefﬁcient alpha = .52
Table 35
Work Reward Practices: Bonuses
Communality
Items Factor Loadings Estimates
R2. Bonuses for performing quality work .784
R3. Bonuses for performance related to budgets and
schedules .780
Coefﬁcient alpha = .75
Table 36
Factor Analysis Results
Work Reward Practices: Status and Security
Communality
Items Factor Loadings Estimates
R20. Membership in high status task forces .694 .554
R19. Expenses paid to present papers at conferences .641 .527
R18. Formal commendations and rewards .588 .478
R15. Freedom concerning job duties .501 .490
R7. Policy or practice of reassigning staff rather than
laying people off .482 .410
R6. Job security .435 .409

 

Coefﬁcient alpha = .74

143

 

 

Appendix D

Null Results

144

Table 37
Null Results for Regression Equation Hlb2: Use of Professional Factors for Client Service
Decisions in the Nonproﬁt Sector

 

 

 

Predictor B Beta Sig.
Constant 21.896 .000
Private F or-proﬁt Sector -.5 80 -.022 .653
Public Sector .268 .016 .736

R .028
R2 .001
~R2 -.004

 

F =.849; sig. = .429

Table 38
Null Results for Regression Equation H2b: Use of Legal Factors for Client Service
Decisions in the Child Welfare Industry

 

 

 

Predictor B Beta Sig.
Constant 7.412 .000
Child Welfare .388 .079 .243

R .079

R2 .006

~R2 .002

 

F = .830; sig. = .437

Table 39
Null Results for Regression Equation H2c: Use of Financial Factors for Client Service
Decisions in the Mental Health Industry

 

 

 

Predictor B Beta Sig.
Constant 42.745 .000
Mental Health 1.065 .082 .228

R .082

R2 .007

~R2 .002

 

F = .958; sig. = .385

 

 

 

Table 40
Null Results for Regession EtLuation H3d: Position Predicting Autonomy
Predictor B Beta Sig.
Constant 12.339 .000
Administrative position .492 .075 .120
R
R2
~R2

 

F = 2.423; sig. = .124

145

Table 41
Null Results for Regression Equation: Predicting Job Involvement from Work Practices and

 

 

 

Work Context
F for)
Predictor Beta R R2 F ) R2 R2
Step 1 - Work Practices
Professional Rewards .00273
Financial Rewards .00247
CS Professional Practices .00976
CS Financial Practices -.00543
CS Legal Practices -.00501 .084 .007
Step 2 — Constant (Nonproﬁt ",1 ‘
Sector) [B]
Public Sector -.254
Private For-proﬁt Sector -1.284 .140 .020 1.331 .012 1.020

 

146

BIBLIOGRAPHY

147

l’

BIBLIOGRAPHY

Alperin, D. E. (1992, January). Family service agencies: Responding to change in a
conservative decade. _F_amilies in Society: The Journal of Contemporary Human Services,
pp. 32-39.

Alperin, D. E. (1993). Family service agencies: Responding to need in the 19805. Social
Work, ﬂ, 597-602.

Atkinson, R. (1989). Altruism in nonproﬁt organization_s (PONPO Working Paper No.
145 and ISPS Working Paper No. 2145). New Haven, CT: Yale University, Institute for
Social and Policy Studies.

Bartol, K. M. (1979). Professionalism as a predictor of organizational commitment, role

stress, and turnover: A multidimensional approach. Academy of Management J oumal, g,
815-821.

 

Ben-Shem, I., & Avi-Itzhak, T. E. (1991). On work values and career choice in freshmen
students: The case of helping vs. other professions. Journal of Vocational Behavior. 32,
369-3 79.

Blau, P. M., & Scott, W. R. (1962). Formal gganizations: A comparative approach. San
Francisco: Chandler.

Bowen, D. E., Ledford, G. E. Jr., & Nathan, B. R. (1991). Hiring for the organization not
the job. Academy of Management Executive, 5, 35-51.

Boxx, W. R., Odom, R. Y., & Dunn, M. G. (1991). Organizational values and value
congruency and their impact on satisfaction, commitment and cohesion. Public Personnel
Management, 2_0, 195-205.

Bretz, Jr., R. D., Ash, R. A., & Dreher, G. E. (1989). Do people make the place? An
examination of the attraction-selection-attrition hypothesis. Personnel Psychology, :12,
561-5 8 1 .

Brown, K. M. (1986). Applied empirical research on nonproﬁt oLganization management:
Survey and recommendationa (Working Paper No. 1). San Francisco: University of San
Francisco, Institute for Nonproﬁt Organization Management.

Cable, D. M., & Judge, T. A. (1994). Pay preferences and job search decisions: Person-
organization ﬁt perspective. Personnel Psychology, 42, 317-348.

Chao, G. T., O’Leary-Kelly, A., Wolf, S., Klein, H. J ., & Gardner, P. D. (1994).
Organizational socialization: Its content and consequences. Journal of Applied
Psychology, B, 730-743.

Chatrnan, J. A. (1989). Improving interactional organizational research: A model of
person-organization ﬁt. Academy of Management Review. 14, 333-349.

Chatrnan, J. A. (1991). Matching people and organizations: Selection and socialization in
public accounting ﬁrms. Administrative Science Quarterly. 36, 459-484.

 

148

Clark, R. E., & Dorwart, R. A. (1992). Competition and community mental health
agencies. Journal of Health Politics. Policy and La_v_v_, ﬂ, 517-540.

Clark, R. E., Dorwart, R. A., & Epstein, S. S. (1994). Managing competition in public
and private mental health agencies: Implications for services and policy. Milbank
Quarterly, 12, 653-678.

Cohen, J ., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the
behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.

Cramptom, S. M., & Wagner, J. A. 111(1994). Percept-percept inﬂation in
microorganizational research: An investigation of prevalence and effect. Journal of
Applied Psychology, 12, 67-76.

Dawis, R. V. (1991). Vocational interests, values, and preference. In Dunnette & Hough
(Eds), Hﬂdbtgt of industrr'aLand organizational psychology_(2nd Ed., Vol. 2, pp. 833-
871). Palo Alto, CA: Consulting Psychologists Press.

 

Drucker, P. F . (1974). Managing the public service institution. Public Interest. 3, 43-60.

Erickson, A. G. (1987). Family services. In Minahan (Ed), Encyclopedia of social work
(18th Ed., Vol. 1, pp. 589-593). Silver Spring, MD: National Association of Social
Workers.

Ginsberg, L. (1992). Social work almanac. Washington, DC: National Association of
Social Workers.

Godderis, J. H. (1988). Compensating differentials and self-selection: An application to
lawyers. Journal of Political Economy. 26, 411-427.

Grenbjerg, K. A., Chen, T. H., & Stagner, M. W. (1995). Child welfare contracting:
Market forces and leverage. Social Service Review. Q, 583-613.

Hall, R. H. (1967). Some organizational considerations in the professional-organizational
relationship. Administrative Science Quarterly, g, 461-478.

Hasenfeld, Y. (1983). Human service organizationa. Englewood Cliffs, NJ: Prentice-Hall.

Hodgkin, C. (1991). Policy and paper clips: Rejecting the lure of the corporate model.
Nonproﬁt Management & Leadership, 3, 415-428.

James, L. (1982). Aggregation bias in estimates of perceptual agreement. Journal of
Applied Psycholgy, 6_7, 219-229.

James, L. R., & Brett, J. M. (1984). Mediators, moderators and tests for mediation.
Journal of Applied Psychology, 62, 307-321.

J eavons, T. H. (1992). When the management is the message: Relating values to
management practices in nonproﬁt organizations. Nonproﬁt Management & Leadership,
2, 403-417.

Judge, T. A., & Bretz, R. D. (1992). Effects of work values on job choice decisions.
Journal of Applied Psychology, 77, 261-271.

149

Kanter, R. M., & Summers, D. V. (1987). Doing well while doing good: Dilemmas of
performance measurement in nonproﬁt organizations and the need for a multiple-
constituency approach. In Powell (Ed.), The nonproﬁt sector: A research handbook (pp.
154-166). New Haven, CT: Yale University Press.

 

Kerr, S. (1988). Some characteristics and consequences of organizational reward. In
Schoorman & Schneider (Eds), Facilitating work effectiveness (pp. 43-76). Lexington,
MA: Lexington Books.

Kettner, P. M., & Martin, L. L. (1994). Purchase of service at 20: Are we using it well?
Public Welfara, QB), pp. 15-20 & 42.

Khoj asteh, V. (1993). Motivating the private vs. public sector managers. Pablic Personnel
Management, _2__2, 391-401.

Kramer, R. M. (1994). Voluntary agencies and the contract culture: “Dream or
nightmare?” Social Service Review. 68, 33-60.

r

Kristoff, A. L. (1996). Person-organization ﬁt: An integrative review of its
conceptualizations, measurement, and implications. Personnel Psychology, 42, 1-49.

Kunungo, R. N. (1982). Work alienation: An integrative approapk. New York: Praeger.

Liederman, D. S. (1995). Child welfare overview. In Edwards & Hopps (Eds),
Encyclopedia of social work (19th Ed., Vol. 1, pp. 424-433). Washington, DC: National
Association of Social Workers.

Lin, A.M.P. (1995). Mental health overview. In Edwards & Hopps (Eds), Encyclopedia
of social work (19th Ed., Vol. 2, pp. 1705-1711). Washington, DC: National Association
of Social Workers.

 

Lodahl, T. M., & Kejner, M. (1965). The deﬁnition and measurement of job involvement.
Journal of Applied Psychology, Q, 24-33.

Loscocco, K. A. (1989). The instrumentally oriented factory worker: Myth or reality?
Work and Occupations, _l__6_, 3-25.

 

Majone, G. (1984). Professionalism and nonproﬁt organizations. Journal of Heath
Politics. Policy and Law, ﬂ, 19-40.

McCauley, C. D., & Hughes, M. W. (1991). Leadership challenges for human service
administrators. Nonproﬁt Management & Leadership. 1, 267-281.

Meglino, B. M., Ravlin, E. C., & Adkins, C. L. (1989). A work values approach to
corporate culture: A ﬁeld test of the value congruence process and its relationship to
individual outcomes. Journal of Applied Psychology, 14, 424-432.

Meyer, J. P., & Allen, N. J. (1991). A three component conceptualization of
organizational commitment. Human Resource Mangement, _1_, 61-98.

 

 

Meyer, J. P., Allen, N. J ., & Smith, C. A. (1993). Commitment to organizations and
occupations: Extension and test of a three-component conceptualization. Journal of
Applied Psychology, 18, 538-551.

150

Mowday, R. T., Steers, R. M., & Porter, L. W. (1979). The measurement of
organizational commitment. Journal of Vocational Behavior, ﬂ, 224-247.

Nalbandian, J ., & Edwards, J. T. (1983). The values of public administrators: A
comparison with lawyers, social workers, and business administrators. Review of Public
Personnel Administration, 4, 114-127.

 

Newman, W. H., & Wallender, H. W. I]1(1978). Managing not-for-proﬁt enterprises.
Academy of Management Review, 3, 24-31.

Nord, W. R., Brief, A. P., Atieh, J. M., & Doherty, E. M. (1988). Work values and the
conduct of organizational behavior. Research in Organizational Behavior. m, 1-42.

O'Connell, B. (1988). Values underlying nonproﬁt endeavor. In O'Neill & Young (Eds),
Educating managers of nonproﬁt organizations (pp. 155-162). New York: Praeger.

O’Neill, M., & Young, D. R. (1988). Educating managers of nonproﬁt organizations. In
[editors?] (Eds), Educating managers of nonmroﬁt organizations (pp. 1-21). New York:
Praeger.

O’Reilly, C. A., Chatrnan, J. A., & Caldwell, D. M. (1991). People and organizational
culture: A Q-sort approach to assessing person-organization ﬁt. Academy of Management
Journal, ﬂ.

Perry, J. L., & Wise, L. R. (1990). The motivational bases of public service. Public
Administration Review, _5_Q, 367-373.

 

Porter, L. W., & Smith, R]. (1970). The etiology of organizational commitment
(unpublished manuscript). Irvine: University of California.

Porter, L. W., Steers, R. M., Mowday, R. T., & Boulian, P. V. (1974). Organizational
commitment, job satisfaction, and turnover among psychiatric technicians. Journal of
Applied Psychology, ﬂ, 603-609.

Posner, B. Z. (1992). Person-organization values congruence: No support for individual
differences as a moderating inﬂuence. Human Relations, 9, 351-361.

Posner, B. Z., & Schmidt, W. H. (1996). The value of business and federal government
executives: More different than alike. Public Personnel Management, 25:, 277-289.

Preston, A. E. (1989). The nonproﬁt worker in a for-proﬁt world. Journal of Lam
Economics, 2, 438-463.

Ravlin, E. C., & Meglino, B. M. (1987). Effect of values on perception and decision
making: A study of alternative work values measures. Journal of Applied Psychology, 12,
666-673.

Rawls, J. R., Ullrich, R. A., & Nelson, 0. R. Jr. (1975). A comparison of managers
entering or reentering the proﬁt and nonproﬁt sectors. Academy of Management J oumal,
ﬂ, 616-623.

Reed, C. S. (1992). Report to the Executive Staff of the Children’s Center. Detroit, MI:
Children's Center.

151

Reed, C. S., Young, W. R., & McHugh, P: P. (1994). A comparative look at dual
commitment: An international study. HumaaRelations, ﬂ, 1269-1294.

Roberts, K. H., Hulin, C. L., & Rousseau, D. M. (1978). Developingand interdiscipliargy
science of organizations. San Francisco: Jossey-Bass.

Rokeach, M. (1973). The nature of hummalues. New York: Free Press.

Rynes, S. L. (1987). Compensation strategies for recruiting. Topics in Total
Compensation. 2, 185—196.

Schneider, B. (1987). The people make the place. Personnel Psychology, 49, 437-453.

Scott, W. R. (1987). Organizations: Rational natural and open systems. Englewood
Cliffs, NJ: Prentice-Hall.

 

Smith, D. H. (1991). Four sectors or ﬁve: Retaining a member-beneﬁt sector. Nonproﬁt
and Voluntary Sector Quartem, 29, 137-150.

Smith, S. R. (1989). The changing politics of child welfare services: New roles for the
government and the nonproﬁt sectors. Child Welfare, 62, 289-299.

Sorensen, J. E., & Sorensen, T. L. (1974). The conﬂict of professionals in bureaucractic
organizations. Administrative Science Quarterly. Q, 98-106.

Steel, R. P., & Rentsch, J. R. (1997). The dispositional model of job attitudes revisited:
Findings of a 10-year study. Journal of Applied Psychology, 8, 873-879.

Super, D. E. (1970). Work Values Inventory. Boston: Houghton Mifﬂin.
Super, D. E. (1977). Work Values Inventory. Chicago, IL: Riverside.

Taylor, S. M., & Giannantonio, C. M. (1993). Forming, adapting, and terminating the
employment relationship: A review of the literature from individual, organizational, &
interactionist perspectives. Journal of Management, l_9_, 461-515.

Thompson, J. D. (1967). Organizations inaaction. New York: McGraw-Hill.

Vancouver, J. B., & Schmitt, N. W. (1991). An exploratory examination of person-
organization ﬁt: Organizational goal congruence. Personnel Psychology, 14, 333-352.

Weaver, C. N., & Franz, R. S. (1992). Work-related attitudes of entrepreneurs, public,
and private employees. Psychological Reports, 7Q, 387-390.

Weisbrod, B. A. (1983). Nonproﬁt and proprietary sector behavior: Wage differentials
among lawyers. Journal of Labor Economics. _1_, 246-263.

Wilson, M. I. (1991). The state of nonproﬁt Michigaa 1991. East Lansing: Michigan
State University.

Wilson, P. A., Voth, V., & Hudson, W. W. (1980). Professionals and the bureaucracy:

Measuring the orientations of social workers. Journal of Human Service Research. 5. 15-
3 1 .

152

Winegar, N. (1993, March). Managed mental health care: Implications for administrators
and managers of community-based agencies. Families in Socieg: The Journal of
Contempoaary Human Services (pp. 171-177).

Wittmer, D. (1991). Serving the people or serving for pay: Reward preferences among
government, hybrid sector, and business managers. Public Productivity & Management
Review, _13, 369-383.

Witty, C. J. (1989). Resilience aad survivala: Executive careers in major California
nonproﬁt organiations (Working Paper No. 7). San Francisco: University of San
Francisco, Institute for Nonproﬁt Organization Management.

Young, D. R. (1983). If not for proﬁL for what? Lexington, MA: LexingtonBooks.

Young, D. R. (1984). Performance and remds in the nonproﬁt sector: Evaluation

com aeraation. and personnel incentives (PONPO Working Paper No. 79 and ISPS
Working Paper No. 2079). New Haven: Yale University, Institution for Social and Policy
Studies.

 

 

153

‘-

   

IIIIIIIIIIIIIIIIIIIII

in]11111111111911111111.1I