distributive and procedural justice as predictors of js

7/25/2019 Distributive and Procedural Justice as Predictors of Js

1/13

RESEARCH NOTES

Academy

o

Management /ournai

1992, Vol. 35, No. 3. 626- 637 .

DISTRIBUTIVE AND PROCEDURAL JUSTICE AS

PREDICTORS OF SATISFACTION WITH PERSONAL AND

ORGANIZATIONAL OUTCOMES

DEAN B. McFARLIN

PAUL D. SWEENEY

Marquette Universi ty

Distributive justice was found to be a more important predictor of two

personal outcomes, pay satisfaction and job satisfaction, than proce-

dural justice, whereas the reverse w as true for two organizational out-

comes organizational commitment and subordinate s evaluation of

supervisor. However, procedural and distributive justice also inter-

acted in predicting organizational outcomes. We discuss limitations of

this study and directions for future research.

Folger and Konovsky captured the key distinction regarding justice in

work organizations, noting that distribu tive justice refers to the perce ived

fairness of the amounts of compensation employees receive; procedural jus-

tice refers to the perceived fairness of the means used to determine those

am ou nts (1989: 115). However, few studies have examined how both dis-

tribu tive and p roc ed ura l justice affect ou tcom es (e.g., Folger & Kon ovsky,

1989; Greenberg, 1987a; Konovsky, Folger, & Crop anzan o, 1987). In add i-

tion, most research has focused on legal rather than work-related issues. For

example, Greenberg and Folger (1983) showed that defendants viewed trial

verdicts (distributions) positively if they were seen as the result of fair pro-

cedu res , an effect called the fair process effect (cf. M usante, Gilbert, &

Thibaut, 1983). Folger and Konovsky pointed out that this research has

suggested different predictive roles for procedural and distributive justice.

In particular, studies have found distributive justice to predict satisfaction

with specific, personal outcomes, like case verdicts, better than procedural

justice. The reverse is true, however, when people make more general eval-

uatio ns of, for instanc e, legal institutio ns or their rep resenta tives (Lind &

Tyler, 1988).

In fact, the few studies that have been done in organizational settings

have tended to support the notion that the predictive roles of procedural and

distributive justice depend, at least in part, on the nature of the outcome in


2/13

1992 McFarlin and Sweeney

7

faction, and perceived conflict than distributive justice. Konovsky and col-

leagues (1987] found that procedural justice predicted organizational com-

mitment, but not pay satisfaction, whereas the reverse was true for distrib-

utive justice. Similarly, Folger and Konovsky (1989) found that procedural

justice accounted for more variance in organizational co mm itment an d trust

in a supervisor than distributive justice, wbereas tbe reverse was true for

satisfaction w itb a pay raise. i

Ov erall, tbese re sults suggest tbat pr oce dur al justice may be la more

important predictor tban distributive justice of outcomes related to evaluat-

ing a company as an institution and its representatives, sucb as organiza-

tional commitment and trust in supervisor. In contrast, distributive justice

may be a more important predictor of personal outcomes, like satisfaction

witb pay level, tban procedural justice.

Interestingly, tbese results reflect wbat we migbt call a main effect ap-

proacb to examining tbe predictive roles of distributive and procedural jus-

tice in field research. Tbere are few studies of tbe interactive effects of tbe

two types of justice. In fact, tbe existing researcb evidence about interactive

effects com es from laboratory stu dies. For exam ple, in a recent investigation

by Greenberg (1987a), subjects worked on a task in wb icb distributive justice

(pay levels) and procedural justice (bow pay levels were determined) were

manipulated. He found an interaction of tbe following form; subjects saw

bigb p ay levels as fair regardless of proc edu res bu t saw low pay levels as fair

only when fair procedures were used. Perceptions of unfair treatment were

maximized wben pay was low and procedures were unfair. Similarly, in a

laboratory study, Cropanzano and Folger (1989) found tbat resentnient was

bigbest wben subjects perceived tbat unfair procedures prevented tbem

from receiving big b rew ards for task perform ance. j

Botb Greenberg (1987a) and Cropanzano and Folger (1989) pointed out,

bowever, tbat researcb was needed to determine wbetber tbese laboratory

interactions generalize to organizational settings. An attempt at suqb gener-

alization was tbe focus of our researcb. Anotber key researcb question was

wbetber existing tbeory allows for a precise prediction regarding tbe form

sucb interactions may take. We used referent cognitions tbeoryj (Folger,

1986) to derive just sucb a prediction.

Briefly, according to referent cognitions tbeory, individuals evaluate

tbeir work experiences by reflecting on w bat migbt bave be en un der dif-

ferent circumstances and conditions (Folger, 1986). One way erriployees

migbt accomplisb sucb a comparison would be to cognitively simulate bow

tbeir current work situation migbt be different bad tbeir organization used

fairer procedures. In fact, Cropanzano and Folger (1989) suggested' tbat ref-

erent cognitions tbeory offers a potential conceptual framework for tbe in-

teractive effects of distributive and procedural justice.


3/13

628 Academy

o

Management Journal August

contrast tbis situation to tbe more positive outcomes tbat tbey would bave

obtained bad tbe decision maker used fair allocation procedures. On tbe

otber band, referent cognitions tbeory predicts tbat wben people perceive

procedures to be fair, resentment will be minimal, even when distributive

justice is low. Unde r suc b cond itions, it wo uld be more difficult for em ploy-

ees to envision alternative fairer procedures tbat could bave led to better

outcomes.

To test tbis intera ction pattern in a field setting, we su rveyed a group of

bank employees. In addition to measuring procedural and distributive jus-

tice, we assessed perceptions of two important personal outcomes, pay and

job satisfaction, and two organizational outcomes: organizational commit-

ment and subordinate's evaluation of supervisor. If Cropanzano and Folger's

(1989) application of referent cognitions tbeory is correct, we would expect

tbat tbe evaluation of tbese outcomes would be most negative wben botb

distributive and procedural justice are low. But we would expect positive

evaluations wben procedural justice is bigb, regardless of tbe level of dis-

tributive justice.

A more speculative prediction is tbat tbe interaction pattern referent

cognitions tbeory suggests will be clearest wben tbe focus is on organiza-

tional, as opposed to personal, outcomes. Because tbe former focus on tbe

sources of tbe procedures tbat affect employees, sucb as supervisors and tbe

employing organization as an institution, tbey provide a clearer target for

retaliation tban outcomes of a more personal nature. In otber words, allow-

ing employees to evaluate an organization as an institution gives tbem tbe

opportunity to blame tbe parties responsible for creating tbe rules of tbe

gametbe unfair procedurestbat have resulted in a lack of distributive

justice. Referent cognitions tbeory suggests tbat employees' resentment is

not only a function of low d istributive justice, but also of tbe opp ortu nity to

blam e m isfor tune on tbe actions of otbers ratber tba n on tbeir ow n be-

bavior (Cropanzano

Folger, 1989).

Finally, sbould tbe interactions we predicted fail to materialize, we

determined tbat our focus would sbift to tbe interpretation of main effects.

In view of prior researcb (e.g., Alexander

Ruderman, 1987), we predicted

tbat distributive and procedural justice would eacb produce strong main

effects for a variety of outcomes. Given Folger and Konovsky's (1989) and

Konovsky and colleagues' (1987) results, bowever, we expected procedural

justice to produce a stronger main effect in predicting organizational out-

comes tban distributive justice. We expected tbe reverse to be true witb

personal outcomes.

M THO S


4/13

1992 McFarlin and Sweeney 629

resulting in a response rate of 61 percent. Tbe respondents' mean age was

33.8 years (s.d. = 12.1); tbeir m ean com pan y tenu re was 6.3 years (s.d. =

5.8); and 74 percent were women. Tbese respondent demograpbics com-

pared favorably witb company statistics for tbe work force as wboje, sug-

gesting tbat problems resulting from response bias were unlikely to be a

major tbreat. In tbe total work force, 68 percent of tbe employees were

w om en, w itb an average age of 34.6 years and an average comp any te nur e of

6.9 years.

Measures

Predictors were measures of global distributive and procedural justice.

Criterion variables w ere two m easures reflecting eva luations of tbe surveyed

organization and its representatives and two measures of satisfaction witb

personal outcomes. ;

Distributive justice

We used five of tbe six items from Price and Muel-

ler's (1986) Distributive Justice Index. Tbese items ask workers to indicate

tbe extent to wbicb tbey bave been fairly rewarded in view of tbeir respon-

sibilities, experience, job stress, effort, and performance. Rewards in tbe

index are defined broadly, witb money, praise, and recognition all being

listed as exa m ples . An exam ple of tbe item format is as follows: How fair

bas [company name] been in rewarding you wben you consider tbe amount

of effort tbat you put into your w ork ? ( 1 = very unfair, 5 = very 'fair).

Procedural justice

We constructed a four-item procedural justice scale

tbat use d tbe sam e basic format and five-point respo nse scale as tbeiDistrib-

utive Justice Index. Respondents indicated tbe extent to wbicb tbe general

proc edu res used to com m unicate performance feedback, determine, pay in-

creases, and evaluate performance and prom otability were fair. '

Organizational outcomes

Organizational commitment was assessed

using four items from Cook and Wall's (1980) commitment scale. Tbese

items ask workers to express tbeir agreement or disagreement witb various

statem ents (e.g., I feel myself to be a part of tbis com pany ), using five-point

scales (1 = strongly disagree, 5 = strongly agree). Tbe second organizational

outcome, sub ordina te's e valuation of supervisor, was assessed w itb a single

item: O verall, I tbink my boss is a poor performer (1 = strongly agree, 5 =

strongly disagree). Several nationa l prob ability studies (e.g, Qu inn

Staines,

1979) bave previo usly used an essentially equiva lent item. j

Personal outcomes

We used tbree items to assess pay JeveJ satis/a ction .

Two of tbe items used five-point agree-disagree formats; tbese were : Tb e

pay for my job is exc ellent and I am satisfied w itb my current pay. Tbe

tbird item asked resp ond ents bo w tbey felt about tbeir pay (7 = deligbted,

1 = terrible ; cf. An drew s & W itbey, 1976). Since tbey bad different mea -

surem ent scales, we standa rdized tbese tbree items before com bining tbem .


5/13

630 Academy

of

Management Journa) August

RESULTS

Descriptive Statistics

Table 1 prese nts m ean s, correlations, and reliability coefficients, w bere

app licable, for all study variables. Tbe reliabilities for m ulti-item scales w ere

generally very good, witb alpbas ranging from .75 to .92. One concern tbat

migbt be raised, bowever, is tbe correlation between procedural and distrib-

utive justice (.67). To belp allay tbis concern, we would make tbree points.

First, prior researcb bas successfully sbown tbat tbese scales predict differ-

ent dependent measures, suggesting tbat tbey are independent constructs.

Second, prom inent researcbers bave claimed tbat altbougb tbey are separate

constructs, in practice a sizable correlation sbould be found between tbem

(e.g., Folger, 1987). For example, Folger cited an important study by Tyler

(1984) in wbicb tbe correlation between distributive and procedural justice

was

77

Tbird, we conducted an empirical analysis tbat sbeds some addi-

tional ligbt on tbis issue.^

Regression nalys is Results

Separa te three-s tep, bierarcb ical regression analyses were performed for

eacb outcome variable. At step 1, we entered five control variables: salary,

job type, age, gender, and tenure. Staines, Pottick, and Fudge (1986) cau-

tioned tbat tbese types of variables need to be controlled for, given tbeir

general potential to inflate or suppress relations between otber variables.

Salary and job typ e were of special co nce rn, given tbeir pote ntia l to affect tbe

particular outcome variables used in tbis study.

As Table 2 sbows, age significantly predicted pay level satisfaction, job

satisfaction, and organizational commitment. Consistent witb earlier re-

searcb findings (Weaver, 1980), our finding was tbat older workers te nd ed to

bave bigber commitment and satisfaction tban younger workers. Job type

also significantly predicted job satisfaction, evaluation of supervisor, and

organizational commitment, witb managers being more positive tban cleri-

cal employees.

At step 2, distributiv e and proce dural justice we re entered . Botb types of

justice were significant predic tors of eacb outcome variable. Con sistent w itb

Folger and K onovsky s (1989) conten tions, bow ever, d istributive justice

^

Podsakoff and Organ (1986) reviewed techniques that address method variance, including

a procedure called the single factor approach. The logic of this approach is that if method

variance ac coun ts for the relations betw een two or more variables, a factor an alysis should yield

a single global (method) factor. Since researchers have generally viewed this approach as weak

but as possibly prov iding some useful information, w e combined it with a stronger confirmatory

factor a naly sis, u sing LISREL VI (Joreskog

Sorbom, 1985), which tests models that increase in


6/13

992

Mc arlin and Sweeney

6 3

I

(8

tn

u

I

'ft

(S

o

O5

CO

o

CO

t H

f H

o

o

1

2

o

o

CO

CO

i n

1

CM

1

CO

o

i n

r H

CM

1

CM

o

o

i n

CO

1

i n

i n

CO

CO

i n

o

o

o

o

o

CO

J"

CO

e

00

00

o

1

i n

o

O l

o

CM

CO

CM

o

o

o

B

c

"cr

CO

CM

i n

CM

O

o

1

r H

r H

O l

i n

i n

tv

CO

;o

r H

C

t

o

s

a

B

O

cd

3

te

s

e

n

CO

t v

r H

CO

q

t n

q

C31

O

1

r-*

i n

1

oa

r H

t v

i n

CO

so

v

i n

tv

CM

CO

O l

i n

CO

r H

r H

CM

CM

CM

r H

1

r H

r H

r H

r H

CO

CO

CO

CO

"B

1

[

J

B

U

O

z

iq

CO

i n

CO

CO

i n

q

CM

1

CO

o

CO

o

CO

CO

CO

r H

o:

v .

Z

a

t

9

t v

CO

CJl

" :

o

CO

t v

CO

CO

i n

q

o

CO

o

1

r H

o

t v

o

o

c:i

u

s

c

B

i

e

l

3

B

o

i

i2

B

i

c

01

o

B

cd

B

o

5

II

CM

T3

B

cd

B

cd

2 .

"3 01

oi -o

in 01

o 00

V r?

c

CO

u

g

cn

03

tH

CO

CD

O

CO

x

03

0 0

CO

C

CO

B

II

CM

-a

a

cd

L

cn 01

B a

o >>


7/13

6

Academy ofManagement Journal

TABLE 2

Results of Hierarchical Regression Analyses

August

Independent

Variables

Step 1

Tenure

Gender

Age

Salary

Job type

AR^

Step 2

Procedural justice

Distributive justice

AR2

Step 3

AR^ for interaction

of procedural and

distributive justice

F

f

Personal

Pay Level

Satisfaction

.02

- . 0 2

.16***

.0 0

.0 8

.04***

.14***

.52***

.37***

.0 0

56.29***

8 63

Outcomes

Job

Satisfaction

- . 0 3

.02

.31***

.07

.15***

.12***

.18***

.30***

.18***

.0 0

32.68***

8 6 4

Organizational Outcomes

Subordinate s

Evaluation

of Supervisor

.0 0

.0 5

- . 0 2

- . 0 5

.12**

.03**

.23***

.15**

.11***

.01*

14.74***

8 626

Organizational

Commitment

- . 0 2

- . 0 3

.23***

.0 8

.10*

.08***

.34***

.23***

.26***

.01**

42.19***

8,631

* p < .05

** p < .01

*** p < .001

tended to be a more important predictor of personal outcomes (pay and job

satisfaction) than procedural justice. In addition, procedural justice was a

more important predictor of both organizational outcomes than distributive

justice (see Table 2).^

In step 3, we entered th e interaction of distributive and proc edu ral

justice. As Table 2 shows, the interaction terms for both evaluation of su-

pervisor and organizational commitment were significant. When plotted (cf.

Cohen Cohen, 1983), both outcomes yielded the same interaction pattern.

^ Following a reviewe r's suggestion, we condu cted our regression analyses again, en tering

distributive and procedural justice on separate steps to more closely examine their relative

predictive power. The results of tbese analyses were consistent with those given in Table 2;

procedural justice was a stronger predictor of organizational outcomes tban distributive justice,

with the reverse being true for personal outcomes. Specifically, witb procedural justice con-

trolled, distributive justice accounted for an additional 5 percent of tbe variance in job satis-

faction and 14 percent of tbe variance in pay level satisfaction; wben tbe order of entry was


8/13

1992

McFarlin and Sweeney

6 33

FIGURE I

Interactive EffectsofD istributiveandProcedural Justiceon

Organizational ommitment

Organizational

Commitment

19.0 -

18.5 -

18.0

-

17.5

-

16.5

-

16.0

-

15.5

15.0

-

14.5

Higb procedural justice

Low procedural justice

Low

Higb

Distributive Justice

To illustrate this pattern,wepresent theinteraction fororganizationalcom-

mitment in Figure 1, which shows that although employees who felt that

proced ures w ere fair tende d tohave higher levelsof organizational commit-

ment than thosewho felt procedures were unfair, thisgap wasm uch larger

when distributive justice was low. Specifically, when procedural justice was

low, organizational commitment varied considerably

as a

function

of dis-

tributive justice. When procedural justice washigh, how ever, organizational

commitment varied little

as a

function

of

distributive justice. Again,

an

identical interaction pattern emerged for evaluation of supervisor.^

DIS USSION I

Distributive

and

Procedural Justice

as

Predictors

Our finding that both distributive and procedural justiceareimportant

predictors of work outcomes underscores recent admonitions for organiza-

^

A

reviewer raised

a

concern about

tbe

reliability

of our

single-item supervisor evaluation

measure.

To

address this issue,

we

corrected

tbe

correlation matrix

for

measurement error

and

conducted

the

analysis again using

the

corrected matrix. Wbere possible, used alpha


9/13

634 Academy o Management Journal August

tional researchers to examine both types of justice (Greenberg, 1987a,b), We

also found, as expected, that distributive justice tended to be a stronger

predictor of personal outcomes than procedural justice, whereas the reverse

was true for organizational outco m es. The fairness of a firm s pro ced ure s

may have a greater impact on organizational commitment than the fairness

of personal outcomes that workers receive, perhaps because procedures de-

fine th e org aniza tion s cap acity to treat employee s fairly. Th us , if they see

procedures as fair, employees may view the organization positively, even if

they are currently dissatisfied with such personal outcomes as a low pay

raise. In contrast, a pay cut may c ripp le an emp loyee financially, and a larger

paycheck will buy more, regardless of whether procedures were fair or not

(cf. Folger

Konovsky, 1989; Konovsky et al, 1987). Our findings are con-

gruent with research in the legal arena that suggests that distributive and

proce dural justice have different predictive roles depen ding on whe ther th e

outcome in question is personal or reflects more general evaluations of legal

institutions or their representatives (Lind Tyler, 1988).

However, we also found that distributive and procedural justice had

significant interactive effects on our organ izational outco m es, su bo rdin ate s

evaluation of supervisor and organizational commitment. Consistent with

Crop anzan o an d Folger s (1989) app lication of referent cogn itions theo ry,

these interactions revealed that the combination of unfair procedures and

low distributive justice produced the lowest ratings. In contrast, fair proce-

dures produced high commitment and supervisor evaluations, regardless of

the level of distributive justice. Advocates of referent cognitions theory ar-

gued that, under conditions of procedural fairness, employees will find it

difficult to envision that more positive alternative outcomes could have

occurred.

Referent cognitions theory may also help explain why we found inter-

action effects for organizational, but not personal, outcomes. According to

the theory, employee resentment requires not only low distributive justice,

but also the ability to identify other people and the procedures that they use

as the source of poor outcomes (Cropanzano

Folger, 1989). Thus, organ-

izational outcom es m ay represent a clear target for blame, whereas personal

outcomes do not. By definition, organizational outcomes make salient the

institutional sources of the procedures that affect employees.

Limitations and irections for Future Research

One of the most basic limitations of the present study is its exclusive

reliance on cross-sectional, self-report data. This reliance precludes us from

making strong causal statemen ts about our results. The use of a longitud inal

design wo uld improve the ab ility to make causal stateme nts. In add ition, o ur


10/13

1992 McFarJin an d Sweeney 635

approach raises the issue of common method variance. Although we can at

least partially allay concerns about method variance in our particular case

(see the Results section), we wo uld also encourage future justice resea rchers

to take steps to avoid method variance problems in field settings', by, for

instance, collecting mu ltiple measures. [

Concern might also arise about the small magnitude of our interaction

effects. Since interaction terms are not independent of main effects in re-

gression analyses, Cohen and Cohen (1983) recommended that researchers

test simple main effect models before entertaining more complicated inter-

active models. This procedure tends to underestimate the amount; of vari-

ance for which interaction terms truly account. Furthermore, the literature

on regression ana lysis prov ides several cau tions about interpreting t;he theo-

retical or practical significance of interaction terms on the basis ofjthe pro-

portion s of variance expla ined (cf. Cham poux Peters, 1987; Ped hazu r,

1982; Stone Hollenbeck, 1984). Apparently, a key issue is not how much

variance is explained, but whether the increment is statistically significant,

indicating that the relationship in question is being moderated. However, to

understand the actual nature of an interaction effect, researchers should plot

separate regression equations for different values of the moderator^ variable

and exam ine the resulting pa ttern for its theoretical or practical significance

(Stone Hollenbeck, 1984). This is the approach we took in this research.

Still, we would suggest that researchers be sensitive to the limitations of

moderated regression analysis and follow the recommendations that have

been made to minimize them.

Finally, a basic direction for future research is to determine whether

personal and organizational outcomes other than those examined here and

in previous research will yield similar predictive patterns for distributive

and procedural justice. More research is also needed to explain why distrib-

utive and procedural justice may differentially affect personal and organiza-

tional outcomes. To address this issue may require a better understanding of

w

pe rsonal an d organizational outco me s differ. A s Cropan zano a nd Folger

(1989) suggested, referent cognitions theory may provide a useful guiding

framework for such efforts.

R E F E R E N E S I

Alexander, S., Ruderman, M. 1987. The role of procedural and distributive justice in organ-

izational bebavior. Social Justice Research 1: 177-198 .

And rews, F. M.,

W ithey, S. B. 1976.Social indicators of well-being: Ame ricans] perception

of life quality. New York: Plenum Press.

Ben tler, P., Bone tt, D. G. 1980. Significance tests and goodness of fit in the analysis of

covariance structures. Psychological Rulletin 103: 588-606.


11/13

^36

Academy of

Managem ent Journal August

Cohen, J.,

Gohen, P. 1983.Applied mu ltiple regression/correlation analysis for the beha v-

ioral sciences (2d ed.). Hillsdale, NJ: Erlbaum.

Gook, J.,

Wall, T. 1980. New work attitude measures of trust, organizational commitment and

personal need nonfulfillment.

Joumal of Occupational Psychology

53 : 39-5 2 .

Gropan zano, R., Folger, R. 1989. Referent cognitions and task decision autono my : Beyond

equity theory.Joumal of Applied Psychology 74: 293-299.

Folger, R. 1986. Rethinking equity tbeory: A referent cognitions m odel. In. H. W.Bierhoff R. L.

Gohen,

J. Greenberg (Eds.),Justice in social relations: 145-162. New York: Plenum.

Folger, R. 1987. Distributive and procedural justice in tbe workplace. Social Justice Research

1: 143-159.

Folger, R.,

Konovsky, M. A. 1989. Effects of proc edural and distributive justice on reactions

to pay raise decisions.Academy of Managem ent Journal 32: 115-130 .

Greenberg, J. 1987a. Reactions to procedural injustice in payment distributions: Do the means

justify the ends? Joumol of Applied Psychology 72: 5 5 - 7 1 .

Greenberg, J. 1987b. A taxonomy of organizational justice theories.Academy of Managemen t

Review 12: 9-22.

Green berg, J., Folger, R. 1983. Procedural justice, participation, and the fair process effect in

groups and organizations. In P. Paulus (Ed.), Rasic group processes: 235-256. New York:

Springer-Verlag.

Joreskog, K. G., Sorbom, D. 1985.LISR LVI:Analysis of linear structural relationships by

the method of maximum likelihood. M ooresv ille, IN: Scientific Software.

Konovsky, M. A., Folger, R.,

Gropanzano, R. 1987. Relative effects of procedural and distrib-

utive justice on employee attitudes.

Represen tative Research in Social Psychology

17 :

Lind , E. A., Tyler, T. R. 1988. The social psychology of procedural justice. New York:

Plenum.

Musante, L., Gilbert, M. A.,

Tbibaut, J. 1983. Tbe effects of control on perceived fairness of

procedures and outcomes.Journal of Experim ental Social Psychology 19: 223-23 8.

Pedbazer, E. J. 1982. Multiple regression in behavioral research. New York: Holt, Rinebart,

Winston.

Podsakoff P. M., Organ, D. W. 1986. Self-reports in organizational research: Problems and

prospects

Joumal of Management

12:

531 544.

Price , J. L., M ueller, G. W. 1986. Handbook of organizational measurement. Marshfield,

MA: Pitman.

Qu inn, R. P., Staines, G. L. 1979. The 1977 quality of employm ent survey. Ann Arbor:

Institute for Social Researcb, University of Michigan.

Sta ines , G. L., Pottick, K. J.,

Fudge, D. A. 1986. Wives' employment and husb and s' attitudes

toward work and life.Journal of Applied Psycho logy 71: 118-12 8.

Stone, E. F.,

Hollenbeck, J. R. 1984. Some issues associated w ith the use of m oderated regres-

sion. O rganizational Rehavior and Human Performance 34: 195-21 3.

Tucker, L. R., Lewis, G. 1973. The reliability coefficient for maximum likelihood factor anal-

ysis.

Psychometrika

38: 1 10.


12/13

1992

McFarlin and Sweeney 7

Dean B McFarlin is an associate professor of management in the College of Business

Administration at Marquette University. His research interests include job satisfaction,

participative management, justice in organizations, and cross-cultural differences in

organizational behavior.

Paul D Sweeney

is an associate professor of management in the College of Business

Administration at Marquette University. His research interests include employee per-

ceptions of fairness, control, and satisfaction as well as cross-cultural differences in

organizational behavior.


13/13

distributive and procedural justice as predictors of js

Documents