Relationship between Job Satisfaction & Job Performance: a Meta-analysis

Author(s): S.C. Davar and RanjuBala

Source: Indian Journal of Industrial Relations , October 2012, Vol. 48, No. 2 (October
2012), pp. 290-305
Published by: Shri Ram Centre for Industrial Relations and Human Resources

Stable URL: https://www.jstor.org/stable/23509839

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/23509839?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Shri Ram Centre for Industrial Relations and Human Resources is collaborating with JSTOR to
digitize, preserve and extend access to Indian Journal of Industrial Relations

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
Relationship between Job Satisfaction & Job
Performance: a Meta-analysis

S.C. Davar & RanjuBala

Introduction

Job satisfaction plays an important

Studies suggest that there is a
significant relationship between
job satisfaction and job perfor
mance. A meta-analysis of 48 cor
relations produces a mean cor
relation of the order (p= 0.30
approx.). However, the modera
tor variables viz., foreign stud
ies vs. Indian studies, occupa
tion-type vs. scale of measure
ment for job satisfaction affect
the magnitude of the relationship
between job satisfaction and job
performance. This study shows
that to obtain a valid estimate of
mean correlation and true vari
ance, we must correct correlation
coefficients for the measurement
errors.
some
S. C. Davar is Professor & Ex. Chairman, developed
Department
of Commerce, Kurukshetra University, Kurukshetra.
Ranju Bala (ranjusobti@gmail.com) is Assistant
mance, while others used peer or super
Professor & Off. Principal, M.P.M.S.D. Girls
College, Ismailabad, Kurukshetra.
role for an employee in terms of health
and well being (Kornhaurser, 1965;
Khaleque, 1981) and for an organization
in terms of its productivity, efficiency,
employee relations, absenteeism and turn
over (Vroom, 1964; Locke, 1976;
khaleque, 1984). Job satisfaction is a com
plex variable and influenced by situational
factors of the job as well as the disposi
tional characteristics of the individual
(Sharma & Ghosh, 2006). It can be cap
tured by either a one dimensional con
cept of global job satisfaction or a multi
faceted construct capturing different as
pects of job satisfaction that can vary
independently.
Research examining the relationship
between job satisfaction and job perfor
mance has been conducted since at least
as early as 1945 (e. g., Brody, 1945) and
methodology utilized has varied greatly.
Some researchers used established
scales to measure job satisfaction, while
their own. Some used
self- report ratings to assess perfor
visor ratings.

290 The Indian Journal of Industrial Relations, Vol. 48, No. 2, Octobe

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 Relationship between Job Satisfaction & Job Performance

Keaveney and Nelson (1993) found a

Over the years, scholars examined non-significant correlation coefficient
this idea that a happy worker is a between job satisfaction and job perfor
productive worker; however, evi- manee. Manjunath (2008) found job sat
dence is not yet conclusive in this isfaction of agricultural scientists signifi
regard. cantly correlated with their scientific pro
ductivity. Ravindran (2007) found that job
, ... , . „ , satisfaction is non-significantly correlated
The i dea that job sat. sfaction leads w,thjobperforma„ce.
to better performance is supported by
Vroom's (1964) work which is based on There are confl¡cti viewp
the notion that performance is natural ,hereiatio„shipbetweenjobsatis
product of satisfying the needs of em- and job formance ^
ployees The study relating to the rela- js ,0 theslze a,, results of
tionship betweenjob satisfaction and job s(udies re|atj ,0 the relati
performance has now become a research ,ween job satjsfaction and
tradition in industnal-organizational psy- „
.T., , . , . , F. \ manee,
chology. The relationship between job
satisfaction and job performance has Meta.anaiyti
been described as the "Holy Grail" of
industrial psychologists (Landy, 1989) Petty et al (
Many organizational theories are based meta.analysis
on the notion that orgamzattons that are crformance r
able to make their employees happy will thejr ana,
have more productive employees. Over ,ished jn flve
the years, scholars examined thts idea that Hjgher and mo
a happy worker is a productive worker; betweep overa
however, evidence is not yet conclusive formance were
in this regard Empirical studies have pro- V10us)y
duced several conflicting viewpoints on job descriptiv
the relationship between job satisfaction satisfaction a
and job performance. Strauss (1968) as hj h or as co
commented, "Early human relatiomsts betweenove
viewed the morale - productivity rela- fonnance. T
tionship quite sunple: higher morale would recKd corT
lead to improved productivity". Siegel & constructs
Bowen (1971) and Bagozzi (1980) sug
gested that job performance leads to job Laffaldan
satisfaction but not the reverse. Ander- analyzed 21? c
son (1984) indicated that autonomy and ies and found
feedback from the job is significantly isfacti0n-pe
correlated with the performance, across the job s

The Indian Journal of Industrial Relations, Vol. 48,

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 S.C. Davar & RanjuBala

ing from a mean "true score" correlation It is a way to summarize, integrate and
of 0.06 for pay satisfaction to 0.29 for interpret selected descriptive statistics
overall job satisfaction. For their primary (e.g., sample correlations) produced by
analysis they averaged the facets per- sample studies or experimental outcomes
formance correlations and reported an (e.g., d- statistics). There are different
average true score correlation of 0.17 methods of meta- analysis. The frame
between job satisfaction and job perfor- work of Rosenthal and Rubin's (1978),
manee. In discussing their findings, the Hunter, Schmidt and Jackson (1982),
authors only made reference to the 0.17 Hedges and Olkin (1985); Davar(2004).
correlation, concluding that job satisfac- Hunter, Schmidt and Jackson (1982) is a
tion and job performance were "Only popular method used to compute true
slightly related to each other". variance i.e. observed variance net of the
measurement error, sampling error and
Because of limitations in the prior range-restriction. Davar (2004) modified
analysis, Judge et al. (2001) conducted a the formulas given by HSJ (1982) frame
new meta analysis on 312 samples. The work and provided us with 'An Improved
true correlation between overall job sat- Version' of HSJ ( 1982). The formulas for
isfaction and job performance was esti- two models are given below:
mated to be 0.30. Meta analysis was
conducted by five facets in the job de- The formulas
scriptive index (Smith, Kendall & Hulin,
1969) and found that the average cor- Chart -A:
rected correlation was 0.18 a figure iden
tical to Laffaldano and Muchinsky's The formulas for true variance models
(1985) overall estimate. Even with up Hunter,
Hunter,Schmidt and
Schmidt Davar
Davar
and (2004) (2004)
dated meta analysis the facet substan Jackson
Jackson(1982) (1982)
tially underestimate the relationship of framework
framework
overall job satisfaction to job perfor
mance. ..Z[Ntrt]
IIM ZM
IM
rrZ IN,
Nt k
What is Meta-analysis ?
22ZM(fi-r)2]
I[N,(r,- r)2]
A meta-analysis is used to synthesize Jr
JrZ NtIN,
the results of different studies relating to
the relationship between job satisfaction
a2 ~--
/T2
2=i0¡- py
Swft-
and job performance. Glass (1976) de °r
k
fined meta-analysis as "The statistical
analysis of a large collection of studies
results for the purpose of integrating the
_2 (1 ~f2)2/k
N
findings". Meta-analysis is regarded as
a° N
an accurate and objective way to assimi
late research findings in the present era.

292 The Indian Journal of Industrial Relations, Vol. 48, No. 2, October 2012

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 Relationship between Job Satisfaction & Job Performance

Statistical Terms & Notations

(1- p2y
ol =
Sample statistics - The statis
tics (e.g. correlation co-efficient)
a2 =o? — o% a2 =of —of
based on the sample data is called
a sample statistics.
r~ = Mean correlation; <jf = Ob
served variance;erf = Sampling error
variance; j2 = True variance and k= Meta-analytic statistics - The
statistics computed with the help of
number of studies; N =£ N,; N = number
of observations in a sample study. sample statistics produced by dif
ferent studies is known as meta

Objectives of the Study analytic statistics, e.g., common

correlation (p).
The first objective of the study is to
Measurement error - It is the
generate a meta-analtic estimate for the
error in the measurement of postu
general relationship between job satisfac
lates (variables). Generally, the
tion and job performance. This estimate
measurement error arises on ac
indicates the general magnitude of rela
count of the lack of construct va
tionship and is computed as a mean esti
lidity and attenuates the magnitude
mate from a set of sample studies. Gen
of a sample correlation coefficient.
erally, moderator variables cause differ
ences in sample correlations i.e the ex Observed variance - It is
tent of relationship may vary from one
meta-analytic statistics that mea
setting to another. Therefore, we need sures the extent of variation in the
to know the moderator or contextual fac
sample correlation coefficients
tors. Hence, the second objective is to across studies.
find out the contextual variables that in
fluence the magnitude of relationship.
Sampling error variance - It
is meta-analytic statistics that re
The Data-set
flects the amount of sampling error
in the data-set of sample correla
The data-set for the study has been
tion coefficients.
collected from journals incorporating
business and psychological studies, PsyTrue variance- The observed
chological Bulletin, Indian Appliedvariance
Psy net of the sampling error
chology & Psychological Abstracts from
variance (or2 -oe2) is termed as
the Journals section of Kurukshetra Uni
true variance (see HSJ, 1982). It is
versity, Kurukshetra. Various onlinea jour
meta-analytic statistic.
nals like JSTORE, QUESTA, and The
Free Online Library have also been vis- from 12 studies from 197i t
ited. 49 correlations have been found 0f the studies reported a nega

The Indian Journal of Industrial Relations, Vol. 48, No. 2, Octobe

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 S.C. Davar & RanjuBala

lation which has been ignored for two by taking the figure of mean correlations
reasons. First, we cannot find the logic as computed with Davar (2004) and HSJ
for a negative relationship between job (1982) procedure,
satisfaction and job performance. Sec- Tab,e ,Table
Mean Corre)ation
1 Mean Coefficieilts
Correlation forfor
Coefficients thethe
ond, it is incorrect to perform mathemati- Relationship between
Relationship Job Satisfaction
between and Joband Job
Job Satisfaction
cal operations on bipolar coefficients for Performance
Performance

the determination of mean correlation,

Number
Number of of Uncorrected
Uncorrected Corrected
Corrected
observed variance and sampling error
Studies=48
variance (Davar, 2004).
Non-weighted
Non- weighted 0.29729
0.25083 0.29729
0.25083
Mean Davar
Results & Discussion
(2004)
t-value L757
t-value 1.757 2.112*
2.112*
Table-1 shows the results of a com
Weighted
Weighted Mean Mean
parative analysis of the above two meta 0.24818 0.29222
(HSJ
(HSJ1982)
1982) 0.24818 0.29222
analysis procedures. t-value 1.738
t-value 1.738 2.072*
2.072*

•significant at 5% (a. =1.96)

Mean Correlation: HSJ (1982) Pro
cedure uses weighing- based formula of The respective t-values for non
mean correlation. But Davar (2004) gave weighted corrected mean correlation
non weighing based formula of mean coefficient is 2.112 and for weighted co
correlation. Both methods suggest that rected mean correlation is 2.072 ar
correlation coefficients should be cor- more than the table value i.e. 1.96 at

rected for measurement error. HSJ a.05,Thus we can conclude that t

(1982) procedure produces a mean cor- a significant association between
relation(f) =0.24818 (when sample cor- isfaction and job performance
relations were uncorrected for measure- mates of t-value of mean correlation
ment error) and r~ =0.29222 (when coefficient based on uncorrected
sample correlations were corrected for correlation coefficients in both
measurement error). In the same way, suggest that there is no (insignif
Davar (2004) procedure produces a relationship between job satisfaction
mean correlation (r-bar) =0.25083(when and job performance. These results sug
sample correlations were not corrected gest that in order to obtain a correct pic
for measurement error) and p=0.29729 ture of the mean correlation, we must
(when sample correlations were cor- use the measurement corrected corre
rected for measurement error)'. The lation coefficients,
above analysis shows that correction of
the individual correlation coefficients for
In order to obtain a correct picture
measurement error improves the mean
.... . , . of the mean correlation, we must
correlation. In the same way weighing , ,
r . , . /ttot i J- use the measurement corrected
of sample correlations (HSJ method) dis
correlation coefficients.
torts the mean value. It can be explained

294 The Indian Journal of Industrial Relations, Vol. 48, No. 2, Octob

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 Relationship between Job Satisfaction & Job Performance

True variance estimate tells whether erator or contextual factors. If there is a

there are moderator factors that dilute true variance, it suggests that one or more
or increase the relationship between job moderator/extraneous factors influence
satisfaction and performance. Zero true the relationship between two specified
variance implies that there are no mod- postulates.

Table 2 Davar (2004) vs. HSJ (1982) Procedures: A Comparative Analysis

Number of
of studies
studies (k)
(k) =48
=48
Uncorrected
Uncorrected Correctedfor
for for Corrected for
measurement
measurement error
error measurement
measurementerror
error

Mean Correlation
(non-wei-ghted)
(non-wei- ghted) 0.25083 0.29729
0.25083 0.29729 Davar Davar
(2004)(2004)
Mean
Mean Correlation
Correlation (weighted)
(weighted) 0.24818 0.24818 0.29222 HSJ(1982)
0.29222 HSJ (1982)
Observed
Observed Variance Variance 0.02639 0.02639 0.03772
0.03772 Davar(2004)
Davar(2004)
0.0191 0.02649
0.0191 0.02649 HSJ(1982)
HSJ (1982)
Sampling
Sampling Error Error
Variance Variance
0.01829 0.01829 0.01731 Davar
0.01731 (2004)
Davar (2004)
0.00409 0.00388
0.00409 0.00388 HSJ(1982)
HSJ(1982)
True
True VarianceVariance0.0081 0.0081 0.02041 0.02041 Davar (2004)
Davar (2004)
0.01501 0.02261
0.01501 0.02261 HSJ(1982)
HSJ(1982)

It is clear from Table 2 that meta- Da

analysis of uncorrected correlation coef- a
ficients with Davar (2004) procedure as ca
well as with HSJ (1982) procedure gen- m
erates lesser value of true variance as means there is substantial variation in the
compared to meta-analysis of corrected individual correlations across 48 studies,
correlation coefficients. Table 2 shows that Here, we examine five possible modera
once the observed correlation coefficients tors (for example, foreign vs. Indian stud
are corrected for measurement error, true ies to know the reasons of such varia
variance rises to the level of 0.02041 vis- tion in the individual correlations. These
à-vis 0.008lin the case of Davar (2004) are shown in further tables,
procedure, and 0.02261vis-à-vis0.01501 in
case of HSJ (1982) procedure. Significant Foreign Studies vs. Indian Studies:
true variance indicates that there is a sig- Table 3 shows that out of the 48 studies
nificant variation in effect sizes of job sat- 34 are foreign and 14 are Indian. The
isfaction. It means there could be one or mean correlation for foreign studies is
more moderators which influence the level 0.25715 and for Indian studies it is
of effect size of job satisfaction. 0.39479. Thus Indian studies have a sig
nificant correlation between job satis
Moderator Analysis faction & job performance and foreign
studies find insignificant average corre
As stated earlier, the true variance lation. Further, foreign studies reveal a
in the case of corrected correlations for negative true variance (-0.00773)

The Indian Journal of Industrial Relations, Vol. 48, No. 2, October 2012 295

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 S. C. Davar & RanjuBala

whereas Indian studies show a positive manee relationship across occupations of

true variance (0.0215). It means one sub- employees. The most significant mean
set generates significant true variance and correlation was observed for managers
other subset generate insignificant true and officers i.e. 0.35705. The weakest
variance. Overall, the difference in mean correlation was observed by mixed em
correlation and true variance indicates ployees (executives and non-executives)
that there is great likelihood that foreign i.e. 0.23635. True variance 0.0046 is posi
studies vs. Indian studies act as impor- five for managers and officers and the
tant moderator for the relationship be- rest four subsets (categories) show nega
tween job satisfaction & job perfor- five true variance i.e. -0.32593 for tele
mance. The relationship varies across phone operators,-0.00926 for mixed em
foreign settings vis a vis Indian settings, ployees (executives and non-executives),
-0.05141 for teachers and researchers, -
Studies 0.06616 for nurses. We treat them as in
Table 3 Moderator Analysis: Foreign Studies
vs. Indian Studies
vs. Indian Studies significant estimates (negative value of
true variance is assumed to be close to
Foreign Indian
Studies Studies zero and hence insignificant). Thus one
subset shows significant positive relation
Number of Studies (k) 34 14
Mean Correlation 0.25715 0.39479 ship and other subsets show insignificant
Observed Variance 0.01792 0.0724 relationship between job satisfaction and
0.02565
Sampling Error Variance 0.0509 job performance. It means occupation -
True Variance -0.00773 0.0215
type (managers vs. others act as a mod
erator variable for the relationship be
Occupation Type: Table 4 shows tween j°b satisfaction and job perfor
the differences in the satisfaction-perfor- manee.

Table 4 Moderator Analysis: Occupation- type

Managers
Managers &
& Telephone Mixed Teachers & Nurses
Officers Operators Employees Researchers

Number of Studies 14 3 14 9 7
Mean Correlation 0.35705 0.29721 0.23635 0.3516 0.27234
Observed Variance 0.05898 0.00154 0.03239 0.03393 0.00801
0.05438
Sampling Error Variance 0.32747 0.04165 0.08534 0.07417
True Variance 0.0046 -0.32593 -0.00926 -0.05141 -0.06616

Overall Satisfaction vs Facet individually like security need, social

Satisfaction need, esteem need, autonomy need, self
actualization need, work itself, promotio
In the column of 'Facet Satisfaction' pay, co-workers, supervision e
in Table 5, we include those studies which the column of overall job
found the relationship between different those studies are included
facets of satisfaction and performance sures the relationship betwee

296 The Indian Journal of Industrial Relations, Vol. 48, No. 2, Oc

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 Relationship between Job Satisfaction & Job Performance

Table
Table5 Moderator Analysis: Variable Analysis:
5 Moderator Used for Variable Used for measuring job satisfaction cannot be
Measuring Job Satisfaction taken as moderator variable for the rela
Measuring Job Satisfaction

Overall
tionship between job satis
Overall
Facet Facet c
Satisfaction
Satisfaction Satisfaction Satisfaction performance
Number
Number of Studies
of Studies 21 21 27
27
Mean
Mean Correlation0.35801
Correlation 0.35801 0.25007
0.25007
Variable used for measuring job
Observed
Observed Variance
Variance 0.06495 0.01145
0.06495 0.01145
satisfaction cannot be taken as
Sampling Error
v^aPncegErr0r
Variance 0.03619 0.03619
0.03255 0.03255 moderator variable for t
True
True Variance-0.0632
Variance -0.0632 -0.0211
-0.0211 tionship between job satisfaction
and job performance. action j
satisfaction and job performance. Table
5 shows that both the subsets generate Measurement Scales
negative true variance i.e. -0.0632 for
overall job satisfaction and -0.0211 for The studies measured job satisfac
facet satisfaction. It reveals that both the tion with JSQ, JDI, JSS and MSQ. Both
subsets show insignificant relationship MSQ and JDI show different values of
between job satisfaction and job perfor- mean correlations. True variance is posi
mance. Therefore, variable used for tive only when job satisfaction is mea

Table 6 Moderator Analysis: Measurement of Job Satisfaction with Different Scales

JSQ JDI JSS MSQ MSQ & OTHERS

JDI

Numbers of Studies 10 12 12 7 3 4
Mean Correlation 0.26730 0.19242 0.39926 0.27234 0.29721 0.42471
Observed Variance 0.01198 0.01176 0.06477 0.00801 0.00154 0.09419
Sampling error variance 0.08622 1.56358 0.05888 0.12245 0.27705 0.16794
True variance 0.07424 -1.55182 0.00589 -0.11444 -0.27551 -0.07375

JSQ(PORTER)- Porter(l 96l)'s job satisfaction questionnaire, measure need satisfaction, contained
12 items based upon Maslow's theory of motivation.
MSQ- Minnesota Satisfaction Questionnaire contained 20 items measure overall satisfaction and
intrinsic versus extrinsic satisfaction.
JD1(S, K&H) - Job Description Index developed by Smith, Kendall and Hullin(1969) measure job
satisfaction with five facets of the job: the work itself, their supervisor, pay, promotion opportuni
ties and co-workers. The scores on the five sub scale can be summed to obtain an overall measure of
job satisfaction.
JSS (P&A) -Pelz and Andrew's (1966) job satisfaction scale measure satisfaction through different
aspects of the job satisfaction.
JSS (B&R) -Brayfield - Rothe Scale measures overall job satisfaction.
JAS (J&B) - Job Attitude Scale of Jayan and Balachandran (2004) contains three domains of Job
Attitude: Job Involvement, Job Commitment and Job Satisfaction-The items are designed to include
intrinsic attributes of the job as well as extrinsic attributes.
JSS (Likert) - Likert's five point scale measures both intrinsic and extrinsic job satisfaction.
INDSALES (C, F &W) - INDSALES, a scale created by Churchill, Ford and Walker (1974) measures
job satisfaction over several dimensions.

The Indian Journal of Industrial Relations, Vol. 48, No. 2, October 2012 297

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 S C. Davar & RanjuBala

sured through JSS scale. And it is nega- other scales. As one subset in the last
tive for all other subsets i.e. -0.07424 column of Table 6, scales for two stud
when job satisfaction is measured with ies are not known and for other two stud
JSQ scale, -1.55182 with JDI scale, - ies, JAS (J&B) & INDSALES (C, F&W)
0.11444 withMSQ scale, -0.27551 with scales were used respectively,
both MSQ and JDI and -0.07375 with

Table 7 Moderator Analysis: Measuring Job Performance through Different Raters

Peer Rating Other Measures Supervisor's Self Rating

Rating
Number of Studies 19 3 16 10
Mean Correlation 0.30723 0.31290 0.24602 0.35576
Observed Variance 0.02415 0.14652 0.02674 0.04056

Sampling Error Variance 0.04316 0.27126 0.05516 0.07629

True Variance -0.01901 -0.12474 -0.02842 -0.03573

Table 7 shows that mean correlation is manee. Moderator variables: foreign

stronger i.e. 0.35576 for those studies which studies vs. Indian studies, occupation type
measure job performance through self rat and measurement-scale for job satisfac
ing and weaker i.e. 0.24602 for those stud tion with different scales affect the mag
ies which measure job performance through nitude of the relationship between job
supervisor's rating. True variance is nega satisfaction and job performance. The
tive for all four subsets i.e. -0.01901 forpresent review, however, has been based
peer rating, -0.12474 for other measures, -on a limited number of studies. To gen
0.02842 for supervisor's rating, -0.03573 eralize the results number of studies may
for self rating. There are insignificant variabe increased for a meta-analysis. In the
tions between all the subsets. Thus mea same way, limited moderator analysis
could be made due to limited number of
suring job performance through different
raters cannot be taken as moderator vari studies. Some other moderators like sex
of sample, job complexity, job perma
able for the relationship between job satis
faction and job performance. nency etc. may also be analyzed in fu
ture research. According to Rosenthal
Conclusions (1995:190), the overall goal of a meta
analysis is to answer the question:
The mean corrected correlation im "Where are we now that this meta-analy
proves when it is computed with cor sis has been conducted?".
rected correlations i.e. 0.29222. On the
basis of t-value of mean corrected cor References
relation computed with both HSJ and
Davar formulae, we may say that thereAnderson, Carol H. (1984), "Job Design: Em
ployee Satisfaction and Performance in
is positive and significant relationship
Retail Stores", Journal of Small Business
between job satisfaction and job perfor
Management, 22: 4.

298 The Indian Journal of Industrial Relations, Vol. 48, No. 2, Octo

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 Relationship between Job Satisfaction & Job Performance

Bagozzi. R. P. (1980), "Performance and Satis Judge, Thimothy A., Thoreson C.J., Bono J.E.&
faction in an Industrial Sales Force: An Ex Patton G.K. (2001), "The Job Satisfaction
amination of Their Antecedents and Simul Job Performance Relationship: A Qualita
taneity", Journal of Marketing, 44: 65-77. tive and Quantitative Review", Psycho
logical Bulletin, 127(3): 376-407
Bhuyan, B. & Choudhary, M. (2002), "Corre
lates of Job Satisfaction among College Keaveney, S. M. & Nelson, J. E. (1993), "Coping
Teachers", Indian Journal of Psychometry with organizational Role Stress: Intrinsic
and Education, XXXIII (2): 143-46. Motivational Orientation, Perceived Role
Benefits, and Psychological Withdrawal",
Biswas, P.C. & De, T. (1994), "A Study of Job
Journal of the Academy of Marketing Sci
Satisfaction of Secondary Teachers in Re ence, 21:113-24.
lation to Variables", Journal of Educational
Research and Extension, 33: 153-63 Khaleque, A. (1981), "Job Satisfaction Perceived
Effort and Heart Rate in Light Industrial
Brief, (1998) cited in Weiss, H.M (2002), Work", Ergonomics, 24:735- 42.
"Deconstructing Job Satisfaction Separat
ing Evaluations, Beliefs and Affective Ex Khaleque, A. (1984), Job Satisfaction and Work
periences", Human Resources Management in Industry, (5th edition), Dhaka, Alamgir
Review, 12: 173-94. Art Press.

Brody, M. (1945), The Relationship between Kornhauser, A (1965). Mental Health of the In
Effciency and Job Satisfaction, Unpub dustrial Worker, New York: John Wiley
lished Master's Thesis. New York Univer
sity, New York. Kotharai. C.R. 1(985), Research Methodology:
Methods and Techniques: New Delhi:
Chandraiah, K. (1994), "Effects of Age on Job Wiley Eastern Limited. Laffaldano,
Satisfaction among College Teachers", The
Creative Psychologist, VI (1&2): 53-56.Landy, F. J. (1989), Psychology of Work Behav
ior, Pacific Grove, CA: Brooks/Cole.
Davar, S.C. (2004). "Meta Analysis: An Improved
Version of Hunter, Scmidt and Jackson Locke, A.A. (1976),The Nature and Causes of
( 1982) Framework", Decision, 31 (2): 210
Job Satisfaction. In M.D. Dunnette (ed.),
38. Handbook of Industrial and Organizational
Psychology, Publisher & Place of Publica
Davar, S.C. (2006), "Meta-Analysis of Sample tion?
Correlations: The Validity of Hunter,
Schmidt and Jackson (1982) Framework", Michelle T. & Muchinsky P.M. (1985), "Job Sat
isfaction & Job Performance: A Meta
Decision, 33(2): 111-40.
analysis". Psychological Bulletin, 97(2):
Fried, Y. & Ferris, G.R. (1987), "The Validity of 251-73.
the Job Characteristics Model: A Review
Manjunath L.,Tyagarajan S., Kumar Vasant J. &
and Meta-analysis", Personnel Psychology,
Ansari M.R. (2008), "Determinants of Sci
40(2): 287-322.
entific Productivity of Agricultural Scien
Glass Gene V. (1976), "Primary, Secondary and tists", Journal of Agriculture Science,
Meta-analysis of Research", Educational 21(3): 466-68.
Researcher, 5: 3-8.
Pal, Yash & S.C. Davar (2001). "Meta-analysis
Hunter, John E, Frank L. Schmidt & Gregg B. in Research-An Introduction", in P.P. Arya
Jackson (1982), Meta-analysis: Cumulat Yash Pal (ed), Research Methodology In
ing Research Finding across Studies, New Management, New Delhi, Deep & Deep
Delhi, Sage Publication India Pvt. Ltd. Publications.

The Indian Journal of Industrial Relations, Vol. 48, No. 2, October 2012 299

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 S. C. Davar & RanjuBala

Petty M.M., Mcgee G.W., Cavender J.W. (1984),Sharma, M. & Ghosh, A. (2006), "Perception of
"A Meta Analysis of the Relationships Organizational Climate and Job Satisfac
between Individual Job Satisfaction and tion in Nursing Staff Personnel", Indian
Individual Performance, Academy of Man Journal of Social Work, 67(3): 263-74.
agement Review, 9(4): 712-21.
Siegel. J. P.& Bowen, D. (1971), "Satisfaction
Ravindran C. (2007), "Study of Factors Influ and Performance: Causal Relationships and
encing The Work Output of Human Re Moderating Effects", Journal of Vocational
source in Revised National Tuberculosis Behavior, I: 263-69.
Control Program" (RNTCP)", Pulmón,
9(3): 88-96. Strauss, G. (1968), "Human Relations—1968
Style", Industrial Relations,7: 262-76.
Schmidt, L. & Hunter, J.E. (1999), "Comparison
Vroom, V. H. (1964), Work and Motivation, New
of Three Meta-analysis Methods Revisited:
York: Wiley.
An Analysis of Johnson, Mullen and Salas",
Journal of Applied Psychology, 84:144-48.

300 The Indian Journal of Industrial Relations, Vol. 48, No. 2, October 2012

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 Relationship between Job Satisfaction & Job Performance

00 00 00 00 00 00 00 00 00 B0
00 c c c c c c c c c e K
u v- .S C3 3 3 3 3 c0 5 3 3 i O

3 <♦-« 3 ba b t-c b* u b ba ba ba

e<3 b u ba b ba ba ba
-o u U b
© 6 6 © 0 O O C o

.§
> 38 E§ ©
a.
V
cu
©
Cu
0
ft
©
0
O
fib
©
&
O
a*
P
Ox fiu
o

C C
O ©
>» >»

^&^£3 fc £ '5 '5

00
E
<o j- .£ £ ° ° .52 .2
3 3
5 _, 5 3 3 _. g ^ g ^ 3 3
•5 "2 S • ? 3 ? 3 *j? 3 '?t 3 O 3 O 3 m 3 ^ 3 3
8888 8888 S 8 S 8 38 = 8 vi«i
> 3 E >-» co c vi c v) cvi c UcWc < e < e vi « com

c bo
« Ui
f t_.r"
c

I <2 5
3 "3 go; C/5 (75 C/3C/3 t/3 Cfl C/3C/3 C/3C/3
a. cu Qua. cu a. a. a. a, a.
J5 S 15
♦x

C 00

c o u.
g - .£ U U b U b U ba ba
/—s
ba ba

3^3 G o 00 00
^t-yC -ut ^ t O' o v o ^0C/0
rv S C

►5
±3
C 38Egn-,«!
"OO'oO'o
S O'oO'o
co a, w Oh w o. co a, co o. co Qh
•—* ►—» W •—» W N-» »—» >W •—>
a, co 0,
w >-» w
a
CO
0
Ou
'w'
g|!
»"■»«. W

8 s* 8 8 8 6«1 8 8 5*-g S 8 B« 5
• 5 SPig *£■§ S?3 8P|a SP:§
ES^g-Sssisg-Sei^g-issS^g-isi^g-i
3 O
o •a
£ ca
O a. ESStSSESSfSSESSfSStZSSfSSESSfSS
C c c
C
00 &
C c
00 S) 5) op 00 00 S)
op
2 § 'p 2 2 8 p 2 '§ g
O
0 0 0 0 0 O 0 0 0
bx bx ba bx ba bx ba
U bx bb bx
nnexure I e c c c c c c c c c

u a * * * *
0 0 0 0 0 0 0 0 0
oo ^ 0
c c c
c c c
« >> c c c c
M a*
M M a*
Jf "?
00 6 u c c C c a C c c e
C
D D D D D D D D D D
■3 < < ba Ua ba bx b. ba ba
a bx bx bx
a *6 # <* #
2 2 s s s s s s s 2
<u

s r r r 1^ r r r r
00
r
00
r
00
<2. oj 00 00 00 00 00 00 00

© © © © © © © © ©
0

T3
© © © © © © © © ©
0
-o <2 ^
^ *CO 00 00 00 00 00 00 00 00 00 00
X
© © © © © ©
u> © © © ©
o r OS r SO On 00 OS
co
CN fO
a . a-a x-x a-a a-x <N x rn
©
© © © © ©
o* © © © ©

CO ro
m
r <N fN r <N
r» (N r <N
00 00
Z 00 00 00

xh b b ba bx b bx wS bx bx ba'
••—» •—1 •■n —>
•—» •—> *—) •»—>
<w
£ 6 ^ £ /—V
£ /—, £ E >—S E E ^ E E
3 r* 3 3

1t
a>x
3
«x
3
^■X
3
•—<
3 3
p Is
8
r
8
r r 0
8
r
J
r 0 0
8
0
g On O Os OS Os 0 Os On Os O Os On

3 3 w 55 GO CO CO co C CO w CO
2 co CO >w' •w'

s
o
g Z
3 ©
«rj so 00 Os
so — <N

The Indian Journal of Industrial Relations, Vol. 48, No. 2, October 2012 301

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 S. C. Davar & RcinjuBala

w CO
C/î
y¡ w
JO
CO w CO

Vh u. *H *Ih Ih »H Ih Ih Ih
O O O o O O O O O o
M CO to CO CO
2 to CO CO C/î

> *>Ih *> > *> F '> '> *> > > >
i- bû
ft 00 fe ûû ft Où ft ûû ft où
Ih
ft
V OJO 1)
00
S M fe W> 00 fe
U {H <u
« Q.
C S.2 C 8 .2
c
H
.Sc
o. CU o. .S D. .5 8. .s a. .S CL .5 CL
c v c
CL .5 o CL
3
5B 5B 52 51 51 5 1
3
CO 2 CO
ca
Ih 52 CO
03
Ih 5B
c
C C o
O O £
•-
O .g
O 2
D 2
O-g
r
■g
ur" S
•? Ss,
v L-í
il. ■-*
t_ fe
r\

1*1"!
¿W O -i >!> 2 2 Í
O. CU Oh£0- ¿=
O £ ¿ oC/5
U £ £

CO CO CO CO CO CO CO CO CO CO CO CO
QhOmOhOh CU Oh Oh Oh Oh Oh Oh Oh

/—•S /—V /—S /—V /—v /*"■S /—V

_ KKK^K^xa: X
fvûfvKAis ^ <y ^ «y ^ ^

S^SüSü®"1" Q« û m Q$2, Qm D$2, ûm Q m « D m

«*£ ¿ «O 4 °C H, V ✓ »-» w •—» W ►-» w ■-% w <—> »-» w >-» w ?>-> •—> W
O V o
C i- c t c t
O O o o o O , , , , , , , , ,
■Sífl-Sed-Sed-oO "O _2 T3 _2 "O O "O O -o o "O® T3 _2 T3 .O
§,ûS'uS'ûvCL^!<a^ïa?î><a^!<a^)<a^><a^><a gCL"
—-o---rv— o .£ c¡ .2 e o .2 c 4> .£ c «■> .2 c .£ c o .2 c u .£ e .2 c w
¿r* flj Br* D XT' m C Qj m C 4> m C 3j m C D hh C 4) m C u m G m C «h G
HOhOhOSu "Su !¡,Su "Su >,Su "Sw «u >,2u Su >,

C c c c C C C C C C C C
00 00 00 00
.00 .00 00
.00 .00 op 00 .oo
'S "S
8 8 s Ih Ih
'C
Ih
'C
Ih
'C
Ih
S
Ih
S
Ih
'S
Ih
'S
Ih
o o o O O O O O O O O O
Uh Uh Uh Uh Uh Uh Uh Uh Uh Uh Uh Uh
c c C
* £ £
o o o
c c c
JX .*
c C c
D D D N" N" Tf 5 5
Uh Uh Uh Uh Uh Uh Uh Uh Uh
<*J <a •y <% <%
S s s S S S S S 2

NO NO NO NO vO NO NO nO
00 00 00 ON ON CO ON ON ON ON ON On
O O o d d c O d d d d d
Tf "«t Tj"
00 OO 00
NO m cd ON ON ON
00 00 d 00 O
00
d 00 00 00
C o
o o © #
*
o * * d d d
♦ * *
CO
♦ HH m <N
HH
00 Ü 00 Tf HH HH
m fN CN ro o CN CN CN
O > q »-H
CN
O O O O o 2 O O O o O d
NO NO NO NO
o O O HH »-H H H-H HH H HH HH HH
00 00 00 *T) •—1
«r> m
HH —
«n

<H-N /•S /—H

NO NO NO NO NO NO
T3 T3 NO NO NO
1cd
T3
1
r r^ r T3
v r» u. r r Ih r* Ih
On «5 ON On o o ON o o o
A êo cO 'cd cd cd
w CO vln^ CQ CQ CQ 03 CQ CQ CQ

© —• CN
CN CN CN

302 The Indian Journal of I

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 Relationship between Job Satisfaction & Job Performance

u.
O
1A
«■s «'s ^«g-s «¡«g* ?§?
>
l
O
00
i 1.2 -S 11.2 -S I f S ! -g i i 8 ! -g i ! s !
C
Q.
3
C/3 1
c
.2
O ü 4) V 4) 4>
*>
w. 00 00 00 00 op 00
<u R 03 çd ÇQ Cd Çd
O. fe. fe. fe. fe. fe. fe. 53
3 >1/3 >00 > C/3 > O) >t/3 > C/3 >
W3

^ o o o o o1 o o
o. 0. O. 0. O. 0-. Oh Oh

S¡ ^ ^
^<<<< < < <
* y) ÇA ÇA ÇA ÇA
Q ça ÇA a, ÇA Cu ÇA Cu CA CU ÇA CU ÇA CU <A CU
•—» h-» ■—» >—» w -W *"» '

c c
£ ° vi 2 .2 2 .
. ¿ kL 4> c/2 fli W c/5 fli H
H« S e •§ Si 82 -S g-Ë ¡¡
~
2 E
W s s « "g
><0£ s *H•§
W S >J2
0¿ S ><5
o H 0¿ «i
o

00 e e c ç c c
.1
'S c3 cj ed ra ra _ra
u
O
"O •3 *5 -5 "5 "o -o
c c e c e c c
Cu
c c c = c ç ç
£ í S ï S ï î
o o o o o o o
c c c c e ç c
III I ■ 1
G c c c c c c
D D 3 D D 3 3
tu tu b. U. 0- Oh Oh O.
i*j =¡a <*5 =¡3 ■ü
s S S S S 2 S 2

r- ^ £•
os 00 00 00 00 00 00 00
o o o o o o o o
Os t
°°
o
£
• r-
t- r-
r-
^o
Pr£ r
♦ ° © © * o o
* * * ♦
vO <N c**> **0 * <N 00
ÍN (N "t ÍN CN «Ti «O —;
o o o o o o o o
vo SO
tfr fS —' ^
n ín »— <n es

<s¿ <¡¡¿ <% <¡¿ <% <% *i

•o ct -a c~ "o ty "2 _ 12 _ u _ — "2
"2 hv £ 13 S S oo 2 13 00 2 13 00 g "3 00 g3°®
•5^ 2
03 co2
<¡ 75 3 oí
<ÍC/Dw c o2<tc/Dw
<ÍC/Dw c o2<,C/Dw
522: "<C/3w
2 o 2<C/5w
5 5

ryv ^rt* V) VO 00 Os O
(N (N CN <N <N fN <N ^

The Indian Journal

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 S.C. Davar & RanjuBala

/■—s /—V vj w v* w "

c 5P e ®P "t- "i. V. 2; t •35

o Ç o e o o o o 73000 00 c 00
•— '5 •= '-O «® "> "> c/5 !2 c e ï .5 c

íi 2 5 2 > > '> '> S) <Ü > ^ '5 5 O o

iV
^ i=Vr h?p fe op fe 00 «3 ?Pç5 5P.-U 5P2
— 5 =5 a c « C Se ï E h C 1-" ï ê£ <Z J2
c

3 «Ë =1 3 I 3 I al s I > I 8 3 I ^ T3 «
cuw a-w c« 2 («2 to 2 c«2<2o.c«2c/2 w 3 oo
<%
u u c e c o .2
« 8)
03 03= 2 7Ï
co 2 72
2 a75gH « p p C z p
o - o S . <£« « íí¡ u t ¿ ü. <u t ¿ «
> C/2
< >
—;C/5 >00
o Q <250
Q5 o o
<2 J <2
ü g.SXw> Oh
<*> O-;
>00 -u
£ XO>-;

S ¿ Séo
J «IglffíiS Q 1? ? Í e o i
o* o* ¿ t/5>oûo^®c/3uS<y 5/5 3 § 1/3 3 « P* "S S ■
S; 2; d £:è§oS^£:o^S fe £ t S; a » S; £ o- §
c/5
ce u
® s s S ^ d ^
<*J 3S ,-v o o
C/)
<
<%
<
S ^ ^ <*> 2 -5 •* j-já qU.
ffl 5 ¿ 5
C/3
C/5
•—>
Cu C/3
0;_ •—»
w «¡0
"* u
in
5 5 8 J z u
c o v
c/i
O tfl V U V o
V
h» >» >» >» >» u +~> • c
'S
o -o O 00 fe o -o o <s çn
4> 1)
-C C
u.
Ô. H o. c £ o. S & « 5 =
Cd
c .5 e co £ en •- E ¡i « S.
■*-*

£ w 1 Mixed Bank Emplo Mixed Emplo Mixed [iSuS O £ S u £.S §?

60
.1
•5
1
73 'S
O
U* Indian Indian Indian Indian Foreigi
e C C
z
0 0 0
c C c

1
M
1
24
1 2 Foreigi 2 Foreigi 2 Foreigi
c C C OO OO OO CO © © r
D D D <N es <N rf Tt Tf ro ?
Uo U. u* Uh U U, tu tu tu tu
<% <% <% <% <% <%
S S S 2 2 2 S 2 2 2 2

r r Os r r
00 00 00 00 00 Os 00 00 00 00 00
0 0 © © © © © © © © ©

r ro
00 ro 00
r-> 00 00 00 T} sO Os
00
r © VO SO s© 00 © r 00
0 0 ©
* © © © © # © ©
« * *
Os w-» "*t 00 ♦ ro
O SO © © © •n r © tN
O O © © © © © © © © ©

VO
© «0 «O
24 22 440 00 00 80 204 ro 65 r

<<
03 C

* u.
u. cd
i¡5
CO "77 /—v "Tí TÎ 03 . , 03
3C /-"v
3 "D/—Vií~o «— /— «a b
« -So 730 -fe © 3 o . a g1!
o 5
je _ï ^ Ï « S'fe^o ^.S^lJSi^c©
Ja!
- o
©
u r
r " co
G

o .So -So .So>iO 0«oC«©« 0 â 0 ds ©

>
3© cd
^ 2 Ci S Ci S Ci .2 Ci c« Z Cic« c/ï
Z ciZ zCi ¡2
c¿ Ü h Ci
(N

«s m -t <r> \o p» oc © ©
ro ro ro ro ro ro co ro

The Indian Journa

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms
 Relationship between Job Satisfaction & Job Performance

00
oo 00 00 00 00\ 00 G e
.c a c G G G
£
2 eG
k
<*-*
1 ci
1
«M
2
M
o
G
— <£ 73
73
—

73 73 73 73 73 CO
C
CO CO CO CO CO CO D
.£•

ll § Ss
Le
XA
G
— •- o .© « O
M , ,

■3 t g ..sa "*3 O "3

u ¡K S . •£ T3 — £
U.
U

!<«§• 2 «Cu¡2
Oh§O■§
JJG
73 >
O •—» CO «-i o¡$ e o

c
*
o

<<<<<<< C
CLCUCUCUOuû-ÛhD ÍN

o a a a a a O 0¿
c/n 00 co on on on on CO «tí

2 2 2 2 2 2 2 22 œ

C/) (A XA t/J ws tA
1> t> Ü u o M «
M (A C/i «A tfl XA t/i o
W Ut 1 U» «M Ut t->
u
3 3 3 3 3 3 3 X
z z z Z Z Z Z tú

G G G
00 00 M §> S 00 s»
s
o
"S
o o
s s
o
S
o
s
o
Si
o
T3
c
lU lu lu u tu ÜH lu

«r> «n •r* V> «O *Tt «O V">

m m fO co CO CO CO
N?
0s
tu tu tu tu tu tu lu tu
c* # <* *5
s 2 2 2 2 2 2 2

m co co co CO CO
00 00 00 00 00 00 00 00 C
dû
o o o o o o o o

f
00 00 r r 00 00 »r»
00 00 00 r
o o o o
o O o O
* * ♦ *
* * * * # * ♦ *
V~t
rf * 00 * <N r Os
CO <N <N o <N CO <N s©
o O O o O o o O

co co CO m CO CO O
<N fS es <N es <S es *n
Os Os Os Os Os Os Os

— ... —

w w -T3 T3 "0 '"0 T3 vP

cg cg c3 <0 ccS A e
m
E E E E E E E_S¿
^ oo" ÍÍ So 5 0? ^ 0? *5 S? 00 *5 O? C H
<:©<©<©<©<®-<©<© «•« c
'©•0'©'0'o_i.oj.© S-S
<CÍ<Ci.<Ci.<:Ci-<e<CÍ<Ci(2íS
cG
O

S
'S
dû

csmTí-»nsor^00Os

The Indian Journal of Industrial Relations, Vol. 48, No. 2, October 2012 305

This content downloaded from

62.201.243.244 on Wed, 28 Sep 2022 16:36:48 UTC
All use subject to https://about.jstor.org/terms