File 1

185

Murwaningsari—The Role of Organizational Commitment and Procedural Justice in Moderating the Relationship ...

Gadjah Mada International Journal of BusinessMay-August 2008, Vol. 10, No. 2, pp. 185–210

THE ROLEOF ORGANIZATIONAL COMMITMENT ANDPROCEDURAL JUSTICE IN MODERATING

THE RELATIONSHIP BETWEENBUDGETARY PARTICIPATION AND

MANAGERIAL PERFORMANCE

Etty Murwaningsari*

The objective of this research is to identify the role of organiza-tional commitment and procedural justice in moderating the rela-tionship between budgetary participation and managerial perfor-mance.

This research utilizes the mail survey method to collect primarydata from September 2007 to mid-November 2007 by sendingquestionnaire forms to corporate managers involved in the budget-ing process.

The empirical result of this research indicates that budgetparticipation has a positive and significant influence on managerialperformance. The subsequent test finds that organizational commit-ment can strengthen the relation between budget participation andmanagerial performance. The last test shows that procedures haveno influence on the relationship between budget participation andmanagerial performance.

Keywords: budgetary participation; organizational commitment; managerial per-formance; procedural justice

*The author wrote this article while pursuing the doctoral degree in accounting from PIAUniversity of Indonesia, Jakarta.

186

Gadjah Mada International Journal of Business, May - August 2008, Vol. 10, No. 2

IntroductionBudget is one of the important

components which a firm has. An or-ganization needs budget to translateall the organization’s strategies intoshort-term and long-term plans andobjectives (Hansen and Mowen 1997).

A budgeting process is an impor-tant activity involving all manageriallevels from top managers to lowerones whose roles are preparing andevaluating various alternatives of thebudgetary objectives, and that budgetis always used as the best measure ofmanagers’ performance (Kren 1992).

One function of budgetary par-ticipation is that it is used as a mediumof communications between higher andlower managers to discuss not onlybudgetary issues but also other relatedissues (Brownell and Hirst 1998). Theeffects of participative budgeting onmanagerial performance have been atopic of interest and are widely exam-ined in the studies of managementaccounting (Lukka 1998). Brownelsuggests two reasons: (a) budgetaryparticipation is viewed as a manage-rial approach able to improve the per-formance of an organization’s mem-bers, and (b) many studies testing therelationship between budgetary par-ticipation and managerial performanceshowed mixed results.

Many studies in management ac-counting are concerned with budget-ary participation issues (Brownell1981). This concern has been drivenby a view that participative budgeting

has a significant consequence to thebehavior of an organization’s mem-bers (Murray 1990). Lower managers/management whose aspirations aretaken into account and affect the bud-geting scheme will show greater re-sponsibility and moral consequence toperform better in compliance with thebudgetary targets (Supomo 1998).

Strong organizational commit-ment; as a form of acceptance of orga-nizational objectives; and a willing-ness to make efforts for the organi-zation’s best interest increase mana-gerial performance (Nouri and Parker1998).

Procedural justice is associatedwith the perception of employees lowerdown the hierarchy of all processesused by their superiors to evaluatetheir performance and as a mean ofcommunicating feedbacks and deter-mining the reward they deserve to re-ceive, such as promotion or a raise(McFarlin and Sweeney 1992). Tangand Sarfield-Baldwin (1996) suggestthat if managers apply corporate rulesequally and consistently to all employ-ees and reward them due to their per-formance and achievement without anypersonal biases, employees will per-ceive procedural justice positively.

Motivated by previous studies, thestudy reaffirms whether budgetaryparticipation has significant effects onmanagerial performance (Brownell1982). Moreover, it also tests whetherthe organizational commitment vari-able (Nouri and Parker 1996) and theprocedural justice variable (Lou and

187


Lim 2002) may moderate the relation-ship between budgetary participationand managerial performance. Basedon the aforementioned suggestions, theobjective of this study is to know:1. The budgetary participation in-

creases managerial performance.2. The degree of correlation between

budgetary participation and orga-nizational commitment corre-sponds to increased managerialperformance.

3. The degree of correlation betweenbudgetary participation and proce-dural justice corresponds to in-creased managerial performance.

Review of Relevant Studies

The Basic Theory of BudgetaryStudies

Shield and Shield (1998) examinestudies on budgetary participation andinfer that nearly all budgetary studiesare based on the following theories:a. Economic Theory. It is based on

the assumption that budget pro-vides a fair ground to select andshare information among those in-volved in a budgeting process. Theeconomic theory suggests that in-dividuals engaged in a budgetingprocess are motivated by two stimu-lants: (1) information sharing and(2) task coordination.

b. Psychological Theory. It suggeststhat budgetary participation pro-vides information sharing betweentop and lower managers/manage-

ment (Hopwood 1976). Psychologi-cal theory argues that two reasonsmake budgetary participation nec-essary (Hopwood 1976): (1) theinvolvement of top and lower man-agers in budgetary participationcontrols of asymmetric informa-tion and task uncertainty, (2) bud-getary participation allows indi-viduals to have reduced tasks andjob satisfaction, and these in turnlead to reduced budgetary gaps.Psychological theory introducesthree main factors in the top andlower managers’ involvement inbudgetary participation (Locke andLatham 1990), namely (1) valueattainment; (2) cognition; and (3)motivation.

c. Sociological Theory. Shield andShield (1998) suggest that psycho-logical theory assumes more un-certain organizational external en-vironments, indicated by increaseddifferences in an organization’sunits in terms of number and type.Consequently, an increased bud-getary participation is required tocoordinate the units’ activities andalign all activities in the organiza-tion. Sociological theory is relatedto such organizational contexts asenvironmental uncertainty, organi-zational structures, and functionaldifferentiation that affect budget-ary participation. Sociologicaltheory on which budgetary partici-pation based on the organizationalcontingency theory (Hopwood1976).

188


Contingency Approach

Contingency theory predicts thatan organization’s external environmentis bound to many uncertainties. Con-tingency theory has a principle thatthere is no a single type of organiza-tional structure nor management sys-tem which is more efficient and effec-tive for all organizations. Therefore, adifferent ecological context shouldconsider main contextual factors suchas an organization’s size, technology,and environment (Lawrence andLorsch 1967)

One function of management ac-counting system is to provide budgetthat helps managers control their ac-tivities and lessen environment uncer-tainties to achieve the organizationalobjectives (Gordon and Millar 1976).A management accounting system isgenerally a contingency approach of aconditional factor as a variable thatmoderates a relationship. Brownell(1982) examines several studies andfinds the effects of conditional factorsthat serve as moderating variables tothe relationship between independentand dependent variables. The condi-tional factors may be categorized into4 variables, i.e. cultural, organizational,interpersonal, and individual.Govindarajan (1986) suggests that it isnecessary to use the contingency ap-proach to identify various conditionsleading to more effective participativebudgeting. Researchers have docu-mented that the effectiveness of bud-getary participation depends on theorganizationally contextual factors and

the characteristics of employees’ psy-chology (Brownell 1981). In this study,the author used the contingency theoryto evaluate the relationship betweenbudgetary participation and manage-rial performance. The author choosethe contextual of the contingency fac-tors, i.e. corporate managers’ Proce-dural Justice and Organizational Com-mitment.

The Concept of FitSome approaches can be used to

define the concept of fit in the litera-ture of strategies, i.e. selection, inter-action, and system approaches (Vande Vend and Drazin 1985). The selec-tion approach does not address theperformance implications of a system(that is, the study’s main focus is toexamine the characteristics of organi-zations that implement certain sys-tems). The interaction approach, how-ever, primarily examines conditionalfactors that determine or affect theimpact of a (participative) system onperformance. Finally, the system ap-proach defines fit as the degree ofconsistence; that is, the internal con-sistence of many contingencies, struc-tures, and performance characteristics(Van de Ven and Drazin 1985). Thus,a system’s effectiveness is determinedby the extent to which conditional fac-tors meet the system’s conditional pre-requisites. This approach allows re-searchers to study the effects of condi-tional factors in the relationship be-tween a (participative) system and per-formance (Govindarajan 1986). Thisstudy extends those studies in which it

189


tests the effects or the degree of con-sistence (fit) of the conditional fac-tors, i.e. Procedural Justice and Orga-nizational Commitment, on the effectsof budgetary participation on manage-rial performance.

The Development of theHypothesis

Budgetary Participation andManagerial Performance

Budgetary participation is ex-pected to increase managers’ perfor-mance; that is, when an objective hasbeen designed and approved in aparticipatory way, employees will in-ternalize the defined objective and theywill have personal responsibility toachieve it due to their involvement inthe budgeting process (Milani 1975).Greenberg and Folger (1983) arguethat budgetary participation increasesperformance because (i) it allows lowermanagement to communicate theirneeds to the upper or top management

and (ii) it allows lower management toselect amongst available options, andthis selection exercises generate com-mitment as a form of responsibility forwhat they have chosen.

The studies of budgetary partici-pation on managerial performance haveshown mixed results. Some studiesindicate evidence that budgetary par-ticipation has a strong positive effecton managerial performance (Argyris1952). Other studies have also reportedpositive results (Merchant 1981;Brownell and Hirst 1998). Bryan andLocke (1967), however, find that bud-getary participation has a negative ef-fect on performance. Milani (1975)and Brownell and Hirst (1998) findthat budgetary participation has an in-significant effect on managerial per-formance. Brownell and McInnes(1986), who investigated 224 respon-dents of mid-level managers in manu-facturing firms, also find that highparticipation in budgeting increasesmanagerial performance.

Figure 1. The Framework

Budget ManagerialParticipation Performance

Organizational Commitment

Procedural Justice

190


The inconsistent results have ledsome Indonesian researchers to inves-tigate the variables in question.Indriantoro (1993), for example, findsa significant and positive associationbetween budgetary participation andmanagerial performance. Riyanto(1997) and Supomo (1998) find thatbudgetary participation has an insig-nificant effect on managerial perfor-mance. Poerwati (2002), in a studyinvolving 700 managers of manufac-turing firms listed in the Jakarta StockExchange and appeared in the Indone-sian Capital Market Directory, findsthat budgetary participation has an in-direct effect on managerial perfor-mance. Hariyanti and Nasir (2002), ina sample of 900 managers and heads ofdivisions comparable to managers inrank who chair functional departmentsin manufacturing firms in Indonesia,find a positive association betweenbudgetary participation and manage-rial performance. Accordingly, thefollowing hypothesis is developed:H1: Budgetary participation has a

positive effect on managerial per-formance.

Organizational CommitmentModerates BudgetaryParticipation and ManagerialPerformance

Govindarajan (1986) reconcilesthe mixed results of those studies us-ing the contingency approach thatevaluates various conditional factorsaffecting the effectiveness of budget-ing participation on managerial per-

formance. Brownell (1982) categorizesthe conditional factors (moderatingvariables) into four groups of vari-ables, i.e. cultural, organizational,cross-individual, and individual. Thecontingency approach is adopted inthis study to evaluate the effects ofbudgeting participation on managerialperformance.

In Wiener (1982), organizationalcommitment is defined as an indi-vidual’s force to support the orga-nization’s success in compliance withthe stipulated objectives and prioritizethe organization’s interests. Strong or-ganizational commitment motivatesindividuals to make efforts to achievethe organization’s objectives and goals(Angel and Perry 1981). Brownell andHirst (1986) categorizes the condi-tional factors (the moderating vari-ables) into four groups of variables:cultural, organizational, cross-indi-vidual, and individual.

The Relationship betweenBudgetary Participation andOrganizational Commitment

Nouri and Parker (1998) use asample of 135 managers of U.S. mul-tinational corporations in the chemicalindustry. The study concludes that thereis a positive association between ef-fective organizational commitment andmanagerial performance. High orga-nizational commitment is a form ofacceptance to the organization’s ob-jectives and goal and willingness tomake organizational efforts which in-crease managerial performance. Bud-

191


getary participation is not only associ-ated with budgetary targets but it alsoincludes acceptance and trust in theorganization’s values and objectivesand goal; thus, budgeting participationhas an indirect effect on organizationalcommitment.

Hariyanti and Nasir (2002), whouse a sample of 900 managers or headsof divisions comparable to managersin capacity that manage functionaldepartments in Indonesian manufac-turing companies, find a positive cor-relation between budgetary participa-tion and organizational commitment.They also find a positive relationshipbetween organizational commitmentand managerial performance.

The Relationship betweenOrganizational Commitment andManagerial Performance

Nouri and Parker (1998) suggestthat strong organizational commitmentas a form of acceptance of the or-ganization’s objectives and goal andwillingness to make efforts in the or-ganizations best interest increasesmanagerial performance. Sumarno(2005), using a sample of 90 managersresponsible for budgeting of the mainbanks’ branch offices in Jakarta, findsthat organizational commitment haspositive significant effects on the rela-tionship between budgetary participa-tion and managerial performance.Based on the findings that the relation-ship between budgetary participationand managerial performance in whicha commitment variable serves as a

moderating variable, the followinghypothesis is suggested:H2: Organizational commitment has

positively significant effects onthe relationship between budget-ary participation and managerialperformance.

Procedural Justice ModeratesBudgetary Participation andManagerial Performance

While some authors believe thatbudgetary participation is related toprocedural justice (Tibaut and Walker1975), budgetary participation and pro-cedural justice are actually different innature. Procedural justice bears abroader notion than budgetary partici-pation and is affected by many factors.

The Relationship betweenProcedural Justice andManagerial Performance

Studies of the effects of proce-dural justice on managerial perfor-mance have shown inconsistent re-sults. Some studies argue that they arepositively correlated (Earley and Lind1987), while other studies suggest thatthey are negatively correlated (Kanferet al. 1987). The conflicting resultsmay be attributed to complex and indi-rect relationship between proceduraljustice and managerial performance(Lind and Tyler 1988).

The indirect relationship betweenthe perceived procedural justice andmanagerial performance has beenpreviously examined by Lau and Lim

192


(2002) that finds a correlation betweenprocedural justice and managerial per-formance through an intervening vari-able, i.e. participative budgeting orbudgetary participation. Lau and Lim(2002) extend previous studies byLindquist (1995) and Libby (2001).Wasisto and Solihin (2004), who rep-licate the findings of Lau and Lim(2002) by using a sample of 229 man-agers of manufacturing companieslisted in the Jakarta Stock Exchangeand appeared in the 2002 IndonesianCapital Market Directory; find a sig-nificant relationship between proce-dural justice and managerial perfor-mance. The relationship is, however,indirect as it is fully mediated by bud-getary participation.

The Relationship betweenProcedural Justice andBudgetary Participation

Procedural justice may be wellregarded as the antecedent of budget-ary participation. Based on the premiseof the extent in which an organizationweighs on maintaining procedural jus-tice, this leads upper management toconsider the lower management’s par-ticipation in the budgeting process(Friedland et al. 1973).

As equality or justice is a keyfactor in a procedural selection, and asa participation-involving procedurepromotes fairness and equality, it isvery likely that an organization com-mitted to maintaining high proceduraljustice tends to select a participative

procedure (Thibaut and Walker 1975).This leads to a hypothesis of the

relationship between the three vari-ables:H3: Procedural justice has positively

significant effects on the relation-ship between budgetary partici-pation and managerial perfor-mance.

The Research Methodology

Variables and Measurement

The Dependent VariableThis study uses Managerial Per-

formance as a dependent variable. Thisdependence variable includes, amongstothers: planning, investigation, coor-dination, supervision, staff arrange-ment, negotiation, and representation(Mahoney et al. 1963). Following theprevious studies, the author use aninstrument developed by Milani (1975)to measure the budgetary participationvariable. Each of the survey’s partici-pants was asked to respond to six ques-tions to measure the degree of partici-pation and the perceived effects onthem as well as their contribution tothe budgeting process. Responses weregiven by choosing points of the ques-tions, ranging from 1 (all budgetingprocesses) to 7 (no budgeting pro-cess). The answers to the questions arearranged using the Likert scale from 1to 7. The instrument has been welltested and used by the previous re-searchers (authors) such as Brownell

193


(1982b), Brownell and McInnes(1986), Indriantoro (1993), and Gul etal. (1995). Some Indonesian research-ers have also used this instrument, forexample, Supomo (1998), Poerwati(2002), Hariyanti and Nasir (2002),Wasisto and Solihin (2004), andSumarno (2005).

The Independent VariableThe author uses the budgetary

participation as the independent vari-able in the study, indicating individu-als’ involvement and influence in thebudgeting process (Brownell 1982);and the “self rating” instrument devel-oped by Mahoney et al. (1963) to mea-sure managerial performance. Authoruse self-rating as (1) it provides anony-mous assessment, and (2) specifically,the upper management provides lessand subjective information (Heneman1974). Mahoney’s measure has twoadvantages, i.e. (1) independent as-sessment of the measure’s reliabilityand validity has generated satisfactoryresults in other studies, (2) the mea-sure explicitly reveals the multidimen-sional characteristics of managerialperformance (Govindarajan 1986 inLau and Lim 2002). Each of the re-spondents was asked to measure his orher performance. The answers to thequestions were arranged using theLikert scale from 1 to 9. The value ofthe scale indicates the value of thescore of each question. The instrumenthas been proven robust and has beenused in previous studies such as inBrownell (1982), Brownell andMcInnes (1986), and Indriantoro

(1993). It has also been used in Poerwati(2002), Hariyanti and Nasir (2002),Wasisto and Solihin (2004), andSumarno (2005).

The Moderating VariableThe moderating variable is an in-

dependent variable included in themodel as it has the contingency effectsof the dependent and independent vari-ables. In the study, the moderatingvariable includes:a. The organizational commitment,

which is defined as individual’sinternal drive to contribute to theorganization’s success and to makeefforts in the organization’s bestinterest (Wiener 1982).The organizational commitment ismeasured using the interval scale.The list of questions developed byMowday (1979) for the organiza-tional commitment is used. Theanswers to the questions were ar-ranged using the Likert scale from1 (highly disagree) to 7 (highlyagree). The scale’s value indicatesthe score value of each question.The Cronbach’s alpha > 0.6 scoreindicates the acceptance of theinstrument’s reliability used in thestudy. The reasonable degree of themeasure’s validity/robustness andreliability has been reported by pre-vious authors (Nouri and Parker1996).

b. The procedural justice, which isrelated to the lower management’sperception of all the processesimplemented by the upper manage-ment to evaluate their performance,

194


as a mean to communicate perfor-mance feedback and determine ap-propriate rewards such as promo-tion or a raise (McFarlin andSweeny 1992).

The procedural justice is mea-sured using the four-item interval scaleinstrument used by Lau and Lim (2002)and Wasisto and Solihin (2004) tomeasure the lower management’s per-ception of the procedural justice. Theanswers to the questions were arrangedusing the Likert scale from 1 to 5. Therespondents answered the questionsby choosing scales from 1 (highly un-fair), 2 (unfair), 3 (neutral), 4 (fair),and 5 (highly fair). The scale’s valueindicates the score’s value of eachquestion. The Cronbach’s alpha > 0.6score suggests the reasonableness ofthe instrument used in the study.

The Data Collecting TechniqueThe author used the mail survey

method to collect primary data fromSeptember 2007 to mid-November2007 by sending forms of question-naire to corporate managers involved

in the budgeting process. Looking atthe results of the previous studies inIndonesia, especially those using pri-mary data, the response rate rangesfrom 10 percent to 20 percent(Hariyanti and Nasir 2002).a. The Robustness Test. The robust-

ness test is used to show how real atest measures what should be mea-sured. The test is conducted byfinding the correlation of each indi-cator to its total score using thePearson Correlation test. It is ro-bust when the correlation is signifi-cant at the level of <0.05.

b. The Reliability Test. The test isused to measure the internal con-sistence of a construct’s indicators,in that how far each of the indica-tors indicate a construct. The reli-ability test is conducted by calcu-lating the magnitude of theCronbach’s coefficient alpha ofeach instrument in a variable, thatis, a positive relationship acrossitems or questions. If alpha > 0.6,the data are thus reliable.

Table 1. The Process of Distributing and Receiving the Questionnaire

Definition Amount Percentage

Questionnaire distributed 150 100%Questionnaire received 35 23 %Unusable questionnaire 3 2 %Usable questionnaire 32 21 %

195


The Normality TestThe normality test is used to test

whether the dependent and indepen-dent variables have normal distribu-tion in a regression model. The authorused the Kolmogorov-Smirnov datanormality test in the study. The testcriteria suggest that if it is significantof > 0.05, then H0 applies––indicatingthat the data distribution is normal andotherwise.

The Hypothesis Testa. An analysis regression model to

answer the Hypothesis 1. In theHypothesis Test 2, the ModerateRegression Analysis (MRA) is usedto determine the effects of interac-tions. Mathematically, the Equa-tion 1 is as follows:

Y= βo + β1 X1 + ε………… (1)

The Equation 2 is shown by theEquations (2a) dan (2b)

Y = βo + β1 X1+ β2 X2 + ε ..... (2a)

Y = βo + β1 X1+ β2 X2 + β3 X1X2 +ε.................................. ....(2b)

The hypothesis 3 is shown by theEquations (3a) dan (3b)

Y= βo + β1 X1+ β3 X3 + ε … (3a)

Y= βo + β1 X1+ β3 X3 + β4 X1X3 + ε …………………........ (3b)

Where:Y= dependent variable= Managerial

Performance;

β1, β2, β3, and β4= regression coeffi-cients;

X1= independent variable= BudgetaryParticipation;

X2, and X3= moderating variable

X2: Organizational Commitment;

X3: Procedural Justice;

X1X2= the interactions of X1 and X2;

X1X3= the interactions of X1 and X3;

ε= error

b. The Determination Coefficient.The author used the coefficient (R2)to find out how good the dependentvariable is in explaining the inde-pendent variable. The measurementexplains the Goodness of Fit of themodel in which the closer it is to 1,then the model has more Goodnessof Fit; but if it is closer to 0, then themodel has less Goodness of Fit.

c. The t-test (the Individual test) is aregression coefficient test of eachof the dependent over the indepen-dent variables to gauge the effectsof one on the other.H0 = β1 = 0 has no effect if p-value< (α = 0,05), then H0 is unaccept-able.

196


Analyses and Results

The Descriptive StatisticsThe author observed the descrip-

tive statistics based on the means andstandard deviations. The Table 2 is theresults of four variables used in thestudy.

The Table 2 shows that the re-spondents’ answer to the BudgetaryParticipation variable has a mean of4.4900. The scale 1 for this variablesuggests high participation. Therefore,the mean of 4.4900 indicates that thesample corporate managers are quiteparticipative and involved in the bud-geting process.

The Managerial Performance vari-able has a mean of 5.8094. Dividingperformance into three categories, themean value suggests the managers ofthe manufacturing companies haveaverage managerial performance. Theobserved managerial activities includeplanning, investigation, coordination,supervision, staff arrangement, nego-tiation, and representation.

The Organizational Commitmentvariable has a mean of 4.6919, indicat-ing that the managers are quite com-mitted to the organizations where theywork. The managers show that theyare proud, happy, and they care fortheir organizations as indicated by theirwillingmess to work harder, their ac-ceptance of any jobs assigned, and telltheir friends and people in generalabout their organizations.

The Table 2 also provides infor-mation about the Procedural Justicevariable that has a mean of 3.6250.Using Likert scale point 5, it suggeststhat managers find procedural justicein their workplaces. The proceduresused by the organizations include pro-motion and a raise, employee perfor-mance evaluation as well as perfor-mance feedbacks.

The standard deviation valuessuggest a variation (heterogeneity) ofdata distribution of each variable. Thestandard deviation values of the fourvariables used in the study indicatevaried data distribution.

Table 2. The Descriptive Statistics of the Variables Used in the Study

Variable Number of Scale Mean DeviationIndicator Standard

Budgetary Participation 6 1-7 4,4900 0,70442Managerial Performance 9 1-9 5,8094 0,95898Organizational Commitment 9 1-7 4,6919 1,09442Procedural Justice 4 1-5 3,6250 0,51977

197


The Data Quality Test and theClassical TestThe Robustness Test. The results ofthe robustness test for each of theconstruct are as follow:· The results of the validity test show

that there are six indicators to mea-sure the Budgetary Participation con-struct whose p-value is less than0.05, indicating that PA1 – PA6 indi-cators are robust. The correlationcoefficients range from 0.405 –0.775.

· There are nine indicators that can beused to measure the Managerial Per-formance construct whose p-valueis less than 0.05. This indicates thatKM1 – KM9 are robust. The correla-tion coefficients range from 0.510 –0.950.

· The nine indicators used to measurethe Organizational Commitment con-struct show p-value that is less than0.05. This indicates that KO1– KO9are robust. The correlation coeffi-cient is at least 0.547 and 0.934 at themost.

· The robustness test for the Proce-dural Justice construct indicates thatfour indicators have p-value less than0.05. This suggests that KP1 – KP4are robust. The minimal correlationcoefficient is of 0.694 and the maxi-mal is 0.831.

The Reliability Test. The results of thereliability test for the four constructsin the study are shown in the Table 3.

The Cronbach’s alpha coefficientin the four constructs meets the reli-ability criteria, that is, it is more than0.60. Therefore, it can be concludedthat the data used (that is, the respon-dents’ answers to each of the con-structs) are reliable.The Normality Test. The hypothesesof the normality test can be defined as:H0: The data are derived from the nor-

mal population.Ha: The data are derived from the ab-

normal population.The decision taking: If sig. < 0.05, thenH0 is unacceptable, and otherwise.

Table 3. The Realiability Test Results

Construct Number of Cronbach’sItems Coefficient Alpha

Budgetary Participation 6 0,639Managerial Performance 9 0,822Organizational Commitment 9 0,928Procedural Justice 4 0,758

198


The Kolmogorov-Smirnov basednormality test indicates that the fourvariables used have significance valueof more than 0.05. This makes H0 failand unacceptable, indicating that thevariables used in the study distributednormally.The Multicolinearity Test. TheMulticolinearity test is conducted us-ing a correlation method amongst in-dependent variables. The method ischosen as in the moderate regressionanalysis with an interaction methodthe probability of multicolinearity isvery high when VIF values are ob-served as multicolinearity indicators.The criterion showing multicolinearityexists in a correlation method is when

the correlation coefficient is > 0.80(Gujarati 2003). The results of themulticolinearity test are presented inthe Table 5.

The Table 5 suggests that the cor-relation coefficients amongst variablesare less than the maximum limit of0.80. This leads one to infer that thereis no strong relationship amongst vari-ables; in other words, it complies withmulticolinearity.The Heteroscedasticity Test. Theheteroscedasticity test used thescatterplot method. The dependentvariable at the X axis is ZPRED andthe dependent variable at the Y axis isthe residual of SRESID. The decisionis based on:

Table 4. The Normality Test for Each Variable

Variable Sig. Decision on H0

Budgetary Participation 0,177 H0 failed and rejectedManagerial Performance 0,807 H0 failed and rejectedOrganizational Commitment 0,628 H0 failed and rejectedProcedural Justice 0,090 H0 failed and rejected

Table 5. The Results of the Multicolinearity Test

PPA KO KP

PPA -KO 0,088ns -KP 0,378* 0,034ns -

* significant at the 5% level (α= 0,05)ns=not signifiicant

199


Figure 2. Result of Heteroscedasticity Test Using the Scatter Plot Method

Model 2b: Heteroscedasticity Test

Reg

ress

ion

Stud

entiz

ed R

esid

ual

Dependent Variable: KM

-2

-1

0

1

2

-3 -2 -1 1 2Regression Standrdized Predicted Value

Scatterplot

0

Model 1: Heteroscedasticity Test

Regression Standrdized Predicted Value

Reg

ress

ion

Stud

entiz

ed R

esid

ual


-4

-2

0

2

4

-4 -4 0 2 4

Scatterplot

200


Model 3b: Heteroscedasticity Test

If there is a certain pattern, say, theexisting dots form a regularly cer-tain pattern (it is wavy, it first wid-ens and then it narrows), then itindicates that the heteroscedasticityhas been violated).If no clear patterns are found, in thatthe dots are scattered above and

Source: Data processed using SPSS

Table 6. The Heteroscedasticity Results

Sig.

Variable Model 1 Model 2b Model 3b

PPA0.183 0.654 0.945KO - 0.842 -KP - - 0.729PPA*KO - 0.759 -PPA*KP - - 0.819

Reg

ress

ion

Stud

entiz

ed R

esid

ual


-2

-1

0

1

2

-3 -2 -2 0 1Regression Standrdized Predicted Value

Scatterplot

-1

-3

2

below zero at the Y axis, then theheteroscedasticity assumption ismet.

Looking at the scatterplot in theModel 1, it is inferred that the model2b and 3b have no heteroscedasticityproblems as the scatterplots do notshow clear patterns, or it is observable

201


from the scattered dots above and be-low zero at the Y axis.

In addition, the heteroscedasticitytest can also be conducted using theGlejser test by regressing each of theindependent variables to the absoluteresiduals as the dependent variables.The Heteroscedasticity hypotheses:Ho: No heteroscedasticity.Ha: Heteroscedasticity exists.

The decision is based on: if sig. <0.05, then Ho is rejected (hetero-scedasticity exists) and otherwise. TheTable 6 shows the heteroscedasticityresults.

The Table 6 indicates that thesignificance of the independent vari-ables Model 1, 2b and 3b is more than0.05. This makes Ho failed and re-jected. As a result, it is inferred that noheteroscedasticity problems exist.

The Autocorrelation TestThe autocorrelation test was con-

ducted using Durbin Watson. If DurbinWatson values hover around the ceil-

ing values (dU), it is expected that thereis no autocorrelation.

The autocorrelation hypotheses:H0: No autocorrelation series.Ha: Autocorrelation series exists.

The results of Durbin Watson es-timate are as Table 8. The Table 8suggests that:1. Durbin Watson value for model 1 is

of 1.917. Durbin Watson valuearound dU is ≤ d ≤ 4-dU, indicatingthat Ho fails to be rejected and thereis no autocorrelation in the Model 1.

2. Durbin Watson value for the model2b is of 1.930. Durbin Watson valuearound dU ≤ d ≤ 4-dU, indicating thatH0 fails to be rejected and it meansthat there is no autocorrelation inthe Model b2.

3. Durbin Watson value for the model3b is of 1.928. Durbin Watson valuearound dU ≤ d ≤ 4-dU, indicating thatH0 fails to be rejected and it meansthat there is no autocorrelation inthe model 3b.

Table 7. The Decision Criteria of the Autocorrelation Test

Zero Hypothesis Decision Criteria

No positive correlation Rejected 0 < d < dL

No positive correlation No Decision dL ≤ d ≤ dU

No negative autocorrelation Rejected 4-dL < d < 4

No negative autocorrelation No Decision 4-dU ≤ d ≤ 4-dL

No autocorrelation ( +)/(-) Failed and Rejected dU < d < 4-dU

202


Table 8. The Autocorrelation Test

Model N K’ dL dU 4-dU 4-dL DW Decision

1 32 1 1,373 1,502 2,498 2,627 1,917 No Autocorrelation

2b 32 3 1,244 1,650 2,350 2,756 1,930 No Autocorrelation

3b 32 3 1,244 1,650 2,350 2,756 1,928 No Autocorrelation

The regression test shows that, assummarized in the Table 9, p-value forthe PPA variable is of 0.000 < alpha0.05, indicating a rejected Ho1, that is,budgetary participation has a signifi-cant effect on managerial performance.The coefficient value is positive (β1=0,931), suggesting that the highermanagerial participation in the bud-geting process is, the higher the mana-gerial performance will be. Statisti-cally, this supports the test for thehypothesis 1 at the significance levelof 5 percent.

Goodness of Fit (Adj. R2)The Table 9 shows the perceived

determination coefficient of Adj.R2 isof 0.450. This suggests that the PPAindependent variable (Budgetary Par-ticipation) is able to explain the varia-tion in the Managerial Performancedependent variable by 45.0 percent,while the remaining variation is ex-plained by other factors not includedin the model.

The regression test shows that, assummarized in the Table 10, p-valuefor the interaction variable of PPA*KOis of 0.078 < alpha 0.10. This makesHo2 rejected, suggesting that organi-zational commitment has a significant

effect on the relationship between bud-getary participation and managerialperformance. The coefficient is posi-tive (β3= 0.251), indicating that theinteraction between budgetary partici-pation and organizational commitmentincreases managerial performance, orin other words the effect of the organi-zational commitment strengthens therelationship between budgetary par-ticipation and managerial performance.Statistically, this supports the test forthe hypothesis 2 at the significancelevel of 10 percent.

To gather evidence of whetherorganizational commitment enhancesthe relationship between budgetaryparticipation and managerial perfor-mance, however, further tests shouldbe conducted. The moderated regres-sion analysis in the Model 2 resulted inan estimate:KM= 5,353 - 0,130PPA - 0,909KO +

0,251PPA*KOThe total effects of the PPA vari-

able interacting with the KO variableand substituted from the KO variablegenerate this Equation 4.

βtotal = β1 + β3KO.....................(4)

203


Table 9. The Results of the Regression Test for the Hypothesis 1

KM = βββββ0 + βββββ1PPA + εεεεε

t-Test

Variable B t-value p-value Adj. R2 F Test(p-value)

(Constant) 1.628 1.977 0.057 26.3740.450

PPA 0.931 5.136 0.000 (0.000)

Source: Data processed using SPSS

Table 10. The Regression Test Results for the Hypothesis 2

t-Test


Model 2a: KM = βββββ0 + βββββ1PPA + βββββ2KO + εεεεε

(Constant) 0.876 0.944 0.353 15.204PPA 0.906 5.110 0.000 0.478 0000KO 0.185 1.617 0.117

Model 2b: KM = βββββ0 + βββββ1PPA + βββββ2KO + βββββ3PPA*KO + εεεεε

(Constant) 5,353 2.055 0.049PPA -0.130 -0.096 0.827 0.517 12.073KO -0.909 -1.037 0.146 (0.000)PPA*KO 0.251 1.535 0.078

Source: The data are processed using the SPSS

204


The regression estimate makes β1=-0,130 dan β3= 0,251. The KO vari-able scale ranges from 1–7. If the KOis in the scale 1 is (β2= 1), the totaleffect will be of 0.121 {βtotal= -0,130 +0,251(1). However, if the KO is in thescale 2, then the total effect is of 0.372{βtotal= -0,130 + 0,251(2). Similarly, ifthe KO is in the maximum scale 7 thetotal effect increases accordingly. Thisprovides evidence that the effect ofbudgetary participation on managerialperformance is higher when organiza-tional commitment increases.

The determination coefficient(Adj. R2) in the Model 1 is of 0.478,while Adj. R2 in the Model 2 is of0.517. This indicates that Adj. R2 in-creases to 0.039 as PPA (BudgetaryParticipation) interacts with KO (Or-ganizational Commitment) over KM(Managerial Performance).

Table 11 shows that p-value forthe PPA*KP interaction variable is of0.589 > alpha 0.05, indicating thatHo3 fails and unacceptable––the pro-cedural justice has no significant ef-fect on the relationship between bud-getary participation and managerialperformance. The coefficient is, how-ever, negative (β3 = -0,197); this sug-gests that the interaction between bud-getary participation and proceduraljustice decreases managerial perfor-mance.

The determination coefficient(Adj. R2) in the model 3a is of 0.432,while Adj. R2 in the model 3b is of0.421. This suggests that Adj. R2 de-creases to 0.014 when PPA interactswith KP over the dependent variable,KM.

Table 11. The Regression Test Results for Hypothesis 3

t-Test


Model 3a: KM = βββββ0 + βββββ1PPA + βββββ2KP + εεεεε

(Constant) 1.351 1.285 0.209PPA 0.899 4.527 0.000 0.435 12.925KP 0.117 0.435 0.667 (0.000)

Model 3b : KM = βββββ0 + βββββ1PPA + βββββ2KP + βββββ3PPA*KP + εεεεε(Constant) -1.808 -0.308 0.761PPA 1.597 1.235 0.227 0.421 8.508KP 1.013 0.610 0.547 (0.000)PPA*KP -0.197 -0.547 0.589

Source: The Data are processed using the SPSS

205


Conclusion, Limitation, andDirections

ConclusionThe results of the three hypoth-

eses tested in the study infer that:1. Budgetary participation has posi-

tive effects on managerial perfor-mance, that is, the higher the man-agers’ participation in the budget-ing process, the better managerialperformance will be. When bud-getary objectives and goals are de-signed and agreed by the manage-ment of all levels, this generatesemployees’ personal responsibil-ity to help achieve the objectives asthey are involved in the budgetingprocess (Milani 1975). Greenbergand Folger (1983) argue that bud-getary participation increases per-formance as (i) budgetary partici-pation allows lower managementto communicate their thoughts andopinions to the upper managementand (ii) budgetary participation al-lows lower management to chooseoptions, and this involvementmakes them committed to and re-sponsible for what they have cho-sen. This supports Brownell (1982),Brownell and McInnes (1986),Indriantoro (1993), and Hariyantiand Nasir (2002). The hypotesis 1is therefore supported.

2. The test for the hypothesis 2 showsthat organizational commitmentserves as the moderating variable.The regression coefficient in theinteraction variable (PPA and KO)indicates a positive direction, sug-

gesting that organizational com-mitment increases the relationshipbetween budgetary participationand managerial performance. Thisis consistent with Govindarajan(1986) that used a contingency ap-proach to evaluate various condi-tional factors affecting the effec-tiveness of budgetary participationon managerial performance, andone of the conditional factors isorganizational. This supports Nouriand Parker (1998), Hariyanti andNasir (2002), and Sumarno (2005).The Hypothesis 2 is thus supported.

3. The test for the hypothesis 3 doesnot find evidence that proceduraljustice is a moderating variable.This suggests that the relationshipbetween budgetary participationand managerial performance is af-fected by procedural justice. Thisis derived from the negatively andinsignificantly signed coefficients.This finding does not support theprevious study by Wasisto andSolihin (2004).

LimitationsThe study, as other empirical stud-

ies, has several limitations:1. The survey method of question-

naire is the only instrument used inthe study. The problem that mayarise from using this method is thelow response rates as compared tothe interviewing method or directlyinvolved in a firm’s activities. An-other limitation is the internal va-lidity, such as the use of self-ratingmethod (Mahoney et al. 1963) in

206


the managerial performance mea-surement tends to generate leniencybiases.

2. Sample firms are of those firmswhose employees are > 100, sug-gesting that the results cannot begeneralized for firms with employ-ees < 100.

3. Test instruments: the tests for hy-potheses 2 and 3 used the Moder-ated Regression Analysis with aninteraction approach; this gener-ates very high multicolinearity inone (or more) independent vari-able.

Directions for Future StudiesThe limitations in this study sug-

gest that:1. Data should be gathered through a

survey method as well as inter-viewing methods or researchersshould be directly involved in afirm’s activities. The interviewingmethod is useful to further analyzespecific factors related to a firm’sactivities to increase the firm’s op-erational and financial perfor-mance.

2. To analyze the interaction effectsamongst independent variables,other test instruments should beemployed such as the ModeratedRegression Analysis (MRA)method with a (residual) fit ap-proach developed to overcome themulticolinearity problems in theinteraction approach. One can alsouse an MRA method with the Ab-solute Value.

3. Consistent with Govindarajan(1986) that used the contingencyapproach to evaluate various con-ditional factors affecting the affec-tivity of budgetary participation onmanagerial performance, the futurestudies should use an organization’sconditional factors other than orga-nizational commitment. Such fac-tors are management accounting ororganizational structures observedfrom delegating control (decentrali-zation or centralization). Other fac-tors are the organization’s culturaland individual.

Managerial ImplicationsIt is hoped that the findings of this

study may contribute to the organiza-tional practices in general, primarilythose related to the interaction imple-mentation between budgetary partici-pation and conditional factors, i.e. or-ganizations and the interactions be-tween budgetary participation and pro-cedural justice.

Even though the hypothesis 3 doesnot support the previous studies, theprocedural justice is expected to be afactor able to affect the relationshipbetween budgetary participation andmanagerial performance. Future stud-ies should improve the limitationsfound in the study. The effects of orga-nizational commitment and proceduraljustice on the relationship betweenbudgetary participation and manage-rial performance need to be reexam-ined to find their consistencies withfuture studies.

207


ReferencesAbdul Rahman, F., and B. Supomo. 2003. Pengaruh partisipasi anggaran dan keterlibatan

kerja terhadap senjangan anggaran dengan komitmen organisasi sebagai variabelmoderating. Jurnal Bisnis dan Akuntansi 5 (2 August): 127-146.

Angel, H. L., and J. L. Perry. 1981. An empirical assessment of organizational commitmentand organizational effectiveness. Administrative Science Quarterly 26: 1-14.

Argyris, C. 1952. The impact of budgets on people. The School of Business and PublicAdministration. Cornell University

Brownell, P. 1981. Participation in budgeting, locus of control and organizationaleffectiveness. The Accounting Review 56 (October): 844-860

Brownell, P. 1982. The role of accounting data in performance evaluation, budgetaryparticipation, and organization effectiveness. Journal of Accounting Research 20(1): 12-27.

Brownell, P., and M. McInnes. 1986. Budgetary participation, motivation and managerialperformance. The Accounting Review 61 (4): 587-600.

Brownell, P., and M. Hirst. 1986. Reliance on accounting information, budgetary partici-pation, and task uncertainity: Test of three-way interaction. Journal of AccountingResearch 24 (Autumn): 241-251.

Bryan, J. F., and E. A. Locke. 1976. Goal setting as a man of increasing motivation. Journalof Applied Psychology 51: 274-277.

Dunk, A. S. 1993. The effect of budget emphasis and information asymmetry on therelationship between budgetary participation and slack. The Accounting Review.

Dunk, A. S. 1989. Budget emphasis, budgetary participation and managerial performance:A note, accounting. Organizations and Society 14 (4): 321-355.

Early, P. C., and E. A. Lind. 1987. Procedural justice and participation in task selection:The role of control in mediating justice judgment. Journal of Personality and SocialPsychology: 1148-1160.

Friedland, N., J. Thibaut, and L. Walker. 1973. Some determinants of the violation of rules.Journal of Applied Social Psychology 3: 103-118.

Ghozali, I. 2001. Aplikasi Analisis Multivariate dengan Program SPSS. Semarang,Indonesia: Badan Penerbit Universitas Diponegoro.

Gordon, L. A., and Millar. 1976. A contingency framework for the design of accountinginformation system. Accounting, Organizations and Society.

Govindarajan, V. 1986. Impact of parcipation in the budgetary process on managerialattitudes and performance: Universalistic and contingency perspectives. DecisionSciences 17: 496-516.

Greenberg, J., and R. Folger. 1983. Procedural justice, participation and the fair processeffect in groups and organization. In Paulus (ed.), Basic Group Process. NJ:Springer-Verlag: 235-256.

208


Gujarati, D. N. 2003. Basic Econometrics (4th ed.). New York: Mc-Graw Hill.Heneman, H. G. 1974. Comparison of self and superior ratings of managerial performance.

Journal of Applied Psychology 59: 638-642.Hansen, D. R., and M. M. Mowen. 2004. Management Accounting (7th ed.). New York:

Thomson, Jakarta: Salemba Empat.Haryanti, W., and M. Nasir. 2002. Pengaruh partisipasi penyusunan anggaran terhadap

kinerja manajerial: Peran kecukupan anggaran dan komitmen organisasi sebagaivariabel intervening. SNA.

Hopwood, A. G. 1976. Accounting and Human Behavior. NJ: Prentice Hall.Indriantoro, N. 1993. The effect of participative budgeting on job performance and job

satisfaction with locus of control and cultural dimensions Lexington as moderatingvariables. Ph.D. Dissertation. Lexington: University of Kentucky.

Kanfer, R, J. Sawyer, P. C. Early, and E. A. Lind. 1987. Participation in task evaluationprocedures: The effects of influential opinion expression and knowledge evaluativecriteria on attitudes and performances. Social Justice Research.

Kaplan, R. S. 1984. The evolution of management accounting. The Accounting Review.Kren, L. 1992. Budgetary participation and managerial performance: The impact of

information and environmental volatility. The Accounting Review 67.Lau, C. M., and E. W. Lim. 2002. The intervening effects of participation on the

relationship between procedural justice and managerial performance. British Ac-counting Review 34.

Lawrence, P. R., and J. W. Lorsch. 1967. Differentiation and integration in complexorganizations. Administrative Science Quarterly 12: 1-30.

Libby, T. 2001. Referent cognition and budgetary fairness: A research note. Journal ofManagement Accounting Research 13: 91-105.

Lind, E. A., and T. R. Tyler. 1988. The Social Psychology of Procedural Justice (2th ed.):ISBN 0306427265. Springer.

Lindquist, T. M. 1995. Fairness as an antecedent to participative budgeting: Examining theeffects of distributive justice. Journal of Management Accounting Research 17: 119-141.

Locke, E. A. 1967. Motivational effect of knowledge of results: Knowledge or goal-setting? Journal of Applied Psychology 51 (4): 324-329.

Locke, E. A., and G. P. Latham. 1990. Work motivation and satisfaction: Light at the endof the tunnel. Psychological Sciences. 1 (4): 240-246.

Lukka, K. 1988. Budgetary biasing in organization: Theoretical framework and empiricalevidence. Accounting, Organization and society 13. New York: John Wiley & Sons:281-301.

Merchant, D. 1981. The design of corporate budgeting system: Influence on managerialbehavior and performance. The Accounting Review 56: 813-829.

209


Mahoney, T. A, T. H. Jerdee, and S. J. Carrol. 1963. Development of ManagerialPerformance: A Research Approach. Cincinnati : South Western Publishing.

McFarlin, D. B., and P. D. Sweeney. 1992. Distributive and procedural justice aspredictors of satisfaction with personal and organizational outcomes. Academy ofManagement Journal 35 (3): 626-637.

Milani, K. 1975. The relationship of participation in budget-setting to industrial supervisorperformance and attitudes: A field study. The Accounting Review.

Mowday, R. T. 1979. The measurement of organizational commitment. Journal ofVocational Behavior 14 (2): 224-247.

Murray, D. 1990. The Performance Effects of participative budgeting: An integration ofintervening and moderating variables. Behavior Research in Accounting.

Nouri, H., and R. Parker. 1996. The effect of organizational commitment on the relationbetween budgetary participation and budgetary slack. Behavioral Research inAccounting 8: 76-90.

Nouri, H., and R. Parker. 1998. The relationship between budget participation and jobperformance: The roles of budget adequacy and organizational commitment. Ac-counting, Organization and Society 23 (5-6): 467-483.

Poerwati, T. 2002. Pengaruh partisipasi penyusunan anggaran terhadap kinerja manajerial:Budaya organisasi dan motivasi sebagai variabel moderating. Paper presented onSimposium Nasional Akuntansi 5: 737-755.

Riyanto, B. 1997. Strategic uncertainity, management accounting and performance: Anempirical investigation of a contingency theory at the firm level. In B. Riyanto (2001)Jurnal Riset Akuntansi, Managemen, Ekonomi 1 (1): 13-32.

Riyanto, B. 1999. The effect of attitude, strategy and decentralization on the effectivenessof budget participation. Jurnal Riset Akuntansi Indonesia 2 (2) (Juli): 136-153.

Shields, J. F., and M. D. Shields. 1998. Antecedents of participate budgeting. Accounting,Organizations and Society 23 (1): 49-76.

Sumarno, J. 2005. Pengaruh komitmen organisasi dan gaya kepemimpinan terhadaphubungan antara partisipasi anggaran dan kinerja manajerial. Paper presented onSimposium Nasional Akuntansi VIII (15-16 September). Solo, Indonesia: 586-606.

Supomo, B. 1998. Pengaruh struktur dan kultur organisasional terhadap efektifitasanggaran partisipatif dalam peningkatan kinerja studi empiris pada perusahaanmanufaktur di Indonesia. Thesis unpublished. Yogyakarta, Indonesia: UniversitasGajah Mada Program Pasca Sarjana Magister Sains Akuntansi.

Tang, T. L., and L. J. Sarfield-Baldwin. 1996. Distributive and procedural justice as relatedto satisfaction and commitment. SAM Advance Management Journal.

Tibaut, J., and L. Walker. 1975. Procedural Justice: A Psychological Analysis. Hillside,NJ: Lawrence Erlbaum.

210


Van de Ven, and R. Drazin. 1985. The concept of fit in Contingency Theory. In B. Riyanto(2001) Jurnal Riset Akuntansi dan Bisnis. Surabaya, Indonesia.

Wasisto, A., and M. Sholihin. 2004. Peran partisipasi penganggaran dalam hubunganantara keadilan prosedural dengan kinerja manajerial dan kepuasan kerja. Paperpresented on Simposium Nasional Akuntansi VII: 565-580.

Wiener, Y. 1982. Commitment in organization: A normative view. Academy of Manage-ment Review 7: 418-428

Waterhouse, J. H., and P. Tiessen. 1978. A contingency framework for managementaccounting system research. Accounting, Organization and Society 3 (1).

Documents

File 1