An Analytical Approach of Cluster Analysis towards Various Manifestations of Organizational Conflict in Public and Private Sector Commercial Banks

Abstract

Conflict exists throughout environments of all kinds. So every employee wants a healthy and competitive working environment in their organization. The main objective of this research article is to explore the factors concerning destructive or dysfunctional impacts of conflict in public and private sector banking organizations selected under study. After the extraction of factors concerning dysfunctional impacts, hierarchical cluster analysis has been employed to classify various observations or cases in clusters resulted into two-cluster solution and k-means cluster analysis has been further applied to find out the cluster membership according to various observations towards manifestations or impacts of organizational conflict in the selected banks under study. The sample includes 541 bank employees from 20 commercial banks situated in Amritsar, Jalandhar and Ludhiana cities of Punjab. Eight major factors have been extracted with the help of exploratory factor analysis, namely, (a) declining performance, profits and management creditability, (b) health and psychological problems, (c) detrimental impact upon working environment and organization, (d) declining cooperation, collaboration and efficiency among employees, (e) disappointments and widened gap of misunderstanding, (f) declining trust, morale and motivating power, (g) interruptions in organizational operations and (h) job dissatisfaction and resultant frustration. Further, analysis of variance (ANOVA) revealed significant results towards various clusters so formed with the help of the multivariate technique of cluster analysis.

Keywords

Organizational Conflict Dysfunctional Impacts Exploratory Factor Analysis Hierarchical Cluster Analysis K-means Cluster Analysis Public Sector Banks and Private Sector Banks

Introduction

Life is not a grand harmony, Conflict exists.

– Newman and Warsen

The uncertainty that accompanies organizational change heightens prospects for intra-organizational conflict. The importance of human factor working in organizations in current scenario cannot be ignored. In modern times, conflict has been multiplying at a very fast rate; whether in a manufacturing enterprise or the service organizations like banks, conflict is neither prevented nor resolved. It is apt to take a turn for the worse which may pose the challenge to the very existence of the organizations (Robbins et al., 2012). Keeping this factor in present view, the present study seeks to estimate the various dysfunctional impacts of organizational conflict. With the reduction in the destructive impacts of conflict in any organization, the work environment will improve and benefit both the employees and organization. So, variables have been derived from literature review, intuitive knowledge of the researchers and consultancy from the employees of the selected banks under study, having dysfunctional impacts upon human resource as well as upon banking organizations that have been undertaken. Various authors contribute towards the area of analyzing the destructive consequences of organizational conflict. The major consequences explored by Bahadur (1993) in Nepalese organizations were poor coordination followed by duplication of efforts, poor performance and wrong communication. Holton (1998) provides the 3-step model to resolve conflict by identifying the major consequences of organizational conflict. Friedman et al. (2000) in their study ‘What goes around comes around: The impact of personal conflict style on work conflict and stress’, focused on staling the effect of personal conflict on work and stress resulted thereof. The investigation had been made for various consequences of conflict in rapidly changing medical environment. Tonder et al. (2008) revealed the major impacts of organizational conflict that were: (a) performance decline, (b) hostility towards colleagues, (c) decline cooperation, (d) experience depression, (e) contemplate job change, (f) withdrawing from colleagues and (g) health problems. The results further revealed that the major effect and least effect of conflicts were performance decline (48.3 per cent) and decline in cooperation (17.7 per cent) with their respective percentages. Riaz and Junaid (2011) further identified low morale, reduced productivity, excessive employee turnover, quality problems, inability to meet deadlines, increased supervision cost, employee theft and damage, increasing psychological problems, decreasing collaboration, undesired behaviour, group think, trust deficit, fractional activities, damaged management creditability, dissatisfied customer and diminishing profits as major dysfunctional impacts of conflict. Even though review of literature is quite exhaustive in nature, certain gaps in empirical as well as theoretical grounds are still prevalent. The present study will try to cover some of the research gaps as suggested by Lkeda et al. (2005) and Adebile and Ojo (2012) pertaining to empirical work on exploring dysfunctional impacts of organizational conflict in public and private sector banks selected under study.

Objectives and Research Methodology

The main objectives of research article are (a) to explore the factors concerning destructive or dysfunctional impacts of conflict in public and private sector banking organizations selected under study and (b) to probe the consistency of various manifestations of conflict perceived across various clusters of respondents so formed with the help of cluster analysis. The sample of the study includes 541 bank employees from 20 commercial banks situated in Amritsar, Jalandhar and Ludhiana cities of Punjab. Ten banks each from public sector and private sector have been selected on the basis of highest number of employees as depicted in Table 1 (Prowess Software and annual reports of these banks March, 2013). Convenience cum judgement sampling technique had been chosen for the purpose of study. In this type of sampling, items for the sample are selected deliberately by the researcher; his choice concerning the items supreme. In other words, the organizers of the inquiry purposively choose the particular units of the universe for constituting a sample on the basis that the small mass that they so select out of the huge one will be typical or representative of the whole (Kothari, 2004). The pretested structured questionnaire has been utilized under the study and responses were recorded on 5-point Likert scale. Perquisites of reliability and validity have also been ensured. Data analysis has been done with the help of SPSS V. 18. Various statistical tools such as exploratory factor analysis (EFA), hierarchical cluster analysis, k-means cluster analysis and ANOVA have been applied to analyze the data (Gupta and Gupta, 2011).

Table 1.

Eigenvalues and Variance Explained by Consequences of Conflict in Selected Public and Private Sector Banks

Factor Coding	Initial Eigenvalues			Extraction Sums of Squared Loadings			Rotation Sums of Squared Loadings
	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %
EF-1	4.160	19.811	19.811	4.160	19.811	19.811	2.280	10.857	10.857
EF-2	2.285	10.883	30.694	2.285	10.883	30.694	1.980	9.431	20.288
EF-3	1.969	9.376	40.069	1.969	9.376	40.069	1.969	9.375	29.662
EF-4	1.901	9.051	49.120	1.901	9.051	49.120	1.951	9.288	38.951
EF-5	1.456	6.933	56.053	1.456	6.933	56.053	1.818	8.656	47.606
EF-6	1.210	5.764	61.817	1.210	5.764	61.817	1.749	8.329	55.935
EF-7	1.122	5.341	67.158	1.122	5.341	67.158	1.746	8.314	64.249
EF-8	1.008	4.801	71.958	1.008	4.801	71.958	1.619	7.709	71.958

Notes: EFA: principle component method; KMO: 0.65; Barlett’s test of sphericity (χ²) = 4,088.432; df = 210; p = 0.000* [indicative of sample adequacy and significant p value indicates that the correlation matrix of variables under study is not an identity matrix and we can proceed further with application of factor analysis. (Public Sector Banks under sample: State Bank of India, Punjab National Bank, Canara Bank, Bank of Baroda, Bank of India, Central Bank of India, Union Bank of India, Syndicate Bank and Indian Overseas Bank; Private Sector Banks under sample: ICICI Bank, HDFC Bank, AXIS Bank, Kotak Mahindra Bank, Jammu & Kashmir Bank, ING Vysya Bank, Indusind Bank, Karnataka Bank, South Indian Bank and Karur Vysya Bank)].

Exploratory Factor Analysis of Manifestations of Conflict at Workplace

In order to analyze the dysfunctional impacts or manifestations of organizational conflict in selected public and private sector banks under study, reliability analysis has been conducted first to identify and eliminate irreverent variables. The variables e12 and e19 have been excluded and the overall composite value of alpha has increased to 0.777 (Nunnally and Bernstein, 1994). So, the process of checking reliability of the scale provided 21-item scale named as E-scale. Again consistent with the exploratory aims of the study, factor analysis has been performed in order to examine the sample adequacy and appropriateness of data so collected. For identification of various factors concerning consequences or manifestation of conflict, 22 statements were further put to EFA with the help of principal component method and VARIMAX (Varimax is the name of particular type of rotation during the data reduction technique or factor analysis. For more information, refer http://www.utdallas.edu/~herve/Abdi-rotations-pretty.pdf) approach of orthogonal rotation to simplify the columns of the factor matrix providing simplified and more meaningful factors. Items having loadings less than 0.40 have been dropped (Hair et al., 1998). Factor analysis resulted into eight factor solution (Table 1). The value of KMO measure of sampling adequacy comes out 0.65, indicating adequacy of the sample size for applying factor analysis (Kaiser, 1970). Diagonals of anti-image correlation matrices reveal KMO statistic for each variable that varies within 0.5 to 0.7 and found sufficiently high for the items under study. Anti-image matrices concerning consequences of conflict at workplace displayed in Tables 2a and 2b clearly reveal anti-image correlations that indicate the existence of true factors. Partial correlations are low as compared to others. So, the true factors exist in the data under analysis. The Bartlett’s test of sphericity is a test statistic used to examine the hypothesis that the variables considered in the study are uncorrelated in the population. It also examines that the null hypothesis that correlation matrix of variables under study is an identity matrix, that is, each variable correlates perfectly with itself but no correlation with other variables. In the current study, the test value (χ²) = 4,088.432, which is found highly significant and p = 0.000* that indicates the rejection of null hypothesis and concluded that correlation matrix of variables under study is not an identity matrix. The correlation matrix of consequences or impacts of organizational conflict computed mean correlations that vary between 0.010 and 0.920. High correlation has been found between the variables under study. So, factor analysis can be performed. Visual examination of correlation matrix examines high correlation among the variables. This indicates that the variables can be categorized into homogeneous sets and each set of variables measures the same underlying dimension.

Table 2a.

Anti-image Matrices (Final) of Consequences of Conflict at Workplace; Sampling Adequacy Analysis (Kaiser-Meyer-Olkin Measure; Barlett’s Test)

	e1	e2	e3	e4	e5	e6	e7	e8	e9	e10	e11
e1	0.610*	−0.250	−0.136	−0.090	−0.159	0.094	−0.059	−0.243	−0.013	−0.013	0.094
e2	−0.250	0.564*	−0.120	0.136	0.162	0.027	−0.063	−0.055	−0.046	−0.030	0.194
e3	−0.136	−0.120	0.615*	0.043	−0.031	0.014	0.009	0.048	−0.046	0.042	−0.153
e4	−0.090	0.136	0.043	0.681*	0.102	0.026	0.107	−0.195	−0.117	−0.087	0.172
e5	−0.159	0.162	−0.031	0.102	0.574*	0.061	0.038	−0.136	−0.006	−0.401	0.056
e6	0.094	0.027	0.014	0.026	0.061	0.739*	−0.010	−0.096	0.051	−0.179	−0.084
e7	−0.059	−0.063	0.009	0.107	0.038	−0.010	0.634*	−0.390	−0.165	−0.153	−0.235
e8	−0.243	−0.055	0.048	−0.195	−0.136	−0.096	−0.390	0.540*	0.037	0.261	0.114
e9	−0.013	−0.046	−0.046	−0.117	−0.006	0.051	−0.165	0.037	0.592*	−0.061	−0.110
e10	−0.013	−0.030	0.042	−0.087	−0.401	−0.179	−0.153	0.261	−0.061	0.658*	0.181
e11	0.094	0.194	−0.153	0.172	0.056	−0.084	−0.235	0.114	−0.110	0.181	0.547*

Table 2b.

Anti-image Matrices (Final) of Consequences of Conflict at Workplace; Sampling Adequacy Analysis (Kaiser-Meyer-Olkin Measure; Barlett’s Test)

	e13	e14	e15	e16	e17	e18	e20	e21	e22	e23
e13	0.680*	0.070	−0.210	0.163	0.005	−0.243	−0.047	−0.079	−0.163	0.115
e14	0.070	0.755*	−0.116	0.008	0.031	−0.321	0.007	−0.086	0.043	0.098
e15	−0.210	−0.116	0.655*	−0.137	−0.228	0.295	−0.030	0.145	−0.068	−0.572
e16	0.163	0.008	−0.137	0.536*	0.019	−0.158	0.198	0.163	−0.028	0.084
e17	0.005	0.031	−0.228	0.019	0.680*	−0.062	0.166	0.040	0.044	−0.100
e18	−0.243	−0.321	0.295	−0.158	−0.062	0.542*	0.021	−0.089	0.048	−0.423
e20	−0.047	0.007	−0.030	0.198	0.166	0.021	0.527*	0.200	0.110	−0.243
e21	−0.079	−0.086	0.145	0.163	0.040	−0.089	0.200	0.661*	−0.020	−0.250
e22	−0.163	0.043	−0.068	−0.028	0.044	0.048	0.110	−0.020	0.560*	−0.201
e23	0.115	0.098	−0.572	0.084	−0.100	−0.423	−0.243	−0.250	−0.201	0.594*

Notes: *Measures of sampling adequacy (MSA); KMO: 0.65; Barlett’s test of sphericity (χ²) = 4,088.432; df = 210; p = 0.000* (indicative of sample adequacy and significant p value indicates that the correlation matrix of variables under study is not an identity matrix and we can proceed further with application of factor analysis).

Hence, the results above revealed that the data set is fit enough to perform factor analysis. The selected 21 variables scale (Table 3) has been subjected to EFA. Also based upon the thumb rule, the sample size which should be at least five times the number of variables to be analyzed has been ensured and items for which factor loadings were found less than 0.40 have been dropped to purify scale and judging construct validity. According to criteria given by Kaiser (1960), factors having eigenvalue greater than 1 were kept only. It depicts that the variance explained by that factor is more than unexplained variance. Table 2 displayed the summary of extracted eight factors, their initial eigenvalues and percentage of variance explained by each factor and total explained variance. The extracted factors depicted that total explained variance is 71.958 per cent which is considered highly sufficient as these eight factors together explain 71.958 per cent of the total variance. Further, Table 3 provides a summary of the factor analysis results displaying the mean importance, factor loadings, percentage of variance explained, eigenvalues and Cronbach’s alpha (α) for various factor extracted with the help of EFA. Coefficients having loading less than 0.40 have been suppressed in the analysis because loading less than 0.40 represents low correlation and would be insufficient. The criteria for convergent validity have been satisfied because all the variables within a single factor are highly correlated. Convergent validity can be very much evident from factor loadings mentioned in Table 3 as they are highly correlated. Convergent validity has been satisfied by all three aspects, that is, construct reliability (α > 0.60); all dominant variables within factor have factor loading greater than 0.45 (λ > 0.45) and variance extracted is greater than 0.5 for all constructs. Discriminant validity has also been satisfied as the variables are more strongly related to its own factors rather than to other factors (λ > 0.45 in own particular construct and λ < 0.15 in other constructs). The labelling of various factors and their description have been mentioned as follows.

Table 3.

Summary of Mean Importance, Factor Loading, % of Variance Explained, Eigenvalues and Cronbach’s Alpha of Extracted Factors

Factor Coding	Factor Statements	Mean Importance	Factor Loading	Construct Reliability (a)
EF-1	Declining Performance, Profits & Management Creditability	3.8453		0.808
E23-Damaged management creditability		3.82	0.842
E15-Employee’s performance decline		3.69	0.767
E18-Unfavourable impacts on profits		4.03	0.517
EF-2	Health & Psychological Problems	3.0462		0.781
E8-Health problems		2.88	0.842
E16-Psychological problems		3.21	0.820
EF-3	Detrimental Impact upon Working Environment & Organization	3.1950		0.762
E5-Withdrawing from colleagues		3.75	0.689
E10-Growing absenteeism		2.66	0.620
E4-Job change		3.58	0.573
E20-Strikes and lockout scenario		2.79	0.424
EF-4	Declining Cooperation, Collaboration & Efficiency among Employees	3.1633		0.741
E11-Decreasing collaboration		2.70	0.807
E7-Declining employee’s cooperation		2.65	0.537
E13-Adverse impact on efficiency & effectiveness of employees		4.15	0.532
EF-5	Disappointments and Widened Gap of Misunderstanding	3.5702		0.702
E9-Employee experiences disappointments		3.75	0.802
E14-Widened misunderstandings due to conflict		3.39	0.568
EF-6	Declining Trust, Morale & Motivating Power	3.3346		0.755
E22-Low morale & Motivating power		3.01	0.792
E3-Trust deficit		3.66	0.559
EF-7	Interruptions in Organizational Operations	3.4948		0.712
E6-Delays in working activities		3.65	0.716
E17-Disorders in working environment of bank		3.65	0.680
E21-Hindering bank operations		3.19	0.567
EF-8	Job Dissatisfaction and Resultant Frustration	3.5619		0.726
E1-Dissatisfaction among employees & resultant frustration		3.58	0.727
E2-Dissatisfaction among customers		3.55	0.724

Source: Variables selected from literature survey: Bercovitch (1982), Wall and Callistor (1995), Lkeda et al. (2005), Tonder et al. (2008), Hotepo et al. (2010), Riaz and Junaid (2011), Adebile and Ojo (2012) and Kaur (2012).

Factor-1 Declining Performance, Profits and Management Creditability

This is the first factor that explains 19.811 per cent of variance (maximum) and has the highest eigenvalue of 4.160. Three statements have been loaded on this factor. The highest loading of 0.842 is for variable ‘Damaged Management Creditability’ followed by ‘Employee’s Performance Decline’ (0.767) and ‘Unfavourable Impacts on Profits’ (0.517) with their respective loadings. This factor covers all the statements concerning decline in profits, performance and management creditability and named as ‘Declining Performance, Profits & Management Creditability’.

Factor-2 Health and Psychological Problems

This is the second factor that explains 10.833 per cent of variance explained and has eigenvalue of 2.285. This factor consists of two variables out of which variable ‘Health Problems’ has a highest factor loading of 0.842 followed by variable ‘Psychological Problems’ with loading of 0.820. ‘Health and Psychological Problems’ are the major impacts of conflict that causes further destructions to organization.

Factor-3 Detrimental Impact upon Working Environment and Organization

This factor explains 9.376 per cent of total variance and has eigenvalue of 1.969. Four statements have been loaded on this factor. Variable ‘Withdrawing from Colleagues’ has a maximum loading of 0.689 followed by ‘Growing Absenteeism’ (0.620), ‘Job Change’ (0.573) and ‘Strikes and Lockout Scenario’ (0.424) with their respective loadings. This factor totally demonstrates negative impact upon employees working in the organization.

Factor-4 Declining Cooperation, Collaboration and Efficiency among Employees

This factor explains 9.051 per cent of total variance and has an eigenvalue of 1.901. This factor consisted of three items. ‘Decreasing Collaboration’ has the highest loading of 0.807 followed by ‘Declining Employee’s Cooperation’ (0.537) and ‘Adverse Impact on Efficiency and Effectiveness of Employees’ (0.532) with their respective loadings. This factor covers all the statements regarding decline in efficiency, effectiveness, cooperation and collaboration among employees working in these organizations, hence named as ‘Declining Cooperation, Collaboration & Efficiency among Employees’.

Factor-5 Disappointments and Widened Gap of Misunderstanding

This factor named as ‘Disappointments and Widened Gap of Misunderstanding’ explains 6.933 per cent of total variance explained and has an eigenvalue of 1.456. Two variables have been loaded on this factor. ‘Employee Experiences Disappointments’ has the highest factor loading of 0.802 followed by ‘Widened Misunderstandings due to Conflict’ with the loading of 0.568. All the statements under this very factor state the disappointments that the respondents experienced and further growing status of misunderstandings among themselves.

Factor-6 Declining Trust, Morale and Motivating Power

This factor explains 5.764 per cent of variance and has eigenvalue of 1.210. This factor consists of two statements. ‘Low Morale & Motivating Power’ has factor loading of 0.792 followed by ‘Trust Deficit’ with the loading of 0.559. This factor revealed the decline in trust, experiencing low morale and motivating power among the employees working in the selected banks under study.

Factor-7 Interruptions in Organizational Operations

Three items have been loaded on this factor. ‘Delays in Working Activities’ has the factor loading of 0.716 followed by ‘Disorders in Working Environment’ (0.680) and ‘Hindering Bank Operations’ (0.567) with their respective factor loadings. This factor has been named as ‘Interruptions in Organizational Operations’ because of delays and interruption in working environment of banks.

Factor-8 Job Dissatisfaction and Resultant Frustration

This is last identifiable factor having an eigenvalue of 1.008 and explains 4.801 per cent of total variance explained. This factor covers two variables. The variable ‘Dissatisfaction among Employees & Resultant Frustration’ has a loading of 0.727 followed by ‘Dissatisfaction among Customers’ with a loading of 0.724. ‘Job Dissatisfaction and Resultant Frustration’ has been found due to conflict generation in the selected banks under study.

Major eight factors concerning dysfunctional impacts of organizational conflict have been explored, that is, (a) declining performance, profits and management creditability, (b) health and psychological problems, (c) detrimental impact upon working environment and organization, (d) declining cooperation, collaboration and efficiency among employees, (e) disappointments and widened gap of misunderstanding, (f) declining trust, morale and motivating power, (g) interruptions in organizational operations and (h) job dissatisfaction and resultant frustration.

Cluster Analysis of Manifestations of Experienced Conflict

Factor analysis has been used to analyze the manifestations or effects of organizational conflict in selected public and private sector banks under study. Again with the exploratory aims of the study, cluster analysis has been used to classify various cases or respondents in clusters and the computed factor scores have become the platform for conducting cluster analysis. Table 4 displays agglomeration schedule along with coefficients in the fourth column. The agglomeration schedule from stage 1 to stage 540 indicated the sequences in which cases got combined with others. In the last stage 540, all clusters were combined into one cluster. The last stage 540 represents 2-cluster solution, stage 539 represents 3-cluster solution, stage 538 represents 4-cluster solution and so on. Table 5 shows re-formed agglomeration schedule along with the differences or change between rows of coefficients (fourth column of Table 4) in order to determine the feasible number of clusters. The final column headed change clearly indicates or determines optimum number of clusters. There is a highest difference of 992.477* (4,876.886–3,884.409) in the coefficients between stage 540 and stage 539, hence providing two-cluster solution. The next differences are 540.813, 428.417, 268.199, 258.911, 215.959, 165.492 and 140.205 for upward stages, respectively. Further, the next differences are smaller between subsequent rows of coefficients. A large difference between any two rows of coefficients shows a solution pertaining to number of clusters. The highest difference between values of coefficients is 992.477* which provide two-cluster solution.

Table 4.

Agglomeration Schedule of Various Consequences of Conflict

Stage	Cluster Combined		Coefficients	Stage Cluster First Appears		Next Stage
Stage	Cluster 1	Cluster 2	Coefficients	Cluster 1	Cluster 2	Next Stage
507	31	94	701.334	483	491	519
508	35	145	720.732	473	477	526
509	50	313	740.555	506	458	521
510	37	39	762.067	500	459	526
511	13	16	784.541	451	499	522
512	4	308	807.564	505	487	529
513	8	19	832.782	495	471	532
514	280	292	859.897	494	448	527
515	24	273	887.820	409	480	524
516	6	245	917.274	503	367	525
517	33	96	947.048	504	470	533
518	1	3	980.548	479	472	525
519	5	31	1,016.695	498	507	524
520	27	30	1,053.000	486	488	523
521	50	282	1,091.028	509	496	533
522	13	293	1,129.827	511	501	535
523	27	310	1,181.735	520	489	534
524	5	24	1,235.608	519	515	528
525	1	6	1,293.614	518	516	538
526	35	37	1,353.147	508	510	530
527	280	290	1,414.279	514	490	529
528	5	274	1,475.808	524	485	531
529	4	280	1,548.669	512	527	536
530	35	97	1,638.525	526	502	535
531	5	314	1,747.437	528	497	534
532	8	28	1,866.413	513	460	536
533	33	50	2,006.618	517	521	537
534	5	27	2,172.110	531	523	537
535	13	35	2,388.069	522	530	539
536	4	8	2,646.980	529	532	538
537	5	33	2,915.179	534	533	540
538	1	4	3,343.596	525	536	539
539	1	13	3,884.409	538	535	540
540	1	5	4,876.886	539	537	0

Source: Primary data analysis.

Table 5.

Re-formed Agglomeration Schedule Table

Stage	No. of Clusters	Agglomeration Last Step	Coefficients This Step	Change/Difference
540	2	4,876.886	3,884.409	992.477*
539	3	3,884.409	3,343.596	540.813
538	4	3,343.596	2,915.179	428.417
537	5	2,915.179	2,646.98	268.199
536	6	2,646.98	2,388.069	258.911
535	7	2,388.069	2,172.11	215.959
534	8	2,172.11	2,006.618	165.492
533	9	2,006.618	1,866.413	140.205

Note: *A clear demarcation point seems to be here to determine feasible number of clusters.

K-means cluster analysis further provides more stable clusters and it is best suited in continuation with hierarchical cluster analysis. After determining number of final clusters, further k-means cluster analysis has been applied in order to find out the cluster membership according to various cases or observations, which has been depicted in Table 6. This table also revealed the final cluster centres or centroid means based on continuous variables, that is, manifestations or consequences of conflict. Results indicated two-cluster solution and the descriptions of these clusters have been mentioned below.

Cluster-1: 59.33 per cent of total respondents belongs to the first cluster. Respondents belonging to this cluster experienced ‘Disappointments and Widened Gap of Misunderstanding’ (m = 4.34), ‘Declining Performance, Profits & Management Creditability’ (m = 4.26), ‘Job Dissatisfaction and Resultant Frustration’ (m = 4.02), ‘Interruptions in Organizational Operations’ (m = 3.76), ‘Detrimental Impact upon Working Environment & Organization’ (m = 3.67), ‘Declining Trust, Morale & Motivating Power’ (m = 3.51), ‘Declining Cooperation, Collaboration & Efficiency among Employees’ (m = 3.40) and ‘Health & Psychological Problems’ (m = 3.20) as major impacts of conflict experienced at workplace.

Cluster-2: 40.67 per cent of total respondents belongs to the second cluster. Respondents belonging to this cluster experienced ‘Interruptions in Organizational Operations’ (m = 3.28), ‘Declining Performance, Profits & Management Creditability’ (m = 3.24), ‘Declining Trust, Morale & Motivating Power’ (m = 3.07) and ‘Health & Psychological Problems’ and ‘Declining Cooperation, Collaboration & Efficiency among Employees’ (m = 2.82) as major impacts of experienced conflict at workplace with scant attention to others impacts.

Further, one way ANOVA has been applied in order to probe the consistency of clusters of respondents so formed towards various explored manifestations of conflict experienced in public and private sector banks selected under research.

Hypotheses framed under study: H₀1 (Null Hypothesis): Insignificant difference has been found across various impacts of conflict experienced by the various clusters of respondents so formed.

H_a1 (Alternate Hypothesis): Significant difference has been found across various impacts of conflict experienced by the various clusters of respondents so formed.

As shown in Table 7, one way ANOVA depicted the significance of all eight factors across these clusters so emerged. The significance (sig.) column indicates that the variables EF1, EF2, EF3, EF4, EF5, EF6, EF7 and EF8 are statistically significant at 0.01 level (p < 0.01). H₀1 has been rejected and alternate hypothesis has been accepted which further concluded that significant difference has been found across various impacts of conflict experienced by the various clusters of respondents so formed. Hence, clusters of respondents under multivariate technique of cluster analysis have significant differences of opinions towards major manifestations of conflict experienced in these organizations selected under the sample of the research.

Table 6.

Cluster Membership Table and Final Cluster Centres; Centroid Means Based on Continuous Variables (Consequences of Conflict)

Variables	Variable Names	Clusters
Variables	Variable Names	1	2
EF1	Declining performance, profits & management creditability	4.26	3.24
EF2	Health & psychological problems	3.20	2.82
EF3	Detrimental impact upon working environment & organization	3.67	2.50
EF4	Declining cooperation, collaboration & efficiency among employees	3.40	2.82
EF5	Disappointments and widened gap of misunderstanding	4.34	2.45
EF6	Declining trust, morale & motivating power	3.51	3.07
EF7	Job dissatisfaction and resultant frustration	4.02	2.73
EF8	Interruptions in organizational operations	3.76	3.28
	Number of cases in each cluster (f)	321	220
	Per cent of cases in each cluster (%)	59.33	40.67

Source: Primary data results: Two-cluster solution.

Table 7.

Analysis of Variance (ANOVA) for Indicating the Significance of Clusters

Variables	Cluster		Error		F-value	Sig.
Variables	Mean Square	Df	Mean Square	Df	F-value	Sig.
EF1	135.463	1	0.769	539	176.109	0.000**
EF2	19.666	1	1.697	539	11.589	0.001**
EF3	177.348	1	0.599	539	296.217	0.000**
EF4	42.619	1	0.531	539	80.27	0.000**
EF5	467.194	1	0.619	539	754.767	0.000**
EF6	25.42	1	0.777	539	32.699	0.000**
EF7	216.683	1	0.666	539	325.329	0.000**
EF8	30.043	1	1.322	539	22.723	0.000**

Note: **Significant at 0.01 level.

Concluding Observations

This research article mainly deals with the analysis of two major objectives of the study, that is, (a) to explore the factors concerning destructive or dysfunctional impacts of conflict in public and private sector banking organizations selected under study and (b) to probe the consistency of manifestations of conflict experienced across various clusters of respondents so formed with the help of cluster analysis. Eight major factors have been extracted with the help of EFA, that is, (a) declining performance, profits and management creditability, (b) health and psychological problems, (c) detrimental impact upon working environment and organization, (d) declining cooperation, collaboration and efficiency among employees, (e) disappointments and widened gap of misunderstanding, (f) declining trust, morale and motivating power, (g) interruptions in organizational operations and (h) job dissatisfaction and resultant frustration. After the extraction of factors concerning dysfunctional impacts, hierarchical cluster analysis has been employed to classify various observations or cases in clusters resulted into two-cluster solution and k-means cluster analysis has been further applied to find out the cluster membership according to various observations towards manifestations or impacts of organizational conflict in the selected banks under study. At last, findings revealed statistically significant clusters across all the extracted factors concerning dysfunctional impacts of conflict in these selected banks under study.

Limitations and Managerial Implications

The sample of the study incorporates only three cities of Punjab and it does not incorporate the foreign banking sector. Limitations have also been found on the basis of shortage of time and resources and subjective biasness of the respondents. Conflict resolution in any organization is a shared priority which needs greater involvement of all human resource effected or likely to be affected by its dysfunctional impacts. So, it has been recommended that management should follow regular interactive activities to ensure a good degree of functionality of organization and sincere efforts should be made towards understanding of various dysfunctional impacts of organizational conflict so that necessary and remedial actions can be taken at the earliest time. It will automatically help in cultivating a good atmosphere of mutual acceptance and better understanding. Management should have open communication policy so that the human resource can come closer, collaborate and make compromises where possible with the authorities concerned. If the destructive impacts of workplace conflict are managed properly, then it helps the management achieve its strategic objectives with the better work performance of banking staff; positive working environment will automatically lead towards high organizational productivity.

References

Adebile

O. A.

Ojo

T. O.

(2012). Management of organizational conflict in Nigeria polytechnics, an empirical study of the federal polytechnic, Ede Osun State. International Journal of Asian Social Science, 2(3), 229–243.

Bahadur

(1993). Conflict management in Nepalese organizations. Tribhuvan University Journal, 26(2), June, 19–28.

Bercovitch

(1982). Conflict and conflict management in organizations: A framework for analysis. Working Paper. Retrieved 15 June 2013, from aspheramedia.com/v2/wp-content/uploads/2011/02/5000214.pdf

Friedman

R. A.

Tidd

S. A.

Currall

S. C.

Tsai

J. C.

(2000). What goes around comes around: The impact of personal conflict style on work conflict and stress. The Journal of Conflict Management, 11(1), 32–55.

Gupta

S. L.

Gupta

(2011). SPSS 17.0 for researchers. New Delhi: International Book Publishing House Pvt. Ltd.

Hair

J. F.

Anderson

R. E.

Tathan

R. L.

Black

W. C.

(1998). Multivariate data analysis (5th ed.). USA: Prentice Hall International, Inc.

Holton

S. A.

(1998). Academic mortar to mend the cracks: The Holton model for conflict management. Mending the Cracks in the Ivory Tower: Strategies for Conflict Management in Higher Education. Bolton, M.A: Anker Publishing. Retrieved from http://college.emory.edu/home/assets/documents/faculty/conflict-management.pdf

Hotepo

O. M.

Asokere

A. S. S.

Azeez

I. A. A.

Ajemunigbohun

S. S. A.

(2010). Empirical study of the effect of conflict on organizational performance in Nigeria. Business and Economics Journal, 2(September), 1–9.

Kaiser

(1960). The application of electronic computers to factor analysis. Educational and Psychological Measurement, 20(1), 141–51.

10.

Kaiser

(1970). A second generation little jiffy. Psychometrika, 35(4), 401–415.

11.

Kaur

(2012). Conflict in organizations causes and management. Biz and Bytes, 2(1), July. Retrieved December 2014, from http://cbsmohali.org/e-%E2%80%93-journal-biz-n-bytes/volume-ii-year-2012.html

12.

Kothari

C. R.

(2004). Research methodology: Methods and techniques (2nd ed.). New Delhi: New Age International Publishers.

13.

Lkeda

A. A.

Oliveira

T. M.

Campomar

M. C.

(2005). Organizational conflict perceived by marketing executives. Electronic Journal of Business Ethics and Organization Studies, 10(1), 22–28.

14.

Nunnally

Bernstein

(1994). Psychometric theory (3rd ed., pp. 251–261). New Delhi: McGraw Hill Humanities/Social Sciences/Languages.

15.

Riaz

M. K.

Junaid

F. A.

(2011). Types, sources, costs and consequences of workplace conflict. Asian Journal of Management Research, 2(1), 600–611.

16.

Robbins

S. P.

Judge

T. A.

Sanghi

(2012). Organizational behaviour. New Delhi: Pearson Education.

17.

Tonder

C. V.

Havenga

Visagie

(2008). The causes of conflict in public and private sector organizations in South Africa. Managing Global Transitions, 6(4), 373–401.

18.

Wall

J. A.

Callister

R. R.

(1995). Conflict and its management. Journal of Management, 21(3), 515–558.