Organizational role stress: confirmatory factor analysis approach.

Yasmeen, Haider ; Supriya, M.V.

Introduction

Work related stress is of growing concern today, than it was two decades ago. This has become a major problem not only for individuals working within an organization but also for the organization itself. Cooper and Marshall (1978) suggested classification of stress in terms of six different stressors. Role stress is considered very important among these as it has a negative impact on organizational outcomes. The stress arising due to person's role is termed as role stress (Pareek, 1993). Role is defined as a set of functions, which an individual performs in response to the expectations of others as well as his own expectations (Khan et al., 1964). Therefore it becomes extremely important both for the individual and the organization to work in the area of role stress management.

Role-based research carried out by researchers has paid more attention only to a few dimensions of role stress like role conflict, role ambiguity and role overload (Richard, 2001; Bhatia and Kumar, 2003; Fried et al., 2003). However, the other important dimensions of role stress were ignored. Some researchers in India have made an attempt to study certain other dimensions of role stress like inter-role distance, role isolation, role stagnation, role-overload and role erosion using Pareek's Organizational Role Stress (ORS) Scale (1982). This scale considers all the dimensions of role stress. A study conducted by Aziz, (2004), investigated the intensity of organizational role stress among women informational technology professionals in the Indian private sector. Resource inadequacy has emerged as the most potent role stressor, followed by role overload and personal inadequacy. The research finds differences in the level of stress between married and unmarried employees on several role stressors. However, level of education did not emerge as a significant differentiator of stressors.

Based on the fact that stressors vary from one job category to another depending on the personality characteristics of focal person, a study was conducted by Pestonjee, (1987) exploring role stresses on three categories of management personnel, namely top management, middle management

and IAS officers using correlational analysis. The average age of the three sample categories was 48.22, 41.60 and 42 .90 years, respectively. The findings of this study indicated that inter-role distance and role erosion were found to be dominant contributors of role stress whereas role ambiguity and personal inadequacy were the least dominant contributors of role stress in all the three job categories.

Satyanarayana (1995) investigated stressors among executives and supervisors. The analysis of the data revealed that role erosion, personal inadequacy, resource inadequacy, and role stagnation were identified as dominant contributors of role stress in executives and supervisors. Kumar (1989) studied the relationship between role-stress, role-satisfaction and role efficacy using a sample of lower and middle level executives from different functional areas of an oil company. The major findings indicated that marketing executives experienced maximum role stress in comparison to finance, production and personnel executives. Personnel executives obtained lowest scores on total role stress.

Pandey (1997) examined the relationship between role stress and role efficacy using a sample of personnel of Indian Railways. The findings of the study indicated that coefficients of correlation between the first dimensions of role efficacy namely, centrality and all the 10 dimensions of role stress were found to be negative. The second dimension of role efficacy i.e. integration was correlated negatively and significantly with all the dimension of role stress except role erosion. Creativity, the third dimension of role efficacy was found to have non-significant but positive correlations with all dimensions of role stress except role overload and self-role distance. The relationship of inter-role linkage with role stress was found to be negative in seven cases but was not statistically significant. A similar finding was reported by Sen (1982).

Pestonjee and Singh (1988) investigated the type--A pattern of behavioral disposition on the relationship between role stresses and state- trait anxiety. The findings revealed that stresses, type-A behavior, state and trait anger were correlated positively and most of the coefficients of correlation were statistically significant. Srinivasan (1988) investigated, in organizations with differing corporate ownership, the effect of executive's sense of competence on the relationship between organizational role stress and job satisfaction. Significant differences in main and interaction effects, and experienced role stresses, were noticed between the three sectors.

In a study conducted by Sen (1981) background variables in relation to role stress was studied. The variables chosen were age, sex, income and marital status. Some of the conclusions drawn showed that role stagnation decreased as people advanced in age. Age is negatively related with role stress. Women experience more role stress as compared to men. Role stress is inversely related to income; the higher the income, the less is the level of reported role stress. Unmarried persons experience more stress than married persons. This may be due to their comparative lack of security need, resulting in higher self-esteem, autonomy and self- actualization needs. Similar studies exploring the relationship between demographic variables and dimensions of role stress are carried out by many other researchers (Sharma and Mahajan, 2003; Aditya and Sen, 1993).

The review of literature points towards the importance of research in the area of role stress and the use of ORS in examining different dimensions of role stress. The earlier research studies where ORS scale was used to examine role stress have proved the reliability and validity of the scale through retest reliability, cronbach's alpha, and exploratory factor analysis approach (Srivastav, 1993; Srinivasan, 1988; Sen, 1981). The present research work attempts to understand the use of ORS scale in assessing the role stress of respondents by using confirmatory factor analysis approach. The ORS scale is administered to the respondents, and they indicated whether an item is a source of stress to them. The 5-point scale ranges from, 1 (never feel this way) to 5, (always feel this way). Thus the total score on each role stress ranges from 5 to 25. The higher the scores, greater is the stress due to a particular source.

Method

Hypothesis

The confirmatory factor analysis will prove the goodness of fit for the role stress factor model comprising of ten factors.

Respondents

The respondents comprised of officers of the Indian Administrative Services, belonging to the Tamil Nadu Cadre. In all there are 296 officers, of these 115 officers responded to the questionnaire. The response rate is 39%. It is observed that the majority of officers have a post graduate qualification and most of them are males. The numbers of older IAS officers are more than younger officers.

Dimensions of Organizational Role Stress Scale

According to Pareek, (1983), the ORS scale can be used to investigate the nature and dynamics of role stress in various organizations and to develop interventions for the use of the individual, groups and for the organization as a whole. In the past, some Indian researchers (Sen, 1981; Srinivasan, 1988; Mittal, 1992; Raju and Madhu, 1994; Pandey, 1997) have adopted this scale as it exhibits an acceptable level of reliability. It is considered to be an apt instrument for the current study also, as almost every dimension of role stress included in ORS has relevance to the IAS. The scale consists of 50 items divided into 10 subscales. The subscales are inter-role distance, role stagnation, role expectation conflict, role erosion, role overload, role-isolation, personal inadequacy self-role distance, role ambiguity and resource inadequacy.

The concept of inter-role distance refers to the stress due to conflicts of not being able to share time between the multiple roles that the role occupant undertake while performing their usual role at work. In addition to this, imbalance between work and family role may also induce stress of this type. The other stressor arises when the opportunities for learning do not include growth. This leads to role stagnation. Role expectation conflict is said to exist when the role occupant experiences conflicting expectations or demands by different role senders. Role erosion is a feeling that some important functions have been given to some other roles or it could be a feeling that there is not much challenge in the functions given to the role. Role overload is also an important determinant of stress in organizations. Role occupants experience overload when they are compelled to perform beyond their capacity.

When the linkages between the departments forming the sub-system are strong, role isolation will be low and in the absence of strong linkages, role isolation will be high. Personal inadequacy is experienced when a role occupant feels that he is not prepared to undertake the role effectively. He feels that he does not have enough knowledge, skills, or training to perform the assigned role. The stress arising due to a mis-match between role occupants own values and what is expected of the role may lead to self-role conflict. One may have to make a compromise between these two. They may be doing many things out of compulsion that they would not like to do. Another source of role stress is a situation when an individual has inadequate information about his work role and there is lack of clarity about the objectives to be achieved, and about the scope and responsibilities of the job. This is termed as role ambiguity. Stress is also experienced due to resource inadequacy that is when the resources required by the role occupant for performing the role effectively is not sufficiently available.

Reliability

Cronbach's alpha (Cronbach, 1951) is used to test the statistical reliability of the scale. Reliability of a measure is the ability to yield consistent results (Nunnally, 1988). When compared to other reliability testing methods, the internal consistency method is simple and easy for computation. An alpha coefficient of 0.60 and above is considered to be a good reliability estimate (Nunnally, 1988). The organization role stress scale yields sample alpha coefficients of 0.93.

Confirmatory Factor Analysis

Kerlinger, (1986) described factor analysis as the most powerful tool for the study of complex behavioral areas. Factor analysis may be exploratory (EFA) or confirmatory (CFA). The exploratory factor analysis method is appropriate for those areas where no prior analyses have been done so far (Ahire et al., 1996). Gorsuch (1983) stated that confirmatory methods, after specifying the priori factors, seek to optimally match the observed and latent factor structures for a given data set in order to determine the "goodness of fit" of the predetermined factor model. As this is the case in this study, the CFA was considered more appropriate than EFA.

Results and Discussion

Unidimensionality, Construct and Convergent validity of the scale adopted for the purpose of study are analyzed, using Confirmatory Factor Analysis. Construct and Convergent validity are carried out to confirm if the various measures used are valid. Content analysis is also done. The organizational role stress scale is tested using Confirmatory Factor Analysis. Confirmatory factor analysis replaces the more traditional technique of exploratory factor analysis (EFA) as it allows a more precise test of an instrument's factor structure (Long, 1988).

Unidimensionality checks if individual items in the model represent the same construct. It helps in checking the construct validity and reliability of the measure (Anderson and Gerbin, 1991). It provides evidence to show that all the items included in each dimension of a scale are related and proves to reflect that particular dimension. Convergent validity is evidenced when the measures that are theoretically supposed to be highly interrelated, are highly interrelated in practice also. It refers to the degree to which different approaches to construct measurement is similar. Each item in a scale can also be treated as a different approach to measure the construct (Ahire et al., 1996).

Construct Validity

Construct Validity is established through unidimensionality. In order to check for unidimensionality, a measurement model is first proposed. This suggestive model shows the variables that are likely to correlate with given factors. The CFA is then run for all the constructs. The model is based on a strong theoretical and empirical foundation (Stevens, 1996). The unidimensionality is determined using the Comparative fit index, which ranges from 0-1 with a larger value indicating better model fit. Acceptable model fit is indicated by a CFI value of 0.90 or greater (Hu & Bentler, 1999). For the role stress scale, 8 out of 10 dimensions are found to have a CFI value ranging from 0.91 to 1.00 thus proving the hypothesis true to a larger extent. These dimensions are inter-role distance, role expectation conflict, role overload, role isolation and resource inadequacy, role erosion, personal inadequacy and role ambiguity. The CFI for two other dimensions of role stress, namely role stagnation and self-role distance is found to have a CFI value of 0.72.

Results indicate that there is evidence of unidimensionality for role stress scale. Unidimensionality is a precondition for ensuring convergent validity. Results also satisfy this purpose. Similar evidence of the validity of the scale using exploratory factor analysis, was observed in the study conducted by Srinivasan et al., (1988). Organizational role stress was found as a configuration of related reliable multiple factors comprising role expectation conflict

Several fit indices were used to assess the constructs like Comparative fit index (CFI) and Bentler Bonett Normed fit index (BFI). Comparative fit index is recommended when using smaller sample size because it corrects the under estimation of model-fit using the Bentler Bonnett Normed fit index (Bentler, 1990). Costruct validity is established when the CFI is 0.90 and above. Convergent (Variance 27.2%), inter role distance (Variance 8.6%), role ambiguity (Variance 3.9), role stagnation (Variance 3.5), resource inadequacy (Variance 3.1%), role isolation (Variance 3.0) and role overload (Variance 2.6). Role underload (Variance 2.6) was also suggested in the factor. Srivastava (1993) got similar results from factor analysis of data on ORS of about 400 executives in a public sector company.

Convergent Validity

The convergent validity is established using a coefficient called Bentler Bonett Coefficient. A scale with values of 0.90 or above is an evidence of strong convergent validity. The values of different dimensions of each scale are summarized in table II. Results indicate that the model provides evidence of moderate fit.

Content Validity

The measure is tested for content validity. Expert opinion obtained show evidence of content validity. Therefore, it can be suggested that the ORS scale is a good measure to understand role stress.

References

Aditya, S.M and Sen, A.K. (1993), "Executives under Stress: A Comparison between Men and Women". Journal of the Indian Academy of Applied Psychology. Vol. 19, pp. 1-6.

Ahire, S.L., Golhar, D.Y., and Waller, M.A. (1996), "Development and Validation of TQM Implementation Constructs". Decision Sciences, Vol. 27, pp. 23-56.

Anderson, J.C., and Gerbin, D.W. (1991), "Predicting the Performance of measures in a Confirmatory Factor Analysis with a pretest assessment of their substantive validities". Journal of Applied Psychology, Vol. 76, pp. 732-740.

Azeez, M. (2004), "Role stress among women in the Indian information technology sector". Women in Management Review, Vol. 19, pp. 356-363.

Bentler, P.M. (1990), "Comparative fit indices in structural models". Psychological Bulletin, Vol. 107, pp. 238-246.

Bhatia, P., and Kumar, A. (2003), "Occupational Stress and Burnout". Indian Psychological Review, Vol. 60, pp. 145-153.

Cooke, R.A., and Rousseau, D.M. (1984), "Stress and strain from family roles and work roles expectations". Journal of applied psychology, Vol. 69, pp. 252-260.

Cooper, C. L., and Marshall, J. (1978), Understanding Executive Stress. London: Macmillan.

Cronbach, L. J. (1951), Coeffecient Alpha and the Internal Structure of Tests, Psychometrika, Vol.16, pp. 297-334.

Fried, Y., Slowik, L.H., Shperling, Z., Franz, C., and David, H.A. (2003), "The moderating effect of job security on the relation between role clarity and job performance: A longitudinal field study". Human Relations, Vol. 56, pp. 787-805.

Gerbin, D.W. and Anderson, J.C. (1988), "An Updated Paradigm for Scale Development Incorporating Unidimensionality and its Assessment". Journal of Marketing Research, Vol. 25, pp. 186-92.

Gorsuch, R.L. (1983), Factor Analysis (2nd ed.). Hillsdale, NJ: Lawrence Earl Baum Associates.

Kerlinger, F.N. (1986), Foundations of behavioral research (3rd ed.). New York: Holt, Rhinehart and Winston.

Khan, R.L., Wolfe, D.M., Quinn, R.P., Snoek, J.D., & Rosenthal, R.A. (1964), Occupational Stress: Studies in Role Conflict and Role Ambiguity. Newyork: Wiley.

Mittal, U. (1992), Coping Styles as Related to Role Stress, Doctoral Thesis, University of Rajasthan, Jaipur.

Nunnally, J.C. (1988), Psychometric Theory. Eaglewood Cliffs, NJ: McGraw Hill Pub.

Pandey, A. (1995), "Role efficacy and role stress relationship: Some experience with workers", Vol. 31, pp. 193-208.

Pareek, U. (2002), Training Instruments in HRD and OD. New Delhi: McGraw Hill Pub.

Pareek, U. (1993), Making Organizational Roles Effective, New Delhi: Tata McGraw Hill Pub.

Pestonjee, D.M. (1999), Stress and Coping: The Indian Experience, Newyork: Sage Publishers.

Pestonjee, D.M., Singh, V.B. (1982), "Job Satisfaction as a Function of Role Stress, Locus of control, participation and organizational climate in an electric supply company." Monograph, Ahmedabad: Indian Institute of Management.

Raju, M.V.R., and Madhu, K. (1994), "Organizational Level and Role Stress". Journal of Indian Psychology, Vol. 12, pp. 1-2.

Richard, T. (2001), "Role stress, The Type A behavior Pattern, and External Auditor Job Satisfaction and Performance". Behavioral Research in Accounting, pp. 1-27.

Satyanarayana,K.(1995), "Stressors among Executives and Supervisors: A Comparative Study in Public Sector Undertaking". Osmania Journal of Psychology, Vol. 19, pp. 1-9.

Sen, P.C. (1982), A Study of Personal and Organizational Correlates of Role Stress and Coping Strategies in Some Public Sector Banks. Doctoral Thesis, Gujarat University.

Sharma, K., and Mahajan, S. (2003), "Gender differences in stress as affected by some demographic variables in bank employees". Indian journal of applied psychology, Vol. 40, pp. 17-24.

Srinivasan, P.T. (1988), Sense of Competence as a Moderator of the Relationship between Organizational Role Stress and Satisfaction. Doctoral Thesis, Indian Institute of Technology, Madras.

Haider Yasmeen, Department of Management Studies, Crescent Engineering College, Vandalur, Chennai, India, E-mail: yarahai@yahoo.co.in

M.V. Supriya, Department of Management Studies, Anna University, Chennai, India

Table 1: Demographic profile of respondents

Variables Number of respondents Percentage
 (%)

Education
 UG 18 16
 PG 90 78
 Ph.D 7 6
Age
 25-35 30 26
 36-46 32 28
 47 and above 53 46
Gender
 Male 101 88
 Female 14 12

It is observed that the majority of officers have a post
graduate qualification and most of them are males. The
numbers of older IAS officers are more than younger
officers.

Table 2: Confirmatory Factor Analysis to check the Dimensions
of Variables

Variables BFI CFI

Inter Role Distance 0.946 0.962
Role Stagnation 0.764 0.784
Role Expectation Conflict 0.914 0.948
Role Erosion 0.899 0.933
Role Overload 0.955 0.98
Role Isolation 0.924 0.959
Personal Inadequacy 0.886 0.929
Self Role Distance 0.728 0.751
Role Ambiguity 0.889 0.917
Resource Inadequacy 0.964 1.000

Variables Chi square Probability
 value

Inter Role Distance 15.16 0.009
Role Stagnation 32.36 0.001
Role Expectation Conflict 11.22 0.047
Role Erosion 12.84 0.024
Role Overload 8.65 0.124
Role Isolation 9.89 0.078
Personal Inadequacy 11.48 0.043
Self Role Distance 30.21 0.001
Role Ambiguity 16.46 0.005
Resource Inadequacy 3.03 0.695