Abstract
Police organizations are tasked with a wide variety of duties, and officers often encounter stressful situations. Past studies have indicated that job stressors are negatively related to job involvement, which, in turn, is positively related to several beneficial outcomes. The present study empirically tested the job demands model with data obtained from a police agency in India. Survey data was collected from 827 police officers in the Indian state of Haryana. Analysis was carried out to determine the impact of job stressors (role overload, role underload, repetitiveness, role ambiguity, fear of victimization, and role conflict) on job involvement. Findings indicate that all stressors except fear of victimization had a negative effect on job involvement. The implications of the findings, for organizations in general and the police in particular, include the need to reduce workplace stressors and improve the work environment.
The primary job of the police is the enforcement of laws and the preservation of the peace; this requires that the police deal with different responsibilities concurrently. Goldstein (1977) noted that the role of the police covers a broad framework, which includes controlling serious crimes, aiding individuals in danger of physical harm (such as victims, the intoxicated, the addicted, or the mentally ill), facilitating the movement of traffic, resolving conflicts, identifying issues that have the potential to become serious societal problems, and creating a feeling of security in the community. Police organizations are manpower intensive. The most significant and costly resource for police agencies is personnel; therefore, optimal performance of officers is a highly desirable objective (Roberg et al., 2012). An important officer concept for the success of police organization is job involvement. Job involvement is the cognitive (psychological) identification with the job and is the opposite of job alienation (Kanungo, 1982a, 1982b).
Research to date indicates that job involvement is an important concept for officers and police agencies. Among officers in an unnamed country, job involvement was reported to have a significant relationship with how officers spent their leisure time (Orpen, 1982). Among Canadian police officers, job involvement was positively correlated with both job satisfaction and organizational commitment, and was theorized to be positively linked to work performance (DeCarufel and Schaan, 1990). In a study of U.S. Midwestern police officers, job involvement was significantly correlated with organizational commitment and turnover intent (McElroy et al., 1999). In another study of U.S. Midwestern police, tenure was negatively associated with officers’ job involvement (Hazer and Alvares, 1981). Among New Zealand police officers, there was no difference in level of job involvement between male and female respondents (Love and Singer, 1988). College-educated, older, and supervisory New York City officers were reported to have higher job involvement than younger, nonsupervisory officers without a college degree (Lefkowitz, 1974).
In light of the importance of job involvement, there is a need to study the variables that affect it; however, a limited body of research has explored the relationship between a variety of workplace factors and police officer job involvement. Lord (1996) reported that, among North Carolina officers, there was a significant negative relationship between job involvement and the stressors of lack of social support, role conflict, and role ambiguity, whereas role change, communication, participation in decision-making, recognition from others, and overall quality of work all had nonsignificant associations. In a study of U.S. Midwestern police officers, supervisory initiation of structure in the workplace was positively correlated with the level of self-reported job involvement (Brief et al., 1976). Job autonomy was found to be a negative predictor of job involvement, and job variety, perceptions of quality training, and quality supervision were found to be positive predictors of job involvement among Indian police officers (Lambert et al., 2015). Role conflict was reported to have a positive relationship with British officer job involvement, role ambiguity had a negative association, and social support and perceived age discrimination had no significant effect (Redman and Snape, 2006).
Three conclusions can be gleaned from the past job involvement research focusing on police officers. First and foremost, additional research is needed. There has been only limited research to date on job involvement among police officers. Second, the handful of studies looking at the effects of workplace variables indicate that some variables appear to help shape officer job involvement, including stressors (e.g., role ambiguity and role conflict). There are, however, other workplace variables, including stressors, which have not been studied. Third, the vast majority of studies have been conducted using officers from Western, industrialized nations, particularly the USA (Buker and Dolu, 2010; Lambert et al., 2015; Martinussen et al., 2007). Most countries in the world are developing and emerging nations. Bennett (1997: 296) noted that “most police forces in the developing world operate with very limited financial and personnel resources, so they must understand how to use those resources most effectively and efficiently”. As such, research on police officer job involvement in developing and emerging nations should aid police administrators and scholars.
The job demands model suggests that job demands, particularly workplace stressors, lead to job strain and other negative outcomes. The current study used the job demands model to examine the effects of role stressors, such as role overload, role underload, repetitiveness, role ambiguity, fear of victimization, and role conflict on the job involvement of Indian police officers assigned to the police agency in the Indian state of Haryana. The Republic of India [henceforth, India] is a large and important developing nation. Although its landmass is one-third the size of the USA, its population is almost four times as large (i.e., 1.2 billion). India has a vibrant and rapidly growing market-based economy, estimated to be the third largest economy in the world (Ranasinghe, 2014). It is the most populous democracy on earth (Unnithan, 2009). Even though India is a highly populated nation with a growing and dynamic economy, little has been published in Western journals concerning how workplace stressors are linked to job involvement among police officers. Not only is there been little research on how workplace stressors effect job involvement among police officers in general, the current study is the first study of workplace variables influence involvement in the job among Indian police officers.
Literature review
Job involvement
Lodahl and Kejner (1965) proposed the concept of job involvement. They argued that job involvement was “the degree to which a person is identified psychologically with his[/her] work, or the importance of work in his[/her] total self-image” and how “work performance affects a person’s self-esteem” (p. 25). There was disagreement over whether the definition of job involvement should include both job motivation and psychological identification with the job (Brown, 1996). Ultimately, work motivation was removed from the definition. Lawler and Hall (1970: 310–311) refined the conceptualization of job involvement as the “psychological identification with one’s work” and “the degree to which the job situation is central to the person and his [or her] identity”. Kanungo (1979, 1982a, 1982b) further developed the concept of job involvement and solidified the definition of it being a cognitive identification with the job. Essentially, job involvement refers to the centrality of the job to a person’s life (Brown and Leigh, 1996). As noted by DeCarufel and Schann (1990: 86), “An individual with a high degree of job involvement would place the job at the center of his/her life’s interests. The well-known phrase ‘I live, eat, and breathe my job’ would describe someone whose job involvement is very high…Persons with low job involvement would place something other than their jobs (e.g., family, hobbies) at the center of their lives”. For the current study, the conceptualization of job involvement was the cognitive (i.e., psychological) identification with the job as proposed by Kanungo (1982a, 1982b).
Job involvement is a different concept from work ethic, job satisfaction, and organizational commitment (Brown, 1996; Lawler and Hall, 1970). Work ethic is a moral belief that work is a critical element of life, makes an individual a better and stronger person, and as such, a person needs to work as hard as possible (Ekmekçi, 2011; Kanungo, 1982b). Having a high work ethic does not translate into being psychologically connected with one’s job. It means that a person feels work in general is an important aspect of life, and he or she may or may not be connected to their current job (DeCarufel and Schann, 1990). Job satisfaction is an affective (emotional) response to the job (Locke, 1976). Job satisfaction is an emotional response resulting from job experiences and refers to the degree a person likes or dislikes his/her job (Spector, 1996). Job involvement is a cognitive identification with the job (Brooke et al., 1988; Kanungo, 1982a, 1982b). It is possible, therefore, for a person to identify with the job psychologically, regardless of whether pleasure is derived from the job (Kanungo, 1982b).
Organizational commitment is the bond between the employee and the agency (Mowday et al., 1982). It is a bond with the entire employing organization and not with the job itself or a particular part of the organization (Lambert et al., 1999). Organizational commitment is, therefore, the identification with the overall employing organization, and job involvement is the identification with the job (Blau, 1987; Kanungo, 1982a, 1982b). Although closely related, work ethic, job satisfaction, organizational commitment, and job involvement have been empirically demonstrated to be distinct concepts (Blau, 1987; Brooke et al., 1988; Kanungo, 1982a; Lawler and Hall, 1970; Morrow and McElroy, 1986).
Job involvement is an important concept for organizations (Diefendorff et al., 2002). Job involvement is important in motivating employees and performance (Ekmekçi, 2011; Hackman and Lawler, 1971; Pfeffer, 1994). Chen and Chiu (2009: 478) noted that, in general, workers with “high job involvement are more independent and self-confident – they not only conduct their work in accordance with the job duties required by the company but are also more likely to do their work in accordance with the employees’ perception of their own performance”. Job involvement is linked to higher job satisfaction and organizational commitment (Adams et al., 1996; Lambert, 2008). In a meta-analysis of workers across different occupations, job involvement was reported to result in higher work performance, job satisfaction, organizational commitment, evaluations by supervisors, and life satisfaction; further, it was linked to lower job stress, work–family conflict, absences from work, and turnover intent/turnover (Brown, 1996). Because of the reported positive benefits of job involvement, it is surprising that there has been so little research on this important concept among police.
Job demands model
The job demands model provides a foundation for why workplace stressors would be inversely related to job involvement. The job demands model, which is also referred to as the job strain model, was developed by Karasek (1979). Karasek (1979: 291) defined job demands as “the psychological stressors involved in accomplishing the workload, stressors related to unexpected tasks, and stressors of job-related personal conflict”. Workplace stressors place strain on workers, which increases the chances of negative outcomes (Hall et al., 2010; Lambert, Hogan et al., 2013; Wong and Laschinger, 2015). Under the job demands model, job demands that the worker cannot avoid or has little control over result in heightened psychological strain, which, in turn, raises the level stress from the job (Sargent and Terry, 2000; Wall et al., 1996). Policing is an occupation with high demands and often with little job latitude. The major job demands (stressors) faced by officers and explored in this current study are role overload, role underload, repetitiveness, role ambiguity, fear of victimization, and role conflict.
Role overload occurs when a worker must deal with too many job tasks in a limited amount of time or without sufficient resources (e.g., proper equipment) to accomplish the assigned duties (Lambert et al., 2005; Sargent and Terry, 2000). Police officers are at times required to handle multiple situations at the same time or deal with issues without sufficient resources (Lambert et al., 2016). These pressing situations can result in frustration and strain for an officer. Feeling overwhelmed can result in psychological strain. As such, role overload was hypothesized to have a negative association with job involvement (Hypothesis 1).
Role underload occurs when a person is not mandated to perform enough work tasks to keep them sufficiently busy or to provide stimulation (Melamed et al., 1995; Sales, 1970). Sales (1970: 593) contended that role underload was a “condition in which the person is faced with obligations which, taken as a set, require him to do considerably less than he is able to do in the time available”. Role underload can be a job demand. Being underwhelmed at work can result in boredom and frustration, resulting in psychological strain, which, in turn, can result in lower job involvement (Lambert, Kelley and Hogan, 2013). Some law enforcement positions can result in role underload. Role underload was predicted to have a negative effect on job involvement of Indian police officers (Hypothesis 2).
Some jobs involve great variety, whereas other jobs are highly repetitive (Lambert, Kelley and Hogan, 2013; Price and Mueller, 1986). Repetitive jobs require the same routine tasks with little variation in job skills (Lambert, 2004). These types of jobs allow for little mental stimulation and skill growth. Repetitive positions can result in tedium, which can result in frustration (Lambert, Kelley and Hogan, 2013). Repetitiveness can result in psychological strain, which, in turn, is likely to result in reduced job involvement. Repetitiveness was, therefore, postulated to have a negative effect on job involvement among police officers (Hypothesis 3).
Role ambiguity occurs when there is a lack of clear job expectations and confusion over how job tasks are to be accomplished (Lambert et al. 2005; Rizzo et al., 1970). For example, an officer may be told to do their job effectively without guidance on how to do so. Role ambiguity can hamper officers being successful in their jobs, as well as making the job itself trying. Role ambiguity can make work frustrating. In the end, this job demand can raise the level of psychological strain faced by a person (Michel et al., 2011; Schaufeli and Taris, 2014), and, therefore, the likelihood of psychologically withdrawing from the job increases. Role ambiguity was predicted to have negative effects on police officer job involvement (Hypothesis 4).
Fear of victimization refers to feeling at risk of being hurt or injured on the job (Cullen et al., 1985; Gordon et al., 2013; Griffin, 2001). Fear of injury or violent death are recognized as stressors (Burke et al., 2011) and this fear may be more common for police officers, where officers face a higher possibility of being injured or even killed than the average worker. Feeling at risk can wear on a person, resulting in apprehension and fear while on the job, which, in turn, raises the psychological tension and strain experienced (Lambert et al., 2005). Over time, fear of victimization raises the chances of negative job outcomes, including reduced involvement in the job. For the current study, fear of victimization was hypothesized to be a negative predictor of Indian officer job involvement (Hypothesis 5).
Role conflict occurs when the job duties, directions, and/or work orders for a position are inconsistent with one another (Poole and Regoli, 1983; Rizzo et al., 1970). For example, an officer may receive conflicting directions or orders from different supervisors, which can result in not knowing what course of action to take, ultimately increasing the level psychological strain from the job (Lambert et al., 2016). Role conflict make a job more difficult than it should be, lowering job performance and effectiveness and raising work frustration (Howard et al., 2004). In the end, psychological strain is likely to increase, resulting in a lower bond with the job. In the current study, role conflict was postulated to have a negative relationship with job involvement (Hypothesis 6).
Brief overview of the Haryana police
India is divided into 28 states. India’s state governments are authorized to make laws relating to the police, who are responsible for maintaining law and order. The Indian federal government supplements the efforts of the state governments by providing funds for police modernization and through specialized agencies that tackle serious cases or those involving multiple states or other countries; however, day-to-day policing is delegated to the states, each of which has its own police force. Each state police agency is headed by a director general of police who is responsible for the operation and control of the police in his/her respective state. The state police are further divided into districts, with a police force under the direction of a superintendent of police. The district unit is responsible for all policing activities including investigations, patrol, traffic, security of important government officials, and intelligence gathering. Larger cities (e.g., Delhi and Mumbai) operate a metropolitan police force, answering to a commissioner of police. The seven Union territories also have their own police forces (Raghavan, 1999). Indian Police Service (IPS) officers function as police managers and are appointed (and removed) by the federal government, and, to that extent, the federal government has some control. The majority of Indian police officers occupy the position of constable, which is the entry level for most officers and is a line (nonsupervisory) position.
The present study was conducted in the northern Indian State of Haryana. The state has an area of 17,070 square miles and a population of 25.5 million (Census2011.co.in, 2015). Haryana is adjacent to the national capital of New Delhi, and is considered a relatively well-developed state with a literacy rate of 77% and life expectancy of 66 years; however, it suffers from an unofficial caste-based social administration system (Sutar, 2012). The Haryana Police is divided into 21 districts and employs approximately 56,750 officers. The present study was conducted in Rohtak and Sonipat districts, which have a force of about 3000 total officers between the two districts. The Rohtak and Sonipat districts are each divided into 21 police stations (Haryana Police, 2017). The two districts serve a total population of about 2.5 million.
Being an Indian police officer is difficult (Raghavan, 1999). Police in India have a public image problem of being corrupt, ineffective, and sometimes abusive (Rathi and Barath, 2013). The negative image of the police not only wears on officers but can lessen the quality of interactions with citizens. Further, police in India, including Haryana, often face other work demands (Lambert et al., 2015; Rathi and Barath, 2013). Indian police agencies are often short-staffed and under-equipped, which can impede their performance (Raghavan, 1999; Rathi and Barath, 2013). In India, most police officers work more than 10 hours on their scheduled days and are on call 24 hours a day if needed; they also sometimes need to work in physically unpleasant environments (Saha et al., 2010). Indian police officers are not unionized and the work structure is paramilitary in structure. The challenges faced by a typical Indian police officer are many, and the consequences of job demands are typically harmful to both the officer and the organization. As such, investigating how job demands impact job involvement is important.
The Haryana police has embarked on several innovative programs driven by a professional approach. Haryana police has prepared a citizen’s charter to crystallize its commitment to the service of people. The charter lays down the motto of Haryana police as sewa suraksha sahyog (service, security, and cooperation). It assures residents of impartiality and transparency in the functioning of the police (Menon and Banerjea, 2012). A patrol program with community partnership was recently launched in Faridabad district (Razdan, 2016). To reduce corruption and to facilitate the delivery of services, an online portal called Harsamay was started by Haryana police where 31 citizen services including complaint registration, missing property reports, etc., are available (Sehrawat, 2017).
Method
Participants
The data was collected using systematic sampling of 1000 of the 3000 police officers in the Rohtak and Sonipat districts of Haryana State. A roster of all officers assigned to the districts of Rohtak and Sonipat was used, and every third office from the roster received a study packet that contained an informed consent form, a cover letter, the survey, and a return envelope. The back method of translation was used, wherein survey materials (informed consent, cover letter, and survey) were translated into Hindi, the national language of India, and then a second scholar translated the survey back into English to determine whether there were any translation problems. The sampled officers received the survey packet during roll call for the officers. The cover letter provided information about the study, indicated that participation was voluntary, that there were no rewards for taking the survey, that there were no punishments for not taking the survey, that they could stop taking the survey at any time, that any question could be skipped, that responses would be anonymous and kept confidential, and directions of how to return the survey. Officers were allowed to take the survey at a time and location of their choosing. Completed surveys were returned in an unmarked envelope included in the survey packet.
A total of 827 surveys were returned, a response rate of approximately 83%. Among the participants, the median age was 34 years, and ranged from 21 to 57. About 88% of surveyed officers were men, and 12% were women. In terms of work position, 69% were line officers (i.e., held the rank of constable). Approximately 50% of the participants were assigned to work in the Rohtak district, while the other half were assigned to the Sonipat district. The mean tenure with the police agency was 2.44 years, with a standard deviation of 3.05. In terms of highest educational level, 23% had completed grade 10 of schooling (matric degree), 35% had completed grade 12 (senior degree), 35% had an undergraduate college degree, and 7% had a postgraduate or professional degree. In terms of supervisory status, about 14% of the participants reported being supervisors of other police officers. Finally, 20% of the responding officers indicated they were part of the Scheduled castes, 33% indicated they were part of the Backward castes, and 47% indicated that they were part of the General castes. Caste is a form of social organization in India, although its hold is now weakening. Many times, the caste identity of an individual becomes significant in social interactions. In this sense, caste may be treated roughly equivalent to the concept of race, although race is more associated with the color of skin, whereas caste is associated with being born in a specific social group. The Scheduled castes occupy the lowest social status in the hierarchical caste division of the society. They have been given constitutional protection from discrimination and preference in jobs to improve their socio-economic status. The Backward castes comprise the next groups who are economically and socially disadvantaged, and the General category includes all other castes. Although it was not possible to obtain information on all the personal characteristics, such as educational level and caste, for the entire group of officers, the responding officers appeared to be representative to the overall complement of officers in terms of gender, age, position, supervisory status, and district.
Variables
Dependent variable
Job involvement, the dependent variable, was measured using four items from Kanungo (1982a). The job involvement items and their response categories are presented in the Appendix. The four items had an internal reliability, as measured by Cronbach’s alpha, of 0.63, which is on the low side, but above the rule of thumb cut-off of 0.60 (Gronlund, 1981). The factor loading scores were acceptable (see Appendix for the factor loading scores). The general rule of thumb is to use factor loading scores of at least 0.40 or higher, and those above 0.50 are viewed as good, and the factor loading scores for the job involvement items were above 0.50 (Kim and Mueller, 1978). The factor analysis results suggested that the items had unidimensionality, and, as such, the responses to the four items were summed together to form an additive index.
Independent variables
Personal demographics
Age, gender, position, district, tenure, educational level, and caste were included more as control variables than explanatory variables. These variables have often been included in past studies (Buker and Dolu, 2010; Gordon et al., 2013; Frank et al., 2017; Griffin, 2001; Lambert, 2008; Lambert et al., 2005, 2015, 2016; Lord, 1996; Martinussen et al. 2007; Redman and Snape, 2006). Moreover, personal demographic characteristics may influence perceptions of the work environment and work attitudes (Lambert et al., 2015). Age was measured in continuous years. Gender was measured as 0 = female and 1 = male. Position was measured as 1 = line officer (nonsupervisory) and 0 = other rank. Tenure at the current position was measured in continuous years. Educational level was coded as 0 = no college degree and 1 = college degree. Supervisory status was coded as 0 = not a supervisor of other staff and 1 = supervisor of other officers. The caste question asked whether the participant was in the Scheduled, Backward or General caste. In order to use this variable in multivariate analysis, dichotomous variables representing whether the participant was in the Scheduled castes (Scheduled = 1 and Backward and General = 0) or the Backward caste (Backward = 1 and Scheduled and General = 0) were created. The reference group for these two dichotomous variables was General caste.
Job demands
The job demand variables were role overload, role underload, repetitiveness, role ambiguity, fear of victimization, and role conflict. The job demand items are presented in the appendix, along with their response options. The items for role overload (two items), role underload (two items), role ambiguity (two items), and role conflict (four items) are from Triplett et al. (1996). Repetitiveness was measured using two items from Curry et al. (1986). Fear of victimization was measured using four items from Cullen et al. (1985). As there were only two items apiece for role overload, role underload, repetitiveness, and role ambiguity, no Cronbach’s alpha was calculated (Eisinga et al., 2013). The role conflict items had a Cronbach’s alpha of 0.62. The fear of victimization had a Cronbach’s alpha of .067. The job demand items were entered into a factor analysis. Each of the six job demand items (role overload, role underload, job repetitiveness, role ambiguity, fear of victimization and the role conflict) loaded on a single factor. The factor loading scores ranged from 0.58 to 0.82. The responses to the items were summed together to form their respective indexes (e.g., the responses to four fear of victimization items were summed together to form an index measuring fear of victimization).
Analytical plan
The univariate descriptive statistics are presented first to ensure that there was sufficient variation in the study variables. Next, bivariate relationships were examined to explore the associations between independent variables of interest and job involvement using calculating the Pearson product-moment correlation coefficient statistic, which is a common statistic to determine the association of two variables with one another. If a case was missing, pairwise deletion was used for the bivariate analysis. Ordinary least squares (OLS) multivariate analysis with all the independent variables entered in as one block was examined to determine which variables had a significant relationship with job involvement, independent of the shared variance with the other independent variables. Multivariate regression controls for the shared variance between independent variables and estimates the effect of a particular variable on the dependent variable once this shared variance is taken into account. Listwise deletion was used for missing responses in the multivariate analysis. There are different estimation techniques than OLS, such as maximum likelihood using structural equation modeling. OLS regression was selected because it tends to be robust and is a common estimation method in the policing literature. This least squares estimation technique is said to produce the best linear unbiased estimators (Berry, 1993). OLS regression also produces efficient estimates (i.e., the amount of variance in the ample estimates around the population value) (Berry, 1993). Further, the focus of the current study was to examine the direct effects of the stressor variables on job involvement, rather than a path/structural model (i.e., mediating and moderating effects). As such, OLS multivariate regression was determined to be the most appropriate statistical technique. Finally, to meet power of analysis, there should ideally be at least 20 cases (participants) and no fewer than 10 cases for each independent variable. With 15 total independent variables, there should be 300 (and no fewer than 150 cases); this criterion is met, even with listwise deletion, in the current study with 827 participants.
Results
The univariate statistics for the study variables are presented in Table 1. There appeared to be significant variation in the dependent and independent variables (i.e., none were constants). The data conformed approximately to a normal distribution based on the skewness and kurtosis statistics. Likewise, the median and mean values for the variables are similar to one another, also suggesting a normal distribution. All the alpha values were above 0.60, and the items loaded on the predicted factor in factor analysis.
Univariate statistics.
Note. Min, minimum value; Max, maximum value; Med, median value; α, Cronbach’s alpha, a measure of internal reliability. The total number of participants was 827.
A correlation matrix of the study variables is presented in Table 2. Three of the personal characteristics (control) variables had a statistically significant correlation with job involvement; they were gender, educational level, and Scheduled caste. On average, male officers reported lower job involvement than their female counterparts. Officers with a college degree tended to report lower involvement than did officers without a college degree. Officers who indicated that they were a member of the Scheduled caste tended to report higher job involvement than officers who indicated that they were part of the Backward or General castes. The other personal characteristics (age, position, district, tenure, supervisory status, and Backward caste) all had nonsignificant correlations with the dependent variable. All the job demand variables had significant correlations with job involvement, which means Hypotheses 1, 2, 3, 4, 5, and 6 were supported by the bivariate correlations. Role underload, repetitiveness, role overload, role ambiguity, and role conflict had negative correlations, which means a rise of any of these job demand variables was associated with lower job involvement. Interestingly, fear of victimization had a positive correlation with job involvement, which means that an increase in the fear of victimization variable was associated with higher job involvement.
Correlation matrix.
Note. Educ level, educational level; Scheduled, Scheduled caste; Backward, Backward caste; Role under, role underload; Repetitive, repetitiveness; Role over, role overload; Role Amb, role ambiguity; Fear Vict, fear of victimization; Role Conf, role conflict. The total number of participants was 827. For a description of variables, how they were coded, and their univariate statistics, see Table 1.
*p ≤ .05, **p ≤ .01.
An OLS regression equation was estimated with job involvement as the dependent variable and the personal (control) and the demand measures as the independent variables. The results are reported in Table 3. Collinearity and multicollinearity were judged not to be an issue. Collinearity is seen as a problem when two variables have a correlation higher than 0.70, and multicollinearity is seen as a problem when variance inflation factor scores (VIF) exceed 6 (Tabachnick and Fidell, 2013). The highest correlation and the highest VIF score were 0.68 (between age and position) and 2.23 (again between age and position), respectively (see Tables 2 and 3). The issues of outliers, influential cases, normality, linearity and homoscedasticity of residuals, and independence of errors (all of which can affect the regression results) were tested and judged not an issue (Berry, 1993; Tabachnick and Fidell, 2013). The R2 value was 0.21, which means the independent variables as a group explained about 21% of the observed variance in the job involvement index.
Regression analysis of the job strain model with job involvement as the dependent variable.
Note. Ordinary Least Squares regression was conducted. The dependent variable was job involvement. B represents the unstandardized regression coefficient, β the standardized regression coefficient, and VIF the variance inflation factor score, a measure of multicollinearity. See Table 1 for a description of the variables and how they were measured.
*p ≤ .05, **p ≤ .01
Among the personal characteristics, gender, educational level, and Scheduled caste had significant effects. Female officers in general reported greater involvement in the job compared with male officers. In general, officers with a college degree tended to report lower job involvement than officers without a college degree. Officers who reported being part of Scheduled caste tended to report higher job involvement compared with officers who reported being part of the Backward or General castes. Age, position, district, tenure, and supervisory status all had nonsignificant effects. Role overload, role underload, repetitiveness, role ambiguity, and role conflict each had significant negative effects, which means that an increase in any of these job demands was associated with a reduction in involvement in the job. As such, the multivariate regression results supported Hypotheses 1, 2, 3, 4, and 6. Conversely, fear of victimization had a nonsignificant effect on the dependent variable in the multivariate analysis, which means the multivariate regression analysis results did not support Hypothesis 5. Using the standardized regression slope coefficients (i.e., β in Table 3), the size of the effects can be ranked. Role ambiguity had the largest effect, followed by role overload, role underload, repetitiveness, gender, role conflict, educational level, and Scheduled caste.
Discussion and conclusion
Overall, the results of the current study support the job demands model in terms of workplace stressors having negative effects on Indian police officer job involvement. Five of the six hypotheses were supported, and all observed relationships were in the expected directions. Specifically, the workplace job demands of role underload, repetitiveness, role overload, role ambiguity, and role conflict had statistically significant negative effects on job involvement in the multivariable regression analysis. Fear of victimization, however, had nonsignificant effects in the regression model. It would appear that as a group, workplace stressors result in lower involvement in the job.
As postulated, role overload appears to result in less job involvement. Being overwhelmed can result in mental fatigue, enhancing the chances of mistakes on the job. Officers were less likely to bond with a job that overwhelms them. Likewise, role underload results in a weaker bond to the job among the surveyed Indian police officers. Being bored by the job is not a pleasant feeling. Officers are unlikely to bond with a job that does not stimulate them or offer them the chance to grow. These two job demands have not been researched in terms of their effects on job involvement. Our results suggest research among different police agencies in other countries should include role underload and role overload as possible predictors of police officer job involvement. Additionally, police administrators need to be aware of the job demands of role underload and role overload. Job redesign, which is within the control of police administrators, might ensure that the job duties and tasks of agency positions are more balanced. If job redesign for positions is not immediately possible, rotating officers between high demand and low demand positions may be a temporary solution. Administrators should ask supervisors to speak with officers and ask them about role underload and overload and obtain their suggestions for improvement.
Repetitiveness, as predicted, had a significant negative effect on job involvement among the studied Indian police officers. Repetitive jobs are monotonous, which is not likely to provide officers mental stimulation or personal growth. It makes sense that officers who perceive their jobs to be dull and routine would not bond with them. Moreover, repetitive jobs could place officers into a lull, placing them at risk, including making errors. No other study could be found that examined the relationship between repetitiveness and job involvement. Lambert et al. (2015), however, did report job variety, the opposite of repetitiveness, had a positive association with job involvement. Based on the current results, police scholars and administrators should be aware of the negative outcome of repetitiveness and undertake efforts to address it. While likely difficult to accomplish, changing the job duties and tasks of positions to minimize repetitiveness should be undertaken. Contrary to what is often presented in the mass media, the role of some officer positions, such as directing traffic, involves performance of repetitive tasks, while remaining prepared to employ other skills depending on the situation. Nevertheless, feedback from officers should be encouraged to determine what other things could be done to address this workplace stressor. Rotation of positions or patrol assignments should also be explored.
As hypothesized, role ambiguity and role conflict were associated with reductions in involvement in the job. These job demands likely make the job more difficult and unpleasant, resulting in psychological strain. Our finding of negative relationships with job involvement for both variables is consistent with the observed relationship of both job demands on the job involvement of U.S. police officers reported by Lord (1996). Our finding for role conflict, however, varied from that reported by Redman and Snape (2006), where role ambiguity had a negative relationship with job involvement, but role conflict had a positive relationship among British police officers. Role conflict having a positive effect on job involvement is not an intuitively obvious finding. There has been little research on how role conflict is linked to job involvement. Additional studies on the relationship between role conflict and job involvement should be undertaken in light of the mixed findings. The effects of role conflict could be contextual and vary across cultures. Alternatively, the findings could be due to random chance. Based on the negative effects of role ambiguity found by Redman and Snape (2006) and here, we recommend that police administrators be aware that this workplace stressor can reduce officer job involvement. Focus groups could be undertaken with officers to learn what aspects of their jobs cause role ambiguity. Additionally, supervisors should be made aware of the issue of role ambiguity and be charged with reducing its occurrence. It is important to note in our study that role ambiguity had the greatest effect on job involvement.
Contrary to our hypothesis, fear of victimization was not a significant predictor of job involvement in the multivariate analysis. No other published study could be found that explored the relationship between these two variables; therefore, we must consider that the current result could simply be due to random chance. Alternatively, this job demand might not play a direct role in shaping the job involvement of Indian police officers. Frank et al. (2017) observed that fear of victimization was not a significant predictor of police officer job stress. Those individuals who take a job as an Indian police officer may already be aware of the nature of the job and the risks it involves. The possibility of being the victim of physical harm may be construed as part of “real police work” and may not exert the type of influence on job involvement that it would for other occupations. It is unclear if the current finding of nonsignificant effects is universal or is unique to the current Indian police districts. Additional research is clearly warranted.
Among the personal variables, gender, educational level, and Scheduled caste had significant effects in the multivariate analysis. Female officers in our study reported higher involvement in the job compared with their male counterparts. Because there have been few studies, additional research is needed to determine whether the effects of gender on job involvement among officers differ across regions of the globe. In our study, officers with a college degree reported lower job involvement. This finding may be the result of status inconsistency. Officers with a college degree may wish to have more say in the police agency and control of their jobs than is typically found. Like many other nations, Indian police agencies are paramilitary, top-down organizations. This finding is the opposite as was observed by Lefkowitz (1974), who found college-educated officers in the USA reported greater job involvement. Officers who belonged to the Scheduled caste reported higher job involvement compared with Backward or General Castes. Being an Indian police officer for Scheduled caste members is a relatively new opportunity. As such, this new opportunity could result in being more involved in the job than those in the General castes. The Scheduled castes occupy the lower rungs in India’s caste-stratified society. It is possible that these officers put in extra effort in producing better performance to offset any biases that their colleagues and supervisors may have. More research is needed to explain this finding. We must point out that the issue of caste status is unique to India.
Although not reported in the results section, it is important to note that the personal characteristic variables only explained about 3% of the variance of job involvement. This means that the job demand variables explained a greater proportion of the variance of involvement in the job. This is good news for police administrators. Most of the personal characteristics are beyond the control of administrators. The workplace stressor variables, however, involve supervisory and managerial behavior over which administrators have control and, with focus and dedication, can be changed in an effort to influence officer job involvement.
The study has limitations. First, it is a single study. Additional research is needed across different Indian police locations to determine whether the results can be replicated. For example, the districts of Rohtak and Sonipat are both part rural and part urban, and the findings may not be generalizable to fully urban or fully rural areas in India. Similarly, research on the job demand model involving job involvement is required across various police agencies across the globe. Results could be contextual and situational. There has been too little research on the predictors of police officer job involvement to date, regardless of nation. Another limitation is that our study was cross-sectional and not longitudinal in design. As such, we are unable to show causality, even though our findings are theoretically consistent with the job demand model. An additional limitation is that role underload, repetitiveness, role overload, and role ambiguity were measured using two items. Future studies should measure these latent concepts using more items.
This study examined the relationship between various job demands and job involvement, and found support for the job demands model. Future research may also examine the effect of decision latitude on mental strain. Decision latitude relates to how much control a person has over the job demands, and greater control may result in lower strain and, in turn, less negative effects on job involvement (Taris et al., 2010). As previously indicated, the current study used OLS regression to examine the direct effects of stressor variables on police job involvement, which was acceptable for the purpose of this study. Nevertheless, it is important to note that there are other multivariate analysis techniques. New research may wish to use structural equation modeling of path models to determine whether variables, such different forms of social support, moderate or mediate the effects of the stressor variables on job involvement.
About 21% of the variance in the job involvement index was explained by the independent variables in the multivariate regression analysis. This means other variables help shape job involvement of the surveyed Indian police officers. These variables need to be explored, tested, and identified by future researchers. Finally, there is a need for additional studies on the possible consequences on officer job involvement. While the existing literature suggests that higher levels of job involvement are associated with more effective and efficient policing, there is a lack of police studies that examine this relationship. In any case, there is clearly a need for future research on police officer job involvement not only among Indian officers, but also across a myriad of nations. Studies in other nations should present a more coherent picture of how workplace stressors effect the job involvement of police officers.
In closing, the job demands model contends that negative outcomes, such as reduced job involvement are likely. Drawing from a survey of police officers from the state of Haryana, India, the present study supports this model and found that job stressors reduce job involvement. Specifically, the job demands of role overload, role underload, repetitiveness, role ambiguity, and role conflict had negative effects on job involvement among Indian police officers. Fear of victimization was the only job demand studied not to have a significant association with job involvement. There are many job demands in the occupational field of policing within the control of administrators, such as how officers interact with citizens, training for expected and common situations, and department and government support for officers. The job demands examined in this current study are changeable with time and efforts by police administrators.
Policing is already a demanding occupation; police agencies should not make it more difficult by failing to minimize workplace stressors, such as those studied here. Efforts aimed at reducing workplace job demands and improving the work environment of police officers not only benefits individual officers but also the police agency. Police managers in India and possibly other nations should pay attention to reducing the above noted workplace stressors to improve job involvement. Job involvement is known to improve performance, independence, and job satisfaction of employees. At the very least, it is hoped that the results this study will spur further research on how workplace factors are associated with job involvement of police officers not only in India but other nations. With additional research findings, law enforcement organizations may be in a better position to understand and eventually improve job involvement levels and other workplace factors. Without further research, changes and interventions would likely be haphazard. Police administrators need to appreciate the importance of job involvement and how workplace variables affect it. Additional research across the globe will provide information about which workplace factors are contextual in their effects on job involvement and which ones are universal. This would be an important step to help employees enhance their performance levels and ultimately aid organizations to achieve their aims and objectives effectively.
Footnotes
Acknowledgement
The authors thank Janet Lambert for proofreading and editing the paper. The authors also thank the reviewers for their comments and suggestions. All three authors contributed roughly equally to the paper and are listed in random order.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Appendix
The below index items were measured using a Likert response scale ranging from strongly disagree (coded 1) to strongly agree (coded 5). The factor loading score based on a factor analysis is given after each item in brackets.
Job Involvement: (1) I live, eat, and breathe my job [0.71]; (2) The major satisfaction in my life comes from work [0.72]; (3) The most important things that happen to me in my life usually occur at work [0.74]; and (4) Most of my interests are centered around my job [0.57].
Role Underload: (1) There is so little for me to do on a typical day at work that I am bored [0.78]; and (2) Haryana police expects too little of me in terms of work demands [0.78].
Repetitiveness: (1) My job requires that I be very creative (reverse coded) [0.78]; and (2) My job requires that I must constantly learn new things (reverse coded) [0.78].
Role Overload: (1) Haryana police expects too much of me in terms of work demands [0.82]; and (2) I consider myself overworked on my job [0.82].
Role Ambiguity: (1) I do not always understand what is expected of me at work [0.76]; and (2) The rules that we’re supposed to follow seem to be very clear (reverse coded) [0.76].
Fear of Victimization: (1) In my job, a person stands a good chance of getting hurt [0.66]; (2) I work at a dangerous job [0.83]; (3) My job is a lot more dangerous than most jobs in the community [0.76]; and (4) A lot of people I work with have been physically injured on the job [0.58].
Role Conflict: (1) I am ordered to do things differently than they should be done [0.73]; (2) Depending on the supervisor, my job orders differ greatly [0.58]; (3) I get criticized by one supervisor for doing things ordered by another supervisor [0.72]; and (4) At work, I am often caught between conflict orders from different supervisors [0.71].