In this study, the authors analyzed job satisfaction in community-based HIV/AIDS organizations, using variables measuring organizational characteristics and environmental influences. Hypothesized predictors of job satisfaction were computed in each of four employee groupings: supervisors, case managers and case management technicians, follow-up workers, and administrative support. The authors found that organizational characteristics do predict job satisfaction among employees with varying intensity based on position within the organization. Environmental influences had modest or no predictive power. Practical implications for managerial practice are made.
Key words: AIDS care; case management; community-based organizations; job satisfaction; organizational commitment
HIV/AIDS community-based organizations serve as the conduit for the delivery of social and health services for many clients from special needs populations. Employees of community-based organizations are challenged daily by obstacles such as razor-thin operating budgets, burgeoning caseloads, structural barriers to service access, and emotional turmoil surrounding clients' deaths. In this setting, employee stress levels are high and monetary compensation modest. Consequently, employee job satisfaction is of concern to practitioners responsible for organizational management.
For decades, scholars have studied predictors of job satisfaction in a variety of organizations and occupational settings. However, there is a paucity of research on job satisfaction and related concepts as applied to community-based organizations. This article expands existing knowledge by analyzing organizational and environmental predictors of job satisfaction within a population of HIV/AIDS community-based organizations. Respondents were stratified by a four-tiered employee classification system that included management, case managers and case management technicians, community follow-up workers, and administrative staff.
Job satisfaction, as a construct, has been subject to much debate and interpretation. Much of the debate has centered on operationalization of the job satisfaction variable and its association with other organizational influences. What is agreed on, however, is that job satisfaction is a complex and interrelated concept.
Spector (1997) defined job satisfaction as "simply how people feel about their jobs and different aspects of their jobs" (p. 2). He suggested that job satisfaction is important for three reasons:
"First, the humanitarian perspective is that people deserve to be treated fairly and with respect. Job satisfaction is to some extent a reflection of good treatment ... second, the utilitarian perspective is that job satisfaction can lead to behavior by employees that affects organizational functioning.... Furthermore, job satisfaction can be a reflection of organizational functioning" (p. 2).
Using job satisfaction as a predictor of job performance has produced mixed results in the literature. Iaffaldano and Muchinsky (1985) published a meta-analysis that found the predictive power of job satisfaction on job performance to be rather modest. However, researchers such as Barber (1986), Tziner and Vardi (1984), and Marriott, Sexton, and Staley (1994) established a direct relationship between job satisfaction and performance in human services work.
Job satisfaction is cited most often in the literature as a predictor of employee turnover (Agho, Mueller, & Price, 1993; Bluedorn, 1982; Gleason-Wynn & Mindel, 1999; Martin & Schinke, 1998; Poulin, 1995), and employee burnout (Lee & Ashforth, 1993; Martin & Schinke, 1998). Both outcomes negatively affect the operation of community-based organizations and consume precious resources.
In studying the influence of supervision on job satisfaction, Yost (1998) suggested that communication between employee and supervisor is important and explains some variance in job satisfaction. Agho et al. (1993) found that employees who had the autonomy to make job-related decisions and supervisors who actively assisted them were more satisfied on the job than those who did not. Lawler (1994), in ranking job attitudes, found supervision and communication to be among the highest ranked variables of employee attitudes.
In studying the influence of incentives on employee job satisfaction, Agho et al. (1993) found that employees who perceive that they were fairly rewarded for their contributions, have the opportunity for mobility within the organization, and receive adequate monetary compensation were more satisfied than those who were not. In addition, McNeely and Meglino (1994) introduced recognition for desirable behavior and reward equity as incentive variables. It is noteworthy, however, that the relationship between incentives and job satisfaction has not been consistently clear across studies, as Glisson and Durick (1988) and Poulin (1995) were unable to establish a statistically significant relationship between the variables in their related studies.
Job involvement, which has captured a literature stream of its own, may be defined as the degree to which individuals identify psychologically with their jobs (Blau, 1985; Kanungo, 1982). Liou (1997a) provided evidence to suggest that employees of community-based HIV/AIDS organizations are very involved in their jobs.
PURPOSE OF THE STUDY
In this study, we address two fundamental research questions:
1. Do organizational and environmental characteristics positively predict job satisfaction within community-based HIV/AIDS organizations?
2. Does the strength of these predictors differ based on employee level in community-based HIV/AIDS organizations?
We define environmental characteristics as those generally outside the control of social services organizations, such as the acuity of the clients they serve or the geographic regions in which they operate. Addressing these questions provides the opportunity to understand predictors of job satisfaction and their implication in community-based HIV/AIDS organizations. This understanding may enable managers and supervisors to engage in behaviors that have the potential for positively influencing employee satisfaction and behavior.
A model of the study framework is included as Figure 1. The model includes factors internal to the organization as well as environmental factors that might influence employee satisfaction. Internal factors were gleaned from the literature and include the influence of supervision, organizational commitment, incentives, and job involvement on job satisfaction. Environmental factors were derived primarily from discussion with case management employees and include the extent to which an agency's clients need assistance linking with services and resolving their problems, the geographic location of the case management agency, and the degree to which the agency serves particularly ill clients.
[FIGURE 1 OMITTED]
Organizational commitment has been defined in terms of the strength of an individual's identification with and involvement in a particular organization (Porter, Steers, Mowday, & Boulian, 1974). As an antecedent, organizational commitment has been used to predict employees' absenteeism, job performance, turnover, job satisfaction, and other behaviors (Mathieu & Zajac, 1990).
In 1983 the New York State Department of Health established the AIDS Institute as directed by state statute. The Institute was, and continues to be, charged with coordinating and implementing a statewide response to the disease. The Case Management Section of the AIDS Institute coordinates client services through, and provides oversight of, four models of delivery that differ in service intensity and participant eligibility.
This study focuses on employees of community-based organizations participating in one of these four programs titled the Community Follow-up Program (CFP). This program provides family-centered intensive case management services to targeted Medicaid-eligible populations, including HIV-infected people, HIV antibody-positive infants, and for a temporary period of time, high-risk individuals (AIDS Institute, 1999). Case management teams, each carrying approximately the same caseload, provide direct assistance to clients in obtaining access to medical care and supporting social services through frequent personal contact or home visitation with clients.
Data for this study come from two sources: (1) CFP program case management outcomes data collected yearly by the AIDS Institute; (2) a survey of individuals employed by CFP agencies collected in August 2000.
Since 1998 the Case Management Section of the AIDS Institute has collected outcomes data on all CFP agencies on a yearly basis. For this study we used data from fiscal year 2000, collected between April 1, 1999, and March 30, 2000. At each agency a team of trained research assistants review charts for clients who have been in case management for no less than six months and no longer than 12 months at the time of data abstraction. Sequentially, research assistants read the client's initial intake form (listing demographic data), initial assessment form, progress notes, and periodic reassessment forms. From these sources a wide range of individual client- and agency-level data are abstracted, including client demographic data (such as disease status), number and type of client needs identified through the case management process, and agency location.
In addition to yearly outcomes data, a survey was developed and administered anonymously to employees of each community-based organization affiliated with the Community Follow-up Program in August 2000. Each community-based organization CFP director solicited a volunteer employee to facilitate survey administration during a staff meeting. Each employee was provided a letter describing the purpose of the study and requested to complete a survey and deposit it in a single envelop for the organization. After surveys were collected, the envelop was sealed during the meeting and subsequently mailed to the authors. To reduce employee fear of reprisal, individual and aggregate results were kept confidential. A second mailing was sent only to organizations that did not respond to the first distribution of surveys. Forty-five agencies were participating in the CFP during the time of survey administration. In total, 34 of the 47 community-based organizations, or 72.3 percent of the population, administered the survey to staff in accordance with protocol. A total of 528 surveys were returned with a employee response rate that ranged from a low of 75 percent to a high of 100 percent within each participating community-based organization; when aggregated, the overall survey response rate for the study was 92 percent. Of the 34 community-based organizations that participated in the study, 28 were physically located in a geographic region identified as downstate New York, and six were located upstate.
Survey questions were designed to measure employee satisfaction and organizational characteristics related to satisfaction in four areas: job involvement, organizational commitment, incentives, and supervision. Individual survey questions were extracted from previously published studies and tested through the multivariate statistical techniques of factor analysis. Using SPSS software, we factor analyzed the aggregate collection of all survey responses through principal component extraction and Varimax rotation using Kaiser's normalization criterion for truncation at 1.0. The four factors, with eigenvalues ranging from 5.09 to 1.08, explained 58.8 percent of the total variance. The four factors and corresponding survey questions are included as Table 1.
Individual survey questions were presented using a Likert-type scale. Questions designed to solicit respondent agreement included a five-point scale where 5 = strong agreement and 1 = strong disagreement. Questions designed to solicit respondent frequency of specific activities also included a five-point scale where a score of 5 = almost always or always and a score of 1 = almost never or never. Although surveys were anonymous in nature, respondents were requested to indicate their position within the organization.
Three sets of variables were extracted from these two data sources: stratification variables, independent variables, and the dependent variable.
Stratification Variables. Because this study addresses differences in satisfaction between levels of case management workers, CFP employees were classified into four groups: management, case managers and case management technicians, community follow-up workers, and billing/administrative staff. A separate analysis was conducted on each employee classification.
The employee classification of management is defined in this study rather loosely. Specifically, the classification includes program directors, site supervisors, and senior case managers. Employees in the management classification are occupationally considered professional exempt employees, the majority from social work, and they typically hold earned graduate degrees.
The employee classification of case managers and case management technicians refers to individuals directly responsible for planning, coordinating, and implementing the case management of clients. The majority of case managers hold bachelor's degrees, and technicians have earned, or are in the process of attaining, associate degrees in human services.
The employee classification of community follow-up workers refers to a group of paraprofessionals who maintain close contact with clients while supporting implementation of case management service plans. A college degree is not required for community follow-up workers.
The employee classification of billing/administrative staff is broadly defined as any other support personnel employed within the community-based organization who work with the CFP. Many of these employees are not directly involved with clients and sometimes removed geographically from the case management site itself.
Independent Variable. Independent variables representing "organizational characteristics" as described in Figure 1, were crafted to reflect the four factors identified in the literature as contributing to employee satisfaction: supervision, organizational commitment, incentives, and job involvement.
The first factor, supervision, consisted of five questions designed to address employee perception of supervisory communication and authority within the community-based organization. One of the questions, related to employee participation in the community-based organization was reverse-ordered. The alpha coefficient of reliability for the factor was .86.
Factor two, organizational commitment, consisted of five questions designed to solicit employee's loyalty to the organization, belief that the priorities of the organization were important, acceptance of organizational values and assessment of trust and cooperation in the workplace. The alpha coefficient of reliability for this factor was .86.
The third factor, incentives, consisted of three questions designed to address employee perception of promotional opportunity, job security, and job recognition within the community-based organization. Two of the questions were reverse-ordered. The alpha coefficient of reliability for the factor was .81.
The fourth factor, job involvement, consisted of two questions designed to address employee perception of their involvement with their job in their community-based organization. The questions were extracted and modified from Lodahl and Kejner's (1965) factoral structure of job involvement. The alpha coefficient of reliability for the factor was .82.
Environmental influences included in the model are the average number of client needs, the geographic location of the community-based organization, and percent of sicker clients (that is, those HIV symptomatic or living with AIDS).
As part of yearly outcomes data collection, client needs are identified by research assistants. When a need is identified during chart review, either based on a documented client request for assistance or the case manager's report of a problem, that need is noted on the abstraction form. The variable included in our model represents a simple numerical count of needs identified.
The geographic location of the community-based organization was assumed to be related to the number of structural barriers influencing successful case management and optimal client outcomes that would in turn affect employee job satisfaction. The geographic location of participating community-based organizations were included and coded as 1 for organizations located downstate and 0 for organizations located upstate. (For the purposes of this study, downstate New York includes the five boroughs of New York City and Long Island.)
The "percent of sicker clients" measure represents the percentage of clients at each community-based organization who were HIV symptomatic or living with AIDS. The base assumption when first developing the model was that "sicker" clients consumed greater case management resources and could affect both organizational influences and job satisfaction. Data are reported on a 0 percent to 100 percent scale.
Dependent Measure. The dependent measure in this study is employee perception of job satisfaction. This measure was derived from the factoring of two employee survey questions that were analyzed separately from the factor analytic procedure used on the independent measures. The first survey question of job satisfaction, "I am very satisfied with my job in this organization," addressed the measure rather directly. The second, a reverse-ordered question, "Time seems to drag while I'm on the job," attempted to view job satisfaction from an alternative perspective. The percent of variance explained within the factor was 65.6 percent with an alpha coefficient of reliability of .84.
ANALYSIS AND RESULTS
The descriptive statistics (that is, means and standard deviations) for all environmental influences are included in Table 2. As data representing the four organizational characteristics are composite factor scores, no meaningful univariate comparisons of the variables are appropriate. Whereas the dependent measure of job satisfaction is a factor score, both questions comprising the factor are descriptively represented in Table 3. From both an aggregate and individual group comparison, employee job satisfaction is relatively high among participants, ranging from 3.94 to 4.40 on a five-point scale.
The average number of client needs produced a mean of 7.44 with standard deviation of 1.46, suggesting that the average client being served by the community-based organizations presented over seven major needs at the three months of case management assessment period. The number of needs per client often relates to the person hours expended by the case management team in service coordination and delivery. This also may be a proxy measurement of the amount of stress experienced by employees on the job.
The percentage of "sicker" clients, better defined as community-based organization clients who were HIV symptomatic or living with AIDS, produced a mean of 77.9 percent with a standard deviation of 12.8 percent. This statistic was measured and included in the model as an environmental influence because anecdotal evidence suggested that sicker clients consumed greater case management resources and may have affected employee job satisfaction.
Four separate OLS multivariate regression equations were analyzed from the stratified pool of survey respondents. There was variance in the number of respondents in each group. There were 74 surveys analyzed in the management subgroup, 286 surveys in the case manager and case management technician subgroup, 108 surveys in the community follow-up worker subgroup, and 52 surveys in the billing/administrative staff group. The variance may be attributed to the gross number of employees working in each of the respective employment classifications.
The OLS multivariate regression equation for the management subgroup produced an adjusted [R.sup.2] of .573 (Table 4). Each of the four organizational characteristics measured by the survey produced test statistics that exceeded critical values with significant findings. Comparing standardized regression coefficients for this subgroup, the measures for job involvement and organizational commitment predict at approximately twice the strength of the measure for incentives, where supervision predicted at approximately half of this level. None of the environmental influences produced significant results.
The OLS multivariate regression equation for the case manager and case management technician subgroup produced an adjusted [R.sup.2] of .303 (Table 5). Each of the four organizational characteristics measured by the survey produced test statistics that exceeded critical values with significant findings. Comparing standardized regression coefficients for this subgroup, the measure for organizational commitment predicted at approximately three times the strength of the measure for incentives, and job involvement predicted at approximately twice the strength of the measure. Supervision predicted at a level only modestly stronger than incentives. Whereas two of the environmental influences produced nonsignificant results, the measure for geographic location of the community-based organization was significant with a relatively weak standardized regression coefficient compared with organizational characteristics. The finding does suggest, however, that case managers and technicians employed in upstate community-based organizations are slightly more satisfied with their job than those employed in downstate organizations.
The OLS multivariate regression equation for the community follow-up worker subgroup, produced an adjusted [R.sup.2] of .350 (Table 6). Three of the four organizational characteristics measured by the survey produced test statistics that exceeded critical values with significant findings. Comparing standardized regression coefficients for this subgroup, the measure for organizational commitment predicted at approximately three times the strength of the measure for incentives, and supervision predicted at a similar level. The job involvement factor produced a nonsignificant finding. Whereas two of the environmental influences produced nonsignificant results, the measure for the average number of client needs was significant with a standardized regression coefficient comparatively exceeding that for supervision and incentives as organizational predictors. The finding suggests that the average number of client needs negatively affect the job satisfaction of community follow-up workers.
The OLS multivariate regression equation for the billing/administrative staff subgroup produced an adjusted [R.sup.2] of .316 (Table 7). Only two of the four organizational characteristics measured by the survey produced test statistics that exceeded critical values with significant findings. Comparing standardized regression coefficients for this subgroup, the measure for incentives predicts at approximately twice the strength of the measure for job involvement. None of the environmental influences produced significant results.
The residuals for each of the four OLS multivariate regression equations were subjected to various tests for distribution normalcy (that is, skewness and kurtosis). Although some evidence of other than normal distribution was uncovered in several of the residuals (that is, data skewed positively), findings were within tolerable limits.
Each of the four regression equations were subjected to tests for multicollinearity (that is, variance inflation factor (VIF) tests and collinearity tolerance tests). The results are presented in Tables 8 and 9 and do not suggest the presence of problematic multicollinearity.
Each of the four regression equations were subjected to tests for heteroscedasticity using a graphical method forwarded by Gujarati (1995) where the regression residuals were squared and plotted against the predicted values. No abnormal or curvilinear patterns emerged suggesting problematic heteroscedasticity.
DISCUSSION AND IMPLICATIONS FOR SOCIAL WORK
It should be noted that, although we captured over 72 percent of a population of AIDS community-based organizations as research participants, our findings are subject to external validity limitations. Generalizing findings beyond HIV/AIDS community-based organizations may be problematic because of the study framework that limited survey participation to HIV/AIDS community-based organizations operating in New York State. In addition, readers should be cautioned that sample size for the management subgroup and billing and administrative subgroup were less than optimal. Application of OLS regression on small samples may lead to problematic results.
Job satisfaction among each of the four employee groups studied in this population of HIV/ AIDS community-based organizations was relatively high, measuring 3.94 to 4.40 on a five-point scale. This finding is consistent with anecdotal evidence and personal observations made by the authors during the past decade. As many of the employees of the community-based organizations were social workers, this finding may run counter to the hypothesis of Barber (1986) that many social workers are dissatisfied with their jobs. It should be noted, however, that this comparison is a rather loose fit in that not all survey respondents were degreed social workers.
Data derived from this study found clear differences between predictors of job satisfaction for employees who interact directly with AIDS clients compared with those often removed from the "front lines" or any direct contact with clients. This suggests the need for different strategies when designing, supervising, and rewarding different categories of employees.
For employees who directly serve clients (that is, management, case managers and technicians, and community follow-up workers), organizational commitment is the dominating predictive factor of job satisfaction. This finding may be representative, as Liou (1997b) suggested, of the altruistic motivation for seeking employment (that is, to fight against AIDS or help the community) or personal beliefs and feelings. From a managerial perspective, it suggests the importance of supervisory techniques and organizational rituals that help increase direct service employees' feelings of association with and pride in their organization and its mission. It also may imply the need for managers to take special care to ensure consistency between stated mission and organizational strategies and tactics.
Supervision is also a significant predictor in all employee groups providing direct services to clients. This finding complements the work of Samantrai (1992) and Rauktis and Koeske (1994) who found that supportive supervision was positively correlated to job satisfaction. As suggested by Kadushin (1985), supportive supervision provides the psychological and interpersonal resources that enable the worker to cope with work-related stresses while at the same time mobilizing the emotional energy needed for effective job performance. Scarce training dollars in community-based organizations often are focused on the provision of skills training for direct care providers. Our findings suggest the additional importance of providing supervisory training and coaching for management personnel.
Job involvement is a significant predictor of job satisfaction for employees charged with case management responsibility, although this was not true for the other two groups of employees. Our finding appears to support Tannenbaum's (1966) hypothesis that workers holding higher-level positions within an organization show greater job involvement than lower-level employees. Alternatively, however, the finding also supports research that suggests that job involvement is in part correlated with an individual's autonomy on the job (Bass, 1965) and level of achievement (Patchen, 1970). In this occupational setting, management personnel, case managers, and case management technicians hold higher organizational positions within the community-based organizations and have greater autonomy or discretion in their interactions with clients.
In terms of environmental factors, the data show that the average number of needs per client negatively influences the job satisfaction of community follow-up workers charged with directly assisting clients in resolving these needs, for example by escorting clients to physician office visits. Managers, in considering this finding, might improve community follow-up workers' satisfaction by ensuring equitable distribution of high-need clients. Indeed, a number of case management agencies, as well as government organizations, have recently undertaken initiatives to better measure client need levels toward the goal of more appropriately assigning and managing case management loads.
Geographic location of the community-based organization is a predictor of job satisfaction in the case manager and case management technician employee group. The findings show that members of this group located in upstate New York were modestly more satisfied with their job than their downstate counterparts. Geographic location was not a significant predictor for any other group. Although, given their mission, most case management organizations cannot realistically relocate to other regions, understanding differing dynamics between geographic areas may be a first step in developing strategies to compensate for regional deficiencies. For example, qualitative information suggests that upstate New York case managers and case management technicians operate in a simpler referral environment where there is a greater likelihood that case managers will be personally acquainted with individuals employed by the agencies to which they refer clients. Thus, strategies aimed at building linkages between downstate case managers and referral sources might help to partially compensate for the more complex city referral environment.
For those employees typically removed from direct contact with clients (that is, billing/administrative staff), incentives are most predictive of job satisfaction. Managers desiring to increase the job satisfaction for this group of employees may opt to carefully examine financial remuneration schedules, job security, and frequency of individual recognition. At a minimum, managers may see gain from identifying the personal benefits of work-goal accomplishment while clarifying the path to reward (Bowditch & Buono, 1997).
We agree with Liou (1997a) in suggesting that research on job satisfaction and related concepts among employees in HIV/AIDS community-based organizations is still in its infancy. To fully understand the application of job satisfaction, organizational commitment, supervision and incentives requires researchers to engage in qualitative research within the structure of the HIV/AIDS community-based organizations. This may be especially helpful in explaining differences in the predictive strength of organizational characteristics to job satisfaction and identifying specific strategies for improving satisfaction for various employee levels within the organization. Qualitative research also may uncover additional variables for inclusion in a new model for predicting job satisfaction.
TABLE 1--Factor Loadings of Community-Based HIV/AIDS Services Organization Survey Questions on Variable of Job Satisfaction Organiza- tional Question Supervision Commitment I get coaching from my supervisor to help me do a better job. .79 .13 My supervisor asks for my opinions and thoughts about how to solve problems. .77 .24 I can talk freely to my supervisor when- ever I have difficulties at work. .76 .23 I receive all of the information I need to do my job effectively. .54 .18 I have enough authority in my position to get the job done. .45 .28 I am proud to tell others that I am part of this organization. .22 .73 This organization provides valuable public services. .11 .67 I believe that the priorities of this organization are quite important. .18 .67 Employees where I work trust each other. .22 .66 A spirit of cooperation and teamwork exists here. .29 .66 Doing my job well does little to improve my chances for promotion. -.09 -.05 Doing good work doesn't guarantee I will keep my job here. -.04 .24 When I improve my performance, my accom- plishments are recognized. .42 .30 I do more at work than my job requires of me. -.01 -.02 I do the best I can to get the job done, regardless of the difficulties. .14 .18 Eigenvalues 5.09 1.38 % variance explained 33.9 9.18 Cronbach's [alpha] .86 .86 Job Question Incentives Involvement I get coaching from my supervisor to help me do a better job. -.01 -.02 My supervisor asks for my opinions and thoughts about how to solve problems. .10 -.03 I can talk freely to my supervisor when- ever I have difficulties at work. -.10 -.01 I receive all of the information I need to do my job effectively. .35 .34 I have enough authority in my position to get the job done. .24 .21 I am proud to tell others that I am part of this organization. .16 .18 This organization provides valuable public services. -.09 .29 I believe that the priorities of this organization are quite important. .11 .27 Employees where I work trust each other. .22 -.19 A spirit of cooperation and teamwork exists here. .14 -.11 Doing my job well does little to improve my chances for promotion. .81 -.03 Doing good work doesn't guarantee I will keep my job here. .73 -.06 When I improve my performance, my accom- plishments are recognized. .58 -.09 I do more at work than my job requires of me. -.02 .77 I do the best I can to get the job done, regardless of the difficulties. -.09 .75 Eigenvalues 1.27 1.08 % variance explained 8.43 7.20 Cronbach's [alpha] .81 .82 (Principal component extraction with Varimax rotation using Kaiser normalization truncated at 1.0) TABLE 2--Univariate Statistics for Environmental Influences for Job Satisfaction for Community-Based HIV/AIDS Services Organizations Influence M SD Min. Max. Average number of needs per client 7.4 1.46 4.6 9.8 % "sicker" clients (those HIV symptomatic or living with AIDS) 77.9 12.82 56.0 100.0 Location: 28 Down- state 6 New York Up- state New York NOTE: Downstate New York is defined as the boroughs of New York City and Long Island. TABLE 3--Univariate Statistics for Dependent Measure of Job Satisfaction at Community-Based HIV/AIDS Organizations Subgroup Questions M Aggregate I am very satisfied with my job in this organization. 4.25 Time seems to drag while I'm on the job. 4.07 Management I am very satisfied with my job in this organization. 4.26 Time seems to drag while I'm on the job. 4.38 Case managers and I am very satisfied with my technicians job in this organization. 4.20 Time seems to drag while I'm on the job. 4.02 Community follow-up I am very satisfied with my workers job in this organization. 4.20 Time seems to drag while I'm on the job. 3.94 Billing/administrative I am very satisfied with my staff job in this organization. 4.40 Time seems to drag while I'm on the job. 4.31 Subgroup Questions SD Aggregate I am very satisfied with my job in this organization. .90 Time seems to drag while I'm on the job. 1.15 Management I am very satisfied with my job in this organization. 1.16 Time seems to drag while I'm on the job. .75 Case managers and I am very satisfied with my technicians job in this organization. .89 Time seems to drag while I'm on the job. 1.14 Community follow-up I am very satisfied with my workers job in this organization. 1.03 Time seems to drag while I'm on the job. 1.44 Billing/administrative I am very satisfied with my staff job in this organization. .72 Time seems to drag while I'm on the job. 1.08 NOTE: The actual measure of job satisfaction used in OLS multivariate regression equations was a factor score derived from both of these questions in each employee group. The individual questions are only reported for descriptive and interpretive review. TABLE 4--Management Subgroup in Community-Based AIDS Services Organizations: ANOVA for Job Satisfaction Variable (N = 74) Model Summary: [R.sup.2] .614 Adjusted [R.sup.2] .573 SE .640 ANOVA df Sum of Mean Squares Square Regression 7 42.978 6.140 Residual 66 27.070 .410 F= 14.970 Signif. F = .000 Coefficients Variable B SE B Beta Supervision .288 .081 .304 Org. commitment .417 .091 .395 Incentives .238 .087 .219 Job involvement .397 .085 .432 Number client needs -.010 .065 -.014 CBO location .026 .274 .010 % "sicker" clients -.001 .008 .001 Constant .098 .574 Coefficients Variable t Sig. T Supervision 3.536 .001 Org. commitment 4.559 .000 Incentives 2.734 .008 Job involvement 4.665 .000 Number client needs -.154 .878 CBO location .097 .923 % "sicker" clients .005 .996 Constant .172 .864 TABLE 5--Case Manager and Case Management Technician Subgroup in Community-Based HIV/AIDS Services Organizations: ANOVA for Job Satisfaction Variable (N = 286) Model Summary: [R.sup.2] .320 Adjusted [R.sup.2] .303 SE .793 ANOVA df Sum of Mean Squares Square Regression 7 82.179 11.740 Residual 278 174.834 .629 F= 18.667 Signif. F = .000 Coefficients Variable B SE B Beta Supervision .190 .046 .205 Org. commitment .376 .048 .397 Incentives .145 .047 .153 Job involvement .229 .047 .246 Number client needs -.054 .038 -.077 CBO location -.361 .157 -.149 % "sicker" clients .006 .005 .087 Constant .211 .346 Coefficients Variable t Sig. t Supervision 4.102 .000 Org. commitment 7.763 .000 Incentives 3.068 .002 Job involvement 4.927 .000 Number client needs -1.412 .159 CBO location -2.304 .022 % "sicker" clients 1.316 .189 Constant .610 .542 TABLE 6--Community Follow-up Worker Subgroup in Community-Based HIV/AIDS Services Organizations: ANOVA for Job Satisfaction Variable (N = 108) Model Summary: [R.sup.2] .393 Adjusted [R.sup.2] .350 SE .904 ANOVA Sum of Mean df Squares Square Regression 7 52.869 7.553 Residual 100 81.810 .818 F = 9.232 Signif. F = .000 Coefficients Variable B SE B Beta Supervision .250 .098 .203 Org. commitment .562 .089 .506 Incentives .189 .081 .188 Job involvement -.096 .094 -.083 Number client needs -.204 .066 -.259 CBO location .217 .327 .059 % "sicker" clients .004 .009 .046 Constant .873 .660 Variable t Sig. t Supervision 2.535 .013 Org. commitment 6.320 .000 Incentives 2.327 .022 Job involvement -1.022 .309 Number client needs -3.088 .003 CBO location .663 .509 % "sicker" clients .486 .628 Constant 1.323 .189 TABLE 7--Billing/Administrative Staff Subgroup in Community-Based IV/AIDS Services Organizations: ANOVA for Job Satisfaction Variable (N = 52) Model Summary: [R.sup.2] .410 Adjusted [R.sup.2] .316 SE .817 ANOVA Sum of Mean df Squares Square Regression 7 20.419 2.917 Residual 44 29.373 .668 F = 4.370 Signif. F = .001 Coefficients Variable B SE B Beta Supervision -.089 .119 -.091 Org. commitment .175 .119 .181 Incentives .560 .141 .483 Job involvement .345 .147 .279 Number client needs -.020 .096 -.028 CBO location -.311 .404 -.130 % "sicker" clients -.004 .015 -.052 Constant .850 .910 Variable t Sig. t Supervision -.757 .453 Org. commitment 1.469 .149 Incentives 3.976 .000 Job involvement 2.356 .023 Number client needs -.212 .833 CBO location -.770 .445 % "sicker" clients -.298 .767 Constant .933 .356 TABLE 8--Analysis of Residuals for Normalcy of Distribution for Subgroups in Community-Based HIV/AIDS Services Organizations Regression Equation Skewness SE Kurtosis SE Management--[e.sub.i1] -.440 .279 1.41 .552 Case managers and technicians--[e.sub.i2] -.430 .144 .270 .278 Community follow-up workers--[e.sub.i3] -.636 .233 .638 .461 Billing/administrative staff--[e.sub.i4] -.743 .330 .254 .650 TABLE 9--Collinarity Diagnosis of Independent Variables for Subgroups in Community-Based HIV/AIDS Services Organizations Regression Equation Subgroup Range of Tolerance Range of VIF Management .488-.912 1.097-2.051 Case managers and technicians .565-.989 1.011-1.769 Community follow-up workers .687-.948 1.055-1.456 Billing/administrative staff .435-.957 1.045-2.301 NOTE: VIF = variation inflation factor.
Agho, A. O., Mueller, C. W., & Price, J. L. (1993). Determinants of employee job satisfaction: An empirical test of a causal model. Human Relations, 46, 1007-1018.
AIDS Institute. (1999). About the AIDS Institute. Albany: New York State Department of Health.
Barber, G. (1986). Correlates of job satisfaction among human service workers. Administration in Social Work, 10, 25-39.
Bass, B. (1965). Organizational psychology. Boston: Allyn & Bacon.
Blau, G. (1985). A multiple study investigation of the dimensionality of job involvement. Journal of Vocational Behavior, 27, 19-36.
Bluedorn, A. C. (1982). A unified model of turnover from organizations. Human Relations, 35, 135-153.
Bowditch, J., & Buono, A. (1997). A primer on organizational behavior (4th ed.). New York: John Wiley & Sons.
Gleason-Wynn, P., & Mindel, C. H. (1999). A proposed model for predicting job satisfaction among nursing home social workers. Journal of Gerontological Social Work, 32(3), 65-79.
Glisson, C., & Durick, M. (1988). Predictors of job satisfaction and organizational commitment in human service organizations. Administrative Science Quarterly, 33, 61-81.
Gujarati, D. (1995). Basic econometrics (3rd ed.). New York: McGraw-Hill.
Iaffaldano, M. T., & Muchinsky, P. M. (1985). Job satisfaction and job performance: A meta-analysis. Psychological Bulletin, 97, 251-273.
Kadushin, A. (1985). Supervision in social work (2nd ed.). New York: Columbia University Press.
Kanungo, R. N. (1982). Measurement of job and work involvement. Journal of Applied Psychology 67, 341-349.
Lawler, E. E. III. (1994). Motivation in work organizations. San Francisco: Jossey-Bass.
Lee, R. T., & Ashforth, B. E. (1993). A further examination of managerial burnout: Toward an integrated model. Journal of Organizational Behavior, 14, 3-20.
Liou, K. T. (1997a). An analysis of employees' job involvement in a community-based AIDS service organization. Journal of Health and Human Services Administration, 20, 304-318.
Liou, K. T. (1997b). Employee job stress in the AIDS service organization: A study of personal and job factors. Journal of Health and Human Services Administration, 20, 393-408.
Lodahl, T., & Kejner, M. (1965). The definition and measurement of job involvement. Journal of Applied Psychology, 49, 24-33.
Marriott, A., Sexton, L., & Staley, D. (1994). Components of job satisfaction in psychiatric social workers. Health & Social Work, 19, 199-205.
Marriott, A., Staley, D., & Sexton, L. (1989). Social work professional development in Canadian psychiatric teaching facilities. Canada's Mental Health, 37(3), 20-23.
Martin, U., & Schinke, S. P. (1998). Organizational and individual factors influencing job satisfaction and burnout of mental health workers. Social Work in Health Care, 28(2), 51-62.
Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108, 171-194.
McNeely, B. L., & Meglino, B. M. (1994). The role of dispositional and situational antecedents in prosocial organizational behavior: An examination of the intended beneficiaries of prosocial behavior. Journal of Applied Psychology, 79, 836-844.
Patchen, M. (1970). Participation, achievement, and involvement on the job. Englewood Cliffs, NJ: Prentice Hall.
Porter, L. W., Steers, R. M., Mowday, R. T, & Boulian, P. V. (1974). Organizational commitment, job satisfaction, and turnover among psychiatric technicians. Journal of Applied Psychology, 59, 603-609.
Poulin, J. E. (1995). Job satisfaction of social work supervisors and administrators. Administration in Social Work, 19(4), 35-49.
Rauktis, M. E., & Koeske, G. F. (1994). Maintaining social worker morale: When supportive supervision is not enough. Administration in Social Work, 18(1), 39-60.
Samantrai, K. (1992). Factors in the decision to leave: Retaining social workers with MSWs in public child welfare. Social Work, 37, 454-458.
Spector, P. E. (1997). Job satisfaction: Application, assessment, cause, and consequences. Thousand Oaks, CA: Sage Publications.
Tannenbaum, A. (1966). Social psychology of the work organization. Belmont, CA: Brooks/Cole.
Tziner, A., & Vardi, Y. (1984). Work satisfaction and absenteeism among social workers. Work and Occupations, 11, 461-470.
Yost, C. J. (1998). A study of organizational and individual factors which are associated with greater job satisfaction as reported by registered nurses at one midwest hospital. Master's thesis, University of Minnesota, Bloomington.
This study was sponsored by the New York State Department of Health AIDS Institute, supported by Ryan White CARE Act Title II funding, and administered through Health Research, Inc. The authors are especially grateful to the CFP Quality Improvement Committee, the leadership and staff of the participating COBRA Community-Based Organizations located throughout the state of New York, and John Rohrbaugh, PhD, and Glenn Deane, PhD, from the State University of New York at Albany.
Original manuscript received March 19, 2001
Final revision received July 19, 2001
Accepted September 10, 2001
Ronald W. Gimbel, MA, is a doctoral candidate, Nelson A. Rockefeller College of Public Affairs, State University of New York, Albany e-mail: email@example.com. Sue Lehrman, PhD, is associate professor and director, Martin A. Strosberg, PhD, is professor of management, and Veronica Ziac, MBA, is research assistant, Graduate Management Institute, Union College, Schenectady, NY. Jay Freedman, BA, is director, Case Management Section; Karen Savicki, BA, is director, Bureau of Community Support Services; and Lisa Tackley, MPH, is administrator, Health Program, AIDS Institute, New York State Department of Health, Albany.…