- Split View
-
Views
-
Cite
Cite
Diane Brannon, Teta Barry, Peter Kemper, Andrea Schreiner, Joe Vasey, Job Perceptions and Intent to Leave Among Direct Care Workers: Evidence From the Better Jobs Better Care Demonstrations, The Gerontologist, Volume 47, Issue 6, December 2007, Pages 820–829, https://doi.org/10.1093/geront/47.6.820
- Share Icon Share
Abstract
Purpose: We assess how perceived rewards and problems with caregiving work and supervision relate to intent to leave among direct care workers who are employed in provider organizations participating in the Better Jobs Better Care (BJBC) demonstration; we also examine how these relationships vary by provider type. Design and Methods: Direct care workers from 50 skilled nursing facilities, 39 home care agencies, 40 assisted living facilities, and 10 adult day services in five states completed a paper survey administered prior to the implementation of the BJBC interventions in each organization. We include direct care workers (n = 3,039) with complete data in the analyses using multinomial regression clustered by provider organization to compare those not at all likely to leave and those very likely to leave in the next year with a middle referent group who are somewhat likely to leave. Results: Logistic regression results were that work overload and lack of upward mobility increased intent to leave. Respondents with positive assessments of their supervisor, who valued helping others, and for whom the income was perceived as rewarding were less likely to be in the very likely to quit category and more likely to be in the stable category. Some differences between provider types are observed, especially between home care workers and those employed in facilities. Implications: These findings provide support for many of the management-practice improvements taking place in the field, including those implemented in the BJBC demonstration. Follow-up surveys will provide insight into their effectiveness.
In this study we examine the first wave of data collected from direct care workers in the Better Jobs Better Care (BJBC) demonstration in order to understand how job perceptions relate to direct care workers' intent to leave across the range of long-term-care provider organizations. The BJBC demonstration is a $15.5 million project funded by the Robert Wood Johnson Foundation and The Atlantic Philanthropies as a national demonstration to investigate, implement, and disseminate strategies to improve recruitment and retention of direct care workers across the main types of long-term-care provider organizations. The relationship between the quality of caregiving as work and the quality of long-term care is central to having the BJBC project implemented in five states. Bringing together broad coalitions of providers, policymakers, and advocacy organizations, the demonstrations are intended to identify practical strategies for caregiving workforce development.
Underlying the initiative and the diverse projects being implemented is an appreciation for the nature of the work itself and how it is perceived by job incumbents. Nonprofessional direct care workers provide intimate, life-altering care that is subject to many of the same pressures faced by health care professionals. Their caregiving role, however, is not defined, protected, and buffered by professional identity. Consequently, understanding the positive and negative responses these direct care workers have toward their respective jobs can provide insight into the design of effective retention efforts, which is a persistent challenge for both policymakers and industry leaders.
The Turnover and Retention Challenge
Retaining qualified direct care workers is a challenge for all types of long-term-care provider organizations. Direct care worker turnover in nursing facilities ranged from 48% to over 100% in 2001, depending on the region of the country (American Health Care Association, 2002). Although the structure of the home care industry has made data collection on turnover more challenging, one recent report indicated that vacancy rates in home care organizations are even higher than those in skilled nursing facilities (Brannon, Barry, Angelelli, & Weech-Maldonado, 2005).
High direct care worker turnover in institution-based long-term-care settings has been linked to a variety of negative outcomes, including disruptive resident behaviors, increased rates of infection, and increased presence of pressure sores and restraint use (Burgio, Fisher, Fairchild, Scilley, & Hardin 2004; Castle & Engberg, 2005). As public “report cards” that focus on clinical outcomes have stimulated competition based on public perceptions of quality, long-term-care employers are increasingly motivated to improve their caregiving workforce. At the same time, demographic forces increase the need for long-term-care services while decreasing and diversifying the pool of the persons who have traditionally assumed this work role (Harmuth & Dyson, 2005; Harris-Kojetin, Lipson, Fielding, Kiefer, & Stone, 2004; Montgomery, Holley, Deichert, & Kosloski, 2005).
Conceptual Framework
Using a baseline cross-sectional survey of individual direct care workers employed across the continuum of long-term-care settings, the present study's primary contribution is that it draws on two complementary themes of management theory and research, each of which has its roots in well-established social-psychological principles. The first theme is the role of job rewards and job concerns in influencing a variety of individual outcomes including psychological well-being, intent to leave, and physical health. Karasek and colleagues (Karasek 1979; Karasek, Triantis & Chaudry (1982), by using their Job Content Questionnaire across occupational types and in many different cultural and political settings, established that negative health effects are more prevalent among workers who have both high job demands and low decision latitude. Marshall, Barnett, Baruch, and Pleck (1991; also see Marshall & Barnett, 1993) adapted this model to focus on the job rewards and job concerns of female workers in caregiving occupations dominated by women. Most notably, they added the concept of “helping others” to the job characteristics literature, which had its origins in survey work with male samples in primarily industrial settings (cf. Hackman & Oldham, 1980).
The second theoretical theme is the critical role that the quality of relationships with supervisors plays in turnover decisions of individuals. This employee–supervisor relationship has been most thoroughly developed conceptually and empirically tested as leader–member exchange theory (Dansereau, Graen, & Haga, 1975; van Breukelen, Schyns, & LeBlanc 2006), which challenged the construct of “average leadership style” with a focus, both conceptual and methodological, on dyadic relationships between leaders and subordinates. The relationship between individual workers and their supervisors has been shown over a large number of studies to promote increased subordinate satisfaction, performance, and career outcomes and to reduce the propensity to quit (Gerstner & Day, 1997). Perceived supervisor support has been shown to be significantly related to perceived organizational support and employee retention (Eisenberger, Stinglhamber, Vandeneberghe, Sucharski, Rhoades, 2002; Walborn, 1996).
Both the job rewards–job concerns model and leader–member exchange theory address the role of individual variation in person–job fit and in subordinate–supervisor relationships. The documentation by Bowers, Esmond, and Jacobson (2003) of the turnover-producing negative influence of social behavior within the nursing home hierarchy highlights the importance of understanding how individual direct care staff members perceive their work and their role in the organization. This is particularly important in studying the cognition of intent to turnover, which is a personal behavioral intention.
An additional contribution of this study is our analysis of these direct care worker perceptions across different provider types within the long-term-care sector with a relatively large sample. It complements prior studies of direct care worker turnover that used the organization as the level of analysis (Banaszak-Holl and Hines, 1996; Brannon, Zinn, Mor, & Davis, 2002). Although it has been shown that organizational and market factors such as staffing ratios, ownership, size, number of employment opportunities, and the like influence turnover rates experienced by nursing facilities (Castle & Engberg, 2005), the process by which caregiving staff move from dissatisfaction to intent to leave to actual turnover is not well documented. A recent study by Castle, Anderson, and Engberg (in press) of 1,779 nursing assistants from 72 nursing facilities found weak evidence for the role of facility and market characteristics, in comparison with perceptions of job satisfaction and role characteristics in addition to personal factors in predicting either turnover intention or actual turnover within 1 year.
Thus, in this analysis, we focus on how individual direct care workers perceive their job rewards and problems and their assessment of the quality of supervision they receive, and we control for their perceptions of how organizational and market factors, such as their pay and opportunities elsewhere, influence their intent to quit. We examine these factors across and within provider types.
Addressing the background just discussed and controlling for other factors, we posed the following research questions: First, are direct care workers with greater perceived job problems more likely to intend to quit within the next year? Second, are direct care workers with greater perceived job rewards less likely to intend to quit within the next year? Third, are direct care workers who perceive better quality supervision less likely to intend to quit within the next year?
Methods
Participants
Researchers at the Penn State University Survey Research Center administered the survey. We had direct care workers recruited from long-term-care organizations participating in the BJBC demonstration in Iowa, North Carolina, Oregon, Pennsylvania, and Vermont. These organizations include nursing facilities (all were skilled nursing facilities), home care agencies (including both Medicare-certified and non-Medicare-certified agencies), assisted living facilities and other residential personal care homes, and adult day services. (Assisted living facilities operate under different nomenclature in different states but serve the same client population and have the same organizational structure. Therefore, in this present study we refer to all of these facilities as assisted living facilities for the purposes of data analysis and interpretation.)
Different methods were used by each state to recruit and select facilities. Although the sample of provider organizations can only best be described as a convenience sample, it does reflect the range of types of long-term-care provider organizations. Nonprofit or government-owned organizations, however, are overrepresented at 64% percent. A comparison of the sample skilled nursing facilities' quality indicators from the Centers for Medicare and Medicaid Services Web site revealed that the indicators for the participating facilities in each state were consistent with the mean indicators for that state (data not shown).
We generated the sampling frame for the direct care worker survey from a management information system developed to track direct care worker turnover in each of the 132 organizations in the BJBC demonstration. After the list of current employees was provided to the Penn State Survey Research Center, we had personalized survey packets including a cover letter, the survey, a financial incentive, and a business reply envelope sent to the provider. Surveys were distributed by the provider prior to the initiation of the BJBC project. Response rates are based on the total number of direct care workers employed at each facility at the time of the survey administration. Overall, we achieved a response rate of 54.4% (n = 3,468); the range of response rates among provider types was 51.9% for skilled nursing facilities to 72.0% for adult day services. A total of 3,468 direct care workers completed the survey, with complete data needed for this analysis available for 3,039. The distribution of the total 139 providers and their respective direct care worker respondents is as follows: 50 enrolled skilled nursing facilities yielded 1,262 (M = 25.2) respondents; 39 home care agencies yielded 1,306 (M = 33.5); 40 assisted living facilities yielded 425 (M = 10.6); and 10 adult day service providers yielded 46 respondents (M = 4.6).
Nursing and assisted living facilities had the lowest response rates (both 52%) and home care and adult day services providers had the highest response rates (57% and 72%, respectively). Because the number of workers in adult day service providers (n = 54) was too small for us to analyze separately, we included them in our total sample but did not analyze them separately. Using data from the information system, the Survey Research Center analyzed differences in response rates by several characteristics (e.g., state, provider type and size, and respondent tenure) and researchers reweighted the data so that the distribution of the sample matched the population distribution on these characteristics.
Measures
The data we used in this analysis come from the 86-item BJBC Direct Care Worker Survey, with the exception of the variable used to code the type of facility; we took this variable from a separate survey of each provider organization's chief clinical manager.
Dependent Variable
We used an established single item labeled intent to leave that was adopted from the Michigan Organizational Assessment Battery (Cammann, Fichman, Jenkins, & Klesh, 1983). This item focuses on behavioral intent by asking workers, “How likely is it you will leave this job in the next year?” We scored this item on a scale of 1 to 3, with higher values indicating a greater likelihood of leaving (1 = not at all; 2 = somewhat; 3 = very). Intent to leave has been found to moderately, consistently, and significantly correlate with actual turnover (Griffeth, Hom, & Gaertner, 2000; Harris, James, & Boonthanom, 2005). Although this cognitive measure of withdrawal intent within a specific time period is not a proxy measure of actual turnover, meta-analytical findings show it to be the strongest predictor of individual voluntary turnover (Tett & Meyer, 1993). Furthermore, even though attitudes such as job satisfaction have actually been found to predict actual quits, the cognition of turnover intention is an important intervening mechanism in the path from affect and job perceptions to actual turnover behavior.
Independent Variables
As we already stated, the independent variables in this study measure the direct care workers' perceptions of job and related work-systems characteristics that have been associated with a variety of outcomes, including job turnover, absenteeism, loyalty, and satisfaction, in extensive research on organizational behavior (see Kiefer et al., 2005).
Theoretically, whenever a given employee is considering whether or not to resign, job problems are balanced against job rewards. The job problems and job rewards measures that we used in this study were adapted from the Job Role Quality Survey (JRQ), developed by Marshall and colleagues (1991). The JRQ has been used to examine occupational stress and job attitudes and outcomes among female health care workers (Marshall & Barnett, 1993). We operationalized all dimensions as mean scale scores (range, 1 = not at all rewarding to 4 = extremely rewarding), with higher scores indicating greater perceived rewards or problems. The one exception was discrimination, which was a binary judgment that we coded as yes = 1 and no = 0 regarding the experience of discrimination at the current job. Acknowledging that some items may not be perceived to fit jobs across all settings, we also included a “does not apply to my job” option, with those respondents dropped from the analysis of that item.
Job Problems
We used the following subscales to measure four job problems. The Hazard Exposure subscale measured physical risks associated with the work, including getting hurt, catching an illness, being in poor physical conditions, and experiencing physical strain. The Dead-End Job subscale measured lack of upward mobility in the position, including finding the job boring or repetitive, with little skill development or chance for promotion. The Overload subscale measured job demands such as having too much work, being isolated and lacking help, and experiencing emotional and physical demands. The Discrimination subscale measured discrimination on the job, based on race or ethnicity.
Job Rewards
We measured job rewards by using scales from the JRQ, constructed as previously described, with some modifications in wording made after pilot testing as follows. Challenge is the extent to which skill variety and skill application are interesting and promote a sense of development and accomplishment. Recognition is the degree of reward associated with feeling valued by management and others and getting credit for one's work. Helping others describes the extent to which altruistic aspects of the job such as helping, being needed, and making a difference in others' lives is rewarding. Decision authority is the degree to which autonomy and empowerment are perceived as a rewarding characteristic of the work. Income is the extent to which the income derived is viewed as relatively rewarding compared with that of other jobs in the field. Team spirit describes coworker support and affiliation with the work group.
Quality of Supervision
For the purposes of comparability with other data, we measured quality of supervision by using eight items from the supervisor subscale on the National Nursing Assistant Survey that was administered as part of the National Nursing Home Survey (2004). This scale was developed to assess the quality of supervision as perceived by individual direct care staff. We added one item to capture the extent to which the respondent felt that her supervisor offered useful criticism. The revised scale measures both of the commonly noted aspects of supervisory leadership, that is, support (e.g., listening, feedback, providing opportunities) and structure (e.g., clarity of instructions, appropriate use of discipline and useful criticism). We measured items on a scale ranging from strongly disagree (scored 1) to strongly agree (scored 4) and combined them to create a mean score. Higher values indicate that the direct care worker perceives higher quality supervision.
Control Variables
We also included individual-level characteristics that have been found to influence turnover, such as job tenure, race, educational level, and self-efficacy (Harris-Kojetin et al., 2004)). For this analysis, we used tenure as a direct care worker rather than tenure in the current setting, and we calculated this as the total number of months the individual reported having worked as a direct care worker anywhere. Age was reported in categories; we use it here as a continuous variable, using the midpoint of each category with higher values indicating increased age. We measured race with a single question with response categories of White, Hispanic or Latino, African American or Black, American Indian or Native American, Asian, Native Hawaiian or Pacific Islander, and Other. For analysis purposes we collapsed these into White and non-White race because there were small frequencies in two of four categories. We collapsed educational levels as secondary education after high school verses high school alone. Self-efficacy was the mean of two items that asked about having learned the skills necessary to do the job well and self-confidence. We measured these on a 4-point scale, with higher scores indicating greater self-efficacy. Finally, we assessed the perceived availability of other better-paying jobs, referred to hereafter as “job alternative,” with a single item: “I could get a job that paid more than this job.” We measured this on a scale of 1 to 4, with higher scores indicating higher agreement.
Analysis
We tested the aforementioned model by using clustered multinomial logistic regression comparing those not at all likely to quit and those very likely to quit with the middle, somewhat likely category. We calculated scales with missing items if at least two thirds of the items were completed and others were dropped, yielding a remaining sample of 3,039. To adjust for lack of independence of the observations, we cluster respondents by provider organization and we use robust standard error estimates in the analysis.
We ran the full logistic regression model on the total sample and then ran it separately for nursing facilities, home care agencies, and assisted living facilities. As stated previously, the sample of adult day services was too small for separate analysis.
Results
Descriptive Findings
As we can see in Table 1, the sample had nearly equal numbers of home care agency workers and skilled nursing facility workers, with fewer assisted living and adult day staff. Of the total sample, 1,734 individuals (57.1%) reported that they are not likely to leave in the next year and 404 (13.2%) indicated a high probability of quitting, with the remaining 903 (29.7%) in the middle referent category.
The home care aides appear to be most committed to staying in their jobs. Home care workers in the sample are also older. The respondents with the longest tenure, however, were those employed in skilled nursing facilities. Two thirds of the participants were White and 25.6% were African American (data not shown). Fewer than 1 in 10 respondents reported education beyond high school.
Table 2 shows total sample mean scores as well as Cronbach's alphas for the scales. Alphas ranged from.70 to.80 for rewards and problems to.92 for quality of supervision. The reward with the highest mean was helping others, which indicates that direct care workers found this to be the most positive aspect of their job. Income as a job reward received low ratings.
Differences with regard to job rewards and concerns among direct care workers working in different provider types were not large, but a pattern was evident in that home care workers distinguished themselves from others in reporting less concern with job problems, greater rewards, and more positive perceptions of supervision. The problem of work overload was expressed most forcefully in skilled nursing facilities.
All the independent and control variables were significantly correlated with the three-level intent-to-leave outcome variable (Table 2). The dead-end job measure had the largest correlation with intent to leave among the perceived job problems (.27), closely followed by overload (.24). Among the rewards, the correlations with income as rewarding (−.23), helping others (−.20), and team spirit (−.22) were largest. Perceived quality of supervision had the greatest correlation (−.30) with intent to leave.
Intercorrelations among scales were significant, suggesting that a reduced number of factors might be analytically appropriate. For this baseline analysis, however, our goal was to maintain as much detail as possible in describing how direct care workers perceive their work situations in formulating intentions to leave, consistent with the theoretical framework adopted.
Multinomial Logistic Regression Results
Results addressing the research questions are reflected in the total sample, multinomial logistic regression model results shown in Table 3. The differences among provider types are shown in the subsample models for skilled nursing facilities, home care agencies, and assisted living facilities. The number of responses from adult day services was too small to allow for multivariate analysis, but this group is included in the total sample model.
Job Problems
With regard to job problems, direct care workers who perceived a lack of opportunity for advancement, that is, a dead-end job, are significantly less likely to be in the very stable group (those not at all likely to leave in the next year). The dead-end job impact is most clearly evident in the home care workers, in which both the not at all likely group and the very likely group are significantly different in the expected respective direction from the middle group. Overall, direct care workers who experience greater work overload are significantly less likely to be in the most stable group with regard to intent to quit. The home care respondents provide the exception to this finding, however, as overload was not a significant predictor of either high or low intent to quit for these workers. This likely reflects the difference in the amount of pacing control inherent in one-to-one home care, which does not exist in institutional settings. For the skilled nursing facility workers, the odds of being in the very stable group were reduced by 50% with each 1-point increase on the 4-point overload scale. The strongest impact in the assisted living facility workers, by contrast, was that the odds of being in the very likely to quit group were more than doubled with each 1-point increase in perceived overload. In sum, the association between perceived overload and intent to quit is significant for facility-based direct care workers, but not for home care workers; thus the magnitude of the effect in the total sample model is understated for facility-based caregivers and overstated for those working in home care. Other job-related problems, such as health and safety hazards and discrimination, did not distinguish among intent-to-leave categories.
Job Rewards
With regard to job rewards, the value of helping others was positively related to being in the stable, not at all likely to quit group and negatively related to being in the very likely to quit group. The effect was similar across provider types and significant in all but two comparisons. For the total sample, a 1-point increase on the measure of helping others increased the odds by 45% of being in the stable group with a low intent to quit. Similarly, a 1-point increase in a respondent's score on helping others was associated with a 35% decrease in the odds of being in group with a high intent to quit.
For the total sample, the extent to which direct care workers perceive their income to be a rewarding part of their job significantly distinguished the middle, somewhat likely to quit group from both the not at all likely and the very likely to quit groups. The odds ratios indicated a smaller impact than the other significant job problems and rewards, however. Examination of the three provider-specific samples showed that this effect was not equally distributed. Direct care workers in the skilled nursing facility sample showed less of a connection between the value of income earned and intent to quit than did the home care agency and assisted living workers. A 1-point increase on this perceived rewards scale increased the likelihood of being in the group with a low intent to quit by 20% for home care workers, 17% for those in assisted living facilities, but only 4% for skilled nursing facility workers.
Controlling for other factors, we found that team spirit, the value placed on coworker relationships, was not a factor in skilled nursing facilities, but it was in the other settings, significantly so in assisted living, where higher scores were associated with being not at likely to leave within 1 year. Recognition, decision authority, and challenge may be important job characteristics, but they were not significant predictors of intent to leave with the exception that assisted living workers who perceived their jobs as more challenging were much more likely to be in the very likely to leave category. This is counter to what we expected and warrants further investigation, as this subsample also rated its skill self-efficacy lower than did the others.
Supervisor Quality
With regard to the role of perceptions of supervision, the data show that both low and high intent to quit are associated with respondents' assessment of the quality of supervision they receive, which is consistent with the theory that the supervisor–subordinate relationship influences personal work outcomes. It is worth noting that the respondents did not rate supervision as particularly low (M = 3.12, with 4 indicating highest quality). For every scale 1-point increase on the supervision quality measure for the total sample, however, the risk of being in the group with the high intent to leave was reduced by 30%, when we controlled for other factors in the model. This observed relationship was evident in all three subsamples.
Control Variables
Among the control variables, tenure as a direct care worker and age increased the likelihood of being in the not at all likely to quit category. As we expected, workers with more education (those with greater than a high school degree or its equivalent) were twice as likely to be in the very likely intent to quit group as in the middle referent category. This effect was strongest in skilled nursing facilities.
For every 1-point increase on the 4-point perceived better paying job alternative measure, the odds of being in the group with a very high intent to leave was 40% higher than the odds of being in the somewhat likely to quit group. This effect was not significant in home care, but it was particularly large in the assisted living sample. In each of the provider-specific samples, higher scores on perceptions of job alternatives significantly reduced the odds of being in the not at all likely to leave category.
Discussion
Consistent with previous studies of women in caregiving occupations (Marshall & Barnett, 1993; Marshall et al., 1991), the intensity with which direct care workers perceived the negative aspects of their work to be problems was considerably less than the degree to which they perceived the positive aspects to be rewarding. Overall, mean subscale scores were notably lower for the problems than for the rewards, with the exception of income, which was rated very low as a reward.
However, after controlling for all other variables, we found two job problems to be significantly related to intent to leave. Specifically, the perceived lack of opportunity for advancement and the perception of work overload were most significant, particularly among home care agency and skilled nursing facility workers, respectively.
These findings raise the question of whether changes in management practices in long-term-care provider organizations can be made that will improve retention by mitigating these factors. The BJBC projects include initiatives that address the lack of upward mobility faced by direct care workers. In Iowa, for example, direct care workers in pilot sites are being given a more defined role in orientation of new workers through a designated peer-mentoring role. North Carolina has recently recognized a specialist medication aide job, and Oregon's project includes the development of career ladders that increase opportunities. Pennsylvania's demonstration is identifying “Best Practice” sites with regard to work redesign. Workload is addressed in the Vermont Gold Star Standards for Nursing Homes and in the North Carolina NOVA designation, a special license designation that recognizes providers who have met standards for workforce development.
The findings also indicate that there are positive aspects of caregiving work that management may want to build on to increase retention. Altruism (helping others) contributes to a lowering of the intent to quit. This finding is in keeping with other studies that have found that job satisfaction or commitment among direct care workers is strongly related to their relationships with their clients (Freidman, Daub, Cresci, & Keyser, 1999; Parsons et al., 2003). Strategies for enhancing the value of helping others might include stable assignments that encourage relationships, thoughtful matching of clients and caregivers, rewards and recognition, and training and care planning participation that broaden the scope of helping opportunities. This kind of practice is evident in Oregon's and Iowa's adoption of “Person-Centered Care” training and in Pennsylvania's and Vermont's standardized curriculum for direct care workers.
As the quality of supervision was a significant influence on intent to leave across settings, it is worth noting that training of supervisors in methods of supervision is part of the demonstrations in several states. The Paraprofessional Healthcare Institute is offering its “Coaching Supervision Program,” a program that teaches nurses to use a nonpunitive, problem-solving approach to supervision (Paraprofessional Healthcare Institute, 2005).
That the income is less than a rewarding part of direct care work continues to impede efforts to improve both jobs and care. To address this pervasive problem, several state governments are experimenting with Medicaid reimbursement enhancement which is explicitly marked for improving direct care worker wage rates, but other responses to the income problem are clearly needed.
Organizational behaviors such as turnover are complex phenomena influenced by environmental and systemic factors (Anderson, Corazzini, & McDaniel, 2004; Barry, Brannon & Mor, 2005), if most directly driven by individual cognitions. This study's contribution to understanding turnover and retention is limited in several ways. First, although intent to leave is an important cognitive precursor to turnover behavior, it should not be viewed as a proxy for actual separation from the job. Second, the organizations from which the sample is drawn were not randomly selected and it cannot be assumed that the findings can be generalized to all long-term-care provider organizations. In addition, the job problems and rewards are measured as perceptions and thus reflect individual responses to the job rather than objective measures of job characteristics. Even though this limits, to some extent, the usefulness of the findings for prescribing best practices for work design, it does promote consideration of person–job fit as a relatively unexplored approach to enhancing the long-term-care workforce. Finally, analyses of cross-sectional data are inherently limited in what they contribute to the understanding of complex and dynamic phenomena such as the turnover process. However, results of this study suggest that management strategies such as those being tested in the BJBC demonstration are appropriately conceived in light of what this sample of direct care workers reported as problems and rewards of caregiving work in each setting. A follow-up survey of direct care workers in the BJBC demonstration sites will provide further insight into the effectiveness of a number of these workforce development strategies.
This research was supported by The Atlantic Philanthropies and The Robert Wood Johnson Foundation as part of the National Study of the Better Jobs Better Care demonstration.
Department of Health Policy and Administration, The Pennsylvania State University, University Park.
Decision Editor: Linda S. Noelker, PhD
. | . | Nursing Facility . | Home Care . | Assisted Living . | Adult Day Services . | Total . |
---|---|---|---|---|---|---|
Variable . | α . | M (SD) . | M (SD) . | M (SD) . | M (SD) . | M (SD) . |
Not at all likely to leave | 0.56 (—) | 0.61 (—) | 0.49 (—) | 0.43 (—) | 0.57 (—) | |
Very likely to leave | 0.13 (—) | 0.12 (—) | 0.20 (—) | 0.15 (—) | 0.13 (—) | |
Hazards | .74 | 2.00 (0.69) | 1.55 (0.60) | 1.83 (0.67) | 1.92 (0.78) | 1.78 (0.69) |
Dead end | .77 | 1.86 (0.69) | 1.71 (0.71) | 1.78 (0.68) | 1.96 (0.79) | 1.79 (0.70) |
Overload | .78 | 2.11 (0.60) | 1.44 (0.45) | 1.84 (0.59) | 1.81 (0.65) | 1.78 (0.62) |
Discrimination (1 vs 0) | 0.09 (—) | 0.04 (—) | 0.08 (—) | 0.07 (—) | 0.06 (—) | |
Challenge | .80 | 3.16 (0.57) | 3.36 (0.53) | 3.25 (0.57) | 3.17 (0.63) | 3.26 (0.56) |
Recognition | .70 | 3.23 (0.68) | 3.40 (0.56) | 3.20 (0.71) | 3.18 (0.65) | 3.30 (0.64) |
Helping others | .72 | 3.45 (0.53) | 3.63 (0.45) | 3.55 (0.48) | 3.45 (0.62) | 3.54 (0.50) |
Decision authority | .72 | 3.06 (0.66) | 3.38 (0.54) | 3.17 (0.60) | 3.16 (0.59) | 3.22 (0.62) |
Team spirit | .77 | 2.89 (0.69) | 3.04 (0.68) | 2.88 (0.72) | 3.04 (0.66) | 2.96 (0.69) |
Income | 2.03 (0.91) | 2.13 (0.93) | 1.78 (0.85) | 1.87 (0.78) | 2.04 (0.91) | |
Supervisor quality | .92 | 2.89 (0.76) | 3.41 (0.64) | 2.93 (0.83) | 3.12 (0.72) | 3.12 (0.77) |
Job tenure (months) | 118.33 (104.65) | 104.56 (95.09) | 98.00 (89.45) | 110.30 (97.50) | 109.45 (98.74) | |
Age (category midpoint) | 39.39 (12.96) | 46.56 (13.33) | 41.44 (13.19) | 44.39 (13.69) | 42.83 (13.58) | |
Non-White race (1 vs 0) | 0.31 (—) | 0.32 (—) | 0.50 (—) | 0.24 (—) | 0.34 (—) | |
Post-HS education (1 vs 0) | 0.08 (—) | 0.08 (—) | 0.08 (—) | 0.24 (—) | 0.08 (—) | |
Self-efficacy | .68 | 3.82 (0.39) | 3.86 (0.35) | 3.80 (0.44) | 3.64 (0.55) | 3.83 (0.39) |
Job alternative | 3.24 (0.87) | 3.43 (0.80) | 3.59 (0.65) | 3.57 (0.75) | 3.38 (0.82) |
. | . | Nursing Facility . | Home Care . | Assisted Living . | Adult Day Services . | Total . |
---|---|---|---|---|---|---|
Variable . | α . | M (SD) . | M (SD) . | M (SD) . | M (SD) . | M (SD) . |
Not at all likely to leave | 0.56 (—) | 0.61 (—) | 0.49 (—) | 0.43 (—) | 0.57 (—) | |
Very likely to leave | 0.13 (—) | 0.12 (—) | 0.20 (—) | 0.15 (—) | 0.13 (—) | |
Hazards | .74 | 2.00 (0.69) | 1.55 (0.60) | 1.83 (0.67) | 1.92 (0.78) | 1.78 (0.69) |
Dead end | .77 | 1.86 (0.69) | 1.71 (0.71) | 1.78 (0.68) | 1.96 (0.79) | 1.79 (0.70) |
Overload | .78 | 2.11 (0.60) | 1.44 (0.45) | 1.84 (0.59) | 1.81 (0.65) | 1.78 (0.62) |
Discrimination (1 vs 0) | 0.09 (—) | 0.04 (—) | 0.08 (—) | 0.07 (—) | 0.06 (—) | |
Challenge | .80 | 3.16 (0.57) | 3.36 (0.53) | 3.25 (0.57) | 3.17 (0.63) | 3.26 (0.56) |
Recognition | .70 | 3.23 (0.68) | 3.40 (0.56) | 3.20 (0.71) | 3.18 (0.65) | 3.30 (0.64) |
Helping others | .72 | 3.45 (0.53) | 3.63 (0.45) | 3.55 (0.48) | 3.45 (0.62) | 3.54 (0.50) |
Decision authority | .72 | 3.06 (0.66) | 3.38 (0.54) | 3.17 (0.60) | 3.16 (0.59) | 3.22 (0.62) |
Team spirit | .77 | 2.89 (0.69) | 3.04 (0.68) | 2.88 (0.72) | 3.04 (0.66) | 2.96 (0.69) |
Income | 2.03 (0.91) | 2.13 (0.93) | 1.78 (0.85) | 1.87 (0.78) | 2.04 (0.91) | |
Supervisor quality | .92 | 2.89 (0.76) | 3.41 (0.64) | 2.93 (0.83) | 3.12 (0.72) | 3.12 (0.77) |
Job tenure (months) | 118.33 (104.65) | 104.56 (95.09) | 98.00 (89.45) | 110.30 (97.50) | 109.45 (98.74) | |
Age (category midpoint) | 39.39 (12.96) | 46.56 (13.33) | 41.44 (13.19) | 44.39 (13.69) | 42.83 (13.58) | |
Non-White race (1 vs 0) | 0.31 (—) | 0.32 (—) | 0.50 (—) | 0.24 (—) | 0.34 (—) | |
Post-HS education (1 vs 0) | 0.08 (—) | 0.08 (—) | 0.08 (—) | 0.24 (—) | 0.08 (—) | |
Self-efficacy | .68 | 3.82 (0.39) | 3.86 (0.35) | 3.80 (0.44) | 3.64 (0.55) | 3.83 (0.39) |
Job alternative | 3.24 (0.87) | 3.43 (0.80) | 3.59 (0.65) | 3.57 (0.75) | 3.38 (0.82) |
Note: HS = high school; Total sample, N = 3,039; nursing facility, n = 1,262; home care, n = 1,306; assisted living, n = 425; adult day services, n = 46.
. | . | Nursing Facility . | Home Care . | Assisted Living . | Adult Day Services . | Total . |
---|---|---|---|---|---|---|
Variable . | α . | M (SD) . | M (SD) . | M (SD) . | M (SD) . | M (SD) . |
Not at all likely to leave | 0.56 (—) | 0.61 (—) | 0.49 (—) | 0.43 (—) | 0.57 (—) | |
Very likely to leave | 0.13 (—) | 0.12 (—) | 0.20 (—) | 0.15 (—) | 0.13 (—) | |
Hazards | .74 | 2.00 (0.69) | 1.55 (0.60) | 1.83 (0.67) | 1.92 (0.78) | 1.78 (0.69) |
Dead end | .77 | 1.86 (0.69) | 1.71 (0.71) | 1.78 (0.68) | 1.96 (0.79) | 1.79 (0.70) |
Overload | .78 | 2.11 (0.60) | 1.44 (0.45) | 1.84 (0.59) | 1.81 (0.65) | 1.78 (0.62) |
Discrimination (1 vs 0) | 0.09 (—) | 0.04 (—) | 0.08 (—) | 0.07 (—) | 0.06 (—) | |
Challenge | .80 | 3.16 (0.57) | 3.36 (0.53) | 3.25 (0.57) | 3.17 (0.63) | 3.26 (0.56) |
Recognition | .70 | 3.23 (0.68) | 3.40 (0.56) | 3.20 (0.71) | 3.18 (0.65) | 3.30 (0.64) |
Helping others | .72 | 3.45 (0.53) | 3.63 (0.45) | 3.55 (0.48) | 3.45 (0.62) | 3.54 (0.50) |
Decision authority | .72 | 3.06 (0.66) | 3.38 (0.54) | 3.17 (0.60) | 3.16 (0.59) | 3.22 (0.62) |
Team spirit | .77 | 2.89 (0.69) | 3.04 (0.68) | 2.88 (0.72) | 3.04 (0.66) | 2.96 (0.69) |
Income | 2.03 (0.91) | 2.13 (0.93) | 1.78 (0.85) | 1.87 (0.78) | 2.04 (0.91) | |
Supervisor quality | .92 | 2.89 (0.76) | 3.41 (0.64) | 2.93 (0.83) | 3.12 (0.72) | 3.12 (0.77) |
Job tenure (months) | 118.33 (104.65) | 104.56 (95.09) | 98.00 (89.45) | 110.30 (97.50) | 109.45 (98.74) | |
Age (category midpoint) | 39.39 (12.96) | 46.56 (13.33) | 41.44 (13.19) | 44.39 (13.69) | 42.83 (13.58) | |
Non-White race (1 vs 0) | 0.31 (—) | 0.32 (—) | 0.50 (—) | 0.24 (—) | 0.34 (—) | |
Post-HS education (1 vs 0) | 0.08 (—) | 0.08 (—) | 0.08 (—) | 0.24 (—) | 0.08 (—) | |
Self-efficacy | .68 | 3.82 (0.39) | 3.86 (0.35) | 3.80 (0.44) | 3.64 (0.55) | 3.83 (0.39) |
Job alternative | 3.24 (0.87) | 3.43 (0.80) | 3.59 (0.65) | 3.57 (0.75) | 3.38 (0.82) |
. | . | Nursing Facility . | Home Care . | Assisted Living . | Adult Day Services . | Total . |
---|---|---|---|---|---|---|
Variable . | α . | M (SD) . | M (SD) . | M (SD) . | M (SD) . | M (SD) . |
Not at all likely to leave | 0.56 (—) | 0.61 (—) | 0.49 (—) | 0.43 (—) | 0.57 (—) | |
Very likely to leave | 0.13 (—) | 0.12 (—) | 0.20 (—) | 0.15 (—) | 0.13 (—) | |
Hazards | .74 | 2.00 (0.69) | 1.55 (0.60) | 1.83 (0.67) | 1.92 (0.78) | 1.78 (0.69) |
Dead end | .77 | 1.86 (0.69) | 1.71 (0.71) | 1.78 (0.68) | 1.96 (0.79) | 1.79 (0.70) |
Overload | .78 | 2.11 (0.60) | 1.44 (0.45) | 1.84 (0.59) | 1.81 (0.65) | 1.78 (0.62) |
Discrimination (1 vs 0) | 0.09 (—) | 0.04 (—) | 0.08 (—) | 0.07 (—) | 0.06 (—) | |
Challenge | .80 | 3.16 (0.57) | 3.36 (0.53) | 3.25 (0.57) | 3.17 (0.63) | 3.26 (0.56) |
Recognition | .70 | 3.23 (0.68) | 3.40 (0.56) | 3.20 (0.71) | 3.18 (0.65) | 3.30 (0.64) |
Helping others | .72 | 3.45 (0.53) | 3.63 (0.45) | 3.55 (0.48) | 3.45 (0.62) | 3.54 (0.50) |
Decision authority | .72 | 3.06 (0.66) | 3.38 (0.54) | 3.17 (0.60) | 3.16 (0.59) | 3.22 (0.62) |
Team spirit | .77 | 2.89 (0.69) | 3.04 (0.68) | 2.88 (0.72) | 3.04 (0.66) | 2.96 (0.69) |
Income | 2.03 (0.91) | 2.13 (0.93) | 1.78 (0.85) | 1.87 (0.78) | 2.04 (0.91) | |
Supervisor quality | .92 | 2.89 (0.76) | 3.41 (0.64) | 2.93 (0.83) | 3.12 (0.72) | 3.12 (0.77) |
Job tenure (months) | 118.33 (104.65) | 104.56 (95.09) | 98.00 (89.45) | 110.30 (97.50) | 109.45 (98.74) | |
Age (category midpoint) | 39.39 (12.96) | 46.56 (13.33) | 41.44 (13.19) | 44.39 (13.69) | 42.83 (13.58) | |
Non-White race (1 vs 0) | 0.31 (—) | 0.32 (—) | 0.50 (—) | 0.24 (—) | 0.34 (—) | |
Post-HS education (1 vs 0) | 0.08 (—) | 0.08 (—) | 0.08 (—) | 0.24 (—) | 0.08 (—) | |
Self-efficacy | .68 | 3.82 (0.39) | 3.86 (0.35) | 3.80 (0.44) | 3.64 (0.55) | 3.83 (0.39) |
Job alternative | 3.24 (0.87) | 3.43 (0.80) | 3.59 (0.65) | 3.57 (0.75) | 3.38 (0.82) |
Note: HS = high school; Total sample, N = 3,039; nursing facility, n = 1,262; home care, n = 1,306; assisted living, n = 425; adult day services, n = 46.
Variable . | Intent . | Hazards . | Dead End . | Over-Load . | Discrimination . | Challenge . | Recognition . | Helping Others . | Decision Authority . | Team Spirit . | Income . | Supervisor quality . | Age . | Non-White . | Tenure . | College . | Self-Efficacy . | |||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Intent | ||||||||||||||||||||||||||||||||||
Hazards | .17** | |||||||||||||||||||||||||||||||||
Dead end | .27** | .55** | ||||||||||||||||||||||||||||||||
Overload | .24** | .66** | .50** | |||||||||||||||||||||||||||||||
Discrimination | .09** | .15** | .17** | .15** | ||||||||||||||||||||||||||||||
Challenge | −.18** | −.16** | −.12** | −.24** | −.05* | |||||||||||||||||||||||||||||
Recognition | −.16** | −.110** | −.10** | −.15** | −.07** | .67** | ||||||||||||||||||||||||||||
Helping others | −.20** | −.15** | −.11** | −.20** | −.07** | .68** | .59** | |||||||||||||||||||||||||||
Decision authority | −.13** | −.17** | −.08** | −.26** | −.05** | .65** | .59** | .53** | ||||||||||||||||||||||||||
Team spirit | −.22** | −.11** | −.15** | −.18** | −.08** | .58** | .59** | .46** | .49** | |||||||||||||||||||||||||
Income | −.23* | −.12** | −.19** | −.16** | −.05** | .29** | .29** | .19** | .22** | .34** | ||||||||||||||||||||||||
Supervisor quality | −.30** | −.28** | −.29** | −.41** | −.16** | .29** | .30** | .24** | .24** | .39** | .26** | |||||||||||||||||||||||
Age | −.14** | −.12** | −.12** | −.19** | −0.02 | −0.00 | −.04* | 0.03 | .04* | 0.02 | 0.03 | .13** | ||||||||||||||||||||||
Non-white | .08** | .09** | .15** | 0.00 | .17** | −0.02 | −.10** | −.04* | 0.01 | −.06* | −.07** | −0.02 | −.07* | |||||||||||||||||||||
Tenure | −.11** | −.04* | −.04* | −0.02 | 0.00 | −0.023 | −.04* | −0.00 | −0.00 | −0.04 | 0.00 | −0.03 | .41** | −0.01 | ||||||||||||||||||||
College | .11** | .05** | .10** | 0.02 | .04* | −0.02 | −.04* | −.04* | −0.03 | −0.03 | −.05** | −0.01 | −0.00 | .04* | −.08** | |||||||||||||||||||
Self-efficacy | −.10** | −.11** | −.10** | −.13** | −.04* | .23** | .17** | .24** | .20** | .16** | .07** | .23** | 0.02 | 0.03 | .06** | −.01 | ||||||||||||||||||
Job Alternative | .20** | 0.03 | .13** | 0.00 | .04* | 0.03 | 0.00 | .05** | .05** | −0.03 | −.27** | −.07** | −.07* | .09** | −.10** | .06** | .08** |
Variable . | Intent . | Hazards . | Dead End . | Over-Load . | Discrimination . | Challenge . | Recognition . | Helping Others . | Decision Authority . | Team Spirit . | Income . | Supervisor quality . | Age . | Non-White . | Tenure . | College . | Self-Efficacy . | |||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Intent | ||||||||||||||||||||||||||||||||||
Hazards | .17** | |||||||||||||||||||||||||||||||||
Dead end | .27** | .55** | ||||||||||||||||||||||||||||||||
Overload | .24** | .66** | .50** | |||||||||||||||||||||||||||||||
Discrimination | .09** | .15** | .17** | .15** | ||||||||||||||||||||||||||||||
Challenge | −.18** | −.16** | −.12** | −.24** | −.05* | |||||||||||||||||||||||||||||
Recognition | −.16** | −.110** | −.10** | −.15** | −.07** | .67** | ||||||||||||||||||||||||||||
Helping others | −.20** | −.15** | −.11** | −.20** | −.07** | .68** | .59** | |||||||||||||||||||||||||||
Decision authority | −.13** | −.17** | −.08** | −.26** | −.05** | .65** | .59** | .53** | ||||||||||||||||||||||||||
Team spirit | −.22** | −.11** | −.15** | −.18** | −.08** | .58** | .59** | .46** | .49** | |||||||||||||||||||||||||
Income | −.23* | −.12** | −.19** | −.16** | −.05** | .29** | .29** | .19** | .22** | .34** | ||||||||||||||||||||||||
Supervisor quality | −.30** | −.28** | −.29** | −.41** | −.16** | .29** | .30** | .24** | .24** | .39** | .26** | |||||||||||||||||||||||
Age | −.14** | −.12** | −.12** | −.19** | −0.02 | −0.00 | −.04* | 0.03 | .04* | 0.02 | 0.03 | .13** | ||||||||||||||||||||||
Non-white | .08** | .09** | .15** | 0.00 | .17** | −0.02 | −.10** | −.04* | 0.01 | −.06* | −.07** | −0.02 | −.07* | |||||||||||||||||||||
Tenure | −.11** | −.04* | −.04* | −0.02 | 0.00 | −0.023 | −.04* | −0.00 | −0.00 | −0.04 | 0.00 | −0.03 | .41** | −0.01 | ||||||||||||||||||||
College | .11** | .05** | .10** | 0.02 | .04* | −0.02 | −.04* | −.04* | −0.03 | −0.03 | −.05** | −0.01 | −0.00 | .04* | −.08** | |||||||||||||||||||
Self-efficacy | −.10** | −.11** | −.10** | −.13** | −.04* | .23** | .17** | .24** | .20** | .16** | .07** | .23** | 0.02 | 0.03 | .06** | −.01 | ||||||||||||||||||
Job Alternative | .20** | 0.03 | .13** | 0.00 | .04* | 0.03 | 0.00 | .05** | .05** | −0.03 | −.27** | −.07** | −.07* | .09** | −.10** | .06** | .08** |
Variable . | Intent . | Hazards . | Dead End . | Over-Load . | Discrimination . | Challenge . | Recognition . | Helping Others . | Decision Authority . | Team Spirit . | Income . | Supervisor quality . | Age . | Non-White . | Tenure . | College . | Self-Efficacy . | |||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Intent | ||||||||||||||||||||||||||||||||||
Hazards | .17** | |||||||||||||||||||||||||||||||||
Dead end | .27** | .55** | ||||||||||||||||||||||||||||||||
Overload | .24** | .66** | .50** | |||||||||||||||||||||||||||||||
Discrimination | .09** | .15** | .17** | .15** | ||||||||||||||||||||||||||||||
Challenge | −.18** | −.16** | −.12** | −.24** | −.05* | |||||||||||||||||||||||||||||
Recognition | −.16** | −.110** | −.10** | −.15** | −.07** | .67** | ||||||||||||||||||||||||||||
Helping others | −.20** | −.15** | −.11** | −.20** | −.07** | .68** | .59** | |||||||||||||||||||||||||||
Decision authority | −.13** | −.17** | −.08** | −.26** | −.05** | .65** | .59** | .53** | ||||||||||||||||||||||||||
Team spirit | −.22** | −.11** | −.15** | −.18** | −.08** | .58** | .59** | .46** | .49** | |||||||||||||||||||||||||
Income | −.23* | −.12** | −.19** | −.16** | −.05** | .29** | .29** | .19** | .22** | .34** | ||||||||||||||||||||||||
Supervisor quality | −.30** | −.28** | −.29** | −.41** | −.16** | .29** | .30** | .24** | .24** | .39** | .26** | |||||||||||||||||||||||
Age | −.14** | −.12** | −.12** | −.19** | −0.02 | −0.00 | −.04* | 0.03 | .04* | 0.02 | 0.03 | .13** | ||||||||||||||||||||||
Non-white | .08** | .09** | .15** | 0.00 | .17** | −0.02 | −.10** | −.04* | 0.01 | −.06* | −.07** | −0.02 | −.07* | |||||||||||||||||||||
Tenure | −.11** | −.04* | −.04* | −0.02 | 0.00 | −0.023 | −.04* | −0.00 | −0.00 | −0.04 | 0.00 | −0.03 | .41** | −0.01 | ||||||||||||||||||||
College | .11** | .05** | .10** | 0.02 | .04* | −0.02 | −.04* | −.04* | −0.03 | −0.03 | −.05** | −0.01 | −0.00 | .04* | −.08** | |||||||||||||||||||
Self-efficacy | −.10** | −.11** | −.10** | −.13** | −.04* | .23** | .17** | .24** | .20** | .16** | .07** | .23** | 0.02 | 0.03 | .06** | −.01 | ||||||||||||||||||
Job Alternative | .20** | 0.03 | .13** | 0.00 | .04* | 0.03 | 0.00 | .05** | .05** | −0.03 | −.27** | −.07** | −.07* | .09** | −.10** | .06** | .08** |
Variable . | Intent . | Hazards . | Dead End . | Over-Load . | Discrimination . | Challenge . | Recognition . | Helping Others . | Decision Authority . | Team Spirit . | Income . | Supervisor quality . | Age . | Non-White . | Tenure . | College . | Self-Efficacy . | |||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Intent | ||||||||||||||||||||||||||||||||||
Hazards | .17** | |||||||||||||||||||||||||||||||||
Dead end | .27** | .55** | ||||||||||||||||||||||||||||||||
Overload | .24** | .66** | .50** | |||||||||||||||||||||||||||||||
Discrimination | .09** | .15** | .17** | .15** | ||||||||||||||||||||||||||||||
Challenge | −.18** | −.16** | −.12** | −.24** | −.05* | |||||||||||||||||||||||||||||
Recognition | −.16** | −.110** | −.10** | −.15** | −.07** | .67** | ||||||||||||||||||||||||||||
Helping others | −.20** | −.15** | −.11** | −.20** | −.07** | .68** | .59** | |||||||||||||||||||||||||||
Decision authority | −.13** | −.17** | −.08** | −.26** | −.05** | .65** | .59** | .53** | ||||||||||||||||||||||||||
Team spirit | −.22** | −.11** | −.15** | −.18** | −.08** | .58** | .59** | .46** | .49** | |||||||||||||||||||||||||
Income | −.23* | −.12** | −.19** | −.16** | −.05** | .29** | .29** | .19** | .22** | .34** | ||||||||||||||||||||||||
Supervisor quality | −.30** | −.28** | −.29** | −.41** | −.16** | .29** | .30** | .24** | .24** | .39** | .26** | |||||||||||||||||||||||
Age | −.14** | −.12** | −.12** | −.19** | −0.02 | −0.00 | −.04* | 0.03 | .04* | 0.02 | 0.03 | .13** | ||||||||||||||||||||||
Non-white | .08** | .09** | .15** | 0.00 | .17** | −0.02 | −.10** | −.04* | 0.01 | −.06* | −.07** | −0.02 | −.07* | |||||||||||||||||||||
Tenure | −.11** | −.04* | −.04* | −0.02 | 0.00 | −0.023 | −.04* | −0.00 | −0.00 | −0.04 | 0.00 | −0.03 | .41** | −0.01 | ||||||||||||||||||||
College | .11** | .05** | .10** | 0.02 | .04* | −0.02 | −.04* | −.04* | −0.03 | −0.03 | −.05** | −0.01 | −0.00 | .04* | −.08** | |||||||||||||||||||
Self-efficacy | −.10** | −.11** | −.10** | −.13** | −.04* | .23** | .17** | .24** | .20** | .16** | .07** | .23** | 0.02 | 0.03 | .06** | −.01 | ||||||||||||||||||
Job Alternative | .20** | 0.03 | .13** | 0.00 | .04* | 0.03 | 0.00 | .05** | .05** | −0.03 | −.27** | −.07** | −.07* | .09** | −.10** | .06** | .08** |
. | SNF Sample . | . | HCA Sample . | . | ALF Sample . | . | Total Sample . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Variables . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | ||||||||
Job problems | ||||||||||||||||
Hazards | 1.05 | 0.99 | 1.20 | 0.69 | 1.04 | 0.88 | 1.11 | 0.85 | ||||||||
Dead-end job | 0.84 | 1.07 | 0.59*** | 2.55* | 0.56* | 0.89 | 0.67*** | 1.18 | ||||||||
Overload | 0.50*** | 1.11 | 0.89 | 1.16 | 1.03 | 2.17* | 0.77* | 1.18 | ||||||||
Discrimination | 0.99 | 1.22 | 0.68 | 1.17 | 1.47 | 0.75 | 1.02 | 1.15 | ||||||||
Job rewards | ||||||||||||||||
Challenge | 1.12 | 1.00 | 1.36 | 1.33 | 0.83 | 2.48** | 1.18 | 1.33 | ||||||||
Recognition | 1.02 | 1.33 | 1.02 | 1.22 | 1.18 | 0.94 | 1.01 | 1.18 | ||||||||
Helping others | 1.49* | .62* | 1.47* | 0.81 | 1.40 | 0.43* | 1.45** | 0.65** | ||||||||
Decision authority | 1.10 | 1.13 | 0.86 | 1.04 | 1.03 | 0.90 | 0.94 | 1.05 | ||||||||
Team spirit | 1.18 | 1.15 | 1.30 | 0.93 | 1.93* | 1.19 | 1.33** | 1.05 | ||||||||
Income | 1.04 | .87 | 1.20* | 0.83 | 1.17 | 0.76 | 1.14* | 0.82* | ||||||||
Supervisory quality control | 1.28* | 0.63** | 1.56*** | 0.67*** | 1.58* | 0.77 | 1.38*** | 0.70*** | ||||||||
Total months as a DCW | 1.00** | 1.00 | 1.00 | 0.99 | 1.00 | 1.00 | 1.00** | 1.00 | ||||||||
Age category | 1.02* | 1.00 | 1.00 | 1.00 | 1.02** | 1.00 | 1.01*** | 1.00 | ||||||||
Non-White race | 0.93 | 1.13 | 0.71* | 0.78 | 1.14 | 1.29 | 0.87 | 1.00 | ||||||||
Some post-HS education | 0.75 | 2.52** | 0.68 | 1.54 | 0.98 | 2.23 | 0.81 | 2.00*** | ||||||||
Self-efficacy | 0.85 | 1.23 | 1.18 | 1.00 | 1.69 | .64 | 1.10 | 0.94 | ||||||||
Job alternative | 0.68*** | 1.39* | 0.74** | 1.07 | 0.54** | 2.60* | 0.68*** | 1.40** |
. | SNF Sample . | . | HCA Sample . | . | ALF Sample . | . | Total Sample . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Variables . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | ||||||||
Job problems | ||||||||||||||||
Hazards | 1.05 | 0.99 | 1.20 | 0.69 | 1.04 | 0.88 | 1.11 | 0.85 | ||||||||
Dead-end job | 0.84 | 1.07 | 0.59*** | 2.55* | 0.56* | 0.89 | 0.67*** | 1.18 | ||||||||
Overload | 0.50*** | 1.11 | 0.89 | 1.16 | 1.03 | 2.17* | 0.77* | 1.18 | ||||||||
Discrimination | 0.99 | 1.22 | 0.68 | 1.17 | 1.47 | 0.75 | 1.02 | 1.15 | ||||||||
Job rewards | ||||||||||||||||
Challenge | 1.12 | 1.00 | 1.36 | 1.33 | 0.83 | 2.48** | 1.18 | 1.33 | ||||||||
Recognition | 1.02 | 1.33 | 1.02 | 1.22 | 1.18 | 0.94 | 1.01 | 1.18 | ||||||||
Helping others | 1.49* | .62* | 1.47* | 0.81 | 1.40 | 0.43* | 1.45** | 0.65** | ||||||||
Decision authority | 1.10 | 1.13 | 0.86 | 1.04 | 1.03 | 0.90 | 0.94 | 1.05 | ||||||||
Team spirit | 1.18 | 1.15 | 1.30 | 0.93 | 1.93* | 1.19 | 1.33** | 1.05 | ||||||||
Income | 1.04 | .87 | 1.20* | 0.83 | 1.17 | 0.76 | 1.14* | 0.82* | ||||||||
Supervisory quality control | 1.28* | 0.63** | 1.56*** | 0.67*** | 1.58* | 0.77 | 1.38*** | 0.70*** | ||||||||
Total months as a DCW | 1.00** | 1.00 | 1.00 | 0.99 | 1.00 | 1.00 | 1.00** | 1.00 | ||||||||
Age category | 1.02* | 1.00 | 1.00 | 1.00 | 1.02** | 1.00 | 1.01*** | 1.00 | ||||||||
Non-White race | 0.93 | 1.13 | 0.71* | 0.78 | 1.14 | 1.29 | 0.87 | 1.00 | ||||||||
Some post-HS education | 0.75 | 2.52** | 0.68 | 1.54 | 0.98 | 2.23 | 0.81 | 2.00*** | ||||||||
Self-efficacy | 0.85 | 1.23 | 1.18 | 1.00 | 1.69 | .64 | 1.10 | 0.94 | ||||||||
Job alternative | 0.68*** | 1.39* | 0.74** | 1.07 | 0.54** | 2.60* | 0.68*** | 1.40** |
Notes: Somewhat likely to leave is the referent category. DCW = direct care worker; SNF = skilled nursing facility; HCA = home care agency; ALF = assisted living facility. For the SNF sample, n = 1,262; HCA, n = 1,306; ALF, n = 425; total, n = 3.039 (total DCW sample includes adult day service, which was too small of a sample for us to run independently).
*p ≤.06; **p ≤.01; ***p ≤.001.
. | SNF Sample . | . | HCA Sample . | . | ALF Sample . | . | Total Sample . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Variables . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | ||||||||
Job problems | ||||||||||||||||
Hazards | 1.05 | 0.99 | 1.20 | 0.69 | 1.04 | 0.88 | 1.11 | 0.85 | ||||||||
Dead-end job | 0.84 | 1.07 | 0.59*** | 2.55* | 0.56* | 0.89 | 0.67*** | 1.18 | ||||||||
Overload | 0.50*** | 1.11 | 0.89 | 1.16 | 1.03 | 2.17* | 0.77* | 1.18 | ||||||||
Discrimination | 0.99 | 1.22 | 0.68 | 1.17 | 1.47 | 0.75 | 1.02 | 1.15 | ||||||||
Job rewards | ||||||||||||||||
Challenge | 1.12 | 1.00 | 1.36 | 1.33 | 0.83 | 2.48** | 1.18 | 1.33 | ||||||||
Recognition | 1.02 | 1.33 | 1.02 | 1.22 | 1.18 | 0.94 | 1.01 | 1.18 | ||||||||
Helping others | 1.49* | .62* | 1.47* | 0.81 | 1.40 | 0.43* | 1.45** | 0.65** | ||||||||
Decision authority | 1.10 | 1.13 | 0.86 | 1.04 | 1.03 | 0.90 | 0.94 | 1.05 | ||||||||
Team spirit | 1.18 | 1.15 | 1.30 | 0.93 | 1.93* | 1.19 | 1.33** | 1.05 | ||||||||
Income | 1.04 | .87 | 1.20* | 0.83 | 1.17 | 0.76 | 1.14* | 0.82* | ||||||||
Supervisory quality control | 1.28* | 0.63** | 1.56*** | 0.67*** | 1.58* | 0.77 | 1.38*** | 0.70*** | ||||||||
Total months as a DCW | 1.00** | 1.00 | 1.00 | 0.99 | 1.00 | 1.00 | 1.00** | 1.00 | ||||||||
Age category | 1.02* | 1.00 | 1.00 | 1.00 | 1.02** | 1.00 | 1.01*** | 1.00 | ||||||||
Non-White race | 0.93 | 1.13 | 0.71* | 0.78 | 1.14 | 1.29 | 0.87 | 1.00 | ||||||||
Some post-HS education | 0.75 | 2.52** | 0.68 | 1.54 | 0.98 | 2.23 | 0.81 | 2.00*** | ||||||||
Self-efficacy | 0.85 | 1.23 | 1.18 | 1.00 | 1.69 | .64 | 1.10 | 0.94 | ||||||||
Job alternative | 0.68*** | 1.39* | 0.74** | 1.07 | 0.54** | 2.60* | 0.68*** | 1.40** |
. | SNF Sample . | . | HCA Sample . | . | ALF Sample . | . | Total Sample . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Variables . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | Not at all Likely . | Very Likely . | ||||||||
Job problems | ||||||||||||||||
Hazards | 1.05 | 0.99 | 1.20 | 0.69 | 1.04 | 0.88 | 1.11 | 0.85 | ||||||||
Dead-end job | 0.84 | 1.07 | 0.59*** | 2.55* | 0.56* | 0.89 | 0.67*** | 1.18 | ||||||||
Overload | 0.50*** | 1.11 | 0.89 | 1.16 | 1.03 | 2.17* | 0.77* | 1.18 | ||||||||
Discrimination | 0.99 | 1.22 | 0.68 | 1.17 | 1.47 | 0.75 | 1.02 | 1.15 | ||||||||
Job rewards | ||||||||||||||||
Challenge | 1.12 | 1.00 | 1.36 | 1.33 | 0.83 | 2.48** | 1.18 | 1.33 | ||||||||
Recognition | 1.02 | 1.33 | 1.02 | 1.22 | 1.18 | 0.94 | 1.01 | 1.18 | ||||||||
Helping others | 1.49* | .62* | 1.47* | 0.81 | 1.40 | 0.43* | 1.45** | 0.65** | ||||||||
Decision authority | 1.10 | 1.13 | 0.86 | 1.04 | 1.03 | 0.90 | 0.94 | 1.05 | ||||||||
Team spirit | 1.18 | 1.15 | 1.30 | 0.93 | 1.93* | 1.19 | 1.33** | 1.05 | ||||||||
Income | 1.04 | .87 | 1.20* | 0.83 | 1.17 | 0.76 | 1.14* | 0.82* | ||||||||
Supervisory quality control | 1.28* | 0.63** | 1.56*** | 0.67*** | 1.58* | 0.77 | 1.38*** | 0.70*** | ||||||||
Total months as a DCW | 1.00** | 1.00 | 1.00 | 0.99 | 1.00 | 1.00 | 1.00** | 1.00 | ||||||||
Age category | 1.02* | 1.00 | 1.00 | 1.00 | 1.02** | 1.00 | 1.01*** | 1.00 | ||||||||
Non-White race | 0.93 | 1.13 | 0.71* | 0.78 | 1.14 | 1.29 | 0.87 | 1.00 | ||||||||
Some post-HS education | 0.75 | 2.52** | 0.68 | 1.54 | 0.98 | 2.23 | 0.81 | 2.00*** | ||||||||
Self-efficacy | 0.85 | 1.23 | 1.18 | 1.00 | 1.69 | .64 | 1.10 | 0.94 | ||||||||
Job alternative | 0.68*** | 1.39* | 0.74** | 1.07 | 0.54** | 2.60* | 0.68*** | 1.40** |
Notes: Somewhat likely to leave is the referent category. DCW = direct care worker; SNF = skilled nursing facility; HCA = home care agency; ALF = assisted living facility. For the SNF sample, n = 1,262; HCA, n = 1,306; ALF, n = 425; total, n = 3.039 (total DCW sample includes adult day service, which was too small of a sample for us to run independently).
*p ≤.06; **p ≤.01; ***p ≤.001.
References
American Health Care Association. (
Anderson, R. A., Corazzine, K. N., & McDaniel, R. R., (
Banaszak-Holl, J., & Hines, M., (
Barry, T. T., Brannon, D., & Mor, V., (
Bowers, B., Esmond, S., & Jacobson, N., (
Brannon, D., Barry, T., Angelelli, J., & Weech-Maldonado, R., (
Brannon, D., Zinn, J., Mor, V., & Davis, J., (
Burgio, L. D., Fisher, S. E., Fairchild, J. K., Scilley, K., & Hardin, J. M., (
Cammann, C., Fichman, M., Jenkins, D., & Klesh, J. R., (
Castle, N., & Engberg, J., (
Castle, N. G., Anderson, R., & Engberg, J., in press. Job satisfaction of nurse aides in nursing homes: Intent-to-leave and turnover. The Gerontologist..
Dansereau, F., Graen, G., & Haga, W., (
Eisenberger, R., Stinglhamber, F., Vandeneberghe, C., Sucharski, I. L., & Rhoades, L., (
Gerstner, C. R., & Day, D. V., (
Griffeth, R. W., Hom, P. W., & Gaertner, S., (
Harmuth, S., & Dyson, S., (
Harris, K. J., James, M., & Boonthanom, R., (
Harris-Kojetin, L, Lipson, D., Fielding, J., Kiefer, K., & Stone, R., (
Karasek, R. A., (
Karasek, R. A., (
Karasek, R. A., Triantis, K., & Chaudry, S., (
Kiefer, K. M., Harris-Kojetin, L., Brannon, D., Barry, T., Vasey, J., & Lepore, M., (
Marshall, N., Barnett, R., Baruch, G., & Pleck, J., (
Marshall, N., & Barnett, R. C., (
Montgomery, R. J., Holley, L. P, Deichert, J., & Kosloski, K., (
National Nursing Home Survey. (
Paraprofessional Health Care Institute. (
Parsons, S., Simmons, W., Penn, K., & Furlough, M., (
Tett, R. P., & Meyer, J. P., (
Van Breukelen, W., Schyns, B., & LeBlanc, P., (