Research on staff supporting individuals with intellectual and developmental disabilities (IDD) tends to focus on negative aspects of the work. This study expanded on previous research on the positive consequences that work in the IDD field has on staff using a brief version of the Staff Positive Contributions Questionnaire with 926 staff. Factor analysis suggested two factors: General positive contributions and Positive work motivation. Positive work motivation was associated with high levels of personal accomplishment, but shared limited variance with the other two burnout dimensions (emotional exhaustion, depersonalization). Findings lend support to the idea that we need to consider both positive and negative aspects of work life. This brief scale may be a useful index of how staff benefit from their work.
Significant attention has been given to developing a strong workforce of committed direct care staff in the developmental disabilities sector in recent years (Hewitt & Larson, 2007). However, relatively little research attention has focused on direct care staff, particularly in North America, which is concerning given the critical role staff play in promoting the well being of those they support (Hastings, 2010). The research that has been done has focused primarily on problematic aspects of working in the field with the topics of staff stress and job burnout being most frequently studied (see Devereux, Hastings, & Noone, 2009; Skirrow & Hatton, 2007 for reviews).
Job burnout has popularly been conceptualized as having three components: emotional exhaustion, depersonalization, and reduced sense of personal accomplishment which develop in settings of ongoing high emotional demands at work (Maslach et al., 2001). Emotional exhaustion represents an emotional drain and accompanying fatigue leading to the feeling that there is little left to give at a psychological level. Depersonalization refers to a state of detachment and cynicism towards service recipients. Finally, a reduced sense of personal accomplishment occurs when one loses the personal reward and achievement experienced from successes at work. Burnout is most commonly measured in this sector by the Maslach Burnout Inventory–Human Services Survey (MBI-HSS; Chao et al., 2011) which consists of three subscales (emotional exhaustion, depersonalization, and personal accomplishment) representing the three theoretical dimensions.
Far fewer studies have considered the positive aspects of work in the IDD field. Two recent commentaries have argued the importance of considering positive as well as negative work experiences (Hastings, 2010; Rose, 2011). Hastings (2010) has discussed the importance of considering the bi-directionality of relationships that form between support staff and their service recipients and the positive benefits that staff derive from these relationships. Rose (2011), on the other hand, reviewed how staff psychological factors can impact service outcomes. He suggested that staff cognitions and attitudes, both positive and negative, can play a part in both staff outcomes and results for the people supported.
Hastings and Horne (2004) published what were probably the first quantitative data on positive perceptions held by support staff in community services. They defined positive contributions as the caring gratification that may arise from the positive characteristics of the person being supported and give rise to positive caregiver outcomes such as life satisfaction. They modified the Positive Contributions Scale component of the Kansas Inventory of Parental Perceptions (Behr, Murphy, & Summers, 1992). This scale focused on the different ways children with various disabilities contributed positively to the life of their parents. Hastings and Horne (2004) maintained the format and response scale but modified 43 of the original 50 items from the parent-based scale to be appropriate for paid staff. They then piloted this scale with 32 staff working in community agencies. The removed items were not considered relevant to non-parents. Based on staff feedback on the 43-item measure, a 41 item version of this modified scale (henceforth referred to as the Staff Positive Contributions Questionnaire, or SPCQ) was completed by 105 staff from four community-based residential staff in the United Kingdom. The scale had high internal consistency and test-retest reliability and it was correlated positively, albeit weakly with the personal accomplishment scale of the MBI-HSS (in contrast to the other dimensions, higher scores on this scale suggest higher personal accomplishment and are less indicative of burnout), but was not associated with the other aspects of burnout (emotional exhaustion and depersonalization) or staff mental health problems. They concluded from this study, as Lawton argued previously with parents (Lawton, Moss, Kleblan, & Glickman, 1991), that the positive and negative outcomes of caregiving are somewhat independent of one another and that both should be studied.
In an era of diminishing resources and high staff turnover, community agencies are interested in recruiting and supporting staff who enjoy their work, who perform well, and who foster a good quality of life for those they support (Hewitt & Larson, 2007; Larson, Hewitt, & Anderson, 1999). Agencies are also interested in efficient ways to measure how staff feel about their work and predictors of work satisfaction, particularly if this can lead to interventions that enhance satisfaction and actively reduce staff stress (Hastings & Horne, 2004). Since its publication in 2004, no studies appear to have adopted or adapted the SPCQ even though it is one of the few measures that focuses on positive dimensions of the staff experience. One explanation for its limited use might be its length (41 items).
The purpose of our study, therefore, like the original Hastings and Horne (2004) research, was two-fold. First, we wanted to examine the factor structure of a briefer version of the SPCQ and describe staff responses to this measure. Next, we wanted to examine whether there was an association between positive contributions and personal accomplishment, along with other burnout dimensions. Based on the original study, we hypothesized that positive contributions would predict personal accomplishment, but not emotional exhaustion or depersonalization.
A cross-sectional survey was conducted with 926 direct staff who provide community services to adults with developmental disabilities in Ontario, Canada. The demographic and employment characteristics of the sample are summarized in Table 1. Further details on the study sample are reported elsewhere (Hensel, Lunsky, & Dewa, 2012).
The Staff Positive Contributions Questionnaire included information on age, gender, marital status, and years of experience in the field, along with a description of the types of services the staff member provided.
The Staff Positive Contributions Scale—Brief Form (SPCQ-BF) was derived from the 41-item scale described by Hastings and Horne (2004), using the 11 items with the highest item-total score correlations from the 2004 dataset. These 11 items were selected on a pragmatic basis—having strong associations with the total SPCQ score whilst reducing the scale to a number of items that could be completed quickly (in 2–3 min). The brief version of the scale reported here consisted of 11 items, with each item scored on a 4-point Likert scale ranging from strongly disagree to strongly agree. Each item is worded as a statement for potential endorsement by staff. For example, “I consider working with people with developmental disabilities to be the reason I am able to cope better with stress and problems.”
The MBI-HSS (Maslach, Jackson, & Leiter, 1996) consists of 22 items addressing feelings that human service workers have towards their work and the people that they support. Each item is rated for frequency on a 6-point Likert scale (never to every day). The MBI-HSS has 3 subscale scores: emotional exhaustion (9 items, e.g., “I feel emotionally drained from my work”), depersonalization (5 items, e.g., “I treat some clients as if they were impersonal objects”), and personal accomplishment (8 items, e.g., “I have accomplished many worthwhile things in this job.”). Note that the personal accomplishment scale is interpreted in the opposite direction, with lower scores more suggestive of burnout. Chao et al. (2011) and Hastings et al. (2004) conducted factor analyses on the MBI-HSS with staff in the developmental disabilities sector and found that it is both reliable and valid, although there has been some critique of the lower reliability of the depersonalization subscale.
Requests to participate were circulated across the province of Ontario through the three provincial direct staff unions and over 150 community agencies. All of the agencies provided community developmental disability services, including residential services (group homes and supported independent living), vocational supports, and other types of day programs. Community agencies and networks received a standardized information letter explaining the project and inviting them to participate in the study. When an agency agreed, additional details were provided for dissemination to their staff including the survey as an online link or printable version. Agencies or individual staff could e-mail their responses, submit them through the web or mail them back to the study staff. Staff participation in the study was voluntary and anonymous. Ethics approval for this study was granted by the Hospital's Research Ethics Board.
Approach to Statistical Analysis
First, as a preliminary step, an exploratory factor analysis was conducted using polychoric correlations (suitable for ordinal data) and a geomin factor rotation. Geomin is an oblique factor rotation method, suitable for identifying simple structure in a solution when the extracted factors are expected to be correlated. At the second step of analysis, our confirmatory factor analysis was carried out using Mplus software. The dependent variables in this investigation were treated as ordered categorical variables and a robust weighted least squares estimator with polychoric correlations was used. This corresponds to a series of probit regressions in which the relationship between the individual items and latent factors can be explicitly tested. Overall goodness of fit statistics (RMSEA, CFI, and TLI), modification indices, residual matrices, and clinical utility and interpretability were used to further refine our preliminary model to develop an objective and stable factor solution (see Schmitt, 2011 for review on factor analysis methods).
The third step of analysis involved descriptive statistics on the final scale. This final scale was then examined in terms of its association with burnout as measured on the MBI-HSS. The association between positive contributions and personal accomplishment was examined through hierarchical linear regression, adjusting for key demographic/background variables: age, degree/diploma, gender, marital status, and years of experience. Similar regression analyses were conducted with positive contributions predicting emotional exhaustion and depersonalization. Given the number of comparisons, and to minimize Type I errors, associations were treated as statistically significant only at the probability level of 1% (Perneger, 1998).
The exploratory factor analysis suggested a two-factor solution based on eigenvalues, scree plot, number of items per factor, and factor interpretability which was further refined and objectively tested using a confirmatory factor analysis approach. This confirmatory factor analysis solution is presented in Figure 1. This model suggests a two-factor solution to the SPCQ-BF, with five items loading on the first factor and three items loading on the second. The remaining three items did not load significantly on to either of the two identified factors and were subsequently removed. (These items were “I consider working with people with developmental disabilities to be a career in which affection is given and received”; “Because of working with people with developmental disabilities, our staff group has become closer”; and “Because of working with people with developmental disabilities, I have many unexpected pleasures.”) The two factor model fitted the data well (RMSEA = 0.038, CFI = 0.998, TLI = 0.997) and provided a clinically interpretable solution. Based on the confirmatory analysis, we finalized a 2-subscale measure: General positive contributions (5 items) and Positive work motivation (3 items). Cronbach's alpha for General positive contributions was .828 and Positive work motivation was .875. Table 2 presents the final items and the percentage of staff who endorsed each item. Table 3 presents initial norms for these two subscales of the SPCQ-BF. These norms have been produced by generating a total sum of the scores for the individual items contributing to each subscale.
Association Between Positive Contributions and Burnout
We expected to find a significant positive correlation between the two SPCQ-BF subscales and the personal accomplishment subscale of the MBI-HSS, and this hypothesis was supported (General positive contributions: r (740) = .186, p < .001; Positive work motivation: r (751) = .294, p < .001). We were interested in knowing whether these two constructs (positive contributions and personal accomplishment) remained positively associated when we controlled for other background factors. We examined this through a hierarchical linear regression where the first step included demographic variables, the second step introduced SPCQ-BF subscale scores, and the criterion variable was personal accomplishment. We then repeated this analysis for the other two dimensions of burnout (emotional exhaustion and depersonalization) as a comparison.
As indicated in Table 4, the model including only demographic variables was not significant, accounting for less than 1% of the variance in personal accomplishment scores. When the two factors of the SPCQ-BF were added to the model, the model was statistically significant, F(7, 708) = 12.02, p < .001), with a significant change in R square of 9.8%. The only significant independent predictor of personal accomplishment was the 3-item subscale of Positive work motivation; those reporting greater Positive work motivation also reported higher rates of personal accomplishment.
For both emotional exhaustion and depersonalization, demographic variables alone (first step) accounted for a significant proportion of total variance. In addition, in the final models including the SPCQ-BF scores, Positive work motivation was also significantly associated with these two burnout subscales. However, in contrast to its relationship with personal accomplishment, the shared variance between the SPCQ-BF and the other aspects of burnout was lower. Significant predictors of emotional exhaustion included age, years of experience, and Positive work motivation. Significant predictors of depersonalization included age, gender, and Positive work motivation. For each of the three burnout components (emotional exhaustion, depersonalization, and personal accomplishment), the other SPCQ-BF subscale, General positive contributions was not a significant predictor.
This study explored the factor structure of the Staff Positive Contributions Questionnaire–Brief Form (SPCQ-BF), a short staff survey focused on positive aspects of work within a large Canadian sample of community staff. These aspects of work have been regarded as equally important to the commonly investigated negative aspects of work and require greater research attention (Hastings, 2010). Factor analysis supports two subscales of the SPCQ-BF which encompass General positive contributions and Positive work motivation.
This is the first study to address the underlying factor structure of positive contributions in staff working in the Intellectual Disability (ID) field. The original Hastings and Horne study (2004) treated positive contributions as a single dimension, but this study suggests that there are two subdomains which have unique patterns of associations with other variables. Only one subdomain (Positive work motivation) is independently associated with personal accomplishment. Future research should explore whether other variables not included in this study are more strongly associated with the other subdomain (General positive contributions) such as support worker skills, empathy, or the ability to cope in times of stress.
As was reported by Hastings and Horne in their original study (2004), positive contributions as measured here accounts for significant variance in personal accomplishment, one of three components of burnout measured with the MBI-HSS. Although both subscales of the SPCQ-BF correlated significantly with personal accomplishment, only Positive work motivation was independently associated after controlling for other factors in the regression analyses. The high intercorrelation between the two subscales should be noted (r = 0.71), and the stronger bivariate correlation between Positive work motivation and the outcome may be masking the overlapping contribution of General positive contributions. Positive contributions are conceptually related to personal accomplishment and so the new SPCQ-BF may be a practically useful and brief method that could be used to assess support staff well-being or work-related satisfaction. As a note of caution, although we found statistically significant and independent (after controlling for demographic variables and General positive contributions scores) associations between Positive work motivation and all three aspects of staff burnout, these associations were small and accounted for only small proportions of variance particularly in two of the staff burnout dimensions which were more related to demographic variables. Perhaps most important, the regression results offer some validation to the concept of staff positive contributions. The SPCQ-BF scores were associated most strongly with the most closely related construct within burnout (personal accomplishment). In contrast, the negative associations with the other dimensions of burnout (emotional exhaustion and depersonalization) were likely statistically significant because of the large sample. These latter associations have very little clinical meaning because the associated relevant effect sizes (variance explained) are small. Thus, our data provide further support for the notion that positive and negative impacts of working with people with IDD are somewhat independent of each other and highlight the need to measure both when studying staff.
It is important for us to understand further the theory behind how positive contributions may contribute to staff well-being at work. Neither the current data nor those reported by Hastings and Horne (2004) suggest that positive contributions have a compensatory function by directly reducing work stress. There remains an intriguing possibility that positive contributions may serve a moderator function. Thus, in the face of high levels of work demands/stressors those staff who also maintain more positive perceptions may be less affected (i.e., work demands may impact much less on work-related well-being such as burnout). At present, we are merely hypothesizing that positive contributions serve an important function for care staff. Further research is needed and the development of the SPCQ-BF may contribute to that research effort.
The brief form of the SPCQ takes fewer than 5 minutes to complete and with further research, might also have some utility with regard to recruitment, and ongoing quality improvement initiatives, particularly the three item Positive work motivation subscale. For example, it might be a helpful index when recruiting staff, if an organization's goal is to hire staff who are more receptive to how this type of work positively impacts their own well being. Staff who do not see the work having that broader impact may not be as suited to working in this field. It could also function as a gauge of how content staff are on the job. When staff report on the job that the positive contributions they experience are low, management could investigate what contributes to this situation: (a) client variables (significant behavioral concerns, interpersonal skills), (b) organizational factors (amount of required documentation, staff: client ratio, proportion of time allocated to direct client interaction), or (c) the interaction between the staff personality and the client or organization.
It might also be important to track staff sense of positive contributions combined with burnout. Staff who report a reduction in positive contributions over time would be a particularly important population to identify. This reduction might reflect the presence of burnout or be a precursor to burnout, requiring intervention to prevent job absenteeism or turnover. A third use of this measure might be as an indicator of success of organizational initiatives targeted to increase staff satisfaction or retention (e.g., following staff training or staff appreciation initiatives). At the very least, encouraging discussions between employees and employers about how staff benefit from the work that they do might be particularly important. We provide preliminary norms for the SPCQ-BF here and future research could compare these norms with norms from other staff groups. It is equally important to examine whether staff who report more positive contributions of/from their work demonstrate better quality of care provision, based on objective measures.
This study has a number of limitations which should be considered in interpreting results. First of all, only staff interested in volunteering to be part of the study completed this measure and it is possible that staff who opted not to participate would have different experiences. With regard to the factor structure, we found a clear conceptual and meaningful distinction between general positive contributions of caring for people with developmental disabilities and perceptions of positive work motivation. A potentially confounding factor is that the item stems for the questions within each of these two factors were different. For General positive contributions, the items began with either: “Working with people with developmental disabilities is…” or “I consider working with people with developmental disabilities to be….” The three items under Positive work motivation began with: “The presence of people with developmental disabilities….” It is impossible to establish in the present study whether this different presentation of the items has driven the results of the factor analysis or if the meaning of the items as we have interpreted them has led to the factor structure. The scale is still likely to be useful for research and practice purposes, but this limitation should be born in mind. Future administration of the scale could change the order of item presentation.
This study demonstrates that it is important to measure how care work with people with ID positively impacts staff, rather than just focusing on negative outcomes in the workplace. This is because the absence of feelings of burnout is not the only indicator of staff outcome and in fact may be a poor indicator of how happy staff are in their work. If indeed our goal as organizations is to recruit and retain satisfied, empathic, and responsible staff, we need to start measuring the positively focused variables associated with the multidimensional construct of work satisfaction. This type of measurement can occur at the hiring stage and on an ongoing basis, and should inform us about which staff would be most successful in this line of work and when interventions are required to support staff better. The SPCQ-BF is a brief measure and seems to work well in this regard. Further research using measures such as the SPCQ-BF is required so we can better understand what contributes to work satisfaction and how to maintain this satisfaction over time.
The authors would like to acknowledge the direct support staff members from the province of Ontario who responded to the survey used in this study. We would also like to extend thanks to the agencies and individuals from the Ministry of Community and Social Services of Ontario who supported us in these research efforts. Dr. Dewa's Applied CHIR/PHAC Chair supported this work.
Yona Lunsky, Centre for Addiction and Mental Health Dual Diagnosis Program, Toronto, Ontario Canada; Richard Patrick Hastings, University of Warwick, UK; Jennifer Hensel, Tamara Arenovich, and Carolyn S. Dewa, Centre for Addiction and Mental Health, Toronto, Ontario Canada.
Address correspondence concerning this article to Yona Lunsky, Centre for Addiction and Mental Health Dual Diagnosis Program, 501 Queen Street West, Toronto, Ontario M5V 2B4 Canada (e-mail: firstname.lastname@example.org).