Printer Friendly

A four factor model predicting likelihood of foster carer retention Anne Eaton and Marie Caltabiano.

Over the past 20 years the number of children coming into care in Australia has been steadily increasing (Commission for Children and Young People, 2003). In Queensland alone, between 2002 and 2004 the number of children in care rose from 3,287 to 4,813, an increase of 35.5% in the two-year period (Department of Child Safety, 2004a). By June 2008, 7,134 Queensland children were living in out-of-home placements (Department of Child Safety, 2008). Other states also have high numbers of children in care. For example, NSW Department of Community Services (2008) reported a total of 14,244 children and young people in care at the end of March 2008, an increase of 1,984 from 12 months prior. In contrast to the increasing numbers of children in care, the number of foster carers approved to provide care has not kept pace with the increase in the number of children requiring care, resulting in a chronic shortage of foster carers (CMC, 2004; Department of Child Safety, 2004a).

Most foster children in home-based care are likely to have endured abuse, neglect and had multiple placements (O'Neill, 2000). Delfabrro, Barber and Cooper (2000) have reported that more than 60% of a sample group of 235 young people in care displayed behavioural problems, ranging from tantrums and unacceptable behaviours at school, to more serious problems such as stealing, vandalism, substance abuse, and/or sexualised behaviours. Continued efforts caring for very difficult, troubled children can leave caters feeling drained of energy and stressed (Wilson, Sinclair & Gibbs, 2000).

Research indicates that there are difficulties in both recruiting and retaining foster carers. Recruiting new foster carets is both time-consuming and expensive, while retention of qualified foster carets reduces the reliance on recruitment; increases the chances of young people being placed with skilled, experienced caters; and, ultimately, increases the chances of placement stability and permanence (Brown & Calder, 2000; Rhodes, Orme & Buehler, 2001). Although foster caters are coping with some of the most vulnerable and difficult children in the community, research on the effects of fostering on carets is minimal. Most foster carer studies focus on the young people being cared for, or on systematic problems with selecting, recruiting and training foster carers (Seaberg & Harrigan, 1997).

Studies in the United Kingdom (Sinclair, Gibbs & Wilson, 2004; Wilson et al., 2000) have identified a number of stressful events that can make or break foster carets. Stressful events place strain on caters beyond the normal daily hassles, mostly building up over time to the point where carets begin to feel that the negatives of fostering outweigh the positive benefits and rewards. Stress is also associated with anxiety, and psychologists are generally agreed that the root cause of anxiety are feelings of uncertainty and unpredictability (Statt, 2004). The very nature of foster care means that there is a high level of uncertainty for the carets. For example, when a young person first comes into care, assessments about the degree of problems within the child's own family are usually still being undertaken (Department of Child Safety, 2004a; Thorpe, 2004) and foster carers are often "in the dark" about factors such as: the character of the young person in care; departmental staff expectations; and the reactions of the young person's own family (Briggs & Broadhurst, 2005). The ability of foster carers to cope with unexpected and stressful events relies on how well they are prepared, and the supports and resources that they are able to draw upon (Sinclair et al., 2004; Wilson et al., 2000).

Research focusing on paid employment suggests employees who experience a lack of control experience higher levels of stress (Muhonen & Torkelson, 2004; Warr, 2002). Along with individual characteristics, job control--the "perceived ability to exert some influence over one's work environment" (Bond & Bunce, 2003 p. 1058)--not only makes a job less threatening and more rewarding, but has been credited with contributing to an individual's mental health, job performance and job satisfaction. (Bond & Bunce, 2003; Dollard, Winefield, Winefield & Jonge, 2000). Dollard et al., (2000) found that the lowest levels of worker satisfaction occurred when there was a perceived lack of control. If 'job control' plays such an important role in the paid work environment, the ability to exert control over the foster caring role would likewise appear relevant. However, no research investigating the effects of perceived control in foster caring was able to be located for inclusion in this review.

Social support networks provide families with the necessary resources for successfully coping with stressful situations (McCubbin, Thompson & McCubbin, 1996; Pearlin & Skaff, 1998). Two types of social support can be identified: formal supports that are received from government organizations, service providers or support groups, and informal support received from extended family members, neighbours, friends and co-workers. In regard to formal support, it is mandated in the Queensland Child Protection Act, 1999 and the Statement of Commitment (Queensland Government, 2000), that it is the responsibility of Child Safety Officers (CSOs) to ensure that: foster carers are supported and given training; partnerships with carers are promoted; their integral role in caring for children is recognized; and relevant information is shared with carets about children in their care (Queensland Government, 2000; 2005). In a survey of foster carers by Briggs and Broadhurst (2005), 56.3% of foster carers reported low levels of support following incidents of violence or threatening behaviour towards them by foster children or birth families.

Foster carer satisfaction has been closely linked to carers' perceived support and to retention of carers (Denby, Rindfleisch & Bean, 1999; Fisher, Gibbs, Sinclair & Wilson, 2000; Triseliotis, Sellick & Short, 1995). While only a small number of studies have addressed the issue of foster carer satisfaction, there is an abundance of research focusing on job satisfaction--"a pleasurable or positive emotional state resulting from the appraisal of one's job or job experiences" (Locke, 1976; p. 1304)--in the paid workforce. While foster carers are not paid for the 'job' that they do, they are expected to conduct themselves with increasing professionalism, work in partnerships with departmental staff, advocate for the young people in their care, and attend regular training workshops (CMC, 2004). Given the expectation that foster carers conduct themselves as if they were in the paid workforce, job satisfaction theories are expected to be relevant for foster carers.

Among other attributes, Warr's Job Content model of job satisfaction (Warr, 1987) espouses that opportunities for personal control and skill use are necessary to remain satisfied with our jobs. Given that foster carers are expected to behave in a professional manner it is logical to suggest that job control would have a similar function for foster carer satisfaction. In Locke's Value Theory, job satisfaction is an emotional state, mostly determined by how closely job rewards and outcomes match with what the worker wants, values or desires. The closer the match between the two, the higher the job satisfaction. Just as individuals bring their own beliefs and values into a workplace situation, foster carers each have their own beliefs and value systems especially in terms of how a child should be cared for.

In one Australian study involving 812 foster carers, support was rated as either 'absolutely essential' or 'very important' by 84% of the participants (Australian Foster Care Association, 2001) and another study found that less than 50% of foster carers surveyed were satisfied overall with the support they received (Colmar Brunton Social Research, 2001). Support for foster carers was also identified as a key issue at a Crime and Misconduct Commission (CMC) inquiry into abuse of children in foster care. The CMC report stated that when foster carers "are adequately supported, they are more likely to be retained in the system" (p. 207) and recommended that support services to foster carers be enhanced (CMC, 2004).

Lack of support and/or positive recognition is often raised as a prominent factor in carets choosing to quit fostering (Australian Foster Care Association, 2001; Baum, Crase & Crase, 2001; Brown & Calder, 2000; Carter, 2004; McHugh, 2002; Sinclair et al., 2004). Intention to quit has also been found to correlate with actual departure from fostering, but is tempered by a commitment to the foster children (Sinclair et al., 2004). In a large study of fostering in the UK, Sinclair and his colleagues (2004) confirmed the role that the provision of support plays in the retention of foster carers. A lack of support has also been shown to have a substantial, negative impact on foster carets especially after stressful events (Sinclair et al., 2004; Triseliotis, Borland & Hill, 1998) which can affect a carer's self-perceptions and erode their self-confidence (Sinclair et al., 2004). At these times, Sinclair et al.. (2004) suggests that effective support structures actually become a moral imperative.

Based on the research that espouses the importance of personal control in relation to job satisfaction or dissatisfaction (Dollard et al., 2000; Ernst, Franco, Messmer & Gonzalez, 2004; Muhonen & Torkelson, 2004; Warr, 2002), and the role played by support in both retention and satisfaction (Australian Foster Care Association, 2001; Baum et al., 2001; Brown & Calder, 2000; Carter, 2004; Fisher et al., 2000; McHugh 2002; Sinclair et al., 2004) as well as findings that commitment to children is related to both enjoyment of fostering and perseverance (Thorpe, 2004), the following conceptual model for determining foster caret satisfaction and likelihood of retention was proposed. Firstly, that satisfaction levels would be determined by a foster carer's perceived level of support and control; and secondly that the likelihood of a foster carer deciding to stay or leave would be determined by their level of satisfaction and whether they have made a personal commitment to a child/ren currently in their care (see Figure 1).


From the proposed model, it was hypothesized that: 1) foster carers

with high internal locus of control scores and high scores on emotional and practical support would score high on overall satisfaction; and 2) foster carers who had made a personal commitment to care for a foster child/ren in their care until they turn 18 (years), and who reported high levels of satisfaction, would be more likely to indicate a likelihood of staying in fostering in the next 18 months, than those who reported low overall satisfaction scores and who had not made a personal commitment to a particular child in their care.


A survey methodology was used to predict which foster carer families would be likely to stay in fostering in the next 18 months. A total of 185 foster carer families voluntarily participated in the research. Participants were recruited through Foster Care Queensland (FCQ) a peak body for foster carers. Participants were defined as approved person/s within a foster carer family and therefore may refer to a single individual or a couple. For example, a husband/ wife team represented only one participant response. The sample consisted of 135 male/female couples, one female couple (sisters), 47 single female carers, two single male carers, and they ranged in age between 28 and 87 years (M = 49.24, SD = 9.90). Participants represented carers from all of the Department of Child Safety's seven different zonal districts across Queensland and were almost evenly divided between Departmental supported carers (50.5% or 94 participants) and Agency supported carers (48.4% or 90 participants) with only one participant being a 'Life Without Barriers' carer, a program that offers enhanced support for carers who provide placements for young people with especially challenging behaviours (Department of Child Safety, 2004b).

Experience as foster carers ranged from nine months to 41 years (M = 9.95 years, SD = 8.73). Only seven (3.8%) participants had fostered for less than one year, and 20 (10.8%) had fostered between one and two years. The remaining majority of 158 (85.4%) participants had fostered for two or more years. Overall, 60% of the participants had been foster carers for six or more years. Based on current ages of participants and length of time fostering, foster carers with less than two years experience began at a mean age of 43 years, and carers with 20 or more years experience began at a mean age of 31. The total number of children who had been fostered by any one foster carer family ranged between one and 300 (M = 43.53, SD = 53.55). Most participants (88.5%) had at least one child of their own (M = 2.51, SD = 1.72). One hundred and twenty-nine participants (69.4%) answered 'yes' to a question about whether they had made a personal commitment to foster children currently in their care.


The main section of the questionnaire contained the Foster Carer Locus of Control Scale (FCLOC), the Foster Carer Significant Others Support Scale (FCSOS), and the Foster Carer Satisfaction Scale (FCSS) which were all adapted from similar existing measures to suit the specificity of the foster carer's situation since relevant scales for foster carers could not be located. Participants were also asked to indicate their overall level of satisfaction with fostering and their likelihood of staying fostering or leaving in the next 18 months. Qualitative data was collected from participants' responses to three open-ended questions.

The FCLOC scale is an 18-item instrument designed to assess control beliefs of foster carers. The scale was adapted from Spector's (1988) Work Locus of Control Scale (WLCS) to suit the foster carer context and is consistent with the notion that domain-specific measures are better predictors than global measures of locus of control (Lefcourt, 1992). Ratings are made using six-point Likert scales from 1 (disagree very much) to 6 (agree very much). As with Spector's WLCS, half of the items in the scale were written in each direction--external and internal. A FCLOC score is calculated by adding the response scores to each item after adjusting for negatively worded items. Total scores can range between 18 and 108 with a high score representing external control. Sample scores for foster caters ranged from 32 to 102 with a mean locus of control score for the sample of 61.36 (SD = 13.34). The FCLOC Scale had moderately high internal consistency with a Cronbach's 0t of .79. While this scale does not measure 'actual' control a foster caret may or may not have, it does measure their perception of control.

The FCSOS was adapted from the Significant Others Scale (SOS) (A) of social support (Weinman, Wright and Johnston, 1995). As with the SOS, participants were required to rate support functions from seven different individuals in terms of the actual level of support received and the ideal level of support. Ratings are made using seven-point Likert scales from 1 (never) to 7 (always). The FCSOS primarily differs from the SOS with respect to some of the individual support persons listed. For example, 'mother', 'father', 'brother or sister' were changed to 'another foster cater', 'extended family member', 'agency support person'. The words 'about fostering' were also added to the initial question relating to support. For example, "Can you trust, talk to frankly and share your feelings (about fostering) with your spouse/partner?" The scale provides an index of likely satisfaction with available support in both the emotional and practical support areas. The SOS scale has shown satisfactory validity and reliability over a six-month interval ranges from 0.73 to 0.83 across the four summary support scores (actual versus ideal x emotional versus practical) (Weinman et al., 1995).

The FCSS was based on Spector's (1985) Job Satisfaction Survey (JSS), a 36 item scale used to assess employee attitudes about their job and aspects of it. However, while the Foster Carer Satisfaction Survey retains the same number of questions, most of the nine-subscales in the JSS were changed to suit the foster care arena. For example, questions based on the subscale of 'pay' were changed to questions about 'financial support'. The content of the questions were drawn from foster care research (Baum et al., 2001; Queensland Government, 2000; Sinclair et al., 2004). As with the JSS, the FCSS contains six response choices ranging from "strongly disagree' to "strongly agree" with items written in both directions requiring reverse scoring. A total FCSS score is calculated by adding the response scores to each item after adjusting for negatively worded items. Total scores can range between 36 and 216 with the higher the overall score, the stronger the level of satisfaction. Scores for the sample of foster carers ranged from 64 to 207 with a mean scale score of 122.17 (SD = 29.83). The FCSS was found to have high internal reliability with a Cronbach's [alpha] of .90.


Prior to conducting statistical analyses, a number of procedures were carried out to check for violations of test assumptions, with no serious violations noted. The scale scores for the sample across all designated 'significant others' were: actual emotional (M = 10.1, SD = 2.0); ideal emotional (M = 12.5, SD = 1.4); discrepancy emotional (M = -2.5, SD 1.9); actual practical (M = 8.5, SD = 2.1); ideal practical (M = 11.1, SD = 1.6); discrepancy practical (M = -2.6, SD = 1.9). The negative discrepancies for both emotional and practical support areas suggest that overall participant satisfaction with support is lower than what they believe is the ideal level of support. As can be seen from Table 1, results for each of the 'significant other' discrepancy scores were all negative. The largest discrepancy between actual and ideal support was from Child Safety Officers [emotional (M =-4.71, SD = 3.57); practical (M =-4.06, SD -2.91)]. The second largest discrepancy scores were from Agency Support Persons [emotional (M = -3.27, SD = 3.48); practical (M = -3.37, SD = 3.78)]. The smallest discrepancy scores were from 'spouse/partner' supports [emotional (M =-0.41, SD = 1.31); practical (M =-1.31, SD = 1.77)]. The second smallest discrepancy scores were from 'best friends' [emotional (M =-1.49, SD = 2.37); practical (M =-1.75, SD = 2.51)1.

In relation to the Foster Carer Satisfaction Scale, a review of item scores for satisfaction indicated that, on average, participants agreed with only seven of the 36 items (M > 4). Three questions relating to the actual care of the children showed the highest level of satisfaction. Mean scores revealed that participants were dissatisfied with eleven other items (M < 3). Of these, four items were related to emotional support, one with practical support, two with training, two with policy and processes, and two concerned the trustworthiness of, and relationships with, departmental workers.

In the last 18 months 39 (31.7%) participants indicated that they had felt like giving up fostering 'often' or 'very often' and 51 (27.4%) participants indicated that they had felt like giving up in the last two months. Responses from a question asking how satisfied the foster carers felt overall with their current fostering situation showed that 99 (53.2%) participants were 'satisfied' or 'very satisfied' and 44 (23.6%) were 'unsatisfied' or 'very unsatisfied'. When asked about the likelihood of giving up fostering in the next 18 months, 37 (19.9%) participants indicated that they were either 'very likely' or 'somewhat likely' to give up; 21 (11.3%) participants indicated that they were 'unsure'; and 126 (67.7%) participants indicated that they were either 'very unlikely' or 'somewhat unlikely' to give up fostering in that period.

After removal of previously identified outliers, a multiple regression analysis was performed between Total Satisfaction as the dependent variable (DV) and Locus of Control and Perceived Support as the independent variables (IVs) (See Figure 1 for model). Bivariate correlations between the two independent variables was r = .43 which is less than the upper limit recommended by Tabachnick and Fidell (2001), therefore all variables were retained. This was supported by a high tolerance value (0.81) and a low Variance Inflation Factor (VIF) value, 1.23, indicating that the multicollinearity assumption was not violated. A check of the Mahalanobis distances revealed that there were no outliers as none of the cases exceeded the critical value of 13.82 as recommended by Tabachnick and Fidell (2001) for two IVs. Locus of control and perceived support predicted carer satisfaction [F(2,173 )= 87.24,]/) < .001, with 50% of the total variance for Satisfaction (R2 =.50) being explained by this analysis. Both IVs contributed significantly to the prediction of the DV (Satisfaction). The strongest contribution was from the Locus of Control variable, [beta].54, p < .005. The two IVs uniquely explained 32% of the variance in Satisfaction scores. The combination of the two IVs contributed another 18% in shared variability.

As with the previous analysis outliers were removed prior to conducting a Logistic Regression Analysis. The categorical variable of Likelihood to Give up (fostering) was recoded to become a dichotomous dependent variable (DV) to suit the analysis. Participant responses indicating that they were either 'very likely' or 'somewhat likely' to give up fostering in the next 18 months were recoded as '0' and labeled 'leave'; responses of 'very unlikely' or 'somewhat unlikely' were recoded as '1' and labeled 'stay'; and responses from 21 participants who indicated that they were 'unsure' were excluded from the analysis. After exclusions, data from 154 Foster Carers were available for analysis.

A goodness of fit test using the two predictors of 'Satisfaction' and 'Commitment' against the dependent variable--'Likelihood of Staying (in fostering in the next 18 months)'-was statistically reliable, [chi square] (2, N = 154)= 26.64, p < .0001, indicating that the predictors, as a set, reliably distinguished between the likelihood of staying in fostering in the next 18 months or leaving. This result is supported by the Hosmer-Lemeshow test, [chi square] (8, N = 154)= 5.605, p > .005. The pseudo R square statistics reveal that between 15.9% and 24.4% of the variability in the dependent variable is explained by this model.

In this model 20.6% of those likely to leave (specificity), and 95% of those likely to stay (sensitivity) were correctly predicted, for an overall percentage accuracy rate of 78.6% of the cases. The positive predicted value, or the percentage of cases classified by the model "as having the characteristic that is actually observed" (Pallant, 2005, p. 168), showed that 80.85% of cases were accurately predicted as likely to stay in fostering. The negative predicted value, or the percentage of cases predicted by the model not to have the characteristic that is actually observed in the participants, showed that 53.8% were accurately predicted as likely to leave fostering.

As can be seen by the Wald statistics and the beta values in Table 2, both predictor variables made a significant positive prediction to the model indicating that an increase in either satisfaction or commitment will result in foster carers being more likely to decide to stay in fostering. The values of the coefficients reveal that a 'yes' response for commitment is associated with an increase in likelihood of staying in fostering by a factor of 2.54 with a 95% C.I. (6.03, 1.07), and that each unit increase in the satisfaction score is associated with an increase in the odds of staying by a factor of 1.04 with a 95% C.I. (1.05, 1.02).


The results of this study supported the proposed model that satisfaction levels for foster carers would be determined by the amount of support and control perceived by the carer, and that the likelihood of a foster carer deciding to stay or leave is determined by their level of satisfaction and whether they have made a personal commitment to a child/ren currently in their care. The Logistic regression analysis demonstrated that an increase in either satisfaction or commitment will result in foster carers being more likely to decide to stay in fostering. However, based on the participants who indicated that they were either 'very likely' or 'somewhat likely' to give up fostering, the results have also shown that the Queensland foster care system can expect to loose up to 20% (N = 317) of their existing carers over the next 18 months if, as reported by Sinclair et al., (2004), actual departure from fostering is correlated with intention to quit. Replacing these carets will be difficult. Not only is the number of children coming into care increasing, but this study indicates that the age at which foster carers are being recruited is also increasing. This suggests that the total length of time that carers will be able to continue fostering is likely to decrease as carers age. It therefore seems that retention of existing carets is even more important now than it was 20 years ago.

Of the two predictor variables, satisfaction and commitment, commitment was shown to have the greatest impact on carets likely to stay in fostering. This finding was further supported by the qualitative data. Not only was commitment the most cited theme, but a number of foster carers actually made comments relating to commitment being the only reason for staying. For example, one cater wrote, "The children I have are the only reason I will not give up fostering." Other comments from carers reflect the strength of their commitment, "I am .. 150 % committed to the two children I have in care. Their needs are foremost in my life...". It seems that even when carers are low in satisfaction they will sometimes 'stick it out' for the sake of the children that they care for. This result supports the findings of Sinclair et al., (2004) which indicated that some carets are likely to quit fostering when a particular child leaves their care. Once again, this situation is not helpful to the already pressing need for foster carers.

While not as strong a predictor of retention as commitment, satisfaction still plays a substantial part in the likelihood of staying or leaving fostering. Parallels with this finding have been identified in both the fostering literature and in that of organizational psychology focusing on satisfaction and retention of paid workers (Denby et al., 1999; Fisher et al., 2000; Spector, 1985, 1997; Triseliotis et al., 1995). Therefore, if the services of existing foster carets are to be retained, results of this study serve to reinforce the need for relevant departments and agencies to consider the satisfaction levels of their carers. Particpant responses have clearly shown that there is scope for improved foster carer support (especially formal supports) and control, the two predictors of satisfaction.

Results from both the quantitative and qualitative data revealed that one of the most satisfying aspects of fostering for this sample was related to the actual care and enjoyment of the children themselves. Comments from one participant have aptly summed this up, "We will not be leaving fostering as we enjoy doing it and looking after children". This finding is not surprising considering the role that commitment has been shown to play in retention, and it is also consistent with previous research (Thorpe, 2004). However, satisfaction with caring for children was also tempered by low satisfaction results in other areas for some carers, for example, "Despite the massive inadequacies at the department our unconditional love and support to our foster child, and the joy she has brought to our family makes it all worthwhile".

Results from the FCCS found the lowest satisfaction in: areas of support; training; policy and practice; and relationships with departmental workers. These findings were further supported by the qualitative data, for example, "Relationship with the department workers needs VAST improvements"; and "We will be giving up fostering NEXT WEEK. Sick and tired of departmental bureaucracy.., we are treated like 'cattle' ..." (capitalisation by participant). The role of control and perceived support as a predictor of satisfaction for foster carers was shown in the multiple regression results of this study. Interestingly, results from the FCSOS revealed that formal supports such as those given by FCQ Delegates, departmental and agency workers resulted in lower satisfaction than informal supports from support people like family and friends.

The notion that locus of control and support are determinants of satisfaction for foster caters, as are those of relationships with departmental workers and respect, is consistent with Warr's Job Content Model of job satisfaction. Wart's Model states that individuals need a minimum daily dosage of nine attributes including control, support, respect, and interpersonal contact to remain satisfied with their jobs (Warr, 2002). However, Warr's content theory does not account for foster carers remaining in fostering out of commitment for children in their care even if they have low satisfaction levels. The commitment variable is more attune with Locke's value theory (Locke, 1990) which sees job satisfaction as an emotional state: the higher the match between job rewards and outcomes with a worker's wants, values or desires; the higher the job satisfaction. Both this study and previous research (Thorpe, 2004) has suggested that one of the main motivations for fostering is the 'love of children' factor. Therefore, if Locke's value theory is accurate in terms of the role played by foster carers, it follows that a satisfying outcome in relation to children could serve to be sufficient 'reward' for some carets to choose to stay in the foster care system.

Strengths and Limitations of the study

A limitation of this study was the method of access to the sample as it could be argued that members of Foster Care Queensland (FCQ) may not be representative of the entire foster carer population. However, it is suggested that foster carers who are not FCQ members might be less informed and feel less supported because they do not receive FCQ's monthly newsletters and may not be aware of the available support from FCQ delegates. Inclusion of these carers in future studies may therefore strengthen the findings rather than weaken them.

Implications and Conclusion

Findings from this research reveal a number of important issues in relation to Queensland foster carets. First, the results have demonstrated that there are at least four key factors that determine the likelihood of retention for foster carers: locus of control, support, satisfaction and commitment. Although no direct link was found between control and support with retention, these two factors were shown to be significant predictors of satisfaction. As proposed, both satisfaction and commitment were predictors of likelihood of retention. Hence, the study has demonstrated significant support for the proposed conceptual model. Existing literature about foster caret satisfaction is minimal, and what is available predominantly originates from other countries. The results of this study are useful in gaining a clearer understanding of what motivates foster carets in Queensland, Australia and, given the lack of this type of research, this study may serve as a stepping-stone for a more indepth exploration of the model proposed here. Knowledge about what motivates foster carers may be useful in determining programs and/or supportive practices beneficial to the retention of carers, thereby assisting in the alleviation of the current crisis in foster care. In turn, retention of qualifed carets decreases the need for recruitment and increases the chances of stability and permanence for young people in the care system.


Australian Foster Care Association (2001) Supporting Strong Parenting in the Australian Foster

Care Sector. Canberra: Department of Family and Community Services. Retrieved March 16, 2006, from the Foster Care Association of the ACT website: http//

Baum, A., Crase, S. & Crase, K. (2001) 'Influences on the decision to become or not become a foster parent', Families in Society, 82, 202-213.

Bond, F & Bunce, D. (2003) 'The role of acceptance and job control in mental health, job satisfaction, and work performance', Journal of Applied Psychology, 88, 1057-1067.

Briggs, E & Broadhurst, D. (2005) 'The abuse of foster carets in Australia', Journal of the Home Economics Institute of Australia, 12, 25-35.

Brown, J. & Calder, E (2000) 'Concept mapping the needs of foster parents', Child Welfare, 79, 729-746.

Carter, J. (2004). Wanted: A New Vision for Foster Care. Australian Foster Care Association Annual Conference. Canberra.

CMC (2004) Protecting Children: An Inquiry into Abuse of Children in Foster Care. Brisbane: The Crime and Misconduct Commission, Queensland.

Colmar Brunton Social Research (2001) Report on Client and Caret Survey. Prepared for Department of Families, Turner, ACT: CBSR. Retrieved February 2, 2006 from

Commission for Children and Young People (2003) Children and Young People in Queensland: a snapshot, Brisbane: Queensland Government.

Delfabbro, P., Barber, J. & Cooper, L. (2000) 'Placement disruption and dislocation in South Australian substitute care', Children Australia, 25, 16-20.

Denby, R., Rindfleisch, N. & Bean, G. (1999) 'Predictors of foster parents' satisfaction and intent to continue to foster', Child Abuse & Neglect, 23, 287-303.

Department of Child Safety (2004a) Child Protection Queensland: 2004 Child Protection System 'Baseline' Performance Report. Brisbane: Queensland Government.

Department of Child Safety (2004b) Queensland children offered better alternatives. Building Blocks.

Department of Child Safety (2005) Distinct Cater Families with Current Approvals: Departmental Zone by Approval Type Queensland, as at 30 June 2005.

Department of Child Safety (2008) Department of Child Safety Annual Report 2007-08. Queensland Government.

Dollard, M., Winefield, H., Winefield, A. & Jonge, J. (2000) ' Psychosocial job strain and productivity in human service workers: A test of the demand--control--support model', Journal of Occupational and Organizational Psychology, 73, 501-574.

Ernst, M., Franco, M., Messmer, P. & Gonzalez, J. (2004) 'Nurses' job satisfaction, stress, and recognition in a pediatric setting', Pediatric Nursing, 30, 219-228.

Fisher, T., Gibbs, I., Sinclair, I. & Wilson, K. (2000) 'Sharing the care: the qualities sought of social workers by foster carets', Child & Family Social Work, 5,225-233.

Lefcourt, H. (1992) 'Durability and impact of the locus of control construct', Psychological Bulletin, 112, 411-414.

Locke, E. (1976) 'The nature and causes of job satisfaction'. In M. Dunnette (ed.), Handbook of industrial and organizational psychology, (pp. 1297-1349), Chicago: Rand McNally.

Locke, E. (1990) 'What is job satisfaction?', Organisational Behaviour and Human Performance', 4, 309-336.

McCubbin, H., Thompson, A. & McCubbin, M. (1996) Family assessment: Resiliencey, coping and adaptation - Inventories for research and practic, Madison, WI: University of Wisconsin System.

McHugh, M. (2001) The Costs of Caring: A Study of Appropriate Foster Care Payments for Stable and Adequate Out-of-Home Care in Australia'. Report prepared for Child and Family Welfare Association of Australia, Australian Foster Care Association, and the Association of Children's Welfare Agencies: Social Policy Research Centre, University of New South Wales.

Muhonen, T & Torkelson, E. (2004) 'Work locus of control and its relationship to health and job satisfaction from a gender perspective', Stress and Health, 20, 21-28.

O'Neill, C. (2000) 'Adoption, permanent care and foster care: Home-based care in and beyond the 1990's', Journal of Paediatrician's: Child Health, 36, 415-417.

NSW Department of Community Services. (2008). Child Protection Quarterly Data. Retrieved March 10, 2009 from assets/main/documents/docs_data/quarterlyjuly06_june08.pdf

Pallant J, (2005) SPSS Survival Manual: A step by step guide to data analysis using SPSS. Crows Nest: Allen & Unwin.

Pasztor, E. & Wynne, S. (1995) Foster Parent Retention and Recruitment: The State of the Art in Practice and Policy, Washington: Child Welfare League of America.

Pearlin, L. & Skaff, M. (1998) 'Perspectives on the family and stress in late life'. In J. Lomranz (ed.), Handbook of aging and mental health. (pp. 323-340), New York, Plenum.

Queensland Government (2000) Statement of Commitment between Families, Youth and Community Care Queensland and the Foster Carers of

Queensland. Queensland Government.

Queensland Government (2005) Child Protection Act 1999. Reprint No 3E

Queensland: Government Printer. Retrieved June 30, 2005 from

Rhodes, K., Orme, J. & Buehler, C. (2001) 'A comparison of family foster parents who quit, consider quitting and plan to continue fostering', The Social Service Review, 75, 84-114.

Seaberg, J. & Harrigan, M. (1997) 'Family Functioning in Foster Care', Families in Society, 78,463-470.

Sinclair, I., Gibbs, I. & Wilson, K. (2004) Foster Carers: Why They Stay and Why They Leave. London, Jessica Kingsley Publishers.

Spector, P. (1985) 'Measurement of human service staff satisfaction: Development of the Job Satisfaction Survey', American Journal of Community Psychology, 13, 693-713.

Spector, P. (1988) 'Development of the work locus of control scale', Journal of Occupational Psychology, 61,335-340.

Spector, P. (1997)Job satisfaction: Application, assessment, cause, and consequences. Thousand Oaks, CA, Sage.

Statt, D. (2004) Psychology and the world of work. New York, Palgrave Macmillan.

Sultmann, C. & Testro, P. (2001) Directions in Out of Home Care: Challenges and Opportunities. Paddington: PeakCare Queensland.

Tabachnick, B. & Fidell, L. (2001) Using Multivariate Statistics. Needham Heights, MA,AIIyn & Bacon.

Thorpe, R. (2004) 'You Have to Be Crackers'. National Foster Care Conference 2004, Walking together: People, Policy and Practise, Canberra.

Triseliotis, J., Borland, M. & Hill, M. (1998) 'Foster Carers who cease to foster', Adoption & Fostering, 22, 60-64.

Triseliotis, J., Sellick, C. & Short, R. (1995) Foster care : theory and practice, London: B.T. Batsford in association with British Agencies for Adoption and Fostering.

Warr, P. (1987) Work, unemployment, and mental health, New York, Oxford University Press

Warr, P. (2002) Psychology at work. London: Penguin.

Weinman, J., Wright, S. & Johnston, M. (1995) Measures in Health Psychology: A User's Portfolio. Windsor: Nfer-Nelson.

Wilson, K., Sinclair, I. & Gibbs, I. (2000) 'The trouble with foster care: the impact of stressful 'events' on foster carers', British Journal of Social Work, 30, 193-209.
Table 1. Foster Carers' Significant Others Scale
Scores Across all Designated Areas

                                    N   Minimum   Maximum     Mean

Actual Emotional SOS score        185      3.67     13.71    10.07
Ideal Emotional SOS score         183      7.00     15.00    12.54
Discrepancy Emotional SOS score   183     -8.50     86.00     2.46
Actual Practical SOS score        185      3.50     14.00     8.48
Ideal Practical SOS score         182      5.50     14.17    11.11
Discrepancy Practical SOS score   181     -7.67      1.17     2.60


Actual Emotional SOS score             2.04
Ideal Emotional SOS score              1.37
Discrepancy Emotional SOS score        1.88
Actual Practical SOS score             2.06
Ideal Practical SOS score              1.59
Discrepancy Practical SOS score        1.85

Table 2. Logistic Regression Analysis of Likelihood
to Stay as a Dependant of Satisfaction and Commitment

                                                     95% Confidence
                                                    Interval for Odds
                      Wald                 Odds
Variables       B     Test    df    Sig   Ratios    Upper     Lower

Commitment     .93    4.49    1    .034    2.54     1.07      6.03
Satisfaction   .04   16.63    1    .000    1.04     1.02      1.05
COPYRIGHT 2009 Australian Council of Social Service
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Eaton, Anne; Caltabiano, Marie
Publication:Australian Journal of Social Issues
Article Type:Report
Geographic Code:8AUQU
Date:Dec 22, 2009
Previous Article:Early childhood services and support for vulnerable families: lessons from the Benevolent Society's Partnerships in Early Childhood program.
Next Article:Archiving qualitative data in Australia: an introduction.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |