For this first draft, focus on getting all of your research written down. This is Step 1 of the Three Steps of History – gathering as much evidence as possible. Purpose: A first complete draft is a

For this first draft, focus on getting all of your research written down. This is Step 1 of the Three Steps of History – gathering as much evidence as possible. Purpose: A first complete draft is a
social work Social workers, who are paid to provide social services and counseling to those in need, descended from the usually female paid agents and friendly visitors of 19th-century charitable societies. In New York City the New York Association for Improving the Condition of the Poor (1845) was the first organization in the nation to use male volunteers or “paternal guardians.” Buffalo, which after Boston (1851) established the nation’s second YMCA (1854) and its first Charity Organization Society (1878), was the first city where paid agents placed dependent children in families. In 1907 the Hawthorne School for Delinquent Boys (Westchester Co), begun by the Jewish Protectory and Aid Society, pioneered psychiatric social work. At the turn of the century, elite women volunteers, so-called Lady Bountifuls, dominated voluntary societies, providing services that mostly amounted to moralizing among the poor. New York City led the way with the professionalization of the occupation: with 27 students in attendance, the New York Charity Organization Society (1882) inaugurated the first formal training program in social work at its Summer School of Philanthropy in 1898. Renamed the New York School of Philanthropy in 1903, by 1919 it had become the New York School of Social Work and in 1963 the Columbia University School of Social Work. In 1916 New York City counted 3,968 social workers in 368 organizations, mostly in the private sector. They included settlement house residents and playground workers, visiting nurses, medical and psychiatric social workers, as well as industrial social workers who oversaw the social lives of employees for companies such as the Metropolitan Life Insurance Co. Approximately 72% were women, and only 162 had any formal social work training. In the 1920s social workers moved to establish their professional standing in order to raise their status and pay and to differentiate themselves from unpaid volunteers. The new professional association, the American Association of Social Workers (1921), required formal training for admission. The activism and moral reform orientation of the settlement houses and the moralizing zeal of the older charities was replaced by an image of the caseworker as an objective, neutral investigator trained in casework and increasingly in Freudian theory. When the depression placed new pressures on social workers and clients, the Freudian shift from an emphasis on social conditions to therapeutic diagnosis divided the profession, especially in New York City, where a radical presence was strong. The depression saw a vast expansion of the public sector and the development of a trade union movement in social work, with New York City leading the nation in both. A rank-and-file movement, inspired by the Communist Party’s class and race analysis of the depression, galvanized many New York City social workers. Federation of Jewish Philanthropies of New York (1917) social workers constituted the largest and most active membership of the left-wing United Office and Professional Workers of America (1937), a branch of the Social Service Employers Union (SSEU). Social workers in New York City’s Department of Welfare (DW) formed the United Public Workers (1946), similarly distinguished by their advocacy for and identification with clients. These unions and the work of social workers would suffer, however, from an anticommunist purge of Congress of Industrial Organizations (CIO) unions in 1950 as work became more routinized and the focus shifted to social worker productivity. Supervisors encouraged midnight raids to find “men under the beds” in the homes of women who would then be removed from the welfare rolls. At the same time, reduced staff was expected to manage “efficiently” increased caseloads. New Left insurgents within welfare agencies would revitalize New York City’s SSEU in the 1960s. These left-wing social workers lent their support to an emergent national welfare rights movement, and in January 1965 some 8,000 DW workers went successfully on strike on behalf of client services and a union contract. Following a series of unsuccessful 1967 strikes, New York City’s public sector social workers moved into the less militant District Council 37 of the AFL-CIO-affiliated American Federation of State, County and Municipal Employees (AFSCME). After 1970 the introduction of computers into offices and the conservative backlash against Great Society programs shaped social work in New York City. Computers relegated public service workers to deprofessionalized clerical workers or welfare-service aides, which were unskilled jobs that paid little more than relief checks. Women remained at approximately 70% from 1940 to the end of the century. Most were white, and many began using a master’s in social work as an entry degree, seeking to become therapists. Men were more likely either to become administrators or to leave the field, and increasing numbers of minority women took the new clerical jobs. In 1998 New York State had 33,840 licensed social workers registered to practice, of whom 13,951 were in New York City (6,618 in Manhattan alone). Over 40% of the state total were clinically trained mental health workers, either psychiatric social workers or therapists. Ironically the new social worker jobs as therapists and welfare-service aides found themselves constrained as the century drew to a close, while the traditional family and child caseworkers continued to staff child welfare and private agencies. Congressional passage of the Personal Responsibility and Work Opportunity Reconciliation Act (1996) and federal efforts to contain rising health insurance costs reduced payments to social worker therapists at the end of the century, and welfare jobs contracted. Author: Daniel J. Walkowitz Editors: Peter R. Eisenstadt and Laura-Eve Moss Date: 2005
For this first draft, focus on getting all of your research written down. This is Step 1 of the Three Steps of History – gathering as much evidence as possible. Purpose: A first complete draft is a
Status of women in social work education Abstract:  This invited study sought to determine the current status of women in social work education for the special section of the Journal of Social Work Education. Analysis of the latest data available indicate that gender differences remain pervasive across many aspects of social work education, including pay, rank, job duties, and tenure. Women appear disadvantaged in almost all areas analyzed when compared to men. Additionally, pilot study data collected in Canada suggest that women academics are similarly disadvantaged. The article concludes with a discussion of the limitations of available data, broader contextual issues for women in social work, and suggestions for implementing change. AS AN INVITED STUDY for this special section that addresses women in social work education, the authors were charged with the task of assessing and reporting on the current status of female faculty within the profession. To place this assessment in context, a brief history of the status of women within the profession precedes the more current literature review. To ascertain the current status of women in social work academia, an analysis of secondary data from the Council on Social Work Education (CSWE) was conducted. In the interest of inclusivity and gaining international perspectives, a pilot project was undertaken in Canada to collect preliminary data about the status of female academics, deemed particularly necessary due to the absence of ongoing national data collection. The pilot study was designed to gather preliminary data to inform future directions for research. The ultimate goal of the analyses is to identify and address equity issues between female and male faculty, which seems especially important in a profession that values social and economic justice. The article concludes with an assessment of the current status of women in social work education in the United States and Canada, a discussion of the limitations of available data, the broader contextual issues for women in social work, and suggestions for implementing change. Literature Review U.S. and Canadian Women in Social Work Social work has long been termed “a woman’s profession” (Kadushin, 1976; Scotch, 1971), a “female dominated” profession (Stromberg, 1988), and a “semi-profession” (Etzioni, 1969). In the United States, women comprise 79% of the membership of the National Association of Social Workers (NASW) and 62% of clients served (NASW, 2003, 2005 respectively). In 2001, 86% of social work baccalaureate and 84.8% of master’s graduates were female (Lennon, 2004). Almost two thirds of social work faculty are female (Lennon, 2005). In Canada, according to the 1996 census, 75.9% of social workers were estimated to be women (Statistics Canada, 2001). Further, close to two thirds of social work faculty are women (Canadian Association of University Teachers, 2006). History of Gender Inequity Although women played a major role in founding and shaping the nascent profession of social work (Carlton-LaNey, 2001; Chambers, 1986; Costin, 1983; Deegan, 1988; Muncy, 1991; Rauch, 1975; Vandiver, 1980), as early as 1890, female social workers voiced concern about the lack of women in leadership positions at the National Conference of Charities and Corrections (cited in Giovannoni & Purvine, 1974). In the early history of social welfare, male workers were more likely to be paid professionals and women unpaid volunteers (Dressel, 1987). In the 1950s, attempts were made to “professionalize” and “defeminize” social work by actively recruiting men (Chafetz, 1972). The examination of the status of women within the social work profession intensified during the second wave of the women’s movement during the 1960s and 1970s revealing sexism (Chafetz, 1972; Kravetz, 1976; Rosenblatt, Turner, Patterson, & Rollosson, 1970; Scotch, 1971; Valentich & Gripton, 1978), sex role stereotyping (Brager & Michael, 1969), sexist assumptions about the nature of the profession (Meyer, 1982), and gendered pay inequities (Becker, 1961). This brief history suggests that women have historically occupied a second-class status within social work, ironically in a profession they are said to “dominate,” leading McPhail (2004) to term social work a “female majority, male dominated” profession. More currently, four topical areas have emerged in discussions about the status of women within social work education: (1) professional advancement, (2) pay inequities, (3) scholarship conducted by female faculty, and, most recently, (4) work-family balance. Each will be briefly reviewed. Professional advancement. Studies of social work practitioners reveal that men are more likely than women to hold managerial positions, assume these positions earlier in their careers, and earn more money than women in these positions (Fortune & Hanks, 1988; Gibelman & Schervish, 1993; Zunz, 1991). Williams (1992) found that men in the social work profession are likely to encounter a “glass escalator” rather than a “glass ceiling.” Consistent patterns of female faculty being less likely than men to be tenured and for women to be more numerous at the lowest ranks of academia has been established in the United States (Petchers, 1996; Schiele, 1992) and internationally (DiNitto, Martin, & Harrison, 1984), including in Canada (Stromberg, 1988). Di Palma (2005) found that compared to other disciplines, female social work faculty hold a higher percentage of positions at all ranks that she attributed to: (1) the large number of women in social work graduating with doctoral degrees, (2) the low status of the profession, (3) shared social work values of social and economic justice, and (4) the efforts of CSWE and the Council on the Role and Status of Women in Social Work Education in monitoring and advocating for women’s advancement. However, Di Palma also found that female social work faculty were less likely to be well-represented in prestigious Tier I research institutions and that social work educators are doing worse than other disciplines in the number of women in non-tenure-track faculty positions. A qualitative study (DiNitto, Aguilar, Franklin, & Jordan, 1995) revealed female social work faculty’s concerns with the tenure-track process, including the perception that scholarship is valued over teaching; quantitative methods valued over qualitative methods; female faculty are held to a higher standard than male faculty; female faculty are responsible for more of the school’s “organizational housework” leaving less time for research; both race and gender play a role in tenure decisions, with some respondents believing candidates of color were held to a lower standard and others believing they were held to a higher standard; and many female faculty attested to the stress and anxiety of the tenure process. Although the majority of female faculty were able to maintain a balance between family and work lives, it was achieved with much effort and planning. Pay inequities. Researchers have documented consistent pay inequities between salaries paid to male and female social work faculty and social workers (Gould & Kim, 1976; Norman 1986) ranging from more than $3,000 to $8,000 differentials (Huber & Orlando, 1995; Gibelman & Schervish, 1993; Gibelman & Schervish, 1997; Koeske & Krowinski, 2004). In an analysis of the broader category of “service industries” workers using U.S. Bureau of Labor Statistics data, Gibelman (2003) identified a consistent pattern–as the percentage of women in a service industry occupational group increased, their weekly salaries decreased. Scholarship. Several studies have found male scholars more likely than female scholars to successfully publish in social work journals (Bentley, Hutchison, & Green, 1994; Kirk & Rosenblatt, 1984; Rosenblatt et al., 1970; Schiele, 1992, 1995). However, a more recent study found no gender differences in the number of articles, books, or books chapters published and grants awarded (Holley & Young, 2005). Work-family balance. Recently, within and outside of academia, there is growing concern for how working women can balance family and career responsibilities (Blades & Rowe-Finkbeiner, 2006; Crittenden, 2001). Mason and Goulden (2004) believe that assessing gender equity in academia should include family as well as professional outcomes. A qualitative survey of social work faculty found that more than 80% of female and male faculty spoke of the importance of family-friendly policies (Young & Holley, 2005). Nonetheless, there were indications that “caregiving continues to be more ‘women’s work’ than ‘shared work'” (p. 148), because more mothers than fathers and more women without children than men without children noted the importance of family-friendly policies. Montgomery (1989) found that career demands had an effect on decisions about parenting, causing some female social work faculty to choose to have fewer children, no children, or to delay having children in order to meet the demands of academia and the tenure clock. In summary, the literature documents a history of gender inequities in the social work profession. However, assessments of the recent status of women educators in social work varies, ranging from identifying a lack of progress (Petchers, 1996) to noting improvement in some areas with inequities continuing in others (Sowers-Hoag & Harrison, 1991) to one of optimism and progress (Di Palma & Topper, 2001; Di Palma, 2005). Research questions are: (1) What is the current status of female faculty in social work? (2) Which assessment in the literature of the status of women in social work education do the latest data from the CSWE annual survey support–a lack of progress, some improvement, or optimism?, and (3) In light of the lack of annual national data collection in Canada, what is the current status of female faculty in Canada, and what should be examined in future studies? Available Data on Social Work Faculty Given the ongoing discussion in the social work literature about the role and status of women in social work education, research data are appallingly lacking. One social work organization, CSWE, collects annual descriptive data on faculty job characteristics (as well as student enrollment) for accredited U.S. schools and departments of social work. Authors of this article are members of the Council on the Role and Status of Women in Social Work Education (referred to as the Women’s Council [WC] of CSWE). The WC is considered one of three “advocacy” committees of CSWE; it provides input to CSWE’s staff and Board of Directors, but does not have formal decision-making power. Upon reviewing the 2005 CSWE report on social work education (Lennon, 2005), these authors requested access to this data file to conduct a refined analysis on gender differences taking into account various job characteristics of social work faculty. The corresponding organization in Canada, the Canadian Association of Schools of Social Work (CASSW), does not collect or analyze comparable data, which necessitated data collection by a coauthor of this article rather than a secondary analysis of data collected by a national organization. Because of the distinctly different scope and methods of the study used for each country, the data sets could not be combined and, therefore, methods and analyses are reported separately. A Note About Canadian and U.S. Education Canadian social work education resembles U.S. social work education in many ways in terms of the knowledge base, pedagogy, curriculum, and structure of the universities in which social work programs are located. At the same time, the understanding of diversity issues in social work reflect different political and social histories in Canada. For example, the Canadian Charter of Rights and Freedoms in 1982 (Department of Justice Canada, n/d) and the Multiculturalism Act in 1988 (Department of Justice Canada, 2007a) are considered the hallmarks of legislative efforts protecting the rights (including Aboriginal rights) and promoting the equity of every individual as well as designated “equity” groups. These laws affect social work in several ways including the coexistence of Francophone and Anglophone social work programs, many schools of social work have programs focusing on Aboriginal social work, and the social work curriculum has a focus on anti-oppressive practice (as opposed to multicultural social work in the United States, Sakamoto & Pitner, 2005). The United States has more than 10 times as many accredited social work programs as Canada, perhaps consistent with the 10-fold difference in populations. (For example, there are 628 BSW and MSW programs in the United States compared to 59 in Canada; the population of the United States is 296 million, while Canada has approximately 32 million. See Canadian Association of Schools of Social Work, 2005; Personal communication with Jessica Holmes of CSWE on August 2, 2006; Statistics Canada, 2006a; U.S. Census Bureau, 2005.) Method: U.S. Study The Council on Social Work Education made the raw data files describing social work faculty for 2003 available to the authors for reanalysis (see Lennon, 2005, for information about the methodology). Although these data are limited by low response rates, missing data across variables, and potentially inaccurate responses, these are the most comprehensive data available on accredited social work educational programs in the United States–their enrollments, their students, and their faculty members. The purpose of this analysis was to refine the information available on gender differences in pay equity and professional advancement in social work education in Lennon (2005). Case Selection For this analysis, 3,606 cases that represent full-time rather than part-time or adjunct faculty were initially selected based on being reported as “100% time” in the social work program. Cases in which gender was not recorded were also excluded, resulting in a sample of 3,567. However, because of variations in how program administrators reported the characteristics of their faculty, it is likely that our efforts to include only full-time faculty was not wholly successful (e.g., some cases with very low salaries still remained in the data set, which may or may not indicate part-time faculty status). Therefore, findings should be viewed with some caution. In addition, as indicated by the differing sample sizes for various analyses reported, the numbers of faculty records with data on all relevant variables varied widely. Study Variables Demographics. Gender was measured using a categorical variable with two categories: male and female. Age was measured continuously in years. CSWE measured race/ ethnicity using the following categories (some of which were collapsed for our analyses): White, non-Hispanic, African American/Other Black, Latino/Hispanic, American Indian/Other Indigenous, Asian American and Pacific Islanders, “Foreign” (no resident visa), Multiple/Other, and unknown. Data on sexual orientation, disability status, and other equity statuses were not available. Salary. CSWE calculates an “adjusted salary” variable to create a common metric for those faculty and administrators on 9-month and 12-month contracts. The adjusted salary variable reflects rate of pay for 8 or 9 months; it thus under-estimates total salaries but makes pay comparable for faculty members despite length of contract. Status-related variables. Highest degree earned was measured ordinally as MSW, other master’s, doctorate in social work/welfare, other doctorate, medicine or law, and other. MSW degree indicated whether or not a faculty member held an MSW degree (with or without an additional degree). Rank was measured in categories of full, associate, assistant, instructor, and clinical. Tenure status is indicated as with or without tenure. Years at rank and years at full-time status were also recorded. School/department characteristics. One innovation in this analysis was to link characteristics of the program setting to individual faculty data. Based on institutional information also reported annually, level of program is defined as graduate only, undergraduate only, and joint/both. Auspice indicates whether the program is in a public, private religious, or private “other” category. Institution size is measured by full-time equivalent student enrollment for the college or university as a whole and is categorized as under 2,000; 2,001-4,999; 5,000-9,999; 10,000-19,000; or 20,000 and over. Professional activities. Work assignments were measured as percentages of time devoted to teaching, field-related activities, and research as well as the percentage of time assigned to the doctoral program, master’s program, and the bachelor’s program. Percent of salary externally funded was also reported. Publication type recorded the number of journal articles, books, book chapters, monographs, and reviews published in that year, and a “total publications” variable was created by summing these values. The variables on publication scholarship were only available for the specific year of data collection and were recorded for only about 70% of the sample. No data were available to examine work-family factors that affect women in academia. Data Analysis Bivariate statistical tests for gender differences were performed as appropriate to the level of measurement of each variable. (Any reported t or one-way F tests used the appropriate versions for equal or unequal variances as indicated by the prior Levene’s test for equality of variances.) Because there were gender differences on several variables related to adjusted salary, multivariate analysis, specifically hierarchical linear regression, was performed to control for these effects. A fully refined analysis was not possible because of limitations in the data available, which excluded, for example, the length and nature of each faculty member’s experience and accomplishments in the profession prior to hiring, scholarly productivity over time, and quality of teaching, to name a few obvious factors that can affect faculty compensation. However, this multivariate approach was able to control for some factors related to salary, such as age, rank, and tenure status, that are also correlated with gender. Race/ethnicity was also examined. Results: U.S. Study As expected, women represented the majority of faculty in this sample, with 64.0% females and 35.0% males (n=3,567). Bivariate analyses indicated that statistically-significant gender differences in faculty work faculty advancement, and faculty compensation were pervasive (see Table 1). Some of these differences were related to the employing institution and some to the individual faculty member. Female faculty members were more likely to be employed in undergraduate only programs with males more likely to be employed in joint programs. Women were over-represented in the two smallest categories of institution size and under-represented in the largest. Women were also somewhat over-represented in religiously-affiliated programs. On average, male faculty members were older and had a higher mean number of years at their current rank. Men predominated among those with tenure and those at the rank of full professor. To some extent, this may represent hiring practices of the past, but women were over-represented among clinical faculty (non-tenure track) as well as among those at the assistant and instructor levels (tenure track). Women were over-represented among those whose highest degree was the MSW, likely reflecting their over-representation in baccalaureate-only settings. Women were under-represented among those with an “other” master’s, “other” doctorate, “other” degree, or a degree in medicine or law, suggesting that men were more likely to be perceived as having expertise to offer a faculty even without the doctorate in social work or social welfare. Similarly, 7.3% of males in this faculty sample did not have an MSW, while only 3.5% of females did not. Work assignments also varied slightly but significantly by gender. Women had lower percentages of time working in doctoral programs, master’s programs, classroom teaching, and research. Men had lower percentages of time reported in the bachelor’s programs, in field, and in liaison work. Significant gender differences were also found in total publications reported for the year, with men having the greater number of publications than women in refereed articles and books, but not the other categories. However, publication of refereed articles in particular is more likely to be rewarded in rank and salary than other forms of scholarly work. Salary As reported in Table 1, there was a statistically significant difference in adjusted salary by gender in this sample, with males on average earning $9,000 more than females. Adjusted salary was in turn correlated with a number of other variables on which there were gender differences, including age (r=.351, p<.001), years full-time (r=.448, p<.001), years at rank (r=.335, p<.001), percentage of time spent in research (r=.226, p<.001), the number of articles published (r=.220, p<.001), and the total publications for the year (r=.226, p<.001). Adjusted salary was negatively correlated with percent of time spent in field (r=-.149, p<.001), percent of time spent in liaison work (r=-.126, p<.001), and percent of time spent in the bachelor’s program (r=-.314, p<.001). Institutional factors also related to adjusted salary, which was modestly correlated with institution enrollment (r=.220, p<.001). Adjusted salary also differed by level of program (F=172.6, df=-2, 3,241, p<.001), with mean adjusted salaries of $44,720 in bachelor’s-only settings, $55,858 in joint programs, and $64,298 in graduate-only settings (posttests indicating that all 3 means were significantly different from each other). The percentage of time spent in research (F=106.367, df=-2, 3,582, p<.001) differed across all 3 program types, with a mean of 4.0% for faculty in bachelor’s programs, 11.1% for faculty in joint programs, and 16.0% for faculty in graduate-only programs. However, for some correlates of salary, like total publications for the year (F=28.446, df=2, 2,330, p<.001), article publications for the year (F=40.898, df=2, 2,624, p<.001), percentage of time spent in field (F=25.705, df=-2, 3,582, p<.001), percentage of time spent in liaison work (F=4.136, df=2, 3,582, p=.016), and age (F=3.611, df=2, 3,224, p=.027), post-hoc tests indicated that the significant difference was between faculty in bachelor’s programs compared to those in graduate or joint programs. In addition, there was a significant difference in mean adjusted salary by institutional auspice (F=52.420, df=2, 3,227, p<.001), with post-hoc tests indicating that faculty in private religiously-affiliated schools earned significantly less than others. These data are further evidence of a segmented employment structure for faculty in social work education in the United States (Anastas, 2006). Multivariate Analysis Because of gender differences in salary and in the correlates of salary, and because some of the correlates of salary were modestly correlated with each other, an exploratory hierarchical multiple regression analysis was performed using only predictors of adjusted salary that showed bivariate gender differences. Table 2 shows the bivariate correlations among the variables used in the equation; almost no prohibitive multi-collinearity was found. In the first step in the regression model, the variables entered were institutional factors–program level, auspice, and enrollment. In the next step, the individual factors that were correlated with salary and that showed gender differences were entered. In the last step, gender was entered to determine if there was any correlation with adjusted salary net of all of the gender-related predictors already in the equation. The results are given in Table 3 below, showing the variables newly entered at each step in italics. The results in the first step show that institutional factors such as size, religious auspice, and level of program did correlate with adjusted salary but explained only 13% of the overall variance in salary. Program level was the strongest predictor in this group of variables. Much more variance was explained by the next set of predictors–selected individual characteristics of faculty members. The beta weights indicate the size and statistical significance of each variable’s effect on salary independent of every other factor in the equation. Women are disadvantaged in all of these factors correlated with salary: age, rank, time at rank, tenure status, being more often assigned to work in the bachelor’s and master’s programs as opposed to doctoral programs, spending less time in research, and publishing fewer articles and books in the year studied. Note that the variable “rank” is crudely ordinal and reverse-coded, with higher values indicating not just assistant professor but also instructor and clinical ranks where women are over-represented. In the third step, gender itself was entered as a variable. Even when controlling for the above gender-related factors affecting salary, gender showed a small but significant correlation with salary, indicating an “excess” effect beyond other personal factors like age, rank, and publications. The good news is that this effect is very small; the bad news is that it is equal to or greater than the effects of age, the percentage of time assigned to the doctoral program, the number of books published for the year, the percentage of time spent in research, and the size and auspice of the employing program. Overall, the full model explained almost half of the variance in salary ([R.sup.2]=.530). However, these findings should be interpreted with caution because the listwise deletion of cases with missing variables resulted in a smaller sample of 2,064 faculty members. Race/ethnicity. One remaining question is about the effects of race/ethnicity, if any, on gender and salary issues. Although there were no gender differences in race/ethnicity in this sample, there were some differences in compensation by race/ethnicity (F=7.910, df=7, 3,236, p<.001). Post-hoc tests showed that those whose race was recorded as “unknown” or as foreign-born and without a resident visa earned the least; those whose race was recorded as Asian American or Pacific Islander (combined for this analysis) earned the most; and the other groups fell together in the middle range of salaries. Thus a 3-group variable describing ethnicity was created on this basis and the stepwise regression procedure described above was repeated with ethnicity entered on step 3 and gender entered in step 4. There was no significant F change or [R.sup.2] change when the ethnicity variable was entered in step 3; in addition, the beta weights for this variable were not significant in step 3 or step 4. Gender was significant to exactly the same degree when entered in a fourth step ([beta]=.067). Thus in this sample, gender has its effect on faculty salaries independent of race and ethnicity, which is similar to what has been reported in other social work samples (see for example, Gibelman & Schervish, 1997). Methods: Canadian Pilot Study There is a dearth of literature in Canada on the role and status of women social work educators. As the Canadian Association of Schools of Social Work (CASSW) does not collect any data on faculty, there is no information available on the status of social work educators who are women or members of other equity groups, or both. Examining not only the role and status of women but also the intersection of identifies is important (Sakamoto & Pitner, 2005). The only recent statistical data that seem to answer at least part of our demographic questions were those provided by the Canadian Association of University Teachers (CAUT, 2006). The purpose of the current investigation, therefore, was to further analyze the CAUT data and collect the preliminary data on the role and status of women in social work education in Canada with special attention to the intersection of different minority (equity) statuses. This pilot study will serve as the foundation for designing a more comprehensive study in the future. Design and Procedures The pilot study used three methods. First, a secondary analysis of the most recent public data on university teachers in social work (CAUT, 2006; collected by the Statistics Canada) was conducted by gender and rank (no other variables were available). Second, a pilot survey study was designed to collect descriptive data on the faculty demographics in each school of social work. The study received the University of Toronto’s institutional research board approval. Further, CASSW’s Women’s Caucus officially endorsed the project at its annual meeting. In 2006, this short questionnaire was sent via e-mail to the deans or directors of all 35 CASSW-accredited schools of social work (CASSW, 2005). The French version of the questionnaire was sent to the nine Francophone schools. Participation was voluntary. It should be noted that, because questionnaire data were collected through administrators and not through individual faculty members, the number of faculty members who belong to each equity status should only be seen as estimates. Third, to supplement the other two methods, publicly available data were collected through Internet sites to obtain an estimate of the number of male and female social work faculty members at various ranks and the number of male versus female deans and directors at the 35 accredited schools of social work in Canada. The researchers used multiple methods to identify the gender of the deans and directors, first by their first names (e.g., Nancy for woman, John for man), and second by the pronouns used in the Web sites referring to the respective deans and directors (e.g., mention of “she” or “he”). Measures Demographics. Gender was asked in three categories of women, men, and other. Race/ ethnicity was asked as part of the equity statuses reflecting the unique Canadian context, whereby people of color would be identified as either “Aboriginal persons” or “visible minorities.” The Canadian Federal Employment Equity Act defines women, Aboriginal peoples, “visible minorities,” and people with disabilities as four designated equity groups, for whom employment equity is of particular concern (for explanation see Department of Justice Canada, 2007b). Under the Employment Equity Act, members of “visible minorities” are dignified as “being non-Caucasian in race or non-White in color,” other than Aboriginal peoples (Statistics Canada, 2006b). The terms, “Aboriginal people” or “Aboriginal peoples” are commonly used to signify the diverse groups of indigenous people(s) of Canada (for more information see Indian and Northern Affairs Canada Communications Branch, 2004). Also, an optional question asked whether any of the faculty members publicly identified themselves as lesbian, gay, bisexual, transgender, transsexual, two-spirited, intersex, or queer (LGBTT2IQ; two-spirited people were traditionally believed to carry two spirits, both male and female, in Aboriginal cultures; Queen’s University Positive Space Program, n/d). Status-related variables. Rank/title was asked in categories of full, associate, assistant, instructor/lecturer, and other full-time faculty. Gender distributions of leadership positions were asked with the options of dean/director, associate dean/director, assistant dean/director, program directors (i.e., BSW, MSW, and PhD directors/coordinators), field directors (i.e., BSW and MSW field directors), research director, and other leadership positions. School/department characteristics. Questions about the respective universities and schools of social work were asked, including the estimated population of the city/town in which the school is located, student enrollment in the university as well as school of social work, and whether the school of social work has a special program or initiative focusing on the recruitment/retention of students who belong to any of the equity groups or specific studies of equity group populations. Results: Canadian Pilot Study Census Data on University Teachers in Social Work There were 312 full-time university teachers in social work according to the individual data collected by Statistics Canada (CAUT, 2006). The data were only broken down by rank (full, associate, assistant professors, and other) and gender (female, male) with no institutional data. Of 312 university teachers, 58.7% were women. However, among 63 full professors, only 38.5% were women, while among 114 assistant professors, 75% were women. The data give strong evidence supporting an unequal rank distribution of men and women faculty ([chi square]=35.46, df=3, p<.001). Pilot Survey and Internet-Based Data Institutional information. Within the study’s timeframe, 34.3% (n=12) of schools completed the questionnaire. Overall, the disclosure of the individual demographic data of employees was seen as a privacy concern and limited our data collection, which will be explained further in the discussion section. To further ensure the confidentiality and anonymity of the data, the variables were aggregated where needed, verifying that no individual or institution would be identifable. Five of the respondent schools of social work belonged to universities that are categorized as primarily undergraduate according to the commonly-cited Macleans (2006) university ranking, and the rest were considered comprehensive or medical/doctoral. The participating universities had student populations ranging from under 2,000 to 20,000 and over, and were located in cities of various sizes. The student enrollment in social work programs ranged from what is considered in Canada as small programs (e.g., 50 full-time students) to large programs (e.g., more than 300 students). Notably, half of the participating schools offered special degree programs or recruitment/support initiatives geared toward Aboriginal students. The completed questionnaires yielded data on 184 full-time individual faculty members in social work. However, because of the low cell counts, some of the demographic information obtained (e.g., gender breakdown of those in “equity” groups) will be withheld to protect the anonymity of participating institutions and individuals. Gender. Of 184 faculty members, approximately two thirds (117) were women and one third (67) were men, including full professors, associate professors, assistant professors, lecturers, instructors, and other full-time teaching faculty members. No faculty members were identified as “other” in the gender category. Some of the lecturers and instructors held tenure-track positions but were in the rank of lecturers and instructors until they completed their doctoral degrees. Due to the small sample size, the data on rank was aggregated and analyzed using only two categories: those who are usually in tenured positions (i.e., full professors and associate professors) and those who hold un-tenured positions (assistant professors and other full-time teaching faculty including full-time lecturers and instructors). The analyses revealed strong evidence for an unequal rank distribution of men and women: 43.6% (51) of female faculty were categorized as full or associate professors, as opposed to 77.6% (52) of male faculty ([chi square]=20.012, df=1, p<.0001). The Internet search revealed that 48.6% (17) of 35 deans and directors of schools are women. The questionnaire data revealed no statistical difference in gender distribution of faculty members in administrative leadership positions. However, when the field directors were excluded from these senior faculty positions, men were more likely to be in senior administrative positions (29.9% of men) than women (14.5% of women; [chi square].=6.225, df=l, p=.013). In other words, only 1 in 7 women were in senior leadership positions when field directors were excluded from the analysis, as opposed to close to I in 3 male faculty members. In relation to other markers of “equity” status, there were no gender differences in representation of faculty of color (i.e., visible minority and Aboriginal faculty members) or among those faculty members identified as lesbian, gay, bisexual, or transgender (LGBT). However, faculty members with disabilities were more likely to be female than male (Fisher’s Exact Test p=.049), although because of a particularly small number of respondents from this population in this study, this result should be interpreted with caution. Other equity groups. As a result of the small sample size, it was not possible to test whether female faculty with other intersecting equity (minority) statuses were more likely to be non-tenured or in lower academic ranks than men. It is noted that there were higher proportions of Aboriginal peoples represented in the social work faculty (8.7%) than in the general population (4.4% reporting some Aboriginal ancestry in 2001 Census; Statistics Canada, 2004). However, when the data on each equity group were examined, Aboriginal full-time faculty members were more likely to hold assistant professor or lecturers/instructor positions than full and associate professor positions ([chi square]=4.348, df=l, p=.037). There were 7.1% of faculty members who were identified as lesbian, gay, bisexual, or trangender, but none were identified as two-spirited, intersexed, or queer/questioning. Faculty who are visible minorities (10.3%) or LGBT-identified were no more likely to be in assistant professor or lecturer positions, than in full or associate professor positions when the data were analyzed separately for these variables. However, when all the equity statuses were aggregated, faculty members who are visible minorities, with disabilities, Aboriginal, or LGBT-identified were more likely to be in assistant professor or lecturer positions, than in full or associate professor positions ([chi square]=4.126, df=1, p=.042). Discussion There is persistent gender disparity across many aspects of social work education in both Canada and the United States. Our purpose is not to conduct a cross-national comparative analysis, as the variables and methods are too divergent across studies. Nonetheless, examining these two countries separately highlights the need for future cross-national data collection to examine the prevalence of gender inequity, and the intersection of gender with other social categories. Additionally, presenting pilot data from Canada brings to light the fact that there are no national monitoring or data collection mechanisms to date, but points to the need to implement one. The U.S. analyses show that women receive lower salaries even after controlling for numerous other job characteristics that influence salary. Furthermore, gender differences were found across most of the job characteristics that were used to predict salary, indicating a pervasive problem for women in status, rank, and job duties, as well as in salary. Women in these data are over-represented in undergraduate-only programs, schools with small enrollment, and religiously-affiliated schools, as well as in clinical and instructor positions. Women are under-represented in the largest schools, joint BSW/MSW programs, as tenured faculty, as full professors, and in journal and book publications. In terms of percentages of time assigned to work tasks, women more often perform field and liaison roles, while men more often perform administrative and research roles and teach more doctoral and master’s classes and fewer BSW classes. In the Canadian study, the pilot data showed that men were more likely to hold higher academic ranks than women faculty members, and that equity is an issue particularly for Aboriginal faculty members. At this point, there is little evidence that Canadian legislation targeting the increased representation of equity groups has had an impact. Meanwhile, the higher proportion of women and other minorities, as well as the presence of Aboriginal-focused academic programs in half of the participating schools, could be interpreted as resulting from schools’ successful efforts to increase the numbers of women and other minorities in social work academia, but these efforts are only truly successful if these individuals are retained and promoted. As a pilot study, the analyses presented here raise important questions and suggest that the status of women in Canadian social work education is a pressing concern that requires close monitoring. Further, variables other than gender need to be examined to take into consideration the diversity within the class of women and promote a more nuanced examination of women’s status in social work education (e.g., the status of women of color, Aboriginal women, women with disabilities). One of the most striking experiences we encountered was the various limitations of available data on social work faculty members by gender, rank, and minority status as well as on their earnings and the nature of their work. Although social work programs are mandated to include diversity concerns, data about social work faculty equity are not being accurately collected by accrediting organizations in the U.S. or in Canada. In Canada, no systematic data are being collected by social work organizations, and public data are only available on gender distributions by rank, but not by other intersecting identities such as race/ethnicity, Aboriginal heritage, and disability status. Further, in the United States, the data are severely limited because of low response rates; the use of one, often unidentified, reporter from each school about all the faculty within that school (rather than a direct survey of individual faculty); high levels of missing or inaccurate data; and lack of survey follow-up and data cleaning. In Canada, the recent implementation of the Freedom of Information and Protection of Privacy Act (information and Privacy Commissioner Ontario, 2007) puts limits on the collection of comprehensive institutional data for research purposes. Protecting faculty members’ privacy and confidentiality is important, especially for those who do not wish to disclose particular social category membership or who fear discrimination based on their particular social status. When administrators report on individual faculty members, the potential for inaccuracies and violation of privacy is greater than if individuals report their own information in a way that is protected by human subjects laws and by aggregating data so that individuals can not be linked to particular schools or programs, and principle investigators are responsible for protecting participants’ confidentiality. The need to obtain improved data through systematic, on-going research that uses sophisticated design, sampling, and data collection methods is apparent and urgent. We recommend that data be collected from all U.S. and Canadian schools using probability sampling of faculty and administrators in all ranks and job positions. Individual participants should report their own job characteristics and salaries, rather than using a single reporter per school to “estimate” entire faculties and individuals’ demographic/diversity characteristics, job duties, and pay. A single reporter, such as an administrator, could provide separate information about overall school characteristics such as size, enrollment, auspice, etc. Survey measures should encompass a wide variety of variables, at least all those mentioned in the literature, and allow for analysis of multiple demographic categories and their interaction. A large-scale study of all educators should be conducted at least every 5 years to monitor change over time. Data should be available upon request for secondary analysis, as well as reports released periodically and systematically, and results used for school accreditation assessments. Research on the profession as a whole should obtain similar data for comparison across academic and nonacademic jobs. Finally, we want to note that some schools in the United States and Canada have implemented progressive family leave and other equity policies to correct for various forms of discrimination. A more in-depth data collection about policies, in addition to the outcome variables, would allow for an evaluation of whether these policies are producing the intended outcomes or better outcomes than schools without such policies. Broader implications and Future Directions We should note that while the proportion of women and men on social work faculties is similar in the United States and Canada (approximately 58-65% women), it is not reflective of the profession as a whole where there is a much higher proportion of women (approximately 80%). To understand these data within a broader social context and compared to other research findings, we note that male advantage has also been found to exist among non-academic social workers (Anastas, 2007; NASW, 2002). It is also important to note that the status of women in social work education is similar to the status of women in higher education in general. While women students are now the majority in higher education, female faculty and administrators have not achieved equity in numbers, pay, position, salary, or status (Fogg, 2003; Twombly & Rosser, 2002). Why are gender differences so enduring? Some scholars have held institutional sexual discrimination and patriarchal norms and practices responsible for female faculty’s second-class status (DiNitto, Martin, & Harrison, 1982; Dressel, 1987; Kravetz, 1976), while others downplayed institutional sexism but highlighted differences in productivity and rank (Rubin, 1982, 1988). Given that salary differences exist even after controlling for factors such as rank and productivity, it seems likely that a multitude of direct and covert discrimination, disparate impact, and the continuing primary role of women as caregivers, wives, spouses, and mothers influence salary and job status. For instance, the fact that tenure clocks often run during women’s prime reproductive period and most universities do not hire their own graduates disparately affect women who seem to be less geographically mobile than their male counterparts, often out of deference to their spouse’s career (Young & Holley, 2005). Further, numerous studies have shown that gender stereotyping and cultural beliefs about women’s lower job competencies in relation to men influence decision making on a variety of levels that affect advancement. However, when specific individuals’ behaviors are measured and rated, women often equal or outperform men, but beliefs about women’s lower competence in general persist, especially in men’s schemas (Catalyst, 2005; Valian, 1999). Further, gender stereotypes interact with other bias perceptions about ethnic, racial, and other minority groups to compound discrimination for minority women (Task Force on Women in Academe, 2000). The “accumulation of advantage” over time results in male dominance in administrative and decision-making positions (Catalyst, 2005; Valian, 1999). DiNitto et al. (1982) outline five levels of social organization that affect women: (1) societal, (2) institutional, (3) organizational, (4) role, and (5) individual, while suggesting that interventions must be addressed at all levels in order to create change. Solutions require a systemic focus and include: mentoring and networking, including cross gender and racial mentoring (Simon, Bowles, King, & Roff, 2004); flexible time for tenure application and the creation of part-time, tenure-track positions (Sowers-Hoag & Harrison, 1991); a more transparent tenure process, greater value placed on diverse research methods and topics, equalizing the type and amount of work assigned to male and female faculty as well as the available resources, increasing women’s leadership (DiNitto et al., 1995); and using the CSWE accreditation process and the Council on the Role and Status of Women in Social Work Education to continue to monitor and improve women’s status (Bentley, Valentine, & Haskett, 1999; Trolander, 1997); and paid family leave, reduced workload after childbirth, job sharing, well and sick drop-in childcare, adult day care, long-term care insurance, and assistance in locating services (Young & Holley, 2005). With the recent emphasis on advocating for family-friendly policies, and some schools already adopting such policies, it is imperative that nontraditional families, such as LGBTT2IQ families, are included in the definition of family. Finally, a greater emphasis on understanding how stereotyping and cognitive schemas that influence decision making should be placed in curriculum as well as scheduling regular faculty and administrative work groups to review how these processes might be influencing faculty hiring, promotion, job duties, and the distribution of resources. Accumulating advantage by gender occurs on a variety of levels from differential promotions, salary, job assignments, and resources (e.g., assignment of student assistants, office space, course releases, summary salary, discretionary funds)–all of which must be transparent, systematically monitored, and open for discussion. Social workers must be willing to actively engage in practices that support the profession’s stated goals, values, and ethics related to social justice and equity, which requires a willingness to monitor and change everyday practices and policies that disadvantage women. When educators do not model a commitment to social and economic justice, students note the disparity and hypocrisy between social work rhetoric and reality. Authors: Izumi Sakamoto, Jeane W. Anastas, Beverly A. McPhail and Lisa G. Colarossi Date: Jan. 1, 2008
For this first draft, focus on getting all of your research written down. This is Step 1 of the Three Steps of History – gathering as much evidence as possible. Purpose: A first complete draft is a
The teaching and learning of communication skills for social work students: a realist synthesis protocol. Abstract:  Background Good interpersonal communication is the cornerstone of social work practice, enhancing the outcomes of people in receipt of its services. Social workers’ communication skills are often identified as an area of concern. Communication skills can be developed and refined through training or education. The teaching and learning of communication skills is firmly embedded in many social work qualifying courses; however, considerable heterogeneity exists regarding such complex interventions and the theoretical underpinnings of which have not been made explicit. Realist synthesis can help explain how, why, for whom and in what circumstances an intervention might work, which is an important first step for helping educators to tailor courses to meet the needs of different learner groups and, where applicable, the employing agencies and government departments who fund them. Methods Realist synthesis is an interpretive, theory-driven and explanatory approach that aims to explain the interplay between the context, mechanisms and outcomes of interventions. This realist synthesis seeks to understand and explain to what extent, how, why, for whom and in what circumstances complex educational interventions aimed at teaching communication skills to social work students produces its effects. A five-step process will be followed iteratively. In step 1, the initial programme theory will be developed. Step 2 will involve searching for evidence. In step 3, selection and appraisal will take place. Step 4 requires data to be extracted and organised, and in step 5, data will be analysed and synthesised. Discussion The teaching and learning of communication skills in social work education is under theorised. The findings from this realist synthesis aim to help policymakers and educators make informed decisions about the design and delivery of complex educational interventions aimed at improving the communication skills of social work students. The realist synthesis will be conducted and reported in accordance with the RAMESES guidelines and standards. Systematic review registration The review is registered with the Open Science Framework. https://doi.org/10.17605/OSF.IO/BYHC7 Keywords: Realist review, Realist synthesis, Social work education, Social work students, Communication skills Author(s): Emma Reith-Hall1 Background There is considerable consensus within the literature that good interpersonal communication is the cornerstone of social work practice [1, 2], enhancing the outcomes of people in receipt of its services [3]. Serious case reviews and commissioned reports commonly identify social workers’ communication as an area of concern. Since interpersonal communication is a goal-driven and goal-directed process ‘undergirded by perceptual, cognitive, affective, and behavioural operations’ [4], communication skills can be developed and refined through training or education. Communication skills are firmly embedded within the curriculum of social work qualifying courses in a number of different countries including Australia, the UK and the USA [5-7]. In the UK, teaching communication skills became mandatory following the introduction of the degree programme 20 years ago [8]. The content, sequencing and pedagogy underpinning the educational interventions were not prescribed; hence, considerable variation exists both within the UK and further afield. Knowledge and practice reviews have identified that the outcomes evidence underpinning these interventions is limited, and that the theoretical underpinnings of the teaching and learning of communication skills have not been made explicit [9-11]. Some time has passed since these reviews were undertaken, during which considerable research activity has taken place and new routes into the profession have proliferated. In the UK, for example, Think Ahead and Frontline seek to recruit high-achieving graduates, whilst Step Up and the new social work apprenticeship degrees recruit experienced support staff into undergraduate programmes. The time is ripe to revisit the literature on the teaching and learning of communication skills in social work education to update our knowledge so that policy and practice decisions can be better informed. To address the first gap within the literature — the outcomes evidence — a systematic review aimed at investigating whether or not the teaching and learning of communication skills is effective has recently been undertaken [12, 13]. Notwithstanding significant methodological challenges, there was overall consistency in the direction of mean change for the development of communication skills of social work students following training [13]. To address the second gap within the literature — the need to theorise the intervention — a broader range of study designs is required which can explain how and why interventions might work [14, 15]. Realist synthesis is particularly suited to this purpose since programme theories help explain how the intervention is supposed to work. Preliminary searching indicates that the body of evidence has grown in the last two decades, suggesting that fresh insights into the mechanisms underpinning communication skills courses in social work education should be reinvestigated. The explanation a realist synthesis can provide about how, why and for whom an intervention might work is an important first step for helping educators tailor courses to meet the needs of different learner groups and, where applicable, the employing agencies and government departments who fund them. The review protocol is registered on the OSF database (https://doi.org/10.17605/OSF.IO/BYHC7). Methodology Realist synthesis Realist synthesis is an interpretive, theory-driven approach [16] which reviews different types of information, evidence and literature about complex social interventions. Methodological inclusivity and pluralism are encouraged. Realist synthesis applies a realist philosophy of science, that is an external (real world) reality exists, but this can only be understood through human interpretation (senses, language and culture) ‘to the synthesis of findings from primary studies’ ([15], p.93) that aims to explain causation within interventions through context-mechanism-outcome configurations. The realist approach recognises that no theory can always explain or predict the outcomes of a complex social intervention in every context. Whilst programmes provide opportunities and resources, the outcomes are ultimately determined by the choices and decisions of its participants. Yet, the realist approach assumes that because only a limited number of options are available in any given context, individuals are likely to, though will not always, make similar choices about the resources they use. In realist terms, these semi-predictable reoccurring patterns of behaviour are known as ‘demi-regularities’ [16]. Realist synthesis seeks to ‘uncover the underlying theories that explain these demi-regularities by critically scrutinising the interaction between context, mechanism and outcome in a sample of primary studies’ [17], which are commonly expressed as ‘context-mechanism-outcome configurations’ (‘CMOCs’). Mechanisms, defined as ‘underlying entities, processes, or structures which operate in particular contexts to generate outcomes of interest’ [18], are a defining feature of realist research. They help us understand that it is not the intervention itself which produces outcomes but people’s reactions, reasoning and responses to it that are important. In realist research, the relationship between context, mechanism and outcome is explored through a variation of the question, ‘What works, for whom, in what circumstances, in what respects and why?’ From this, ‘the reviewer constructs one or more middle-range theories to account for the findings’ ([15], p. 94). Through an iterative process, realist synthesis seeks ‘to gradually develop and refine the programme theory so that it is more detailed, realist in nature and the inferences within it are supported by data’ [19]. In later stages of the inquiry, following a series of different iterations, a number of C-M-O configurations are developed and then tested, using the data gathered in the review. The configurations seek to explain in which context(s) and which mechanism(s) are ‘triggered’ to produce which outcomes(s). The refined realist programme theory should be in the ‘middle range’, that is it should be specific enough to permit empirical testing but abstract enough to provide useful explanations transferable to other situations where the same mechanisms may be operating. The realist approach is particularly suited to education research, where multicomponent interventions are complex and outcomes are highly context dependent and influenced by the reactions, responses and reasoning of both educators and learners. In relation to medical education, Wong et al. (2012, p. 90) [15] explain that ‘the impact of the “same” intervention will vary considerably depending on who delivers it, to which learners, in which circumstances and with which tools and techniques’. The same point can be made about social work education generally and the teaching and learning of communication skills more specifically. It is for this reason that a realist approach was deemed appropriate for investigating this topic. Objectives The intended objectives of this realist synthesis are as follows: 1. To understand and explain to what extent, how, why, for whom and in what circumstances the teaching and learning of communication skills for social work students produces its effects. Theory adjudication (determining which theories best explain the intervention) and cross-case comparison (comparing how the intervention works for different groups or in different settings) will be investigated, provided sufficient detail is included within the primary studies. 2. To provide recommendations, based on the above explanation, and to help educators make informed decisions about the design and delivery of pedagogic practices. Review questions 1. To what extent does the teaching and learning of communication skills for social work students produce its intended outcomes? 2. What formal substantive theories are used to inform the teaching and learning of communication skills for social work students? 3. What are the mechanisms by which the teaching and learning of communication skills for social work students are believed to result in their intended outcomes? 4. What are the important contexts which determine whether the different mechanisms produce their intended outcomes? 5. In what circumstances are such interventions likely to be effective? These questions, deemed important by key stakeholders, will be iteratively refined once the exultant literature is better understood. Whilst question 1 has been addressed by the aforementioned systematic review [12, 13], the author wonders whether additional outcomes, and a greater understanding of the complexity of the intervention, might be demonstrated through a broader range of study designs. Questions 4 and 5 may not be answerable through existing studies and may need to be addressed subsequently, through realist evaluation. Ethical considerations Ethical approval was not required for this synthesis because the literature is in the public domain. Synthesis structure and features The synthesis is informed by Pawson’s (2006) five stages (identifying the review question, searching for primary studies, quality appraisal, extracting the data, synthesising the data and disseminating the findings) [16]. An initial explanatory theory will be developed, after which the ‘programme theory’ will be tested and refined against data from empirical studies. A visual representation, informed by Duddy and Wong’s (2018) depiction [20], which outlines the approach underpinning this realist synthesis, is provided in Fig. 1 below. Fig. 1: Steps of a realist synthesis Planning and preparation Background reading As a researcher, who is also an academic in the field, I regularly read and review the literature for teaching, research and other scholarly purposes to ensure that my content knowledge and expertise are current and relevant. Through sustained immersion, familiarisation with the literature was already well established before this particular research project began. Identifying and involving stakeholders Citing Ryan and Hood (2004) [21] and Schwandt (2005) [22], Suri and Clarke (2009) [14] suggest that ‘the knowledge construction of educational practices can be considered incomplete and oppressive if it undermines the rich knowledge of different stakeholders, especially teachers and students, whose practices and experiences are the sites for educational research’ (p. 412). In realist research, identifying and involving stakeholders is encouraged from the outset. In addition to policymakers, the key stakeholders involved in social work education are students, academics, practitioners, and people with lived experience (sometimes referred to as service users and carers). My commitment to, and experience of, collaboration and partnership working [23-25] supports my ability to work with different stakeholder groups, using their ‘lived experience’ and/or content expertise to focus the review and inform the development and refinement of the programme theory. Deviating slightly from the more established practice of bringing different stakeholder groups together, I have met different stakeholder groups separately. This was partly to ensure that the voices of more powerful groups do not become privileged above less powerful groups and also for more practical reasons — getting everyone together in one place has not been feasible when there is no funding available to reimburse expenses and was not deemed responsible in light of a global pandemic given the existing health conditions of some collaborators. Using a reflexive approach, I, as the researcher, will consider whether this strategy needs to be adapted as the research project progresses. Stakeholders will be involved throughout the research process, as shown in Fig. 1. Step 1 — Develop the initial programme theory Realist inquiry begins (and ends) with a programme theory [20]. The initial programme theory tends to operationalise a set of assumptions of the programme designers about how the programme is expected to work. Preliminary literature searches and stakeholder consultations allow the programme theory to be iteratively developed and help determine the priorities of the realist synthesis. Discussions with stakeholders have influenced the nature and form of this realist synthesis. For example, the lack of a coherent theoretical framework to inform the teaching and learning of communication skills [9] is an issue of particular interest for social work academics, which influenced the decision to place more emphasis on identifying candidate substantive theories within the literature. Formal substantive theories ‘provide a bridge to a wealth of existing research and knowledge about a topic’ and operate at a higher level of abstraction than programme theories [26]. Preliminary searches A series of preliminary scoping searches aimed at retrieving substantive theories from the literature have been undertaken. The first of the preliminary searches entailed searching the Social Care Institute of Excellence (SCIE) website for any grey literature sources by selecting ‘communication skills’ from the subject topic menu of the resources and services section. Two more structured searches were also undertaken: a database search of the Web of Science and a discipline-specific journal search of Social Work Education, the British Journal of Social Work and the Journal of Social Work Education. The searches were guided by the BeHEMoTh (behaviour of interest, health context, exclusions, model, theory) approach [27], using various terms to describe the behaviour of interest (communication/interpersonal), adapting the health context (social work education) alongside the suggested terms for theory or model (theor*/model*/framework*, concept*). No exclusions were applied, and the theory concepts were not restricted to title and abstract as it was anticipated that information about theories might be located in the main text and reference lists. The combined searches produced a total of 39 records. Ten grey literature resources were retrieved from the SCIE website, including the aforementioned knowledge reviews. Six records were retrieved through the web of science search and twenty-three through the discipline-specific journal search (4 from Social Work Education, 6 from the British Journal of Social Work and 13 from the Journal of Social Work Education). Each record was added to an EndNote group folder. Fifteen records were excluded for the following reasons: duplication (N = 1), the record was a book review (N = 4), the topic was not about the teaching and learning of communication skills (N = 6) and the population did not comprise social work students (N = 4). The remaining 24 records were read in full. A PRISMA flow diagram of the preliminary searches is depicted in Fig. 2. Fig. 2: PRISMA 2020 flow diagram for preliminary searches to identify substantive theories Reference and citation tracking Recognising that theory might be contained within a sibling paper, reference and citation tracking were undertaken manually and using Google Scholar, which led to the identification of two additional records. Discussions with stakeholders Following the preliminary searches, discussions with stakeholders took place regarding substantive theories. Contact was made with social work academics involved in the teaching and/or researching of communication skills. Their content expertise confirmed that relevant candidate theories had been identified. Substantive theories The key substantive theories underpinning the teaching and learning of communication skills which were explicitly referred to in the literature found through the theory search outlined above are included below: * Experiential learning theory * Reflective practice * Adult learning theory * Theory of living human systems * Relational/cultural theory * The postmodern and post-structural approach * Task-centred and behavioural approaches * Humanistic/person-centred counselling approaches, including microskills training * Psychosocial theory Experiential learning theory and humanistic person-centred approaches were most frequently mentioned in the studies identified through the theory search, a finding supported by a recent systematic review [13]. Experiential learning theory synthesises the contribution of scholars, including educational psychologists and philosophers, who positioned experience as playing a fundamental role in learning, training and educational development of adult learners who bring their personal and professional experiences with them. Experiential learning involves learning by experience, in which the learner takes on an active role, followed by reflection and analysis of that experience, which further develops their learning. Kolb’s (1984) experiential learning cycle [28] was the most frequently cited reference to theory in the records identified through the preliminary searches, followed by Donald Schön’s (1983) [29] work on reflective practice and Knowles’ (1978) [30] adult learning theory. These theories are associated with a constructivist view of education, ideas that can be traced back to John Dewey. Ivey and Authier’s (1971, 1978) [31, 32] microskills approach provides a systematic method for teaching beginning communication skills to counsellors and therapists. It shares similarities with the above theories, although its roots lie in psychotherapy, particularly in humanistic person-centred counselling approaches, developed by Carl Rogers (1951, 1961) [33, 34]. The other theories were mentioned in just one study. These included Agazarian’s (1997) system-centred therapy for groups [35], Miller and Stiver’s (1997) ‘relational/cultural theory’ [36] and Jessup and Rogerson’s (1999) postmodern and post-structural approach [37]. A brief overview of the substantive theories identified through the preliminary searches are provided in Table 1. Table 1: Overview of substantive theories The theories are not mutually exclusive, and there is significant overlap between them. The role of the substantive theories will be considered in the realist synthesis. Development of programme theory From the combined preliminary searches and discussions with stakeholders, an initial programme theory was developed. A wide range of teaching and learning activities were identified, which involved combining formal input (on theory or background) from an instructor with experiential or practice-based activities such as group exercises, group discussions, role-plays, simulations and skills laboratories, video work, observation, feedback and reflection. Service user and carer involvement and shadowing experienced social workers in practice were also identified within the literature identified through the preliminary searches. The different activities and components that communication skills courses comprise contributed to a ‘theory of action’. Some of the resources also described how the intervention is thought to work. Pedagogic methods that encouraged learning by doing and reflection were common, explaining why assessment, evaluation, feedback and reinforcement were frequently commented on in the literature. In terms of the intended programme outcomes, the studies referred to confidence [43, 44] and interpersonal communication skills [45]. Knowledge and attitudes were also mentioned within the domains of ‘knowing’ and ‘being’ [46]. Discussions with different stakeholder groups consulted during the development of the systematic review protocol suggested that self-efficacy, knowledge, attitudes and skills were the outcomes of importance [12, 13] and are captured within Carpenter’s (2005) framework for social work educational outcomes [47]. Pawson (2006, p. 74) [16] suggests that the initial rough theory should contain some key features of realist explanation, comprising ‘conjectures on the generative mechanisms that change behaviour, ideas on the contexts that influence its operation, and hunches about the different outcome patterns that ensue’. Successful outcomes appeared to be dependent upon students engaging in ‘learning by doing’ and ‘learning through reflection’, tenets which are supported by two of the substantive theories (experiential learning theory and reflective practice). A safe learning environment was deemed to be an important context for the learning by doing component to take place. Figure 3 shows the initial rough programme theory. Fig. 3: Initial rough program theory From the studies identified through the preliminary search and discussions with stakeholders (students, people with lived experience and social work academics), it was also possible to identify some tentative and provisional context-mechanism-outcome configurations, as stated below, where C = context, M = mechanism and O = outcome. 1. In a safe learning environment [C], students will experience a sense of trust [M] and manage performance-related anxiety, fear and embarrassment [M], enabling them to engage [M] in practice opportunities to improve their communication skills [O]. 2. Students are more likely to demonstrate effective communication skills [O] when practice scenarios are authentic [C] because they perceive it to be believable [M] and/or useful [M], which motivates them [M] to perform well. 3. In the context of supportive and constructive working relationships [C], students will take feedback on board [M], evaluate [M] and reflect on their skills [M], developing knowledge [O] and confidence [O] to demonstrate communicative improvements [O] in subsequent practice opportunities. 4. In a ‘containing’ and attuned reflective space [C], students will make sense of their own internal worlds and those of others [M], developing self-awareness [O], use of self [O] and emotional capacity [O] to communicate effectively within the helping process [O]. Elements of the first three CMOCs featured repeatedly in the studies. A microskills approach underpinned by humanistic/person-centred counselling, behavioural psychology and experiential learning theory appears to underpin various different participative teaching and learning activities, whereby skills are practised, evaluated and reflected on and then reinforced through further practice. The fourth CMOC, underpinned by a psychosocial approach, appeared to operate at a deeper level and considered how understanding and using ‘the self’ develops in relationships with others. Although this CMOC was most evident in the literature about communicating with children and young people and referred to some very specific activities including tutor modelling, child observation and reflective groupwork, it was possible to glean from the more generic studies that this CMOC would also operate in learning how to communicate with adults. This idea will be tested in the realist synthesis. Themes relating to power and control also emerged in the studies reviewed for the preliminary search. The role of service user and carer involvement in social work education was highlighted in some of the studies, whereas others considered the importance of students feeling in control of their learning. Although it was not possible to develop a relevant CMOC, this may become possible in the review of the wider body of literature located through the main search. The purpose of realist research is to ‘gradually develop and refine the programme theory so that it is more detailed, realist in nature and the inferences within it are supported by data’ (Wong, 2015, p. 2) [19]. To support this endeavour, the initial programme theory will be further developed and refined through the subsequent steps in the approach underpinning this realist synthesis, as shown in Fig. 1. Step 2 — Searching for evidence The main systematic search of the literature aimed at identifying relevant documents and articles from which the programme theory will be developed and tested takes place in step 2. Academics, service users and carers, students and practitioners have been asked for suggestions for key words for the intervention, programme recipients and intended outcomes. An information specialist has helped the researcher formulate the search string for this particular search (based on population and intervention concepts only) using a combination of subject headings and free text, adapted for each database. The basic search string is as follows: * (“social work student*” OR “student social worker*”) * AND * (communicat* OR interpersonal OR interview*) * AND * (train* OR educat* OR teach* OR learn* OR curricul*) Study design and features will not form part of the search criteria of the realist synthesis since ‘nuggets’ of information [48] can be gleaned from quantitative, qualitative and mixed methods research as well as policy documents, opinion pieces and other grey literature. Social science and education databases are less well indexed than those used in medicine and health sciences, so a comprehensive and inclusive search has been constructed. The databases and platforms comprising the formal search are as follows: 1. Education Abstracts (EBSCO) 2. ERIC (EBSCO) 3. MEDLINE (OVID) 4. PsycINFO (OVID) 5. Web of Science Database Social Science Citation Index 6. Social Services Abstracts (ProQuest) 7. ASSIA Applied Social Sciences Index and Abstracts (ProQuest) 8. ClinicalTrials.gov 9. Database of abstracts of reviews of effectiveness 10. The Campbell Library 11. Cochrane Collaboration Library 12. Evidence for Policy Practice Information and Coordinating Centre (EPPI-Centre) 13. Google Scholar — using a series of searches and screening the first 5 pages of results for each search 14. ProQuest Dissertations and Theses Study selection will not be restricted by language, publication date or publication status. To supplement the systematic search, more emergent techniques of reference and citation tracking will be used, alongside contacting content experts (leading authors and researchers in the field) who will be asked to recommend additional empirical studies or other relevant grey literature. Documents will be screened by title and abstract (and full text if required) by the researcher and co-reviewer. Discrepancies will be discussed, and a consensus reached, involving other stakeholders if required. Background searching has revealed that studies tend to focus on one small aspect of the anticipated initial programme theory; therefore, inclusion criteria will be broad, typically containing information about the following: * Theories (substantive or informal) relevant to learning and teaching communication skills * Intervention components, e.g. videoing and feedback * Outcomes (including proximal outcomes) — knowledge, attitudes and values, self-efficacy and skills * Different student groups, e.g. undergraduate or postgraduate students * Information about the learning environment and/or relationships including the role of tutors and peers Exclusion criteria include the following: * Records that do not relate to an empirical study about a relevant intervention for the teaching and learning of communication skills * Studies that are not about students on social work qualifying courses * Studies that focus on learning in placements/practicums only It is expected that a series of additional searches will take place during the later stages of the review, particularly as gaps in the literature ascertained through the main systematic literature search are identified. Different inclusion/exclusion criteria will apply, and these will be devised iteratively with support from an information specialist. Literature from counselling training and medical education is not currently the focus of this review but may be drawn upon later, to elucidate further information about how particular intervention components work or to explore where similar mechanisms are in operation. Refinements to the programme theory will inform the nature of these additional searches and will be reported in full in the project write-up. The results of each search will be presented using PRISMA style flow diagrams. Step 3 — Selection and appraisal The researcher will review all of the documents that meet inclusion criteria during screening, assessing them against the two criteria of the following: * Relevance (i.e. whether there is any information contained within the documents which can be used to support, refute or refine the programme theory) * Rigour (i.e. whether the data is trustworthy — or not) Rigour can be difficult to apply in realist research because it is the ‘nugget’ of information [48] which needs to be assessed rather than the methodological quality of the whole study. For example, a methodologically weak study is less problematic if the relevant section for the review is simply a description of the programme’s components compared to using the findings of a study where the internal validity is questionable. Recognising that ‘different types of data will be subject to different judgements of methodological coherence and plausibility’, Duddy and Wong (2018) recommend recording the assessment of each piece of information [20]. This is the approach that will be adopted here to provide a clear and transparent audit trail. Identifying more than one source of data relevant to a programme theory is another strategy suggested by Wong (2018) [49] which will be adhered to at this stage, to enhance trustworthiness. The focus will be on finding sufficient data that is relevant, coherent and supports the aim of developing programme theory. It is intended that student social workers undertaking an evidence-based practice module will be involved in the selection and quality appraisal process, which will also be overseen and supported by the PhD supervisory team. Step 4 — Extracting and organising data All of the documents from every search will be uploaded into the EndNote reference manager software. Documents which meet the inclusion criteria will be copied into a separate group folder, into which the full-text PDF files will be uploaded. The researcher will extract the main characteristics from each document in the included study group, using a data extraction template. Data from each document will be coded according to the contribution it makes to the developing programme theory. Initially, data will be organised into broad ‘bucket codes’, based on the initial programme theory. The realist logic of analysis developed by Pawson and Tilley (1997) [50] will then be applied. As the data extraction process continues, and the programme theory is gradually and iteratively refined, the data will be recoded and organised into potential C-M-O configurations. The use of data to refine programme theory will be recorded and reported in the project write up. A 10% random subsample of documents will be checked by a second reviewer. Again, discrepancies will be discussed and brought to the attention of an academic acting as an independent adjudicator. Step 5 — Analyse and synthesise data Realist analysis and synthesis entail ‘juxtaposing, adjudicating, reconciling, consolidating and situating the evidence’ [16], with a view to refining the programme theory. In realist synthesis, the analysis and synthesis of the selected data in step 5 occur concurrently with data selection and appraisal in step 3 and data extraction and organisation in step 4. Through inductive and deductive reasoning, the researcher will move back and forth between the steps, using the data to build and test the CMOCs, iteratively refining the programme theory, as shown in Fig. 1. Additional searches will be conducted as gaps in the literature materialise or where other disciplines can inform our understanding of how particular mechanisms might operate. Stakeholders will be consulted about the development and refinement of programme theory, hopefully adding their own insights and amendments as they see fit. Retroductive reasoning will be used in the later stages, allowing the refinements to programme theory to be ‘made on the basis of what can plausibly be inferred by all the data available’ [20]. The final synthesis will provide an overview of some of the underlying causal mechanisms which are fired in specific contexts to produce particular patterns of outcomes. Dissemination The dissemination strategy will be developed with stakeholder involvement. Findings will be translated into evidence-based recommendations that can be shared with and applied by policymakers and educators. Findings will also be made available to students and experts by experience. Discussion The teaching and learning of communication skills in social work education is under theorised. This realist synthesis will provide theory-based explanations to determine to what extent the teaching and learning of communication skills in social work education work and for whom, how, why and in what circumstances. The findings from this realist synthesis will help policymakers and educators make informed decisions about the design and delivery of complex educational interventions aimed at improving the communication skills of social work students. One limitation of this realist synthesis is that it is being undertaken by a PhD student with no recourse to funding stakeholder involvement activities. Although undergradute students may readily take up the screening opportunities, resource constraints will inevitably have an impact on the extent to which stakeholder collaboration evolves. The PhD student’s supervisory team has content and methodological expertise, including conducting systematic reviews and realist syntheses. Their involvement will add further rigour to the conduct of this research. Another limitation concerns the state of the extant literature. Although there is a reasonable body of literature about the teaching and learning of communication skills in social work education, it is possible that there are gaps, particularly in terms of outcomes and contextual factors, which means some of the research questions might be unanswerable. A comprehensive search will be undertaken with citation and reference harvesting seeking to locate sibling and kinship papers. Authors will be contacted for further information, although now several appear to have research interests in other areas. Stakeholders will be asked to consider any identifiable gaps, which will add to the development and refinement of the programme theory, with the caveat that programme theory can only ever be partial and is of course open to further testing. Despite these limitations, stakeholder interest indicates that a realist synthesis still has much to offer. Immersion, meticulous data collection, systematic analysis and reflexive thinking are fundamental to the realist approach [15]. Transparency of methods and decision-making is an essential part of realist synthesis to ensure rigour and credibility [20, 51]. To assist this endeavour, the relevant quality and reporting standards and publication standards for realist synthesis will be followed [51-53]. The PRISMA-P statement (included as Additional file 1) has been used to develop this protocol. Abbreviations: CMO: Context, mechanism, and outcome; PRISMA-P: Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols; RAMESES: Realist and meta-narrative evidence syntheses: evolving standards Author: Emma Reith-Hall Date: Dec. 12, 2022
For this first draft, focus on getting all of your research written down. This is Step 1 of the Three Steps of History – gathering as much evidence as possible. Purpose: A first complete draft is a
Improving Citizen Engagement Through Human-Centered Design: Our Partnership with the New York City Department of Social Services The New York City Department of Social Services (DSS) serves three million vulnerable New Yorkers. The agency has a workforce of close to 18,000 and delivers a wide range of services from public assistance and employment services to a large homeless shelter system. In the Supplemental Nutrition Assistance Program (SNAP) alone, caseworkers process close to 70,000 applications and re-certifications per month. Like many social services agencies, the New York City DSS wanted to improve its citizen engagement process in a way that would reduce the burden on its staff while simultaneously enhancing the customer experience. Over the last three years, faced with rising caseloads and static resources in the SNAP program, the agency decided to invest in a new service delivery model. The goal was to shift the client interaction from a heavily bricks and mortar, in-person experience to a more self-directed service model using new digital channels, including an online portal that uses IBM Curam Universal Access. Called ACCESS HRA, the portal is designed to provide an improved client experience and to reduce foot traffic at their service centers. In addition to online applications for SNAP, the portal allows a client to view more than 100 real-time case-related data points, including case status, account balances, and e-notices. IBM Watson Health[TM] was well suited to partner with the city to address these needs, with 42 design studios around the world and 1,600 formally trained designers. The Watson Health Government Health and Human Services (GHHS) team is tackling the most difficult challenges facing our customers, applying human-centered design and agile development practices across our products and services, to enable us to greatly enhance the value we provide to our customers and their clients. Universal Access is a social enterprise management platform that is used to connect people to social services across five countries in eight different languages. Government services accessed with Universal Access run the gamut from disability benefits to income support to food assistance to health insurance. The GHHS team engaged in a partnership with DSS to help solve the department’s pain points–and do so in a scalable way that would maximize the value delivered across the entire GHHS customer base. In order to achieve this mission, the team agreed on three core values that would enable them to push the boundaries in terms of user research, product design, and software development: * Observe, listen, understand * Build, measure, learn * Influence over control Observe, Listen, Understand The DSS invited the GHHS team to use their PC banks, available at 15 service centers, as user-feedback labs to help gain a better understanding of people’s needs as they interact with the government online. A multidisciplinary team comprised of a product manager, designers, and architects traveled to New York to spend time with people as they went through the process of applying for benefits at these service centers. The team observed, listened, and very clearly understood their needs as a result. So, what did they learn? That all kinds of people find themselves asking for help. Some don’t speak English, some have low literacy levels, others have to usher tired and hungry children with them to the service center. However, despite their diversity, their needs are common and relatively simple: * They need an easy, clear path to help. * They need the ability to choose the channel they used to interact with the agency, including smartphone, desktop, in person, or on paper. * They need to know upfront what was involved in the application process and get answers to questions such as, “How long is this going to take?” and, “Can I save and resume this process later, or should I not start at all if I can’t stay for the duration?” * When they are finished with the online application, they need to know what comes next in the overall path to help. Understanding these needs, which become obvious only through firsthand observation with the users, helped the GHHS team discern the kind of user experience the product designers needed to create to meet these needs. Build, Measure, Learn These insights led to the creation of a prototype of a transformed citizen engagement experience, accessible from any device. The prototype was brought to New York for feedback. The team then iterated that prototype to tweak and adjust the user experience as the solution progressed. The DSS team (a mix of public- and private-sector business and tech experts) also traveled to an IBM design studio where, together with the GHHS team, they co-designed elements of the new experience. Taking this approach resulted in a very tight collaboration, with public and private teams working shoulder to shoulder throughout the process. Influence Over Control The solution was to build a design system that incorporated government-specific design principles and guidelines as well as templates, examples, and a set of components that could be assembled to build the custom experience. And, unlike many other design systems used mainly for in-house development, the team delivered that design system along with the product for customers and partners to use, achieving that influence over control. Looking Forward Investing in good design practices and taking a lean, agile approach to development within the context of a public-private partnership meant that just months after defining the first prototype of these ideas, the GHHS team has designed the next generation of Universal Access. This next-generation product has already been validated by its users, and later this year, a simple, modern responsive version of ACCESS HRA will be released to New Yorkers. Author: Claire McLoughlin Date: June 2018
For this first draft, focus on getting all of your research written down. This is Step 1 of the Three Steps of History – gathering as much evidence as possible. Purpose: A first complete draft is a
Get ready … get set … go! A new entrant programme for social workers in mental health. INTRODUCTION: Previous research has found that many social work new graduates do not feel that they come to their work in mental health with sufficient knowledge. One way of remedying this is using post-qualifying New Entrant to Specialist Programmes (NESP), where social workers gain additional knowledge and skills in a particular field of practice. METHODS: A small-scale research project was conducted with five social workers who had recently completed a NESP qualification within a district health board (DHB) setting in Aotearoa New Zealand. Semi-structured interviews were used to obtain deep information from participants in line with a subjectivist perspective. FINDINGS: Participants indicated that they had found the NESP qualification useful in improving their confidence and competence in their early practice years. They found the programme was useful in decreasing a sense of isolation by bringing them together with other social workers, or members of other disciplines, and they appreciated the skills and support they had received from their group supervision experience. The greatest challenges revolved around time constraints and they wished for more social-work-specific and cultural content within their programmes. CONCLUSIONS: The programmes appear to serve an important function in improving competence and confidence for social workers new to the field of mental health. Social work’s commitments to social justice and Te Tiriti o Waitangi need to continue to be pulled through into post-qualifying programmes and a sense of professional identity needs to be nurtured within medically oriented settings. KEYWORDS: New entrant programme; social work; mental health; education Social work education in Aotearoa New Zealand is in a constant state of flux. The recent development of the mandatory registration of social workers through amendments to the 2003 Social Workers Registration Act has meant that all registered social workers educated in Aotearoa New Zealand are required to have graduated from either a prescribed 4-year undergraduate or 2-year postgraduate qualifying social work qualification in Aotearoa New Zealand (SWRB 2021), or be assessed as having a substantially equivalent overseas qualification to an Aotearoa New Zealand prescribed qualification. There is currently a review being undertaken of the 2016 Social Workers Registration Board (SWRB) Programme Recognition Standards (SWRB, 2022a) and, in 2023, all of the programmes being delivered within the polytechnic sector will be amalgamated into one qualifying programme (Te Pukenga, 2022). There has been ongoing tension, both in Aotearoa New Zealand and overseas, regarding the aims of a basic social work qualification vis-a-vis seeing it as an end in preparing social workers for employment, or as a step in a process of ongoing development and training (Beddoe et al., 2018). One of the suggestions bridging these views is to enable new graduates to continue their development through a programme of ongoing post-qualifying academic programmes, while working in a particular field of practice in an environment with supports such as intensive supervision, a reduced caseload and time given for study. New graduate social workers entering mental health practice often have to grapple with the tensions of working in this medically dominated setting (Brown, 2021). There can be conflict between social work values aligned with recovery-oriented practice, social justice and empowerment, and the demands of clinical practice that can be deficit-focused, risk-averse and coercive (Davidson et al., 2016). Brown (2021) has explored the impact of neoliberalism and managerialism on mental health services, and the negative impact this has had on social work professional identity, along with the co-option of social workers into the dominant medical discourse and away from social justice. Brown (2021) suggested that work is needed to further develop critical clinical social work to resist these inequities. This approach unpacks power, both for clients, and for social workers within hierarchical service contexts. The challenge remains for new social workers to navigate these challenges (Smith, 2018). This article explores a New Entry to Specialist Practice (NESP) programme for social workers entering the field of mental health within a district health board (DHB) and attempts to answer the research question: “What helped and what hindered your experience and progression through the NESP programme?” We have written this article with several audiences in mind. We believe that the findings may be useful for organisations who are considering the development of new entrant programmes, for education providers who are considering provision of post qualifying programmes and students/new social workers who are considering undertaking a new entrants’ programme. The article may also be useful to funders in considering contract requirements for funding of new entrant programmes. A brief history and context of this initiative will be provided followed by a review of the literature to provide an initial summary of some of the previous research in regard to new entrant programmes, particularly what has been shown to help or hinder students/new employees experiences of these programmes. Data from the literature review will then be enhanced by the findings of a small scale research project within a DHB conducted with three graduate cohorts of a NESP programme through 2020 and 2021. Finally, a discussion will explore some of the implications of the findings in the current and future environment of social work education and practice. Context Social work qualifications are prescribed by the Social Workers Registration Board who stipulate the academic standards required for social work programmes (NB: currently in review). There is an expectation that all social work programmes deliver from a generalist social work orientation, with specialisms to be considered post-qualification (Hunt et al., 2019). This is consistent with there only currently being a General Scope of Practice being defined by the SWRB. In 2016, a research team of social work academics came together to develop an evidence-based capabilities framework for social work practice. This research occured in phases and included consideration of content in social work programmes, and the experiences of new graduates and employers around readiness to practise (Ballantyne et al., 2019a). Fourteen of the 17 social work programmes in Aotearoa New Zealand took part in the study, which undertook a curriculum mapping project. Of the 402 course descriptors submitted for analysis, only six courses contained the key term of mental health. The report also indicated that “Knowledge of mental health, addictions and trauma has also been identified as a curriculum area of concern” (Ballantyne et al., 2019a, p. 23). In phase two of the study, looking at preparedness for newly qualified social workers, 72% of participants indicated that mental health conditions and their likely progress was a specialist area within their current practice (3rd highest rating), while only 44% indicated that they knew as much about the area as was expected of them (Ballantyne et al., 2019b). The research seems to indicate that there is a greater demand for more mental health content in qualifying social work programmes. With limited space in already overcrowded qualifying programmes, the NESP programme aims to respond to this need. NESP programme The academic component of the NESP programme was developed in 2003, in response to research by the Ministry of Health that identified lack of support and mental health knowledge as factors in poor allied mental health graduate staff retention (Pack, 2010). There were low numbers and high turnover of social workers and occupational therapists in mental health services, and so an academic programme was developed to further educate social workers and occupational therapists in mental health during their first year of practice. Victoria University of Wellington (VUW) delivered this Postgraduate Certificate in Health (Allied Mental Health) from 2003. Margaret Pack coordinated the NESP programme at Victoria University from 2007. She used problem-based learning to mimic the multidisciplinary team to help social workers to develop sound clinical reasoning skills and the confidence to flexibly engage in day-today clinical decision making (Pack, 2010). The programme ceased being delivered by VUW and a new contract was initiated with Auckland University of Technology (AUT). AUT’s Postgraduate Certificate in Mental Health and Addictions began in 2010. Both of these programmes have focused on adult mental health. The current AUT programme is taught via block course and covers mental health assessment, intervention, and recovery-oriented practice, with a focus on discipline-specific professional identity development. The programme initially took in social workers and occupational therapists but, since 2018, nurses have also been included. There are lecturers representing each of these disciplines teaching on the programme. There is also a NESP programme that is specifically for those working in child and adolescent mental health services (CAMHS). The Postgraduate Certificate in Health Science (Child and Adolescent Mental Health) is offered by the University of Auckland I Waipapa Taumata Rau. This is delivered via block course and online learning, covering child development, psychopathology and clinical skills in assessment. Trainees on the programme are social workers, occupational therapists and nurses, and the lecturers on the programme have predominantly been psychologists from within the Department of Psychological Medicine. Both the general mental health programme at AUT and the CAMHS programme at the University of Auckland are funded by Te Pou. Funding includes tuition fees, travel and attendance at block courses (Te Pou, 2022). Trainees complete the study while working in mental health services, with assignments based on their work. Workplace support The funded academic NESP programmes are available for social workers from DHBs or NGOs who are in their first two years of mental health practice. There are wider eligibility criteria for those who self-fund. There is variation between the DHBs regarding what other supports are provided. Currently there are three DHBs that have NESP Coordinators who provide group supervision and clinical support to staff on the programme. The research described in this article was conducted within one of these DHBs. At the research site, the DHB had mandated that all social workers who were new to mental health practice (whether new graduates or experienced in another field of practice) were to undertake study within the NESP programme. Most of the trainees secured Te Pou funding, and those who did not were funded by the DHB to complete the programme. Social workers on the NESP programme were provided with fortnightly or weekly study days, regular individual supervision, group supervision, and a mentor who provided feedback on their practice. The NESP Coordinator (J. Appleby) advocated for a developmental approach to caseload allocation, with NESP social workers starting with about 50% of a full caseload which increased over time as their knowledge and confidence increased. Literature review Being a new graduate is challenging Across disciplines, it is hard to be a new graduate in mental health (Gunning et al., 2019; Spence et al., 2011), particularly for social workers (Agllias, 2010; Beddoe et al., 2018; Hay et al., 2017; Pack, 2009). The first year of practice can be frenetic, learning to navigate workplace processes and practical matters such as office access, dress codes, parking and room bookings (Donnellan & Jack, 2015). In the Aotearoa New Zealand context, social worker trainees have a generalist education, focusing on foundational social work skills without specialising in a particular field of practice. Specialist knowledge is then gathered during their first years of practice, learning on the job, often in informal ways (Beddoe et al., 2018). There are additional challenges for new graduates working in the field of mental health. There can be a wide gap between qualifying knowledge and mental health expertise, and new graduates may lack confidence and feel shame in the workplace (Pack, 2009). Mental health social workers can experience role conflict with few opportunities for professional decisionmaking and autonomous practice in under-resourced mental health services (Acker, 2004). This is complicated by social workers in Aotearoa New Zealand having a strong awareness of a hierarchy of professions in healthcare services, where social work is often lowly rated (Beddoe, 2013; Keen, 2016). Social workers often struggle to establish their professional identity within DHB mental health teams dominated by a medical model at odds with social work values (Acker, 2004; Appleby et al., 2020; Beddoe, 2013; Brown, 2021, Pack, 2010). Developmental approach Support for new graduates needs to be formalised to assist social workers as they transition from study into the workplace (Moorhead et al., 2020). This formalised support should be underscored by a developmental approach to new graduates, viewing them as developing practitioners working towards expert status as they gain skills and knowledge over several years (Beddoe et al., 2018; Hay et al. 2017; Moorhead et al., 2020; Spence et al., 2001). There can be tensions between social work educators and employers regarding expectations of new graduates, and Moriarty et al. (2011) argued that it is the distinction between viewing qualifying education as a developmental process versus an end product that fuels these tensions. This support should also be responsive to emerging needs, such as increasing caseloads commensurate with ability and confidence (Spence et al., 2001). It is important to provide a supportive environment with non-evaluative feedback to assist social workers to develop their clinical reasoning (Ladyshewsky, 2010). There is a formalised support and development programme for new graduates in social work in England (Moorhead et al., 2020). The Assessed and Supported Year in Employment (ASYE) assists the development of newly qualified social workers, including support for fewer and less demanding cases to be assigned in the first year of practice (Beddoe et al., 2018). Suzanne Patterson has undertaken research with social work and occupational therapy NESP trainees in Aotearoa New Zealand who completed the NESP programme between 2015 and 2017 (S. Patterson, personal communication, February 24, 2022). The research findings are yet to be published, but preliminary data show that participants have indicated the importance of feeling supported in making the leap from student to clinician, and to be recognised by their team as a novice in this learning period, affirming the developmental approach. Supervision Good supervision is essential for social workers as they adapt to mental health practice, and makes a difference to the work satisfaction of new graduate social workers (Agllias, 2010). It is the responsibility of the supervisor to create a safe and nurturing supervision environment for reflection, particularly given that shame is a common experience for new graduates (Pack, 2009; Spence et al., 2001). New graduate social workers are often looking for help to apply theory to practice within supervision (Beddoe et al., 2020). Ideally, clinical supervisors are available, knowledgeable and can connect practice to theory, support professional growth, serve as a professional role model, and communicate in a collaborative and warm style (Bogo & McKnight, 2006). Clinical supervision needs to be separate from managerial line supervision and allow space for reflective supervision (Beddoe et al., 2020). Pack (2015) investigated the clinical supervision experiences of early-career mental health social workers and occupational therapists on the NESP programme. In that research, 12 NESP supervisees were asked how they had resolved their most challenging cases or issues in clinical supervision. The main themes that emerged concerned the importance of ongoing peer review and critique within a trusted supervisory relationship, addressing transferences with clients and colleagues, identifying training needs, and exploring self-care. The importance of the supervisory relationship was highlighted as a supportive structure to facilitate these resolutions. Group supervision Group supervision is provided by NESP Coordinators in some DHBs. There are some elements of group supervision that differentiate it from individual supervision (Beddoe & Davys, 2016). Members can learn from each other with a greater exposure to a variety of cases and perspectives. It can also be reassuring for members to see that others are in similar positions to them, particularly as new graduates. There can be several challenges for group supervision which centre around the impact of group dynamics. Unhelpful dynamics can include problems between members, poor group time management and competitiveness for time, potential for domination by one person, irregular attendance, lack of structure and a focus on tasks over group process (Beddoe & Davys, 2016; Enyedy et al., 2003). In thinking about group structure and process, Beddoe and Davys (2016) suggested that it can be helpful to think about the stages of group development, such as Tuckman’s (1965) model of forming, storming, norming and performing. It is also important to have a supervision contract and working agreement for the group and to set a session agenda each time (Beddoe & Davys, 2016). Staniforth and McNabb (2004) have written about the development of a supervision group for NESP social workers in a DHB. The group began in 2001 and met monthly, facilitated by the two authors. The group supervision included checking in with members, didactic presentations and training, and case presentations by participants. This was offered in addition to existing individual supervision arrangements. Staniforth and McNabb (2004) have reflected on this process and highlight the importance of contracting clearly at the beginning, ensuring individual supervisors were aware of the purpose of the group, sticking to task, including cultural components, and attending to group process issues alongside content. Methodology The research was conceptualised and conducted by the authors who both subscribe to a constructivist ontology–the view that knowledge is created subjectively by those engaged in interpreting and producing it (Bryman, 2015). A qualitative methodology is congruent with this view, seeking to engage with participants and develop a deep understanding of their perspective in a particular area. The researchers in this case were both located at different points of an insider research approach. Staniforth supervised a group of social workers while employed at the DHB in the early 2000s, then provided individual clinical supervision of students/social workers in the programme through 2004-2010 and then finally was contracted to provide social work and therapeutic perspectives within one of the educational programmes from 2012-2020. Further, some of the NESP participants had also graduated from social work programmes where she was employed as a lecturer. Appleby undertook the NESP programme herself in 2007, was employed as the DHB NESP Coordinator from 2019, facilitated the supervision group (other than when on parental leave in 2021) and more recently has been contracted to provide social work education and support within one of the academic programmes affiliated with the NESP programme. Insider research provides multiple perspectives and some of the ethical considerations of this will be discussed further later. The research was approved by both the Auckland Health Ethics Research Committee and the ethics committee of the participating DHB in 2020, and covered three cohorts of students who completed their programmes throughout 2020 to 2022. An initial literature review was conducted in 2020, which informed the interview questions. Participants were recruited via an email sent by Appleby as NESP Coordinator, inviting them to make contact with Staniforth if they wanted further information, or to participate. Participants were then interviewed by the first author and were given the opportunity to review and amend their transcripts. Identifying information was removed by Staniforth and analysed thematically (Braun & Clarke, 2006) through use of an NVivo software package. Themes and deidentified data were then made available to Appleby for joint write-up of the project. We interviewed five participants. Unfortunately, the research ocurred through the various phases of Covid-19, where health services were stretched to the limit and new graduates at the time were likely impacted by increased demands on their time. The research also occured on the back of completion of the NESP programme which had placed increased time pressures upon them. All participants had recently completed the NESP programme and, due to the very small pool of potential participants, demographic information is not presented to maintain privacy. Four of the participants had undertaken the child and adolescent mental health education qualification while only one of the participants had engaged in the more generalist mental health programme. Ethical implications The researchers’ various roles associated with NESP needed to be considered carefully. As Appleby had current and ongoing involvement in the programme, the perceived possibility for coercion or power differential was minimised through her not interviewing candidates and not being made aware of who had (or had not) participated due to deidentification of the interview data. Participants were provided information by her, but all further communication occured via Staniforth, with participants needing to “opt in” to the research. While participants were made aware that their identities would be confidential, they were also informed that, due to the small pool of potential participants, there was a possibility that someone may be able to identify them through comments that they made. It was also a requirement from the participating DHB that the specific DHB would not be named, however there is a possibility that it could be identified due to the researchers’ affiliation with it. Limitations of the research This is a small-scale research project and, as such, the results need to be taken in that context. Initial findings however, indicate that a more in-depth study of the area may be useful. As discussed, both researchers have been heavily involved in the NESP project. There was some tension between our insider knowledge about the programme and of the participants which needed to be balanced with wanting to maintain an effective inquisitive researcher stance. We hope that this has been resolved somewhat through use of the participants’ voices and allowing their experiences to shine through. Findings Findings from the research interviews are loosely based along themes with responses to particular questions in the interviews and are provided within the following subthemes: expectations for the programme; the benefits of undertaking NESP; what worked well in the process; what got in the way and suggested changes; and advice to others considering engagement in a NESP Programme. Expectations for the programme Some of the participants had a very clear idea of what the purpose of NESP was, their expectations and the goals that they hoped to achieve. The programme was a condition of employment for all the participants, and one of the participants had not given much prior consideration to it. I just thought it was something that, you know, new grads had to do and it was just a process of the DHBs bringing in new graduates who would have just finished their programmes …. So, prior to doing NESP I didn’t know what it was. For others, the purpose was very clear. I think in this job, as well having a sense of mastery like you actually know, because it is such difficult complicated work, sort of going in with not having some specialist skills is quite daunting and you could feel like … “I don’t know what I’m doing, I’m not being helpful”. So, I think it [NESP] gives some structure as to what the work is. So, I think as much as we talk down to young people that we work with about gaining a sense of mastery and how important that is, I think it is also really important for us as social workers, as clinicians to also have that sense of mastery to know what we are doing, to have a structure around us. So, I think it provides us with that. One of the participants did not have specific goals prior to the programme, while others had specific areas in mind that they wanted to focus upon. So I think my goal was to really make it meaningful to the work that I was doing and to have it beneficial, to not only my practice, but to people that I was seeing. Building knowledge and confidence Participants noted various benefits of being involved in the NESP programme, which were considered in response to the question “What was the best thing about the NESP programme?” Two areas were identified across this question, with some consistency. These included improved confidence in participants’ work role and getting to meet with other people engaged in NESP or through their postgraduate courses, particularly getting to work with people from other disciplines. Participants shared how the increased mental health knowledge positively impacted on their confidence in the workplace: I think my confidence definitely increased … first I had worries “how do I present this person in MDT?” or “how do I do this person’s assessment”? I think now I am more familiarised with words and the structure of assessment and knowing what resources are there makes me more confident. … it made it less overwhelming to work in a space that is A) very fast paced and B) it is a medical model which social work doesn’t teach you a huge amount about, just that it is a very medical model and being a social worker is hard in that space. Participants also spoke about the benefits of networking with others. Two participants spoke about how they enjoyed meeting people from other disciplines on the programme. Best things about the NESP programme is the networking. I loved the networking and being able to make new friends. The best things about the NESP programme–learning new things out of my bachelor’s degree, supervision group. I think that was really the best thing about it and knowing different new grads from different professions. Group supervision supports NESP success The most consistent response to the question, “What helped to succeed in the programme?” was participation in the supervision group. Four of the five participants engaged in the supervision group and felt that it was essential in supporting them through their time in NESP. We would do like group supervision altogether. So honestly, I think that was essential as well because it meant that you could problem solve as well and talk about the assignment or to share ideas and eventually, we could all come together for supervision which I think was pretty essential to doing the paper. I think it would be really hard doing it in isolation and not having that comradery to support you through it. We covered I think a good balance of the content and academic side of things and like managing the workload and systems to choose all of that type of thing, [they] have been a very supportive group to actually keep going and get the work done and also supporting one another. I think people were worried about passing the papers, rather than thinking about how we reflect in supervision group. So we kind of started out more procedural and then people started to open up a little bit about their own experiences. Participants also spoke about the role of the group supervision facilitator. They noted it was helpful for the facilitator to liaise with team managers, to have previous experience of working in the teams, and to be available in between supervision sessions. It was very helpful to have someone who was very experienced and us bringing our stuff to supervision and they were able to support us with it … if there was anything we found hard our coordinator was actually able to reach out to our team managers and also the uni people and ask questions and they actually got things done. I think having somebody who like for [group facilitator], she used to work here and she had really good insight into the practicality of the day-to-day job and what that looks like and trying to navigate all those kind of things, and actually having her experience and expertise was really helpful because we were all new to the workforce and to the course and trying to balance the two things was quite messy a lot of the time. I think [group facilitator] was really, really supportive in our process. Like we were really easily able to contact her on email and she would respond really quickly and if we had some problem she would really guide us. In discussing the academic programmes, participants identified that they liked assessments that related directly to practice and involved linking theory to practice. Examples were case studies of assessment and intervention and students recording themselves conducting assessments. Other things that participants identified as helpful were having realistic expectations of themselves (“I don’t have to get A+ all of the time”), establishing some work life balance and having flexibility with how they took study leave (“taking one day a fortnight rather than a half day a week”). The need for supportive work environments The findings here are presented according to areas that participants indicated could be improved, with their corresponding suggestions for improvements. The suggestions for improvements from the service included protected regular study leave, capped caseloads and better coordination between the service and the universities. Participants were all provided with paid study leave, but the frequency of this was variable between services, and there were challenges in taking this time. It was also difficult to manage work and study commitments. Some people faced challenges with getting their study leave approved by their manager, which was connected with service pressures. Service under intense pressure–high turnover, had to do too many things, too early in career and was isolated. Social workers were also meant to have capped caseloads during the NESP programme, although some participants reported that this was not their reality. They reported extra stress on services as a result of Covid-19, which impacted on their caseload numbers. Having set time off and maybe even having a capped caseload would have been helpful as well for us to be able to focus, to perform well in the academic programme as well. Participants suggested that better communication was required between different parts of the programme, particularly regarding the academic workload expectations. There are no check ins with the organisation, like my manager, I mean I’m sure she knows I’ve passed, but no check in, no idea, you know. I think of the work that you do, all the assignments or actually what you are doing or the course work, maybe she does, but there is no conversation about it. It does feel quite, I mean I am linking it, but it feels quite separate actually in practice. Participants also had recommendations for the academic programme, including increased focus on cultural competency, and having “more lectures from social workers” and “having more of a social work lens for the CAMHS NESP students in the academic programme”. Discussion Beddoe et al. (2018) and Ballantyne et al. (2019c), have identified some of the tensions which exist in relation to expectations of new graduates to come “work ready” to the field. While this is in no way limited to the area of mental health, the Ballantyne et al. (2019b) R2P research identified that 72% of graduate participants stated they required mental health knowledge in their post-qualifying positions with only 44% feeling that they were adequately prepared upon completing their qualifying programme. The authors have argued elsewhere (Appleby et al., 2020) that a post-qualifying specialist scope of practice may be one way forward, and the NESP programme provides one model for how that may be considered. While many of the social workers undertaking the NESP programme are new graduates, a number have also come to mental health from other fields. For those social workers interested in moving into mental health, or for new graduates who do not have a NESP programme available to them, there is an issue about how they may become experienced enough in mental health to enable them to be employable in the field. This is an area that could benefit from further NESP-type training becoming available, particularly at times where there is a worker shortage that has been identified (McConnell, 2022). Much of the focus of the NESP programme is on mental health assessment. While there is some broad consideration of interventions and recovery-oriented practice, particularly within AUT’s adult mental health programme, there is no focused training on specific intervention approaches and therapeutic modalities from a social work lens. As we have argued elsewhere (Appleby et al., 2020), this is also an important component of a post-qualifying clinical scope of practice. The NESP programme is supported by government funding via Te Pou. This level of investment into a post-qualifying programme is crucial for its success. However, structural strengthening of the social work role within DHB mental health services is necessary to retain social workers in these settings. There are currently structural disincentives including low pay, increased caseloads and limitations on the use of professional judgement, that we suggest have contributed to significant social work vacancies in mental health services around the country (Peters, 2022). This has added stress for social workers learning the role in teams without the benefit of senior social work presence. Mental health social workers are faced with much uncertainty, including ongoing and prolonged pay negotiation between the DHBs and the union, imminent changes to the health structure with the introduction of Health New Zealand and the Maori Health Authority, and the ongoing impacts of Covid-19 on a stretched healthcare system. The current research findings and existing literature illustrate how challenging the NESP year can be, managing study and work, learning workplace processes, mental health knowledge, and developing social work professional identity. More recently, there have been new graduate social workers who completed all their practicums during the pandemic, and now are working and studying while navigating self-isolation and online learning. Working remotely and having smaller teams onsite has meant that there are fewer opportunities for informal support and debriefing from colleagues. The NESP programmes were developed with the expectation from Te Pou and education providers that trainees would have capped caseloads, regular clinical supervision and be able to take allocated study time. As the findings showed however, many students struggled to access these conditions. These, and other challenges futher highlight the need for managers and supervisors to create nurturing spaces for new graduates in mental health. Allowing space to be learners may alleviate some of the stress of the role and contribute to retention of social workers in mental health. The added challenges of Covid-19, health system reform and ongoing pay equity negotiation challenges highlight the importance of having ringfenced funding and protection to support new graduates with capped caseloads and study leave as they gain experience and develop skills. The changes to the health system in Aotearoa New Zealand are intended to address inequities in healthcare provision. This is particularly important for Maori health, given the higher rates of mental distress for Maori (Ministry of Health, 2018), use of the Mental Health Act (Elder & Tapsell, 2013), Maori experiences of racism within the healthcare system (Graham & Masters-Awatere, 2020), and the important role of anti-racism work as a suicide prevention strategy (Clark et al., 2011). It is important that all social workers have a strong grounding in the application of Te Tiriti o Waitangi within mental health and awareness of culturally responsive assessment and treatment models for Maori. The Maori Health Action Plan (Ministry of Health, 2020) sets a vision of reducing health loss for Maori by valuing Maori solutions to achieve Maori aspirations and wellbeing. This is also needed in the NESP programme. Participants spoke about the need to include cultural considerations of mental health in both the academic and the group supervision components of the NESP programme. Monocultural teaching and supervision continues the colonial legacy that has already brought so much inequity to mental health service provision (Government Inquiry into Mental Health and Addiction, 2018). A critical clinical social work approach brings this awareness of power and equity to mental health services (Brown, 2021). While it may not hold a privileged position, social work is much-needed in mental health services to work alongside people, understanding them in their family context, culture and life experiences. Social workers have an important role to bring a social justice and advocacy lens to clinical work, challenging the medical model of mental health and service structures that are not person-centred (Appleby et al., 2020). Canadian social worker Vikki Reynolds (2019, pp. 37-38) explains: “we need to prioritise being person-centred teams and not staff-centred teams. We want to be in the Zone of Fabulousness–because we are fabulous. Fabulousness happens when we enact collective ethics, and are collaborative, innovative, and justice-doing”. Having social work input into mental health training and supervision can help new social workers to develop their professional identity, to resist dominant, deficit-based mental health paradigms, to integrate their clinical and social justice work, and to recognise and celebrate the fabulousness of the social work approach in mental health. Conclusions Social work education in Aotearoa New Zealand occurs in a multiply contested space where employers hope for work-ready graduates to emerge from generalist social work programmes. In the field of mental health, the government has recognised the need for social workers to be able to access more specialised education postqualification and has funded New Entry to Specialist Practice programmes which combine academic and practice components in teaching values, skills and knowledge required to work in mental health settings. This article has presented some of the growing body of literature in this area as well as the context of the NESP programme in Aotearoa New Zealand. Results from a small-scale study of participants who had completed the NESP programme demonstrate that participants found the programmes helpful in building their confidence and competence in the field of mental health and that they were particularly appreciative of the experience of decreasing isolation through coming together with other new entrants (social workers as well as other disciplines) and the learning and support that they had taken away from group supervision. Participants also indicated an interest in having a stronger discipline focus in some of the training and more access to training from various cultural perspectives. The researchers call for ongoing demonstration of consideration of te Tiriti o Waitangi, critical consideration of power and oppression in the field of mental health and having awareness and pride in a social work identity. Authors: Barbara Staniforth and Jo Appleby Date: Dec. 2022
For this first draft, focus on getting all of your research written down. This is Step 1 of the Three Steps of History – gathering as much evidence as possible. Purpose: A first complete draft is a
RESEARCH LEARNING ATTRIBUTES OF GRADUATE STUDENTS IN SOCIAL WORK, PSYCHOLOGY, AND BUSINESS Abstract:  Although research teaching literature suggests that social work students may have different research learning attributes than students in other disciplines, minimal data exist to support this view. This study compared the self-reported research anxiety, computer anxiety, and research orientations of 149 full-time graduate social work, psychology, and business students at a research university. Social work students reported more research and computer anxiety and generally believed that research was less important to their profession than students in the comparison groups. Implications for teaching research to social work graduate students are discussed. WHEN SUMMARIZING OBSERVATIONS and reactions to having taught the required research and statistics courses, social work faculty have consistently emphasized students’ low levels of preparation, motivation, and achievement (Bogal & Singer, 1981; Epstein, 1987; Forte, 1995; Grasso & Epstein, 1992; Smith, DeWeaver, & Kilpartrick, 1986). Irwin Epstein, for example, a 30-year veteran of teaching research courses, describes his students as “research reluctant” and observes that “no other part of the social work curriculum has been so consistently received by students with as much groaning, moaning, eye-rolling, hyperventilation, and waiver strategizing as the research course” (p. 71). In response to faculty observation and experience, a significant proportion of social work’s research teaching literature has been devoted to methods and techniques for encouraging and motivating students, as well as helping them manage their anxiety about learning, consuming, and conducting research. This literature encourages research faculty to anticipate the importance of student research reluctance in the planning and teaching of their courses. Accordingly, faculty are advised to emphasize group process variables in their teaching and presentation of research content, to facilitate and model humor and self-disclosure, to focus their courses on the practical rather than the theoretical, and to replace the traditional deductive emphasis of research teaching with more inductive experiential approaches to student learning (Epstein, 1987; Forte, 1995; Glisson & Fisher, 1987; Kirk & Kolevzon, 1978; Rabin, 1985; Royse, 1999; Royse & Rompf, 1992; York, 1997, 1998). Overall, the consistency of these descriptions of student learning and prescriptions for teaching suggest that social work students may be unique when compared to students in other professions and disciplines. Surprisingly, however, only a single study has systematically compared the research learning attributes of social work students to those of students in other disciplines and professions. In this study, David Royse and Elizabeth Rompf investigated whether the reluctance and anxiety they had observed among their BSW students exceeded that experienced by undergraduate students in other disciplines and professions. It was their view that undergraduate social work students, when compared to students in other disciplines, experienced more math anxiety or “feelings of tension and anxiety that interfere with the manipulation of numbers and the solving of mathematical problems in a variety of ordinary life and academic situations” (Royse & Rompf, 1992). To test this proposition, they administered the Mathematics Anxiety Rating Scale-Revised (MARS-R) to a sample of 419 undergraduate students enrolled at a large state university on the first day of class. The social work sample was comprised of 163 undergraduate majors enrolled in either statistics (n=54) or research methods (n=109) courses. The comparison group (n=256) included undergraduate students from a variety of other majors who had enrolled in an undergraduate statistics course in the same university. Royse and Rompf hypothesized that the social work students would report (1) higher levels of math anxiety on the MARS-R, (2) having completed fewer mathematics courses, and (3) more bad experiences in math courses when compared with reports of students in the undifferentiated degree group. Although the study revealed no differences in the number of bad experiences reported, the social work undergraduates, as predicted, had higher scores on the MARS-R and reported taking fewer math courses than the undifferentiated undergraduate majors. Royse and Rompf’s study affirms and begins to provide data-based support for the continuing use of specialized research teaching techniques for undergraduate social work students. There is no similar empirical evidence, however, to suggest that these techniques are also more applicable to social work graduate students than to graduate students in other fields and disciplines. Indeed, it is conceivable that differences in learning attributes observed at the undergraduate level between social work and other students may dissipate over time due to cohort maturation and selection processes. Because elevated research expectations almost singularly distinguish the American graduate curriculum from the baccalaureate in the social and behavioral sciences, as well as related professions (Boyer, 1990), the graduate student cohort in social work might very well be comprised of substantially different kinds of learners than social work students at the baccalaureate level. If this is the case, the specialized teaching procedures advocated in the social work literature may be less applicable to social work graduate students than it is to undergraduates. Further, deductive teaching methods, traditionally associated with graduate education in the United States, may be more applicable in graduate social work education than in undergraduate education in social work. In fact, if social work graduate students on the whole do not exhibit the research reluctance and anxiety reported by Royse and Rompf’s undergraduates, the use of the specific teaching procedures and anxiety reduction techniques advocated in the literature may be unnecessary and even counterproductive to effective teaching at the graduate level. Study Description This study extends Royse and Rompf’s work by focusing specifically on MSW students’ attitudes and anxieties about research and their research courses. Unlike Royse and Rompf, who assessed math anxiety exclusively, this study assessed more general attitudes toward the value of research within social work and anxieties associated with computers and other research activities and requirements. In the study, MSW students were compared to three purposefully assembled comparison groups of graduate students. In the first group of analyses, MSW students were compared with a mixed group of graduate students in either psychology or business. Then, to specifically compare measures completed by the MSW students with those completed by graduate students in a related human service profession, the MSW student reports were compared with those of a group comprised exclusively of graduate students in clinical or counseling psychology programs. Finally, the third analysis compared the MSW students’ attitudes and reactions solely with those of candidates for the MBA and MS degree in business, a group generally pursuing careers outside of the human service professions, and therefore frequently recruited as a contrast group in studies of social work occupational choice (Feld, 1987; Marsh, 1987; Moran, Frans, & Gibson, 1995; Segal, 1992). Methodology Sample The study’s participants were 149 full-time, continuing students in good standing enrolled in one of three graduate programs at Virginia Commonwealth University, a large, urban research institution located in Richmond, Virginia. Sixty-two (42%) of the study’s participants were in social work, 49 (33%) in clinical or counseling psychology, and 38 (26%) in business. All social work students were clinical concentrators completing the second year of an MSW program and were enrolled in one of three different sections of their third and final required research course. These students were completing the fieldwork on their required research project. To provide comparability with the social work clinical concentrators, only psychology students pursuing degrees in either a clinical or counseling psychology program were included. The business students included in the study were completing requirements for either MBA or MS degrees in business at Virginia Commonwealth University’s School of Business. Although housed in different schools or colleges within the university, all three programs offer the PhD as well as the master’s degree, and students admitted to all three programs are subject to the admission requirements of Virginia Commonwealth University’s Graduate School. The 62 social work students who completed the three page instrument represented 83% of all students enrolled (n=75) in three sections of the required second year clinical research course sampled. The psychology and business students in the study responded to a mail survey which was administered with a follow-up letter and one replacement questionnaire. Forty-nine (64%) of 77 psychology students and 38 (53%) of 72 business students completed and returned questionnaires. Comparison of the three groups of students on age, minority composition, and marital status failed to reveal any statistically significant differences (p [is less than] .05). The mean age for the study was 28.39 years; means for the ages of the social work students (28.05), psychology students (28.79), and business students (28.19) were almost identical. One hundred and six (71%) of the students reported their racial/ethnic background as white and 43 (29%) reported African American, Asian American, or “other.” Ninety-six (64%) of the students were single, 45 (30%) were married, and 8 students (5%) were separated or divorced at the time of the study; 45 (73%) of the social work students, 36 (76%) of the psychology students, and 25 (66%) of the business students were single. Statistical analysis also revealed that the study’s participants reflected the distinctly different gender compositions of these three occupational groups in American society: 90% (n=56) of the social work students in the study, 71% (n=35) of the psychology students, and 50% (n=19) of the business students were female ([Chi]=20.04, p [is less than] .001). Measures Research Anxiety. The 11 items comprising the Graduate Student Research Anxiety Scale (GRAS) used in this study were adapted from a version of the Mathematics-Anxiety Scale-Revised (MARS-R) modified by Walsh (1997) to assess the general research anxiety experienced by social work students. Walsh constructed 24 questions about learning apprehension, anxiety, and tension to parallel the more narrow questions about mathematics and statistics courses taken directly from MARS-R (Richardson & Suinn, 1972). Eleven of the 24 statements, those judged to be most relevant to research and statistics, were selected for the GRAS and placed on a five point Likert-type continuum ranging from 1=strongly disagree to 5=strongly agree. Higher scores were assumed to represent higher levels of research anxiety. Representative statements from the GRAS include, “The thought of registering for a research course makes me apprehensive or anxious,” “I get anxious reading and interpreting charts and graphs,” and “I get uncomfortable or anxious looking through the pages of a research textbook.” As summarized in Table 1, the internal consistency reliability for the GRAS was estimated by calculating Cronbach’s alpha for the 11-item scale. The resulting coefficient of .93 suggests excellent internal consistency for this sample of graduate students. Table 1. Measures of Research Anxiety (GRAS), Computer Anxiety (GCAS), and Research Importance (GRIM) among Graduate Students in Social Work, Psychology, and Business (N=149) Number Cronbach’s Variable/Measure of Items Alpha M Range SD Research Anxiety 11 .93 24.25 11-51 9.60 (GRAS) Computer Anxiety 7 .79 12.44 7-29 4.53 (GCAS) Research Importance 5 .76 17.15 8-25 3.39 (GRIM) Computer Anxiety. The seven-item Graduate Student Computer Anxiety Scale (GCAS) was constructed to reflect the amount of tension and anxiety with which the graduate students approached computers and computer use, required within each of the three professional curriculums. Responses to all seven items were placed on the same five-point, strongly agree-strongly disagree continuum so that higher scores represented higher degrees of computer anxiety. Representative items include, “Computers intimidate and threaten me,” “I am confident I can improve my computing skills during my graduate program,” and “I have avoided computers because they are unfamiliar to me.” The alpha coefficient of .79 for the seven-item GCAS (Table 1) is similar to previously reported alpha coefficients and indicates adequate internal consistency for the GCAS. Research Influence. The variation in the influence or importance the graduate students attributed to research within their profession was assessed through the use of the Graduate Students Research Influence Measure (GRIM), comprised of five items modified for this study from the longer Kirk-Rosenblatt Research Inventory (KRRI) (Kirk & Rosenblatt, 1981). For the present study, the five statements were prepared in parallel forms modified to reflect each of the three different disciplines represented. Utilizing the same five-point, strongly agree-strongly disagree scale used in the GRAS and the GCAS, students were asked to indicate their level of agreement or disagreement with statements asserting that professionals (i.e., either social workers, business executives, or psychologists) “should rely on knowledge gained from research,” that their chosen discipline (social work, business, or psychology) should be “more science than art,” that a “major part of [their] professional education should consist of research training,” that professionals should “keep abreast of research in the field,” and that continuation of a program or strategy “should be contingent on effectiveness demonstrated by research.” Cronbach’s alpha for this scale, .76, was also within acceptable limits for summated scales. Higher scores on the GRIM indicate a stronger interest or valuation of research within the students’ respective professions. Results The means, standard deviations, and the range of scores for all graduate students on the GRAS, the GCAS, and the GRIM are presented in Table 1. There was considerable variation on all three scales. Some students highly valued and endorsed the role of research in their respective disciplines while others did not. Similarly, on the two anxiety scales, some students reported very high levels of apprehension about their research courses and the use of computers in fulfilling their research requirements, while others reported only minimal amounts of apprehension. As expected, student responses to the computer (GCAS) and research anxiety scales (GRAS) were moderately and positively correlated (r=.44, p [is less than] .001). Graduate students who reported higher levels of anxiety about course requirements also tended to report higher levels of anxiety about computer use. There was a somewhat weaker, inverse correlation between the GRIM and the GRAS (r=-.29, p [is less than] .001). Students who reported higher levels of anxiety about research tended to be less positive about the importance or influence of research within their discipline. The Pearson correlation coefficient between the GRIM and the GCAS was not statistically significant (r=-.11, p=.181). Because Multivariate Analysis of Variance (MANOVA) procedures are advisable when independent variables are moderately correlated, the data were first screened for outliers and then diagnostic tests for linearity, homogeneity of the variance covariance matrices, and multi-collinearity were performed with the Statistical Package for the Social Sciences. Because these tests supported the suitability of the data, each of the three statistical comparisons was performed with a factorial MANOVA. In all analyses student age, gender, and minority group membership were treated as covariates; multivariate and univariate effects of the independent variable on the dependent variables were observed only after the effects of the covariates had been considered statistically. Because of the research questions, the relatively small sample size, and the uneven number of cases in each cell, interactions were not assessed. Comparing Social Work and Undifferentiated Graduate Students Table 2 summarizes the results of the MANOVA procedures. The probability level of Wilks’s Lambda (p [is less than] .05) was used to assess the multivariate effect in the initial comparison of the means of the social work group with those of the mixed group of psychology and business students and for the two subsequent statistical comparisons. The same alpha level was used for the resulting univariate F tests performed for each dependent variable. As can be seen in Table 2, the tests revealed that the social work graduate students reported more research anxiety, more computer anxiety, and less interest in research when compared with the undifferentiated group of graduate students (Wilks’s Lambda-.77, p [is less than] .0001). The univariate F values were statistically significant for research anxiety (GRAS) (F=40.04, p [is less than] .001), computer anxiety (GCAS) (F=10.23, p=.002), and research importance (GRIM) (F=8.25, p=.005). Table 2. Comparison of Social Work Graduate Students with Psychology Graduate Students, Business Graduate Students, and an Undifferentiated Group of Business and Psychology Graduate Students on Measures of Research Anxiety, Computer Anxiety, and Research Importance Group Means Graduate Research Research Student Anxiety Anxiety Importance Wilks’s Group n (GRAS) (GCAS) (GRIM) Lambda(*) p Social Work 62 29.67 13.98 16.34 N/A N/A Psychology 49 18.06 10.69 19.87 .57 .000 Business 38 23.36 12.18 14.95 .90 .022 Psychology /Business 87 20.37 11.35 17.72 .77 .000 (*) For comparison between social work students and other students. Comparing Social Work and Psychology Students As summarized in Table 2, the social work graduate students reported more research anxiety, more computer anxiety, and less interest in research than did the group of psychology graduate students (Wilks’s Lambda=.57, p [is less than] .001). The univariate F tests were statistically significant for the GRAS (F=46.66, p [is less than] .001), the GCAS (F=10.56, p=.002), and the GRIM (F=8.01, p [is less than] .001). Comparing Social Work and Business Graduate Students As Table 2 shows, multivariate F was also statistically significant for comparisons of the social work (n=62) and business (n=38) students (Wilks’s Lambda=.90, p=.022). The univariate F tests for research anxiety (F=5.77, p [is less than] .001) and research importance (F=6.21, p=.008) were also statistically significant. The univariate F for the computer anxiety scale (GCAS), however, was not statistically significant (F=1.77, p [is less than] .186). Discussion Limitations In spite of the strength and consistency of these findings, discussion and interpretation of the implications for social work education must proceed with considerable caution. This study of research learning attributes among social work and other students is only the second reported in the social work literature and the first conducted exclusively with graduate students. Although acceptable alpha reliabilities for the scales were obtained, these modified instruments were not subjected to validity testing. In addition, the sample involved students from a single research institution. This institution’s programs in social work, psychology, and business comprise a small fraction of the hundreds of graduate training programs in these disciplines in the wide variety of U.S. colleges and universities. Although some uniformity and comparability may be expected among students enrolled in the same degree programs at different colleges and universities, variability in these research learning attributes must be assumed as well. Because the social work sample was restricted to students with clinical concentrations, it is not known whether the same results would have been obtained if nonclinical social workers or macro concentrators had been involved in the study. Indeed, an earlier comparison of graduate students in social work and business revealed that MSW students in administrative concentrations were more similar to MBAs than to clinical social work MSWs in some aspects of their practice (Feld, 1987). Finally, although statistically significant differences between groups were obtained even after gender was controlled in the analysis, the small sample size and unevenly distributed cells did not permit a more comprehensive analysis of potential interactions between gender and other variables. Similarly, we do not know whether the same results would have been obtained if the study had been restricted exclusively to students who were in the process of earning doctoral degrees. Although each of the master’s students in the study was enrolled full-time, and each of the three different programs in which they were enrolled offered the PhD degree, a larger percentage of graduate students in psychology earn their profession’s terminal degree than students in business or social work. Future studies of these research learning attributes among social work students should include exclusive samples of social workers who have completed the MSW degree and are enrolled in social work doctoral programs. Implications The research learning attribute differences among various groups of graduate students remained evident even after statistical controls for gender, minority group membership, and age were applied. Although none of these demographic characteristics were associated with or appeared to confound the measures of learning attributes in the present study, the findings neither support nor challenge the influence of these variables on students’ learning. Indeed, the importance of these demographic variables in shaping the learning patterns of high school and college students has been demonstrated in earlier educational studies (Anderson, 1996; Llabre & Suarez, 1985; Whiteley, 1996). Rather, the findings of this study suggest only that variation in age, gender, and minority group membership did not explain the differences in learning attributes reported by the graduate and professional students examined. The purpose of this study was description rather than explanation. Its aim was to understand the degree to which the research learning attributes of social work students were similar to or different from those of full-time graduate students enrolled in psychology and business. Although social work students have consistently been characterized as hesitant, reluctant, and resistant by their research professors, it was assumed that most social work research faculty had not had the opportunity to closely observe the reactions of research students in other disciplines. This study was conducted, in part, to fill this gap. If the study’s findings had suggested that social work students’ level of research anxiety and their beliefs about the importance of research were similar to other graduate students’ levels of research anxiety and beliefs about the importance of research, the specialized teaching strategies in the social work research teaching literature might be applicable to fewer social work students than previously assumed, and clearly inapplicable to some social work students. The findings of this study, however, suggest otherwise. The social work MSW students in this study, enrolled at a major research university, consistently reported more computer and research anxiety than both business and psychology students. This finding may provide empirical support for the consistent, albeit largely anecdotal, social work knowledge base about the research learning attributes of social work graduate students. Two related lines of future inquiry into research teaching are therefore recommended. First, the findings concerning this sample of graduate students and Royse and Rompf’s (1992) undergraduate students need to be evaluated within the context of other samples of graduate and undergraduate students at other colleges and universities. If subsequent findings continue to suggest different research learning attributes for social work students when compared to other students, the specialized research teaching procedures advocated in the social work literature might more appropriately organize, rather than merely supplement or complement the research curriculum. Second, future evaluations should be directed towards systematically identifying the relative strengths, weaknesses, and effectiveness of the numerous research teaching methods and techniques proposed in the social work literature. Although this study may provide general support for these specialized techniques, the particular circumstances and conditions under which some might be more useful than others remains to be evaluated and reported within the research teaching literature. Authors: ROBERT G. GREEN, ANTOINETTE BRETZIN, CHRISTINE LEININGER and ROSE STAUFFER Date: Spring-Summer 2001