Chapter 1 of your text, Supervision in Social Work (linked in Resources),introduces you to many aspects of social work supervision including definitions, functions and objectives,

Transformational Leadership in the Social Work Context: The Importance of Leader Continuity and Co-Worker Support

Susanne Tafvelin*, Ulf Hyvönen, and Kristina Westerberg

Susanne Tafvelin is a Ph.D. candidate at the Department of Psychology at Umeå University in Sweden. Her thesis is devoted to leadership in social work with a focus on transformational

leadership. Ulf Hyvönen, Ph.D., is research director at the Field Research and Development Unit at Umeå Social Services in Sweden. Besides his interest in questions concerning the

interplay between research and practice, he has been engaged in national and international research on child welfare and child protection for many years. Kristina Westerberg, Ph.D., is currently Associate Professor at the Department of Psychology at Umeå University. She also

holds a part-time position as research leader at the Field Research and Development Unit at Umeå Social Services in Sweden, where her area of research concerns organisation, learning

and leadership.

*Correspondence to Susanne Tafvelin, Department of Psychology, Umeå University, 901 87 Umeå, Sweden. E-mail: [email protected]

Abstract

Social work leadership has attracted growing attention in both social work practice and

research. As social service organisations have changed in a variety of ways during the last

decades, knowledge of how leaders should act in these transformed organisations is

crucial. However, few empirical studies have examined what kind of leadership these

changed organisations benefit from and how the continuing organisational change

might affect the impact leaders have. The present study aimed at exploring the effect

of transformational leadership of first line managers in a social work setting. We used

a randomised sample of 158 employees in a Swedish social service organisation, and

examined the direct and indirect effect of transformational leadership on two important

employee attitudes—commitment and role clarity. The results demonstrate the contri-

bution of transformational leadership in creating a workplace where employees are

committed and know what their assignment is. Interaction effects of leader continuity

and co-worker support indicate the need for some stability in the organisation in

order to increase the positive influence transformational leaders have on employees.

This study has implications for leadership training in social work and is a contribution

# The Author 2012. Published by Oxford University Press on behalf of

The British Association of Social Workers. All rights reserved.

British Journal of Social Work (2014) 44, 886–904 doi:10.1093/bjsw/bcs174 Advance Access publication November 19, 2012

to the co-operative knowledge development of leadership in social service

organisations.

Keywords: Transformational leadership, social service organisations, leader continuity,

co-worker support

Accepted: August 2012

Introduction

Leadership in social work has attracted growing attention from both re- search and practice (Lawler, 2007; Rank and Hutchison, 2000; Walter et al., 2004). One reason for this newly awakened interest in leadership may be the many changes social service organisations have undergone during the last decades. With the introduction of flat organisations, new public management and the aim to develop an evidence-based organisation (Alexanderson et al., 2009; Lambers, 2002; Wolmesjö, 2005), the need for knowledge of how to lead these transformed organisations has increased. With even more anticipated organisational changes in the future due to changing legal, social, technological and competitive circumstances, good leadership is seen as being the key to retain employees and to handle the rapid pace of change in today’s social service organisations (Lawler, 2007).

Although the importance of social work leadership has been recognised by many scholars, knowledge of effective and successful leadership in these organisations is still scarce. The few studies that have been conducted have examined the effects of leadership in social work on a very limited set of outcomes (e.g. satisfaction and service effectiveness) and are mostly from the USA (e.g. Elpers and Westhuis, 2008; Yoo and Brooks, 2005), whereas effects of leadership in European social service organisations remain more or less unexplored. Also, while the methods in social work are debated in the spirit of evidence-based practice in order to see its effects (e.g. Mullen and Shuluk, 2011), leadership has, surprisingly, been left out of the discussion. The need for further knowledge about good lead- ership for social service organisations is clear (Lawler, 2007).

The purpose of this study was to examine effects of leadership in social work on employee attitudes in a Swedish context. We did this by studying the impact of first line managers’ transformational leadership, a leadership model based on vision and empowerment (Bass and Riggio, 2006), on em- ployee role clarity and commitment in a Swedish social service organisa- tion, which includes social welfare, elderly care and care of the disabled. Role clarity refers to the extent to which employees clearly understand what is expected of them at work (e.g. King and King, 1990). Commitment is defined as ‘the relative strength of an individual’s identification with and involvement in a particular organization’ (Mowday et al., 1982, p. 27).

Transformational Leadership in the Social Work Context 887

Given the challenging future with a changing environment, we will argue that role clarity and commitment of employees will be essential for handling the changes and preventing turnover in these organisations.

Leadership in social work

The discussion of social work leadership more or less began with Brilliant’s (1986) analysis of social workers’ resistance to take on leadership roles. She saw leadership as the missing ingredient in social work education but emphasised that leadership is an important aspect of the professional role for social workers. Analysing the roots of leadership as a non-theme in social work, she suggested that social work students were passionate about direct practice with clients, not with assuming leadership roles. She also put forward ideological constraints, a sense of powerlessness and a general lack of status in society as barriers for social workers to become leaders. Her conclusion was that schools of social work must take on re- sponsibility in developing leadership potential among social workers.

Since then, there have been dramatic changes within social service orga- nisations with consequences for the leadership role. During the two last decades, a wave of new ideologies, referred to as new public management or managerialism, has found its way into the public sector. By importing business models into the public sector and making its organisations more market-oriented, the goal was to increase effectiveness and control (Edwards, 1998). The idea was to modernise the public sector in terms of making these organisations accountable, flexible and transparent. However, the usefulness of business models in social service organisations has been debated and many scholars argue that public sector organisations differ too much from their private counterparts to make the continuing import of models worthwhile (Langan, 2000; Persson and Westrup, 2009). For leaders, the shift to managerialism resulted in increased responsibility in general and more specifically in responsibility to reach goals (Lawler, 2007). A study of Swedish first line managers in elderly care discovered that, with increased responsibility for budget and staff, the leadership role took an administrative turn, with increased loyalty upwards in the or- ganisation (Karlsson, 2006). Healy (2002) argues that, with managerialism and its ideals of money and efficiency at heart, it becomes even more im- portant for social workers to take on leadership roles to promote the softer values of human service organisations.

Another line of research has investigated what kind of leadership differ- ent stakeholders would like to see in social work. Wolmesjö (2005) com- pared the expectations of politicians on leaders on the one hand with expectations of employees on the other hand. The results demonstrated that politicians expected leaders to be managers, in that sense that they wanted leaders to focus on the administrative tasks. On the contrary,

888 Susanne Tafvelin et al.

employees asked for a leadership role with focus on the relational aspects, which is in line with what is traditionally meant by leadership. Rank and Hutchison (2000) asked leaders themselves what kind of leadership they thought was appropriate in social work. The answers painted a picture of a leadership based on vision, promoting values of the profession, motivating and stimulating employees, and a leader who was able to facilitate change. The authors also reported that the leaders saw social work leadership as dif- ferent from other professions in terms of being more inclusive and altruistic.

Even though these kinds of studies give valuable and important insights into the premises of social work leadership, they do not offer any informa- tion of what kind of leadership would be effective, or have positive effects for employees and clients. One approach to study effects of leadership in social work used by a small body of recent studies has been to apply trans- formational leadership theory.

Transformational leadership in social work

Transformational leadership theory was introduced by Bernard Bass in the mid-1980s and is today one of the most thoroughly researched leadership models (Bass, 1999; Bass and Riggio, 2006). In essence, transformational leadership is based on a strong identification of the follower with the leader and the social unit in which the leadership takes place. In this process, the leader raises follower awareness and understanding of moral values and inspiring visions, and encourages followers to transcend their own personal goals and interests for the collective good (Bass, 1985). The concept of transformational leadership is composed of four dimensions: idealised influence (charismatic role modelling), inspirational motivation (articulating an appealing vision), intellectual stimulation (promoting cre- ativity and innovation) and individualised consideration (coaching and mentoring) (Bass and Riggio, 2006). Research has demonstrated the useful- ness of transformational leadership in many different types of organisations and countries and established a range of positive effects of transformational leadership, such as on organisational effectiveness, performance, innov- ation, trust and job satisfaction (Fuller et al., 1996; Lowe et al., 1996). At first, transformational leadership was associated with top leadership in the private sector, but research evidence points in a different direction. In a meta-analysis of transformational leadership and effectiveness, the authors found, to their surprise, that transformational leadership was more common in the public sector and among first line managers, not the CEOs (Lowe et al., 1996).

Even though the numbers of studies using transformational leadership theory in social work settings are increasing, these studies are still scarce, scattered and rarely refer to each other. Most of the studies originate from the USA and Canada and have explored the influence of transformational

Transformational Leadership in the Social Work Context 889

leadership in social work with respect to job satisfaction and satisfaction with the leader, and found positive relationships (Elpers and Westhuis, 2008; Gellis, 2001; Mary, 2005). Another set of studies examined the effect of transformational leadership on service effectiveness. In a study from the USA, service effectiveness was measured by the number of out-of-home placements of children and it was found that a transformation- al leader together with support and good work routines decreased the number of children being placed out of home (Yoo and Brooks, 2005). An Israeli study, which measured service effectiveness in terms of imple- mentation of plans, goal attainment and client empowerment, found weak but positive correlations between transformational leadership and effectiveness (Boehm and Yoels, 2009). In Europe, studies of transformation- al leadership in social work are rare. A few studies have been conducted in Danish elderly care and supported the usefulness of transformational lead- ership with regard to employee well-being (Nielsen and Munir, 2009; Nielsen et al., 2008).

We suggest that employee role clarity and commitment will be two im- portant employee attitudes for social service organisations in the future, for at least two reasons. First, social service organisations face a challenging future with a rapidly changing environment where, in order to keep high performance and service quality levels, it will be important for employees to know what their specific role in the organisation is (Mukherjee and Mal- hotra, 2006) and to be committed to the organisation, as this buffers against stress and job displeasure during organisational change (Begley and Czajka, 1993). Second, studies of turnover among social workers have demon- strated the importance of both commitment and role clarity in retaining employees (Mor Barak et al., 2001; Tham, 2007), which was suggested to be one of the most important challenges for social work (Nissly et al., 2004; Tham, 2007). High turnover rates among social workers were reported in the USA, Great Britain and Sweden (Tham, 2007), which may result in higher costs, reduced effectiveness and poorer outcomes (Balfour and Neff, 1993; Drake and Yadama, 1996; Powell and York, 1992). Both commitment and role clarity have been positively related to transformational leadership (Viator, 2001; Dumdum et al., 2002). By being visionary and formulating clear goals, the transformational leader helps employees to understand their role in the organisation (Viator, 2001). Transformational leaders are able to increase organisational commit- ment by promoting higher levels of intrinsic motivation (Shamir et al., 1993), inspiring loyalty, and recognising and appreciating the different needs of each follower to develop his or her personal potential (Bass and Avolio, 1994; Yammarino et al., 1993). Based on the above arguments, we propose:

Hypothesis 1: Transformational leadership is positively related to role clarity and organisational commitment.

890 Susanne Tafvelin et al.

In addition to examining the direct effect of transformational leadership on role clarity and commitment, we also want to explore the influence of organisational factors that may facilitate or hinder the effect of transform- ational leadership in social work. In general, less is known of the processes by which transformational leaders exert their influence on employee atti- tudes (Avolio et al., 2004). One such factor is the continuing change in social service organisations both in the past and anticipated in the future (Lambers, 2002; Lawler, 2007; Wolmesjö, 2005). At the employee level, these changes can have a range of consequences in terms of changing work tasks, leader or work group. We examined the interaction effect of two aspects that may be affected by a turbulent organisational environment: leader continuity and perceived co-worker support. Leader continuity, in terms of time with current leader, has been suggested to be one of the most important but overlooked situational variables affecting transform- ational leadership (Hughes et al., 2006). Transformational leadership does not happen overnight and it takes time for leaders to build trusting relation- ships, to develop and articulate their vision, to heighten followers’ emotion- al level and empower them to fulfil the vision. These arguments are supported by a study of Ling and colleagues (2008), who found that leader- ship tenure moderated the relationship between transformational leader- ship and firm performance. Therefore, we suggest that the longer the employee is exposed to a transformational leader, the larger the effect will be on commitment and role clarity.

Co-worker support refers to the extent to which employees can count on their colleagues to help and support them when needed (Liao et al., 2004). Co-worker support may be hampered by organisational change, as it needs connectedness to develop (Langford et al., 1997), which is hindered if work groups are frequently reorganised. Co-worker support, which includes caring, tangible aid and information (Ducharme and Martin, 2000; Parris, 2003), may increase employees’ comfort within the organisation by fulfilling needs for esteem, approval and affiliation (Stinglhamber and Vanden- berghe, 2003), and thereby enhancing commitment by making employees having an emotionally satisfying work experience, which, over time, trans- lates into an emotional attachment to their employing organisation (Rous- seau and Aubé, 2010). Empirical studies also confirm the positive relationship between co-worker support and commitment (Ng and Soren- sen, 2008; Rousseau and Aubé, 2010). Support from co-workers has also been suggested to protect or shield the individual employee from the worst aspects of role ambiguity (Cobb, 1976) or, in other words, spending time with co-workers may help employees to better understand their role in the organisation, and therefore enhance their role clarity. Empirical support for this suggestion can be found in a meta-analysis that showed that satisfaction with co-workers decreases role ambiguity (Fischer and Gitelson, 1983) and thereby increases role clarity. In line with this, we expect that co-worker support will strengthen and enhance the effect of

Transformational Leadership in the Social Work Context 891

transformational leadership on commitment and role clarity. Therefore, we suggest:

Hypothesis 2: The positive effect of transformational leadership on role clarity and commitment is moderated by leader continuity and co-worker support.

Method Sample and procedure

The context of this study is a social service organisation in a larger Swedish municipality, which provides three main services—social welfare, elderly care and care of the disabled. The organisation had in the beginning of 2010 around 2,700 employees. Three hundred and eighty-two participants with permanent or long-term temporary employment were randomly selected from the staff records and a questionnaire was distributed to the participants’ home address by mail. All respondents were informed that the survey was anonymous, participation was voluntary and they could withdraw at any time. A postage-paid, self-addressed envelope was pro- vided so that they could mail their surveys back to the researchers directly. Forty participants asked to be excluded because of circumstances as mater- nity leave, longer sick periods or leave of absence. In the end, 158 out of the remaining 342 answered the questionnaire, yielding a response rate of 46 per cent. The sample in our study included 125 (79 per cent) women and 33 (21 per cent) men. The average age for women was forty-four years, and for men it was forty-one years. Fifty per cent of both men and women had completed a university degree and 91 per cent held a perman- ent position. The sample was representative of the population, the entire social service organisation, with respect to age, gender and employment terms (permanent or temporary).

Measures

Transformational leadership

We measured transformational leadership with the Multifactor Leadership Questionnaire (MLQ) developed by Bass and Avolio (1995). The MLQ is the most common measure of transformational leadership (Yukl, 1998) and operationalises the four theoretically identified dimensions of transform- ational leadership—idealised influence, inspirational motivation, intellec- tual stimulation and individualised consideration. Respondents were asked to rate how often their leader engages in behaviours specific to each dimension on a five-point scale ranging from 0 (not at all) to 4 (often, if not always). The present study employed a Swedish translation

892 Susanne Tafvelin et al.

of the MLQ 5X (Form 5x—Short) provided by Mind Garden Inc.; however, a few items were slightly changed to get a more accurate translation. The reliability of the twenty items that together measure transformational lead- ership was in our study 0.94.

Role clarity

We used a three-item scale to measure role clarity taken from the QPS Nordic, which is a well validated and much used questionnaire measuring psychological and social factors at work (Dallner et al., 2000). The scale includes the items ‘Are there clearly defined goals in your work?’, ‘Do you know which areas of responsibility you have?’ and ‘Do you know exactly what is demanded of you at work?’. The response categories ranged from 1 (very seldom or never) to 5 (very often or always). The reli- ability of this subscale was 0.74.

Commitment

The scale to measure organisational commitment was also taken from the QPS Nordic (Dallner et al., 2000). It contained three items: ‘The organiza- tion inspires me to do my best’, ‘I tell my friends that the organization is a very good place to work at’ and ‘My own values are very close to the values of the organization’. The response scale ranged from 1 (strongly disagree) to 5 (strongly agree). The reliability of this subscale was 0.85.

Moderators

Leader continuity was measured by a single item with the question ‘How many years have you had your current leader?’. The scale to measure co- worker support comes from the Learning Climate Scale (Westerberg and Hauer, 2009) and contains three items: ‘When I need I get support and help from my colleagues’, ‘I can count on help and support to learn from my mistakes at work’ and ‘It is acceptable to have a bad day’. The response scale ranged from 1 (very seldom or never) to 5 (very often or always) and the reliability of this subscale was 0.78.

Data analysis

Since we did not have any a priori expectations that any of the dimensions of transformational leadership would differentially be associated with role clarity or commitment, we combined these subscales into one higher order

Transformational Leadership in the Social Work Context 893

factor, which is consistent with previous research (e.g. Bono et al., 2007; Lim and Ployhart, 2004). We then examined the bivariate correlation between our study variables to gain preliminary support of our hypotheses. To test the interactional and main effect models described earlier, we followed the hierarchical multiple regression procedures described by Aiken and West (1991). In a first step, our independent variable transformational lead- ership was entered into the regression. In the second step, co-worker support or leader continuity was added to the regression. In the third and final step, the interaction term, the cross-product of the two independent variables (transformational leadership and either co-worker support or leader continuity) was included into the regression, now reflecting the joint effect of the two variables. The change in R2 from step 2 to step 3 pro- vides a test of whether the interaction term is making a significant contribu- tion to the equation. This three-step procedure was repeated in four separate regressions, two with commitment and two with role clarity as de- pendent variable, to test the interaction effect of the two moderators: leader continuity and co-worker support. Assessing interaction terms in field re- search is known to be difficult, and we therefore retained interactions that accounted for more than 1 per cent of the variance for further analyses. Analyses of field studies suggest that, on average, interactions account for less than 3 per cent of the variance, and Monte Carlo findings indicate that interactions accounting for 1 per cent of the variance are meaningful in re- gression analyses (Aiken and West, 1991). Before running the regressions, all independent variables were mean centred in order to decrease the risk of collinearity (Aiken and West, 1991). As one of the moderators, leader con- tinuity, was not normally distributed, a median split (Mdn ¼ 1.5 years) was used for this variable instead of mean centering. This variable has two levels: 0 representing a short time with the current leader and 1 representing a longer time with the current leader. This variable was then entered as a dummy variable into the regressions. When the interaction term was signifi- cant, we calculated and plotted the simple slopes for high and low levels of co-worker support or leader continuity. All statistical analyses were per- formed using PASW Statistics, version 18.0 (SPSS, Inc., Chicago, IL).

Results

Table 1 presents means, standard deviations and correlations among the variables. As expected, transformational leadership was positively associated with both role clarity and commitment (r ¼ 0.35, p , 0.01 and r ¼ 0.44, p , 0.01, respectively), both associations significantly supporting hypothesis 1. Examination of the relationship between transformational leadership and co-worker support revealed a strong positive association (r ¼ 0.48, p , 0.01). No association was found between transformational leadership and leader continuity. Positive associations were found between

894 Susanne Tafvelin et al.

the two dependent variables role clarity and commitment (r ¼ 0.35, p ,

0.01). Further examination of the data revealed that co-worker support was positively associated with both role clarity (r ¼ 0.30, p , 0.01) and commitment (r ¼ 0.38, p , 0.01), while leader continuity was not associated with either role clarity, commitment or co-worker support (r ¼ 0.03, p .

0.05, r ¼ 0.02, p . 0.05, r ¼ –0.05, p . 0.05, respectively). Next, moderated hierarchical regressions were conducted to test our hy-

potheses. In support of hypothesis 1, we found significant, positive relation- ships between transformational leadership and role clarity as well as between transformational leadership and commitment (see Table 2). Co- worker support was significantly related to both role clarity and commit- ment, while leader continuity was not significantly related to any of the outcome variables. In hypothesis 2, co-worker support and leader continu- ity were predicted to moderate the relationship between transformational leadership and outcomes, such that stronger relationships with role clarity and commitment would occur when employees experienced high co-worker support and had worked with their leader for a longer period of time. Table 2 reports a significant moderator effect of co-worker support on com- mitment, but not on role clarity. A longer time with the leader moderated

Table 2 Results of moderated hierarchical regression analyses

Role clarity b Commitment b

1 2 3 1 2 3

Step 1: Transformational leadership 0.32** 0.32** 0.06 0.43** 0.43** 0.20 Step 2: Leader continuity 0.00 0.01 –0.03 –0.03 Step 3: Interaction 0.31** 0.28* Adjusted R2 0.10 0.09 0.12 0.18 0.18 0.20 DR2 at last step 0.03* 0.02* Step 1: Transformational leadership 0.35** 0.27** 0.26** 0.45** 0.33** 0.34** Step 2: Co-worker support 0.17* 0.16 0.26** 0.28** Step 3: Interaction –0.09 0.16* Adjusted R2 0.11 0.13 0.13 0.19 0.24 0.26 DR2 at last step 0.00 0.02*

* p , 0.05; ** p , 0.01.

Table 1 Descriptive statistics and zero-order correlations for all study variables (n ¼ 158)

Variable M SD 1. TL 2. LC 3. CS 4. RC

1. Transformational leadership 2.18 0.81 2. Leader continuity 2.87 3.74 –0.01 3. Co-worker support 3.73 0.84 0.48* –0.05 4. Role clarity 4.24 0.66 0.35* 0.03 0.30* 5. Commitment 3.10 0.92 0.44* 0.02 0.38* 0.35*

* p , 0.01.

Transformational Leadership in the Social Work Context 895

the effect of transformational leadership on both role clarity and commit- ment as the changes in the multiple squared correlation coefficient (DR2) associated with transformational leadership and its interaction with leader continuity were both statistically significant.

In order to interpret the interactions, all three significant two-way inter- actions were plotted with cut values of one standard deviation below the mean and one standard deviation above the mean on each moderator vari- able. The first plot revealed that the positive effect of transformational leadership on role clarity was stronger among employees with longer time with their leader (see Figure 1). A second plot revealed that the posi- tive effect of transformational leaders on commitment was also stronger among employees who had spent longer time with their leader (see Figure 2). The third and last plot revealed that the positive effect of trans- formational leadership on commitment was stronger among employees with higher experience of co-worker support (see Figure 3). These plots are consistent with hypothesis 2.

Discussion

This study has tested the direct and moderated effects of transformational leadership on role clarity and commitment in a Swedish social service or- ganisation. Our results supported a direct and positive effect of transform- ational leadership in social work on both employee role clarity and commitment. The findings further revealed interaction effects by two

Figure 1 Role clarity predicted by interaction of transformational leadership and leader continuity.

896 Susanne Tafvelin