Validating a model for indicating the professionalisation of social work in China

Introduction

Social work is a global profession, with professional organisations in 129 countries (International Federation of Social Workers, n.d.). Increasing scholarly interest in social work professionalisation has explored how it can be distinguished from other occupations (Hugman, 2003; Hutchings & Taylor, 2007; Leung, 2007; Weiss-Gal & Welbourne, 2007, 2008; Yip, 2007). As Weiss-Gal and Welbourne (2008) point out, studying the professionalisation of social work not only enables the discovery of common ground in a global context, but also identifies local variations from international social work standards.

Following the 2006 proposal by the Central Committee of the Communist Party of China (CCCPC) to ‘build a large team of social workers’, an unprecedented increase in the number of social workers from 200,000 in 2012 to 1.45 million in 2020 was ordered (Organisation Department of the CCCPC, 2012). This would result in a ratio of approximately 10.4 social workers per 10,000 population compared to the current ratio of 17.3 registered social workers per 10,000 population in England (Cromarty, 2018). By July 2019 (the most recent date for which figures are available), the number of Chinese social workers had grown to over 1.2 million (Ministry of Civil Affairs (MCA), 2019).

This ambitious project has attracted serious attention in the literature. The question of whether the international standards of social work could be applied in the Chinese context has been raised and debated (e.g. Hutchings & Taylor, 2007; Leung, 2007). For example, in contrast to the competency frameworks developed in the United States, England and Hong Kong, Chinese experts appeared to favour only de-politicising techniques that would not highlight power relationships within an authoritarian regime (Lei & Huang, 2018). A collaborative Master of Social Work programme imported the curriculum design and pedagogy from a distinguished American university, but the professional commitment of students – measured by their willingness to seek employment in social work – consistently declined throughout classroom teaching and practice learning (Wang et al., 2019).

In order to investigate further the unique characteristics distinguishing Chinese social work from its international counterparts, this study adopted Weiss-Gal and Welbourne’s (2007, 2008) conceptual framework which includes eight dimensions of theoretical relevance to the professionalisation of social work: public recognition, knowledge base, professional education, professional organisations, ethical standards, monopoly, professional autonomy, and prestige and remuneration. Weiss-Gal and Welbourne (2007, 2008) used each of these dimensions in their own global comparative studies, while some have been chosen for case studies in specific countries (Beddoe & Duke, 2009; Bertotti, 2016; Fouché et al., 2014; Hines et al., 2010; Ng & Sim, 2012; Payne, 2002; Scholar et al., 2014).

Based on this framework, the present study aims to address the following research question: What indicators developed from the eight dimensions identified by Weiss-Gal and Welbourne can be validated to define the professionalisation of social work in China?

The conceptual framework

Weiss-Gal and Welbourne’s (2007, 2008) conceptual framework is based on two influential approaches which have commonly been used to define professionalisation, ‘attributes’ and ‘power’. The ‘attributes’ approach follows the functionalist idea of identifying specific traits which could characterise a profession and maintain its functions in society (Greenwood, 1957); the ‘power’ approach stresses how an occupation could become a profession by monopolising its entry and practice delivery (Johnson, 1972). By identifying common factors from these two approaches, Weiss-Gal and Welbourne (2007, 2008) developed eight dimensions to measure the extent of social work professionalisation. It was hypothesised that higher degrees of each of the eight dimensions should be associated with better development of professionalisation. The connections between these dimensions and professionalisation were empirically demonstrated in subsequent studies.

First, public recognition is mainly concerned with government regulation of the profession, in particular barring those without qualifications from practice. Beddoe and Duke (2009) argued that the social work registration system in New Zealand had ‘strengthen[ed] the profession by building its credibility alongside other similar professions’ (p. 787). Second, social work should be built on a unique knowledge base. Otherwise, depreciation of social workers’ skills, such as when migrating and working in a different country, can pose a serious risk to professional identity (Fouché et al., 2014). Third, professional education ensures that social workers possess relevant knowledge that can also be used to control entry to the profession. Even within unusual non-statutory placement settings in England, it has been reported that practice learning remains central for helping social work students develop their professional identities (Scholar et al., 2014).

Fourth, professional organisations usually play a dual role in representing their members’ interests and regulating the profession. For example, Payne (2002) illustrated the British Association of Social Work’s success in enhancing social work’s status through promoting the establishment of a professional accreditation body. Fifth, ethical standards which are developed from professional values can guide and regulate the behaviours of social workers, ‘reflect[ing] what is unique to the social work profession’ (National Association of Social Workers, n.d.). Sixth, monopoly refers to a primary or even exclusive role for social workers in undertaking certain tasks. For example, identification of what types of work would need a social worker was a preliminary step for introducing the profession in Vietnam (Hines et al., 2010). Seventh, professional autonomy explores the extent to which social workers’ decision-making is independent of influences that are external to the profession. Bertotti (2016) demonstrated how Italian child protection social workers’ professional identities changed after resource reductions limited their professional autonomy. Eighth, increasing prestige and remuneration is always considered as a means of facilitating professionalisation, as illustrated in Singapore, where salary improvements initiated in 2007 aimed to advance the status of social workers (Ng & Sim, 2012).

Weiss-Gal and Welbourne’s (2007, 2008) framework has been employed to study social work professionalisation in 10 countries. However, as they themselves admitted, extant research had not included China, ‘where social work is re-emerging after a period of severe repression and is finding its own distinctive identity’ (Weiss-Gal & Welbourne, 2007, p. 225). It must be noted that the application of Weiss-Gal and Welbourne’s framework for assessing social work professionalisation in China should be undertaken with caution by carefully examining the relevance of all dimensions in practice.

Regarding public recognition, the licensure system is supposed to build professional identity and guarantee social workers’ competence. However, a survey of 2108 social workers in various locations in China revealed that licensed social workers expressed even lower levels of perceived competence than unlicensed social workers (Li et al., 2019).

Education is generally considered as the foundation of a profession. However, empirical studies in China suggest its lack of relevance regarding social work professionalisation. Li et al.’s (2019) study also revealed no differences in the perceived levels of competence and professional identity between professionally trained and untrained social workers. Similarly, Lei et al.’s (2019) study of social workers in Integrated Family Services Centres (IFSCs) in Guangzhou revealed that professional training had no impact on their intentions to remain employed in their posts.

Current research evidence has identified the difficulties Chinese social workers face in achieving a monopoly role in practice. Although the government has officially identified 13 activities to be undertaken by social workers (Organisation Department of the CCCPC, 2012), no legislation prescribes social work task-exclusivity. Moreover, it is reported that social workers are routinely compelled by governments or secondary settings to undertake professionally irrelevant tasks, such as clerical assignments (Leung, 2012), grass-roots administrative duties (Chan & Lei, 2017) or even Communist Party campaigns (Cho, 2017). Therefore, the perception that social workers can easily be replaced by other professionals effectively discourages graduates from entering the profession (Zeng et al., 2016).

Last but not least, the prestige and remuneration of social workers tend to be remarkably low in China. Cho’s (2017) ethnographic study of social work practice in Foxconn town, Shenzhen, concluded that front-line social work had been proletarianised by China’s neo-liberal outsourcing system. Front-line social work was mainly undertaken by rural migrant youth, who were considered ‘unqualified’ and thus offered low wages, resulting in high turnover rates. Other research has reported that low salaries both deter students from entering the social work profession (Zeng et al., 2016) and reduce new recruits’ professional commitment (Yan et al., 2013), although Lei et al.’s (2018) recent research failed to identify any effects of salary on IFSC social workers’ organisational or occupational turnover intentions.

Method

Questionnaire

The authors developed a questionnaire comprising 11 indicators, based on Weiss-Gal and Welbourne’s (2007, 2008) framework, 6 of which were subjective and assessed using a five-point Likert scale, and 5 were objective. Before launching the study, this questionnaire was reviewed by two experienced social work professors who suggested minor edits, and then randomly piloted with 20 social workers, resulting in amendments to improve its readability.

First, adhering to public recognition focusing on licensure, the indicator of qualification level (X2) assumed that higher qualifications would be reflected in higher public recognition. The response alternatives not only included the national qualifications such as ‘junior’ and ‘intermediate’ certificates, but also recognised local qualifications such as ‘social work assistant’ and ‘supervisor’ in Guangzhou. Meanwhile, a subjective indicator was also included in this dimension because, as Weiss-Gal and Welbourne (2008) observed, public recognition may comprise not only ‘recognition by the administration’, but also ‘recognition by the public’ (p. 283). In other words, formal regulation does not necessarily contribute to public recognition. Based on the United Kingdom, Weiss-Gal and Welbourne (2007) even concluded that ‘it remains unclear whether the existence of licensing will be effective in changing public opinion in social work’ (p. 230). Therefore, respondents’ self-perception of the public status of social work was included (X1). Since the second dimension, knowledge base, explores the knowledge mastered by social workers, respondents were asked about their general perceptions of their knowledge adequacy (X3) for managing their jobs. The third dimension, professional education, was measured by the highest social work degree (if any) held by respondents (X4). Response alternatives ranged from ‘without social work degree’, ‘diploma in social work’, ‘bachelor degree of social work’ and ‘master degree of social work’.

The fourth dimension focused on two major roles of professional organisations – continued training and regulation. The objective indicator measured the frequency of respondents’ attendance at training activities arranged by various professional organisations (X6) in the previous six months. The subjective indicator gauged respondents’ satisfaction (X5) with the Guangzhou Social Work Association, the major regulator of the IFSCs and which was responsible for designing and implementing annual appraisals. The fifth dimension referred to respondents’ perceptions of their congruence with ethical standards (X7) in practice. This indicator was designed to address Weiss-Gal and Welbourne’s (2008) concern about ethical practice and respondents were asked how thoroughly the ethical structures in the various codes are embedded in day-to-day practice. The sixth dimension, monopoly, was measured by respondents’ perceptions of the necessity for the IFSCs to employ professional social workers (X8) and the entrance criteria for social work job applicants (X9). For the latter indicator, it was standard practice for IFSCs to consider qualification certificates as equivalent to professional education. Therefore, response alternatives comprised: ‘neither certificate nor professional education required’, ‘either certificate or professional education required’ and ‘both certificate and professional education required’. The seventh and eighth dimensions related to professional autonomy and prestige and remuneration, respectively. Professional autonomy was measured by respondents’ perception of autonomy (X10) in working in the IFSCs, while prestige and remuneration were indicated by respondents’ self-report of their average after-tax salary (X11) in the previous three months.

This questionnaire is summarised in Table 1. All response alternatives were re-coded in subsequent analysis, so that a higher value represented a higher degree of professionalisation. The results of all the Likert scales were also standardised.

OPEN IN VIEWER

Table 1.

Questionnaire.

Dimension	Indicator	Response alternatives
Public recognition	X1 Self-perception	Subjective Likert scale
	X2 Qualification rank	Without/Assistant-Level/Junior/Intermediate/Supervisor certificates
Knowledge base	X3 Knowledge adequacy	Subjective Likert scale
Professional education	X4 Educational background	Without/Diploma/Bachelor/Master social work degrees
Professional organisations	X5 Satisfaction	Subjective Likert scale
	X6 Involvement	Number of professional organisations with which respondents have been involved for training in the previous six months.
Ethical standards	X7 Congruence	Subjective Likert scale
Monopoly	X8 Necessity	Subjective Likert scale
	X9 Recruitment criteria	Neither certificate nor professional education required/Either certificate or professional education required/Both certificate and professional education required
Professional autonomy	X10 Self-perception	Subjective Likert scale
Prestige and remuneration	X11 Salary	Average after-tax salary in the previous three months.

Analytic strategy

Formative measurement was used for analysis. This measurement is often confused with reflective measurement. For example, Jarvis et al. (2003) reviewed 1192 articles published in top marketing journals during 1977–2000 and found that 29% modelled constructs incorrectly and an overwhelming number should have used formative instead of reflective measures. Therefore, it is essential to decide whether a formative or a reflective model should be adopted in the research design.

The current literature identifies four criteria to differentiate formative and reflective models. First, the causal relationship in a formative model flows from the indicators to the latent construct, instead of the other way around as in a reflective model. Second, formative indicators cannot be interchangeable, which means that omitting any of the indicators in a formative model may change the meaning of the latent construct. Third, formative indicators should not be highly correlated to each other as is required for reflective indicators. This is the reason why the concept of reliability (e.g. internal consistency) and construct validity (e.g. convergent and discriminant validities) may not be used to evaluate a formative model. Fourth, in contrast to the reflective model, a formative model is not expected to have the same antecedents and consequences (Arnett et al., 2003; Diamantopoulos & Winklhofer, 2001; Jarvis et al., 2003; MacKenzie et al., 2005).

Using the formative measure to indicate the concept of professionalisation in this study can be justified as follows. First, the 11 indicators described above were identified to define the nature of professionalisation, rather than being used to manifest its features. Second, since these 11 indicators were developed from Weiss-Gal and Welbourne’s (2007, 2008) eight-dimensional framework, omission of any dimension would hinder a comprehensive understanding of the meaning of professionalisation. Third, these indicators may not be covaried with one another. For example, an increase of perceived public recognition (X1) does not necessarily relate to a high level of satisfaction with professional organisations (X5). Fourth, not all 11 indicators could simultaneously produce effects on professionalisation. For example, satisfaction with professional organisations (X5) may not have the same effect as involvement with professional organisations (X6).

Following a recommendation in the current literature, two reflective indicators that were considered as the effects or consequences of the latent construct of professionalisation were added to show respondents’ self-efficacy for model identification (e.g. Diamantopoulos et al., 2008; Diamantopoulos & Winklhofer, 2001). This has two advantages. First, it avoids the need for additional constructs for model identification. Second, using two reflective indicators only is less sensitive to changes in the structural relationships caused by the formative model (Diamantopoulos et al., 2008; Jarvis et al., 2003; MacKenzie et al., 2005).

The concept of self-efficacy refers to respondents’ belief and confidence in achieving satisfactory outcomes through execution of their skills (Bandura, 1986). This concept has been used in social work literature to examine the perceived abilities of social workers (or students) to complete professional tasks. For example, feelings of self-efficacy are enhanced by practice learning (Parker, 2006), professional education (Holden et al., 2002) or continuing professional development (Carpenter et al., 2015; Jack & Donnellan, 2010), and impaired by limited professional autonomy in terms of heavy workloads and strict time-management (Jack & Donnellan, 2010).

In accordance with Bandura’s (1986) concept of self-efficacy, respondents’ agreement with two reflective indicators – Y1 (‘I believe in the value of my job’) and Y2 (‘I have accomplished many worthwhile things in this job’) – were measured by a five-point Likert scale. In short, this study assumed that a higher level of professionalisation could result in a higher level of work-related self-efficacy.

Using single-worded items as reflective indicators is commonly found in existing studies (e.g. Diamantopoulos & Siguaw, 2006; Diamantopoulos & Winklhofer, 2001). Moreover, both indicators not only showed better model fits in preliminary tests, but also generated strong loadings on the latent construct of professionalisation in the final results of this study (see Table 2).

OPEN IN VIEWER

Table 2.

PLS results.

Measurement path
Formative indicators		VIF	Weight	Standard error	T statistics	p value
Public recognition	X1 Self-perception	1.283	0.310	0.146	2.117	0.035*
	X2 Qualification level	1.407	0.193	0.183	1.054	0.292
Knowledge base	X3 Knowledge adequacy	1.228	0.509	0.138	3.692	0.000*
Professional education	X4 Educational background	1.310	−0.261	0.159	1.647	0.100
Professional organisations	X5 Satisfaction	1.113	0.197	0.146	1.349	0.178
	X6 Involvement	1.190	−0.266	0.133	2.009	0.045*
Ethical standards	X7 Congruence	1.451	0.309	0.128	2.415	0.016*
Monopoly	X8 Necessity	1.166	−0.030	0.169	0.176	0.860
	X9 Recruitment criteria	1.290	0.073	0.142	0.515	0.607
Professional autonomy	X10 Self-perception	1.469	0.179	0.144	1.240	0.216
Prestige and remuneration	X11 Salary	1.274	−0.152	0.150	1.007	0.314
Measurement path			Loading	Standard error	T statistics	p value
Professionalisation →Y1 (I believe in the value of my job)			0.918	0.031	29.418	0.000
Professionalisation→Y2 (I have accomplished many worthwhile things in this job)			0.728	0.090	8.103	0.000

*Denotes significant formative indicators (p < 0.05).

In short, this study hypothesises a positive relationship between each of the 11 formative indicators and the latent construct of professionalisation which can be reflected by self-efficacy (see Figure 1).

OPEN IN VIEWER

Figure 1.

Measurement models.

Next, the multi-collinearity of all the formative indicators was tested using the variance inflation factor (VIF) because excessive collinearity hinders detection of each indicator’s effect on the latent construct. Previous studies suggest the VIF value should be less than 10 (Diamantopoulos & Winklhofer, 2001), 5 (Hair et al., 2011), 3.3 (Wang et al., 2011) or even 3 (Diamantopoulos & Siguaw, 2006).

In order to establish the formative model, the study adopted the tool of partial least squares (PLS). PLS analysis is a non-parametric structural equation modelling technique that is appropriate for establishing a formative model, especially when the research objective is to predict or develop theories (Arnett et al., 2003). As a causal modelling approach, PLS not only aims to maximise the explained variance of the dependent latent constructs, but also works effectively with a smaller sample size and non-normal data. Therefore, this technique can avoid the problem of indeterminacy. As some authors conclude, PLS analysis, if appropriately applied, is ‘indeed a silver bullet’ for estimating formative models in empirical research (Hair et al., 2011).

The student version 3.0 of SmartPLS was used to process and analyse the data. After running the bootstrapping algorithm with 5000 samples, three steps were taken to determine the validated formative indicators. First, indicators with significant values or weights should be retained (Hair et al., 2011; Wang et al., 2011). Second, indicators with significant loadings and above the benchmark of 0.5 (Hair et al., 2012a) should be retained even if their weights were insignificant.

Third, the implications for the theoretical model of dropping any formative indicator whose weight and loading are each non-significant should be carefully considered because, as previously elaborated, any change of formative indicators may affect the content validity of the construct (Diamantopoulos et al., 2008; Diamantopoulos & Winklhofer, 2001; Hair, et al., 2011).

Results

Discussion

The contribution of this study may be summarised as follows. From a methodological point of view, this study is the first to extend Weiss-Gal and Welbourne’s (2007, 2008) global analysis to social work in China, which has experienced substantial development in the last decade. Second, Weiss-Gal and Welbourne’s (2007, 2008) original framework was improved by developing a questionnaire comprising 11 indicators and then using formative measurement to validate 5 indicators for China: public recognition (X1 – self-perception), knowledge base (X3 – knowledge adequacy), professional organisations (X6 – involvement), ethical standards (X7 – congruence) and professional autonomy (X10 – self-perception). Third, by careful consideration of formative and reflective measures in the research design, the study properly employed a formative model to generate findings. Fourth, since Guangzhou is held up as epitomising social work development in China, the study provides important implications for other areas in China where social work professionalism may be less well advanced.

The significance of perceived public recognition (X1) echoes previous studies demonstrating that social workers’ perceptions of public status could affect other professionally relevant issues, such as graduates’ job intention (Wang et al., 2019) or turnover among social workers (Lei et al., 2018). This finding clearly demonstrates the importance of including subjective evaluation in assessing public recognition, instead of simply focusing on the licensure system.

Similarly, validation of the knowledge base dimension is also consistent with current literature which indicates the importance of translating theories into practice. For example, in a Delphi study of social work educators most agreed on the primary importance of knowledge for the profession (Yan & Tsang, 2005). Although students may have doubts about the usefulness of knowledge, they come to realise the value of increasing their knowledge once they enter the field (Wong & Pearson, 2007). Early-career social workers also acknowledged the contribution of knowledge as a reference framework in practice (Yan et al., 2013).

Although there are no state sanctioned codes of conduct for social workers in China, this study showed the importance respondents attached to acting in accordance with ethical standards. As Leung (2007) maintains, ‘China would have few problems in accepting the importance of those social work values prescribed by the international definition (p. 395)’. However, such acceptance may be limited to those values and ethics that are perceived as compatible with the political ideologies of the authoritarian state (Hutchings & Taylor, 2007; Lei & Huang, 2018).

Professional autonomy is another topic highlighted in current literature. Existing studies are mostly concerned about the relationship between the independence of social work and the ‘enduring and strong’ political domain (Yip, 2007, p. 60). As research on IFSC service providers indicated, these NGOs were not only assigned extra tasks that exceeded the specifications of the service contracts, they also had limited autonomy in deciding service delivery (Chan & Lei, 2017). In Shenzhen, social workers from NGOs sent to work in statutory settings by local governments may be compelled to undertake non-professional duties in exchange for being accepted by their host agencies (Leung, 2012).

The remaining formative indicators that were not statistically significant for indicating social work professionalisation in China need thorough discussion. The issue regarding professional education is perhaps a result of Chinese students not being adequately prepared to become a professional during the learning process. For example, the integral elements of field education – such as policy support, field instructors, placement sites and relevant research activities – are generally absent from social work education in China (Ting & Zhang, 2012). Even in a well-regarded university in Shanghai where social work is considered to be better developed, field education still suffered deficiencies of ill-defined learning objectives in the curriculum and insufficient support from agencies and supervisors (Cai et al., 2018). As a result, practicum experiences were described by graduates as ‘informal, short and shallow’ (Yan et al., 2013, p. 543) and it was even reported that the placement made no difference to the students’ perceived competencies (Wang et al., 2019).

The dilemma of social work development in China is more exposed by the absence of monopoly role and public recognition in terms of qualifications. The Chinese government intended to professionalise 1.45 million social workers by means of licensure exams and creating job positions (Organisation Department of the CCCPC, 2012). However, this study revealed that qualification levels, recruitment criteria and self-perceptions of a monopoly role for social work have not contributed to professionalisation. These findings merit immediate attention from policy makers, because it not only reveals the unsatisfactory practical outcomes of policies related to social work professionalisation, but also exposes the paradox in which an elastic boundary of professional activities is being built by the ambitious state project of professionalisation.

The lack of impact of salary for professionalisation is contrary to the conceptual framework and previous empirical findings (Cho, 2017; Yan et al., 2013; Zeng et al., 2016). A possible reason for this is that the salary range for social workers in the IFSCs – less than two-thirds of average salaries in Guangzhou – was simply too low to have any influence on professionalisation (Lei et al., 2015, p. 84).

The radical critique of professionalisation has always been concerned about professional power overriding the rights of service users. For Wilding (1982), for example, the overwhelming influence of the caring professions on policy making needs assessment and resource allocation was to neglect service users’ rights, impair their agency or abrogate accountability for services. As a profession with ‘the most developed strand of “anti-professional” theorizing’, social work in developed countries has often been criticised for prioritising its own privileges over the interests of service users (Hugman, 2003, pp. 183–184).

In contrast to these critiques, the findings of this study reflect a distinctive feature of contemporary professionalisation in China: relatively weak professional power. As conceptualised by Weiss-Gal and Welbourne (2008), internal and external professional power refer to the influence of the professional community inside or outside the profession, respectively. As this study demonstrates, some indicators that could be related to internal power (e.g. public recognition in terms of qualification levels, professional education, satisfaction with professional organisations) and external power (e.g. both subjective perceptions and objective recruitment criteria in service monopoly) proved to have insignificant impact on social work professionalisation in China. The irrelevance of these indicators suggests that the pressing focus for social work professionalisation in China should centre on insufficient professional power rather than over-concern about social workers’ hegemony, as suggested in radical critiques.

The disempowerment of the professional social work community in China may be related to the overriding power of the authoritarian state. Yip (2007) described the professional domain in China as ‘budding but weak’ (p. 59) in knowledge, accreditation and education. He argued that it had been subordinate to both the political and market domains within the tri-dimensional dynamics. Therefore, it is not surprising to discover that the social work profession has become a tool of government, identifying individual treatment of personal failings as the means of solving social problems (Leung et al., 2012).

Despite this, the scope for the professional community to influence professionalisation in China should not be ignored. On the one hand, Chinese social work experts may not choose ‘upholding human rights and social justice’ as a core competency, more because of concerns about the political constraints in applying these values than perceiving them as unimportant (Lei & Huang, 2018). As Huang and Zheng (2019) describe, situating the principle of human rights into Chinese social work is ‘hard, impossible or even dangerous . . . given the current social and political contexts in China’ (p. 616). Hutchings and Taylor (2007) also share a deep concern about the principle of social justice, because its implementation usually demands social action which is quite difficult for Chinese social workers to engage with ‘in a conservative country’ (p. 387).

On the other hand, there are positive factors which can help the professional community to generate incremental progress despite within political constraints. The Chinese governments have recognised the urgent needs of developing social services and this is why they initiated and endorsed the professionalisation of social work with significant administrative and financial supports (Leung, 2007). This open-minded and supportive attitude towards social work may leave certain space for the professional community to advocate for changes. Moreover, a large amount of over 1.2 million social workers existing in China (MCA, 2019) means that the collective voices of the profession can be influential within both the statutory settings and the non-governmental sector. The influences of the professional community, however, will depend on the contextually bounded strategies developed by Chinese social workers. This means that ‘clear rules and expectations may be set to avoid potential ethical and other conflicts’ (Huang & Zheng, 2019, p. 616).

Conclusion

This study applied Weiss-Gal and Welbourne’s (2007, 2008) conceptual framework to social work development in China. By using formative measures, this study validated five indicators, including perception of public recognition, perceived adequacy of the knowledge base, involvement in professional organisations (in terms of continued training), congruence with ethical standards, and perception of professional autonomy. The unimportance of qualification levels for public recognition, professional education, satisfaction with professional organisations, monopoly, and prestige and remuneration require serious consideration by social work academia and policy makers.

The achievement of the Chinese government in ‘building a large team of social workers’ in the past decade (over 1.2 million by 2019) must be acknowledged. However, the invalidated indicators as mentioned above suggest that the current measures implemented by the Chinese governments to professionalise social workers, such as launching licensure exams, promoting professional education, creating job positions and remuneration systems, may not impact on professionalisation as intended. These findings also imply that social work professionalisation in China may depart from the international standards of social work, because some indicators that are perceived as integral to social work development elsewhere appear to be inconsequential for social work professionalisation in China. Lastly, as social work professionalisation in China is a state-led project, the findings reflect the professional community’s relative weakness. The issue of how the professional community can be empowered to influence professionalisation in an authoritarian context remains a continuing but significant challenge for Chinese social workers.