Psychometric evaluation of the Chinese version of the service user psychological empowerment scale in a sample of youth service users

ABSTRACT

This study aimed to evaluate the psychometric properties of the translated Chinese version of the Service User Psychological Empowerment Scale (C-SUPES) in a youth service setting in Hong Kong. Data were collected from a cross-sectional survey of 381 youth residing in Hong Kong and participating in services provided by community-based children and youth services centres. The results of confirmatory factor analysis indicate a good fit for the proposed threefold factor structure with the original Positive Attitude and Involved Attitude subscales combined. The scale’s criterion validity was supported by significant correlations between creative self-efficacy, youth – adult partnerships, and civic engagement in the community, and its internal consistency reliability was satisfactory. Overall, youth service users with a higher level of participation in youth services reported a higher level of psychological empowerment. Altogether, the results suggest that C-SUPES is a reliable, valid instrument for measuring psychological empowerment among youth service users in Hong Kong.

KEYWORDS:

• Psychological empowerment
• youth service
• service user
• empowerment
• scale validation

Introduction

In recent years, empowerment has provided an attractive conceptual and practical framework for developing proactive, critical youth services (e.g. Jennings et al., 2006). The concept challenges the deficit approach of working with youth by perceiving young people as having their own strengths, assets, and resources (Planas-Lladó & Úcar, 2022; To, 2009; To et al., 2021; Zimmerman et al., 2018) and acknowledges that young people have a distinct capacity to engage in community-based activities for change and achieve positive development (Nicholas et al., 2019, Zimmerman et al. 2018). Going beyond the personal dimension, empowerment facilitates youths’ critical reflection on sociopolitical processes as a means to address institutional and/or structural problems of society that adversely affect their lives (Jennings et al., 2006). Although many practitioners have indeed adopted the empowerment-oriented approach in youth services, more work is needed to develop suitable measures for assessing the dimensions and outcomes of empowerment among users of youth services (Planas-Lladó & Úcar, 2022).

Such work is important, for though scholars examining youth empowerment agree that empowerment can be applied among individuals, groups, or communities to support young people in gaining control and power over their lives in different social circumstances (Zimmerman, 2000), research on developing indicators of empowerment at the level of youth participation or in youth service settings is particularly needed (Planas-Lladó & Úcar, 2022, van Dop et al., 2016). Many youth organisations strive to connect with young people and actualise their competencies to pursue innovative practices and develop collaborative youth – adult partnerships (To et al., 2021). Because youth organisations worldwide have proven to benefit youth participation and empowerment, there is a pressing need to develop and utilise psychometrically valid and reliable tools for measuring youth empowerment from the perspective of youth service users, who are the chief target for youth empowerment.

Service providers and youth service users often work together to generate empowerment by collaborating with each other in designing and organising youth programmes (To & Liu, 2021). From the perspective of individual youth service users, a major focus of empowerment is to create opportunities for them to recognise and draw upon their abilities, heighten their consciousness of sociopolitical influences, and participate in the civic life of their communities and society. All of those possibilities can be understood and characterised as different dimensions and outcomes of empowering young people as they participate in youth services (Planas-Lladó & Úcar, 2022, Zimmerman et al., 2018).

According to Zimmerman (2000), empowerment at the individual level can be termed psychological empowerment, which includes beliefs about one’s strengths and competence to exert control and effect changes, one’s participation in democratic decision-making processes, and one’s understanding of and work within the sociopolitical environment via individual and collective actions. Zimmerman (2000) has also distinguished three components of psychological empowerment: intrapersonal, interactional, and behavioural. The intrapersonal component refers to how people affirm their individual capabilities and level of control to produce an intended effect or make differences in personal situations. The interactional component, by contrast, refers to how people develop critical awareness so that they can continue to explore the impact of the environment on individual and social problems and learn how to access social services and various resources. Last, the behavioral component refers to how people put forth individual or collective effort into impacting environmental forces that can help mitigate individual or social problems.

Adopting the service user psychological empowerment scale (SUPES) to measure youth service users’ psychological empowerment

Based on Zimmerman’s conceptualisation of psychological empowerment, van Dop et al. (2016) have developed and validated the Service User Psychological Empowerment Scale (SUPES) with the intention to measure the intrapersonal, interactional, and behavioural dimensions of psychological empowerment among service users. Seventy items were generated, among which 28 remained after a series of online surveys and focus groups. The 28-item SUPES was tested in the form of a self-reported online questionnaire with a sample of 349 service users in the context of the Public Centres for Social Welfare in Belgium. The participants were all adults receiving subsistence income benefits. The mean age of the sample was 37.7. The results of an exploratory factor analysis indicated a 20-item fourfold factor structure, namely social attitude, positive attitude, involved attitude, and professional client attitude. Social attitude reflects cooperation, specifying the behavioural and interactional dimensions of psychological empowerment; Positive Attitude reflects a positive view of personal competence and future, specifying the intrapersonal dimension of psychological empowerment; Involved attitude reflects group belonging, helping, and sharing behaviour, specifying the intrapersonal and behavioural dimensions of psychological empowerment; and Professional Client Attitude reflects the service user’s relationship with the social worker, specifying the interactional dimension of psychological empowerment. The results of a confirmatory factor analysis yielded a 14-item fourfold factor structure with a good model fit. van Dop et al. (2016) also found empirical support for the scale’s internal consistency and construct validity.

We adopted the SUPES in our study given its satisfactory psychometric properties as well as findings from a recent scoping review of instruments for measuring empowerment in social work (Noordink et al., 2021), which highlighted the SUPES as the only instrument targeting psychological empowerment of service users in social service organisations. Although the SUPES is an instrument with satisfactory psychometric properties, it has not been validated for use in Chinese societies. The SUPES does not have a Chinese translation, and there is a lack of data on how well it works in this cultural context. Furthermore, while it is common for social scientists to use Western-developed scales to study Chinese cognitions and behaviours, it is important to consider any potential challenges associated with these constructs because the cultural framework of youth empowerment can be different or even dichotomous between Western and Chinese societies (Shek, 2012, Wong et al., 2021). For instance, the Western view of the self differs from the traditional Chinese view. The Western perspective emphasises personal goals and self-differentiation, while the Chinese perspective prioritises social roles and achieving social harmony (Harter 2012; Wong et al., 2021). The development of a young person is considered to be holistic and relational and is shaped by their social surroundings (Chai et al., 2022). Due to these cultural differences, it is possible that the factor structure of the SUPES, especially regarding the relations between the factors of positive attitude, involved attitude, social attitude, and professional client attitudes, may differ from the original scale when used in a Chinese sociocultural context. Since there has been no attempt to validate the scale in a sample of youth service users or in a Chinese context, our research can be regarded as pioneering in its evaluation of the psychometric properties of the SUPES in a Chinese youth service setting. Despite the importance of highlighting the possible context-specificity of the measure of empowerment, we cannot rule out the possibility that the SUPES can be applied in other sociocultural contexts following a rigorous scale validation process and if the items are adapted to another specific group of service users (Noordink et al., 2023).

Correlations between the SUPES and related measures of psychological empowerment

Van Dop et al. (2016, p. 661) found that ‘The SUPES has proven capable of generating accurate assessments of all three dimensions of psychological empowerment’, which were grouped into four factors: social attitude, positive attitude, involved attitude, and professional client attitude. While there is a lack of another measure of the same construct of psychological empowerment, in order to evaluate the criterion validity of the SUPES, this study explored the correlations between the SUPES and measures that align with the three dimensions of psychological empowerment. These measures include the Creative Self-Efficacy Scale (CSES), the Youth – Adult Partnership Questionnaire (YAPQ), and the Civic Engagement Scale (CES).

Creative self-efficacy

Psychological empowerment at the intrapersonal dimension constitutes a positive, proactive approach to youth development in which young people are believed to have self-efficacy and coping abilities to solve their problems and pursue positive self-transformation (Pearrow, 2008, Zimmerman et al., 2018). A constructive socioecological context that facilitates the enhancement of youths’ level of control is also beneficial to building their confidence in developing competencies to become active agents of change and innovation (Christens et al., 2011, Nicholas et al., 2019). Perceiving themselves as being in control of how they can foster change, young people who feel more empowered psychologically have a higher tendency to tackle creative tasks and produce creative outputs (To & Liu, 2021; Torres-Harding et al., 2018). Thus, this study hypothesised that youth with a higher level of psychological empowerment would have a higher level of creative self-efficacy, meaning a subjective belief that one can apply skills creatively and produce creative outcomes (Tierney & Farmer, 2002).

Youth–adult partnerships

Youth – adult partnerships refer to the various acts of collaboration and co-creation between youth and adults that result in young people’s having higher levels of participation and greater personal power in making decisions at an organisational or community level (Gallant et al., 2010). Youth – adult partnerships have been found to generate psychological empowerment at the interactional dimension. For example, Akiva et al. (2014) found that encouraging youth to interact with positive adult role models, involving them in decision-making processes, and supporting them to make real contributions to organisational and community life were positively correlated with their psychological empowerment. Zimmerman et al. (2018) also found that school and youth service settings designed to empower youth to share power with adults in designing and implementing meaningful, prosocial organisational activities can nurture youths’ self-efficacy and leadership skills. Therefore, this study hypothesised that youth with a higher level of psychological empowerment would have a higher level of youth – adult partnerships.

Civic engagement in the community

According to Christens et al. (2011), civic engagement can be regarded as a behavioural dimension of psychological empowerment. By extension, youths’ civic engagement can be understood as how young people participate in the life of a community in order to enhance collective well-being, strengthen networks to improve the quality of community life, and/or help shape the community’s future (Pearrow, 2008). Representing Spanish youth at risk of social exclusion, Funes and Robles’s (2016) qualitative findings indicate the possible effects of psychological empowerment as a result of being involved in civic engagement. Another qualitative study undertaken by Torres-Harding et al. (2018) revealed that students who engaged in social activism projects showed an enhanced sense of belonging to the community as a result of their participation, which in turn contributed to their psychological empowerment. Thus, this study hypothesised that youth with a higher level of psychological empowerment would have a higher level of civic engagement in the community.

Possible differences in psychological empowerment among youth service users with different levels of youth participation

Given the growing popularity of the concepts of user participation and empowerment, many scholars and practitioners have placed those concepts at the centre of youth studies and practice. Although the two concepts are closely connected and although youth participation has been found to benefit youth development and empowerment (e.g. Nicholas et al., 2019), the ways in which different types and levels of youth participation relate to youth empowerment need further investigation. As argued by To (2009), along with user participation, youth empowerment places emphasis on the complexity of power with respect to youth services. Along those lines, it is necessary to deepen current understandings of power in youth organisations, synthesise personal and institutional perspectives of youth services, and explicate the roles of service providers involved in the empowerment process. More work should also be done to explore the possible differences in empowerment among youth with different types or levels of participation in youth service settings, which may show that important aspects of empowerment have been heeded during the enhancement of youth participation. Against this backdrop, this study hypothesised that the psychological empowerment of youth service users is stronger among users with a higher level of participation in youth organisations.

Method

Setting and participants

In Hong Kong, an ICYSC has been established in each district or catchment area with a population of 12,000 youths. In 2014–2015, the total number of members of all ICYSCs was 322,233 (Commission on Youth, 2018), and as of 2023, there were 139 ICYSCs across Hong Kong (Social Welfare Department, 2023) that provide professional social work interventions to children and youth, their significant others, and the community. As a centre-based service, ICYSC adopts a total person and community approach to the diverse needs of youth aged 6 to 24 years old (Social Welfare Department, 2024). The 21 ICYSCs involved in our study are managed by seven local NGOs and located across nine areas within 18 districts in Hong Kong. In our study, a service user is defined as someone who identifies themselves as a member or participant of an ICYSC and uses the facilities or participates in the programmes provided by the ICYSC. In addition to organising programmes, it is worth noting that ICYSCs in Hong Kong provide various non-programme-based services, such as sports and leisure facilities, study rooms, and libraries. Also, the services provided by ICYSCs are not limited to their physical locations. They offer outreach programmes and activities for youth online, especially since the COVID pandemic. An example of this is real-time streaming activities through Facebook.

For a sample, 381 youth aged 11–24 who resided in Hong Kong were openly recruited by 21 Integrated Children and Youth Services Centers (ICYSCs) that participated in the study. All youth included in this study were self-identified as a member or participant of one of those ICYSCs. This study adopted a broad age range of youth for three purposes: (a) representing diverse groups of youth service users in Hong Kong, (b) encompassing the diverse definitions of ‘youth’, ‘adolescence’, and ‘young people’, and (c) understanding how empowerment programmes should intervene with young people. The first purpose was to represent the wide range of youth who have been utilising services and programmes offered by youth services, particularly through ICYSCs, as shown in official records (Home and Youth Affairs Bureau, 2022, Social Welfare Department, 2024). Capturing such a representation is essential as it is the stepping stone to understanding the diversity of young people’s experiences and perspectives within youth services in Hong Kong. The second purpose was to cover the different definitions and understandings of ‘young people’, ‘adolescence’, and ‘youth’. These terms vary across different disciplines and cultures. The World Health Organization defines young people as those aged 10–24 years old (World Health Organization, 2024). In the discussion of adolescent health, Sawyer et al. (2012) used the terms ‘young people’, ‘adolescence’, and ‘youth’ interchangeably as they focused on the development of an individual who goes through biological changes and transitions of social role. The adoption of such a broad age range in this study, thus, ensures the inclusion of individuals who fall under different definitions of ‘young people’, ‘adolescence’, and ‘youth’ and allows us to examine the changing needs and characteristics of youth service users more comprehensively. The third reason was to understand how empowerment programmes should intervene with young service users. Studies have shown that youth-adult partnership has significantly contributed to the fostering and flourishing of positive youth development in empowerment programmes in Western and Asian settings (Jones & Perkins, 2006, Krauss et al., 2014, Zeldin et al., 2014), the adoption of a broad age range in this study hence allows the exploration of measuring the effectiveness of empowerment programmes across different developmental stages among young people.

The respondents were recruited and the data were collected from November to December 2022.

Table 1 details the sociodemographic characteristics of the youth participants. Overall, 52.8% were females, 58.0% were less than 17 years old, and 69.3% had completed senior secondary school or more.

Table 1. Sociodemographic characteristics of youth respondents as a percentage of the sample (n = 381).

Sociodemographic characteristics	%
Gender
Male	39.90
Female	52.76
Not willing to disclose	7.35
Age
11–13	14.70
14–16	43.31
17–20	32.28
21–24	9.71
Service Users’ Highest Education Level
Primary school	2.89
Junior secondary school	27.82
Senior secondary school	45.93
Post-secondary school	23.36

The approximate indication of age in different education levels in Hong Kong is as follows.

1. Primary school: Students in primary school are typically between the ages of 5 to 12 years old. Primary 1 usually starts at 5 to 6 years old, and Primary 6 is completed at 11 to 12 years old.

2. Secondary school.

• junior secondary: Students in junior secondary are generally between the ages of 11 to 15 years old, covering Secondary 1 to Secondary 3.

• senior secondary: Senior secondary covers Secondary 4 to Secondary 6, with students typically aged between 14 to 18 years old.

3. Post-secondary education: Students move on to higher education or other pathways typically around 18 years old.

Measures

The translated Chinese version of the service user psychological empowerment scale (C-SUPES)

The 14-item, four-factor SUPES (van Dop et al., 2016) was adopted for validation in our sample of young service users. The items were shown in Table 3, which were rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree), instead of the original 5-point Likert scale by van Dop et al. (2016). By presenting consistent and clear answering instructions and scale categories throughout the self-administered online questionnaire interface, we attempted to minimise the possible cognitive overload associated with confusing displays (Evans & Mathur, 2005, Ray & Tabor, 2003). Statistically, a 7-category rating scale has also been advocated by several researchers as having the highest reliability (e.g. Miller, 1956, Oaster, 1989). The original four factors were Social Attitude, Positive Attitude, Involved Attitude, and Professional Client Attitude. The ranges of the total scores of the original four subscales were 5–35 for Social Attitude and 3–21 for Positive Attitude, Involved Attitude, and Professional Client Attitude.

Mr. Jan Depauw, the scale’s developer, granted us permission to translate and validate the C-SUPES. To validate the 14-item SUPES in our Chinese context, we translated the SUPES from English to Chinese with the aid of two research assistants who are proficient in both languages and familiar with youth services in Hong Kong. The items were then back-translated into English by another research assistant. The translated version was subsequently reviewed by an expert panel consisting of seven senior practitioners at ICYSCs and four researchers.

The translated Chinese version was pilot-tested with and reviewed by 20 youth aged 12–23. According to their feedback, the original English wordings of the SUPES were modified if determined to be unsuitable for the context of youth social services in Hong Kong.

In our validation study, the composite scores of the C-SUPES and its subscales were the mean scores, ranging from 1 to 7, computed by dividing the total scores by the respective number of items.

Creative self-efficacy

The Creative Self-Efficacy Scale (CSES), originally developed by Tierney and Farmer (2002), measures one’s beliefs about their ability to be creative using three items (e.g. ‘I have confidence in my ability to solve problems creatively’), each rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). The scale was translated into Chinese by Gong et al. (2009). In our study, Cronbach’s alpha of the CSES was 0.86.

Youth–adult partnerships

The Youth – Adult Partnership Questionnaire (YAPQ) was developed by Zeldin et al. (2014) to measure young people’s perception of their participation and partnership with youth workers. The YAPQ has nine items, including ‘Youth and adult youth workers learn a lot from working together in this center’. The scale, rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree), was translated into Chinese and validated by To et al. (2021). In our study, Cronbach’s alpha of the YAPQ was 0.93.

Civic engagement in the community

We also adopted the civic behaviours component of the Civic Engagement Scale (CES) developed by Doolittle and Faul (2013), which measures respondents’ level of participation in actively attempting to engage in and make a difference in their own community. The CES’s six items (e.g. ‘I stay informed of events in my community’) are rated on a 7-point Likert scale ranging from 1 (never) to 7 (always). The scale was translated into Chinese and validated by Fong and To (2022). In our study, Cronbach’s alpha of the CES was 0.90.

Level of participation in youth programmes

According to Wong et al. (2010), there are three broad types of youth participation in youth programmes: adult-driven, shared control, and youth-driven. Regarding the type of adult-driven participation coined ‘vessel’, programmes are adult-driven and demand little to no input from young people. Because of the lack of youth involvement, low empowerment potential can be expected from this type of participation. In our study, in order to quantify youth’s participation levels in youth programmes apart from using centre facilities or joining online activities, respondents were asked to select from a spectrum of eight descriptors that best describe their face-to-face participation in youth programmes organised by ICYSCs in the past year (i.e. from 2021 December to 2022 December). The first three descriptors, from ‘I have no face-to-face participation in the related programs’ to ‘I only participated in related programs’, were defined as having a ‘lower level of participation in youth programs’, which reflects the type of adult-driven participation argued by Wong et al. (2010). Those who selected from ‘I assisted in the development of a program’ to ‘I played a leading role in the design and delivery of a program’ were defined as having a ‘higher level of participation in youth programs’, which reflects the types of shared control and youth-driven participation (Wong et al., 2010).

Data collection procedures

Prior to data collection, ethics approval was obtained from the Chinese University of Hong Kong. Permission was also obtained from the participating ICYSCs, the youth service users, and the parents of participants less than 18 years old. Youth service users were invited by their social workers to complete a self-administered online questionnaire. Upon completing the questionnaire, the respondents were given a cash incentive of HK $50.

Data analysis procedures

Data analyses were conducted using SPSS version 28 and Amos version 28. To investigate whether the original four-factor structure was applicable to our sample, a preliminary confirmatory factor analysis (CFA) was conducted to investigate the model fit. As recommended by several researchers, a ratio of chi-square to the degrees of freedom (CMIN/df) of 2 to 3 indicates acceptable model fit (Hair et al., 2019). Meanwhile, the comparative fit index (CFI) and Tucker – Lewis Index (TLI) values range from 0 to 1, with values exceeding 0.95 are considered to indicate good fit (Hu & Bentler, 1999). Last, the root mean square error of approximation (RMSEA) values range from 0 to 1, with values of 0.06 or less indicative of good fit (Hu & Bentler, 1999). The standardised estimates and the correlations between factors from the CFA were also examined.

Although gender differences may have been considered in youth development and empowerment in the Chinese cultural context, measurement invariance was established across gender groups in previous related Chinese studies (e.g. Chai et al., 2022, Shek & Ma, 2010). To confirm that the C-SUPES is a stable measure across male and female youth in a Chinese context, a multi-group confirmatory factor analysis (MGCFA) was also conducted to assess the increasingly stringent models from configural, metric, and scalar models in a stepwise manner. A non-significant Chi-square difference test result with changes in values of CFI < .01 (Cheung & Rensvold, 2002), and RMSEA < .015 (Chen, 2007) indicate invariance at that level.

After confirming the model fit of the proposed structure, we examined item – total correlations and internal consistency values for evidence of construct validity. Cronbach’s alpha coefficient and McDonald’s omega coefficient of 0.8 or more are recommended as indications of acceptable reliability (Nunnally & Bernstein, 1994). Meanwhile, factor loading estimates should be 0.5 or higher and statistically significant (Hair et al., 2019), while values of item – total correlation exceeding 0.3 are considered to contribute sufficiently to a scale’s internal consistency (Polit & Beck, 2006). The criterion validity of the proposed structure was also explored by examining correlations between the C-SUPES, its subscales, and the empowerment-related measures. Last, the applicability of the C-SUPES was explored using an independent sample t test to examine whether youth service users with a higher level of participation in ICYSC feel more psychologically empowered.

Results

Confirmatory factor analysis

A preliminary CFA was performed to investigate the model fit. The results show that the standardised loading estimates of all items were statistically significant at the 0.001 level, and the loading values exceeded 0.5, in a range from 0.62 to 0.89. The goodness-of-fit statistics also indicated a good model fit. CMIN/df (2.36) fell between 2 and 3, which indicates a good fit, as did CFI (0.97) and TLI (0.96) values exceeding 0.95 and an RMSEA (0.06) that met the cut-off value of 0.06. However, the correlation between Positive Attitude and Involved Attitude was exceptionally high (0.96), which suggests possible multicollinearity (Hair et al., 2019) (The results can be provided by the authors upon request).

Because the high correlation between Positive Attitude and Involved Attitude suggests that those two subscales may measure the same constructs in our sample, an exploratory factor analysis was conducted to illuminate the factor structure of the scale. The results, which can be provided by the authors upon request, support a three-factor model with the two subscales combined. A restructured factor model was thus proposed, and another CFA was conducted to test whether a three-factor structure would have a good model fit but be free of the risk of multicollinearity. The new path diagram, shown in Figure 1, has all items with factor loadings above the recommended cut-off value of 0.5 (Hair et al., 2019). Moreover, the correlations between the three factors, ranging from 0.67 to 0.68, reflect that they are related but distinct constructs. The goodness-of-fit statistics, including CMIN/df (2.48), TLI (0.95), CFI (0.96), and RMSEA (0.06), indicate a good model fit. Therefore, we combined the Positive Attitude and Involved Attitude subscales. Because all six items involved concerned how youth service users view and position themselves with an emphasis on their own character and conditions, the combined factor was labelled ‘Personal Attitude’.

short-legendFigure 1.

A MGCFA was then conducted to confirm that the resulting factor structure was stable across male and female youth. As shown in Table 2, the values of changes in CFI and RMSEA as the models were progressively constrained from configural model to metric model, and from metric model to scalar model with non-significant Chi-square differences suggest that the model was invariant across gender at scalar level.

Table 2. Model comparisons: Impact of metric and scalar invariance constraints on model fit in the male and female samples.

Model	χ2	∆χ2	df	χ2/df	CFI	RMSEA
Configural invariance model	330.08		148	2.23	0.934	0.059
Metric invariance model	354.81	24.73	159	2.23	0.929	0.059
Scalar invariance model	362.65	7.84	165	2.20	0.928	0.058

Item analysis and reliability analysis

Apart from the adequate factor loadings, the item-total (factor) correlations from 0.54 to 0.81, shown in Table 3, also suggest that the items measure the corresponding factors. The Cronbach’s alpha coefficient and McDonald’s omega coefficient of the full scale and its three subscales all exceeded 0.8, as shown in Table 4, which suggests the three-factor model’s good reliability.

Table 3. Item analysis of the proposed 3-factor structure.

	M	SD	Item-total Correlation
Factor 1: Social Attitude
Q1_SA. Working with others in community can help to change things for the better.	5.71	1.07	0.61
Q2_SA. I experience peace and security in my community.	5.69	1.16	0.61
Q3_SA. I take satisfaction in what goes well.	5.68	1.04	0.70
Q4_SA. The only way I can have power is by connecting to others.	5.22	1.30	0.56
Q5_SA. This society allows space for my way of doing things.	5.39	1.27	0.54
Factor 2: Personal Attitude
Q6_PA. I know my strengths.	5.39	1.35	0.61
Q7_PA. I have a structure in my life.	5.43	1.39	0.80
Q8_PA. I am generally optimistic about the future.	5.15	1.51	0.81
Q9_IA. I feel as if I am part of a larger whole.	5.10	1.52	0.80
Q10_IA. I can share my experience with others.	5.52	1.24	0.64
Q11_IA. I feel supported by the people I love.	5.59	1.28	0.63
Factor 3: Professional Client Attitude
Q12_PCA. My social worker and I have a good cooperative relationship.	6.09	0.93	0.69
Q13_PCA. I can state my thoughts freely to my social worker.	5.86	1.13	0.79
Q14_PCA. The help provided by social workers fits my life well.	5.79	1.09	0.78

Table 4. Internal consistency based on the proposed 3-factor structure.

	M (1–7)	SD	Cronbach’s alpha	McDonald’s omega
Whole Scale	5.54	0.86	0.92	0.92
Factor 1: Social Attitude	5.54	0.88	0.81	0.81
Factor 2: Personal Attitude	5.36	1.12	0.89	0.90
Factor 3: Professional Client Attitude	5.91	0.94	0.87	0.88

Criterion validity

As hypothesised, the C-SUPES correlated significantly with the related measures on creative self-efficacy, youth – adult partnership, and civic engagement in the community at the full-scale level. Furthermore, correlations were found at the subscale level (see Table 5). The subscale of Social Attitude, which consists of items tapping the interactional and behavioural dimensions of empowerment, correlated significantly with youth – adult partnerships (i.e. strong correlation) and civic engagement in the community (i.e. moderate correlation). The subscale of Personal Attitude, consisting of items tapping into intrapersonal and behavioural dimensions of empowerment, correlated significantly with creative self-efficacy (i.e. strong correlation) and civic engagement in the community (i.e. moderate correlation). The subscale of Professional Client Attitude, by contrast, consisting of items from the interactional dimension, had a significantly strong correlation with youth – adult partnerships.

Table 5. Correlations among the whole scale & 3 subscales of SUPES and related constructs.

	M	SD	1	2	3	4	5	6
1. Service User Empowerment	5.54	0.87
2. Social Attitude	5.54	0.88	0.84**
3. Personal Attitude	5.36	1.12	0.92**	0.62**
4. Professional Client Attitude	5.91	0.94	0.78**	0.57**	0.62**
5. Creative Self-Efficacy	5.13	1.16	0.58**	0.48**	0.56**	0.40**
6. Youth-adult Partnerships	5.94	0.78	0.70**	0.68**	0.55**	0.62**	0.40**
7. Civic Engagement in the community	4.51	1.20	0.48**	0.41**	0.45**	0.35**	0.49**	0.37**

**Correlation is significant at the 0.01 level (2-tailed).

Applicability of the scale in testing the difference in psychological empowerment among youth service users with different levels of participation in youth programmes

The applicability of the C-SUPES in testing the difference in psychological empowerment between youth service users with higher versus lower levels of participation in youth programmes was also analysed. In our survey, youth service users with a lower level of participation indicated that they did not have face-to-face participation in the youth programmes organised by the service centres, only browsed information about youth programmes on offer, or only took part in the youth programmes as participants. By contrast, youth service users with a higher level of participation indicated taking part in the organisation of the youth programmes more actively, from giving advice on the service design to playing a leading role in the design and implementation of youth programmes. The independent sample t test (see Table 6) showed that youth with a higher level of participation in youth programmes indeed perceived themselves as being more empowered in general and in all three aspects, and the effect sizes in terms of Cohen’s d all exceeded 0.8, which are considered to be large (Cohen, 1988).

Table 6. Independent sample t test comparing the service user empowerment between youth service users having higher (n = 189) and lower (n = 192) levels of participation.

Scales	Participation Level	Mean (SD)	t	Effect size (Cohen’s d)
Service User Empowerment	Lower	5.42 (0.88)	−2.78**	0.86
	Higher	5.67 (0.84)
Social Attitude	Lower	5.44 (0.90)	−2.09*	0.88
	Higher	5.63 (0.85)
Personal Attitude	Lower	5.24 (1.10)	−2.11*	1.11
	Higher	5.48 (1.13)
Professional Client Attitude	Lower	5.74 (0.99)	−3.66***	0.92
	Higher	6.09 (0.84)

*p ≤ .05.**p ≤ .005.***p ≤ .001.

Discussion

Our findings indicate that the factor structure of the C-SUPES differs slightly from the original SUPES, thereby implying that the underlying dimensions of psychological empowerment among youth service users in Hong Kong require further interpretation. The results of CFA indicate that instead of the original four-factor model, a three-factor structure combining the original subscales of Positive Attitude and Involved Attitude into a new subscale of Personal Attitude demonstrates a better model fit free from the risk of multicollinearity. The scalar invariance established from the MGCFA suggests that the three-factor model applies to both male and female youth. In addition to the statistical evidence of its applicability, this study reveals that the original structure of SUPES needs modification when being adopted in Chinese societies. The result can be explained from developmental, cultural, and contextual perspectives.

From a developmental perspective, youth encounter two dimensions of psychosocial development in fulfilling their life tasks and challenges: competence for younger youth and fidelity for older youth (Erikson, 1993). Competence refers to the psychosocial strength arising from a sense of industry that enables a person to deal effectively with the environment. Fidelity refers to the psychosocial virtue that enables a young person to acquire a personal identity by consolidating all the knowledge they have gained about themselves.

The construct of competence for younger youth can be understood as closely relating to the involved attitudes of youth empowerment. All three items of the Involved Attitude subscale of the C-SUPES concern how young people connect with their significant others and the surrounding environment. These include feeling part of a larger whole, sharing experiences with others, and feeling supported by loved ones. When young people’s existence is accepted by others, they can develop an expanded and healthy self-concept (To et al., 2014). This helps them to develop a deeper level of self-understanding, identify their personal goals, and embrace hope for the future, as older youth are challenged to consolidate their sense of fidelity. This can be reflected in all three items of the Positive Attitude Subscale, comprising knowing their strengths, having structure in their life, and being optimistic about the future. Because our sample involves both younger and older youth respondents, our findings support the combination of the Positive Attitude and Involved Attitude subscales to become the Personal Attitude subscale. This better reflects the developmental tasks of young people as a whole.

From a cultural perspective, the impact of Chinese cultural elements on the psychological empowerment of youth service users can exemplify various tenets of the collectivistic culture rooted in Confucianism, including upholding social order, prioritising human connections, and striving for harmony (Chai et al., 2022; Hwang, 2001; Wong et al., 2022; Yip, 2004). In contrast to the Positive Attitude and Involved Attitude constructs, Chinese culture does not separate positive self-development and a sense of belonging. Instead, these concepts are integrated under the notion of self-cultivation, which means people should be socially conscious during self-development (Hwang, 2001). This integration elucidates why the original Positive Attitude and Involved Attitude factors combined to form a single factor during the C-SUPES validation process.

From a contextual perspective, although van Dop et al. (2016)’s results of factor analysis on SUPES revealed a relatively high correlation (r = 0.84) between positive attitude and involved attitude, the researchers made the decision not to combine the two subscales. It should be noted that unlike our sample, which was taken from the general youth population, van Dop et al. (2016)’s Belgium sample featured adult subsistence recipients. Under this context, empowerment in terms of self-development and sense of belonging were independently called for, making the adaptation of distinct psychological empowerment subsets of both positive attitudes and involved attitude appropriate.

It is worth noting that, although the Personal Attitude subscale is formed by combining the Positive Attitude and Involved Attitude subscales, our current findings suggest that Personal Attitude and Social Attitude remain two distinct constructs and subscales. Hong Kong is a society that has been strongly influenced by both Western and Chinese cultures as a result of globalisation. The education system and curriculum in Hong Kong have predominantly emphasised Western knowledge and practices, which may account for why young people in Hong Kong are more aware of their social rights and community involvement (To, 2009). This could also explain why they prioritise building good social relationships to achieve positive development on the one hand, and actively engaging in their community’s civic lives on the other.

Our findings also show that the construct validity of the C-SUPES was established at the full-scale level and that its subscales were significantly correlated with measures in relation to youth empowerment. It is important to note that in the development and validation of the original SUPES, the construct validity was evidenced by the high, significant loadings of the items on the four factors of the scale (van Dop et al., 2016). In our study, the criterion validity was further evidenced by the correlations between the whole scale and subscales of the C-SUPES and empowerment-related measures, as explained below.

We hypothesised that psychological empowerment would emerge with the three closely related measures: creative self-efficacy, youth-adult partnerships, and civic engagement in the community. Our finding supports the hypothesised relationships at both whole scale and subscale levels, as significant correlations were demonstrated between the whole scale C-SUPES and the three related measures, and also between C-SUPES subscales encompassing items related to the underlying dimensions of psychological empowerment and the respective measures. Specifically, consistent with our hypothesis, creative self-efficacy was found to correlate significantly with Personal Attitude (e.g. I know my strengths), which consists of indicators representing the intrapersonal and behavioural dimensions of psychological empowerment; civic engagement in the community correlated significantly with Social Attitude (e.g. working with others in one’s community can help to change things for the better), which encompasses interactional and behavioural dimensions of psychological empowerment; and youth-adult partnerships had a significant correlation with Professional Client Attitude (e.g. I can state my thoughts freely to my social worker), which consists of items representing the interactional dimension of psychological empowerment. Consistent with van Dop et al. (2016)’s remark that SUPES was capable of assessing the three dimensions of psychological empowerment, our results of criterion validity assessment illuminate that C-SUPES, which was grouped into three components for our Chinese youth sample, measures psychological empowerment as a whole and is connected to the underlying dimensions of psychological empowerment (Zimmerman, 2000).

On the other hand, it was to our surprise that in addition to the hypothesised correlations, all the proposed related measures also correlated significantly to the other subscales of C-SUPES. For instance, the measure of youth-adult partnerships was found to correlate significantly not only to Professional Client Attitude but also to Personal Attitude and Social Attitude. This suggests that while representing a specific subscale, each component of C-SUPES is absorbed into the scale and may represent psychological empowerment as a whole. The correlations also indicate that while an aspect of C-SUPES serves to measure a particular component of psychological empowerment, all the components of psychological empowerment and related youth empowerment constructs are closely connected on their own.

Last, there were substantial differences in all three dimensions of psychological empowerment (i.e. personal attitude, professional client attitude, and social attitude) between youth with a higher versus lower level of participation in youth services. That finding suggests that youth who feel more psychologically empowered feel more willing and confident to contribute to the design and implementation of youth programmes. The finding is consistent with previous results concerning the relationship between those two variables. For example, youth involved in activities in organisations or their communities have shown a higher level of intrapersonal psychological empowerment (Zimmerman et al., 2018). Forenza’s (2016) study on a youth engagement programme has also suggested that participating in civic engagement can enhance the interactional and behavioural dimensions of youths’ psychological empowerment.

Because an effect size demonstrates a finding’s substantive significance, our finding’s large effect size suggests that youth participation can considerably enhance participants’ psychological empowerment. One explanation may be that the level of youth participation and empowerment has changed enormously as a result of youth service design and delivery in Hong Kong in the past two decades. On the one hand, far more emphasis has been placed on young people’s initiation and participation in youth services (To & Liu, 2021). Such emphasis on youth participation has prompted the proliferation of various youth empowerment projects that provide platforms for young people to enhance their engagement in their communities and society (To & Liu, 2021). On the other hand, some young people continue to assume the role of service recipients and do not actively participate in youth services. Under the influence of Chinese culture, such young people may prefer to play a more passive role when among authority figures such as social workers and may even question the value or purpose of their participation (To & Liu, 2021). That trend can explain the remarkable difference in psychological empowerment between youth with a higher versus lower level of participation.

Limitations

Our findings have several limitations. First, given the lack of a randomised representative sample in our study, the representativeness of the population and the generalisability of the findings are subject to scrutiny. Second, the target participants in our study were exclusively youth residing in Hong Kong. Future studies need to examine and analyse the psychometric properties and model fit of the questionnaire in other Chinese populations. Third, in view of the context-specificity of measures of empowerment, a bottom-up approach should be used to develop items for a scale on youth service users’ empowerment or to expand the pool of items for the translated scale so that youth service users’ perspectives are comprehensively recognised by researchers.

Despite its limitations, our findings can propel research on the structure of psychological empowerment in general and on youths’ psychological empowerment in particular, as well as its connections with other empowerment-related measures in Chinese contexts.

Implications

The findings of our study can contribute to research and practice in several ways. First, a validated Chinese version of the SUPES can guide the systematic assessment of young people’s psychological empowerment. The instrument can also provide evidence-based insights for youth work practitioners and other helping professionals in conducting comprehensive assessments of youth empowerment.

Second, based on a refined understanding of the empowerment of young service users in a Chinese context, a psychological empowerment framework with three dimensions (Personal Attitude, Social Attitude, and Professional Client Attitude) has been developed from this study to guide the work of youth service agencies. This framework provides practitioners with a tool to measure the current level of psychological empowerment of young people. By identifying the lacking dimensions, practitioners can set short-term service targets when designing programmes to support the development of young people.

Third, because our results show the practical significance of service users’ participation in youth programmes, social service agencies now have more empirical evidence to support their promotion of user participation in youth service settings. In that vein, how user participation can help to empower service users could also be investigated, as could whether user participation in settings other than community-based youth service centres, such as services for marginalised youth and youth with psychological and behavioural concerns (e.g. Forenza, 2016, Funes & Robles, 2016), might empower service users.

Last, having established the validity of the C-SUPES, it can be used to evaluate the outcomes of youth empowerment programmes. In response to a lack of measures that have operationalised the theoretical construct of psychological empowerment, the C-SUPES can be used to address the identified gap in research by providing a psychometrically valid scale to help practitioners assess the effectiveness of youth empowerment programmes and identify areas for improvement. Additional research can be conducted to gain a better understanding of the components and process of empowerment. For example, it would be valuable to investigate over time whether Professional Client Attitude serves as a significant protective factor for disadvantaged youth who have not yet developed strong Personal Attitude and Social Attitude.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was funded by the Hong Kong Jockey Club Charities Trust [Project Code: 6906496].

Notes on contributors

Siu-Ming To

Dr. Siu-Ming To is the Chairperson and an Associate Professor of the Department of Social Work at The Chinese University of Hong Kong.

Ji-Kang Chen

Dr. Ji-Kang Chen is the Head of Graduate Division and an Associate Professor of the Department of Social Work at The Chinese University of Hong Kong.

Johnson Chun-Sing Cheung

Dr. Johnson Chun-Sing Cheung is a Senior Lecturer of the Department of Social Work and Social Administration at the University of Hong Kong.

Siu-Ming Chan

Dr. Siu-Ming Chan is an Assistant Professor of the Department of Social and Behavioral Sciences at the City University of Hong Kong.

Ming-Wan Yan

Ms. Ming-Wai Yan is a Research Assistant of the Department of Social Work at The Chinese University of Hong Kong.

Man-Yuk Chan

Mr. Man-Yuk Chan is a Research Assistant of the Department of Social Work at The Chinese University of Hong Kong.

Alex Fong

Mr. Alex Fong is a PhD student of the School of Social Work at York University.