Introduction
The cruciality of institutional trust, particularly in its role in promoting voluntary compliance with legal authorities’ laws and orders, has sparked significant interest in understanding its origins (Marien & Hooghe, Reference Marien and Hooghe2011; Tyler, Reference Tyler1990). Procedural justice (Tyler, Reference Tyler2003) and performance theories (Kampen et al., Reference Kampen, Van De and Bouckaert2006; Van de Walle & Bouckaert, Reference Van de Walle and Bouckaert2003), developed separately in criminal justice and public management literature, offer crucial theoretical insights into the foundation of institutional trust. The core assumption in these research strands is that improving the quality of public services and regulatory institutions will boost citizens’ satisfaction with those services, thereby leading to an increase in institutional trust. These quality or performance dimensions encompass citizens’ assessments of justice, fairness, and effectiveness, collectively impacting institutional trust.
A nation’s CPS includes institutions, actors, and services aimed at protecting children from harm, be it from parents, caregivers, or the children themselves (Berrick, Gilbert et al., Reference Berrick, Gilbert and Skivenes2023; Gilbert et al., Reference Gilbert, Parton, Skivenes, Gilbert, Parton and Skivenes2011). The United Nations Convention on the Rights of the Child also obliges CPS of signatory countries to intervene in family situations when children’s rights are at risk or violated (Juhasz & Skivenes, Reference Juhasz and Skivenes2017). From this perspective, a CPS is mandated to carry out functions akin to law enforcement agencies and the judicial system in maintaining order and safeguarding children’s rights. At the same time, CPS, as an integral component of the welfare system in Western democracies, plays a role in distributing various social and economic benefits to different target groups. Therefore, it is crucial for a CPS to be seen as trustworthy to effectively intervene in families, distribute benefits efficiently, and secure compliance with its order (Gilbert et al., Reference Gilbert, Parton, Skivenes, Gilbert, Parton and Skivenes2011).
Increasing backlash against CPS from various social institutions and groups (Stang, Reference Stang2018) warrants deeper understanding of the factors associated with the legitimacy of CPS. However, prior research either on, for instance, when the child welfare interventions occur (Bennett et al., Reference Bennett, Schlüter, Melis, Webb, Reddy, Barr, Wickham and Taylor-Robinson2022) or support for child welfare interventions varies (Berrick, Skivenes et al., Reference Berrick, Skivenes and Roscoe2023; Loen & Skivenes, Reference Loen and Skivenes2023; Skivenes et al., Reference Skivenes, Falch-Eriksen and Hassan2023). Comparative research on legitimacy of CPS and its origins is very nascent. For instance, in a 2016 four-country study, 50.2 per cent of Norwegians and 40.6per cent of Britons expressed significant confidence in their CPS (Juhasz & Skivenes, Reference Juhasz and Skivenes2017). In a related nine-country study in 2022, Norwegians demonstrated higher trust in their CPS (mean 2.69) compared to their British counterparts (mean 2.21) on a scale ranging from 1 ‘very little confidence’ to 4 ‘a great deal of confidence’ (Skivenes & Benbenishty, Reference Skivenes and Benbenisthy2022).
Furthermore, while a limited number of comparative studies have been undertaken, they predominantly offer insights into the legitimacy and trustworthiness of CPS based solely on socio-demographic factors and welfare typologies. One line of inquiry links trust in CPS with variables such as gender, age, left-right orientations, income, educational attainment, and parental status (Juhasz & Skivenes, Reference Juhasz and Skivenes2017). Another stream of research indicates that trust in CPS varies according to different child welfare typologies, with systems focused on child well-being garnering higher levels of trust, followed by those emphasizing family services and risk-oriented approaches (Skivenes & Benbenishty, Reference Skivenes and Benbenisthy2022). Moreover, additional research demonstrates a relatively elevated level of public confidence in CPS centred on family services compared to risk-oriented CPS (Loen & Skivenes, Reference Loen and Skivenes2023). This study takes a further step by examining the relationship between perceived performance and trust in CPS through a comparative analysis of the English and Norwegian contexts.
England and Norway are chosen for comparison due to their differing democratic systems, child welfare programs, and family welfare systems (Berrick et al., Reference Berrick, Dickens, Pösö and Skivenes2016; Berrick et al., Reference Berrick, Skivenes and Roscoe2021; Skivenes & Benbenishty, Reference Skivenes and Benbenisthy2022). Norway is characterised by social welfare democratic regimes, child-centred protection, and de-familialised systems, whereas England is marked by liberal democratic regimes, risk-oriented child protection, and familialised systems. In Norway, the connection between performance and trust may be particularly prominent due to the advantages that Norwegians derive from their child protection system. The suggested association is likely to be robust in Norway, primarily due to the nation’s higher degree of social protection.
This study contributes significantly to the research on trust in CPS and performance theory. It combines insights from political science (Norris, Reference Norris, Rosema, Denters and Arrts2011; Rothstein & Teorell, Reference Rothstein and Teorell2008), public management (Van de Walle & Bouckaert, Reference Van de Walle and Bouckaert2003; Van Ryzin, Reference Van Ryzin2011), and policing literature (Sunshine & Tyler, Reference Sunshine and Tyler2003; Tyler, Reference Tyler2003) to propose a positive association between perceived performance (distributive justice, procedural fairness, and functional effectiveness) and trust in CPS. Secondly, it operationalises performance through distributive justice, procedural fairness, and functional effectiveness, providing clarity within CPS context. Thirdly, it introduces a basic scale for functional effectiveness based on citizens’ perceptions of the knowledge, abilities, and financial resources needed to assist needy children (Skivenes, Reference Skivenes2021).
Lastly, based on cross-sectional survey data from English and Norwegian populations, and consistent with prior research (Juhasz & Skivenes, Reference Juhasz and Skivenes2017; Skivenes & Benbenisthy, Reference Skivenes and Benbenisthy2022), the paper indicates that Norwegians view their CPS as more trustworthy than their English counterparts. Further examination shows that, unlike procedural fairness and the competences aspect of functional effectiveness, which consistently correlate with trust in CPS in both countries, distributive justice does not affect trust in CPS in either country. Together, these three elements account for approximately 58 per cent of the variance in trust in CPS of both nations. Consequently, this study provides additional validation for the performance theory within the unique context of CPS.
Like the police and social services, CPS serves to safeguard children from abuse and aid vulnerable families in times of need. A CPS with higher trust is more likely to receive support for intervening in family matters (Loen & Skivenes, Reference Loen and Skivenes2023) and achieving compliance with their directives. On the other hand, lower levels of trust create challenges in achieving the fundamental goal of ensuring a fulfilling life for children, regardless of their family situations, including protection from abuse and neglect (Burns et al., Reference Burns, Helland, Križ, Sánchez-Cabezudo, Skivenes and Strömpl2021; Skivenes & Benbenisthy, Reference Skivenes and Benbenisthy2022). Simultaneously, understanding citizens’ views of CPS is crucial for comprehending trust in the political system. Nonetheless, even in well-developed welfare states, child protection authorities encounter severe criticism from various sectors of society (Stang, Reference Stang2018). Therefore, it is crucial to discern the factors that underlie trust in CPS.
The rest of this paper is divided into four sections: literature review, data and methods, findings, and conclusion.
Literature review
It is surprising that trust in the child protection system has been understudied, given the common state intervention in family lives in developed welfare states. This review fills this gap by incorporating insights from policing and public management literature to highlight the connection between perceived performance and trust in CPS.
Prior studies have considered perceived performance based on citizens’ views of various indicators, including impartiality (Rothstein & Teorell, Reference Rothstein and Teorell2008), processes, outcomes (Van de Walle & Migchelbrink, Reference Van de Walle and Migchelbrink2020; Van Ryzin, Reference Van Ryzin2011), and notions of justice, fairness, and effectiveness (Sunshine & Tyler, Reference Sunshine and Tyler2003; Tankebe, Reference Tankebe2008; Tyler, Reference Tyler2003). In the context of CPS, distributive justice pertains to citizens’ evaluations of the degree to which child protection authorities discriminate, such as based on socio-economic and cultural backgrounds, or perceptions of the distribution of rewards and punishments by these authorities. Procedural fairness denotes citizens’ evaluation of whether child protection authorities adhere to rules and procedures when making decisions. Functional effectiveness assesses citizens’ views regarding the competencies and resources that child protection authorities possess to fulfill their obligations. The utility of these performance indicators lies in their capacity to influence citizens’ perceptions of the trustworthiness of a range of public services (Van Ryzin, Reference Van Ryzin2011).
Trust is the psychological state driven by positive expectations about others’ intentions and behaviors (Rousseau et al., Reference Rousseau, Sitkin, Burt and Camerer1998). Institutional trust involves citizens believing that institutions will consistently achieve desired outcomes without constant scrutiny (Easton, Reference Easton1975; Kramer & Lewicki, Reference Kramer and Lewicki2010) and reflects their confidence in state institutions working in the best interests of society and the population (Thomas, Reference Thomas1998). Trust in CPS denotes citizens’ expectations that the system will persist in securing, protecting, and promoting children’s rights without ongoing oversight.
The performance perspective suggests institutions gain trust by meeting citizens’ performance expectations (Bouckaert et al., Reference Bouckaert, Walle, Maddens and Kampen2002; Mishler & Rose, Reference Mishler and Rose2001; Van de Walle & Bouckaert, Reference Van de Walle and Bouckaert2003). There are two performance perspectives on institutional trust. The macro perspective assesses regime performance across various areas, including economic growth, inflation control, employment, anti-corruption efforts, and the rule of law. In contrast, the micro perspective emphasizes citizens’ interactions with public services as the primary drivers of institutional trust. This perspective posits that improving public service quality enhances satisfaction with the public sector, leading to increased trust in the government (Kampen et al., Reference Kampen, Van De and Bouckaert2006). This concept is recurrent in literature related to public services, and in discussions on the criminal justice system.
Hypothesis 1 ( H1 ) is based on the premise that trust in CPS is strengthened through their role in ensuring distributive justice within the realm of child protection. Distributive justice pertains to how institutions, including child protection authorities, allocate rewards and punishments as perceived by citizens (McLean, Reference McLean2020; Sunshine & Tyler, Reference Sunshine and Tyler2003; Zmerli & Castillo, Reference Zmerli and Castillo2015). These rewards encompass equitable treatment of individuals, irrespective of their socioeconomic status, race/ethnicity, religion, or country of origin. In the context of CPS, distributive justice could entail, for instance, that decisions of care orders of authorities requiring placing children in foster care against the will of biological parents should be based purely on their developmental needs and the availability of suitable placements rather than on the religious or ethnic backgrounds of the children and their parents.
According to the distributive justice model, the perception of fair allocation of services to diverse individuals, groups, and communities plays a pivotal role in generating institutional trust. Institutions are more likely to be considered trustworthy when authorities administer services without discrimination (Levi et al., Reference Levi, Sacks and Tyler2009; Sunshine & Tyler, Reference Sunshine and Tyler2003). Conversely, consistent discriminatory treatment may breed feelings of insecurity, indicating that institutions serve specific group interests and are therefore untrustworthy. Recent research, unlike earlier studies with mixed findings (Sunshine & Tyler, Reference Sunshine and Tyler2003), consistently supports distributive justice as a predictor of trust (Tankebe, Reference Tankebe2013; Zmerli & Castillo, Reference Zmerli and Castillo2015). Given this, it is reasonable to assume that citizens’ trust in CPS relies on their perception of impartial treatment, with trust being contingent on the absence of bias. Thus, it is synthesized:
H1 Perceived distributive justice will be positive associated with trust in CPS.
Hypothesis 2 ( H2 ) suggests a positive relationship between procedural fairness and trust in CPS. Procedural fairness involves citizens’ assessments of the fairness, impartiality, honesty, and respect demonstrated by authorities in reaching decisions (Tyler, Reference Tyler1990: p. 7). Procedural fairness within CPS may encompass various aspects, such as providing parents with opportunities to voice their concerns, ensuring their voices are heard when authorities issue care orders, and granting them avenues to contest the decisions made by authorities regarding care orders if they disagree with them.
Procedural fairness (Sunshine & Tyler, Reference Sunshine and Tyler2003; Tankebe, Reference Tankebe2008; Tyler, Reference Tyler2003) and micro-performance accounts (Van de Walle & Bouckaert, Reference Van de Walle and Bouckaert2003; Van Ryzin, Reference Van Ryzin2011) propose that public perceptions of institutions are influenced by their operational effectiveness, albeit with some nuanced differences. Trust in institutions from a procedural perspective is established through the adherence to normative standards of justice, which are more objective and rational compared to culturally influenced behavioural expectations (Forero & Gómez, Reference Forero and Gómez2017; Manski, Reference Manski2004).
Both the procedural and micro-performance views are grounded in the rational choice assumption that performance drives institutional trust. Numerous studies consistently show that procedural fairness is a robust predictor of institutional trust across diverse cultures and time periods (Nalla & Nam, Reference Nalla and Nam2021; Sunshine & Tyler, Reference Sunshine and Tyler2003; Tankebe, Reference Tankebe2008; Van de Walle & Migchelbrink, Reference Van de Walle and Migchelbrink2020). These studies emphasise that institutional trust is founded on qualities such as fairness, honesty, impartiality, dignity, the opportunity for individuals to express their concerns, and respect. Considering the similarities in the structures and functions of governing bodies, it is reasonable to assume that the way authorities treat individuals will enhance trust in CPS in the public’s eyes. Therefore, it is expected:
H2 There will be a positive correlation between perceived procedural fairness and trust in CPS.
Hypothesis 3 ( H3 ) aligns with the instrumental account, which underscores that the primary determinant of institutional trust is their effectiveness in fulfilling their mandated tasks. In this model, institutions earn trust by delivering utility functions. For example, the police gain trust by ensuring safety, security, and apprehending criminals (Sunshine & Tyler, Reference Sunshine and Tyler2003), schools gain trust by preparing students for the job market, and hospitals gain trust through their role in promoting well-being, among other examples. Instances of effectiveness within the domain of CPS include promptly addressing cases of child abuse to protect children from harm inflicted by their parents and caregivers, fostering collaboration among diverse social, educational, and health services to offer assistance to vulnerable children, and supervising placement, adoption, and care order activities to ensure wellbeing of the children.
Assessing institutional effectiveness lacks a universal method. However, it is essential to emphasise that,
a governor’s legitimacy is therefore evaluated not only upon the basis of his aptitude at deciding and acting in conformity with a society’s current laws and with its fundamental principles but also upon the basis of his capacity to obtain effective results (Coicaud, Reference Coicaud2004: p. 36).
Within this perspective, institutional capacity, which encompasses the quality of human resources and financial resources, is seen as a potentially influential factor for trust. Within this context, it is proposed that:
H3 Perceived functional effectiveness will show a positive association with trust in CPS.
Several factors influenced the testing of the three hypotheses in England and Norway. Notably, the type of general welfare state regimes, the child welfare system, and familialisation have all contributed to the Norwegian CPS having higher trust than its English counterparts (Berrick et al., Reference Berrick, Dickens, Pösö and Skivenes2016; Berrick et al., Reference Berrick, Skivenes and Roscoe2021; Skivenes & Benbenisthy, Reference Skivenes and Benbenisthy2022). Both countries operate under social welfare and liberal democratic systems (Esping-Andersen, Reference Esping-Andersen1990). Norway is categorised as a social welfare democracy, characterised by generous and universal social benefits, strong government intervention for income equality through the social security system. In contrast, England is considered a liberal country with a social security system that is comparatively weaker, featuring stricter qualifications and providing modest benefits.
Furthermore, England’s system is characterised by a focus on individual and familial responsibility in child rearing, with limited resources available for those in need. In contrast, Norway operates under a child-rights system that prioritises a wide range of health and care services with a strong emphasis on each child’s rights. Moreover, the two countries diverge in terms of familialism and de-familialisation. Familialism involves public policy assuming that households bear the primary responsibility for their members’ welfare, while de-familialisation aims to reduce individuals’ reliance on kinship for welfare (Esping-Andersen, Reference Esping-Andersen1990: p. 51). From this perspective, England is categorised as a familialised country, where childcare is seen as primarily a family responsibility. In contrast, Norway is considered a de-familialised country, where families rely on the state for care and protection.
Existing research underscores the importance of contextual differences that make the Norwegian CPS more trustworthy than its English counterpart. However, due to the limited quantitative research on the relationship between CPS performance and trustworthiness, it is unclear to what extent contextual variables across nations impact this connection. Moreover, it is plausible that citizens’ awareness of their child protection system, influenced by contextual factors, positively affects their evaluations of performance, thus enhancing trustworthiness. Consequently, the positive correlation between performance and trust is likely stronger in Norway than in England. Conversely, if people perceive weaknesses in the systems, they may become critical. Furthermore, the previously presented three hypotheses may perform relatively well in England, especially given the current skepticism towards the Norwegian system (Stang, Reference Stang2018). Based on research in criminal justice and public services (Hinds & Murphy, Reference Hinds and Murphy2007; Van Ryzin, Reference Van Ryzin2011), procedural justice is expected to be the most significant predictor of trust in CPS, followed by functional effectiveness and distributive fairness.
Data and method
Procedures
This study, conducted as part of the Discretion project funded by the European Research Council, explores the potential relationship between public perceptions of CPS performance and trust in CPS in Norway and England using cross-sectional survey data. The survey instrument, developed in standard English, was translated into Norwegian by an expert and two doctoral students, all of whom are native Norwegians conducting research on the trust and legitimacy of CPS. The translated version of the instrument underwent thorough review on multiple occasions. Response Analyse (https://responsanalyse.no/) conducted the survey in England (N = 981) and Norway (N = 1,140) during June and July 2022. Survey participants, aged 18 and above, were diverse in terms of demographic characteristics, including gender, age, education level, income, religious affiliation, and left-right orientations. If a specific demographic group was underrepresented, Response Analyse conducted additional surveys to ensure adequate representation. This research encountered no ethical concerns as the data provided by Response Analyse remained anonymous and could not be traced back, either directly or indirectly, to specific individuals or groups of individuals.Footnote 1
Measurement of variables
Respondents were first asked to indicate their degree of agreement with each of the seven statements assessing both the performance of and trust in CPS, using a 5-point scale: 1 strongly agree, 2 agree, 3 neither agree nor disagree, 4 disagree, and 5 strongly disagree. The literature has a longstanding tradition of assessing trust through either single item (Craen, Reference Craen2013; MacDonald & Stokes, Reference MacDonald and Stokes2006) or multiple items, each approach carrying its own advantages and disadvantages (Bradford et al., Reference Bradford, Jackson and Stanko2009; Jackson & Bradford, Reference Jackson and Bradford2010). Employing a multiple-item approach could potentially introduce complexity and ambiguity in differentiating predictors associated with performance evaluation derived from past assessments and those associated with trust concerning future expectations (Craen, Reference Craen2013). Hence, the dependent variable, trust in CPS across the two countries, was measured through the following statements:
-
The English child protection authorities are trustworthy, and
-
The Norwegian child protection authorities are trustworthy.
The scale was reverse-coded, with higher scores indicating increased trust in CPS. This approach aligns with previous literature on trust in CPS (Hsieh & Boateng, Reference Hsieh and Boateng2015; Juhasz & Skivenes, Reference Juhasz and Skivenes2017; Skivenes and Benbenisthy, Reference Skivenes and Benbenisthy2022).
Six items from the criminal justice literature were adapted to measure three facets of CPS performance: distributive justice (Reisig et al., Reference Reisig, Flippin, Meško and Trinkner2021; Tyler & Wakslak, Reference Tyler and Wakslak2004), procedural fairness (Jackson & Bradford, Reference Jackson and Bradford2019; Sunshine & Tyler, Reference Sunshine and Tyler2003; Wolfe et al., Reference Wolfe, Nix, Kaminski and Rojek2016), and functional effectiveness. Literature provides multiple indicators for assessing distributive justice, including but not limited to considerations of racial/ethnic and economic prejudices (Chory-Assad & Paulsel, Reference Chory-Assad and Paulsel2004; Hassan, Reference Hassan2021). Research on policing indicates that just over 60 per cent of individuals in Norway and Britain perceive equal treatment of people irrespective of their wealth. In contrast, approximately 50 per cent of Norwegians and about 30 per cent of Britons believe that individuals from diverse racial backgrounds are subject to inferior treatment (Hassan, Reference Hassan2021: pp. 153–154). This implies that race and ethnicity are more salient indicators of distributive justice than economic prejudices. Therefore, distributive justice was measured by asking respondents to rate their agreement on a scale from 1 (strongly agree) to 5 (strongly disagree) with the statement, ‘The [country] child protection authorities discriminate against some families because of their race and/or ethnicity.’ Higher scores indicate a more positive assessment of distributive justice by child protection authorities.
Procedural justice was assessed by having respondents express their level of agreement on a scale from 1 (1 strongly agree) to 5 (strongly disagree) with the following statements:
-
The [country] child protection authorities allow families to express their views and feelings during the decision-making process.
-
The [country] child protection authorities treat families with dignity and respect.
-
The [country] child protection authorities are fair when making decisions.
As these three items demonstrated high reliability (Cronbach’s alpha: England 0.88; Norway 0.85), they were reverse-coded and averaged to create a procedural fairness index, with higher scores indicating a more favourable perception of procedural fairness concerning child protection authorities among citizens.
Effectiveness was gauged by assessing citizens’ perceptions of institutions’ performance in fulfilling their mandated responsibilities. For example, in the case of the police, this responsibility involves protecting citizens from crimes and responding to their needs (Nalla & Nam, Reference Nalla and Nam2021; Sunshine & Tyler, Reference Sunshine and Tyler2003). Given the complexity of evaluating child protection authorities’ mandated tasks, a resource-based approach to enhancing their effectiveness has been advocated by child rights scholars (Skivenes, Reference Skivenes2021). Thus, respondents were asked to rate their agreement on a scale from 1 (strongly agree) to 5 (strongly disagree) with the following statements:
-
The [country] child protection authorities possess the necessary knowledge and skills to protect children who need help.
-
The [country] child protection authorities possess enough financial resources to protect children who need help.
These two items assess the citizens’ perceptions of the competences and resourcefulness of child protection authorities in assisting vulnerable children. Since these items could reliably gauge the construct of functional effectiveness only in England (Cronbach’s alpha: England 0.61; Norway 0.34), they were consequently treated as distinct independent variables. These items were reverse-coded, with higher scores denoting a more favourable evaluation of the competences and resourcefulness of child protection authorities in both countries.
To address potential confounding factors (e.g., Juhasz & Skivenes, Reference Juhasz and Skivenes2017; Reisig et al., Reference Reisig, Flippin, Meško and Trinkner2021; Skivenes and Benbenisthy, Reference Skivenes and Benbenisthy2022; Van Craen & Skogan, Reference Van Craen and Skogan2015), six socioeconomic variables were included in the regression analysis. Gender was coded as ‘male’ (1), and ‘female’ (2), while age was treated as a continuous variable. Education was ordinal, ranging from ‘secondary school’ (1) to ‘postgraduate’ (4). Personal income, categorised into eight bands in England and six in Norway, was treated as continuous. Religious denominations were coded as Christians (1), atheists (2), or ‘other faiths’. Left-right orientations were continuous, from ‘left’ (1) to ‘right’ (10). Detailed descriptions of the coding scheme are provided in the online supplementary material, while Table 1 presents their descriptive statistics.
Descriptive statistics
Analytical strategy
The data analysis consisted of five steps. Initially, a descriptive analysis was conducted to provide an overview of the data. Following this, a one-way multivariate analysis of variance (MANOVA) was employed to identify systematic variations in the measures of trust in CPS and performance measures. Subsequently, performance measures, age, income, and left-right orientations were transformed into their respective z-scores to facilitate comparisons between these variables. Afterward, both bivariate and multivariate ordinary least squares (OLS) analyses were conducted using pooled data and separately within each country to examine the correlation between distributive justice, procedural fairness, functional effectiveness, and trust in CPS.
Results
Descriptive analysis
Table 1 presents descriptive statistics. Norwegian respondents display higher trust in CPS (mean 3.03) than English counterparts (mean 2.89). Norwegian participants rate their child protection authorities (mean 2.82) higher than English authorities (mean 2.91) in terms of distributive justice. Norwegian respondents perceive higher procedural fairness (mean 2.91) in their authorities compared to English respondents (mean 2.88). In terms of functional effectiveness, Norwegian respondents regard their authorities as more competent (mean 2.90) and resourceful (mean 3.42) compared to their English counterparts (means 2.81, and 3.23, respectively). The MANOVA results indicated statistically significant differences between trust in and performance of CPS in Norway and England (Wilks’ λ 0.985; F < 0.001; p < 0.001 for all comparisons), justifying further comparative analysis of these child protection systems.
Figure 1 presents estimates of the bivariate association between performance measures and trust in CPS (see Appendix A), which are highly significant in both pooled analysis and individual analyses for England and Norway (p < 0.001). Distributed justice demonstrates a significant association with trust in CPS in pooled analysis (B = 0.34). However, this association is more than four times stronger in Norway (B = 0.37) compared to England (B = 0.08, p < 0.001). Similarly, the relationship between procedural justice and trust in CPS is slightly more pronounced in Norway (B = 0.70) and slightly less pronounced in England (B = 0.64) compared to the pooled analysis (B = 0.67). In terms of functional effectiveness indicators, the correlation between competences and trust in CPS is slightly stronger in Norway (B = 0.63) than in the pooled analysis (B = 0.60) and England (B = 0.56). Additionally, while resources and trust in CPS are significantly correlated in the pooled analysis (B = 0.24), this relationship is over twice as strong in Norway (B = 0.33) compared to England (B = 0.16). These bivariate estimates strongly confirm the hypotheses that distributive justice (H1), procedural fairness (H2), and functional effectiveness are positively linked with trust in CPS (H3).
Bivariate association between performance and trust in CPS.
Note: The horizontal bars show the 95% confidence intervals. **p ≤ .01; ***p ≤ .001. Detailed estimates are in Appendix A.
Effects of performance on trust in CPS
Figure 2 displays coefficients from three of the six OLS models (see Appendix B). The first three models (M1-3), not detailed here, introduced performance-related variables stepwise, while the fourth model (M4: pooled model) included control variables. These four models incorporated England as a dummy variable to assess its significance in both England and Norway, demonstrating significant associations between performance measures and trust in CPS.
Effects of performance on trust in CPS.
Note: The horizontal bars show the 95% confidence intervals. †p ≤ .10; *p ≤ .05; **p ≤ .01; ***p ≤ .001. Detailed estimates are in Appendix B.
Significant association patterns between performance and trust in CPS are evident in pooled analysis. Distributive justice exhibits a significant yet very weak positive association with trust in CPS (B = 0.03, p < 0.10), while procedural justice emerges as the most consistent predictor of trust in CPS (B = 0.48, p < 0.001). Functional effectiveness, particularly its competences aspect, demonstrates a significant positive association with trust in CPS (B = 0.24, p < 0.001), contrasting with the resources dimension, which shows a notably weak association (B = 0.03, p < 0.10). Lastly, Norway’s significant association with trust in CPS (B = 0.11, p < 0.001) enables the model’s replication in both nations. These findings strongly support the hypotheses, confirming distributive justice, procedural fairness, and functional effectiveness as substantial predictors of trust in CPS (H1-3).
Following this, the pooled model was separately examined in England and Norway. It is noteworthy that the weak but significant positive relationship between distributive justice and trust in CPS in the pooled model disappears within both countries. Thus, empirical support for H1 is not found in England and Norway. However, the results indicate a strong, positive, and statistically significant association between procedural fairness and trust in CPS in both countries, with Norway demonstrating stronger performance (B = 0.49, p < 0.001) compared to England (B = 0.46, p < 0.001). These findings provide robust support for H2, which suggests a significant association between procedural fairness and trust in CPS in both countries.
Furthermore, concerning the two indicators of functional effectiveness, the impact of competences on trust in CPS is highly significant and positive, though slightly stronger in Norway (B = 0.25, p < 0.001) than in England (B = 0.22, p < 0.001). However, the resources indicator of functional effectiveness shows a weak but statistically significant correlation with trust in CPS, observed only in Norway (B = 0.05, p < 0.05). Hence, the competences indicator of functional effectiveness provides substantial support in both countries for H3, which hypothesises a significant association between functional effectiveness and trust in CPS. Nevertheless, the resources indicator can only confirm H3 in Norway. Combined, these three models account for 58 per cent of the total variance in trust in CPS in the pooled analysis and across the two countries.
The analysis considered the influences of gender, age, education, income, religious affiliations, and left-right political orientations (see Appendix C). In Norway, age shows a negative association with trust in CPS (B = −0.04, p < 0.10), while income is negatively associated with trust in the pooled analysis (B = −0.02, p < 0.10). Moreover, higher education demonstrates a positive association with trust in CPS in the pooled analysis (B = 0.07, p < 0.10). Lastly, left-right political orientations are positively correlated with trust in the CPS in both the pooled model (B = 0.03, p < 0.10) and Norway (B = 0.05, p < 0.001), suggesting that trust tends to increase as respondents lean towards the right on the left-right scale. The remaining variables did not show significant associations with trust in the CPS.
Conclusion
Numerous cross-disciplinary studies highlight the significance of institutional trust for democracy and the quality of governance. Although research on trust in CPS is still emerging, empirical evidence indicates that even in Europe’s well-established welfare nations, individuals harbor doubts about their authorities’ performance (Skivenes & Benbenisthy, Reference Skivenes and Benbenisthy2022; Stang, Reference Stang2018). Nonetheless, the impact of this performance on public trust in CPS remains relatively unexplored. This study aimed to fill the research gap by proposing that three performance factors – distributive justice, procedural fairness, and functional effectiveness – would positively influence trust in CPS (H1-3). Distributive justice pertains to equitable treatment, procedural fairness involves fairness in decision-making, and functional effectiveness relates to agency competences.
The study provides valuable insights, drawing from cross-sectional survey data gathered from respondents in England and Norway in 2022. Notably, it underscores the greater prominence of trust in the Norwegian CPS in comparison to its English counterpart. This observation aligns with previous research emphasising the influence of institutional designs on trust in CPS (Juhasz & Skivenes, Reference Juhasz and Skivenes2017; Skivenes & Benbenisthy, Reference Skivenes and Benbenisthy2022). Second, in line with H2, procedural fairness significantly impacted trust in CPS in pooled analysis and in both countries. H1, however, received exclusive support in pooled model. As for H3, the competences indicator of functional effectiveness received unequivocal support in both countries, whereas its resources indicator found validation solely in Norway. These findings validate the study’s hypotheses, suggesting that child protection agencies are more likely to garner trust when people perceive them as treating everyone fairly, operating fairly, and performing their functions effectively. From a theoretical perspective, these results are significant as they contribute new evidence to the micro performance theory, previously well-supported in organisational and criminal justice literature, now applied to the context of CPS.
These findings hold policy relevance. First, child protection authorities deal with individuals from diverse backgrounds, and certain factors, such as low income and low education, have been associated with skepticism toward the effectiveness of child protection systems (Skivenes & Benbenisthy, Reference Skivenes and Benbenisthy2022). In these interactions, treating individuals fairly and impartially is a vital means of gaining public support, regardless of demographic and institutional differences among societies. Additionally, child protection systems can enhance their trustworthiness by appointing staff based on their knowledge and skills and securing sufficient financial resources to fulfill their service obligations. Another potential insight, consistent with the policing literature (Graham et al., Reference Graham, Kulig and Cullen2020; Murphy & Barkworth, Reference Murphy and Barkworth2014), suggests that citizens are more likely to support institutions by complying with the directives of child protection authorities and reporting child abuses when they trust these authorities.
Secondly, distributive justice did not demonstrate a significant association with trust in CPS in either country, while the resources indicator of functional effectiveness was unable to predict trust in the CPS in England. An explanation for the lack of significant impact of distributive justice, as suggested in prior research (Reisig et al., Reference Reisig, Flippin, Meško and Trinkner2021; Zmerli & Castillo, Reference Zmerli and Castillo2015), could be that distributive justice interacts with income levels and neighbourhood contexts, thus affecting trust. Consequently, distributive justice is more likely to mediate the association between procedural fairness and trustworthiness (McLean, Reference McLean2020). This intricate interplay among the three performance measures may systematically mitigate the effects of distributive justice on trust in CPS. Whether distributive injustice matter is subject to debate, and CPS’s ability to counter negative perceptions may be limited. However, improved communication about poverty-alleviation efforts reducing CPS’s societal impact could help. Perhaps if CPS were to better communicate to the public that efforts to alleviate poverty would likely reduce their fingerprint on society. Nevertheless, the reasons for the lack of significance in the association between the resources indicator of functional effectiveness and trust in CPS have yet to be identified.
Third, procedural fairness is the most influential predictor of trust in CPS. It means that when child protection authorities ensure that their processes are perceived as fair and just, this has a greater influence on fostering trust among the public than their overall effectiveness or the equitable distribution of their services. This finding aligns with previous research in the field of criminal justice (Sunshine & Tyler, Reference Sunshine and Tyler2003), which has also shown that the perception of fair and just procedures is a powerful factor in building trust. In practical terms, it suggests that child protection agencies can enhance public trust by emphasising not only their effectiveness but also the fairness and justice of their actions and decision-making processes.
Finally, aside from individuals’ left-right orientations, standard demographic factors like gender, age, education, income, and religious affiliation had no significant impact on trust in CPS. This aligns with previous research on trust in the police (Murphy et al., Reference Murphy, Hinds and Fleming2008). However, these findings deviate from two influential studies on trust in CPS (Juhasz & Skivenes, Reference Juhasz and Skivenes2017; Skivenes & Benbenisthy, Reference Skivenes and Benbenisthy2022). One potential explanation for this inconsistency is the methodological difference: previous studies used correlation analysis, while the current study employed multivariate analysis.
This study has several important limitations. One limitation of this study is that the causal direction of the relationships cannot be determined due to the use of cross-sectional data analysis. Furthermore, the data represent trust and performance perceptions of citizens in both England and Norway, both highly developed societies. The significance of one’s personal experiences in evaluating child protection services (Petersen, Reference Petersen2018) raises questions about the generalisability of this study’s findings to contexts where respondents possess diverse experiences with CPS, including both positive and negative encounters. Nor do these results claim to be true for all post-communist and post-colonial societies as a whole. Future research could replicate the current study to further investigate the generalisability of the micro- performance theory across contexts and cultures. Finally, considering the multidimensionality of justice, fairness, effectiveness, and trustworthiness (Beugre & Baron, Reference Beugre and Baron2001; Bolger & Walters, Reference Bolger and Walters2019; Sunshine & Tyler, Reference Sunshine and Tyler2003; Tankebe, Reference Tankebe2013), future research could utilize structural equation modeling to further investigate the generalisability of these findings (e.g., Jackson et al., Reference Jackson, Bradford, Hough, Myhill, Quinton and Tyler2012; Jackson & Bradford, Reference Jackson and Bradford2019).
In conclusion, the study recommends that child protection authorities in England and Norway prioritise improving procedural fairness and functional effectiveness to enhance trust among citizens. Additionally, Norwegian authorities should pay more attention to distributive justice to gain public trust. However, it is important to acknowledge the challenges involved in changing long-standing organisational practices and addressing budgetary and logistical concerns, which can be substantial obstacles for policymakers seeking to improve trust in CPS.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S0047279424000114
Acknowledgments
Special appreciation due to Prof. Marit Skivenes, the principal investigator of the Discretion project and director of the Centre for Discretion and Paternalism (DIPA) at the University of Bergen, for her invaluable contribution to developing the survey instrument and providing comprehensive feedback on earlier versions of this paper. I extend my gratitude to Barbara Ruiken and Mathea Loen, PhD students at DIPA, for their assistance with translation. Additionally, I would like to acknowledge the anonymous reviewers and the audience at the 2022 Conference of the Netherlands Institute of Governance at Tilburg University, Netherlands, for their valuable feedback.
Funding
This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 724460) and from the Research Council of Norway under the Research Programme on Welfare, Working Life and Migration (VAM II) (grant no. 302042). The results of this research reflect only the author’s view and the funding Agency is not responsible for any use that may be made of the information it contains.
Competing interests
The author(s) declare none.
Disclaimer
Publications from the project reflect only the authors’ views and the funding agency is not responsible for any use that may be made of the information contained therein. The author declares no conflict of interest.
Appendices
Appendix A. Bivariate OLS estimates of trust in CPS
Note: Coefficients (B) are followed by standard errors (SE) in parenthesis. **p ≤ .01; ***p ≤ .001.
Appendix B. Multivariate OLS estimates of trust in CPS
Note: Coefficients (B) are followed by standard errors (SE) in parenthesis. †p ≤ .10; *p ≤ .05; **p ≤ .01; ***p ≤ .001.
Open access
