I want to study social work social servicesCommunity and Society
How each model performed on this task. Every chart states its N, and the reward carries a 95% confidence interval once there is more than one run to measure.
Core metrics
Check Pass %
Mean share of the task's checks passing, averaged over runs. N = 1 run per model.
95%
Claude Opus 5.5
20%
GPT-5.6 sol
5%
GPT-6 sol
0%
Claude Opus 5
Extended metrics
Median Time
Wall-clock minutes, median across runs. N = 1 run per model.
60m
Claude Opus 5
9m
GPT-5.6 sol
Tool Calls
Tool invocations, median across runs — counted from the trajectories. N = 1 run per model.
113
Claude Opus 5
31
GPT-6 sol
14
GPT-5.6 sol
11
Claude Opus 5.5
Verifier × model pass rate
How often each verifier passed, per model. Green is passing, red is failing — a red band across a row is a verifier nothing clears.
| Verifier | Claude Opus 5.5n=1 | GPT-6 soln=1 | GPT-5.6 soln=1 | Claude Opus 5n=1 |
|---|---|---|---|---|
| checks (20) | ||||
| 100% | 0% | 0% | 0% | |
How to passBegin by establishing precise definitions of social work, social services, social welfare, social protection, community development, humanitarian assistance, charity and social policy. Explain the different levels at which social work operates, from direct work with individuals and families to organisational, community and policy-level intervention. Distinguish professional social workers from other helping professions and explain how multidisciplinary teams operate in healthcare, education, justice, housing and community settings. Establish the idea that social problems are rarely caused by a single factor and that effective intervention often requires cooperation between multiple institutions. | ||||
| 100% | 0% | 0% | 0% | |
How to passExamine how societies historically responded to poverty, disability, illness, homelessness, child abandonment and other forms of social vulnerability before professional social work existed. Then trace the emergence of organised charity, settlement movements, public assistance, social reform and professional social work. Explain how industrialisation, urbanisation, labour movements, major wars, economic crises and human-rights movements transformed ideas about poverty and social responsibility. Compare the development of welfare systems in different parts of the world and show how historical circumstances continue to influence contemporary social-service institutions. | ||||
| 100% | 0% | 0% | 0% | |
How to passExamine major theoretical frameworks used to understand people and their environments, including systems theory, ecological perspectives, person-in-environment approaches, strengths-based practice, empowerment theory, trauma-informed approaches, crisis intervention and anti-oppressive practice. Explain the assumptions behind each framework and demonstrate how the same client situation could be interpreted differently depending on the theoretical lens used. Critically assess the strengths and limitations of these approaches and distinguish evidence-supported practice principles from theories that remain more conceptual or context dependent. | ||||
| 100% | 0% | 0% | 0% | |
How to passExplore how individuals develop within families, communities, institutions and wider societies. Examine childhood development, adolescence, adulthood and ageing while considering psychological, social, cultural and economic influences. Explain how family relationships, education, employment, poverty, discrimination, neighbourhood conditions, trauma and social networks can influence behaviour and wellbeing. Show why social workers must consider both individual characteristics and environmental conditions rather than automatically interpreting social difficulties as personal failures. | ||||
| 100% | 0% | 100% | 0% | |
How to passInvestigate poverty as a multidimensional phenomenon involving income, housing, education, health, employment, food security, social participation and access to opportunities. Explore absolute and relative poverty, intergenerational poverty, inequality, unemployment, informal work and structural exclusion. Examine how social-service systems attempt to address these problems through benefits, employment programmes, housing assistance, food support, education and community intervention. Critically evaluate the limits of individual casework when underlying economic and political structures continue reproducing disadvantage. | ||||
| 100% | 0% | 0% | 0% | |
How to passDevelop a comprehensive examination of services supporting children and families. Cover child safeguarding, abuse and neglect, family preservation, foster care, adoption, kinship care, early intervention and family-support programmes. Examine the difficult professional decisions surrounding removal of children from unsafe environments, balancing child protection with family autonomy and determining the appropriate level of state intervention. Compare preventive approaches with crisis-driven systems and investigate how poverty, housing insecurity, parental mental health, substance use and social isolation can complicate child-welfare cases. | ||||
| 100% | 0% | 0% | 0% | |
How to passInvestigate the role of social workers within hospitals, primary healthcare, mental-health services, rehabilitation, palliative care and community health programmes. Explain psychosocial assessment, discharge planning, care coordination, family support, access to benefits and referrals to community services. Then examine mental-health social work, including crisis intervention, community-based support, recovery-oriented approaches and substance-use services. Explore the relationship between clinical treatment and social conditions such as housing, employment, family support and social isolation. | ||||
| 100% | 0% | 0% | 0% | |
How to passExamine disability through medical, social and rights-based perspectives and explain why the way disability is conceptualised affects service design. Cover accessibility, independent living, employment, education, assistive services, family caregiving and protection against exploitation. Then investigate the rapidly changing needs of ageing populations, including dementia, loneliness, long-term care, retirement, social isolation and changing family structures. Compare institutional care with community-based and person-centred approaches and evaluate the challenges governments face as populations age. | ||||
| 100% | 0% | 100% | 0% | |
How to passExplore homelessness as a complex social phenomenon influenced by housing costs, unemployment, mental health, family breakdown, disability, substance use, domestic violence and structural economic conditions. Compare emergency shelters, transitional housing, permanent supportive housing, prevention programmes and housing-first approaches. Examine how social workers coordinate with housing agencies, healthcare providers, employment services and local governments. Evaluate why treating homelessness solely as an individual behavioural problem often fails to address the structural conditions that produce housing insecurity. | ||||
| 100% | 0% | 0% | 0% | |
How to passInvestigate the social-service needs of migrants, refugees, asylum seekers and internally displaced populations. Cover access to housing, healthcare, education, employment, language services, legal assistance and community integration. Examine the roles of governments, NGOs, humanitarian organisations and social workers in supporting displaced people before, during and after resettlement. Compare short-term humanitarian assistance with long-term integration strategies and discuss the ethical challenges involved when legal status, cultural differences, trauma and limited resources intersect. | ||||
| 100% | 0% | 0% | 0% | |
How to passExamine services for people affected by domestic violence, human trafficking, substance use, incarceration, homelessness, severe poverty and other forms of vulnerability. Explore rehabilitation, reintegration, victim support, harm-reduction approaches, community supervision and employment assistance. Investigate why people can repeatedly cycle between social-service systems, hospitals, shelters and criminal-justice institutions and examine whether coordinated intervention can reduce this fragmentation. | ||||
| 100% | 0% | 0% | 0% | |
How to passConduct a comparative examination of how countries organise social protection. Contrast universal programmes with means-tested assistance, social insurance with tax-funded benefits, and highly centralised systems with decentralised community services. Examine pensions, unemployment support, healthcare, child benefits, disability assistance, housing programmes and cash transfers across different regions. Analyse how political ideology, economic development, demographics, taxation, labour markets and cultural expectations influence welfare-system design and why a model that works in one country may not transfer directly to another. | ||||
| 100% | 0% | 100% | 0% | |
How to passMove from individual casework to community-level transformation. Explore community organising, participatory development, empowerment, social capital, local leadership, asset-based community development and collective action. Explain how social workers can identify community strengths rather than focusing exclusively on deficits and how residents can participate in defining problems and designing solutions. Examine successful community interventions while critically considering issues of representation, unequal power, participation fatigue and the danger of professionals claiming to speak on behalf of communities without genuine participation. | ||||
| 100% | 0% | 0% | 0% | |
How to passInvestigate the role of non-governmental organisations, charities, foundations, community groups and international organisations in delivering social services. Examine how these organisations obtain funding, design programmes, coordinate with governments and measure outcomes. Compare situations where NGOs effectively fill service gaps with situations where fragmented NGO systems create duplication, inconsistent standards or weak accountability. Include the roles of major international actors in humanitarian response, poverty reduction, child protection, health, refugee support and international development. | ||||
| 100% | 100% | 0% | 0% | |
How to passAnalyse social work through a human-rights framework and examine how discrimination can influence access to housing, education, healthcare, employment, justice and social protection. Explore issues involving gender, disability, ethnicity, race, migration status, poverty and other forms of social exclusion without reducing these complex issues to individual prejudice alone. Examine how social workers balance respect for individual autonomy with professional responsibilities to protect vulnerable people and challenge unjust institutional practices. | ||||
| 100% | 0% | 0% | 0% | |
How to passDevelop a rigorous examination of ethical practice in social work. Cover confidentiality, informed consent, professional boundaries, safeguarding, mandatory reporting where applicable, conflicts of interest, dual relationships, cultural competence and documentation. Explore situations where ethical principles conflict, such as protecting a vulnerable person while respecting their autonomy or maintaining confidentiality while addressing serious risk. Examine the power imbalance between social workers and service users and explain how ethical practice requires awareness of institutional authority as well as personal professional conduct. | ||||
| 0% | 0% | 0% | 0% | |
How to passInvestigate how social-service programmes are financed through government budgets, taxation, grants, philanthropy, international development funding and other mechanisms. Examine how funding structures influence programme priorities, staffing, eligibility rules and service continuity. Explore the economics of prevention and early intervention and investigate whether social programmes can reduce future costs in healthcare, criminal justice, homelessness and child protection. Distinguish genuine cost-effectiveness evidence from politically motivated claims that a programme simply saves money. | ||||
| 100% | 0% | 100% | 0% | |
How to passDevelop a rigorous framework for evaluating social interventions. Explain programme evaluation, outcome measurement, qualitative research, quantitative research, longitudinal studies, quasi-experimental approaches, cost-effectiveness analysis and community-level indicators. Examine the difference between correlation and causation and explain why an improvement after a programme does not automatically prove that the programme caused it. Investigate how researchers account for selection effects, incomplete participation, changing circumstances and unintended consequences when evaluating social-service interventions. | ||||
| 100% | 0% | 0% | 0% | |
How to passAnalyse contemporary pressures facing social workers and service systems, including excessive caseloads, staff shortages, burnout, secondary trauma, inadequate funding, administrative burdens, fragmented services, professional safety and growing complexity among service users. Then investigate emerging developments such as artificial intelligence, digital case management, tele-social work, predictive analytics, remote service delivery and integrated data systems. Examine the potential benefits while giving equal attention to privacy, algorithmic bias, surveillance, digital exclusion and the danger of replacing human judgement with automated systems. | ||||
| 100% | 0% | 0% | 0% | |
How to passBring the entire subject together by examining how social interventions operate across individual, family, community, institutional, national and international levels. Construct several hypothetical case studies in which a complex social problem requires simultaneous intervention in areas such as housing, healthcare, education, employment, mental health, family support and public policy. Trace how a social worker or multidisciplinary organisation would respond, identify the relevant government and community institutions, determine appropriate interventions and establish measurable outcomes. Finally, evaluate what makes a social-service system genuinely transformative rather than merely reactive, considering prevention, accessibility, dignity, equity, community participation, evidence, sustainability, accountability and structural change. Conclude by examining the central question of modern social work: whether social services should primarily help individuals adapt to difficult circumstances, or whether the profession must also participate in changing the social, economic and institutional conditions that create those difficulties in the first place. | ||||
Failure modes on this task
56 tagged verifier failures, by the stage they were assigned to. Higher is worse.
| Failure stage | Claude Opus 5.5 | GPT-6 sol | GPT-5.6 sol | Claude Opus 5 |
|---|---|---|---|---|
| Delivering the result1 mode | 1 | 5 | 5 | 5 |
| Standing up the evidence1 mode | — | 3 | 3 | 1 |
| No signal in the judge’s words1 mode · the grader | — | 11 | 8 | 14 |