Human and Machine: A Journey - Part 10

Evidence and Practice: A Complex Relationship

Throughout our journey exploring modern welfare systems, we've used the metaphor of an hourglass to understand how knowledge flows between research, policy, and practice. Looking down from the top of this hourglass, we find two fundamentally different ways of synthesizing knowledge from practice - ways that perfectly illustrate our ongoing exploration of Human and Machine approaches to welfare work.

Two Views from the Top

The first view, representing the Machine, is the meta-analysis: aggregative, formalized, and focused on measuring intervention effects. This approach has dominated evidence-based practice, using statistical methods to combine results from multiple quantitative studies. Like a machine processing data, meta-analysis follows strict protocols to produce standardized knowledge about "what works."

The second view, emerging more recently and representing the Human, is the meta-synthesis: configurative, associative, and focused on understanding meaning and context. This approach seeks to synthesize qualitative insights about how interventions work in different contexts, what they mean to different stakeholders, and how they transform through implementation. Rather than simply aggregating data, meta-synthesis configures insights into new understanding.

These contrasting approaches reflect fundamentally different ways of knowing:

• The Machine approach through meta-analysis:

  • Aggregates quantitative data

  • Focuses on measurable effects

  • Follows standardized procedures

  • Seeks statistical significance

  • Aims for generalizable knowledge

  • Informs practice guidelines and policy

• The Human approach through meta-synthesis:

  • Configures qualitative insights

  • Focuses on meaning and context

  • Follows interpretive processes

  • Seeks deeper understanding

  • Aims for transferable wisdom

  • Guides practice adaptation

This distinction becomes particularly crucial in complex welfare services, where success often depends as much on how interventions are implemented as on the interventions themselves. Meta-analysis follows a deductive path, starting with predetermined criteria to test specific hypotheses about what works. Meta-synthesis works inductively, allowing patterns and understanding to emerge from practice experiences. While meta-analysis can tell us if an intervention generally works, meta-synthesis helps us understand why it works and how to adapt it effectively in different contexts.

Multiple Knowledge Realities

Care practices require sophisticated integration of multiple forms of knowledge, each operating with its own logic and reality. Drawing from Mol's¹ analysis of how different versions of disease exist simultaneously in medical practice, we can understand how systematic evidence, professional wisdom, and lived experience must be woven together in welfare work.

Care practices involve at least three distinct but interrelated realities:

The Research Reality

• Centers on systematic knowledge production

• Built through controlled studies and meta-analyses

• Uses standardized measurements

• Provides crucial foundations for practice

• Can miss nuanced complexity of individual situations

The Practice Reality

• Emerges through professional experience

• Develops pattern recognition abilities

• Enables contextual understanding

• Allows adaptation of principles to specific situations

• Notices what standardized assessments might miss

The Client Reality

• Encompasses lived experience

• Shaped by cultural backgrounds

• Influenced by personal aspirations

• Embedded in family relationships

• Constrained by practical limitations

These realities aren't simply different perspectives on a uniform underlying truth. Each produces its own version of what is real and relevant, operating with its own logic and priorities. Good practice requires moving skillfully between them.

Knowledge Translation in Practice

This integration of different knowledge types plays out daily in welfare practice. Consider how a skilled social worker navigates multiple forms of knowledge when assessing complex family situations:

• Research knowledge helps identify critical risk factors

• Professional judgment enables recognition of subtle patterns

• Family expertise provides essential context and history

• Practice wisdom guides intervention adaptation

• Systematic frameworks ensure consistent assessment

• Relational skills enable meaningful engagement

These knowledge types transform each other in practice, but they also shape how professionals understand and approach their work. Consider child welfare services, where different knowledge traditions have profoundly influenced how stability for placed children is understood and pursued:

Research findings about attachment and child development have shaped assessment frameworks and timelines, leading to increasing emphasis on early permanency decisions. Professional experience reveals the complex realities of working with both biological and foster families, while family expertise highlights the ongoing importance of relationships and cultural connections. These different perspectives create tensions in practice - between needs-based and relationship-oriented approaches, between protecting children and preserving family bonds.

A common challenge emerges when documentation systems and organizational routines emphasize placement stability over family reunification work. Despite policies prioritizing reunification, practice often drifts toward focusing on maintaining placement stability - not because professionals lack understanding, but because tools and frameworks shape attention and effort in particular directions. The very instruments we use to ensure quality can sometimes work against broader practice wisdom about supporting sustainable change in children's home environments.

This illustrates how knowledge isn't just applied in practice - it actively shapes how professionals understand their role and what they perceive as possible or important. While professionals maintain their judgment, they're also inevitably shaped by the frameworks, tools and systems within which they work.

Building Local Evidence

Modern welfare organizations face a paradox: while they collect vast amounts of data through documentation systems, they often struggle to transform this information into actionable knowledge for practice development. Consider how child welfare services document countless assessments, interventions, and outcomes - yet this rich practice experience rarely translates into systematic learning about what approaches work best in different contexts.

The challenge isn't just technical - it's about bridging different forms of knowledge and understanding. Local practices develop their own wisdom about what works, shaped by:

• Direct experience with families and interventions

• Professional interpretations of research evidence

• Local organizational conditions and resources

• Cultural and community contexts

• Practical constraints and opportunities

Yet this local knowledge often remains tacit and personal rather than becoming part of organizational learning. While documentation systems excel at tracking compliance and individual cases, they rarely capture the nuanced understanding that emerges from practice - the patterns professionals recognize, the adaptations that prove successful, the contextual factors that influence outcomes.

To bridge this gap, we need frameworks that:

• Support systematic reflection on practice experiences

• Connect individual cases to broader patterns

• Document not just what was done but why and how

• Enable sharing of practice wisdom across contexts

• Create dialogue between research and practice knowledge

• Build understanding of local adaptation processes

This isn't just about better documentation - it's about creating spaces where different forms of knowledge can meet and inform each other. Where research evidence can guide practice while practice experience enriches our understanding of how interventions work in real contexts.

Looking Forward: Bridging Machine and Human Approaches

As welfare services grow more complex, the challenge isn't just managing increasing demands - it's about fundamentally rethinking how we configure our systems to support professional practice. The future requires both machine-like precision in certain areas and human wisdom in others. The key lies in understanding where each approach adds the most value.

Consider how this might work in practice:

Documentation systems could shift from primarily tracking compliance to supporting learning - capturing not just what happened but the professional reasoning behind decisions. Artificial intelligence might handle routine documentation, freeing professionals to focus on deeper analysis and relationship building. Assessment frameworks could evolve to combine standardized measures with space for professional judgment about context and meaning.

But this transformation requires careful attention to:

• How we design technology to augment rather than replace professional judgment

• Ways to document complexity without creating overwhelming bureaucracy

• Methods for integrating systematic and experiential knowledge

• Approaches for developing professional capabilities that combine technical and human wisdom

• Spaces for reflection and learning amid practical demands

Success means moving beyond the false choice between human judgment and systematic approaches. It requires thoughtful integration that enhances both - where systematic approaches provide reliable foundations while human wisdom guides adaptation and meaningful engagement.

Before we explore how professional communities enable this integration in Part 11, consider:

• How do you preserve professional judgment while using systematic tools?

• What helps you maintain focus on human connection amid system demands?

• Where do you see opportunities for better integration of different approaches?

• What would help you better combine systematic and experiential knowledge?

The path forward isn't about choosing between Human and Machine approaches - it's about thoughtfully combining them to create welfare systems that truly serve their purpose while maintaining professional wisdom and human dignity.

---

This is part 10 in our ongoing series exploring the intersection of human judgment and systematic knowledge in modern welfare systems. Join the conversation by sharing your thoughts and experiences in the comments below.

1. Mol, A. (2008). The Logic of Care: Health and the problem of patient choice. Routledge.