Child poverty pathfinders in Dundee and Glasgow: phase two evaluation

1. Introduction and methods

This report presents findings from the phase 2 external evaluation of the child poverty pathfinders in Glasgow and Dundee. The evaluation was commissioned by the Scottish Government and conducted by Ipsos, an independent research company.

Policy background

Tackling child poverty

The Scottish Government has put tackling child poverty at the heart of its policy agenda. The Child Poverty (Scotland) Act 2017 introduced new statutory targets to reduce child poverty and the Tackling Child Poverty Delivery Plan^[1] (TCPDP) was developed to help meet these targets. The second Delivery Plan, covering 2022-2026, was released in March 2022^[2] .

The last Scottish Government TCPDP progress report in 2024^[3] acknowledges that “child poverty rates have shown little change in recent years”. It highlights the ongoing impact of the cost-of-living crisis and limitations on the Scottish Government’s ability to mitigate UK Government welfare policies as contributing factors to high child poverty levels. However, while acknowledging the challenging context, it reaffirms the priority attached to tackling child poverty, with the First Minister John Swinney declaring the eradication of child poverty the Scottish Government’s “single most important objective”.

Public service reform

Alongside addressing child poverty, the Scottish Government has a long-standing focus on Public Service Reform (PSR). The report of the Christie Commission in 2011^[4] has been highly influential in shaping thinking on this issue. Christie argued that “a radical change in the design and delivery of public services is necessary, irrespective of the current economic challenges, to tackle the deep-rooted social problems that persist in communities across the country.” In particular, it strongly recommended shifting resources to prioritise preventative approaches, and addressing systemic issues of fragmentation, complexity and lack of accountability that hamper joint working to deliver effective prevention. The 2022-26 TCPDP highlights the importance of PSR to tackling child poverty, arguing that the transformational change needed to tackle child poverty requires working differently to focus on the needs of families and ensure that systems work for the people who need them most.

Introduction to the child poverty pathfinders

The second TCPDP set out an intention to identify “a small number of pathfinder areas to commence work on a new phased approach to whole system change … bringing together partners in a locality to overcome barriers and maximise the use of resources to provide holistic support tailored to the individual needs of families.” The Scottish Government would invest in and work with pathfinder partners to “refine, test, adapt and scale different approaches to provide person-centred solutions”, with the aim of generating evidence to “ensure we are making the most of our investments and assets to have the maximum impact on child poverty.”

Two local pathfinder projects, in Dundee and Glasgow, were initially identified for Scottish Government investment, primarily through a combination of Tackling Child Poverty Delivery Fund and Whole Family Wellbeing Fund support. The two pathfinders differ significantly in scale, core partners and design.

Dundee pathfinder

The Dundee pathfinder is centred around a place-based, multi-agency key worker model, focused on the Linlathen and Mid Craigie area of the city. Other key features are:

• Targeted outreach, involving door-knocking at households identified (using Council Tax Reduction data) as having children under 16 and no earned income from employment
• A weekly drop-in session at the Brooksbank community centre in Linlathen, run by the key worker team and attended by a range of third and public sector partners.

Delivery in Linlathen began in October 2022. Support provided by key workers is flexible and open-ended, and covers a very wide range of issues. The pathfinder is intended to embody ‘no wrong door’ and ‘tell your story only once’ principles in how it engages with people. At the time of writing, the core delivery team comprised five key workers, three employed by Dundee City Council (DCC) Employability Team, one by Social Security Scotland, and one by DWP, and an operational manager (employed by Dundee City Council).

The pathfinder is overseen by a strategic Oversight Board, comprising senior staff from the four core partner organisations contributing resources to the pathfinder (Dundee City Council, DWP, Scottish Government, and Social Security Scotland).

Glasgow pathfinder

The Glasgow pathfinder also began in 2022, and has influencing and embedding “whole system change” as its main focus. A key feature has been the establishment of a multi-agency change team (MACT) of staff with a range of backgrounds and skills to drive forward system change. The embedding of a ‘no wrong door’ model in Glasgow, so that citizens (including families with children) can access person-centred, relational, holistic case management and support regardless of when, where and how they engage with services, is also central to the vision of the Glasgow pathfinder.

The Glasgow pathfinder is much bigger in scale and scope than the Dundee pathfinder. This is reflected in the number of interlinked workstreams it includes, initially six and now nine, each with priority aims and activities. These workstreams are summarised in box 1.1, below (and discussed in more detail in chapter 7). The Glasgow pathfinder, now known as the ‘Glasgow Child Poverty Programme’, has since agreed a new plan with the Scottish Government covering 2024-2027. This builds on learning from the first ‘phase’ of the pathfinder and introduces a number of new elements, including place-based ‘Demonstration of Change’ projects in wards identified as having high numbers of children at risk of poverty.

Box 1.1: Glasgow pathfinder workstreams

Pathfinder workstreams (2022-24)

• Accountability and culture
• Data
• Funding and commissioning
• No wrong door
• Programme management and support
• Services and campaigns

Programme workstreams (2024-27)

• City Governance
• Commissioning and procurement
• Data and insights
• Data sharing
• Design and innovation
• No wrong door
• Performance management
• Place-based programme
• Policy alignment and funding flexibility

Since the development of these initial pathfinders, the Scottish Government has invested in five new ‘Fairer Futures Partnerships’, place-based partnerships aimed at providing locally-based wrap-around support to address child poverty. These partnerships, in Inverclyde, North Ayrshire, East Ayrshire, Aberdeen and Perth and Kinross, are expected to build on the learning from the pathfinder projects in Dundee and Glasgow, as well as the Family Wellbeing Partnership in Clackmannanshire (which is being evaluated separately).

Evaluation aims and hypotheses

This phase 2 evaluation was commissioned in Autumn 2023. It follows a phase 1 early implementation and process evaluation and evaluability assessment (Scottish Government 2023a and b). The main aims of this phase 2 evaluation, as set out in the Scottish Government’s Invitation to Tender (ITT) were to:

• Understand the short to medium term impacts of the child poverty pathfinders for families
• Examine how the pathfinders are working and how and whether they are leading to any system change
• Understand the value for money of the pathfinders
• Distil learning to inform the development, further implementation and expansion of child poverty pathfinders in the current sites and in other local authorities.

More detailed evaluation questions were also included in the ITT, and are listed in Appendix E of this report.

As part of the early scoping phase of the phase 2 evaluation, a number of ‘hypotheses’ for how the pathfinders were expected to impact on both families and ‘systems’ were developed, which have also helped to shape the evaluation.

Table 1.1: Hypotheses

Focus: Family

Dundee: By offering place-based, open-ended, targeted support, the pathfinder is able to engage and form trusting relationships with families in poverty who have not been able to access the support they need in the past

Glasgow: By supporting the development and delivery of person-centred, holistic, ‘no wrong door’ approaches to support, the pathfinder is able to engage families in poverty who have not been able to access the support they need in the past

Focus: Family

Dundee: By forming trusting relationships with families, key workers are able to help them meet their short- and medium-term needs, which both helps to address some of the drivers of poverty (e.g. cost of living, income from benefits) and to remove barriers to longer-term routes out of poverty (e.g. barriers to employment, training, etc.)

Glasgow: By working with a NWD approach, Glasgow Helps and wider organisations are able to effectively help families in poverty meet their short- and medium-term needs, which both helps to address some of the drivers of poverty (e.g. cost of living, income from benefits) and to remove barriers to longer-term routes out of poverty (e.g. barriers to employment, training, etc.)

Focus: Family

Dundee: By addressing families’ short and medium-term needs, the pathfinder sets families on a pathway to longer-term and sustainable routes out of poverty, by increasing employability and empowering families to make change for themselves

Glasgow: By addressing families’ short and medium-term needs, the pathfinder sets families on a pathway to longer-term and sustainable routes out of poverty, by increasing employability and empowering families to make change for themselves

Focus: System

Dundee: By investing in a place-based approach, pathfinder partners learn more about how to work effectively with families in poverty, which informs improvement to wider services, joint working and development/strengthening of shared values and processes

Glasgow: By investing in a multi-agency team to drive change, the pathfinder is able to be a catalyst for wider systemic change across the city in terms of how agencies and organisations work with families in poverty, engaging wide ranging partners in addressing systematic and cultural barriers to working more effectively for families in poverty.

Focus: System

Dundee: Ultimately, the need for specific pathfinder services is reduced as fully integrated, person-centred, NWD, place-based approaches to supporting people in poverty become embedded across all partners AND families become better equipped to navigate this themselves – so the system is easier to navigate, and families are more confident about doing so.

Glasgow: Ultimately, the need for specific pathfinder services is reduced as fully integrated, person-centred, NWD, place-based approaches to supporting people in poverty become embedded across all partners AND families become better equipped to navigate this themselves – so the system is easier to navigate, and families are more confident about doing so.

Evaluation design and methods

The phase 2 evaluation took a mixed method approach, combining and triangulating data from primary data collection (interviews and workshops) and analysis of secondary data (documentation, cost data, monitoring data). The intention was that the evaluation would include:

• Theory-based impact evaluation, assessing, as far as is possible, the difference that the pathfinders have made for families and their progress towards achieving systems change. This element was structured around the evaluation theories of change (see below), while the hypotheses above were used to consider overall progress against aims.
• Process evaluation to understand how the pathfinders have worked and the factors that led to (or acted as barriers to) positive outcomes for families and to system change
• A value for money evaluation
• Learning for pathfinder stakeholders and beyond.

Three main stages of primary data collection and analysis provided the data that feeds into this report:

• A scoping phase, from mid-September 2023 to March 2024, involving interviews, meetings, workshops and desk-based review of documents from Glasgow and Dundee. Key outputs from this scoping phase were:
  • Revised theories of change for each pathfinder
  • A scoping phase report for the Scottish Government, setting out the planned approach to the main stage of the phase 2 evaluation
• Two waves of main stage data collection and analysis, from April to August 2024, and from October 2024 to early January 2025. This included:
  • Two waves of qualitative interviews and workshops with professional stakeholders and parents
  • Analysis of monitoring data collected by the Dundee pathfinder and by Glasgow Helps, and shared with the evaluation team
  • Development of a ‘break-even’ analysis framework for Dundee to inform consideration of value for money and exploratory work around how to measure the value for money of whole system change in Glasgow
  • Review of key documentation (e.g. amended project plans, meeting papers and minutes, cost estimates for Dundee, key internal research reports) from both pathfinders.

Table 1.2, below, summarises the primary qualitative data that informs this report. In addition to these formal interviews and workshops, the evaluation has involved a very large number of scheduled and ad hoc meetings with the Scottish Government and the pathfinders, reflecting the fact that both initiatives were (intentionally) evolving over time and the evaluation needed to respond to this evolution. It is worth noting that the evaluation (necessarily) reflects a point in time for these evolving programmes (fieldwork dates are shown in Table 1.2) and they have continued to evolve subsequently.

Qualitative interviews were conducted using topic guides developed by the evaluation team and agreed with the Scottish Government. These are included in Appendices B and C. Qualitative data was summarised using a thematic framework structured around the theories of change.

Quantitative monitoring data was analysed using a combination of SPSS and Excel.

The approach to value for money analysis is discussed in more detail at the start of the relevant chapters.

Table 1.2: Primary (qualitative) data collection for phase 2 evaluation

Participant type: Professional stakeholders

Dundee interviews: Scoping phase (Sept 2023-Feb 2024):

• 1-1/small group interviews with 19 professional stakeholders
• Theory of change workshop (17 participants)

Wave 1 (May/June 2024):

• 1-1/small group interviews with 16 professional stakeholders
• Findings workshop, September 2024 (12 participants)

Wave 2 (Nov/Dec 2024):

• 1-1/small group interviews with 18 stakeholders
• Costs and value for money workshop (12 participants)

Glasgow interviews: Scoping phase (Sept 2023-Feb 2024):

• 1-1/small group interviews with 10 professional stakeholders
• Theory of change workshop (18 participants)

Wave 1 (June-August 2024):

• 1-1/small group interviews with 16 professional stakeholders
• Findings workshop, October 2024 (17 participants)

Wave 2 (Nov 2024-Jan 2025):

• 1-1/small group interviews with 20 professional stakeholders
• Costs and value for money workshop (10 participants)

Participant type: Families (parents)

Dundee interviews: Wave 1: 14 interviewees

Wave 2: 11 interviewees (8 new families, 3 follow-up interviews)

Glasgow interviews: Wave 1: 11 interviewees (Glasgow Helps clients)

Wave 2: 8 interviewees (primarily ILM participants) (Note that only 5 of these were with parents of children aged 16 and under)

In addition to the data included in this report, the phase 2 evaluation included a feasibility assessment for a Quasi-Experimental (QED) pilot study in Dundee only. This would provide a more robust assessment of medium-term impacts for families on employment and income, by enabling comparison of outcomes for pathfinder clients with those for a control group. The feasibility report was submitted to Scottish Government in February 2024. It concluded that a QED pilot was feasible if relevant administrative data could be shared and linked. A final decision on whether or not to go ahead with the QED pilot was still pending at the time of writing (March 2025), depending on the outcome of negotiations between the Scottish Government, DWP, HMRC and Social Security Scotland over relevant data access.

Scope and limitations

The scoping phase report highlighted a number of significant challenges likely to impact on the ability of the evaluation to deliver comprehensive and robust impact evaluation of the two pathfinders (and by extension, value for money evaluation). These were:

• The evolving nature of the pathfinders: Which meant that individuals might receive different interventions if they engaged at different times.
• Expected timelines for impacts: Early interviews in Dundee highlighted uncertainty about the expected timelines for impacts for both families and system change. This included a concern that impacts might not be evident within the evaluation time frame. In Glasgow, it was repeatedly emphasised that ‘whole system change’ is expected to require at least a 10-year project. There was a clear strand of opinion among those interviewed in both pathfinders that the timing of this evaluation was too early to meaningfully assess their full impact.
• Burden on the pathfinder teams: Balancing the needs and timing of the evaluation with the impact on the pathfinder teams was difficult at times. In Glasgow, in particular, there were some significant delays in agreeing the approach to the evaluation, reflecting the fact that the phase 2 evaluation began at a transitional point for the Glasgow pathfinder when the pathfinder team needed to focus on finalising their 2024-2027 plans. In Dundee, there was limited resource among the key worker team to retrospectively make improvements to their data monitoring processes to support the evaluation.
• Limits to the availability of, and access to, quantitative data to support the evaluation: Three potential sources of quantitative data to support the evaluation were considered at the outset, of which two proved unfeasible due to timescale and data protection issues:
• First, it was hoped that the evaluation would be able to access administrative data from HMRC, Social Security Scotland and DWP in order to analyse any changes in families’ income from employment and benefits. As discussed above, at the time of writing this report, discussions on data access were still ongoing.
• Second, at the scoping phase the feasibility of surveying parents who had accessed the pathfinders was considered. However, neither pathfinder had collected consent to share contact details of parents the pathfinder had engaged with outside agencies. It was agreed between the research team and Scottish Government that collecting this consent retrospectively would be burdensome on pathfinder teams and that the response rate would likely be very low. This ruled out surveys as a feasible option within the evaluation timeframe.
• A third source of data was monitoring data collected by the Dundee pathfinder team and, in Glasgow, by Glasgow Helps. While these data provide some useful insights into the Dundee pathfinder and Glasgow Helps as an exemplar of the NWD approach in Glasgow, they were not developed with the Scottish Government’s evaluation questions in mind and therefore were missing some key data, including around client characteristics and outcomes.

Further detail on the content of the monitoring data is discussed in later chapters. However, overall, the quantitative data available on outcomes for families was limited.

The issues detailed above also impact on how comprehensive and robust any impact evaluation can be at this stage. In general, for a ‘gold standard’ impact evaluation, there needs to be a clearly specified intervention, aimed at a defined, identifiable target group, with agreed outcomes and a timeline for achieving these that enables the collection of baseline and final outcome data within the evaluation timeframe (or where robust baseline data has already been collected). Ideally, equivalent data is also required for a comparison group in order to robustly establish attribution. These criteria did not apply in either Dundee or Glasgow when this phase 2 evaluation took place.

This also has consequences for the approach it was possible to take to value for money evaluation; without robust quantitative impact data it is not possible to conduct formal cost benefit analysis. Given this, alternative approaches were developed, in consultation with the Scottish Government, for each area. These are discussed in more detail in the relevant chapters.

Finally, in terms of qualitative data from families, the evaluation only interviewed parents. Parent interviews touched on the impacts of the support received for their children, but there is less evidence on direct impacts for children than on the impacts for parents (although most of these could be expected to have indirect impacts on their children’s wellbeing).

In addition, the boundaries and scope of each of the pathfinders had to be to considered carefully:

• It was clear from the scoping phase that it would not be possible or appropriate (given timeframe and resources) for this evaluation to attempt to consider every element of the Glasgow pathfinder in detail due to its large scale and scope. As such, the evaluation of the Glasgow pathfinder has adopted more of a ‘case study’ approach, focusing on specific elements of the pathfinder and drawing out learning from their progress at mid to late 2024, while recognising that the pathfinder is expected to take longer to fully realise its aims.
• For Dundee, the evaluation focused on the key worker model and multi-agency drop-in in Linlathen and Mid-Craigie, which was the core element of the pathfinder during the period of the evaluation. The aim was to deliver, as far as possible, a theory-based impact and process evaluation of this intervention, while acknowledging the limits of the data (particularly quantitative data) available to support definitive conclusions. However it should be noted that following consideration of early findings by the Dundee Oversight Board, the approach in Linlathen/Mid Craigie was subsequently expanded to different areas of the city, but these are not covered by this evaluation.

Reporting approach and conventions

Given the differences in approach between Dundee and Glasgow – both in terms of the pathfinders themselves, and differences in evaluation approach stemming from this – it is not appropriate to compare findings between the two directly. The remainder of this report therefore largely presents findings from Dundee (chapters 2-6) and Glasgow (chapters 7-10) separately. The main exception is the final chapter (chapter 11), which draws together cross-cutting themes and learning from across the two.

Each of the Dundee and Glasgow sections of this report include chapters on:

• The design and development of the pathfinders (chapters 2 and 7), which cover this in more detail, providing context for understanding and interpreting subsequent findings.
• Evidence on the impacts of the pathfinders on families (chapters 3-4 and 8)
• Perceived impacts to date on system change (chapters 5 and 9)
• And the pathfinders’ value for money (chapters 6 and 10).

Summaries of evidence for the evaluation hypotheses are included at the end of chapters on impacts for families and on system change. Each findings chapter also includes a boxed summary of key learning and reflections, aimed primarily at future ‘pathfinders’ or similar projects but also relevant to Glasgow, Dundee, and their national partners. While a broad overview of the limitations on the evaluation is given above, each chapter also includes discussion of key data limitations to keep in mind when interpreting the findings presented, as relevant to that chapter.

Chapters are structured thematically, with quantitative and qualitative data interspersed. When discussing qualitative data, the report generally avoids using ‘quantifying’ language, such as ‘most’ or ‘a few’, as the purpose of qualitative research is to map the range of views rather than to measure prevalence. To protect professional anonymity, we have avoided including participant identification numbers when including quotes from professional interviewees, given the small number of professionals in specific roles within each pathfinder.

Finally, the authors recognise that this report is long, but the scope of the evaluation aims and of pathfinder activities necessitated a certain level of detail. Anyone keen to identify the key messages could skip directly to the ‘conclusions and learning’ sections at the end of each findings chapter. There is also a shorter, standalone summary version of this report. However, it is also important to acknowledge that, although this report is long, given the scope of the pathfinders, there is much more that could have been added. While it is hoped that the report strikes a balance between sufficient detail for the reader to understand key themes and a manageable overall length, it must be recognised that no report of this type can be completely comprehensive.