Official Statistics

Employment Data Lab Analysis: Durham County Council DurhamWorks programme

Published 9 July 2026

This Employment Data Lab report presents estimates of the impact of the DurhamWorks programme on the Employment and Education outcomes of the programme participants. The DurhamWorks programme is aimed at supporting 16 to 24 year olds who are not in employment, education or training (NEET).

The results in this report have been generated using quasi-experimental techniques which introduce some uncertainty. The results should be used with a degree of caution. Further information can be found in Section 7, and in an associated methodology report.

Headline results

Increase in employment

Between 4 and 7 percentage points more programme participants were classed as employed 2 years after starting the programme, than had they not participated. This result was statistically significant.

Increase in continuous employment

Between 7 and 10 percentage points more programme participants achieved continuous employment (six months or more in a row) in the two years after starting the programme, than had they not participated. This result was statistically significant.

Increase in training and education courses passed

The percentage of participants who passed an education or training course at any time during the 2 years after start was between 10 and 13 percentage points higher than had they not participated. This result was statistically significant.

• The main analysis focuses on a sub-group of 4,868 evaluated participants (out of 8,139) who started the programme between August 2015 and October 2021 and were between the ages of 18 to 24.
• Additional sub-analyses were carried out exploring the impact of the programme on those with special educational needs (SEN) provisions and those of different genders. There was also analysis on the impact of the COVID-19 pandemic and analysis of the longer-term impacts up to four years after start.
• For this report the Employment Data Lab team used administrative data to analyse labour market and education outcomes for two to four years after starting the programme.
• Participants were compared to a comparison group of “similar” individuals to evaluate the programme.
• The three headline statistics were chosen before starting the analysis as the primary outcome measures to assess the success of the programme.

1. What you need to know

What is the Employment Data Lab?

The Employment Data Lab is a service provided by a team of analysts at the Department for Work and Pensions (DWP). The Data Lab provides group-level benefits and employment information to organisations who have worked with people to help them into employment. The purpose is to provide these organisations with information to help them understand the impact of their programmes.

What was the DurhamWorks programme?

DurhamWorks was a partnership programme led by Durham County Council, which aimed to break down barriers to work and build employability skills for young people. The programme targeted 16 to 24 year olds who were not in Education, Employment or Training (NEET) in County Durham.

DurhamWorks integrated 3 strands of activity, through which each participant received tailored support:

Strand 1: Transition, Peer Mentor and Employment Support

In this strand participants met with DurhamWorks advisors. The objective was to ensure participants were guided to the right initiatives and projects available within the programme. Peer mentors were used to fill potential knowledge gaps between participants.

Strand 2: Engagement and Progression

This strand shifted towards more targeted delivery of flexible and tailored interventions that met the individual needs of a participant. Delivery partners and subcontractors were used to support progression towards the labour market.

Strand 3: New Employment Zone

The purpose of this strand was to engage employers and create work related opportunities such as jobs or apprenticeships. Delivery partners specialised in apprenticeships, social enterprise, entrepreneurship, work experience and volunteering.

DurhamWorks gave participants access to one-to-one support from both transitional and specialist advisors. This ensured participants progressed onto suitable delivery partner schemes and addressed barriers both in and outside of employment, education or training. Information provided by Durham County Council showed that participants most commonly had meetings with advisors ranging from once or twice a week to fortnightly with weekly meetings the most common, and meetings lasted 34 minutes on average. Some examples of support offered to participants were:

• Information, advice and guidance about available education, employment and training opportunities
• Support to understand and manage barriers to progression e.g. confidence
• Support to apply for employment opportunities
• Help to prepare for interviews
• Help to access financial support for clothing and equipment required for education and employment
• Support to access transport and to travel independently
• Support to access benefits they may be entitled to
• Help to access volunteering and work experience opportunities

DurhamWorks also provided Employer Grants to small and medium enterprises (SMEs) who hired programme participants. The Employer Grant provided up to £2,500 to help employers cover the initial cost of hiring a young person. To qualify for an Employer Grant, the job opportunity had to be for at least 30 contracted hours per week and have some form of education or training attached to it.

The DurhamWorks programme consisted of three phases, which have been delivered in full. Delivery of phase one ended in July 2018. Phase two began in August 2018 and ended in December 2021, and phase three began in January 2022 and ended in December 2023. Key differences between phase one and phase two included a reduction in delivery partners used from 16 to 8, as well as disruption to phase two delivery caused by the COVID-19 pandemic.

As the Department for Work and Pensions only holds data on DurhamWorks participants up to October 2021 (when European Social Fund^{[footnote 1]} (ESF) funding ended), this analysis solely investigates outcomes for participants who enrolled during phase one and phase two.

Who was evaluated as part of this analysis?

Data was shared on 8,141 participants who took part in the programme between August 2015 and October 2021. The main impact analysis focusses on a subset of 4,868 participants who were between 18 and 24 when they started the programme to ensure that participants were unlikely to be in compulsory full-time education one year after starting the programme. For more details see Appendix A.

Participant information

Of the 4,868 participants included in this analysis, the available administrative data indicated:

• 62% were male and 38% were female
• Average age on joining the programme was 20 years
• 96% were white
• 39% had previously been eligible for Free School Meals (FSM)
• 45% had previously had SEN provisions
• 51% had been employed at some point in the two years before starting the programme

Further information on those who were and were not included as part of the analysis and missing markers in the administrative data can be found in Appendix A and Appendix B.

The analysis in this report

This report presents analysis on the impact of the DurhamWorks programme by comparing education, employment and benefit outcomes of participants to those of a matched comparison group who did not participate. The comparison group is used to estimate the outcomes of participants had they not participated in the programme and was created using a method called propensity score matching (PSM).  Further information about how the analysis was conducted can be found in the associated methodology document.

The following primary outcome measures were selected for this evaluation before the analysis was undertaken.

Primary outcome measures

• The percentage of the group classed as employed two years after starting the programme.

• The percentage of the group that achieved sustained employment at any time during the two years after starting the programme. Sustained employment is defined as being employed for six months or more in a row.

• The percentage of the group who passed an education or training course at any time during the two years after starting the programme.

The main analysis in this report (Section2 and Section 3) presents the impact for all individuals in Durham aged 18 to 24 who participated in the DurhamWorks. Section 4 explores the longer-term impacts of the programme for the subset who joined before 1 April 2021. Finally, Appendix F examines the impact of the provision on other subgroups (gender, those with SEN and those affected by the COVID-19 pandemic).

2. The labour market impacts of the programme over time

The results show that the programme led to:

• More classed as employed, at 2 years
• More classed as looking for work, at 2 years
• Fewer classed as inactive, at 2 years
• Fewer classed as ‘other’, at 2 years

An increase in employment, at 2 years

Between 4 and 7 percentage points more programme participants were classed as employed 2 years after starting the programme, than had they not participated.

This result was statistically significant.

More classed as looking for work, at 2 years

Between 6 and 9 percentage points more programme participants were classed as looking for work 2 years after starting the programme, than had they not participated.

This result was statistically significant.

Fewer classed as inactive, at 2 years

Between 2 and 5 percentage points fewer participants were classed as being on inactivity benefits, 2 years after start.

This result was statistically significant.

Fewer classed as ‘other’, at 2 years

Between 6 and 8 percentage points fewer participants were classed as ‘other’, 2 years after start.

This result was statistically significant.

Employment Data Lab reports use four categories of labour market status (see Section 7 for more details).

The figures and tables in this section show the number of people in each labour market category over the two years before and after starting the intervention. The participants are compared to a comparison group used to estimate the outcomes they would have achieved had they not participated in the programme. The difference between the groups can be interpreted as the impact of the programme.

The results in Figure 1 and Table 1 show that the programme led to a statistically significant increase in both employment and looking-for-work benefit claims at one and two years after programme start. The results also show a statistically significant decrease in both inactivity benefit claims and the number of participants in the ‘other’ category at one and two years after programme start.

Figure 1: Plots showing the impact of the programme on the numbers in each labour market category over the two years after programme start

The plots on the left (in orange) show the percentages of the participant and comparison groups in each category. The difference (or impact of the programme) is shown on the right in blue. The darker blue line shows the central estimate, and the shaded blue area is the 95% confidence interval. 

1(a & b) – Employed: The impact plot (b) shows the programme had a positive and statistically significant effect on the percentage of participants who were employed. The outcomes plot (a) shows the percentage of both groups employed at start is similar, but the percentages increase at a faster rate for the participant group over the 2 years after start. 

1(c & d) – Looking for work: The impact plot (d) shows the programme resulted in a statistically significant increase in those on looking-for-work benefits. The outcomes plot (c) shows the percentage of both groups in the looking for work category increasing up to programme start, and the percentage in the category decreasing at a lesser rate for those in the participant group in the 2 years after start.

1(e & f) – Inactive: The impact plot (f) shows the programme led to a statistically significant reduction in the percentage of participants in receipt of ‘inactive’ benefits. The outcomes plot (e) shows the percentage of both groups on inactive benefits.

1(g & h) – Other: The impact plot (h) shows the programme led to a statistically significant and sustained reduction in the percentage of participants in the ‘other’ category. The outcomes plot (g) shows the percentage categorised as ‘other’ fell at a faster rate for the participant group initially. In the 2 years after start, the impact tended towards 0 but remained significant.

Table 1: Showing the percentage of each group in each category at one and two years after starting the programme.  The impact, or difference, is shown along with an indication of statistical significance:

The impact, or difference, is shown along with an indication of statistical significance. The “upper” and “lower” values give the 95% confidence interval around the central estimate of the impact. Percentage points are denoted by ppt.

Percentage of group in category:	Participant group (%)	Comparison group (%)	Impact: Central (ppt)	Impact: Lower (ppt)	Impact: Upper (ppt)	Sig.
Employed (1 year)	51	44	6	5	8	yes
Looking for work (1 year)	37	25	12	10	13	yes
Inactive (1 year)	15	19	-4	-5	-2	yes
Other (1 year)	10	21	-11	-12	-10	yes
Employed (2 years)	56	50	5	4	7	yes
Looking for Work (2 years)	29	21	8	6	9	yes
Inactive (2 years)	18	22	-3	-5	-2	yes
Other (2 years)	9	16	-7	-8	-6	yes

Note: Values are rounded to the nearest whole number, so statistically significant impacts may have a 95% confidence interval with a lower bound of zero.

Note: Categories are not mutually exclusive, so percentages in category do not sum to 100% and impacts do not sum to 0.

Table 2 shows the effect of the programme on the percentage of participants who had at least one sustained employment spell at any point during the two years after start. Sustained employment is defined as being employed for six months or more in a row. During the two years after start 7 and 10 percentage points more participants had a sustained employment spell.

Table 2: Showing the percentage of the participant and comparison groups who had a sustained employment spell at any point during the two years after start.

Percentage of group with sustained employment at any point…	Participant group (%)	Comparison group (%)	Impact: Central (ppt)	Impact: Lower (ppt)	Impact: Upper (ppt)	Sig.
…in the 2 years after start	61	53	9	7	10	yes

To explore the labour market impacts further, Table 3 shows an additional labour market impact measure - the percentage of each group who were in work at any point during the one and two years after starting the programme. The results indicate the programme led to a statistically significant increase in the number of participants in work at any point during:

• The year after programme start. Employment during this period increased between 8 and 12 percentage points.
• The two years after programme start. Employment during this period increased between 7 and 10 percentage points.

Table 3: Showing the percentage of the participant and comparison groups in the Employed category at any point during the 12 and 24 months after start.

Percentage of group employed at any point…	Participant group (%)	Comparison group (%)	Impact: Central (ppt)	Impact: Lower (ppt)	Impact: Upper (ppt)	Sig.
…in the year after start	64	54	10	8	12	yes
…in the 2 years after start	74	65	9	7	10	yes

3. Impact on education and training 

The Table 4 and Figure 2 in this section show the impact of the programme on the percentage of participants passing education or training courses during the two years after programme start.

Table 4 shows the percentage of participants and comparators who passed an education or training course at any time during the one and two years after starting the programme. The results suggest the impact of the programme was a statistically significant increase of between 10 and 13 percentage points in courses passed during the two years after start.

Table 4: Showing the percentage of the participant and comparison groups classed as passing a course during the year after programme start and during the 2 years after programme start. 

Percentage who passed a course at any point…	Participant group (%)	Comparison group (%)	Impact: Central (ppt)	Impact: Lower (ppt)	Impact: Upper (ppt)	Sig.
… in the year after start	23	18	6	4	7	yes
… in the 2 years after start	35	23	11	10	13	yes

Figure 2 breaks down the levels^{[footnote 2]} of qualifications passed by the participant and comparison groups at any point during the two years after programme start. The results suggest DurhamWorks had a positive and significant impact on those achieving entry level, level one and level two qualifications in the year after starting the provision. The results also suggest the programme led to a statistically significant reduction in the number of people obtaining level 3 or higher qualifications over the same period.

Figure 2: Showing the percentage of the participant and comparison groups who passed at least one Entry Level, Level 1, Level 2, and Level 3 or higher course at any point during the two years after programme start.

Note: Level 3 and higher passed course outcomes were combined for statistical disclosure control reasons.

Specific data limitations mean that these figures should be treated with a greater degree of caution.  See Section 8 for more detail.

4. Longer term impacts

For a subset of 4,611 programme participants, it is possible to observe longer term labour market outcomes. For participants who started the programme before 1 April 2021 it is possible to observe four years of labour market outcomes. The impacts of the programme on employment and benefit receipt three and four years after start can be estimated for this subgroup, allowing for additional years of labour market impacts to be observed. Tables and figures below represent this subgroup of the main cohort. For most of this subgroup, the four-year outcomes period overlaps the COVID-19 pandemic and associated lockdowns (see Appendix F for more details).

The results in Figure 3 and Table 5 show that the statistically significant increase in employment at two years after programme start was sustained up to four years after programme start. The results also show the initial statistically significant decrease in inactivity benefit claims was not sustained and became statistically insignificant between three and four years after programme start. The impacts on the number of participants in the ‘other’ and employed categories declined but were still statistically significant four years after programme start.

Figure 3: Plots showing the impact of the programme on the numbers in each labour market category over the four years after programme start

The plots on the left (in orange) show the percentages of the participant and comparison groups in each category. The difference (or impact of the programme) is shown on the right in blue. The darker blue line shows the central estimate, and the shaded blue area is the 95% confidence interval. 

3(a & b) – Employed: The impact plot (b) shows the programme had a positive and statistically significant effect on the percentage of participants who were employed up to four years after start. The outcomes plot (a) shows that while the impact remained similar between two and four years after start the percentage of participants in employment increased. 

3(c & d) – Looking for work: The impact plot (d) shows the programme resulted in a statistically significant increase in those on looking-for-work benefits, but that the impact continued to decrease between two and four years after start. The outcomes plot (c) shows the percentage of participants receiving looking-for-work benefits decreased at a faster rate than the comparison group.

3(e & f) – Inactive: The impact plot (f) shows four years after start the impact on percentage of participants in receipt of ‘inactive’ benefits is not statistically significant. The outcomes plot (e) shows between two and four years after start the percentage of participants on ‘inactive’ benefits converged with the comparison group.

3(g & h) – Other: The impact plot (h) shows impact of the programme on the percentage of participants in the other category decreased between two and four years after start. The outcomes plot (g) shows the percentage categorised as ‘other’ fell at a faster rate for the participant group initially but that the comparison group then began to converge.

Table 5: Showing the percentage of each group in each category at three and four years after starting the programme.  The impact, or difference, is shown along with an indication of statistical significance:

The impact, or difference, is shown along with an indication of statistical significance. The “upper” and “lower” values give the 95% confidence interval around the central estimate of the impact. Percentage points are denoted by ppt.

Percentage of group in category:	Participant group (%)	Comparison group (%)	Impact: Central (ppt)	Impact: Lower (ppt)	Impact: Upper (ppt)	Sig.
Employed (3 years)	57	52	6	4	7	yes
Looking for work (3 years)	25	20	6	4	7	yes
Inactive (3 years)	21	24	-3	-4	-1	yes
Other (3 years)	8	14	-6	-7	-5	yes
Employed (4 years)	60	54	5	4	7	yes
Looking for Work (4 years)	23	19	3	2	5	yes
Inactive (4 years)	24	25	-1	-3	0	no
Other (4 years)	6	11	-5	-6	-4	yes

Note: Values are rounded to the nearest whole number, so statistically significant impacts may have a 95% confidence interval with a lower bound of zero.

Note: Categories are not mutually exclusive, so percentages in category do not sum to 100% and impacts do not sum to 0.

Table 6 shows the effect of the programme on the percentage of participants who had at least one sustained employment spell at any point during the three and four years after start. Sustained employment is defined as being employed for six months or more in a row. This shows the statistically significant increase in sustained employment was maintained during the three and four years after start. Table 7 also shows that the statistically significant increase in the percentage of people employed at any point was sustained during the three and four years after start.

The impact of the programme on other subgroups is explored in Appendix F.

Table 6: Showing the percentage of the participant and comparison groups who had a sustained employment spell at any point during the three and four years after start.

Percentage of group with sustained employment at any point…	Participant group (%)	Comparison group (%)	Impact: Central (ppt)	Impact: Lower (ppt)	Impact: Upper (ppt)	Sig.
…in the 3 years after start	69	61	8	7	10	yes
…in the 4 years after start	74	66	7	6	9	yes

Table 7: Showing the percentage of the participant and comparison groups in the Employed category at any point during the 36 and 48 months after start.

Percentage of group employed at any point…	Participant group (%)	Comparison group (%)	Impact: Central (ppt)	Impact: Lower (ppt)	Impact: Upper (ppt)	Sig.
…in the 3 years after start	80	72	8	7	9	yes
…in the 4 years after start	83	76	7	6	9	yes

5. How to use the results of this report?

Three primary outcome measures were chosen to assess the success of this programme.  The results suggest that the programme had positive, and statistically significant impacts on all three primary outcome measures. This suggests the programme has been successful at:

• Increasing the percentage of the group classed as employed two years^{[footnote 3]} after starting the programme.

• Increasing the percentage of the group that achieved continuous employment at any time during the two years after starting the programme.

• Increasing the percentage of the group passing an education qualification at any time during the two years after starting the programme.

A range of secondary outcome measures were also analysed in this report in Appendix E and can be used to learn more about the impacts of the programme. A range of outcomes for other subgroups were also analysed in Appendix F. Results marked as statistically significant indicate an estimate that is unlikely to have occurred by chance (and is more likely to be a causal impact of the programme). If a result is not statistically significant it does not mean that there was no impact, it just means there was insufficient evidence to verify this to the required threshold.

The estimates in this report were generated using quasi-experimental methods that can be less reliable than experimental methods such as a randomised control trial. The results should be used with a degree of caution.

The estimates were also generated using a subset of DurhamWorks programme participants, notably those aged between 18 and 24. Care should be taken in generalising the results to those outside of this group.

The estimates relate to a programme working in a particular context. This report makes no assessment as to whether these impacts are generalisable to different contexts. The estimates were also made in a “business as usual” setting where participants and comparators were free to go on to access other support.

6. Durham County Council in their own words

Durham County Council have provided a description of their programme in their own words and a response to the analysis.

7. About these statistics

This report presents estimates of the impact of a programme.  This is achieved by comparing the outcomes of the programme participants to a credible estimate of their outcomes had they not participated in the programme. This is often referred to as the counterfactual.  In this report the counterfactual was generated using a quasi-experimental technique called PSM.  This involves constructing a comparison group of individuals who did not participate in the programme but who are matched on key characteristics that affect whether an individual takes part in the programme and the outcomes that they experience as a result of participation.

Once this comparison group has been constructed the outcomes of the two groups can be compared to generate the estimate of the impact of the programme.  More information about this technique and how it is used in the Data Lab can be found in the methodology report.

Categorisation

The analysis in this report is based on the labour market outcomes of the participants (and a matched comparison group) in the two years before and after starting the intervention.  This report uses four categories of labour market status for the analysis, detailed below. 

• Employed:  People who are either employed or self-employed

• Looking for Work:  People who are in receipt of Jobseeker’s Allowance (JSA), or in the Universal Credit (UC) “intensive work search”, “light touch out of work”, “light touch in work”, or “working enough” conditionality regimes.

• Inactive:  People who are in receipt of inactive benefits such as Employment and Support Allowance (ESA) or in the UC “no work requirements” or “work focussed interview” conditionality regimes. Several other benefits also fall into this category, though the numbers of people on these benefits is small. See methodology report for details.

• Other:  People who do not fall into the above three categories, this could include people who are in full-time education and not working or receiving benefits or those who are in custody.

These categories are not mutually exclusive, and it is possible to be in more than one category. For example, someone working fewer than 16 hours a week may also be in receipt of JSA and would be classed as “employed” and “looking for work”. 

Statistical significance

The report highlights if the results are statistically significant or not.  A statistically significant result is one that is unlikely to have occurred by chance because of sampling error.  If a result is not statistically significant it does not mean that the intervention has no impact, it simply means that there is not enough evidence to verify this to a required threshold.  In this report, unless otherwise stated, the threshold for significance is 95%. 

This report sometimes presents the central estimate of a result along with the upper and lower confidence values.  These upper and lower values create a range that you would expect the estimate to fall within if the test was to be redone, within a certain level of confidence.  This level is set at 95 per cent unless otherwise stated. The confidence intervals will typically be stated in the tables of results and be presented on graphs and plots as either error bars or shaded regions.

Limitations

The validity of the technique used in this report rests on the assumption that all the characteristics that are linked to a person’s participation in the programme and the outcome variables of interest have been sufficiently accounted for in the analysis, either explicitly or otherwise.  This is a strong assumption that cannot be tested and depends on the data available and on the nature of each programme and its participants.  This is reviewed on a case-by-case basis in the Data Lab and impact evaluations are only carried out where the validity of this assumption is plausible. That said, these are quasi-experimental techniques that tend to be less robust than true experimental methods, such as a randomised control trial, and the results must be treated with a degree of caution.

Throughout this report it is highlighted that particular caution should be applied when using and interpreting the education related results. The reason for this stems from the fact that the education spells data relates to enrolment on a course, and actual attendance and/or drop-outs are not always captured and accounted for. This has the potential for someone to appear as though they are in education when in practice they are not (for example if they enrolled on a course and subsequently dropped out).  Since these programmes are aimed at people who are NEET, i.e. Not in Employment, Education or Training, it is possible that this issue would be more likely to affect the participant group than the comparison group, therefore introducing some bias into the results.  Whilst there are indications in the data that this sort of bias may be present to a degree, sensitivity analyses have been conducted providing reassurance that the benefit and employment related impact estimates are robust. The education related impacts are more exposed to this potential issue and therefore should be treated with more caution.

Where to find out more

Read the Employment Data Lab analysis, information and guidance.

8. Statement of compliance with the Code of Practice for Statistics

The Code of Practice for Statistics (the Code) is built around 3 main concepts, or pillars: 

• Trustworthiness – is about having confidence in the people and organisations that publish statistics
• Quality – is about using data and methods that produce statistics
• Value – is about publishing statistics that support society’s needs

The following explains how we have applied the pillars of the Code in a proportionate way. 

 Trustworthiness 

Employment Data Lab reports, such as this, are published to provide User Organisations with an estimate of the impact of their programmes that support employment. Releasing them via an ad hoc publication will give equal access to all those with an interest in them.

Quality 

The methodology used to produce the information in this report has been developed by DWP analysts in conjunction with the Institute for Employment Studies. The information is based on data from the User Organisation and Government administrative data.  The calculations have been quality assured by DWP analysts to ensure they are robust.

Value 

Producing and releasing these estimates provides User Organisations and the public with useful information about employment support provision that they may not have otherwise been able to generate or obtain.

Appendix A: Exclusions from the treatment group

Durham County Council shared data on 8,141 participants who were between the ages of 16 to 24 and took part in the programme between August 2015 and October 2021. The main impact analysis focussed on a subset of these participants who were aged 18 to 24 and lived in County Durham when they started the programme. The age restriction was to ensure that participants were unlikely to be in compulsory full-time education one year after starting the programme.

Figure 4 shows the distribution of programme start dates for all 8,139 programme participants matched to administrative data, along with reasons for exclusion from the analysis. Blue bars show the start dates for the participants who were used in the main analysis. The start date distribution shows gaps in provision start for Summer 2018 and Spring 2020. These reflect the ramping down of phase one and COVID disruption respectively.

Figure 5 shows the stages that individuals were excluded from the analytical process. The final group of 4,868 participants, comprising the main analysis group used in the PSM, represent approximately 60% of the matched participants who took part during the relevant period.

Figure 4: Showing the distribution of start dates of the programme participants.

Figure 5: Presenting a diagram showing the numbers of participants and the stages at which they were excluded from the analysis.

Appendix B:  Participant group information

Table 8 displays the participant group information for the full analysis sample who could be matched to administrative data. This is then broken down into the 4,868 “evaluated” participants; those who were selected for the evaluation, the 3,271 “non-evaluated”; those who were excluded from the evaluation based on age and geography, and “all” of the participants, for whom data was available.

Table 8: Showing characteristics, benefits and employment information for the participant group (%)

Variable	Evaluated	Non-Evaluated	All
Observations	4,868	3,271	8,139
Age (mean years)	20	16.6	18.6
18 years or under (%)	27	97	55
Over 18 years (%)	73	3	45
Male (%)	62	58	60
SEN marker set (%)	45	37	42
FSM marker set (%)	39	44	41
Care leaver/adopted marker (%)	4	9	6
Child in need marker (%)	10	20	14
Exclusion marker (%)	14	27	19
Permanent exclusion marker (%)	1	3	2
Employed marker (%)	51	18	37
Child on CHB claim marker (%)	55	93	71
White ethnicity (%)	96	94	95
Other ethnicity (%)	4	6	5
Entry level qualification (%)	98	86	93
Level 1 qualification (%)	97	85	93
Level 2 qualification (%)	91	68	82
Level 3 qualification (%)	43	5	27
Level 4 qualification (%)	8	1	5
Level 5 qualification (%)	6	1	4
Level 6 qualification (%)	5	1	3
Level 7 qualification (%)	x	x	1
Level 8 qualification (%)	x	x	x
Enrolled in ‘School’* at start (%)	19	46	30
Enrolled in Further Education at start (%)	14	17	15
Enrolled in Higher Education at start (%)	x	x	1
Number of months ‘Employed’ in the previous two years	6	1	4
Number of months ‘Looking for Work’ in the previous two years	4	0	2
Number of months ‘Inactive’ in the previous two years	3	0	2
Number of months ‘Other’ in the previous two years	2	11	6

Note: Some figures which have been suppressed for disclosure control purposes are denoted by an x.

*The ‘School’ category covers any education or training spells captured in the School Census, Pupil Referral Unit Census, Alternative Provision Census, Key Stage 4 or Key Stage 5 datasets

Appendix C: Matching the comparison group

PSM is used to construct a comparison group of individuals that are matched on key characteristics that are linked to a person’s participation in the Programme and the outcome variables of interest.  More information about this technique and how it is used in the Data Lab can be found in the methodology report.

Before proceeding with the analysis, the Data Lab team assessed the plausibility of constructing a comparison group that satisfies the conditional independence assumption that underlies PSM (see methodology report for more details). The programme was targeted at individuals with some characteristics that were well represented in the available data.

The comparison pool was selected from the Department for Education’s (DfE) administration data and was restricted to only include individuals who were in the same age range as the participants at the time of the programme start. This group was then assigned a pseudo-start date (in lieu of an actual start date) in a way that matched the distribution of participant start dates.

This group was then reduced further in additional steps:

• Firstly, individuals who were known to have participated in other ESF programmes were excluded. 

• The group was then restricted to geographic locations that were similar to, but excluded, the locations that the programme was delivered in. The selection of these locations was based on cluster analysis that used local labour market characteristics and indicators to group Local Authorities in Great Britain into 14 groups.  See Appendix D for more info.
• The comparison pool was restricted further by stratified sampling that matched the distributions of the comparison and participant groups on their age and labour market status at programme start/pseudo-start.

These steps resulted in a comparison pool of approximately 54,000 individuals who were then used in the matching process.

The matching estimator used to generate the impact estimates presented in this report was nearest neighbour matching using 100 nearest neighbours and a bandwidth of 0.01. Nearest neighbour matching involves running through each participant and matching them with the closest eligible individuals from the comparison pool, determined by closeness of the propensity scores.  The sensitivity of the impact estimates to the choice of matching estimator was tested using a range of estimators and found to be insensitive. Further information about matching estimators can be found in the methodology and literature review documents.

Table 9 below shows a sample of the variables used in the matching process and the mean values of these variables both before and after matching.  The table shows that before matching, the participant and comparison groups are not well matched, or balanced, as shown by sizeable differences in the mean values.  After matching, the mean values of the participant and comparison groups are much closer. The percent bias and p-value columns provide information on how big the residual difference is and if this difference is statistically significant.  Ideally one would like the percent biases to be small (below 5%) and there to be no statistically significant differences i.e., p-values above 0.05 (the 95 percent confidence level threshold).

Table 10 also presents summary statistics that relate to how well matched the participant and comparison groups are for the main run. It shows values for Rubin’s B, Rubin’s R and the maximum and median percent biases, all of which meet commonly accepted thresholds for the selected approach (see the methodology report for more details). The percent biases for each variable after matching were also small (below 5%) and statistically insignificant (p-values above 0.05). The table also shows there were 12 participants (0.25 percent) who were off support (had no matched comparator). This is a sufficiently small percentage so as not to raise concerns about the representativeness of the results.

Table 9: Showing mean value of each control variable for the main run, before and after matching

Variable	Unmatched Comparison Group (%)	Unmatched participant group	Matched comparison group	Matched participant group	Percent bias after matching	P value
CHAR_AGE	20.1	20.0	20.0	20.0	3.6	0.07
CHAR_AGE_SQ	408.3	404.5	401.6	404.4	3.6	0.07
DFE_FSM	27.8	38.7	40.3	38.7	-3.3	0.12
dfe_level_3_start	48.7	42.6	41.0	42.5	3	0.13
dfe_level_4_start	7.5	7.6	6.8	7.5	2.8	0.16
PIT_INT_START	1.6	3.9	4.3	3.9	-2.8	0.27
DFE_CIN	9.9	9.7	10.5	9.7	-2.7	0.19
SPELL_HIST_CHB_CHILD	47.1	55.2	56.5	55.2	-2.5	0.22
SPELL_HIST_HB	11.5	16.0	16.7	15.9	-2.3	0.3
SPELL_HIST_JSA	8.2	18.8	19.5	18.7	-2.2	0.34
DFE_A_start	10.3	13.9	13.1	13.9	2.2	0.3
dfe_level_5_start	6.5	6.4	5.8	6.3	2.2	0.26
SPELL_HIST_DLA	6.2	6.5	7.1	6.5	-2.2	0.3
dfe_level_2_start	85.5	90.8	90.1	90.8	2.1	0.25
dfe_level_6_start	5.8	5.3	4.8	5.2	2.1	0.29
CHAR_INT_YEAR_2019	20.3	20.2	19.3	20.2	2	0.31
SPELL_LFW_m3	9.7	28.0	28.8	28.0	-1.9	0.43
PIT_SANC_HIST	5.2	11.5	12.0	11.5	-1.9	0.42
SPELL_LFW_m2	10.2	33.9	34.7	33.9	-1.9	0.44
DFE_exclusion	13.3	14.3	14.9	14.3	-1.8	0.38
SPELL_HIST_EMPLOYMENT	50.1	50.9	51.7	50.8	-1.8	0.37
dfe_level_7_start	0.7	0.9	0.7	0.8	1.8	0.36
SPELL_WORK_m24	29.5	21.0	20.3	21.0	1.8	0.35
SPELL_HIST_IS	7.3	8.0	8.5	8.0	-1.8	0.4
CHAR_INT_YEAR_2020	13.6	11.7	12.3	11.7	-1.8	0.38
SPELL_LFW_m9	7.2	15.9	16.4	15.9	-1.7	0.46
SPELL_HIST_H	15.5	7.6	7.0	7.5	1.6	0.35
dfe_level_0_start	95.0	97.9	97.6	97.9	1.6	0.34
dfe_level_8_start	x	x	x	x	1.5	0.5
SPELL_LFW_m6	8.2	20.0	20.5	20.0	-1.5	0.52
dfe_level_1_start	93.6	97.4	97.0	97.4	1.5	0.35
SPELL_HIST_PIP	6.9	7.1	7.5	7.1	-1.5	0.48
CHAR_INT_YEAR_2017	24.6	26.2	26.8	26.1	-1.5	0.48
SPELL_WORK_m21	31.4	23.6	23.0	23.6	1.4	0.47
CHAR_INT_MONTH_2	6.7	8.2	7.8	8.2	1.3	0.53
SPELL_HIST_C	46.5	48.5	49.2	48.5	-1.3	0.52
DFE_CLA	2.8	3.8	4.1	3.8	-1.3	0.56
SPELL_WORK_m18	32.2	25.3	24.7	25.3	1.2	0.54
SPELL_OTHER_m24	58.1	64.3	64.9	64.3	-1.2	0.56
CHAR_INT_MONTH_11	7.6	7.8	8.1	7.8	-1.1	0.6
CHAR_INT_YEAR_2021	8.6	7.8	8.1	7.8	-1.1	0.59
SPELL_HIST_A	44.7	67.2	67.7	67.2	-1.1	0.58
CHAR_INT_MONTH_10	9.0	10.4	10.1	10.5	1.1	0.61
SPELL_LFW_m1	11.0	45.2	45.5	45.1	-1	0.68
SPELL_WORK_m1	29.2	22.7	22.3	22.7	0.9	0.63
CHAR_INT_YEAR_2018	12.0	13.6	13.3	13.6	0.9	0.66
SPELL_OTHER_m21	55.2	61.0	61.4	61.0	-0.9	0.65
CHAR_INT_MONTH_3	6.1	7.6	7.9	7.7	-0.9	0.68
CHAR_INT_YEAR_2016	20.8	20.4	20.1	20.4	0.9	0.66
SPELL_Inactive_m1	16.1	13.0	13.3	13.0	-0.9	0.66
SPELL_Inactive_m24	9.5	8.3	8.5	8.3	-0.8	0.67
SPELL_HIST_CTC	8.1	6.8	7.0	6.8	-0.8	0.68
SPELL_HIST_ICA	2.9	2.2	2.3	2.2	-0.8	0.68
CHAR_INT_YEAR_2015	x	x	x	x	0.8	0.68
DFE_ETHNICITY_WHITE	86.8	96.2	96.0	96.2	0.7	0.61
SPELL_Inactive_m9	13.8	12.3	12.6	12.3	-0.7	0.71
SPELL_Inactive_m18	11.1	9.5	9.7	9.5	-0.7	0.71
SPELL_OTHER_m3	45.8	37.8	37.5	37.8	0.7	0.72
SPELL_Inactive_m12	13.0	11.5	11.8	11.5	-0.7	0.73
SPELL_Inactive_m3	15.4	13.2	13.4	13.1	-0.7	0.73
SPELL_OTHER_m2	45.7	34.8	34.5	34.8	0.7	0.74
CHAR_INT_MONTH_4	5.1	6.4	6.3	6.4	0.6	0.77
DFE_SEN	33.1	45.2	45.4	45.1	-0.6	0.76
CHAR_INT_MONTH_5	5.1	6.5	6.6	6.5	-0.6	0.77
DFE_ETHNICITY_MISSING	7.1	2.9	3.0	2.9	-0.6	0.69
DFE_C_start	13.4	18.8	18.6	18.8	0.6	0.77
SPELL_LFW_m24	4.4	9.0	8.8	9.0	0.6	0.8
SPELL_HIST_CHB_PARENT	9.7	7.2	7.3	7.2	-0.6	0.76
SPELL_HIST_WTC	2.6	2.1	2.2	2.1	-0.5	0.78
SPELL_Inactive_m21	10.3	9.0	9.2	9.0	-0.5	0.79
SPELL_Inactive_m6	14.6	12.9	13.1	12.9	-0.5	0.79
CHAR_INT_MONTH_12	3.8	4.3	4.2	4.3	0.5	0.8
CHAR_INT_MONTH_9	12.0	11.6	11.8	11.6	-0.5	0.8
CHAR_INT_MONTH_7	15.9	9.9	10.0	9.9	-0.5	0.79
SPELL_LFW_m12	6.4	13.5	13.6	13.4	-0.5	0.84
CHAR_CHILDREN	10.0	7.9	8.1	7.9	-0.5	0.81
SPELL_OTHER_m18	53.2	57.5	57.8	57.5	-0.5	0.81
SPELL_OTHER_m9	47.3	47.7	47.5	47.7	0.4	0.83
SPELL_WORK_m15	32.9	26.6	26.4	26.6	0.4	0.82
SPELL_WORK_m9	33.9	28.8	28.5	28.7	0.4	0.83
DFE_ETHNICITY_OTHER	x	x	x	x	-0.4	0.64
SPELL_LFW_m15	5.8	11.8	11.8	11.7	-0.4	0.86
DFE_H_start	11.3	0.8	0.9	0.8	-0.4	0.61
DFE_ETHNICITY_ASIAN	x	x	x	x	-0.4	0.6
SPELL_Inactive_m2	15.7	13.1	13.1	13.0	-0.4	0.84
SPELL_WORK_m12	33.8	28.2	28.0	28.2	0.4	0.85
PIT_INT_HIST	2.0	4.9	4.8	4.9	0.4	0.88
SPELL_OTHER_m12	48.9	51.0	50.9	51.0	0.3	0.88
SPELL_HIST_ESA	9.7	11.5	11.5	11.4	-0.3	0.9
DFE_ETHNICITY_BLACK	x	x	x	x	0.3	0.78
SPELL_WORK_m3	31.3	28.0	27.8	27.9	0.2	0.9
CHAR_INT_MONTH_6	7.2	8.3	8.3	8.3	0.2	0.91
DFE_permanent	0.5	0.9	0.9	0.9	0.2	0.93
DFE_ETHNICITY_CHINESE	x	x	x	x	-0.2	0.88
SPELL_OTHER_m6	46.6	43.7	43.7	43.8	0.2	0.92
SPELL_HIST_UC	17.1	43.7	43.7	43.6	-0.2	0.93
CHAR_INT_MONTH_8	13.7	9.6	9.6	9.7	0.2	0.94
SPELL_LFW_m18	5.2	10.4	10.3	10.4	0.1	0.95
CHAR_INT_MONTH_1	7.7	9.2	9.2	9.2	-0.1	0.96
SPELL_WORK_m2	30.6	25.7	25.6	25.7	0.1	0.97
SPELL_Inactive_m15	12.0	10.6	10.7	10.7	-0.1	0.98
DFE_ETHNICITY_MIXED	1.9	0.6	0.6	0.6	0	0.98
SPELL_WORK_m6	32.8	29.3	29.2	29.2	0	0.99
SPELL_OTHER_m15	51.2	54.6	54.7	54.7	0	0.99
SPELL_OTHER_m1	46.0	28.7	28.7	28.7	0	0.99
SPELL_LFW_m21	4.7	9.6	9.5	9.6	0	1
CHAR_SEX	53.5	62.1	62.0	62.0	0	1
SPELL_HIST_BB	x	x	x	x	0	1
SPELL_HIST_BSP	x	x	x	x	0	1
SPELL_HIST_IB	x	x	x	x	0	1
SPELL_HIST_PIB	x	x	x	x	0	1
SPELL_HIST_SDA	x	x	x	x	0	1
SPELL_HIST_WB	x	x	x	x	0	1
PIT_REF_START	x	x	x	x	0	1

Note: Some figures which have been suppressed for disclosure control purposes are denoted by an x.

Note: The definition of the matching variables can be found in the methodology document.

Table 10: PSM summary statistics used to assess the success of the matching for the main analytical run

	Summary Statistics
Matching estimator	100 Nearest Neighbours
bandwidth/calliper	0.01
Rubin’s B	11.19
Rubin’s R	0.89
Max % bias	3.58%
Median % bias	0.80%
Number on support	4,868
Number off support	12
Percent off support	0.25%

Appendix D: Regional cluster analysis

For all of the analysis presented in this report the comparison pool was selected from outside the regions where the programme was implemented. As discussed in the methodology report, care must be taken when doing this as local factors such as the availability of employment and public transport, levels of disadvantage, etc. all have an impact on the likelihood of someone finding and maintaining employment.

This analysis took advantage of cluster analysis carried out within DWP that groups the Local Authorities of Great Britain into 14 groups based on a range of variables about key features of the local labour market. These include local employment rates, unemployment related benefit caseload, qualification levels, variables related to mental and physical health / disability characteristics of the local population.

This cluster information was used as a way of selecting the comparison pool from “similar” regions of Great Britain, and as a control variable in the propensity score matching.

Appendix E: Tables of results

Table 11: Showing the full list of results featuring participants who were between the ages of 18 and 24 at programme start

See Table 11 in Tables: DurhamWorks programme.

Note: Some figures which have been suppressed for disclosure control purposes.

Table 12: Showing the full list of results featuring participants who were between the ages of 18 and 24 at programme start and started before 1 April 2021

See Table 12 in Tables: DurhamWorks programme.

Note: Some figures which have been suppressed for disclosure control purposes.

Appendix F: Additional subgroup analysis

This section provides additional information on other subgroups. Those subgroups include:

• Pre-COVID and COVID impacted participants
• Participants with and without SEN provisions
• Male and female participants

The programme overlaps the period of the COVID-19 pandemic. The pandemic and the associated lockdowns which took place between March 2020, and the summer of 2021 had significant impacts on the labour market in England, with large swings in employment, unemployment, and inactivity rates. See Coronavirus: Impact on the labour market (House of Commons Library briefing) for further details.

To explore the impacts of the pandemic on the participants and their outcomes, a sub-analysis was carried out that split the participants into two cohorts; one where the programme delivery and two-year outcomes were not impacted by the pandemic, and another where they were. The pre-COVID group was made up of participants who started the programme before 1 April 2018, and the COVID impacted group was made up of participants who started on or after 1 April 2018.

Table 13 presents the impact of the programme on the percentage of participants in employment two years after programme start for all evaluated participants as well as each of the above subgroups. Table 11 provides a full list of impacts for all evaluated participants. Table 14 through Table 19 provide a full list of results for each of the subgroups.

Table 13: Showing the percentage in employment two years after programme start for all participants between the ages of 18 and 24 alongside a number of subgroups.

	Number of Observations	Participant group (%)	Comparison group (5)	Impact: Central (ppt)	Impact: Lower (ppt)	Impact: Upper (ppt)	Statistically significant
All Participants	4,868	56	50	5	4	7	yes
Pre-COVID	2,539	55	50	4	2	6	yes
COVID impacted	2,327	57	49	8	5	10	yes
SEN	2,193	44	41	2	0	5	no*
Non-SEN	2,667	65	57	8	6	10	yes
Female	1,842	53	46	6	4	9	yes
Male	3,020	57	52	5	3	7	yes

*This result is statistically significant at the 90% level of confidence

Table 14: Showing the full list of results featuring a subset of participants who were between the ages of 18 and 24 at programme start and started before 1 April 2018

See Table 14 in Tables: DurhamWorks programme.

Note: Some figures which have been suppressed for disclosure control purposes.

Table 15: Showing the full list of results featuring a subset of participants who were between the ages of 18 and 24 at programme start and started on or after 1 April 2018

See Table 15 in Tables: DurhamWorks programme.

Note: Some figures which have been suppressed for disclosure control purposes.

Table 16: Showing the full list of results featuring a subset of participants who were between the ages of 18 and 24 at programme start and had a SEN provision at any point between the ages of 14 and 18

See Table 16 in Tables: DurhamWorks programme.

Note: Some figures which have been suppressed for disclosure control purposes.

Table 17: Showing the full list of results featuring a subset of participants who were between the ages of 18 and 24 at programme start and did not have a SEN provision at any point between the ages of 14 and 18

See Table 17 in Tables: DurhamWorks programme.

Note: Some figures which have been suppressed for disclosure control purposes.

Table 18: Showing the full list of results featuring a subset of participants who were female and between the ages of 18 and 24 at programme start

See Table 18 in Tables: DurhamWorks programme.

Note: Some figures which have been suppressed for disclosure control purposes.

Table 19: Showing the full list of results featuring a subset of participants who were male and between the ages of 18 and 24 at programme start

See Table 19 in Tables: DurhamWorks programme.

Note: Some figures which have been suppressed for disclosure control purposes.

Appendix G: Glossary of Terms

Term	Definition
Care experienced	Refers to individuals who have spent any amount of time in the care system at any point.
Common support/ On Support/ Off support	Once propensity scores have been assigned for each observation, the overlap of propensity scores between the participants and comparison group is called ‘common support’. Those who fall in the overlap are referred to as ‘on support’, those who do not fall into the overlap are ‘off support’.
Comparison group	Carefully selected subset of the comparison pool, selected to have outcomes as similar as possible, to act as a counterfactual.
CIN	Child in Need
DfE	Department for Education
DLA	Disability Living Allowance
DWP	The Department for Work and Pensions
ESA	Employment and Support Allowance
ESF	European Social Fund
FSM	Free School Meals
JSA	Jobseeker’s Allowance
NEET	Not in Employment, Education or Training
NUTS	Nomenclature of Territorial Units for Statistics
Participant group	The people who took part in the programme being evaluated.
PIP	Personal Independence Payment
Programme	The employment support provision under investigation.
Pseudo-start date	Dates assigned to the comparison pool in lieu of the real programme start dates of the participant group.
PSM	Propensity Score Matching
Quasi-Experimental	An experimental technique that looks to establish a cause and effect relationship between two variables, where the assignment to the participant or comparison group is not random.
Rubin’s B & R	Tests used to evaluate the matching in PSM
SEN	Special Educational Needs
SME	Small and medium enterprises
Statistically significant	Describes a result where the likelihood of observing that result by chance, where there is no genuine underlying difference, is less than a set threshold. In the Data Lab reports, this is set at 5 per cent.
UC	Universal Credit
User Organisation	The organisation using the employment data lab service.

1. For more information see European Social Fund 2014 to 2020 programme: 2023 booklet - GOV.UK ↩

2. For more information see What qualification levels mean: England, Wales and Northern Ireland - GOV.UK ↩

3. Analysis of longer-term outcomes suggests employment and sustained employment impacts persist up to four years after programme start. ↩