Our innovations have been assessed against Nesta's Standards of Evidence framework, which has five levels:

Level 1

Level 1 is the minimum Standard of Evidence that we required grantees to have in order to be funded by the programme. At this level organisations can give an account of their impact. By this we mean providing a logical reason, or set of reasons, for why their intervention could have an impact and why that would be an improvement on the current situation.

At the most basic level, organisations can do this themselves by producing a convincing theory of change, and drawing upon existing data and resources from other sources. Some of the more advanced Level 1 evaluations include data captured from beneficiaries. However, this is only data derived from subjective questions captured after the intervention (e.g. did our intervention make you feel less isolated?)

The Level 1 evaluations currently published by the Fund are:

• Bedfordshire Rural Communities Charity
• Get Connected
• North London Cares
• Spice
• Task Squad
• Team Up
• TLG
• British Lung Foundation
• Stroke Association
• Carers UK

Level 2

Level 2 evaluations has data that shows some change amongst those receiving or using an intervention. At this stage, data can begin to show effect but it will not evidence direct causality. Some of the most common methods to gather this data include: pre and post-survey evaluation; cohort/panel study, regular interval surveying.

The Level 2 evaluations currently published by the fund are:

• Body & Soul
• British Red Cross
• d2 Digital
• GoodGym
• Green Gym by the Conservation Volunteers
• London Ambulance Service
• MySupportBroker
• Parents 1st
• The Royal National Institute of the Blind (RNIB)
• Smart Works
• StreetDoctors
• Vi-Ability
• VIY by Cospa
• uk active
• User Voice

Level 3

Level 3 evaluations can demonstrate that an organisation’s intervention is causing the impact, by showing less impact amongst those who don’t receive the product or service.

To reach this level, evaluations need robust methods that use a control group (or another well justified method) that begin to isolate the impact of the product or service. Random selection of participants strengthens evidence at this level, and there needs to be a sufficiently large sample at hand (scale is important in this case).

The Level 3 evaluations currently published by the fund are:

• Action Tutoring
• The Access Project

Level 4

Evidence that reaches the fourth level of the Standards of Evidence allows organisations to explain why and how their intervention is having the impact they have observed and evidence.

An independent evaluation validates the impact. In addition, the intervention can deliver impact at a reasonable cost, suggesting that it could be replicated and purchased in multiple locations. The kinds of methods deployed at this level include multiple replication evaluations, future scenario analysis and fidelity evaluation.

This is a very advanced form of evaluation. None of the innovations in this fund have reached this level yet.

Level 5

This is the most advanced level of evidence in the Standards. To reach this level, organisations need to be able to show that their intervention could be operated by someone else, somewhere else and scaled up, whilst continuing to have a positive and direct impact on the outcome. And whilst remaining a financially viable proposition.

Like Level 4, the kinds of methods deployed at this level include multiple replication evaluations, future scenario analysis and fidelity evaluation.

None of the innovations in this fund have reached this level yet.