Brief resilience and coping scale: Assesses resilience and coping strategies. Designed originally in the health and psychology fields to measure individuals’ tendencies to cope with, and recover from, stress in an adaptive manner. It involves self-assessment on statements on behaviour and actions. For example, “I look for creative ways to alter difficult situations”, or “Regardless of what happens to me, I believe I can control my reaction to it.”

Useful in assessing resilience. Refined measures are designed to measure individuals’ tendencies to cope with stress in a highly adaptive manner; particularly useful for assessing resilience of people affected by trauma or shock.

Often not adapted to context, needs tailoring to specific context and group needs and experience.

Perceptions survey: A survey used when trying to find out how people understand or feel about their situations, institutions or services. They can be used to assess needs, establish baselines, analyse trends, and inform design.

Perceptions of individuals and communities related to key VE vulnerability or resilience factors, such as attitudes towards state institutions or sense of grievance, can uncover people’s attitudes and beliefs related to specific VE drivers, and can show potential change if repeated at a later time.

Responses are subjective based on respondents’ individual understanding at that specific time; the responses can be influenced by circumstances on the day.

Knowledge, Attitudes and Practice (KAP) survey: Quantitative method (predefined questions formatted in standardised questionnaires) that provides access to quantitative and qualitative information used to measure the extent of a known situation, confirm or disprove a hypothesis, enhance understanding of particular themes, identify what is known and done about various subjects. A KAP survey records what was said, the respondents’ opinions (rather than actions), and is based on declarative statements.

KAP surveys reveal perceptions or misperceptions, as well as potential barriers to behaviour change. In the PVE context, KAP surveys can be used for assessing perceptions of violence, religion, exclusion, etc. They provide quantitative measures which can be tracked as indicators of change. They can help measure the effectiveness of activities to change behaviours. They can also be used to identify potential intervention strategies and activities that reflect local circumstances and cultural factors.

Training in surveying is necessary. Costs of employing a survey team and analysing can be high. KAP surveys record opinion, what was said by the respondent, and may not reflect actions or behaviour. This carries a risk of bias, as participants may not give honest answers to sensitive PVE questions or may not participate (especially written forms). Data needs to be triangulated and contextualised with qualitative methods (e.g. FGD or KII).

Grievance, activism, and radicalism scales: Aims to test the relationship between grievance, activism and radicalism through questions about past and future intention to engage in activism-radicalism and questions about grievance (e.g. against a government or a particular group). This has been used to assess whether past grievance, activism and radicalism can be a predictor for future.

Intention to engage in activist activity in the future has been found to be correlated with future radical intention. Assesses personal or group grievance, which in some cases (not all) can be predictive of both past activism-radicalism and of future activism.

Often not adapted to context, needs tailoring to specific context and group needs and experience.

Most significant change (MSC): A participative approach developed for complex evaluations involving generating and analysing personal accounts of change and deciding which of these accounts is the most significant – and why. There are three steps:

1. Deciding the types of stories that should be collected.
2. Collecting the stories and determining which stories are the most significant.
3. Sharing the stories and discussion of values with stakeholders for learning.

MSC is particularly useful when you need different stakeholders to understand the different values that other stakeholders have in terms of “what success looks like” – criteria and standards for outcomes, processes and the distribution of costs and benefits. In a PVE context, this allows for a nuanced understanding of the VE dynamics from a community perspective.

MSC works best in combination with other data collection and analysis methods which capture broader social or structural change, as it is limited in collecting impact-level data. MSC requires access to communities to collect their change stories; when working remotely or with hard-to-reach populations in PVE this access may be hampered. Can be resource-intensive to gather.

Perceptions survey: A survey used when trying to find out how people understand or feel about their situations, institutions or services. They can be used to assess needs, establish baselines, analyse trends, and inform design.

Perceptions of individuals and communities related to key VE vulnerability or resilience factors, such as attitudes towards state institutions or sense of grievance, can uncover people’s attitudes and beliefs related to specific VE drivers, and can show potential change if repeated at a later time.

Responses are subjective based on respondents’ individual understanding at that specific time; the responses can be influenced by circumstances on the day.

Community score cards (CSCs): A quantative participatory tool most often used to solicit community members’ ‘perceptions on quality, efficiency and transparency’ of community service providers and their performance at the local level. Service providers may include implementers, donors, companies, or local institutions, such as police departments, government and district officials, and judiciary procedures. CSCs provide a mechanism for actors associated with an intervention to receive feedback on their behaviour, attitude or conduct.

This tool can promote and empower local perspectives through enabling communities to comment directly on services. Useful for assessing sensitivities to local dynamics, gathering perspectives within the community and identifying changes towards particular groups. For example, CSCs could reveal that services are only being provided in one part of a community, thereby exacerbating local tensions. Can be useful for remote monitoring, for example, the results of CSCs can be used to ensure that programmes are being implemented as intended by directly consulting the beneficiaries.

A tool best suited to the local level. It is not recommended at the national or macro-scale due to the degree of facilitation, mobilisation, follow-up, and analysis required.

Experimental design, or randomised Control Trial: RCT or randomised impact evaluation designs assign participants randomly to either intervention/treatment or comparison groups. Assigning people randomly (using specific methods of randomisation) to treatment or comparison minimises group differences that may bias results: the only difference between groups should be the intervention you wish to evaluate.

Minimises bias and isolates the impact of your intervention. Gives the best estimate of the counterfactual. Randomisation can be to different interventions, and/or intervention vs. no intervention. Randomisation can be at cluster level (such as school or district) if suitable for the programme.

Requires significant planning, expertise and can be costly. Must be designed into the programme from the start. RCTs do not allow for programme adaptation and limit flexibility, which is a challenge in conflict and fragile contexts. Random assignment of participants is often impractical within PVE contexts where it is hard to access target groups (e.g. issues of sufficient numbers or attrition). Large sample sizes can be needed to give statistical power and therefore reliable results. Implementers often question the ethics of assigning people to treatment not based on need but randomly, however there are ways to manage this concern (e.g. using ‘waitlist’ design where individuals assigned to the control group will participate in the programme at a later date).

Quasi-experimental design: Participants are assigned to treatment and comparison groups through a matched design rather than at random. Either individuals, communities or other units are matched on various factors to ensure that any attributable differences between the groups post-programme are due to the treatment and not another variable.

A well selected comparison group gives most of the advantages of an RCT. Additionally, it is possible to match people later in programme implementation, post-treatment assignment, increasing practicality.

Non-random assignment means that there may be bias built into group selection, for example, assignment by the time of day people visit a clinic may accidentally sample groups with different demographics or working patterns.

Non-experimental design: Does not use a comparison group, but takes a measurement before and after an intervention.

An effective way of measuring change that occurred during an intervention. Less complex than other approaches.

The lack of a control or comparison group makes it harder to detect if circumstances have affected results, any change is harder to attribute to your intervention.

Media content and discourse analysis: Monitoring and analysis of different media sources and outlets, looking at content, language, visuals and tone. Discourse analysis then examines the key assumptions underpinning the discourses that influence social constructions and interactions.

Helps measure the unmeasurable by analysing the relationship between discourse and societal effects. This could be suitable for programmes working with media or public figures and on counter- narratives.

Requires significant expertise to be able to undertake the analysis. Requires building up a picture of analysis over time to identify patterns. Data protection, online safety, and rights to expression. Distinction needs to be made between online/offline persona (whom people interact with online and in the‘real world’).

Contribution analysis: Assesses programmes’ achievement towards an outcome. Contribution analysis focuses on questions of ‘contribution’, specifically to what extent observed results (positive or negative) are the consequence of a programme. It involves 6 steps:

1. Set out the attribution problem to be assessed,
2. Develop a ToC/logic model,
3. Populate the model with existing data and evidence,
4. Assemble and assess the ‘performance story’,
5. Seek out additional evidence,
6. Revise the ‘performance story’.

Contribution analysis is particularly useful for situations where designing an ‘experiment’ to test cause and effect is impractical, which is likely in PVE contexts. Contribution analysis is also useful when the programme has been funded on the basis of a relatively clearly articulated ToC and where there is little or no scope for varying how the programme is implemented.

Contribution analysis requires experience and expertise, as well as resources to go through the 6-step process.

It does not uncover an implicit or inexplicit ToC. It does not provide definite proof but provides evidence for whether the programme has made an important contribution to the documented results.

Qualitative coding: Coding is a fundamental task in most qualitative projects – it involves gathering all the material about a particular theme or case into a node (branch/pattern/ theme) for further exploration.

Enhances qualitative analysis, allows for more nuanced and systematic analysis of qualitative data.

Requires time (transcribing, coding and analysing). Software, such as NVivo, reduce time and increase efficiency, but licences can be expensive.