Perceived Safety and Dignity
English: % of the target group members reporting a feeling of safety and dignity regarding the protection risks that were identified and which the project deals with
French: % des membres du groupe cible faisant état d’un sentiment de sécurité et dignité en relation aux risques de protection qui ont été identifiés et dont le projet traite
Portuguese: % de membros do grupo-alvo que relatam um sentimento de segurança e dignidade em relação aos riscos de protecção que foram identificados e com os quais o projecto lida
Czech: % členů cílové skupiny uvádějící pocit bezpečí a důstojnosti ve vztahu k identifikovaným a řešeným rizikům
What is its purpose?
In the emergency contexts, humanitarian interventions are frequently responsible for identifying and mitigating various risks that threaten people’s safety and dignity. This indicator therefore measures the percentage of the target group members who reported a feeling of safety and dignity regarding the key risks they identified and which the project decided to address.
How to Collect and Analyse the Required Data
The indicator’s value can be determined by using the following methodology:
1) If you have not done so earlier, at the beginning of the project, conduct a participatory protection risk analysis that is specific to the area you operate in and identifies i) who is at risk, ii) from what/whom, and iii) why. You can take advantage of the guidance provided by the Protection Mainstreaming Toolkit (see below, page 28) and the already conducted analyses (e.g. regional or national analysis conducted by the cluster). The definition of the main risks must be based primarily on the perceptions and experiences of the target group members. Ensure that the analysis does not focus only on factors that threaten people’s safety but also includes those that negatively affect their dignity. See the following examples of three safety and three dignity issues:
- Girls and women in the camps feel uncomfortable about going to bathrooms after dark.
- Children are unsupervised because their parents are looking for income generating activities.
- Boys and girls are at risk of being forcibly recruited by armed groups.
- People have to wait for distributed aid in the sun on a hot day.
- Victims of gender-based violence (GBV) receive health assistance without psychosocial counselling.
- People receive food assistance instead of cash where markets are still functioning and people want to and are able to shop.
2) Select the risks your intervention can and intends to address in the course of its duration (for example, while you likely cannot control the movement of armed groups, you can decide on the place of your distributions). Be very specific – for example, instead of measuring the risk of GBV in general, focus specifically on the risk of GBV when women move around a camp / visit the bathrooms. Choose a limited number (3 - 4) of risks that you realistically can and will measure.
3) Define the interview questions required for measuring how big an impact the selected risks have on the target group members’ feeling of safety or dignity. Your focus must be on how safe and dignified people themselves feel – not on how someone else thinks they feel. Consider asking them to show you how safe / dignified they feel on a visual scale whose meaning the enumerators explain in advance (see example below; the scale should have several options ranging from feeling “very safe/dignified” to “very unsafe/undignified”). Examples of such interview questions might include: Can you please show me on this scale how safe you feel when going after dark to the bathrooms?
4) Incorporate the questions into your questionnaire and conduct individual interviews with a representative sample of the target group members who face the risks your survey measures. If you assess different risks faced by different groups of people (e.g. women facing certain risks and men facing other risks), you will have to have separate samples for each group (this takes time, so consider focusing on one group only).
5) To calculate the indicator’s value, divide the number of respondents who replied to all questions as feeling “very” or “rather” safe/dignified by the total number of respondents. Multiply the result by 100. If you used two (or more) samples, divide the number of respondents from both samples who replied to all questions as feeling “very” or “rather” safe/dignified by the total number of respondents from both samples. For example 220 women + 180 men = 400 people feeling “very” or “rather” safe/dignified, divided by a total sample of 500 respondents equals 0.8, meaning 80% of the total target population.
Disaggregate the data by gender, age, ethnicity, persons with disability and other categories relevant to the local context.
1) DG ECHO recommends using a similar version of this indicator: “% of the target group members reporting an improved feeling of safety and dignity by the end of the intervention compared to at the beginning”. This indicator is also based on using the results of the protection risk analysis; however, since you have to compare people’s feelings at the time of the baseline and endline surveys, you either have to:
- collect baseline and endline data from exactly the same respondents which in many emergency contexts can be very difficult; or
- at the time of the endline survey, ask people to recall whether they currently feel safer that they felt when the project started. However, this approach involves a high risk of several biases and therefore is not recommended
For this reason, IndiKit proposed the above-described methodology that provides very similar data but does not require having the same respondents.
2) Use this indicator only when your project either applies a protection integration approach or properly mainstreams protection. When preparing the logical framework’s targets and assumptions always consider and acknowledge that external factors that are beyond your control might influence your ability to address the identified protection risks.
3) Be careful about the number and type of risks your intervention can realistically address. Many of them might be caused by factors you have only limited control over, so avoid being too ambitious and focus on measuring a limited number of risks your intervention can realistically mitigate against. In the case that you decide to measure a larger number of risks (e.g. more than 4), it might be too ambitious to expect people to feel “safe and dignified” about all the situations you assess. Therefore, consider defining “feeling safe and dignified” as a situation when people feel safe and dignified about a pre-defined minimum of the assessed situations/contexts (for example, at least 4 out of 5 measured situations/contexts). Otherwise, you might manage to mitigate against the vast majority of the measured risks but the indicator will still appear as if you did not succeed.
4) The risk analysis is likely to identify that different groups of people (e.g. women, young men, or children) face different risks. If you need to gain accurate data about their feelings of safety and dignity, you would have to conduct several separate surveys, each with a representative number of people from the given group. Since this would be quite time consuming, you might have to decide that:
- your representative survey will focus on assessing the feelings of one or two groups only (e.g. only women or women and their children); and
- the feelings of other groups (e.g. youth, men) would be assessed through qualitative, less representative methods, such as key informant interviews or well-facilitated focus group discussions - however, it will not be possible to use the qualitative data for calculating the value of this indicator (it will be used “only” for informing your programming and for qualitative reporting)
5) If you measure people’s feelings about several different risks, you might also want to report individually on the percentage of the target group members feeling safe/dignified with regard to each of these risks. For example:
- % of women feeling safe when going to the bathrooms
- % of people stating than receiving cash, as compared to receiving food, was a more dignified type of support
6) It is recommended that instead of asking people whether they feel “very safe”, “rather safe”, “rather unsafe” or “very unsafe”, you use a visual scale (see example below) where the enumerators ask people to indicate the symbol that best represents their feelings (the same also applies to measuring the feeling of dignity). It is important that you ensure that the data collectors explain the meaning of each face on the picture / scale in the same way, for example: "The happiest face means that you feel very safe. The face with the smaller smile means that you feel quite safe. The face ... ". Let each data collector practice and ask others to observe whether s/he explains the meaning of the faces correctly. Before you decide to use the scale, pre-test it extensively, so that you are sure that it is appropriate to the local context and people easily understanding the correct meaning of each symbol.
7) If you are able to use such data for your current or planned programming, also assess why some people did not feel safe or dignified and why it was difficult for some people to access to the provided assistance.
8) To be able to recognize whether any recorded positive changes in people’s feeling of safety and dignity can be attributed to your intervention, consider asking people who reported feeling safe whether there were any positive activities or measures implemented that made them to feel safer (train the enumerators on how to keep probing). Then calculate the percentage of such respondents who mentioned at least one of the protection measures that were supported by the intervention.
9) It is important that the enumerators are able to explain to the respondents (if required) what they mean by “feeling safe” and “feeling dignified”. This must be done in a clear and neutral way, so that it does not influence the respondents’ answer.
Access Additional Guidance
- Global Protection Cluster (2017) Protection Mainstreaming Toolkit (.pdf)