The CRiTERIA Human security Filter bridges a critical gap in the humanitarian dimension of border security. Existing risk assessment methodology is primarily focused on securing borders with little consideration for the well-being of migrants. However, the importance of identifying and addressing migrant-specific vulnerabilities and the potential of expanding the humanitarian impact to individual or group level to provide better first responses to the needs of migrants is increasingly recognized by leading organizations in the field. The CRiTERIA Human Security Filter brings the humanitarian dimension of security such as the well-being of migrants into focus. It serves as mainstreaming the human rights approach throughout the entire framework and guides border authorities in adequately assessing and addressing migrant-specific vulnerabilities.
The Human Security Filter as a Composite Indicator
The Human Security Filter (HSF) aims to measure a complex, multidimensional phenomenon, which cannot be measured directly. It aggregates several individual sub-indicators into a single composite measure, which is then seen as a reflection of the developments of the individual indicators.
Currently, the HSF compromises 29 sub-indicators grouped into three categories (individual/embodied factors, situational factors, and structural factors) whereas the overall filter corresponds to the weighted averages of the sub-indicators. The weights are estimated in a so-called path regression model which includes both direct and indirect effects and provides estimates of the direction of the subsequent effects of the sub-indicators and the magnitude of the combined direct and indirect effects of the sub-indicators on the three main categories and the overall composite indicator. The sub-indicators and the three corresponding categories have been defined based on a comprehensive review of the available models in the literature (such as the IOM vulnerability framework and different European and national legal provisions on vulnerability) and expert consultations with 52 experts from various humanitarian organizations working in the field of migration.
Defining Migrants’ Vulnerability in the Context of the Human Security Filter
To build the composite indicator for migrants’ vulnerability, we go beyond existing formal definitions of the concept of vulnerability and investigate how it is actually employed in practice and what are the different types of vulnerabilities. We observed that vulnerability is commonly characterized as the product of:
- innate/natural characteristics (innate vulnerability),
- past, present, or future situations and experiences (situational vulnerability), and
- structural characteristics and dynamics (structural vulnerability).
Considering all these three dimensions, we define migrants’ vulnerability as follows:
“migrants who are unable to effectively enjoy their human rights and who are at increased risk of violence and abuse as a results of innate characteristics, past, present, or future situations and experiences and/or structural characteristics and dynamics.”
The Sub-Indicators of the Human Security Filter
The figure on the right presents the set of sub-indicators used to build the CRITERIA Human Security Filter. It includes both input indicators (e.g., policies requiring detection on arrival) and output indicators (e.g., attitudes towards authorities). It also includes a combination of subjective indicators (e.g., attitudes towards health practitioners) and objective indicators (e.g., policies, gender, race).
The Human Security Filter describes each indicator including the corresponding scoring (the likelihood that the vulnerability exists in that particular group). The score is using a four-point scale to “force” an opinion and discourage analysis to overly on the neutral middle option: 1 – unlikely, 2 – possible, 3 – likely, and 4 – almost certain. The scores are built on existing research and statistics, but they need to be constantly reviewed and updated to remain relevant. While the scores are not comprehensive, they provide examples supported by literature and validated by expert in the field which should be completed with further information and evidence.
Example Scoring: I1. Age
Age as an indicator for migrants’ vulnerability refers to both minors and elderly migrants. There is ample evidence that minors face greater risks of sexual exploitation and abuse, military recruitment, child labor, and detention among others. In many countries, both minors and elderly are routinely denied entry or detained by border or immigration officials. In other cases, they are admitted but are denied access to asylum procedures, or their needs are not handled in an age-sensitive manner (e.g., additional health assistance may be required in the case of elderly migrants or specific mental/physical health assistance required for children, especially those who have experienced distinct types of psychological and physical violence such as child soldiers or child brides).
| Score 1 – unlikely: | None of the below |
| Score 2 – possible: |
|
| Score 3 – likely: | Country of origin: Afghanistan, Syria, Somalia |
| Score 4 – almost certain: | Multimedia analysis detects minors/elderly people |
Handling Missing Data
In general, a composite indicator should be robust to incomplete data or other data issues as it is not always possible to obtain data for individual sub-indicators; this should be ideally estimated by imputation. The four-point scale starts with 1 – unlikely representing a situation where there is not sufficient information in the data collection to support a strong decision. To compensate for this shortcoming, analysts could also employ a combination of intelligence (e.g., OSINT, HUMINT, SIGINT) in addition to resorting estimations. It is also important to remember that potential outliers should be documented and threated separately since they could represent the most vulnerable migrants (e.g., belonging to very marginalized groups, lacking access to communication technologies or other type of critical resources).
Indicator Correlation
Following the expert consultations, a number of correlations between the indicators have been identified.
Example Indicator Correlations: I1. Age
| Correlations | Scoring |
|---|---|
| I6. Mental health: minors and elderly are more likely to experience mental health issued due to abuse/separation of families | Equal scoring (age is highly correlated with mental health problems) |
| I7. Physical health: minors and elderly are more likely to experience physical health issues, lack of healthcare or lack of adequate food/shelter and other resources | Equal scoring (age is highly correlated with physical health problems) |
| I12. Exposure to violence: minors and elderly are likely to experience a high exposure to violence due to abuse or lack of adequate facilities | Equal scoring (age is highly correlated with exposure to violence) |
| I13. Availability of food: minors and elderly are less likely to have access to adequate food | +2 (correlation is likely but not certain) |
| I14. Availability of water: minors and elderly are less likely to have access to adequate water resources | +2 (correlation is likely but not certain) |
| I16. Availability of shelter: minors and elderly are less likely to have access adequate shelter | +2 (correlation is likely but not certain) |
| I19. Knowledge of legal rights: minors are less likely to possess a good knowledge of their rights | +1 (correlation is likely but not sufficient evidence) |
| I21. Availability of material resources: minors and elderly are less likely to have the necessary material resources | +2 (correlation is likely but not certain) |
| I27. Policies aimed at separating families: minors are likely to be directly impacted by such policies | Equal scoring (age is highly correlated with this indicator) |
Aggregation of Sub-Indicators in the Human Security Filter
In the case that the Human Security Filter is applied manually, the analysts can use a weighted mean of the three categories. Each of the three categories (dimensions, di) – innate, situational, and structural indicators – should be weighted accordingly to reflect the high variance in the number of sub-indicators and the partially correlation between some of the indicators (e.g., d1: 30%, d2: 25%, and d3: 45%) .
The design of the Human Security Filter takes into consideration the existing level of knowledge and data access in the field of migration. However, the composite indicator must be continuously monitored and adjusted over time in line with both seasonal developments and phenomenon changes. As the definition of indicators may change in relation to the route employed by migrants (e.g., what exposure to violence means), it is important to periodically review the composite indicators to consider vulnerability factors that may have emerged, evolved, or disappeared.
Aitana Radu
Dr. Aitana Radu is the Security Research Coordinator within the Department of Information Policy & Governance. Her research focuses on different aspects of security science, from violent radicalisation to intelligence oversight. Since 2013, Dr Radu has carried out extensive EU-funded research focused on radicalization, law enforcement practices, the implementation of the European Investigation Order, developing security science (ESSENTIAL), fake news (DOMINOES) and intelligence analysis in the contxt of border security (MIRROR and CRITERIA projects). Dr. Radu obtained her M.A. (Comparative Political Science) from the University of Bucharest with a thesis on democratic transitions in the Middle East, her M.A. in the Management of Intelligence Activities for National Security from the National Intelligence Academy with a thesis on the security risks posed by the radical Islamic discourse, and her PhD in Intelligence and National Security from the National Intelligence Academy with a thesis on the transformation of intelligence organizations.
