Gender-sensitive indicators for early warning of violence and conflict

Part V

New and Modified Gender-Sensitive Early Warning Indicators

Gender-Sensitive Indicators for Early Warning of Violence and Conflict

Incidents involving women, men, boys, and girls that may be overlooked in existing early warning systems but could signal potential violence or conflict. IFES’ literature review found that, where gender was included in early warning systems, it was primarily focused on measuring gender-based violence. This approach led to the consideration of gender only as an outcome (dependent variable) of violence rather than a predictor of violence (independent variable). It is certainly important to monitor increases in gender-based violence (including conflict-related sexual violence), and this is included in this framework’s list of indicators; however, omitting other gender indicators risks discounting key factors that could enhance the prediction of violence. Attempts to restrict or rescind women’s political, social, economic, and legal rights should also be seen as potential precursors to violence and conflict, as should any public narrative aimed at cementing narrowly defined gender roles, as these point to an increase in exclusionary or discriminatory practices. It is equally important that gender-sensitive indicators capture changes in behavior among or incidents involving men and boys as gender norms are also driving their behavior and can contribute in some instances to young men joining armed groups⁹ or being persecuted for defying traditionally prescribed gender roles based on societal definitions of masculinity.¹⁰

It is important to distinguish between the dynamic indicators suggested in this framework and more structural gender inequality indicators such as rates of women’s political and workforce participation, literacy and maternal mortality. Structural indicators are important to monitor as part of longer-term conflict prevention, stabilization, and sustainable development efforts. These structural indicators tend to include more readily available (although often time-lagged) associated public data sets on a global level. However, unless there are extreme sudden changes in these structural indicators, they do not provide early warning analysts with sufficiently rapid-changing data that could signal an imminent outbreak or escalation of violence or conflict. For example, a national structural indicator such as “number of women in parliament” might only change whenever there is an election (generally every three to five years). Monitoring this structural indicator would not give early warning analysts who monitor daily, weekly, or monthly changes in an environment relevant information for their analysis. Indicators that monitor longer-term structural measures are still critical; in this framework, however, we have chosen to focus on more dynamic indicators. While the indicators in this framework focus on changes in behavior and incidents that might indicate upcoming violence or conflict, indicators that measure changing attitudes on gender equality and gender norms, when available, can serve as a bridge between dynamic indicators – which measure shorter-term changes – and more long-term structural indicators.

As there are different types of early warning systems with varying geographic scopes, purposes, and data collection methodologies, the proposed gender-sensitive indicators may need to be adapted to ensure their compatibility with specific early warning systems (acknowledging that not all the indicators will be suitable in every context). They have been drafted to maximize applicability to different early warning systems and violence monitoring efforts across varying regional and country contexts. There may be additional indicators beyond this list that reflect more context-specific metrics of changing attitudes, growing hostilities and volatility, or increased community radicalization or support for armed groups, which can be monitored through tracking behavior changes or incidents based on gender norms. Implementors can ensure the context-specific adaption of these indicators, their definitions, and data collection strategies by holding consultations with local women and civil society groups in advance of the monitoring period to identify how gender behaviors and incidents could be monitored in a safe and efficient way by local monitors or through existing monitoring efforts.

As well as having different geographic scopes, time horizons, and data collection strategies, early warning systems can have different categorizations of indicators. The grouping of these indicators may vary but tend to be categorized as: security and justice; politics and governance; human rights; health; and economic factors. For gender to be comprehensively integrated in early warning systems, at least one gender-sensitive indicator must be included in each of these categories, and these gender-sensitive indicators should be analyzed contemporaneously. It is unlikely that variances in only one gender-sensitive indicator would sufficiently point to a changing security environment within a community. However, if noticeable changes appeared across several of the gender-sensitive indicators – such as a rapid decrease in women’s freedom of mobility, a sudden decrease in their presence in common places, and their increased efforts to acquire cash – these sudden changes could be indicative of growing insecurity or impending violence or conflict and highlight the need for more careful analysis. Another example of a sudden change could be the closing of factories, which could lead to a sudden spike in male youth unemployment, then an increase in gatherings of young men at religious sites and an upsurge in hypermasculine propaganda. Changes in these gender behaviors and incidents involving women, men, girls, and boys could signal an impending uptick in violence or conflict.

The table below includes the new or modified gender-sensitive indicators, rationale for inclusion in this framework, considerations to inform development of indicator parameters for specific contexts, and suggested data collection strategies.
A step-by-step guide on how to integrate these indicators in early warning systems and violence monitoring efforts is provided in Section VII; however, when considering the table below, it is important to highlight that:

The indicators should be tailored to the specific context and existing early warning system or violence monitoring effort prior to integration.
The definitions and data collection strategies should be adapted to reflect the local context and supplemented with additional context-specific information to ensure indicators, definitions, and data collection strategies are feasible and contextually appropriate.
The indicator table presents a menu of indicators so that implementors can choose ones most relevant to their local context. The expectation is not that every single indicator mentioned here will be integrated in an early warning system or monitoring effort.
Baseline data needs to be recorded for each of the gender-sensitive indicators at the beginning of the monitoring period to provide analysts with comparative data.
This table focuses on gender-sensitive indicators to monitor and does not make suggestions about responses to changes in that indicator. Existing early warning systems or violence monitoring efforts often already have specified actions to undertake in response to a sudden change in risk.
Indicators marked with an asterisk (*) in the table were tested as part of the Nigeria pilot – and include more detail than others as a result. Future pilots and other efforts to integrate indicators beyond those piloted in Nigeria during this project will help provide additional lessons learned.

Gender-Sensitive Early Warning Indicators

While endnotes for the global framework can be found at the end of this document, endnotes for this table are found on page 28.

CATEGORY 1: Security and Justice

Indicator	Rationale for Indicator	Definition Considerations	Possible Data Collection Strategy
Percentage of individuals who are women present in designated common places*	To monitor changes in women and girlsâ mobility and movement. A sudden change, particularly a sudden decrease, in women appearing in common areas they usually frequent could signal a fear of imminent violence, conflict or insecurity.	Implementors will need to define âdesignated common places.â Considerations and suggested definitions to tailor to the local context: "Designated common places" refers to locations in target states that are highly frequented by the public, such as markets, places of worship, locations for political events, public transportation, eateries, or places of leisure (e.g., parks or pools). These locations should be designated at the beginning of the monitoring period and should not change. For each location, the numerator will refer to the number of women present. The denominator will be the total number of people present at that location, generally ascertained by a rough count by the monitors. It is important to ensure the common place being monitored reflects local customs and norms. For example, in northern Nigeria, it is not the custom for women to go to markets, so monitoring their presence there would be ineffective.	It is critical for monitors to visit common places for monitoring on the same day of the week and at the same time to keep external variables as consistent as possible. In the reporting form, monitors should be requested to provide the following information: Location Type (e.g., market, townhall meeting, place of worship, political rally/event, other) Number of people (e.g., 0-25; 25-50; 50-100; more than 100) Percentage of women present out of total number of people (e.g., 0-25%; 25-50%; 50-75%; mostly or all women) Change in % from previous week (e.g., about the same; fewer women than previously; more women than previously; unknown) Reasons for changes to number of women (open-ended) Total number of people present (open-ended)
Number of male gatherings and presence of non-local men	Increases in all-male gatherings and unusual movements of all-male groups are possible signs of community unrest or mobilization for the purpose of committing violence or causing conflict. In some contexts, there may be increased presence of men in a community ahead of a rally or campaign event during an election period. Supporters of some political parties may resort to violence and intimidation to show dominance, strength and superiority, as well as to intimidate potential voters supporting an opposing party. Instances of male groups descending on a community for violent purposes frequently occur around the world. For example, local women told the authors of this framework that they did not recognize the majority of hardliner Sinhalese men who were responsible for the March 2018 violence against Muslims in central Sri Lanka. It was suspected that the violence had been pre-organized with members of the group mobilizing on social media and travelling to the town for violent purposes.	Implementors will need to define âmale gatheringsâ and ânon-local men.â Considerations and suggested definitions to tailor to the local context: Unusual gathering sizes may differ depending on the context; for example, sporting events often expectedly draw larger crowds of men, so the emphasis should be on monitoring gatherings which are out of the ordinary for the time and day that the gathering is observed. Non-local men can include anyone suspected of not being a local resident. This, of course, is easier to ascertain in smaller towns that it is in urban settings; however, even in urban settings an influx of men from different areas may be noticeable.	Incident reporting forms should include estimated number of gatherings; estimated age range of those gathering; and number of men suspected of not being local residents of that area per gathering. Primary sources could include observation, eyewitness accounts or reports from a community reporting phone line; secondary sources could include weekly police reports and media reports. Some gatherings that do not lead to violence or lead to only minor scuffles may not be reported officially, but should be included where there is sufficient anecdotal information available or where the incident was observed by a local monitor or another eyewitness.
TBD	IFES will fill the rest of the table out

Indicator

Rationale for
Indicator

Definition
Considerations

Possible Data
Collection Strategy

Percentage of individuals who
are women present
in designated common places*

To monitor changes in women and girlsâ mobility and movement. A sudden change, particularly a sudden decrease, in women appearing in common areas they usually frequent could signal a fear of imminent violence, conflict or insecurity.

Implementors will need to define âdesignated common places.â

Considerations and suggested definitions to tailor to the
local context:

"Designated common places" refers to locations in target states that are highly frequented by the public, such as markets, places of worship, locations for political events, public transportation, eateries, or places of leisure (e.g., parks or pools).

These locations should be designated at the beginning of the monitoring period and should not change. For each location, the numerator will refer to the number of women present. The denominator will be the total number of people present at that location, generally ascertained by a rough count by the monitors. It is important to ensure the common place being monitored reflects local customs and norms. For example, in northern Nigeria, it is not the custom for women to go to markets, so monitoring their presence
there would be ineffective.

It is critical for monitors to visit common places for monitoring on the same day of the week and at the same time to keep external variables as consistent as possible.
In the reporting form, monitors should be requested to provide the following information:
Location Type (e.g., market, townhall meeting, place of worship, political rally/event, other)

Number of people (e.g., 0-25; 25-50; 50-100; more than 100)
Percentage of women present out of total number of people (e.g., 0-25%; 25-50%; 50-75%; mostly or all women)
Change in % from previous week (e.g., about the same; fewer women than previously; more women than previously; unknown)
Reasons for changes to number of women (open-ended)
Total number of people present (open-ended)

Number
of male
gatherings
and presence
of non-local
men

Increases in all-male gatherings
and unusual movements of
all-male groups are possible
signs of community unrest or
mobilization for the purpose of
committing violence or causing
conflict. In some contexts, there
may be increased presence of
men in a community ahead of a
rally or campaign event during an election period.

Supporters of some political
parties may resort to violence and intimidation to show dominance, strength and superiority, as well
as to intimidate potential voters
supporting an opposing party.

Instances of male groups
descending on a community
for violent purposes frequently
occur around the world. For
example, local women told
the authors of this framework
that they did not recognize the
majority of hardliner Sinhalese
men who were responsible for
the March 2018 violence against Muslims in central Sri Lanka. It was suspected that the violence had been pre-organized with members of the group mobilizing on social media and travelling to the town for violent purposes.

Implementors will need to
define âmale gatheringsâ and
ânon-local men.â

Considerations and suggested
definitions to tailor to the local
context:

Unusual gathering sizes may
differ depending on the context;
for example, sporting events often
expectedly draw larger crowds of
men, so the emphasis should be
on monitoring gatherings which
are out of the ordinary for the
time and day that the gathering is
observed.

Non-local men can include
anyone suspected of not being
a local resident. This, of course,
is easier to ascertain in smaller
towns that it is in urban settings;
however, even in urban settings
an influx of men from different
areas may be noticeable.

Incident reporting forms should
include estimated number of
gatherings; estimated age range
of those gathering; and number
of men suspected of not being
local residents of that area
per gathering.

Primary sources could include
observation, eyewitness accounts
or reports from a community
reporting phone line; secondary
sources could include weekly
police reports and media reports.
Some gatherings that do not
lead to violence or lead to
only minor scuffles may not be
reported officially, but should be
included where there is sufficient
anecdotal information available
or where the incident was
observed by a local monitor
or another eyewitness.

TBD

IFES will fill the rest of the table out

CATEGORY 2: Social Norms and Gender Equality

TBD

CATEGORY 3: Politics and Governance

TBD

CATEGORY 4: Economy

TBD

Table Endnotes

UN News, “Belarus Targets Women Human Rights Defenders After Disputed August Elections”, November 3, 2020. Available at: https://news.un.org/en/story/2020/11/1076822

Julia Steers, “Women Say They Are Being Raped as Part of the President’s Fight to Keep Power”, Time, January 14, 2016. Available at: https://time.com/4179101/rape-burundi/https://time.com/4179101/rape-burundi/

TBD - IFES team to add the rest

Download Report

ifes gender-sensitive indicators for early warning of violence and conflict a global framework may 2021.pdf PDF

2.67 MB