AMAR Project

The original MAR dataset was created to identify communal (i.e., culturally-defined) groups "at risk" of human rights violations and/or of protracted conflict behavior.

When researchers generalized from findings derived from the MAR dataset to the larger population of ethnic groups, concerns arose about the effects of selection bias on the accumulation of knowledge about ethnic group behavior.

MAR ethnic groups were selected on criteria that are likely to be correlated with a propensity for conflict, since politically mobilized minorities and/or groups that are discriminated against tend to be more involved in violence than non-mobilized minorities. These criteria may result in a selection bias problem, and findings from the MAR data cannot necessarily be generalized to ethnic groups as a whole.

The AMAR project was designed to address this issue.

The AMAR Sample Frame

As introduced in Birnir et al. (2015), the AMAR sample frame is the first attempt at constructing a list of socially relevant ethnic groups that is not defined by any political criteria, such as being 'at risk', as in the original MAR dataset. The inclusion criteria are consistent with the original MAR data, but significantly broader.

Specifically, the inclusion criterion for AMAR is based on groups that are socially relevant without any necessary political activation. By 'socially relevant', as described in Fearon 2006 (Fearon, James D. 2006. "Ethnic mobilization and ethnic violence." In Oxford Handbook of Political Economy, edited by Barry R Weingast & Donald Wittman, 852-868. Oxford: Oxford University Press, 853), we mean 'when people notice and condition their actions on ethnic distinctions in everyday life.' This contrasts to the politicization of ethnicity, that is, 'when political coalitions are organized along ethnic lines, or when access to political or economic benefits depends on ethnicity' (Fearon, 2006: 853).

Importantly, social relevance of an identity does not refer to political mobilization and does not have inherent political connotations; instead, it only refers to the salience of the identity in guiding an individual's actions in her life.

Based on this concept of "social relevance", the new criteria for inclusion in AMAR are defined as follows:

1. Membership in the group is determined primarily by descent by both members and non-members. The group may be a caste if membership is determined by descent and precludes public social mobility.
2. Membership in the group is recognized and viewed as important by members and/or non-members. The importance may be psychological, normative, and/or strategic.
3. Members share some distinguishing cultural features, such as common language, religion, occupational niche, and customs.
4. One or more of these cultural features are either practiced by a majority of the group or preserved and studied by a set of members who are broadly respected by the wider membership for so doing.
5. The group has at least 100,000 members or constitutes one percent of a country's population.

Applying these selection criteria to the world's ethnic groups resulted in the enumeration of 1202 ethnic groups, over 900 of which were not included in the original MAR dataset. These groups are organized by region and can be downloaded in the Microsoft Excel file below.

Furthermore, the primary lists are accompanied by suggestive supplementary lists intended to evoke the multiple underlying ethnic structural dimensions that researchers may wish to incorporate into alternative sample frames as directed by their research agenda. The supplementary lists are not comprehensive.

Download the AMAR Ethnic Groups List (Sample Frame)

Citing the AMAR Sample Frame (Ethnic Groups List)

Birnir, Jóhanna K., Jonathan Wilkenfeld, James D. Fearon, David Laitin, Ted Robert Gurr, Dawn Brancati, Stephen Saideman, Amy Pate, and Agatha S. Hultquist. 2015. "Socially relevant ethnic groups, ethnic structure and AMAR." Journal of Peace Research 52(1): 110-115.

From the list of over 900 new AMAR socially relevant ethnic groups that were not in the MAR data, the AMAR data project Phase I identified a random sample of 74 groups. These 74 groups were coded on all MAR variables, and with proper statistical weighting, combined with the original MAR data to create a new dataset -- the AMAR dataset. The resultant data constitutes an unbiased sample that is more representative of the universe of socially relevant ethnic groups. The AMAR dataset will soon be available publicly.

Download the AMAR Phase I codebook.

Download the AMAR Phase I data (coming soon)

Citing the AMAR Phase I data

Birnir, Jóhanna K; David D Laitin; Jonathan Wilkenfeld; Agatha Hultquist; David Waguespack; Ted Gurr. 2016. "Socially Relevant Identity: Addressing Selection Bias Issues and Introducing the AMAR (All Minorities at Risk) Data." CIDCM Working Paper.