Codebook for Stephen M. Saideman, “Discrimination in International Relations: Examining Why Some Ethnic Groups Receive More External Support Than Others.” Journal of Peace Research, forthcoming.
Levels of Support: Coded by Increasing Intensity [SUPINT90, SUPINT92, SUPINT94, SUPINT96, SUPINT97, SUPINT98] The first three are biennial: 1990-91, 1992-93, 1994-95. The latter three are annual measures.
|
Value |
Label |
Minorities at risk
labels |
|
0 |
None |
No support received |
|
1 |
Low |
Ideological
encouragement, diffuse support,* other unspecified support |
|
2 |
Moderate |
Non-military financial
support, access to external communications, markets, transport, including the
hosting of nonviolent exile organizations |
|
3 |
Strong |
Funds for military
supplies, provision of military equipment and supplies, military training in
exile, advisory military personnel, peace-keeping observers |
|
4 |
Intense |
Blockades,
interdiction against regime, cross-border sanctuaries for armed fighters,
rescue missions in country, cross-border raids in support of dissidents,
active combat units in country. |
* Diffuse support is a rather open category, referring to relatively weak forms of support.
Breadth of Support: [TOTSUP90, TOTSUP92, TOTSUP94, TOTSUP96, TOTSUP 97, TOTSUP98] The first three are biennial: 1990-91, 1992-93, 1994-95. The latter three are annual measures.
A count of the total number of countries supporting a particular minority at risk.
Is group separatist? [SEPARTSM]: Re-coded MAR’s SEPX variable into a dichotomous variable: Is a group actively separatist? 1=Yes, 0= No
Other separatists in host state: [OTHSEP9X]: Using MAR data on separatism, Number of other groups in same state that are actively separatist
Separatists in nearby states [NRSEP9X]: Using MAR data on separatism, Number of separatist groups in adjacent states
African states [Africa]: Is a group located in an African country? 1=Yes, 0=No
Relative Power of host state [POWER90, POWER92]: From Correlates of War data on on each country’s military personnel, military expenditure, energy use, production of iron and steel, and urban and total population. The actual equation for country x would be: power of x relative to the world={[(military personnel of x/world total)+(military expenditure of x/world total]/2+[(energy use of x/world total)+(x’s production of iron and steel/world total)]/2+[(x’s urban population/world total)+(x’s total population/world total)]/2}/3.
Does ethnic kin dominate adjoining state? [IDOMSEG]: From Phase I of MAR and updated for 1990s—does group dominate or is majority in state adjacent to host state. 1=Yes, 0=No
Racial differences [RACE]: From MAR: Different physical appearance. 0=None or Unknown; 1=Physically distinguishable subtype of same racial stock; 2=Different racial stock from the dominant group with substancial intermixture; 3=Different racial stock, little or no intermixture.
Religious
differences [BELIEF]: From MAR:
Different Group Religion. 0=Unknown or
None; 1=Different sect within same religion as the dominant group; 2= Multiple
sects; some different from dominant group; 3=Different religion than dominant
group.
Linguistic
differences [LANGFMI]: Linguistic
differences between group and host state.
I use data from Ethnologue
(Grimes & Grimes, 1996) that codes groups by common supersets. The data are coded from 1 to 20 with 20
reflecting groups whose language is considered identical to that of the
comparison group. For the analyses
here, we divide one by the language family score to put more weight on greater
differentials.
Regime Type
[REGTYP90, REGTYPE94, REGTYP98]:
Using Polity98 data, I subtracted each host country’s autocracy score from its democracy score,
producing an indicator ranging from -10 (most autocratic) to 10 (most democratic).
Rebellion [REB89, REB93, REB97]: From MAR, Measures level of conflict between group and host state.
0=None
1=Political Banditry, sporadic terrorism
2=Campaigns of terrorism
3=Local Rebellions
4=Small-scale guerrilla activity
5=Intermediate scale guerrilla activity
6=Large-scale guerrilla activity
7=Protracted civil war
Other variables (handy for analyses, used for unreported analyses)
Numcode: Unique identifying number for each minority at risk, using country code as starting point.
Ccode: Unique identifying number for each country (same coding as Correlates of War)
Name: Three letter abbreviation of group name
Group: Full group name
Region: In which region is the group/country located?
FSU: Dummy variable for whether group is located in Eastern Europe or former USSR.
Roma: Dummy variable for whether a group consists of Roma (also known as Gypsies) or not.
Maxreb: The maximum level of rebellion for a group between 1945-1984. Range is the same as rebellion. To test whether historically violent movements have different properties.
The Minorities at Risk data and codebooks can be found at: www.bsos.umd.edu/cidcm/mar/. Data I coded and adjustments made to the MAR data can be found at: http://www.ttu.edu/~polisci/saidres.htm.
For questions, I can be reached via email at steve.saideman@ttu.edu or 806-742-3022