- 3 The Following Two Tables Appear In The Paper Can Hearts And Minds By Bought The Economics Of Counterinsurgency In I 1 (92.43 KiB) Viewed 30 times
- 3 The Following Two Tables Appear In The Paper Can Hearts And Minds By Bought The Economics Of Counterinsurgency In I 2 (121.44 KiB) Viewed 30 times
3. The following two tables appear in the paper "Can Hearts and Minds by Bought: The Economics of Counterinsurgency in Iraq" by Eli Berman, Jacob N. Shapiro and Joseph H. Felter. The first table (TABLE 2) gives summary statistics and the second table (TABLE 3) gives results from six different linear regressions. O 0 CERP projets 552.812 1.040 69.214 ERP projects TABLE 2 SUMMARY STATISTICS Standard Variable Observations Weight Mean Deviation Minimum Maximum Incidents per 1.000 1.000 288,023,728 1.958 22.754 Ethnicity: Sunni vole share 18 28,104,187 210 0 .917 Shia wote share 18 28,104,187 466 351 .902 Kurdish vote share 18 28,104,187 .171 0 .999 Reconstruction spending (sper capita): 1.000 288,025,728 10.562 17.150 0 CERP excluding CHRRP and OHDACA 1.040 288,023.728 10,281 16.956 0 552.812 of which > 50,000 projects 288,023,728 9.043 15.302 0 527.450 of which s 50,000 projects 1,040 288,023,728 3.007 Non-CERP 1,040 288,023,728 68.856 124.632 Larve projects 1.040 288,023,728 115.168 Education projects 1,040 461 2.127 Electricity projects .003 .105 9.944 Health projects 1.040 288,023,728 872 2.450 45.795 projects 1.000 288,023,728 .745 2.858 99.415 Water and sanitation projects 1.040 288,023,728 6.981 68204 10,830,4 Economy Unemployment rate 312 84,647,702 096 06 .495 Proportion of population in 1st or 2nd income 304 84,181.884 39.686 14.886 30 Income quintile change. 2003-4 100 26,637,385 -012 -1.6 1.925 Natural resources Oil and gas reserves, price weighted 1.040 288,023,728 0012 0012 017 Pipeline volume, price weighted 1,040 288,023,728 13.143 29.218 250.968 Nori-Means weighted by WFP district population estimates. Vote shares are from December 2005 clections at the governorate level. Unit of observation for timecarving data is the district/half year. Summary statistics for variables based on survey data are weighted by population during survey, Economic variables are based on COSIT surveys fielded in 2004:1, 2005:11, and 2007:1 1.238 57.750 10,840,8 10,818,0 288,023,728 288,023,728 33.107 1,040 0 0 0 0 0 D Transportation 5 quintile
TABLE 3 PREDICTORS OF VIOLENT INCIDENTS AGAINST COALITION AND IRAQI FORCES DEPENDENT VARIABLE: INCIDENTS PER 1,000 (2) (4) (5) (6) Sunni share 2.123. 2.114 915 1.113 1.381*** .965 (29) (29) (.15) (.12) (139) 2005 .191 -.0745 - 0761 285 -.0009 (064) (.043) (042) (190) 2006 (056) 531 -0215 -0231 243 .085 (.15) (12) (.12) (207) 2007 (.071) .608 230 402 .135 (.12) (10) (.10) (149) 0.074) 2008 .0279 0246 .0229 .110 -.090 (.040) (.034) (.034) (059) (.054) Sunni share x 2005 1.321 1.328 1.172*** .027 (29) (29) (29) (244) Sunni share < 2006 2.689*** 2.696*** 2.651*** .425 (.78) (.78) (179) (.431) Sunni share x 2007 1.857 1.86 1.886 -1.313 (40) (40) (376) (.489) Sunni stare x 2008 0713 0777 .141 -1.270 (25) (26) (285) (253) Shia share 251 .083 (.14) (208) (.050) Income quintile change, - 215 - 030 20024 (.178) (.036) Unemployment rate -3.252 - 783 (1.694) (.475) Proportion of popula- - 0101 -.00006 tion in Ist or 2nd in- (.0042) 0.0009) come quintile Incidents/1,000 lagged half year (0258) Constant - 129 .110 -.0479 .305 0271 (.067) (.059) (.035) (.069) (120) (.0346) Observations 1,040 1,040 1.010 1.040 1.000 900 .18 22 26 27 32 .77 NOTE--Robust standard errors are in parentheses, clustered by district. Results are robust to clustering by governorate instead. Regressions are weighted by estimated pop- ulation. * Significant at the 10 percent level. Significant at the 5 percent level. *** Significant at the I percent level. a. Interpret the estimated coefficient of 2.123 on the Sunni vote share given in equation 1 of TABLE 3 (5 marks) b. Assess the magnitude of the coefficient that you interpreted in part a. of this question. (10 marks)