*1A Basic Descriptive Statistics sum housereg drive elecbill affidavit witness adddoc income male age literacy educ occup cityyears if control==1 Variable Obs Mean Std Dev Min Max --- housereg 21 2380952 4364358 0 1 drive 21 0952381 3007926 0 1 elecbill 21 9047619 3007926 0 1 affidavit 21 0 0 0 0 witness 21 0 0 0 0 --- adddoc 21 1904762 4023739 0 1 income 21 1890476 1051145 0 40 male 21 7619048 4364358 0 1 age 21 3680952 9474276 20 53 literacy 21 6666667 7958224 0 2 --- educ 21 2666667 321455 0 10 occup 21 152381 8728716 0 3 cityyears 21 1919048 563577 12 40 sum housereg drive elecbill affidavit witness adddoc income male age literacy educ occup cityyears if ngo==1 Variable Obs Mean Std Dev Min Max --- housereg 18 0555556 2357023 0 1 drive 18 0555556 2357023 0 1 elecbill 18 9444444 2357023 0 1 affidavit 18 0 0 0 0 witness 18 0 0 0 0 --- adddoc 18 2777778 4608886 0 1 income 18 2105556 6812378 12 36 male 18 7777778 4277926 0 1 age 18 3972222 8470055 29 60 literacy 18 6111111 6978023 0 2 --- educ 18 1777778 2839958 0 8 occup 18 1666667 9074852 1 3 cityyears 18 1627778 5210911 8 30 sum housereg drive elecbill affidavit witness adddoc income male age literacy educ occup cityyears if rti==1 Variable Obs Mean Std Dev Min Max --- housereg 23 2608696 4489778 0 1 drive 23 0434783 2085144 0 1 elecbill 23 9130435 2881041 0 1 affidavit 23 0 0 0 0 witness 23 0 0 0 0 --- adddoc 23 3478261 4869848 0 1 income 23 2004348 8519914 0 32 male 23 826087 3875534 0 1 age 23 3891304 8490171 24 55 literacy 23 7826087 9513876 0 2 --- educ 23 3652174 4396369 0 11 occup 23 126087 6887004 0 3 cityyears 23 1617391 4458416 10 30
sum housereg drive elecbill affidavit witness adddoc income male age literacy educ occup cityyears if bribe==1 Variable Obs Mean Std Dev Min Max --- housereg 24 125 337832 0 1 drive 24 0416667 2041241 0 1 elecbill 24 9583333 2041241 0 1 affidavit 24 0 0 0 0 witness 24 0 0 0 0 --- adddoc 24 0833333 2823299 0 1 income 24 2195833 8185242 12 42 male 24 8333333 3806935 0 1 age 24 37125 101438 23 60 literacy 24 8333333 8164966 0 2 --- educ 24 2875 3639308 0 12 occup 24 1708333 8586727 1 3 cityyears 24 17 4836456 10 30 *1B Nonparametric descriptive statistics - cardreceive summarize cardreceive if bribe ==1, detail cardreceive - Percentiles Smallest 1% 66 66 5% 72 72 10% 73 73 Obs 24 25% 75 74 Sum of Wgt 24 50% 82 Mean 8375 Largest Std Dev 1084776 75% 90 97 90% 97 97 Variance 1176739 95% 101 101 Skewness 8242581 99% 113 113 Kurtosis 3459718 summarize cardreceive if rti ==1, detail cardreceive - Percentiles Smallest 1% 90 90 5% 94 94 10% 98 98 Obs 23 25% 104 98 Sum of Wgt 23 50% 120 Mean 1503913 Largest Std Dev 7943536 75% 163 176 90% 343 343 Variance 6309976 95% 343 343 Skewness 1876776 99% 343 343 Kurtosis 4999459 summarize cardreceive if ngo ==1, detail cardreceive - Percentiles Smallest 1% 119 119
5% 119 137 10% 137 331 Obs 18 25% 336 331 Sum of Wgt 18 50% 343 Mean 3228333 Largest Std Dev 7175715 75% 355 356 90% 364 362 Variance 5149088 95% 367 364 Skewness -2376489 99% 367 367 Kurtosis 6888763 summarize cardreceive if control ==1, detail cardreceive - Percentiles Smallest 1% 331 331 5% 336 336 10% 336 336 Obs 21 25% 343 342 Sum of Wgt 21 50% 343 Mean 3484762 Largest Std Dev 1063306 75% 353 360 90% 361 361 Variance 1130619 95% 371 371 Skewness 6924268 99% 371 371 Kurtosis 2917844 *1C Nonparametric descriptive statistics - resreceive summarize resreceive if bribe ==1, detail resreceive - Percentiles Smallest 1% 9 9 5% 9 9 10% 9 9 Obs 24 25% 14 9 Sum of Wgt 24 50% 17 Mean 16 Largest Std Dev 4191088 75% 195 21 90% 21 21 Variance 1756522 95% 21 21 Skewness -401798 99% 22 22 Kurtosis 2164494 summarize resreceive if rti ==1, detail resreceive - Percentiles Smallest 1% 28 28 5% 32 32 10% 35 35 Obs 23 25% 35 35 Sum of Wgt 23 50% 37 Mean 3734783 Largest Std Dev 4754029 75% 39 41 90% 41 41 Variance 2260079 95% 42 42 Skewness 163429 99% 54 54 Kurtosis 8043052
summarize resreceive if ngo ==1, detail resreceive - Percentiles Smallest 1% 25 25 5% 25 29 10% 29 30 Obs 18 25% 35 32 Sum of Wgt 18 50% 37 Mean 3694444 Largest Std Dev 6940692 75% 38 40 90% 43 43 Variance 481732 95% 58 43 Skewness 128994 99% 58 58 Kurtosis 6082519 summarize resreceive if control ==1, detail resreceive - Percentiles Smallest 1% 28 28 5% 32 32 10% 32 32 Obs 21 25% 36 32 Sum of Wgt 21 50% 37 Mean 512381 Largest Std Dev 669641 75% 38 43 90% 43 43 Variance 448419 95% 45 45 Skewness 4224812 99% 343 343 Kurtosis 1892271 *1D Probit to check randomization probit control drive housereg elecbill income male age educ occup cityyears islam Iteration 0: log likelihood = -47803724 Iteration 1: log likelihood = -42433173 Iteration 2: log likelihood = -42263501 Iteration 3: log likelihood = -42262509 Iteration 4: log likelihood = -42262509 Probit regression Number of obs = 86 LR chi2(10) = 1108 Prob > chi2 = 03511 Log likelihood = -42262509 Pseudo R2 = 01159 control Coef Std Err z P> z [95% Conf Interval] - drive 802875 6288174 128 0202-4295845 2035334 housereg 7056139 5268467 134 0180-3269867 1738215 elecbill 1370645 7422781 018 0854-1317774 1591903 income -0140208 029651-047 0636-0721358 0440941 male 0701465 4639213 015 0880-8391225 9794154 age -0378079 0214758-176 0078-0798997 004284 educ -0311466 0621447-050 0616-1529479 0906548 occup 0423707 2547152 017 0868-4568619 5416034 cityyears 0765702 0366658 209 0037 0047066 1484339 islam -1416773 1075919-132 0188-3525536 6919899 _cons 6579391 169556 039 0698-2665298 3981177
probit ngo drive housereg elecbill income male age educ occup cityyears islam Iteration 0: log likelihood = -44120652 Iteration 1: log likelihood = -38506631 Iteration 2: log likelihood = -37880111 Iteration 3: log likelihood = -37751456 Iteration 4: log likelihood = -3771586 Iteration 5: log likelihood = -37705243 Iteration 6: log likelihood = -37701951 Iteration 7: log likelihood = -37700901 Iteration 8: log likelihood = -37700559 Iteration 9: log likelihood = -37700446 Iteration 10: log likelihood = -37700409 Iteration 11: log likelihood = -37700396 Iteration 12: log likelihood = -37700392 Iteration 13: log likelihood = -3770039 Iteration 14: log likelihood = -3770039 Iteration 15: log likelihood = -37700389 Iteration 16: log likelihood = -37700389 Iteration 17: log likelihood = -37700389 Iteration 18: log likelihood = -37700389 Iteration 19: log likelihood = -37700389 Iteration 20: log likelihood = -37700389 Probit regression Number of obs = 86 LR chi2(10) = 1284 Prob > chi2 = 02327 Log likelihood = -37700389 Pseudo R2 = 01455 ngo Coef Std Err z P> z [95% Conf Interval] - drive 2257225 7399285 031 0760-1224511 1675956 housereg -6405723 elecbill -5852504 6785383-863 0000-7182415 -4522593 income 007824 0309924 025 0801-05292 068568 male 1801828 4741488 038 0704-7491318 1109497 age 0257219 020593 125 0212-0146396 0660835 educ -1248846 0679556-184 0066-2580752 008306 occup 2074001 247265 084 0402-2772304 6920307 cityyears -0359132 0361326-099 0320-1067319 0349054 islam -133752 1173755-114 0254-3638038 9629976 _cons 5744684 1804743 318 0001 2207454 9281915 Note: 9 failures and 0 successes completely determined probit rti drive housereg elecbill income male age educ occup cityyears islam note: islam!= 1 predicts failure perfectly islam dropped and 2 obs not used Iteration 0: log likelihood = -49308938 Iteration 1: log likelihood = -42417018 Iteration 2: log likelihood = -42155602 Iteration 3: log likelihood = -42152908 Iteration 4: log likelihood = -42152908 Probit regression Number of obs = 84 LR chi2(9) = 1431 Prob > chi2 = 01116 Log likelihood = -42152908 Pseudo R2 = 01451
rti Coef Std Err z P> z [95% Conf Interval] - drive -1701683 779185-022 0827-1697343 1357006 housereg 2902553 4881474 059 0552-6664961 1247007 elecbill -3991897 7226107-055 0581-1815481 1017101 income -0184942 0277749-067 0506-072932 0359437 male -1937685 4604067-042 0674-1096149 7086121 age 0187205 0207936 090 0368-0220343 0594753 educ 1723741 0622252 277 0006 0504149 2943333 occup -4559441 2771581-165 0100-9991639 0872758 cityyears -0595372 0371938-160 0109-1324357 0133613 _cons 678041 1380969 049 0623-2028608 338469 probit bribe drive housereg elecbill income male age educ occup cityyears islam note: islam!= 1 predicts failure perfectly islam dropped and 2 obs not used Iteration 0: log likelihood = -50254645 Iteration 1: log likelihood = -48303737 Iteration 2: log likelihood = -48281256 Iteration 3: log likelihood = -4828124 Probit regression Number of obs = 84 LR chi2(9) = 395 Prob > chi2 = 09149 Log likelihood = -4828124 Pseudo R2 = 00393 bribe Coef Std Err z P> z [95% Conf Interval] - drive -6165723 6975217-088 0377-198369 7505452 housereg -0966785 5319642-018 0856-1139309 9459522 elecbill 5883868 8616424 068 0495-1100401 2277175 income 0276465 0275865 100 0316-0264221 0817151 male -1791836 443847-040 0686-1049108 6907406 age -0076042 0179148-042 0671-0427165 0275082 educ -0407024 0588258-069 0489-1559989 0745941 occup 1574731 2226772 071 0479-2789662 5939124 cityyears 0042961 0324957 013 0895-0593942 0679865 _cons -1419107 1266775-112 0263-3901941 1063727 *1E LPM to check randomization reg control drive housereg elecbill income male age educ occup cityyears islam Source SS df MS Number of obs = 86 ---+ F( 10, 75) = 097 Model 181304476 10 181304476 Prob > F = 04789 Residual 140590483 75 187453977 R-squared = 01142 ---+ Adj R-squared = -00039 Total 15872093 85 186730506 Root MSE = 43296 control Coef Std Err t P> t [95% Conf Interval] - drive 2300534 2127013 108 0283-1936693 6537761 housereg 1724249 1627813 106 0293-151852 4967018 elecbill 0285887 2395725 012 0905-4486641 5058415 income -0039438 0083967-047 0640-0206709 0127834 male 0370424 1377431 027 0789-237356 3114408
age -0083178 0057571-144 0153-0197866 003151 educ -0083759 0177714-047 0639-0437785 0270266 occup 0046016 0730621 006 0950-1409454 1501487 cityyears 0200828 0101906 197 0052-0002179 0403836 islam -4277064 3506691-122 0226-1126275 2708622 _cons 6314622 5263865 120 0234-4171535 1680078 reg ngo drive housereg elecbill income male age educ occup cityyears islam Source SS df MS Number of obs = 86 ---+ F( 10, 75) = 099 Model 166017627 10 166017627 Prob > F = 04594 Residual 125723819 75 167631758 R-squared = 01166 ---+ Adj R-squared = -00011 Total 142325581 85 16744186 Root MSE = 40943 ngo Coef Std Err t P> t [95% Conf Interval] - drive 0160971 2011412 008 0936-3845966 4167909 housereg -2468735 1539342-160 0113-5535262 0597793 elecbill -0968941 2265519-043 0670-5482086 3544205 income 0018604 0079404 023 0815-0139577 0176784 male 038339 1302569 029 0769-2211461 2978241 age 0055987 0054442 103 0307-0052468 0164442 educ -0277315 0168056-165 0103-0612099 005747 occup 0561917 0690912 081 0419-081445 1938284 cityyears -0077889 0096368-081 0422-0269863 0114086 islam -3799907 3316105-115 0255-1040593 2806113 _cons 5557963 4977778 112 0268-435828 1547421 reg rti drive housereg elecbill income male age educ occup cityyears islam Source SS df MS Number of obs = 86 ---+ F( 10, 75) = 134 Model 254751555 10 254751555 Prob > F = 02274 Residual 143013217 75 190684289 R-squared = 01512 ---+ Adj R-squared = 00380 Total 168488372 85 198221614 Root MSE = 43667 rti Coef Std Err t P> t [95% Conf Interval] - drive -0510316 2145262-024 0813-4783896 3763265 housereg 1018704 1641778 062 0537-2251886 4289294 elecbill -0806862 2416279-033 0739-5620336 4006612 income -0056127 0084688-066 0510-0224834 0112579 male -0194195 1389249-014 0889-2961721 2573331 age 004857 0058065 084 0406-0067102 0164242 educ 0456639 0179239 255 0013 0099576 0813701 occup -1107058 0736889-150 0137-2575016 03609 cityyears -014328 010278-139 0167-0348029 0061469 islam 4406101 3536776 125 0217-2639519 1145172 _cons 1310005 5309026 025 0806-9266117 1188613 reg bribe drive housereg elecbill income male age educ occup cityyears islam Source SS df MS Number of obs = 86 ---+ F( 10, 75) = 040 Model 872240508 10 087224051 Prob > F = 09435
Residual 164300851 75 219067801 R-squared = 00504 ---+ Adj R-squared = -00762 Total 173023256 85 203556772 Root MSE = 46805 bribe Coef Std Err t P> t [95% Conf Interval] - drive -1951189 2299387-085 0399-6531803 2629425 housereg -0274218 1759731-016 0877-3779783 3231346 elecbill 1489916 2589875 058 0567-366938 6649212 income 0076962 0090772 085 0399-0103866 0257789 male -0559619 1489059-038 0708-3525977 2406739 age -0021379 0062237-034 0732-0145362 0102603 educ -0095565 0192116-050 0620-047828 0287151 occup 0499125 078983 063 0529-1074298 2072548 cityyears 002034 0110165 018 0854-0199119 0239799 islam 367087 3790875 097 0336-3880939 1122268 _cons -318259 5690451-056 0578-1451855 8153369 *2 Plots gen treatment = 1 if bribe ==1 (62 missing values generated) replace treatment = 2 if rti ==1 (23 real changes made) replace treatment = 3 if ngo ==1 (18 real changes made) replace treatment = 4 if control ==1 (21 real changes made) *2A Dotplots scatter treatment cardreceive [aweight = dotsizecr], ytitle("treatment Group") xtitle("processing Time (days)") ylabel(1 "Bribe" 2 "RTI" 3 "NG > O" 4 "Control") m(oh)
Treatment Group Bribe RTI NGO Control 0 100 200 300 400 Processing Time (days) scatter treatment resreceive [aweight = dotsizeres], ytitle("treatment Group") xtitle("resver Time (days)") ylabel(1 "Bribe" 2 "RTI" 3 "NGO" 4 > "Control") m(oh) Treatment Group Bribe RTI NGO Control 0 100 200 300 400 ResVer Time (days) *2B Boxplots graph hbox cardreceive, over(treatment, relabel (1 "Bribe" 2 "RTI" 3 "NGO" 4 "Control")) ytitle("processing Time (days)") caption("figure 1 - > Ration Card Processing Time by Treatment Group", size(medlarge))scheme(s2mono)
Bribe RTI NGO Control 0 100 200 300 400 Processing Time (days) Figure 1 - Ration Card Processing Time by Treatment Group graph hbox resreceive, over(treatment, relabel (1 "Bribe" 2 "RTI" 3 "NGO" 4 "Control")) ytitle("residency Verification Time (days)") Bribe RTI NGO Control 0 100 200 300 400 Residency Verification Time (days)
*2C Survival rates by group label define treatment 1 "Bribe" 2 "RTI" 3 "NGO" 4 "Control" label value treatment treatment ltable cardreceive, intervals(20) by(treatment) gr overlay xlabel(0(100) 400) xtitle("processing time") ytitle("proportion without ration card")msymbol(t)legend(off) proportion without ration card 0 2 4 6 8 1 Bribe RTI NGO Control 0 100 200 300 400 processing time *3A Mann-Whitney nonparametric difference of means tests - cardreceive *Bribe versus RTI sort brnum ranksum cardreceive in 1/47, by(bribe) bribe obs rank sum expected 0 23 8095 552 1 24 3185 576 combined 47 1128 1128 unadjusted variance 220800 adjustment for ties -153 adjusted variance 220647 Ho: cardre~e(bribe==0) = cardre~e(bribe==1) z = 5482 Prob > z = 00000 *Bribe versus NGO sort bnnum
ranksum cardreceive in 1/42, by(bribe) bribe obs rank sum expected 0 18 603 387 1 24 300 516 combined 42 903 903 unadjusted variance 154800 adjustment for ties -113 adjusted variance 154687 Ho: cardre~e(bribe==0) = cardre~e(bribe==1) z = 5492 Prob > z = 00000 *Bribe versus control sort bcnum ranksum cardreceive in 1/45, by(bribe) bribe obs rank sum expected 0 21 735 483 1 24 300 552 combined 45 1035 1035 unadjusted variance 193200 adjustment for ties -815 adjusted variance 192385 Ho: cardre~e(bribe==0) = cardre~e(bribe==1) z = 5745 Prob > z = 00000 *RTI versus NGO sort rnnum ranksum cardreceive in 1/41, by(rti) rti obs rank sum expected 0 18 544 378 1 23 317 483 combined 41 861 861 unadjusted variance 144900 adjustment for ties -341 adjusted variance 144559
Ho: cardre~e(rti==0) = cardre~e(rti==1) z = 4366 Prob > z = 00000 *RTI versus Control sort rcnum ranksum cardreceive in 1/44, by(rti) rti obs rank sum expected 0 21 6915 4725 1 23 2985 5175 combined 44 990 990 unadjusted variance 181125 adjustment for ties -2170 adjusted variance 178955 Ho: cardre~e(rti==0) = cardre~e(rti==1) z = 5177 Prob > z = 00000 *NGO versus Control sort ncnum ranksum cardreceive in 1/39, by(ngo) ngo obs rank sum expected 0 21 454 420 1 18 326 360 combined 39 780 780 unadjusted variance 126000 adjustment for ties -1811 adjusted variance 124189 Ho: cardre~e(ngo==0) = cardre~e(ngo==1) z = 0965 Prob > z = 03346 *3B Mann-Whitney nonparametric difference of means tests - residency verification *Bribe versus RTI sort brnum ranksum resreceive in 1/47, by(bribe) bribe obs rank sum expected
0 23 828 552 1 24 300 576 combined 47 1128 1128 unadjusted variance 220800 adjustment for ties -1813 adjusted variance 218987 Ho: resrec~e(bribe==0) = resrec~e(bribe==1) z = 5898 Prob > z = 00000 *Bribe versus NGO sort bnnum ranksum resreceive in 1/42, by(bribe) bribe obs rank sum expected 0 18 603 387 1 24 300 516 combined 42 903 903 unadjusted variance 154800 adjustment for ties -1116 adjusted variance 153684 Ho: resrec~e(bribe==0) = resrec~e(bribe==1) z = 5510 Prob > z = 00000 *Bribe versus control sort bcnum ranksum resreceive in 1/45, by(bribe) bribe obs rank sum expected 0 21 735 483 1 24 300 552 combined 45 1035 1035 unadjusted variance 193200 adjustment for ties -2087 adjusted variance 191113 Ho: resrec~e(bribe==0) = resrec~e(bribe==1) z = 5764 Prob > z = 00000
*RTI versus NGO sort rnnum ranksum resreceive in 1/41, by(rti) rti obs rank sum expected 0 18 3695 378 1 23 4915 483 combined 41 861 861 unadjusted variance 144900 adjustment for ties -3307 adjusted variance 141593 Ho: resrec~e(rti==0) = resrec~e(rti==1) z = -0226 Prob > z = 08213 *RTI versus Control sort rcnum ranksum resreceive in 1/44, by(rti) rti obs rank sum expected 0 21 482 4725 1 23 508 5175 combined 44 990 990 unadjusted variance 181125 adjustment for ties -4723 adjusted variance 176402 Ho: resrec~e(rti==0) = resrec~e(rti==1) z = 0226 Prob > z = 08211 *NGO versus Control sort ncnum ranksum resreceive in 1/39, by(ngo) ngo obs rank sum expected 0 21 4345 420 1 18 3455 360 combined 39 780 780
unadjusted variance 126000 adjustment for ties -4132 adjusted variance 121868 Ho: resrec~e(ngo==0) = resrec~e(ngo==1) z = 0415 Prob > z = 06779 end of do-file *Duration Model Cox Model *Bribe, RTI and NGO vs control *Bribe, RTI and NGO vs control sort brnum stcox bribe rti control, nohr efron Cox regression -- Efron method for ties No of subjects = 86 Number of obs = 86 No of failures = 46 Time at risk = 18598 LR chi2(3) = 12344 Log likelihood = -12817849 Prob > chi2 = 00000 _t Coef Std Err z P> z [95% Conf Interval] - bribe 5346693 8405906 636 0000 3699166 699422 rti 2769576 7494007 370 0000 1300778 4238374 control -452671 stcox bribe in 1/47, nohr efron failure _d: censor analysis time _t: cardreceive Iteration 0: log likelihood = -13501096 Iteration 1: log likelihood = -11618645 Iteration 2: log likelihood = -11604485 Iteration 3: log likelihood = -11604448 Refining estimates: Iteration 0: log likelihood = -11604448 Cox regression -- Efron method for ties No of subjects = 47 Number of obs = 47 No of failures = 44 Time at risk = 5469 LR chi2(1) = 3793 Log likelihood = -11604448 Prob > chi2 = 00000 _t Coef Std Err z P> z [95% Conf Interval] - bribe 2523528 4489662 562 0000 164357 3403486