* note: Shenzen survey excluded from the Chinese sample set more off * generating variables generate byte female=gender replace female = . if gender>3 replace female = female-1 gen knowent_dum = knowent replace knowent_dum=0 if knowent_dum==8 replace knowent_dum=. if knowent_dum==9 gen educ_secpost = gemeduc replace educ_secpost = 0 if gemeduc<200 replace educ_secpost = 1 if gemeduc==1212 | gemeduc==1316 | gemeduc==1720 replace educ_secpost = . if gemeduc==9999 gen educ_post = gemeduc replace educ_post = 0 if gemeduc<1250 replace educ_post = 1 if gemeduc==1316 | gemeduc==1720 replace educ_post = . if gemeduc==9999 gen gemwork_dum = gemwork3 replace gemwork_dum = 1 if gemwork_dum == 10 replace gemwork_dum = 0 if gemwork_dum == 20 | gemwork_dum == 30 replace gemwork_dum = . if gemwork_dum == 99 gen age_cleaned = age replace age_cleaned = . if age_cleaned == 998 | age_cleaned == 999 gen age_squared = age_cleaned^2 gen fearfail_dum = fearfail replace fearfail_dum = 0 if fearfail_dum == 8 replace fearfail_dum = . if fearfail_dum == 9 gen suskill_dum = suskill replace suskill_dum = 0 if suskill_dum == 8 replace suskill_dum = . if suskill_dum == 9 * Generating means of individual-level variables: country-year egen mage_cy=mean(age_cleaned), by(country_year) gen magesq_cy=mage_cy^2 egen mestb_cy=mean(estbbuso), by(country_year) egen mbaby_cy=mean(babybuso), by(country_year) egen mfemale_cy=mean(female), by(country_year) egen mgemwork_cy=mean(gemwork_dum), by(country_year) egen meduc_secp_cy=mean(educ_secpost), by(country_year) egen meduc_post_cy=mean(educ_post), by(country_year) egen mbusang_cy=mean(busangyy), by(country_year) egen mknowent_cy=mean(knowent_dum), by(country_year) egen mfearfail_cy=mean(fearfail_dum), by(country_year) egen msuskill_cy=mean(suskill_dum), by(country_year) gen lag_ln_gdppc = ln(lag_gdppc_p) gen lag_gdppc_sq = lag_gdppc_p^2 * SOCIAL ENTREPRENEURSHIP set more off * specific variable for involvement in social entrepreneuship (owners-managers plus nascent) gen seomstup_dum = sestart replace seomstup_dum = . if sestart>5 replace seomstup_dum = 0 if sestart==4 replace seomstup_dum = 1 if sestart==1 | sestart==2 | sestart==3 * specific variable for involvement in social entrepreneuship (owners-managers) gen seom_dum = sestart replace seom_dum = . if sestart>5 replace seom_dum = 0 if sestart==4 | sestart==1 replace seom_dum = 1 if sestart==2 | sestart==3 * specific variable for involvement in social entrepreneuship (nascent) gen sestup_dum = sestart replace sestup_dum = . if sestart>5 replace sestup_dum = 0 if sestart==4 | sestart==2 replace sestup_dum = 1 if sestart==1 | sestart==3 egen msseomst_cy=mean(seomstup_dum), by(country_year) egen msseom_cy=mean(seom_dum), by(country_year) egen mssest_cy=mean(sestup_dum), by(country_year) * correcting nascent, baby and established for overlap with social enterprises gen suboanw_nse = suboanw replace suboanw_nse = 0 if sestdif==1 gen babybuso_nse = babybuso replace babybuso_nse = 0 if seowndif==1 gen estbbuso_nse = estbbuso replace estbbuso_nse = 0 if seowndif==1 egen msunse_cy=mean(suboanw_nse), by(country_year) egen mbbnse_cy=mean(babybuso_nse), by(country_year) egen mestbnse_cy=mean(estbbuso_nse), by(country_year) * young and established businesses together gen baesbuso_nse = babybuso_nse replace baesbuso_nse = 1 if estbbuso_nse == 1 egen mbaesbnse_cy=mean(baesbuso_nse), by(country_year) * nascent, young and established businesses together gen subaes_nse = babybuso_nse replace subaes_nse = 1 if estbbuso_nse == 1 replace subaes_nse = 1 if suboanw_nse == 1 egen msubaesns_cy=mean(estbbuso_nse), by(country_year) * government spending gen l_gov_spend = ((100 - l_govt_size)/0.03)^0.5 * generating interative effects gen exec_ss = mssest_cy*l_xconst gen gov_ss = mssest_cy*l_govt_size gen exec_coms = msunse_cy*l_xconst gen gov_coms = msunse_cy*l_govt_size gen spend_ss = mssest_cy*l_gov_spend gen spend_coms = msunse_cy*l_gov_spend * LABELS set more off label variable suboanw_nse "Commercial startup" label variable msunse_cy "Commercial startup - country mean" label variable sestup_dum "Social entrepreneurship startup" label variable mssest_cy "Social entrepren. startup - country mean" label variable babybuso_nse "Commercial young business" label variable mbbnse_cy "Commerc. young business - country mean" label variable estbbuso_nse "Commercial established business" label variable mestbnse_cy "Commerc. establ. bus. - country mean" label variable baesbuso_nse "Commercial young & estab. bus." label variable mbaesbnse_cy "Commer. young & estab. bus. - country mean" label variable msubaesns_cy "Commerc. startup, young & est. - country mean" label variable msseomst_cy "Soc.ent. startup, young & est. - ctry mean" label variable seom_dum "Social entrepren. young & established" label variable msseom_cy "Soc. entr. young & establ. - country mean" label variable busangyy "Business angel in last 3 years" label variable mbusang_cy "Business angel in last 3 y - country mean" label variable knowent_dum "Know somebody who started a business" label variable mknowent_cy "Know somebody who started - coutry mean" label variable female "Female" label variable mfemale "Female - country mean" label variable educ_secpost "Education secondary or higher" label variable meduc_secp_cy "Education sec or higher - country mean" label variable educ_post "Education postsecondary" label variable meduc_post_cy "Education postsecondary - country mean" label variable gemwork_dum "In employment" label variable mgemwork_cy "In employment - country mean" label variable age_cleaned "Age" label variable mage_cy "Age - country mean" label variable age_squared "Age squared" label variable magesq_cy "Age squared - country mean" label variable fearfail_dum "Would fear of failure prevent startup" label variable mfearfail_cy "Fear of failure - country mean" label variable suskill_dum "Believes has skills for startup" label variable msuskill_cy "Skills for startup - country mean" label variable l_xconst "Effective constraints on executive (t-1)" label variable l_govt_size "Government size, Heritage score (t=1)" label variable lag_ln_gdppc "Log of GDP per capita ppp (t-1)" label variable lag_gdppc_p "GDP per capita ppp (t-1)" label variable lag_gdppc_sq "GDP per capita ppp squared (t-1)" label variable lag_dgdp "GDP growth (t-1)" label variable l_gov_spend "Government spending / GDP" label variable exec_ss "Constraints on exec. x Soc. entr. country mean" label variable exec_coms "Constraints on exec. x Com. entr. country mean" label variable gov_ss "Gov. size x Soc. entr. country mean" label variable gov_coms "Gov. size x Com. entr. country mean" label variable spend_ss "Gov. spending x Soc. entr. country mean" label variable spend_coms "Gov. spending x Com. entr. country mean" label variable trust "Most people can be trusted (WVS)" label variable member_assoc "Membership in associations" * ---------------------------------------------------------------------------------------------------------------- * GRAPHS (if not needed: skip and go to estimations) set more off collapse (mean)m*, by(country) label values country country label define country 387 "Bosnia", modify label define country 971 "UAE", modify label define country 970 "West Bank", modify label define country 27 "RSA", modify label define country 1 "US", modify label define country 1809 "Dominican R.", modify *--------------------------------- list country list country, nolabel sort msunse_cy gen byte pos1=3 replace pos1=2 if country==1 replace pos1=2 if country==6 replace pos1=4 if country==21 replace pos1=9 if country==22 replace pos1=11 if country==27 replace pos1=12 if country==30 replace pos1=9 if country==31 replace pos1=9 if country==33 replace pos1=10 if country==34 replace pos1=4 if country==39 replace pos1=5 if country==40 replace pos1=11 if country==60 replace pos1=2 if country==86 replace pos1=12 if country==212 replace pos1=4 if country==213 replace pos1=8 if country==216 replace pos1=2 if country==354 replace pos1=2 if country==358 replace pos1=10 if country==371 replace pos1=4 if country==385 replace pos1=2 if country==387 replace pos1=8 if country==507 replace pos1=6 if country==582 replace pos1=2 if country==593 replace pos1=12 if country==598 replace pos1=7 if country==852 replace pos1=4 if country==876 replace pos1=2 if country==962 replace pos1=1 if country==963 replace pos1=12 if country==966 replace pos1=9 if country==967 replace pos1=4 if country==972 replace pos1=2 if country==1809 * without outliers lv mssest_cy msunse_cy table country mssest_cy if mssest_cy > .1298751 scatter msunse_cy mssest_cy if country!=676 & country!=967, mlabel(country) mlabcolor(gs0) legend(off) mlcolor(gs0) mfcolor(gs4) ytitle("Prevalence rate of commercial entrepreneurship") xtitle("Prevalence rate of social entrepreneurship") mlabv(pos1) * all countries scatter msunse_cy mssest_cy, mlabel(country) mlabcolor(gs0) legend(off) mlcolor(gs0) mfcolor(gs4) ytitle("Prevalence rate of commercial entrepreneurship") xtitle("Prevalence rate of social entrepreneurship") mlabv(pos1) gen byte pos2=3 scatter msseom_cy mbaesbnse_cy, ytitle("Rates of ownership & management of existing soc. enterprises") xtitle("Rates of ownership & management of existing commercial enterprises") mlabel(country) legend(off) mlabv(pos2) pwcorr mssest_cy msunse_cy msseom_cy mbaesbnse_cy * reversed axes gen byte pos3=3 replace pos3=2 if country==7 replace pos3=1 if country==27 replace pos3=1 if country==40 replace pos3=2 if country==44 replace pos3=1 if country==51 replace pos3=9 if country==60 replace pos3=2 if country==358 replace pos3=2 if country==381 replace pos3=4 if country==386 replace pos3=41 if country==593 replace pos3=4 if country==852 replace pos3=11 if country==962 replace pos3=2 if country==963 replace pos3=4 if country==966 replace pos3=4 if country==972 * with regression line scatter mssest_cy msunse_cy if country!=676 & country!=967, mlabel(country) mlabcolor(gs0) legend(off) mlcolor(gs0) mfcolor(gs4) xtitle("Prevalence rate of commercial entrepreneurship") ytitle("Prevalence rate of social entrepreneurship") mlabv(pos3) || lfit mssest_cy msunse_cy if country!=676 & country!=967 * without regression line scatter mssest_cy msunse_cy if country!=676 & country!=967, mlabel(country) mlabcolor(gs0) legend(off) mlcolor(gs0) mfcolor(gs4) xtitle("Prevalence rate of commercial entrepreneurship") ytitle("Prevalence rate of social entrepreneurship") mlabv(pos3) * ---------------------------------------------------------------------------------------------------------------- * CHECKING SOURCES OF MULTICOLLINEARITY set more off reg seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg msseom_cy seom_dum suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg suboanw_nse seom_dum msseom_cy msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg msunse_cy seom_dum msseom_cy suboanw_nse baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg baesbuso_nse seom_dum msseom_cy suboanw_nse msunse_cy mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg mbaesbnse_cy seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg female seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg mfemale_cy seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg educ_secpost seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg meduc_secp_cy seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg educ_post seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg meduc_post_cy seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg age_clean seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg mage_cy seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg age_squared seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy magesq_cy l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg magesq_cy seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared l_xconst l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg l_xconst seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_govt_size lag_ln_gdppc lag_dgdp, cluster(country_year) reg l_govt_size seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst lag_ln_gdppc lag_dgdp, cluster(country_year) reg lag_ln_gdppc seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_dgdp, cluster(country_year) reg lag_dgdp seom_dum msseom_cy suboanw_nse msunse_cy baesbuso_nse mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy l_xconst l_govt_size lag_ln_gdppc, cluster(country_year) * ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ * BIVARIATE MODEL: RESUBMISSION TO ETP *** MLwiN commands (replace with path on your computer) global MLwiN_path "C:\Program Files (x86)\MLwiN v2.24\mlwin.exe" gen byte cons=1 egen i = fill(1 2) sort country_year i * correlations pwcorr sestup_dum suboanw_nse seom_dum baesbuso_nse mssest_cy msunse_cy msseom_cy mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy gemwork_dum mgemwork_cy busangyy mbusang_cy knowent_dum mknowent_cy l_xconst l_gov_spend lag_ln_gdppc lag_dgdp * note that to run the following you will need to change parameters in MLwiN to allow big size of data & variance-covariance etc. * 1/ model with existing businesses means but no entry means runmlwin (suboanw_nse seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy gemwork_dum mgemwork_cy busangyy mbusang_cy knowent_dum mknowent_cy l_xconst l_gov_spend lag_ln_gdppc lag_dgdp cons, equation(1)) (sestup_dum seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female mfemale_cy educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy gemwork_dum mgemwork_cy busangyy mbusang_cy knowent_dum mknowent_cy l_xconst l_gov_spend lag_ln_gdppc lag_dgdp cons, equation(2)), level1(i:) level2(country_year: cons) discrete(distribution(binomial binomial) link(logit) denominator(cons cons)) outreg2 using "ses-tmi-ust results entry 2012 01 12.doc", alpha(0.001,0.01,0.05,0.10) symbol(***,**,*,+)eform replace * testing for differences between individual and mean effects test seom_dum_1 = msseom_cy_1 test seom_dum_2 = msseom_cy_2 test baesbuso_nse_1 = mbaesbnse_cy_1 test baesbuso_nse_2 = mbaesbnse_cy_2 test female_1 = mfemale_cy_1 test female_2 = mfemale_cy_2 test educ_secpost_1 = meduc_secp_cy_1 test educ_secpost_2 = meduc_secp_cy_2 test educ_post_1 = meduc_post_cy_1 test educ_post_2 = meduc_post_cy_2 test age_cleaned_1 = mage_cy_1 test age_cleaned_2 = mage_cy_2 test age_squared_1 = magesq_cy_1 test age_squared_2 = magesq_cy_2 test gemwork_dum_1 = mgemwork_cy_1 test gemwork_dum_2 = mgemwork_cy_2 test busangyy_1 = mbusang_cy_1 test busangyy_2 = mbusang_cy_2 test knowent_dum_1 = mknowent_cy_1 test knowent_dum_2 = mknowent_cy_2 * testing differences between equations test [FP1]seom_dum_1 = [FP2]seom_dum_2 test [FP1]msseom_cy_1 = [FP2]msseom_cy_2 test [FP1]baesbuso_nse_1 = [FP2]baesbuso_nse_2 test [FP1]mbaesbnse_cy_1 = [FP2]mbaesbnse_cy_2 test [FP1]female_1 = [FP2]female_2 test [FP1]mfemale_cy_1 = [FP2]mfemale_cy_2 test [FP1]educ_secpost_1 = [FP2]educ_secpost_2 test [FP1]meduc_secp_cy_1 = [FP2]meduc_secp_cy_2 test [FP1]educ_post_1 = [FP2]educ_post_2 test [FP1]meduc_post_cy_1 = [FP2]meduc_post_cy_2 test [FP1]age_cleaned_1 = [FP2]age_cleaned_2 test [FP1]mage_cy_1 = [FP2]mage_cy_2 test [FP1]age_squared_1 = [FP2]age_squared_2 test [FP1]magesq_cy_1 = [FP2]magesq_cy_2 test [FP1]gemwork_dum_1 = [FP2]gemwork_dum_2 test [FP1]mgemwork_cy_1 = [FP2]mgemwork_cy_2 test [FP1]busangyy_1 = [FP2]busangyy_2 test [FP1]mbusang_cy_1 = [FP2]mbusang_cy_2 test [FP1]knowent_dum_1 = [FP2]knowent_dum_2 test [FP1]mknowent_cy_1 = [FP2]mknowent_cy_2 test [FP1]l_xconst_1 = [FP2]l_xconst_2 test [FP1]l_gov_spend_1 = [FP2]l_gov_spend_2 test [FP1]lag_ln_gdppc_1 = [FP2]lag_ln_gdppc_2 test [FP1]lag_dgdp_1 = [FP2]lag_dgdp_2 * 2/ model with existing businesses means but no entry means - country means no different from ind. effects in both equations gone runmlwin (suboanw_nse seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy gemwork_dum busangyy mbusang_cy knowent_dum l_xconst l_gov_spend lag_ln_gdppc lag_dgdp cons, equation(1)) (sestup_dum seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy gemwork_dum busangyy mbusang_cy knowent_dum l_xconst l_gov_spend lag_ln_gdppc lag_dgdp cons, equation(2)), level1(i:) level2(country_year: cons) discrete(distribution(binomial binomial) link(logit) denominator(cons cons)) outreg2 using "ses-tmi-ust results entry 2011 01 12.doc", alpha(0.001,0.01,0.05,0.10) symbol(***,**,*,+)eform * 3/ model with existing businesses means but no entry means - country means no different from ind. effects gone in corresponding equation runmlwin (suboanw_nse seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy gemwork_dum busangyy mbusang_cy knowent_dum l_xconst l_gov_spend lag_ln_gdppc lag_dgdp cons, equation(1)) (sestup_dum seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post age_clean mage_cy age_squared magesq_cy gemwork_dum busangyy knowent_dum l_xconst l_gov_spend lag_ln_gdppc lag_dgdp cons, equation(2)), level1(i:) level2(country_year: cons) discrete(distribution(binomial binomial) link(logit) denominator(cons cons)) outreg2 using "ses-tmi-ust results entry 2011 01 12.doc", alpha(0.001,0.01,0.05,0.10) symbol(***,**,*,+)eform * 4/ with trust added runmlwin (suboanw_nse seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy gemwork_dum busangyy mbusang_cy knowent_dum l_xconst l_gov_spend lag_ln_gdppc lag_dgdp trust cons, equation(1)) (sestup_dum seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post age_clean mage_cy age_squared magesq_cy gemwork_dum busangyy knowent_dum l_xconst l_gov_spend lag_ln_gdppc lag_dgdp trust cons, equation(2)), level1(i:) level2(country_year: cons) discrete(distribution(binomial binomial) link(logit) denominator(cons cons)) outreg2 using "ses-tmi-ust results entry 2011 01 12.doc", alpha(0.001,0.01,0.05,0.10) symbol(***,**,*,+)eform * 5/ with associations added runmlwin (suboanw_nse seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post meduc_post_cy age_clean mage_cy age_squared magesq_cy gemwork_dum busangyy mbusang_cy knowent_dum l_xconst l_gov_spend lag_ln_gdppc lag_dgdp member_assoc cons, equation(1)) (sestup_dum seom_dum baesbuso_nse msseom_cy mbaesbnse_cy female educ_secpost meduc_secp_cy educ_post age_clean mage_cy age_squared magesq_cy gemwork_dum busangyy knowent_dum l_xconst l_gov_spend lag_ln_gdppc lag_dgdp member_assoc cons, equation(2)), level1(i:) level2(country_year: cons) discrete(distribution(binomial binomial) link(logit) denominator(cons cons)) outreg2 using "ses-tmi-ust results entry 2011 01 12.doc", alpha(0.001,0.01,0.05,0.10) symbol(***,**,*,+)eform