This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| *********************************** | |
| * Ryan Quan * | |
| * Multigenerational Households * | |
| * August 18, 2014 * | |
| *********************************** | |
| /* We are excluding children under 18 from the | |
| analysis to capture only multigenerational family | |
| households of interest. | |
| The variable hhtype1 allows us to do so. | |
| household type (condensed) | Freq. Percent Cum. | |
| ----------------------------+----------------------------------- | |
| married couple, no children | 9,127 27.92 27.92 | |
| single parent | 2,159 6.60 34.52 | |
| other fam., no children | 1,322 4.04 38.57 | |
| single adult | 7,857 24.03 62.60 | |
| cohab couple, no children | 705 2.16 64.76 | |
| non-family, no children | 778 2.38 67.14 | |
| unsure, no children | 444 1.36 68.49 | |
| married couple w children | 8,872 27.14 95.63 | |
| other family w children | 749 2.29 97.92 | |
| cohab couple w children | 373 1.14 99.06 | |
| non-family w children | 76 0.23 99.30 | |
| unsure w children | 230 0.70 100.00 | |
| ----------------------------+----------------------------------- | |
| Total | 32,692 100.00 | |
| */ | |
| drop if inlist(hhtype1, 2, 11, 13, 15, 16, 18) & inlist(famgen_collapsed, 1, 2) | |
| *(12,459 observations) | |
| drop if hhtype1 == .n | |
| *(138 observations) | |
| save Data/famgen_data_for_analysis.dta, replace |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| *********************************** | |
| * Ryan Quan * | |
| * Multigenerational Households * | |
| * August 18, 2014 * | |
| *********************************** | |
| clear | |
| use Data/famgen_data_for_analysis.dta | |
| ************************************************** | |
| ********* LaTeX Tables ************* | |
| ************************************************** | |
| #delimit; | |
| la var famgen_collapsed "No. of Generations Living in Household"; | |
| la def fam 1 "1" 2 "2" 3 "3+", modify; | |
| la val famgen_collapsed fam; | |
| la var born "Imm. Status (%)"; | |
| la def imm 1 "Native Born" 2 "Immigrant"; | |
| la val born imm; | |
| la var race "Race (%)"; | |
| la def rce 1 "White" 2 "Black"; | |
| la val race rce; | |
| la var caedu "Education (%)"; | |
| la def educa 0 "<12 Years" 1 "12 Years" 2 "12-16 Years" 3 ">16 Years"; | |
| la val caedu educa; | |
| la var sex "Sex (%)"; | |
| la def mf 1 "Male" 2 "Female"; | |
| la val sex mf; | |
| la var viq "Verbal IQ (%)"; | |
| la def lh 0 "Low" 1 "High"; | |
| la val viq lh; | |
| recode income (1/11 = 1 "<$25,000") | |
| (12 = 2 ">$25,000"), gen(caincome); | |
| la var caincome "Income (%)"; | |
| recode age (18/24 = 1 "18-24") (25/34 = 2 "25-34")(35/44 = 3 "35-44") | |
| (45/54 = 4 "45-54")(55/64 = 5 "55-64")(65/89 = 6 "65+"), gen(caage); | |
| la var caage "Age (%)"; | |
| /*Descriptives by FAMGEN*/ | |
| tabout caage sex race born viq caincome caedu famgen_collapsed using table1.txt, | |
| cells(col) format(1) clab(_) ptotal(none) | |
| replace | |
| style(tex) bt font(bold) cl1(2-5) npos(row) layout(rb) h3(nil) | |
| topf(top.tex) botf(bot.tex) topstr(14cm) botstr(GSS-NDI); | |
| /*Verbal IQ by Immigration Status*/ | |
| tabout viq born using table2.txt, | |
| cells(col) format(1) clab(_) ptotal(none) layout(rb) h3(nil) | |
| replace | |
| style(tex) bt font(bold) cl1(2-5) npos(row) | |
| topf(top.tex) botf(bot.tex) topstr(14cm) botstr(GSS-NDI); | |
| /*Verbal IQ by Race*/ | |
| tabout viq race using table3.txt, | |
| cells(col) format(1) clab(_) ptotal(none) layout(rb) h3(nil) | |
| replace | |
| style(tex) bt font(bold) cl1(2-5) npos(row) | |
| topf(top.tex) botf(bot.tex) topstr(14cm) botstr(GSS-NDI); | |
| #delimit cr | |
| */ |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| *********************************** | |
| * Ryan Quan * | |
| * Multigenerational Households * | |
| * August 18, 2014 * | |
| *********************************** | |
| clear | |
| use Data/famgen_data_for_analysis.dta | |
| stset studytime, failure(death) | |
| #delimit; | |
| xi: | |
| outreg2 using famgen_main_analysis, | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2) replace: | |
| stcox i.famgen_collapsed; | |
| xi: | |
| outreg2 using famgen_main_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.race i.sex i.caedu; | |
| *Transpose hazard tables to long format; | |
| insheet using famgen_main_analysis.txt, nonames clear; | |
| sxpose, clear firstnames force; | |
| dataout, excel save(famgen_main_analysis) replace; | |
| # delimit cr | |
| /*Model Selection | |
| xi: stcox i.famgen_collapsed age income i.race i.sex i.caedu hompop | |
| /*hompop is not a significant predictor*/ | |
| xi: stcox i.famgen_collapsed age income i.race i.sex i.caedu i.caregion | |
| /*caregion is not a significant predictor*/ | |
| xi: stcox i.famgen_collapsed age income i.race i.sex i.caedu i.famgen_collapsed*i.sex | |
| /*the interaction term between famgen and sex is not significant*/ | |
| xi: stcox i.famgen_collapsed age income i.race i.sex i.caedu i.caregion hompop i.famgen_collapsed*i.sex | |
| */ |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| *********************************** | |
| * Ryan Quan * | |
| * Multigenerational Households * | |
| * August 18, 2014 * | |
| *********************************** | |
| /********************************** | |
| *To run this file:* | |
| Save this .do file to the folder where you wish to do your work. | |
| Save the following .do file to a subfolder named "Programs" | |
| famgen_prepare_data.do | |
| famgen_summary_stats.do | |
| famgen_main_analysis.do | |
| famgen_sens_analysis.do | |
| Save the following datasets to a subfolder named "Data" | |
| GSSNDIFinalRelease.dta | |
| famgen_data_for_analysis.dta | |
| You may download the above .dta file by following the link below: | |
| http://publicdata.norc.org/GSS/DOCUMENTS/OTHR/GSSNDIFinalRelease_stata.zip | |
| Edit the local "switches" below to choose which tables to run. For example, | |
| leaving the code as "local summary_stats_switch = 1" will run the summary statistics | |
| program. Changing the code to "local T1_alt_switch=0" will not run that program. | |
| The full log file outputs to famgen_master.log. | |
| **************************************/ | |
| qui capture log close | |
| log using famgen_master.log, replace | |
| qui set more off, permanently | |
| qui clear mata | |
| qui clear | |
| qui clear matrix | |
| qui set maxvar 8000 | |
| ******************************** | |
| * SWITCHES | |
| ******************************** | |
| local exclude_children_switch = 1 // Exclude households w/ children under 18 | |
| // for 1st & 2nd gen households ONLY. | |
| local summary_stats_switch = 1 // Summary Statistics | |
| local main_analysis_switch = 1 // Main Analysis | |
| local sens_analysis_switch = 1 // Sensitivity Analysis | |
| local latex_tables_switch = 0 // Output Tables in LaTeX | |
| ************************** | |
| * GLOBALS | |
| ************************** | |
| global desc_variables age year race income sex educ hhrace region hompop marital hhtype hhtype1 | |
| global key_variables famgen health yeardeath death agedeath | |
| global sens_variables stress strsswrk strsshme wkstress happy hapmar wordsum born parborn granborn | |
| global new_variables famgen_collapsed castrsshme cahappy viq caparborn caregion caedu hlth children | |
| ************************** | |
| * PREPARE DATA | |
| ************************** | |
| ** This piece - Programs/prepare_data.do - does the following | |
| ***** 1. collapses and recodes variables as necessary | |
| ***** 2. creates any other variables needs for the analysis (e.g. categorical) | |
| ***** 3. prepares data for Cox models in downstream survival analyses | |
| ***** 4. saves a limited local dataset called "famgen_data_for_analysis.dta" | |
| include Programs/famgen_prepare_data.do | |
| *********SAMPLE SIZE IS N=28,350************ | |
| if `exclude_children_switch' == 1 { | |
| include Programs/famgen_exclude_children.do | |
| *********SAMPLE SIZE IS N=17,970************ | |
| } | |
| *************************** | |
| * SUMMARY STATISTICS | |
| **************************** | |
| if `summary_stats_switch' == 1 { | |
| include Programs/famgen_summary_stats.do | |
| } | |
| *************************** | |
| * MAIN ANALYSIS | |
| **************************** | |
| if `main_analysis_switch' == 1 { | |
| include Programs/famgen_main_analysis.do | |
| } | |
| *************************** | |
| * SENSITIVITY ANALYSIS | |
| **************************** | |
| if `sens_analysis_switch' == 1 { | |
| include Programs/famgen_sens_analysis.do | |
| } | |
| *************************** | |
| * LaTeX Tables | |
| **************************** | |
| if `latex_tables_switch' == 1 { | |
| include Programs/famgen_latex_tables.do | |
| } | |
| use Data/famgen_data_for_analysis.dta, clear | |
| log close | |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| *********************************** | |
| * Ryan Quan * | |
| * Multigenerational Households * | |
| * August 18, 2014 * | |
| *********************************** | |
| ************************************************ | |
| *GSS-NDI Final Release (Nov 2011) | |
| ************************************************ | |
| clear | |
| use Data/GSSNDIFinalRelease.dta | |
| keep id $desc_variables $key_variables $sens_variables | |
| *****PREP DATA FOR ANALYSIS*****************; | |
| *Create studytime variable for survival analysis | |
| codebook yeardeath | |
| capture drop studytime | |
| gen studytime =. | |
| replace studytime = yeardeath-year if death==1 | |
| replace studytime =2008-year if death==0 | |
| sum studytime | |
| codebook studytime | |
| tab studytime, missing | |
| /*Adds +1 to studytime so respondents who died | |
| in the same year don't get recorded as 0*/ | |
| replace studytime = studytime +1 | |
| codebook studytime | |
| *****DATA TRANSFORMATIONS*****************; | |
| ****************************** | |
| *FAMGEN | |
| ****************************** | |
| /*There are 7 levels for FAMGEN. | |
| Freq. Numeric Label | |
| 15304 1 1 gen | |
| 11931 2 2 gens, children | |
| 183 3 2 gens, parents | |
| 203 4 2 gens, grandchldrn | |
| 512 5 3 gens, grandchldrn | |
| 205 6 3 gens, chld, par | |
| 12 7 4 gens | |
| Because we are interested in the number of generations | |
| living within a household, we collapse the variable into | |
| three categories: 1 generation, 2 generation, and 3+ gens. | |
| */ | |
| tab famgen, missing | |
| capture drop famgen_collapsed | |
| gen famgen_collapsed =. | |
| replace famgen_collapsed = 1 if famgen == 1 | |
| replace famgen_collapsed = 2 if inlist(famgen, 2, 3, 4) | |
| replace famgen_collapsed = 3 if inlist(famgen, 5, 6, 7) /*let's collapse 4 gen households into 3*/ | |
| replace famgen_collapsed =. if famgen ==.d | famgen ==.i |famgen ==.n | |
| label var famgen_collapsed "Number of Family Generations in Household" | |
| label define fam 1 "1 Gen" 2 "2 Gens" 3 "3+ Gens" | |
| label values famgen_collapsed fam | |
| tab famgen_collapsed, missing | |
| /* | |
| famgen_coll | | |
| apsed | Freq. Percent Cum. | |
| ------------+----------------------------------- | |
| 1 | 15,304 53.98 53.98 | |
| 2 | 12,317 43.45 97.43 | |
| 3 | 729 2.57 100.00 | |
| ------------+----------------------------------- | |
| Total | 28,350 100.00 | |
| */ | |
| ****************************** | |
| *REGION | |
| ****************************** | |
| /*New England = Maine, Vermont, New Hampshire, Massachusetts, Connecticut, Rhode Island | |
| Middle Atlantic = New York, New Jersey, Pennsylvania | |
| East North Central = Wisconsin, Illinois, Indiana, Michigan, Ohio | |
| West North Central = Minnesota, Iowa, Missouri, North Dakota, South Dakota, Nebraska, Kansas | |
| South Atlantic = Delaware, Maryland, West Virginia, Virginia, North Carolina, South Carolina, Georgia, Florida, District of Columbia | |
| East South Central = Kentucky, Tennessee, Alabama, Mississippi | |
| West South Central = Arkansas, Oklahoma, Louisiana, Texas | |
| Mountain = Montana, Idaho, Wyoming, Nevada, Utah, Colorado, Arizona, New Mexico | |
| Pacific = Washington, Oregon, California, Alaska, Hawaii*/ | |
| gen caregion=. | |
| replace caregion=0 if inlist(region, 6, 7) /*W. South Central and E. South Central*/ | |
| replace caregion=1 if inlist(region, 1, 2, 3, 4, 5, 8, 9) /*Not Southern Region*/ | |
| label variable caregion "Categorical Region" | |
| label define southern 0 "Southerner" 1 "Non-Southerner" | |
| label values caregion southern | |
| ****************************** | |
| *EDUCATION | |
| ****************************** | |
| gen caedu=. | |
| replace caedu=0 if edu<12 | |
| replace caedu=1 if edu==12 | |
| replace caedu=2 if edu>12 & edu<16 | |
| replace caedu=3 if edu>=16 | |
| replace caedu=. if edu==.d | edu==.n | |
| la var caedu "Education Level" | |
| la def edl 0 "<12 Years" 1 "12 Years" 2 "13-16 Years" 3 "16+ Years" | |
| la val caedu edl | |
| ****************************** | |
| *WORDSUM (Verbal IQ) | |
| ****************************** | |
| sum wordsum | |
| /*60% of population scored 6 or below. We will use that as a cutoff between low and high VIQ*/ | |
| gen viq=. | |
| replace viq = 0 if wordsum > 0 & wordsum <=6 | |
| replace viq = 1 if wordsum >= 7 | |
| label variable viq "Verbal IQ" | |
| label define lowhigh 0 "Low" 1 "High" | |
| label values viq lowhigh | |
| ****************************** | |
| *PARBORN (Parents born in the United States?) | |
| ****************************** | |
| /*We exclude folks who do not know where at least one parent is born | |
| because we cannot be sure how to categorize them. N = 129)*/ | |
| gen caparborn=. | |
| replace caparborn=0 if parborn==0 | |
| replace caparborn=1 if parborn==1 | parborn==2 | |
| replace caparborn=2 if parborn==8 | |
| replace caparborn=. if inlist(parborn,3, 4, 5, 6, 7, .n) | |
| label define cpb 0 "Both Parents Born in US" 1 "One Parent Born in US" 2 "Neither Parents Born in US" | |
| label values caparborn cpb | |
| label variable caparborn "Parents' Immigration Status" | |
| ****************************** | |
| *BORN (Respondent born in the United States?) | |
| ****************************** | |
| label define brn 1 "R Born in US" 2 "R Not Born in US" | |
| label values born brn | |
| ****************************** | |
| *HEALTH | |
| ****************************** | |
| **self-rated health, hlth==1 are the healthy people | |
| recode health (1/2=1)(3/4=0), gen(hlth) | |
| ****************************** | |
| *STRESS | |
| ****************************** | |
| **stress at home low stress == 1 | |
| gen castrsshme=. | |
| replace castrsshme=0 if inlist(strsshme, 1, 2) | |
| replace castrsshme=1 if inlist(strsshme, 3, 4, 5) | |
| la var castrsshme "Stress at Home" | |
| la val castrsshme lowhigh | |
| ****************************** | |
| *HAPPINESS | |
| ****************************** | |
| **general happiness | |
| gen cahappy=. | |
| replace cahappy=0 if inlist(happy, 1, 2) | |
| replace cahappy=1 if happy == 3 | |
| replace cahappy=. if inlist(happy, .d, .i, .n) | |
| la var cahappy "General Happiness" | |
| la val cahappy lowhigh | |
| ****************************** | |
| *HAPPINESS | |
| ****************************** | |
| **are children under 18 present in the household? | |
| gen children =. | |
| replace children = 1 if inlist(hhtype1, 2, 11, 13, 15, 16, 18) | |
| replace children = 0 if inlist(hhtype1, 1, 3, 4, 5, 6, 8) | |
| label variable children "Children Under 18 Present in Household" | |
| label define chld 1 "Yes" 0 "No" | |
| label values children chld | |
| *****DROPPING MISSING VALUES FROM CONTROL VARIABLES*****************; | |
| #delimit cr | |
| drop if age==.n | |
| *(49 observations deleted) | |
| drop if income==.a | |
| *(3125 observations deleted) | |
| drop if sex ==. | |
| *(0 observations deleted) | |
| *Dropping "other" category in race; | |
| drop if race==. | race == 3 | |
| *(O observations deleted) | |
| drop if hompop ==.n | |
| *(1 observation deleted) | |
| drop if caedu ==. | |
| *(39 observations deleted) | |
| drop if born ==.d | born ==.n | |
| *(77 observations deleted) | |
| ****************************** | |
| save Data/famgen_data_for_analysis.dta, replace | |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| *********************************** | |
| * Ryan Quan * | |
| * Multigenerational Households * | |
| * August 18, 2014 * | |
| *********************************** | |
| clear | |
| use Data/famgen_data_for_analysis.dta | |
| stset studytime, failure(death) | |
| #delimit; | |
| /* *********************************************** | |
| * FULL SAMPLE | |
| * ***********************************************/ | |
| xi: | |
| outreg2 using famgen_sens_analysis, | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2) replace: | |
| stcox i.famgen_collapsed age income i.race i.sex i.caedu; | |
| /* *********************************************** | |
| * HEALTH | |
| * ***********************************************/ | |
| /*including only respondents who have high self-rated health*/ | |
| xi: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.race i.sex i.caedu if hlth ==1; | |
| /* *********************************************** | |
| * VIQ | |
| * ***********************************************/ | |
| /*stratified by high/low VIQ*/ | |
| xi: bysort viq: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.sex i.caedu if viq !=.; | |
| /* *********************************************** | |
| * RACE | |
| * ***********************************************/ | |
| /*stratified by white/black*/ | |
| xi: bysort race: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.sex i.caedu if race !=.; | |
| /*stratified by white/black, high/low VIQ*/ | |
| xi: bysort race viq: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.sex i.caedu if race !=. & viq !=.; | |
| /* *********************************************** | |
| * IMMIGRATION STATUS | |
| * ***********************************************/ | |
| /*stratified by born in the US/not born in the US*/ | |
| xi: bysort born: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.race i.sex i.caedu if born !=.; | |
| /*stratified by born in the US/not born in the US, high/low VIQ*/ | |
| xi: bysort born viq: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.race i.sex i.caedu if born !=. & viq !=.; | |
| /*stratified by born in the US/not born in the US for both respondent and his or her parents*/ | |
| xi: bysort born caparborn: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.race i.sex i.caedu if born !=. & caparborn !=.; | |
| /* *********************************************** | |
| * STRESS | |
| * ***********************************************/ | |
| /*stratified by low stress/high stress*/ | |
| xi: bysort castrsshme: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.race i.sex i.caedu if castrsshme !=.; | |
| /* *********************************************** | |
| * HAPPINESS | |
| * ***********************************************/ | |
| /*stratified by low happiness/high happiness*/ | |
| xi: bysort cahappy: | |
| outreg2 using famgen_sens_analysis, append | |
| stats(coef ci) addstat(N, e(N_sub), Deaths, e(N_fail)) | |
| keep(i.famgen_collapsed) eform noobs nor2 bdec(2) dec(2): | |
| stcox i.famgen_collapsed age income i.race i.sex i.caedu if cahappy !=.; | |
| *Transpose hazard tables to long format; | |
| insheet using famgen_sens_analysis.txt, nonames clear; | |
| sxpose, force firstnames clear; | |
| dataout, excel save(famgen_sens_analysis) replace; | |
| #delimit cr |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| *********************************** | |
| * Ryan Quan * | |
| * Multigenerational Households * | |
| * August 18, 2014 * | |
| *********************************** | |
| *********************************** | |
| *TWO-WAY TABULATIONS | |
| *********************************** | |
| foreach var in $desc_variables $key_variables $sens_variables $new_variables{ | |
| tab `var' famgen_collapsed, missing row col chi2 | |
| } | |
| *********************************** | |
| *POWER ANALYSIS | |
| *********************************** | |
| //store SD of famgen | |
| qui summarize famgen_collapsed | |
| scalar std_famgen = r(sd) | |
| //OVERALL | |
| qui summarize famgen_collapsed | |
| scalar N_famgen = r(N) | |
| qui tab death if death == 1 | |
| scalar N_death = r(N) | |
| scalar failprob = (`=N_death' / `=N_famgen') | |
| stpower cox, n(`=N_famgen') power(0.8) alpha (0.05) sd(`=std_famgen') failprob(`=failprob') hr | |
| //SUB-GROUPS | |
| forval i = 1/3{ | |
| //stores N for each famgen group | |
| qui tab famgen_collapsed if famgen_collapsed == `i' | |
| scalar N_famgen = r(N) | |
| //stores number of deaths for each famgen group | |
| qui tab death if famgen_collapsed == `i' & death == 1 | |
| scalar N_death = r(N) | |
| //calculates probability of death | |
| scalar failprob = (`=N_death' / `=N_famgen') | |
| stpower cox, n(`=N_famgen') power(0.8) alpha (0.05) sd(`=std_famgen') failprob(`=failprob') hr | |
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment