##thoughts of suicide wave 1
mergeddata4$sui_1<-Recode(mergeddata4$H1SU1, recodes="1=1; 0=0; else=NA")
## age, if friend attempted suicide, if family attempted, if live alone, education, thoughts of suicide, sex, race, closeness to mentor, and thoughts on likelihood of divorce at wave 3
mergeddata4$age_3<-(mergeddata4$IYEAR3 - mergeddata4$H3OD1Y)
mergeddata4$own_gun<-Recode(mergeddata4$H3DS14, recodes="1=1; 0=0; else=NA")
mergeddata4$H3TO27 <- as.numeric(mergeddata4$H3TO27)
mergeddata4$try_nth<-Recode(mergeddata4$H3TO27, recodes= "1='not_true';
2:3='true'; 4:5='very_true'; else=NA", as.factor=T)
mergeddata4$H3TO28<- as.numeric(mergeddata4$H3TO28)
mergeddata4$l4excite<-Recode(mergeddata4$H3TO28, recodes= "1='not_true';
2:3='true'; 4:5='very_true'; else=NA", as.factor=T)
mergeddata4$H3TO30<- as.numeric(mergeddata4$H3TO30)
mergeddata4$dthfmoment<-Recode(mergeddata4$H3TO30, recodes= "1='not_true';
2:3='true'; 4:5='very_true'; else=NA", as.factor=T)
mergeddata4$H3TO31<- as.numeric(mergeddata4$H3TO31)
mergeddata4$excite_lc<-Recode(mergeddata4$H3TO31, recodes= "1='not_true';
2:3='true'; 4:5='very_true'; else=NA", as.factor=T)
mergeddata4$H3TO32<- as.numeric(mergeddata4$H3TO32)
mergeddata4$like_no_res<-Recode(mergeddata4$H3TO32, recodes= "1='not_true';
2:3='true'; 4:5='very_true'; else=NA", as.factor=T)
mergeddata4$H3TO33<- as.numeric(mergeddata4$H3TO33)
mergeddata4$follow_ins<-Recode(mergeddata4$H3TO33, recodes= "1='not_true';
2:3='true'; 4:5='very_true'; else=NA", as.factor=T)
mergeddata4$H3ED1 <- as.numeric(mergeddata4$H3ED1)
mergeddata4$educ_re3<-Recode(mergeddata4$H3ED1, recodes= "6:11='lesshigh';
12='high'; 13:15='somecol'; 16:22='col'; else=NA", as.factor=T)
mergeddata4$sui_3<-Recode(mergeddata4$H3TO130, recodes="1=1; 0=0; else=NA")
mergeddata4$male_3<-Recode(mergeddata4$BIO_SEX3, recodes="1=1; 2=0; else=NA")
mergeddata4$H3IR4 <- as.numeric(mergeddata4$H3IR4)
mergeddata4$race_3<-Recode(mergeddata4$H3IR4, recodes= "1='nwhite';
2='nhblack'; 3:5='other'; else=NA", as.factor=T)
## thoughts of suicide at wave four
mergeddata4$sui_4<-Recode(mergeddata4$H4SE1, recodes="1=1; 0=0; else=NA")
## use scale ( ) function for continuous, makes a z score
## in order to make an interaction term you do the following (variable name * variable name 2) within the regression model, do not attempt to make a unique variable
## no strata value, does not effect standard errors that much##Selecting my variables
mergeddata4$d.event<-ifelse(mergeddata4$H3TO130==1,1,0) #did they think about it
mergeddata4$d.age<-ifelse(mergeddata4$H3TO130==1,mergeddata4$IYEAR3-(mergeddata4$H3OD1Y+1), NA)
myvars<-c( "AID", "age_3", "male_3", "race_3", "educ_re3", "own_gun", "try_nth", "l4excite", "dthfmoment", "excite_lc", "like_no_res", "follow_ins", "sui_1", "sui_3", "sui_4", "d.event", "d.age", "GSWGT134", "CLUSTER2")
mergeddata4<-mergeddata4[,myvars]
##subsetting data
sam <- mergeddata4 %>%
filter(complete.cases(.))
##suicide transition variable
sam<-sam %>%
mutate(suitran_1 =ifelse(sui_1==0 & sui_3 ==0, 0,1),
sui_tran2 =ifelse(suitran_1==1, NA,
ifelse(sui_3==0 & sui_4==0,0,1)))
#suicide thought transition based on self reported education at wave 3]
table1(~ d.age + male_3 + educ_re3 + race_3 + own_gun + try_nth + l4excite + dthfmoment + excite_lc + like_no_res + follow_ins| d.event, data=sam, overall="Total")## Warning in table1.formula(~d.age + male_3 + educ_re3 + race_3 + own_gun + :
## Terms to the right of '|' in formula 'x' define table columns and are expected
## to be factors with meaningful labels.| 1 (N=261) |
Total (N=261) |
|
|---|---|---|
| d.age | ||
| Mean (SD) | 20.7 (1.91) | 20.7 (1.91) |
| Median [Min, Max] | 21.0 [18.0, 26.0] | 21.0 [18.0, 26.0] |
| BIOLOGICAL SEX-W3 | ||
| Mean (SD) | 0.421 (0.495) | 0.421 (0.495) |
| Median [Min, Max] | 0 [0, 1.00] | 0 [0, 1.00] |
| educ_re3 | ||
| col | 30 (11.5%) | 30 (11.5%) |
| high | 79 (30.3%) | 79 (30.3%) |
| lesshigh | 39 (14.9%) | 39 (14.9%) |
| somecol | 113 (43.3%) | 113 (43.3%) |
| race_3 | ||
| nhblack | 52 (19.9%) | 52 (19.9%) |
| nwhite | 202 (77.4%) | 202 (77.4%) |
| other | 7 (2.7%) | 7 (2.7%) |
| S26Q14 DO YOU OWN A HANDGUN-W3 | ||
| Mean (SD) | 0.111 (0.315) | 0.111 (0.315) |
| Median [Min, Max] | 0 [0, 1.00] | 0 [0, 1.00] |
| try_nth | ||
| not_true | 51 (19.5%) | 51 (19.5%) |
| true | 105 (40.2%) | 105 (40.2%) |
| very_true | 105 (40.2%) | 105 (40.2%) |
| l4excite | ||
| not_true | 39 (14.9%) | 39 (14.9%) |
| true | 99 (37.9%) | 99 (37.9%) |
| very_true | 123 (47.1%) | 123 (47.1%) |
| dthfmoment | ||
| not_true | 10 (3.8%) | 10 (3.8%) |
| true | 88 (33.7%) | 88 (33.7%) |
| very_true | 163 (62.5%) | 163 (62.5%) |
| excite_lc | ||
| not_true | 93 (35.6%) | 93 (35.6%) |
| true | 107 (41.0%) | 107 (41.0%) |
| very_true | 61 (23.4%) | 61 (23.4%) |
| like_no_res | ||
| not_true | 50 (19.2%) | 50 (19.2%) |
| true | 105 (40.2%) | 105 (40.2%) |
| very_true | 106 (40.6%) | 106 (40.6%) |
| follow_ins | ||
| not_true | 67 (25.7%) | 67 (25.7%) |
| true | 114 (43.7%) | 114 (43.7%) |
| very_true | 80 (30.7%) | 80 (30.7%) |
##You must form a person-period data set
subpp<-survSplit(Surv(d.age, d.event)~.,
data = sam, cut=seq(19,28,1),
episode ="age")
options(survey.lonely.psu = "adjust")
des<-svydesign(ids =~CLUSTER2, weights = ~GSWGT134,
data = subpp, nest = T)m0<-svyglm(d.event~male_3,
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!summary(m0)##
## Call:
## svyglm(formula = d.event ~ male_3, design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.7575 0.0888 -8.530 7.92e-14 ***
## male_3 -0.1611 0.1124 -1.434 0.154
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1.001355)
##
## Number of Fisher Scoring iterations: 5m1<-svyglm(d.event~as.factor(educ_re3),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!summary(m1)##
## Call:
## svyglm(formula = d.event ~ as.factor(educ_re3), design = des,
## family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.34498 0.06666 -20.177 < 2e-16 ***
## as.factor(educ_re3)high 0.47035 0.12834 3.665 0.000382 ***
## as.factor(educ_re3)lesshigh 0.61338 0.16572 3.701 0.000337 ***
## as.factor(educ_re3)somecol 0.68674 0.12976 5.292 6.23e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1.001355)
##
## Number of Fisher Scoring iterations: 5m2<-svyglm(d.event~as.factor(d.age),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!summary(m2)##
## Call:
## svyglm(formula = d.event ~ as.factor(d.age), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.826798 0.012690 222.756 < 2e-16 ***
## as.factor(d.age)19 -4.411070 0.191521 -23.032 < 2e-16 ***
## as.factor(d.age)20 -3.956200 0.202632 -19.524 < 2e-16 ***
## as.factor(d.age)21 -3.482702 0.168172 -20.709 < 2e-16 ***
## as.factor(d.age)22 -3.738893 0.230003 -16.256 < 2e-16 ***
## as.factor(d.age)23 -3.133563 0.205580 -15.243 < 2e-16 ***
## as.factor(d.age)24 -2.971796 0.301665 -9.851 < 2e-16 ***
## as.factor(d.age)25 -1.988362 0.419982 -4.734 6.9e-06 ***
## as.factor(d.age)26 -0.005027 0.061349 -0.082 0.935
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9497283)
##
## Number of Fisher Scoring iterations: 14m3<-svyglm(d.event~as.factor(race_3),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!summary(m3)##
## Call:
## svyglm(formula = d.event ~ as.factor(race_3), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.9982 0.1524 -6.548 1.87e-09 ***
## as.factor(race_3)nwhite 0.1852 0.1705 1.086 0.280
## as.factor(race_3)other 0.2001 0.2311 0.866 0.388
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1.001355)
##
## Number of Fisher Scoring iterations: 5m4<-svyglm(d.event~own_gun,
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!summary(m4)##
## Call:
## svyglm(formula = d.event ~ own_gun, design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.83404 0.07303 -11.421 <2e-16 ***
## own_gun -0.03617 0.17134 -0.211 0.833
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1.001355)
##
## Number of Fisher Scoring iterations: 5m5<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurredsummary(m5)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5729 0.3082 8.348 4.29e-13 ***
## male_3 -0.1900 0.1594 -1.192 0.236018
## as.factor(educ_re3)high 0.2550 0.2170 1.175 0.242693
## as.factor(educ_re3)lesshigh 0.4980 0.2205 2.259 0.026085 *
## as.factor(educ_re3)somecol 0.5857 0.1746 3.354 0.001129 **
## as.factor(d.age)19 -4.6991 0.3416 -13.757 < 2e-16 ***
## as.factor(d.age)20 -4.2263 0.3442 -12.279 < 2e-16 ***
## as.factor(d.age)21 -3.7334 0.3094 -12.068 < 2e-16 ***
## as.factor(d.age)22 -3.9453 0.3497 -11.283 < 2e-16 ***
## as.factor(d.age)23 -3.3103 0.3228 -10.255 < 2e-16 ***
## as.factor(d.age)24 -3.1469 0.4126 -7.627 1.49e-11 ***
## as.factor(d.age)25 -2.0343 0.5173 -3.933 0.000156 ***
## as.factor(d.age)26 0.2433 0.1062 2.291 0.024110 *
## as.factor(race_3)nwhite 0.2487 0.2498 0.996 0.321745
## as.factor(race_3)other 0.5629 0.3340 1.685 0.095054 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9310614)
##
## Number of Fisher Scoring iterations: 15m6<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+own_gun,
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m6)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + own_gun, design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.56857 0.30716 8.362 4.25e-13 ***
## male_3 -0.18575 0.15858 -1.171 0.244292
## as.factor(educ_re3)high 0.25737 0.21942 1.173 0.243651
## as.factor(educ_re3)lesshigh 0.50110 0.22477 2.229 0.028073 *
## as.factor(educ_re3)somecol 0.58874 0.17672 3.331 0.001220 **
## as.factor(d.age)19 -4.69142 0.34203 -13.717 < 2e-16 ***
## as.factor(d.age)20 -4.21857 0.34469 -12.239 < 2e-16 ***
## as.factor(d.age)21 -3.72523 0.31339 -11.887 < 2e-16 ***
## as.factor(d.age)22 -3.93610 0.35135 -11.203 < 2e-16 ***
## as.factor(d.age)23 -3.30149 0.33092 -9.977 < 2e-16 ***
## as.factor(d.age)24 -3.13980 0.41686 -7.532 2.49e-11 ***
## as.factor(d.age)25 -2.02994 0.52246 -3.885 0.000186 ***
## as.factor(d.age)26 0.24109 0.10829 2.226 0.028289 *
## as.factor(race_3)nwhite 0.24590 0.25148 0.978 0.330565
## as.factor(race_3)other 0.55569 0.33963 1.636 0.105018
## own_gun -0.05687 0.23750 -0.239 0.811269
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9309247)
##
## Number of Fisher Scoring iterations: 15m7<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+as.factor(try_nth),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m7)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + as.factor(try_nth), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.724105 0.369750 7.367 5.78e-11 ***
## male_3 -0.194585 0.162902 -1.194 0.235197
## as.factor(educ_re3)high 0.223005 0.231920 0.962 0.338662
## as.factor(educ_re3)lesshigh 0.521436 0.222537 2.343 0.021164 *
## as.factor(educ_re3)somecol 0.585238 0.187733 3.117 0.002401 **
## as.factor(d.age)19 -4.815657 0.388382 -12.399 < 2e-16 ***
## as.factor(d.age)20 -4.340652 0.368237 -11.788 < 2e-16 ***
## as.factor(d.age)21 -3.842095 0.338760 -11.342 < 2e-16 ***
## as.factor(d.age)22 -4.051494 0.366484 -11.055 < 2e-16 ***
## as.factor(d.age)23 -3.415553 0.356324 -9.586 1.05e-15 ***
## as.factor(d.age)24 -3.244226 0.456353 -7.109 1.99e-10 ***
## as.factor(d.age)25 -2.133057 0.531514 -4.013 0.000118 ***
## as.factor(d.age)26 0.128747 0.226320 0.569 0.570759
## as.factor(race_3)nwhite 0.270489 0.234967 1.151 0.252488
## as.factor(race_3)other 0.637521 0.322423 1.977 0.050848 .
## as.factor(try_nth)true -0.131745 0.182067 -0.724 0.471049
## as.factor(try_nth)very_true 0.007263 0.176127 0.041 0.967191
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.934304)
##
## Number of Fisher Scoring iterations: 15m8<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+as.factor(l4excite),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m8)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + as.factor(l4excite), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.2782 0.3373 6.755 1.06e-09 ***
## male_3 -0.2962 0.1777 -1.667 0.09872 .
## as.factor(educ_re3)high 0.2815 0.2365 1.190 0.23692
## as.factor(educ_re3)lesshigh 0.4577 0.2260 2.026 0.04556 *
## as.factor(educ_re3)somecol 0.5777 0.2054 2.813 0.00594 **
## as.factor(d.age)19 -4.6895 0.2912 -16.106 < 2e-16 ***
## as.factor(d.age)20 -4.2057 0.2580 -16.299 < 2e-16 ***
## as.factor(d.age)21 -3.7014 0.2772 -13.354 < 2e-16 ***
## as.factor(d.age)22 -3.9103 0.2918 -13.402 < 2e-16 ***
## as.factor(d.age)23 -3.2675 0.2984 -10.951 < 2e-16 ***
## as.factor(d.age)24 -3.0786 0.3841 -8.015 2.50e-12 ***
## as.factor(d.age)25 -1.8348 0.4147 -4.424 2.53e-05 ***
## as.factor(d.age)26 0.6177 0.2974 2.077 0.04042 *
## as.factor(race_3)nwhite 0.1076 0.2204 0.488 0.62642
## as.factor(race_3)other 0.3572 0.3192 1.119 0.26592
## as.factor(l4excite)true 0.4953 0.2427 2.041 0.04399 *
## as.factor(l4excite)very_true 0.5583 0.2429 2.299 0.02365 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9237147)
##
## Number of Fisher Scoring iterations: 15m9<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+as.factor(dthfmoment),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m9)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + as.factor(dthfmoment), design = des,
## family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.11543 0.51194 4.132 7.63e-05 ***
## male_3 -0.33076 0.16615 -1.991 0.04932 *
## as.factor(educ_re3)high 0.22958 0.22572 1.017 0.31162
## as.factor(educ_re3)lesshigh 0.40657 0.23706 1.715 0.08952 .
## as.factor(educ_re3)somecol 0.55552 0.19768 2.810 0.00599 **
## as.factor(d.age)19 -4.89135 0.39754 -12.304 < 2e-16 ***
## as.factor(d.age)20 -4.40391 0.38292 -11.501 < 2e-16 ***
## as.factor(d.age)21 -3.90076 0.35967 -10.845 < 2e-16 ***
## as.factor(d.age)22 -4.11254 0.39394 -10.440 < 2e-16 ***
## as.factor(d.age)23 -3.45211 0.39117 -8.825 4.60e-14 ***
## as.factor(d.age)24 -3.23887 0.44810 -7.228 1.13e-10 ***
## as.factor(d.age)25 -2.05059 0.57716 -3.553 0.00059 ***
## as.factor(d.age)26 0.26437 0.09395 2.814 0.00592 **
## as.factor(race_3)nwhite 0.26336 0.24922 1.057 0.29325
## as.factor(race_3)other 0.55370 0.33860 1.635 0.10523
## as.factor(dthfmoment)true 0.60265 0.37927 1.589 0.11532
## as.factor(dthfmoment)very_true 0.84359 0.32731 2.577 0.01146 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9285278)
##
## Number of Fisher Scoring iterations: 15m10<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+as.factor(excite_lc),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m10)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + as.factor(excite_lc), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.4606 0.3493 7.044 2.70e-10 ***
## male_3 -0.2205 0.1631 -1.352 0.17949
## as.factor(educ_re3)high 0.2337 0.2230 1.048 0.29711
## as.factor(educ_re3)lesshigh 0.4285 0.2317 1.849 0.06751 .
## as.factor(educ_re3)somecol 0.5362 0.1932 2.776 0.00661 **
## as.factor(d.age)19 -4.6680 0.2833 -16.475 < 2e-16 ***
## as.factor(d.age)20 -4.1921 0.2964 -14.144 < 2e-16 ***
## as.factor(d.age)21 -3.6925 0.2777 -13.298 < 2e-16 ***
## as.factor(d.age)22 -3.8984 0.3169 -12.301 < 2e-16 ***
## as.factor(d.age)23 -3.2609 0.3046 -10.705 < 2e-16 ***
## as.factor(d.age)24 -3.0866 0.4011 -7.695 1.19e-11 ***
## as.factor(d.age)25 -1.9304 0.3986 -4.843 4.84e-06 ***
## as.factor(d.age)26 0.4074 0.2208 1.845 0.06804 .
## as.factor(race_3)nwhite 0.2234 0.2385 0.937 0.35106
## as.factor(race_3)other 0.6379 0.3418 1.866 0.06500 .
## as.factor(excite_lc)true 0.2378 0.1800 1.322 0.18944
## as.factor(excite_lc)very_true 0.1946 0.2570 0.757 0.45088
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9322157)
##
## Number of Fisher Scoring iterations: 15m11<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+as.factor(l4excite),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m11)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + as.factor(l4excite), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.2782 0.3373 6.755 1.06e-09 ***
## male_3 -0.2962 0.1777 -1.667 0.09872 .
## as.factor(educ_re3)high 0.2815 0.2365 1.190 0.23692
## as.factor(educ_re3)lesshigh 0.4577 0.2260 2.026 0.04556 *
## as.factor(educ_re3)somecol 0.5777 0.2054 2.813 0.00594 **
## as.factor(d.age)19 -4.6895 0.2912 -16.106 < 2e-16 ***
## as.factor(d.age)20 -4.2057 0.2580 -16.299 < 2e-16 ***
## as.factor(d.age)21 -3.7014 0.2772 -13.354 < 2e-16 ***
## as.factor(d.age)22 -3.9103 0.2918 -13.402 < 2e-16 ***
## as.factor(d.age)23 -3.2675 0.2984 -10.951 < 2e-16 ***
## as.factor(d.age)24 -3.0786 0.3841 -8.015 2.50e-12 ***
## as.factor(d.age)25 -1.8348 0.4147 -4.424 2.53e-05 ***
## as.factor(d.age)26 0.6177 0.2974 2.077 0.04042 *
## as.factor(race_3)nwhite 0.1076 0.2204 0.488 0.62642
## as.factor(race_3)other 0.3572 0.3192 1.119 0.26592
## as.factor(l4excite)true 0.4953 0.2427 2.041 0.04399 *
## as.factor(l4excite)very_true 0.5583 0.2429 2.299 0.02365 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9237147)
##
## Number of Fisher Scoring iterations: 15m12<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+as.factor(like_no_res),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m12)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + as.factor(like_no_res), design = des,
## family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.3820 0.3424 6.957 4.09e-10 ***
## male_3 -0.3151 0.2128 -1.481 0.141954
## as.factor(educ_re3)high 0.1758 0.2407 0.731 0.466822
## as.factor(educ_re3)lesshigh 0.3490 0.2483 1.405 0.163093
## as.factor(educ_re3)somecol 0.5248 0.1923 2.728 0.007557 **
## as.factor(d.age)19 -4.7731 0.4172 -11.440 < 2e-16 ***
## as.factor(d.age)20 -4.2848 0.4090 -10.475 < 2e-16 ***
## as.factor(d.age)21 -3.7677 0.3667 -10.273 < 2e-16 ***
## as.factor(d.age)22 -3.9748 0.3760 -10.572 < 2e-16 ***
## as.factor(d.age)23 -3.3509 0.3393 -9.876 2.49e-16 ***
## as.factor(d.age)24 -3.1508 0.4240 -7.431 4.25e-11 ***
## as.factor(d.age)25 -1.8497 0.4661 -3.968 0.000139 ***
## as.factor(d.age)26 0.6385 0.2069 3.086 0.002644 **
## as.factor(race_3)nwhite 0.2053 0.2413 0.851 0.396920
## as.factor(race_3)other 0.6186 0.3067 2.017 0.046465 *
## as.factor(like_no_res)true 0.3886 0.1793 2.168 0.032620 *
## as.factor(like_no_res)very_true 0.6335 0.2675 2.368 0.019868 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9230586)
##
## Number of Fisher Scoring iterations: 15m13<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+as.factor(follow_ins),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m13)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + as.factor(follow_ins), design = des,
## family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.3804 0.3547 6.711 1.30e-09 ***
## male_3 -0.2962 0.1666 -1.778 0.078471 .
## as.factor(educ_re3)high 0.1343 0.2388 0.562 0.575189
## as.factor(educ_re3)lesshigh 0.3796 0.2295 1.654 0.101332
## as.factor(educ_re3)somecol 0.4790 0.2123 2.256 0.026316 *
## as.factor(d.age)19 -4.7174 0.3782 -12.473 < 2e-16 ***
## as.factor(d.age)20 -4.2307 0.3643 -11.614 < 2e-16 ***
## as.factor(d.age)21 -3.7225 0.3316 -11.224 < 2e-16 ***
## as.factor(d.age)22 -3.9241 0.3740 -10.492 < 2e-16 ***
## as.factor(d.age)23 -3.2680 0.3550 -9.207 6.92e-15 ***
## as.factor(d.age)24 -3.0453 0.4366 -6.975 3.76e-10 ***
## as.factor(d.age)25 -1.9307 0.5680 -3.399 0.000982 ***
## as.factor(d.age)26 0.2459 0.1139 2.158 0.033384 *
## as.factor(race_3)nwhite 0.2449 0.2597 0.943 0.347974
## as.factor(race_3)other 0.5354 0.3740 1.431 0.155568
## as.factor(follow_ins)true 0.4169 0.1765 2.362 0.020197 *
## as.factor(follow_ins)very_true 0.5095 0.1947 2.616 0.010308 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.928674)
##
## Number of Fisher Scoring iterations: 15m14<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(race_3)+own_gun+as.factor(try_nth)+
as.factor(l4excite)+as.factor(dthfmoment)+as.factor(excite_lc)+as.factor(l4excite)
+as.factor(like_no_res)+as.factor(follow_ins),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!
## Warning in eval(family$initialize): glm.fit: fitted probabilities numerically 0
## or 1 occurredsummary(m14)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(race_3) + own_gun + as.factor(try_nth) + as.factor(l4excite) +
## as.factor(dthfmoment) + as.factor(excite_lc) + as.factor(l4excite) +
## as.factor(like_no_res) + as.factor(follow_ins), design = des,
## family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.08700 0.59688 3.497 0.000748 ***
## male_3 -0.47571 0.18796 -2.531 0.013195 *
## as.factor(educ_re3)high 0.11206 0.25250 0.444 0.658316
## as.factor(educ_re3)lesshigh 0.26147 0.28206 0.927 0.356520
## as.factor(educ_re3)somecol 0.50698 0.23188 2.186 0.031504 *
## as.factor(d.age)19 -5.04616 0.39842 -12.665 < 2e-16 ***
## as.factor(d.age)20 -4.52966 0.40767 -11.111 < 2e-16 ***
## as.factor(d.age)21 -3.98427 0.35246 -11.304 < 2e-16 ***
## as.factor(d.age)22 -4.16667 0.41169 -10.121 2.59e-16 ***
## as.factor(d.age)23 -3.47135 0.39807 -8.720 1.82e-13 ***
## as.factor(d.age)24 -3.19330 0.43846 -7.283 1.46e-10 ***
## as.factor(d.age)25 -1.87264 0.54015 -3.467 0.000824 ***
## as.factor(d.age)26 0.54212 0.37776 1.435 0.154891
## as.factor(race_3)nwhite 0.24168 0.23885 1.012 0.314452
## as.factor(race_3)other 0.51466 0.37211 1.383 0.170211
## own_gun -0.04117 0.28331 -0.145 0.884811
## as.factor(try_nth)true -0.23238 0.18741 -1.240 0.218363
## as.factor(try_nth)very_true -0.28079 0.23269 -1.207 0.230850
## as.factor(l4excite)true 0.41641 0.22748 1.831 0.070634 .
## as.factor(l4excite)very_true 0.29408 0.25807 1.140 0.257642
## as.factor(dthfmoment)true 0.34498 0.38407 0.898 0.371575
## as.factor(dthfmoment)very_true 0.49455 0.32962 1.500 0.137178
## as.factor(excite_lc)true -0.14562 0.22092 -0.659 0.511557
## as.factor(excite_lc)very_true -0.42322 0.32824 -1.289 0.200735
## as.factor(like_no_res)true 0.38428 0.20192 1.903 0.060364 .
## as.factor(like_no_res)very_true 0.68307 0.28073 2.433 0.017036 *
## as.factor(follow_ins)true 0.27083 0.19573 1.384 0.170034
## as.factor(follow_ins)very_true 0.38824 0.22741 1.707 0.091384 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9313429)
##
## Number of Fisher Scoring iterations: 15exp(coef(m0))## (Intercept) male_3
## 0.4688331 0.8511983exp(coef(m1))## (Intercept) as.factor(educ_re3)high
## 0.2605446 1.6005564
## as.factor(educ_re3)lesshigh as.factor(educ_re3)somecol
## 1.8466621 1.9872232exp(coef(m2))## (Intercept) as.factor(d.age)19 as.factor(d.age)20 as.factor(d.age)21
## 16.89128939 0.01214218 0.01913570 0.03072427
## as.factor(d.age)22 as.factor(d.age)23 as.factor(d.age)24 as.factor(d.age)25
## 0.02378042 0.04356230 0.05121125 0.13691952
## as.factor(d.age)26
## 0.99498572exp(coef(m3))## (Intercept) as.factor(race_3)nwhite as.factor(race_3)other
## 0.3685483 1.2034672 1.2215796exp(coef(m4))## (Intercept) own_gun
## 0.4342926 0.9644762exp(coef(m5))## (Intercept) male_3
## 13.104289821 0.826959221
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.290479611 1.645451473
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.796286840 0.009103809
## as.factor(d.age)20 as.factor(d.age)21
## 0.014607054 0.023912177
## as.factor(d.age)22 as.factor(d.age)23
## 0.019345003 0.036503499
## as.factor(d.age)24 as.factor(d.age)25
## 0.042985751 0.130769855
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.275474088 1.282381224
## as.factor(race_3)other
## 1.755756263exp(coef(m6))## (Intercept) male_3
## 13.047131320 0.830478003
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.293523509 1.650540294
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.801710248 0.009173614
## as.factor(d.age)20 as.factor(d.age)21
## 0.014719659 0.024107635
## as.factor(d.age)22 as.factor(d.age)23
## 0.019524146 0.036828423
## as.factor(d.age)24 as.factor(d.age)25
## 0.043291476 0.131343107
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.272632114 1.278773102
## as.factor(race_3)other own_gun
## 1.743138940 0.944720932exp(coef(m7))## (Intercept) male_3
## 15.242759298 0.823176071
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.249827120 1.684444177
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.795418084 0.008101895
## as.factor(d.age)20 as.factor(d.age)21
## 0.013028032 0.021448610
## as.factor(d.age)22 as.factor(d.age)23
## 0.017396359 0.032858232
## as.factor(d.age)24 as.factor(d.age)25
## 0.038998744 0.118474598
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.137402290 1.310605297
## as.factor(race_3)other as.factor(try_nth)true
## 1.891786022 0.876564364
## as.factor(try_nth)very_true
## 1.007289654exp(coef(m8))## (Intercept) male_3
## 9.759379151 0.743655497
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.325136533 1.580503844
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.781950450 0.009190914
## as.factor(d.age)20 as.factor(d.age)21
## 0.014909664 0.024689107
## as.factor(d.age)22 as.factor(d.age)23
## 0.020034743 0.038100992
## as.factor(d.age)24 as.factor(d.age)25
## 0.046023350 0.159644824
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.854666216 1.113657476
## as.factor(race_3)other as.factor(l4excite)true
## 1.429286093 1.641035686
## as.factor(l4excite)very_true
## 1.747761282exp(coef(m9))## (Intercept) male_3
## 8.293175352 0.718375958
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.258072902 1.501658606
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.742850209 0.007511284
## as.factor(d.age)20 as.factor(d.age)21
## 0.012229421 0.020226497
## as.factor(d.age)22 as.factor(d.age)23
## 0.016366098 0.031678630
## as.factor(d.age)24 as.factor(d.age)25
## 0.039208172 0.128658355
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.302614509 1.301298985
## as.factor(race_3)other as.factor(dthfmoment)true
## 1.739672380 1.826947279
## as.factor(dthfmoment)very_true
## 2.324690217exp(coef(m10))## (Intercept) male_3
## 11.71181143 0.80214358
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.26329413 1.53491448
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.70948302 0.00939070
## as.factor(d.age)20 as.factor(d.age)21
## 0.01511421 0.02491032
## as.factor(d.age)22 as.factor(d.age)23
## 0.02027469 0.03835229
## as.factor(d.age)24 as.factor(d.age)25
## 0.04565784 0.14508760
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.50293331 1.25037401
## as.factor(race_3)other as.factor(excite_lc)true
## 1.89244351 1.26849838
## as.factor(excite_lc)very_true
## 1.21481010exp(coef(m11))## (Intercept) male_3
## 9.759379151 0.743655497
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.325136533 1.580503844
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.781950450 0.009190914
## as.factor(d.age)20 as.factor(d.age)21
## 0.014909664 0.024689107
## as.factor(d.age)22 as.factor(d.age)23
## 0.020034743 0.038100992
## as.factor(d.age)24 as.factor(d.age)25
## 0.046023350 0.159644824
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.854666216 1.113657476
## as.factor(race_3)other as.factor(l4excite)true
## 1.429286093 1.641035686
## as.factor(l4excite)very_true
## 1.747761282exp(coef(m12))## (Intercept) male_3
## 10.826732346 0.729721933
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.192242840 1.417681531
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.690100665 0.008453884
## as.factor(d.age)20 as.factor(d.age)21
## 0.013776586 0.023105989
## as.factor(d.age)22 as.factor(d.age)23
## 0.018783100 0.035052347
## as.factor(d.age)24 as.factor(d.age)25
## 0.042818784 0.157277729
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.893654954 1.227897183
## as.factor(race_3)other as.factor(like_no_res)true
## 1.856373668 1.474977486
## as.factor(like_no_res)very_true
## 1.884172193exp(coef(m13))## (Intercept) male_3
## 10.809199486 0.743642095
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.143738817 1.461667211
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.614463187 0.008938456
## as.factor(d.age)20 as.factor(d.age)21
## 0.014542384 0.024174400
## as.factor(d.age)22 as.factor(d.age)23
## 0.019759135 0.038080813
## as.factor(d.age)24 as.factor(d.age)25
## 0.047581648 0.145049566
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.278737430 1.277479753
## as.factor(race_3)other as.factor(follow_ins)true
## 1.708065210 1.517241229
## as.factor(follow_ins)very_true
## 1.664515043exp(coef(m14))## (Intercept) male_3
## 8.060711242 0.621444487
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.118574891 1.298842119
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.660271902 0.006434008
## as.factor(d.age)20 as.factor(d.age)21
## 0.010784354 0.018605929
## as.factor(d.age)22 as.factor(d.age)23
## 0.015503833 0.031075037
## as.factor(d.age)24 as.factor(d.age)25
## 0.041036148 0.153717787
## as.factor(d.age)26 as.factor(race_3)nwhite
## 1.719649773 1.273385610
## as.factor(race_3)other own_gun
## 1.673077319 0.959669140
## as.factor(try_nth)true as.factor(try_nth)very_true
## 0.792642657 0.755189239
## as.factor(l4excite)true as.factor(l4excite)very_true
## 1.516509698 1.341892440
## as.factor(dthfmoment)true as.factor(dthfmoment)very_true
## 1.411957811 1.639760946
## as.factor(excite_lc)true as.factor(excite_lc)very_true
## 0.864486779 0.654937453
## as.factor(like_no_res)true as.factor(like_no_res)very_true
## 1.468559449 1.979949084
## as.factor(follow_ins)true as.factor(follow_ins)very_true
## 1.311045835 1.474388806#interaction terms
m15<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+male_3*as.factor(educ_re3),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!summary(m15)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## male_3 * as.factor(educ_re3), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.83580 0.32148 8.821 4.36e-14 ***
## male_3 -0.28664 0.19057 -1.504 0.1358
## as.factor(educ_re3)high 0.28188 0.24756 1.139 0.2576
## as.factor(educ_re3)lesshigh 0.35046 0.31383 1.117 0.2668
## as.factor(educ_re3)somecol 0.48707 0.22304 2.184 0.0314 *
## as.factor(d.age)19 -4.68666 0.31997 -14.647 < 2e-16 ***
## as.factor(d.age)20 -4.21397 0.33552 -12.559 < 2e-16 ***
## as.factor(d.age)21 -3.72616 0.32233 -11.560 < 2e-16 ***
## as.factor(d.age)22 -3.94887 0.34982 -11.288 < 2e-16 ***
## as.factor(d.age)23 -3.31261 0.35733 -9.270 4.64e-15 ***
## as.factor(d.age)24 -3.12911 0.41538 -7.533 2.48e-11 ***
## as.factor(d.age)25 -2.04093 0.44153 -4.622 1.16e-05 ***
## as.factor(d.age)26 0.12441 0.06423 1.937 0.0556 .
## male_3:as.factor(educ_re3)high -0.07418 0.37258 -0.199 0.8426
## male_3:as.factor(educ_re3)lesshigh 0.30791 0.44711 0.689 0.4927
## male_3:as.factor(educ_re3)somecol 0.23027 0.27565 0.835 0.4055
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9321839)
##
## Number of Fisher Scoring iterations: 15m16<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+male_3*as.factor(d.age),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!summary(m16)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## male_3 * as.factor(d.age), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.50759 0.20903 11.996 < 2e-16 ***
## male_3 0.03832 0.03998 0.959 0.34025
## as.factor(educ_re3)high 0.27864 0.21362 1.304 0.19530
## as.factor(educ_re3)lesshigh 0.51511 0.22254 2.315 0.02281 *
## as.factor(educ_re3)somecol 0.62668 0.17463 3.589 0.00053 ***
## as.factor(d.age)19 -4.37873 0.23241 -18.840 < 2e-16 ***
## as.factor(d.age)20 -3.95035 0.28378 -13.920 < 2e-16 ***
## as.factor(d.age)21 -3.54076 0.26498 -13.362 < 2e-16 ***
## as.factor(d.age)22 -3.62195 0.34060 -10.634 < 2e-16 ***
## as.factor(d.age)23 -3.25644 0.36695 -8.874 4.55e-14 ***
## as.factor(d.age)24 -2.89513 0.40378 -7.170 1.69e-10 ***
## as.factor(d.age)25 0.07841 0.05043 1.555 0.12340
## as.factor(d.age)26 0.05672 0.06330 0.896 0.37255
## male_3:as.factor(d.age)19 -0.34951 0.31770 -1.100 0.27409
## male_3:as.factor(d.age)20 -0.23711 0.30267 -0.783 0.43537
## male_3:as.factor(d.age)21 -0.07814 0.33550 -0.233 0.81633
## male_3:as.factor(d.age)22 -0.38785 0.43136 -0.899 0.37088
## male_3:as.factor(d.age)23 0.11683 0.44021 0.265 0.79128
## male_3:as.factor(d.age)24 -0.23319 0.56325 -0.414 0.67982
## male_3:as.factor(d.age)25 -2.31872 0.51623 -4.492 2.01e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9275231)
##
## Number of Fisher Scoring iterations: 15m17<-svyglm(d.event~male_3+as.factor(educ_re3)+as.factor(d.age)+as.factor(educ_re3)*as.factor(d.age),
design=des, family=binomial (link="cloglog"))## Warning in eval(family$initialize): non-integer #successes in a binomial glm!summary(m17)##
## Call:
## svyglm(formula = d.event ~ male_3 + as.factor(educ_re3) + as.factor(d.age) +
## as.factor(educ_re3) * as.factor(d.age), design = des, family = binomial(link = "cloglog"))
##
## Survey design:
## svydesign(ids = ~CLUSTER2, weights = ~GSWGT134, data = subpp,
## nest = T)
##
## Coefficients: (5 not defined because of singularities)
## Estimate Std. Error t value
## (Intercept) 5.48893 0.62443 8.790
## male_3 -0.15791 0.15576 -1.014
## as.factor(educ_re3)high -2.42266 0.53622 -4.518
## as.factor(educ_re3)lesshigh -2.45248 0.54340 -4.513
## as.factor(educ_re3)somecol -2.44116 0.53944 -4.525
## as.factor(d.age)19 -22.85937 0.64575 -35.400
## as.factor(d.age)20 -7.65932 0.82860 -9.244
## as.factor(d.age)21 -7.00199 0.80946 -8.650
## as.factor(d.age)22 -6.62759 0.80884 -8.194
## as.factor(d.age)23 -5.16536 0.59930 -8.619
## as.factor(d.age)24 -2.39999 0.53067 -4.523
## as.factor(d.age)25 0.04710 0.06520 0.722
## as.factor(d.age)26 0.02898 0.06109 0.474
## as.factor(educ_re3)high:as.factor(d.age)19 17.79087 0.71484 24.888
## as.factor(educ_re3)lesshigh:as.factor(d.age)19 18.62796 0.72690 25.627
## as.factor(educ_re3)somecol:as.factor(d.age)19 18.76302 0.66739 28.114
## as.factor(educ_re3)high:as.factor(d.age)20 3.76836 0.84889 4.439
## as.factor(educ_re3)lesshigh:as.factor(d.age)20 3.62357 0.93096 3.892
## as.factor(educ_re3)somecol:as.factor(d.age)20 3.59633 0.80465 4.469
## as.factor(educ_re3)high:as.factor(d.age)21 3.74307 0.79479 4.710
## as.factor(educ_re3)lesshigh:as.factor(d.age)21 3.58306 0.85962 4.168
## as.factor(educ_re3)somecol:as.factor(d.age)21 3.26507 0.88336 3.696
## as.factor(educ_re3)high:as.factor(d.age)22 2.13880 0.95860 2.231
## as.factor(educ_re3)lesshigh:as.factor(d.age)22 2.39471 1.09530 2.186
## as.factor(educ_re3)somecol:as.factor(d.age)22 3.32540 0.84527 3.934
## as.factor(educ_re3)high:as.factor(d.age)23 1.23014 0.80869 1.521
## as.factor(educ_re3)lesshigh:as.factor(d.age)23 1.81943 0.94066 1.934
## as.factor(educ_re3)somecol:as.factor(d.age)23 2.01356 0.73055 2.756
## as.factor(educ_re3)high:as.factor(d.age)24 -1.91503 0.78772 -2.431
## as.factor(educ_re3)lesshigh:as.factor(d.age)24 -0.30781 0.89958 -0.342
## as.factor(educ_re3)high:as.factor(d.age)25 -2.24607 0.46788 -4.800
## as.factor(educ_re3)lesshigh:as.factor(d.age)25 -0.14730 0.20467 -0.720
## Pr(>|t|)
## (Intercept) 1.88e-13 ***
## male_3 0.313636
## as.factor(educ_re3)high 2.07e-05 ***
## as.factor(educ_re3)lesshigh 2.11e-05 ***
## as.factor(educ_re3)somecol 2.02e-05 ***
## as.factor(d.age)19 < 2e-16 ***
## as.factor(d.age)20 2.36e-14 ***
## as.factor(d.age)21 3.57e-13 ***
## as.factor(d.age)22 2.88e-12 ***
## as.factor(d.age)23 4.12e-13 ***
## as.factor(d.age)24 2.04e-05 ***
## as.factor(d.age)25 0.472162
## as.factor(d.age)26 0.636477
## as.factor(educ_re3)high:as.factor(d.age)19 < 2e-16 ***
## as.factor(educ_re3)lesshigh:as.factor(d.age)19 < 2e-16 ***
## as.factor(educ_re3)somecol:as.factor(d.age)19 < 2e-16 ***
## as.factor(educ_re3)high:as.factor(d.age)20 2.79e-05 ***
## as.factor(educ_re3)lesshigh:as.factor(d.age)20 0.000201 ***
## as.factor(educ_re3)somecol:as.factor(d.age)20 2.49e-05 ***
## as.factor(educ_re3)high:as.factor(d.age)21 9.99e-06 ***
## as.factor(educ_re3)lesshigh:as.factor(d.age)21 7.57e-05 ***
## as.factor(educ_re3)somecol:as.factor(d.age)21 0.000394 ***
## as.factor(educ_re3)high:as.factor(d.age)22 0.028400 *
## as.factor(educ_re3)lesshigh:as.factor(d.age)22 0.031639 *
## as.factor(educ_re3)somecol:as.factor(d.age)22 0.000174 ***
## as.factor(educ_re3)high:as.factor(d.age)23 0.132069
## as.factor(educ_re3)lesshigh:as.factor(d.age)23 0.056537 .
## as.factor(educ_re3)somecol:as.factor(d.age)23 0.007204 **
## as.factor(educ_re3)high:as.factor(d.age)24 0.017233 *
## as.factor(educ_re3)lesshigh:as.factor(d.age)24 0.733093
## as.factor(educ_re3)high:as.factor(d.age)25 7.02e-06 ***
## as.factor(educ_re3)lesshigh:as.factor(d.age)25 0.473773
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9009256)
##
## Number of Fisher Scoring iterations: 16exp(coef(m15))## (Intercept) male_3
## 17.044022494 0.750781353
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.325623746 1.419724567
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.627543856 0.009217463
## as.factor(d.age)20 as.factor(d.age)21
## 0.014787479 0.024085168
## as.factor(d.age)22 as.factor(d.age)23
## 0.019276559 0.036420839
## as.factor(d.age)24 as.factor(d.age)25
## 0.043756771 0.129907984
## as.factor(d.age)26 male_3:as.factor(educ_re3)high
## 1.132477347 0.928506602
## male_3:as.factor(educ_re3)lesshigh male_3:as.factor(educ_re3)somecol
## 1.360578145 1.258943363exp(coef(m16))## (Intercept) male_3
## 12.27533306 1.03906649
## as.factor(educ_re3)high as.factor(educ_re3)lesshigh
## 1.32132913 1.67382562
## as.factor(educ_re3)somecol as.factor(d.age)19
## 1.87139494 0.01254125
## as.factor(d.age)20 as.factor(d.age)21
## 0.01924798 0.02899125
## as.factor(d.age)22 as.factor(d.age)23
## 0.02673039 0.03852542
## as.factor(d.age)24 as.factor(d.age)25
## 0.05529196 1.08156175
## as.factor(d.age)26 male_3:as.factor(d.age)19
## 1.05835835 0.70503212
## male_3:as.factor(d.age)20 male_3:as.factor(d.age)21
## 0.78890760 0.92483091
## male_3:as.factor(d.age)22 male_3:as.factor(d.age)23
## 0.67851553 1.12393310
## male_3:as.factor(d.age)24 male_3:as.factor(d.age)25
## 0.79200452 0.09839919exp(coef(m17))## (Intercept)
## 2.419971e+02
## male_3
## 8.539245e-01
## as.factor(educ_re3)high
## 8.868542e-02
## as.factor(educ_re3)lesshigh
## 8.607978e-02
## as.factor(educ_re3)somecol
## 8.705951e-02
## as.factor(d.age)19
## 1.181145e-10
## as.factor(d.age)20
## 4.716274e-04
## as.factor(d.age)21
## 9.100699e-04
## as.factor(d.age)22
## 1.323349e-03
## as.factor(d.age)23
## 5.710994e-03
## as.factor(d.age)24
## 9.071872e-02
## as.factor(d.age)25
## 1.048224e+00
## as.factor(d.age)26
## 1.029406e+00
## as.factor(educ_re3)high:as.factor(d.age)19
## 5.326907e+07
## as.factor(educ_re3)lesshigh:as.factor(d.age)19
## 1.230329e+08
## as.factor(educ_re3)somecol:as.factor(d.age)19
## 1.408236e+08
## as.factor(educ_re3)high:as.factor(d.age)20
## 4.330916e+01
## as.factor(educ_re3)lesshigh:as.factor(d.age)20
## 3.747099e+01
## as.factor(educ_re3)somecol:as.factor(d.age)20
## 3.646407e+01
## as.factor(educ_re3)high:as.factor(d.age)21
## 4.222744e+01
## as.factor(educ_re3)lesshigh:as.factor(d.age)21
## 3.598338e+01
## as.factor(educ_re3)somecol:as.factor(d.age)21
## 2.618200e+01
## as.factor(educ_re3)high:as.factor(d.age)22
## 8.489255e+00
## as.factor(educ_re3)lesshigh:as.factor(d.age)22
## 1.096506e+01
## as.factor(educ_re3)somecol:as.factor(d.age)22
## 2.781017e+01
## as.factor(educ_re3)high:as.factor(d.age)23
## 3.421721e+00
## as.factor(educ_re3)lesshigh:as.factor(d.age)23
## 6.168326e+00
## as.factor(educ_re3)somecol:as.factor(d.age)23
## 7.489951e+00
## as.factor(educ_re3)high:as.factor(d.age)24
## 1.473373e-01
## as.factor(educ_re3)lesshigh:as.factor(d.age)24
## 7.350516e-01
## as.factor(educ_re3)high:as.factor(d.age)25
## 1.058145e-01
## as.factor(educ_re3)lesshigh:as.factor(d.age)25
## 8.630360e-01