setwd("C:/Users/jacob/OneDrive - University of Texas at San Antonio/Courses/Stats for Demographic Data 2/Homework 4")Stats II Homework 5
Initialize
library(purrr)
library(haven)
library(janitor)
Attaching package: 'janitor'The following objects are masked from 'package:stats':
chisq.test, fisher.testlibrary(car)Loading required package: carData
Attaching package: 'car'The following object is masked from 'package:purrr':
somelibrary(usmap)
library(ggthemes)
library(tigris)To enable caching of data, set `options(tigris_use_cache = TRUE)`
in your R script or .Rprofile.library(tidyverse)── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2
──✔ ggplot2 3.4.0 ✔ dplyr 1.0.10
✔ tibble 3.1.8 ✔ stringr 1.5.0
✔ tidyr 1.3.0 ✔ forcats 0.5.2
✔ readr 2.1.3
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag() masks stats::lag()
✖ dplyr::recode() masks car::recode()
✖ car::some() masks purrr::some()library(tidycensus)
library(lme4)Loading required package: Matrix
Attaching package: 'Matrix'
The following objects are masked from 'package:tidyr':
expand, pack, unpacklibrary(haven)
library(foreign)
library(FactoMineR)
library(ggrepel)
library(margins)
library(emmeans)
library(stargazer)
Please cite as:
Hlavac, Marek (2022). stargazer: Well-Formatted Regression and Summary Statistics Tables.
R package version 5.2.3. https://CRAN.R-project.org/package=stargazer library(stringr)
library(gtsummary)#BlackLivesMatterData Import
brfss21 <- haven::read_xpt("LLCP2021.XPT_")
attr(brfss21, which = "labels")NULLnames(brfss21) <- str_replace(names(brfss21),"_","")Explore
tabyl(brfss21$SEXVAR) %>% gt::gt()| brfss21$SEXVAR | n | percent |
|---|---|---|
| 1 | 203810 | 0.4645846 |
| 2 | 234883 | 0.5354154 |
tabyl(brfss21$EDUCAG) %>% gt::gt()| brfss21$EDUCAG | n | percent |
|---|---|---|
| 1 | 25991 | 0.059246443 |
| 2 | 111545 | 0.254266651 |
| 3 | 120102 | 0.273772319 |
| 4 | 178577 | 0.407065989 |
| 9 | 2478 | 0.005648597 |
tabyl(brfss21$RFSMOK3) %>% gt::gt()| brfss21$RFSMOK3 | n | percent |
|---|---|---|
| 1 | 359891 | 0.82037097 |
| 2 | 53832 | 0.12270996 |
| 9 | 24970 | 0.05691908 |
tabyl(brfss21$CHCSCNCR) %>% gt::gt()| brfss21$CHCSCNCR | n | percent | valid_percent |
|---|---|---|---|
| 1 | 41112 | 9.371474e-02 | 0.0937151663 |
| 2 | 396140 | 9.030005e-01 | 0.9030046206 |
| 7 | 1151 | 2.623703e-03 | 0.0026237146 |
| 9 | 288 | 6.564955e-04 | 0.0006564985 |
| NA | 2 | 4.558997e-06 | NA |
Codebook
Skin Cancer (IV)
Educational Attainment (DV)
Smoking Behavior (DV)
Lifetime Depression Diagnosis (IV)
Cleaning
#Remove invalid values. Creates new dataframe.
brfss21.c <- brfss21 %>% filter(EDUCAG %in% c(1:4))
brfss21.c <-filter(brfss21.c, AGEG5YR != 14)
brfss21.c <- filter(brfss21.c, !CHCSCNCR %in% c(7:9))
brfss21.c <- filter(brfss21.c, RFSMOK3 !=9)
brfss21.c <- filter(brfss21.c, !ADDEPEV3 %in% c(7:9))#Data check.
tabyl(brfss21.c$SEXVAR) %>% gt::gt()| brfss21.c$SEXVAR | n | percent |
|---|---|---|
| 1 | 186342 | 0.4638315 |
| 2 | 215403 | 0.5361685 |
tabyl(brfss21.c$EDUCAG) %>% gt::gt()| brfss21.c$EDUCAG | n | percent |
|---|---|---|
| 1 | 23654 | 0.05887814 |
| 2 | 102092 | 0.25412140 |
| 3 | 111065 | 0.27645646 |
| 4 | 164934 | 0.41054400 |
tabyl(brfss21.c$RFSMOK3) %>% gt::gt()| brfss21.c$RFSMOK3 | n | percent |
|---|---|---|
| 1 | 349171 | 0.8691359 |
| 2 | 52574 | 0.1308641 |
tabyl(brfss21.c$CHCSCNCR) %>% gt::gt()| brfss21.c$CHCSCNCR | n | percent |
|---|---|---|
| 1 | 38426 | 0.09564774 |
| 2 | 363319 | 0.90435226 |
[1] "There were 401745 observations found."Recode
brfss21.c$SEXVAR.female <- car::recode(brfss21.c$SEXVAR, recodes = "1 = 0;2=1")
brfss21.c$CHCSCNCR.skincancer<- car::recode(brfss21.c$CHCSCNCR, recodes = "1=1;2=0; else=NA")
brfss21.c$RFSMOK3.smoke <- car::recode(brfss21.c$RFSMOK3, recodes= "1=0;2=1; else=NA")
brfss21.c$EDUCAG.graduate <- car::recode(brfss21.c$EDUCAG, recodes= "1:3 = 0; 4=1")
brfss21.c$ADDEPEV3.depressed <-car::recode(brfss21.c$ADDEPEV3, recodes = "1=1; 2=0")tabyl(brfss21.c$SEXVAR.female) %>% gt::gt()| brfss21.c$SEXVAR.female | n | percent |
|---|---|---|
| 0 | 186342 | 0.4638315 |
| 1 | 215403 | 0.5361685 |
tabyl(brfss21.c$EDUCAG.graduate) %>% gt::gt()| brfss21.c$EDUCAG.graduate | n | percent |
|---|---|---|
| 0 | 236811 | 0.589456 |
| 1 | 164934 | 0.410544 |
tabyl(brfss21.c$RFSMOK3.smoke) %>% gt::gt()| brfss21.c$RFSMOK3.smoke | n | percent |
|---|---|---|
| 0 | 349171 | 0.8691359 |
| 1 | 52574 | 0.1308641 |
tabyl(brfss21.c$CHCSCNCR.skincancer) %>% gt::gt()| brfss21.c$CHCSCNCR.skincancer | n | percent |
|---|---|---|
| 0 | 363319 | 0.90435226 |
| 1 | 38426 | 0.09564774 |
mean(brfss21.c$AGE80)[1] 54.74752mean(brfss21.c$SEXVAR.female)[1] 0.5361685mean(brfss21.c$CHCSCNCR.skincancer)[1] 0.09564774mean(brfss21.c$EDUCAG.graduate)[1] 0.410544mean(brfss21.c$ADDEPEV3.depressed)[1] 0.2006074Specifications of Age
brfss21.c$AGE80SQ<- brfss21.c$AGE80**2
brfss21.c$AGE80CU<- brfss21.c$AGE80**3Lifetime Depression
ageprobs<- brfss21.c %>%
group_by(AGE80) %>%
summarize(p=mean(ADDEPEV3.depressed),n=n())
ageprobs$num <- ageprobs$p*(1-ageprobs$p)
ageprobs$sep <- sqrt(ageprobs$num/ageprobs$n)
ageprobs$me <- 2*ageprobs$sep
ggplot(data =ageprobs, aes(x = AGE80, y = p, ymin=p-me, ymax=p+me)) +
geom_line() + ylim(0,1) + geom_ribbon(alpha=0.3,aes(color=NULL)) +
labs(title="Lifetime Depression by Age",
subtitle="Respondents in Sample",
caption="Source: 2021 BRFSS Data") +
xlab(label="Age at Survey") +
ylab(label="Proportion of Respondents with Lifetime Depression Diagnosis") 
model0 <- glm(ADDEPEV3.depressed ~ AGE80, family = "binomial", data = brfss21.c)
summary(model0)
Call:
glm(formula = ADDEPEV3.depressed ~ AGE80, family = "binomial",
data = brfss21.c)
Deviance Residuals:
Min 1Q Median 3Q Max
-0.8295 -0.7008 -0.6224 -0.5657 1.9552
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.6399188 0.0121749 -52.56 <2e-16 ***
AGE80 -0.0138933 0.0002207 -62.95 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 402744 on 401744 degrees of freedom
Residual deviance: 398783 on 401743 degrees of freedom
AIC: 398787
Number of Fisher Scoring iterations: 4model0$coefficients = exp(model0$coefficients)
summary(model0)
Call:
glm(formula = ADDEPEV3.depressed ~ AGE80, family = "binomial",
data = brfss21.c)
Deviance Residuals:
Min 1Q Median 3Q Max
-0.8295 -0.7008 -0.6224 -0.5657 1.9552
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.5273353 0.0121749 43.31 <2e-16 ***
AGE80 0.9862028 0.0002207 4468.11 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 402744 on 401744 degrees of freedom
Residual deviance: 398783 on 401743 degrees of freedom
AIC: 398787
Number of Fisher Scoring iterations: 4
Note
Interpretation:
There is a 1.5% reduction in the odds of reporting depression for every additional year of life, holding all else constant.
Marginal Effects at Representative Values
data <- brfss21.c %>% select(AGE80,ADDEPEV3.depressed)
model0 <- glm(ADDEPEV3.depressed ~ AGE80, family = "binomial", data = data)
summary(model0)
Call:
glm(formula = ADDEPEV3.depressed ~ AGE80, family = "binomial",
data = data)
Deviance Residuals:
Min 1Q Median 3Q Max
-0.8295 -0.7008 -0.6224 -0.5657 1.9552
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.6399188 0.0121749 -52.56 <2e-16 ***
AGE80 -0.0138933 0.0002207 -62.95 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 402744 on 401744 degrees of freedom
Residual deviance: 398783 on 401743 degrees of freedom
AIC: 398787
Number of Fisher Scoring iterations: 4summary(model0)
Call:
glm(formula = ADDEPEV3.depressed ~ AGE80, family = "binomial",
data = data)
Deviance Residuals:
Min 1Q Median 3Q Max
-0.8295 -0.7008 -0.6224 -0.5657 1.9552
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.6399188 0.0121749 -52.56 <2e-16 ***
AGE80 -0.0138933 0.0002207 -62.95 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 402744 on 401744 degrees of freedom
Residual deviance: 398783 on 401743 degrees of freedom
AIC: 398787
Number of Fisher Scoring iterations: 4MERs<-margins(model0,variables="AGE80", at=list(AGE80=c(18,20,30,40,50,60,70,80)))
MERS.Depressed <- summary(MERs)
MERS.Depressed factor AGE80 AME SE z p lower upper
AGE80 18.0000 -0.0029 0.0001 -52.3316 0.0000 -0.0030 -0.0028
AGE80 20.0000 -0.0028 0.0001 -52.6052 0.0000 -0.0029 -0.0027
AGE80 30.0000 -0.0027 0.0000 -54.4802 0.0000 -0.0028 -0.0026
AGE80 40.0000 -0.0025 0.0000 -57.2950 0.0000 -0.0026 -0.0024
AGE80 50.0000 -0.0023 0.0000 -61.2403 0.0000 -0.0024 -0.0022
AGE80 60.0000 -0.0021 0.0000 -66.6283 0.0000 -0.0022 -0.0020
AGE80 70.0000 -0.0019 0.0000 -73.9670 0.0000 -0.0020 -0.0019
AGE80 80.0000 -0.0018 0.0000 -84.0966 0.0000 -0.0018 -0.0017plot(MERS.Depressed$AGE80,MERS.Depressed$AME)
#Will if not using subset data.
Note
Interpretation
MERs.depressed1YR <-margins(model0,variables="AGE80", at=list(AGE80=c(39,40,41)))
MERs.depressed1YR <- summary(MERs.depressed1YR)
plot(MERs.depressed1YR$AGE80,MERs.depressed1YR$AME)
Note
Interpretation
The marginal effect of aging on lifetime depression diagnosis diminishes across the lifespan.
ageprobs<- brfss21.c %>%
group_by(AGE80) %>%
summarize(p=mean(ADDEPEV3.depressed),n=n())
ageprobs$num <- ageprobs$p*(1-ageprobs$p)
ageprobs$sep <- sqrt(ageprobs$num/ageprobs$n)
ageprobs$me <- 2*ageprobs$sep
ggplot(data =ageprobs, aes(x = AGE80, y = p, ymin=p-me, ymax=p+me)) +
geom_line() + ylim(0,1) + geom_ribbon(alpha=0.3,aes(color=NULL)) +
labs(title="Lifetime Depression by Age",
subtitle="Respondents in Sample",
caption="Source: 2021 BRFSS Data") +
xlab(label="Age at Survey") +
ylab(label="Proportion of Respondents with Lifetime Depression Diagnosis")+
theme_bw()
AMEs<-margins(model0,variables=c("AGE80"))
summary(AMEs) factor AME SE z p lower upper
AGE80 -0.0022 0.0000 -63.5499 0.0000 -0.0023 -0.0021
Note
Interpretation:
The average marginal effect of age across the lifespan is 0.22% reduction in odds of reporting depression.
model1 <- glm(ADDEPEV3.depressed ~ AGE80 + AGE80SQ, family = "binomial", data = brfss21.c)
summary(model1)
Call:
glm(formula = ADDEPEV3.depressed ~ AGE80 + AGE80SQ, family = "binomial",
data = brfss21.c)
Deviance Residuals:
Min 1Q Median 3Q Max
-0.7591 -0.7348 -0.6338 -0.5154 2.0418
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.466e+00 3.321e-02 -44.14 <2e-16 ***
AGE80 2.328e-02 1.399e-03 16.64 <2e-16 ***
AGE80SQ -3.669e-04 1.363e-05 -26.92 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 402744 on 401744 degrees of freedom
Residual deviance: 398045 on 401742 degrees of freedom
AIC: 398051
Number of Fisher Scoring iterations: 4model1 <- glm(ADDEPEV3.depressed ~ AGE80 + AGE80SQ+AGE80CU, family = "binomial", data = brfss21.c)
summary(model1)
Call:
glm(formula = ADDEPEV3.depressed ~ AGE80 + AGE80SQ + AGE80CU,
family = "binomial", data = brfss21.c)
Deviance Residuals:
Min 1Q Median 3Q Max
-0.8032 -0.7222 -0.6537 -0.4861 2.0949
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.269e-01 8.632e-02 -1.471 0.141
AGE80 -7.274e-02 5.906e-03 -12.315 <2e-16 ***
AGE80SQ 1.697e-03 1.244e-04 13.646 <2e-16 ***
AGE80CU -1.366e-05 8.195e-07 -16.667 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 402744 on 401744 degrees of freedom
Residual deviance: 397767 on 401741 degrees of freedom
AIC: 397775
Number of Fisher Scoring iterations: 4
Note
Interpretation:
ageprobs<- brfss21.c %>%
group_by(EDUCAG.graduate, SEXVAR.female,AGE80) %>%
summarize(pDepressed =mean(ADDEPEV3.depressed),n=n())`summarise()` has grouped output by 'EDUCAG.graduate', 'SEXVAR.female'. You can
override using the `.groups` argument.ggplot(data =ageprobs, aes(x = AGE80, y = pDepressed, color = factor(EDUCAG.graduate), group = EDUCAG.graduate)) + geom_line() + facet_wrap(~ SEXVAR.female) + ylim(0,.5)+
ggtitle("Lifetime Depression Diagnosis by Sex and College Graduate Status")+
theme_bw()+
theme(legend.position="bottom")
Note
Interpretation: College education is protective against depression for both men and women. The probability of reporting depression diminishes across the lifespan for both men and women. Women report depression at higher rates at corresponding ages to that of men. The effect of college education on mental health is greater for women than men.
model <- glm(ADDEPEV3.depressed ~ AGE80 + EDUCAG.graduate + SEXVAR.female, family = "binomial", data = brfss21.c)
summary(model)
Call:
glm(formula = ADDEPEV3.depressed ~ AGE80 + EDUCAG.graduate +
SEXVAR.female, family = "binomial", data = brfss21.c)
Deviance Residuals:
Min 1Q Median 3Q Max
-0.9986 -0.7126 -0.5969 -0.4619 2.2178
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.8925466 0.0133414 -66.90 <2e-16 ***
AGE80 -0.0154828 0.0002238 -69.19 <2e-16 ***
EDUCAG.graduate -0.2386695 0.0082599 -28.89 <2e-16 ***
SEXVAR.female 0.7349933 0.0083700 87.81 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 402744 on 401744 degrees of freedom
Residual deviance: 389959 on 401741 degrees of freedom
AIC: 389967
Number of Fisher Scoring iterations: 4modelExp = model
modelExp$coefficients = exp(modelExp$coefficients)
summary(modelExp)
Call:
glm(formula = ADDEPEV3.depressed ~ AGE80 + EDUCAG.graduate +
SEXVAR.female, family = "binomial", data = brfss21.c)
Deviance Residuals:
Min 1Q Median 3Q Max
-0.9986 -0.7126 -0.5969 -0.4619 2.2178
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.4096113 0.0133414 30.70 <2e-16 ***
AGE80 0.9846364 0.0002238 4400.10 <2e-16 ***
EDUCAG.graduate 0.7876751 0.0082599 95.36 <2e-16 ***
SEXVAR.female 2.0854679 0.0083700 249.16 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 402744 on 401744 degrees of freedom
Residual deviance: 389959 on 401741 degrees of freedom
AIC: 389967
Number of Fisher Scoring iterations: 4
Note
Interpretation:
MEMS
emmeans(model, specs = "EDUCAG.graduate",
regrid = "response") EDUCAG.graduate prob SE df asymp.LCL asymp.UCL
0 0.209 0.000826 Inf 0.207 0.210
1 0.173 0.000917 Inf 0.171 0.174
Results are averaged over the levels of: SEXVAR.female
Confidence level used: 0.95
Note
Interpretation: The probability of reporting depression is associated 0.036 pp reduction when evaluated at the means. That is, across all respondents, the effect of being a college graduate is protective and associated with a reduced probability of reporting depression.
for (i in c(20,40,50,60,80))
{
print(paste("At age = ", i))
emmeans(model, specs = "EDUCAG.graduate", by = "SEXVAR.female", at = list(AGE80=i),
regrid = "response") |>
contrast(method = "revpairwise") |>
confint() %>% print()
}[1] "At age = 20"
SEXVAR.female = 0:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0397 0.00136 Inf -0.0423 -0.0370
SEXVAR.female = 1:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0548 0.00188 Inf -0.0585 -0.0511
Confidence level used: 0.95
[1] "At age = 40"
SEXVAR.female = 0:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0327 0.00112 Inf -0.0349 -0.0305
SEXVAR.female = 1:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0491 0.00168 Inf -0.0524 -0.0458
Confidence level used: 0.95
[1] "At age = 50"
SEXVAR.female = 0:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0294 0.00100 Inf -0.0313 -0.0274
SEXVAR.female = 1:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0458 0.00157 Inf -0.0489 -0.0427
Confidence level used: 0.95
[1] "At age = 60"
SEXVAR.female = 0:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0262 0.000897 Inf -0.0280 -0.0245
SEXVAR.female = 1:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0423 0.001448 Inf -0.0452 -0.0395
Confidence level used: 0.95
[1] "At age = 80"
SEXVAR.female = 0:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0206 0.000713 Inf -0.0220 -0.0192
SEXVAR.female = 1:
contrast estimate SE df asymp.LCL asymp.UCL
EDUCAG.graduate1 - EDUCAG.graduate0 -0.0353 0.001212 Inf -0.0376 -0.0329
Confidence level used: 0.95
Note
Interpretation
Holding all else constant, for 40 year old women, the effect of college graduation is -0.0327 pp reduction in the odds of reporting depression.
grad_effect <- predict(model, newdata = data.frame(EDUCAG.graduate=1, SEXVAR.female=brfss21.c$SEXVAR.female,AGE80=brfss21.c$AGE80),
type = "response")
nongrad_effect <- predict(model, newdata = data.frame(EDUCAG.graduate=0, SEXVAR.female=brfss21.c$SEXVAR.female, AGE80=brfss21.c$AGE80),
type = "response")
mean(grad_effect - nongrad_effect)[1] -0.03659929
Note
Interpretation:
The marginal effect graduating college is -0.03659929 pp reduction in the odds of reporting depression.
Excel
Marginal Effects at Representative Values
| Coefficients: | |||
|---|---|---|---|
| b | x | bx | |
| (Intercept) | -0.89255 | 1 | -0.89255 |
| X_AGE80 | -0.01548 | 40 | -0.61931 |
| EDUCAG.graduate | -0.23867 | 0 | 0 |
| SEXVAR.female | 0.734993 | 1 | 0.734993 |
| Graduate | Nongraduate | ||
| Prob | 0.265898 | 0.314996 | |
| Hand | R studio | ||
| Delta | -0.0491 | -0.0491 |
Marginal Effects at Mean
| Coefficients: | |||
| b | x | bx | |
| (Intercept) | -0.89255 | 1 | -0.89255 |
| AGE80 | -0.01548 | 57.74 | -0.89398 |
| EDUCAG.graduate | -0.23867 | 0 | 0 |
| SEXVAr.female | 0.734993 | 0.535316 | 0.393453 |
| 0.77687 | -1.39307 | ||
| 0.459854 | 0.248312 | ||
| 0.314996 | 0.198918 | ||
| Graduates | Nongraduates | ||
| Prob | 0.163592 | 0.198918 | |
| Hand | R studio | ||
| Delta | -0.03533 | -0.0366 |
| Note. |
|---|
Note
Interpretation:
Excel produced very similar output to that of R. This shows that the methods between R and Excel are convergent.