Pandemic’s Long-Term Effects on Young Children’s Social-emotional Developments
Luis Anunciacao, PhD
0.0.1 This Markdown is part of the manuscript
📝 Unseen Consequences: The Pandemic’s Long-Term Effects on Child
Development
đź“Ś Created by: Luis Anunciacao, PhD
Codes are below and data is at https://osf.io/2y7gp/ (available if authorized by the
publisher) 🕒 Last update: 03 março, 2025
1 MS ASQSE pre and post pandemic
2 Load df
2.1 Packages
2.2 Table 1
First table (currently at page 3) Results match! (Lucky me - On Feb 28 2025)
df_full %>%
filter(!is.na(asqse_quest)) %>%
filter(year_completed_cat %in% c("19","20","21","22","23") )%>%
mutate(asqse_quest = as.factor(asqse_quest)) %>%
select(asqse_quest, year_completed_cat, momage, gender, momed, momage, income, race_wbho) %>%
tableby(year_completed_cat ~ ., data =.) %>%
summary(digits = 2, text = T) %>% as.data.frame()2.4 Number of respondents per questionnaire
df_full %>% filter(!is.na(asqse_quest)) %>% group_by(year_completed_cat) %>% count(asqse_quest) %>%
pivot_wider(id_cols = asqse_quest, names_from = year_completed_cat , values_from = n) %>% adorn_totals()2.5 Table with means
df_full %>%
filter(!is.na(asqse_quest)) %>%
select(asqse_quest,year_completed, quest, csum, gmsum, fmsum, cgsum, ps_sum,asqse_total,esq_total) %>%
tableby(year_completed ~ asqse_total, strata = asqse_quest,
control=tableby.control(total=TRUE, cat.simplify=TRUE,
numeric.stats = c("meansd", "N","Nmiss")),
data = .) %>%
summary(text = T, digits=2) %>% as.data.frame()df_full %>%
filter(!is.na(asqse_quest)) %>%
select(asqse_quest,year_completed_cat, quest, csum, gmsum, fmsum, cgsum, ps_sum,asqse_total,esq_total) %>%
tableby(year_completed_cat ~ asqse_total, data = .) %>%
summary(text = T, digits=2) %>% as.data.frame()2.6 Plot with means
df_full %>%
filter(!is.na(year_completed_cat), !is.na(asqse_quest)) %>%
select(asqse_quest, year_completed_cat, asqse_total ) %>%
#filter(asqse_quest == 2) %>%
# get statistics
summarise(mean = mean(asqse_total),
count = n(),
sd = sd(asqse_total),
se = sd/sqrt(n()), .by = year_completed_cat) %>% mutate_if(is.numeric,round,1) %>%
# plot
ggplot(., aes(x = year_completed_cat, y = mean, group = 1)) +
stat_summary(geom = "pointrange", fun = mean, size=.5) +
stat_summary(geom = "line", fun = mean, size=1, linetype=3) +
geom_errorbar(aes(ymin = mean - se, ymax = mean + se), position = position_dodge2(padding = 0.5), width=0.2 ) +
# add text
#geom_col(alpha = .2, aes(fill = year_completed_cat)) +
geom_label(aes(label = paste0( mean,"±",sd, "\n(N=",count, ")") ) , fill = "white", vjust = 0.1) +
labs(y = "ASQ:SE-2", x = "Year the questionnaire was completed") +
theme_bw() ## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## â„ą Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.## Warning: Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
2.7 Plot overall trend among years
for(i in 1:(unique(df_full$asqse_quest))) {
p = df_full %>% #get data
# filter(!is.na(year_completed_cat), !is.na(asqse_quest)) %>% #keep complete obs
# Plot different graphs
filter(asqse_quest == asqse_quest[[i]]) %>%
# get statistics
summarise(mean = mean(asqse_total),
quest_name = paste(unique(asqse_quest), collapse = ","),
count = n(),
sd = sd(asqse_total),
se = sd/sqrt(n()), .by = year_completed_cat) %>% mutate_if(is.numeric,round,1) %>%
# plot
ggplot(., aes(x = year_completed_cat, y = mean, shape = quest_name,
group = 1)) + #color here is a trick to have titles
annotate("rect", xmin = 3.5, xmax = 5.5, ymin = 0, ymax = Inf , fill= "orange",
alpha = 0.2) +
geom_col(alpha = .2) +
stat_summary(geom = "pointrange", fun = mean, size=.5) +
stat_summary(geom = "line", fun = mean, size=1, linetype=3) +
geom_errorbar(aes(ymin = mean - se, ymax = mean + se), position = position_dodge2(padding = 0.5), width=0.2 ) +
# add text
geom_label(aes(label = paste0( mean,"±",sd, "\n(N=",count, ")") ) , fill = "white", vjust = 0.2, size = 3.5) +
labs(y = "ASQ:SE-2", x = "Year the questionnaire was completed", shape = "Age group") +
# add colors
theme_minimal() + theme(legend.position = "top",
legend.justification='left')
#print(p)
ggsave(paste0("myPlot",i,".png"), bg = 'white',plot = p)
}
unique(df_full$asqse_quest)2.8 Deltas
2.8.1 Tables
df_full %>%
filter(!is.na(asqse_quest)) %>%
select(year_completed_cat, asqse_total, asqse_quest) %>%
pivot_wider(id_cols = asqse_quest, names_from = year_completed_cat, values_from = asqse_total,
values_fn = {mean}) %>%
as.data.frame() %>%
setNames(., c("quest","2018","2019",
"2020", "2021", "2022",
"2023", "2024")) %>%
#compute deltas
mutate(delta_1918 = cur_data()[["2019"]] - cur_data()[[ "2018" ]] ) %>%
mutate(delta_2019 = cur_data()[["2020"]] - cur_data()[[ "2019" ]] ) %>%
mutate(delta_2120 = cur_data()[["2021"]] - cur_data()[[ "2020" ]] ) %>%
mutate(delta_2221 = cur_data()[["2022"]] - cur_data()[[ "2021" ]] ) %>%
mutate(delta_2322 = cur_data()[["2023"]] - cur_data()[[ "2022" ]] ) %>%
mutate(delta_2423 = cur_data()[["2024"]] - cur_data()[[ "2023" ]] ) %>%
.[gtools::mixedorder(.$quest), ] %>%
adorn_totals(name = "mean") %>%
mutate(across(where(is.numeric),
~ replace(., n(), .[n()]/(n()-1)))) %>%
mutate(across(where(is.numeric),
~ round(., 2)))## Warning: There was 1 warning in `mutate()`.
## â„ą In argument: `delta_1918 = cur_data()[["2019"]] - cur_data()[["2018"]]`.
## Caused by warning:
## ! `cur_data()` was deprecated in dplyr 1.1.0.
## â„ą Please use `pick()` instead.df_full %>%
filter(!is.na(asqse_quest)) %>%
select(year_completed_cat, asqse_total, asqse_quest) %>%
pivot_wider(id_cols = asqse_quest, names_from = year_completed_cat, values_from = asqse_total,
values_fn = {mean})%>%
as.data.frame() %>%
setNames(., c("quest","2018","2019",
"2020", "2021", "2022",
"2023", "2024")) %>%
#compute deltas
mutate(delta_1918 = cur_data()[["2019"]] - cur_data()[[ "2018" ]] ) %>%
mutate(delta_2019 = cur_data()[["2020"]] - cur_data()[[ "2019" ]] ) %>%
mutate(delta_2120 = cur_data()[["2021"]] - cur_data()[[ "2020" ]] ) %>%
mutate(delta_2221 = cur_data()[["2022"]] - cur_data()[[ "2021" ]] ) %>%
mutate(delta_2322 = cur_data()[["2023"]] - cur_data()[[ "2022" ]] ) %>%
mutate(delta_2423 = cur_data()[["2024"]] - cur_data()[[ "2023" ]] ) %>%
.[gtools::mixedorder(.$quest), ] %>%
# plot
select(quest, delta_1918:delta_2423) %>%
pivot_longer(-quest) %>%
ggplot(., aes(x = name, y = value, group=1)) +
stat_summary(geom = "pointrange", fun = mean, size=.5) +
stat_summary(geom = "line", fun = mean, size=1, linetype=3) +
stat_summary(geom = "errorbar", width = 0.3) +
geom_hline(yintercept = 0) +
theme_bw() ## No summary function supplied, defaulting to `mean_se()`## Warning: Removed 6 rows containing missing values or values outside the scale range
## (`geom_segment()`).
2.8.2 Plot all age groups
df_full %>%
filter(!is.na(year_completed_cat), !is.na(asqse_quest)) %>%
mutate(Quest = as.factor(asqse_quest)) %>%
ggplot(., aes(x = year_completed_cat, y = asqse_total,
group = Quest, color = Quest )) +
stat_summary(geom = "line", fun = mean, size=1) +
#stat_summary(geom = "errorbar", width = 0.1) +
gghighlight:: gghighlight(year_completed_cat %in% c("21", "22", "23", "24") ) +
labs(y = "ASQ:SE-2", x = "Year the questionnaire was completed") +
theme_bw() + theme(legend.position = "bottom")## Warning: Tried to calculate with group_by(), but the calculation failed.
## Falling back to ungrouped filter operation...
3 Risk
3.1 Tables
3.2 Table w. totals
df_full %>%
filter(!is.na(asqse_quest)) %>%
filter(year_completed_cat %in% c("19","20","21","22","23")) %>%
select(year_completed_cat, asqse_quest, asqse_risk) %>%
filter(asqse_risk == "1") %>%
pivot_wider(id_cols = asqse_quest, names_from = year_completed_cat, values_from = asqse_risk,
values_fn = {function(x) sum(!is.na(x))}, values_fill = 0) %>%
adorn_totals(where = c("row", "col")) %>%
.[gtools::mixedorder(.$asqse_quest), ] 3.3 Table w. percentages
df_full %>%
filter(!is.na(asqse_quest)) %>%
filter(year_completed_cat %in% c("19","20","21","22","23") )%>%
select(year_completed_cat, asqse_quest, asqse_risk) %>%
pivot_wider(id_cols = asqse_quest, names_from = year_completed_cat, values_from = asqse_risk,
values_fn = {mean}) %>%
.[gtools::mixedorder(.$asqse_quest), ] %>%
#percentages
adorn_totals(name = "mean") %>%
mutate(across(where(is.numeric),
~ replace(., n(), .[n()]/(n()-1)))) %>%
mutate(across(where(is.numeric), ~ round ( .*100 ,2) ))3.4 Plots
df_full %>%
filter(!is.na(asqse_quest)) %>%
select(year_completed_cat, asqse_quest, asqse_risk) %>%
ggplot(., aes(x = year_completed_cat, y = asqse_risk, group=1)) + #color here is a trick to have titles
annotate("rect", xmin = 4, xmax = 5, ymin = 0, ymax = Inf , fill= "orange",
alpha = 0.2) +
stat_summary(geom = "pointrange", fun = mean, size=.5) +
stat_summary(geom = "line", fun = mean, size=1, linetype=3) +
labs(x = "Year", y = "Risk") +
scale_y_continuous(labels = scales::percent)+
theme_bw() ## Warning: Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
3.5 Plot all age groups
df_full %>%
filter(!is.na(asqse_quest)) %>%
filter(asqse_quest < 60) %>%
select(year_completed_cat, asqse_quest, asqse_risk) %>%
ggplot(., aes(x = year_completed_cat, y = asqse_risk, group=1)) + #color here is a trick to have titles
annotate("rect", xmin = 4, xmax = 5, ymin = 0, ymax = Inf , fill= "orange",
alpha = 0.2) +
stat_summary(geom = "pointrange", fun = mean, size=.5) +
stat_summary(geom = "line", fun = mean, size=1, linetype=3) +
labs(x = "Year", y = "Risk") +
scale_y_continuous(labels = scales::percent)+
theme_bw() +
facet_wrap(~asqse_quest)## Warning: Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
## Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
## Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
## Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
## Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
## Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
## Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
## Removed 7 rows containing missing values or values outside the scale range
## (`geom_segment()`).
3.6 Table 2
Same results (on Sep 24, 2024)
3.7 Chi square Compare all years
suppressWarnings(
for (i in c("19","20","21","22","23")) {
cat("#### tab", i,'\n')
j = df_full %>%
filter(!is.na(asqse_quest)) %>%
filter(!is.na(momed)) %>% # Here is where we find differences across results.
filter(year_completed_cat == i) %>%
mutate(sex_male = as.factor(sex_male)) %>%
mutate(asqse_quest = as.factor(asqse_quest)) %>%
compareGroups::compareGroups(asqse_risk ~ asqse_quest + race_wbho + sex_male +
momage + momed + income, data = .,
chisq.test.perm = TRUE, chisq.test.seed = 123) %>%
compareGroups::createTable(.)
print(j)
}
)## #### tab 19
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ________________________________________________
## 0 1 p.overall
## N=4677 N=1886
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## asqse_quest: <0.001
## 2 316 (6.76%) 115 (6.10%)
## 6 763 (16.3%) 163 (8.64%)
## 12 473 (10.1%) 215 (11.4%)
## 18 522 (11.2%) 204 (10.8%)
## 24 455 (9.73%) 248 (13.1%)
## 30 308 (6.59%) 181 (9.60%)
## 36 488 (10.4%) 171 (9.07%)
## 48 622 (13.3%) 272 (14.4%)
## 60 730 (15.6%) 317 (16.8%)
## race_wbho: 0.719
## 1W 2913 (62.3%) 1148 (60.9%)
## 2B 503 (10.8%) 217 (11.5%)
## 3H 499 (10.7%) 206 (10.9%)
## Other 762 (16.3%) 315 (16.7%)
## sex_male: <0.001
## 0 2230 (47.7%) 634 (33.7%)
## 1 2443 (52.3%) 1250 (66.3%)
## momage 29.7 (5.53) 27.7 (5.65) <0.001
## momed: <0.001
## 1 82 (1.75%) 76 (4.03%)
## 2 1288 (27.5%) 916 (48.6%)
## 3 636 (13.6%) 310 (16.4%)
## 4 2671 (57.1%) 584 (31.0%)
## income: <0.001
## 1 366 (8.24%) 294 (16.7%)
## 2 360 (8.10%) 274 (15.6%)
## 3 652 (14.7%) 378 (21.5%)
## 4 2466 (55.5%) 563 (32.0%)
## 5 599 (13.5%) 249 (14.2%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 20
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ________________________________________________
## 0 1 p.overall
## N=5353 N=1976
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## asqse_quest: <0.001
## 2 394 (7.36%) 142 (7.19%)
## 6 998 (18.6%) 247 (12.5%)
## 12 764 (14.3%) 329 (16.6%)
## 18 714 (13.3%) 279 (14.1%)
## 24 598 (11.2%) 254 (12.9%)
## 30 410 (7.66%) 156 (7.89%)
## 36 518 (9.68%) 174 (8.81%)
## 48 456 (8.52%) 188 (9.51%)
## 60 501 (9.36%) 207 (10.5%)
## race_wbho: 0.055
## 1W 3412 (63.7%) 1226 (62.0%)
## 2B 531 (9.92%) 213 (10.8%)
## 3H 556 (10.4%) 181 (9.16%)
## Other 854 (16.0%) 356 (18.0%)
## sex_male: <0.001
## 0 2585 (48.3%) 753 (38.1%)
## 1 2766 (51.7%) 1223 (61.9%)
## momage 29.4 (5.57) 28.1 (5.74) <0.001
## momed: <0.001
## 1 145 (2.71%) 102 (5.16%)
## 2 1661 (31.0%) 907 (45.9%)
## 3 853 (15.9%) 298 (15.1%)
## 4 2694 (50.3%) 669 (33.9%)
## income: <0.001
## 1 475 (9.41%) 319 (17.3%)
## 2 440 (8.72%) 243 (13.2%)
## 3 889 (17.6%) 403 (21.9%)
## 4 2501 (49.6%) 616 (33.5%)
## 5 742 (14.7%) 259 (14.1%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 21
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ______________________________________________
## 0 1 p.overall
## N=1586 N=520
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## asqse_quest: 0.001
## 2 94 (5.93%) 36 (6.92%)
## 6 227 (14.3%) 46 (8.85%)
## 12 212 (13.4%) 102 (19.6%)
## 18 255 (16.1%) 83 (16.0%)
## 24 190 (12.0%) 63 (12.1%)
## 30 144 (9.08%) 38 (7.31%)
## 36 174 (11.0%) 42 (8.08%)
## 48 151 (9.52%) 50 (9.62%)
## 60 139 (8.76%) 60 (11.5%)
## race_wbho: 0.620
## 1W 980 (61.8%) 332 (63.8%)
## 2B 191 (12.0%) 59 (11.3%)
## 3H 176 (11.1%) 48 (9.23%)
## Other 239 (15.1%) 81 (15.6%)
## sex_male: <0.001
## 0 789 (49.7%) 191 (36.7%)
## 1 797 (50.3%) 329 (63.3%)
## momage 29.6 (5.64) 28.4 (5.76) <0.001
## momed: <0.001
## 1 43 (2.71%) 14 (2.69%)
## 2 518 (32.7%) 255 (49.0%)
## 3 252 (15.9%) 71 (13.7%)
## 4 773 (48.7%) 180 (34.6%)
## income: <0.001
## 1 159 (10.6%) 71 (14.8%)
## 2 146 (9.75%) 49 (10.2%)
## 3 236 (15.8%) 107 (22.2%)
## 4 751 (50.2%) 186 (38.7%)
## 5 205 (13.7%) 68 (14.1%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 22
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ______________________________________________
## 0 1 p.overall
## N=1170 N=736
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## asqse_quest: <0.001
## 2 69 (5.90%) 34 (4.62%)
## 6 150 (12.8%) 43 (5.84%)
## 12 156 (13.3%) 90 (12.2%)
## 18 165 (14.1%) 132 (17.9%)
## 24 152 (13.0%) 107 (14.5%)
## 30 100 (8.55%) 67 (9.10%)
## 36 121 (10.3%) 81 (11.0%)
## 48 131 (11.2%) 102 (13.9%)
## 60 126 (10.8%) 80 (10.9%)
## race_wbho: 0.019
## 1W 655 (56.0%) 456 (62.0%)
## 2B 217 (18.5%) 101 (13.7%)
## 3H 136 (11.6%) 75 (10.2%)
## Other 162 (13.8%) 104 (14.1%)
## sex_male: <0.001
## 0 549 (46.9%) 265 (36.1%)
## 1 621 (53.1%) 470 (63.9%)
## momage 29.6 (5.71) 28.0 (5.68) <0.001
## momed: <0.001
## 1 43 (3.68%) 36 (4.89%)
## 2 408 (34.9%) 367 (49.9%)
## 3 198 (16.9%) 116 (15.8%)
## 4 521 (44.5%) 217 (29.5%)
## income: <0.001
## 1 134 (12.2%) 113 (16.2%)
## 2 102 (9.26%) 59 (8.45%)
## 3 169 (15.3%) 157 (22.5%)
## 4 77 (6.99%) 17 (2.44%)
## 5 279 (25.3%) 203 (29.1%)
## 8 341 (30.9%) 149 (21.3%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 23
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ________________________________________________
## 0 1 p.overall
## N=4110 N=2480
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## asqse_quest: <0.001
## 2 202 (4.91%) 93 (3.75%)
## 6 521 (12.7%) 173 (6.98%)
## 12 615 (15.0%) 444 (17.9%)
## 18 659 (16.0%) 474 (19.1%)
## 24 587 (14.3%) 346 (14.0%)
## 30 355 (8.64%) 211 (8.51%)
## 36 465 (11.3%) 262 (10.6%)
## 48 391 (9.51%) 272 (11.0%)
## 60 315 (7.66%) 205 (8.27%)
## race_wbho: 0.145
## 1W 2724 (66.3%) 1595 (64.3%)
## 2B 427 (10.4%) 258 (10.4%)
## 3H 347 (8.44%) 249 (10.0%)
## Other 612 (14.9%) 378 (15.2%)
## sex_male: <0.001
## 0 1851 (45.1%) 928 (37.4%)
## 1 2257 (54.9%) 1552 (62.6%)
## momage 29.5 (5.44) 28.1 (5.72) <0.001
## momed: <0.001
## 1 83 (2.02%) 129 (5.20%)
## 2 1362 (33.1%) 1258 (50.7%)
## 3 662 (16.1%) 371 (15.0%)
## 4 2003 (48.7%) 722 (29.1%)
## income: <0.001
## 1 294 (7.50%) 310 (13.3%)
## 2 240 (6.12%) 259 (11.1%)
## 3 522 (13.3%) 464 (19.9%)
## 5 1158 (29.5%) 756 (32.4%)
## 8 1707 (43.5%) 545 (23.4%)
## ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯– Comparegroups runs a t test grouping the predictors
df_full %>%
filter(!is.na(asqse_quest)) %>%
filter(!is.na(momed)) %>%
filter(year_completed == "22") %>%
#t.test(momage ~ asqse_risk, data =.) %>%
{gmodels::CrossTable(.$race_wbho, .$asqse_risk, prop.r = T, expected = T, prop.chisq = F, prop.c = F, prop.t = F)} ##
##
## Cell Contents
## |-------------------------|
## | N |
## | Expected N |
## | N / Row Total |
## |-------------------------|
##
##
## Total Observations in Table: 1906
##
##
## | .$asqse_risk
## .$race_wbho | 0 | 1 | Row Total |
## -------------|-----------|-----------|-----------|
## 1W | 655 | 456 | 1111 |
## | 681.988 | 429.012 | |
## | 0.590 | 0.410 | 0.583 |
## -------------|-----------|-----------|-----------|
## 2B | 217 | 101 | 318 |
## | 195.205 | 122.795 | |
## | 0.682 | 0.318 | 0.167 |
## -------------|-----------|-----------|-----------|
## 3H | 136 | 75 | 211 |
## | 129.523 | 81.477 | |
## | 0.645 | 0.355 | 0.111 |
## -------------|-----------|-----------|-----------|
## Other | 162 | 104 | 266 |
## | 163.284 | 102.716 | |
## | 0.609 | 0.391 | 0.140 |
## -------------|-----------|-----------|-----------|
## Column Total | 1170 | 736 | 1906 |
## -------------|-----------|-----------|-----------|
##
##
## Statistics for All Table Factors
##
##
## Pearson's Chi-squared test
## ------------------------------------------------------------
## Chi^2 = 9.932958 d.f. = 3 p = 0.0191447
##
##
## And its run a chi-square for categorical variables
df_full %>%
filter(!is.na(momed)) %>% # 5 is I don't know
filter(year_completed == "22") %>%
{chisq.test(.$asqse_risk, .$race_wbo)} ##
## Pearson's Chi-squared test
##
## data: .$asqse_risk and .$race_wbo
## X-squared = 9.933, df = 3, p-value = 0.019143.8 Chi square all Compare years
suppressWarnings(
for (i in c(2,6,12,18,24,30,36,48,60)) {
cat("#### tab", i,'\n')
j = df_full %>%
filter(!is.na(momed)) %>% # 5 is I don't know
filter(asqse_quest == i) %>%
mutate(sex_male = as.factor(sex_male)) %>%
compareGroups::compareGroups(asqse_risk ~ year_completed_cat + sex_male +
race + momage + momed + income, data = .,
chisq.test.perm = TRUE, chisq.test.seed = 123) %>%
compareGroups::createTable(.)
print(j)
}
)## #### tab 2
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## _____________________________________________________
## 0 1 p.overall
## N=1078 N=423
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: 0.304
## 18 1 (0.09%) 0 (0.00%)
## 19 316 (29.3%) 115 (27.2%)
## 20 394 (36.5%) 142 (33.6%)
## 21 94 (8.72%) 36 (8.51%)
## 22 69 (6.40%) 34 (8.04%)
## 23 202 (18.7%) 93 (22.0%)
## 24 2 (0.19%) 3 (0.71%)
## sex_male: 0.348
## 0 531 (49.3%) 196 (46.4%)
## 1 546 (50.7%) 226 (53.6%)
## race: 0.276
## Black 134 (14.3%) 39 (11.2%)
## Hispanic_Latino 97 (10.3%) 42 (12.0%)
## White 707 (75.4%) 268 (76.8%)
## momage 29.2 (5.74) 29.0 (5.89) 0.656
## momed: <0.001
## 1 25 (2.32%) 19 (4.49%)
## 2 384 (35.6%) 195 (46.1%)
## 3 173 (16.0%) 69 (16.3%)
## 4 496 (46.0%) 140 (33.1%)
## income: <0.001
## 1 112 (10.7%) 74 (18.7%)
## 2 97 (9.31%) 55 (13.9%)
## 3 168 (16.1%) 75 (18.9%)
## 4 358 (34.4%) 95 (24.0%)
## 5 210 (20.2%) 67 (16.9%)
## 8 97 (9.31%) 30 (7.58%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 6
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ______________________________________________________
## 0 1 p.overall
## N=2670 N=674
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: 0.013
## 18 5 (0.19%) 1 (0.15%)
## 19 763 (28.6%) 163 (24.2%)
## 20 998 (37.4%) 247 (36.6%)
## 21 227 (8.50%) 46 (6.82%)
## 22 150 (5.62%) 43 (6.38%)
## 23 521 (19.5%) 173 (25.7%)
## 24 6 (0.22%) 1 (0.15%)
## sex_male: 0.297
## 0 1294 (48.5%) 311 (46.1%)
## 1 1375 (51.5%) 363 (53.9%)
## race: 0.599
## Black 241 (10.7%) 52 (9.29%)
## Hispanic_Latino 259 (11.5%) 65 (11.6%)
## White 1743 (77.7%) 443 (79.1%)
## momage 29.5 (5.43) 28.2 (5.72) <0.001
## momed: <0.001
## 1 55 (2.06%) 29 (4.30%)
## 2 882 (33.0%) 314 (46.6%)
## 3 398 (14.9%) 97 (14.4%)
## 4 1335 (50.0%) 234 (34.7%)
## income: <0.001
## 1 239 (9.43%) 92 (14.8%)
## 2 207 (8.17%) 74 (11.9%)
## 3 414 (16.3%) 114 (18.4%)
## 4 992 (39.1%) 186 (30.0%)
## 5 417 (16.4%) 107 (17.3%)
## 8 266 (10.5%) 47 (7.58%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 12
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ______________________________________________________
## 0 1 p.overall
## N=2232 N=1187
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: <0.001
## 18 2 (0.09%) 1 (0.08%)
## 19 473 (21.2%) 215 (18.1%)
## 20 764 (34.2%) 329 (27.7%)
## 21 212 (9.50%) 102 (8.59%)
## 22 156 (6.99%) 90 (7.58%)
## 23 615 (27.6%) 444 (37.4%)
## 24 10 (0.45%) 6 (0.51%)
## sex_male: <0.001
## 0 1102 (49.4%) 447 (37.7%)
## 1 1128 (50.6%) 738 (62.3%)
## race: 0.008
## Black 170 (9.09%) 127 (12.7%)
## Hispanic_Latino 195 (10.4%) 110 (11.0%)
## White 1505 (80.5%) 764 (76.3%)
## momage 30.1 (5.41) 28.9 (5.59) <0.001
## momed: <0.001
## 1 42 (1.88%) 35 (2.95%)
## 2 553 (24.8%) 502 (42.3%)
## 3 316 (14.2%) 173 (14.6%)
## 4 1321 (59.2%) 477 (40.2%)
## income: <0.001
## 1 151 (7.15%) 124 (11.2%)
## 2 136 (6.44%) 102 (9.17%)
## 3 250 (11.8%) 215 (19.3%)
## 4 851 (40.3%) 277 (24.9%)
## 5 357 (16.9%) 246 (22.1%)
## 8 367 (17.4%) 148 (13.3%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 18
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ______________________________________________________
## 0 1 p.overall
## N=2325 N=1177
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: <0.001
## 18 1 (0.04%) 1 (0.08%)
## 19 522 (22.5%) 204 (17.3%)
## 20 714 (30.7%) 279 (23.7%)
## 21 255 (11.0%) 83 (7.05%)
## 22 165 (7.10%) 132 (11.2%)
## 23 659 (28.3%) 474 (40.3%)
## 24 9 (0.39%) 4 (0.34%)
## sex_male: <0.001
## 0 1092 (47.0%) 436 (37.0%)
## 1 1233 (53.0%) 741 (63.0%)
## race: <0.001
## Black 194 (9.90%) 145 (15.1%)
## Hispanic_Latino 238 (12.1%) 120 (12.5%)
## White 1528 (78.0%) 698 (72.5%)
## momage 30.0 (5.35) 28.3 (5.71) <0.001
## momed: <0.001
## 1 57 (2.45%) 56 (4.76%)
## 2 591 (25.4%) 587 (49.9%)
## 3 354 (15.2%) 192 (16.3%)
## 4 1323 (56.9%) 342 (29.1%)
## income: <0.001
## 1 171 (7.73%) 163 (14.8%)
## 2 144 (6.51%) 145 (13.2%)
## 3 297 (13.4%) 214 (19.4%)
## 4 821 (37.1%) 189 (17.2%)
## 5 429 (19.4%) 267 (24.2%)
## 8 351 (15.9%) 124 (11.3%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 24
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ______________________________________________________
## 0 1 p.overall
## N=1987 N=1023
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: <0.001
## 18 3 (0.15%) 0 (0.00%)
## 19 455 (22.9%) 248 (24.2%)
## 20 598 (30.1%) 254 (24.8%)
## 21 190 (9.56%) 63 (6.16%)
## 22 152 (7.65%) 107 (10.5%)
## 23 587 (29.5%) 346 (33.8%)
## 24 2 (0.10%) 5 (0.49%)
## sex_male: <0.001
## 0 911 (45.8%) 348 (34.0%)
## 1 1076 (54.2%) 675 (66.0%)
## race: <0.001
## Black 188 (11.3%) 136 (16.1%)
## Hispanic_Latino 177 (10.7%) 122 (14.5%)
## White 1293 (78.0%) 585 (69.4%)
## momage 29.8 (5.40) 27.7 (5.65) <0.001
## momed: <0.001
## 1 43 (2.16%) 44 (4.30%)
## 2 593 (29.8%) 511 (50.0%)
## 3 311 (15.7%) 158 (15.4%)
## 4 1040 (52.3%) 310 (30.3%)
## income: <0.001
## 1 148 (7.84%) 164 (17.0%)
## 2 130 (6.89%) 123 (12.8%)
## 3 276 (14.6%) 205 (21.3%)
## 4 667 (35.3%) 145 (15.1%)
## 5 373 (19.8%) 225 (23.4%)
## 8 293 (15.5%) 100 (10.4%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 30
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## _____________________________________________________
## 0 1 p.overall
## N=1324 N=655
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: <0.001
## 18 4 (0.30%) 0 (0.00%)
## 19 308 (23.3%) 181 (27.6%)
## 20 410 (31.0%) 156 (23.8%)
## 21 144 (10.9%) 38 (5.80%)
## 22 100 (7.55%) 67 (10.2%)
## 23 355 (26.8%) 211 (32.2%)
## 24 3 (0.23%) 2 (0.31%)
## sex_male: <0.001
## 0 589 (44.6%) 214 (32.7%)
## 1 733 (55.4%) 441 (67.3%)
## race: 0.548
## Black 143 (12.9%) 83 (14.8%)
## Hispanic_Latino 127 (11.5%) 61 (10.9%)
## White 838 (75.6%) 416 (74.3%)
## momage 29.7 (5.70) 28.0 (5.61) <0.001
## momed: <0.001
## 1 33 (2.49%) 27 (4.12%)
## 2 377 (28.5%) 348 (53.1%)
## 3 223 (16.8%) 124 (18.9%)
## 4 691 (52.2%) 156 (23.8%)
## income: <0.001
## 1 91 (7.21%) 99 (16.3%)
## 2 96 (7.61%) 78 (12.8%)
## 3 202 (16.0%) 145 (23.8%)
## 4 435 (34.5%) 100 (16.4%)
## 5 271 (21.5%) 132 (21.7%)
## 8 167 (13.2%) 54 (8.88%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 36
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ______________________________________________________
## 0 1 p.overall
## N=1775 N=731
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: <0.001
## 18 2 (0.11%) 0 (0.00%)
## 19 488 (27.5%) 171 (23.4%)
## 20 518 (29.2%) 174 (23.8%)
## 21 174 (9.80%) 42 (5.75%)
## 22 121 (6.82%) 81 (11.1%)
## 23 465 (26.2%) 262 (35.8%)
## 24 7 (0.39%) 1 (0.14%)
## sex_male: <0.001
## 0 814 (45.9%) 246 (33.7%)
## 1 961 (54.1%) 485 (66.3%)
## race: 0.398
## Black 213 (14.3%) 78 (13.1%)
## Hispanic_Latino 185 (12.4%) 86 (14.5%)
## White 1091 (73.3%) 430 (72.4%)
## momage 29.3 (5.44) 27.7 (5.72) <0.001
## momed: <0.001
## 1 41 (2.31%) 37 (5.06%)
## 2 575 (32.4%) 369 (50.5%)
## 3 294 (16.6%) 123 (16.8%)
## 4 865 (48.7%) 202 (27.6%)
## income: <0.001
## 1 129 (7.67%) 111 (16.3%)
## 2 156 (9.27%) 81 (11.9%)
## 3 296 (17.6%) 180 (26.4%)
## 4 559 (33.2%) 101 (14.8%)
## 5 325 (19.3%) 145 (21.2%)
## 8 217 (12.9%) 65 (9.52%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 48
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## _____________________________________________________
## 0 1 p.overall
## N=1757 N=890
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: <0.001
## 18 2 (0.11%) 1 (0.11%)
## 19 622 (35.4%) 272 (30.6%)
## 20 456 (26.0%) 188 (21.1%)
## 21 151 (8.59%) 50 (5.62%)
## 22 131 (7.46%) 102 (11.5%)
## 23 391 (22.3%) 272 (30.6%)
## 24 4 (0.23%) 5 (0.56%)
## sex_male: <0.001
## 0 846 (48.2%) 304 (34.2%)
## 1 910 (51.8%) 586 (65.8%)
## race: <0.001
## Black 289 (19.3%) 106 (13.9%)
## Hispanic_Latino 233 (15.5%) 75 (9.82%)
## White 978 (65.2%) 583 (76.3%)
## momage 28.9 (5.65) 27.4 (5.56) <0.001
## momed: <0.001
## 1 39 (2.22%) 44 (4.94%)
## 2 666 (37.9%) 452 (50.8%)
## 3 258 (14.7%) 111 (12.5%)
## 4 794 (45.2%) 283 (31.8%)
## income: <0.001
## 1 197 (11.9%) 137 (16.4%)
## 2 162 (9.81%) 115 (13.8%)
## 3 283 (17.1%) 179 (21.5%)
## 4 520 (31.5%) 132 (15.8%)
## 5 306 (18.5%) 194 (23.3%)
## 8 183 (11.1%) 77 (9.23%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## #### tab 60
##
## --------Summary descriptives table by 'asqse_risk'---------
##
## ______________________________________________________
## 0 1 p.overall
## N=1814 N=870
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ
## year_completed_cat: <0.001
## 18 0 (0.00%) 1 (0.11%)
## 19 730 (40.2%) 317 (36.4%)
## 20 501 (27.6%) 207 (23.8%)
## 21 139 (7.66%) 60 (6.90%)
## 22 126 (6.95%) 80 (9.20%)
## 23 315 (17.4%) 205 (23.6%)
## 24 3 (0.17%) 0 (0.00%)
## sex_male: <0.001
## 0 854 (47.1%) 284 (32.6%)
## 1 959 (52.9%) 586 (67.4%)
## race: <0.001
## Black 302 (19.4%) 86 (11.4%)
## Hispanic_Latino 211 (13.6%) 79 (10.5%)
## White 1042 (67.0%) 589 (78.1%)
## momage 29.0 (5.89) 26.8 (5.76) <0.001
## momed: <0.001
## 1 61 (3.36%) 68 (7.82%)
## 2 638 (35.2%) 436 (50.1%)
## 3 285 (15.7%) 124 (14.3%)
## 4 830 (45.8%) 242 (27.8%)
## income: <0.001
## 1 195 (11.6%) 145 (17.6%)
## 2 165 (9.77%) 112 (13.6%)
## 3 288 (17.1%) 185 (22.5%)
## 4 604 (35.8%) 161 (19.6%)
## 5 308 (18.2%) 162 (19.7%)
## 8 128 (7.58%) 58 (7.05%)
## ÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻÂŻ3.9 Table 3
3.10 Chi square for trend
Cochran-Armitage test
for (i in c(2,6,12,18,24,30,36,48,60)) { #child's age
j = df_full %>%
filter(!is.na(asqse_quest)) %>%
filter(year_completed_cat %in% c("19","20","21","22","23") )%>%
filter(asqse_quest == i) %>%
select(year_completed_cat, asqse_risk) %>%
# trend analysis
{table(.$year_completed_cat, .$asqse_risk)} %>%
DescTools::CochranArmitageTest(., alternative = c("one.sided"))
print(i)
print(j)
rm(j,i)
}## Registered S3 method overwritten by 'DescTools':
## method from
## reorder.factor gdata## [1] 2
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -1.9876, dim = 7, p-value = 0.02343
## alternative hypothesis: one.sided
##
## [1] 6
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -3.8014, dim = 7, p-value = 7.194e-05
## alternative hypothesis: one.sided
##
## [1] 12
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -5.9052, dim = 7, p-value = 1.761e-09
## alternative hypothesis: one.sided
##
## [1] 18
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -7.832, dim = 7, p-value = 2.442e-15
## alternative hypothesis: one.sided
##
## [1] 24
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -2.4228, dim = 7, p-value = 0.0077
## alternative hypothesis: one.sided
##
## [1] 30
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -1.5075, dim = 7, p-value = 0.06584
## alternative hypothesis: one.sided
##
## [1] 36
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -5.1863, dim = 7, p-value = 1.073e-07
## alternative hypothesis: one.sided
##
## [1] 48
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -5.5076, dim = 7, p-value = 1.819e-08
## alternative hypothesis: one.sided
##
## [1] 60
##
## Cochran-Armitage test for trend
##
## data: .
## Z = -3.9487, dim = 7, p-value = 3.93e-05
## alternative hypothesis: one.sideddf_full %>%
filter(!is.na(asqse_quest)) %>%
filter(year_completed_cat %in% c("19","20","21","22","23") )%>%
filter(asqse_quest == 2) %>%
select(year_completed_cat, asqse_risk) %>%
{table(.$year_completed_cat, .$asqse_risk)} %>%
DescTools::CochranArmitageTest(., alternative = c("two.sided", "one.sided"))##
## Cochran-Armitage test for trend
##
## data: .
## Z = -1.9876, dim = 7, p-value = 0.04686
## alternative hypothesis: two.sideddf_full %>%
filter(!is.na(asqse_quest)) %>%
filter(year_completed_cat %in% c("19","20","21","22","23") )%>%
select(year_completed_cat, asqse_risk) %>%
mutate_all(., ~as.factor(.)) %>%
tableby(year_completed_cat ~ asqse_risk, data = .) %>% summary(digits = 2, text = T) %>%
as.data.frame()df_full %>%
filter(momed %in% (1:4)) %>% # 5 is I don't know
filter(asqse_quest == 30) %>%
mutate_at(vars(asqse_risk,sex_male,year_completed_cat),~as.factor(.)) %>%
glm(asqse_risk ~ year_completed_cat ,
family = binomial(link = "logit"),
data = .) %>% anova(.)3.11 Matching sampling Table 4
3.11.1 Dataset
df_full %>%
filter(!is.na(asqse_quest)) %>%
filter(year_completed_cat == "19") %>%
filter(asqse_quest == 24) %>%
select(asqse_quest,sex_male, momage, momed, income) %>%
mutate_all(., ~as.numeric(.)) %>%
mutate_all(~ifelse(is.na(.x), median(.x, na.rm = TRUE), .x))df_full_sampling <- df_full %>%
select(id,
year_completed_cat, asqse_quest, race, sex_male, momage, momed, income_poor , asqse_total, asqse_risk) %>%
mutate(momed = case_when(
momed == "1" ~ "1",
momed == "2" ~ "2",
momed == "3" ~ "3",
momed == "4" ~ "3",
TRUE ~ NA_character_
)) %>%
mutate(race = case_when(
race == "White" ~ "1W",
race == "Black" ~ "2B",
TRUE ~ "Other"
)) %>%
na.omit()
write.csv(df_full_sampling, "df_full_sampling.csv", row.names = F)df_full_sampling %>%
tableby(year_completed_cat ~ asqse_risk + race + momage + momed + income_poor, data =.) %>% summary(text=T, digits=2) %>% as.data.frame()3.11.2 1 year
3.11.3 Matching - work with dynamic dataframe
library(MatchIt)
# Filter datasets for 2019, 2020, 2021, 2022, and 2023
df_2019 <- df_full_sampling %>%
filter(year_completed_cat == "19", asqse_quest == "12")
cat("Number of rows for 2019: ", nrow(df_2019), "\n")## Number of rows for 2019: 635df_2020 <- df_full_sampling %>%
filter(year_completed_cat == "20", asqse_quest == "12")
cat("Number of rows for 2020: ", nrow(df_2020), "\n")## Number of rows for 2020: 964df_2021 <- df_full_sampling %>%
filter(year_completed_cat == "21", asqse_quest == "12")
cat("Number of rows for 2021: ", nrow(df_2021), "\n")## Number of rows for 2021: 279df_2022 <- df_full_sampling %>%
filter(year_completed_cat == "22", asqse_quest == "12")
cat("Number of rows for 2022: ", nrow(df_2022), "\n")## Number of rows for 2022: 215df_2023 <- df_full_sampling %>%
filter(year_completed_cat == "23", asqse_quest == "12")
cat("Number of rows for 2023: ", nrow(df_2023), "\n")## Number of rows for 2023: 933# Function to perform matching and return matched data
perform_matching <- function(df_main, df_other, main_year_label, other_year_label) {
set.seed(1)
data <- bind_rows(df_main, df_other, .id = "year") %>%
mutate(year = ifelse(year == "1", 1, 0)) # 1 for main year, 0 for the other year
match_obj <- matchit(year ~ asqse_quest + race + sex_male + momage + momed + income_poor,
data = data,
ratio = 1,
exact = ~ momed + race ,
method = "optimal")
#plot(match_obj, type = 'density')
matched_data <- match.data(match_obj)
matched_data <- matched_data %>% mutate(compared_year = other_year_label)
return(matched_data)
}
# Perform matching for each year compared to 2021
matched_data_2019 <- perform_matching(df_2021, df_2019, "2021", "2019")## Warning: Fewer control units than treated units in some `exact` strata; not all
## treated units will get a match.matched_data_2020 <- perform_matching(df_2021, df_2020, "2021", "2020")
matched_data_2022 <- perform_matching(df_2021, df_2022, "2021", "2022")## Warning: Fewer control units than treated units in some `exact` strata; not all
## treated units will get a match.matched_data_2023 <- perform_matching(df_2021, df_2023, "2021", "2023")
# Combine all matched datasets
all_matched_data_1 <- bind_rows(matched_data_2019, matched_data_2020, matched_data_2022, matched_data_2023)
# Print the first few rows of the combined matched data to verify
#head(all_matched_data)3.11.4 Fine tuning
3.11.5 Compare results
all_matched_data_1 %>%
tableby(year_completed_cat ~ as.factor(asqse_risk) + race + momage + momed + as.factor(sex_male) + income_poor, data =.) %>% summary(text=T, digits=2) %>% as.data.frame()df_full_sampling %>%
select(momage, year_completed_cat) %>%
ggplot(., aes(year_completed_cat, momage, group=1)) +
stat_summary(geom ="line")## No summary function supplied, defaulting to `mean_se()`
a=df_full_sampling %>%
select(momage, year_completed_cat,income_poor) %>%
ggplot(., aes(year_completed_cat, momage, group=income_poor,color=income_poor)) +
stat_summary(geom ="line")
b=all_matched_data %>%
select(momage, year_completed_cat,income_poor) %>%
ggplot(., aes(year_completed_cat, momage, group=income_poor,color=income_poor)) +
stat_summary(geom ="line")
a+b
rm(a,b)3.11.6 2 years
# Filter datasets for 2019, 2020, 2021, 2022, and 2023
df_2019 <- df_full_sampling %>%
filter(year_completed_cat == "19", asqse_quest == "24")
cat("Number of rows for 2019: ", nrow(df_2019), "\n")## Number of rows for 2019: 645df_2020 <- df_full_sampling %>%
filter(year_completed_cat == "20", asqse_quest == "24")
cat("Number of rows for 2020: ", nrow(df_2020), "\n")## Number of rows for 2020: 762df_2021 <- df_full_sampling %>%
filter(year_completed_cat == "21", asqse_quest == "24")
cat("Number of rows for 2021: ", nrow(df_2021), "\n")## Number of rows for 2021: 229df_2022 <- df_full_sampling %>%
filter(year_completed_cat == "22", asqse_quest == "24")
cat("Number of rows for 2022: ", nrow(df_2022), "\n")## Number of rows for 2022: 228df_2023 <- df_full_sampling %>%
filter(year_completed_cat == "23", asqse_quest == "24")
cat("Number of rows for 2023: ", nrow(df_2023), "\n")## Number of rows for 2023: 851# Function to perform matching and return matched data
perform_matching <- function(df_main, df_other, main_year_label, other_year_label) {
set.seed(1)
data <- bind_rows(df_main, df_other, .id = "year") %>%
mutate(year = ifelse(year == "1", 1, 0)) # 1 for main year, 0 for the other year
match_obj <- matchit(year ~ asqse_quest + race + sex_male + momage + momed + income_poor,
data = data,
exact = ~ momed + income_poor + sex_male,
method = "optimal")
matched_data <- match.data(match_obj)
matched_data <- matched_data %>% mutate(compared_year = other_year_label)
return(matched_data)
}
# Perform matching for each year compared to 2021
matched_data_2019 <- perform_matching(df_2021, df_2019, "2021", "2019")
matched_data_2020 <- perform_matching(df_2021, df_2020, "2021", "2020")
matched_data_2022 <- perform_matching(df_2021, df_2022, "2021", "2022")## Warning: Fewer control units than treated units in some `exact` strata; not all
## treated units will get a match.matched_data_2023 <- perform_matching(df_2021, df_2023, "2021", "2023")
# Combine all matched datasets
all_matched_data_2 <- bind_rows(matched_data_2019, matched_data_2020, matched_data_2022, matched_data_2023)
# Print the first few rows of the combined matched data to verify
#head(all_matched_data)3.11.7 Fine tuning
3.11.8 Compare results
all_matched_data_2 %>%
tableby(year_completed_cat ~ as.factor(asqse_risk) + race + momage + momed + as.factor(sex_male) + income_poor, data =.) %>% summary(text=T, digits=2) %>% as.data.frame()3.11.9 3 years
# Filter datasets for 2019, 2020, 2021, 2022, and 2023
df_2019 <- df_full_sampling %>%
filter(year_completed_cat == "19", asqse_quest == "36")
cat("Number of rows for 2019: ", nrow(df_2019), "\n")## Number of rows for 2019: 614df_2020 <- df_full_sampling %>%
filter(year_completed_cat == "20", asqse_quest == "36")
cat("Number of rows for 2020: ", nrow(df_2020), "\n")## Number of rows for 2020: 615df_2021 <- df_full_sampling %>%
filter(year_completed_cat == "21", asqse_quest == "36")
cat("Number of rows for 2021: ", nrow(df_2021), "\n")## Number of rows for 2021: 199df_2022 <- df_full_sampling %>%
filter(year_completed_cat == "22", asqse_quest == "36")
cat("Number of rows for 2022: ", nrow(df_2022), "\n")## Number of rows for 2022: 184df_2023 <- df_full_sampling %>%
filter(year_completed_cat == "23", asqse_quest == "36")
cat("Number of rows for 2023: ", nrow(df_2023), "\n")## Number of rows for 2023: 664# Function to perform matching and return matched data
perform_matching <- function(df_main, df_other, main_year_label, other_year_label) {
set.seed(1)
data <- bind_rows(df_main, df_other, .id = "year") %>%
mutate(year = ifelse(year == "1", 1, 0)) # 1 for main year, 0 for the other year
match_obj <- matchit(year ~ asqse_quest + race + sex_male + momage + momed + income_poor,
data = data,
exact = ~ momed + income_poor,
method = "optimal")
matched_data <- match.data(match_obj)
matched_data <- matched_data %>% mutate(compared_year = other_year_label)
return(matched_data)
}
# Perform matching for each year compared to 2021
matched_data_2019 <- perform_matching(df_2021, df_2019, "2021", "2019")
matched_data_2020 <- perform_matching(df_2021, df_2020, "2021", "2020")
matched_data_2022 <- perform_matching(df_2021, df_2022, "2021", "2022")## Warning: Fewer control units than treated units in some `exact` strata; not all
## treated units will get a match.3.11.10 Fine tuning
3 years old
3.11.11 Compare results
all_matched_data_3 %>%
tableby(year_completed_cat ~ as.factor(asqse_risk) + race + momage + momed + as.factor(sex_male) + income_poor, data =.) %>% summary(text=T, digits=2) %>% as.data.frame()3.11.12 4 years
# Filter datasets for 2019, 2020, 2021, 2022, and 2023
df_2019 <- df_full_sampling %>%
filter(year_completed_cat == "19", asqse_quest == "48")
cat("Number of rows for 2019: ", nrow(df_2019), "\n")## Number of rows for 2019: 826df_2020 <- df_full_sampling %>%
filter(year_completed_cat == "20", asqse_quest == "48")
cat("Number of rows for 2020: ", nrow(df_2020), "\n")## Number of rows for 2020: 588df_2021 <- df_full_sampling %>%
filter(year_completed_cat == "21", asqse_quest == "48")
cat("Number of rows for 2021: ", nrow(df_2021), "\n")## Number of rows for 2021: 181df_2022 <- df_full_sampling %>%
filter(year_completed_cat == "22", asqse_quest == "48")
cat("Number of rows for 2022: ", nrow(df_2022), "\n")## Number of rows for 2022: 214df_2023 <- df_full_sampling %>%
filter(year_completed_cat == "23", asqse_quest == "48")
cat("Number of rows for 2023: ", nrow(df_2023), "\n")## Number of rows for 2023: 594# Function to perform matching and return matched data
perform_matching <- function(df_main, df_other, main_year_label, other_year_label) {
set.seed(1)
data <- bind_rows(df_main, df_other, .id = "year") %>%
mutate(year = ifelse(year == "1", 1, 0)) # 1 for main year, 0 for the other year
match_obj <- matchit(year ~ asqse_quest + race + sex_male + momage + momed + income_poor,
data = data,
exact = ~ momed + income_poor,
method = "optimal")
matched_data <- match.data(match_obj)
matched_data <- matched_data %>% mutate(compared_year = other_year_label)
return(matched_data)
}
# Perform matching for each year compared to 2021
matched_data_2019 <- perform_matching(df_2021, df_2019, "2021", "2019")
matched_data_2020 <- perform_matching(df_2021, df_2020, "2021", "2020")
matched_data_2022 <- perform_matching(df_2021, df_2022, "2021", "2022")## Warning: Fewer control units than treated units in some `exact` strata; not all
## treated units will get a match.3.11.13 Fine tuning
4 years old
3.11.14 Compare results
all_matched_data_4 %>%
tableby(year_completed_cat ~ as.factor(asqse_risk) + race + momage + momed + as.factor(sex_male) + income_poor, data =.) %>% summary(text=T, digits=2) %>% as.data.frame() 3.11.15 5 years
5 years old
# Filter datasets for 2019, 2020, 2021, 2022, and 2023
df_2019 <- df_full_sampling %>%
filter(year_completed_cat == "19", asqse_quest == "60")
cat("Number of rows for 2019: ", nrow(df_2019), "\n")## Number of rows for 2019: 967df_2020 <- df_full_sampling %>%
filter(year_completed_cat == "20", asqse_quest == "60")
cat("Number of rows for 2020: ", nrow(df_2020), "\n")## Number of rows for 2020: 631df_2021 <- df_full_sampling %>%
filter(year_completed_cat == "21", asqse_quest == "60")
cat("Number of rows for 2021: ", nrow(df_2021), "\n")## Number of rows for 2021: 175df_2022 <- df_full_sampling %>%
filter(year_completed_cat == "22", asqse_quest == "60")
cat("Number of rows for 2022: ", nrow(df_2022), "\n")## Number of rows for 2022: 184df_2023 <- df_full_sampling %>%
filter(year_completed_cat == "23", asqse_quest == "60")
cat("Number of rows for 2023: ", nrow(df_2023), "\n")## Number of rows for 2023: 456# Function to perform matching and return matched data
perform_matching <- function(df_main, df_other, main_year_label, other_year_label) {
set.seed(1)
data <- bind_rows(df_main, df_other, .id = "year") %>%
mutate(year = ifelse(year == "1", 1, 0)) # 1 for main year, 0 for the other year
match_obj <- matchit(year ~ asqse_quest + race + sex_male + momage + momed + income_poor,
data = data,
exact = ~ momed + race + sex_male + income_poor,
method = "optimal")
matched_data <- match.data(match_obj)
matched_data <- matched_data %>% mutate(compared_year = other_year_label)
return(matched_data)
}
# Perform matching for each year compared to 2021
matched_data_2019 <- perform_matching(df_2021, df_2019, "2021", "2019")
matched_data_2020 <- perform_matching(df_2021, df_2020, "2021", "2020")
matched_data_2022 <- perform_matching(df_2021, df_2022, "2021", "2022")## Warning: Fewer control units than treated units in some `exact` strata; not all
## treated units will get a match.## Warning: Fewer control units than treated units in some `exact` strata; not all
## treated units will get a match.3.11.16 Fine tuning
3.12 Trend lines
Ungrouped
bind_rows(
all_matched_data_1 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="1-year-old"),
all_matched_data_2 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="2-year-old"),
all_matched_data_3 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="3-year-old"),
all_matched_data_4 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="4-year-old"),
all_matched_data %>% select(asqse_risk, year_completed_cat) %>% mutate(age="5-year-old")) %>%
ggplot(., aes(x = year_completed_cat, y = asqse_risk, group = 0)) + #+ color = age)
stat_summary(geom = "line", fun = mean, size=1) +
stat_summary(geom = "errorbar",fun.data = mean_se, width = .1) +
labs(x = "Year the ASQ-SE was completed", y = "Percentage of Children at Risk", color = "Child's age") +
theme_bw() + theme(legend.position = "bottom")
grouped
bind_rows(
all_matched_data_1 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="1-year-old"),
all_matched_data_2 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="2-year-old"),
all_matched_data_3 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="3-year-old"),
all_matched_data_4 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="4-year-old"),
all_matched_data %>% select(asqse_risk, year_completed_cat) %>% mutate(age="5-year-old")) %>%
ggplot(., aes(x = year_completed_cat, y = asqse_risk, group = age, color = age)) +
stat_summary(geom = "line", fun = mean, size=1) +
stat_summary(geom = "errorbar",fun.data = mean_se, width = .1) +
labs(x = "Year the ASQ-SE was completed", y = "Percentage of Children at Risk", color = "Child's age") +
theme_bw() + theme(legend.position = "bottom")
## Warning: pacote 'ggrepel' foi compilado no R versĂŁo 4.4.2bind_rows(
all_matched_data_1 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="1-year-old"),
all_matched_data_2 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="2-year-old"),
all_matched_data_3 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="3-year-old"),
all_matched_data_4 %>% select(asqse_risk, year_completed_cat) %>% mutate(age="4-year-old"),
all_matched_data %>% select(asqse_risk, year_completed_cat) %>% mutate(age="5-year-old")
) %>%
ggplot(aes(x = as.numeric(as.character(year_completed_cat)), y = asqse_risk, group = age, color = age)) +
stat_summary(geom = "line", fun = mean, size = 1) +
stat_summary(geom = "errorbar", fun.data = mean_se, width = .1) +
# Create a separate data frame for the labels at year 19
geom_label_repel(
data = . %>%
group_by(age, year_completed_cat) %>%
summarise(asqse_risk = mean(asqse_risk), .groups = "drop") %>%
filter(year_completed_cat == 19),
aes(label = age),
nudge_x = -0.5,
direction = "y",
segment.size = 1,
box.padding = 0.5,
point.padding = 0.1,
force = 5,
show.legend = FALSE
) +
# Set a distinct color palette that is colorblind-friendly
scale_color_manual(values = c(
"1-year-old" = "black", # Orange
"2-year-old" = "blue", # Blue
"3-year-old" = "darkgreen", # Green
"4-year-old" = "purple", # Pink/magenta
"5-year-old" = "red" # Red/brown
)) +
labs(x = "Year the ASQ-SE was completed",
y = "Percentage of Children at Risk",
color = "Child's age") +
theme_bw() +
theme(legend.position = "bottom")
! done