Crosstabs
Dan Yurovsky
September 30, 2014
Load Required Libraries
## Loading required package: Matrix
## Loading required package: Rcpp
##
## Attaching package: 'dplyr'
##
## The following objects are masked from 'package:stats':
##
## filter, lag
##
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
##
## Loading required package: quadprog
## Loading required package: DBI
##
## Attaching package: 'RMySQL'
##
## The following objects are masked from 'package:RSQLite':
##
## dbBuildTableDefinition, isIdCurrent, safe.writeLoad Database
Load WG productive/receptive data
wg.vocab.words <- as.data.frame(select(
wg.table,basetable_ptr_id,col_baabaa:col_some))
names(wg.vocab.words)[1] <- "id"
wg.bykid.longform <- melt(wg.vocab.words,id.vars="id",
variable.name = "word",
value.name="produces") %>%
mutate(understands = produces >= 1,
produces = produces == 2)
wg.scores <- wg.bykid.longform %>%
group_by(id) %>%
summarise(productive = sum(produces),
receptive = sum(understands))Load WS productive data
ws.vocab.words <- as.data.frame(select(
ws.table,basetable_ptr_id,col_baabaa:col_connthen))
names(ws.vocab.words)[1] <- "id"
ws.bykid.longform <- melt(ws.vocab.words,id.vars="id",
variable.name = "word",
value.name="produces")
ws.scores <- ws.bykid.longform %>%
group_by(id) %>%
summarise(productive = sum(produces))Load child-specific data
admins <- as.data.frame(select(admin.table,data_id,child_id,age,source_id))
children <- as.data.frame(select(child.table,id,gender,mom_ed,birth_order))
child.data <- merge(children,admins,by.y="child_id",by.x = "id")
wg.word.data <- merge(child.data,wg.bykid.longform)
wg.child.scores <- merge(child.data,wg.scores)
ws.word.data <- merge(child.data,ws.bykid.longform)
ws.child.scores <- merge(child.data,ws.scores)Compute crosstabs
wg.age.gender.data <- wg.child.scores %>%
gather(measure,score,c(receptive,productive)) %>%
group_by(age,gender,measure) %>%
summarise(
mean = mean(score),
ci.l = ci.low(score),
ci.h = ci.high(score),
n = n())
#Show WG table
wg.age.gender.data## Source: local data frame [44 x 7]
## Groups: age, gender
##
## age gender measure mean ci.l ci.h n
## 1 8 F receptive 44.094 19.3453 21.3781 32
## 2 8 F productive 2.375 1.0938 1.3758 32
## 3 8 M receptive 41.800 13.7436 17.8721 35
## 4 8 M productive 1.657 0.8000 0.9429 35
## 5 9 F receptive 44.688 10.1885 11.5630 48
## 6 9 F productive 2.375 0.8130 0.8542 48
## 7 9 M receptive 32.902 9.2488 9.6104 41
## 8 9 M productive 1.512 0.6104 0.6341 41
## 9 10 F receptive 63.328 12.5473 16.8914 64
## 10 10 F productive 5.219 1.6723 2.1098 64
## 11 10 M receptive 49.167 9.7920 10.0014 72
## 12 10 M productive 3.208 0.8337 0.8476 72
## 13 11 F receptive 95.267 23.1589 27.9372 45
## 14 11 F productive 11.844 6.3778 10.4250 45
## 15 11 M receptive 60.840 13.5830 13.0210 50
## 16 11 M productive 5.200 1.9205 2.5010 50
## 17 12 F receptive 92.709 11.0142 10.7725 79
## 18 12 F productive 10.544 2.2165 2.3560 79
## 19 12 M receptive 77.064 11.5385 12.3247 78
## 20 12 M productive 9.821 2.7311 3.0519 78
## 21 13 F receptive 129.263 18.2303 21.0553 57
## 22 13 F productive 17.912 3.6145 3.4390 57
## 23 13 M receptive 106.733 19.3779 19.3904 60
## 24 13 M productive 13.267 4.7842 6.1854 60
## 25 14 F receptive 157.655 20.3047 22.5267 58
## 26 14 F productive 28.414 8.9147 9.0207 58
## 27 14 M receptive 149.170 21.4920 21.0953 53
## 28 14 M productive 25.057 7.2642 10.0014 53
## 29 15 F receptive 172.857 28.9832 29.8429 49
## 30 15 F productive 57.388 19.9597 23.5005 49
## 31 15 M receptive 164.477 26.5017 28.3489 44
## 32 15 M productive 36.795 13.7977 17.0256 44
## 33 16 F receptive 218.571 28.4107 28.9048 42
## 34 16 F productive 85.262 22.1946 23.8929 42
## 35 16 M receptive 165.643 21.9917 20.3595 42
## 36 16 M productive 35.095 8.3345 9.5512 42
## 37 17 F receptive 224.818 23.5212 23.0386 33
## 38 17 F productive 71.212 20.3348 23.8303 33
## 39 17 M receptive 199.634 23.6732 21.4244 41
## 40 17 M productive 59.073 17.9518 20.3921 41
## 41 18 F receptive 261.027 29.7595 30.8797 37
## 42 18 F productive 104.514 24.0838 27.5405 37
## 43 18 M receptive 241.138 29.5759 27.8284 29
## 44 18 M productive 63.345 18.1397 18.1750 29ws.age.gender.data <- ws.child.scores %>%
group_by(age,gender) %>%
filter(age>= 16, age <= 30) %>%
summarise(
mean = mean(productive),
ci.l = ci.low(productive),
ci.h = ci.high(productive),
n = n())
#Show WS table
ws.age.gender.data## Source: local data frame [30 x 6]
## Groups: age
##
## age gender mean ci.l ci.h n
## 1 16 F 73.02 15.00 16.71 95
## 2 16 M 53.32 14.44 15.79 93
## 3 17 F 84.63 19.60 21.74 70
## 4 17 M 66.19 14.32 15.44 63
## 5 18 F 135.95 17.76 18.48 153
## 6 18 M 100.88 13.45 14.17 180
## 7 19 F 173.41 22.71 22.50 126
## 8 19 M 134.07 21.48 21.93 101
## 9 20 F 194.89 28.50 28.46 88
## 10 20 M 162.16 28.36 28.57 88
## 11 21 F 201.06 41.38 42.93 53
## 12 21 M 188.00 34.30 36.78 61
## 13 22 F 290.69 42.69 42.71 62
## 14 22 M 238.18 47.63 49.97 49
## 15 23 F 304.98 44.02 42.70 66
## 16 23 M 303.04 33.59 34.33 71
## 17 24 F 359.43 23.30 22.99 195
## 18 24 M 284.81 22.31 22.64 208
## 19 25 F 394.01 30.10 27.63 114
## 20 25 M 302.84 29.38 27.18 109
## 21 26 F 430.46 43.62 38.84 65
## 22 26 M 348.49 43.00 42.12 69
## 23 27 F 433.86 42.33 40.20 65
## 24 27 M 380.00 39.50 38.23 73
## 25 28 F 455.77 34.42 36.03 48
## 26 28 M 381.53 48.75 44.75 55
## 27 29 F 480.96 44.48 44.06 54
## 28 29 M 420.16 41.49 41.10 64
## 29 30 F 538.16 21.05 18.91 113
## 30 30 M 500.30 25.04 22.57 115Plot WG
quartz(width=7,height=4)
ggplot(wg.age.gender.data,
aes(x=age, y=mean,colour=gender,linetype=measure,
group=interaction(gender,measure)))+
geom_pointrange(aes(ymin = mean-ci.l,
ymax = mean+ci.h),
size = .8,
show_guide = FALSE) +
geom_line(size=1) +
scale_x_continuous(breaks=seq(8,18,1),
name = "Age (months)")+
scale_y_continuous(name = "Vocabulary Size (words)") +
theme_bw(base_size=14) +
scale_color_manual(values = c("#e41a1c","#377eb8"))
Plot WS
quartz(width=6,height=4)
ggplot(ws.age.gender.data,
aes(x=age, y=mean,colour=gender,label=gender))+
geom_pointrange(aes(ymin = mean-ci.l,
ymax = mean+ci.h),
size = .8,
show_guide = FALSE) +
geom_line(size=1) +
scale_x_continuous(breaks=seq(16,30,1),limits=c(16,30.5),
name = "Age (months)")+
scale_y_continuous(name = "Vocabulary Size (words)",limits=c(0,680)) +
theme_bw(base_size=14) +
theme(legend.position="none") +
geom_dl(method=list("last.qp",cex=1,hjust=-.5)) +
scale_color_manual(values = c("#e41a1c","#377eb8"))## Loading required package: proto