childes_weird
https://rpubs.com/gloukatou/733656
R Markdown
library(childesr)
library(wordbankr)
library(ggplot2)
library(dplyr)
library(tidyr)
library(purrr)
require(scales)
library(RColorBrewer)
myPalette <- brewer.pal(5, "Set2") Load annotations
#####TODOS first part is the descriptive section, to what extent the full range of populations is represented, what are some codes that are missing often,
##### descriptive section and representation of the population
##########systematic review of features
##################confront eg proportion of child population versus proportion of data by eg continent
#############second goal: point out strengths and limitations of the current CHILDES, explain what is missing
annotations1<- read.csv("Childes_corpora-Clara1.csv")
annotations2<- read.csv("Childes_corpora-Camila.csv")
annotations3<- read.csv("Childes_corpora-Lara.csv")
annotations <-rbind(annotations1, annotations2, annotations3) #merge annotations from 3 annotators
annotations<- rename(annotations, lang = Language)
annotations$Bilingualism.in.corpus..yes..no.[annotations$Bilingualism.in.corpus..yes..no. == "yes"] <- "Bilinguals"
annotations$Bilingualism.in.corpus..yes..no.[annotations$Bilingualism.in.corpus..yes..no. == "no"] <- "Monolinguals"
annotations <- annotations %>% filter(!(Inclusion=="")) #only keep annotated corpora
annotations[annotations==""]<-NA
annotations[annotations=="N/A"]<-NA
annotations[annotations=="N/A (0,33)"]<-NA
annotations[annotations=="N/A (0)"]<-NA
annotations[annotations=="NA (0)"]<-NA
#head(annotations)253 corpora already annotated.
Optional chunk merging with childesr for extra information
d_transcripts <- get_transcripts()
d_statistics <- get_speaker_statistics()
mergedChiAnnot <- annotations %>% left_join(d_transcripts)
mergedChiAnnot <- mergedChiAnnot %>% left_join(n_utt_d)
mergedChiAnnot <- mergedChiAnnot %>% left_join(n_utt_speaker)
mergedChiAnnot<-mergedChiAnnot[!(is.na(mergedChiAnnot$n_utt) | mergedChiAnnot$n_utt=="" ), ]
if ( length(unique(mergedChiAnnot$corpus_name)) < nrow(annotations) ) {
annotations$corpus_name[!annotations$corpus_name %in% mergedChiAnnot$corpus_name]
} #print non merged corporaCorpus inclusion
Based on “Inclusion” annotation column
How many corpora should be included?
inclu_data<- annotations %>%
group_by(Inclusion) %>%
count()
ggplot(inclu_data,aes(y=Inclusion, x=n)) +
geom_bar(stat="identity", width=1, color="white") +
theme_minimal() +
ggtitle("Corpus inclusion") 
annotations <- annotations %>%
filter(Inclusion=="yes") From the total amount of corpora, 141 have been included in this study. The following analyses will continue only on these included corpora.
Bilinguals
Based on “Bilingualism” annotation column
How many corpora are bilingual?
group_count <-function(data_, column_){
data_<- data_ %>%
group_by({{column_}}) %>%
count()
return(data_)
}
byBilinguals<-group_count(annotations, Bilingualism.in.corpus..yes..no.)
lbls <- paste(byBilinguals$Bilingualism.in.corpus..yes..no., byBilinguals$n, sep=" ") # add numbers to labels
pie(byBilinguals$n, labels=lbls, border="white", col=myPalette)
head(byBilinguals)## # A tibble: 4 x 2
## # Groups: Bilingualism.in.corpus..yes..no. [4]
## Bilingualism.in.corpus..yes..no. n
## <chr> <int>
## 1 Bilinguals 27
## 2 Monolinguals 29
## 3 trilinguism 1
## 4 <NA> 8484 corpora have NA in “Bilingualism in corpus”.
Siblings
Based on “Number of children with siblings” annotation column
How many of the target children have siblings?
byPart <- annotations %>%
group_by(lang, Bilingualism.in.corpus..yes..no.) %>%
summarise(numpar = sum(as.numeric(Number.of.participants), na.rm=TRUE))
bySibl<- annotations %>%
group_by(lang) %>%
summarise(numChildrenwithSiblings = sum(as.numeric(X..of.children.with.siblings)))
ggplot(bySibl[!is.na(bySibl$numChildrenwithSiblings),], mapping = aes(y=lang, x=numChildrenwithSiblings)) +
geom_point() +
ggtitle("Languages where number of children with siblings is not unknown") +
geom_col()
head(bySibl)## # A tibble: 6 x 2
## lang numChildrenwithSiblings
## <chr> <dbl>
## 1 Afrikaan 1
## 2 Arabic NA
## 3 Basque NA
## 4 Brazilian Portuguese NA
## 5 Cantonese 4
## 6 Cantonese-English 6Out of 1687 total target children, only 49 have siblings. No information for children in 65 languages (marked as NAs).
Based on “Average number of siblings” annotation column
How many siblings in average?
bySiblnum<-annotations %>%
group_by(Average.number.of.siblings) %>%
count()
ggplot(bySiblnum, mapping = aes(x=Average.number.of.siblings, y=n)) +
geom_point() +
ggtitle("Average number of siblings per corpus") +
geom_col()
#ggplot(bySiblnum[!is.na(bySiblnum$Average.number.of.siblings),], mapping = aes(x=Average.number.of.siblings, y=n))
annotations$Average.number.of.siblings <- as.numeric(as.character(annotations$Average.number.of.siblings))
head(bySiblnum)## # A tibble: 6 x 2
## # Groups: Average.number.of.siblings [6]
## Average.number.of.siblings n
## <chr> <int>
## 1 0 10
## 2 0.25 1
## 3 0.33 1
## 4 0.66 1
## 5 0.75 1
## 6 0.875 1The mean number of siblings has been left NA for 88 out of 141 corpora. In the remaining corpora, the mean number of siblings per corpus is 0.992451 with min 0 and max 2.
Based on “Average number of older siblings” annotation column
How many older siblings in average?
byoldSiblnum<-annotations %>%
group_by(Average.number.of.older.siblings) %>%
count()
ggplot(byoldSiblnum, mapping = aes(x=Average.number.of.older.siblings, y=n)) +
geom_point() +
ggtitle("Average number of older siblings per corpus") +
geom_col()
annotations$Average.number.of.older.siblings <- as.numeric(as.character(annotations$Average.number.of.older.siblings))
head(byoldSiblnum)## # A tibble: 6 x 2
## # Groups: Average.number.of.older.siblings [6]
## Average.number.of.older.siblings n
## <chr> <int>
## 1 0 27
## 2 0,5 3
## 3 0,6 1
## 4 0.25 1
## 5 0.3 2
## 6 0.5 1The mean number of older siblings has been left NA for 89 out of 141 corpora. In the remaining corpora, the mean number of older siblings per corpus is 0.4288043 with min NA and max 2.
Sessions
Based on “Mean sessions per child” annotation column
How many sessions per language?
annotations <- annotations[!is.na(annotations$Mean...sessions.per.child), ]
bySess<- annotations
bySess$Mean...sessions.per.child <- sub(",",".",bySess$Mean...sessions.per.child)
bySess$Mean...sessions.per.child <- as.numeric(bySess$Mean...sessions.per.child)
bySess<- bySess %>%
mutate(sessions = as.numeric(Number.of.participants) * Mean...sessions.per.child) %>%
group_by(lang, Bilingualism.in.corpus..yes..no.) %>%
summarise(numses = sum(as.numeric(sessions)))
ggplot(bySess, mapping = aes(y = reorder(lang, numses), numses)) + geom_point() + ggtitle("Number of sessions per language") + geom_col()+ facet_wrap(~Bilingualism.in.corpus..yes..no., scale="free")
Location
Based on “Neighbour.village.city.province.state.country” annotation column
For now only based on continents and countries. Lots of handcoding.
annotations$continent <-NA
annotations$continent[grep("England|Sweden|Spain|Madrid, Spain ; Tenerife, Canary Islands|Madrid, Spain|Navarra, Spain|Estonia|Tartu, Estonia ; Rapla, Estonia|Budapest, Hungary|West Coast, Ireland|Arfon area Gwynedd, North Wales|Netherlands|Belfast, Northern Ireland|Nottingham/Manchester, England|Sweden ; Portugal|Czech Republic|Bucharest, Romania|Postdam, Germany|Stuttgart, Germany|France|Alt penedes, region of catalonia|spain|Naples|Spain, Hungary, Lloret de mar|Spain, Salamanca|England, Brighton|France (Normandy, Marseille, + places visited)|Portugal|italy|Barcelona|Switzerland|Stockholm, Sweden|Iceland|Poznań, Poland|Romania|Antwerp, Belgium|Ireland|Hungary|Athens, Greece|Southern Estonia|Tartu, Estonia|A Coruña, Santiago de Compostela ; Lugo ; Pontevedra, A Estrada ; SPAIN|Moscow, Russia|Spain, Salamanca|Wales", annotations$Neighbour.village.city.province.state.country)] <- "Europe"
annotations$continent[grep("Rio Cuarto, Cordoba, Argentina|Sao Paulo|Patzcuaro, Michoacan, iMexico|Jamaica", annotations$Neighbour.village.city.province.state.country)] <- "Central and South America"
annotations$continent[grep("Iran|Israel|Bombay, India|China|Bangkok, Thailand|Tokyo area|Nagoya area|osaka area|Kuwait|Yahud, Israel|Nagoya center|Kusatsu City, Shiga Pref|Hong-Kong|Taiwan|Turkey", annotations$Neighbour.village.city.province.state.country)] <- "Asia"
annotations$continent[grep("Michigan, USA|USA, Northern Virginia|Canada|California, USA|washington dc|United States|USA", annotations$Neighbour.village.city.province.state.country)] <- "North America"
annotations$continent[grep("Alexandria, Egypt|Mokhotlong, Lesotho", annotations$Neighbour.village.city.province.state.country)] <- "Africa"
annotations$country <-NA
annotations$country[grep("England|West Coast, Ireland|Arfon area Gwynedd, North Wales|Belfast, Northern Ireland|Nottingham/Manchester, England|England, Brighton|Ireland|Wales", annotations$Neighbour.village.city.province.state.country)] <- "UK/Ireland"
annotations$country[grep("Sweden|Sweden ; Portugal|Stockholm, Sweden", annotations$Neighbour.village.city.province.state.country)] <- "Sweden"
annotations$country[grep("Spain|Madrid, Spain ; Tenerife, Canary Islands|Madrid, Spain|Navarra, Spain|Alt penedes, region of catalonia|spain|Spain, Hungary, Lloret de mar|Spain, Salamanca|Barcelona|A Coruña, Santiago de Compostela ; Lugo ; Pontevedra, A Estrada ; SPAIN|Spain, Salamanca", annotations$Neighbour.village.city.province.state.country)] <- "Spain"
annotations$country[grep("Estonia|Tartu, Estonia ; Rapla, Estonia|Southern Estonia|Tartu, Estonia", annotations$Neighbour.village.city.province.state.country)] <- "Estonia"
annotations$country[grep("Budapest, Hungary|Hungary|Spain, Hungary, Lloret de mar", annotations$Neighbour.village.city.province.state.country)] <- "Hungary"
annotations$country[grep("Sweden ; Portugal|Portugal", annotations$Neighbour.village.city.province.state.country)] <- "Portugal"
annotations$country[grep("Czech Republic", annotations$Neighbour.village.city.province.state.country)] <- "Czech Republic"
annotations$country[grep("Bucharest, Romania|Romania", annotations$Neighbour.village.city.province.state.country)] <- "Romania"
annotations$country[grep("Postdam, Germany|Stuttgart, Germany", annotations$Neighbour.village.city.province.state.country)] <- "Germany"
annotations$country[grep("France|France (Normandy, Marseille, + places visited)", annotations$Neighbour.village.city.province.state.country)] <- "France"
annotations$country[grep("Naples|italy", annotations$Neighbour.village.city.province.state.country)] <- "Italy"
annotations$country[grep("Switzerland", annotations$Neighbour.village.city.province.state.country)] <- "Switzerland"
annotations$country[grep("Iceland", annotations$Neighbour.village.city.province.state.country)] <- "Iceland"
annotations$country[grep("Poznań, Poland", annotations$Neighbour.village.city.province.state.country)] <- "Poland"
annotations$country[grep("Antwerp, Belgium", annotations$Neighbour.village.city.province.state.country)] <- "Belgium"
annotations$country[grep("Athens, Greece", annotations$Neighbour.village.city.province.state.country)] <- "Greece"
annotations$country[grep("Moscow, Russia", annotations$Neighbour.village.city.province.state.country)] <- "Russia"
annotations$country[grep("Rio Cuarto, Cordoba, Argentina", annotations$Neighbour.village.city.province.state.country)] <- "Argentina"
annotations$country[grep("Sao Paulo", annotations$Neighbour.village.city.province.state.country)] <- "Brazil"
annotations$country[grep("Patzcuaro, Michoacan, iMexico", annotations$Neighbour.village.city.province.state.country)] <- "Mexico"
annotations$country[grep("Jamaica", annotations$Neighbour.village.city.province.state.country)] <- "Jamaica"
annotations$country[grep("Iran", annotations$Neighbour.village.city.province.state.country)] <- "Iran"
annotations$country[grep("Israel|Yahud, Israel", annotations$Neighbour.village.city.province.state.country)] <- "Israel"
annotations$country[grep("Bombay, India", annotations$Neighbour.village.city.province.state.country)] <- "India"
annotations$country[grep("China", annotations$Neighbour.village.city.province.state.country)] <- "China"
annotations$country[grep("Bangkok, Thailand", annotations$Neighbour.village.city.province.state.country)] <- "Thailand"
annotations$country[grep("Tokyo area|Nagoya area|osaka area|Nagoya center|Kusatsu City, Shiga Pref", annotations$Neighbour.village.city.province.state.country)] <- "Japan"
annotations$country[grep("Kuwait", annotations$Neighbour.village.city.province.state.country)] <- "Kuwait"
annotations$country[grep("Hong-Kong", annotations$Neighbour.village.city.province.state.country)] <- "Hong-Kong"
annotations$country[grep("Taiwan", annotations$Neighbour.village.city.province.state.country)] <- "Taiwan"
annotations$country[grep("Turkey", annotations$Neighbour.village.city.province.state.country)] <- "Turkey"
annotations$country[grep("Michigan, USA|USA, Northern Virginia|California, USA|washington dc|United States|USA", annotations$Neighbour.village.city.province.state.country)] <- "USA"
annotations$country[grep("Canada", annotations$Neighbour.village.city.province.state.country)] <- "Canada"
annotations$country[grep("Alexandria, Egypt", annotations$Neighbour.village.city.province.state.country)] <- "Egypt"
annotations$country[grep("Mokhotlong, Lesotho", annotations$Neighbour.village.city.province.state.country)] <- "Lesotho"Corpora by countries and continents
byLocation <- annotations %>%
select(country, continent) %>%
group_by(country, continent) %>%
count()
ggplot(byLocation, aes(x=n, y=country)) +
geom_bar(stat="identity", width=1, color="white") +
ggtitle("Corpora by Countries and Continents") +
facet_wrap(~continent, scale="free") 
head(byLocation)## # A tibble: 6 x 3
## # Groups: country, continent [6]
## country continent n
## <chr> <chr> <int>
## 1 Argentina Central and South America 1
## 2 Belgium Europe 2
## 3 Brazil Central and South America 1
## 4 Canada North America 2
## 5 China Asia 3
## 6 Czech Republic Europe 136 countries are represented in childes. 5 continents. 68 of the 102 corpora are European, and 10 of the corpora are North American.
Based on “Neighbour.village.city.province.state.country” annotation column
Corpora by countries and continents
byLocation1 <- group_count(annotations, continent)
lbls <- paste(byLocation1$continent, byLocation1$n, sep=" ") # add numbers to labels
pie(byLocation1$n, labels=lbls, border="white", col=myPalette, main="Corpora by continent")
head(byLocation1)## # A tibble: 6 x 2
## # Groups: continent [6]
## continent n
## <chr> <int>
## 1 Africa 2
## 2 Asia 20
## 3 Central and South America 4
## 4 Europe 66
## 5 North America 8
## 6 <NA> 2Place of recording
Based on “Place.of.recordings.” annotation column
Will need handcoding.
byPlaceofRecording <- group_count(annotations, Place.of.recordings...home..nursery....)
lbls <- paste(byPlaceofRecording$Place.of.recordings...home..nursery...., byPlaceofRecording$n, sep=" ") # add numbers to labels
pie(byPlaceofRecording$n, labels=lbls, border="white", col=myPalette)
head(byPlaceofRecording)## # A tibble: 6 x 2
## # Groups: Place.of.recordings...home..nursery.... [6]
## Place.of.recordings...home..nursery.... n
## <chr> <int>
## 1 6 in nursery, 3 at home 1
## 2 home 69
## 3 home (mostly) 1
## 4 home (TYA) ; home of mother, of paternal grandparents or maternal grand… 1
## 5 home or daycare 1
## 6 home or kindergarten 1Recording method
Based on “Recording.method..static..portable..both..diary..other..” annotation column
Will need handcoding. For now only prints column
byRecord <- annotations %>%
group_by(Presence.of.investigator..yes.no.parent.is.investigator.half.half., Recording.method..static..portable..both..diary..other..) %>%
summarise(n = n()) %>%
arrange(desc(n))
byRecord %>%
filter(!is.na(Presence.of.investigator..yes.no.parent.is.investigator.half.half.), !is.na(Recording.method..static..portable..both..diary..other..)) %>%
rename( number_of_corpora = n)## # A tibble: 38 x 3
## # Groups: Presence.of.investigator..yes.no.parent.is.investigator.half.half.
## # [9]
## Presence.of.investigator..yes.… Recording.method..static..p… number_of_corpo…
## <chr> <chr> <int>
## 1 yes portable 7
## 2 yes static 5
## 3 yes audio 3
## 4 yes videotape 3
## 5 no audio recorded 2
## 6 no portable 2
## 7 no static 2
## 8 no videotape 2
## 9 yes audio & videotape 2
## 10 yes audiotape 2
## # … with 28 more rows#lbls <- paste(byRecord$Presence.of.investigator..yes.no.parent.is.investigator.half.half., byRecord$Recording.method..static..portable..both..diary..other.., byRecord$n, sep=" & ") # add numbers to labels
#pie(byRecord$n, labels=lbls, border="white", col=myPalette)21 corpora have NA in “Presence.of.investigator” column.
33 corpora have NA in “Recording.method” column.
Recording age
Based on “Mean.child.age.at.beginning.of.recordings, Mean.child.age.at.end.of.recordings” annotation columns
byAge <- annotations %>%
select(Mean.child.age.at.beginning.of.recordings, Mean.child.age.at.end.of.recordings, country, continent) %>%
group_by(country, continent) %>%
summarize(min = min(Mean.child.age.at.beginning.of.recordings), max = max(Mean.child.age.at.end.of.recordings)) %>%
arrange(desc(min))
ggplot(byAge, aes(y=country))+
geom_linerange(aes(xmin=min,xmax=max),linetype=2,color="blue")+
geom_point(aes(x=min),size=3,color="red")+
geom_point(aes(x=max),size=3,color="red")+
ggtitle("Nim and Max age per country") +
theme_bw()+
facet_wrap(~continent, scale="free") 
Yes/No Tiers
Based on “CDS.annotation, English.translation, Morphological.annotation, Only.directed.to.child.transcribed, Correction.of.speech.errors” annotation columns
Mostly NAs. For now just prints column
group_select<-function(data_, column_){
data_ <- data_%>%
group_by(!!column_) %>%
select(Corpus, !!column_)
data_ <- data_[ order( data_[,3] ),]
return(data_)
}
byCdsTier <-group_select(annotations, "CDS.annotation..yes.no.")
byCdsTier <- byCdsTier %>% group_by(CDS.annotation..yes.no.) %>% summarise(n = n())
head(byCdsTier)## # A tibble: 3 x 2
## CDS.annotation..yes.no. n
## <chr> <int>
## 1 no 11
## 2 yes 1
## 3 <NA> 90byannotTier <-group_select(annotations, "English.translation..yes.no")
byannotTier <- byannotTier %>% group_by(English.translation..yes.no) %>% summarise(n = n())
head(byannotTier)## # A tibble: 6 x 2
## English.translation..yes.no n
## <chr> <int>
## 1 "-" 3
## 2 "- " 1
## 3 "no" 39
## 4 "no (from what I can see in the transcripts)" 5
## 5 "yes" 9
## 6 <NA> 45bymorphTier <- group_select(annotations, "Morphological.annotation..yes.no.")
bymorphTier <- bymorphTier %>% group_by(Morphological.annotation..yes.no.) %>% summarise(n = n())
head(bymorphTier)## # A tibble: 3 x 2
## Morphological.annotation..yes.no. n
## <chr> <int>
## 1 no 13
## 2 yes 28
## 3 <NA> 61byCDStransTier <- group_select(annotations, "Only.directed.to.child.transcribed..i.e..speech.between.adults.not.transcribed...yes.no")
byCDStransTier <- byCDStransTier %>% group_by(Only.directed.to.child.transcribed..i.e..speech.between.adults.not.transcribed...yes.no) %>% summarise(n = n())
head(byCDStransTier)## # A tibble: 5 x 2
## Only.directed.to.child.transcribed..i.e..speech.between.adults.not.tran… n
## <chr> <int>
## 1 mostly 1
## 2 no 3
## 3 yes 3
## 4 yes (long conversations between adults were not transcribed) 1
## 5 <NA> 94byCorrections<- group_select(annotations, "Correction.of.speech.errors..i.e..syntactic...transcriptions.not.exactly.matching.speech...yes.no")
byCorrections <- byCorrections %>% group_by(Correction.of.speech.errors..i.e..syntactic...transcriptions.not.exactly.matching.speech...yes.no) %>% summarise(n = n())
head(byCorrections)## # A tibble: 3 x 2
## Correction.of.speech.errors..i.e..syntactic...transcriptions.not.exactl… n
## <chr> <int>
## 1 no 3
## 2 yes 8
## 3 <NA> 91Weird variables
Based on “Fertility, Interbirth intervals, Schooling, Community, Healty, Culture, Mortality, Household…” annotation columns
Mostly NAs. For now just prints column
byFertility <- group_select(annotations, "Fertility.rate.of.community")
byFertility <- byFertility %>% group_by(Fertility.rate.of.community) %>% summarise(n = n())
byFertility## # A tibble: 2 x 2
## Fertility.rate.of.community n
## * <chr> <int>
## 1 no 1
## 2 <NA> 101byInterbirth <- group_select(annotations, "Interbirth.intervals")
byInterbirth <- byInterbirth %>% group_by(Interbirth.intervals) %>% summarise(n = n())
byInterbirth## # A tibble: 3 x 2
## Interbirth.intervals n
## * <chr> <int>
## 1 Na 1
## 2 no 1
## 3 <NA> 100bySchooling <- group_select(annotations, "Access.to.schooling.for.children..yes.no.only.elementary.")
bySchooling <- bySchooling %>% group_by(Access.to.schooling.for.children..yes.no.only.elementary.) %>% summarise(n = n())
bySchooling## # A tibble: 2 x 2
## Access.to.schooling.for.children..yes.no.only.elementary. n
## * <chr> <int>
## 1 yes 8
## 2 <NA> 94byCommunity <- group_select(annotations, "Type.of.community..hunter.forager.herder.farmer.work.for.pay.industrial.")
byCommunity <- byCommunity %>% group_by(Type.of.community..hunter.forager.herder.farmer.work.for.pay.industrial.) %>% summarise(n = n())
byCommunity## # A tibble: 3 x 2
## Type.of.community..hunter.forager.herder.farmer.work.for.pay.industrial. n
## * <chr> <int>
## 1 industrial 3
## 2 rural 1
## 3 <NA> 98byHealth <- group_select(annotations, "Access.to.health.for.children...yes.no.")
byHealth <- byHealth %>% group_by(Access.to.health.for.children...yes.no.) %>% summarise(n = n())
byHealth ## # A tibble: 2 x 2
## Access.to.health.for.children...yes.no. n
## * <chr> <int>
## 1 yes 2
## 2 <NA> 100byCulture <- group_select(annotations, "Type.of.culture..minority..majority.")
byCulture <- byCulture %>% group_by(Type.of.culture..minority..majority.) %>% summarise(n = n())
byCulture## # A tibble: 4 x 2
## Type.of.culture..minority..majority. n
## * <chr> <int>
## 1 majority 1
## 2 majority? 2
## 3 minority 2
## 4 <NA> 97byMortality<- group_select(annotations, "Mortality.rate.of.children.in.the.community")
byMortality<- byMortality %>% group_by(Mortality.rate.of.children.in.the.community) %>% summarise(n = n())
byMortality## # A tibble: 1 x 2
## Mortality.rate.of.children.in.the.community n
## * <chr> <int>
## 1 <NA> 102byLangSpeakers<- group_select(annotations, "Number.of.speakers.of.the.language")
byLangSpeakers<- byLangSpeakers %>% group_by(Number.of.speakers.of.the.language) %>% summarise(n = n())
byLangSpeakers## # A tibble: 3 x 2
## Number.of.speakers.of.the.language n
## * <chr> <int>
## 1 4 million 1
## 2 70% of population 1
## 3 <NA> 100byHousehold<- group_select(annotations, "Household.structure..nuclear..extended.")
byhousehold<- byHousehold %>% group_by(Household.structure..nuclear..extended.) %>% summarise(n = n())
byhousehold## # A tibble: 3 x 2
## Household.structure..nuclear..extended. n
## * <chr> <int>
## 1 extended 1
## 2 nuclear 10
## 3 <NA> 91byTimespent<- group_select(annotations, "Where.children.spend.their.time..home..nursery..playing.by.the.river....")
byTimespent<- byTimespent %>% group_by(Where.children.spend.their.time..home..nursery..playing.by.the.river....) %>% summarise(n = n())
byTimespent## # A tibble: 15 x 2
## Where.children.spend.their.time..home..nursery..playing.by.the.river..… n
## * <chr> <int>
## 1 "assistante maternelle francophone de 2;4 a 3;1 ans a mi-temps (20H pa… 1
## 2 "crèche, home" 1
## 3 "crèche/grandmother 1 day a week" 1
## 4 "daycare" 1
## 5 "daycare ; grandparents/aunt " 1
## 6 "grandparents ; family with 2 children & a dog " 1
## 7 "home, nursery" 1
## 8 "home+dutch babysitter, and when school started: english speaking pres… 1
## 9 "kindergarten" 1
## 10 "mother for the first year and then daycare 3h/day" 1
## 11 "nursery" 1
## 12 "nursery school" 1
## 13 "nursery school ; home with babysitter " 1
## 14 "playgroups & nursery school" 1
## 15 <NA> 88Parental education, profession
Based on “Parental.education, Parental.profession” annotation columns
byParentEducation<- annotations[!is.na(annotations$Parental.education),] %>%
group_by(lang, continent, Parental.education) %>%
count()
byParentEducation <- byParentEducation %>%
rename(
number_of_corpora = n,
language = lang,
Parental.education.NAs_excluded = Parental.education)
byParentEducation## # A tibble: 17 x 4
## # Groups: language, continent, Parental.education.NAs_excluded [17]
## language continent Parental.education.NAs_excluded number_of_corpo…
## <chr> <chr> <chr> <int>
## 1 Basque Europe "Fathers : primary education ; … 1
## 2 catalan Europe "yes" 2
## 3 Czech Europe "high school ; university " 1
## 4 dutch/english Europe "university graduates" 1
## 5 Estonian Europe "4 mothers w/ higher education … 1
## 6 French Europe "college degree" 1
## 7 French Europe "university education" 2
## 8 french/russi… Europe "mother has 2 PhDs and father i… 1
## 9 German Europe "university degree or some kind… 1
## 10 Hebrew Asia "high education" 1
## 11 Hebrew Asia "well-educated" 2
## 12 Hungarian/ca… Europe "Both parents possess universit… 1
## 13 japanese Asia "Na" 1
## 14 Polish Europe "highly-educated" 1
## 15 Spanish Central and … "university" 1
## 16 Spanish Europe "higher degrees" 1
## 17 Spanish Europe "higher education college degre… 1102 corpora have NA for parental education.
byParentProfession<- annotations[!is.na(annotations$Parental.profession),] %>%
group_by(lang, continent, Parental.profession) %>%
count()
byParentProfession <- byParentProfession %>%
rename(
number_of_corpora = n,
language = lang,
Parental.profession.NAs_excluded = Parental.profession)
byParentProfession## # A tibble: 25 x 4
## # Groups: language, continent, Parental.profession.NAs_excluded [25]
## language continent Parental.profession.NAs_excluded number_of_corpo…
## <chr> <chr> <chr> <int>
## 1 Basque Europe "Fathers : industrial operator … 1
## 2 Brazilian P… Central and S… "fahter linguist professor, mot… 1
## 3 Cantonese Asia "varied (mass transport company… 1
## 4 Cantonese-E… Asia "varied" 1
## 5 catalan Europe "mother: researcher" 2
## 6 Czech Europe "university scientist ; cultura… 1
## 7 Dutch Europe "researcher" 2
## 8 dutch/engli… Europe "mother: part time free lance j… 1
## 9 English Europe "researcher" 1
## 10 English/Spa… Europe "mother: linguist (investigator… 1
## # … with 15 more rows102 corpora have NA for parental education.
Parental SES
Based on “Parental.ses” annotation column
byParentSES<- annotations[!is.na(annotations$Parental.socioeconomic.status),] %>%
group_by(lang, continent, Parental.socioeconomic.status) %>%
count()
byParentSES <- byParentSES %>%
rename(
number_of_corpora = n,
language = lang,
Parental.socioeconomic.status.NAs_excluded = Parental.socioeconomic.status)
byParentSES ## # A tibble: 26 x 4
## # Groups: language, continent, Parental.socioeconomic.status.NAs_excluded
## # [26]
## language continent Parental.socioeconomic.sta… number_of_corpo…
## <chr> <chr> <chr> <int>
## 1 Basque Europe medium 1
## 2 Brazilian Por… Central and Sout… upper-middle class 1
## 3 Cantonese Asia working class 1
## 4 catalan Europe middle class 1
## 5 Dutch Europe lower-middle to middle-mid… 1
## 6 dutch/english Europe upper middle class 1
## 7 English Europe middle-class 1
## 8 English Europe predominantly middle-class 1
## 9 English Europe upper working class 1
## 10 English/Spani… Europe middle class 1
## # … with 16 more rows102 corpora have NA for parental education.
Number of children in CHILDES
Based on “Number.of.participants” annotation column
Number of target children per language
byPart %>%
ggplot(mapping = aes(y = reorder(lang, numpar), numpar)) +
geom_point() +
ggtitle("Number of target children per language") +
geom_col()+
facet_wrap(~Bilingualism.in.corpus..yes..no., scale="free") +
labs(y= "Languages", x = "Number of target children")
byPart_nona <- byPart[!is.na(byPart$Bilingualism.in.corpus..yes..no.),]
byPart_mono <- byPart_nona %>%
filter(Bilingualism.in.corpus..yes..no.=="Monolinguals")
byPart_bili <- byPart_nona %>%
filter(Bilingualism.in.corpus..yes..no.=="Bilinguals")
head(byPart)## # A tibble: 6 x 3
## # Groups: lang [5]
## lang Bilingualism.in.corpus..yes..no. numpar
## <chr> <chr> <dbl>
## 1 Afrikaan <NA> 2
## 2 Arabic <NA> 10
## 3 Basque Bilinguals 38
## 4 Basque Monolinguals 8
## 5 Brazilian Portuguese Monolinguals 1
## 6 Cantonese <NA> 80 corpora have NA in “Number of participants”. There is data from 153 target children in Monolingual corpora. There is data from 1318 target children in corpora which have not been classified as Monolingual/Bilingual. There is data from 215 target children in Bilingual corpora. Overall, the mean number of target children per language is 23.109589 with min 1 and max 475.
Based on “Number.of.participants” annotation column
Number of target children per continent
byPartCont <- annotations %>%
group_by(continent) %>%
select(Number.of.participants, continent) %>%
summarise(numpar = sum(as.numeric(Number.of.participants), na.rm=TRUE)) ## Warning in mask$eval_all_summarise(quo): NAs introduced by coercionbyPartCont %>%
ggplot(mapping = aes(y = reorder(continent, numpar), numpar)) +
ggtitle("Number of target children per continent") +
geom_col()+
labs(y= "Continents", x = "Number of target children")
head(byPartCont)## # A tibble: 6 x 2
## continent numpar
## <chr> <dbl>
## 1 Africa 14
## 2 Asia 274
## 3 Central and South America 5
## 4 Europe 301
## 5 North America 42
## 6 <NA> 4301 target children in Europe, 14 in Africa, 42 in North America, 274in Asia, 5 in Central and South America.