Parenting behaviors
Emily
2017-09-06
1 Data preprocessing
Preliminaries.
## [1] "dplyr" "langcog" "tidyr" "ggplot2" "lme4"##
## Attaching package: 'langcog'## The following object is masked from 'package:base':
##
## scale## Loading tidyverse: ggplot2
## Loading tidyverse: tibble
## Loading tidyverse: tidyr
## Loading tidyverse: readr
## Loading tidyverse: purrr
## Loading tidyverse: dplyr## Conflicts with tidy packages ----------------------------------------------## %+%(): ggplot2, psych
## alpha(): ggplot2, psych
## filter(): dplyr, stats
## lag(): dplyr, stats##
## Attaching package: 'ggthemes'## The following objects are masked from 'package:langcog':
##
## scale_color_solarized, scale_colour_solarized,
## scale_fill_solarized## Loading required package: Matrix##
## Attaching package: 'Matrix'## The following object is masked from 'package:tidyr':
##
## expand##
## Attaching package: 'lmerTest'## The following object is masked from 'package:lme4':
##
## lmer## The following object is masked from 'package:stats':
##
## step## This is lavaan 0.5-23.1097## lavaan is BETA software! Please report any bugs.##
## Attaching package: 'lavaan'## The following object is masked from 'package:langcog':
##
## sem## sROC 0.1-2 loaded##
## Attaching package: 'MVN'## The following object is masked from 'package:psych':
##
## mardiaRead in participant data.
data <- read.csv("../cdm_paq.csv", header =TRUE)
dem <- read.csv("../cdm_paq_dem.csv", header =TRUE)
labels <- read.csv("sent_forms.csv")
labels$sent <- as.character(labels$sent)
questions <- data %>%
filter(Status == "Response Type")%>%
select(Q1:Q28)%>%
gather("item", "sent", Q1:Q28)
d <- data %>%
filter(Finished == 1)%>%
mutate(sid = ResponseId)%>%
select(-Status, -StartDate, - EndDate, - IPAddress, -Progress, -Duration..in.seconds., -Finished, - ResponseId, - RecordedDate, -RecipientLastName, -RecipientFirstName, - RecipientEmail, -ExternalReference, -LocationLatitude, -LocationLongitude, - DistributionChannel)%>%
select(sid, Q1:Q28)%>%
gather("item", "rating", Q1:Q28)%>%
left_join(questions)
subinfo <- dem %>%
filter(Finished == "True")%>%
transmute(sid = ResponseId, ethnicity = Q25.1, parent_ed = Q26.1, parent_age = Q27.1, parent_gender = Q28.1, num_kids = Q29, oldest_kid = Q30, youngest_kid = Q32, only_kid = Q33)Make data frames.
d$sent <- stringr::str_replace_all(d$sent, "’", "")
dq <- d %>%
left_join(labels)
#rescore reverse coded items
dq$rating <- as.numeric(dq$rating)
dq$rating[dq$reverse_code == 1] <- 8 - dq$rating[dq$reverse_code == 1]2 PAQ
Test for normality.
dq_wide <- dq %>%
select(sid, short_sent, rating)%>%
spread(short_sent, rating)
x_vars <- dq_wide %>%
select(-sid)
uniPlot(x_vars[1:10], type = "histogram")
uniPlot(x_vars[11:20], type = "histogram")
uniPlot(x_vars[21:24], type = "histogram")
#histograms for CDM
#get subset for plotting
hist_items <- c("Q8", "Q22", "Q12", "Q16")
hist <- d %>%
select(sid, item, rating)%>%
filter(item %in% hist_items)%>%
spread(item, rating)
ggplot(hist, aes(Q8, fill = Q8)) +
geom_histogram(stat = "count") 
ggplot(hist, aes(Q22, fill = Q22)) +
geom_histogram(stat = "count") 
ggplot(hist, aes(Q12, fill = Q12)) +
geom_histogram(stat = "count") 
ggplot(hist, aes(Q16, fill = Q16)) +
geom_histogram(stat = "count") 
#get highest and lowest rated items
d$sent <- stringr::str_replace_all(d$sent, "’", "")
dqg <- d %>%
left_join(labels)
dqg$rating <- as.numeric(dqg$rating)
ms <- dqg %>%
group_by(category, short_sent, reverse_code) %>%
multi_boot_standard(col = "rating", na.rm = TRUE) Get mean ratings for sentences.
dq$rating <- dq$rating - 1
ms <- dq %>%
group_by(category, short_sent, reverse_code) %>%
multi_boot_standard(col = "rating", na.rm = TRUE) %>%
arrange(category, desc(mean))
ms$short_sent_ord <- factor(ms$short_sent,
levels = ms$short_sent)
short_sent_ord <- ms$short_sent_ordPlot responses to individual questionnaire items.
qplot(short_sent_ord, mean, col = category,
ymin = ci_lower, ymax = ci_upper, pch = factor(reverse_code),
geom = "pointrange",
data = ms) +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = .5)) +
xlab("") +
ylab("Mean Rating") +
ylim(c(0,6)) +
scale_colour_solarized()
Compare to Mturk samples. We are comparing the data collected from CDM parents to experiment 9 (parents and non-parents) of the questionnaire norming study and the data from the parenting behaviors study (self-reported parents).
cdm <- ms %>%
ungroup()%>%
select(category, short_sent, mean, ci_upper, ci_lower)%>%
mutate(sample = "cdm")
load("e9.RData")
load("beh.RData")
load("e9_full.RData")
load("beh_dq.RData")
beh_d <- beh %>%
ungroup()%>%
mutate(category = category_paq)%>%
select(category, short_sent, mean, ci_upper, ci_lower)%>%
mutate(sample = "mturk1")
atts <- c("AA", "EL","RR")
e9_d <- e9 %>%
filter(category %in% atts)%>%
ungroup()%>%
select(category, short_sent, mean, ci_upper, ci_lower)%>%
mutate(sample = "mturk2")
ss_cdm <- dq %>%
dplyr::group_by(sid, category) %>%
dplyr::summarise(rating = mean(rating))%>%
mutate(sample = "cdm")
ss_e9 <- d_full_e9 %>%
filter(category %in% atts)%>%
mutate(sid = workerid)%>%
select(-workerid)%>%
dplyr::group_by(sid, category) %>%
dplyr::summarise(rating = mean(rating)) %>%
mutate(sample = "mturk1")
ss_beh <- beh_dq %>%
dplyr::group_by(sid, category_paq) %>%
dplyr::summarise(rating = mean(rating)) %>%
mutate(category = category_paq)%>%
select(-category_paq)%>%
mutate(sample = "mturk2")
ss_cdm$rating <- as.numeric(ss_cdm$rating)
ss_beh$rating <- as.numeric(ss_beh$rating)
ss_e9$rating <- as.numeric(ss_e9$rating)
ss_cdm$category <- as.factor(as.character(ss_cdm$category))
ss_beh$category <- as.factor(as.character(ss_beh$category))
ss_e9$category <- as.factor(as.character(ss_e9$category))
ss_compare<- bind_rows(ss_cdm, ss_beh)%>%
bind_rows(ss_e9)
samp_compare <- e9_d %>%
bind_rows(cdm)%>%
bind_rows(beh_d)
samp_compare$short_sent <- factor(samp_compare$short_sent, levels = short_sent_ord)
samp_compare$sample <- factor(samp_compare$sample, levels = c("cdm","mturk1","mturk2"))
ss_compare$sample <- factor(ss_compare$sample, levels = c("cdm","mturk1","mturk2"))
ss_compare$category <- factor(ss_compare$category, levels = c("AA","EL","RR"))
qplot(short_sent, mean, col = sample,
ymin = ci_lower, ymax = ci_upper,
geom = "pointrange",
data = samp_compare) +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = .5)) +
xlab("") +
ylab("Mean Rating") +
ylim(c(0,6)) +
scale_colour_solarized()
#plot subset of questions for delivery to CDM parents.
cdm_items <- c("learn before speaking", "read to kids", "learn by playing", "explore and experiment")
ggplot(filter(samp_compare, short_sent %in% cdm_items), aes(short_sent, mean, fill=sample)) +
geom_bar(stat="identity", position = "dodge")
CDM parents agree more with EL items related to exploration and play but less on teaching math before school compared to Mturk. They also rate RR items quite a bit lower than Mturk samples.
In general there is a bit more variability in agreement with items within subscales compared to Mturk samples.
Plot mean subscale scores.
atts_m <- dq %>%
group_by(category) %>%
multi_boot_standard(col = "rating", na.rm = TRUE) %>%
arrange(category, desc(mean))
ggplot(atts_m, aes(x = category, y = mean)) +
geom_bar(stat="identity") +
geom_linerange(aes(ymin = ci_lower, ymax = ci_upper),
position = position_dodge(width = .9))+
ylim(c(0,6)) 
3 Scale reliability
3.1 Whole Scale
wide.paq <- dq %>%
select(sid, short_sent, rating) %>%
spread(short_sent, rating)
alpha.rr <- as.matrix(select(wide.paq, -sid))
psych::alpha(x = alpha.rr)## Some items ( calm children when upset explore and experiment learn by playing too much attention does not spoil ) were negatively correlated with the total scale and
## probably should be reversed.
## To do this, run the function again with the 'check.keys=TRUE' option##
## Reliability analysis
## Call: psych::alpha(x = alpha.rr)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd
## 0.68 0.68 0.72 0.08 2.1 0.017 5 0.41
##
## lower alpha upper 95% confidence boundaries
## 0.64 0.68 0.71
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc)
## calm children when upset 0.68 0.68 0.72
## close bonds for relationships 0.67 0.66 0.70
## comfort children 0.67 0.67 0.71
## consequences break rules 0.66 0.67 0.71
## do as told 0.65 0.66 0.70
## explain rules 0.68 0.67 0.72
## explore and experiment 0.68 0.68 0.72
## grateful to parents 0.66 0.67 0.71
## help deal with emotions 0.66 0.66 0.70
## learn about math before school 0.65 0.66 0.70
## learn before speaking 0.67 0.67 0.71
## learn by playing 0.68 0.68 0.72
## need to feel emotionally close 0.67 0.66 0.71
## not ok to boss around caregivers 0.66 0.66 0.71
## not ok to see adults as equals 0.65 0.66 0.70
## pay attention to likes 0.66 0.66 0.70
## read to kids 0.67 0.67 0.71
## respect adults 0.64 0.65 0.69
## should not make decisions 0.68 0.68 0.72
## talk to babies 0.67 0.66 0.70
## teach kids to prepare for school 0.65 0.66 0.70
## too much affection does not make weak 0.67 0.67 0.71
## too much attention does not spoil 0.69 0.68 0.71
## worry about misbehavior 0.66 0.66 0.71
## average_r S/N alpha se
## calm children when upset 0.084 2.1 0.017
## close bonds for relationships 0.078 2.0 0.018
## comfort children 0.080 2.0 0.018
## consequences break rules 0.080 2.0 0.018
## do as told 0.079 2.0 0.019
## explain rules 0.082 2.1 0.017
## explore and experiment 0.083 2.1 0.017
## grateful to parents 0.080 2.0 0.018
## help deal with emotions 0.078 1.9 0.018
## learn about math before school 0.078 1.9 0.019
## learn before speaking 0.082 2.1 0.018
## learn by playing 0.083 2.1 0.017
## need to feel emotionally close 0.079 2.0 0.018
## not ok to boss around caregivers 0.079 2.0 0.018
## not ok to see adults as equals 0.077 1.9 0.019
## pay attention to likes 0.078 1.9 0.018
## read to kids 0.082 2.1 0.017
## respect adults 0.074 1.8 0.020
## should not make decisions 0.084 2.1 0.018
## talk to babies 0.078 1.9 0.018
## teach kids to prepare for school 0.078 1.9 0.019
## too much affection does not make weak 0.080 2.0 0.018
## too much attention does not spoil 0.083 2.1 0.017
## worry about misbehavior 0.079 2.0 0.018
##
## Item statistics
## n raw.r std.r r.cor r.drop mean
## calm children when upset 680 0.20 0.22 0.13 0.061 4.3
## close bonds for relationships 680 0.33 0.39 0.35 0.229 5.4
## comfort children 680 0.29 0.33 0.27 0.180 5.3
## consequences break rules 680 0.42 0.34 0.29 0.282 3.9
## do as told 680 0.48 0.38 0.35 0.353 3.5
## explain rules 680 0.22 0.26 0.18 0.116 5.4
## explore and experiment 680 0.15 0.25 0.16 0.089 5.8
## grateful to parents 680 0.44 0.35 0.30 0.291 3.5
## help deal with emotions 680 0.36 0.42 0.37 0.272 5.6
## learn about math before school 680 0.48 0.42 0.39 0.349 5.0
## learn before speaking 680 0.18 0.28 0.19 0.125 5.9
## learn by playing 680 0.17 0.25 0.16 0.090 5.8
## need to feel emotionally close 680 0.29 0.36 0.31 0.205 5.6
## not ok to boss around caregivers 680 0.39 0.36 0.31 0.285 5.2
## not ok to see adults as equals 680 0.51 0.43 0.42 0.377 4.0
## pay attention to likes 680 0.37 0.40 0.36 0.263 5.2
## read to kids 680 0.21 0.28 0.20 0.128 5.8
## respect adults 680 0.59 0.52 0.52 0.489 4.7
## should not make decisions 680 0.28 0.22 0.14 0.152 2.9
## talk to babies 680 0.32 0.41 0.36 0.268 5.9
## teach kids to prepare for school 680 0.49 0.42 0.40 0.352 4.7
## too much affection does not make weak 680 0.26 0.35 0.29 0.192 5.8
## too much attention does not spoil 680 0.19 0.25 0.18 0.049 4.8
## worry about misbehavior 680 0.41 0.37 0.32 0.304 4.9
## sd
## calm children when upset 1.32
## close bonds for relationships 1.09
## comfort children 1.16
## consequences break rules 1.50
## do as told 1.41
## explain rules 1.01
## explore and experiment 0.65
## grateful to parents 1.64
## help deal with emotions 0.97
## learn about math before school 1.45
## learn before speaking 0.54
## learn by playing 0.78
## need to feel emotionally close 0.86
## not ok to boss around caregivers 1.08
## not ok to see adults as equals 1.60
## pay attention to likes 1.15
## read to kids 0.79
## respect adults 1.39
## should not make decisions 1.30
## talk to babies 0.52
## teach kids to prepare for school 1.64
## too much affection does not make weak 0.67
## too much attention does not spoil 1.42
## worry about misbehavior 1.15
##
## Non missing response frequency for each item
## 0 1 2 3 4 5 6
## calm children when upset 0.01 0.02 0.05 0.20 0.26 0.25 0.21
## close bonds for relationships 0.01 0.01 0.00 0.06 0.08 0.16 0.69
## comfort children 0.01 0.01 0.01 0.07 0.08 0.24 0.59
## consequences break rules 0.02 0.05 0.09 0.28 0.18 0.21 0.18
## do as told 0.02 0.06 0.12 0.36 0.18 0.16 0.10
## explain rules 0.00 0.01 0.01 0.04 0.08 0.22 0.64
## explore and experiment 0.00 0.00 0.00 0.01 0.02 0.14 0.82
## grateful to parents 0.04 0.07 0.11 0.34 0.14 0.13 0.17
## help deal with emotions 0.01 0.01 0.00 0.02 0.05 0.16 0.75
## learn about math before school 0.02 0.02 0.02 0.11 0.13 0.15 0.55
## learn before speaking 0.00 0.00 0.00 0.01 0.00 0.05 0.93
## learn by playing 0.01 0.00 0.00 0.01 0.02 0.08 0.88
## need to feel emotionally close 0.01 0.00 0.00 0.02 0.04 0.17 0.75
## not ok to boss around caregivers 0.00 0.01 0.01 0.06 0.13 0.25 0.54
## not ok to see adults as equals 0.03 0.06 0.06 0.21 0.23 0.21 0.20
## pay attention to likes 0.01 0.01 0.00 0.08 0.10 0.24 0.56
## read to kids 0.01 0.00 0.00 0.00 0.01 0.05 0.93
## respect adults 0.01 0.02 0.04 0.14 0.16 0.21 0.42
## should not make decisions 0.05 0.09 0.13 0.47 0.15 0.08 0.03
## talk to babies 0.00 0.00 0.00 0.00 0.01 0.03 0.95
## teach kids to prepare for school 0.02 0.04 0.04 0.18 0.09 0.11 0.52
## too much affection does not make weak 0.01 0.00 0.00 0.01 0.03 0.09 0.86
## too much attention does not spoil 0.01 0.02 0.05 0.13 0.12 0.23 0.43
## worry about misbehavior 0.00 0.01 0.01 0.10 0.21 0.29 0.37
## miss
## calm children when upset 0
## close bonds for relationships 0
## comfort children 0
## consequences break rules 0
## do as told 0
## explain rules 0
## explore and experiment 0
## grateful to parents 0
## help deal with emotions 0
## learn about math before school 0
## learn before speaking 0
## learn by playing 0
## need to feel emotionally close 0
## not ok to boss around caregivers 0
## not ok to see adults as equals 0
## pay attention to likes 0
## read to kids 0
## respect adults 0
## should not make decisions 0
## talk to babies 0
## teach kids to prepare for school 0
## too much affection does not make weak 0
## too much attention does not spoil 0
## worry about misbehavior 03.2 Rules and Respect Subscale
wide.paq <- dq %>%
filter(category == "RR") %>%
select(sid, short_sent, rating) %>%
spread(short_sent, rating)
alpha.rr <- as.matrix(select(wide.paq, -sid))
psych::alpha(x = alpha.rr)##
## Reliability analysis
## Call: psych::alpha(x = alpha.rr)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd
## 0.73 0.73 0.72 0.25 2.7 0.015 4.1 0.82
##
## lower alpha upper 95% confidence boundaries
## 0.7 0.73 0.76
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N
## consequences break rules 0.71 0.70 0.69 0.25 2.4
## do as told 0.68 0.68 0.66 0.23 2.1
## grateful to parents 0.71 0.71 0.69 0.26 2.4
## not ok to boss around caregivers 0.71 0.71 0.69 0.26 2.4
## not ok to see adults as equals 0.67 0.67 0.65 0.23 2.1
## respect adults 0.68 0.68 0.66 0.23 2.1
## should not make decisions 0.73 0.73 0.71 0.27 2.6
## worry about misbehavior 0.71 0.71 0.70 0.26 2.5
## alpha se
## consequences break rules 0.017
## do as told 0.018
## grateful to parents 0.017
## not ok to boss around caregivers 0.017
## not ok to see adults as equals 0.019
## respect adults 0.019
## should not make decisions 0.016
## worry about misbehavior 0.016
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## consequences break rules 680 0.58 0.57 0.46 0.39 3.9 1.5
## do as told 680 0.67 0.67 0.61 0.53 3.5 1.4
## grateful to parents 680 0.59 0.55 0.45 0.38 3.5 1.6
## not ok to boss around caregivers 680 0.50 0.55 0.44 0.36 5.2 1.1
## not ok to see adults as equals 680 0.70 0.69 0.64 0.54 4.0 1.6
## respect adults 680 0.68 0.67 0.62 0.53 4.7 1.4
## should not make decisions 680 0.46 0.47 0.34 0.29 2.9 1.3
## worry about misbehavior 680 0.49 0.53 0.41 0.34 4.9 1.1
##
## Non missing response frequency for each item
## 0 1 2 3 4 5 6 miss
## consequences break rules 0.02 0.05 0.09 0.28 0.18 0.21 0.18 0
## do as told 0.02 0.06 0.12 0.36 0.18 0.16 0.10 0
## grateful to parents 0.04 0.07 0.11 0.34 0.14 0.13 0.17 0
## not ok to boss around caregivers 0.00 0.01 0.01 0.06 0.13 0.25 0.54 0
## not ok to see adults as equals 0.03 0.06 0.06 0.21 0.23 0.21 0.20 0
## respect adults 0.01 0.02 0.04 0.14 0.16 0.21 0.42 0
## should not make decisions 0.05 0.09 0.13 0.47 0.15 0.08 0.03 0
## worry about misbehavior 0.00 0.01 0.01 0.10 0.21 0.29 0.37 03.3 Affection and Attachment Subscale
wide.paq <- dq %>%
filter(category == "AA") %>%
select(sid, short_sent, rating) %>%
spread(short_sent, rating)
alpha.aa <- as.matrix(select(wide.paq, -sid))
psych::alpha(x = alpha.aa)##
## Reliability analysis
## Call: psych::alpha(x = alpha.aa)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd
## 0.58 0.6 0.58 0.16 1.5 0.024 5.2 0.56
##
## lower alpha upper 95% confidence boundaries
## 0.53 0.58 0.63
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc)
## calm children when upset 0.58 0.60 0.57
## close bonds for relationships 0.53 0.54 0.52
## comfort children 0.53 0.55 0.53
## help deal with emotions 0.56 0.58 0.55
## need to feel emotionally close 0.55 0.57 0.55
## pay attention to likes 0.55 0.57 0.55
## too much affection does not make weak 0.55 0.56 0.53
## too much attention does not spoil 0.54 0.55 0.52
## average_r S/N alpha se
## calm children when upset 0.17 1.5 0.024
## close bonds for relationships 0.14 1.2 0.027
## comfort children 0.15 1.2 0.027
## help deal with emotions 0.16 1.4 0.025
## need to feel emotionally close 0.16 1.3 0.026
## pay attention to likes 0.16 1.3 0.026
## too much affection does not make weak 0.15 1.3 0.026
## too much attention does not spoil 0.15 1.2 0.027
##
## Item statistics
## n raw.r std.r r.cor r.drop mean
## calm children when upset 680 0.48 0.42 0.25 0.20 4.3
## close bonds for relationships 680 0.56 0.57 0.49 0.36 5.4
## comfort children 680 0.56 0.54 0.43 0.34 5.3
## help deal with emotions 680 0.45 0.48 0.34 0.25 5.6
## need to feel emotionally close 680 0.45 0.50 0.37 0.28 5.6
## pay attention to likes 680 0.51 0.50 0.37 0.28 5.2
## too much affection does not make weak 680 0.45 0.53 0.42 0.32 5.8
## too much attention does not spoil 680 0.60 0.55 0.45 0.33 4.8
## sd
## calm children when upset 1.32
## close bonds for relationships 1.09
## comfort children 1.16
## help deal with emotions 0.97
## need to feel emotionally close 0.86
## pay attention to likes 1.15
## too much affection does not make weak 0.67
## too much attention does not spoil 1.42
##
## Non missing response frequency for each item
## 0 1 2 3 4 5 6
## calm children when upset 0.01 0.02 0.05 0.20 0.26 0.25 0.21
## close bonds for relationships 0.01 0.01 0.00 0.06 0.08 0.16 0.69
## comfort children 0.01 0.01 0.01 0.07 0.08 0.24 0.59
## help deal with emotions 0.01 0.01 0.00 0.02 0.05 0.16 0.75
## need to feel emotionally close 0.01 0.00 0.00 0.02 0.04 0.17 0.75
## pay attention to likes 0.01 0.01 0.00 0.08 0.10 0.24 0.56
## too much affection does not make weak 0.01 0.00 0.00 0.01 0.03 0.09 0.86
## too much attention does not spoil 0.01 0.02 0.05 0.13 0.12 0.23 0.43
## miss
## calm children when upset 0
## close bonds for relationships 0
## comfort children 0
## help deal with emotions 0
## need to feel emotionally close 0
## pay attention to likes 0
## too much affection does not make weak 0
## too much attention does not spoil 03.4 Early Learning Subscale
wide.paq <- dq %>%
filter(category == "EL") %>%
select(sid, short_sent, rating) %>%
spread(short_sent, rating)
alpha.el <- as.matrix(select(wide.paq, -sid))
psych::alpha(x = alpha.el)##
## Reliability analysis
## Call: psych::alpha(x = alpha.el)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd
## 0.41 0.41 0.43 0.081 0.71 0.031 5.5 0.44
##
## lower alpha upper 95% confidence boundaries
## 0.35 0.41 0.47
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r
## explain rules 0.42 0.41 0.42 0.089
## explore and experiment 0.40 0.40 0.42 0.087
## learn about math before school 0.24 0.35 0.33 0.071
## learn before speaking 0.41 0.40 0.41 0.086
## learn by playing 0.41 0.40 0.41 0.086
## read to kids 0.41 0.39 0.41 0.084
## talk to babies 0.38 0.35 0.37 0.070
## teach kids to prepare for school 0.30 0.37 0.34 0.077
## S/N alpha se
## explain rules 0.69 0.029
## explore and experiment 0.67 0.031
## learn about math before school 0.54 0.043
## learn before speaking 0.65 0.031
## learn by playing 0.66 0.030
## read to kids 0.64 0.030
## talk to babies 0.53 0.033
## teach kids to prepare for school 0.58 0.040
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## explain rules 680 0.37 0.40 0.19 0.087 5.4 1.01
## explore and experiment 680 0.29 0.41 0.20 0.109 5.8 0.65
## learn about math before school 680 0.70 0.50 0.46 0.381 5.0 1.45
## learn before speaking 680 0.26 0.42 0.22 0.109 5.9 0.54
## learn by playing 680 0.31 0.41 0.22 0.091 5.8 0.78
## read to kids 680 0.32 0.43 0.25 0.103 5.8 0.79
## talk to babies 680 0.38 0.51 0.38 0.240 5.9 0.52
## teach kids to prepare for school 680 0.69 0.47 0.41 0.306 4.7 1.64
##
## Non missing response frequency for each item
## 0 1 2 3 4 5 6 miss
## explain rules 0.00 0.01 0.01 0.04 0.08 0.22 0.64 0
## explore and experiment 0.00 0.00 0.00 0.01 0.02 0.14 0.82 0
## learn about math before school 0.02 0.02 0.02 0.11 0.13 0.15 0.55 0
## learn before speaking 0.00 0.00 0.00 0.01 0.00 0.05 0.93 0
## learn by playing 0.01 0.00 0.00 0.01 0.02 0.08 0.88 0
## read to kids 0.01 0.00 0.00 0.00 0.01 0.05 0.93 0
## talk to babies 0.00 0.00 0.00 0.00 0.01 0.03 0.95 0
## teach kids to prepare for school 0.02 0.04 0.04 0.18 0.09 0.11 0.52 04 Demographics
Create a data frame that has subscale scores.
ss <- dq %>%
dplyr::group_by(sid, category) %>%
dplyr::summarise(rating = mean(rating))
ss <- left_join(ss, subinfo)4.1 By Gender
| Estimate | Std. Error | df | t value | Pr(>|t|) | |
|---|---|---|---|---|---|
| (Intercept) | 5.5227273 | 0.0625985 | 1953.093 | 88.224637 | 0.0000000 |
| categoryAA | -0.5946970 | 0.0827199 | 1344.001 | -7.189288 | 0.0000000 |
| categoryRR | -1.5631313 | 0.0827199 | 1344.001 | -18.896686 | 0.0000000 |
| parent_genderFemale | 0.0085771 | 0.0677735 | 1953.093 | 0.126555 | 0.8993056 |
| categoryAA:parent_genderFemale | 0.3601318 | 0.0895583 | 1344.001 | 4.021199 | 0.0000611 |
| categoryRR:parent_genderFemale | 0.1279139 | 0.0895583 | 1344.001 | 1.428276 | 0.1534449 |
Women agree with AA items more than men.
4.2 By Parent Age
| Estimate | Std. Error | df | t value | Pr(>|t|) | |
|---|---|---|---|---|---|
| (Intercept) | 5.4379790 | 0.1537098 | 1908.464 | 35.3782235 | 0.0000000 |
| categoryAA | 0.1489978 | 0.2041008 | 1308.000 | 0.7300207 | 0.4655083 |
| categoryRR | -1.5921409 | 0.2041008 | 1308.000 | -7.8007573 | 0.0000000 |
| parent_age_con | 0.0024748 | 0.0040775 | 1908.464 | 0.6069451 | 0.5439595 |
| categoryAA:parent_age_con | -0.0116295 | 0.0054143 | 1308.000 | -2.1479255 | 0.0319029 |
| categoryRR:parent_age_con | 0.0037600 | 0.0054143 | 1308.000 | 0.6944592 | 0.4875176 |
There is a small effect of age such that older parents agree less with AA items.
4.3 By Education
Unfortunately there is not much variability in Parent Education- most respondants have at least a college education.
| Estimate | Std. Error | df | t value | Pr(>|t|) | |
|---|---|---|---|---|---|
| (Intercept) | 5.5336181 | 0.0238715 | 1962.014 | 231.8083231 | 0.0000000 |
| categoryAA | -0.2891902 | 0.0317771 | 1342.000 | -9.1005740 | 0.0000000 |
| categoryRR | -1.4578380 | 0.0317771 | 1342.000 | -45.8769460 | 0.0000000 |
| scale(parent_ed_con) | 0.0295716 | 0.0238774 | 1962.014 | 1.2384750 | 0.2156881 |
| categoryAA:scale(parent_ed_con) | -0.0224081 | 0.0317850 | 1342.000 | -0.7049885 | 0.4809397 |
| categoryRR:scale(parent_ed_con) | -0.0697710 | 0.0317850 | 1342.000 | -2.1950927 | 0.0283277 |
There is a small effect of parent education such that parents with higher levels of education agree more with EL items.
4.4 By Number of Kids
| Estimate | Std. Error | df | t value | Pr(>|t|) | |
|---|---|---|---|---|---|
| (Intercept) | 5.6180129 | 0.0622564 | 1929.9 | 90.2398592 | 0.0000000 |
| categoryAA | -0.2995080 | 0.0814558 | 1340.0 | -3.6769403 | 0.0002454 |
| categoryRR | -1.7459085 | 0.0814558 | 1340.0 | -21.4338219 | 0.0000000 |
| num_kids | -0.0485468 | 0.0321300 | 1929.9 | -1.5109493 | 0.1309651 |
| categoryAA:num_kids | 0.0065732 | 0.0420386 | 1340.0 | 0.1563606 | 0.8757723 |
| categoryRR:num_kids | 0.1639263 | 0.0420386 | 1340.0 | 3.8994272 | 0.0001012 |
With more kids, parents agree more with RR itmes and less with AA and EL items.
4.5 By Ethnicity
| Estimate | Std. Error | df | t value | Pr(>|t|) | |
|---|---|---|---|---|---|
| (Intercept) | 5.5780347 | 0.0330930 | 1700.939 | 168.5565220 | 0.0000000 |
| categoryAA | -0.2742052 | 0.0441219 | 1162.001 | -6.2147204 | 0.0000000 |
| categoryRR | -1.5516618 | 0.0441219 | 1162.001 | -35.1676209 | 0.0000000 |
| ethnicityAsian | -0.1276670 | 0.0543378 | 1700.939 | -2.3495076 | 0.0189121 |
| ethnicityHispanic or Latino | -0.0522994 | 0.1106338 | 1700.939 | -0.4727251 | 0.6364700 |
| categoryAA:ethnicityAsian | -0.0726085 | 0.0724470 | 1162.001 | -1.0022297 | 0.3164413 |
| categoryRR:ethnicityAsian | 0.2109756 | 0.0724470 | 1162.001 | 2.9121372 | 0.0036582 |
| categoryAA:ethnicityHispanic or Latino | 0.2190581 | 0.1475049 | 1162.001 | 1.4850907 | 0.1377910 |
| categoryRR:ethnicityHispanic or Latino | 0.3494560 | 0.1475049 | 1162.001 | 2.3691144 | 0.0179935 |
4.6 By Sample
| Estimate | Std. Error | df | t value | Pr(>|t|) | |
|---|---|---|---|---|---|
| (Intercept) | 5.2406250 | 0.0301878 | 2677.965 | 173.600863 | 0.0000000 |
| categoryEL | 0.2895221 | 0.0331790 | 2364.681 | 8.726062 | 0.0000000 |
| categoryRR | -1.1610294 | 0.0331790 | 2364.681 | -34.992896 | 0.0000000 |
| samplemturk1 | -0.4699247 | 0.0580625 | 2795.563 | -8.093436 | 0.0000000 |
| samplemturk2 | -0.5129448 | 0.0578955 | 2794.330 | -8.859840 | 0.0000000 |
| categoryEL:samplemturk1 | 0.1184921 | 0.0641817 | 2364.681 | 1.846196 | 0.0649887 |
| categoryRR:samplemturk1 | 0.7890536 | 0.0641817 | 2364.681 | 12.294051 | 0.0000000 |
| categoryEL:samplemturk2 | 0.0759779 | 0.0639933 | 2364.681 | 1.187279 | 0.2352369 |
| categoryRR:samplemturk2 | 0.8100294 | 0.0639933 | 2364.681 | 12.658026 | 0.0000000 |
##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "cdm" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk1" & ss_compare$category == "AA"] and "AA"]
## t = 7.6792, df = 315.7, p-value = 2.025e-13
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.3544233 0.5986009
## sample estimates:
## mean of x mean of y
## 5.240625 4.764113##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "cdm" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk2" & ss_compare$category == "AA"] and "AA"]
## t = 7.6647, df = 305.97, p-value = 2.391e-13
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.3847366 0.6505134
## sample estimates:
## mean of x mean of y
## 5.240625 4.723000##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "mturk1" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk2" & ss_compare$category == "AA"] and "AA"]
## t = 0.47474, df = 491.96, p-value = 0.6352
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.1290411 0.2112669
## sample estimates:
## mean of x mean of y
## 4.764113 4.723000##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "cdm" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk1" & ss_compare$category == "EL"] and "EL"]
## t = 6.1837, df = 294.74, p-value = 2.085e-09
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.2440749 0.4719652
## sample estimates:
## mean of x mean of y
## 5.530147 5.172127##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "cdm" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk2" & ss_compare$category == "EL"] and "EL"]
## t = 6.449, df = 282.26, p-value = 4.87e-10
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 0.3068447 0.5764495
## sample estimates:
## mean of x mean of y
## 5.530147 5.088500##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "mturk1" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk2" & ss_compare$category == "EL"] and "EL"]
## t = 0.96789, df = 481.23, p-value = 0.3336
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.08614422 0.25339826
## sample estimates:
## mean of x mean of y
## 5.172127 5.088500##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "cdm" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk1" & ss_compare$category == "RR"] and "RR"]
## t = -4.6215, df = 391.22, p-value = 5.181e-06
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.4455013 -0.1795817
## sample estimates:
## mean of x mean of y
## 4.079596 4.392137##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "cdm" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk2" & ss_compare$category == "RR"] and "RR"]
## t = -4.1598, df = 380.44, p-value = 3.94e-05
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.4306145 -0.1541943
## sample estimates:
## mean of x mean of y
## 4.079596 4.372000##
## Welch Two Sample t-test
##
## data: ss_compare$rating[ss_compare$sample == "mturk1" & ss_compare$category == and ss_compare$rating[ss_compare$sample == "mturk2" & ss_compare$category == "RR"] and "RR"]
## t = 0.23195, df = 495.01, p-value = 0.8167
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.1504370 0.1907111
## sample estimates:
## mean of x mean of y
## 4.392137 4.372000