library(tidyverse)
## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
## ✓ ggplot2 3.3.5 ✓ purrr 0.3.4
## ✓ tibble 3.1.4 ✓ dplyr 1.0.7
## ✓ tidyr 1.1.4 ✓ stringr 1.4.0
## ✓ readr 2.0.2 ✓ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
prefrosh = read.csv("~/Downloads/Stanford Communities Project - Prefrosh 2022-2023_September 28, 2022_10.26.csv", header=T, na.strings=c("","NA"))
# Removing unceccessary rows
prefrosh <- prefrosh[-c(1, 2, 548), ]
prefrosh = prefrosh %>%
distinct(RecipientEmail, .keep_all = TRUE)
For presentation, (1) an overview of what’s included in prefrosh
survey, (2) some sample demographics (gender, political ideology, urm
etc), (3) some temporal trends (e.g., how is this class rate in
happiness, stress, depression compared to past prefrosh samples), and
(4) some hypothesis testing (e.g., for the intervention perception
questions, we can look at whether difference in other vs. self predicts
their social behaviors like ERQ)
Visualisations
prefrosh_race = prefrosh %>%
filter(Race != "Unknown") %>%
filter(Race != "Other. Please specify:")
# Race
p<-ggplot(prefrosh_race, aes(x=Race)) +
geom_bar(color="black", fill="white")
p + theme(axis.text.x = element_text(size = 7, angle = 45, hjust = 1))
# Gender
prefrosh_gender = prefrosh %>%
filter(Gender != "Unknown")
p<-ggplot(prefrosh_gender, aes(x=Gender)) +
geom_bar(color="black", fill="white")
p + theme(axis.text.x = element_text(size = 7, angle = 45, hjust = 1))
prefrosh = read.csv("~/Downloads/Stanford Communities Project - Prefrosh 2022-2023_October 7, 2022_15.20.csv", header=T, na.strings=c("","NA"))
prefrosh <- prefrosh[-c(1, 2, 548), ]
prefrosh = prefrosh %>%
distinct(RecipientEmail, .keep_all = TRUE)
prefrosh_poli = prefrosh %>%
filter(PoliIdeology_general != "NA")
# Political Ideology
p = ggplot(prefrosh_poli, aes(x=PoliIdeology_general)) +
geom_bar(color="black", fill="white")
p + scale_x_discrete(name = "Political Ideology", labels=c("1" = "Extremely Liberal", "2" = "Liberal",
"3" = "Moderately Liberal", "4" = "Moderate", "5" = "Moderateley Conservative", "6" = "Conservative", "7" = "Extremely Conservative")) + theme(axis.text.x = element_text(size = 7, angle = 45, hjust = 1))
Family Income
# We have to take steps to re-order the labels
income_data = data.frame(
ID <- c(1:17),
labels = c("$0-$20,000", "$20,001-$40,000 ", "$40,001-$60,000", "$60,001-$80,000", "$80,001-$100,000", "$100,001-$120,000", "$120,001-$140,000", "$140,001-$160,000", "$160,001-$180,000", "$180,001-$200,000", "$200,001-$250,000", "$250,001-$300,000", "$300,001-$350,000", "$350,001-$400,000", "$400,001-$450,000", "$450,001-$500,000", "$500,001+"),
income = c(40, 76, 50, 45, 41, 41, 26, 25, 15, 21, 43, 37, 26, 23, 13, 15, 85))
#barplot(income_data$income, names.arg=labels,xlab="Income Bracket",
#ylab="Count")
Temporal Trends
prefrosh_21 = read.csv("~/Downloads/df.prefrosh2021 (5).csv")
- some temporal trends (e.g., how is this class rate in happiness,
stress, depression compared to past prefrosh samples)
Brief Preprocessing
library(ggsci)
df1 = read.csv("~/Downloads/df1.csv")
df1
## Survey Mean
## 1 TAI_5_21 12.866720
## 2 TAI_5_22 13.284450
## 3 EROS_21 3.722222
## 4 EROS_22 3.758606
## 5 loneliness_21 2.367347
## 6 loneliness_22 2.414708
## 7 CESD_21 16.354050
## 8 CESD_22 16.750940
p<-ggplot(data=df1, aes(x=Survey, y=Mean)) +
geom_bar(stat="identity")
p + theme(axis.text.x = element_text(size = 7, angle = 45, hjust = 1)) + scale_fill_jco()
t.test(prefrosh_21$CESD, prefrosh$CESD)
##
## Welch Two Sample t-test
##
## data: prefrosh_21$CESD and prefrosh$CESD
## t = -1.7511, df = 1688.2, p-value = 0.0801
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.84143832 0.04764637
## sample estimates:
## mean of x mean of y
## 16.35405 16.75094
t.test(prefrosh_21$TAI_5,prefrosh$TAI_5)
##
## Welch Two Sample t-test
##
## data: prefrosh_21$TAI_5 and prefrosh$TAI_5
## t = -2.7059, df = 1648.4, p-value = 0.006882
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.7205224 -0.1149353
## sample estimates:
## mean of x mean of y
## 12.86672 13.28445
- some hypothesis testing (e.g., for the intervention perception
questions, we can look at whether difference in other vs. self predicts
their social behaviors like ERQ)
library(jtools)
# miscalibration_1 is "If someone else is feeling bad, how much do you expect a typical Stanford Student to try and help them feel better by listening to them and acting kindly?"
# miscalibration_3 is "If someone else is feeling bad, how likely are you to try and help them feel better by listening to them and acting kindly?"
prefrosh = prefrosh %>%
mutate(diff_self_other = miscalibration_1 - miscalibration_3) %>%
mutate(abs_diff_self_other = abs(miscalibration_1 - miscalibration_3))
model = lm(EROS ~ 1 + diff_self_other + miscalibration_3, prefrosh)
summary(model)
##
## Call:
## lm(formula = EROS ~ 1 + diff_self_other + miscalibration_3, data = prefrosh)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.32135 -0.48310 0.01404 0.51404 1.76506
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.48186 0.12904 19.233 <2e-16 ***
## diff_self_other -0.03972 0.02456 -1.617 0.106
## miscalibration_3 0.25103 0.02666 9.415 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7006 on 758 degrees of freedom
## (230 observations deleted due to missingness)
## Multiple R-squared: 0.1373, Adjusted R-squared: 0.135
## F-statistic: 60.31 on 2 and 758 DF, p-value: < 2.2e-16
summ(model, vifs = TRUE)
|
Observations
|
761 (230 missing obs. deleted)
|
|
Dependent variable
|
EROS
|
|
Type
|
OLS linear regression
|
|
F(2,758)
|
60.31
|
|
R²
|
0.14
|
|
Adj. R²
|
0.13
|
|
|
Est.
|
S.E.
|
t val.
|
p
|
VIF
|
|
(Intercept)
|
2.48
|
0.13
|
19.23
|
0.00
|
NA
|
|
diff_self_other
|
-0.04
|
0.02
|
-1.62
|
0.11
|
1.17
|
|
miscalibration_3
|
0.25
|
0.03
|
9.42
|
0.00
|
1.17
|
|
Standard errors: OLS
|
model = lm(EROS ~ 1 + abs_diff_self_other + miscalibration_3, prefrosh)
summary(model)
##
## Call:
## lm(formula = EROS ~ 1 + abs_diff_self_other + miscalibration_3,
## data = prefrosh)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.29784 -0.44265 0.03549 0.50513 1.81828
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.39894 0.12489 19.209 <2e-16 ***
## abs_diff_self_other 0.06568 0.02755 2.384 0.0174 *
## miscalibration_3 0.26093 0.02474 10.549 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6992 on 758 degrees of freedom
## (230 observations deleted due to missingness)
## Multiple R-squared: 0.1407, Adjusted R-squared: 0.1385
## F-statistic: 62.08 on 2 and 758 DF, p-value: < 2.2e-16
summ(model, vifs = TRUE)
|
Observations
|
761 (230 missing obs. deleted)
|
|
Dependent variable
|
EROS
|
|
Type
|
OLS linear regression
|
|
F(2,758)
|
62.08
|
|
R²
|
0.14
|
|
Adj. R²
|
0.14
|
|
|
Est.
|
S.E.
|
t val.
|
p
|
VIF
|
|
(Intercept)
|
2.40
|
0.12
|
19.21
|
0.00
|
NA
|
|
abs_diff_self_other
|
0.07
|
0.03
|
2.38
|
0.02
|
1.01
|
|
miscalibration_3
|
0.26
|
0.02
|
10.55
|
0.00
|
1.01
|
|
Standard errors: OLS
|
# ERQ cognitive reappraisal (ERQ_C) and expressive suppression (ERQ_E)
prefrosh <- prefrosh %>%
rowwise() %>%
mutate(ERQ_C = sum(c(ERQ_1,
ERQ_3,
ERQ_5,
ERQ_7,
ERQ_8,
ERQ_10)),
ERQ_E = sum(c(ERQ_2,
ERQ_4,
ERQ_6,
ERQ_9)))
model = lm(ERQ_C ~ 1 + diff_self_other + miscalibration_3, prefrosh)
summary(model)
##
## Call:
## lm(formula = ERQ_C ~ 1 + diff_self_other + miscalibration_3,
## data = prefrosh)
##
## Residuals:
## Min 1Q Median 3Q Max
## -22.3745 -3.6627 0.5219 3.9165 14.1635
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 24.7323 1.0623 23.281 < 2e-16 ***
## diff_self_other 0.6573 0.2023 3.249 0.00121 **
## miscalibration_3 0.9404 0.2195 4.285 2.07e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.762 on 757 degrees of freedom
## (231 observations deleted due to missingness)
## Multiple R-squared: 0.02746, Adjusted R-squared: 0.02489
## F-statistic: 10.69 on 2 and 757 DF, p-value: 2.647e-05
summ(model, vifs = TRUE)
|
Observations
|
760 (231 missing obs. deleted)
|
|
Dependent variable
|
ERQ_C
|
|
Type
|
OLS linear regression
|
|
F(2,757)
|
10.69
|
|
R²
|
0.03
|
|
Adj. R²
|
0.02
|
|
|
Est.
|
S.E.
|
t val.
|
p
|
VIF
|
|
(Intercept)
|
24.73
|
1.06
|
23.28
|
0.00
|
NA
|
|
diff_self_other
|
0.66
|
0.20
|
3.25
|
0.00
|
1.17
|
|
miscalibration_3
|
0.94
|
0.22
|
4.28
|
0.00
|
1.17
|
|
Standard errors: OLS
|
model = lm(ERQ_C ~ 1 + abs_diff_self_other + miscalibration_3, prefrosh)
summary(model)
##
## Call:
## lm(formula = ERQ_C ~ 1 + abs_diff_self_other + miscalibration_3,
## data = prefrosh)
##
## Residuals:
## Min 1Q Median 3Q Max
## -22.2493 -3.6583 0.4769 4.0679 13.7941
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 25.8923 1.0325 25.078 < 2e-16 ***
## abs_diff_self_other -0.5910 0.2279 -2.593 0.009695 **
## miscalibration_3 0.7262 0.2045 3.552 0.000406 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.776 on 757 degrees of freedom
## (231 observations deleted due to missingness)
## Multiple R-squared: 0.02258, Adjusted R-squared: 0.02
## F-statistic: 8.744 on 2 and 757 DF, p-value: 0.000176
summ(model, vifs = TRUE)
|
Observations
|
760 (231 missing obs. deleted)
|
|
Dependent variable
|
ERQ_C
|
|
Type
|
OLS linear regression
|
|
F(2,757)
|
8.74
|
|
R²
|
0.02
|
|
Adj. R²
|
0.02
|
|
|
Est.
|
S.E.
|
t val.
|
p
|
VIF
|
|
(Intercept)
|
25.89
|
1.03
|
25.08
|
0.00
|
NA
|
|
abs_diff_self_other
|
-0.59
|
0.23
|
-2.59
|
0.01
|
1.01
|
|
miscalibration_3
|
0.73
|
0.20
|
3.55
|
0.00
|
1.01
|
|
Standard errors: OLS
|
model = lm(ERQ_E ~ 1 + diff_self_other + miscalibration_3, prefrosh)
summary(model)
##
## Call:
## lm(formula = ERQ_E ~ 1 + diff_self_other + miscalibration_3,
## data = prefrosh)
##
## Residuals:
## Min 1Q Median 3Q Max
## -14.189 -3.582 0.418 3.418 13.326
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.1239 0.9276 21.696 < 2e-16 ***
## diff_self_other 0.1179 0.1767 0.667 0.505
## miscalibration_3 -0.9084 0.1917 -4.739 2.57e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.034 on 758 degrees of freedom
## (230 observations deleted due to missingness)
## Multiple R-squared: 0.03773, Adjusted R-squared: 0.03519
## F-statistic: 14.86 on 2 and 758 DF, p-value: 4.667e-07
summ(model, vifs = TRUE)
|
Observations
|
761 (230 missing obs. deleted)
|
|
Dependent variable
|
ERQ_E
|
|
Type
|
OLS linear regression
|
|
F(2,758)
|
14.86
|
|
R²
|
0.04
|
|
Adj. R²
|
0.04
|
|
|
Est.
|
S.E.
|
t val.
|
p
|
VIF
|
|
(Intercept)
|
20.12
|
0.93
|
21.70
|
0.00
|
NA
|
|
diff_self_other
|
0.12
|
0.18
|
0.67
|
0.51
|
1.17
|
|
miscalibration_3
|
-0.91
|
0.19
|
-4.74
|
0.00
|
1.17
|
|
Standard errors: OLS
|
model = lm(ERQ_E ~ 1 + abs_diff_self_other + miscalibration_3, prefrosh)
summary(model)
##
## Call:
## lm(formula = ERQ_E ~ 1 + abs_diff_self_other + miscalibration_3,
## data = prefrosh)
##
## Residuals:
## Min 1Q Median 3Q Max
## -14.4778 -3.5174 0.4694 3.4826 13.7569
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.1639 0.8980 22.454 < 2e-16 ***
## abs_diff_self_other 0.2875 0.1983 1.450 0.148
## miscalibration_3 -0.9868 0.1779 -5.546 4.03e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 5.028 on 758 degrees of freedom
## (230 observations deleted due to missingness)
## Multiple R-squared: 0.03983, Adjusted R-squared: 0.0373
## F-statistic: 15.72 on 2 and 758 DF, p-value: 2.041e-07
summ(model, vifs = TRUE)
|
Observations
|
761 (230 missing obs. deleted)
|
|
Dependent variable
|
ERQ_E
|
|
Type
|
OLS linear regression
|
|
F(2,758)
|
15.72
|
|
R²
|
0.04
|
|
Adj. R²
|
0.04
|
|
|
Est.
|
S.E.
|
t val.
|
p
|
VIF
|
|
(Intercept)
|
20.16
|
0.90
|
22.45
|
0.00
|
NA
|
|
abs_diff_self_other
|
0.29
|
0.20
|
1.45
|
0.15
|
1.01
|
|
miscalibration_3
|
-0.99
|
0.18
|
-5.55
|
0.00
|
1.01
|
|
Standard errors: OLS
|