You have 7 small tasks. You can work in groups to complete them in-class. The answer provided here may not be the only right answer. You can have your own ways of coding.
Import anchor1_50percent_Eng data
#Question: Import anchor1_50percent_Eng data
library(tidyverse) #use the tidyvese package
library(haven) #use the haven package
wave1 <- read_dta("anchor1_50percent_Eng.dta") #import the dataKeep variables of id, age, sex_gen, cohort, homosex_new, yedu, relstat, as well as one variable that reflects the attitude towards family (e.g. val1i7), and one variable that reflects subjective wellbeing (e.g. sat6). The function that allows you to do this job is called “______”. Make these as a new dataset.
We show the data in this tab.
#Question: Keep variables of id, age, sex_gen, cohort, homosex_new, yedu, relstat, as well as one variable that reflects the attitude towards family, and one variable that reflects subjective wellbeing. The function that allow you to do this job is called "______". Make these as a new dataset.
(wave1a <- select(wave1, id, age, sex_gen, cohort, homosex_new, yeduc,
relstat, val1i7,sat6)) #select the variables to create a new dataset named wave1a.## # A tibble: 6,201 × 9
## id age sex_gen cohort homose…¹ yeduc relstat val1i7 sat6
## <dbl> <dbl+lbl> <dbl+lb> <dbl+l> <dbl+lb> <dbl+lb> <dbl+lb> <dbl+l> <dbl>
## 1 267206000 16 2 [2 Fe… 1 [1 1… -1 [-1 … 0 [0 c… 1 [1 N… 3 7
## 2 112963000 35 1 [1 Ma… 3 [3 1… -1 [-1 … 10.5 1 [1 N… 2 6
## 3 327937000 16 2 [2 Fe… 1 [1 1… 0 [0 H… 0 [0 c… -7 [-7 … 4 8
## 4 318656000 27 2 [2 Fe… 2 [2 1… 0 [0 H… 11.5 4 [4 M… 5 [5 A… 9
## 5 717889000 37 1 [1 Ma… 3 [3 1… 0 [0 H… 11.5 4 [4 M… 4 7
## 6 222517000 15 1 [1 Ma… 1 [1 1… -1 [-1 … 0 [0 c… 1 [1 N… 5 [5 A… 9
## 7 144712000 16 2 [2 Fe… 1 [1 1… -1 [-1 … 0 [0 c… 1 [1 N… 4 8
## 8 659357000 17 2 [2 Fe… 1 [1 1… 0 [0 H… 0 [0 c… 2 [2 N… 5 [5 A… 7
## 9 506367000 37 1 [1 Ma… 3 [3 1… 0 [0 H… 10.5 4 [4 M… 1 [1 D… 9
## 10 64044000 15 2 [2 Fe… 1 [1 1… -1 [-1 … 0 [0 c… 1 [1 N… 1 [1 D… 7
## # … with 6,191 more rows, and abbreviated variable name ¹homosex_newChange the variables into numeric and factors appropriately: 1)Remove labels for continuous variable; 2)make categorical variable as factors. Think about the functions for numeric and for categorical variables. Write and run the code for this task
#Question: Change the variables into numeric and factors appropriately: 1)Remove labels for continuous variable; 2)make categorical variable as factors.For that, you need the two functions, "______" for numeric and "______" for categorical variables..
wave1a <- mutate(wave1a,
id=zap_labels(id), #remove label for id
age=zap_labels(age), #remove label of age
yeduc=zap_labels(yeduc), #remove label of education years
sat6=zap_labels(sat6), #remove label for life satisfaction, it is good to view it as continuous.
sex_gen=as_factor(sex_gen), #make sex variable as factor
cohort=as_factor(cohort), #make cohort variable as factor
homosex_new=as_factor(homosex_new), #make homosex_new variable as factor
relstat=as_factor(relstat), #make relstat variable as factor
val1i7=as_factor(val1i7) #make val1i7 as factor
)
wave1a## # A tibble: 6,201 × 9
## id age sex_gen cohort homosex_new yeduc relstat val1i7 sat6
## <dbl> <dbl> <fct> <fct> <fct> <dbl> <fct> <fct> <dbl>
## 1 267206000 16 2 Female 1 1991-1993 -1 No partner 0 1 Neve… 3 7
## 2 112963000 35 1 Male 3 1971-1973 -1 No partner 10.5 1 Neve… 2 6
## 3 327937000 16 2 Female 1 1991-1993 0 Hetero 0 -7 Inc… 4 8
## 4 318656000 27 2 Female 2 1981-1983 0 Hetero 11.5 4 Marr… 5 Agr… 9
## 5 717889000 37 1 Male 3 1971-1973 0 Hetero 11.5 4 Marr… 4 7
## 6 222517000 15 1 Male 1 1991-1993 -1 No partner 0 1 Neve… 5 Agr… 9
## 7 144712000 16 2 Female 1 1991-1993 -1 No partner 0 1 Neve… 4 8
## 8 659357000 17 2 Female 1 1991-1993 0 Hetero 0 2 Neve… 5 Agr… 7
## 9 506367000 37 1 Male 3 1971-1973 0 Hetero 10.5 4 Marr… 1 Dis… 9
## 10 64044000 15 2 Female 1 1991-1993 -1 No partner 0 1 Neve… 1 Dis… 7
## # … with 6,191 more rows#Question: Z-standardize yeduc and age.
# First. check whether values on those variables make sense.
summary(wave1a$yeduc) #summary the yeduc variable to see if there are missing cases## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -7.000 0.000 11.000 8.933 13.000 20.000summary(wave1a$age) #summary the age variable to see if there are missing cases## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 14.00 17.00 26.00 25.84 35.00 38.00wave1a <- mutate(wave1a,
yeduc=case_when(
yeduc<0 ~ as.numeric(NA), #when yeduc <0, i make it NA
TRUE ~ as.numeric(yeduc) #the rest of yedu as it is and make it numeric
),
z_yeduc=(yeduc- mean(yeduc,na.rm=TRUE))/sd(yeduc,na.rm = TRUE), #z-standardization of yeduc
z_age=(age- mean(age))/sd(age) #z-standardization of age
)
summary(wave1a$z_yeduc) #summary the standardized yeduc## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## -1.4513 -1.4513 0.3226 0.0000 0.6451 1.7739 26summary(wave1a$z_age) #summary the standardized age## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -1.41586 -1.05703 0.01946 0.00000 1.09595 1.45478Show the frequency of relstat. Recode relstat to “no partner” when the person is never married regardless of the cohabiting status; to “with partner” when the person is married regardless of the cohabiting status. Show the frequence of the relstat again. What is the frequency for “no partner” and the frequency for “with partner”?
#Question: Show the frequency of relstat. Recode relstat to "no partner" when the person is never married regardless of the cohabiting status; to "with partner" when the person is married regardless of the cohabiting status. Show the frequence of the relstat again. What is the frequency for "no partner" and the frequency for "with partner"?
table(wave1a$relstat)##
## -7 Incomplete data 1 Never married single
## 34 2448
## 2 Never married LAT 3 Never married COHAB
## 1012 660
## 4 Married COHAB 5 Married noncohabiting
## 1735 23
## 6 Divorced/separated single 7 Divorced/separated LAT
## 146 63
## 8 Divorced/separated COHAB 9 Widowed single
## 76 3
## 10 Widowed LAT 11 Widowed COHAB
## 1 0wave1a <- mutate(wave1a,
relstat = case_when(
relstat %in% c("1 Never married single" ,"2 Never married LAT", "3 Never married COHAB") ~ "no partner", #when the value of relstat belong to the categories mentioned, replace them with 'no partner'
relstat %in% c("4 Married COHAB" ,"5 Married noncohabiting") ~ "with partner",
#when the value of relstat belong to the categories mentioned, replace them with 'with partner'
TRUE ~ as.character(relstat) #the rest remain as it is and make it a character
),
relstat = factor(relstat) #change the relstat back to factor
)
table(wave1a$relstat) #get the frequence table of relstat##
## -7 Incomplete data 10 Widowed LAT
## 34 1
## 6 Divorced/separated single 7 Divorced/separated LAT
## 146 63
## 8 Divorced/separated COHAB 9 Widowed single
## 76 3
## no partner with partner
## 4120 1758#Remember, %in% is a operator means "belonging to"Now check out prop.table() and find out how to use. Check which cohort has the highest proportion of people reporting homosexual orientation
#Question: Now check out prop.table() and find out how to use. Check which cohort has the highest proportion of people reporting homosexual orientation
prop.table(table(wave1a$homosex_new, wave1a$cohort), margin = 2) # when margin=2 means calculating percentage by column##
## -7 Incomplete data 0 former capikid first interview
## -7 Incomplete data
## -1 No partner
## 0 Hetero
## 1 Gay
## 2 Lesbian
##
## 1 1991-1993 2 1981-1983 3 1971-1973 4 2001-2003
## -7 Incomplete data 0.000000000 0.000000000 0.000000000
## -1 No partner 0.747814082 0.311972181 0.175682382
## 0 Hetero 0.247583985 0.674118231 0.815384615
## 1 Gay 0.001380580 0.006458023 0.004466501
## 2 Lesbian 0.003221353 0.007451565 0.004466501
##
## 9 former capikid re-interview
## -7 Incomplete data
## -1 No partner
## 0 Hetero
## 1 Gay
## 2 LesbianShow a proportion table for variable “val1i7” for the whole sample, and then show a proportion table for “val1i7” only among people who are aged >30
#Question: Show a frequency table for variable "val1i7" for the whole sample, and then show a frequency table for people who are aged >30
prop.table(table(wave1a$val1i7)) #proportion table for variable "val1i7" based on the all sample##
## -5 Inconsistent value -4 Filter error / Incorrect entry
## 0.000000000 0.000000000
## -3 Does not apply -2 No answer
## 0.000000000 0.001612643
## -1 Don't know 1 Disagree completely
## 0.008547009 0.147234317
## 2 3
## 0.129011450 0.214159007
## 4 5 Agree completely
## 0.185776488 0.313659087prop.table(table(wave1a$val1i7[wave1a$age > 30])) #proportion table for variable "val1i7" based on those aged>30##
## -5 Inconsistent value -4 Filter error / Incorrect entry
## 0.000000000 0.000000000
## -3 Does not apply -2 No answer
## 0.000000000 0.002481390
## -1 Don't know 1 Disagree completely
## 0.006947891 0.211910670
## 2 3
## 0.133995037 0.216377171
## 4 5 Agree completely
## 0.151364764 0.276923077#or you can
wave1b <- filter(wave1a,age > 30 ) #generate a new data containing only those aged >30
prop.table(table(wave1b$val1i7)) #then generate a proportion table ##
## -5 Inconsistent value -4 Filter error / Incorrect entry
## 0.000000000 0.000000000
## -3 Does not apply -2 No answer
## 0.000000000 0.002481390
## -1 Don't know 1 Disagree completely
## 0.006947891 0.211910670
## 2 3
## 0.133995037 0.216377171
## 4 5 Agree completely
## 0.151364764 0.276923077