Inhwan Ko, PhD Student in Political Science
University of Washington, Seattle
inhwanko at uw dot edu
Introduction of this markdown
This code is to display results of preliminary analyses of the relationship between the responses to climate change / environment- related survey questions and basic covariates of the respondants, as collected in the Korean General Social Survey (KGSS) in 2010, 2013, 2014. If data are available over more than 2 years, I used fixed effect model for panel analysis. If data are available only for 1 year (2010), I used a simple linear regression model. One can correct the panel data analyses with HC2 standard error using “clubSandwich” package. A research project based on this analysis is ongoing.
For detailed information on labels and their explanation, see the codebook. All data and a codebook are downloaded from: http://kgss.skku.edu/?page_id=39
Relevant questions for climate change are:
1233(CAPA2): Please indicate the extent to which you think Korean society is coping well with each of the possible risks listed below: Radical Climate Change (2013, 2014)
1262(VULNERAB2): Please indicate the extent to which you think Korean society is vulnerable to each of the possible risks listed below: Radical climate change (2013, 2014)
1291(HAZA2): Please indicate the extent to which you think each of the possible risks listed below is likely to occur to yourself: Radical Climate Change (2013, 2014)
1320(EXPOS2): Lastly, please indicate the extent to which you think Korean society is exposed to each of the possible risks listed below: Radical climate change (2014, 2014)
2252~2311 questions are also relevant, for example:
2261(ENTRBUS) Here is a list of some different environmental problems. Which problem, if any, do you think is the most important for Korea as a whole? (2010, 2014):
2262(ENPRBFAM) Here is a list~. Which problem, if any, affects you and your family the most?
1-air pollution 2- chemicals and pesticates 3- water shortage 4- water pollution 5- nuclear waste 6- domestic waste disposal 7- climate change 8- GMO 9- using up our resource 77-OTHERS –8 IDK
2279 there are more important things to do in life than protect the environment
2310(ATOMSHOL): Assessing the truth level of “Climate change is caused by a hole in the earth’s atmosphere” (2010)
2011(COALCAUZ): Assessing the truth level of “Everytime we use coal or oil or gas, we contribute to climate change” (2010)
Relevant covariates are: AGE, SEX, EMPLOYMENT, EDUCATION, INCOME, IDEOLOGY, PARTY SUPPORT. Other covariates can also be included depending on the scope of the research.
Covariates:
AGE: numeric
SEX: 1-male, 2-female
EMPLY: 1-employed, 2-not employed
EDUC: 0-no formal school, 1-elem, 2-juniorhigh, 3-high, 4-2yearcollege, 5-4yearcollege, 6-masters, 7-phd, 8-other
RINCOME(monthly): 0-no income, 1-less than 500k WON, 2-500k~990k, 3-1000k~1490k, 4-1500k~1990k, 5-2000k~2490k, 6-2500k~2990k, 7-3000k~3490k, 8-3500k~3990k, 9-4000k~4490k, 10-4500k~4990k, 11-5000k~
*1k WON = 1USD (approximately)
PARTYLR: 1-liberal, 2-somewhat lib, 3-neither, 4-somewhat conserv, 5-conserv
#setwd("~")
# read raw data of 2003-2018 KGSS
library(haven)
rawdata <- read_sav("2003-2018_KGSS_eng_08092019.sav") # put your own data file name
# extract year and interested variables for preliminary analysis
library(tidyverse)## -- Attaching packages ------------------------------- tidyverse 1.3.0 --## √ ggplot2 3.2.1 √ purrr 0.3.3
## √ tibble 2.1.3 √ dplyr 0.8.3
## √ tidyr 1.0.0 √ stringr 1.4.0
## √ readr 1.3.1 √ forcats 0.4.0## -- Conflicts ---------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()varlist <- c("YEAR", "AGE", "SEX", "EMPLY", "EDUC", "RINCOME", "PARTYLR", "PRTYID10", "PRTYID13", "PRTYID14", "CAPA2", "VULNERAB2", "HAZA2", "EXPOS2", "GRNCON", "ENTRBUS", "ENPRBFAM", "ATMOSHOL", "COALCAUZ", "HARMGOOD", "GRWTHELP", "GRWTHARM", "TOODIFME", "IMPGRN", "OTHSSAME", "GRNEXAGG", "HELPHARM", "TEMPGEN", "GRNINTL", "ECONGRN")
predata <- rawdata %>%
filter(YEAR==2010 | YEAR==2013 | YEAR==2014) %>%
select(YEAR, RESPID, AGE, SEX, EMPLY, EDUC, RINCOME, PARTYLR, PRTYID10, PRTYID13, PRTYID14, CAPA2, VULNERAB2, HAZA2, EXPOS2, GRNCON, ENTRBUS, ENPRBFAM, ATMOSHOL, COALCAUZ, HARMGOOD, GRWTHELP, GRWTHARM, TOODIFME, IMPGRN, OTHSSAME, GRNEXAGG, HELPHARM, TEMPGEN, GRNINTL, ECONGRN)
## data cleansing for KGSS
predata[predata==-8 | predata==-1 | predata==77 | predata==666] <- NA##delete educ-8
predata <- predata %>%
mutate(EDUC=replace(EDUC, EDUC==8, NA))Panel & ANOVA analysis (Careful with the interpretation: some variables have different orders in their responses.)
library(plm)##
## Attaching package: 'plm'## The following objects are masked from 'package:dplyr':
##
## between, lag, leadlibrary(broom)
library(stargazer)##
## Please cite as:## Hlavac, Marek (2018). stargazer: Well-Formatted Regression and Summary Statistics Tables.## R package version 5.2.2. https://CRAN.R-project.org/package=stargazer## how korea is coping well with climate change (1badly ~ 7well)
capa2plm <- plm(CAPA2 ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(capa2plm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = CAPA2 ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
## data = predata, model = "within", index = c("YEAR", "RESPID"))
##
## Unbalanced Panel: n = 2, T = 699-779, N = 1478
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -2.686704 -0.869978 -0.083804 0.815535 3.767378
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE 0.0083420 0.0027700 3.0115 0.002644 **
## SEX -0.3439911 0.0664902 -5.1736 2.614e-07 ***
## EMPLY 0.8112809 0.4538609 1.7875 0.074061 .
## EDUC -0.0635604 0.0269561 -2.3579 0.018508 *
## RINCOME -0.0234799 0.0086626 -2.7105 0.006797 **
## PARTYLR 0.0213890 0.0309464 0.6912 0.489573
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 2152.8
## Residual Sum of Squares: 2064.7
## R-Squared: 0.040921
## Adj. R-Squared: 0.036354
## F-statistic: 10.4533 on 6 and 1470 DF, p-value: 2.228e-11## how korea is vulnerable to climate change (1high ~ 7low)
vulnerab2plm <- plm(VULNERAB2 ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(vulnerab2plm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = VULNERAB2 ~ AGE + SEX + EMPLY + EDUC + RINCOME +
## PARTYLR, data = predata, model = "within", index = c("YEAR",
## "RESPID"))
##
## Unbalanced Panel: n = 2, T = 698-779, N = 1477
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -2.491106 -0.946563 -0.049117 0.843441 4.248102
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE 0.00009502 0.00268168 0.0354 0.971739
## SEX -0.20168684 0.06437787 -3.1329 0.001765 **
## EMPLY 0.03230199 0.43936223 0.0735 0.941402
## EDUC -0.06632253 0.02611055 -2.5401 0.011185 *
## RINCOME -0.01278995 0.00838632 -1.5251 0.127450
## PARTYLR 0.05313017 0.02996904 1.7728 0.076463 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 1965.9
## Residual Sum of Squares: 1933.6
## R-Squared: 0.016426
## Adj. R-Squared: 0.011739
## F-statistic: 4.08867 on 6 and 1469 DF, p-value: 0.00044784## how likely climate change occurs to yourself (1likely ~ 7unlikely)
haza2plm <- plm(HAZA2 ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(haza2plm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = HAZA2 ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
## data = predata, model = "within", index = c("YEAR", "RESPID"))
##
## Unbalanced Panel: n = 2, T = 698-779, N = 1477
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -3.119088 -0.868212 0.079576 1.082494 3.359173
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE 0.00096951 0.00300100 0.3231 0.74669
## SEX -0.17045867 0.07204361 -2.3660 0.01811 *
## EMPLY 0.61182765 0.49167892 1.2444 0.21356
## EDUC 0.02644620 0.02921964 0.9051 0.36557
## RINCOME -0.01526143 0.00938491 -1.6262 0.10413
## PARTYLR 0.05126108 0.03353758 1.5285 0.12661
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 2439.3
## Residual Sum of Squares: 2421.5
## R-Squared: 0.0072853
## Adj. R-Squared: 0.0025549
## F-statistic: 1.79678 on 6 and 1469 DF, p-value: 0.096219## how much korean society is exposed to climate change (1exposed ~ 7not)
expos2plm <- plm(EXPOS2 ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(expos2plm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = EXPOS2 ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
## data = predata, model = "within", index = c("YEAR", "RESPID"))
##
## Unbalanced Panel: n = 2, T = 699-779, N = 1478
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -2.41830 -0.97669 -0.10743 0.81112 4.11797
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE -0.00022943 0.00275857 -0.0832 0.93373
## SEX -0.14100520 0.06621484 -2.1295 0.03338 *
## EMPLY 0.28027875 0.45198153 0.6201 0.53528
## EDUC -0.05309799 0.02684447 -1.9780 0.04812 *
## RINCOME -0.01084602 0.00862674 -1.2573 0.20886
## PARTYLR 0.04550018 0.03081830 1.4764 0.14005
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 2067.6
## Residual Sum of Squares: 2047.7
## R-Squared: 0.0096475
## Adj. R-Squared: 0.0049315
## F-statistic: 2.38666 on 6 and 1470 DF, p-value: 0.026761## how much concerned with environmental issues (1no ~ 5much)
grnconplm <- plm(GRNCON ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(grnconplm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = GRNCON ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
## data = predata, model = "within", index = c("YEAR", "RESPID"))
##
## Unbalanced Panel: n = 2, T = 778-859, N = 1637
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -3.12631 -0.67651 0.18038 0.58434 1.70949
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE 0.0098821 0.0021691 4.5558 5.608e-06 ***
## SEX 0.2016929 0.0522020 3.8637 0.0001161 ***
## EDUC 0.0567793 0.0209077 2.7157 0.0066831 **
## RINCOME 0.0076385 0.0066949 1.1409 0.2540628
## PARTYLR -0.0781118 0.0244887 -3.1897 0.0014512 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 1602.8
## Residual Sum of Squares: 1562.3
## R-Squared: 0.025254
## Adj. R-Squared: 0.021666
## F-statistic: 8.44595 on 5 and 1630 DF, p-value: 6.6791e-08## what type of environmental issues most important?
# 1-air pollution 2- chemicals and pesticates 3- water shortage 4- water pollution
# 5- nuclear waste 6- domestic waste disposal 7- climate change 8- GMO
# 9- using up our resources
library(stats)
entrbusaov <- aov(ENTRBUS ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata)
summary(entrbusaov)## Df Sum Sq Mean Sq F value Pr(>F)
## AGE 1 63 62.78 10.431 0.00126 **
## SEX 1 14 13.98 2.323 0.12769
## EDUC 1 6 6.12 1.017 0.31339
## RINCOME 1 4 4.43 0.736 0.39121
## PARTYLR 1 15 15.46 2.569 0.10915
## Residuals 1621 9756 6.02
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2613 observations deleted due to missingness## what issue affects family?
enprbfamaov <- aov(ENPRBFAM ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata)
summary(enprbfamaov)## Df Sum Sq Mean Sq F value Pr(>F)
## AGE 1 20 20.26 2.979 0.0845 .
## SEX 1 38 37.72 5.548 0.0186 *
## EDUC 1 0 0.37 0.055 0.8149
## RINCOME 1 2 1.96 0.288 0.5913
## PARTYLR 1 0 0.01 0.001 0.9754
## Residuals 1609 10940 6.80
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 2625 observations deleted due to missingness## HARMGOOD: modern science does more harm than good (1agree ~ 5disagree)
harmgoodplm <- plm(HARMGOOD ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(harmgoodplm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = HARMGOOD ~ AGE + SEX + EMPLY + EDUC + RINCOME +
## PARTYLR, data = predata, model = "within", index = c("YEAR",
## "RESPID"))
##
## Unbalanced Panel: n = 2, T = 696-856, N = 1552
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -2.69473 -0.57123 0.23239 0.65777 2.22320
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE 0.0013610 0.0023296 0.5842 0.559162
## SEX -0.1650779 0.0544174 -3.0335 0.002457 **
## EMPLY -0.3245814 0.3786126 -0.8573 0.391417
## EDUC 0.1100901 0.0217746 5.0559 4.795e-07 ***
## RINCOME 0.0141669 0.0073417 1.9296 0.053833 .
## PARTYLR 0.0108715 0.0255907 0.4248 0.671026
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 1572.8
## Residual Sum of Squares: 1508.7
## R-Squared: 0.040766
## Adj. R-Squared: 0.036417
## F-statistic: 10.9361 on 6 and 1544 DF, p-value: 5.8726e-12## GRWTHELP: in order to protect the environmental Korea needs more economic growth (1agree~ 5disagree)
grwthelpplm <- plm(GRWTHELP ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(grwthelpplm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = GRWTHELP ~ AGE + SEX + EMPLY + EDUC + RINCOME +
## PARTYLR, data = predata, model = "within", index = c("YEAR",
## "RESPID"))
##
## Unbalanced Panel: n = 2, T = 778-858, N = 1636
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -1.77907 -0.69055 -0.19125 0.65886 3.23047
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE -0.0027487 0.0022651 -1.2135 0.22512
## SEX -0.0566826 0.0545127 -1.0398 0.29858
## EDUC 0.0984977 0.0218293 4.5122 6.876e-06 ***
## RINCOME 0.0135396 0.0069891 1.9373 0.05289 .
## PARTYLR -0.0581983 0.0255706 -2.2760 0.02298 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 1769.1
## Residual Sum of Squares: 1702.1
## R-Squared: 0.037869
## Adj. R-Squared: 0.034325
## F-statistic: 12.8232 on 5 and 1629 DF, p-value: 2.9896e-12## GRWTHARM: economic growth always harms the environment (1agree ~ 5disagree)
grwtharmplm <- plm(GRWTHARM ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(grwtharmplm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = GRWTHARM ~ AGE + SEX + EMPLY + EDUC + RINCOME +
## PARTYLR, data = predata, model = "within", index = c("YEAR",
## "RESPID"))
##
## Unbalanced Panel: n = 2, T = 779-859, N = 1638
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -2.233367 -0.834886 0.054387 0.954431 2.655900
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE 0.0030471 0.0023484 1.2975 0.194640
## SEX 0.1327250 0.0565031 2.3490 0.018943 *
## EDUC 0.1093225 0.0226297 4.8309 1.486e-06 ***
## RINCOME 0.0059520 0.0072477 0.8212 0.411635
## PARTYLR 0.0853271 0.0264985 3.2201 0.001307 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 1880.8
## Residual Sum of Squares: 1832.4
## R-Squared: 0.025722
## Adj. R-Squared: 0.022138
## F-statistic: 8.6119 on 5 and 1631 DF, p-value: 4.5798e-08## TEMPGEN: How dangerious a rise in the world temperature caused by climate change (1 danger ~ 5 not)
tempgenplm <- plm(TEMPGEN ~ AGE + SEX + EMPLY + EDUC + RINCOME + PARTYLR,
data=predata,
index=c("YEAR", "RESPID"),
model = "within")
summary(tempgenplm)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = TEMPGEN ~ AGE + SEX + EMPLY + EDUC + RINCOME +
## PARTYLR, data = predata, model = "within", index = c("YEAR",
## "RESPID"))
##
## Unbalanced Panel: n = 2, T = 779-855, N = 1634
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -1.59514 -0.79923 -0.08420 0.74330 3.13436
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## AGE 0.0074423 0.0018777 3.9634 7.708e-05 ***
## SEX -0.0733405 0.0451897 -1.6229 0.1048
## EDUC -0.0241392 0.0181239 -1.3319 0.1831
## RINCOME 0.0030537 0.0058022 0.5263 0.5988
## PARTYLR 0.0118429 0.0212565 0.5571 0.5775
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 1191.9
## Residual Sum of Squares: 1167.3
## R-Squared: 0.020638
## Adj. R-Squared: 0.017026
## F-statistic: 6.85705 on 5 and 1627 DF, p-value: 2.4248e-06Some variables only have 2010 data and missing variables: therefore, these DVs below are only conducted with linear model regression: (all 1 agree ~ 5 disagree index)
Linear model regression
## ATMOSHOL: climate change is cuased by a hole in the earth's atmosphere
atmoshollm <- lm(ATMOSHOL ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(atmoshollm)##
## Call:
## lm(formula = ATMOSHOL ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
## data = predata)
##
## Residuals:
## <Labelled double>: Assessing the truth level of 'Climate change is caused by a hole in the earth’s atmosphere'
## Min 1Q Median 3Q Max
## -1.45444 -0.29716 -0.13128 0.64366 2.10497
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Definitely true
## 2 Probably true
## 3 Probably not true
## 4 Definitely not true
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.000970 0.205187 9.752 < 2e-16 ***
## AGE 0.006701 0.002434 2.753 0.00604 **
## SEX 0.070540 0.057271 1.232 0.21842
## EDUC -0.040635 0.022813 -1.781 0.07525 .
## RINCOME 0.013210 0.007542 1.751 0.08024 .
## PARTYLR -0.044518 0.027105 -1.642 0.10089
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7586 on 817 degrees of freedom
## (3417 observations deleted due to missingness)
## Multiple R-squared: 0.02629, Adjusted R-squared: 0.02033
## F-statistic: 4.412 on 5 and 817 DF, p-value: 0.0005697## COALCAUZ: everytime we use coal or oil or gas, we contribute to climate change
coalcauzlm <- lm(COALCAUZ ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(coalcauzlm)##
## Call:
## lm(formula = COALCAUZ ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
## data = predata)
##
## Residuals:
## <Labelled double>: Assessing the truth level of 'Every time we use coal or oil or gas, we contribute to climate change'
## Min 1Q Median 3Q Max
## -1.11668 -0.08210 0.05699 0.14131 2.20691
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Definitely true
## 2 Probably true
## 3 Probably not true
## 4 Definitely not true
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.1319154 0.1707071 12.489 <2e-16 ***
## AGE 0.0003739 0.0020296 0.184 0.8539
## SEX 0.0292956 0.0474769 0.617 0.5374
## EDUC -0.0473758 0.0189075 -2.506 0.0124 *
## RINCOME -0.0033322 0.0062779 -0.531 0.5957
## PARTYLR -0.0220699 0.0224337 -0.984 0.3255
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.636 on 839 degrees of freedom
## (3395 observations deleted due to missingness)
## Multiple R-squared: 0.01517, Adjusted R-squared: 0.009298
## F-statistic: 2.584 on 5 and 839 DF, p-value: 0.02487## TOODIFME: it is just too difficult for someone like me to do much about the environment
toodifmelm <- lm(TOODIFME ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(toodifmelm) #only 2010 data available #employment not available##
## Call:
## lm(formula = TOODIFME ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
## data = predata)
##
## Residuals:
## <Labelled double>: Agree or disagree: It is just too difficult for someone like me to do much about the environment
## Min 1Q Median 3Q Max
## -2.38731 -0.90559 0.00950 0.89394 2.76807
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Agree strongly
## 2 Agree
## 3 Neither agree nor disagree
## 4 Disagree
## 5 Disagree strongly
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.219564 0.286948 7.735 2.91e-14 ***
## AGE -0.003436 0.003403 -1.010 0.3130
## SEX 0.180209 0.080079 2.250 0.0247 *
## EDUC 0.145243 0.031749 4.575 5.47e-06 ***
## RINCOME 0.016673 0.010573 1.577 0.1152
## PARTYLR 0.010482 0.037795 0.277 0.7816
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.082 on 854 degrees of freedom
## (3380 observations deleted due to missingness)
## Multiple R-squared: 0.05409, Adjusted R-squared: 0.04855
## F-statistic: 9.767 on 5 and 854 DF, p-value: 4.345e-09## IMPGRN: there are more important things to do in life than protect the environment
impgrnlm <- lm(IMPGRN ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(impgrnlm)##
## Call:
## lm(formula = IMPGRN ~ AGE + SEX + EDUC + RINCOME + PARTYLR, data = predata)
##
## Residuals:
## <Labelled double>: Agree or disagree: There are more important things to do in life than protect the environment
## Min 1Q Median 3Q Max
## -1.83823 -0.63445 -0.33987 0.62342 2.83643
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Agree strongly
## 2 Agree
## 3 Neither agree nor disagree
## 4 Disagree
## 5 Disagree strongly
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.170444 0.275904 7.867 1.1e-14 ***
## AGE -0.003779 0.003280 -1.152 0.24955
## SEX 0.215781 0.076877 2.807 0.00512 **
## EDUC 0.064454 0.030499 2.113 0.03487 *
## RINCOME -0.008149 0.010133 -0.804 0.42151
## PARTYLR -0.001155 0.036408 -0.032 0.97469
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.036 on 850 degrees of freedom
## (3384 observations deleted due to missingness)
## Multiple R-squared: 0.02369, Adjusted R-squared: 0.01795
## F-statistic: 4.126 on 5 and 850 DF, p-value: 0.001043## OTHSSAME: There is no point in doing what I can for the environmentl unless others do the same
othssamelm <- lm(OTHSSAME ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(othssamelm)##
## Call:
## lm(formula = OTHSSAME ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
## data = predata)
##
## Residuals:
## <Labelled double>: Agree or disagree: There is no point in doing what I can for the environment unless others do the same
## Min 1Q Median 3Q Max
## -2.82592 -1.04306 0.17287 0.95354 2.40924
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Agree strongly
## 2 Agree
## 3 Neither agree nor disagree
## 4 Disagree
## 5 Disagree strongly
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.104632 0.323432 6.507 1.30e-10 ***
## AGE 0.008459 0.003836 2.205 0.0277 *
## SEX 0.034530 0.090260 0.383 0.7021
## EDUC 0.187649 0.035786 5.244 1.99e-07 ***
## RINCOME 0.013304 0.011917 1.116 0.2646
## PARTYLR -0.062287 0.042601 -1.462 0.1441
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.219 on 854 degrees of freedom
## (3380 observations deleted due to missingness)
## Multiple R-squared: 0.04512, Adjusted R-squared: 0.03953
## F-statistic: 8.071 on 5 and 854 DF, p-value: 1.885e-07## GRNEXAGG: Many of the claims about environmental threats are exaggerated
grnexagglm <- lm(GRNEXAGG ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(grnexagglm)##
## Call:
## lm(formula = GRNEXAGG ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
## data = predata)
##
## Residuals:
## <Labelled double>: Agree or disagree: Many of the claims about environmental threats are exaggerated
## Min 1Q Median 3Q Max
## -2.48385 -1.08181 -0.11559 0.79517 2.14385
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Agree strongly
## 2 Agree
## 3 Neither agree nor disagree
## 4 Disagree
## 5 Disagree strongly
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.871954 0.287770 9.980 <2e-16 ***
## AGE -0.003992 0.003419 -1.168 0.2433
## SEX 0.119955 0.080258 1.495 0.1354
## EDUC 0.059138 0.031785 1.861 0.0632 .
## RINCOME 0.008348 0.010573 0.790 0.4300
## PARTYLR 0.019579 0.037967 0.516 0.6062
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.079 on 847 degrees of freedom
## (3387 observations deleted due to missingness)
## Multiple R-squared: 0.01544, Adjusted R-squared: 0.009624
## F-statistic: 2.656 on 5 and 847 DF, p-value: 0.02158## HELPHARM: I find it hard to know whether the way I live is helpful or harmful to the environment
helpharmlm <- lm(HELPHARM ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(helpharmlm)##
## Call:
## lm(formula = HELPHARM ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
## data = predata)
##
## Residuals:
## <Labelled double>: Agree or disagree: I find it hard to know whether the way I live is helpful or harmful to the environment
## Min 1Q Median 3Q Max
## -2.44995 -0.94380 -0.00842 0.83788 2.27422
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Agree strongly
## 2 Agree
## 3 Neither agree nor disagree
## 4 Disagree
## 5 Disagree strongly
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.306635 0.283564 8.134 1.46e-15 ***
## AGE 0.005572 0.003366 1.655 0.0982 .
## SEX 0.009446 0.079248 0.119 0.9051
## EDUC 0.156429 0.031393 4.983 7.59e-07 ***
## RINCOME 0.010756 0.010452 1.029 0.3038
## PARTYLR -0.023670 0.037348 -0.634 0.5264
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.068 on 850 degrees of freedom
## (3384 observations deleted due to missingness)
## Multiple R-squared: 0.04002, Adjusted R-squared: 0.03437
## F-statistic: 7.087 on 5 and 850 DF, p-value: 1.666e-06## GRNINTL: There should be international agreements that Korea and other countries should be made to follow
grnintllm <- lm(GRNINTL ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(grnintllm)##
## Call:
## lm(formula = GRNINTL ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
## data = predata)
##
## Residuals:
## <Labelled double>: Agree or disagree: For environmental problems, there should be international agreements that Korea and other countries should be made to follow
## Min 1Q Median 3Q Max
## -0.9416 -0.6095 0.1891 0.3827 3.2794
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Agree strongly
## 2 Agree
## 3 Neither agree nor disagree
## 4 Disagree
## 5 Disagree strongly
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.404027 0.181111 13.274 < 2e-16 ***
## AGE -0.007066 0.002156 -3.278 0.00109 **
## SEX -0.047952 0.050529 -0.949 0.34289
## EDUC -0.108649 0.019975 -5.439 7e-08 ***
## RINCOME -0.003633 0.006653 -0.546 0.58515
## PARTYLR 0.014660 0.023954 0.612 0.54069
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6801 on 849 degrees of freedom
## (3385 observations deleted due to missingness)
## Multiple R-squared: 0.04104, Adjusted R-squared: 0.03539
## F-statistic: 7.267 on 5 and 849 DF, p-value: 1.12e-06## ECONGRN: Economic progress in Korea will slow down unless we look after the environment better
econgrnlm <- lm(ECONGRN ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
data=predata)
summary(econgrnlm)##
## Call:
## lm(formula = ECONGRN ~ AGE + SEX + EDUC + RINCOME + PARTYLR,
## data = predata)
##
## Residuals:
## <Labelled double>: Agree or disagree: Economic progress in Korea will slow down unless we look after the environment better
## Min 1Q Median 3Q Max
## -1.37485 -0.32591 -0.21598 0.73066 2.84099
##
## Labels:
## value label
## -8 DK
## -1 IAP
## 1 Agree strongly
## 2 Agree
## 3 Neither agree nor disagree
## 4 Disagree
## 5 Disagree strongly
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.406875 0.255463 9.422 <2e-16 ***
## AGE 0.001059 0.003034 0.349 0.727
## SEX -0.088670 0.071144 -1.246 0.213
## EDUC -0.031119 0.028146 -1.106 0.269
## RINCOME 0.002653 0.009374 0.283 0.777
## PARTYLR 0.003110 0.033666 0.092 0.926
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9585 on 850 degrees of freedom
## (3384 observations deleted due to missingness)
## Multiple R-squared: 0.004811, Adjusted R-squared: -0.001043
## F-statistic: 0.8219 on 5 and 850 DF, p-value: 0.5342