Isabel’s input for Mira
Isabel Pacheco
2025-10-15
Sample descriptives
Building
Overview over variables overall and per country
## overview over new variables
describe(dEU[, c("build_pa", "build_se", "build_oe", "build_pn", "willdown", "willkitchen", "willbath")])## vars n mean sd median trimmed mad min max range skew kurtosis
## build_pa 1 5651 3.09 1.68 3.0 2.95 1.48 1 7 6 0.40 -0.70
## build_se 2 5651 3.41 1.83 3.5 3.31 2.22 1 7 6 0.24 -0.96
## build_oe 3 5651 3.08 1.75 3.0 2.94 2.22 1 7 6 0.39 -0.82
## build_pn 4 5651 2.68 1.70 2.0 2.47 1.48 1 7 6 0.72 -0.46
## willdown 5 5651 3.24 2.03 3.0 3.05 2.97 1 7 6 0.42 -1.09
## willkitchen 6 5632 2.01 1.71 1.0 1.62 0.00 1 7 6 1.65 1.59
## willbath 7 5624 1.89 1.63 1.0 1.49 0.00 1 7 6 1.85 2.35
## se
## build_pa 0.02
## build_se 0.02
## build_oe 0.02
## build_pn 0.02
## willdown 0.03
## willkitchen 0.02
## willbath 0.02#per country
describeBy(dEU[, c("build_pa", "build_se", "build_oe", "build_pn", "willdown", "willkitchen", "willbath")], group = dEU$country) ##
## Descriptive statistics by group
## group: UK
## vars n mean sd median trimmed mad min max range skew kurtosis
## build_pa 1 1053 3.26 1.73 3.0 3.14 1.48 1 7 6 0.36 -0.72
## build_se 2 1053 3.91 1.88 4.0 3.88 2.22 1 7 6 0.00 -1.04
## build_oe 3 1053 3.39 1.81 3.5 3.28 2.22 1 7 6 0.25 -0.91
## build_pn 4 1053 2.91 1.81 2.5 2.71 2.22 1 7 6 0.60 -0.69
## willdown 5 1053 3.38 2.13 3.0 3.22 2.97 1 7 6 0.34 -1.25
## willkitchen 6 1050 2.07 1.84 1.0 1.65 0.00 1 7 6 1.59 1.20
## willbath 7 1050 1.97 1.75 1.0 1.55 0.00 1 7 6 1.73 1.71
## se
## build_pa 0.05
## build_se 0.06
## build_oe 0.06
## build_pn 0.06
## willdown 0.07
## willkitchen 0.06
## willbath 0.05
## ------------------------------------------------------------
## group: Germany
## vars n mean sd median trimmed mad min max range skew kurtosis
## build_pa 1 1100 3.26 1.75 3.5 3.17 2.22 1 7 6 0.18 -1.04
## build_se 2 1100 3.58 1.86 4.0 3.53 2.22 1 7 6 0.03 -1.13
## build_oe 3 1100 3.07 1.80 3.0 2.93 2.97 1 7 6 0.34 -1.05
## build_pn 4 1100 2.50 1.69 2.0 2.26 1.48 1 7 6 0.84 -0.45
## willdown 5 1100 3.27 2.01 3.0 3.12 2.97 1 7 6 0.33 -1.19
## willkitchen 6 1099 1.91 1.64 1.0 1.53 0.00 1 7 6 1.71 1.69
## willbath 7 1099 1.74 1.50 1.0 1.35 0.00 1 7 6 2.04 3.07
## se
## build_pa 0.05
## build_se 0.06
## build_oe 0.05
## build_pn 0.05
## willdown 0.06
## willkitchen 0.05
## willbath 0.05
## ------------------------------------------------------------
## group: Netherlands
## vars n mean sd median trimmed mad min max range skew kurtosis
## build_pa 1 1081 2.92 1.62 3.0 2.77 1.48 1 7 6 0.50 -0.53
## build_se 2 1081 3.35 1.81 3.5 3.25 2.22 1 7 6 0.24 -1.01
## build_oe 3 1081 2.94 1.68 3.0 2.79 2.22 1 7 6 0.44 -0.82
## build_pn 4 1081 2.40 1.58 2.0 2.16 1.48 1 7 6 0.93 -0.06
## willdown 5 1081 3.23 2.10 3.0 3.03 2.97 1 7 6 0.45 -1.15
## willkitchen 6 1075 1.80 1.59 1.0 1.39 0.00 1 7 6 2.04 3.10
## willbath 7 1076 1.69 1.52 1.0 1.27 0.00 1 7 6 2.32 4.34
## se
## build_pa 0.05
## build_se 0.06
## build_oe 0.05
## build_pn 0.05
## willdown 0.06
## willkitchen 0.05
## willbath 0.05
## ------------------------------------------------------------
## group: Italy
## vars n mean sd median trimmed mad min max range skew kurtosis
## build_pa 1 1409 2.98 1.62 3.0 2.84 1.48 1 7 6 0.48 -0.54
## build_se 2 1409 3.24 1.77 3.0 3.10 2.22 1 7 6 0.40 -0.72
## build_oe 3 1409 3.07 1.69 3.0 2.93 2.22 1 7 6 0.43 -0.67
## build_pn 4 1409 2.81 1.65 2.5 2.63 2.22 1 7 6 0.64 -0.41
## willdown 5 1409 3.32 1.95 3.0 3.17 2.97 1 7 6 0.37 -1.01
## willkitchen 6 1408 2.11 1.69 1.0 1.75 0.00 1 7 6 1.50 1.23
## willbath 7 1400 1.97 1.59 1.0 1.62 0.00 1 7 6 1.69 2.01
## se
## build_pa 0.04
## build_se 0.05
## build_oe 0.05
## build_pn 0.04
## willdown 0.05
## willkitchen 0.04
## willbath 0.04
## ------------------------------------------------------------
## group: Lithuania
## vars n mean sd median trimmed mad min max range skew kurtosis
## build_pa 1 1008 3.05 1.70 3.0 2.90 1.48 1 7 6 0.44 -0.61
## build_se 2 1008 3.01 1.69 3.0 2.86 2.22 1 7 6 0.46 -0.63
## build_oe 3 1008 2.95 1.73 3.0 2.78 2.22 1 7 6 0.49 -0.67
## build_pn 4 1008 2.78 1.72 2.5 2.59 2.22 1 7 6 0.60 -0.60
## willdown 5 1008 2.94 1.96 3.0 2.71 2.97 1 7 6 0.65 -0.73
## willkitchen 6 1000 2.16 1.77 1.0 1.78 0.00 1 7 6 1.49 1.18
## willbath 7 999 2.09 1.75 1.0 1.71 0.00 1 7 6 1.57 1.37
## se
## build_pa 0.05
## build_se 0.05
## build_oe 0.05
## build_pn 0.05
## willdown 0.06
## willkitchen 0.06
## willbath 0.06Checking for differences in variables
Country
library(effectsize)##
## Attaching package: 'effectsize'## The following object is masked from 'package:xtable':
##
## display## The following object is masked from 'package:psych':
##
## phi# Problem awareness
aov_build_pa <- aov(build_pa ~ country, data = dEU)
summary(aov_build_pa) #sig## Df Sum Sq Mean Sq F value Pr(>F)
## country 4 114 28.61 10.15 3.5e-08 ***
## Residuals 5646 15919 2.82
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_build_pa, partial = FALSE) # = 0.004## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## -----------------------------------
## country | 7.14e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Self-efficacy
aov_build_se <- aov(build_se ~ country, data = dEU)
summary(aov_build_se) #sig.## Df Sum Sq Mean Sq F value Pr(>F)
## country 4 494 123.50 38 <2e-16 ***
## Residuals 5646 18348 3.25
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_build_se, partial = FALSE) # = 0.03## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## -------------------------------
## country | 0.03 | [0.02, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Outcome efficacy
aov_build_oe <- aov(build_oe ~ country, data = dEU)
summary(aov_build_oe) #sig## Df Sum Sq Mean Sq F value Pr(>F)
## country 4 137 34.14 11.27 4.15e-09 ***
## Residuals 5646 17103 3.03
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_build_oe, partial = FALSE) # = 0.004## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## -----------------------------------
## country | 7.92e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Personal norms
aov_build_pn <- aov(build_pn ~ country, data = dEU)
summary(aov_build_pn) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## country 4 216 54.10 18.97 1.64e-15 ***
## Residuals 5646 16098 2.85
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Downsize
aov_willdown <- aov(willdown ~ country, data = dEU)
summary(aov_willdown) #sig.## Df Sum Sq Mean Sq F value Pr(>F)
## country 4 117 29.337 7.143 9.85e-06 ***
## Residuals 5646 23188 4.107
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_willdown, partial = FALSE) # = 0.003## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## -----------------------------------
## country | 5.04e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Share spaces
aov_willkitchen <- aov(willkitchen ~ country, data = dEU)
summary(aov_willkitchen) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## country 4 100 24.973 8.616 6.28e-07 ***
## Residuals 5627 16310 2.899
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 19 observations deleted due to missingnessaov_willbath <- aov(willbath ~ country, data = dEU)
summary(aov_willbath) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## country 4 125 31.35 11.92 1.2e-09 ***
## Residuals 5619 14777 2.63
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 27 observations deleted due to missingnessOverall, there are hardly any country differences between the variables, especially in the influencing factors. There are slight differences in the willingness.
Dwelling
library(effectsize)
# Problem awareness
aov_build_pa <- aov(build_pa ~ dwelling_messageix, data = dEU)
summary(aov_build_pa) #not sig## Df Sum Sq Mean Sq F value Pr(>F)
## dwelling_messageix 1 13 13.089 4.615 0.0317 *
## Residuals 5649 16020 2.836
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Self-efficacy
aov_build_se <- aov(build_se ~ dwelling_messageix, data = dEU)
summary(aov_build_se) #sig.## Df Sum Sq Mean Sq F value Pr(>F)
## dwelling_messageix 1 24 23.964 7.194 0.00734 **
## Residuals 5649 18818 3.331
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_build_se, partial = FALSE) # = 0.001## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## --------------------------------------------
## dwelling_messageix | 1.27e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Outcome efficacy
aov_build_oe <- aov(build_oe ~ dwelling_messageix, data = dEU)
summary(aov_build_oe) #not sig## Df Sum Sq Mean Sq F value Pr(>F)
## dwelling_messageix 1 2 1.760 0.577 0.448
## Residuals 5649 17238 3.051# Personal norms
aov_build_pn <- aov(build_pn ~ dwelling_messageix, data = dEU)
summary(aov_build_pn) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## dwelling_messageix 1 0 0.1865 0.065 0.799
## Residuals 5649 16315 2.8880# Downsize
aov_willdown <- aov(willdown ~ dwelling_messageix, data = dEU)
summary(aov_willdown) #sig.## Df Sum Sq Mean Sq F value Pr(>F)
## dwelling_messageix 1 137 136.6 33.31 8.29e-09 ***
## Residuals 5649 23169 4.1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_willdown, partial = FALSE) # = 0.006## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## --------------------------------------------
## dwelling_messageix | 5.86e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Share spaces
aov_willkitchen <- aov(willkitchen ~ dwelling_messageix, data = dEU)
summary(aov_willkitchen) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## dwelling_messageix 1 40 39.91 13.73 0.000213 ***
## Residuals 5630 16370 2.91
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 19 observations deleted due to missingnesseta_squared(aov_willkitchen, partial = FALSE) # = 0.002## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## --------------------------------------------
## dwelling_messageix | 2.43e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].aov_willbath <- aov(willbath ~ dwelling_messageix, data = dEU)
summary(aov_willbath) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## dwelling_messageix 1 28 27.784 10.5 0.0012 **
## Residuals 5622 14875 2.646
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 27 observations deleted due to missingnesseta_squared(aov_willbath, partial = FALSE) # = 0.003## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## --------------------------------------------
## dwelling_messageix | 1.86e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].Residence
library(effectsize)
# Problem awareness
aov_build_pa <- aov(build_pa ~ residence_messageix, data = dEU)
summary(aov_build_pa) #sig## Df Sum Sq Mean Sq F value Pr(>F)
## residence_messageix 2 114 57.19 20.29 1.66e-09 ***
## Residuals 5648 15919 2.82
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_build_pa, partial = FALSE) # = 0.007## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## ---------------------------------------------
## residence_messageix | 7.13e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Self-efficacy
aov_build_se <- aov(build_se ~ residence_messageix, data = dEU)
summary(aov_build_se) #not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## residence_messageix 2 13 6.428 1.928 0.145
## Residuals 5648 18829 3.334# Outcome efficacy
aov_build_oe <- aov(build_oe ~ residence_messageix, data = dEU)
summary(aov_build_oe) #sig## Df Sum Sq Mean Sq F value Pr(>F)
## residence_messageix 2 54 27.107 8.909 0.000137 ***
## Residuals 5648 17185 3.043
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_build_oe, partial = FALSE) # = 0.003## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## ---------------------------------------------
## residence_messageix | 3.14e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Personal norms
aov_build_pn <- aov(build_pn ~ residence_messageix, data = dEU)
summary(aov_build_pn) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## residence_messageix 2 126 62.95 21.96 3.16e-10 ***
## Residuals 5648 16189 2.87
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1eta_squared(aov_build_pn, partial = FALSE) # = 0.008## # Effect Size for ANOVA (Type I)
##
## Parameter | Eta2 | 95% CI
## ---------------------------------------------
## residence_messageix | 7.72e-03 | [0.00, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].# Downsize
aov_willdown <- aov(willdown ~ residence_messageix, data = dEU)
summary(aov_willdown) #not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## residence_messageix 2 1 0.553 0.134 0.875
## Residuals 5648 23304 4.126# Share spaces
aov_willkitchen <- aov(willkitchen ~ residence_messageix, data = dEU)
summary(aov_willkitchen) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## residence_messageix 2 220 109.91 38.22 <2e-16 ***
## Residuals 5629 16190 2.88
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 19 observations deleted due to missingnessaov_willbath <- aov(willbath ~ residence_messageix, data = dEU)
summary(aov_willbath) # not sig.## Df Sum Sq Mean Sq F value Pr(>F)
## residence_messageix 2 244 121.82 46.71 <2e-16 ***
## Residuals 5621 14659 2.61
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 27 observations deleted due to missingnessCorrelation table
apa.cor.table(dEU[, c("build_pa", "build_se", "build_oe", "build_pn", "willdown", "willkitchen", "willbath")], filename = "Building_EUcor_Mira.doc")##
##
## Means, standard deviations, and correlations with confidence intervals
##
##
## Variable M SD 1 2 3 4
## 1. build_pa 3.09 1.68
##
## 2. build_se 3.41 1.83 .57**
## [.55, .59]
##
## 3. build_oe 3.08 1.75 .71** .62**
## [.70, .73] [.60, .63]
##
## 4. build_pn 2.68 1.70 .67** .58** .82**
## [.66, .69] [.56, .59] [.81, .83]
##
## 5. willdown 3.24 2.03 .39** .50** .47** .44**
## [.36, .41] [.48, .52] [.45, .49] [.41, .46]
##
## 6. willkitchen 2.01 1.71 .30** .32** .33** .39**
## [.28, .33] [.30, .35] [.31, .36] [.36, .41]
##
## 7. willbath 1.89 1.63 .30** .31** .33** .39**
## [.28, .33] [.29, .33] [.30, .35] [.37, .42]
##
## 5 6
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## .27**
## [.25, .30]
##
## .27** .88**
## [.24, .29] [.87, .88]
##
##
## Note. M and SD are used to represent mean and standard deviation, respectively.
## Values in square brackets indicate the 95% confidence interval.
## The confidence interval is a plausible range of population correlations
## that could have caused the sample correlation (Cumming, 2014).
## * indicates p < .05. ** indicates p < .01.
## Regression analysis
#willingness to downsize
lm_willdown <- lm(willdown ~ build_pa + build_se + build_oe + build_pn, data = dEU)
summary(lm_willdown)##
## Call:
## lm(formula = willdown ~ build_pa + build_se + build_oe + build_pn,
## data = dEU)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.8884 -1.0276 -0.2046 1.1116 5.3042
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.004260 0.053027 18.939 < 2e-16 ***
## build_pa -0.001025 0.020169 -0.051 0.959
## build_se 0.373631 0.016358 22.841 < 2e-16 ***
## build_oe 0.217409 0.024988 8.701 < 2e-16 ***
## build_pn 0.107722 0.023940 4.500 6.94e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.706 on 5646 degrees of freedom
## Multiple R-squared: 0.2952, Adjusted R-squared: 0.2947
## F-statistic: 591.3 on 4 and 5646 DF, p-value: < 2.2e-16#willingness to share spaces
lm_willkitchen <- lm(willkitchen ~ build_pa + build_se + build_oe + build_pn, data = dEU)
summary(lm_willkitchen)##
## Call:
## lm(formula = willkitchen ~ build_pa + build_se + build_oe + build_pn,
## data = dEU)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.8850 -0.9335 -0.3093 0.4733 5.8730
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.70569 0.04862 14.514 <2e-16 ***
## build_pa 0.03709 0.01848 2.007 0.0448 *
## build_se 0.13668 0.01500 9.112 <2e-16 ***
## build_oe -0.02710 0.02287 -1.185 0.2361
## build_pn 0.30171 0.02192 13.764 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.56 on 5627 degrees of freedom
## (19 observations deleted due to missingness)
## Multiple R-squared: 0.1653, Adjusted R-squared: 0.1648
## F-statistic: 278.7 on 4 and 5627 DF, p-value: < 2.2e-16lm_willbath <- lm(willbath ~ build_pa + build_se + build_oe + build_pn, data = dEU)
summary(lm_willbath)##
## Call:
## lm(formula = willbath ~ build_pa + build_se + build_oe + build_pn,
## data = dEU)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.7247 -0.8420 -0.2734 0.1214 5.9488
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.67714 0.04635 14.608 < 2e-16 ***
## build_pa 0.03879 0.01765 2.197 0.0280 *
## build_se 0.11110 0.01430 7.770 9.25e-15 ***
## build_oe -0.05048 0.02180 -2.316 0.0206 *
## build_pn 0.32514 0.02090 15.561 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.487 on 5619 degrees of freedom
## (27 observations deleted due to missingness)
## Multiple R-squared: 0.1668, Adjusted R-squared: 0.1662
## F-statistic: 281.2 on 4 and 5619 DF, p-value: < 2.2e-16Regression for subgroups
Willingness to downsize
Regression analyses depending on country, level of urbanity, and dwelling type
#overview over number of people per group
with(dEU, table(country, dwelling_messageix, residence_messageix))## , , residence_messageix = rural
##
## dwelling_messageix
## country mfh sfh
## UK 23 132
## Germany 58 176
## Netherlands 60 336
## Italy 83 193
## Lithuania 23 96
##
## , , residence_messageix = suburban
##
## dwelling_messageix
## country mfh sfh
## UK 84 384
## Germany 241 156
## Netherlands 134 359
## Italy 406 307
## Lithuania 69 127
##
## , , residence_messageix = urban
##
## dwelling_messageix
## country mfh sfh
## UK 144 286
## Germany 341 128
## Netherlands 89 103
## Italy 291 129
## Lithuania 484 209dEU %>%
count(country, dwelling_messageix, residence_messageix, name = "N")## # A tibble: 30 × 4
## country dwelling_messageix residence_messageix N
## <fct> <chr> <chr> <int>
## 1 UK mfh rural 23
## 2 UK mfh suburban 84
## 3 UK mfh urban 144
## 4 UK sfh rural 132
## 5 UK sfh suburban 384
## 6 UK sfh urban 286
## 7 Germany mfh rural 58
## 8 Germany mfh suburban 241
## 9 Germany mfh urban 341
## 10 Germany sfh rural 176
## # ℹ 20 more rowsgroup_sizes <- dEU %>%
count(country, dwelling_messageix, residence_messageix, name = "N")
group_sizes %>%
filter(N < 60)## # A tibble: 3 × 4
## country dwelling_messageix residence_messageix N
## <fct> <chr> <chr> <int>
## 1 UK mfh rural 23
## 2 Germany mfh rural 58
## 3 Lithuania mfh rural 23#smallest groups:
# UK, mfh, rural -> 23
# Lithuania, mfh, rural -> 23
# Germany, mfh, rural -> 58
# -> Apartment in rural area is very uncommon
describe(group_sizes$N)## vars n mean sd median trimmed mad min max range skew kurtosis
## X1 1 30 188.37 125.25 139 178.04 103.78 23 484 461 0.65 -0.78
## se
## X1 22.87ggplot(group_sizes,
aes(x = dwelling_messageix, y = N, fill = residence_messageix)) +
geom_col(position = "dodge") +
facet_wrap(~ country) +
labs(title = "Group sizes by country, dwelling, and urbanity")
library(dplyr)
library(purrr)
## regression model for each subgroup
#downsizing
model_willdown <- dEU %>%
group_by(country, dwelling_messageix, residence_messageix) %>%
group_modify(~ tibble(
model = list(
lm(willdown ~ build_pa + build_se + build_oe + build_pn, data = .x)
),
model_willdown = mean(.x$willbath, na.rm = TRUE)
)) %>%
ungroup() %>%
mutate(model_id = row_number())
#summary(model_willdown$model[[4]])
library(broom)
library(tidyr)
library(purrr)
model_willdown %>%
mutate(coefs = map(model, tidy)) %>%
unnest(coefs)## # A tibble: 150 × 11
## country dwelling_messageix residence_messageix model model_willdown model_id
## <fct> <chr> <chr> <list> <dbl> <int>
## 1 UK mfh rural <lm> 2.18 1
## 2 UK mfh rural <lm> 2.18 1
## 3 UK mfh rural <lm> 2.18 1
## 4 UK mfh rural <lm> 2.18 1
## 5 UK mfh rural <lm> 2.18 1
## 6 UK mfh suburban <lm> 2.13 2
## 7 UK mfh suburban <lm> 2.13 2
## 8 UK mfh suburban <lm> 2.13 2
## 9 UK mfh suburban <lm> 2.13 2
## 10 UK mfh suburban <lm> 2.13 2
## # ℹ 140 more rows
## # ℹ 5 more variables: term <chr>, estimate <dbl>, std.error <dbl>,
## # statistic <dbl>, p.value <dbl>Willingness to downsize
Regression analyses depending on country, level of urbanity, and dwelling type
model_willbath <- dEU %>%
group_by(country, dwelling_messageix, residence_messageix) %>%
group_modify(~ tibble(
model = list(
lm(willbath ~ build_pa + build_se + build_oe + build_pn, data = .x)
),
mean_willbath = mean(.x$willbath, na.rm = TRUE)
)) %>%
ungroup() %>%
mutate(model_id = row_number())
model_willbath %>%
mutate(coefs = map(model, tidy)) %>%
unnest(coefs)## # A tibble: 150 × 11
## country dwelling_messageix residence_messageix model mean_willbath model_id
## <fct> <chr> <chr> <list> <dbl> <int>
## 1 UK mfh rural <lm> 2.18 1
## 2 UK mfh rural <lm> 2.18 1
## 3 UK mfh rural <lm> 2.18 1
## 4 UK mfh rural <lm> 2.18 1
## 5 UK mfh rural <lm> 2.18 1
## 6 UK mfh suburban <lm> 2.13 2
## 7 UK mfh suburban <lm> 2.13 2
## 8 UK mfh suburban <lm> 2.13 2
## 9 UK mfh suburban <lm> 2.13 2
## 10 UK mfh suburban <lm> 2.13 2
## # ℹ 140 more rows
## # ℹ 5 more variables: term <chr>, estimate <dbl>, std.error <dbl>,
## # statistic <dbl>, p.value <dbl>