library(dplyr)
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
library(ggplot2)
## Warning: package 'ggplot2' was built under R version 3.6.3
library(sjPlot)
## Warning: package 'sjPlot' was built under R version 3.6.3
## Learn more about sjPlot with 'browseVignettes("sjPlot")'.
library(psych)
## Warning: package 'psych' was built under R version 3.6.3
##
## Attaching package: 'psych'
## The following objects are masked from 'package:ggplot2':
##
## %+%, alpha
folder <- "E:/Analyse_Corona_Einkaufen/Exporte/upload/data_purchasing_under_threat.csv" #define path
data_sample <- read.table(folder,
header = TRUE,
sep = ","
)
#data_sample$sem_diff <- as.numeric(as.character(data_sample$sem_diff))
data_sample$householdsize <- as.numeric(as.character(data_sample$householdsize))
#Functions
preprocess_overall_pruchasing <- function(df){
continuous_vars <- select(df,
FreqMarch,
QuaMarch,
age,
education,
householdsize,
SEA,
sem_diff,
IUS,
STAI,
info_verlauf,
Prob_Einkaufen)
df_export <- data.frame(select(df, ID, sex, risk_self, risk_loved),
apply(continuous_vars, 2, scale))
return(df_export)
}
preprocess_overall_pruchasing_filtered <- function(df){
group_fre <- c()
group_fre[df$FreqMarch < 0] <- "less"
group_fre[df$FreqMarch == 0] <- "same"
group_fre[df$FreqMarch > 0] <- "more"
group_qua <- c()
group_qua[df$QuaMarch < 0] <- "less"
group_qua[df$QuaMarch == 0] <- "same"
group_qua[df$QuaMarch > 0] <- "more"
filtered_df <- filter(df, group_qua != "less" & group_fre != "more")
continuous_vars <- select(filtered_df,
FreqMarch,
QuaMarch,
age,
education,
householdsize,
SEA,
sem_diff,
IUS,
STAI,
info_verlauf,
Prob_Einkaufen)
df_export <- data.frame(select(filtered_df, ID, sex, risk_self, risk_loved),
apply(continuous_vars, 2, scale))
return(df_export)
}
preprocess_individual_products <- function(df, t, DV){
df$Konserven_March[df$Konserven_March < 0] <- t
df$Seife_March[df$Seife_March < 0] <- t
df$Toilettenpapier_March[df$Toilettenpapier_March < 0] <- t
df$Nudeln_Reis_March[df$Nudeln_Reis_March < 0] <- t
df$Hefe_March[df$Hefe_March < 0] <- t
df$frisch_March[df$frisch_March < 0] <- t
df$Desinfektion_March[df$Desinfektion_March < 0] <- t
NonPerishableFood <- rowMeans(select(df, Desinfektion_March,
Seife_March,
Toilettenpapier_March), na.rm = TRUE)
HygieneProducts <- rowMeans(select(df, Nudeln_Reis_March,
Konserven_March,
Toilettenpapier_March), na.rm = TRUE)
FreshFood <- df$frisch_March
df_new <- data.frame(select(df, ID:Prob_Einkaufen, info_verlauf),
NonPerishableFood,
HygieneProducts,
FreshFood)
continuous_vars <- select(df_new,
all_of(DV),
age,
education,
householdsize,
SEA,
sem_diff,
IUS,
STAI,
info_verlauf,
Prob_Einkaufen)
df_export <- data.frame(select(df_new, ID, sex, risk_self, risk_loved),
apply(continuous_vars, 2, scale))
return(na.omit(df_export))
}
print_model <- function(df, model, add_variable){
if(model == "Baseline"){
subdata_Freq <- select(df, FreqMarch, sex, age, education, householdsize, SEA)
subdata_Qua <- select(df, QuaMarch, sex, age, education, householdsize, SEA)
Model_Freq <- lm(FreqMarch ~ ., data = subdata_Freq)
Model_Qua <- lm(QuaMarch ~ ., data = subdata_Qua)
table_model <- tab_model(Model_Freq, Model_Qua,
show.intercept = FALSE,
show.ci = FALSE,
show.se = FALSE,
show.stat = TRUE,
#show.df = TRUE,
#p.style = "scientific",
#string.se = "SE",
string.est = "b")
return(table_model)
}
else if(model == "specific"){
subdata_Freq <- select(df, FreqMarch, sex, age, education, householdsize, SEA, add_variable)
subdata_Qua <- select(df, QuaMarch, sex, age, education, householdsize, SEA, add_variable)
Model_Freq <- lm(FreqMarch ~ ., data = subdata_Freq)
Model_Qua <- lm(QuaMarch ~ ., data = subdata_Qua)
table_model <- tab_model(Model_Freq, Model_Qua,
show.intercept = FALSE,
show.ci = FALSE,
show.se = FALSE,
show.stat = TRUE,
#show.df = TRUE,
#p.style = "scientific",
#string.se = "SE",
string.est = "b")
return(table_model)
}
else if(model == "overall"){
subdata_Freq <-select(df, FreqMarch,
sex,
age,
education,
householdsize,
SEA,
sem_diff,
#risk_self,
#risk_loved,
IUS,
STAI,
info_verlauf,
Prob_Einkaufen)
subdata_Qua <-select(df, QuaMarch,
sex,
age,
education,
householdsize,
SEA,
sem_diff,
#risk_self,
#risk_loved,
IUS,
STAI,
info_verlauf,
Prob_Einkaufen)
Model_Freq <- lm(FreqMarch ~ ., data = subdata_Freq)
Model_Qua <- lm(QuaMarch ~ ., data = subdata_Qua)
table_model <- tab_model(Model_Freq, Model_Qua,
show.intercept = FALSE,
show.ci = FALSE,
show.se = FALSE,
show.stat = TRUE,
#show.df = TRUE,
#p.style = "scientific",
#string.se = "SE",
string.est = "b")
return(table_model)
}
else{"Something went wrong!"}
}
#Descriptives
describe(data_sample)
## vars n mean sd median trimmed mad min
## X 1 813 407.00 234.84 407.00 407.00 300.97 1
## ID 2 813 1437.23 734.81 1341.00 1402.01 929.59 348
## sex 3 813 0.22 0.41 0.00 0.14 0.00 0
## age 4 813 42.42 15.00 42.00 42.02 19.27 18
## education 5 813 4.08 0.91 4.00 4.16 1.48 1
## householdsize 6 813 2.40 1.56 2.00 2.21 1.48 1
## risk_self 7 813 0.41 0.49 0.00 0.39 0.00 0
## risk_loved 8 813 0.50 0.50 1.00 0.50 0.00 0
## SEA 9 813 19.86 2.33 20.00 19.88 2.97 13
## IUS 10 813 32.50 9.92 32.00 32.24 10.38 12
## STAI 11 813 40.31 11.13 39.00 39.69 11.86 20
## sem_diff 12 813 4.16 1.36 4.17 4.18 1.48 1
## Prob_Einkaufen 13 813 25.68 25.57 18.00 22.05 23.72 0
## FreqMarch 14 813 -0.86 1.35 -1.00 -0.91 1.48 -3
## QuaMarch 15 813 0.58 1.12 0.00 0.55 1.48 -3
## info_verlauf 16 813 3.20 0.72 3.00 3.26 1.48 1
## Konserven_March 17 666 0.39 0.95 0.00 0.33 0.00 -3
## Seife_March 18 787 0.26 0.84 0.00 0.18 0.00 -3
## Toilettenpapier_March 19 803 -0.02 1.02 0.00 0.07 0.00 -3
## Nudeln_Reis_March 20 797 0.31 0.95 0.00 0.29 0.00 -3
## Hefe_March 21 462 -0.38 1.27 0.00 -0.29 0.00 -3
## frisch_March 22 794 0.18 0.70 0.00 0.11 0.00 -3
## Desinfektion_March 23 489 0.42 1.25 0.00 0.46 1.48 -3
## diff_sorge 24 812 4.29 1.85 5.00 4.36 1.48 1
## diff_angst 25 808 3.49 1.84 4.00 3.40 2.97 1
## diff_denke 26 808 4.53 1.53 5.00 4.60 1.48 1
## diff_hilflos 27 811 3.98 1.82 4.00 3.97 1.48 1
## diff_belastend 28 812 4.50 1.90 5.00 4.63 1.48 1
## diff_nah 29 809 4.16 1.64 4.00 4.22 1.48 1
## STAI1 30 813 2.88 0.72 3.00 2.87 0.00 1
## STAI2 31 813 2.29 0.82 2.00 2.26 1.48 1
## STAI3 32 813 1.74 0.73 2.00 1.66 1.48 1
## STAI4 33 813 1.47 0.72 1.00 1.33 0.00 1
## STAI5 34 813 1.86 0.75 2.00 1.80 1.48 1
## STAI6 35 813 2.52 0.83 3.00 2.53 1.48 1
## STAI7 36 813 2.78 0.82 3.00 2.79 1.48 1
## STAI8 37 813 1.77 0.77 2.00 1.68 1.48 1
## STAI9 38 813 2.25 0.93 2.00 2.19 1.48 1
## STAI10 39 813 2.95 0.81 3.00 2.98 1.48 1
## STAI11 40 813 1.92 0.88 2.00 1.84 1.48 1
## STAI12 41 813 1.99 0.89 2.00 1.89 1.48 1
## STAI13 42 813 2.82 0.93 3.00 2.88 1.48 1
## STAI14 43 813 1.92 0.82 2.00 1.85 1.48 1
## STAI15 44 813 1.81 0.77 2.00 1.73 1.48 1
## STAI16 45 813 3.08 0.82 3.00 3.14 1.48 1
## STAI17 46 813 2.07 0.86 2.00 2.00 1.48 1
## STAI18 47 813 2.06 0.92 2.00 1.97 1.48 1
## STAI19 48 813 2.82 0.85 3.00 2.84 1.48 1
## STAI20 49 813 2.00 0.90 2.00 1.91 1.48 1
## IUS1 50 813 2.77 1.24 3.00 2.71 1.48 1
## IUS2 51 813 3.39 1.29 3.00 3.49 1.48 1
## IUS3 52 813 2.66 1.37 3.00 2.57 1.48 1
## IUS4 53 813 3.18 1.25 3.00 3.23 1.48 1
## IUS5 54 813 2.66 1.31 3.00 2.57 1.48 1
## IUS6 55 813 2.30 1.27 2.00 2.14 1.48 1
## IUS7 56 813 2.67 1.28 3.00 2.59 1.48 1
## IUS8 57 813 2.86 1.36 3.00 2.83 1.48 1
## IUS9 58 813 2.31 1.26 2.00 2.15 1.48 1
## IUS10 59 813 2.22 1.22 2.00 2.06 1.48 1
## IUS11 60 813 3.10 1.29 3.00 3.13 1.48 1
## IUS12 61 813 2.38 1.24 2.00 2.24 1.48 1
## max range skew kurtosis se
## X 813 812 0.00 -1.20 8.24
## ID 2858 2510 0.30 -1.13 25.77
## sex 1 1 1.38 -0.09 0.01
## age 79 61 0.13 -1.20 0.53
## education 5 4 -0.62 -0.47 0.03
## householdsize 20 19 4.44 41.44 0.05
## risk_self 1 1 0.36 -1.88 0.02
## risk_loved 1 1 -0.01 -2.00 0.02
## SEA 27 14 -0.07 0.03 0.08
## IUS 60 48 0.24 -0.40 0.35
## STAI 72 52 0.44 -0.43 0.39
## sem_diff 7 6 -0.14 -0.50 0.05
## Prob_Einkaufen 100 100 1.03 0.22 0.90
## FreqMarch 3 6 0.29 0.00 0.05
## QuaMarch 3 6 0.11 0.84 0.04
## info_verlauf 4 3 -0.44 -0.58 0.03
## Konserven_March 3 6 0.23 2.52 0.04
## Seife_March 3 6 0.30 4.05 0.03
## Toilettenpapier_March 3 6 -0.70 2.97 0.04
## Nudeln_Reis_March 3 6 0.01 2.64 0.03
## Hefe_March 3 6 -0.65 0.63 0.06
## frisch_March 3 6 1.02 5.39 0.02
## Desinfektion_March 3 6 -0.37 1.26 0.06
## diff_sorge 7 6 -0.36 -0.95 0.06
## diff_angst 7 6 0.21 -1.04 0.06
## diff_denke 7 6 -0.42 -0.33 0.05
## diff_hilflos 7 6 0.02 -0.96 0.06
## diff_belastend 7 6 -0.43 -0.91 0.07
## diff_nah 7 6 -0.28 -0.61 0.06
## STAI1 4 3 -0.06 -0.55 0.03
## STAI2 4 3 0.30 -0.37 0.03
## STAI3 4 3 0.70 0.03 0.03
## STAI4 4 3 1.49 1.75 0.03
## STAI5 4 3 0.53 -0.21 0.03
## STAI6 4 3 -0.11 -0.54 0.03
## STAI7 4 3 -0.13 -0.62 0.03
## STAI8 4 3 0.78 0.15 0.03
## STAI9 4 3 0.32 -0.74 0.03
## STAI10 4 3 -0.30 -0.58 0.03
## STAI11 4 3 0.64 -0.41 0.03
## STAI12 4 3 0.69 -0.20 0.03
## STAI13 4 3 -0.29 -0.83 0.03
## STAI14 4 3 0.60 -0.20 0.03
## STAI15 4 3 0.72 0.15 0.03
## STAI16 4 3 -0.48 -0.57 0.03
## STAI17 4 3 0.52 -0.32 0.03
## STAI18 4 3 0.53 -0.57 0.03
## STAI19 4 3 -0.15 -0.78 0.03
## STAI20 4 3 0.54 -0.56 0.03
## IUS1 5 4 0.18 -0.77 0.04
## IUS2 5 4 -0.37 -0.78 0.05
## IUS3 5 4 0.29 -1.07 0.05
## IUS4 5 4 -0.14 -0.75 0.04
## IUS5 5 4 0.30 -0.90 0.05
## IUS6 5 4 0.64 -0.59 0.04
## IUS7 5 4 0.25 -0.88 0.04
## IUS8 5 4 0.10 -1.08 0.05
## IUS9 5 4 0.62 -0.55 0.04
## IUS10 5 4 0.71 -0.38 0.04
## IUS11 5 4 -0.10 -0.91 0.05
## IUS12 5 4 0.52 -0.56 0.04
#Threat of COVID-19 (sem_diff)
df_sem_diff <- select(data_sample, diff_sorge:diff_nah)
#STAI
data_sample$STAI1_rec <- as.numeric(recode(as.character(data_sample$STAI1), "1" = "4", "2" = "3", "3" = "2", "4" = "1", ))
data_sample$STAI6_rec <- as.numeric(recode(as.character(data_sample$STAI6), "1" = "4", "2" = "3", "3" = "2", "4" = "1", ))
data_sample$STAI7_rec <- as.numeric(recode(as.character(data_sample$STAI7), "1" = "4", "2" = "3", "3" = "2", "4" = "1", ))
data_sample$STAI10_rec <- as.numeric(recode(as.character(data_sample$STAI10), "1" = "4", "2" = "3", "3" = "2", "4" = "1", ))
data_sample$STAI13_rec <- as.numeric(recode(as.character(data_sample$STAI13), "1" = "4", "2" = "3", "3" = "2", "4" = "1", ))
data_sample$STAI16_rec <- as.numeric(recode(as.character(data_sample$STAI16), "1" = "4", "2" = "3", "3" = "2", "4" = "1", ))
data_sample$STAI19_rec <- as.numeric(recode(as.character(data_sample$STAI19), "1" = "4", "2" = "3", "3" = "2", "4" = "1", ))
df_STAI <- select(data_sample,
STAI1_rec,
STAI2,
STAI3,
STAI4,
STAI5,
STAI6_rec,
STAI7_rec,
STAI8,
STAI9,
STAI10_rec,
STAI11,
STAI12,
STAI13_rec,
STAI14,
STAI15,
STAI16_rec,
STAI17,
STAI18,
STAI19_rec,
STAI20)
#IUS
df_IUS <- select(data_sample, IUS1:IUS12)
#Cronbach´s Alpha
psych::alpha(df_sem_diff)
##
## Reliability analysis
## Call: psych::alpha(x = df_sem_diff)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.86 0.86 0.85 0.51 6.2 0.0074 4.2 1.4 0.52
##
## lower alpha upper 95% confidence boundaries
## 0.85 0.86 0.88
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r
## diff_sorge 0.82 0.82 0.79 0.47 4.5 0.0100 0.0074
## diff_angst 0.82 0.82 0.79 0.47 4.4 0.0101 0.0067
## diff_denke 0.85 0.85 0.83 0.53 5.5 0.0084 0.0133
## diff_hilflos 0.84 0.84 0.83 0.52 5.4 0.0085 0.0113
## diff_belastend 0.84 0.84 0.82 0.51 5.2 0.0086 0.0146
## diff_nah 0.86 0.86 0.84 0.54 5.9 0.0079 0.0082
## med.r
## diff_sorge 0.50
## diff_angst 0.52
## diff_denke 0.53
## diff_hilflos 0.52
## diff_belastend 0.51
## diff_nah 0.52
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## diff_sorge 812 0.85 0.84 0.83 0.76 4.3 1.8
## diff_angst 808 0.85 0.85 0.84 0.77 3.5 1.8
## diff_denke 808 0.71 0.73 0.64 0.60 4.5 1.5
## diff_hilflos 811 0.75 0.74 0.66 0.62 4.0 1.8
## diff_belastend 812 0.77 0.76 0.69 0.64 4.5 1.9
## diff_nah 809 0.68 0.69 0.59 0.55 4.2 1.6
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## diff_sorge 0.11 0.11 0.10 0.14 0.25 0.18 0.11 0.00
## diff_angst 0.19 0.18 0.13 0.18 0.18 0.08 0.07 0.01
## diff_denke 0.04 0.07 0.12 0.23 0.26 0.19 0.09 0.01
## diff_hilflos 0.11 0.13 0.16 0.22 0.15 0.12 0.11 0.00
## diff_belastend 0.10 0.10 0.07 0.14 0.25 0.16 0.18 0.00
## diff_nah 0.09 0.09 0.12 0.26 0.23 0.14 0.07 0.00
psych::alpha(df_STAI)
##
## Reliability analysis
## Call: psych::alpha(x = df_STAI)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.94 0.94 0.95 0.42 15 0.0033 2 0.56 0.42
##
## lower alpha upper 95% confidence boundaries
## 0.93 0.94 0.94
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## STAI1_rec 0.93 0.93 0.94 0.43 14 0.0034 0.0110 0.42
## STAI2 0.93 0.94 0.95 0.43 14 0.0033 0.0112 0.43
## STAI3 0.93 0.93 0.94 0.42 14 0.0034 0.0114 0.42
## STAI4 0.93 0.93 0.94 0.43 14 0.0034 0.0114 0.42
## STAI5 0.94 0.94 0.95 0.44 15 0.0033 0.0104 0.43
## STAI6_rec 0.93 0.93 0.94 0.43 14 0.0034 0.0110 0.42
## STAI7_rec 0.93 0.93 0.94 0.43 14 0.0034 0.0113 0.42
## STAI8 0.93 0.93 0.94 0.42 14 0.0035 0.0114 0.41
## STAI9 0.93 0.93 0.94 0.43 14 0.0033 0.0105 0.42
## STAI10_rec 0.93 0.93 0.94 0.42 14 0.0035 0.0098 0.41
## STAI11 0.93 0.93 0.94 0.42 14 0.0035 0.0112 0.41
## STAI12 0.93 0.93 0.94 0.43 14 0.0034 0.0118 0.42
## STAI13_rec 0.93 0.93 0.94 0.43 14 0.0034 0.0103 0.42
## STAI14 0.93 0.93 0.94 0.43 14 0.0034 0.0108 0.42
## STAI15 0.93 0.93 0.94 0.42 14 0.0035 0.0109 0.41
## STAI16_rec 0.93 0.93 0.94 0.42 14 0.0035 0.0098 0.41
## STAI17 0.93 0.93 0.94 0.42 14 0.0034 0.0109 0.42
## STAI18 0.93 0.93 0.94 0.43 14 0.0034 0.0117 0.42
## STAI19_rec 0.93 0.93 0.94 0.42 14 0.0035 0.0103 0.41
## STAI20 0.93 0.93 0.94 0.42 14 0.0035 0.0117 0.41
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## STAI1_rec 813 0.64 0.65 0.63 0.60 2.1 0.72
## STAI2 813 0.58 0.58 0.55 0.53 2.3 0.82
## STAI3 813 0.66 0.67 0.65 0.62 1.7 0.73
## STAI4 813 0.64 0.65 0.62 0.60 1.5 0.72
## STAI5 813 0.53 0.53 0.49 0.48 1.9 0.75
## STAI6_rec 813 0.64 0.64 0.62 0.59 2.5 0.83
## STAI7_rec 813 0.65 0.65 0.62 0.60 2.2 0.82
## STAI8 813 0.73 0.73 0.72 0.70 1.8 0.77
## STAI9 813 0.62 0.61 0.60 0.57 2.3 0.93
## STAI10_rec 813 0.72 0.72 0.72 0.68 2.1 0.81
## STAI11 813 0.77 0.76 0.75 0.73 1.9 0.88
## STAI12 813 0.66 0.65 0.63 0.61 2.0 0.89
## STAI13_rec 813 0.64 0.64 0.62 0.59 2.2 0.93
## STAI14 813 0.61 0.61 0.58 0.56 1.9 0.82
## STAI15 813 0.77 0.78 0.77 0.74 1.8 0.77
## STAI16_rec 813 0.77 0.77 0.77 0.74 1.9 0.82
## STAI17 813 0.68 0.67 0.66 0.63 2.1 0.86
## STAI18 813 0.66 0.66 0.63 0.61 2.1 0.92
## STAI19_rec 813 0.78 0.78 0.78 0.75 2.2 0.85
## STAI20 813 0.71 0.70 0.68 0.66 2.0 0.90
##
## Non missing response frequency for each item
## 1 2 3 4 miss
## STAI1_rec 0.19 0.51 0.28 0.01 0
## STAI2 0.15 0.49 0.28 0.08 0
## STAI3 0.41 0.45 0.12 0.02 0
## STAI4 0.64 0.27 0.07 0.02 0
## STAI5 0.34 0.47 0.16 0.02 0
## STAI6_rec 0.10 0.43 0.36 0.11 0
## STAI7_rec 0.20 0.44 0.32 0.05 0
## STAI8 0.41 0.44 0.13 0.03 0
## STAI9 0.22 0.41 0.25 0.11 0
## STAI10_rec 0.26 0.45 0.25 0.03 0
## STAI11 0.37 0.39 0.18 0.06 0
## STAI12 0.32 0.45 0.15 0.08 0
## STAI13_rec 0.27 0.37 0.27 0.09 0
## STAI14 0.33 0.45 0.17 0.04 0
## STAI15 0.38 0.46 0.13 0.03 0
## STAI16_rec 0.35 0.41 0.21 0.03 0
## STAI17 0.27 0.47 0.20 0.07 0
## STAI18 0.31 0.40 0.20 0.08 0
## STAI19_rec 0.24 0.40 0.31 0.05 0
## STAI20 0.34 0.39 0.21 0.07 0
psych::alpha(df_IUS)
##
## Reliability analysis
## Call: psych::alpha(x = df_IUS)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.87 0.87 0.88 0.36 6.8 0.0066 2.7 0.83 0.36
##
## lower alpha upper 95% confidence boundaries
## 0.86 0.87 0.89
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## IUS1 0.86 0.86 0.87 0.36 6.2 0.0073 0.0107 0.35
## IUS2 0.87 0.87 0.88 0.38 6.8 0.0067 0.0096 0.40
## IUS3 0.86 0.86 0.86 0.35 6.0 0.0075 0.0099 0.35
## IUS4 0.87 0.87 0.87 0.38 6.6 0.0068 0.0092 0.38
## IUS5 0.86 0.86 0.87 0.36 6.2 0.0072 0.0119 0.35
## IUS6 0.86 0.86 0.86 0.36 6.2 0.0073 0.0087 0.35
## IUS7 0.86 0.86 0.87 0.37 6.4 0.0071 0.0088 0.36
## IUS8 0.86 0.86 0.87 0.36 6.2 0.0073 0.0116 0.36
## IUS9 0.86 0.86 0.87 0.36 6.3 0.0072 0.0117 0.36
## IUS10 0.86 0.86 0.87 0.36 6.1 0.0073 0.0101 0.35
## IUS11 0.86 0.86 0.87 0.37 6.3 0.0071 0.0106 0.36
## IUS12 0.86 0.86 0.86 0.35 6.0 0.0075 0.0107 0.33
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## IUS1 813 0.67 0.67 0.63 0.59 2.8 1.2
## IUS2 813 0.50 0.50 0.42 0.40 3.4 1.3
## IUS3 813 0.72 0.71 0.69 0.64 2.7 1.4
## IUS4 813 0.54 0.55 0.49 0.45 3.2 1.2
## IUS5 813 0.66 0.66 0.61 0.58 2.7 1.3
## IUS6 813 0.67 0.67 0.65 0.59 2.3 1.3
## IUS7 813 0.62 0.62 0.58 0.53 2.7 1.3
## IUS8 813 0.67 0.67 0.63 0.59 2.9 1.4
## IUS9 813 0.65 0.65 0.60 0.56 2.3 1.3
## IUS10 813 0.68 0.68 0.65 0.60 2.2 1.2
## IUS11 813 0.63 0.63 0.59 0.54 3.1 1.3
## IUS12 813 0.72 0.73 0.70 0.66 2.4 1.2
##
## Non missing response frequency for each item
## 1 2 3 4 5 miss
## IUS1 0.21 0.18 0.38 0.12 0.12 0
## IUS2 0.12 0.08 0.34 0.20 0.26 0
## IUS3 0.29 0.16 0.29 0.11 0.14 0
## IUS4 0.13 0.11 0.40 0.16 0.20 0
## IUS5 0.26 0.18 0.33 0.10 0.13 0
## IUS6 0.37 0.20 0.27 0.07 0.08 0
## IUS7 0.25 0.17 0.34 0.12 0.11 0
## IUS8 0.23 0.15 0.32 0.14 0.16 0
## IUS9 0.37 0.19 0.30 0.06 0.09 0
## IUS10 0.38 0.21 0.27 0.06 0.07 0
## IUS11 0.16 0.13 0.37 0.16 0.19 0
## IUS12 0.34 0.17 0.35 0.05 0.09 0
#Preprocessing
data <- preprocess_overall_pruchasing(df = data_sample) #full range scale data
filtered_data <- preprocess_overall_pruchasing_filtered(df = data_sample)
data_food <- preprocess_individual_products(df = data_sample, t = NA, DV = "NonPerishableFood")
data_hygiene <- preprocess_individual_products(df = data_sample, t = NA, DV = "HygieneProducts")
data_freshFood <- preprocess_individual_products(df = data_sample, t = NA, DV = "FreshFood")
#correlations full range scale
cor_data <- tab_corr(data = select(data,
FreqMarch,
QuaMarch,
sex,
age,
education,
householdsize,
SEA,
sem_diff,
Prob_Einkaufen,
IUS,
STAI,
info_verlauf,
risk_self,
risk_loved), triangle = "lower")
cor_data
Â
|
FreqMarch
|
QuaMarch
|
sex
|
age
|
education
|
householdsize
|
SEA
|
sem_diff
|
Prob_Einkaufen
|
IUS
|
STAI
|
info_verlauf
|
risk_self
|
risk_loved
|
FreqMarch
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
QuaMarch
|
-0.309***
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
sex
|
0.107**
|
-0.035
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
age
|
-0.037
|
-0.121***
|
-0.045
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
education
|
-0.058
|
0.095**
|
0.002
|
-0.099**
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
householdsize
|
-0.002
|
0.022
|
-0.047
|
-0.097**
|
0.039
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
SEA
|
-0.014
|
-0.081*
|
-0.113**
|
0.179***
|
0.001
|
-0.016
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
sem_diff
|
-0.184***
|
0.226***
|
-0.150***
|
-0.110**
|
0.116***
|
0.024
|
-0.083*
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Prob_Einkaufen
|
-0.122***
|
0.141***
|
-0.036
|
-0.083*
|
-0.026
|
-0.038
|
-0.088*
|
0.343***
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
IUS
|
0.018
|
0.149***
|
-0.033
|
-0.179***
|
-0.033
|
-0.026
|
-0.136***
|
0.300***
|
0.237***
|
Â
|
Â
|
Â
|
Â
|
Â
|
STAI
|
-0.021
|
0.109**
|
-0.043
|
-0.231***
|
-0.048
|
-0.006
|
-0.224***
|
0.325***
|
0.232***
|
0.605***
|
Â
|
Â
|
Â
|
Â
|
info_verlauf
|
-0.060
|
0.087*
|
-0.006
|
0.269***
|
-0.025
|
0.039
|
0.092**
|
0.241***
|
0.033
|
0.067
|
-0.059
|
Â
|
Â
|
Â
|
risk_self
|
-0.059
|
-0.094**
|
-0.001
|
0.438***
|
-0.119***
|
-0.134***
|
0.010
|
0.048
|
0.106**
|
-0.042
|
0.011
|
0.090*
|
Â
|
Â
|
risk_loved
|
-0.064
|
-0.004
|
-0.078*
|
0.152***
|
-0.033
|
-0.017
|
-0.039
|
0.042
|
0.008
|
-0.022
|
0.019
|
0.024
|
0.372***
|
Â
|
Computed correlation used pearson-method with listwise-deletion.
|
#correlations
cor_data_filtered <- tab_corr(data = select(filtered_data,
FreqMarch,
QuaMarch,
sex,
age,
education,
householdsize,
SEA,
sem_diff,
Prob_Einkaufen,
IUS,
STAI,
info_verlauf,
risk_self,
risk_loved), triangle = "lower")
cor_data_filtered
Â
|
FreqMarch
|
QuaMarch
|
sex
|
age
|
education
|
householdsize
|
SEA
|
sem_diff
|
Prob_Einkaufen
|
IUS
|
STAI
|
info_verlauf
|
risk_self
|
risk_loved
|
FreqMarch
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
QuaMarch
|
-0.607***
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
sex
|
0.129***
|
-0.068
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
age
|
-0.033
|
-0.108**
|
-0.051
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
education
|
-0.127***
|
0.122**
|
0.017
|
-0.103**
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
householdsize
|
-0.018
|
0.038
|
-0.026
|
-0.077*
|
0.035
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
SEA
|
0.007
|
-0.049
|
-0.111**
|
0.159***
|
-0.011
|
-0.013
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
sem_diff
|
-0.319***
|
0.314***
|
-0.156***
|
-0.104**
|
0.086*
|
0.017
|
-0.095*
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
Prob_Einkaufen
|
-0.189***
|
0.203***
|
-0.048
|
-0.068
|
-0.018
|
0.001
|
-0.068
|
0.358***
|
Â
|
Â
|
Â
|
Â
|
Â
|
Â
|
IUS
|
-0.074
|
0.121**
|
-0.051
|
-0.188***
|
-0.038
|
-0.028
|
-0.133***
|
0.322***
|
0.208***
|
Â
|
Â
|
Â
|
Â
|
Â
|
STAI
|
-0.082*
|
0.111**
|
-0.043
|
-0.230***
|
-0.080*
|
-0.036
|
-0.223***
|
0.306***
|
0.218***
|
0.600***
|
Â
|
Â
|
Â
|
Â
|
info_verlauf
|
-0.174***
|
0.146***
|
-0.007
|
0.273***
|
-0.015
|
0.031
|
0.106**
|
0.238***
|
0.025
|
0.063
|
-0.088*
|
Â
|
Â
|
Â
|
risk_self
|
-0.066
|
-0.047
|
0.006
|
0.464***
|
-0.119**
|
-0.125**
|
0.013
|
0.032
|
0.087*
|
-0.063
|
-0.009
|
0.105**
|
Â
|
Â
|
risk_loved
|
-0.069
|
-0.012
|
-0.076*
|
0.147***
|
-0.026
|
0.015
|
-0.023
|
0.028
|
0.001
|
-0.038
|
0.006
|
0.026
|
0.354***
|
Â
|
Computed correlation used pearson-method with listwise-deletion.
|
subdata_Threat <-select(data,
sem_diff,
sex,
age,
education,
householdsize,
SEA,
Prob_Einkaufen,
#risk_self,
#risk_loved,
IUS,
STAI,
info_verlauf)
Model_Threat <- lm(sem_diff ~ ., data = subdata_Threat)
#tab_model(Model_Threat,
# show.intercept = FALSE,
# show.ci = FALSE,
# show.se = FALSE,
# show.stat = TRUE,
# #p.style = "scientific",
# string.est = "b",
# string.se = "SE")
summary(Model_Threat)
##
## Call:
## lm(formula = sem_diff ~ ., data = subdata_Threat)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.88960 -0.58585 0.04263 0.59182 2.32513
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.070195 0.033610 2.089 0.03707 *
## sex -0.326107 0.073128 -4.459 9.39e-06 ***
## age -0.086391 0.032460 -2.661 0.00794 **
## education 0.134131 0.029960 4.477 8.67e-06 ***
## householdsize 0.006096 0.030051 0.203 0.83931
## SEA -0.027315 0.031058 -0.879 0.37939
## Prob_Einkaufen 0.256445 0.030895 8.301 4.37e-16 ***
## IUS 0.074628 0.038074 1.960 0.05033 .
## STAI 0.210519 0.038699 5.440 7.08e-08 ***
## info_verlauf 0.268601 0.031352 8.567 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8467 on 803 degrees of freedom
## Multiple R-squared: 0.291, Adjusted R-squared: 0.283
## F-statistic: 36.62 on 9 and 803 DF, p-value: < 2.2e-16
print_model(df = filtered_data, model = "Baseline", add_variable = "")
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.32
|
3.46
|
0.001
|
-0.19
|
-2.05
|
0.040
|
age
|
-0.05
|
-1.18
|
0.237
|
-0.09
|
-2.38
|
0.018
|
education
|
-0.13
|
-3.49
|
0.001
|
0.11
|
2.95
|
0.003
|
householdsize
|
-0.01
|
-0.35
|
0.730
|
0.02
|
0.64
|
0.526
|
SEA
|
0.03
|
0.71
|
0.477
|
-0.04
|
-1.09
|
0.278
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.036 / 0.029
|
0.032 / 0.025
|
print_model(df = filtered_data, model = "specific", add_variable = "sem_diff")
## Note: Using an external vector in selections is ambiguous.
## i Use `all_of(add_variable)` instead of `add_variable` to silence this message.
## i See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.19
|
2.14
|
0.032
|
-0.07
|
-0.74
|
0.457
|
age
|
-0.07
|
-1.97
|
0.050
|
-0.07
|
-1.78
|
0.076
|
education
|
-0.11
|
-2.99
|
0.003
|
0.09
|
2.44
|
0.015
|
householdsize
|
-0.01
|
-0.35
|
0.728
|
0.02
|
0.65
|
0.516
|
SEA
|
-0.00
|
-0.08
|
0.939
|
-0.01
|
-0.34
|
0.731
|
sem_diff
|
-0.30
|
-8.21
|
<0.001
|
0.29
|
7.86
|
<0.001
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.124 / 0.116
|
0.114 / 0.106
|
print_model(df = filtered_data, model = "specific", add_variable = "IUS")
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.30
|
3.30
|
0.001
|
-0.17
|
-1.86
|
0.064
|
age
|
-0.06
|
-1.55
|
0.121
|
-0.07
|
-1.86
|
0.063
|
education
|
-0.14
|
-3.62
|
<0.001
|
0.12
|
3.11
|
0.002
|
householdsize
|
-0.02
|
-0.44
|
0.659
|
0.03
|
0.76
|
0.448
|
SEA
|
0.02
|
0.47
|
0.640
|
-0.03
|
-0.78
|
0.438
|
IUS
|
-0.08
|
-2.12
|
0.035
|
0.11
|
2.70
|
0.007
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.042 / 0.034
|
0.043 / 0.034
|
print_model(df = filtered_data, model = "specific", add_variable = "STAI")
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.30
|
3.26
|
0.001
|
-0.17
|
-1.86
|
0.063
|
age
|
-0.07
|
-1.72
|
0.085
|
-0.07
|
-1.80
|
0.072
|
education
|
-0.14
|
-3.76
|
<0.001
|
0.12
|
3.20
|
0.001
|
householdsize
|
-0.02
|
-0.49
|
0.623
|
0.03
|
0.77
|
0.440
|
SEA
|
0.01
|
0.18
|
0.859
|
-0.02
|
-0.58
|
0.565
|
STAI
|
-0.10
|
-2.57
|
0.010
|
0.10
|
2.42
|
0.016
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.045 / 0.037
|
0.040 / 0.032
|
print_model(df = filtered_data, model = "specific", add_variable = "risk_self")
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.32
|
3.51
|
<0.001
|
-0.19
|
-2.06
|
0.040
|
age
|
-0.01
|
-0.18
|
0.858
|
-0.10
|
-2.29
|
0.022
|
education
|
-0.14
|
-3.64
|
<0.001
|
0.11
|
2.97
|
0.003
|
householdsize
|
-0.02
|
-0.53
|
0.594
|
0.03
|
0.67
|
0.503
|
SEA
|
0.02
|
0.58
|
0.559
|
-0.04
|
-1.06
|
0.291
|
risk_self
|
-0.17
|
-1.92
|
0.056
|
0.03
|
0.39
|
0.700
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.041 / 0.033
|
0.032 / 0.024
|
print_model(df = filtered_data, model = "specific", add_variable = "risk_loved")
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.31
|
3.35
|
0.001
|
-0.19
|
-2.05
|
0.041
|
age
|
-0.04
|
-0.95
|
0.341
|
-0.09
|
-2.34
|
0.020
|
education
|
-0.13
|
-3.51
|
<0.001
|
0.11
|
2.95
|
0.003
|
householdsize
|
-0.01
|
-0.31
|
0.757
|
0.02
|
0.64
|
0.525
|
SEA
|
0.02
|
0.63
|
0.528
|
-0.04
|
-1.09
|
0.277
|
risk_loved
|
-0.11
|
-1.46
|
0.145
|
-0.01
|
-0.07
|
0.943
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.039 / 0.030
|
0.032 / 0.023
|
print_model(df = filtered_data, model = "specific", add_variable = "info_verlauf")
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.32
|
3.58
|
<0.001
|
-0.20
|
-2.17
|
0.031
|
age
|
0.00
|
0.06
|
0.952
|
-0.14
|
-3.64
|
<0.001
|
education
|
-0.13
|
-3.49
|
0.001
|
0.11
|
2.94
|
0.003
|
householdsize
|
-0.00
|
-0.10
|
0.920
|
0.01
|
0.38
|
0.705
|
SEA
|
0.04
|
1.03
|
0.303
|
-0.05
|
-1.43
|
0.152
|
info_verlauf
|
-0.18
|
-4.62
|
<0.001
|
0.19
|
4.93
|
<0.001
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.066 / 0.057
|
0.066 / 0.058
|
print_model(df = filtered_data, model = "specific", add_variable = "Prob_Einkaufen")
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.29
|
3.23
|
0.001
|
-0.16
|
-1.79
|
0.074
|
age
|
-0.06
|
-1.52
|
0.130
|
-0.08
|
-2.10
|
0.036
|
education
|
-0.14
|
-3.68
|
<0.001
|
0.12
|
3.13
|
0.002
|
householdsize
|
-0.01
|
-0.38
|
0.705
|
0.03
|
0.68
|
0.500
|
SEA
|
0.02
|
0.40
|
0.686
|
-0.03
|
-0.78
|
0.438
|
Prob_Einkaufen
|
-0.19
|
-5.03
|
<0.001
|
0.19
|
5.20
|
<0.001
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.071 / 0.063
|
0.070 / 0.061
|
subdata_Freq <-select(filtered_data, FreqMarch,
sex,
age,
education,
householdsize,
SEA,
sem_diff,
#risk_self,
#risk_loved,
IUS,
STAI,
info_verlauf,
Prob_Einkaufen)
subdata_Qua <-select(filtered_data, QuaMarch,
sex,
age,
education,
householdsize,
SEA,
sem_diff,
#risk_self,
#risk_loved,
IUS,
STAI,
info_verlauf,
Prob_Einkaufen)
Model_Freq <- lm(FreqMarch ~ ., data = subdata_Freq)
Model_Qua <- lm(QuaMarch ~ ., data = subdata_Qua)
summary(Model_Freq)
##
## Call:
## lm(formula = FreqMarch ~ ., data = subdata_Freq)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.5050 -0.6950 0.1895 0.7284 1.8617
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.046147 0.040839 -1.130 0.25889
## sex 0.208586 0.088555 2.355 0.01879 *
## age -0.044080 0.039247 -1.123 0.26178
## education -0.116909 0.036560 -3.198 0.00145 **
## householdsize -0.007608 0.036117 -0.211 0.83322
## SEA 0.001703 0.037450 0.045 0.96375
## sem_diff -0.239725 0.042787 -5.603 3.09e-08 ***
## IUS 0.047021 0.046100 1.020 0.30811
## STAI -0.039813 0.047347 -0.841 0.40071
## info_verlauf -0.110284 0.039673 -2.780 0.00559 **
## Prob_Einkaufen -0.101949 0.038826 -2.626 0.00884 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.933 on 667 degrees of freedom
## Multiple R-squared: 0.1423, Adjusted R-squared: 0.1295
## F-statistic: 11.07 on 10 and 667 DF, p-value: < 2.2e-16
summary(Model_Qua)
##
## Call:
## lm(formula = QuaMarch ~ ., data = subdata_Qua)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.6772 -0.6880 -0.1595 0.4859 2.8900
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.019108 0.040986 0.466 0.64122
## sex -0.086368 0.088875 -0.972 0.33150
## age -0.096995 0.039389 -2.463 0.01405 *
## education 0.098189 0.036693 2.676 0.00763 **
## householdsize 0.018535 0.036247 0.511 0.60928
## SEA -0.019097 0.037586 -0.508 0.61155
## sem_diff 0.215460 0.042941 5.018 6.72e-07 ***
## IUS -0.007187 0.046267 -0.155 0.87661
## STAI 0.015525 0.047518 0.327 0.74399
## info_verlauf 0.123330 0.039817 3.097 0.00203 **
## Prob_Einkaufen 0.113084 0.038966 2.902 0.00383 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9364 on 667 degrees of freedom
## Multiple R-squared: 0.1361, Adjusted R-squared: 0.1232
## F-statistic: 10.51 on 10 and 667 DF, p-value: < 2.2e-16
tab_model(Model_Freq, Model_Qua,
show.intercept = FALSE,
show.ci = FALSE,
show.se = FALSE,
show.stat = TRUE,
#p.style = "scientific",
string.est = "b",
string.se = "SE")
Â
|
FreqMarch
|
QuaMarch
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.21
|
2.36
|
0.019
|
-0.09
|
-0.97
|
0.332
|
age
|
-0.04
|
-1.12
|
0.262
|
-0.10
|
-2.46
|
0.014
|
education
|
-0.12
|
-3.20
|
0.001
|
0.10
|
2.68
|
0.008
|
householdsize
|
-0.01
|
-0.21
|
0.833
|
0.02
|
0.51
|
0.609
|
SEA
|
0.00
|
0.05
|
0.964
|
-0.02
|
-0.51
|
0.612
|
sem_diff
|
-0.24
|
-5.60
|
<0.001
|
0.22
|
5.02
|
<0.001
|
IUS
|
0.05
|
1.02
|
0.308
|
-0.01
|
-0.16
|
0.877
|
STAI
|
-0.04
|
-0.84
|
0.401
|
0.02
|
0.33
|
0.744
|
info_verlauf
|
-0.11
|
-2.78
|
0.006
|
0.12
|
3.10
|
0.002
|
Prob_Einkaufen
|
-0.10
|
-2.63
|
0.009
|
0.11
|
2.90
|
0.004
|
Observations
|
678
|
678
|
R2 / R2 adjusted
|
0.142 / 0.129
|
0.136 / 0.123
|
model_food <- lm(NonPerishableFood ~ sex
+ age
+ education
+ householdsize
+ SEA
+ risk_self
+ risk_loved
+ info_verlauf
+ sem_diff
+ Prob_Einkaufen
+ IUS
,data = data_food)
model_hygiene <- lm(HygieneProducts ~ sex
+ age
+ education
+ householdsize
+ SEA
+ risk_self
+ risk_loved
+ info_verlauf
+ sem_diff
+ Prob_Einkaufen
+ IUS
,data = data_hygiene)
model_freshFood <- lm(FreshFood ~ sex
+ age
+ education
+ householdsize
+ SEA
+ risk_self
+ risk_loved
+ info_verlauf
+ sem_diff
+ Prob_Einkaufen
+ IUS
,data = data_freshFood)
tab_model(model_food, model_hygiene, model_freshFood,
show.intercept = FALSE,
show.ci = FALSE,
show.se = FALSE,
show.stat = TRUE,
#p.style = "scientific",
string.est = "b",
string.se = "SE")
Â
|
NonPerishableFood
|
HygieneProducts
|
FreshFood
|
Predictors
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
b
|
Statistic
|
p
|
sex
|
0.04
|
0.48
|
0.631
|
0.11
|
1.28
|
0.199
|
0.13
|
1.44
|
0.149
|
age
|
0.09
|
2.30
|
0.022
|
-0.01
|
-0.30
|
0.762
|
-0.08
|
-1.73
|
0.084
|
education
|
0.05
|
1.37
|
0.172
|
0.06
|
1.82
|
0.069
|
-0.00
|
-0.02
|
0.982
|
householdsize
|
0.02
|
0.64
|
0.522
|
-0.02
|
-0.52
|
0.603
|
-0.02
|
-0.52
|
0.607
|
SEA
|
-0.08
|
-2.41
|
0.016
|
-0.06
|
-1.63
|
0.103
|
-0.03
|
-0.72
|
0.470
|
risk_self
|
0.02
|
0.29
|
0.769
|
-0.09
|
-1.07
|
0.284
|
0.02
|
0.21
|
0.836
|
risk_loved
|
0.17
|
2.33
|
0.020
|
0.09
|
1.17
|
0.241
|
0.12
|
1.54
|
0.125
|
info_verlauf
|
0.11
|
3.00
|
0.003
|
0.13
|
3.47
|
0.001
|
0.05
|
1.30
|
0.195
|
sem_diff
|
0.21
|
5.56
|
<0.001
|
0.17
|
4.32
|
<0.001
|
0.10
|
2.29
|
0.023
|
Prob_Einkaufen
|
0.11
|
2.94
|
0.003
|
0.06
|
1.77
|
0.077
|
0.02
|
0.51
|
0.609
|
IUS
|
0.10
|
2.77
|
0.006
|
0.14
|
3.83
|
<0.001
|
0.09
|
2.21
|
0.028
|
Observations
|
789
|
801
|
749
|
R2 / R2 adjusted
|
0.154 / 0.142
|
0.121 / 0.109
|
0.044 / 0.029
|
summary(model_food)
##
## Call:
## lm(formula = NonPerishableFood ~ sex + age + education + householdsize +
## SEA + risk_self + risk_loved + info_verlauf + sem_diff +
## Prob_Einkaufen + IUS, data = data_food)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.4819 -0.6014 -0.2523 0.2938 4.5515
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.10468 0.05491 -1.906 0.05698 .
## sex 0.03962 0.08238 0.481 0.63066
## age 0.09239 0.04012 2.303 0.02157 *
## education 0.04623 0.03385 1.366 0.17242
## householdsize 0.02127 0.03322 0.640 0.52209
## SEA -0.08240 0.03421 -2.409 0.01625 *
## risk_self 0.02402 0.08163 0.294 0.76860
## risk_loved 0.16717 0.07160 2.335 0.01981 *
## info_verlauf 0.10814 0.03602 3.002 0.00277 **
## sem_diff 0.21415 0.03852 5.559 3.73e-08 ***
## Prob_Einkaufen 0.10538 0.03583 2.941 0.00337 **
## IUS 0.09832 0.03545 2.773 0.00568 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9265 on 777 degrees of freedom
## Multiple R-squared: 0.1536, Adjusted R-squared: 0.1416
## F-statistic: 12.82 on 11 and 777 DF, p-value: < 2.2e-16
summary(model_hygiene)
##
## Call:
## lm(formula = HygieneProducts ~ sex + age + education + householdsize +
## SEA + risk_self + risk_loved + info_verlauf + sem_diff +
## Prob_Einkaufen + IUS, data = data_hygiene)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.5650 -0.5980 -0.2450 0.3077 4.6122
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.03145 0.05577 -0.564 0.572975
## sex 0.10730 0.08354 1.284 0.199378
## age -0.01229 0.04063 -0.302 0.762427
## education 0.06274 0.03443 1.822 0.068808 .
## householdsize -0.01761 0.03380 -0.521 0.602507
## SEA -0.05646 0.03460 -1.632 0.103092
## risk_self -0.08835 0.08241 -1.072 0.284031
## risk_loved 0.08503 0.07249 1.173 0.241129
## info_verlauf 0.12644 0.03648 3.466 0.000557 ***
## sem_diff 0.16968 0.03924 4.324 1.73e-05 ***
## Prob_Einkaufen 0.06454 0.03649 1.769 0.077312 .
## IUS 0.13843 0.03615 3.830 0.000139 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9441 on 789 degrees of freedom
## Multiple R-squared: 0.1209, Adjusted R-squared: 0.1086
## F-statistic: 9.861 on 11 and 789 DF, p-value: < 2.2e-16
summary(model_freshFood)
##
## Call:
## lm(formula = FreshFood ~ sex + age + education + householdsize +
## SEA + risk_self + risk_loved + info_verlauf + sem_diff +
## Prob_Einkaufen + IUS, data = data_freshFood)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.0331 -0.5156 -0.3702 -0.0950 4.6539
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.0910427 0.0600216 -1.517 0.1297
## sex 0.1296775 0.0898013 1.444 0.1492
## age -0.0758529 0.0438825 -1.729 0.0843 .
## education -0.0008191 0.0367720 -0.022 0.9822
## householdsize -0.0185583 0.0360158 -0.515 0.6065
## SEA -0.0274428 0.0379964 -0.722 0.4704
## risk_self 0.0183648 0.0884437 0.208 0.8356
## risk_loved 0.1198622 0.0780488 1.536 0.1250
## info_verlauf 0.0509697 0.0393037 1.297 0.1951
## sem_diff 0.0961733 0.0420819 2.285 0.0226 *
## Prob_Einkaufen 0.0200107 0.0390751 0.512 0.6087
## IUS 0.0853787 0.0386950 2.206 0.0277 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9852 on 737 degrees of freedom
## Multiple R-squared: 0.04366, Adjusted R-squared: 0.02938
## F-statistic: 3.059 on 11 and 737 DF, p-value: 0.000507
group_fre <- c()
group_fre[data_sample$FreqMarch < 0] <- "less"
group_fre[data_sample$FreqMarch == 0] <- "same"
group_fre[data_sample$FreqMarch > 0] <- "more"
group_qua <- c()
group_qua[data_sample$QuaMarch < 0] <- "less"
group_qua[data_sample$QuaMarch == 0] <- "same"
group_qua[data_sample$QuaMarch > 0] <- "more"
df_gr <- data.frame(group_fre, group_qua)
df_gr$group_fre <- factor(df_gr$group_fre, levels = c("less", "same", "more"))
df_gr$group_qua <- factor(df_gr$group_qua, levels = c("less", "same", "more"))
Fig1 <- ggplot(data = df_gr, aes(x=group_fre, label = scales::percent(prop.table(stat(count)), accuracy = 0.1L)))+
geom_bar(fill = "#FF0000")+
ylab("Number of Observations")+
xlab("Change in Purchasing Frequency")+
geom_text(stat = 'count',
position = position_dodge(.9),
vjust = -0.5,
size = 3)
Fig2 <- ggplot(data = df_gr, aes(x=group_qua,label = scales::percent(prop.table(stat(count)), accuracy = 0.1L)))+
geom_bar(fill = "#FF0000")+
ylab("Number of Observations")+
xlab("Change in Purchasing Quantity")+
geom_text(stat = 'count',
position = position_dodge(.9),
vjust = -0.5,
size = 3)
Fig1
![](data:image/png;base64,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)
Fig2
![](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAABUAAAAPACAMAAADDuCPrAAAB+FBMVEUAAAAAADUAADoAAF4AAGYANTUANV4ANYQAOjoAOmYAOpAAXqgAZrYzMzM1AAA1NQA1NV41XoQ1Xqg1hKg1hMk6AAA6OgA6Ojo6OmY6OpA6ZmY6ZpA6ZrY6kJA6kLY6kNtNTU1NTW5NTY5Nbm5Nbo5NbqtNjsheAABeNQBeNTVehIRehKheqMleqOtmAABmADpmOgBmOjpmOpBmZjpmZmZmZpBmkGZmkJBmkLZmkNtmtttmtv9uTU1ubk1ubm5ubo5ujo5ujqtujshuq8huq+SENQCEXjWEqMmEyeuOTU2Obk2Obm6Oq6uOq8iOq+SOyOSOyP+QOgCQOjqQZjqQZmaQZpCQkDqQkGaQkLaQtraQttuQ2/+oXgCoXjWohDWoqISoyeuo6+urbk2rbm6rjm6rq46rq8iryOSr5P+2ZgC2Zjq2Zma2kDq2kGa2kJC2tpC2tra2ttu229u22/+2///Ijk3Ijm7Iq27Iq6vI5OTI5P/I///JhDXJqF7JqITJyajJ6+vbkDrbkGbbtmbbtpDbtrbbttvb27bb29vb2//b///kq27kq47kyI7kyKvkyMjk5P/k///rqF7ryYTryajrycnr66jr68nr6+v/AAD/tmb/tpD/yI7/25D/27b/29v/5Kv/5Mj/5OT//7b//8j//9v//+T///+ogE1eAAAACXBIWXMAAB2HAAAdhwGP5fFlAAAgAElEQVR4nO2d+4Nc5Xnfj9eOEUImcZJFjqBmRLg4NxkXkTigNsUFNzgXW4E0NsYtUWntpSbmYjkNkJQ1uNxqtwJqVFUSKGI0/2bnvHPZmdlZzbvPnpfvnO/5fH5AM7Mr7Xm++zwfzpn3nDPVAAAAQlTqDQAAaCsIFAAgCAIFAAiCQAEAgiBQAIAgCBQAIAgCBQAIgkABAIIgUACAIAgUACAIAgUACIJAAQCCIFAAgCCfkEDPd4GOlLmuEL+WjuSPQIvRkTLXFeLX0pH8EWgxOlLmukL8WjqSPwItRkfKXFeIX0tH8kegxehImesK8WvpSP4ItBgdKXNdIX4tHckfgRajI2WuK8SvpSP5I9BidKTMdYX4tXQkfwRajI6Uua4Qv5aO5I9Ai9GRMtcV4tfSkfwRaDE6Uua6QvxaOpI/Ai1GR8pcV4hfS0fyR6DF6EiZ6wrxa+lI/gi0GB0pc10hfi0dyR+BFqMjZa4rxP+J8N7NN7xa//nLe6rEp/92/IVh/u//TlV96vfSl8+//6XqU78xevjk6G94gECL0ZEy1xXi/yT44JFqpMP3bt4l0H8cvfLZ+pVf3nPDq+/dfNPoO/9AtbUFQKDF6EiZ6wrxfxK8Uo0F+m510/xXBpdurm546vwH30vf8Eo1tOaTn/rT82Y7oAi0HB0pc10h/k+Aer9z5MMnq9+c/9Jga/qViTuTRd+tnHZAEWg5OlLmukL85RkewP/u6D3QDx5Je5cz9E+PTZn2TacC/eARqx1QBFqOjpS5rhB/eZ6sbhovIv3ynhv+45eq6teemn6tf3qs1PQdSaD1vqjZDigCLUdHyhRznVXguZdcV4GVvFvVK0Mpzcka0o4e5wU6eQ/UbQcUgZajI2Vqud4q8OxLtqvAQn55z1CRY4G+W1W//ur5//P1anokP9gaLys9Wf+GJvm77YAi0HJ0pEwt11kFHsy+ZLsKLCStG40F+srUlpO1pHoVfujUD743+j9YfQTwe6/WO6AffL2qPm/zK0CgxehImVKutwo8mH3JdhVYxysp+clbKGPerSZPB4NXRgcAvzvzHcPwP3jk00+9f4/N/8QQaDE6UqaS660CD/qzL9muAst47+YU74JA37t58h7KsP3/KS0rzXxHHX56I+UVm/+LIdBidKRMJddbBR5cnX3JdhVYxnj/cu5951mfTtt/5p2UOvxX6uOCdxePF1oLAi1GR8oUct1V4Po9uJ2XbFeBZcwJdHgoMHPSZ2La/jvvpKTwx++j3LT477UUBFqMjpSpY8Uq8IXZl2xXgdWM839yZMSpSIf5b4/EOT7Ur0nhsweKQPPoSJk6VqwCb8+95LoKrGbnCGD4f6thyNMd/OH/wD71++fP/9PN8zug53kPFIHm0ZEyZaxcBd71kuMqsJpJ/uMj+k8/dX680zkYfG/00jTq0d5/vdxnlD8CLUZHylSRsQq8+JLjKrCaaf7p5p+/Pr6oIQn0/D/+alX9yu9PvnPy9rPXEQACLUZHylSRvQo8o1jDVeC1pSPtj0CL0ZEyVaxaBZ6/G1DCcRV4belI+yPQYnSkTDF7rwJv7XrJcRV4belI+yPQYnSkTDF7rwJfWnzJchV4belI+yPQYnSkTDHXWQWefanGchV4belI+yPQYnSkTDHXWwWeeem86yrw2tKR9kegxehImesK8WvpSP4ItBgdKXNdIX4tHckfgRajI2WuK47x/18oQ/xXgkCL0ZEy1xXH+NWesSX+K0GgxehImeuKY/xqz9gS/5Ug0GJ0pMx1xTF+tWdsif9KEGgxOlLmuuIYv9oztsR/JQi0GB0pc11xjF/tGVvivxIEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCdCI1BoKepGt4X4pTQ2IOyBNodjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEINBiOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnogmB/uIbvd7xvzg3enLt6VO93kMvLXmCQNuPutFtIX4p8YloQKDv9BJ3vlw/+fjR9OSO3U8QqAHqRreF+KXEJ+LgAv3o1PFvDgZXvtG7u352tnfXS4MrZ3p3nVt8gkANUDe6LcQvJT4RBxfo2d7DI4/WO5qj/w53PY//YOEJAnVA3ei2EL+U+EQcXKBjPn60tuU7o/3Q4Z8PLzxBoA6oG90W4pcSn4jGBPrRqfpA/WzvW+nZxdqdc08QqAPqRreF+KXEJ6Ipgf78VG3La2fGR+u1TueejL7rc2OiPwXEqBvdFuKX0tiABAV6ttc7/v0BArVH3ei2EL+UxgYkJtBr/+lfn+od/8s5gd7x8tyT+b8Q32NuEY5lqhvdFuKXEp+IRgRa84v6GH7VHigCbTnqRreF+KXEJ6IxgQ4u9haciUDVW9A86ka3hfilxCeiOYEmTbIKv4NjmepGt4X4pcQn4sACvXZmrMkk0Mkpn+PzQGeeIFAH1I1uC/FLiU/EgQU63M+8e+dPrkTawbFMdaPbQvxS4hNxcIF+dKr3tXODaz/p1Zoc7o/eOb38fe4JAnVA3ei2EL+U+EQcXKCDi6O7MR1PR/JXZm/AdIW7MZmhbnRbiF9KfCIaEOjgyp8N9Tm56+eVp4fKfOjckicItP2oG90W4pcSn4gmBLp/4tvbIhzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IvYn0P7PnnsegWbiWKa60W0hfinxicgV6OW/fnswuHpvVVU3fheBZuFYprrRbSF+KfGJyBTodvWZHw8GW1VN/QiBrsaxTHWj20L8UuITkSfQC0mblzarw29fPlHdjkBzcCxT3ei2EL+U+ETkCXRraM5aoxvfrf9bP0agK3EsU93othC/lPhEZAm0f7o251ijV08c/Bg+vr0twrFMdaPbQvxS4hORJdCRM6+eqI4MEGg2jmWqG90W4pcSn4h9CPTSZvUgAs3HsUx1o9tC/FLiE5El0NEh/HZ6C5T3QHNxLFPd6LYQv5T4RGQJdLBVHamX32tzsgqfi2OZ6ka3hfilxCciT6AX0gmgwyP4/mPVaD8Uga7EsUx1o9tC/FLiE5En0OHh+5AjaSFp48ED+9PQLEtwLFPd6LYQv5T4RGQKdHD58fufGP5x9cu3vnhwfxqaZQmOZaob3RbilxKfiFyBNkt8e1uEY5nqRreF+KXEJwKBFsOxTHWj20L8UuITsQ+BvjnhLQSag2OZ6ka3hfilxCciU6CXH6t24ET6LBzLVDe6LcQvJT4ReQK9eqJCoPvFsUx1o9tC/FLiE5En0O2quvGPnpvwPFci5eBYprrRbSF+KfGJyBJo/3S6jUhzxLe3RTiWqW50W4hfSnwisgR69UQDVx8hUAPUjW4L8UuJT0SmQBv4GA8EaoC60W0hfinxicgS6PiGygh0XziWqW50W4hfSnwisgQ62G7gDkxggLrRbSF+KY0NyB4CHe6CPtDYzxg47potwbFMdaPbQvxS4hORJdD+4/dV1caxk2Pu5zSmHBzLVDe6LcQvJT4RWQKdP4+eE+nzcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IrIE2jjx7W0RjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCciW6AfPnu0qjaOfvXgNwNFoO1F3ei2EL+U+ETkCnR7uoTUxCn18e1tEY5lqhvdFuKXEp+ITIHW/jx07OR9tzRj0Pj2tgjHMtWNbgvxS4lPRJ5AL21Wh0efxnn5NJ8Ln4ljmepGt4X4pcQnIk+gW9XhydVHjdwbNL69LcKxTHWj20L8UuITkSXQubsxXdo8zKWcOTiWqW50W4hfSnwisgQ6dz/QJm4OGt/eFuFYprrRbSF+KfGJQKDFcCxT3ei2EL+U+ERkCbR/unpw+uRCxSF8Fo5lqhvdFuKXEp+ILIGyiBTBsUx1o9tC/FLiE5En0Eub1Y0vpEdv3MtpTJk4lqludFuIX0p8IvIEOroQ6ejRow1dihTf3hbhWKa60W0hfinxicgU6OC1zfGVnI18tkd8e1uEY5nqRreF+KXEJyJXoIP+6/cN90CPPXHgBSQE2mLUjW4L8UuJT0S2QBslvr0twrFMdaPbQvxS4hOBQIvhWKa60W0hfinxiVgh0P7j9WdwDv87C5/KmYVjmepGt4X4pcQnYoVAr56oP0KOD5WL4FimutFtIX4p8YlAoMVwLFPd6LYQv5T4RKwQaCHi29siHMtUN7otxC8lPhEItBiOZaob3RbilxKfiCyB9h+fWTe6dN9vsYiUg2OZ6ka3hfilxCciS6Dczi6CY5nqRreF+KXEJ2L/Ar20iUCzcCxT3ei2EL+U+ESsEujCAnyC+4Fm4VimutFtIX4p8YlYJdDBhd0CfXDXNyHQJTiWqW50W4hfSnwiVgq0/19Pnrxvc+PY9DqkP3rhwP40NMsSHMtUN7otxC8lPhErBVrTxLoRAnVA3ei2EL+U+ERkCXTuNCYEmoljmepGt4X4pcQnIkugjRPf3hbhWKa60W0hfinxicgXaP/NMa//S05jysGxTHWj20L8UuITkSnQy49xM5H94limutFtIX4p8YnIE+j82aCHEWgOjmWqG90W4pcSn4g8gW5X1cax+mSm+zYb+VS5+Pa2CMcy1Y1uC/FLiU9ElkD7p+urj4b/fbB26cEvRDI0yxIcy1Q3ui3ELyU+EVkCvXpi47uD2p31R8JvcSVSHo5lqhvdFuKXEp+ITIGmdaML1ZHpfxHoShzLVDe6LcQvJT4R+xNoffR+9QQ3E8nCsUx1o9tC/FLiE5El0P7pdAg/upEd9wPNxLFMdaPbQvxS4hORJdDBVnr3c/RWKPcDzcSxTHWj20L8UuITkSfQS5vVrS/Wy/C31zLlED4LxzLVjW4L8UuJT0SeQIfWrK8/ulBVG5tV2htFoCtxLFPd6LYQv5T4RGQKdPDT+sC9v9XMDekNzbIExzLVjW4L8UuJT0SuQAeD/zH0Zv+1o0e/2sCd7eLb2yIcy1Q3ui3ELyU+EfkCbZL49rYIxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnIkugV0/c+iIC3S+OZaob3RbilxKfiEyBVtVGE+99ItCWo250W4hfSnwisgTaf2azXn7/whMIdB84lqludFuIX0p8IrIEOuT1r9QK3WjoUD6+vS3CsUx1o9tC/FLiE5Er0PoUpltqhx56gNOY8nAsU93othC/lPhE5At0yIfPpkP53+JSzhwcy1Q3ui3ELyU+EfsS6GD06XLcTCQLxzLVjW4L8UuJT8T+BNp/9hY+lTMXxzLVjW4L8UuJT8Q+BNp//d6m3gSNb2+LcCxT3ei2EL+U+ERkC7TRZfj49rYIxzLVjW4L8UuJT0SeQOt3Pofc2NSJoPHtbRGOZaob3RbilxKfiCyB1lciNXopUnx7W4RjmepGt4X4pcQnIlegt77QmD0RaHtRN7otxC8lPhFZAu0/3+B18Ai0xagb3RbilxKfiCyBDra4G9P+cSxT3ei2EL+U+ERkCXR4CP8gAt0vjmWqG90W4pcSn4hMgR783HkE6oC60W0hfinxicgSaP90/YHwCHR/OJapbnRbiF9KfCKyBDrYrg43+iZofHtbhGOZ6ka3hfilxCciT6Af/rCqDh07OeZ+7saUg2OZ6ka3hfilxCciS6DpRPoduJlIFo5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJT0SWQBsnvr0twrFMdaPbQvxS4hOBQIvhWKa60W0hfinxidiPQPtvIdB94FimutFtIX4p8YnIFmh9P9DP/Pjqlxu5J1N8e1uEY5nqRreF+KXEJyJToP1nRqtHV09UNzZwUVJ8e1uEY5nqRreF+KXEJyJToFtVdeMfbn7mx/2/qqrDfKRHFo5lqhvdFuKXEp+IPIFeqKoHxlfEv7bZwI1F4tvbIhzLVDe6LcQvJT4ReQLdqm6f3lJkuzqCQHNwLFPd6LYQv5T4RGQJdHQzkbFAL21yIn0WjmWqG90W4pcSn4gsgY7UORZoE/e2i29vi3AsU93othC/lPhEINBiOJapbnRbiF9KfCKyBNo/XS8cjc15oYFl+Pj2tgjHMtWNbgvxS4lPRJZARwtHI4EOZcoiUhaOZaob3RbilxKfiDyBXtqsbns7CfTyvVUDd6ePb2+LcCxT3ei2EL+U+ETkCXS4C1pVRzc3jt0y/PP2A/vT0CxLcCxT3ei2EL+U+ERkCnTw083J3UAb8KehWZbgWKa60W0hfinxicgV6ODDZ48O7XmomQ+Ij29vi3AsU93othC/lPhEZAu0UeLb2yIcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp+I/Qm0/7PnnkegmTiWqW50W4hfSnwicgV6+a/fHgyu3ltV1Y0HP4vJ0CxLcCxT3ei2EL+U+ERkCnQ7fRTn1vIP5fzFn/V6xx96afTk2tOner2lTxBo+1E3ui3ELyU+EXkCvZC0eWmzOvz25ROLJzL9pJc4/oP6ycePpid3vLzrCQI1QN3othC/lPhE5Al0K13/fiFdhLR4LfzF3vFvDgZXzow8ebZ310v1k7vOLT5BoAaoG90W4pcSn4gsgY7uBzrW6MLdmK6d6X1rvLc5/POjU0mjHz9a74/OPUGgDqgb3RbilxKfiCyBTm5nl24jsiDQjx8dH6Gf7T08GLzTuzs9eWfXEwTqgLrRbSF+KfGJ2IdAL40+DWmv+4EmgZ4d7Y4Oj+vvXniCQB1QN7otxC8lPhFZAh0dwm+P7sO0x/1A04H6tTPjo/WPTt11bu7J6Js+N2bpT4H1R93othC/lMYGZM9FpCP18nttzt2r8CPS8ToCNUfd6LYQv5TGBmTv05hqHhz0H6uW3w/0YjqNacaZd7w892T+u+N7zC3CsUx1o9tC/FLiE5En0HQ/0NFd6auNZR8Lf/HU8fr9zlV7oAi05agb3RbilxKfiEyBDi4/fv8Twz+ufnnp/ezeGZ9Gj0BncCxT3ei2EL+U+ETkCvS6/KQ3OdOTVfgdHMtUN7otxC8lPhH7EWj/zaUvXzvbu3PyHufklM/xeaAPz72IQFuPutFtIX4p8YnIFujrX0nrSEuO4M/OXKrJlUg7OJapbnRbiF9KfCIyBdo/PflIpOq2hbNA35m91P3amd6d08vf554gUAfUjW4L8UuJT0SeQGt/bhz79nN/d99oMX6G8R2Xaup3Oq/M3oDpCndjMtAUAqsAACAASURBVEPd6LYQv5T4ROQJ9ML0wzj7P6yqufOYLvbmBDq48vTw0UPjXc65Jwi0/agb3RbilxKfiCyBDndAd64+2lrYBY0Q394W4VimutFtIX4p8YnIEujVEzNXH13aXH4zEQS6gGOZ6ka3hfilxCciU6AzztzrbkwIdAHHMtWNbgvxS4lPRJZAxzdUHnFpc+ndmBDoIo5lqhvdFuKXEp+ILIEOtmfe99xefjcmBLqIY5nqRreF+KXEJyJPoFdPVF98e+LPgx/BG5plCY5lqhvdFuKXEp+IFQLtP34ycd/oPNAfndysqmP3cwifg2OZ6ka3hfilxCdihUCHu567YREpC8cy1Y1uC/FLiU8EAi2GY5nqRreF+KXEJ2KFQAsR394W4VimutFtIX4p8YlAoMVwLFPd6LYQv5T4RCDQYjiWqW50W4hfSnwicgXaf/PNNw+8+I5A24660W0hfinxicgS6OXHNqu9bqeMQPfCsUx1o9tC/FLiE5Eh0PqjjKc0o9D49rYIxzLVjW4L8UuJT8RqgV5Ke59HT548mT7UY+mnwiPQJTiWqW50W4hfSnwiVgq0PhN044Hx25+vbTZxFqijWZbgWKa60W0hfinxiVgp0K2hMncO2y+fWPxIDwS6F45lqhvdFuKXEp+IVQIdHsDPHbQvPkege+JYprrRbSF+KfGJWCXQ7cU9zq2K29nl4VimutFtIX4p8YlYIdD64zjnPkOu/nw5bqichWOZ6ka3hfilxCdihUCvnlhcNNr9CgJdjmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJWCHQ/unFRfdLm7wHmodjmepGt4X4pcQnYoVAdy+671qWR6B74FimutFtIX4p8YlYJdALC5ce1RcmPbj4TQh0GY5lqhvdFuKXEp+IVQKtz2Oa2ePsbzVxFpOhWZbgWKa60W0hfinxiVgl0HQvkdsmyrx8upm7icS3t0U4lqludFuIX0p8IlYKtH7Ts6qOfeeFN9/8Ubqv3QMH96ehWZbgWKa60W0hfinxiVgt0HQLpikbTfjT0CxLcCxT3ei2EL+U+ERkCHTQf2ai0I2vNvOxHvHtbRGOZaob3RbilxKfiByBDvnwub85ef933mrEngi0vagb3RbilxKfiEyBNkx8e1uEY5nqRreF+KXEJwKBFsOxTHWj20L8UuITgUCL4VimutFtIX4p8YlAoMVwLFPd6LYQv5T4RCDQYjiWqW50W4hfSnwiEGgxHMtUN7otxC8lPhEItBiOZaob3RbilxKfCARaDMcy1Y1uC/FLiU/ECoH2T6d7h3zY2Cn0CLTVqBvdFuKXEp+IFQK9eqK+G+jovwh0fziWqW50W4hfSnwiVgq03gNFoBEcy1Q3ui3ELyU+ESsE2j9dHX7+zTdOfOaFN3c4+PF8fHtbhGOZ6ka3hfilxCdihUBHdwNdgE/lzMKxTHWj20L8UuITsUqg/WcQaBDHMtWNbgvxS4lPxCqBDhX65nM/2tz4znM7PM/HGufgWKa60W0hfinxiVgt0AGLSDEcy1Q3ui3ELyU+EVkC7T9+fzN3okegLUfd6LYQv5T4RGQJtHHi29siHMtUN7otxC8lPhHZAv3w2aNVtXH0q41ckxTf3hbhWKa60W0hfinxicgV6M7pTLcj0Dwcy1Q3ui3ELyU+EZkCrf156NjJ+25pxqDx7W0RjmWqG90W4pcSn4g8gV7arA6/mB5dPl2l24sg0JU4lqludFuIX0p8IvIEulUdnizD909XRxBoDo5lqhvdFuKXEp+ILIGOb2o34tLmYU6kz8GxTHWj20L8UuITkSXQuRPpmzirPr69LcKxTHWj20L8UuITgUCL4VimutFtIX4p8YnIEmj/dPXg9MmFikP4LBzLVDe6LcQvJT4RWQJlESmCY5nqRreF+KXEJyJPoJc2qxtfSI/euJfTmDJxLFPd6LYQv5T4ROQJdHQh0tGjRxu6FCm+vS3CsUx1o9tC/FLiE5Ep0MFrm+MrOTceOLg/Dc2yBMcy1Y1uC/FLiU9ErkAH/dfvG+6BHnuikfvaxbe3RTiWqW50W4hfSnwisgXaKPHtbRGOZaob3RbilxKfCARaDMcy1Y1uC/FLiU8EAi2GY5nqRreF+KXEJwKBFsOxTHWj20L8UuITgUCL4VimutFtIX4p8YlAoMVwLFPd6LYQv5T4RCDQYjiWqW50W4hfSnwiEGgxHMtUN7otxC8lPhEItBiOZaob3RbilxKfiDyBbt36IgLdL45lqhvdFuKXEp+ILIFePTFzP1AEmoljmepGt4X4pcQnIlOgB78JPQJ1QN3othC/lPhEZAl07kPlEGgmjmWqG90W4pcSn4gsgQ62Jx8Lj0DzcSxT3ei2EL+U+ETkCfTDH1bVoWMnx9zPZyLl4FimutFtIX4p8YnIEujVE9UsfCpnFo5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJT0SWQBsnvr0twrFMdaPbQvxS4hOBQIvhWKa60W0hfinxidiPQPtvIdB94FimutFtIX4p8YnIFujrX6nf/Lz65a828aly8e1tEY5lqhvdFuKXEp+ITIH2nxmtHl09Ud3YwEVJ8e1tEY5lqhvdFuKXEp+ITIFuVdWNf7j5mR/3/6qqDh98HzS+vS3CsUx1o9tC/FLiE5En0AtV9cD4ivjXNhu4sUh8e1uEY5nqRreF+KXEJyJPoFvV7dNbimxXRxBoDo5lqhvdFuKXEp+ILIGObiYyFuilTU6kz8KxTHWj20L8UuITkSXQkTrHAm3i3nbx7W0RjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPRJZA+6frhaOxOS80sAwf394W4VimutFtIX4p8YnIEuho4Wgk0KFMWUTKwrFMdaPbQvxS4hORJ9BLm9VtbyeBXr63avju9NAi1I1uC/FLaWxA9jqRfruqqqObG8duGf55+8F/TFz4LcKxTHWj20L8UuITkSnQwU83J3cDbcCfhmZZgmOZ6ka3hfilxCciV6CDD589OrTnoWY+ID6+vS3CsUx1o9tC/FLiE5Et0EaJb2+LcCxT3ei2EL+U+ETsR6D9NxHoPnAsU93othC/lPhEZAu0vh/oEA7hs3EsU93othC/lPhEZAq0f3r6kXK3NXBH5fj2tgjHMtWNbgvxS4lPRJ5Aa39uHPv2c39339CgBz+P3tAsS3AsU93othC/lPhE5An0wvTspf4PK+4HmodjmepGt4X4pcQnIkugwx3QnbM/t7iUMw/HMtWNbgvxS4lPRJZAr56YuXyT+4Fm4limutFtIX4p8YnIFOiMM7mdXSaOZaob3RbilxKfiCyBju5IP+bSJrezy8KxTHWj20L8UuITkSXQuc9B2m7gavj49rYIxzLVjW4L8UuJT0SeQK+eqL749sSfBz+CNzTLEhzLVDe6LcQvJT4RKwTaf/xk4r7ReaA/OrlZVcfu5xA+B8cy1Y1uC/FLiU/ECoEOdz13wyJSFo5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJT8QKgRYivr0twrFMdaPbQvxS4hOBQIvhWKa60W0hfinxiUCgxXAsU93othC/lPhEZAv0Z89NeZ5V+Bwcy1Q3ui3ELyU+EZkC3flMORaRcnEsU93othC/lPhE5An0Aqvw+8exTHWj20L8UuITkSXQ/ulq44k3p7x1UH8ammUJjmWqG90W4pcSn4gsgV490cBNlBGoAepGt4X4pcQnIlOgDVz/jkANUDe6LcQvJT4RWQLtn0ag+8exTHWj20L8UuITkSXQwTaH8PvHsUx1o9tC/FLiE5En0LnP9ECgeTiWqW50W4hfSnwi8gQ6uHyiuvHkBG5nl4VjmepGt4X4pcQnIlOgz3Ae6L5xLFPd6LYQv5T4ROQJdJsT6fePY5nqRreF+KXEJyJLoOlE+gMftyPQ9qNudFuIX0p8IrIE2vQakqFZluBYprrRbSF+KfGJyBQo54HuH8cy1Y1uC/FLiU9ElkA5kT6CY5nqRreF+KXEJyJLoI18FjwCNUDd6LYQv5T4ROQJtH964wEEuk8cy1Q3ui3ELyU+EVkC7T+ePheeE+n3hWOZ6ka3hfilxCciS6ALH27MeaBZOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EVkCbZz49rYIxzLVjW4L8UuJTwQCLYZjmepGt4X4pcQnAoEWw7FMdaPbQvxS4hORJ9AP35yFD5XLwrFMdaPbQvxS4hORJVAWkSI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RWQLt/+y5MX9zb7Xxnec5kT4HxzLVjW4L8UuJT0SWQGe5tHn44HcGjW9vi3AsU93othC/lPhE7FugjdxYJL69LcKxTHWj20L8UuITsX+BNrELGt/eFuFYprrRbSF+KfGJ2L9Am7i7cnx7W4RjmepGt4X4pcQnYv8CvbSJQLNwLFPd6LYQv5T4ROxboP2tikP4LBzLVDe6LcQvJT4RWQLtPz65FejJ+zYrFpHycCxT3ei2EL+U+ERkCXT+RHpOY8rDsUx1o9tC/FLiE7FvgR76agMfEB/f3hbhWKa60W0hfinxicgSaOPEt7dFOJapbnRbiF9KfCIQaDEcy1Q3ui3ELyU+EQi0GI5lqhvdFuKXEp8IBFoMxzLVjW4L8UuJT8QKgc6cwLQDH2uchWOZ6ka3hfilxCdihUAX7gTK/UD3gWOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE7FCoIu8vllxJVImjmWqG90W4pcSn4h9CbT/2FCfN754YH8ammUJjmWqG90W4pcSn4j9CPS14e7nRhMXIhmaZQmOZaob3RbilxKfiHyBNrb7iUDbi7rRbSF+KfGJyBZo2v3840b06WiWJTiWqW50W4hfSnwiMgV6ucHdTwTaXtSNbgvxS4lPRJ5AG939RKDtRd3othC/lPhE5Aj08un6LqCN7X4i0PaibnRbiF9KfCIyBJp2Px9oUJ+OZlmCY5nqRreF+KXEJ2KlQJvf/USg7UXd6LYQv5T4RKwS6E+b3/1EoO1F3ei2EL+U+ESsECjXwsdxLFPd6LYQv5T4RCDQYjiWqW50W4hfSnwiVgiU+4HGcSxT3ei2EL+U+ESsEGgh4tvbIhzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSn4iGBPrxo3ePH117+lSv99BLS54g0PajbnRbiF9KfCIaEujZ3t0Tk/Zq7nh51xMEaoC60W0hfinxiWhEoNfO9iYCPdu766XBlTO9u84tPkGgBqgb3RbilxKfiCYE+otv9CYC/ehU2t38+NHjP1h4gkAdUDe6LcQvJT4RDQj0nV7vaz8fC/Sd6Z8PLzxBoA6oG90W4pcSn4gmBHrn9wcXx6482/tW+jM9n3uCQB1QN7otxC8lPhENCHTGkdfOjI/WPzp117m5J6Nv+9yY4E8BNepGt4X4pTQ2IAgUroe60W0hfimNDUhzAr3j5bkn898e32NuEY5lqhvdFuKXEp+IYgJdugeKQFuOutFtIX4p8YlAoMVwLFPd6LYQv5T4RDQrUFbhZ3AsU93othC/lPhENCzQySmf4/NAH557EYG2HnWj20L8UuIT0bBAuRJpB8cy1Y1uC/FLiU9EwwK9dqZ35/Ty97knCNQBdaPbQvxS4hPRsEAHV2ZvwHSFuzGZoW50W4hfSnwimhbo4MrTQ2U+dG7JEwTaftSNbgvxS4lPREMC3Sfx7W0RjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAi+FYprrRbSF+KfGJQKDFcCxT3ei2EL+U+EQg0GI4lqludFuIX0p8IhBoMRzLVDe6LcQvJT4RCLQYjmWqG90W4pcSnwgEWgzHMtWNbgvxS4lPBAIthmOZ6ka3hfilxCcCgRbDsUx1o9tC/FLiE4FAG+D936mq6vNPzb70wfdurqpf+f3Rl79Ufeo3Xk0Pn7zh1U9+8xpE3ei2EL+U+EQg0IPz3tCVQz71pzsvvX9Peqn69eHjX95zw6vv3XzT6Dv/QLSNDaFudFuIX0p8IhDogfngkeqGp86/P/zvqzMvffapQf8/JKm+Ug2t+WTSa9t3QJngUhC/lPhEINAD897Nn/7b8/WO5s4u6LvV8KVhma9UN43dmSz6btXyHVAmuBTELyU+EQj0wLxbpcPz4V7nRI+jh5MypwL94JG274AywaUgfinxiUCgB2b3Hugv76lfmRPok0OBtn8HlAkuBfFLiU8EAj0w0/dAb5q88t7NN7z6j79aVZ/9d/WzyXugBjugTHApiF9KfCIQ6MH54HujFfepHocC/fpoFf43z++swhvsgDLBpSB+KfGJQKAH570vJVl+dnoi6LtJp4P+90anNtXngf7eq/UO6AdDr36+zbuh6ka3hfilxCcCgR6Y926udz4/+N7OiaDvpl3PYZlP7hzW1zugHzzy6afev6fNB/LqRreF+KXEJwKBHpgn04H6+Rlbvndz7dJhmfWboePX6h3QdDr9K20+klc3ui3ELyU+EQj0oHzwyHjPc8eWo3X5JNC0QF9TvwP6Sq3ad8e+bSXqRreF+KXEJwKBHpQlAh29NCzz3WrnpeGj8dn0N+3xL7UAdaPbQvxS4hOBQA/M7kP40cP0HuhkbzMtwbMHCntA/FLiE4FAD8y71eIiUr2u9Pmnpqvw58c7oOd5DxT2gPilxCcCgR6cV6qdkz5H60fn3x3foGniytE5oPWhPavwsATilxKfCATaAP+9vh/or6XTQMcCPf/+12+evHT+/PQqeM4DheUQv5T4RCDQYjiWqW50W4hfSnwiEGgxHMtUN7otxC8lPhEItBiOZaob3RbilxKfCARaDMcy1Y1uC/FLiU8EAi2GY5nqRreF+KXEJwKBFsOxTHWj20L8UuITgUCL4VimutFtIX4p8YlAoMVwLFPd6LYQv5T4RLRMoOqgbSF/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQkKgkCB/LcQvJSwkBAoJ8tdC/FLCQhIJNIw6aFvIXwvxSwkLaRH2QDsK+WshfilhIXEIDwny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCQqCQIH8txC8lLCQECgny10L8UsJCKizQa0+f6vUeegmBrj3kr4X4pYSFVFagHz/aq7njZQS67pC/FuKXEhZSWYGe7d310uDKmd5d5xDomkP+WohfSlhIRQX60am07/nxo8d/gEDXHPLXQvxSwkIqKtB3eneP/3wYga455K+F+KWEhVRUoGd730p/XhyLFIGuL+SvhfilhIVUUqDXzowP3T86NXkT9HNjwv+mOmhbyF8L8UsJC2mRdRcoAMDaUkigiycyxfeYW0RHylxXiF9LR/L/ZAS6eB6TuupPhI6Uua4Qv5aO5I9Ai9GRMtcV4tfSkfwLCrTEKnyb6EiZ6wrxa+lI/iUFOjn/s8HzQNtER8pcV4hfS0fyLynQAlcitYmOlLmuEL+WjuRfUqDXzvTubPha+DbRkTLXFeLX0pH8Swp0cKXxuzG1iY6Uua4Qv5aO5F9UoIMrTw/9+dDi/mdnolVvQachfi0dyb+sQPdCXfUnQkfKXFeIX0tH8kegxehImesK8WvpSP4ItBgdKXNdIX4tHckfgRajI2WuK8SvpSP5I9BidKTMdYX4tXQkfwRajI6Uua4Qv5aO5I9Ai9GRMtcV4tfSkfwRaDE6Uua6QvxaOpI/Ai1GR8pcV4hfS0fyR6DF6EiZ6wrxa+lI/gi0GB0pc10hfi0dyR+BFqMjZa4rxK+lI/kj0GJ0pMx1hfi1dCR/BFqMjpS5rhC/lo7kj0CL0ZEy1xXi19KR/BFoMTpS5rpC/Fo6kj8CLUZHylxXiF9LR/JHoMXoSJnrCvFr6Uj+CLQYHSlzXSF+LR3JH4EWoyNlrivEr6Uj+SPQYnSkzHWF+LV0JH8EWoyOlLmuEL+WjuSPQIvRkTLXFeLX0pH8EWgxOlLmukL8WjqSPwItRkfKXFeIX0tH8kegxehImesK8WvpSP4ItBgdKXNdIX4tHckfgRajI2WuK8SvpSP5I9BidKTMdYX4tXQkf41AO8HnPqfegk5D/Fq6mT8CbY5udtDaQPxaupk/Am2ObnbQ2kD8WrqZPwJtjm520NpA/Fq6mT8CbY5udtDaQPxaupk/Am2ObnbQ2kD8WrqZPwJtjm520NpA/Fq6mT8CbY5udtDaQPxaupk/AgUACIJAAQCCIFAAgCAIFAAgCAIFAAiCQAEAgiBQAIAgCBQAIAgCjfHxo3e8rN4GABCDQGMgUABAoEEQKAAg0CAIFAAQaJCxQK88farX+5OX0ktX/qzX6/3rb+56DI0yF+0v6ie//Rfnhg/P9r710Td6x/9yMPifp3q//f3R9878eqAhlgc9/D08lIK+2Hv456d6d/6gI+kj0BgjgX407JAhx781mD7u3b3wGBplLtqf9HaenO39efrSw2fTr+QHg/lfDzTFkqAvzgR9sfdvhs/uOteR9BFojCTQjx/tfe3c4Nrf94aPr53p/dvhjtAvTg07avYxNMpctBd7x78/Xhp6jgAACX1JREFUfO3naYqH4/y1c8PfRe/4NwdXztROnf31QGPsDnqoyof+9+DaT3qjX0ptz86kj0BjJIG+M97FfKf38PCFHVvOPoZGmYv27DD3QZLqt+on9dwOZ7l+7WL9ZPbXA42xd9Bn6z8vjnZKu5I+Ao1RC/TamfEwf3TqrnPDKb7zH8ZfnH0MjbIr2n/+X//5G72RQOtBHQu2nuu5X49gU11ZEnRvdJiebHpxtAPalfQRaIyRQHsT0u7okDv/S2qW2cfQKHPRXvnGOP/lAp399UBT7Ap6elSQDsumAu1G+gg0Rt0rHz861yM//1fp4dfq9pl9DI0yE229THH8T/78H8aH8PVu0Nxcd2OEP2mWBD3OdyzQu0ePu5E+Ao0xEujCW53X/lt9Vs3Dux5Do0yiHe7k3D3e2VkuUN6JLsHeQc8LtBvpI9AY40P4XadoXPv73vQtn9nH0Cgp2smMDnd2lgl02a8HDs7eQY/fA00C7Ur6CDTGeBV+JMi6cT46NTpSqd8zn32s3EhDZqOdCPRi2tPfNddzvx7dFtuxLOi5VfjJ6nsn0kegMSbngd710mDw81PDlhr+H7d+nE6Mm30MjbIQ893pTMM9BDr764HG2B306DzQf356fB7oqOk7kj4CjTF643x8BUbva4OdS2TufHn+MTTKbLQXRw/v+vt6ZnfP9dyvB5pi76DHVyKN9xq6kT4CjTF3Lfz300vp8ei67LnH0Ciz0f7iG8OH3xxN8ZK5nvv1QEMsD7q3cy385LCrE+kjUACAIAgUACAIAgUACIJAAQCCIFAAgCAIFAAgCAIFAAiCQAEAgiBQAIAgCBQAIAgCBQAIgkABAIIgUACAIAgURrz+2GZVVV944u3R06snqtuL/Jyt6vDb1/3yhI0vPLG/f3m/m7xiS5byxmO3DDft0P0v7vcvNvLTYe1AoFDz+i1Tbz2QXtALdMj+FFNeoK/fO920G8MKfe233p796eOn0E4QKAzZntVWstBaCHR/hisu0LmUxv+j2TeTnzr+kx3RdoNAYWSGQ199a/joja8M1fDdQUGBrmBHKP0fDrfqwX381dKbXKe0MUrpsWqf27bDgjERaLtBoDC4tFlVt03GeHu036cXaNqUI/v4q4U3eS6ly8OD+c/8OPLPIFArECjUh81HZp/Vu6DrINChsvZjl8KbPJ/S8IfFfhoCtQKBwtAF6ah9zKUv/NGLIxv1n72lPrSfDPgbaZ3+6Oj5kq+/Xh/+3/b29mTX7HL9/Ru3Lqy2bE13cBf++uyXpxs2fLLjxdG/3D9dPViv5hz67uRnjJfrV23y+Lsn3773liwWMvnX53Y5LyzuqU+//Xo5bU/fZk4/ffJ0+L+K2yc/ZT873aAGgcKFZUfKw0m+9cTsSk7/mekC9I9HX/8X9848r72W+MxjY5FMl1xum/uHp9qa/+vzXx4x1MqR3Yoa/qCT6Qf9eOdn3JazyYPX5p/usSW7CpmmNLfHOfy2+T31sUCvn9MeAh39r2JUcuytVdCAQGF72cFofYi6Mdxpunx6/NXt0TuA9Zt/RyZf/+OZr2+lryd9JJEMv7/e+fzwmQWDTrU1/9fnvzy7YUsEWqvr8hOj92tfTD/z9oxNrt8ReHG0OHXkOluyWMjOxsyrbfTCLoGuzmnZKvzW+CBgO/jOKohAoLBLDTX14KdXh74ae+bI5AsT78x+fXoQuj3yzmjvcfTC7BsEM9qa/esLX66pl7qX7ePVAn1wdlPqV+ovrNrkqZy2Rg+Wb8liITsbNlfGXnpfldMeAr0w+q7hN3EE3yoQKOwl0LHJRrq5MPHHjq3mvj610/i9wp1dqQUpTLU199dnvjzLeB9uUaDjv3FhsuEXJqa97iYvmHyPLVksZOe7FwR6YalAV+W0h0DH3uUIvm0gUNhLoNM9yPmjyq35Pa3R12csubXwfGGheWt+73Dxn58T6G1jtywKdPzvLfzVVZt8oV7SemHFliwWMvPdu/dAlx3CXz+nPU+kX7A3tAQECovLI4mlYvjwjecev6WaiGH26zOnEE2ezzAvoj0WrydfnnBovH6/RKBHZv+p7E0ejJZ3RlcM7LUli4VMWfIe6NJFpOvntKdA0940R/CtA4HC4umWP/32W8vE8Notsz4sJ9DF8yIPIND5TR6dn5TWw5/Yc0v2FOiSVfilgrx+TnsKNO2pcgTfOhAoLJwHuufZl0OOnvz2W1t7CHSy77T4fIFPTqCLm5x44/Hktwf32pLFQmb/9fHT/nPjZZ893qG9Xk57XwtfP7jAEXzbQKCwcMnk8tXl7el1jEvFsOQ90D2usDmAQLd2CXSitNPLThhd3OQJ9XlM9Q9ZuiV7vge6cyXScD/x1hcn/9dZ3LoVOe0t0Hr1aYsj+LaBQGH+Ku/Xlh567ghxvCe221bzi9c7ay4LLo0I9MjkH5oX6HQVfnTou2KTd6lx+ZbstQq/k9L/u7eqNm7ZuWvV7NatzGlPgQ6/8YsnOIJvGwgUZu4z1E/vEo7PSl9uo+3l7+0tOQ90ums1p4V9C3T6oy9UCwKdniK0107x/CZvz5xhtOce6J7ngc7ejen1zara8ebs1q3Mae/b2W1VhzY5gm8bCBQGk/XpyclD9St7HA+/UV+WuHT1eXv3lUj1Ws3keHnKvgU6ud7o2c1Fgabr0eeuRFq1yRsP1BcJnb7Oe6C7C5nbkBnGNl7YuoycNr7bf2ta6ORpKoYj+NaBQKHmp5tTL/xxemHXFTaTm7Hf+sPl5z/2xycg3fjvx96Z6mbeiPsX6OTi9MM/XBTo9Gcsdf7iJqeD8BEzF1WuLmSHSzt3pK/54tu7t25lTrUn6wJ2LkEae3N6zRK0CAQKif7ffWXuxkW7vVLfUqg+EX30rt8SA+6+G1O94L34yUb7F+joRx96YNcq/GD33Ziut8n1C0eHW3TotreuvyVL78Y0YvyZSLe+kHY5RxeSzm3d6pxeG/7FI29PCx0/HexatII2gEChYWzucLmikMuPNbzDuM0RfPtAoNAIC6cUtRdZIfV9Tj/JnwdNgEChEbZnTimS3Mq+KWSFXGINvoUgUGiEoW7qz6nsv7bZ8nfyVIVcPt3u//F0FAQKzXBhssQd/8T09UBSyPbuk6agDSBQaIgPnxkturd+BUlRyIX2/4+nmyBQAIAgCBQAIAgCBQAIgkABAIIgUACAIAgUACAIAgUACIJAAQCCIFAAgCAIFAAgCAIFAAiCQAEAgiBQAIAgCBQAIMj/B5QsL1SadmCyAAAAAElFTkSuQmCC)