在进行调查后，我以数据框的形式收集了结果。这是实际数据框架的可重现版本。
library(dplyr)
library(tidyr)
df=data.frame(ID=c("1101","1102","1103","1104",
"1105","1106","1107","1108",
"1109","1110","1111","1112",
"1113","1114","1115","1116",
"1117","1118","1119","1120",
"1121","1122","1123","1124",
"1125","1126","1127","1128",
"1129","1130","1131","1132",
"1133","1134","1135","1136",
"1137","1138","1139","1140",
"1141","1142","1143","1144",
"1145","1146","1147","1148",
"1149","1150","1151","1152",
"1153","1154","1155","1156"),
Country=c("US","UK","Canada","Mexico",
"India","US","Peru","China",
"US","UK","Canada","Mexico",
"Portugal","India","Portugal","Mexico",
"Peru","India","Canada","Mexico",
"India","UK","India","Canada",
"US","UK","China","India",
"US","Mexico","Canada","Mexico",
"Canada","China","Canada","Canada",
"China","China","India","Mexico",
"Portugal","Portugal","Portugal","Portugal",
"UK","UK","UK","Peru",
"Peru","Mexico","US","US",
"Peru","Mexico","Peru","Mexico"),
Gender=c("Male","Male","Male","Female",
"Female","Female","Male","Female",
"Female","Female","Male","Female",
"Male","Male","Female","Female",
"Female","Male","Female","Female",
"Female","Female","Male","Female",
"Male","Female","Male","Female",
"Female","Male","Female","Female",
"Male","Male","Male","Female",
"Male","Male","Female","Female",
"Male","Female","Male","Female",
"Male","Female","Male","Female",
"Male","Female","Male","Female",
"Male","Male","Male","Male"),
Age=c("<25","25-35","25-35","36-45",
">55",">55","25-35",">55",
"<25","25-35","25-35","36-45",
"25-35","25-35","25-35","36-45",
">55","36-45","46-55","36-45",
">55","46-55","25-35","46-55",
"<25","46-55","25-35","46-55",
"25-35","25-35","46-55","36-45",
"<25","<25",">55","36-45",
"36-45","46-55","<25","<25",
"<25",">55","36-45","46-55",
"<25",">55","36-45","46-55",
"36-45",">55","36-45","46-55",
"<25","46-55","<25","46-55"),
Score_Q1=c(4,4,3,2,
1,1,4,2,
1,1,1,2,
2,1,4,3,
4,3,1,1,
1,2,1,1,
1,4,1,4,
3,4,3,3,
1,3,3,1,
1,1,2,1,
1,2,1,2,
1,1,1,1,
2,2,2,2,
1,2,3,4),
Score_Q2=c(1,4,1,1,
1,2,1,1,
1,4,4,4,
2,1,1,3,
4,3,1,1,
1,3,3,3,
2,4,1,2,
4,4,4,4,
1,1,1,1,
1,2,3,4,
4,4,2,1,
1,2,3,2,
1,2,1,2,
4,3,2,1))
数据框可以拆分为以下部分 -
1）ID：受访者ID
2）国家：被投诉人的原籍国
3）性别：受访者的性别
4）年龄：受访者年龄
5）Score_Q1：Q1的满意度得分，从1（非常满意）到4（非常不满意）。
6）Score_Q2：Q2的满意度得分，从1（非常满意）到4（非常不满意）。
首先清理一些数据 -
#convert to factor
df$Country=as.factor(df$Country)
df$Gender=as.factor(df$Gender)
df$Age=as.factor(df$Age)
现在我在我的数据集中检查年龄和性别的比率 -
性别由Country
#1) Gender by Country
split_gender=df %>% select(Country,Gender) %>%
group_by(Gender,Country) %>%
summarise(n=n()) %>%
ungroup() %>%
select(Country,Gender,n) %>%
group_by(Country,add=TRUE) %>%
spread(Country,n)
split_gender=data.frame(apply(split_gender, 2, as.numeric))
split_gender_sample=as.data.frame(sweep(split_gender,2,colSums(split_gender),`/`))
split_gender_sample[1,1]="Female"
split_gender_sample[2,1]="Male"
Age的Country
#2) Age by Country
split_age=df %>% select(Country,Age) %>%
group_by(Age,Country) %>%
summarise(n=n()) %>%
ungroup() %>%
select(Country,Age,n) %>%
group_by(Country,add=TRUE) %>%
spread(Country,n)
split_age=data.frame(apply(split_age, 2, as.numeric))
split_age[is.na(split_age)] <- 0
split_age_sample=as.data.frame(sweep(split_age,2,colSums(split_age),`/`))
split_age_sample[1,1]="<25"
split_age_sample[2,1]=">55"
split_age_sample[3,1]="25-35"
split_age_sample[4,1]="36-45"
split_age_sample[5,1]="46-55"
#Clean up unwanted dataframes
rm(list=c('split_age','split_gender'))
以上两个步骤给了我两个数据框架 - split_age_sample和split_gender_sample。这些数据框包含56个受访者的国家年龄和性别样本比率。
我的目标：基于人口统计计算抽样权重
为了使我的数据框架更能代表现实，我想根据国家的年龄和性别的官方人口比率将权重归因于我的受访者。
这些是我在调查的国家找到的官方人口比率。
#Gender by Country
split_gender_official=data.frame(Gender=c("Female","Male"),
Canada=c(0.4,0.6),
China=c(0.3,0.7),
India=c(0.3,0.7),
Mexico=c(0.5,0.5),
Peru=c(0.6,0.4),
Portugal=c(0.5,0.5),
UK=c(0.4,0.6),
US=c(0.4,0.6))
#Age by Country
split_age_official=data.frame(Age=c("<25",">55","25-35","36-45","46-55"),
Canada=c(0.1,0.3,0.3,0.2,0.1),
China=c(0.3,0.05,0.35,0.1,0.2),
India=c(0.5,0.05,0.35,0.05,0.05),
Mexico=c(0.2,0.3,0.2,0.1,0.2),
Peru=c(0.1,0.3,0.2,0.2,0.2),
Portugal=c(0.2,0.1,0.05,0.05,0.6),
UK=c(0.2,0.3,0.1,0.3,0.1),
US=c(0.2,0.3,0.1,0.3,0.1))
期望的输出
根据我的样本比率和年龄与性别的官方人口比率，我想将权重归因于我的受访者，在另一个名为qazxsw poi的专栏中。
目前我无法弄清楚如何进行此计算。
然后，一旦计算出权重，我想使用weights列总结得分。聚合看起来像这样（除了计算中的权重） -
示例：英国的加权汇总分数
weights我真的很感激我可以在重量计算及其在后面的加权聚合步骤中使用的任何帮助。
不确定它是否正是您正在寻找的，但这是我的想法。您需要将国家权重与数据框合并，然后才能计算KPI。
#Calculate weighted overall scores by Country & Gender: example UK
weighted_aggregated_scores_gender=df %>%
select(-Age) %>%
group_by(Country,Gender) %>%
filter(Country=='UK') %>%
summarise(Q1_KPI=round(sum(Score_Q1 %in% c(1,2)/n()),2),
Q2_KPI=round(sum(Score_Q2 %in% c(1,2)/n()),2))
希望能帮助到你。