跨越两个不同的数据集以将结果从一个数据集变异并汇总到另一个

如何解决跨越两个不同的数据集以将结果从一个数据集变异并汇总到另一个

说我有这个假数据框

，并且在此也有此“ external_table”：

我想通过以下过程“同步”两个数据集，以将新列添加到external_table的新列中：

1. 访问fake_ds
2. 对来自external_table％>％itens_fator的项目执行功能
3. 将结果映射到外部表中

我想要的输出（带有假结果）

这是脚本应执行的伪函数：

fake_ds %>%  #get my ds
  mutate(cronbach_alpha = fake_ds %>% 
           select(external_table,itens_fator) %>% 
           alpha(.)$total[1]) #get variables from external table

如果难以实现此功能，我将介绍其他也能产生所需输出的方法。

我想留在tidyverse。

重新创建数据的代码：

library(dplyr)
library(tidyr)

x <- paste0("y",seq(1:96)) #create X
y <- rep(0:5,96*2) #create values
fake_ds <- data.frame(x,y) #dataframe
fake_ds %>% 
  pivot_wider(names_from = x,values_from=y,values_fn = {mean}) -> fake_ds
fake_ds <- fake_ds %>% slice(rep(1:n(),each = 50)) #replicate
fake_ds <- rbind(fake_ds,seq(1:96)) #add variability
    
#external table
external_table <- structure(list(
  name = c("X5","X1","X2","X0","X3","X4"),itens_fator = c("y1,y12,y59,y76,y78,y92,y93,y94,y96","y5,y14,y15,y16,y17,y18,y20,y24,y40,y60,y62,y64,y75","y10,y19,y32,y34,y36,y37,y47,y56,y58,y72,y80,y85","y13,y30,y39,y53,y54,y55,y66,y73,y84,y91","y42,y43,y45,y63,y69,y77,y87,y88","y44,y49,y50,y68,y82,y89")),row.names = c(NA,-6L),groups = structure(list(name = c("X0","X4","X5"),.rows = structure(list(4L,2L,3L,5L,6L,1L),ptype = integer(0),class = c("vctrs_list_of","vctrs_vctr","list"))),class = c("tbl_df","tbl","data.frame"),.drop = TRUE),class = c("grouped_df","tbl_df","data.frame"))

解决方法

正如我在评论中提到的那样，您要实现的目标是不可能的....要获得cronbach的alpha，您需要做的就是对strsplit使用lapply。.

获取cronbach的字母

lapply(strsplit(external_table$itens_fator,","),function(x) fake_ds %>% 
           select(all_of(x)) %>% 
           alpha(.)%>% .$total %>% .$raw_alpha)

整洁的方式

external_table$itens_fator %>% strsplit(",") %>% purrr::map_dbl(function(x) fake_ds %>% 
           select(all_of(x)) %>% 
           psych::alpha(.)%>% .$total %>% .$raw_alpha)

数据

library(psych)
library(dplyr)
library(tidyr)

x <- paste0("y",seq(1:96)) #create X
y <- rep(0:5,96*2) #create values
fake_ds <- data.frame(var=x,val=y) #dataframe
fake_ds %>% 
  pivot_wider(names_from = var,values_from=val,values_fn=list(val=mean)) -> fake_ds
fake_ds <- fake_ds %>% slice(rep(1:n(),each = 50)) #replicate
fake_ds <- rbind(fake_ds,seq(1:96)) #add variability
    
#external table
external_table <- structure(list(
  name = c("X5","X1","X2","X0","X3","X4"),itens_fator = c("y1,y12,y59,y76,y78,y92,y93,y94,y96","y5,y14,y15,y16,y17,y18,y20,y24,y40,y60,y62,y64,y75","y10,y19,y32,y34,y36,y37,y47,y56,y58,y72,y80,y85","y13,y30,y39,y53,y54,y55,y66,y73,y84,y91","y42,y43,y45,y63,y69,y77,y87,y88","y44,y49,y50,y68,y82,y89")),row.names = c(NA,-6L),groups = structure(list(name = c("X0","X4","X5"),.rows = structure(list(4L,2L,3L,5L,6L,1L),ptype = integer(0),class = c("vctrs_list_of","vctrs_vctr","list"))),class = c("tbl_df","tbl","data.frame"),.drop = TRUE),class = c("grouped_df","tbl_df","data.frame"))