英文:
How do I alculate 2 means per subject differentiated by a variable in R
问题
这是我在stack overflow上的第一个问题。我尝试了一些研究，但没有取得进展。我试图在R中计算不同变量区分的每个子项目的两个不同的均值。
我创建了这个示例数据集：
airquality <- data.frame(subject = c("CityA", "CityA", "CityA", "CityA", "CityA", "CityA", "CityA", "CityA",
"CityB", "CityB", "CityB", "CityB", "CityB", "CityB", "CityB", "CityB",
"CityC", "CityC", "CityC", "CityC", "CityC", "CityC", "CityC", "CityC"),
code = c("AA_high", "BB_high", "AA_low", "BB_low", "AB_high", "AB_low", "BA_high", "BA_low",
"AA_high", "BB_high", "AA_low", "BB_low", "AB_high", "AB_low", "BA_high", "BA_low",
"AA_high", "BB_high", "AA_low", "BB_low", "AB_high", "AB_low", "BA_high", "BA_low"),
measure = c(2, 6, 5, 3, 5, 5, 4, 6,
7, 8, 6, 7, 4, 12, 9, 7,
8, 12, 11, 9, 15, 11, 10, 16))
print(airquality)
目标是每个主题有2个均值：第一个均值是所有代码以"AA"或"BB"开头的测量的均值。每个主题的第二个均值应该是以"AB"或"BA"开头的代码的测量均值。
我从这里开始：
df1 <- filter(airquality, substr(code, 1, 2) == "AA" | substr(code, 1,2) == "BB")
df1 <- df1 %>%
group_by(subject) %>%
mutate(mean_AABB = mean(measure))
View(df1)
df2 <- filter(airquality, substr(code, 1, 2) == "AB" | substr(code, 1,2) == "BA")
df2 <- df2 %>%
group_by(subject) %>%
mutate(mean_ABBA = mean(measure))
View(df2)
但这似乎不起作用，我需要将这些均值添加到数据框中，以便我可以进行更多的均值计算。
所以下一步是将这些类型的均值添加到数据框中。
非常感谢帮助！
非常感谢，Maike
英文:
This is my first question on stack noverflow. I have tried some reasearch, but I am not getting there. I am trying to calculate two differents means per subjet which are differentiated by a variable in R.
I created this example data set:
airquality <- data.frame(subject = c("CityA", "CityA","CityA","CityA", "CityA", "CityA", "CityA", "CityA",
"CityB","CityB","CityB", "CityB", "CityB", "CityB", "CityB", "CityB",
"CityC", "CityC", "CityC", "CityC", "CityC", "CityC", "CityC", "CityC"),
code = c("AA_high", "BB_high", "AA_low", "BB_low", "AB_high", "AB_low", "BA_high", "BA_low",
"AA_high", "BB_high", "AA_low", "BB_low", "AB_high", "AB_low", "BA_high", "BA_low",
"AA_high", "BB_high", "AA_low", "BB_low", "AB_high", "AB_low", "BA_high", "BA_low"),
measure = c("2", "6", "5", "3", "5", "5", "4", "6",
"7", "8", "6", "7", "4", "12", "9", "7",
"8", "12", "11", "9", "15", "11", "10", "16"))
print(airquality)
The aim is to have 2 means per subject: The first mean is the mean of all the measures which codes start with "AA" or "BB".
The second mean per subject should be the mean of the measures of all the codes which start with "AB" or "BA".
I started with this:
df1 <- filter(airquality, substr(code, 1, 2) == "AA" | substr(code, 1,2) == "BB")
df1 <- df1 %>%
group_by(subject) %>%
mutate(mean_AABB = mean(measure))
View(df1)
df2 <- filter(airquality, substr(code, 1, 2) == "AB" | substr(code, 1,2) == "BA")
df2 <- df2 %>%
group_by(subject) %>%
mutate(mean_ABBA = mean(measure))
View(df2)
But this doesn't seem to work and I need to have columns added to the dataframe, so I can do more calculations with the means.
So the next step would be to add those type of means to the dataframe.
I am very grateful for help!
Thanks very much,
Maike
答案1
得分: 1
怎么样：
library(dplyr)
airquality |>
group_by(first_two_equal = substr(code, 1, 1) == substr(code, 2, 2)) |>
mutate(measure = as.numeric(measure), ## !
mean = mean(measure)
)
（请注意，你示例数据中的`measure`是类型为"character"，所以你需要将其转换为"numeric"）
**编辑**
对于扩展的分组，将额外的因素包括在`group_by`中，如下所示：
airquality |>
group_by(subject,
first_two_equal = substr(code, 1, 1) == substr(code, 2, 2)
) |>
mutate(measure = as.numeric(measure), ## !
mean = mean(measure)
)
输出：
-
A tibble: 24 x 5
Groups: subject, first_two_equal [6]
subject code measure first_two_equal mean
<chr> <chr> <dbl> <lgl> <dbl>
1 CityA AA_high 2 TRUE 4
2 CityA BB_high 6 TRUE 4
3 CityA AA_low 5 TRUE 4
4 CityA BB_low 3 TRUE 4
5 CityA AB_high 5 FALSE 5
6 CityA AB_low 5 FALSE 5
<details>
<summary>英文:</summary>
What about:
library(dplyr)
airquality |>
group_by(first_two_equal = substr(code, 1, 1) == substr(code, 2, 2)) |>
mutate(measure = as.numeric(measure), ## !
mean = mean(measure)
)
(mind that `measure` of your example data is mode "character" so you'll need to convert it to "numeric")
**edit**
For extended grouping, include the additional factors in the `group_by` like so:
airquality |>
group_by(subject,
first_two_equal = substr(code, 1, 1) == substr(code, 2, 2)
) |>
mutate(measure = as.numeric(measure), ## !
mean = mean(measure)
)
output:
-
A tibble: 24 x 5
Groups: subject, first_two_equal [6]
subject code measure first_two_equal mean
<chr> <chr> <dbl> <lgl> <dbl>
1 CityA AA_high 2 TRUE 4
2 CityA BB_high 6 TRUE 4
3 CityA AA_low 5 TRUE 4
4 CityA BB_low 3 TRUE 4
5 CityA AB_high 5 FALSE 5
6 CityA AB_low 5 FALSE 5
</details>
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论