R语言中的factor

对于初学者来说，R语言中的factor有些难以理解。如果直译factor为“因子”，使得其更加难以理解。我倾向于不要翻译，就称其为factor，然后从几个例子中理解：

<span style="font-size:12px;">data <- c(1,2,2,3,1,2,3,3,1,2,3,3,1)
data
</span>

显示结果：

<span style="font-size:12px;"> [1] 1 2 2 3 1 2 3 3 1 2 3 3 1</span>

然后运行：

<span style="font-size:12px;">fdata <- factor(data)
fdata </span>

显示结果：

<span style="font-size:12px;"> [1] 1 2 2 3 1 2 3 3 1 2 3 3 1
Levels: 1 2 3</span>

继续查看class

<span style="font-size:12px;">class(fdata)
[1] "factor"
class(data)
[1] "numeric"</span>

可以看到，factor()函数将原来的数值型的向量转化为了factor类型。factor类型的向量中有Levels的概念。Levels就是factor中的所有元素的集合（没有重复）。我们可以发现Levels就是factor中元素排重后且字符化的结果！因为Levels的元素都是character。

<span style="font-size:12px;">levels(fdata)
[1] "1" "2" "3"</span>

我们可以在factor生成时，通过labels向量来指定levels，继续上面的程序：

<span style="font-size:12px;">rdata <- factor(data,labels=c("I","II","III"))
rdata
</span>

显示结果：

<span style="font-size:12px;">[1] I   II  II  III I   II  III III I   II  III III I
Levels: I II III</span>

也可以在factor生成以后通过levels函数来修改：

<span style="font-size:12px;">rdata <- factor(data,labels=c("e","ee","eee"))
rdata
</span>

显示结果：

<span style="font-size:12px;"> [1] e   ee  ee  eee e   ee  eee eee e   ee  eee eee e
Levels: e ee eee</span>

看到这里，我们马上就会意识到，为什么factor要有levels？因为factor是一种更高效的数据存储方式。对于不同的变量，只需要存储一次就可以，具体的数据内容只要存储相应的整数内容就可以了。因此，read.table()函数会默认把读取的数据以factor格式存储，除非你指定类型。

并且，factors可以指定数据的顺序：

<span style="font-size:12px;"> mons <- c("March","April","January","November","January", "September","October","September","November","August", "January","November","November","February","May","August", "July","December","August","August","September","November", "February","April")</span><pre tabindex="0" class="GCWXI2KCJKB" id="rstudio_console_output" style="font-family: 'Lucida Console'; font-size: 10pt !important; outline: none; border: none; word-break: break-all; margin: 0px; -webkit-user-select: text; white-space: pre-wrap !important; line-height: 15px; color: rgb(0, 0, 0); font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: -webkit-left; text-indent: 0px; text-transform: none; widows: auto; word-spacing: 0px; -webkit-text-stroke- 0px; background-color: rgb(255, 255, 255);"><pre name="code" class="html"><span style="font-size:12px;">mons <- factor(mons)
</span><pre name="code" class="html"><span style="font-size:12px;">table(mons)
</span>

显示结果：

<span style="font-size:12px;">mons
    April    August  December  February   January      July     March       May  November
        2         4         1         2         3         1         1         1         5
  October September
        1         3 </span>

显然月份是有顺序的，我们可以为factor指定顺序

mons = factor(mons,levels=c("January","February","March","April","May","June","July","August","September","October","November","December"),ordered=TRUE)

现在运行：

table(mons)
mons
  January  February     March     April       May      June
        3         2         1         2         1         0
     July    August September   October  November  December
        1         4         3         1         5         1

需要注意的是数值型变量与factor的互相转化：

fert = c(10,20,20,50,10,20,10,50,20)
mean(fert)
[1] 23.33333

转化后：

mean(factor(fert))
Warning message:
In mean.default(factor(fert)) : 参数不是数值也不是逻辑值：回覆NA

那我们这里，是不是可以直接用as.numeric() 转化呢？

mean(as.numeric(factor(fert)))
[1] 1.888889

发现上面是错误的！
这里需要这么转回去：

ff <- factor(fert)
mean(as.numeric(levels(ff)[ff]))
[1] 23.33333