For the data set:
$$18,15,12,6,8,2,3,5,20,10$$
Find the value of the $5$th decile, $D_5$.
I computed this 2 ways and each time I got a different answer.
- If the sample size is $n$, then the rank of the $m$th decile is $\frac{mn}{10}$. So in our case it would be $\frac{5(10)}{10}=5$. So it would be the fifth value in the data after we arrange them from the smallest to the largest.
$$2,3,5,6,8,10,12,15,18,20$$
So it would be $D_5=8$.
- We know that $D_5=Q_2$, where $Q_2$ is the second quartile. $Q_2$ is the value with $50%$ of the data below it, so it is the mean of the data. In this case the mean is $9$
So is it $8$ or $9$? Any help would be appreciated. Thanks!
If you use the nearest rank method you came with this inconsisty, just as pointed in quintile from wikipedia
If you use linear interpolation, you determine the position $p$ of the $q$-quantile in the partition $k$ in your set $X$ as
$$p:={k\over q}(N+1) $$
and the value in the position $p$ computing
$$f(p) = \begin{cases} X_{\lfloor p\rfloor}+\left(X_{\lceil p\rceil}-X_{\lfloor p\rfloor}\right)\cdot (p-\lfloor p\rfloor) & ,p\notin\Bbb N \\[2ex] X_p & ,p\in\Bbb N \end{cases}$$
where $X_p$ is the $p$-element of your ordered set and $\lfloor p\rfloor$ is the integer part of your real number and $\lceil p\rceil = \lfloor p\rfloor +1$.
Appying to your case goes like this. The position $p$ of $D_5$ is calcuted as
$$p={5\over 10}(11)=5.5 $$
then $f(5.5)$ is equal to
$$f(5.5)=8+(10-8)\cdot(0.5)=8+2\cdot0.5=9 $$
If you do the same for the median, gives the same result making consistent.