The Student Room Group

Median

Why do we use the formula n+1/2 for working out the median of discrete data and then we just use n/2 for continuous data?
Reply 1
If you mean (n+1)/2 for discrete data then that is making the assumption that the discrete data itself begins at 1. Because in that case you are adding the highest and smallest numbers and dividing by 2 to get the middle of the whole range. In the case of continuous data the median isn't necessarily the midpoint of the data. Its where F(x) = 0.5 is the median.
Reply 2
Original post by Dingo749
If you mean (n+1)/2 for discrete data then that is making the assumption that the discrete data itself begins at 1. Because in that case you are adding the highest and smallest numbers and dividing by 2 to get the middle of the whole range. In the case of continuous data the median isn't necessarily the midpoint of the data. Its where F(x) = 0.5 is the median.

Right really simple question but when working out the median from stem and leaf do you do (n+1)/2?
And say if you were working out quartiles and you had 40 pieces of data and you did (n+1)/4 for LQ. You would get 10.25? Now are you looking for the 10th term or the 11th? I've always been confused about this so it is better I am told rather than mess up in an exam. Thanks :smile:
Reply 3
Original post by Super199
Right really simple question but when working out the median from stem and leaf do you do (n+1)/2?
And say if you were working out quartiles and you had 40 pieces of data and you did (n+1)/4 for LQ. You would get 10.25? Now are you looking for the 10th term or the 11th? I've always been confused about this so it is better I am told rather than mess up in an exam. Thanks :smile:


n means the highest number in the data not the number of data. so n isn't equal to 40 its equal to the highest of the 40 numbers :smile:

The lower quartile would be the 10th term, the upper quartile the 30th :smile:

Imagine the numbers 1 - 10

Immediately you might think ahhh yes the median is 5, bu infact it's 5.5, exactly 4.5 away from the highest and lowest values :smile:
Reply 4
Original post by Dingo749
n means the highest number in the data not the number of data. so n isn't equal to 40 its equal to the highest of the 40 numbers :smile:

The lower quartile would be the 10th term, the upper quartile the 30th :smile:

Imagine the numbers 1 - 10

Immediately you might think ahhh yes the median is 5, bu infact it's 5.5, exactly 4.5 away from the highest and lowest values :smile:


How is it 5.5?
Reply 5
Reply 6


Yeah, a good thing to look up if you haven't come across it already would be the rectangular distribution :smile:

I was thinking yesterday that it could explain this well :smile:

To give a quick insight if you a rectangular distribution is where a function of x can take any value between two values a and b

so a<= x <=b

Each of these values takes a further value but when they are put into the function but are all the same :smile: so in actual fact the function is just a number k, specifically (1/(b-a))

The mean value of the rectangular distribution, defined as 0.5(b+a) is also the median.

Have a search about and check it out :smile: It appears in S2 of the AQA syllabus, not sure of other exam boards :smile:

Quick Reply

Latest