This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Need help with statistics: Quartiles and Median

Hello everybody,

I need some help with statistical anaylsis of biological properties. Assume, that for some biological properties A, B, C and D, we have their occurrence given as m/n-ratios. We have these data for three example species S1 to S3 in a table like shown below (My real data set contains 14 properties for over 100 species):

    A    B    C     D
S1  0.7  0.3  0.25  NA
S2  NA   0.4  0.1   0.6
D3  0.3  NA   NA    0.3

As seen above, in some cases no data is available (denoted as NA), so my question now is how to define the quartiles Q1 and Q3 and the Median Q2 for a certain property? Can somebody explain to me how to handle this for property A? Unfortunately I never encountered this case with non-available data.

Best regards

quartile median statistics
• 2,220 views • link
updated by Ram Ram • written by thefirstrealace • thefirstrealace

1 answer

When you calculate statistics, just use the rm.na=TRUE option (that's in R - there should be equivalent options in other tools). These data points are to be excluded in your analysis anyway.

• 0 views • link
written by Ram Ram

I actually use R, but im still new to it. If is understand correctly, rm.na=TRUE filters out the NA and only uses the remaining data points for the calculation of quartiles and median?

• 0 views • link
written by thefirstrealace • thefirstrealace

Yes, that is correct. The option is available in a lot of functions and is quite useful.

• 0 views • link
written by Ram Ram

Log in to answer this question.

More posts like this