Interquartile Range

[StringJunky]

Ten samples are taken and an average reading of 6.1 resulted. The IQR was 0.6. Does this mean the sample readings were pretty close in range about the mean?

[imatfaal]

Possibly - but not enough infomation.

0.8 1.8 5.8 5.8 5.8 6.4 6.4 6.4 10.4 11.4 I would call that clustered around the mean

[StringJunky]

It's a fibroscan liver stiffness result. The 10 samples would need to be + or - 1 or 2 within each other to be a valid test series I think, otherwise it would have been done again. I just wondered what the IQR of 0.6 meant WRT to this.

Edited April 17, 2015 by StringJunky

[imatfaal]

It's a fibroscan liver stiffness result. The 10 samples would need to be + or - 1 or 2 within each other to be a valid test series I think, otherwise it would have been done again. I just wondered what the IQR of 0.6 meant WRT to this.

 

It means that you have majority of results within 0.3 of the median. But it is difficult to say anything else. I have just realised that I also made the rookie mistake of assuming that the mean and median were close/same. Without that info the information you gave could be even more extreme.

 

 

Would you like to act on the mean of this set of data 0.9 1.0 1.0 1.1 1.2 1.4 1.5 1.6 10.3 41

I am trying to think how bad it could be with the added proviso that "10 samples would need to be + or - 1 or 2 within each other"

[StringJunky]

 

It means that you have majority of results within 0.3 of the median.

Yeah, I thought it would be something like that..Maybe, I'm assuming it was an 'average' reading as opposed to a median one.

 

Thanks anyway. I was just curious.