Help needed please

[Rheologist]

Hello All,

I am new to this forum and actually new to statistics.

I am comparing the results of measurements made using 20 different viscosity correlations. The fact is each correlation gives me a different result. Please, what kind of analysis can I do with such data as I can not even be sure which is my 'true' viscosity.

Your help will be greatly appreciated.

Cheers

Moderator Note: Please only post a question in a single thread. I deleted the thread in 'Regression Analysis' because it seemed more appropriate here - Dason

[scubapuppy]

Hi Rheologist,

I think I need more information about what your data looks like and what you are trying to do before I can help at all.

When you say you have measurments from 20 different correlations, what do you mean by that?

[Outlier]

You mean this?
"
There are a number of correlations for predicting crude oil viscosity. These correlations can be
categorized into three main groups: dead oil viscosity (μod), bubble point viscosity (μob) and below
saturation oil viscosity (μb). In general, the correlations utilize oil density and temperature to determine μod.
"

[scubapuppy]

You mean this?
"
There are a number of correlations for predicting crude oil viscosity. These correlations can be
categorized into three main groups: dead oil viscosity (μod), bubble point viscosity (μob) and below
saturation oil viscosity (μb). In general, the correlations utilize oil density and temperature to determine μod.
"

So, it SOUNDS like you actually have 20 observations split across 3 groups? I think you need to look at your data in more detail to figure out what you are being asked to do. From the information you've provided its unclear what your actual question is.... (for example, are you being asked to tell how reliable the 3 measurements are? or are you supposed to be looking at whether there are differences between the three groups?)