Jump to content
Sign in to follow this  

Lab Report IB-acceptable methods of estimating total uncertainty when dealing with MSE?

Recommended Posts

In the writing of my Extended Essay, I'm analysing the data of a bunch of runs I did of a reaction involving ammonium nitrate and water. I made the temperature graphs in Pasco DataStudio (which also happens to be poorly made and a huge pain in the ass to use), and compensated for the imperfect calorimeter and reaction by extrapolating the rate of cooling and finding the intersection between the line and the time the nitrate was added.

Which is all well and good, but out of the three uncertainties I'd need to take into account (slope, init time, y-intercept), I'm missing two. Pasco DataStudio doesn't seem to have a function for finding the slope and y-intercept uncertainty the IB way, and it won't be possible by hand because there simply isn't a pencil that small in existence.

What the program offers instead for its linear fit are the following, using numbers from one of the runs as an example: MSE (or Mean Squared Error) 9.01*10^-4, Root MSE 0.03 and Pearson's r as 0.983.

I did go and read about MSE beforehand, but considering all of my Group 4 teachers (esp. the qualified examiner teachers) were very particular about the exact methods to calculate uncertainties, I'm left completely in the dark as to how I should take these into account - or how any of it is supposed to work without relevant uncertainties to the values I used to calculate temperature.

Anyone have some suggestions?

Share this post


Link to post
Share on other sites

This is very interesting to me, because when I did Chem HL we never had to deal with data this complex! From what I understand you're trying to run a regression? Why do you care about the MSE? From what I know that's a criterion about the opportunity cost of bias vs efficiency. Are you trying to analyse the distribution of your data?

This seems suspiciously advanced for IB to me.

Share this post


Link to post
Share on other sites

Ok, as confused as I am, I believe the MSE is the measure of the "fit" of the slope with respect to the data. I'm not sure why you're trying to do an error analysis on the slope and y-intercept of the function either (Ask your teacher!!!).

I have only used DataStudio for physics, so I'm not entirely sure what your problem is. But for past calorimetry labs, I've found that it was sufficient to determine the uncertainty on initial time as you need that for future enthalpy calculations.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.