# Practical on determinig if a burette is more accurate

1. Oct 13, 2013

### lionely

1. The problem statement, all variables and given/known data

Practical on determining if a burette is more accurate than a measuring cylinder

Method: The burette was filled to a mark above 45cm3 with water and the reading recorded. The burette was opened and 30 drops of water were allowed to be delivered into the measuring cylinder, and the reading on the burette recorded. The volume of water in the measuring cylinder was recorded. The readings were recorded to two decimal places, these steps were then repeated allowing 60,90,120 and 150 drops of water to be delivered and the results tabulated.

Now I was asked to plot three graphs

i) Volume of liquid delivered from burette vs # of drops and Volume of liquid collected by cylinder vs # of drops on the same graph leaf.

ii) A graph of burette volume delivered Vs Volume of liquid collected in cylinder.

I don't see the point of the third graph, what's the point of plotting the volume of water delivered vs the volume of liquid collected? Could someone please explain this to me?

Oh and for the first graph , I took two gradients for each line, and compare how they close they were. For the burette I got .046cm3 and .047cm3 and for the cylinder .049 and 0.040. So since the gradients for the burettes were closer it should be more accurate right? Oh and the reason why my measuring cylinder gradient is so messed up is due to errors, in doing the experiment my partner was impatient so the rate at which the drops fell from the burette were not constant.

Last edited: Oct 13, 2013
2. Oct 13, 2013

### UltrafastPED

If the devices were perfect the two volumes should be identical ... but this graph will highlight any discrepancy.

3. Oct 13, 2013

### lionely

You mean by taking the gradient it should come to say.. 1? But if it doesn't how does it show which is more accurate , i don't get it :S

4. Oct 13, 2013

### UltrafastPED

If the graph is perfectly linear there may not be a way to tell - but for your lab report you must carry out an analysis of the actual data.

Data is king!

5. Oct 13, 2013

### lionely

So I wasn't supposed to draw a line of best fit for the 3rd graph?

6. Oct 13, 2013

### UltrafastPED

I would just plot the points ... the best fit line may be interesting, so also calculate the R^2 for it.

Then write study it and decide what it means. Good luck!

7. Oct 13, 2013

### lionely

Umm What is the R^2, I'm only in grade 13, I haven't done any labs needing R^2 before, and I don't recall seeing anything like this in maths.

8. Oct 13, 2013

### UltrafastPED

The least squares best linear fit will calclate a correlation coefficient called R^2. If not, forget it.

9. Oct 13, 2013

### lionely

Okay just to be clear, I shouldn't think about the R^2 value and for the graph of the volumes of the drops against each other, should I just find the gradient? If the gradient is almost the same that should indicate the precision to which the experiment was carried out with? I'm still a bit lost with what to do with the third graph!

10. Oct 15, 2013

### epenguin

I would say the first two graphs are capable of telling you whether the drops are uniform (same volume/drop) so look for any trends in the graph like systematic slight curvature. Though if there is any that could have more than one explanation!

The last graph just compares now well the two devices give the same result. Notice any drop size variation is cancelled out. So although in theory the information is all in the first two graphs, something is visible in the third one that is less evident than in the first two.

Maybe you thought they were looking for something more clever that stopped you having these simple thoughts?

One mistake students of experimental sciences make is to think of the procedures as rituals, that if you are terribly careful about the procedures and careful exact right volumes or number of drops or whatever that guarantees you truth and success. Instead they need to learn to think about validation. What really guarantees that this and that experiment or measurement is telling me what I think/they me what I think it tells? That my instruments are telling me the truth, if if necessary what do I have to do to know this? Maybe they are trying to put over this idea.

You might or you might not be able to say something (which you pretty well guess anyway) about which device is better or worse that the other and in what way. How would you decide how good either of these devices are anyway?

I hope this compensates you a bit for such a dull experiment!