Mr. Rutt’s Lab Tips

Uncertainty Measured quantities should be ranges, not exact values

If you think the value is about 5 cm, and you’re sure it’s between 4 and 6, record 5 +/- 1cm

Uncertainty cont. Don’t forget about uncertainty when performing calculations on your data

E.g. Hooke’s Law Lab - let’s say m = 0.2 +/- 0.01 kg, x = .10 +/- .01 m

k = mg/x Average k = 0.2(9.8)/.1 = 19.6

Largest possible k = 0.21(9.8)/.09 = 22.9 Smallest possible k = 0.19(9.8)/.11 = 16.9 So k = 19.6 +/- approximately 3

Uncertainty cont. The actual accepted ways of incorporating uncertainty are more complicated than this

When can we use uncertainty? Rulers, length measurements Scales: if a scale gives you a reading of 5.0 kg, what is the uncertainty?

Stopwatch, Voltmeter, force meter, anything that gives a reading

More about Uncertainty - “Human Error” NEVER write “human error” If you can’t use traditional uncertainty, there’s always a better way to account for this

Consider timing a mass falling to the floor - you won’t be able to time it precisely, so how do you incorporate uncertainty? Research human reaction time Perform multiple trials to find a range of uncertainty

Uncertainty vs. Experimental Error If you’ve incorporated uncertainty, you CANNNOT use those measurements as experimental errors

Think about Hooke’s Law Lab - you can incorporate uncertainty to account for ruler measurement error, inaccurate masses, oscillation of spring. What errors are left?

Errors for the sake of errors Even though lab reports often ask for 3 experimental errors, the goal is to have none

Therefore, don’t write things like “we could have waited for the mass to stop oscillating” if you EASILY could have done them!

Data Visualization

x (m) m (kg) F (=mg) k0.1 0.2 1.96 19.6

0.16 0.25 2.45 15.31250.24 0.3 2.94 12.250.32 0.4 3.92 12.250.18 0.1 0.98 5.44444444

Or…

Hooke's Law

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

x

F

Data Visualization Always better to visually represent your data

Gives you a better sense of how good your data is

Helps to identify outlier data points

Reinforces your understanding of the relationships involved

What is the most significant experimental error? Explain why you think it’s a significant error

Explain why you think it’s more significant than other errors

If possible, prove that it’s the most significant error empirically Hooke’s Law Lab - how well does the data fit?

Show your Work! Hooke’s Law:

k = rise/run k = 22.4

Where did this number come from? What two points did you use? Why did you use these points as opposed to other points? Details!

Don’t go into too much detail though - showing one calculation is enough

The Experiment is NEVER Over In university, labs are often much longer

If you’re halfway through, and it’s not working, fix it!

It can be annoying to start the lab over from the beginning, but it’s worth it for good data

Also, your TA will be impressed if you can identify problems and solve them on the spot

Inaccurate data is NOT a bad thing So you get bad data. What do you do? Fudge it so the experiment works? NO!

First task is to figure out what went wrong, fix it, and do the experiment over

If this isn’t possible, the experiment didn’t work, nothing you can do about it

A good prof/TA will focus on your process and analysis, not your results

Don’t get caught up in technology You will be required to learn a lot of different software - graphing, statistical, diagramming, etc.

Don’t let the technology become the focus

Don’t explain in detail exactly the commands you used to find the line of best fit

Rather, focus on what the technology is telling you about your results