Download presentation
Presentation is loading. Please wait.
Published byAusten Terry Modified over 9 years ago
1
Unit 1 Accuracy & Precision
2
Data (Singular: datum or “a data point”): The information collected in an experiment. Can be numbers (quantitative) or qualities (qualitative). Known: Substance that has specific (known) values. Used to calibrate experiments. Unknown: Substance to be tested in an experiment.
3
True Value: The “real” value of some property – this is often impossible to really know. Accepted Value: The closest approximation to a True Value; agreed upon by most experts.
4
Trial: A single run of an experiment. Sample Size (n): The number of times an experiment is performed (# trials). Should be at least 3 to ensure valid results. The larger n, the more confident we can be in the data.
5
Mean (x): The average of all the trials. When reporting results, this is usually the final answer that is published. This is also called the Experimental value. Median: The central value in a sample. If there is an even number of trials, it is the average of the two central values.
6
Maximum: Highest data point. Minimum: Lowest data point. Range (for data): A data set bordered by the Minimum and Maximum values. Range (for a result): The reported result (average) with a certain amount of error (± some number). e.g. 7.1±0.2
7
Accuracy: A measure of how close a result is to the true or accepted value. Measured by percent error. One data point can be accurate, but it tells you nothing of the likelihood that the next value will be accurate. When you have a group of results, the accuracy is based off of the mean (average) of your results.
8
Precision: A measure of how close together data points are to each other. It is measured by the standard deviation. You must have > 1 data point in order to have precision. You must have a group or population of results in order to be precise. The greater the precision, the greater the likelihood that your next value will be close to the previous one. It is also called reproducibility or repeatability.
9
C.A.B.D.
10
Percent Error: This is a measure of accuracy. A percent error of <5% is usually considered accurate. Its formula is: |True Value – Experimental | * 100 True Value -or- |True – x| * 100 True
11
The Standard Deviation (σ): The standard deviation can be considered a range of values where the true answer lies. It is a measure of precision. The smaller relative to x, the greater the precision. e.g. if your average result is 4.1 g and σ is ±0.5 g, your true answer is 4.1 ±0.5 g. This is 3.6 g to 4.6 g.
12
You will not be responsible for calculating the Standard Deviation (yet), but its formula is: = sqrt( (x – x) 2 / n)
13
Relative Deviation: This is a measure of relative precision. The lower the relative deviation, the more precise. An experiment can be considered precise if Rel Dev < 10%. Its formula is Rel Dev = / x *100
14
Anomaly: A result that cannot be explained by current scientific understanding. Often times, anomalies are simply the result of random errors. Outlier: An anomalous result that lies outside of the specified range of acceptable values. It may be discarded under certain circumstances.
15
Interpolation : Using a set of data to infer an unknown value between the measured quantities. “Between the points.” Extrapolation: using a set of data to infer an unknown value by extending the trend of the measured quantities. “Outside the points.”
16
What is the value when X = 2.5? Y = 12.5
17
What is the value when X = 0.5? Y = 2.5
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.