11.3 - Evaluating & Improving a Method
Sources of error and improvements
- A source of error is something in the method that makes a reading less accurate.
- Any error must be paired with a specific improvement that reduces it.
Common errors and their fixes
| Source of error | Improvement that fixes it |
|---|---|
| A short time measured by hand with a stopwatch, so your reaction time is a large share of the reading | Time twenty swings and divide by twenty, or use light gates instead of a stopwatch |
| A scale read from an angle, so the mark you line up with isn't the one directly in front of you (parallax) | Put your eye level with the scale and read it straight on |
| Hot water losing heat to the room during the run | Insulate the container and fit a lid |
Anomalies and repeats
- An anomaly is a result that doesn't fit the pattern of the others.
- Identify anomalies and leave them out when calculating the mean.
Why you repeat and take a mean
- It lets you spot anomalies that a single reading wouldn't show.
- Calculating the mean of several repeats reduces the effect of random error, so the mean is likely to be closer to the true value.
Representative samples
- A representative sample is needed when the property being measured isn't uniform.
- For example, a wire's thickness varies, so measure its diameter at several points along its length and at different angles, then calculate a mean.
TermDefinition
Source of errorSomething in the method or equipment that makes measurements differ from the true value.
Parallax errorAn error caused by reading a scale from an angle instead of with your eye level with it, so a pointer or liquid surface appears against the wrong mark.
Anomalous resultA result that does not fit the pattern of the other results.
Random errorAn unpredictable error that makes readings vary above and below the true value, so repeat readings differ from each other.
PreciseDescribes repeated measurements that are close to each other, with very little spread about the mean.
True valueThe value that would be obtained in an ideal measurement with no errors.
Representative sampleA sample taken from several places on an object, so that it shows what the whole object is like even where it varies from place to place.
Key points to remember
How were these notes?