Term
| No matter how carefully it was made, a measurement could have turned out differently. This reflects what |
|
Definition
| Chance Error. You need to replicate the measurement to estimate the likely size of the chance error. |
|
|
Term
| The likely size of the chance error in a single measurement can be estimated by |
|
Definition
| the Standard Deviation of a sequence of repeated measurements made under the same conditions |
|
|
Term
| Bias, or systematic error, causes measurements to be systematically too high or systematically too low. What is the equation |
|
Definition
| individual measurement = exact value + bias + chance error |
|
|
Term
| Chance error changes from measurement to measurement, but the bias |
|
Definition
| stays the same. Bias can’t be estimated just by repeating the measurements. |
|
|
Term
| Even in careful measurement work, a small percentage of |
|
Definition
| outliers (extremes) can be expected |
|
|
Term
| The average and the SD can be strongly influenced by |
|
Definition
| outliers. Then the histogram will not follow the normal curve at all well. The median is not effected |
|
|
Term
| Histograms are used to summarize |
|
Definition
|
|
Term
| When a set of numbers are give as observed data arrange them in ascending order first. The position of the median is given by (n+1)/2. If n is an odd number, n+1 is even and (n+1)/2 is an integer. This integer value gives the position of the |
|
Definition
|
|
Term
|
Definition
|
|
Term
| Does bias affect all measurements the same way |
|
Definition
| Yes. Bias pushes them in the same direction |
|
|
Term
| Does Chance Error change from measurement to measurement |
|
Definition
| Yes. sometimes up and down |
|
|
Term
| what does it mean to calibrate something |
|
Definition
|
|
