Accuracy: A measure of how close the data is to the actual true value. Not the difference between accuracy precision. If a man is 1. 81m tall, a measurement of 1. 743 is precise but not accurate. Precision: The closeness of repeated measurements to one another. Precision involves choosing the right apparatus and using it properly. Precise readings are not necessarily accurate. A faulty piece of equipment or incorrectly used apparatus may give preide readings (all repeated values are close together) but inaccurate (not true) results.
Eg in an experiment with a colorimeter, using a dirty of scratched cuvette (sample tube) might give precise readings but they will be highly inaccurate. Reliability: If a measurement or test is reliable, it gives consistent results each time the activity is repeated. When undertaking an investigating a large number of repeats should ideally be taken, and any readings that vary considerably from the others (anomalous) should be repeated.
Validity: The confidence that researchers put in a set of results and the conclusions drawn from those results. Results are valid if they measure what they are supposed to, and if they are
precise, accurate and reliable. Many people may have ideas about why materials have different properties, but these opinions are not very useful if they are not supported by data.
To justify an explanation, you need to have data to support it. This data may be obtained by taking measurements. The accuracy of each measurement depends on the quality of the measuring apparatus and the skill of the scientists taking the measurement. If the apparatus is faulty, or the scientists make a mistake, the measurement may be inaccurate. For the data to be reliable, the variation within the values must be small.
There is always some variation in any set of measurements, whatever is being measured. There may be small differences in the composition of the rubber or the way the measuring apparatus is used. In this set of data, each measurement is only slightly different from the others. The results are repeatable, meaning that each time a measurement is taken it has approximately the same value. We can say that this set of data is reliable.