Evaluation Key Words Flashcards
Accuracy
How close in agreement the result is and the true value. The true value is usually unknown.
Anomaly/outlier
A value in a set of results that is judged not to be part of the inherent variation and does not follow the trend.
Confidence
A statement expressing how strong a conclusion is based on the quality of evidence.
Control variable
Something in an experiment which must be kept the same in order to test the relationship between the independent and dependent variables and ensure the validity of the experiment.
Dependent variable
The variable which is the “output” or the “effect” in the experiment. It is usually the variable that you measure. It is plotted on the y axis of the graph.
Error of measurement
The difference between an individual measurement and the true value.
Independent variable
The variable which is the “input” or “cause” in the experiment. It is usually the variable that you change. It is plotted on the X axis of the graph.
Negative controls/ Negative control experiment
These will produce no effect on the experiment. It ensures that there is no effect when there should be no effect e.g. the use of water in place of a drug to prove that any effects are as a result of the drug and nothing else within the experiment.
Positive controls/ Positive control experiment
These will produce an expected effect in the experiment in order to compare this to the independent variable that is being investigated in the experiment e.g. comparing an existing working drug to a new drug with unknown effects.
Precision
A measure of how close the repeats are to one another.
Repeatability (used to be called reliability)
The precision of results collected by one person with the same equipment on the same day.
Reproducibility
The precision of results collected by different people using equivalent equipment in different places over a wider period of time.
Resolution
The smallest change in quantity that can be detected by a piece of measuring equipment e.g. the resolution of ruler is 0.5mm.
Uncertainty
An estimate of the range of values within which the true value is thought to lie. This is normally expressed e.g. 44.0 (plus or minus sign) 0.4.
Validity (i.e. Fair testing - does it answer the Q? Did you control all the control variables?)
A measure is valid if it measures what it is supposed or to be measuring.
An experiment is valid if it is suitable to answer the question being asked and the variables are controlled.