3.2 Experimentation

Question 1

Validity

Answer 1

When variables are controlled so that any measured effect is likely to be due to the independent variable.

Question 2

Reliability

Answer 2

When consistent data is obtained in repeats and in replicates carried out independently

Question 3

Accuracy

Answer 3

Closeness to the true value

Question 4

Precision

Answer 4

Closeness to other results obtained

Question 5

Pilot Study

Answer 5

Used to plan procedures, assess validity and check techniques. It allows for evaluation and modification of experimental design and can ensure an appropriate range of values for independent variable

Question 6

Simple Experiment

Answer 6

Experiment with one independent variable

Question 7

Multifactorial Experiment

Answer 7

Experiment with more than one independent variable

Question 8

Observational Studies

Answer 8

Good to detect correlation but not causation

Question 9

How can uncontrollable confounding variables be dealt with?

Answer 9

Randomised Block Design

Question 10

Positive Control

Answer 10

Used to check that the experimental system is able to detect a positive result.

Question 11

Placebos

Answer 11

Negative control treatments without the independent variable

Question 12

Placebo Effect

Answer 12

A measurable change in the dependent variable as a result of a human patient’s expectations, rather than changes in the independent variable

Question 13

Random Sampling

Answer 13

All members of the population have an equal chance of being selected in the sample.

Question 14

Systematic Sampling

Answer 14

Members of a population are selected at regular intervals

Question 15

Stratified Sampling

Answer 15

A non-homogeneous population is divided into categories called strata that are then sampled proportionally.

Question 16

Representative Sample

Answer 16

Should have the same mean and same degree of variation about the mean as the whole population. Larger sample size in more variable populations