Experimental Design 2

Question 1

Define the terms;
-randomised
- double blind
- single blind
- triple blind

Answer 1

Randomised - subjects are assigned to either treatment or control group in random fashion

Double blind - subjects aren’t aware of the treatment they’re receiving and neither is the investigator

Single blind - subjects aren’t aware of the treatment they’re receiving but the investigator is

Triple blind - belt and braces approach

Question 2

Define;
- placebo controlled
- cross over

Answer 2

Placebo controlled - A false treatment is necessary to measure the proportion of any observed effect accountable to the placebo effect

Cross over - Subjects are swapped over from treatment to control group, assumes the effects of the drug are completely reversible

Question 3

What are some obstacles to be overcome ?

Answer 3

Biological variability

Chances - experimental errors, faults in methods, differences in reagents

Bias in experimenter

Experimental design

Question 4

What are the positives and negatives of statistics ?

Answer 4

Positives;
- overcome variation in natural systems by analysing whether a difference can be ascribed to chance events
- enable hypothesis to be tested
- large volumes of otherwise confusing data can be organised, analysed and compared

Negatives;
- can be abused
- used to confirm desired results

Question 5

Define what is meant by “sham” procedures in animal experiments

Answer 5

Refer to a control intervention designed to mimic the experimental procedure as closely as possible, without implementing the specific treatment or variable being tested

Used to control for the potential effects of the surgical or procedural aspects of an experiment, ensuring that any observed outcomes are due to the experimental treatment itself rather than the act of performing the procedure

Ethical considerations as may cause pain to animals without offering potential benefits from the treatmen

Question 6

Discuss the issues involved in obtaining representative samples of a population:

Answer 6

Sampling bias - some members of the population are more likely to be included in the sample than others

Too small a sample; inaccurate

Too large a sample - costly and inefficient

Population can migrate or emigrate, significant deaths or births

Samples must be representative of the entire population for valid conclusions to be drawn

Question 7

Explain how randomisation helps reduce or prevent bias

Answer 7

Ensures that groups are comparable, minimizes confounding factors, and enhances the internal validity of the study

Eliminates selection bias, which occurs when researchers influence group allocation, leading to systematic differences

Balances known and unknown confounding variables, ensuring that they are evenly distributed across groups

Supports valid statistical analysis by ensuring that observed differences between groups are likely due to the intervention rather than bias

Prevents chronological bias by preventing researchers from knowing or predicting the next allocation

Question 8

Define the term Populations, samples, parameters

Answer 8

Population - entire group a researcher is studying

Samples - a subset of the population that should be representative of the population

Parameter - numerical value that describes a characteristic of a population

Question 9

What does matching do ?

Answer 9

The technique of matching (“randomised block design”) can further reduce the effect of known variability