How Science Works

Question 1

Bad evaluation of conclusion points

Answer 1

There is no control group.
Only one type studied which makes a generalised conclusion.
The standard deviations overlap meaning the data is not significantly different.
There may be some untruthful responses or they are subjective.
Quote the data that doesn’t fit the pattern.
It might not have tested the intermediate values.
If done on animals then their tissue culture may not be comparable to humans.
Small sample size.
A short experiment, meaning the long term effects are unknown.
Something important might not have been measured or tested.
Correlation doesn’t always mean causation as it can be due to another factor.
It might not account for all results.

Question 2

Good evaluation of conclusion points

Answer 2

There tends to be some data that fits the pattern. Quote it.

Question 3

Stats tests: t-test

Answer 3

It is used when looking for a difference between two means.

The difference is significant when the probability is less than 0.05 (5%) that the difference is due to chance.

Question 4

Stats test: Spearmans rank correlation

Answer 4

It is used when looking for a correlation between two sets of continuous data.
1 = perfect correlation
0 = no correlation
The correlation will be significant or not significant.

Question 5

Stats test: Chi squared

Answer 5

It is used when looking at frequencies (e.g. Numbers).
It uses categoric data with the same dependent variable.
The expected and observed are compared.
The difference will be significant or not significant.