Probability & Type 1 & 2 errors

probability

• When psychologists have collected their data from an investigation they want to know how likely it is that the results are ‘due to chance factors’ or whether there is a ‘true difference’

• The end result of all inferential tests is a figure that indicates the probability of the results being due to chance

How to calculate probability

• Probability is abbreviated to p. It is used to indicate that the probability is less than, greater than, or equal to 0.05

• Probability is expressed as a number between 1 and 0.0, means an event definitely will not happen, and 1 means an event definitely will happen

• Divide a particu

Significance levels

• Significance levels are assigned to establish the probability of the results being due to chance

• If this is acceptably low, we can reject the null hypothesis

• In psychology, we accept a 1/20 likelihood that the results are down to chance

• If our results meet this level of significance then we would be 95% confident that we have a true difference or relationships between variables

• Whilst p<0.05 is the baseline level of significance in psychology, there are other levels . Sometimes more stringent (stricter) significance levels are set, where more certainty is needed in the effects of the independent variable e.g. in studies that have more dangerous implications i.e. drug testing or in replications of studies where the aim is to verify the original findings

Type I error

This is where we wrongfully accept the experimental hypothesis. This means we believe there is a difference or relationship, when no such relationship exists. This is sometimes known as a false positive or an error of optimism.

Type II

This is where we wrongly accept the null hypothesis. We believe there is no difference between conditions or no relationship when in fact a relationship does exist. This is a false negative, or an error of pessimism