type one errors
a type one error is when we say there's a significant difference or correlation when infact, there isn't. this also means we have accepted the alternate hypothesis incorrectly. therefore we have rejected the null hypothesis incorrectly.
type two errors
a type two error is when we say there isn't a significant difference or correlation when infact there is. this also means we have rejected the alternate hypothesis incorrectly. therefore we have accepted the null hypothesis incorrectly.
we can minimise type one errors by decreasing the probability to 0.01. but whilst this minimises type one errors it can increase the likelihood of type two errors.
to minimise type two errors the probability has to increase to 0.1, however this can increase the hances of type one errors.
usually p=0.05 is used as a probability level as it is a compramise between type one and type two errors