PROBABILITY- Levels of statistical significance Flashcards
expressing probability
e.g
1 in 2
50%
50:50
0.5
Probability in psychology
Psychologists interested to see if results of their studies show real differences or correlations, or if results are due to chance factors.
A significant result
One where there is LOW probability that chance factors were responsible for any observed difference, correlation or association in variables tested.
Level of significance
To say whether results are significant and not due to chance factors we need to use a measure of the level of significance.
Also need to decide how large an effect/relationship is required for us to conclude it is unlikely to be due to chance, and to decide where to draw the line. (this decision is reflected in level of significance we choose to use).
Level of significance as decimal value in the form:
P < 0.00
‘P’ stands for
The probability that chance factors are responsible for the results.
5% level of significance
p < 0.05
(psychologists concluded 5% level of significance is appropriate)
5 in 100 might go against preduction so 5% of results may be due to chance, however you are still going to accept alternative hypothesis and claim effect is supported using this level.
P < 0.05 significance level
-The probability of our results occurring by chance is LESS 5%. (1 in 20 chance of being a fluke)
-Or chance results will have occurred in less than 1 out of 20 ps.
-95% confident that results obtained are due to manipulation of IV rather than chance factors.
(NULL HYPOTHESIS REJECTED)
If choosing a well-established theory/ research finding with contradictory results…
You may choose a higher level of significance like
P <0.01
reduces the risk of false positives but increases the risk of false negatives.
If a result is significant ( P < 0.05 )
NULL is REJECTED
If a result is NOT significant ( P > 0.05 )
NULL is RETAINED
Significant means…
The result is unlikely to have occurred by chance at the level set.
