probability n sig Flashcards
probability
- refers to likelihood results were obtained by chance
- this likelihood shouldnt take into account ur own intended manipulation of the IV
- u will c probability written as ‘p’
- dont want results to happen bc of chance we want them to happen bc of us, the researchers, have manipluated IV
- its expressed w a number from 0 to 1
0= event wont happen
1= event will happen
significance
- its a stats term which indicates that the association between 2 (or more) variables is strong enough for us to accept the experimental hypothesis
- want results to be sig
- want high probability r results r down to r manipulations n not chance
null hypothesis
No relationship between the two variables
non directional hypothesis
Will predict an effect but the direction is not specified
experimental hypothesis
directional hypothesis
Predicts an effect and the direction of the effect
experimental hypothesis
probability=
number of particular outcomes/number of possible outcomes
significant levels
- most used is : p=0.05 or p<0.05
- A p value less than 0.05 is considered to be statistically significant
rejecting n accepting hypothesis
- if result is over 0.05, we have to reject r experimental hypothesis n accept null
- if its under 0.05=accept experimental n reject null
Why is probability hard to calculate?
natural probability
- may be one of eg 1/9 ppl will develop cancer at some point in their lifetime
- but as theres also lots of other factors influencing this eg smoking, diet it makes it difuclut to say for sure what porbability is
- ppl influenced by these factors may have to be put into groups by experimenter according to each factor
conditional probability
- the probability of something happening if something else occurs
errors
- Sometimes, we calculate/use the wrong significance level
- This is referred to as an error
- There are two types of errors: type 1 n 2
type 1 errors
- ‘false positive’
- null hypothesis is rejected n experimental hypothesis is accepted when it shouldnthave been
- consequences:
1. Publishing false data- other researchers waste time and money trying to replicate
1. Scientific journals waste resources - if study=replicated again n they all get same result which is diff to original=indicate an initial tye 1 error
type 2 errors
- A ‘false negative’
- uve kept the null hypothesis n rejected the alternative, but u should have kept the alternative hypothesis.
-
Consequences:
1. Researcher thinks the study is not good enough and will spend more money trying to improve it eg more participants
1. The actual findings might be important but will not be published
When are errors more likely?
p=0.05 sig level
is fine as long as research is not life or death e.g. medical research (5% possibility that results are due to chance)
When are errors more likely?
p=0.1 sig level
- more likely to make type 1 error (10% possibility that results are due to chance)