Inferential Stats- Probabllity and Significance Flashcards
what do inferential stats involve?
analysing data using statistical tests to make conclusions about whether a hypothesis is supported by the findings
what are inferential stats compared with descriptive stats?
more objective way of analysing data than descriptive statistics which only describe or summarise data which must then be interpreted by the researcher which can be more subjective
What is a statistical test used to do to analyse data obtained?
to calculate the calculated value (sometimes called observed value or test statistic).
What is the calculated value compared with and why?
a critical value from an appropriate critical values table to objectively determine whether the calculated value is significant or not significant.
What is a significant result?
Indicates that the result was due to the psychological phenomenon being investigated and not due to chance.
-null hypothesis rejected
-the alternative hypothesis is accepted
What is a non significant result?
Indicates that the result does not support the psychological phenomenon, or the result was due to chance.
-null hypothesis is therefore accepted -alternative hypothesis is rejected
What do researchers do to deal with the fact that whatever the outcome of a study, they can never be 100% certain that the result wasn’t due to chance?
they assume a level of significance, which accounts for chance causing the result, but which researchers are prepared to accept.
what do researchers set for assuming the level of significance?
a probability value (p-value) which is the probability that the result was due to chance. It is usually written as a decimal (or percentage).
what does a smaller p-value mean?
the result is less likely to be due to chance and the null hypothesis can be more confidently rejected.
What is the most commonly chosen level of significance in Psychology?
p≤0.05 - (up to a 5% probability that the result is due to chance- means that psychologists are generally satisfied with being 95% certain of a result.
When is the outcome of a statistical test deemed as significant?
if the calculated value is appropriately greater or less than the critical value at the given level of significance (indicated by the p-value)
In psychology why is it convention to use p<=0.05 as a significance level?
it provides the best chance of reaching a balance between Type 1 and Type 2 errors.
when may psychologists choose a stricter level of significance such as p<=0.01?
If the outcome from a study could have a significant effect on people’s lives, psychologists will often want to be more confident in the results
What does a level of significance of p<=0.01 mean?
there is only up to a 1% probability the results are due to chance making researchers 99% certain of the outcome.
