Statistical testing Flashcards
The aim of statistical testing
To determine whether a hypothesis should be accepted or rejected by seeing if the differences/relationships between variables are statistically significant or have occurred by chance
When is the sign test used?
1) Looking for a difference in scores.
2) Experiments have used repeated measures design (same person is tested twice) or matched pairs design as p’s are paired and thus count as one person being tested twice.
3) When data is nominal
How to do the sign test
1) state hypothesis
2) Record data and work out the sign, each pair of data has a plus or minus.
3) Find a calculated value. S = calculated test statistic, add up the +s and -s and take the smaller value.
4) Find critical value of S using table
Calculated value must be EQUAL TO or LESS THAN critical value for significance to be shown, and thus hypothesis to be accepted.
0.05 probability/ significance level
5% chance that results did occur by chance- 95% sure results are due to manipulation of IV.
0.01 probability/ significance level
1% chance that results occurred by chance. Used when drugs are trialled and when studies affect human lives OR if the study will never be repeated again.
Directional hypothesis
requires one-tailed test, used when previous research gives researcher confidence in what will happen.
Non-directional hypothesis
requires a two-tailed test, used when either there is no previous research or previous results are unclear or conflicting.
Null hypothesis
the independent variable does not have an effect on the dependent variable ie. there is no change
Table of critical values
contains numbers used to judge significance. the calculated value is compared to the critical value to see if it is significant.
What is the sign test?
A statistical (inferential) test to determine the significance of a sample of data.
