Model answer outline:
A significance test is one important tool researchers use to draw conclusion about population based upon research conducted on samples. Research hypothesis is confirmed disconfirming the null hypothesis-by showing that it’s not supportted by data. Null hypothesis is that the population mean is the same as the sample mean.
If we reject the null hypothesis, the observed difference is statistically significant. Who decides what probability represents statistically significance? By convention, researchers decides before collecting data by establishing a criterion of significance. It’s usually 5%. While testing a null hypothesis, if the observed difference is equal or less than 5%, the result is statistically significant and they reject the null hypothesis. If greater than 5%, the result is not statistically significant and they must accept the null hypothesis. this criterion of significance is also called alpha level.
The three kinds of significance tests are t-test, ANOVA and chi- square test. T-tests are used to compare the means of two groups. For more than two groups, analyses of variance are used. A chi square test tests the equality of two frequencies or proportions.
suppose that females do better on spelling tests on average if given a high protein breakfast and male do the same with a low protein breakfast. Now if someone asks if it’s good to have a high protein breakfast before a spelling test , your answer would be it depends who is the person – male or female.This is because there is an interaction between the two independent variables. ANOVAs can assess interactions among IVs and this technique helps ascertain if the IV influenced the DV.
Chi- square tests are significance tests that work with categorical , rather than numerical data. Categorical data are on the nominal scale and we often have frequency.