STATS 7 Lecture Notes - Lecture 4: Null Hypothesis, Categorical Variable, Test Statistic
Document Summary
Row percentages: use the total row count as the basis for computing a percentage. Column percentages: use the total column count as the basis for computing a percentage. Correlation is a measure of the strength of a linear relationship b/w two quantitative variables. Verify that the necessary data conditions are met then summarize the data into an appropriate test statistic. The the p value, the probability the test statistic would be as extrmeme as it is, or more so, calculated assuming the null hypothesis is true. Make a decision as to whether or not the result is statistically significant based on the p value. Every hypothesis test begins with stating a set of hypotheses about the population. The null hypothesis h0 is the hypothesis of no relationship, no difference, nothing going on, the status quo, etc, in the population. The alternative hypothesis ha is what researchers hope to show, that something interesting is going on in the populations.
