Hypothesis testing simplified
Objective: Describe a variable - Mean, Median, Mode, Standard Deviation Examples: What is the percentage of totally satisfied customers in our database? What is the most common income in our data base?
Objective: compare one group to a hypothetical/ set value Z-test (rare) or T-test (one sample t-test; very often) Good with interval and ratio data Examples: – Santa Fe grill wants to check whether the prices they have are perceived as reasonable or not? – I am doing a better job at satisfying customers than average?
What does it mean?
P-value and alpha-level The P-value (probability value) is a measure of how confident we can be that what we observe in the sample is also true for the population. How confident we want to be? 90%? 95%? 100? The alpha level is the P-value that we as researchers decide to accept before we will be confident enough to release a finding.
Statistical test of significance a result is called statistically significant if it is unlikely to have occurred by chance.
Objective: Compare two unpaired groups Unpaired t-test (independent samples t-test) Example: – Is the satisfaction level of women different than the satisfaction level of men?
Comparing Two Means with Independent Samples t-Test
Objective: comparing two paired groups The Paired Samples T-Test compares the means of two variables. It computes the difference between the two variables for each case, and tests to see if the average difference is significantly different from zero. Satisfaction pre-recovery Satisfaction post-recovery Difference