Chapter 8: Introduction to Hypothesis Testing, a fundamental technique of inferential statistics. Hypothesis testing differentiates between real, systematic patterns and random, chance occurrences. What is real vs what is random? By combining concepts of z-score, probability, and the distribution sample means, we create a hypothesis test.
-
-
-
Directional Hypothesis Tests (one-tailed) specifies either an increase or decrease in the population mean. (Makes a statement about the direction the effect)
-
Critical region for directional tests (unlikely to occur if the null hypothesis is true - the treatment has no effect)
-
Concerns
Focus of a hypothesis test is on the data rather than the hypothesis (specifically when the null hypothesis is rejected)
Demonstrating a significant treatment effect does not necessarily indicate a substantial treat effect
Size
Measure of effect size is intended to provide a measurement of the absolute magnitude of a treatment effect, independent of the size of the samples(s) being sued
-
Alpha Level (levels of significance) : boundaries that separate the high-probability samples from the low-probability samples
Critical region - extreme values in the tails of the distribution defines the outcomes that are not consistent with the null hypothesis (very unlikely to occur if the null hypothesis is true.
Z-score statistic that's used in the hypothesis is the call test statistic indicating that sample data is converted into a single, specific statistic used to test the hypothesis.
Power of a statistical test is the probability that the test will correctly reject a false null hypothesis