SOCI 328 Chapter Notes - Chapter 8-14: Confidence Interval, Chi-Squared Distribution, Scatter Plot
Document Summary
Chapter 8: hypothesis testing ii the two-sample case. Identify + cite situations to use two sample test hypothesis. Concerned with significance of difference b/w 2 separate populations. Is difference b/w samples large enough to conclude that the populations represented by samples are different? . Tests for significance of difference b/w sample means and sample populations. Second major difference null hypothesis: instead of saying that population of sample has certain characteristic, null hypothesis states no difference between populations. When sample size > 100, use z distribution, when sample size < 100, use t-distribution. Alpha = 0. 05, z(critical) = +/- 1. 96. Degrees of freedom needed in step 3 when doing t-test degrees of freedom = n1 + n2 2. Hypothesis testing with sample proportions (large samples) When confidence intervals don"t overlap, confidence interval and hypothesis testing lead. In two-sample case, there is close, but not direct relationship between the techniques to the same result rejection of null hypothesis.
