Sociology 3306A/B Lecture Notes - Lecture 7: Univariate, Null Hypothesis, Missing Data

• School: Western University
• Department: Sociology
• Course: Sociology 3306A/B
• Professor: Teresa Abada

Document Summary

Two purposes: estimate population characteristics from sample data (description, test hypotheses (explanation) If the confidence level is 95%, and the confidence interval is +/- -2. 5% (tails: we are 95% confident that the population figure is b/w 47. 5% and 52. 5% Precautions for making inferences: samples must be drawn from the population about which the inferences are being made. Inferential statistics make assumptions: simple random sampling, sampling with replacement, 100% completion rate. Inferential stats are addressed to sampling error, not non-sampling error. Bivariate analysis and unmarried: comparison of 2 variables, simultaneously assess the relationship b/w the two, of discrete variables crosstabs/contingency tables. Indicates the strength of association b/w 2 continuous variables: comparison of means continuous dv, categorical iv. Should have title that succinctly describes the analysis presented. The attributes of each variable should be clearly indicated/labeled. When percentages are reported in the table, the base on which they are computed should be indicated.