COMM 2002 Lecture Notes - Lecture 5: Ordinal Data, Univariate, Central Tendency
Document Summary
Univariate to bivariate: formulated a research question, developed/discovered survey question, stated h1, described each variable (recoded/missing values) Bivariate analysis (analysis of two variables): explores relationship between variables, search for co-variance and correlations, cannot establish causality. Crosstabulation: purpose: connects the frequencies of two variables, helps you identify any patterns of association, uses: suitable for nominal and ordinal data. Independent variable always goes in the column position: dependent variable always goes in the row position, column, row and total percentage all give different information, column percentage reveal the relationship of the iv to the dv. The nature of the relationship: what is the direction of the relationship + or , how strong is the relationship. Bivariate statistical tests for nominal variables: chi-square and cramer"s v. Strength relationship between two nominal variables or one nominal and one ordinal variable. K = smallest number of rows/columns in the table. Cramer"s v doesn"t tell us whether it"s males or females.