POLI 30 Lecture Notes - Lecture 10: Contingency Table, Confounding, Ordinal Data
Document Summary
Hypotheses: what we should see in the data if our theory is true. So we need methods for looking at data to see if our hypotheses are true. Crosstabs: relationships between nominal and/or ordinal variables. Use when both independent and dependent variables are nominal and/or ordinal. Calculate separately for each value of the independent variable. If different values of the iv are in rows, calculate row percentages. Compare percentages for different categories of the independent variables. Note: calculating the other way answers a different question. We use three-way and n-way tables to address possible confounds. Three-way/n-way tables: cross tab but you separate the data according to the other variable(s) With these we see if the relationship between our iv and dv persists when holding a confound constant. We call this controlling for a confounding variable. We will breakdown our original crosstab and look at it separately for each value.