PADP 7110 Lecture Notes - Lecture 12: Theoretical Definition, Univariate, Type I And Type Ii Errors

• School: University of Georgia
• Department: Public Administration and Policy
• Course: PADP 7110
• Professor: Abraham

Document Summary

Lecture 12 : data analysis: outline, measurement, hypothesis testing. Type i & type ii errors: univariate statistics. Describe the relationship between two variables: multivariate analysis. Describe the relationship among three or more variables. Classification > variables with values that are mutually exclusive (nominal, ordinal, interval, & ratio) Order > can be logically ranked as having more or less of a thing (ordinal, interval, & ratio) Distance > distances spreading values are standardized/known (interval & ratio) Origin > the values composing a variable are based on a true zero point; can create meaningful ratios between two or more values (ra- tio) Similar results on repeated observations of same condition/event (consistent) Methods for assessment: test-retest reliability, inter-observer/inter-rater reliability, alternative or equivalent forms, interitem reliability (cronbach"s alpha) Extent to which indicator/variable capture the concept it aims to measure. Methods for assessing validity: face validity, content validity, criterio validity, construct validity, measurement sensitivity. Precision of measure (can it detect relevant variation among re-