ADMS 3490 Lecture Notes - Lecture 2: Statistical Inference, Statistical Significance, Sample Size Determination

• School: York University
• Department: Administrative Studies
• Course: ADMS 3490
• Professor: Anita Boey

Document Summary

Issue #2- measurement issues in r&s: variables. Nominal: cannot average data values or run t-tests. Nominal measurement: characteristics, numerical values just name, label, or classify attributes, no ordering of cases implied, examples, ethnicity, others, examples, caucasian, asian-american, african-american, middle eastern, pacific islander. Ordinal measurement: characteristics, attributes can be rank-ordered, distances between attributes do not mean anything, example. Less than hs: some hs, hs diploma, some college, college degree, ma/ms/mba, ph. d. Ide(cid:374)tify the le(cid:448)el of (cid:373)easure(cid:373)e(cid:374)t of the follo(cid:449)i(cid:374)g (cid:448)ariables : annual income, favorite rock band, satisfaction with campus safety. Inferential statistics test hypotheses about relations between variables and generalize beyond sample observations. Infer population characteristics from samples: there are restrictions on inference, sample size and representativeness, statistical significance of findings, significant = probably true of the population and was not due to chance. Pearson product moment correlation coefficient: correlation coefficient, ranges from -1 to 0 to +1. Indicates the direction of influence: + or -