POLI 210 Chapter Notes - Chapter 15: Analysis Of Variance, Statistical Parameter, Contingency Table

• School: McGill University
• Department: Political Science
• Course: POLI 210
• Professor: Dietlind Stolle

Document Summary

Summarize the conceptual nature of significance tests and the roles that inferential statistics play in hypothesis-testing. Distinguish between type i and type ii errors and appreciate the factors to consider in choosing between the two types. Explain the difference between parametric and non-parametric statistics. Find critical value (expected value) in table for test statistics and select confidence level. If calculated statistic > critical value, reject null hypothesis. If calculated statistic < critical value, accept null hypothesis. Allows to assess whether sample statistics are acceptable estimate of pop parameters. But statisti(cid:272)al sig(cid:374)ifi(cid:272)a(cid:374)(cid:272)e does(cid:374)"t al(cid:449)ays (cid:373)ea(cid:374) su(cid:271)sta(cid:374)ti(cid:448)e (cid:894)i(cid:373)porta(cid:374)t(cid:895) sig(cid:374)ifi(cid:272)a(cid:374)(cid:272)e: sometimes weak significance is still representative, significance tests affected by sample size. Substantive significance: only if statistic/relationship between variables is: theoretically important, plays a role in elaborating, modifying, or rejecting a theory. Confidence(alpha) levels: used to operationalize the extent to which research is due to chance. Probability that sample statistic is an accurate estimate of population parameter.