This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Incidental Sample : sample of certain available group, only given option of one population Biased Sample : systematic error, self selection MEAN of the Means • Better than any one mean, more samples • To see how well sample mean estimates true mean, do confidence interval HYPOTHESI Null Hypothesis = Ho • Do not expect any difference • To be nullified Alternative Hypothesis = reject Ho • Assume correct and disprove then favor alternative ALPHA : .001 .01 .05 .10 • The less strict (.05 less strict than .001) the easier to have a "winner" • Importance/involvement determines appropriate strictness Ex: medical procedure - very strict, little margin for error *every industry has different "norm", mean of mean that is acceptable 2-Tail Test : don't know direction of preference/difference (can't predict difference) 1-Tail Test can predict direction of difference (predict winner)...
View Full Document
This note was uploaded on 04/25/2011 for the course MKT 202 taught by Professor Hillman during the Spring '10 term at DePaul.
- Spring '10