COMM 2002 Lecture Notes - Lecture 8: Multivariate Analysis, Confidence Interval, Statistical Parameter

• School: Carleton University
• Department: Communication Studies
• Course: COMM 2002
• Professor: Heather Pyman

Document Summary

March 18 th multi-variate analysis controls. Statistical significance sampling distributions and the normal curve. All possible samples of a population will produce a normal curve. Statistics calculated on the sampling distribution will equal the population parameters. Calculation of the mean of a sampling distribution and standard error allow us to calculate the confidence interval. Confidence will rise based on the size of the sample and the strength of the relationship. The probability that our statistic (gamme, tau, etc. ) is within an acceptable distance (usually 2 standard deviations 95% of the time) of the population parameter. How confident you want to be is related to your significance level. Type i or ii error possible based on rejecting or not rejecting the null. 95% confidence level = . 05 significance level or less hypothesis. When the globe and mail say that conservatives have 35% of the decided vote, they also say that the confidence interval or margin of error is + or .