202notes3__Ttests

# 202notes3__Ttests - Chapter 9 The Significance of the...

This preview shows pages 1–2. Sign up to view the full content.

Chapter 9 The Significance of the Difference between Means The t-test for comparing the difference between two means Tests the hypothesis that the mean scores on some interval- or ratio-scaled variables will be significantly different for the two independent samples or groups How far apart do the two mean s have to be so that one can say the difference is not due to chance/sampling error Answer: resort to sampling distribution Null Hypothesis : difference between sample means Assume there is no difference in the means of the populations from which the 2 samples were drawn: i.e. assume the means of the 2 populations are the same Ho : υ 1 - υ 2 = 0 Slight difference between means will occur due to sampling error t = Mean 1 - Mean 2 ______ Variability of random means t = X1 - X2 = difference between means___________ S combined standard error of difference between means X1-X2 *If one surveys many different samples, there will be a whole array of differences between means: X1 - X2, X3 - X4, X5 - X6 and so forth 1. The differences between means are arrayed as a sampling distribution 2. The mean of the sampling distribution would be 0 3. The batch of differences would be normally distributed around the mean of the differences The distribution of differences in a normal distribution, therefore: 1. 68.23% of the differences would fall between -1 and +1 standard deviations from the mean 2. 95% of all the differences would fall between -1.96 and +1.96 standard deviations from the mean 3. 99% of the differences would fall between -2.58 and +2.58 standard deviations from the mean Probability statements concerning the normal curve would apply 1. p = .68 that any difference would fall between u diff + or - standard deviation 2. p = .95 that any difference would fall between u diff + or - 1.96 standard deviations 3. p = .05 that any difference would fall outside the same interval, u diff = or - 1.96 standard deviations The question we ask then is where in the sampling distribution of differences does our obtained difference (between the two means) lie? If it is near the mean, i/e. between -1.96 and +1.96, we say the difference is due to sample error. If the difference is large and falls outside the area of -1.96 and +1.96, we think the difference may not be do to sample error, i.e. it is a rare occurrence The procedure: calculate the z-score for the difference between the two means and find its location in the sampling distribution of differences *USE TABLE A. FIND THE X-SCORE, THEN LOOK TO THE RIGHT TO FIND

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the 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.

### Page1 / 4

202notes3__Ttests - Chapter 9 The Significance of the...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online