SPSS_ANOVA

SPSS_ANOVA - SPSS ANOVA 1. One-way Anova You should not use...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
SPSS ANOVA 1. One-way Anova You should not use the chi-squared test if the dependent variable is interval-scaled. Instead, ANOVA should be used. In this example, we want to investigate if there is any educational difference in "Likelihood of Buying" the Woolworth dog biscuit. To use ANOVA, click on Analyze from the main menu, select Compare Means, then One-way ANOVA to go into the dialog box for one-way ANOVA. In the dialog box, select 'Likelihood of buying', the probability of buying Woolworth dog biscuit at $15 per packet, and specify it as the dependent variable. Then select Education and put it under Factor. Finally, click on Options to pop up another menu in which you should click on Descriptive and Homogeneity of Variance:
Background image of page 1

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

View Full DocumentRight Arrow Icon
Click Continue, then Post Hoc. From the list of Equal Variances Assumed, select Scheffe. Select Tamhane’s T2 from the list of Equal Variances Not Assumed: Click Continue. Click OK in the One-Way ANOVA dialog box to get the output: Oneway
Background image of page 2
The above is just a table summarizing the data. The N column shows the number of respondents in each educational category. The Mean is the average value of the variable “likelihood of buying” for each group. Similarly, the standard deviation of the “likelihood of buying” for each group is shown in the Std. Deviation column. STD. Error for each group is obtained by dividing the Mean by the Std. Deviation. For example, the Std. Error for the primary group is = 5.67/2.875 = 1.174. The Std. Error (standard error) is an estimate of the error in using the sample mean as an estimator of the population mean. For example, we may use the mean of the primary education group 5.67 as an estimator of the corresponding population mean that is unknown (the population mean is unknown unless you conduct a census of the population), and in so doing, you can expect an error of 1.174. The 95% confidence interval is obtained by “Mean + 1.96*Std. Error”. For example, using the primary group as an example, we have “5.67 + 1.96*1.174 = 2.65 and 8.68”. Thus, we are 95% sure that the unknown population mean is between 2.65 and 8.68. The 95% confidence interval is known as an interval estimator of the unknown population mean, while the sample mean is a point estimator. For the output, look at the following ANOVA table. Note that the total sum of squares (SST) is partitioned into the sum of squares between groups (SSB) - the variation in buying probability that can be attributable to difference in education, and the sum of squares within groups - the variable in buying probability that is not due to educational difference. Also look at the column labelled as D.F., or degrees of freedom. The total D.F. is equal to the D.F. for between groups and the D.F. for within groups., and the D.F. for between groups is just the number of groups less one. The numbers in the mean squares column are the ratios of the sum of squares and the
Background image of page 3

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

View Full DocumentRight Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 17

SPSS_ANOVA - SPSS ANOVA 1. One-way Anova You should not use...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online