Download presentation
Presentation is loading. Please wait.
Published byCandace Hawkins Modified over 9 years ago
1
Analysis of Variance ANOVA
2
Example Suppose that we have five educational levels in our population denoted by 1, 2, 3, 4, 5 We measure the hours per week that each level watches TV and we repeat this experiment for 6 weeks. So we will enter our data in the following table
3
Week 1Week 2Week 3Week 4Week 5Week 6 1202318293023 224 293119 33125 262028 42729 341927 5232934322729
4
We want to know that whether educational level affect on the weekly hours watching TV or not. So we define this hypothesis
5
For testing this hypothesis we use ANOVA in SPSS by using following order 1- Entering data in the variable view 2- Analyze 3- Compare Means 4- One-Way ANOVA
10
Correlation
11
In this part we are going to calculate the Correlation between two variables In SPSS we can get the Correlation Coefficients between two variables like -Pearson Correlation Coefficient -Spearman Correlation Coefficient
12
For getting these coefficients we use this order in SPSS 1- Entering data in variable view 2- Analyze 3- Correlate 4-Bivariate Correlation
13
Example XY 1232 1424 1334 2632 2333 1526 1828 XY 2030 1931 1925 1827 1826 2226 2530
18
Results If correlation coefficient between two variables is 1 or near to 1, they are dependent. If it is zero or near to zero, they are independent. So, in this example X & Y are independent
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.