OUR ANALYSIS
Here is the picture of our data, it consists a total of 30 datas.
In total we have 4 variables, which are : circumference of waist, circumference of neck, sex and the difference between waist’s circumference and 2 times of neck’s circumference.
However, we are only using 2 variables in deciding whether to reject or accept the null hypothesis.
And the 2 variables are:
1) circumference of neck
2) circumference of waist
Since our 2 variables are scales, we decided to use Pearson ‘s R test
Pearson's R test[rationale]: this test is used for analyzing only scale datas.
Before start using Pearson’s R test, we have to ensure that all 4 assumptions are not violated.
Assumption1 states that all observations must be independent of each other
Assumption 2 states that the dependent variables should be normally distributed at each value of the independent variables.
Assumption 3 states that the dependent variables should have the same variability at each value of the dependent variable.
Assumption 4 states that: the relationship between the dependent and independent variables should be linear
From explore, we know that the skewness of both the independent and dependent value are very small. So, it can be considered normally distributed. (assumption 2 is not violated)
Scatter plot are used to check for linearity and homogeneous variance.( Assumption 3 and 4)
Scatter Plot
[rationale]:
1) To ensure our datas are fit for using Pearson’s R test
2) To find the correlation coefficient for measurement of linearity (if there is a relationship between 2 variables, we can find out that whether it is a positive or negative relationship.)
From the graph shown:
1) association of 2 variables is a straight line,it means assumptions 3 and 4 are not violated
2) it is a positive line, indicates a positive relationship
By now, we can start using Pearson’s R
Result:
There is a positive, very strong and significant association between a person' circumference of neck and circumference of waist.
November 2008
