Shapiro-Wilk test is a statistical test used to check if a continuous
variable follows a normal distribution. The null hypothesis (H 0) states that the variable is normally distributed, and the alternative hypothesis (H 1) states that the variable is NOT normally distributed. So after running this test: If p ≤ 0.05: then the null hypothesis can be rejected (i.e. the variable is NOT normally distributed). If p > 0.05: then the null hypothesis cannot be rejected (i.e. the variable MAY BE normally distributed).
So in this examples no.1, Im going to demonstrate how to
perform Shapiro wilk test. Before we ran the set we must make sure our data is continuous or they are under interval or ratio level of measurement. so as you can see this the formula on Shapiro wilk test. In column 3 which is equivalent for the denominator in the given formula. And this 2 column which is the column 4 and 5 this is equivalent for the numerator. In my example here’s the internet usage of the respondents in terms of the number of hours they spent on internet. So you can see here data. The first thing you would do is to arrange the number in ascending order or from smallest to largest. After arranging compute now for the mean of the data set just add and divide, and we now have the mean which is 9.8.
We are going to use this mean to get the value of column 3. So
to do that we need to get the difference of (number of hrs-mean) so (6−9.8) so we now have 14.44. and so on. So, for the a sub i 2
column we will be needing this table which I found on the net.
So you can see here the Coefficients a sub i for the Shapiro-Wilk test for normality. So in the columns the number of samples are indicated and on the row the number of pairing that we have. So in our example I have 40 respondents so I need to locate the column that contains 40 then divide by 2 that’s 20. Our reference is 20 and 40 so all this values under 40 will be used, as you can see I just copy pasted all the values under 40 in the column 4 to fill out the remaining copy first the values then arrange this values in ascending order in another sheet of paper. Then transform them in negative numbers. -1x0.0049=- 0.0049 and so on. Then just copy paste everything e sunod lang sya.
For the last column which is the column 5. To get the values is we have to multiply the values in column 4 x column 2. So 0.3964x6=2.3784 and so on and do forth.
To get the value of numerator we just need to the total in
column 5 so its equal to (-10.3806)^2= 107.7569 and for the denominator just get the total in column 3 which is equal to 112.4 to get the W just divide numerator/denominator= 0.958691 or 0.96 then identify the pvalue and remember that it’s a test that null should not be rejected. So we can say sample belongs to a normal distribution. And for that, the p value for Shapiro wilk test should be greater than the significance level in this case if 5% it should be greater than 5% to say that this sample is normally distributed. And to compare the p-value to the test statistics that we got we are going to need the Quantiles of the Shapiro-Wilk test for normality which is I found on internet. So n should be 40 since it has 40 respondents and the reference value is 0.940 which under 5% column. So we have now to compare the test statistics value and the reference value. The test statistics is greater than the reference value so we can now say that p>0.05 so the decision is We fail to reject ho, therefore we accept null hypothesis. (Basa ang conclusion)