In: Statistics and Probability
PROBLEM
We have sample scores (out of 15) of a quiz for 20 students of a class as follows.
9.3 | 8.8 | 10.7 | 11.5 | 8.2 | 9.7 | 10.3 | 8.6 | 11.3 | 10.7 |
11.2 | 9.0 | 9.8 | 9.3 | 9.9 | 10.3 | 10.0 | 10.1 | 9.6 | 10.4 |
Test whether population median score of all students of the course is 9.9 using one sample sign test at level of significance .
SOLUTION
We have to test for null hypothesis
against the alternative hypothesis
We put - sign if sample value is lesser than 9.9, + sign if sample value is greater than 9.9 and ignore otherwise.
So the sequence og signs is as follows.
-, -, +, +, -, -, +, -, +, +, +, -, -, -, +, +, +, -, +
Number of + sign
Number of - sign
So,
We define Bernoulli variate
Under Binomial distribution,
we have to test for null hypothesis
against the alternative hypothesis
Under ,
In case of two tailed test (not equal type alternative hypothesis) we reject our null hypothesis if .
Here we observe that .
So, we cannot reject our null hypothesis.
Hence, based on the given data we can conclude that there is not significant evidence that the population median quiz score differs from 9.9.