In: Statistics and Probability
Assignment Direction:
Q1) One-Sample t-Test (21 points total)
I am curious about how many hours per week my students spend on their online coursework. So, I surveyed 9 of my students on how many hours per week they spend on my online PSY 230 course. According to a survey of online college students (in all courses), students studied an average of 9 hours a week for a 3-credit course. I assume that the national study time scores are normally distributed, and I set the significance level at α = .05.
Student |
Study hours |
1 |
12 |
2 |
19 |
3 |
21 |
4 |
14 |
5 |
11 |
6 |
9 |
7 |
12 |
8 |
10 |
9 |
7 |
One sample t test is conducted when you have a known values for your sample data.In this you might want to compare the population mean with the sample mean you have, and you dont know the population variance .or in the case you have a small sample
a) In a testing of hypothesis the dependent variable is the one which is being tested or neasured , here we are studying about the study time so the dependent variable is the study time .
Here the sample is the 9 study time score data of 9 students collected .
b) The Null hypothesis H0: There is no significant difference between the population mean and the sample mean
H0:
and the alternative hypothesis will be H1: There is difference between the population mean and the sample mean
H1:
c) To calculate sample mean we ahve the formula
Here and n= 9 so = sample mean = 12.77
d) We have the formula for estimating the population SD as
Where is the population standard devation and = 9 hours (given)
So we have the Population SD=
X | X- | |
12 | 3 | 9 |
19 | 10 | 100 |
21 | 12 | 144 |
14 | 5 | 25 |
11 | 2 | 4 |
9 | 0 | 0 |
12 | 3 | 9 |
10 | 1 | 1 |
7 | -2 | 4 |
Sum= 296/9=32.88
Square root of 32.88 = 5.73 which is the estimated value of population SD
e) Standard error is the standard devation of the sample
in the previous formula we need to cange the population mean as sample mean and the equation will become
The Standard error
x | x-xbar | |
12 | -.7 | .49 |
19 | 6.3 | 39.69 |
21 | 8.3 | 68.89 |
14 | 1.3 | 1.69 |
11 | -1.7 | 2.89 |
9 | -3.7 | 13.69 |
12 | -.7 | .49 |
10 | -2.7 | 7.29 |
7 | -5.7 | 32.49 |
Sum =167.61/9=18.62 Square root = S=4.31
f) from b we have taken the null hypothesis as Population mean equal smaple mean and the alternate hypothsis are population mean not equal to the sample mean to it is a two tailed test , if we would have taken greater than or less than we would have got a one tailed test
g) The test statistic will be = 12.77-9/4.31/3=3X3/4.31=9/4.31=2.088
h) since we have the 2 tailed test we have p value is the probability that the t statistic with degrees of freedom 8 (n-1) is less than -2.088 or greater than +2.088
From the t distribution calculator applying the values of T score that is 2.088, Degree of freedom =8 two tailed and also significance level =.05
P value = .0702> .05
If we have the pa value greater than the significance level we need to reject the null hypothesis
So we have to conclude that there is significance difference between the mena value of study time of the sample and national average
Effect size