Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

GIO_SS_ DecisionScience

Subjective Question
Q1. A school conducted special classes to help students with poor verbal English. Each student was
required to give a test twice, once before starting the classes (Test 1) and once again after
completing 4 weeks in the class (Test 2).

Let Y1 and Y2 be the scores in Test 1 and Test 2 respectively.

Let Y = Y2 - Y1. Hence, if the population mean μ for Y is equal to zero, then the special classes
have no effect on the average.

It is planned to test the null hypothesis of the class having no effect, against the alternative
hypothesis that there is a positive effect due to the classes.

Student Test 1 Test 2

Abhay 5 8

Bhumika 3 10

Charan 2 5

Dinesh 4 7

Variable Number of Students Mean Standard Deviation Standard Error

Y 4 4.000 2.000 1.000

1. When we make a decision using Significance level = 0.05, this means that the special
classes are truly beneficial, but there is a 5% chance that the classes are not beneficial.
Set up the null and alternative hypothesis
Answer:

H0: µ = 0 (Null hypothesis), so that the classes are not beneficial.


Ha: µ > 0 (Alternate hypothesis), coaching class are truly beneficial.

2. Are the classes beneficial to the students? Make a decision using Significance level as
0.05.

Answer:
Test 2 (Y2)
Student Test 1 (Y1) Y

Abhay 5 8 3
Bhumika 3 10 7
*Charan 2 5 3
Dinesh 4 7 3

Number of
Standard Standard
Variable Students Mean
Deviation Error

Y 4 4 2 1

Significance Level (Alpha) : 0.05

t -Test: Paired Two Sample for Means

Test 1 Test 2
Mean 3.5
Variance 1.666666667 4.333333333 7.5
Observations 4 4
Pearson Correlation 0.372104204
Hypothesized Mean Difference 0
Df 3
t Stat -4
P(T<=t) one-tail 0.014004228
t Critical one-tail 2.353363435
P(T<=t) two-tail 0.028008456

t Critical two-tail 3.182446305


P-Value is 0.014004228, which is less than the significance level 0.05. There is a strong
evidence that the null hypothesis can be rejected (H0 = No effect from coaching class)

3. State whether the following statement is true or false:


When we make a decision using Significance level = 0.05, this means that the special
classes are truly beneficial, but there is a 5% chance that the classes are not beneficial.
Answer:

The statement is FALSE


With the confidence level 5%, the regression statistics is given below

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.372104204
R Square 0.138461538
Adjusted R Square -0.29230769
Standard Error 1.467598771
Observations 4

ANOVA
df SS MS F Significance F
Regression 1 0.692307692 0.692308 0.321429 0.627895796
Residual 2 4.307692308 2.153846

Total 3 5

Standard
Coefficients Error t Stat P-value Lower 95% Upper 95% Lower 5.0% Upper 5.0%
Intercept 1.769230769 3.139743328 0.563495 0.62985 -11.73999443 15.27845597 1.546939351 1.991522187
X Variable 1 0.230769231 0.407038663 0.566947 0.627896 -1.520576785 1.982115246 0.201951206 0.259587256

We cannot conclude that there is only 5% chance that the classes are not beneficial.
The reason is, the decision made using significance level α = 0.05 can be incorrect. And
there are high changes to mistakenly reject the true null hypothesis. This type of error is
called Type 1 error.
TRUE FALSE
Fail to Reject Correct Type II error
Reject Type I error Correct

The probability of committing the type I error is measured by the significance level (α) of a
hypothesis test. The significance level indicates the probability of erroneously rejecting the
true null hypothesis.
One of the most common approaches to minimizing the probability of getting a false
positive error is to minimize the significance level of a hypothesis test.

You might also like