Errors in Hypothesis Testing

Errors in Hypothesis Testing

OPTIMIZE :
Can you still recall how to find the probability of the sample means? Find out by answering the following items below.
Given σ = 9.89, µ = 87, and n = 50
1. P (X > 85) = ______________

2. P (X > 88) = ______________

3. P (X > 90) = ______________

4. P (X < 83) = ______________

5. P (84 < X > 89) = ______________

Rex Curriculum Resource WWW.REX.COM.PH

NAVIGATE :
Example 2.1: Joy wants to test if the average allowance of senior
high school students in a private school is less than ₱2,000 per
From the previous lesson, you were able to differentiate the
week with a standard deviation of ₱750. From a random sample of
errors in hypothesis testing. Type I error is committed if the null 50 senior high school students, the sample mean is ₱1,750.
Supposed that the actual population is ₱2,000, what is the
hypothesis is rejected when it is true, while type II error is
probability that Joy concludes that the average allowance of
committed if the null hypothesis is accepted when it is false. The senior high school students is less than ₱2,000 at a 5% level of
significance?
question now is, when do you commit type I and type II errors?
Solution:
This is also the same as asking the probability of committing type I 1. Formulate the null and alternative hypotheses.
H0: µ = 2,000 H1: µ < 2,000
and type II errors. Take note of the following notations below.
2. The probability that Joy concludes that the average
allowance of senior high school students is less than ₱2,000 refers
Notations:
to the Type I error. Take note that the true population means is
P (Type I Error) = P (Reject H0|H0 is true) = α 2,000, which was rejected.
3. By definition of level of significance, it is the probability of
P (Type II Error) = P (Accept H0|H0 is false) = β rejecting the null hypothesis when it is true, and it is denoted by
a. Thus, the probability that Joy concludes that the average
allowance of senior high school students is less than ₱2,000 is
Take note that the probability of committing a type I error is 5%. In symbols:
denoted by α (alpha), and the probability of committing a type II P (Type I Error) = P (Reject H0|H0 is true) = α = 5%
error is denoted by β (beta).

Rex Curriculum Resource WWW.REX.COM.PH

NAVIGATE : Given:

Example 2.2: Suppose a manufacturer claims that his newly X = 16.23, σ = 3.15, n = 60, and m = 17 since H0 is
developed fishing line has a mean breaking strength of 17 true.
kilograms with a standard deviation of 3.15 kilograms. A random Since the sample size is greater than 30, Central Limit Theorem
sample of 60 fishing lines will be tested to test the hypothesis H0: can be applied, and you can use the formula:
m = 17 against H1: m < 17. The critical region is at X < 16.23. to convert the raw score to its corresponding z-score.
Find the probability of committing a type I error when H0 is true.
Formula to convert a raw score to its
Solution:
corresponding z-score.
1. Formulate the null and alternative hypotheses.
H0: µ = 17
Substitute the variables to
H1: µ < 17 their corresponding values.
2. Since the level of significance is not given and we Simplify.
want to the probability of committing a type I error when H0 is
true, we need to solve for m.
The following are the steps in finding the probability of committing Step 2: Find the probability of the z-score computed
a type I and type II error. from the previous step.
Take note that P(X < 16.23) = P(Z < -
a. Find the corresponding z-score of the sample mean 1.89)
(X). Using the z-table (area between 0 and a z-score),
b. Find the probability of the z-score computed from P(Z < -1.89) = 0.5 - 0.4706
the previous step. P(Z < -1.89) = 0.0294
c. Write the final answer. Step 3: Write the final answer.
P(Type I error) = P(Z < -1.89) = 0.0294
Therefore, there is approximately a 2.94% chance of
committing a type I error.
Rex Curriculum Resource WWW.REX.COM.PH
 Errors in Hypothesis Testing

NAVIGATE :
Example 2.4: Edward wants to test if the average weight
of the Grade 12 students is 68 kilograms with a standard deviation
of 11 kilograms. A random sample of 55 students was taken with a Since the means at the critical values were already computed,
mean of 60 kilograms. If the true population mean is 70 kilograms, we can now compute the z-score given the true population mean.
what is the probability of committing a type II error at 0.05 level of
significance?

Solution:
Step 1. State the null and alternative hypotheses.
H0 = 68 kg
H1 ≠ 68 kg Take note that the test to be used in this example is a two-tailed test.
Step 2. Finding the probability of committing a type II error Take note that committing a type II error means that the null
Find the corresponding z-score of the hypothesis is accepted even if it is false. If we reject the null
sample mean (X). hypothesis even if it is true, the probability will be denoted
The two means corresponding to the critical values under as P(Z < -3.31) or P(Z > 0.61). Therefore, its opposite will be
the normal curve are not given but can be computed since the P(-3.31 < Z < 0.61), which is the probability of committing the type II
level of significance is given at 0.05. If a = 0.05, then the z-values error.
are 1.96 and -1.96 Using the z-table (area between 0 and z-scores)
P(-3.31 < Z < 0.61) = 0.4999 + 0.2291
Compute for the means given the z-values using the formula: P(-3.31 < Z < 0.61) = 0.729 or 72.90%
Step 3: Write the final answer.
Therefore, the probability of committing a type II error is 72.90%.
Were you able to understand how to get the probability of type I and
type II errors? If there are unclear concepts to you, do not hesitate to
Given: ask your teacher.
µ = 68 σ = 11 z1 = 1.96 n = 55 z2 = 1.96
Rex Curriculum Resource WWW.REX.COM.PH
 KEY TAKEAWAYS
Here are the steps in calculating the probability of committing a type I and type II error.

1. Find the corresponding z-score of the sample mean (X).

2. Find the probability of the z-score computed from the previous step.

3. Write the final answer.

Rex Curriculum Resource WWW.REX.COM.PH

CHECKPOINT

Answer the given problem below.

Mary wants to test if the average number of years that students can finish college is less than 6 years with a standard
deviation of 0.70 years. Suppose the actual population mean is greater than 6 years, what is the probability of Mary’s claim that the
average residence of college students is less than 6 years at 10% level of significance?

Rex Curriculum Resource WWW.REX.COM.PH

OPTIMIZE

1. 0.9422
2. 0.2389
3. 0.0162
4. 0.0021
5. 0.9074
NAVIGATE

Checkpoint

a = 10%

Rex Curriculum Resource WWW.REX.COM.PH