HYPOTHESIS TESTING, Z-test, and T-test

➢ Inferential Statistics
- It is a formalized body of techniques used to make conclusions about populations based on
samples taken from the population.
➢ Hypothesis
- It is a conjecture or a statement which aims to explain certain phenomenon in the real world.
Statistical or not, many hypotheses are products of man’s curiosity.
• Null Hypothesis (Ho) - in statistical inference is the original statement about a parameter (or
parameters).
• Alternative Hypothesis (Ha) - is the opposite of the null hypothesis that challenges the status
quo.
➢ Significance Level
- The probability of making Type I error and it is denoted by the symbol 𝜶.
• Type I error – is the probability of accepting the alternative hypothesis (H a) when in fact the
null hypothesis (Ho) is true.
• Type II error – The probability of accepting the null hypothesis (H o) when in fact it is false. It
is denoted by the symbol 𝜷. The most commonly used level of significance is 5%.
➢ One-tailed and Two-tailed Tests
• One-tailed – Ha: 𝑥̅ > 𝜇 or Ha: 𝑥̅ < 𝜇
• Two-tailed – Ha: 𝑥̅ = 𝜇 or Ha: 𝑥̅ ≠ 𝜇
➢ Testing Hypothesis
• The following are the steps in testing the truth of a hypothesis.
1. Formulation of the Null and Alternative Hypothesis
- A null hypothesis (Ho) is a statement being tested. It is assumed as the true value of the
parameter being tested and usually, it is “hoped” to be rejected.
- An alternative hypothesis (Ha) is a set of possible values about the population parameter.
There are three ways of stating the alternative hypothesis.
• Ha: Ө ≠ Өo (two – tailed test)
• Ha: Ө > Өo (one – tailed test)
• Ha: Ө < Өo (one – tailed test)
2. Choosing the statistical test and determining the sampling distribution
- The selection of the statistical test depends on the manner in which the sample is drawn
from the population, the nature of the population, and the level of measurement used in
the collection of data.
- Once the test statistic is chosen, the sampling distribution is determined so that the
probability statements about the occurrence or non-occurrence of its values can be
obtained.
3. Determining the rejection region or critical region
4. Data generation and computation

Z-TEST on the Comparison between the Population Mean and Sample Mean
➢ The formula for z-test when the population mean (𝜇) and the population standard deviation (𝜎) are
known, and 𝜇 is compared to a Sample Mean (𝑥̅ ) is where
̅ − 𝝁)
(𝒙
𝒛= ∙ √𝒏 𝑥̅ = sample mean
𝝈
𝜇 = population mean
n = sample size
𝜎 = population standard deviation

➢ The tabular values or critical values of z is obtained from the following table:

Level of Significance
Test Type 0.10 0.05 0.025 0.01
One-tailed Test ± 𝟏. 𝟐𝟖 ± 𝟏. 𝟔𝟒𝟓 ± 𝟏. 𝟗𝟔 ± 𝟐. 𝟑𝟑
Two-tailed Test ± 𝟏. 𝟔𝟒𝟓 ± 𝟏. 𝟗𝟔 ± 𝟐. 𝟑𝟑 ± 𝟐. 𝟓𝟖
 Note:
z – test (𝒏 > 𝟑𝟎)
t – test (𝒏 < 𝟑𝟎)

̅ >/< 𝝁)
One-tailed test (Ha: 𝒙
̅ =/≠ 𝝁)
Two-tailed test (Ha: 𝒙

Critical region for z-test: refer to the table above

Degrees of freedom for t-test: 𝑑𝑓 = 𝑁 − 1

Example 1 (z-test):

A company who makes children’s battery-operated toy cars, claims that its products have a mean life
span of 5 years with a standard deviation of 2 years. Test the hypothesis that 𝜇 = 5 years against the
alternative hypothesis that 𝜇 = 5 years if a random sample of 40 toy cars is tested and found to have
a mean life span for only 3 years. Use 0.05 level of significance.

Solution:

Steps:
1. Ho: The mean life span of toy cars is for only 5 years. (𝜇 = 5)
Ha: The mean life span of toy cars is not for 5 years. (𝜇 ≠ 5)
2. 𝛼 = 0.05 𝑜𝑟 5%; two-tailed
3. Use the z-test for test statistic
4. Computation
(𝑥̅ − 𝜇)
𝑧= ∙ √𝑛
𝜎
(3 − 5)
𝑧= ∙ √40
2
𝑧 = −6.32
5. Critical region: ±1.96 (given)
6. Decision:
Reject Ho
Accept Ha because the mean life span of the toys is not equal to 5 years
T-TEST on the Comparison between the Population Mean and Sample Mean
➢ When the population standard deviation is unknown but the sample standard deviation (s) can be
computed, the t-test can be used instead of the z-test. The formula follows:
(𝒙
̅ − 𝝁)
𝒕= (√𝒏)
𝒔

where
𝑥̅ = sample mean
𝜇 = population mean
n = random samples
s = standard error / deviation

The denominator of the formula, s, divided by the √𝑛 for t is called the

standard error of the statistics. It is the standard deviation of the sampling
distribution of statistics, for random samples n.
Example 2 (t-test):
The average length of time for people to vote using the old procedure during the presidential election
precinct A is 55 minutes. Using computerization as a new election method, a random sample of 20
registrants was used and found to have a mean length of voting time of 30 minutes with a standard
deviation of 1.5 minutes. Test the hypothesis in which the population mean is greater than the sample
mean.
Solution:
Steps:
1. Ho: 𝑥̅ = 𝜇
Ha: 𝑥̅ < 𝜇
2. 𝛼 = 5%; one-tailed
3. 𝑡−test is the appropriate test statistics
4. Computation:
(𝑥̅ − 𝜇)
𝑡= (√𝑛)
𝑠
(30 − 55)
𝑡= √20
1.5
𝑡 = −74.54
5. 𝑑𝑓 = 𝑛 − 1
𝑑𝑓 = 20 − 1
𝑑𝑓 = 19
6. Tabular Value = 1.729 (one-tailed) (given)
7. Decision:
Accept Ha
Reject Ho since the computed absolute value is greater than the tabular value
8. There is a significant difference between the means.