The Null and Alternative Hypothesis
Hypothesis testing- Is a decision- making process for evaluating claims about a population based on the
characteristics of sample purportedly coming from that population.
Null Hypothesis-No relationship or difference between two or more variables. Alternative
Hypothesis-Having relationship or difference between two or more variables.
ILLUSTRATIVE EXAMPLES:
1.Claim: The average monthly income of Filipino families who belong to low-income bracket is
Php8,000.
Answer:
Null Hypothesis (Ho) Alternative Hypothesis (Ha)
Ho: The average monthly income of Filipino Ha: The average monthly income of Filipino
families who belong to low-income bracket is families who belong to low-income bracket is
Php8, 000. not equal to Php8, 000.
≠
In symbol, In symbol,
Ho: μ = 8,000 Ha: μ
8,000
2.The inventor of the new kind of light bulb claims that all such bulb last 3000 hours.
Answer:
Null Hypothesis (Ho) Alternative Hypothesis (Ha)
Ho: The life of a light bulb is equal to 3000 Ha: The life of a light bulb is less than 3000
hours. hours.
Ha: μ < 3,000
In symbol, In symbol,
Ho: μ = 3,000
In this example, greater than 3000 hours doesn’t make sense since the bulb maximum life is 3000 hours
based on the claim, therefore the alternative hypothesis should be less than 3000 hours.
3. A random sample of 200 students got a mean score of 62 with a standard deviation of 5 in knowledge
test in mathematics. In the standardization of the test,𝜇 = 50 with 𝜎 = 10.
Answer:
Null Hypothesis (Ho) Alternative Hypothesis (Ha)
Ho: The mean score in knowledge test in Ha: The mean score in knowledge test in
In symbol, Ha: μ ≠ 3,000
mathematics is equal to 62. mathematics is not equal to 62.
In symbol, Ho: μ = 62
Definition 1.2: A statistical hypothesis is a prediction
regarding the possible outcome of a study. It
can be shown to be supported or not
supported.
The null hypothesis, denoted by 𝐻𝑂, is the hypothesis to be tested. It has a statement of
There are two types of statistical hypotheses: the null hypothesis and the alternative hypothesis.
equality, such as ≥, ≤, or =. On the other hand, the alternative hypothesis, denoted by 𝐻𝐴, is
the hypothesis that has no statement of equality, such as >, <, or ≠.
In formulating the null and alternative hypotheses, you should first define your research problem.
Using the example in Warm Up!, the research problem can be stated as:
Do Filipinos spend a daily average time of 4 hours on social media?
of 15 Filipino students (the sample). We can then compare the sample mean, which is the mean
In this problem, the two groups being considered are the Filipinos (the population) and the group
measured from the data of 15 students, to the population mean stated in the article, which is 4
hours a day.
Since the null hypothesis always shows equality, we just have to express that there is no
difference between the two means being compared. Similar to possible outcome (a) in Warm
Up!, the null hypothesis should be stated as:
𝐻𝑂: The daily average time spent by Filipinos on social media is 4 hours a day. (𝐻𝑂: 𝜇 = 4)
After stating the null hypothesis, the alternative hypothesis can be formulated. The
alternative hypothesis is the opposite of the null hypothesis; there should be a difference
between the two means being compared. Similar to possible outcome (b) in Warm Up!, it
should be stated as:
𝐻𝐴: The daily average time spent by Filipinos on social media is not equal to 4 hours a day.
(𝐻𝐴: 𝜇 ≠ 4)
Any claim can be classified under either the null hypothesis or the alternative hypothesis. Each
hypothesis is the complement of the other. If the null hypothesis is rejected, the alternative
hypothesis is accepted, and if the null hypothesis is not rejected, it follows that the alternative
hypothesis is not accepted.
The manner in which the alternative hypothesis is stated determines the type of
hypothesis test to be used.
Types of Hypothesis Testing
There are two basic types of hypothesis testing procedures depending on the alternative
hypothesis.
1. The directional test of hypothesis, more commonly referred to as one-tailed test, makes
use of only one side or tail of the statistical model or distribution. It can either be a right-
tailed or left-tailed test.
a. Right-tailed test: It is used when an assertion is made that the difference falls within the
positive end of the distribution. The alternative hypothesis uses comparatives such as
greater than, higher than, better than, superior to, exceeds, above, increased, etc.
b. Left-tailed test: It is used when an assertion is made that the difference falls within the
negative end of the distribution. The alternative hypothesis uses comparatives such as
less than, smaller than, inferior to, lower than, below, decreased, etc.
2. The non-directional test of hypothesis, more commonly referred to as two-tailed test,
makes use of two opposite sides or tails of the statistical model or distribution. It
is used when no assertion is made as to whether the difference falls within the positive or the
negative end of the distribution. The alternative hypothesis uses comparatives such as not
equal to, different from, not the same as, etc.
Using the given research problem in Warm Up!, let us illustrate the different type of hypothesis
test.
1. right-tailed test
𝐻𝑂: The daily average time spent by Filipinos on social media is at most 4 hours a day.
(𝐻𝑂: 𝜇 ≤ 4)
𝐻𝐴: The daily average time spent by Filipinos on social media is more than 4 hours a day.
(𝐻𝐴: 𝜇 > 4)
2. left-tailed test
𝐻𝑂: The daily average time spent by Filipinos on social media is at least 4 hours a day.
(𝐻𝑂: 𝜇 ≥ 4)
𝐻𝐴: The daily average time spent by Filipinos on social media is less than 4 hours a day.
(𝐻𝐴: 𝜇 < 4)
3. two-tailed test
𝐻𝑂: The daily average time spent by Filipinos on social media is 4 hours a day.
(𝐻𝑂: 𝜇 = 4)
𝐻𝐴: The daily average time spent by Filipinos on social media is not equal to 4 hours a
day. (𝐻𝐴: 𝜇 ≠ 4)
Level of Significance
The level of significance, denoted by 𝛼 (alpha), is the probability of rejecting the null
hypothesis in favor of the alternative hypothesis when it is really true.
the test. Conventional significance levels such as 0.05 and 0.01 are frequently used in
In hypothesis testing, the researcher decides what level of significance to use at the beginning of
hypothesis testing because of the desire to maintain a low probability of rejecting the null
hypothesis when it is actually true.
An 𝛼 = 0.05 means that there is a 5% chance of rejecting a true null hypothesis. Here is
another way of putting it: A 5% significance level implies that we are 95% confident that we
have made the right decision of rejecting the null hypothesis. The similar argument follows for 𝛼
= 0.01.
Example 1: A psychologist thinks that playing soft music during a
mean score in statistics tests of students was 82. State the null and
Statistics test can improve scores of students. It is known that the
alternative hypotheses.
Solution: Let s𝜇 be the mean score of students who listened to soft music during statistics tests.
Thus, the hypotheses can be stated as:
𝐻𝑂: The mean score of students who listened to soft music during statistics tests is not higher than
82. (𝐻𝑂: 𝜇 ≤ 82)
𝐻𝐴: The mean score of students who listened to soft music during statistics tests is higher than 82.
(𝐻𝐴: 𝜇 > 82)
The alternative hypothesis requires a one-tailed test, particularly a right- tailed test, since the
psychologist believes that playing soft music improves student grades.
Example 2: A local traffic enforcement officer
believes that the number of vehicular
accidents per day in the city was
reduced after the increase in fine for
traffic violations. Prior
average of 23 vehicular accidents a
to the increase, the city reported an
day. State the null and alternative
hypotheses.
Solution:Let average 𝜇 be the number of vehicular accidents after the increase in fine for traffic
violations. Thus, the hypotheses can be stated as:
𝐻𝑂: The average number of vehicular accidents after the increase in fine for traffic violations is not
less than 23. (𝐻𝑂: 𝜇 ≥ 23)
𝐻𝐴: The average number of vehicular accidents after the increase in fine for traffic violations is less
than 23. (𝐻𝐴: 𝜇 < 23)
The alternative hypothesis requires a one-tailed test, particularly a left- tailed test, since
the enforcement officer believes that the increase in fine for traffic violations reduced
the number of vehicular accidents.