CHAPTER 6

Hypothesis Testing

The topics that we have discussed so far are classified under

descriptive statistics. They give helpful information about the samples
under investigation. The measures of central tendency represent a typical
value, or it describes the central location of the data. Thus, if the mean
grade of the student is 80, it describes his average performance and what
kind of student he is. The measure of dispersion indicates the variability of
the data about a sample.
However, for a particular investigation, we are interested in
describing the data set and would like to go on further analysis. Many
problems require that we decide whether or not a statement about the
mean or the standard deviation is true or false. The statement is usually
called a hypothesis, and the decision-making procedure about the truth or
falsity of the hypothesis is called hypothesis testing.
This module will present the steps involved in the hypothesis
testing procedure for some important investigations related to your major
discipline.

1. Differentiate null hypothesis from alternate hypothesis.

2. Formulate the statistical hypotheses Ho and Ha for a given
situation.
3. Compare a directional test from a nondirectional test.
4. Identify Type I and Type II errors and recognize consequences in
such errors.

STATISTICAL HYPOTHESIS
A statistical hypothesis is a statement about the probability
distribution of a random variable. Likewise, it is a preconceived idea about
the value of a population parameter which is verified through statistical
procedures or tests. Statistical hypothesis involves one or more parameters
of a given probability distribution. For example, suppose we are interested
in the mean performance of principals. Specifically, we are interested in
deciding whether or not the mean performance of principals (say ) is 85.
We may express this formally as:
Ho:  = 85 versus Ha:   85
The statement Ho:  = 85 is called the null hypothesis, and the
statement Ha:   85 is called the alternative hypothesis.
79
 The null hypothesis Ho:  = 85 can be accurately verified by
studying the entire population of principals to determine if the value of the
population mean it specifies is true. If the whole population is studied, the
investigator will not make a wrong decision about the null hypothesis.
Similarly, we can also hypothesize that the two groups of the
population are the same or different, and we may express this as
Ho: 1 = 2 versus Ha: 1 ≠ 2.
In most cases, however, the whole population cannot be studied
due to time, financial, and other constraints. In such a case, investigators
resort to taking or selecting a random sample from the population being
studied. We either accept or reject the null hypothesis based on the results
gathered. If the information gathered is consistent with the null hypothesis,
we would conclude that the null hypothesis is true, leading to the
acceptance of the null hypothesis. However, if the information gathered is
inconsistent with the null hypothesis, we would conclude that the null
hypothesis is false, which leads us to reject the null hypothesis. If the null
hypothesis is rejected, an alternative hypothesis is accepted (the
alternative hypothesis include all values not specified by the null
hypothesis).

DIRECTIONAL AND NON-DIRECTIONAL TESTS OF HYPOTHESIS

When a researcher hypothesizes that individualized instruction is
the same as that of the lecture method, he actually meant that students
under the individualized method perform similarly to those subjected to
the lecture method. Thus, the hypothesis can be stated in the following
manner: the mean score of students under the individualized instruction
method does not differ significantly from that of the lecture method. This
can be expressed as
Ho: 1 = 2
Where 1 represents the mean score of students for the
individualized instruction while 2 represents the mean score of students
under the lecture method.
There are three possible alternative hypotheses (Ha) to the null
hypothesis:
1. Ha: 1 ≠ 2
2. Ha: 1 > 2
3. Ha: 1 < 2
The first alternative hypothesis allows for the possibility that either
1 is greater than or less than 2. In this case we are making a non-
directional (two-tailed) test. The second alternative hypothesis states
specifically that 1 is greater than 2 and similarly for the third alternative
hypothesis where it is stated specifically that 1 is less than 2. In this
case we are making a directional (one-tailed) test.

80
 The null and alternative hypotheses are stated together, and the
null hypothesis contains the equals sign, as shown (where 𝑘 represents a
specific number)
Two-tailed Right-tailed Left-tailed
𝐻0 : 𝜇 = 𝑘 𝐻0 : 𝜇 ≤ 𝑘 𝐻0 : 𝜇 ≥ 𝑘
𝐻1 : 𝜇 ≠ 𝑘 𝐻1 : 𝜇 > 𝑘 𝐻1 : 𝜇 < 𝑘

Level of Significance
In any kind of hypothesis test, we always ask whether the
observed difference between the two groups of respondents is significant
or merely by chance. Answer to such query involves using probability level
 the level of significance of a test which the researcher sets before
gathering the sample data. Most researchers usually set the level of
significance of a test of hypothesis at 0.05 (5%) or 0.01 (1%).
A significance level of 0.05 or 5% means that in a repeated
sampling from a given population of interest, the probability of obtaining
sample results similar to the one presently observed is 95%, and getting
different sample results is 5%. A similar interpretation can be made
regarding a 1% level of significance. Significance at a certain level of 
means that the probability of occurrence of the observed sample results
due to plain chance or sampling error is equal to or less than . The smaller
the probability is (or the smaller the value of ) the more significant are
the sample results.

Types of Errors
In the hypothesis test, we always commit two types of errors: the
Type I error and the Type II error.
A Type I error occurs if the null hypothesis is rejected when it
should be accepted. It is denoted by the  or the level of significance while
1 -  is known as the confidence level. On the other hand, if the null
hypothesis is accepted when it should be rejected, we are committing a
Type II error. We denote it by  or operating characteristic curve of the
test while 1 -  is known as the power of the test.
A correct decision in hypothesis testing is made if the null
hypothesis is accepted when it is true and should therefore be accepted.
The probability of making this correct decision is 1 - . Hence if the adopted
level of significance in conducting a test of hypothesis is 0.05, then the
probability of correctly accepting a true hypothesis is 0.95.
A second correct decision in hypothesis testing is made if the null
hypothesis is rejected when it is really false and should be rejected. The
probability of making this correct decision is 1 - . The power of a test is
its ability to discriminate a true from a false hypothesis. In general, the
power of a test increases as the number of observations in the sample
increases.
81
 𝐻0 True 𝐻0 False

Error Correct
Reject 𝐻0
Type I Decision

Error
Do not Reject 𝐻0 Correct Decision
Type II

Steps of Hypothesis Testing

In performing hypothesis testing, the following suggestive steps
should be followed:
Step 1: State the null and alternative hypotheses and identify the
claim if needed.
Step 2: Find the critical value(s) from the appropriate table.
Step 3: Compute the test value.
Step 4: Make the decision "to reject" or "not to reject" the null
hypothesis.
✓ Reject the null hypothesis if the absolute computed value
(𝐶𝑉) is greater than or equal to the tabular value (𝑇𝑉). In
symbols, (|𝐶𝑉| ≥ 𝑇𝑉).
✓ Do not reject the null hypothesis if the absolute computed
value (𝐶𝑉) is less than the tabular value (𝑇𝑉). In symbols,
(|𝐶𝑉| < 𝑇𝑉).
Step 5: Make a conclusion.
✓ If you reject the null hypothesis, then the conclusion is the
alternative hypothesis.
✓ If you do not reject the null hypothesis, then the conclusion
is the null hypothesis.

82