Computer II – Introduction to Statistical Computing

UNIT 2
Using MS Excel for Hypothesis Testing

Introduction to Hypothesis Testing

When data are gathered with appropriate procedures, they are to be analyzed using appropriate statistical
tools. Again, these series of procedures that is from the selection of appropriate of sampling technique until the
analysis of data, must be religiously done for in its absence your efforts would be useless.
Inferential statistics demands higher order of critical thinking and mathematical methods. Its main purpose
is draw conclusions or judgments regarding an entire set of observation by using only a small representative of the
whole set. One is trying to reach conclusions that extend beyond the immediate data alone. Like for example,
making judgments of the probability that an observed difference between groups is a dependable one or one that
might have happened by chance in a specific study.
In this case, it is understood that you already have a thorough knowledge on the different measures of
central tendency or position and measures of variability.

What is a Hypothesis?
A hypothesis is an educated guess. It is a conjecture or proposition about the solution to a problem, or the
status of a situation based on the available facts or information that the researcher already knows.
Hypotheses are constructed anywhere and anytime of the day. For example, if one of your students fails to
show up of her class on a Monday morning, the teacher could make conjectures as to why she failed to attend her
class.

Possible problem that can be formulated from the situation is:

What are the factors that contribute to the absence of employee A on Monday morning?
From the problem, guesses can be made such as:
*the employee is sick
*the employee meets an accident
*there is a family problem
*etc
Some hypothesis could be rejected outright since they have no important bearing in solving the problems.
Others can also be accepted depending how related they are in explaining or giving answers to the problem.
Consequently, we can either accept or reject a hypothesis
Hypothesis can be derived directly from the statement of the problem or from the related literature
reviewed.

For example the research problem is:

Is there a relationship between attendance and performance of employees in ABC Company?

The corresponding hypothesis can be stated as:

There is a positive relationship between attendance and performance of employees in ABC Company.

Types and Forms of Hypothesis

A research or alternative hypothesis is a tentative explanation for certain behaviors, phenomena or events,
which have occurred or will occur. It is the expectations concerning the relationship between variables or statements
specifying the relationship between two or more variables.
A null hypothesis is a statement of equality between population parameters (or no difference between) or a
statement of no relationship between variables or no interaction between variables. The null hypothesis is usually a

Page 1
Computer II – Introduction to Statistical Computing

negation of the research or alternative hypothesis intended for testing purposes and this is denoted by the symbol,
H o . The null hypothesis is sometimes called the statistical hypothesis.
Technically speaking, when a statistical hypothesis is being tested using inferential statistics, it is the null
hypothesis that is being tested. When the null hypothesis is being tested, it is either accepted or rejected. When it is
rejected, the statement that is accepted is the research or alternative hypothesis.
The research or alternative hypothesis could either be directional or non- directional.
A directional alternative hypothesis states the direction of the difference of two population parameters or
direction of relationship between variables.
A non- directional alternative does not specify the direction of the difference or direction of relationship
between variables.
The research or alternative hypothesis is formulated based on the review of related literature. Hence, a
directional research or alternative hypothesis is formulated if most of the research studies reviewed seems to support
such research hypothesis. On the other hand if research findings are contradictory, the non- directional type of
research hypothesis research may be formulated.

Example:
Research Question
Does attendance to training and seminars make teachers more productive in their work performance than
non- attendance to trainings and seminars?

Directional Alternative Hypothesis

The work performance of teachers attending training and seminars is significantly better than the work
performance of teachers who are not attending trainings and seminars.

Non- directional Alternative Hypothesis

There is significant difference between the work performance of teachers attending training and seminars
and those who are not attending training and seminars.

Criteria for a Good Hypothesis

Four criteria that hypothesis should satisfy (Borg & Gall, 1989)
1. The hypothesis should be an expected relationship between two or more variables.
2. The researcher should have definite reasons based on either theory or evidence for considering the hypothesis
worthy of testing.
3. A hypothesis should be testable.
4. A hypothesis should be brief, consistent and clear.
Type 1 and Type 2 Errors
When a null hypothesis is subjected to a statistical test, the decision maker always runs the risk of
committing error in rejecting or accepting the null hypothesis. The table below shows when these errors committed
are.
Table 4.1. Type I and type II Error
Decision Actual Condition Type of Error
Reject H o H o istrue Type I Error
Accept H o H a istrue Type II Error

Type I error or α-error occurs when the null hypothesis is rejected when in fact it is true and therefore the
alternative or research hypothesis is false. On the other hand, Type II error is committed when the null hypothesis is
accepted when in fact is false, and therefore the alternative or research hypothesis is true.

Page 2
Computer II – Introduction to Statistical Computing

Level of Significance

The probability of making a Type I error is a test called level of significance. It has the maximum value
of the probability of rejecting the null hypothesis H o when in fact it is true.
The customary level of significance used is either 5% or 1%. A 5% level of significance means that we can
accept about 5 chances in 100 that we would reject the null hypothesis when it should be accepted. This implies that
we are 95% confident that we have made the right a decision. A 1% significance level means that we could be
wrong with a probability of 0.01. In other words, a 1% significance level implies that we are 99% confident that we
have made the right decision.

One- tailed and Two-tailed Test

When the rejection region is located at only one extreme of the range of values for the test statistics, the test
is one- tailed. When the rejected region is located on both sides of the tails of the distribution, the test is called two-
tailed test.

Factors to Consider in Choosing Appropriate Statistical Tool

1. Level of measurement used in measuring the variables.
2. Objectives of the study
3. Design used in the study

Further, appropriate tools to be used depend on whether a one-sample, two-sample or multiple-sample

analyses are done.

Review Questions :
1. How does null hypothesis differs from alternative hypothesis ? Give an example of each.
2. What is meant by type 1 error? A type II error?
3. What is meant by statistical test ?
4. Explain the difference between one-tailed and two-tailed test.
5. What are the symbols used to represent the null and alternative hypothesis ?
6. Explain what is meant by level of significance.
7. When can you say that a study is one-tailed test? A two-tailed test ?
8. State the null and alternative hypotheses for each conjecture.

a. A researcher thinks that if expectant mothers use vitamin pills, the birth weight of the babies will increase. The
average birth weight of the population is 8.6 pounds.

b. An engineer hypothesizes that the mean number of defects can be decreased in a manufacturing process of
compact disks by using robots instead of humans for certain tasks. The mean number of defective disks per 1000 is
18.

c. A psychologist feels that playing soft music during a test will change the results of the test. The psychologist is
not sure whether the grades will be higher or lower. In the past, the mean of the scores was 73.

P-Value Method for Hypothesis Testing

• Statisticians usually test hypotheses at the common levels of 0.05 or 0.01 and sometimes at 0.10.
• The P-value (or probability value) is the probability of getting a sample statistic (such as the mean) or a
more extreme sample statistic in the direction of the alternative hypothesis when the null hypothesis is true.
• For example, suppose that an alternative hypothesis is H1: mean= 50 and the mean of a sample is 52.

Page 3
Computer II – Introduction to Statistical Computing

• If the computer printed a P-value of 0.0356 for a statistical test, then the probability of getting a sample
mean of 52 or greater is 0.0356 if the true population mean is 50 (for the given sample size and standard
deviation).
• The relationship between the P-value and the alpha value can be explained in this manner.
• For P = 0.0356, the null hypothesis would be rejected at a 0.05 but not at a 0.01.

Decision Rule When Using a P-Value

• If P-value <= a, reject the null hypothesis.
• If P-value > a, do not reject the null hypothesis.
• For p= 0.0356, the null hypothesis would be rejected at alpha= 0.05 but not at alpha= 0.01.
•  When the hypothesis test is two-tailed, the area in one tail must be doubled.
• In summary, then, if the P-value is less than or equal to the level of significance, reject the null hypothesis.

• State whether the decision is to reject Ho or not.

• P-value 0.258, a = 0.05, one-tailed test - accept Ho
• P-value 0.0684, a = 0.10, two-tailed test – reject Ho
• P-value 0.0153, a = 0.01, one-tailed test – accept Ho
• P-value 0.0232, a = 0.05, two-tailed test – reject Ho
• P-value 0.002, a 0.01, one-tailed test – reject Ho

Page 4