Hypothesis Testing and Some of its Fundamental Concepts
Hypothesis is a statement regarding a single or multiple populations. A
hypothesis often focuses on the parameters of the populations about which the
statement has been developed. On the other hand, hypothesis testing is the
process of determining the probability of a hypothesis to be true.
Hypothesis testing is considered as a very important factor as far as Six Sigma
is concerned. The hypothesis test is designed in such a way that it can make an
inference regarding the actual population value at a significant confidence
level.
Let’s consider a scenario to have clear understanding about the concept. Let’s
assume that the administrator of “ABC Hospital” has hypothesized that in
average, each patient that gets admitted to this hospital stays there for five
days. Now the role of the hypothesis testing would be to reveal that whether the
hypothesis developed by the administrator is compatible or not with the
available data or information.
Hypothesis Testing
Studying the data
It is important to make sure that the nature of the data which forms the basis of
the testing procedure is properly understood considering the fact that this is
something which determines the specific test that is to be used in
six sigma projects. The analysts need to determine whether the data involves
counts or measurements.
Setting up a hypothesis
The first step of conducting hypothesis test is to set up a hypothesis regarding
a population parameter. When it comes to the hypothesis testing, it involves two
statistical hypotheses. The first one is called “null hypothesis” and another
one is the “alternative hypothesis”.
Now the null hypothesis refers to that particular hypothesis which is required to
be tested designated by the symbol H 0. This is considered as a hypothesis of no
difference. In the testing process, a null hypothesis either gets rejected or
not rejected. If the null hypothesis is rejected then it means the data on which
the test is based on is not compatible with the null hypothesis. On the other
hand, the hypothesis not getting rejected will refer that not enough evidence
has been provided by the data (on which the test is based on) which can cause
the rejection.
Alternative hypothesis refers to a particular statement which will be considered
as true in cases where the null hypothesis gets rejected based on the sample
data. The alternative hypothesis is designated by H A. It is important to keep
in mind that the null hypothesis and the alternative hypothesis are
complementary.
Understanding the Level of Significance
The level of significance is the probability of rejecting a true null hypothesis.
Level of significance is designated as α. It is the desired level of
significance based on which decision is made regarding which values will fall
into the rejection region and which will fall into the non-rejection region. It
is important to keep in mind that the hypothesis tests are sometimes termed as
“significance test” and the computed value of the test statistic which falls in
the rejection region is termed as “significant”. In an attempt to minimize the
probability of rejecting a true null hypothesis, a small value of the level of
significance is selected. Some of the commonly used values include .01, .05 and
.10.
|