Page 275 - Area 9 - Relevant Document

22

THE TWO TYPES OF ERROR

The chance exist that rejecting the null hypothesis is wrong .It is always
possible that a statistically rare occurrence of chance may be obtained. The
findings or propositions based on experiment can never be asserted to be true
with complete confidence. The chance of making wrong conclusions always
exists.

From these considerations, there are two types of errors that maybe made in
statistical inference.

1. Type I Error (Type a error)
In a type one error, we reject the null hypothesis when it is actually
true. The probability of making a type one error is a. It is when we set
our rejection point at 5 percent level significance. We will mistakenly
reject H approximately 5 percent of the time.

2. Type II Error (Type b error)
In a type two error, we fail to reject the null hypothesis when it is
actually false. The probability of making a type two error is b. This type
of error is far more common than a type one error.

It is clear, then, that the lower rejection level set, the less is the
likelihood of a type one error, and the greater is the likelihood of type
two error. Conversely the higher we set the rejection level the greater
the likelihood of type I error, and the smaller the likelihood of type II
error.

COMPARISON OF BASIC EXPERIMENTAL DESIGN

Aspect CRD RCBD LSD
Unlimited except If too many, blocking Limited since p= r
when the eu’s are
advantage may be
markedly varied
lost

Ease of Easy Fairly easy as long More difficult
laying out as blocks of considering that
experiment required size are each treatment
set-up appears only once in