Page 56 - Basic Statistics

51

4. HYPOTHESIS TESTING

The main aspect in the application of inferential statistics, after
estimation is hypothesis testing. In statistics, hypothesis testing is an important

part to making a decision. By testing the hypothesis of the researchers will be

able to answer the questions posed, stating the rejection or acceptance of the

hypothesis.

Hypothesis is a temporary answer before the experiment carried out,
based on the results of the study of literature. Hypotheses often contain a

statement that is neutral or common occurrence. Truth of the hypothesis is

certainly never known except if carried out observations of the entire

population. To do this it is extremely inefficient especially when the population

size is very large.
Withdrawal of a random sample from a population, the observed

characteristics and then compared with the hypothesis put forward is a step to

test the hypothesis. If a random sample is an indication that supports the

hypothesis, then the hypothesis is accepted. Conversely, if a random sample

gives an indication that contrary to the hypothesis, then the hypothesis is
rejected.

Definition of a hypothesis is accepted or rejected is not absolute. One

hypothesis is rejected does not mean that the hypothesis is wrong, but the data

does hint that there have been changes in the characteristics of the hypothesized

population. Acceptance of the hypothesis means that is not enough evidence to
accept the alternative hypothesis.

Statistical hypotheses divided into two statement, namely the null

hypothesis (H0) and the alternative hypothesis (H1). Statement wishing rejected

his truth set as the null hypothesis, while his opponent hypothesis set as a

alternative hypothesis.

~~* CHAPTER 4 HYPOTHESIS TESTING *~~