A Hypothesis is an assumption or a statement which may be true or not true. The testing of the hypothesis takes place on the basis of the information that has been obtained from a sample. The use of Hypothesis testing is widely prevalent in the business industry as it assists in decision making. There is some important terminology in hypothesis that needs to be known so as to completely understand the concept.

Null Hypothesis: Those hypotheses that have been proposed with the intention of receiving a rejection are called the null hypothesis. It means that the researcher hypotheses the opposite of what is desired to be proved.

Alternative Hypothesis:  When the null hypothesis is rejected, the alternate hypothesis has to be accepted. The rejection of the null hypothesis means that there is a relationship between the variables. The alternate hypotheses cover a large range of values.

One tailed test:  Any test can be either one tailed or two tailed. A test is put in the category of a one tailed test when the null hypothesis gets rejected and the value of the test statistic comes into a specified tail of the distribution. The test is called a two tailed test when the null hypothesis gets rejected and the statistics of the test fall in either of the two tails of the sampling distribution. There  are often the chances of errors in hypothesis testing which need to be corrected at the right time so as to avoid any major flaws in the analysis and interpretation of the data. There are usually Type I and Type II errors

The study of the relationships existing between the population and samples that are drawn from it is called as sampling theory. It is only the random samples on which the sampling theory is applied. In order to solve this purpose, universe is defined as a complete group of items for which knowledge is wanted which could be finite or infinite.

In the same way, universe is hypothetical or existent. In the case of hypothetical universe, the universe does not exist and one can only imagine the items that constitute it. When a coin is tossed or a dice thrown, it implies examples of hypothetical universe. On the contrary, universe of concrete elements is known as existent universe.

The term sample refers to that part of universe which is selected to investigate the current problem. Therefore, the theory of sampling is studying the relationships existing between the universe and the samples that are drawn from it.

The problem between a parameter and a statistic constitutes the basic problem of a sampling theory. It is concerned with estimating the right features of the population from those of the sample and also to gauge the correctness of the made estimate. This phenomenon of moving from specific towards general is known as statistical inference or statistical induction.

Test of practicality

Economy, convenience and interpretability are the three characteristics on the basis of which the feature of practicality of an instrument could be judged. 

According to economy characteristics, there should be some trade off between the ideal research projects and between what the budget could afford. Economic pressures are fast felt in length of measuring instrument. There surely are other items where the researcher could get more reliability, but in order to limit the interview or observation time, only few minutes are required for study purpose. Even data collection methods that are used are dependent on economic factors.

According to convenience characteristic there should be no complexity in administering the measuring instrument. There should be proper attention given to the right layout of the measuring instrument. For example the questionnaire is more effective and easy to complete if there are clear instructions given.

According to the interpretability characteristic, it is especially significant when other people and not the designers of test are involved to interpret the results. In order to make the measuring instrument interpretable, it must be supplemented by:

I. Complete instructions to administer the test
II. Scoring keys
III. Evidence about the reliability
IV. Proper guidance for using the test and to interpret the results of the test.

Ordinal scale:

Ordinal scale is the lowest level of scale. It places events in order but there is no attempt to make the intervals of the scale in regard to any rule. Order scales are represented by rank orders basically used in research relating to qualitative phenomenon.  An example of student’s rank in his graduation class involves the use of an ordinal scale.

The usage of ordinal scale means a statement of ‘greater than’ or ‘less than’ without being able to tell how much greater or lesser. There could be real difference between ranks 1 and 2 maybe less or more than the actual difference between ranks 5 and 6.

The right measure of central tendency is the median since the number of this scale has only a rank meaning. In order to measure dispersion, a percentile or quartile measure is used. Correlations are restricted to different rank order methods whereas measures of statistical importance are constrained to the non-parametric methods.