T Tests

There are many times when the research is just not confined to one group of data. It is expanded to two groups of data and compares them. The purpose is to see if the two groups are similar or different. For this purpose T tests are often used in order to compare from two different groups of data. T tests help to find out the difference in the means and clearly find out the similarity and difference between them. In the case when the means are different from each other, it can be analysed that the variables are being manipulated. Popularly there are two types of T tests that can be seen in SPSS.

  1. Independent Samples T test: Those populations that are independent of each other and not related in any way are compared with a T Test. In this case the participants of one group do not have a relationship with the second group.
  2. Paired Sample T Tests:  These types of T Tests are used for comparing the groups that have some kind of relation with each other. The participants in two groups can be related to each other in more than one ways. In a situation the population of both the groups is same. The second situation can be that participants in one group are genetically related to the participants of the second group. These are called the repeated measures.

For these repeated measures, the advantage is that the sample size required is not large. For the reason that all the participants in research are involved in the treatment the required number of participants in reduced to half and it also brings down the cost of doing research and the time frame involved to a large extent. One more advantage of this repeated measure is that it helps in a strong control over the individual differences in the populations. The chances of random error are brought down to a great extent. When there is only one group to test the experimenter doesn’t have to worry about the differences in the groups and the technically called “noise” level is brought down to a great level. Though, of course it is nearly impossible to produce data without any noise because some differences within the group would still exist.  The reactions of the individuals in the population get affected by subjectivity in some or the other way.



Leave a Reply