A researcher does not assume anything. In a research, all observations must be noted down and properly analyzed before any conclusion or generalization is drawn. Research is not just done for the sake of it but it presents clearly investigated issues that are then used to make important decision in ones life. Inaccurate information will mean that the decision made will be incorrect, misleading and detrimental. Before a research is conducted, there must be a problem that has already been well identified and defined. All possibilities of getting a solution should also be explored. Having been convinced that no specific solution has yet been gotten, it is the right time to use scientific methods to generate possible solutions that can then be applied to similar problems. This paper explores the role played by scales in research. (Cohen L, Manion L, Morrison K, 2007, pp.586-638; Johnston J.M, Pennypacker H. S, 1993, pp. 133-160)

Scales provide evidence in support or oppose a hypothesis. Many people have developed claims or just mare allegations that have no proof. Such allegations may or may not be true and the may interfere with what decision should be made pertaining a certain issue. It is therefore important to collect data that would then be analyzed using scales to generate information in which conclusions, inferences and generalizations are based. There exist four types of scales in research; ordinal, nominal, interval and ratio scales. (Cohen L, Manion L, Morrison K, 2007, pp.586-638; Johnston J.M, Pennypacker H. S, 1993, pp. 133-160)

Scales provides units of measurements that can be used to label an observation. This can then be plotted to identify the trend of an occurrence and be able to determine what precaution to take incase the trend signals a negative outcome. For instance, a nominal scale may be used to measure the number of males who performed best in an exam. An ordinal measurement scale is then used to grade the studentsâ€™ performance. One can not say that since the percentage of girls and boys in the top ten positions is 5 in each side the students have faired equally well. Of course gender was a key variable to consider and if only one gender occupied the first 5 positions while the other occupied the rest, then it means that those who occupied the first 5 positions did better and something must be done to ensure that none dominates the first positions only. In such a case, the interval scale that measures the difference between the first students in each gender plays a major role. In the same manner, if we were to find out using mean grades which gender did better in the exam, if males got a mean score of 30 while females got a mean score of 60, we can say that the girls mean score was twice that of boys. In such a case there exist a fixed zero and thus ratio scales come into play. After analyzing the data, the researcher can generate information that will act as the base in making decision that will see both boys and girls competing fairly and getting equal opportunities to excel academically. This is the most important part of the research and the role of scales in research. (Cohen L, Manion L, Morrison K, 2007, pp.586-638; Johnston J.M, Pennypacker H. S, 1993, pp. 133-160)

References

Cohen L, Manion L, Morrison K, (2007), RESEARCH METHODS IN EDUCATION, 6th Edd, Routledge pp.586-638

Johnston J.M, Pennypacker H. S, (1993), READINGS FOR STRATEGIES AND TACTICS OF BEHAVIORAL RESEARCH, 2nd Edd, Routledge pp. 133-160