Data evaluation is a complex activity presented by the presence of unreliable data because of the different levels of inconsistency. Hence, scales are used to evaluate data in many perspectives to give the user of the data the relevant and needed information.
1) Baseball or Football2) Baseball or Basketball3) Baseball or Hockey4) Baseball or Soccer5) Football or Basketball6) Football or Hockey7) Football or Soccer8) Basketball or Hockey9) Basketball or Soccer10) Hockey or Soccer
Paired comparison enables individuals to choose between the various elements, more so the elements in pairs. This method is vital in the comparison of only tow choices or preferences for items. In addition, the scale is insufficient in comparison for all the various items at one instance and only allows for comparison of only two elements or items. However, it allows for the simultaneous evaluations and comparison of two elements at the same time. It does not allow for the evaluations of the most preferred elements. People found this as a difficult task in identifying the various preferred sports when they were tasked in choosing the tasks, which they preferred, form a pair of items, which they equally liked. This shows the limitations of the scale of paired comparisons in that it does not allow for the comparison (Shaw, & Wright, 1967).
Rank 2 -Soccer
Rank 3 -Football
Rank 4-BaseballRank 5-Hockey
Rank order enables people to indicate their preferences in an ascending order such that most preferred element within the group is put first while the rest of the items are in levels of their preference from the most preferred to the least preferred. Rank order was relatively easy because people found it easy to identify the items, which the preferred in an ascending order. However, it limits comparison of the elements arranged in the order of preferences.
Rank order scale is limited because it only allows for the comparison of the general data, which is the level of preference for the sporting activities.
On the other hand, constant sum scales are used for the evaluating the levels of importance or preference for the various items in question. The method is preferred because people tend to avoid the long lists associated with other methods such as the paired comparison scales. In addition, the method enables people to give the most preferred sporting activities the level of preference by awarding points in the level of preference. It was relatively easy for people to use this scale as the most preferred elements were awarded points in levels of their preferences. In addition, the scale also allows for validation in that they are accurate. The accuracy of the data provided is evidenced by the ability of people to offer precise and accurate numerical data in relation to their levels of preference. Validations are also evidence by the ability of the information provided to be measure bale in the relevant forms with specific reference to numerical terms of measurement (Miller, & Salkind, 2002).
I preferred using the constant sum scales in evaluating the various preferences. This is because the constant sum because it incorporates numerous qualities. The first quality is that it is reliable because of its consistency in the aspects of measurement. Consistent date is preferred especially if it is numeral as it enables precise and accurate evaluation of the facts or preferences put forth. Consistency of data comes to play because the data presented is identical and can be used for general comparison of the various elements. In addition, the data is also favorable because it also generalization of the elements through the data provided. The data used to differentiate the various elements allow for identical treatment of data using numerical interpretation of preferences for the various sporting activities (Shaw, & Wright, 1967).
Data, which is relevant, has several broad characteristics; it is measurable, reliable, general and more so valid for the purpose, which it has been taken. The various forms of scale are advantageous in their individual means as they allow for the various purposes depending on the need for which the data has been collected. Means of data evaluation, which are relevant, enable the fulfillment of the purposes for which they are intended, making them reliable (Miller, & Salkind, 2002). The three forms of scale vary in their purposes making them reliable for the various needs of interpretation and evaluation. The three scales: paired comparison, rank order and constant sum have their different levels of validity because of the purposes for which they are used, which is different from each. Reliability and validity of data represent different things. The various scales can be valid but not reliable in the measurement of the various attributes of the scored data or preferences of the individual. Reliability on the other hand is determined by the presence of errors of measurement of the data presented. Hence, reliable data is free form material error, which influences the position of the data or the general outcomes of the measurement.
Miller, D. C., & Salkind, N. J. (2002). Handbook of research design & social measurement. Thousand Oaks, Calif: Sage Publications.
Shaw, M. E., & Wright, J. M. (1967). Scales for the measurement of attitudes. New York: McGraw-Hill.