Evaluating Students and Defining Them in Classes - Research Paper Example

Measurement and Instruments for a Quantitative Research Plan The levels of measurement that are important for the study The present study would be instituted to measure the intellectual abilities of given students to assign them to different classes of study. To this end, it would be important to have a measurement in place that can help in the determination of the performance of the students as far as the study of given course content is concerned. This means that students would be given a curriculum or course content and they will be expected to finish learning the course content within a given time frame. Essentially, the measurement would be seeking to know how well the students have learnt the concepts given in the course content (Sanigorski et al, 2008). It is against this background that the levels of measurement to be selected for the study would follow the principle of the ordinal scale of measurement. This means that grading shall be assigned to the students according to their performance rate. The levels of measurement shall, therefore, be a grading system to be tagged or labeled as ‘A’, ‘B’, and ‘F’. As an ordinal scale, these levels of measurement shall be directly linked or related to the performance rate of students. In the given situation, students who score grade ‘A’ shall be given the classes they applied for while students who score ‘B’ shall be placed in one class below the class they applied for. Finally, students scoring ‘F’ shall be rejected a place in the schools. Content validity, empirical validity, and construct validity for your study Given the levels of measurement that will be used in the study, the researcher shall design a multiple-indicator instrument or test. The multiple-indicator instrument will be used because it has the potential of ensuring validity on several scales. Cherry (2012) explains that “When a test has content validity, the items on the test represent the entire range of possible items the test should cover.” This means that the study shall ensure content validity by setting out questions to cover all aspects of the course content to be given to the students (Rodchua, 2009). By so doing, the test items shall represent the entire range of possible items because no aspect of the course content shall be left out. On empirical validity, it has been said that for empirical validity to be achieved, the researcher must create criteria with which the results obtained can be compared for confirmation (Cheng, 2009). In the present study, the empirical validity can be measured using comparing the scores of the test, to the continuous assessment results that the students will bring from their respective schools. What this means is that there is an expectation that students who come into the test with better continuous assessment results would do better than those who have abysmal continuous assessment results. Trochim (2006) holds the opinion that “You might think of construct validity as a "labeling" issue.” What this means is that construct validity shall be ensured because at the end of the whole test, the students shall be labeled using the levels of measurement and the labeling shall be used to grade them. How reliability will be ensured for the measurement. It has been established that in measurement studies such as this one, one of the best ways to ensure reliability is to reduce the chances that there will be an error in the scoring process (Wade, Worthington and Haake, 2009). This is because in the given scenario, if there are errors in the scoring, there shall be the wrongful placement of students in classes and this would affect the overall academic output expected from the school. One of the ways to ensure reliability for the measurement in the study shall be to use internal consistency. It has been established that “Internal consistency gives an estimate of the equivalence of sets of items from the same test (Kimberlin and Winterstein, 2008, p. 2277). This means that to use internal consistency to achieve reliability of the study, it would be important that a mechanism is instituted to ensure that the items that will be used in the test are representative of the very aspects that students were expected in the course content. Once items are set outside the course content for the test, what it would be is that the reliability of the study shall be compromised because there would be students who will be rejected for the wrong assessment criteria and others who will be accepted for the wrong assessment criteria. The strengths the measurement instrument in terms of reliability and validity Generally, is has been indicated a multiple-indicator measurement instrument shall be used. This multiple-indicator instrument has many strengths that ensure that validity and reliability are both achieved. The first fact is that several phenomena in sociology, which are measured in social researches such as this one, have multiple facets (Tanigoshi, Kontos and Remley, 2008). To this end, it is always important to have a measurement instrument with multiple-indicator that can meet the different provisions of the different facets that are measured. When measuring community integration, for instance, it would be virtually impossible to achieve validity with results by using a single-indicator instrument to measure community allegiance. Once this is done, several areas of the facets would be left out. This, therefore, presents itself as a major strength for using the multiple-indicator instrument that has been outlined by the researcher. Again, the use of the multiple-indicator instruments is deemed highly appropriate for ensuring validity and subsequently ensuring reliability because the risk of errors in all social researches is inevitable (Fuchs et al, 1997). Meanwhile, as these risks of errors occur, it is important that the researcher device a means of ensuring that the effect of the errors on the research or study is curtailed. It is in such provision that the use of the multiple-indicator instrument as a measurement instrument becomes highly appropriate and suitable in ensuring that the adverse effect on validity is reduced a smaller margin (Broadwings, 2009). Given a test instrument with fifteen items for instance, if there is an error defect of one, what it means is that only 6.6% of the entire validity of the test in terms of its scale of measurement becomes affected. Finally, when using the alpha coefficient to measure internal validity, it requires that there will be more indicators to score higher alpha points. Since multiple-indicator instruments are used with varying indicators, it becomes possible to achieve internal reliability and subsequently achieve general reliability of the study because when tested on the alpha coefficient scale, there will be higher alpha points scored. Limitations of the measurement instrument in terms of reliability and validity Notwithstanding all the advantages of the multi-dimensional instrument in ensuring and achieving validity and reliability when used for measurement, there are some major weaknesses or limitations with its use also. It is, for instance, argued that there does not exist enough or sufficient standardized tests that follow the multi-dimensional instrument for public research and use (The Blanch Law Firm, 2012). To this end, newly constructed instruments may not have been tested much enough to be proven for its credibility in achieving needed results. What is more, there is a worrying trend whereby the existing instruments that conform to the multiple-indicator system have not seen any major improvements over the years. This finding was made known by Borgatta and Borgatta (1992), who in their research found that the absence of improvement has led to a situation whereby there “there has been no increase in the proportion of multiple-indicator instruments for about twenty-five years” (p. 1237). Consequently, it is very difficult to trust that any given multiple-indicator instrument can pass the test of reliability. Scale appropriate for study As it has been pointed out earlier, the present study would use the ordinal scale of measurement. The ordinal scale is one in which the researcher categories items as to whether or not they possess more or less of a given characteristic (Snider, 2006). With this definition, it can be said outright that the ordinal scale is the best form of scale that can be used in the present situation. The reason for saying this is that the researcher seeks to find the performance level of respondents as far as their acquaintance with given course content is concerned. To this end, it is important that the researcher would have a way of categorizing the respondents to place them in their most appropriate classes. A justification for scale in achieving reliability and validity Much can be said to justify the use of the ordinal scale in achieving validity and reliability for the study. In the first place, internal validity shall be enhanced in that the nature of the test instrument to be used would be one that will make it possible to quantify the scores of respondents in real ordinal terms. Consequently, it would be more valid to use a scale that relates to the ordinal presentation of the results than other forms. Again, the ordinal scale follows the quantitative approach (Pearson and Pearson, 2012). What this means is that it will be very simple and easy for any third party to subject the results to constant scrutiny to ensure that the placements that will be made to follow stipulated criteria for scoring. This would undoubtedly be a major approach in ensuring that the final scores presented by the researcher can be deemed as reliable. Population for the Study and test appropriate for population The test to be used will be performance-related and follow the principles of the criterion-referenced test. What this generally means is that the results to be obtained by respondents will not represent the individual mastery that the respondents have had over the course content but the comparison shall be between individual respondents and the performance of their colleagues. Literarily, one has to perform better than others in the mastery of the course content to score a good grade. This type of test has become necessary because the researcher has several respondents, who need to be placed according to an identifiable criterion. In all this, the population to be used for the test will be a group of students who will each be given the same course content to learn within a specified time frame. After the time frame elapses, the researcher would then conduct the test to enable him to classify the respondents or population into appropriate classes. REFERENCE LIST Borgatta and Borgatta (1992). Measurement Instrument. Encyclopedia of Sociology, Vol 3, pp. 1236 – 1240. NY: Macmillan Publication Company. Broadwings T. G (2009). Elements of Quantitative Research Surveys. Ultimate Press Limited: Texas Cheng, K. W. K. (2009). The effect of web-based collaborative learning methods to the accounting courses in technical education. College Student Journal, 43, 755–765. Cherry K. (2012). What is Validity? http://psychology.about.com/od/researchmethods/f/validity.htm Fuchs, L. S., Fuchs, D., Hamlett, C. L., Phillips, N. B., Karns, K., & Dutka, S. (1997). Enhancing students’ helping behavior during peer-mediated instruction with conceptual mathematical explanations. The Elementary School Journal, 97, 223–249. Kimberlin and Winterstein, Validity and reliability of measurement instruments used in research. Am J Health-Syst Pharm—Vol 65 Dec 1, 2008 Pearson A and Pearson B (2012). Research Problems, Variables and Hypothesis. Accessed September 2, 2012 from http://wps.ablongman.com/ab_mcmillan_edresearch_4/16/4150/1062474.cw/index.html Rodchua, S. (2009). Comparative analysis of quality costs and organization sizes in the manufacturing environment. Quality Management Journal, 16(2), 34–43. Sanigorski, A. M., Bell, A. C., Kremer, P. J., Cuttler, R., & Swinburn, B. A. (2008). Reducing unhealthy weight gain in children through community capacity-building: Results of a quasiexperimental intervention program, Be Active Eat Well. International Journal of Obesity, 32, 1060–1067. Snider D. (2006). Alignment: Do your processes really line up? CMA Management. June/July 2006; 80/40 ABI/INFO Global Tanigoshi, H., Kontos, A. P., & Remley, T. P., Jr. (2008). The effectiveness of individual wellness counseling on the wellness of law enforcement officers. Journal of Counseling & Development, 86, 64–74. The Blanch Law Firm (2012). Computer Tampering. Accessed July 30, 2012 from http://www.criminaldefenseassociates.com/index.php/resources/computer_tampering/ Trochim W.K.M. (2006). Construct Validity. http://www.socialresearchmethods.net/kb/constval.php Wade, N. G., Worthington, E. L., Jr., & Haake, S. (2009). Comparison of explicit forgiveness interventions with an alternative treatment: A randomized clinical trial. Journal of Counseling & Development, 87(2), 143–151. Read More