Levels of Measurement that are Important for the Research Study
There are different levels of measurement in research studies. They include nominal, ordinal, interval and ratio levels. The proposed study will use the nominal and ordinal levels of measurement to study the research phenomenon. The nominal level of measurement would be important in determining whether the public health interventions to manage cholera in Sierra Leone are successful or not. This level of measurement is in line with the views of Groves (1987) who says nominal data often uses common traits to explain a research phenomenon. The common traits for the proposed data include “success” and “failure.” The ordinal level of measurement will also be useful for the proposed study because it would help to highlight the degree of success or failure among the public health interventions introduced in Sierra Leone to manage Cholera. This level of measurement would also help to establish the efficacy of each intervention, based on a comparison of their performance (MacKenzie, 2003). The different levels of measurement would run across different degrees of success or failure, such as “very successful,” “successful,” “very unsuccessful,” “unsuccessful” and similar measures. Collectively, the nominal and ordinal levels of measurement would be important in evaluating the proposed research phenomenon
How to Ensure the Validity of the Paper
Ensuring the validity of a paper’s findings is important in improving the quality of the research paper. In fact, Green, Tull and Albaum (1993) say that the validity of a paper shows the soundness of its quality of research. I will ensure the validity of the proposed study as outlined below
Content Validity
It is important to make sure that the content validity of the proposed study is high because it ensures that the measures of the research represent all facets of the study (Green et al., 1993). Ensuring that the test items on the research paper represent the knowledge needed to evaluate the efficacy of the public health interventions would safeguard the content validity of the proposed paper (Green et al., 1993).
Empirical Validity
Empirical validity would not apply to the proposed study because it is observational. Therefore, there would be no experiments conducted in the study (empirical validity mainly applies to studies that involve experiments).
Construct Validity
According to Kane (2006), construct validity refers to the researchers’ quest to measure what their studies should measure. To ensure that the proposed research has high construct validity, I will examine the correlates of the measure with their associated variables. This method will be consistent with the views of Campbell and Fiske’s (two researchers cited in MacKenzie, 2003) who emphasized the use of the same method for improving the construct validity of a paper. They did so by advocating the method through the multitrait-multimethod matrix (MacKenzie, 2003).
How to Ensure Reliability of the Research
Similar to research validity, the reliability of a study refers to the process of assessing the quality of the measurement instruments used in the research. Reliability comes before validity because it is difficult for a research paper to have valid measurement instruments if they are not reliable (Kane, 2006). The different threats to reliability in the proposed study include researcher (observer) bias and environmental changes. To make sure that the proposed study has a high reliability, the study would adequately explain its methodology, processes, and techniques used in the study to allow other investigators to replicate the same study. Since the research would involve field surveys, the proposed study would also include field processes to help other investigators understand the interests and difficulties experienced by the researcher when collecting data. Triangulation is also another technique that would be applicable in the proposed study to ensure the reliability of the proposed study. This method involves using two or more data collection techniques and comparing them with one another. This method would allow the researcher and observer to look at the research phenomenon from multiple angles, thereby improving the research’s validity.
Strengths and Limitations of Measurement Instrument
There are different types of measurement instruments in research. The most common instruments include tests, questionnaires, observation, secondary data, and surveys (Mora, 2011). In line with the goal of triangulating the study, the proposed study would use a triangulation technique to improve the reliability of the study by using secondary, data, surveys and questionnaires as the main measurement instruments in the research paper. The reliability and validity of these instruments vary as shown below
Reliability
The inability of the researcher to know if respondents are telling the truth, or not, limits the reliability of the sources of data highlighted in the proposed paper. However, to make sure that the questionnaires used in the study are reliable, it would be crucial to conduct a pilot study, first, to establish areas of strength and weaknesses about the questionnaire. Secondly, I will run alpha revisions to improve its validity before it is ready for shipping and use. To improve the reliability of the surveys, the description used will be consistent throughout the paper. For example, there will be a consistent use of success and failure indicators throughout the research process. Testing the reliability of the surveys could involve different techniques including the use of correlations and split sample comparisons. Lastly, the use of secondary data may pose significant limitations, in terms of changes in measurement instruments. Stated differently, while the initial study may use accurate measurements to come up with their findings, these measurement instruments may have changed, thereby undermining the reliability of the instruments used (Wilson, Pan, & Schumsky, 2012).
Validity
One limitation of the validity of the questionnaire is its inability to provide explanations for different research phenomena (Linn, 2000). For example, in the context of the proposed study, it would be difficult to know why some public health interventions are successful and why others are not. Nonetheless, to improve the validity of the questionnaire, it would be important to conduct revisions on readability tests. Thereafter, I would conduct field tests and readability tests to make sure the questionnaires are valid for use. These methods would scientifically prove that the questionnaire would have a high validity. However, exposing the questionnaires to a group of experts would also help to improve their validity (Wilson et al., 2012). Comparatively, improving the validity of surveys would demand that the questionnaires included in the surveys address all the research questions. Furthermore, to improve the face validity, the surveys should show proper organization to make sure that they are presentable to the participants.
Choice and Rationale of Scale
Surveys are the most important scale to use in the proposed study because they align with the quantitative nature of the research study. Stated differently, they could use different numeric measures for assessing the successes, or failures, of public health interventions. For example, they could easily accommodate a 5-point of 7-point Likert scale to assess the successes, or failures, of public health interventions (Lawshe, 1975). Furthermore, they are easy to administer and collect information over a large study area. Moreover, through the advanced survey software, it is easy to develop surveys that properly capture the essence of the proposed study. Since the study is national, this advantage makes surveys a more practical instrument compared to questionnaires. Similarly, they make them more appropriate to undertake the study compared to secondary data because the scope of secondary information may not fit the proposed research.
Justification for Reliability and Validity of Scale
Since the surveys have emerged as the most important instrument for undertaking the proposed study, making sure that they provide reliable and valid data is essential to the development of a sound research paper. To do so, I will make sure that the items presented in the survey address the research phenomenon. There would also be a consistent use of vocabulary and language when formulating the surveys. Improving the internal validity of the surveys would also require the researcher to change the wordings of different questions without changing the meanings to avoid a situation where respondents answer questions based on their memory and not exactly what the research asks of them (Radhakrishna, 2007).
Appropriate Test for the Research Plan
The proposed study would strive to investigate the successes or failures of existing public health interventions to manage cholera in Sierra Leone. Significance testing would be appropriate in undertaking the study because the researcher would be preoccupied in evaluating the significance of the public health interventions that focus on reducing the rate of cholera in the West African country (Dimitrov & Rumrill, 2003). The lower the significance level, the more confidence a researcher should have in replicating the research in a different context (Dimitrov & Rumrill, 2003).
The Population used for the Scale and Test
Health agencies in Sierra Leone would be the most desirable population for the proposed scale and test. This population sample is appropriate because it is often aware of the health issues affecting different communities. Furthermore, they do not only have the scholarly background for articulating the progress of public health interventions, but also the practical knowledge for understanding whether a public health intervention works, or not. Lastly, they know a lot of information about the public and private health efforts for managing cholera in the country. This way, they know the level of government successes and failures in managing the crisis and the same successes or failures of alternative efforts by the private sector to manage the epidemic.
References
Dimitrov D. M., & Rumrill, P. D. (2003). Pretest-posttest designs and measurement of change. A Journal of Prevention, Assessment and Rehabilitation, 20(2), 159–165.
Green, P.E., Tull, D.S., & Albaum, G. (1993). Research methods for marketing decisions. London, UK: Prentice Hall.
Groves, R. M. (1987). Research on survey data quality. Public Opinion Quarterly, 51(1), 156-172.
Kane, M. T. (2006). Validation. Educational measurement, 4(1), 17–64.
Lawshe, C.H. (1975). A quantitative approach to content validity. Personnel Psychology, 28(1), 563–575.
Linn, R. (2000). Assessments and accountability. ER Online, 29(2), 4-14.
MacKenzie, S.B. (2003). The dangers of poor construct conceptualization. Journal of the Academy of Marketing Science, 31(3), 323–326.
Mora, M. (2011). Validity and Reliability in Surveys. Web.
Radhakrishna, R. (2007). Tips for Developing and Testing Questionnaires/Instruments. Journal of Extension, 45(1), 1-10.
Wilson, F.R., Pan, W., & Schumsky, D.A. (2012). Recalculation of the critical values for Lawshe’s content validity ratio. Measurement and Evaluation in Counseling and Development, 45(3), 197-210.