Measurement and Instruments for a Quantitative Sample Research Plan

Most of the non-governmental organizations in South Africa have emerged to play a key role in bridging the gap between the rich and the poor. This has been done through their provision of financial and social aid to the low class earners in South Africa. Besides, the NGOs are seen as key foundations of the civil society where citizens often struggle to earn a living. In most cases, these organizations tend to compete with governments for influence and power and this has brought disagreements between the two in various countries, South Africa being an example. Despite the competition, NGOs have been keen not to overstep the mandate or the supremacy of states. A few cases where NGOs have been seen to overstep their mandates have led to their disbandment as recently seen in some of the African countries. Undoubtedly, the global community has embraced NGOs as they have helped fill the gap between the poor and the rich in the society. People faced with various challenges will always turn to these organizations for support and assistance relating to their social needs. The significance of NGOs has prompted the conduction of a research study that would majorly focus on the importance of these organizations to the South African citizens.

Generally, such a study will always encompass a number of issues for it to be effective. One of these factors is the levels of measurement, which would be of importance for this study. In this case, the two levels of measurement that are of great importance are the ordinal and the interval, as they have been widely used in studies. For the assessment of the tremendous increments in the support offered by these organizations to South African citizens, the measurement level known as would be the most appropriate. This level of measurement would give room for the ranking of the performance of these organizations. Besides, this level of measurement proposes that the presentation of variables or outcomes is done in an orderly manner (Rosenthal, 2011). Interval measurement level would also be applied to this study. This measurement would allow for the assessment of these NGOs based on the degree of difference in performance. The main reason for choosing these two levels of measurements is that, for the case of ordinal measurement, it would be very easy to assess the effectiveness of these NGOs through calculating median and mode. Conversely, the interval if used in measurement, would allow for the calculation of mean, range and standard deviation based on the possible outcome of the study (Adler & Clark, 2008). Therefore, this implies that these two levels of measurements would provide a more complex and elaborate statistical analysis of the data, which will thereafter provide convincing information on the variables during the study. Generally, for any research to be effective, data analysis must be done through encompassing measures of central tendency, variance and standard deviation. Notably, the advantage of applying these two levels of measurement is that they involve all these methods of data analysis.

Other aspects that are of significance in the study include content, empirical, and construct validities. In any study, it is necessary to analyze the outcome based on these validities. For instance, to ensure content validity for this study, it would be important to systematically examine the test content so as to determine whether it covers a representative sample to be measured. An example is a situation where a questionnaire is used in determination of the effectiveness of NGOs in South Africa. In this case, to ensure content validity, it would be important to find out whether the questionnaire covers all the areas of importance based on the hypotheses formed for the study and relating it to other scientific literatures. In ensuring content validity, it has been found out that if the use of any instrument is not effective during the study,  it would be very difficult to determine the content validity. This implies that the content validity evidence will involve the degree in which the content of the study would match the content domain, which is associated with construct (Krishnaswamy et al, 2006). Subsequently, construct validity can be demonstrated in a number of perspectives. Construct validity can be ensured by the accumulation of evidence and this would be demonstrated using content analysis, factor analysis, correlation coefficient, ANOVA studies, which demonstrate differences between differential pretest-posttest studies among others (House et al, 1982). To ensure construct validity, it would be also important that the instrument used for measurement only measures the construct of focus, which for this case would be the effectiveness of NGOs in South Africa. Otherwise, the result could be meaningless or misleading. Additionally, it would be prudent in this kind of study to ensure empirical validity. This is also known as predictive validity, which encompasses the scores on a particular assessment and how they correlate with the same outcomes measured with different assessment instruments (Owusu-Gyapong & Wayne State University, 1987). In ensuring the empirical validity, the statistical evidence suggests and provides that the instrument should only measure what it was intended to measures. If this were not taken into consideration, then a false outcome would be realized making the study null and void. These validities seem to work in coordination and therefore should be ensured in any statistical study for its effectiveness.

Apart from ensuring validity of the study, it would also be necessary to check on the reliability of the study. Reliability is defined as an estimation of how an instrument of measurement in a particular research study would be able to yield the same result when reassessed. Such studies have never yielded a good result upon reassessment due to unreliability of the instrument chosen for the study and in most cases; this has led to provision of in accurate result for the study. Reliability of an instrument is a very important phenomenon since it contributes to and enhances an empirical research. Reliability can also refer to the stability of a measurement even if the measurement is repeated severally (Baumgarten, 2013). The choice of instrument of study would be of significance where in this case the most appropriate would be the GPS. This instrument would record all the data found in the field and store them in a similar manner. Besides, this instrument would ensure randomization and even distribution of data which are key areas of any given research study. The choice of this is what will determine the reliability of the study. The reliability of this instrument would estimate the extent of reality of variance in this study. In general, if the assessment were reliable, its results would be stable and relatively true.

The advantage of using GPS as instrument of measurement is that it would provide an opportunity for quantification of human mobility at fine spatial and temporal scale. This is because this kind of study deals with organizations headed by individuals in various departments who are not stationary. The use of this instrument for this research study would be feasible since they have relatively reduced cost and sizes but with greatly improved technology. Its units are widely accepted by the study populations and this would make it one of the best and reliable instruments in the study of the effectiveness of NGOs in South Africa. GPS- based tracking method would present enormous opportunities that would help in improving individuals’ space –time activities and their influence in the outcomes of the previous studies done. Over the past ten years, this instrument of measurement coupled with others has taken a prominent role in physical and social activities and exposure research (Czerniak et al, 1998). However, just like other instruments, GPS has a number of limitations that should be addressed before opting for it as an instrument of measurement in the study especially for such a mobile research. It has been noted that the rate of data loss in this instrument can go up to 92 percent as a result of signal dropout, participants not wearing the units, dead batteries, signal loss during initialization period or even misuse of the device. When these factors are not considered when using the instrument, then there is likely hood of getting a wrong result after the study and this would limit the possibility of getting the most accurate result.

In this study, the most appropriate scale that would be used is the Likert scale. This scale would provide a uni-dimensional scaling method .The Likert scale allows respondents to provide answers by filling in questionnaires. This scale would capture various intensities and the level of effectiveness of different NGOs and how they operate in South Africa and their possible outcomes to the citizens. This is because it is able to capture the respondent experience and feelings based on the services provided by these organizations. The scale also highlights the level of agreement between the researcher and the respondents. This scale would be more reliable for this study since it entails measurements that would be beneficial during the study. Since the primary goal of this study would be to determine the effectiveness of NGOs as service providers, this scale would be the most appropriate since it will encompass both the participant experience and attitude based on the organization. The Likert scale will also allow the researchers to define and put the variables in categories such as very weak, weak, average, strong and very strong (Ary et al, 2009). Some scales have elements that enhance their use and this is good for any study.  Notably, certain scales can be utilized under special conditions where they are restricted to particular tests in the field of study. The advantage of using this scale is that it assumes the equality of distance or the range of measurements during a study. When appropriate scale is not used for a test, the effect can be inaccurate information and hence useless data. This kind of scaling would allow the individuals doing the test to give responses that they perceive to be right or wrong basing on their own experiences. Additionally, the result of this scale would enable the researcher to collect data that is ordinal and change it to data that is interval.

The most relevant test for this plan would be a five item –criterion referenced test with scores ranging from 0 to 5. The higher scores would simply indicate the increased effectiveness of the organization. The score ranging from 1 to 2 would simply indicate low or poor performance of the organization while 4 and 5 would implies that the organizations have attained the required impact on the South African citizens. For this kind of scale and test, the population to be used is also subject to consideration (Popham, 1971). Since NGOs majorly focus on the welfare of the most vulnerable in the society, it would be important to focus on the slum dwellers. The people living in slums have poor living conditions and the crime rate in such areas is also very high. Therefore, this would be a resourceful population based on the challenges they undergo. This is a very big population that cannot be captured all at once and therefore the GPS as an instrument of measurement would ensure randomization and less biasness in the study. The outcome of the study would be treated and categorized in groups based on the scale of measurement. Since this population almost faces similar challenges, it would be easy to organize the variables based on the Likert scale. In general, for effectiveness of this study, selecting reliable and valid assessment instrument would be very foundational to a valid study design (Brown & Hudson, 2002). This is besides determination of appropriate levels of measurement.

 

 

 

 

 

 

 

 

References

Adler, E. S., & Clark, R. (2008). How it’s done: An invitation to social research. Belmont, CA: Thomson/Wadsworth.

Ary, D., Jacobs, L. C., Razavieh, A., & Ary, D. (2009). Introduction to research in education. Belmont, CA: Wadsworth.

Baumgarten, M. (2013). Paradigm wars – validity and reliability in qualitative research. S.l.: Grin Verlag.

Brown, J. D., & Hudson, T. (2002). Criterion-referenced language testing. Cambridge: Cambridge University Press.

Czerniak, R. J., Reilly, J. P., National Cooperative Highway Research Program., National Research Council (U.S.)., American Association of State Highway and Transportation Officials., & United States. (1998). Applications of GPS for surveying and other positioning needs in departments of transportation. Washington, D.C: National Academy Press.

House, R. J., Levanoni, E., Schuler, R. S., & University of Toronto. (1982). An empirical examination of the construct validity of the Rizzo, House and Lirtzman role scales: Toward a clarification of the nature of role conflict. Toronto: Faculty of Management Studies, University of Toronto.

Krishnaswamy, K. N., Appa, I. S., & Mathirajan, M. (2006). Management research methodology: Integration of principles, methods and techniques. New Delhi: Pearson Education.

Owusu-Gyapong, A., & Wayne State University. (1987). The empirical validity and forecasting performance of three estimating techniques for pooled regressions on strike frequency. Detroit, MI: Department of Economics, Wayne State University.

Popham, W. J. (1971). Criterion-referenced measurement: An introduction. Englewood Cliffs, N.J: Educational Technology Publications.

Rosenthal, J. A. (2011). Statistics and data interpretation for social work. New York, NY: Springer.