Measurement Systems: How to Choose the Right One

In determining the effectiveness of social contact in reducing stigma in mental illness, a ratio scale will be used. Ratio scale is a system of measurement in research that is quantitative (Christensen, 2014). The system allows a researcher to make a comparison of intervals or any identified differences. It is in most ways similar to an interval scale and the only difference is that the zero point of a ratio scale is not arbitrary (Christensen, 2014). The zero point in a ratio scale means that a certain characteristic or phenomenon is absent. A ratio scale is able to show order and determine the actual value between different units which together with the absolute zero makes it possible to be used in most inferential and descriptive research (Christensen, 2014). Using the scale in data collection will make it possible to calculate measures of central tendency such as mean and measures of dispersion that include coefficient of differentiation among others. The variables can among other operations be added and multiplied. The system offers many possibilities in analyzing statistics (Christensen, 2014). 

Baseline data will be gathered before the intervention and another set of data will be gathered after the intervention in order to establish its effectives after analysis and comparison of the two data sets. Other systems of data collection considered include interval, nominal, and ordinal scales. A nominal value lacks a quantitative value, in ordinal scales the difference between values cannot be established while interval scales lack an absolute zero. As a result, they cannot be used for the study. To ensure reliability of the data collected, the same individuals will be tested on two different periods. A less random error will indicate that the data collected is reliable. A high random error however will show that the data collected in not reliable. Although a ratio scale is the most appropriate measurement system for the study, measuring attitudes on a scale of 1 to 6 for example does not indicate that response six is twice a response of two. This shows that there are instances where the scale may not make much sense. 

Reference  

Christensen, L. B. (2014).  Research Methods . Pearson Education Limited.