Data – what we need to do the analysis? But how much we are sure that the data we have collected for analysis is reliable enough! The main agenda of this post is to discuss data reliability, it’s importance and how to achieve it.
Let’s start with a brief example, suppose during a clinical trial “If the data are not taken properly?” And proceed further for analysis. Then “what will be the consequences of it?” Clearly, it may fail later and results in loss of resources and time. Thus, the device measuring the host condition should produce reliable results.
Now, I would like to introduce the term called “Measurement system analysis (MSA)” which helps to determine the reliability of a data for analysis. You must be curious to know about it, right!
Measurement system in Healthcare
Generally, MSA is a method to determine the measurement system in a process. The measurement system is a collection of measurement devices, measurement procedures and operators that are used to obtain a measurement. MSA helps us to detect the variation ( i.e difference or disparity) exists in a measurement system. So we should evaluate our measurement system before conducting statistical process control, design of experiment, or other statistical analysis. By doing so, we can be sure that our data are reliable for analysis.
We can apply MSA for two types of data namely attribute and continuous data.
MSA is classified into Attribute agreement analysis and Gage R&R analysis based on data types used.
Suppose in a wheel manufacturing company, operators use gage to measure the diameter of the wheel. Are we 100% sure that the collected data is measured correctly or not?
With reference from the Gage Run Chart – Three operators measure 10 wheels, three times per wheel randomly. Here, we can examine the differences in measurements between operators. We can see that Operator B does not measure consistently, and Operator C usually measures lower than the other operators.