Question

A) Define reliability, as it relates to selection measures, B) Describe 3 common sources of error, C) Describe situations during which each of these types of error might occur, D) After you have done so, explain how you might prevent this types of error.

NOTE: Support your explanation with the word count of approximately 100 words in each alphabet category (A, B, C, D) with a total of about 400-450 words.

Answer 1

A) Reliability means consistency or stability. One way the reliability of the system can be measured is to record its stability. More the stability, better will be its reliability. So, the extent to which a system is free from errors and is consistent in all its measurements, it is called as reliability. Any system can be reliable if the measurements are consistent and error-free. Such system can be defined as a very reliable system. Let's take a very generic example of a weighing machine. How do we know that the weighing machine is reliable? For the sake of experiment, let's take a 5 lb. bag. As we already know the weight of the bag, you try to check the reliability of the weighing system. So, you place the weight on the system and you get the exact weight. Try and check the weight of the bag for couple of times in different positions and if still get the same weight consistently, then it means system is reliable. But, if the measurements vary each time when bag is measured, then there is an error in the system and it is not a very reliable system. So, if the system is stable, and consistent, then it is said to be a reliable system

B) There are two common sources of error: (and not three)

1. Systematic error
2. Random error

Systematic errors also called as predictable errors in the measurement. They occur in same pattern with same magnitude in a similar direction. Systematic error is said to be present if the second measurement was consistently greater than the first measurement, by a constant amount.

Random error is not a predictable error, these happen very randomly, owing to a chance. Random errors can arise from an equipment or the system that is used to take the measurements. Random errors are a result of inadequate calibration or warm-up, or operation in sub-optimal conditions.

C) Random errors are not attributable to a specific cause. If sufficiently large numbers of observations are made, random errors average to zero, because some readings over-estimate and some under-estimate. The situation where random errors might occur is at calibration test. In calibration trails, if there is a variation in the measurement values when compared to the standard object, then it is called random error. And the reason could be difference in the observation of two different observers.

Systematic errors tend to fall in a particular direction and are likely due to a specific cause. Because systematic errors fall in one direction. For example, if there was an 5km active relay running race conducted between the runners, it is measured based on the speeds of the runners in each round to reach the final destination. If they are supposed to cover five rounds to reach the final point, then their speed in each round will be different. It could be either lesser or more when compared to the previous round, since it depends on the running strategy of the runners. The systematic error would be present if the second measurement was consistently greater or lesser than the previous, by a constant amount. One of the simplest methods to determine whether systematic bias has occurred in measurement is to graph all the five measurements for each runner on a scatter plot.

D) Random errors and systematic errors can be prevented when the uncertainty is measured. Systematic errors can generally be corrected either numerically or technically. Random errors can be corrected by increasing and improving the correlation between the readings and measurements.