Nature and Quality of the Test Data

The quality of the mechanical properties derived from the mechanical testing depends upon various factors. These factors include:

1. Precision
2. Accuracy
3. Authenticity and repeatability
4. Relevance to the test objective
5. Physical significance
   

The factors precision and accuracy can be attributed to statistical analysis. However, they cannot be separated if the data set is small. The remaining three factors can not readily be quantified.
Usually, there will be scatter in the measured data. The scatter is attributed to the combined effect of the factors:

1. Precision with which the measurements are made
2. Accuracy with which the measurements are made
3. Variations in the structure of the test coupon in the set itself

            The mean value and a measure of width of the distribution, like standard deviation or range are the two main statistical factors that are used to characterize the distribution of the values. Apart from their direct role as a measure of the variability in a set of data, the variance and the standard deviation, which is square root of variance, can be used to infer following points:

1. confidence limits for a set of data
2. reliability of apparent differences between sets of data
3. combined uncertainty of measurements when there are several sources of variability
4. separate variabilities when several factors have affected a set of data
5. goodness of fit when correlation between a dependent and an independent variable is derived