Data Analysis

Data analysis takes as input the measured metric values (data) generated by running workloads in a measurement context. This component determines how to analyze the data and interpret the results of the analysis.
A data analysis can be inappropriate, ignored, inconsistent, or irreproducible:
Inappropriate Data Analysis
A common goal of data analysis is to draw conclusions about a population from a small subset of the population (a sample). For this generalization to be valid, the experiment must use appropriate analysis methods.
Ignored Data Analysis
Well‐studied methods for analyzing data are used widely among the experimental sciences. Using these methods reduces the risk of poor data analysis and interpretation.
Inconsistent Data Analysis
Inconsistent data analysis occurs when different data analysis techniques are applied to data pertaining to different objects of an experimental study.
Irreproducible Data Analysis
For a data analysis to be reproducible, the experiment must carefully document the analysis it uses and identify and report any biases in the analysis. If an experimenter does not document their analysis and explicitly state their biases, other researchers have little hope of reproducing the analyses and arriving at the same conclusion.