Terminology in DoE
This post is intended as a quick-reference dictionary for understanding key terminology in Design of Experiments (DoE). If you encounter any terms during your reading on the experimentaldesignhub.com that are unfamiliar, you can easily refer back to this guide for a clear and concise explanation.
Category | Term | Description | Example |
---|---|---|---|
Foundational Principles | Replication | Repeating the same experiment multiple times to ensure accuracy. | Repeating a temperature test at 200°C multiple times to verify results. |
Randomization | Conducting experiments in a random order to avoid bias. | Testing temperatures at random intervals (e.g., 150°C, 200°C, 100°C). | |
Blocking | Grouping experiments by uncontrollable external factors. | Grouping tests by morning and afternoon sessions to account for temperature variations. | |
Variables in DoE | Factor | Controlled variables in an experiment set by the experimenter. | Temperature and pressure in a chemical reaction experiment. |
(Factor) Level | Specific values a factor can take. | Temperature levels at 100°C, 200°C, and 300°C. | |
Response Variable | The main outcome measured in an experiment. | Measuring the tensile strength of a material as a response to heat treatment. | |
Disturbance Variable | Variables not of primary interest but can influence the response. | Ambient temperature or humidity during material testing. | |
Statistical Measures | Variance | Measure of the spread of data points in a dataset. | Different readings in a temperature experiment showing variance. |
Residuals | The difference between observed and predicted values in a model. | Residuals are analyzed to check the fit of a regression model. | |
ANOVA | Analysis of Variance used to compare means and identify significant factors. | Determining if temperature significantly affects filtration rate. | |
P-value | Probability that the observed results are due to chance. | A P-value less than 0.05 indicates statistical significance. |