# Data considerations for Pre-Process Responses for Analyze Variability

To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret your results.

The design must be a 2-level factorial design
If you do not have a 2-level factorial design, use Test for Equal Variances.
The response should contain repeat or replicate measurements
Repeat measurements are taken during the same experimental run or consecutive runs, while replicate measurements are taken during identical but distinct experimental runs. You enter repeats across rows of multiple columns, while you enter replicates down a single column.
You can have precalculated standard deviations of the repeat or replicate measurements in the worksheet. You must also enter a column or a constant indicating the number of repeats or replicates in your experiment.
The data must include at least 2 factors, which can be either continuous or categorical
If you have only one categorical factor and no continuous predictors, use Test for Equal Variances.
Ensure that the measurement system produces reliable response data

If the variability in your measurement system is too great, your experiment may lack the power to find important effects.

Each observation should be independent from all other observations
If your individual observations are dependent, your results might not be valid. Consider the following points to determine whether your observations are independent:
• If an observation provides no information about the value of another observation, the observations are independent.
• If an observation provides information about another observation, the observations are dependent.
The experimental runs should be randomized

Randomization reduces the chance that uncontrolled conditions will bias the results. Randomization also lets you estimate the inherent variation in materials and conditions so that you can make valid statistical inferences based on the data from your experiment.

In some situations, randomization may lead to an undesirable run order. For instance, factor level changes can be difficult, expensive, or take a long time to produce a stable process. Under these conditions, you may want to randomize with a split-plot design to minimize the level changes.

Collect data using best practices
To ensure that your results are valid, consider the following guidelines:
• Make certain that the data represent the population of interest.
• Collect enough data to provide the necessary precision.
• Record the data in the order it was collected.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy