Enter your data for Pre-Process Responses for Analyze Variability

Stat > DOE > Factorial > Pre-Process Responses for Analyze Variability

Select the option that best describes your data. For more information on the difference between repeats and replicates, go to Replicates and repeats in designed experiments.

Compute for repeat responses across rows

Complete the following steps if you have repeat measurements. Repeat measurements are taken within one experimental run. For example, an engineer completes an experimental run on a process that manufactures insulation. Each run of the experiment produces 6 samples of insulation.

  1. Select Compute for repeat responses across rows.
  2. In Repeat responses across rows of, enter 2 or more columns that contain the response measurements.
  3. In Store standard deviations in, enter a name for the storage column of the standard deviations that you will analyze with Analyze Variability. If the name has more than one word or special characters, enclose the name in single quotation marks.
  4. In Store number of repeats in, enter a name for the storage column of the number of repeat measurements for each experimental run. If the name has more than one word or special characters, enclose the name in single quotation marks.
  5. To analyze the means of the repeat measurements, in Store means (optional) in, enter a name for the storage column of the means. If the name has more than one word or has special characters, enclose the name in single quotation marks.
This worksheet contains a factorial design. Yield_1 and Yield_2 represent separate measurements of the yield after a single experimental run. In the fourth row, the researchers took only 1 measurement, so the row contains a missing value symbol, *.
C1 C2 C3 C4 C5 C6 C7 C8
StdOrder RunOrder CenterPt Blocks Time Temp Yield_1 Yield_2
1 1 1 1 20 150 42.7636 43.2976
4 2 1 1 50 200 44.7592 47.3932
2 3 1 1 50 150 45.9131 43.1891
3 4 1 1 20 200 * 48.4665

Compute for replicates in each response column

Complete the following steps if you have replicate measurements. Replicate measurements have the same factor settings but are taken in different experimental runs. For example, an engineer completes an experimental run on a process that manufactures insulation. After more experimental runs, the engineer collects a new sample of insulation that was manufactured with the same factor settings. The engineer measures the strength of the new sample.

  1. Select Compute for replicates in each response column.
  2. In Response, enter at least one column that contains measurements from the response across multiple replicates.
  3. In Store Std Dev in, enter a name for the storage column that you will analyze with Analyze Variability. If the name has more than one word or special characters, enclose the name in single quotation marks.
  4. In Store Counts in, enter a name for the storage column that will contain the number of replicates of each experimental run. If the name has more than one word or special characters, enclose the name in single quotation marks.
  5. If the design in the worksheet has covariates, enter the columns in Adjust for covariates. For more information go to Adjust for covariates in replicates in factorial designs.
This worksheet contains a factorial design. The factors are in the columns Time and Temp. The response is in the Yield column. In run order, the first row and the fifth row have the same factor settings but are different experimental runs, so these rows are replicates. The standard deviation for a set of replicates is stored only in the first row with that factor setting. Missing value symbols are stored in the remaining rows with the same factor settings.
C1 C2 C3 C4 C5 C6 C7
StdOrder RunOrder CenterPt Blocks Time Temp Yield
2 1 1 1 50 150 43.0306
3 2 1 1 20 200 46.0762
4 3 1 1 50 200 44.1911
1 4 1 1 20 150 50.7655
6 5 1 2 50 150 43.2277
8 6 1 2 50 200 43.8761
7 7 1 2 20 200 46.0187
5 8 1 2 20 150 45.4380

Standard deviations already in worksheet

Complete the following steps if you have a column that contains pre-calculated standard deviations and a column of the count of measurements.

  1. Select Standard deviations already in worksheet.
  2. In Use Std Devs in, enter at least one column that contains standard deviations. For repeats, the standard deviations should be stored in the same row as the factor settings that produced them. For replicates, the standard deviations should be stored only in the first row of its corresponding factor setting.
  3. In Use Counts, enter a column or a constant that indicates how many repeat or replicate measurements were used to calculate each standard deviation. These numbers of measurements must be integers greater than or equal to 0. If the number of measurements is 0 or 1, then the corresponding row in the standard deviation column must contain a missing value symbol, *.
This worksheet contains a factorial design. The standard deviation column contains the standard deviations of the response measurements taken at each combination of factor settings. The count column is the number of measurements that are used in the standard deviation calculation. In the fourth row, the researchers took only 1 measurement, so the row contains a missing value symbol, *.
C1 C2 C3 C4 C5 C6 C7 C8
StdOrder RunOrder CenterPt Blocks Time Temp Standard Deviation of Yield Count
1 1 1 1 20 150 0.37760 5
4 2 1 1 50 200 1.86252 5
2 3 1 1 50 150 1.41704 5
3 4 1 1 20 200 * 1
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy