Balanced Two-level Experiment Effects

How to use Balanced Two-level Experiment Effects

Name the factors and enter every +/− setting exactly once with equally sized response arrays. Form the orthogonal sign-product contrasts, report twice each coefficient as its effect, and verify both cell-mean reconstruction and the complete sum-of-squares identity.

1. Use the example as the schema; replace its labels and entered model assumptions.
2. Name the factors and enter every +/− setting exactly once with equally sized response arrays. Form the orthogonal sign-product contrasts, report twice each coefficient as its effect, and verify both cell-mean reconstruction and the complete sum-of-squares identity.
3. Calculate, inspect the complete witnesses and limitations, then copy or download the JSON report.

Good to know

• Supports 1–5 distinct factors, all 2^k settings, and 1–10 responses per setting. Responses are exact decimal strings of magnitude at most 1,000,000,000 with up to six places. Settings use literal ASCII + and -, one per listed factor. Missing, duplicate, unequal or center-point cells are refused.
• Includes every interaction, not a screened reduced model. One response per cell provides zero pure-error degrees of freedom and a null error mean square. No F test, p-value, causality, randomization assessment, missing-data imputation or selection of production settings.
• Local calculation with no upload, input persistence or outside lookup. Model JSON is limited to 30,000 characters and 10,000 values. Decimal quantities are strings with up to six places; count fields use plain JSON integers. Exact fractions are capped at 8,192 bits. The complete JSON output is limited to 1,000,000 characters; approximate numbers preserve their stored values. Unsupported states fail explicitly.