The actual experimental response data varied somewhat from what a Belt had predicted them to be. This is the result of which of these?

Question

Options

• AInefficiency of estimates
• BResiduals
• CConfounded data
• DGap Analysis

• A
  9% (3)
• B
  82% (27)
• C
  3% (1)
• D
  6% (2)

Explanation

Residuals are defined as the differences between observed (actual) values and the values predicted by a statistical model - exactly what's described here. When a Six Sigma Belt builds a regression or predictive model, the predicted values rarely match the actual experimental results perfectly; those gaps are the residuals, and they're a normal, expected part of any model.

Why the distractors are wrong:

• A (Inefficiency of estimates): "Efficiency" in statistics refers to how well an estimator uses available data relative to the best possible estimator - not the gap between predicted and actual values.
• C (Confounded data): Confounding occurs when the effect of one variable is entangled with another, distorting causal interpretation - it's a design problem, not a description of prediction error.
• D (Gap Analysis): Gap analysis is a business/process tool comparing current performance to a desired target state - not a statistical concept describing model prediction errors.

Memory tip: Think of "residual" like residue - it's what's left over after your model explains as much as it can. If your prediction were perfect, nothing would remain; residuals are the leftovers.

No community discussion yet for this question.