A00-240 · Question #2
The following LOGISTIC procedure output analyzes the relationship between a binary response and an ordinal predictor variable, wrist_size Using reference cell coding, the analyst selects Large (L) as
The correct answer is D. -1.0415. In reference cell coding, the intercept is the estimated logit for the reference category - here, Large. No additional terms are added because Large has no indicator coefficient in the model, making D (-1.0415) the direct and complete answer. Why the distractors fail: A (0.0819)
Question
Options
- A0.0819
- B0.5663
- C-3.7727
- D-1.0415
How the community answered
(40 responses)- B5% (2)
- C3% (1)
- D93% (37)
Explanation
In reference cell coding, the intercept is the estimated logit for the reference category - here, Large. No additional terms are added because Large has no indicator coefficient in the model, making D (-1.0415) the direct and complete answer.
Why the distractors fail:
- A (0.0819) is the logit for Medium: -1.0415 + 1.1234 = 0.0819 - a common trap if you confuse which category is the reference.
- B (0.5663) is the logit for Small: -1.0415 + 1.6078 = 0.5663 - same error, wrong group.
- C (-3.7727) results from nonsensically summing all three estimates (intercept + M + S coefficients), which has no statistical meaning.
Memory tip: Think of the reference level as "getting nothing extra" - it takes only the intercept, while all other levels add their coefficient on top. If you can remember "reference = intercept only," you'll never chase phantom terms for the baseline group.
Topics
Community Discussion
No community discussion yet for this question.