Apps

🧪 This platform is in early beta. Features may change and you might encounter bugs. We appreciate your patience!

Response Shift Bias

Also Known As: Recalibration bias Scale recalibration
Aspect ID: response_shift_bias

Definition

Response shift bias occurs when a change in the internal reference standard, values, or definition of a construct causes before-after comparisons on the same scale to be misleading. Patients who adapt to chronic illness may redefine what 'good quality of life' means, rating similar objective functioning higher after adjustment than before. This makes interventions appear more or less effective than they are when evaluated by subjective self-report measures.

Examples

Cancer patients rate their quality of life as 7/10 before treatment. After treatment causes partial disability, they rate it as 7/10 again — but this second rating reflects a recalibrated standard. If asked to retrospectively rate their pre-treatment quality of life from their current perspective, they now say it was actually 9/10, revealing a response shift.

A company surveys employee satisfaction before and after a major restructuring. Employees rate their work-life balance as 6/10 both times — but post-restructuring, 60-hour weeks have become the new normal, and employees have simply redefined 'balance' to fit their new reality. The identical scores mask a dramatic decline in objective conditions.

A first-year medical resident rates their stress level as 8/10 during orientation week. Two years later, after grueling overnight shifts and high-stakes decisions, they rate their stress as 6/10 — not because the job got easier, but because their benchmark for what counts as 'stressful' has fundamentally shifted. A naive comparison would falsely suggest they adapted positively.

Verification Steps
Verification Steps
Binary yes/no questions that an AI must answer to detect a reasoning pattern in a text.
Each of the 452 aspects has verification steps — simple yes/no questions designed to systematically detect whether a pattern appears in a text. For ad hominem: "Does the argument attack a person rather than their claim?" For false dichotomy: "Are only two options presented when more exist?" This ensures consistent, reproducible analysis.

Binary (yes/no) questions an LLM must answer to identify this aspect:

  1. 1

    Does the study use before-after comparisons on subjective scales (e.g., quality of life, pain)?

    Type: binary
  2. 2

    Could the participants' internal reference point or definition of the construct have changed between measurements?

    Type: binary
  3. 3

    Was a retrospective pre-test (then-test) used alongside the prospective pre-test?

    Type: binary
  4. 4

    Could adaptation or recalibration explain apparent improvements in the outcome measure?

    Type: binary
Deep Dive
The expandable detail section on each aspect page with examples, psychology, and counter-strategies.
The Deep Dive section provides in-depth information about each aspect: a real-world example showing the pattern in action, an explanation of why it works psychologically, practical advice on how to counter it, alternative names, and links to related aspects.