{{ progress }} %
  • {{ q.id }}. {{ q.short }}
{{ resultText.title }}
{{ score }} / 3 · {{ level }}

{{ resultText.answersHeading }}
# {{ resultText.questionCol }} {{ resultText.yourAnswerCol }} {{ resultText.correctCol }}
{{ a.id }} {{ a.text }} {{ a.responseLabel }} {{ a.correctLabel }}
Tags: Education , Psychology , Psychometrics
Export to PDF Fullscreen

Cognitive Reflection Test questions contrast a quick intuitive guess and a considered solution, showing how often reflection overrides impulse. The three question version offers a fast snapshot of your reasoning habits in a simple setting. Many people find it revealing on a first attempt.

You read each short problem, note the tempting reply, then choose an answer after a brief check. Results report how many times you reached the reflective choice across the set. A practical example is the bat and ball item where five cents is correct and ten cents is the common lure.

Scores help you notice patterns, not talent or worth. Pace yourself and aim for consistent inputs each time so repeats are comparable. Results do not constitute a clinical diagnosis.

Use the takeaways to build a short pause before committing. You can review each item and reuse the test later to see if the habit sticks.

Technical Details:

The Cognitive Reflection Test (CRT) compares an intuitive response with a normative solution across three items. The observed quantities are your selections per question and the resulting count of correct choices. It captures a snapshot of decision style for a brief set of word problems.

Computation turns each response into a binary indicator and sums them. The index S is the number of correct selections, ranging from 0 to 3, which allows simple comparisons across attempts.

Results map to four bands that describe the tendency toward reflective thinking. Crossing a band boundary signals more frequent overrides of the immediate impulse; values near a boundary can feel mixed in everyday use.

Comparisons are most meaningful within the same person over time. Group comparisons can be uneven when familiarity with the items differs or when practice effects are present.

Ci1 = { 1 , if responsei1 matches the correct choice 0 , otherwise }
S = C1+C2+C3
Symbol Meaning Unit/Datatype Source
S Total correct answers across three items Integer 0–3 Derived
Ci Correctness indicator for item i {0, 1} Derived
N Number of items 3 Constant
Worked example. Suppose your choices are incorrect, correct, correct, so C1=0, C2=1, C3=1. Then S=2. The score falls in the High band, indicating frequent overrides of the first impulse.
Threshold band Lower bound Upper bound Interpretation Action cue
Low 0 0 Intuition led most choices. Pause briefly before committing.
Medium 1 1 Some reflection evident. Practice a quick check once per item.
High 2 2 Strong reflection. Keep modeling the structure first.
Very High 3 3 Consistent reflection. Maintain the habit in new contexts.

Subskills surfaced by the items include algebraic difference, proportions and rates, and exponential growth. Incorrect choices often align with intuitive lures like ten cents, one hundred minutes, or twenty four days.

Field Type Min Max Step/Pattern Error text Placeholder
Responses Single choice per item 0 3 Must match one listed choice — —
Score Computed integer 0 3 n/a — —
URL code r String 3 chars 3 chars ^[0-3\-]{3}$ (- means unanswered) Invalid codes are ignored ---
Exports CSV, DOCX — — Enabled when all items answered Export controls remain disabled until finished —
Input Accepted families Output Encoding/precision Rounding
Three multiple‑choice selections Preset labels per item Score 0–3, band label, donut chart, analysis notes Integers and strings; CSV or DOCX export Not applicable

Networking & storage. Processing is client‑only with no network calls. Choices can be encoded in the page URL via the r query parameter to revisit or share the exact state.

Performance & determinism. Scoring is linear in the number of items. The same inputs always yield the same score and band.

Security considerations. Inputs are constrained to predefined options. If you share the URL, the embedded code reveals your selections.

Privacy & compliance. No data is transmitted or stored server‑side. Exports are generated on the device. Developed attribution in‑app notes: Shane Frederick, 2005.

Assumptions & limitations

  • The set contains three canonical items; breadth of reasoning is limited.
  • Not timed; speed is not captured.
  • Familiarity with items can inflate scores. Heads‑up
  • Guessing can produce correct picks without reflection. Heads‑up
  • Band labels are descriptive rather than diagnostic.
  • Practice effects reduce comparability across repeated attempts.
  • Color in charts encodes grouping, not risk.
  • Sharing the URL exposes your selections.

Edge cases & error sources

  • Invalid r codes are ignored and defaults are used.
  • Non‑three‑character r strings are rejected.
  • Partial answers yield no exports until all items are answered.
  • Rapid window resizing before drawing can delay chart layout.
  • Clipboard permissions may block “Copy CSV”.
  • Blocked downloads prevent file saves.
  • Browser history navigation may revert an earlier state via r.
  • Accidental refresh without r loses interim choices.
  • Language interpretation of question text can affect choices.
  • Sharing a copied URL may include stale codes after edits.

Step‑by‑Step Guide:

The Cognitive Reflection Test score summarizes how often reflection overrides the first instinct.

  1. Select Begin Test to start.
  2. Read each problem, note the tempting reply, then choose deliberately.
  3. Use the list to revisit items before finishing.
  4. Review the score, band, chart, and analysis notes.
  5. Copy or download CSV, or export a DOCX summary.
  6. Optionally share the page URL to preserve your selections.
Example. Choose five cents, five minutes, forty seven days to score 2, labeled High.
  • Take ten to fifteen seconds to translate each problem into a simple relationship.
  • Check whether scaling inputs should change time or leave it constant.
  • For growth, use the half or double heuristic to sanity‑check answers.

Finish with a short note on what to try differently next time.

FAQ:

Is my data stored?

No. Processing occurs on the device, and no data is sent to a server. Sharing the URL can reveal your selections to others.

No server storage or external calls.
How accurate is the score?

It counts correct picks on three items. Scores reflect pattern and habit more than ability. Repeated exposure can raise scores without deeper change.

Three‑item snapshot; practice effects apply.
What units or formats are used?

Outputs are integers for the score, a band label, and a simple chart. Exports provide a table of questions, your answers, and correct answers.

CSV and DOCX exports are available after completion.
Can I use this offline?

Yes after loading the page once. The assessment runs entirely on the device and does not require a connection during use.

Connectivity is only needed to access the page initially.
Is there any cost or license requirement?

No fees are indicated in the package. You may use the assessment for personal insight. Cite the original author for academic work.

Attribution: Shane Frederick, 2005.
How do I share results as a link?

Copy the page URL after answering. It includes a compact code that reproduces your selections when opened.

The r query parameter stores choice indexes.
What does a borderline result mean?

Scores near the middle reflect a mix of habits. Look at item‑level notes to see where a short pause or quick model could help most.

Use per‑item highlights to target practice.
Why is export disabled?

Export controls activate only after all three items are answered. Finish the remaining items to enable CSV and DOCX.

Completion is required for consistent summaries.