Entry title
- First pre-test participant.
- The data look really good, just as we expected. Inflation seems like it is going to trip people up, and getting explanations with any depth is going to be a challenge.
Unexpected Observations
New Hypotheses
Current Protocol
Current Materials
- Error, forgot to record the materials.
New Data
Participant 1
- Predicted prices to go up from 51 to 74, consistent with our expectations.
- First estimate took 100 seconds, total estimates 200 seconds. It seems as though once the theory was in mind the estimates were easier to generate. 100 seconds seems like a fair amount of deliberation.
- Explanation with outcome knowledge matched the pattern of observed results exactly (check mark). Confidence increased.
Participant 2
Faults
- Be specific for explaining the prices. (e.g., "Please write 2-3 sentences" or "Write at least 20 words")
- Predictions were increasing, but based on the argument of inflation. This is a serious problem and needs to be corrected.
- Spent only 16 seconds on the explanation.
Corrections
- Added title to demographics page.
- Highlighted MTurk code.
- Added request for "as much detail as possible" for the explanations.
- Added a star on the inflation note to emphasize it.
- Changed pay to $2.50 as it took less than 15 minutes.
- Removed "I don't know" option for judgment of trend.
- Add second attention check to the coal price forecast page.
- Added third check for the initial price estimates to highlight that inflation cannot be a factor.
|