'Comparing different types of collaboration, general knowledge items' (AsPredicted #48,732)
Author(s) Maren Mayer (University of Mannheim) - maren.mayer@iwm-tuebingen.de Daniel W. Heck (University of Marburg) - daniel.heck@uni-marburg.de
Pre-registered on 2020/10/01 08:57 (PT)
1) Have any data been collected for this study already? No, no data have been collected for this study yet.
2) What's the main question being asked or hypothesis being tested in this study? Hypothesis 1: over the course of a sequential collaboration sequence judgments on general knowledge questions become more accurate
Hypothesis 2: judgments derived from sequential collaboration are more accurate than judgments derived from wisdom of crowds
3) Describe the key dependent variable(s) specifying how they will be measured. itemwise z-standardized judgments of general knowledge questions, instead of the item mean the correct answer is used
4) How many and which conditions will participants be assigned to? Two conditions: wisdom of crowds (all participants in this condtion answer all 65 questions presented) vs. sequential collaboration (the first participant provides a wisdom of crowds judgment pattern, following participants are shownn the latest judgments and can correct them)
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. Linear mxied models with judgment as criterion, position in sequence (Hypothesis 1) or condition (Hypothesis 2) as fixed effects and item as random effect are conducted.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. We will exclude participants that provide irregular answer patterns and participants whose judgments are correct for more than 10% of the items answered as we suspect that the answers were looked up during participation.
We will exclude judgments that were given when the time limit for judgments (40 seconds) was reached since some participants may not yet have completed their answers. We will compute median and quartiles for every item before standardizing and exclude judgments that are defined as outliers in a boxlplot (i.e. lower than 0.25-quartile - 1.5 interquartile range or higher than 0.75-quartile + 1.5 interquartile range).
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. We will collect 600 participants through an online panel (respondi).
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) We will use 65 general knowledge items.
Participants are randomly assigned to the conditions. Participants completing the wisdom of crowds condition can serve as first participant for a sequence. Sequences are started from the latest wisdom of crowds data entered. If (usually at the beginning of the experiment) no judgment pattern from wisdom of crowds is available, participants are redirected to the wisdom of crowds condition.
The data will be collected through an online experiment. Since it is possible to look up correct answers, switching the browser window is tracked and excessive switching (more than 5 times while giving judgments) leads to a screenout of this participant. Participants are instructed not to switch the browser window and are warned whenever they do so.
