Christian Kroer's dataset contains all generated notifications for a subset of Instagram users across four notification types within a specific time window. It was collected during an A/B test comparing first-price and second-price auction systems for notification optimization.
Use Cases
- Analyze auction outcomes across four notification types to compare first-price and second-price auction system performance.
- Study fair online allocation and Fisher market equilibrium using notification auction data.
- Model user engagement by correlating notification send decisions with user activity metrics implied by the A/B test framework.
Strengths
- Data originates from a real-world A/B test on the Instagram platform.
- Includes information across four distinct notification types.
- Designed to support research in fair online allocation and market equilibrium.
Limitations
- The dataset scope is limited to a subset of Instagram users, which may not be representative.
- Specific sample size, column details, and time range are not provided in the input.
- Data is tied to a specific experimental auction mechanism, limiting generalizability.
Provenance
- Source
- ICPSR Harvested Dataverse
- Collection Method
- Collected during an A/B test on Instagram comparing auction systems for notification delivery.
- Time Range
- A certain time window (specifics unknown).
- Freshness
- null
- Geography
- null