Empirical Analyses of the Bias in User Based Randomized Control Trials Created Due to Difference in Unit of Randomization and Unit of Measurement in Digital Ad Delivery Platforms

Author(s)	Varun Chivukula
Country	USA
Abstract	impact of ads. However, misalignment between the units of randomization and measurement introduces biases that significantly distort lift outcomes. This paper examines such challenges when the unit of randomization is an ad platform user, but the unit of conversion measurement aligns with advertiser accounts. Using a probabilistic model, we quantify the understatement in measured lift caused by this misclassification, identify drivers, and propose potential solutions.
Keywords	Causal inference, Digital Ad platforms, Randomized control trials, Misclassification Bias, Halo effect
Published In	Volume 3, Issue 2, March-April 2021
Published On	2021-04-26
Cite This	Empirical Analyses of the Bias in User Based Randomized Control Trials Created Due to Difference in Unit of Randomization and Unit of Measurement in Digital Ad Delivery Platforms - Varun Chivukula - IJFMR Volume 3, Issue 2, March-April 2021. DOI 10.36948/ijfmr.2021.v03i02.26696
DOI	https://doi.org/10.36948/ijfmr.2021.v03i02.26696
Short DOI	https://doi.org/g82h5w

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research