When More is Less: Incorporating Additional Datasets Can Hurt Performance By Introducing Spurious Correlations

Rhys Compton, Lily Zhang, Aahlad Puli, Rajesh Ranganath

Research output: Contribution to journalConference articlepeer-review


In machine learning, incorporating more data is often seen as a reliable strategy for improving model performance; this work challenges that notion by demonstrating that the addition of external datasets in many cases can hurt the resulting model’s performance. In a large-scale empirical study across combinations of four different open-source chest x-ray datasets and 9 different labels, we demonstrate that in 43% of settings, a model trained on data from two hospitals has poorer worst group accuracy over both hospitals than a model trained on just a single hospital’s data. This surprising result occurs even though the added hospital makes the training distribution more similar to the test distribution. We explain that this phenomenon arises from the spurious correlation that emerges between the disease and hospital, due to hospital-specific image artifacts. We highlight the trade-off one encounters when training on multiple datasets, between the obvious benefit of additional data and insidious cost of the introduced spurious correlation. In some cases, balancing the dataset can remove the spurious correlation and improve performance, but it is not always an effective strategy. We contextualize our results within the literature on spurious correlations to help explain these outcomes. Our experiments underscore the importance of exercising caution when selecting training data for machine learning models, especially in settings where there is a risk of spurious correlations such as with medical imaging. The risks outlined highlight the need for careful data selection and model evaluation in future research and practice.

Original languageEnglish (US)
Pages (from-to)110-127
Number of pages18
JournalProceedings of Machine Learning Research
StatePublished - 2023
Event8th Machine Learning for Healthcare Conference, MLHC 2023 - New York, United States
Duration: Aug 11 2023Aug 12 2023

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability


Dive into the research topics of 'When More is Less: Incorporating Additional Datasets Can Hurt Performance By Introducing Spurious Correlations'. Together they form a unique fingerprint.

Cite this