Synonym-based word frequency analysis to support the development and presentation of a public health quality improvement taxonomy

Abstract

CONTEXT: Researchers and analysts have not completely examined word frequency analysis as an approach to creating a public health quality improvement taxonomy.

OBJECTIVE: To develop a taxonomy of public health quality improvement concepts for an online exchange of quality improvement work.

DESIGN: We analyzed documents, conducted an expert review, and employed a user-centered design along with a faceted search approach to make online entries searchable for users. To provide the most targeted facets to users, we used word frequency to analyze 334 published public health quality improvement documents to find the most common clusters of word meanings. We then reviewed the highest-weighted concepts and categorized their relationships to quality improvement details in our taxonomy. Next, we mapped meanings to items in our taxonomy and presented them in order of their weighted percentages in the data. Using these methods, we developed and sorted concepts in the faceted search presentation so that online exchange users could access relevant search criteria.

RESULTS: We reviewed 50 of the top synonym clusters and identified 12 categories for our taxonomy data. The final categories were as follows: Summary; Planning and Execution Details; Health Impact; Training and Preparation; Information About the Community; Information About the Health Department; Results; Quality Improvement (QI) Staff; Information; Accreditation Details; Collaborations; and Contact Information of the Submitter.

CONCLUSION: Feedback about the elements in the taxonomy and presentation of elements in our search environment from users has been positive. When relevant data are available, the word frequency analysis method may be useful in other taxonomy development efforts for public health.