Automatic jargon identifier for scientists engaging with the public and science communication educators

Area of Science:

• Science Communication
• Computational Linguistics

Background:

• Effective science communication requires tailoring messages for diverse audiences, including non-experts.
• Scientists often struggle to avoid specialized jargon due to their training, lacking standardized guidance.

Purpose of the Study:

• To develop and validate a scientist-friendly program, the De-jargonizer, for identifying jargon in written texts.
• To assess the De-jargonizer's effectiveness in distinguishing jargon levels across different text types.

Main Methods:

• Developed the De-jargonizer using a large corpus (90+ million words) from BBC News (2012-2015).
• Validated the tool through three studies: correlation with word lists, comparison with spoken language jargon data, and analysis of research abstracts vs. lay summaries.

Main Results:

• The De-jargonizer significantly correlates with existing frequency word lists.
• It replicates known jargon differences between scientific and general lectures.
• Identified high jargon levels even in lay summaries, exceeding public comprehension thresholds.

Conclusions:

• The De-jargonizer is a validated tool to help scientists identify and reduce jargon for broader public understanding.
• It can be implemented in science communication training to evaluate and improve clarity.