comparison_sciences Dataset

<p>The <code>comparison_sciences</code> dataset contains word-frequency and other non-consumptive-use data about 553,699 unique English-language news documents (no duplicate or close-variant documents) that contain the words &quot;science&quot; or &quot;sciences.&quot; The documents came from U.S. mainstream and student news sources published during 1977-2019 (though mostly from 1985-2019). WE1S researchers use this data to understand how public discourse about the humanities compares to public discourse about science.</p> <p>We gathered this data using keyword searches for &quot;science,&quot; which found articles containing either (or both) the words &quot;science&quot; and &quot;sciences.&quot; We took data from the top 10 circulating newspapers in the U.S. and from University Wire sources (student newspapers). Documents in this dataset may also contain the word &quot;humanities,&quot; just as documents in the <a href="https://zenodo.org/record/5068311"><code>humanities_keyword</code></a> dataset may contain the words &quot;science&quot; or &quot;sciences.&quot;</p> <p><em>(See <a href="https://we1s.ucsb.edu/research/we1s-materials/">WE1S Research Materials Overview</a> for the relation between the project&#39;s &quot;datasets&quot; and &quot;collections.&quot;)</em></p>