San Francisco Field Site

Current Research Collaborations

• WordSift: How do visualizations of text support teacher and student understanding of academic language and content area literacy?

  Within the context of our partnership with SFUSD, researchers from Stanford University developed an online tool that visually displays academic language and high-frequency vocabulary found in academic texts. The researchers, working together with a team of SFUSD teachers and district administrators, are developing instructional uses for an online text preview tool (www.wordsift.com) aimed at helping content area teachers work successfully with their students using academic language and developing content area literacy.

  Timeline: November 2008 to June 2010

  District co-developers:
  Jeanne D'Arcy, Supervisor, Mathematics and Science
  Deb Farkas, Instructional Support Specialist, Middle School Science
  Ritu Khanna, Assistant Superintendent, Research, Planning, and Accountability

  Teacher co-developers:
  Joel Austin, Roosevelt Middle School
  Karen Clayman, A.P. Giannini Middle School
  Lisa Ernst, Alice Fong Yu K8 Alternative School
  Michael Fox, Denman Middle School
  Patricia Kudritzi, Aptos Middle School
  Stephie Prout, Hoover Middle School
  Lisa Beth Watkins, A.P. Giannini Middle School

  Researcher co-developers:
  Ed Haertel, Professor of Education, Stanford University
  Kenji Hakuta, Professor of Education, Stanford Univeristy
  Diego Roman, Graduate student, Stanford University
  Karen Thompson, Graduate student, Stanford University
  Greg Wientjes, Graduate student, Stanford University

  The first phase of this work involved exploring and developing instructional uses for WordSift. This online tool has been developed for teachers and students to use in previewing text to increase comprehension. The site takes in text that is pasted into a field and then creates a tag cloud (list of the most common words, with the size of the words related to its frequency in the text). It also marks the “academic vocabulary” such as process and procedure in a different color. The most common word is displayed in the Visual Thesaurus with the results of picture and video searches of the top two common words from Google. These functions are intended to help the teacher and student preview the text, allow teachers to assess student understanding of the vocabulary, and generally talk about the content of the relevant text.

  The second phase of this work involved setting up a 'micro-experiment' to determine the effect of previewing text in WordSift on students' reading comprehension. A team of middle school science teachers with research support has conducted micro-experiments with 498 students in 19 different 6-8th grade science classes. In the “treatment” condition, students previewed the text in WordSift, then read the text and took a comprehension assessment. In another class session, the “control” condition, the students previewed the text without using WordSift, read the text and took a comprehension assessment.

  Controlling for variables that matter (gender, gifted and talented designation, school’s API score, and school’s ethnic diversity), the research team found out that the variation found in reading comprehension assessment scores was not significantly related to differences based on exposure to WordSift. Rather, the variation was due to differences among characteristics of the students and the schools in which they are enrolled.

  More work is still needed to understand how teachers and students are using this tool and under what conditions continued usage could lead to greater vocabulary development and reading comprehension.