LA Pathway A: Analyze Text through Computational Methods (Overview)

These building blocks center around the idea of a mixed literary analysis, teaching students data science methods and tools to analyze texts.

While the mere taught of blending the disciplines of data science and English can sound daunting to many, these building blocks will walk Language Arts (LA) educators through using paper and pen "unplugged activities" to explore and graph textual data then conclude with a lesson that encourages educators to engage in a distance reading, uploading texts to web-based applications to support or enhance literary analysis.

This building block highlights four practices to integrate computational thinking in the language arts classroom:

A1. Analyze text through algorithmic exercises.

A2. Identify patterns in texts by abstracting textual data

A3. Conduct a mixed analysis of entire texts by examining quantitative and qualitative textual data and patterns

A4. Discuss the affordances and limitations of using quantitative data to analyze texts

The next steps in this project will break down each of the four practices a bit further and provide brief examples of how each can look and feel in a classroom.

As stated earlier, the other two projects in this building block will guide you through more in-depth Language Arts activities covering practices A1 and A3.

As you complete this project, and others in this building block, consider the following questions for your own classroom instruction:

• What does it look like to apply computational methods to analyze texts in the classroom?
• Does analyzing text through computational methods support student outcomes? How?

A1 focuses on using algorithms to:

• Identify and analyze sequential or repeating literary devices (eg, sound devices, parallel structure, plot, POV) in a text by using flowcharts, storyboards or other representational forms highlighting sequence.
• Identify, analyze, and critique an author’s word choice by following an algorithm.
• Discuss the strengths and weaknesses of using an algorithmic approach to identify and analyze sequence and repetition in texts.

Practice in Action

An activity could entail having students generate a list of questions and rules to determine if the narrator is a character in the story. Students could then take the questions and rules to create point of view (POV) flowcharts to identify if the narrator is a character in a story. Finally, students could compare and contrast flowcharts and discuss why an author may have selected that particular point of view for the narrator.

Creating flowcharts can be reviewed in a prior introductory Building Block >

A2 focuses on identifying patterns in texts by abstracting textual data through the following methods:

• Collect and organize data from a text.
• Visualize patterns in a text (e.g. word frequency, syllable counts, meter) using tables, charts, and graphs.
• Analyze text by tracing patterns in usage and interactions of specific words or ideas in quantitative textual data and visualizations. Trace patterns in usage and interaction of specific words to analyze a text.

Practice in Action

An activity could start with a class discussing and brainstorming rules to determine what makes a sentence negative, neutral or positive. A class could then break into student pairs. Each pair would be assigned a section of a news article. Student pairs then highlight sentences in a news article and tag them as being overall negative, neutral or positive. Student pairs count sentences and tally them in a classroom chart broken down by article section. The whole class then rates each section with a sentiment score of negative, neutral or positive and discusses any patterns or reevaluates their rules for determining if a word or phrase is negative, neutral or positive.

A3 focuses on conducting a mixed analysis of entire texts by examining quantitative and qualitative textual data and patterns in the following ways:

• Use a computational tool to organize textual data into visualizations in order to identify patterns in the text.

• Use quantitative textual data and visualizations to identify patterns in a text (i.e. phrases, word choices, topics) that shape meaning or tone in order to re-read and analyze a text more closely.

• Use your insights from a close reading of a text to identify notable elements of the text (i.e. phrases, word choices, topics, characters, point of view) to pose questions using quantitative textual data and visualizations.

• Compare and contrast two or more texts using textual data and visualization in order to identify patterns and pose questions about how texts relate to each other, the reader, and the world.

Practice in Action

Continuing the prior example of sentiment analysis in a news article, students could re-read sections that stand out because of tally chart data, perhaps the section where the article shifts from neutral to negative. Students could compare the chart of a different article on the same topic generated by different classes or pre-generated by the teacher and pose questions. Students could take other articles written by the same author, news source or on the same topic and upload them to a web-based application that analyzes sentiment. Finally, students could take data and visualizations generated by a web application and cite them as evidence in an essay about how a given topic is presented in the media or how sentiment analysis can be used to determine author bias