When students interact with information outside of a classroom setting, it's rarely as part of a packet of pre-vetted sources. Instead, students need to search for the truth and decide how far to trust the information they encounter. In other words, they will be determining an author's purpose and evaluating their credibility based on context clues including the type of source and its author's authority.

The techniques introduced in this module are presented for use with non-fiction information sources, but may provide connections for discussing related literary analysis skills. For example, regardless of a source's content, close reading and annotation both require readers to determine the key points within a text, synthesize the information with what they already know, and somehow respond to it. A thorough student may not have to re-read a piece of fiction as many times as an explanation of a scientific process or an argument on school board appropriations, but in both cases they are extracting useful information from a source.

Readers typically expect authors of non-fiction to have a greater knowledge of facts than authors of fiction. Fiction authors are often held to the related standard of verisimilitude: whether something seems real or has subjective, emotional truth. This isn't to suggest that all interpretations of a piece of fiction are equally valid, but rather that the processes for understanding and evaluating these two sources often require different rubrics. These lessons are best left to the experts in your language arts curriculum.

In the following pages we will look at several ways to read and evaluate an author's purpose, with a strong focus on non-fiction works.

A Note on Generative AI

As we mention in Unit 1, many traditional search tools now include elements of generative AI. In the following module, we maintain the stance that using generative AI to evaluate a source can obscure the aspects that are most relevant to a student's growth. Mike Caulfield, author of the SIFT method, is more open to the mediated use of generative AI. His SIFT model, modified for use with generative AI, is an effective demonstration of how searchers can compare different ways of evaluating sources.