Library of Congress

Teachers TPS

The Library of Congress > Teachers > TPS Program > TPS Journal > Feature Article

Assessing Historical Thinking Skills Using Library of Congress Primary Sources

By Joel Breakstone & Mark Smith, TPS-Stanford University

The Common Core State Standards have set forth ambitious goals for student learning. Students are expected to cite textual evidence to support arguments, consider authors’ perspectives, corroborate among competing accounts, and develop written arguments. But what, exactly, can teachers do to help students master these challenging skills?

One proven strategy is to have students analyze primary sources. Research suggests that engagement with historical documents as part of an inquiry-based curriculum can promote learning of these higher-order skills (e.g., Reisman, 2012). A critical component of such a curriculum is assessment. If teachers are to promote complex thinking skills, they must have tools to regularly gauge student learning and adjust instruction to meet the needs of their students. Research has shown that this process, known as formative assessment, promotes student learning (Black and Wiliam, 1998). Teachers implementing this practice in classrooms need assessments that are reliable and precise indicators of whether students are, in fact, mastering these skills. Although the assessments most readily available to teachers today are not ideally suited for formative assessment of higher-order skills, teachers can use primary sources from the Library of Congress to create assessments that address the Common Core.

Neither multiple-choice questions nor document-based questions (DBQs) are designed to provide the quick, detailed feedback about student thinking that formative assessment requires. Students may get multiple-choice questions right for the wrong reasons or wrong for the right reasons, yet teachers only see blackened bubbles on a test form. Without information about student thinking, it is difficult for teachers to identify gaps in student understanding and revise instruction to meet student needs. At the other end of the spectrum, the DBQ is a rich, challenging task, but its length limits its use for formative assessment. Teachers with multiple classes might have to read hundreds of pages of essays to determine whether students have mastered particular skills! To build students’ skills in interrogating primary sources, teachers need tools that target specific aspects of document analysis and provide immediate feedback about student understanding.

History Assessments of Thinking

Drawing upon the unparalleled archives of the Library of Congress, the Stanford History Education Group has developed new types of assessments. History Assessments of Thinking (HATs) fill the void between the simple recall of multiple-choice questions and the complexity of DBQs. HATs address both content and analytical thinking skills, and each task can be completed in under fifteen minutes, some in less than five. Each question requires students to analyze primary sources. Students weigh the merits of competing claims, evaluate the reliability of evidence, and use evidence to support arguments. In the process, students practice the 21st century skills that the Common Core champions and teachers quickly gain insight into student thinking.

One question addresses a key aspect of document analysis: consideration of source information. Students examine a 1932 image of the First Thanksgiving from the Library of Congress archive and are asked to decide whether it would be useful to historians who want to understand the relationship between settlers and the Wampanoag in 1621.i More than three centuries separate the illustration from the event it depicts, but most students ignore this information.

Instead of considering the limitations of the document, most students fixate on the painting’s details. One student wrote: “You can see how they are interacting with each other. Without any picture, you couldn’t really see how Wampanoag Indians and the Puritans interacted with each other.” Other students examined the source critically but still skipped over the source information. Another student explained, “As soon as the settlers arrived, there was mass curiosity which turned into violence and hatred. There was never such a ‘party’ between the two peoples. They couldn’t even understand each other.” The student uses prior knowledge, but like the other student, engages in what we call “matching” – comparing the image to prior beliefs about the event. In both cases, teachers reading these responses quickly receive information about gaps in students’ understanding of document analysis that can be used to revise instruction.

Although many students struggled with this task, others crafted sophisticated answers. Consider this response, “This painting was drawn 311 years after the actual event happened. There is no evidence of historical accuracy, as we do not know if the artist did research before painting this, or if he just drew what is a stereotypical Pilgrim and Indian painting.” The student identifies the problem associated with the gap in time and also notes that additional information about the artist would help in evaluating the reliability of the source. A teacher reading a response like this can be confident that the student has a solid understanding of this particular skill.

The First Thanksgiving assessment focuses on students’ ability to “source” a document. Teachers also need tools for assessing other skills, including historical knowledge. Another HAT gauges understanding of the broad narrative of the Civil Rights Movement without simply asking students to recite facts. Students are given two letters from the NAACP’s archives at the Library of Congress. One letter, written by Eleanor Roosevelt to Walter White, the assistant secretary of the NAACP, describes her husband’s reluctance to intervene in the “lynching situation” ii. The second letter, written by Daisy Bates, a journalist and local NAACP official, to Roy Wilkins, executive secretary of NAACP, describes government efforts to support African American students integrating a previously all-white school in Arkansas.iii The dates are removed from the letters and students are asked a simple question: which letter came first? To successfully answer this question, students must analyze the text of both documents and place them within the broader context of the Civil Rights Movement.

Student responses to this assessment are revealing. Some students maintain that the lynching letter came later because the violence had been prompted by the desegregation of schools. This answer has a certain cause-and-effect logic, but it is historically inaccurate. Lynching had been virtually eradicated by the time of school desegregation in the 1950s. In contrast, strong answers demonstrate a firm grasp of the narrative of the broader Civil Rights Movement.

These types of targeted tasks provide teachers with rich information about student understanding. Rather than staring hopelessly at a sheet of darkened bubbles or wading wearily through stacks of essays, teachers can quickly scan a set of responses and identify patterns in student thinking. Teachers can subsequently use this information to adapt instruction. After teachers have addressed student misconceptions, they can re-assess using another version of the assessment that gauges the same skills but with different documents from the Library of Congress.

Dozens of these assessments are now available for free on a new Stanford University website ( The site also includes annotated student work, relevant Common Core standards, short videos that elaborate on key concepts, and links to documents in the Library of Congress’s digital archive. Yet, teachers should not feel limited to the assessments posted on the site. Instead, we hope the available assessments will serve to stimulate the imagination of teachers seeking to make new assessments.

To develop similar assessments, teachers can follow these steps:

  1. Choose what the assessment will measure. Historical knowledge? Students’ ability to corroborate sources? Students’ ability to construct an argument? Be as precise as possible. This initial step will guide the rest of the design process.
  2. Search the Library of Congress digital archive ( for appropriate documents, or visit the Library’s Teachers Page ( to find already compiled primary source sets relating to commonly taught topics, and create a rough draft of the assessment. For example, an assessment like the First Thanksgiving requires a document with a gap in time between the primary source and the event it depicts.
  3. Craft one or more questions about the selected source(s) that reflect the measure in step 1. Questions can be as simple as “Which came first?”, as in the civil rights example, or more complicated like “Could this source be used...?” as in the First Thanksgiving example.
  4. Before giving the assessment to students, consider what students must do to answer the question. Is the task actually focused on the skill or content identified in step 1? For instance, an assessment about corroboration that requires students to write a lengthy essay may provide more information about students’ writing abilities than their familiarity with corroborating sources.
  5. Try the assessment out with students. Read their responses and consider whether the prompt provided useful feedback about the content or skills identified in step 1. It can take a couple of drafts to produce a strong assessment.
  6. Revise the assessment and pilot it again with students. Review answers and make additional tweaks to the task as needed.

As teachers work to build students’ 21st century skills, they need tools to track student progress. Assessments created with Library of Congress primary sources provide teachers with timely feedback.

JOEL BREAKSTONE and MARK SMITH co-manage Teaching with Primary Sources program activity at Stanford University. Both are doctoral students working under the direction of Dr. Sam Wineburg, Margaret Jacks Professor of Education and History, and Director of the Stanford History Education Group.


Black, P. & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education: Principles, Policy, & Practice, 5(1), 7-73.

Reisman, A. (2012). Reading like a historian: A document-based history curriculum intervention in urban high schools. Cognition and Instruction, 30(1), 86‐112.


i. The First Thanksgiving