Constructed-response questions sit between multiple choice and a full essay. Students write a paragraph or two answering a specific question about a text. They are quick to administer and quicker to grade than essays, at least in theory. In practice, teachers often score them inconsistently because the criteria are not written down.

Franklin's Autobiography suits this format well. Passages are self-contained, and questions can target a specific idea, such as why Franklin decided to work on one virtue at a time. Students must read carefully and support their answers with evidence.

A good scoring guide answers three questions in advance. What does a complete answer include? What counts as partial credit? And what mistakes do you expect to see? Writing these down before grading saves time and reduces arguments.

The sections below give sample questions, a scoring scale, and strategies for handling large stacks.

Sample questions

Effective questions ask students to explain, not just recall. "Why does Franklin add humility to his list of virtues, and what does the way he describes it suggest about his attitude toward pride?" requires understanding and interpretation. "What virtues did Franklin list?" does not. Keep questions focused enough to answer in a paragraph.

• Why does Franklin work on one virtue at a time, and what does this show about his approach to change?
• How does the story of the speckled ax help Franklin explain his view of perfection?
• What does Franklin's description of studying The Spectator reveal about how he thinks writers improve?
• Why might Franklin choose to tell his readers about his early mistakes?
• How does Franklin's tone in the account of the virtues chart affect the reader's trust?

A four-point scoring guide

A four-point scale works for most constructed responses. A four answers the question fully, uses specific evidence, and explains its significance. A three answers correctly with adequate evidence but limited explanation. A two shows partial understanding, with vague or incomplete evidence. A one shows minimal understanding, and a zero is blank or off-topic.

Write a sample answer at each level for every question. Even a sentence or two per level is enough. These become your reference when scoring and can be shared with students afterward as a learning tool.

Handling partial credit

Partial credit is where disagreement grows. Decide ahead of time how to treat an answer that is correct but has no evidence, or one with good evidence but a weak explanation. A simple rule, like capping answers without evidence at a two, keeps things fair.

Watch for answers that are right for the wrong reasons. A student who stumbles into the correct idea through a misreading should not receive full credit. Look at the reasoning and not only the conclusion.

Speed strategies for 120 responses

Score one question at a time across all papers rather than all questions on one paper. You will remember the criteria better and spot patterns faster. Keep a tally of common errors for a quick class-wide follow-up.

Rubric-driven AI grading tools can take a first pass at short responses, applying your scoring guide to each answer and flagging borderline cases. You then focus on the flagged ones and spot-check the rest. That combination keeps both speed and accuracy in reach without handing your judgment to a machine.