A digital essay is still an essay. It makes a claim, develops an interpretation, works with evidence, and asks a reader to move through an idea. The difference is that the claim may be carried by text, image, audio, layout, sequence, annotation, video, or interaction. If the rubric treats those choices as decoration, students will treat them that way too.
The first mistake is to grade the technology instead of the thinking. A student can build a polished website that says very little. Another student can make a plain but careful digital essay that uses two maps, a short voiceover, and quoted primary sources to make a real historical argument. The second project is better, even if it looks less like a portfolio template.
Writing studies gives teachers a useful starting point. The CCCC position statement on writing assessment argues that assessment should be tied to clearly stated learning goals, that criteria shape how students understand writing, and that assessment should respect varied language practices and contexts. That applies directly to multimodal work. Students need to know whether they are being assessed on historical interpretation, rhetorical design, technical execution, collaboration, or all of those at once.
A digital essay rubric should begin with the same questions a teacher would ask of a print essay. What is the claim? What evidence supports it? How does the student handle counterevidence or ambiguity? Is the structure purposeful? After that, the rubric can ask what the added modes do.
Multimodality means using more than one mode to make meaning. The University of California, Irvine's writing program describes multimodal composing as work that may involve written, oral, digital, visual, spatial, or embodied modes, and emphasizes the relationship between mode, audience, purpose, and circulation. That is the important part. A video clip is not automatically meaningful. A map is not automatically analysis. The student has to choose a mode because it can do something the prose alone cannot do.
For a humanities class, I would keep the rubric plain: argument, evidence, modal fit, navigation, and reflection. The project should make an interpretive claim rather than collect materials. Quotations, images, maps, and archival items should be selected and cited with care. Each medium should earn its place. The reader should know how to move through the project without being trapped in a maze of buttons. The student should also be able to explain why the project uses these forms.
One way to make that concrete is to ask students to identify the "load-bearing" element in the project. In a conventional essay, the load-bearing element might be a paragraph that interprets a key passage. In a digital essay, it might be an annotated image, a sequence of two maps, or a clipped interview segment. If the project would say the same thing after that element was removed, the element is probably decorative. If the argument collapses without it, the student has found a real multimodal choice.
The designer's note is worth keeping. Ask students for 300 words with the project. "I used the audio clip here because the cadence of the speech matters." "I placed the newspaper image before the paragraph because I wanted readers to see the headline before my interpretation." "I avoided background music because it made the oral history clip feel manipulative." These comments show whether the student made choices or simply filled a platform.
A 2020 multimodal assessment framework by Jen Ross, Jen Scott Curwood, and Amani Bell argues that multimodal assessment has to account for criticality, creativity, holism, and valuing multimodality. The word "holism" is useful here. Teachers can break a rubric into categories, but they should still read the project as a whole. A digital essay can fail if the parts are individually competent but do not work together.
There is also a labor issue. Digital projects can become grading traps. If every student submits a website, a podcast, and a video, the teacher may spend hours troubleshooting broken links instead of assessing learning. Keep the assignment bounded. Require a limited number of modes. Require an export or backup. Have students submit a process log with links, assets, and citations. Decide in advance how you will handle inaccessible files.
Accessibility belongs in the assessment, but it should be taught before it is graded. Students should know that images need alt text, audio needs a transcript or summary, colors need contrast, and navigation should not depend only on tiny icons. This is part of audience awareness, not a compliance chore pasted onto the end.
Teachers should also decide how much technical failure counts. A broken embed can happen for reasons outside a student's control. A confusing structure, missing citation, or unreadable color choice is part of the student's design. I usually separate "platform failure" from "compositional choice." If a tool breaks, the student should have a backup link or exported file. If the project is hard to follow because the student chose a maze-like layout, that belongs in the grade.
Peer review should happen before the final version. Give reviewers three questions: What claim do you think this project is making? Which non-text element helped you understand it? Where did you get lost? Those questions are better than asking whether the project is "good." They tell the author whether the intended argument is visible to a reader.
The strongest digital essays usually feel restrained. They do not use every feature in the tool. They use enough form to make the argument clearer. A student writing about migration might pair a short essay with an annotated route map and two oral history clips. A student writing about propaganda might use image comparison and close reading. A student writing about a poem might record a reading and annotate sound, line breaks, and typography. None of these projects needs fireworks.
So the grading question is not "Did the student make something digital?" It is: did the digital form help the student think and help the reader understand? If the answer is yes, the project is doing real humanities work.