Click to upload or drag and drop
PDF, DOCX, PPTX, TXT, JPG, JPEG, PNG, HEIC, ODP, ODT, BMP, or TIFF
up to 20MB
Uploading...
No. Not one state that mandates harassment training requires a quiz, and no quiz satisfies the mandate on its own. California counts hours, New York and Illinois count years, and all of them regulate the delivery of the training rather than the result. If you are asking whether skipping the quiz puts you out of compliance, the answer is that it does not.
The more useful question is the one employers usually mean when they ask this: does an attendance record actually protect us? That is where the answer changes. Every one of these laws produces evidence that a session happened and no evidence that anyone understood it, and understanding is what gets argued about later. Building the comprehension check from your own policy with a harassment training quiz generator costs a few minutes and closes the part of the file that is genuinely thin.
Here is the short version of each mandate alongside the specific question this article is about. The pattern is consistent enough to be worth stating plainly: content is sometimes prescribed, duration is sometimes prescribed, assessment never is.
| Jurisdiction | What is mandated | Quiz required? | What the record proves |
|---|---|---|---|
| Federal (Title VII) | Nothing. The EEOC's April 29 2024 harassment guidance treats an effective, widely disseminated, understandable policy and complaint process as part of reasonable care. | No | Whatever you choose to keep. There is no prescribed record at all. |
| California, Gov. Code 12950.1 | Two hours interactive training for supervisors, one hour for everyone else, every two years; new hires within six months; short-term seasonal staff within 30 days or 100 hours worked | No | 2 CCR 11024 asks for the employee name, date, type of training, and provider, kept two years. Attendance, not comprehension. |
| New York, Labor Law 201-g | Annual interactive training meeting or exceeding the state model, for every employer regardless of size | No | That an interactive session covering the required topics was delivered each year. |
| Illinois, 775 ILCS 5/2-109 | Training for all employees at least once every calendar year, for every employer including one-person shops | No | Completion within the calendar year. Restaurants and bars carry a supplemental program under 5/2-110. |
| Connecticut | Two hours for all employees at employers with three or more, and for supervisors at any size; new hires within six months | No | That the two hours happened. The absence of a stated refresh interval is the usual tracking failure. |
| Delaware | Interactive training within one year of hire at employers with 50 or more employees, repeated every two years | No | Completion and date. No hours are specified. |
| Maine | Training within one year of hire at employers with 15 or more employees, covering prescribed content including the complaint process and retaliation protection | No | That the prescribed topics were covered. Content is specified, duration is not. |
Cities layer their own rules on top of several of these, and thresholds are counted differently from state to state, so a multi-state employer cannot run one roster rule across the whole workforce. Teams that operate in more than three of these states usually end up needing something that watches for changes to the underlying rules rather than a spreadsheet somebody updates once a year.
Because the record the law asks for and the record a dispute asks for are different documents.
The training file that gets produced in a harassment matter usually contains a sign-in sheet, a completion export from an LMS, and a copy of the policy with an acknowledgment signature. Read as a set, those establish that the employer delivered something. What they cannot establish is the fact that tends to matter: whether the employee knew there was a second reporting route when the person they would normally report to was the subject of the complaint, and whether the supervisor who heard the first informal remark knew they had to escalate it.
Those two failures account for a large share of the cases where an employer had a perfectly good policy and still lost the argument about reasonable care. A completion record cannot speak to either. A scored quiz that asked both questions can, and it costs one extra step at the end of a session you were already running.
The EEOC's 2024 guidance nudges in the same direction from the liability side rather than the compliance side. It describes an effective policy as one that is widely disseminated and understandable to employees, including employees facing language or literacy barriers. An email log speaks to dissemination. Nothing in a standard training file speaks to understandable.
Weight it by consequence, not by how much text each section occupies in the policy. A generated first draft will do the opposite, because definitions near the front of a document are the easiest thing to convert into a question, so plan on rebalancing it once.
For a 15-item quiz on a typical policy, a distribution that works:
This is the objection worth taking seriously, and the honest answer is that a failed quiz with no documented follow-up is worse than no quiz. It records that you identified a gap and left it open.
The fix is to decide the threshold and the retake path before you run it, not after you see the results. Set the bar, run the quiz, and where someone falls short, document the retake or the short re-brief and the second score. Handled that way the failure becomes evidence of a working process. Handled carelessly it becomes an exhibit.
The same logic argues for short quizzes tied to a specific policy version rather than one long annual assessment. A 10-item check attached to the revision it was written from is easy to re-run when the policy changes, and "trained on the 2024 policy while the 2026 revision is in force" is a quietly common problem in files that otherwise look complete.
Yes. California already treats the two audiences differently by requiring two hours for supervisory employees against one hour for everyone else, and the substantive reason is duty rather than depth.
A supervisor who learns about conduct informally generally puts the employer on notice, whether or not anyone filed anything. That makes their obligation situational, and situational knowledge is tested with scenarios rather than definitions. Keep the shared core of the employee quiz, then add three or four items that put the supervisor in the moment: an employee raises something in a one-on-one and asks you not to escalate it, a complaint concerns a peer manager, a rumor reaches you secondhand.
No. Every state mandate is written against the training itself. California requires two hours of classroom or other effective interactive training for supervisors and one hour for other employees. New York requires annual interactive training. A quiz is a comprehension check layered on top of that session and does not substitute for any part of it.
California's FEHA regulations at 2 CCR 11024 set the clearest floor: at least two years, including the employee name, training date, type of training, and provider name. Webinar delivery adds obligations, including retaining a copy of the webinar and the written questions and responses for two years. Other states are less prescriptive, so most multi-state employers keep everything to the strictest applicable standard.
Online is generally acceptable where it is genuinely interactive. New York's requirement turns on interactivity rather than format, and California expressly contemplates classroom, webinar, and interactive online delivery. A recorded video with no participation element is the delivery style most likely to fall short of what interactive is understood to mean.
Generally yes in the mandating states. Illinois applies to all employees including short-term, part-time, and remote workers. California sets a specific accelerated rule for seasonal and temporary staff hired for under six months: within 30 calendar days of hire or 100 hours worked, whichever comes first. Employee location usually drives which state's rule applies, which is what makes distributed teams complicated.
An acknowledgment proves the employee received the document. A quiz proves they could use it. Those are different claims, and only the second one answers the question that actually gets asked after an incident, which is whether the employee knew what to do. Most employers should keep both, since the acknowledgment is trivial to collect and the quiz is what carries weight.
Run the training your state requires, on its schedule, in an interactive format. Then add a short quiz built from your own policy, score it, attach the score to the named employee and the policy revision, and keep it as long as you keep the attendance record. The law will not thank you for it. The file will be considerably harder to pick apart.
If your policy already exists as a PDF or a handbook section, the fastest way to get a first draft is to upload it and let the questions come from your own language. Related reading: building a quiz from an employee handbook, and compliance training quizzes more broadly.
From the same family of tools