Questions people ask before they sign

Programs that test 500 to 50,000 people a year: certification and licensing bodies, colleges and districts, training providers and employers. Smaller programs can use it too, but the platform starts at $6,000 a year, so a program testing a few dozen people may be better served by simpler tools.

Yes. The candidate interface is built to WCAG 2.1 AA and tested with screen readers each release. Accommodations such as extra time, breaks, larger text and high contrast are set per candidate before the window opens, and the candidate portal is available in English and Spanish.

Yes, if it supports LTI 1.3. Canvas, Moodle, Blackboard and D2L Brightspace all do. Candidates launch the exam from their course, sign in once, and grades pass back when you release scores. Single sign-on through SAML 2.0 or OpenID Connect works alongside it.

Within one business day of a window closing. Each form gets difficulty (the share of candidates answering an item correctly), discrimination (how well an item separates stronger from weaker candidates), distractor counts and reliability. Your lead walks through anything unusual before the next window opens.

Yes. Practice tests run on the same platform with their own settings, such as repeat attempts and instant feedback. Roediger and Karpicke (Psychological Science, 2006) found that students who took recall tests remembered more two days and one week later than students who spent the time restudying.

AI suggests a score for each essay or short answer against your rubric and shows the rubric lines it matched. A trained scorer approves or changes that score before release. Objective items such as multiple choice score instantly. Every suggestion, final score and change is logged with the scorer’s name.

Your scorers: instructors, examiners or subject matter experts from your field. A Rubric Calibration Sprint trains them to apply the rubric the same way, and the platform reports agreement between scorers (inter-rater reliability), so you can see where two people drift apart on the same response.

No. AI can draft items from your blueprint, but every draft is edited and approved by a subject matter expert from your field, then reviewed for bias and accessibility before it can join a form. The item keeps a record that it began as an AI draft. Our responsible AI rules set out the rest.

No. Candidate responses, recordings and personal data are never used to train third-party models. We use model providers whose terms bar training on the data we send them, and you can ask in writing for the current list of providers and the terms that apply to your program.

Yes. A Grading Pilot, from $2,400, runs automated grading on one course or exam for one term. Your scorers work through the same review queues they would use later, and at the end you see how often they accepted suggested scores before you decide whether to continue.

Record and review, from $6 a session, records camera, microphone and screen; software flags moments and a trained person reviews every flag within two business days. Live proctoring, from $18 a session, adds a proctor watching in real time, with up to six candidates per proctor, who can pause a session or help.

No system stops all cheating, and we will not claim ours does. Proctoring deters, detects and documents: it makes misconduct harder, flags unusual moments and keeps a record a person can review calmly. Decisions about a candidate stay with your program, under its own published policy.

A computer with a webcam and microphone, a stable internet connection, a supported browser and photo ID. A short system check in the candidate portal confirms the setup days before the exam, and the candidate-day line helps anyone who runs into trouble during a scheduled live window.

The session pauses and every answer given so far is saved. When the candidate reconnects, they resume where they stopped with their remaining time intact, and the record shows the gap. If they cannot reconnect, the proctor or reviewer notes it, and your program decides on a retake under its policy.

Yes. Before the exam starts, candidates see what will be recorded, who can view it, how long it is kept and how flags are reviewed, and they agree before continuing. Recordings are kept for the period your program sets, then deleted, and the deletion is written to the audit log.

Platform and certification management subscriptions are annual, and larger programs can sign multi-year terms. Programs such as Exam Launch or a Job Task Analysis are fixed-fee pieces of work with an end date. Every proposal is itemized, so you can see which lines renew each year and which do not.

You do. Items, candidate records, results and recordings belong to your program. You can export the item bank as QTI and results as CSV at any time. On exit we hand over a full export and delete our copies within 30 days, unless you ask us in writing to keep them.

In the United States. Data is encrypted in transit with TLS 1.2 or later and at rest with AES-256, access is limited by role and every action is logged. Our security page describes the practices plainly, and we answer security questionnaires in full on request.

Yes. We review your data agreement as part of contracting, including student data terms for schools and colleges that handle records under FERPA. Send it to legal@exampit.com before or after your demo. For questions about your own legal obligations, please ask your counsel; we do not give legal advice.

Support specialists answer support@exampit.com Monday to Friday, 9 a.m. to 5 p.m. Mountain, and staff a candidate-day line during scheduled live windows. Annual platform fees include support, hosting and updates, so there is no separate support contract to buy or renew.