| A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | AA | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | Name | Question for the Group | Notes | Category | |||||||||||||||||||||||
2 | September 22, 2026 | ||||||||||||||||||||||||||
3 | |||||||||||||||||||||||||||
4 | |||||||||||||||||||||||||||
5 | |||||||||||||||||||||||||||
6 | |||||||||||||||||||||||||||
7 | August 25, 2026 | ||||||||||||||||||||||||||
8 | Kara Prichard, Linux Foundation | For those delivering programs in Japan, do you have a preferred browser/add-on combo or workflow for translating English web content into Japanese? I’m familiar with Chrome + Google Translate, but I’d love to hear what else people are using or recommending. (DeepL, localization tools, other?). | SAS - We rely on local staff in Japan to do our translations - We have an internal license for MS Copilot that keeps our content secure and we allow use of that but not open AI Chat assistants. Text translations are solid but it can't translate screen shots of our software, so that is all still manual. We use ITS Item Workshop for item banking and leverage its Translation Assignment functionality to present the ENU and JPN items side-by-side. Additional Q's? mark.stevens@sas.com - sorry that I could not be on the call today. Kryterion - I believe our clients that administer in Japan either go through professional translation services, or have bilingual staff that do their translations for Japanese test forms. Workday - sending out to 3rd party translation service with local review internally. We have a gap and do not hold the English version. ServiceNow - We don't gap, we push our translation vendor to have it launch at the same time. But sometimes that doesn't work; check out DeepL, it was recommended by our Japanese team | ||||||||||||||||||||||||
9 | Juan Aponte, Crowdstrike | What is the cadence programs are using for program health checks? Number of administrations, time length, etc. | InterSystems - We're somewhat low-volume, so we check every 100 exam results or every year (whichever comes first) Salesforce - We typically run over a specific time frame, and pull data for first administration only. CompTIA - uses both, a candidate-volume threshold and a time threshold, such as reviewing an exam after 100 days if it has not reached the volume target. ServiceNow groups exams into tiers by volume and uses weekly reporting to flag items and adjust banks/forms. Kryterion - cadence should align with how often the program updates exams; more frequent updates justify more frequent analysis. Candidate comments can surface issues that statistics may miss, though their usefulness depends on the candidate population. | ||||||||||||||||||||||||
10 | Adam Reid, CrowdStrike | When you are maintaining exams, do you do another standard setting round on any items that have been updated? What about items that haven't changed, do you use existing p-values or previous Angoff/standard setting values for balancing forms and determing whether the cut score needs to change? Or do you re-Angoff everything? | Zach recommends not re-Angoffing unchanged items. Instead, use equating: remove the edited item, replace it with a comparable item with known statistics, and equate the form. That avoids needing SMEs to review unchanged items. Salesforce uses p-values for items that have not changed. They run an Angoff review for any item with even a small change. They use recent p-values where enough data exists, then combine the values; has produced fairly equal forms. | ||||||||||||||||||||||||
11 | Jim Henderson, Validify Consulting | I'm curious if there are programs here today that considered performance-based exams but decided against them, and what factors led to that decision. | Sandy Foderick - We use performance exams.The challenges include cost of development and keeping them current Mark Hary - Key barriers are platform cost, scoring free-form responses, AI-grading trust and EU AI-law compliance, limited internal capacity, and psychometric expertise needed to build a valid performance-based exam. These needs often add significant cost. Performance-based testing is more justified when the skill cannot be measured directly through multiple choice. Some programs combine multiple-choice exams with separate hands-on assessments rather than putting performance tasks inside a timed, proctored exam. End-state scoring is often less fragile than scoring every process step, though process scoring matters when sequence or safety is essential. | ||||||||||||||||||||||||
12 | |||||||||||||||||||||||||||
13 | July 28, 2026 | ||||||||||||||||||||||||||
14 | Matthew Scicchitano, SAS | All certification information into a database for partners (like Yelp for partners) with a rating system. All partner information is self reported so it's difficult. We only know what they tell us. | InterSystems - Rely on partners to submit employee names, then manually checked whether those people were certified. They said that became unreliable because people change companies and those changes are easy to miss. Their newer process is an annual employee affiliation form: certified people opt in to tie their certification to a company, reconfirm each year, and get removed from the partner list if they do not confirm.Using Credly and get the information from Credly via API since it is public, then we can move it to our internal systems. If it is public, it can be shared. SAS - Also uses Credly directory. Google - A lot of our partners get free vouchers for certification - party of the voucher condition is that their certified status is shared with their employer. Linux Foundation - LF has offers company panoramas that can assign a manager to distribute vouchers, etc. People who gain access to their certification through their panorama access will have their results reported and viewable via the manager. It’s part of the conditions. Doesn’t help if the people change companies, were already certified, etc. At that point, it’s up to the candidate to share their certification number with their employer, partner, etc. The “person who cares” can then use our external validation tool to verify the certification. F5 - We don’t report on any individual results, but certifications achieved is public knowledge. It’s spelled out in our candidate agreement. Reporting is always tricky. We want to be helpful, and know how important our partner organizations are, but our loyalty is to our candidates. One suggestion is to work with legal to add a line to the candidate agreement that allows results to be disclosed to whomever the organiozation choses for their purposes. | ||||||||||||||||||||||||
15 | Matthew Scicchitano, SAS | Cost or revenue or breakeven? Do you know how much it costs (without sharing!)? | F5 - Cost, never positioned themselves as revenue generating. It's an investment in our candidates. Keep exam delivery fees low. Return on investment isn't on exam delivery fees, the value comes from stronger customer/account stickiness through the individual who earns the certification. Certification should be seen as an investment in candidates that can influence technology and purchase decisions. SAS - Historically has a large academic presense. EMEA exams have been free as long as they meet certain conditions. Another referenced their previous programs being treated as a sales enablement activity, so they tried to breakeven (cost receovery). It may be difficult to track in some cases because of discounting and other factors. InterSystem - Cost CrowdStrike - Cost, within training/revenue generating program and numbers are looked at carefully but it is understood they won't be generating significant revenue. Another referenced that across 3 enterprise companies, it was always a cost center. In some there was a partner channel that was a means to get to break-even. Suggestion: Something that I typically would do with cost recovery was both to help identify deals that a discount helped (so focusing on revenue enablement), but I also reported on overall training value (our training and certification programs fell under a single accounting center), and showing the non-discounted prices also helped with telling the story for how it helped with sales enablement. That was more training-focused, but if I was discounting $x of revenue to $0 to enable a sale, that was important to identify. The group agreed it is a worthwhile exercise to consider how much exam delivery costs. LF: Each exam we build has different goals, but I do think overall we operate as a revenue generator. LF is non-profit but our education program (training & certification) must be self sufficient (there are about 40 of us around the world from edu sales, development, through support) AND generate revenue to help open source projects fund their teams. | ||||||||||||||||||||||||
16 | |||||||||||||||||||||||||||
17 | ITCC HQ | What are the biggest unknowns facing the certification industry today? What keeps you from being confident about where the industry will be 5–10 years from now? | AI, how to use AI in testing (engine, scoring, etc), ethical AI use in the test itself, AI or no AI in the test taking experience How do people work differently? What tasks are they expected to do individually vs with AI and how to we ensure the exams will capture that evidence? How does AI change the actual nature of work? How will it vary between entry level and senior level roles? The value of entry-level skill certifications, given AI’s impact on those job markets What do the future roles look like and how do we map to them with current certifications and then adapt over time? How to "square the circle" of more high fidelity testing modalities (e.g., simulations, AI chatbot) with accuracy, validity, and comparability of results (i.e., standardized testing). How does AI affect how people learn and how does that affect certification? Reputation of certification - does the reputation of the certification hold the same value and weight as in the past? What does it actually mean and what does it entail? Don't want certification to get lost. Will AI Agents become just as valid/important of a customer as a human? Continue to clarify training vs certification exams Expectations from candidate perspective - is it the same across all programs? Has it changed within certain domains (i.e. security or storage)? Are we actually building what they want and need? And have we asked that? Unknown: How we evolve at the speed of “JIT” learning to address technology that changes faster than we can finalize a blueprint. It may be that we need to figure out how to measure someone's ability to adapt to change | ||||||||||||||||||||||||
18 | ITCC HQ | Topic and Presenter ideas for 2027 Education Sessions | |||||||||||||||||||||||||
19 | |||||||||||||||||||||||||||
20 | |||||||||||||||||||||||||||
21 | June 23, 2026 | ||||||||||||||||||||||||||
22 | Kate Sides, ServiceNow | Does anyone run loft (linear on the fly) testing? | Linear on the fly testing, a version of adaptive testing. Randomize questions that are psychometrically sound, not a fixed form. Delivered in real time. OFT typically takes a lot of items with lots of data (but less than computer adaptive testing). Usually the forms are "designed" at delivery vs CAT which changes mid delivery. LOFT is a dynamic form assembly method that uses item bank metadata to assemble forms according to a test blueprint from the entire item bank (basically what your psychometrics/test development team does). It usually has rules set to limit form difficulty variability, and will usually have equating measures built in to ensure cut scores remain consistent. On the security side, it spreads breaches to potentially your entire item bank instead of a single fixed form. Requires a large item bank to be "worth it" in terms of replacing fixed linear forms that are randomized. Microsoft stopped doing this. Another security consideration is how the exam is delivered and how content harvesting would actually happen. Can be secure in a controlled testing center setting. Not all technologies that support loft are equal. Some exam drivers are way easier to manage the LOFT forms that others. Depends on the engine delivering. Some have automated psychometric analysis. Some require analytics to be entered manually. | ||||||||||||||||||||||||
23 | Jim Lucari, HPE | Is anyone doing anything beyond generating items with AI? | AWS - Facial recognition for proxy test takers. Consider legal hurdles. Interesting to see repeat offenders (one had 50 accounts). Someone asked if people have considered allowing candidates to use AI during exams because they already have access to documentation during exams. InterSystems hopes to allow on their next exam and will report back. They are still using multiple-choice questions, so they need to rethink recall-style questions and do AI stress testing. It is key to give candidates access to AI in a controlled way. A skilled person would be able to see if the AI is right or wrong. We should test candidates on how they use the tool, not just what they recall. | ||||||||||||||||||||||||
24 | Adam Reid, CrowdStrike | What standard setting methods are people using? E.g., Angoff? Modified Angoff? Something else? Do you use the same SMEs or different SMEs when you do standard setting? | Testing standards (e.g., NCCA) generally recommends using a separate group of SMEs, specifically with representatives nearer the target of the exam. Real world often dictates taking who you can get to get as close to a representative group of judges as possible. Angoff is still king, some Bookmark with clients. Bookmark/Hofstee/Concurrent Validity Liberty Method - Mixes data-driven approach with a SME-driven borderline approach. As SMEs to give a 1-5 rating on knowledge. 1. Beginners, 2. Intermediate...etc 5. Expert. Take that from SMEs and look at average Rasch for that item. Also checks against expected beta passing rates. The average usually lands in the right place, with a little extra investigation when needed. When the program runs a beta test and can identify who is "qualified" vs "unqualified" without using the beta score, we've run Contrasting Group Studies alongside the beta test (same administration, 2 different purpose). If SMEs are available, we may add Mod. Angoff or IIT for additional defensibility. If you want to, a modified-modified Angoff (or Bookmark) can be helpful (e.g., have judges answer the items while they rate) and provide actual item difficulty statistics (similar to Liberty's method) as a "reality check" on judges assumptions. That allows you to point out items that the SMEs were getting items wrong themselves (i.e., consider revising your ratings down). Also highlights items with average ratings (in Angoff) that are in excess of the average difficulty (i.e., also consider revising your rating down). | ||||||||||||||||||||||||
25 | Adam Reid, CrowdStrike | Are folks using IRT or CTT for item-level analysis? Or something else (Rasch) For those of you using CTT, what thresholds do you use for item difficulty/p-value and item discrimination/point biserial? | CrowdStrike uses CTT, 0.2 threshold for item discrimation InterSystems CTT SAS CTT, 0.35 - 0.85/0.87, above 0.3 is good. Interaction matters as well. If an item is harder but still has strong discrimation, they may keep it. The stats are also useful for helping SMEs revise weak items, especially by looking at how each answer choice is performing. Stats model choice depends on program volume and how much item-level information you want, generally. CTT generally can be done at lower volumes, on par with Rausch, but CTT arguably gives you both difficulty and Discrim where Rausch is just difficulty. 2PL and 3PL IRT generally requires higher candidate volumes on items to be accurate. Generally thresholds are org centered on risk/fairness tolerance. Kryterion uses thresholds that are a little conservative, <0.30 reviewed for fairness concerns, 0.40-0.80 generally good item performance, >0.85 getting into the "too easy" range. One person's recommendation was to attend more to item discrimination to get at the "bang for your buck" an item is giving you (i.e., low-negative Discrim isn't really doing its job, regardless of difficulty). | ||||||||||||||||||||||||
26 | |||||||||||||||||||||||||||
27 | May 26, 2026 | ||||||||||||||||||||||||||
28 | Nermin - Intersystems | How do programs handle fake certifications on resumes? | -HPE using digital badging to have a verifiable record of those who have earned the certifications - Salesforce points people to a verifiable system or digital badge so others can confirm whether someone is actually certified. They also mentioned a Salesforce communication encouraging people to report suspicious cases through a hotline. Salesforce terms of applying for a job: Yes, applying for a job and submitting a resume at Salesforce requires accepting a formal Terms of Use and Privacy Policy. When you create an account and submit your materials through the official Salesforce Careers portal, you are prompted to agree to specific conditions. [1] These terms generally cover: Data Privacy & Processing: Consent for Salesforce to process, store, and share your personal data, resume, and contact details across their global recruitment teams. Accuracy of Information: A binding agreement that all the education, certifications, and work experience you provide on your resume or application form are true and accurate. Equal Employment Opportunity (EEO): Affirmation of Salesforce's commitment to equality and non-discrimination in the hiring process. [1, 2] - Another person suggested keeping a internal list to track patterns but noted the manual process and how it is time-intensive - Another referenced handling it through HR or job applications to attest their resume is true - Keep an internal list only it is used carefully and fairly - Also watch for unauthorized people or companies in the ecosystem selling questionable certification-related products or services. - Unauthorized training partners are also a consideration. - PMI has a certification registry that an employer can search if the candidate has a certification https://www.pmi.org/certifications/certification-resources/registry | ||||||||||||||||||||||||
29 | Kristin Wall - IBM | Has the rise of digital badges being used for a wide variety of training activities led to a decline in certification? | - Several people said certification declines seem more likely when partner programs require training badges instead of certifications. Where certification is still required, they were not seeing the same drop. - Pne person said they use training badges too, but because their partner program still requires certification, they have not seen a decrease - F5: Drew a clear line visually and structurally: certification badges are reserved in a specific format, while training badges use different shapes. The idea was to embrace badges without letting them carry the same perceived value as certification. - Ping - Badges made sense for emerging products that were too new or fast-changing to support a full certification yet. In that case, badges recognized early learning, but a clear badge taxonomy was important so people understood the difference. - People also said confusion increases when badges are displayed in places like LinkedIn in ways that make them look equivalent to certifications, which can dilute the perceived value of the certification track. - Newer, fast-moving areas like AI may be better suited to lighter badge-style assessments for now, because organizations are hesitant to invest in full certifications when the content changes quickly. | ||||||||||||||||||||||||
30 | Pat Hughes, ITS | Are others seeing AI make it easier for test sponsors to build custom software tools for their program without traditional software development teams? Where do people draw the line? | - ServiceNow: It depends on how its going to be used; we are working on process, governance and visibility for overlap of use case. I am using claude projects to control what the team is doing with the skills I create so they aren't loading them into the instance. - InterSystems: Has their own app library right now - Databricks: Have their own data platform to build internal applications so they built tools for item banking and item workflow management in-house. - Several people said AI makes building feel cheaper and faster, but long-term maintenance, support, and ownership can still make in-house tools more expensive over time. - There is a general concern about security and content exposure. - AI shift is happening but doesn't remove the need for vendors to help with scale, compliance, relability, and operational burden. | ||||||||||||||||||||||||
31 | Kate Sides, ServiceNow | What are programs doing for onsite testing and what unique challenges have been experienced? | - ServiceNow: Onsite testing at an event went well and now leadership wants to see more. Did hardline delivery. - HPE:Do onsite testing at major customer and partner events. Main challenge has been complexity from partnerships and acquisitions. Some need to be offered together even though delivery may be different. - Separate connection from conference connection for internet is vital. own network - Consider FAQ for proctors and team members so they know how to react - Consider a minimum threshold of exams to be delivered - protect your time and resources to ensure it can be managed effectively - Bag check area (remind people to turn off cell phones!) - Closing time and last seating time should be different - Gut readiness check from vendor, have a person or the team onsite to support - For performance based testing, all new technology is released around event time, so consider that as well. Might want to have free retake available. - Hosting the event and onsite exams could shift focus away from the event Kate to report back! | ||||||||||||||||||||||||
32 | |||||||||||||||||||||||||||
33 | April 28, 2026 | ||||||||||||||||||||||||||
34 | Dell | How many anchor items do you need? | Some overlap always recommended to show equated forms. 20% if equating. If pre-equating, anchor items are not strictly necessary but still useful. | ||||||||||||||||||||||||
35 | CrowdStrike | Do you construct your blueprint survey differently for product-based exams vs. job-role based? Do you ask for different inputs than Frequency/Criticality/Importance for one vs. the other? | For Fundamental/Entry level exams, one rating can be used. Liberty provided example from MSFT Fundamentals survey with five ratings from Must Know to Not Needed. | ||||||||||||||||||||||||
36 | Workday | Modularizing training content - Looking into breaking up long-form, multi-day class, video to consume at your own place, independent learning. | - Breaking it into Modules - updating modules may requre exam updates. - Need to define module granularity - It can feel disruptive to learners if they are able to jump around can affect learning experience - Importance of platform to support modularization such as diagnostic pre-test -> personalized learning path | ||||||||||||||||||||||||
37 | CrowdStrike | Curious how (or to what extent) folks are using AI as part of the item writing process? E.g., having SMEs review AI generated items? Encouraging SMEs to use AI tools as part of their item writing? Have you noticed differences in the quality of AI-generated or AI-assisted items compared to human-authored items? | - Microsoft: All of the exams that we created since November have been 100% AI-generated with SME review, edit, revise and approve. Have not noticed a differenece in psychometics from when it was writen by SMEs. Harder using AI in creating case study items. - Has substantially redued costs. - Alpine: designed item bank platform and is only used in-house by psychometricians. Seeing 40% time savings. | ||||||||||||||||||||||||
38 | Microsoft | How to create assessments for soft skills,/human skills, so I'm just curious, are other programs having those same kinds of conversations? | |||||||||||||||||||||||||
39 | CompTIA | How are we thinking about the growing use of AI/LLMs to generate study questions that may inadvertently recreate exam content, especially when those models are trained on or influenced by brain dump sites?” | Candidates claiming AI use may still be hiding actual cheating; De-indexing brain dump sites; Requesting AI chat histories from candidates | ||||||||||||||||||||||||
40 | |||||||||||||||||||||||||||
41 | March 24, 2026 | ||||||||||||||||||||||||||
42 | SPARC | How are you onboarding new hires? | Jim L - new hires are encouraged to engage with ITCC, complete internal technical training, earn company certifications, and pursue project management credentials such as Green Belt or Black Belt. Pat - emphasized foundational training in security, accessibility, and assessment basics, and recommended ITEMS Portal - NCME (https://ncme.org/resources/professional-learning/items/) Adding some accessibility courses we've had employees go through during onboarding. - Microsoft Accessibility Fundamentals (free): Accessibility fundamentals - Training | Microsoft Learn (https://learn.microsoft.com/en-us/training/paths/accessibility-fundamental/) - Deque University (Paid Accessibility Training): Deque University: Web Accessibility Training and Courses (https://dequeuniversity.com/) - ATP technology based assessment guildelines resourse Matt (SAS) - described onboarding as largely informal and hands-on, noting that infrequent hiring makes structured onboarding difficult. Kristin (IBM) - described a shadowing approach where new staff observe multiple exam workshops before leading projects themselves. Heidi (F5) - one effective approach was connecting new hires directly with certified SMEs to understand candidate perspectives and why certification matters. Megan (CrowdStrike)- shared that pairing new hires together on projects helped both onboarding and project efficiency. Adam - highlighted shadowing plus internal process documentation as key for helping new staff understand company-specific development practices. Nermin- added that new hires complete foundational reading in assessment standards, ATP learning paths, and NCME resources, then progress through structured shadowing. Zach - Kryterion has a library of FAQs on test development and psychometric topics that is accessible to our Customer Success, Business Development, etc. teams and can be provided to clients as needed (1-2 pages each, high level overviews) Adam - having process documentation, maintain standard operating procedures are helpful to show the company's apporach for doing JTAs - docs that show the standard settings and methods we use. | ||||||||||||||||||||||||
43 | Are you doing anything interesting to learn more as you grow your career in certification (how are you staying up to date)? | ||||||||||||||||||||||||||
44 | Any upcoming events/webinars/conferences you or your team are attending? | Josephine - We had 3 non-certification program managers (curriculum development, university program) attend the last ATP conference. They were enthusiastic about understanding the link between training and certification, the role of AI, and the products offered by vendors. Keith (Alpine) - I try and send key members of my software engineering team to ATP. They've found it extremely helpful in understanding the products they are working on and how to better service our clients. | |||||||||||||||||||||||||
45 | CrowdStrike | Do you limit the number of certification projects you allow your SMEs to work on, either for a calendar year or for a specific certification? If so, how do you implement that? | Kristin (IBM) - They limit SME participation in certain workshop. Example, Item writing can include many SMEs, but Tech Review is capped because too many participants reduce engagement. SMEs are rotated across workshops to maintain continuity while avoiding overuse. Josephine (Adobe) - We rely on volunteer SMEs and we are limited by the amount of rewards we can give them for a year. We have a set of SMEs who are really active or engaged with the certification program. They want to see what the updates are, and so they want to join either item writing or tech review. Volunteer SMEs are generally limited to about three projects per year. IBM - added that SME selection is handled by exam and workshop, not by prior involvement. Previous participation does not guarantee future selection, allowing new contributors to enter while matching people to their strengths. | ||||||||||||||||||||||||
46 | CrowdStrike | Do you award administrative grants of the credential to SMEs who participate on certification development projects? If so, do you have minimum participation requirements for SMEs to be eligible to receive the administrative grant? | Matt (SAS) SMEs are granted certification if they participate substantially in technical review—typically reviewing at least 80% of exam items. Nermin (InterSystems) Credentials are granted when SMEs review a sufficient subset of questions, with eligibility tied to hours contributed (about 6–8 hours) or number of questions reviewed. Kristin (IBM) SMEs receive certification after participating meaningfully in JTA, item writing, or technical review, plus an exam developer badge. Standard setting alone is not enough. Caitriona (Linux Foundation) SMEs do not receive certification but instead receive an exam developer badge, which often carries strong professional value. Josephine (Adobe) SMEs working on new certifications receive both the certification and an Adobe Subject Matter Expert badge, recognizing significant volunteer effort. Zach (Kryterion) I recommend having SMEs rotate through projects, or portions of different projects (e.g., item writing, item review, standard setting, job analyses). Item writing is a good activity to get a lot of SMEs involved (previous and new), that can also serve to identify/recruit SMEs for other more selective activities (JTAs, standard setting). If you ever want to consider accreditation, those standards also watch for individuals having "undue influence" on validation activities, so recruitment and rotation of SMEs becomes even more critical. David (Dell) we give vouchers for JTAs, administrative certification to item writing workshop participants provided they fully participate in the whole workshop (item writing, reviewing, etc.). Jim H - My experience is similar - if they see a majority of the items, then they'll get a grant. If they sign up for item writing and show up for 25% of it, then that's not generally enough item exposure. It can be a little harder to identify if the IDW is asynchronous how many of the items they've seen (depending on the tool). Josephine (Adobe) for exam updates, we accept only certified participants and their certifications get renewed for participating. Cairtriona (Linux Foundation) shared the Linux Exam Developer Badge - Earners of this designation are subject matter experts selected by the Linux Foundation and Cloud Native Computing Foundation, who contributed to creating the KCNA exam. https://www.credly.com/org/the-linux-foundation/badge/exam-developer-kubernetes-and-cloud-native-associate | ||||||||||||||||||||||||
47 | Kristin, IBM | Has anyone experienced issues using “engineer” in certification titles, given that the term is protected in some regions? | Jim L (HPE) - past certifications at Hewlett Packard Enterprise had to remove “engineer” from titles because it is protected in some states, including possibly California. Jim H (Validify) - similar past issues at Novell, where titles were adjusted because some jurisdictions restricted use of “engineer.” Adam - Palo Alto Networks has used “engineer” in certification titles for years without receiving pushback. | ||||||||||||||||||||||||
48 | Jim Henderson | Call for Volunteers, join website workgroup, subgroup, reviewing document accessibility for ITCC resources, focused on improving current and future document accessibility standards. | If you or someone in your team has experience in accessiblity, please reach out to Jim H. in Slack or email, jhenderson@validify.consulting | ||||||||||||||||||||||||
49 | February 24, 2026 | ||||||||||||||||||||||||||
50 | ITCC Announcements | Looking forward to this weekend's in-person Member Dinner and Member Meeting Shared a peak to a couple of new resources to be announced this weekend: - Recertification Model White Paper created by SPARC - Resource document created by the Strategic plan work group - Future of IT Skill Assessment and Certification | New resources available to members are now posted in the Resource Library | ||||||||||||||||||||||||
51 | Adam Reid, CrowdStrike | How much score detail programs should provide to candidates—such as pass/fail, section-level feedback, or scaled scores—and what risks come with sharing more detailed results. | F5- section-level feedback was previously given only to unsuccessful candidates on longer exams, but shorter one-section exams now receive no section feedback. Practice exams still provide broad section feedback (below, borderline, meets) to support preparation. - noted that item-level feedback is avoided to prevent candidates from memorizing questions for future attempts. Caveon- score reporting should depend on exam stakes and security: high-stakes exams with limited item pools should provide minimal feedback, while practice exams or programs with larger item banks can provide more domain-level detail. | ||||||||||||||||||||||||
52 | |||||||||||||||||||||||||||
53 | January 27, 2026 | Qualtri | |||||||||||||||||||||||||
54 | David Hays, Dell Technologies | Is there an industry standard for security holds on test takers suspected of cheating? Currently Dell does a 90-day suspension on the first offence and a lifetime suspension on a second. We'd like to create a policy that is closer to other tech companies. | The Linux Foundation: When we determine there has been exam misconduct, we cancel any unused exam eligibility records, we revoke/invalidate any active Certification records and Digital badges and we ban the Test Taker from taking any of our exams, for 2yrs HPE: Do have policies in place. Janet will have more specifics. They suspend or remove them for a minimum of 1 year depending on the extent of the offense. Microsoft: Proxy testing - ban for life. For other lessor offenses: Six months. Liberty will share the policy. SAS: Has a written policy to share that is only internal, not published. InterSystems: If there is something where you don't have 100% proof, InterSystem uses a 3 month ban. Different considerations for partners. Engage legal department as needed. CrowdStrike: Recent situation with a candidate who was flagged for proxy testing but the candidate wasn't aware. The question was if someone else used the device. Made the candidate go to a test center. Different considerations are needed for bans/consequences when you don't have actual proof. We don't ban from everything, just online testing based on the situation. F5: Added specific langauge around glasses because some people may be wearing smart glasses. CompTIA: Added language around glasses for exams and workshops. Not a single answer across the industry but have a written policy. Also ensure you understand leadership appetite for how aggressive or lenient they want you to be. | ||||||||||||||||||||||||
55 | David Hays, Dell Technologies | Has anyone done an analysis of the duration of your exams, or are exam times arbitrarily set (or based on the number of items)? Is there a standard time for different item types? | Two minutes per standard multiple choice question. Extra 30 seconds per question for non-native English speakers. Post on LinkedIn from Susan Farago: https://www.linkedin.com/pulse/how-time-exam-determined-susan-farago-m-ed-/ Another had one minute per each multiple choice/multiple select questions. Expert level exams that had more complex questions required a different approach. If you let the test taker know there will be a range of questions, it helps. Make sure a psychometrician is reviewing the exams. Target a number of questions to tell the candidates and then the time may vary. We used to change exam times based on the data but it's really hard to manage candidate expectations so we landed on a stand lard exam time, and i use the data to inform number of questions, which varies from exam to exam. Two minutes has been pretty standard, UNLESS the cognitive load was higher. | ||||||||||||||||||||||||
56 | Kate Sides, ServiceNow | For accessibility, how you handling pdf remediation? Are there tools to remediate? | Jim Henderson: I've found this tool to be good for checking PDFs for accessibility compliance - it's a free tool and is frequently updated. https://pac.pdf-accessibility.org/en That at least gives you a start on identifying issues with the PDF itself. Remediation isn't something I've dug into, though - maybe some of the folks at Adobe would have some ideas (I'm definitely interested!) AmyJune The Linux Foundation: If you already have access to Adobe, I know it’s closed source, but their accessibility tools are approachable and straightforward. The built-in tooltips are helpful when doing remediation work. Section508.gov also has a training video series that walks through accessible PDF creation and review: https://www.section508.gov/create/pdfs/training-videos/ One question I often encourage teams to ask is whether a PDF is actually the right format for the content. PDFs can be difficult for mobile users to view and interact with, and copy-and-paste is often frustrating (everywhere). But I know there are definitely cases where PDFs are the right choice, especially for documents that need a fixed layout or offline access. | ||||||||||||||||||||||||
57 | Megan Gordon, CrowdStrike | Looking for a check on current practices around exam vouchers and retakes. CrowdStrike exam vouchers include a free retake; essentially two attempts for one price. What is your approach? | Periodic for most, varies. Second attempt, some consider "insurance" for a full retake. Most are 1st retake with a discounted 2nd. | ||||||||||||||||||||||||
58 | Heidi Ford, Workday | Does anyone track their certification candidates by customer? | Listed in partner portal, for partners only. Different to track customer employees as they move from company to company. There are ways to look at LinkedIn to see how many people over a certain time are adding a ___ certification vs ours? | ||||||||||||||||||||||||
59 | ITCC Announcements | 2.2 output Spring Member Dinner February 28 and Meeting March 1 | Register for the Spring Member Meeting: https://itcertcouncil.org/members/springmeeting/ | ||||||||||||||||||||||||
60 | |||||||||||||||||||||||||||
61 | |||||||||||||||||||||||||||
62 | |||||||||||||||||||||||||||
63 | |||||||||||||||||||||||||||
64 | |||||||||||||||||||||||||||
65 | |||||||||||||||||||||||||||
66 | |||||||||||||||||||||||||||
67 | |||||||||||||||||||||||||||
68 | |||||||||||||||||||||||||||
69 | |||||||||||||||||||||||||||
70 | |||||||||||||||||||||||||||
71 | |||||||||||||||||||||||||||
72 | |||||||||||||||||||||||||||
73 | |||||||||||||||||||||||||||
74 | |||||||||||||||||||||||||||
75 | |||||||||||||||||||||||||||
76 | |||||||||||||||||||||||||||
77 | |||||||||||||||||||||||||||
78 | |||||||||||||||||||||||||||
79 | |||||||||||||||||||||||||||
80 | |||||||||||||||||||||||||||
81 | |||||||||||||||||||||||||||
82 | |||||||||||||||||||||||||||
83 | |||||||||||||||||||||||||||
84 | |||||||||||||||||||||||||||
85 | |||||||||||||||||||||||||||
86 | |||||||||||||||||||||||||||
87 | |||||||||||||||||||||||||||
88 | |||||||||||||||||||||||||||
89 | |||||||||||||||||||||||||||
90 | |||||||||||||||||||||||||||
91 | |||||||||||||||||||||||||||
92 | |||||||||||||||||||||||||||
93 | |||||||||||||||||||||||||||
94 | |||||||||||||||||||||||||||
95 | |||||||||||||||||||||||||||
96 | |||||||||||||||||||||||||||
97 | |||||||||||||||||||||||||||
98 | |||||||||||||||||||||||||||
99 | |||||||||||||||||||||||||||
100 |