Skip to content

Instructor Accessibility

Principle · All principles
Evidence · 24 claims (9 for, 15 mixed) · 28 studies (6 causal, 5 review, 5 qualitative, 4 quant-synthesis, 4 design, 3 associational, 1 theoretical), q1–q4 · 4 of 28 report an effect size · 19 claims rest on one study

Conditional relationship

This page owns the route to the instructor: whether a learner who is stuck between scheduled contacts knows how to reach the instructor (or the course's teaching team), can do so at low cost and without social risk, and gets a reply soon enough and specific enough to act on while the problem is still live. For a learner of any expertise who meets difficulty on course work outside the moments when the instructor is watching (most often an adult or postsecondary learner in an online-instructor-led or blended course over days-weeks or a term-plus, but also a pupil in a classroom whose difficulty goes unnoticed), published channels, a stated and kept response window, an invitation that makes asking normal, and replies that respond to the learner's actual attempt are expected to raise the share of learners who get help before they disengage, and through that to support persistence and performance on the task they were stuck on (immediate-performance, general-achievement). Access helps only through what happens when it is used: a reachable instructor who replies late, vaguely or with the whole answer has no route to an effect in this model.

No claim in the wiki tests instructor accessibility as such. Nothing here compares learners given more reachable instructors (more channels, faster replies, proactive outreach) with learners given less, on learning, persistence or help-seeking. What the wiki holds is evidence on neighbouring links of the chain, each carried here to a setting it did not test: contingent one-to-one support beats fixed or no support in tutoring studies; human tutoring beats no tutoring by a medium amount; in a simulation of a computing course's office-hours queue, the order in which waiting students are served changes how many students get help at all, but not how long they wait; drop-in tutoring is used by a self-selected minority, and its users' better results are associations; warm teacher–student relationships go with engagement in correlational syntheses; and a brief belonging intervention works only where the setting gives students real opportunities to belong. So the default design below is a labelled proposal, and its evidence status is stated step by step.

How the converted siblings sit beside this: Ask Experts owns what happens inside a consultation once a learner reaches someone with more expertise (have them explain first, least help that works, apply the advice soon); this page owns getting the learner there and keeping the route open. Immediate Feedback owns when a correction should arrive relative to an attempt; this page uses its finding that timing is a weak lever for retention and does not repeat it. Check-ins owns the instructor-initiated prompt for a self-report; this page owns the learner-initiated request and the instructor's outreach when requests stop. Community of Inquiry owns teaching presence inside a discussion community. Scaffolding and Fading owns how much help to give and how to withdraw it.

Default design, while the relationship is untested

The earlier page's guidance (visible channels, predictable responsiveness, a supportive tone, help aligned to the work, sustainable load, more than one channel, proactive outreach), kept as a concrete default a designer can act on and revise. Each step is a proposal; its evidence status is stated beside it.

  1. Publish the routes in the first session, on one page learners can find again. For each channel say what it is for, when it is answered and how fast: a public question forum for task questions, a private message for personal matters, a bookable or drop-in live slot, and what to do if the matter is urgent. A default for a course that meets weekly: replies within one working day on weekdays, one 45–60-minute live help slot a week at a time chosen to suit the learners (for working adults, an evening or lunchtime), and an answer at the next session for anything raised in class. The quantities are untested proposals. A programme-wide course template with pages for due dates and grading turnaround times was credited with high ratings for course organisation in one accelerated online MBA study (A program-wide standard online course template heightens teaching presence and supports learner intentions in accelerated online courses [+W]; survey and interview, no comparison).
  2. Have every learner use a route once before they need it. In the first two weeks, ask for one short contact (a 5-minute booked conversation, or a reply to a personal welcome message asking what they want from the course and what might get in their way). The first use of a channel is the costly one; this moves it to a moment with nothing at stake. Untested proposal. E-teachers in one interview study reported that online community and interaction needed deliberate, planned effort that face-to-face co-presence supplies naturally (Creating community and interacting with students online required deliberate, planned effort unlike spontaneous face-to-face co-presence [+W]; high-school web-based teachers, carried to other online courses).
  3. Say what to bring. Ask learners to bring an attempt: what they tried, where it stopped, and what they expected. Reply to that, not to the topic in general. Answer task questions in the public channel (with the asker's permission, or anonymised) so the reply serves the others with the same question, and keep personal matters private. Untested proposal; how to run the exchange itself is on Ask Experts.
  4. Reply with the least help that moves the learner on. A question back or a hint first, then a worked step, a full explanation last; check with one follow-up item that the learner can now go on. Support adjusted to each response beat fixed or no support in one-to-one tutoring of children on long division (Contingent scaffolding improves learning more than fixed or absent support. [+M]; one-to-one only, classroom scale not shown), and in one small study pupils learned as well when tutors only prompted as when they also explained and gave feedback (Human tutoring has a medium effect over no tutoring, and in one small study students learned as well when tutors only prompted them as when tutors also explained and gave feedback [+M]).
  5. When more learners are waiting than can be seen, serve those not yet helped first. In live help sessions with a queue, take learners who have not yet had help in this session before returning to anyone already helped, and cap each conversation (a default of 10 minutes, then rejoin the queue; untested). In a simulation built on one computing course's office-hours data, that ordering raised the share of students who got any help under busy and normal load (Under busy or normal queue load, the New Student First (NSF) strategy significantly increases the percentage of students who receive help compared with FCFS, LWF, and VLWF [+M]), but no ordering changed the share of requests resolved or the average wait (Choice of scheduling strategy does not significantly affect the number of resolved requests or students' overall wait time [~M]), so long waits need more capacity, not a better queue. The recipe is on NSF for online office-hours scheduling.
  6. Reach out when a learner goes quiet. If a learner misses two deadlines in a row or posts and submits nothing for a week, send one short personal message naming a specific route and a time ("I have a slot on Thursday at 6; reply with a time that works"), not a general "let me know if you need anything". If there is no reply, try a second channel once. Untested proposal. Those who use optional help unprompted are a minority and are not the ones most at risk: in one physics class about a tenth of students attended optional drop-in tutorials, mostly from the middle mark bands (Drop-in one-to-one tutorial attendees improved P1X-to-P1Y marks while the class average declined [~W]).
  7. Move the source of first help across the course. A default arc, untested: in the first third of the course, the instructor answers within the window and is visible weekly (an announcement, a short video reply to common questions, the live slot); in the middle third, learners post an attempt first and peers have a set time (24 hours, say) to answer before the instructor confirms or corrects; in the last third, learners check their work against the stated criteria before asking, and live time goes to applying the work in new cases and to the final task. A learner moves to the later arrangement when their questions shift from "how do I start" to checking a specific step, or when they resolve the next comparable task unaided after a reply. How to withdraw support is modelled on Scaffolding and Fading; comparing one's work with a target is the self-monitoring the earlier page aimed at (Self-regulated learning models treat self-monitoring as a core process through which learners notice gaps and adjust strategy [~W]; a review and a theoretical synthesis, no comparison).
  8. Budget the instructor's time and keep the promise you can keep. Set the response window from the hours available, not from what sounds generous: a default of about two hours a week per 30 learners for replies and the live slot, answered in two or three fixed batches a day rather than continuously, with a running list of frequent questions turned into a page. A slower window that is always met is preferred over a fast one that is met sometimes, because the earlier page's warning about stated accessibility that is not delivered is about trust. Untested proposal. For retention, the timing of a correction is a weak lever: after multiple-choice tests, immediate and delayed feedback were reported equally effective at reducing later errors (Providing feedback after initial multiple-choice tests cut lure intrusions on final short-answer tests roughly in half, with immediate and delayed feedback equally effective [~W]; second-hand, fact recall).
  9. Keep a short log. For each request: who, when, which channel, what about, time to reply, and what the learner did next. Review it every two or three weeks for who never asks, which topics recur and whether the window is being kept, and change the design from that. Untested proposal.

When a condition is not met: if the instructor cannot keep any reply window (a very large course, a volunteer tutor), say so, route task questions to a staffed or peer forum with a worked-answer bank, and keep the instructor's time for the live slot and outreach; if learners have no reliable private channel (shared devices, no email), use in-session time for brief private check-ins instead.

Fitting the design to a situation

The three facts that most change the decision:

  • When and where learners get stuck, and how they can reach anyone then. A learner stuck on a weekday evening in an asynchronous course needs a different route from a pupil stuck in a lesson with the teacher in the room. If unstated, ask: when do learners do the work, on what device, and who could they ask at that moment?
  • How many learners per instructor hour, and how demand peaks. Promises that work for 15 learners fail for 300, and demand piles up before deadlines. If unstated, ask: how many learners, how many staff hours a week for questions, and when are the deadlines?
  • Who is least likely to ask. Learners who most need help are often the ones who do not use optional routes (low confidence, first in family at university, working full-time, different language or culture of schooling). If unstated, ask: who has dropped out or gone quiet in past runs, and why?
If the brief says… Then change… Basis
online-instructor-led, asynchronous, adult learners working or caring Make the forum and private message the main routes with a published reply window of one working day; put live slots outside working hours and record them; accept questions as text or a short voice note Creating community and interacting with students online required deliberate, planned effort unlike spontaneous face-to-face co-presence [+W] (interviews with web-based high-school teachers; carried to adult courses); the channel choices are untested
Mixed mode: some learners in the room, some joining remotely Run one help queue for both groups (a shared ticket list or chat), alternate between room and remote in it, repeat each in-room question aloud and post it with the answer, and give remote learners a turn first after each in-room one untested proposal; the queue-ordering finding (NSF [~M]) is from a simulated online office-hours queue, not a mixed room
Learners work on phones only, or have poor bandwidth Use a text channel that works on a phone (messaging, forum app); never require video for help; send replies as short text with one image at most untested proposal
One instructor for many learners (a large lecture course or an open online course) Route task questions to a public forum with peer answers first and staff confirming; turn repeated questions into a frequently-asked page weekly; in live help, serve learners not yet helped first and cap each conversation; add staff or sessions when waits grow, since ordering does not shorten them NSF [+M] and no strategy changes wait or resolved requests [~M] (simulation from one CS2 course's office hours; outcome is who gets help, not learning)
Deadline or exam surges Add live slots in the two or three days before a deadline, publish a cut-off time for questions, and post a common-errors note the day before; outside peaks, one slot may be enough The NSF advantage over FCFS appears only above an arrival-rate threshold [~M] (simulation: under a relaxed queue no ordering made a difference); the scheduling of extra slots is untested
A single session only (a workshop, a one-off training) State before the end how learners can reach you afterwards, for how long and for what (a two-week reply window by email, say), and collect one question from each learner before they leave untested proposal
Learners are child (primary age) Accessibility means routines in the room: the teacher circulates on a set route, learners have a help signal (a card or cup) so they need not call out, and the teacher holds brief informal talks with each pupil over a fortnight Positive teacher–student relationships are associated with greater student engagement and, more weakly, with achievement [~M] (correlational meta-analyses, preschool to high school; negative relationships mattered more in primary school); the routines are untested
Learners with low confidence or literacy, first in family, or with bad past experiences of school Invite each one personally to a first short contact; offer a private route and say that asking is expected, not remedial; follow up by name after the first quiet week; do not read high use of help as proof that the help works Belonging Interventions Improve Outcomes [~M] (college entrants; worked only where the setting afforded opportunities to belong); Minority students visited the tutoring center more often, yet tutored minority and first-generation students had lower average GPAs [~W] (one college, associational)
Learners study in a second language, or the instructor and learners do not share a language Offer a route in the learners' stronger language (a bilingual assistant, written questions that can be translated, a separate session) rather than only interpreting into the main channel In one participatory budgeting case, Latino immigrant participation lapsed under outreach with interpretation and was sustained only after the activity structure was redesigned [~W] (one case of Spanish-speaking volunteers in civic budgeting committees; carried from civic participation to courses)
Adults in work settings (workplace training, clinical placement, supervision) Name the person to ask on each shift or project and a back-up; agree an escalation route for urgent questions and a slower one for learning questions; schedule short regular meetings rather than relying on ad hoc asking Tutors and trainee researchers most value supportive attitude, mastery of the research field, and accessibility in a research project supervisor [+W] (online doctoral supervision; what people value, not an effect); the arrangement is untested
Learners are novice on the course's tasks Schedule contact (a booked slot or a required check of the first attempt) rather than waiting to be asked, and answer with prompts before explanations Contingent scaffolding improves learning more than fixed or absent support. [+M] (one-to-one tutoring; not a test of access)
Learners are advanced (experienced professionals, final-year students) Make contact on request and use it for critique of their work and for decisions they are unsure of, not explanation; keep the reply window but drop the required first contact untested proposal (the expertise reasoning is on Ask Experts)
The course is pre-recorded video and the budget goes on production Spend effort on replies and live contact rather than on putting the instructor's face in every video Human embodiment in video: perceived social presence benefits learning ratings, but instructor-face inclusion shows no significant learning-performance difference [~W] (second-hand review; learners rated learning higher, performance did not differ)
High stakes for the learner (a licensing exam, a grade that decides progression) Publish what kinds of help are allowed before assessed work (clarifying the task, not checking answers), give every learner the same access to that help, and log it untested proposal

Three of the fourteen rows rest on a claim that tests a link in the chain (contingent support, queue ordering), and none tests accessibility itself; the queue rows come from one simulation whose outcome is who gets help, not learning. Seven other rows cite a claim for a limit or a neighbouring finding (online teachers' reports, the queue simulation carried to a mixed room, relationships, belonging and tutoring-centre use, a civic case, supervision preferences, instructor video) and say how far it is carried. The remaining four are untested proposals.

Observation, state and explanation

What can be observed is the use of a route: who asked, when (relative to the work and the deadline), through which channel, what they asked, how long the reply took, what the reply did, and what the learner did next. "Feels the instructor is approachable", "is avoiding help" and "is dependent on the instructor" are inferred states. A learner who never asks may need nothing, or may need a great deal and not ask.

Response Plausible explanations Observation that could change the interpretation
A learner never asks and their work is weak or missing Does not know the routes; asking feels risky or remedial; time is the problem, not understanding; has already disengaged Send one personal message offering a specific slot; ask privately what would help this week; check whether the routes were ever used, and whether work stops at the same point each week.
Questions arrive only in the hours before each deadline Work is done late because of outside commitments; the route is used only when the stakes force it; earlier questions went unanswered or were answered slowly Look at when learners open the task against when they ask; add a mid-week check of a first attempt and see whether questions move earlier.
The same learner asks for the answer to each new item Help is cheaper than effort; low confidence in their own judgement; a real gap below the current task Reply with a question back and see whether they can go on; give a diagnostic item on the prerequisite; ask them to rate confidence before you check.
Many learners ask the same question The task or materials are unclear; a common misconception; one learner's question has spread Post one public answer and see whether the question stops; give a short item where the misconception and the correct idea predict different answers.
Learners ask each other, not the instructor Peers are quicker and less risky; the instructor seems busy or distant; peer help is working Read the peer exchanges for accuracy; ask a few learners why they did not ask you; check whether peer-helped learners do as well on the next task.
Replies are read but the next attempt shows the same error The reply was too general or too complete; it arrived after the learner had moved on; the learner could not apply it Ask the learner to say what the reply meant; reply with a hint about the step and check one follow-up item.

These rows are proposals, untested with learners. Reaching out is itself an intervention: it changes whether the learner asks next time, so record it, and keep the explanations that remain open rather than choosing the one that fits.

Evidence and qualifications

No claim in the wiki compares more and less accessible instructors, or tests a response window, a set of channels or proactive outreach. The claims below test links of the chain, or limit it.

Learners, settings and outcomes differ across these (children in tutoring, a simulated computing queue, physics undergraduates, school-age correlational samples, college entrants), so do not rank them. None measures learning as a function of how reachable the instructor is.

The evidence by question

Further evidence, not yet read against this model

Claims that bear on instructor accessibility but are not part of the model above, or appear only in a row of the situation table. Each says what it is and how far it bears on this page.

Objective and learner-valued goal

The designer's objective for accessibility is usually that difficulties are resolved early enough that learners keep going and complete the work to the criterion. Elicit separately what the learner wants from contact with the instructor: a quick answer to get unstuck, reassurance that they are on track, flexibility over a deadline, to be known as a person, or to be left alone to work. Agreement and divergence are both observations. A learner who values not being seen to struggle will not use a public forum however fast it is answered; one who values flexibility may need a deadline conversation, not a content reply.

For example, in an online course for working nurses, the designer's objective is that every learner submits a care plan meeting the rubric by week six. A learner on night shifts may value being able to ask at 3 a.m. and find an answer when she wakes, and may never come to a live slot. A route that serves her (a forum answered within the window, a bank of answered questions, a booked call at the start of her day) serves the designer's objective too; a live office hour at 5 p.m. serves neither. Record what each learner said they wanted, so that low attendance at the live slot is not read as low need.

What would revise this model?

The expectation should weaken if comparisons of courses that differ in reachability (a kept reply window, several channels, outreach to quiet learners) against comparable courses without them find no difference in persistence or performance, or a difference only among learners who were already doing well. If outreach to quiet learners raises contact but not completion, the outreach step is a reassurance measure only. If faster replies do no better than slower replies that are reliably kept, the response window should be set by capacity alone. If peer-first answering in the middle of a course lowers the accuracy of the help learners receive, the arc should keep the instructor first for longer. Do not protect the model by calling every failure "not accessible enough" after the fact.

Use of help, satisfaction with the instructor, help received before a deadline, persistence and achievement are separate outcomes; evidence for one does not establish the others. The present evidence establishes no response window, no number of channels or live hours, and no effect of outreach.

Claims

  • Immediate Feedback — accessibility often determines whether timely feedback is actually available.
  • Check-ins — check-ins are one recurring way to make support visible and reachable.
  • Inquiry-based Learning — learners need accessible guidance when working through open-ended tasks.
  • Cultural & Life Experiences Connections — accessibility improves when instructors take learner context seriously.
  • Ask Experts — what to do inside a consultation once a learner reaches the instructor; this page covers getting there.
  • Community of Inquiry — teaching presence in a discussion community, of which responsiveness is one part.
  • Scaffolding and Fading — how much help a reply gives, and how help is withdrawn across a course.
  • Social Presence — learners' sense of others, including the instructor, as real people in an online course.

Examples

  • Clear office-hours model: Learners know when help is available, how to prepare, and what kinds of support to expect.
  • Response-time commitments: A course specifies turnaround times for questions and draft feedback.
  • Multiple help channels: Learners can use discussion boards, messages, short live sessions, or annotated feedback depending on need.
  • Proactive outreach: The instructor checks in when participation drops or confusion patterns appear.
  • Leveraging instructor office hours — a recipe for regular, structured weekly office hours.
  • NSF for online office-hours scheduling — serving students not yet helped first when the queue is busy.
  • MyDigitalHand office-hours ticketing system — a ticket queue for help requests.
  • Maintain presence for students — predictable availability for brief informal talks.
  • Help seeking — the learner's side: recognising a need, forming a request, getting help.

Key Sources

  • Ekmekci, O. (2013). Being there: Establishing instructor presence in an online learning environment. Higher Education Studies, 3(1), 29-38. https://doi.org/10.5539/hes.v3n1p29
  • Lowenthal, P., Dunlap, J., & Snelson, C. (2017). Live synchronous web meetings in asynchronous online courses: Reconceptualizing virtual office hours. Online Learning, 21(4). https://doi.org/10.24059/olj.v21i4.1285
  • Park, C., & Kim, D.-G. (2020). Perception of instructor presence and its effects on learning experience in online classes. Journal of Information Technology Education: Research, 19, 475-488. https://doi.org/10.28945/4611
Page metadata
Field Value
type principle
id instructor-accessibility
title Instructor Accessibility
description For a learner who gets stuck on course work between contacts, published channels, a kept reply window, an invitation that makes asking normal and replies that respond to the learner's attempt are expected to raise the share who get help before they disengage, though no claim here tests instructor accessibility as such.
status review
canonical True
generated by claude/unspecified on 2026-10-05
trust unverified — no one has confirmed this page's content
sources 3 — ekmekci-2013, lowenthal-2017, park-2020

Frontmatter is the page's machine-readable record — see the Schema & Guide.