A hotspot question shows a picture and asks for a place on it: click the nucleus, click the mitral valve, click Kenya. There are no options to read and no word to type. The student knows where the thing is or does not, which a labelled diagram with four names beside it cannot tell you.
EdProcto builds the shape as its hotspot template: one image, one region that you draw, one click, marked by rule. This guide covers what an image-click question measures, how to write one, how large to draw the accepted region, and who the shape leaves out.
The short version: an unlabelled image that reads on a phone, a stem that names the thing and not its position, one target, and a region big enough for a finger yet only a few per cent of the picture.
What does a hotspot question measure that multiple choice cannot?
A hotspot question measures whether a student can locate a thing with no cue: no list of names, no letters on the diagram, no option to recognise. A multiple-choice question on a labelled diagram ("Which structure is marked C?") runs the other way. It supplies the place and asks for the name, and the names on the screen narrow the search before any knowledge is used.
There is evidence that the answer format changes what is tested, most of it from image-based assessment in anatomy. Vorstenbosch et al. (2013) gave 210 students 39 anatomy questions answered either from a labelled image or from a written list. With two cross-sectional images the questions were harder and separated students better than with the list, and the authors suggest an extra ability was being tested. With a schematic of the fetal circulation they were easier and separated students less well, which the authors read as the image cueing the answer. It was one course with volunteers, so treat it as a pointer: the picture is part of the question.
A hotspot also spares you the hardest job in multiple choice. Tarrant et al. (2009) examined 1,542 wrong options on teacher-written nursing papers and counted only 52.2% as functioning. On an image nobody writes the wrong options: every other structure in the picture is one.
What a hotspot does not measure
A click shows where, and nothing more: not the name, its spelling, or what the structure does. Add a short-answer question when the name matters, and use multiple choice to tell two ideas apart. Asking students to label a diagram in full is a different shape again.
What does the research say about image-click questions?
The research is thin, and it says three things: questions answered on the figure can be harder than their multiple-choice twins, they measure much the same knowledge, and they do not clearly yield more information per minute. The measurement literature calls the family figural response: Martinez (1991) described questions in which a figure, such as an illustration or a graph, is the medium the answer is made in. A single click is the simplest kind. The published summaries do not say how many of the questions studied were single clicks, so read them as evidence about the family.
| Study | Compared | Found |
|---|---|---|
| Martinez (1991) | Parallel figural and multiple-choice science questions, grades 4, 8 and 12 | Figural questions generally harder, slightly more discriminating and reliable, left blank more often |
| Jodoin (2003) | Innovative formats and multiple choice in one computer-based testing programme | Innovative formats gave more information at every ability level; multiple choice gave more per unit of time |
| Wan and Henly (2012) | Figural response and multiple choice in a computerised school science test | Figural questions matched multiple choice for information and efficiency, and measured similar constructs |
The disagreement is about time: Jodoin found multiple choice ahead, Wan and Henly found figural questions level with it. Neither study is a classroom quiz on a phone. Zenisky and Sireci (2002), reviewing new computer-based formats for large testing programmes, concluded that substantial measurement concerns still stood in the way of wider use. The safe reading is modest: choose a hotspot because the skill is locating, not because the format is new.
What we could not confirm
We found no tested list of writing rules for hotspot questions, and no study that measured the best size for an accepted region. The next two sections are reasoned from how the shape is marked and from accessibility guidance on pointer targets. Treat them as reasoning.
How do you write a hotspot question?
Write a hotspot question by choosing one target, taking every label off the image, and naming the thing in the stem without saying where it is.
- Choose one target. One question, one click, one structure. A question that needs two clicks but accepts one marks half the knowledge as all of it.
- Use an image without labels. Letters, leader lines, captions and legends turn locating back into reading.
- Make the image read on a phone. If the target is a speck at a phone's width, crop the picture.
- Name the thing, not its position. "Click the left ventricle" is a hotspot stem. "Click the large chamber at the bottom right" has answered itself.
- Draw the accepted region to the sizes in the next section. In EdProcto's hotspot template it is a rectangle drawn on the image.
- Answer it yourself on a phone, with a finger, before a student does.
A stem before and after
Before: a textbook diagram of the heart with its labels on, and the stem "Find the valve between the upper and lower chambers on the left of the picture." It has labels to read and a position in the stem, and two valves fit, depending on whose left is meant.
After: the same diagram without labels, cropped to the heart, and the stem "Click the mitral valve." One target, nothing to read, and which side of a drawn heart is its left is now for the student to know.
How big should the accepted region be?
Draw the accepted region at least as large as a fingertip on a phone, to protect the student who knows, and no larger than a few per cent of the image, to defeat the student who does not.
The chance that a blind click lands inside is the region's area divided by the image's area. EdProcto stores the region as a rectangle in fractions of the image's width and height, so the sum is width share times height share. For the lower limit, borrow the web accessibility standard's sizes for pointer targets: at least 24 by 24 CSS pixels, and 44 by 44 at its highest level (World Wide Web Consortium, 2024). Its reflow rule is written for a width of 320 CSS pixels, so the table assumes a picture shown 320 pixels wide and 240 high.
| Region (width by height, as shares of the image) | Random hit | Pixels on a 320 by 240 image | Verdict |
|---|---|---|---|
| 5% by 5% | 0.25% | 16 by 12 | Too small for a finger |
| 10% by 10% | 1% | 32 by 24 | The least worth drawing |
| 15% by 20% | 3% | 48 by 48 | Comfortable for a finger |
| 50% by 50% | 25% | 160 by 120 | The same as a guess among four options |
Those sizes were written for controls such as buttons, so they are a yardstick here, not a rule the region must pass.
Two limits of the arithmetic
The area figure is a floor. A student who does not know will not click at random: they will click a structure. On a cell diagram with eight organelles, a guess among them is right about one time in eight, whatever the region's area. Count the plausible structures as you would count options.
The rectangle is the other limit. A long structure lying on a diagonal has a bounding box that swallows its neighbours. Rotate or crop the image, ask for a compact part ("the head of the femur"), or draw the rectangle inside the bulk of the target.
Who does a click on an image exclude, and what should you offer instead?
A click on an image excludes students who cannot see the picture and students who cannot point at it, and the fair answer for them is a different question on the same knowledge.
- Students who are blind or have low vision. A description full enough to answer from would give the answer away. The standard allows for this: where a test would be invalid as text, it asks only that the image be identified (World Wide Web Consortium, 2024).
- Students who use a keyboard, a switch or their voice in place of a pointer. The same standard expects every function to work from a keyboard. Check, in whatever tool you use, that a point can be placed without a mouse.
- Students with hand tremors, spasticity or quadriplegia, or an input device less exact than a mouse. The guidance on minimum target size names them as the people small targets fail (World Wide Web Consortium, 2026).
- Students who cannot tell certain colours apart, where colour alone separates the parts.
Offer a parallel question that keeps the objective and changes the channel: "Which chamber does blood enter after it passes the mitral valve?" tests the same map of the heart in words. Decide it before the quiz and give it the same marks.
EdProcto's "Accessibility" settings offer "Screen-reader mode", "Larger text", "High contrast" and "Audio for visual questions", which reads the picture's description aloud. None of them changes what the shape asks for, a position on a picture, so plan the alternative yourself.
Are map questions the same shape?
Yes: a map question is a hotspot whose image is a map, and every rule above applies. EdProcto lists it as a template of its own, find it on the map, and maps bring three problems with them.
- Labels. Use a blank outline map. A map with names on it is a reading test.
- Size. Kenya on a map of Africa is a region a finger can hit. Luxembourg on a map of Europe is not. Crop to the area you taught.
- Shape. Croatia's bounding box takes in most of Bosnia and Herzegovina, so a rectangle round the whole country would mark the wrong country right. Draw it inside the bulk of the country, or pick another target.
Use a projection the class has worked with, so the difficulty is the place and not the map.
What are the common mistakes?
The common mistakes either hand the student the answer or take marks away for the wrong reason.
- A label left in a legend or a caption, or a position left in the stem.
- A region drawn to the pixel. Marking is all or nothing, so a right answer a hair outside scores zero.
- A region that covers two structures. The nearest wrong click is accepted with the right one.
- The textbook's page, unchanged. Students recall the layout. Use another drawing or a photograph of the same thing.
How does EdProcto build and mark a hotspot?
EdProcto is the best tool for image-click questions when the result has to stand as a record, because the accepted region never reaches the student's device and the mark is made on the server.
The hotspot template takes an image and a region that you draw, kept as a rectangle in fractions of the image so the question works at any size. The student's page carries the image and nothing else, and one click is the answer. Marking is by rule: a click inside the rectangle, edges included, earns the question's points, and a click anywhere else earns none. A question with no region drawn is not marked by rule and waits as pending.
A hotspot also leaves little to paste into a chatbot. The stem "Click the mitral valve" has no text answer: the answer is a place in a picture the chatbot has not been shown, and whatever it says must still become a click in the right spot. That makes the shape a first defence, not the only one: a student can photograph the screen or ask a neighbour, and no shape makes cheating impossible. For a quiz that counts, add the integrity level the stakes need.
| What you want to know | Shape in EdProcto |
|---|---|
| Where one thing is on a diagram | Hotspot: ready, marked by rule |
| Where a place is | Find it on the map: ready, marked by rule |
| Whether a young child can point to it | Touch the picture: ready, with big, forgiving zones |
| Which of two images meets a criterion | Compare two images: ready |
| A thing's extent, every instance of it, or the names of its parts | Outline a region, map region, click all that apply, label the parts: ready, marked by rule |
The quiz maker on the home page drafts text questions from a chapter or a handout, and the hotspots go in beside them with your own images. The Free plan covers the core templates, marking by rule and light integrity. Tutor opens all 111 templates, and Educator adds the full analysis, where a hotspot's difficulty and discrimination sit beside every other question's. The pricing page says which plan carries what.
Quick questions
What is a hotspot question?
A hotspot question shows an image and asks the student to click the place where something is. The answer is a position, not a word or an option. EdProcto's hotspot template marks it by rule against a region you draw.
Can students guess a hotspot question?
Less easily than multiple choice. A region covering 3% of the image catches 3% of blind clicks, and a student choosing among eight visible structures is right about one time in eight.
Are hotspot questions accessible to every student?
No. Students who cannot see the image or cannot use a pointer need an equivalent question on the same knowledge, decided before the quiz and worth the same marks.
Can a chatbot answer a hotspot question?
Not from the stem: there are no options and no text answer to paste back, only a point on an image. That lowers the risk without removing it, so a quiz that counts still needs integrity settings that fit its stakes.
References
- Jodoin, M. G. (2003). Measurement efficiency of innovative item formats in computer-based testing. Journal of Educational Measurement, 40(1), 1–15. https://doi.org/10.1111/j.1745-3984.2003.tb01093.x
- Martinez, M. E. (1991). A comparison of multiple-choice and constructed figural response items. Journal of Educational Measurement, 28(2), 131–145. https://doi.org/10.1111/j.1745-3984.1991.tb00349.x
- Tarrant, M., Ware, J., & Mohammed, A. M. (2009). An assessment of functioning and non-functioning distractors in multiple-choice questions: A descriptive analysis. BMC Medical Education, 9, Article 40. https://doi.org/10.1186/1472-6920-9-40
- Vorstenbosch, M. A. T. M., Klaassen, T. P. F. M., Kooloos, J. G. M., Bolhuis, S. M., & Laan, R. F. J. M. (2013). Do images influence assessment in anatomy? Exploring the effect of images on item difficulty and item discrimination. Anatomical Sciences Education, 6(1), 29–41. https://doi.org/10.1002/ase.1290
- Wan, L., & Henly, G. A. (2012). Measurement properties of two innovative item formats in a computer-based test. Applied Measurement in Education, 25(1), 58–78. https://doi.org/10.1080/08957347.2012.635507
- World Wide Web Consortium. (2024). Web Content Accessibility Guidelines (WCAG) 2.2 (W3C Recommendation). https://www.w3.org/TR/WCAG22/
- World Wide Web Consortium. (2026). Understanding SC 2.5.8: Target size (minimum). https://www.w3.org/WAI/WCAG22/Understanding/target-size-minimum.html
- Zenisky, A. L., & Sireci, S. G. (2002). Technological innovations in large-scale assessment. Applied Measurement in Education, 15(4), 337–362. https://doi.org/10.1207/s15324818ame1504_02