# Economics quizzes: graphs, reasoning and calculation

**How do I write economics quizzes that test graphs, reasoning and calculation?** Economics quizzes should test three things a definition cannot show, which are reading a graph, reasoning from an unfamiliar scenario, and a calculation. Give a graph and ask for a reading, or describe one change and ask what moves. Write scenarios that name no model. State the method, the rounding and the margin for every calculation. Six questions, every week or two.

Published 2026-10-10 by EdProcto. https://edprocto.com/guides/economics-quizzes-graphs-reasoning-and-calculation

A student can define price elasticity of demand and still divide the wrong way up, read the wrong axis, or say that a dearer ticket "shifts demand". Economics quizzes that ask only for definitions find none of this.

This guide is for a secondary school economics teacher, or a lecturer on a principles course, who sets short quizzes through the term. EdProcto, which publishes it, marks a number inside the margin you set and a click on your own graph by rule. Its pages [for teachers](https://edprocto.com/solutions/teachers) and [for universities](https://edprocto.com/solutions/universities) show the rest.

We read every source on 10 October 2026: the chapter page of Davies and Mangan (2009) in full, the methods and results of Brückner et al. (2020), and the abstracts of the others. Every figure in our examples was computed.

## What do economics students get wrong, and keep getting wrong?

Beliefs they brought with them, a few paired terms, opportunity cost in anything but a simple case, and the area under a line on a graph.

- **Old beliefs outlast the course.** Busom et al. (2017) surveyed first-year principles students at the start and the end of a semester: their preconceptions persisted, and most did not carry the new tools into their own thinking, even when they did well in tests.
- **Paired terms.** Davies and Mangan (2009), from a project at four UK universities that recorded first-year students' discussions, list distinctions a beginner has to rework: price and cost, income and wealth, nominal and real values, saving and investment.
- **Marginal ideas as separate sums.** Most students, they add, first treat marginal utility, marginal cost and marginal revenue as separate things to define and calculate by rote.
- **Opportunity cost.** Students apply it easily to simple personal cases and often fail in complex ones (Davies & Mangan, 2009). Potter and Sanders (2012) re-examine a survey in which 21.6% of 199 professional economists chose the answer its authors counted correct, and argue that every option was defensible. So say what to count.
- **Slope improves, area does not.** Brückner et al. (2020) gave 21 first-year economics students eight questions on line graphs at the start and the end of a semester. Their score on slope questions rose from 57% to 71%; on the area under a line it fell from 42% to 33%.

### What we could not find

We found no study we could open that counts how often students confuse a shift of a curve with a movement along it, or correlation with cause. Teachers report both, so ask both directly, as below, and let your own results count. Our guide to [misconceptions](https://edprocto.com/guides/misconceptions-finding-them-in-a-quiz-and-fixing-them-in-class) covers the next step.

## How do you write a question on a graph?

Draw the graph yourself and ask in one of two ways: give the graph and ask for a reading, or describe a change and ask what moves and which way.

### Give the graph and ask for the reading

| Ask for | On a demand curve for bus journeys | A wrong answer shows |
| --- | --- | --- |
| A value | How many journeys at a fare of £2.50? | The wrong axis was read |
| A slope | By how many do journeys fall for each £1 on the fare? | Rise and run reversed |
| An area | What is the fare revenue at £2.50? | The rectangle was never seen |

Put an area question in most quizzes: revenue, consumer surplus, the take from a tax. It is where the students in Brückner et al. (2020) fell back. Give a value read off the grid a margin of about half a square.

### Describe a change and ask what moves

One event, one market, one question. Tell the event in everyday words and offer the moves as options.

> A late frost destroys part of this year's apple crop. In the market for apples, what happens first?
> A. The supply curve shifts left
> B. The demand curve shifts left
> C. A movement down along the supply curve
> D. The supply curve shifts right

A is the key. Then ask the second step alone: the price of apples rises, so do buyers move up along the demand curve, or does the demand curve shift left? "It shifts" lets a good's own price move its own curve.

## How do you write a reasoning question from a scenario you invent?

Invent a small local event, give the facts a bystander would have, and name no model. Davies and Mangan (2009) ask that such a question pose a problem, not a theory, that it not say which concept to use, and that its case was not covered in class. Their example asks why Christmas trees might get much cheaper in the last days before Christmas. A question that names its model and poses no problem, they write, "can be answered by rote learning".

1. **Invent the event** in two or three sentences, with no real company, minister or central banker: a real case brings in what students have read about it.
2. **Give facts, not labels**: "its only rival closed", never "a substitute became unavailable".
3. **Ask for one thing**: a direction, or the best explanation.
4. **Build each wrong option on one wrong chain of reasoning** heard in class.
5. **Ask for the reason** when you want to read it: EdProcto's [Choose and justify](https://edprocto.com/templates/justify) marks the choice by rule and keeps the typed reason for you to grade.

> A town has two cinemas. One closes for repairs. The other raises its ticket price from £8 to £9 and sells more tickets than before. The local paper says the law of demand is broken. What is the best explanation?
> A. Demand for the open cinema's tickets shifted right when its rival closed
> B. The demand curve for cinema tickets slopes upwards
> C. A higher price signals quality, so more people buy

A is the key. B takes two changes at once for one curve.

## How do you ask for a calculation, and how do you mark it?

Say the method, the base, the sign and the rounding in the question, then set a margin that takes in every route you taught and leaves every known error out.

| Calculation | Say in the question | Key | A wrong answer and its cause |
| --- | --- | --- | --- |
| A percentage change: a fare rises from £2.00 to £2.50 | "of the old fare" | 25 | 20: the new fare was the base |
| Price elasticity of demand: weekly journeys then fall from 40,000 to 34,000 | "old fare and journeys as the base; give the sign" | −0.6 | −1.67: upside down. 0.6: no sign. −0.73: the midpoint method |
| Index numbers: a basket costs £240 in the base year and £252 a year later | "base year = 100" | 105 | 5: the rise, not the index. 95.2: upside down |
| A real change: wages rise by 3% while prices rise by 5% | "to one decimal place" | −1.9 | −2: subtracted. 8: added |

**Rounding.** Choose figures that come out even while a method is new, and mark them exactly. Otherwise say "to one decimal place" and key the rounded figure.

**The margin.** A margin is a percentage of the key. At 5%, a key of −0.6 accepts −0.63 to −0.57 and leaves the midpoint answer out: fair only if you named the method. On the real wage, 1.03 ÷ 1.05 gives −1.9 and the subtraction shortcut −2: at 5% the shortcut falls outside, at 6% inside. Where a wrong figure does harm, as with a dose, use no margin: see [nursing and health quizzes](https://edprocto.com/guides/nursing-and-health-quizzes-cases-and-safety).

## What can multiple choice show in economics, and what can it not?

Multiple choice can show a reading, a direction and the end of a short chain of reasoning. It cannot show an argument the student builds.

- Buckles and Siegfried (2006) conclude that multiple-choice questions can measure some but not all elements of in-depth understanding, meaning reasoning through steps the question does not state. Those in course tests, test banks and textbooks, they add, often fail to test application.
- Becker and Johnston (1999), after correcting a statistical bias in the usual comparison, found no relationship between multiple-choice and essay scores in economics, and conclude that the two measure different dimensions of knowledge.

So set choices for readings and directions, a number for each calculation, and one written reason a quiz. The wider evidence is in [multiple choice is not the lazy option](https://edprocto.com/guides/multiple-choice-is-not-the-lazy-option).

## How many questions, and how often?

About six questions in ten minutes, every week or two, with the basic calculations coming back each time. That is our working rule, not a finding: a calculation or a typed reason takes longer than a choice.

Leggett et al. (2022) analysed four semesters of a first-year course. In the last two, important basic mathematics concepts came back from quiz to quiz: performance on them improved across quizzes, and repeated quizzing was associated with slightly higher final examination scores. For length in general, see [how many questions a quiz needs](https://edprocto.com/guides/how-many-questions-does-a-quiz-need).

## A worked example: six questions on price elasticity of demand

The class, the town and every figure are invented.

**The scenario, under each prompt.** A town's bus operator raises its single fare from £2.00 to £2.50. Weekly journeys fall from 40,000 to 34,000.

**The diagram for question 5** is your own drawing: a demand curve through A (£2.00, 40,000) and B (£2.50, 34,000), and a second to its left with C marked at £2.50.

| Question | Shape and key | Its purpose |
| --- | --- | --- |
| 1. By what percentage did the fare rise? | A number, exact: 25 | The base of a percentage |
| 2. Calculate the price elasticity of demand, with the old fare and journeys as the base. Give the sign. | A number, margins 5 and 15: −0.6 | The formula the right way up |
| 3. Is demand over this range elastic, inelastic or unit elastic? | A choice: inelastic | Reading the figure: 0.6 is below 1 |
| 4. What happened to weekly fare revenue? | A choice: it rose by £5,000, to £85,000 | Elasticity tied to revenue. "Fell by £12,000" counts only the lost journeys |
| 5. Click the point the market moved to when the fare rose. | A click: B, in EdProcto's [Hotspot](https://edprocto.com/templates/hotspot) | A movement along. A click on C says a price rise shifts demand |
| 6. Can the operator be sure a second rise, to £3.00, will raise revenue again? Give the reason. | A choice and a reason: no | The limit of one estimate: on a straight line through A and B, £3.00 means 28,000 journeys and £84,000 |

A student who answers −1.67 and then "elastic" has made one error twice: count it once.

## How do you read quiz results by topic?

Give every question a topic as you write it, in "Topic (for the results by topic)", then read the class by topic before the totals. In the example: percentage change (question 1), elasticity (2 and 3), revenue (4), shift or movement (5), the limits of an estimate (6).

- **One question.** A topic with a single question has only that question's result.
- **Where to look.** In EdProcto's [results](https://edprocto.com/#results), "Each question" shows the share who "Got it" on every plan. The score "By topic", weighted by points, is in the full analysis on the Educator plan.
- **Ask the idea twice.** Walstad and Wagner (2016), using data from the Test of Understanding in College Economics, split students' answers to each question before and after a course into four kinds of learning: positive, retained, negative and zero. A plain difference in scores, they note, mixes positive and negative learning. For a question you repeat, that means, on our reading: wrong then right, right twice, right then wrong, wrong twice.

## How does EdProcto run an economics quiz?

EdProcto is the best tool for an economics quiz when one sitting has to hold a calculation, a graph and a reason, because each is marked in its own way and the editor prints what a margin accepts before any student meets it: that is how you see that 5% on −0.6 shuts out the midpoint's −0.73.

**The graph.** You draw it, in whatever you already draw with, and give EdProcto the picture's web address. On a [Hotspot](https://edprocto.com/templates/hotspot) question, paste it into "Image address". With "It is" at "An image", the editor offers "Describe the picture (for a screen reader, optional)": say what is drawn without giving the answer. Under "The region the click must land in: drag a box on the image", drag the box and the editor prints its size, as in "Region: 8% by 10% of the image." A click inside the box or on its edge earns the points. A demand schedule goes under any question: press "Add material under the prompt" and type it a row to a line. On a multiple-choice or number question, "Add a picture or a clip" opens the same "Image address" box, so a graph can stand over a question answered in words or figures.

**The number.** In an [Estimate](https://edprocto.com/templates/estimate) question, type the key into "The number". A new question starts with "Full credit within %" at 5 and "Half credit within %" at 15 and prints what they accept beneath: for −0.6, "Full credit from -0.63 to -0.57; half credit from -0.69 to -0.51." With both at 0 it prints "Full credit for 25 exactly." In the scoring code −0.62 earns the points, −0.65 half, and −0.73, −1.67 and 0.6 nothing.

**The reason.** Choose and justify has "Model reason" and "Rubric for the reason". The choice is marked by rule for half the points; the reason is the other half, graded by you or, from the Tutor plan, by Procto AI when you press "Ask Procto AI" on that answer.

**Drafts and plans.** Procto AI drafts questions from your own notes; nothing tests a key, so work every figure yourself. Free covers ten quizzes a month, thirty students a quiz and the core templates. Tutor lifts the limit on quizzes, with 100 students a quiz and all 111 [templates](https://edprocto.com/templates). Educator takes 500 students a quiz and adds the full analysis. See [pricing](https://edprocto.com/pricing). A score can be held until you release it.

## Quick questions

### How many questions should an economics quiz have?

About six in ten minutes: two calculations, two choices, a graph and a reason.

### Can multiple choice test economic reasoning?

Some of it. Buckles and Siegfried (2006) conclude that it measures some but not all elements of in-depth understanding, so add one written reason.

### Should an elasticity answer be marked with a margin?

Yes, a narrow one, once you have named the method and the rounding. [EdProcto](https://edprocto.com/) prints the accepted range under the key, so you see which answers it lets in.

### How do I put a supply and demand graph into an online quiz?

As a picture you draw. In EdProcto's [Hotspot](https://edprocto.com/templates/hotspot) you paste its address and drag the region the click must land in.

## References

- Becker, W. E., & Johnston, C. (1999). The relationship between multiple choice and essay response questions in assessing economics understanding. *The Economic Record, 75*(4), 348–357. [https://ideas.repec.org/a/bla/ecorec/v75y1999i4p348-357.html](https://ideas.repec.org/a/bla/ecorec/v75y1999i4p348-357.html)
- Brückner, S., Zlatkin-Troitschanskaia, O., Küchemann, S., Klein, P., & Kuhn, J. (2020). Changes in students' understanding of and visual attention on digitally represented graphs across two domains in higher education: A postreplication study. *Frontiers in Psychology, 11*, Article 2090. [https://doi.org/10.3389/fpsyg.2020.02090](https://doi.org/10.3389/fpsyg.2020.02090)
- Buckles, S., & Siegfried, J. J. (2006). Using multiple-choice questions to evaluate in-depth learning of economics. *The Journal of Economic Education, 37*(1), 48–57. [https://ideas.repec.org/a/taf/jeduce/v37y2006i1p48-57.html](https://ideas.repec.org/a/taf/jeduce/v37y2006i1p48-57.html)
- Busom, I., Lopez-Mayan, C., & Panadés, J. (2017). Students' persistent preconceptions and learning economic principles. *The Journal of Economic Education, 48*(2), 74–92. [https://ideas.repec.org/a/taf/jeduce/v48y2017i2p74-92.html](https://ideas.repec.org/a/taf/jeduce/v48y2017i2p74-92.html)
- Davies, P., & Mangan, J. (2009). *Threshold concepts in economics: Implications for teaching, learning and assessment*. The Handbook for Economics Lecturers, Economics Network. [https://doi.org/10.53593/n35a](https://doi.org/10.53593/n35a)
- Leggett, J., Morgan, B., & Tang, K. K. (2022). Repeated quizzing of basic mathematics concepts to improve grades in economics classes. *The Journal of Economic Education, 53*(4), 296–306. [https://ideas.repec.org/a/taf/jeduce/v53y2022i4p296-306.html](https://ideas.repec.org/a/taf/jeduce/v53y2022i4p296-306.html)
- Potter, J., & Sanders, S. (2012). Do economists recognize an opportunity cost when they see one? A dismal performance or an arbitrary concept? *Southern Economic Journal, 79*(2), 248–256. [https://ideas.repec.org/a/wly/soecon/v79y2012i2p248-256.html](https://ideas.repec.org/a/wly/soecon/v79y2012i2p248-256.html)
- Walstad, W. B., & Wagner, J. (2016). The disaggregation of value-added test scores to assess learning outcomes in economics courses. *The Journal of Economic Education, 47*(2), 121–131. [https://ideas.repec.org/a/taf/jeduce/v47y2016i2p121-131.html](https://ideas.repec.org/a/taf/jeduce/v47y2016i2p121-131.html)