CBSE Class 10 · Artificial Intelligence, code 417 · MCQ bank 2026-27

Class 10 AI MCQ for 2026-27, sorted by the units you will actually be examined on.

This class 10 AI MCQ bank follows the 2026-27 CBSE 417 unit names, not the older chapter list many question banks still use. There are 36 original questions, each with four options, the answer and one line on why it is right or why the tempting option is wrong. Tap Answer to check yourself. In our Monday and Wednesday batch, sets like these are timed, marked live and the weakest unit comes back the next week.

Book a free demo class Book a priority demo Ask on WhatsApp

Live online · 36 original MCQs below · timed sets marked in class · weak units re-drilled · Section A inside every mock · first class free

Start here

Courses behind this MCQ practice

The 417 course whose Class 10 batch runs these timed sets, the Python course for the practical-only units, and the Orange course for the no-code tools that options in the paper often name.

CBSE Computational Thinking and AI for Teens course thumbnail

The batch / 417 Class 10

CBSE AI for Class 9 to 12

The full 417 course: each unit taught, then tested as a timed objective set that is marked in class with the reason for every option.

Open the course →
Python for Teens course thumbnail

Units 4 and 7 / Python

Python for Teens

NumPy, Matplotlib and the small programs behind the practical exam, where questions like the median and scatter ones below turn into code.

Open the course →
Visual Machine Learning with Orange for Teens course thumbnail

No-code tools / Orange

Visual ML with Orange

Hands-on with the no-code tool the curriculum uses for data, image and text case studies, so tool-based options stop being guesses.

Open the course →

Ans. The short version

Class 10 AI MCQ practice for 2026-27 should follow the current CBSE 417 units: Part A employability skills, then Unit 1 (AI Project Cycle and Ethical Frameworks), Unit 2 (Advanced Concepts of Modeling), Unit 3 (Evaluating Models), Unit 5 (Computer Vision) and Unit 6 (Natural Language Processing). Units 4 and 7 are practical only. In the 2025-26 sample paper, objective questions carried 24 of the 50 theory marks: any 4 of 6 from Part A and any 20 of 24 from the AI units, with no negative marking. Modern Age Coders drills these MCQs live in a Monday and Wednesday 9:00 PM IST batch, or personalised one to one.

Q1. How much are MCQs worth?

How class 10 AI MCQs are weighted in the 417 paper

The 417 sample paper for 2026-27 is not out yet and the curriculum is unchanged, so the honest guide is the 2025-26 sample paper and its blueprint. Here is its objective section.

Group in this bankOne-mark objective items printed (2025-26)You answer
Part A: Communication, Self-Management, ICT, Entrepreneurial and Green Skills6any 4
Unit 1: Revisiting AI Project Cycle and Ethical Frameworks for AI5any 20 of the 24
Unit 2: Advanced Concepts of Modeling in AI4
Unit 3: Evaluating Models6
Unit 5: Computer Vision4
Unit 6: Natural Language Processing5
Unit 4: Statistical Data and Unit 7: Advance Python0practical only

What the numbers mean for practice

  • 24 of 50 theory marks. Almost half the written paper, one mark per item.
  • No negative marking. Leave nothing blank.
  • Built-in choice. Two Part A items and four AI items can be skipped, so one weak unit need not cost a mark.
  • Evaluating Models leads with six of the 24 AI items, several of them calculations.

Why the unit names matter

Many free banks still use the earlier chapter titles, such as Data Science or Evaluation. In 2026-27, data handling sits in Unit 4, which is examined only in the practical, so an old-style bank mixes theory and practical items. This bank uses the current names, so you can see which items count in Section A.

The full blueprint, written questions included, is on CBSE Class 10 AI exam date and paper pattern.

Q2. Part A and Unit 1

Class 10 AI MCQ with answers: Part A and the project cycle unit

Questions 1 to 10. Try each before opening the answer, then read the reason: it names the option most students wrongly pick.

Part A: employability skills (one per unit)

  1. During a group discussion, Arjun nods and keeps eye contact while a classmate speaks. This is
    (a) verbal communication
    (b) written communication
    (c) non-verbal communication
    (d) visual communication
    Answer (c) No words are used, only body language. Visual communication means charts, signs and pictures.
  2. In a SMART goal, the letter M stands for
    (a) Meaningful
    (b) Measurable
    (c) Motivating
    (d) Manageable
    Answer (b) You must be able to track progress. "Manageable" is the trap: being realistic is what the A, achievable, covers.
  3. Which of these protects a computer from viruses that arrive with downloads?
    (a) defragmenting the disk
    (b) installing and updating antivirus software
    (c) raising screen brightness
    (d) deleting temporary files
    Answer (b) The others are care and maintenance tasks; none of them stops harmful software.
  4. Meera sees that students near her school cannot buy healthy snacks, and starts a small stall. Which entrepreneurial quality does she show first?
    (a) spotting an opportunity
    (b) avoiding every risk
    (c) following instructions
    (d) waiting for a job offer
    Answer (a) An entrepreneur starts by noticing an unmet need. Starting a stall involves risk, so (b) contradicts the story.
  5. Using compost and crop rotation in place of chemical fertilisers is part of
    (a) organic farming
    (b) monoculture
    (c) deforestation
    (d) heavy pesticide use
    Answer (a) Organic farming keeps soil healthy without synthetic chemicals; monoculture repeats one crop and drains it.

Every Part A unit, with its short answers as well, is covered on Part A employability skills.

Unit 1: Revisiting AI Project Cycle and Ethical Frameworks for AI

  1. Which stage of the AI project cycle comes straight after Data Acquisition?
    (a) Modelling
    (b) Data Exploration
    (c) Problem Scoping
    (d) Evaluation
    Answer (b) Collected data is explored and visualised before a model is chosen. Jumping to (a) is the usual slip.
  2. In the 4Ws problem canvas, which W asks how solving the problem will benefit the people affected?
    (a) Who
    (b) What
    (c) Where
    (d) Why
    Answer (d) Why records the value of the solution. What describes the problem itself, which is why it tempts.
  3. A set of ethical rules written only for hospitals and medical AI is
    (a) a value-based framework
    (b) a sector-based framework
    (c) a rule-based model
    (d) a utility-based framework
    Answer (b) It is built for one sector, healthcare. Bioethics is the standard example of a sector-based framework.
  4. A city picks the traffic AI that cuts total travel time for the most people, even though a few roads get slower. Which approach is this?
    (a) rights-based
    (b) virtue-based
    (c) utility-based
    (d) sector-based
    Answer (c) It weighs the overall benefit against the harm. A rights-based approach would protect every road user first.
  5. Which of these is NOT one of the four principles of bioethics?
    (a) respect for autonomy
    (b) do not harm
    (c) ensure maximum benefit
    (d) maximise the hospital's income
    Answer (d) The fourth principle is justice, meaning fair treatment. Income appears nowhere in the four.

Q3. Units 2 and 3

Modeling and evaluating models: class 10 MCQ, including the numericals

Questions 11 to 22. Unit 3 carries the most objective items, and five of its seven below need a quick calculation.

Unit 2: Advanced Concepts of Modeling in AI

  1. A spam filter gets better every week as users mark new emails as spam. It follows
    (a) a rule-based approach
    (b) a learning-based approach
    (c) a fixed if-else checklist
    (d) no model at all
    Answer (b) It learns from fresh labelled data. A rule-based system changes only when a person rewrites the rules.
  2. A model predicts tomorrow's highest temperature in degrees Celsius. This is
    (a) classification
    (b) clustering
    (c) regression
    (d) association
    Answer (c) The output is a continuous number. Labels such as "hot" or "cold" would make it classification.
  3. An online shop groups its customers by buying habits, with no labels given. It uses
    (a) supervised learning, classification
    (b) unsupervised learning, clustering
    (c) supervised learning, regression
    (d) reinforcement learning
    Answer (b) No labels means unsupervised, and grouping means clustering. Association finds items bought together, not groups of people.
  4. A perceptron has inputs 1, 0 and 1, weights 0.4, 0.7 and 0.2, and a bias of -0.5. What is the weighted sum plus bias?
    (a) 0.6
    (b) 0.8
    (c) 0.1
    (d) 1.3
    Answer (c) 0.4 + 0 + 0.2 - 0.5 = 0.1. Option (a) forgets the bias; (b) treats the middle input as 1.
  5. Which deep learning network is designed for images, using small filters that slide across the picture?
    (a) ANN
    (b) CNN
    (c) a decision tree
    (d) a rule-based system
    Answer (b) A convolutional neural network. ANN is the general family, so it is close but not the specific answer.

Unit 3: Evaluating Models

Use for questions 16 to 19disease = positive
A model screens 200 X-ray reports.
It flags 60 as positive; 54 of them
really have the disease. Of the 140
it calls negative, 36 have it.

              Reality: Yes  Reality: No
Predicted Yes   TP = 54      FP = 6
Predicted No    FN = 36      TN = 104

Build the matrix from the words first. Every option below that looks right but is not comes from a swapped cell or a skipped formula.

  1. The precision of this model is
    (a) 0.60
    (b) 0.79
    (c) 0.90
    (d) 0.72
    Answer (c) TP / (TP + FP) = 54 / 60 = 0.90. Option (a) is the recall.
  2. The recall of this model is
    (a) 0.60
    (b) 0.90
    (c) 0.27
    (d) 0.79
    Answer (a) TP / (TP + FN) = 54 / 90 = 0.60. Option (c) divides TP by all 200 reports, which no metric does.
  3. The F1 score is
    (a) 0.75
    (b) 0.72
    (c) 0.79
    (d) 0.54
    Answer (b) 2 x 0.90 x 0.60 / (0.90 + 0.60) = 1.08 / 1.50 = 0.72. Option (a) is the plain average, the most common mistake.
  4. The accuracy, and the number of wrong predictions, are
    (a) 0.79 and 42
    (b) 0.90 and 20
    (c) 0.72 and 56
    (d) 0.79 and 36
    Answer (a) (54 + 104) / 200 = 0.79, and wrong = FP + FN = 6 + 36 = 42. Option (d) counts only the false negatives.
  5. For a disease test, which error is most dangerous, so which metric matters most?
    (a) false positives, precision
    (b) false negatives, recall
    (c) true negatives, accuracy
    (d) true positives, precision
    Answer (b) A missed patient goes untreated. For a spam filter the answer flips to precision.
  6. Of 100 patients, 5 have a rare disease. A lazy model calls everyone healthy. Its accuracy is
    (a) 5%
    (b) 50%
    (c) 95%
    (d) 100%
    Answer (c) 95 of 100 are right, yet its recall is 0. Accuracy flatters a model when one class is rare.
  7. Why is a model not tested on the same data it was trained on?
    (a) testing would be slower
    (b) the score would be over-optimistic and hide overfitting
    (c) training data has no labels
    (d) the data would be deleted
    Answer (b) A model can memorise its training data, so only unseen test data shows how it will really perform.

Q4. Units 5 and 6, then the practical units

Computer vision class 10 MCQ and NLP class 10 MCQ, with the counting questions

Questions 23 to 36. Pixel and token counts are where a careful student and a rushed one part ways.

Unit 5: Computer Vision

  1. A self-driving car's camera draws a box around every pedestrian and cyclist in a frame. Which task is this?
    (a) image classification
    (b) object detection
    (c) instance segmentation
    (d) image processing only
    Answer (b) It finds and locates many objects with boxes. Segmentation would trace each object's exact outline in pixels.
  2. How many pixels are there in an image of resolution 640 x 480?
    (a) 1,120
    (b) 307,200
    (c) 921,600
    (d) 30,720
    Answer (b) 640 x 480 = 307,200. Option (c) is three times that, the count of RGB values, not pixels.
  3. A colour (RGB) image is 200 x 100 pixels. How many numbers does it store?
    (a) 20,000
    (b) 300
    (c) 60,000
    (d) 600
    Answer (c) Each pixel holds a red, a green and a blue value: 200 x 100 x 3. Option (a) would be the grayscale version.
  4. In a grayscale byte image, a pixel value of 0 means
    (a) pure white
    (b) pure black
    (c) mid grey
    (d) a missing pixel
    Answer (b) 0 is no brightness and 255 is full brightness, white. Reversing the two is a frequent slip.
  5. Why does a byte image allow exactly 256 pixel values?
    (a) a byte has 8 bits, and 2 to the power 8 is 256
    (b) screens show only 256 colours
    (c) each pixel uses 256 bytes
    (d) cameras record 256 shades
    Answer (a) Eight bits give 256 patterns, which is why values run from 0 to 255.

Unit 6: Natural Language Processing

  1. How many tokens does the sentence "AI helps doctors, and doctors help patients." produce?
    (a) 7
    (b) 6
    (c) 9
    (d) 8
    Answer (c) Seven words plus the comma plus the full stop. Punctuation marks are tokens too.
  2. After lowercasing and dropping punctuation, how many unique words does that sentence give a Bag of Words dictionary?
    (a) 9
    (b) 7
    (c) 6
    (d) 5
    Answer (c) "doctors" repeats: ai, helps, doctors, and, help, patients. Option (d) wrongly merges "help" and "helps", which happens only after stemming.
  3. With the dictionary [rain, go, away, come], the vector for "rain rain go away" is
    (a) [1, 1, 1, 0]
    (b) [2, 1, 1, 0]
    (c) [2, 1, 1, 1]
    (d) [1, 2, 1, 0]
    Answer (b) Bag of Words counts occurrences in dictionary order. Option (a) only marks presence.
  4. "studies" becomes "studi" with one technique and "study" with another. Which pairing is right?
    (a) stemming gives studi, lemmatisation gives study
    (b) lemmatisation gives studi, stemming gives study
    (c) both give study
    (d) tokenisation gives studi
    Answer (a) Stemming chops the ending and may leave a non-word; lemmatisation returns a real dictionary word.
  5. A word appears in all 4 documents of a corpus. With IDF taken as log(N / documents containing the word), its IDF is
    (a) 4
    (b) 1
    (c) 0
    (d) 0.25
    Answer (c) log(4 / 4) = log 1 = 0, so its TFIDF is 0: a word found everywhere tells the documents apart not at all.
  6. "The chair sang a song" is grammatically correct but makes no sense. At which stage does it fail?
    (a) lexical analysis
    (b) syntactic analysis
    (c) semantic analysis
    (d) tokenisation
    Answer (c) The grammar passes the syntactic check; the meaning fails the semantic one.

Units 4 and 7: practical only, not in the theory paper

These units carry no Section A items, but the practical exam and viva ask students to predict what a program prints.

  1. For the data 9, 3, 10, 4, 4, the median is
    (a) 10
    (b) 6
    (c) 4
    (d) 3
    Answer (c) Sort first: 3, 4, 4, 9, 10, so the middle value is 4. Option (a) is the middle of the unsorted list and (b) is the mean.
  2. What does print(np.median([1, 4, 9, 7])) show?
    (a) 9
    (b) 5.5
    (c) 5.25
    (d) 4
    Answer (b) Sorted, the list is 1, 4, 7, 9; with an even count the median is (4 + 7) / 2. Option (c) is the mean.
  3. Which Matplotlib call draws a scatter chart of x against y?
    (a) plt.plot(x, y)
    (b) plt.scatter(x, y)
    (c) plt.bar(x, y)
    (d) plt.hist(x)
    Answer (b) plt.plot joins the points into a line chart by default.

The programs behind these three are written out on Python for CBSE Class 10 AI.

Q5. How do students lose easy marks?

MCQ technique for 417, and the traps that cost the most

Objective marks are rarely lost on unstudied topics. They go to distractors built from half-remembered ideas. These are the patterns we see most when marking class sets.

TrapWhat it looks likeThe habit that beats it
Swapped cellsPrecision and recall both appear as options, from FP and FN reversedDraw the matrix with Reality across the top every time, even for a one-mark item
Average for F1An option equal to (P + R) / 2Write 2PR / (P + R) beside the question before choosing
Forgotten punctuationToken counts one or two shortCircle each comma and full stop before counting
Pixels versus valuesA figure three times too large or too smallAsk: does the question want pixels, or numbers stored?
Family versus memberANN when the answer is CNN; supervised when it is classificationPick the most specific option that is still true
Close definitionsClustering against association; stemming against lemmatisationLearn one example of each that cannot belong to the other
The NOT questionThree true statements and one odd one outUnderline NOT, then mark each option true or false on its own

Read all four first

It catches the option that is true but less specific than the one the examiner wants.

Eliminate, then decide

Strike the plainly wrong ones. With no negative marking, a choice between the last two is always worth making.

Choose your skips early

On a first read, mark the four AI items you are least sure of, answer the rest in order, then return if time allows.

Section A also carries assertion and reason items: they appeared in the 2025 and 2026 board papers and the 2025-26 sample paper. This bank keeps to four-option MCQs; assertion and reason sets have their own page, CBSE Class 10 AI assertion reason questions. Written answers for every unit are on CBSE Class 10 AI important questions.

Q6. How are MCQs drilled in class?

How the live batch turns an MCQ bank into marks

A bank worked alone gives a score. It rarely shows why the misses happened or whether they will happen again in February. That is what a teacher adds.

Timed sets, marked live

After each unit, the Monday class sits a fresh set of ten to twelve items against the clock. Answers are then gone through on screen, and for every option the class says why it is right or why it is a trap.

Weak units come back

Each student's misses are logged by unit. The unit with the most misses returns in the Wednesday mixed round, with new numbers, so a remembered answer is no help and only the method scores.

Section A inside full mocks

From December, each full timed mock opens with a complete objective section. Students practise picking their four skips, then see their score by unit with the reasons still fresh.

Any student can stop a set to ask why an option fails, and the teacher works it through on the shared screen. The mock calendar is on the sample paper and mock tests page, and the term plan that the sets sit inside is on CBSE Class 10 AI board exam preparation.

Batch or personalised: you choose

Drill these MCQs in the Monday and Wednesday batch, or one to one

Same question sets, same marking, same teachers. The batch gives a fixed rhythm with classmates to race against the clock; one to one moves at your child's pace on the units they keep missing. Monthly billing, no admission fee, stop at the end of any month.

Option 1 · Fixed batch timings

Group batch, Monday and Wednesday 9 PM

₹1,499

per month, two live classes a week

  • Monday and Wednesday, 9:00 PM IST, live on video
  • Fifteen to twenty Class 10 AI students, one teacher
  • A timed objective set after every unit
  • Mixed-unit rounds on Wednesdays
  • Full Section A inside every mock from December
Book a free demo in this batch

Option 2 · Choose your own timing

Personalised one to one

₹4,999

per month, one private class a week, at a time set with you

  • Days and time picked by you, in your own time zone
  • Sets built only from the units your child misses most
  • Numerical MCQs repeated with new figures until steady
  • Every wrong option talked through, one student at a time
  • Same CBSE 417 syllabus and teachers
Ask for a personalised plan

A student who scores well on most units but keeps dropping marks on confusion matrices or token counts often fixes it in a few one to one sessions. A student who has not yet covered every unit gains more from the batch, where the whole syllabus is taught and tested in order. Fees are shown in your own currency.

Studying outside India?

Class 10 AI MCQ practice for CBSE students in the Gulf and abroad

A CBSE school in Dubai, Riyadh or Nairobi sets the same 417 paper, so the same objective section is waiting at the end of the year. Here is the batch hour in local time this winter.

Where you liveMon and Wed 9:00 PM ISTBest format
India (IST)
every Indian city		9:00 PMBatch fits
UAE and Oman
Dubai, Abu Dhabi, Sharjah, Muscat		7:30 PMBatch fits
Saudi Arabia, Qatar, Kuwait, Bahrain
Riyadh, Jeddah, Dammam, Doha, Kuwait City, Manama		6:30 PMBatch fits
Nepal
Kathmandu		9:15 PMBatch fits
Ethiopia and East Africa
Addis Ababa and East Africa Time		6:30 PMBatch fits
Singapore and Malaysia
Singapore, Kuala Lumpur		11:30 PMChoose 1-on-1

If 9:00 PM IST lands too late where you live, personalised one to one runs at an hour you choose. More for families abroad is on CBSE Class 10 AI online classes in the Gulf and abroad.

What families say

Rated 4.9 across 547 Google reviews

Real reviews from real families. We neither write nor commission them.

★★★★★

"The one step solution for my son. Modern Age Coders make learning coding so simple that kids love it. The teachers explain complex concepts clearly with practical exercises and interactive content."

Ria Mukherjee

Parent

★★★★★

"Modern Age Coders has been a game-changer for me. I struggled to grasp IT concepts and coding before joining, but their classes transformed everything. I can now confidently write complex programs with ease."

Samriddha Mondal

Student

★★★★★

"One of the most wonderful education centres out there. Education is not limited to school syllabus but focuses on skill development."

Vansh Agarwal

Student

★★★★★

"My child Dhairya is really enjoying the Modern Age Coders classes. This is his first online class and he eagerly looks forward to it. I can already see his improvement, and the teachers are very cooperative."

Sonam Oswal

Parent of Dhairya

★★★★★

"Modern Age Coders have wonderful teachers who teach in a clear, easy and practical way. The teacher boosts students' confidence and inspires them to learn without hesitation."

Sonu Goyal

Parent

★★★★★

"I highly recommend this computer coding class! The teachers are incredibly knowledgeable and passionate about coding."

Ritu Kedia

Parent

The rest of the Class 10 series

More pages for Class 10 board students

Every page in the CBSE Class 10 AI (417) series, grouped by what you need: classes, each unit, question practice, notes and revision, and the exam itself.

Classes, tuition and batches

Unit by unit

Questions and practice

Notes and revision

Exam, marks and practicals

Choosing the subject

ICSE Class 10 Computer Applications (Java): The Tue and Wed ICSE Java batch ICSE Computer Applications syllabus ICSE Java programs practice ICSE CA important questions

Questions about Class 10 AI MCQs

What students and parents ask about class 10 AI MCQs

Where can I find class 10 AI MCQ with answers for 2026-27?

This page has 36 original MCQs for CBSE 417, sorted by the 2026-27 unit names: Part A employability skills, then Units 1, 2, 3, 5 and 6, plus three practical-only items for Units 4 and 7. Each has four options, the answer and a one-line reason.

How many MCQs are there in the CBSE Class 10 AI 417 paper?

In the 2025-26 sample paper, the objective section printed 30 one-mark items and students answered 24: any 4 of 6 from Part A and any 20 of 24 from the AI units. The 2026-27 sample paper for 417 is not out yet.

Is there negative marking in class 10 AI MCQs?

No. The 2025-26 sample paper has no negative marking in the objective section, so attempt every item, using elimination when unsure.

Which unit has the most MCQs in Class 10 AI?

Evaluating Models, with 6 of the 24 AI objective items in the 2025-26 blueprint, many of them confusion matrix calculations. Units 1 and 6 follow with 5 each, and Units 2 and 5 have 4 each.

Are MCQs asked from Statistical Data and Advance Python?

Not in the theory paper. Units 4 and 7 are assessed through the practical: the file, the practical exam and the viva. Examiners there ask what a short NumPy or Matplotlib program will show.

Are assertion reason questions asked in CBSE Class 10 AI?

Yes. Assertion and reason items appeared in Section A of the 2025 and 2026 CBSE Class 10 AI board papers and in the 2025-26 sample paper, alongside two-statement items. Each asks whether both statements are true and whether the reason explains the assertion. The 2026-27 sample paper is not out yet, so we drill the format without promising its exact count.

How should I practise AI class 10 chapter wise MCQ?

One unit at a time first, timed at about a minute per item, then mixed sets once every unit is done. Read the reason for every option, and redo the weakest unit a week later with new numbers.

Are numerical MCQs asked in Class 10 AI?

Yes. Expect confusion matrix metrics (accuracy, precision, recall, F1 score), the perceptron weighted sum, pixel and RGB value counts, token counts and Bag of Words vectors.

How long should I spend on Section A?

Our mocks give the objective section about 45 to 50 minutes of the two hours, leaving the rest for written answers.

Is MCQ practice included in your classes, or is it extra?

It is part of ordinary teaching: timed sets, live marking, re-drills and mocks are all covered by the monthly fee shown on this page, in your own currency, for the batch or one to one. There is no admission fee and the first class is a free demo.

Start

Sit a free live MCQ set with a teacher

In the free first class your child takes a short timed set from two units, then goes through it with the teacher option by option, including the confusion matrix and token-count items that catch most students. You will see how the marking works, and afterwards the mentor tells you plainly which units need the most work and whether the batch or one to one suits your child better.

Want the full picture of the subject first? Read the 417 syllabus explained.

WhatsApp us · +91 91233 66161 · connect@modernagecoders.com

Every class is live video. The form requests a callback and reserves nothing.

We call within a few hours. No card, no enrolment fee.

Want it sooner? Book a priority demo

A full live class of about 45 to 60 minutes at a slot you pick, with a written skill report and learning roadmap on WhatsApp afterwards. The fee is adjusted against your first month if you enrol.

Book a priority demo

Keep exploring Modern Age Coders

By class and age

Learn more

Free resources

From the blog

Start here