Social Studies and Nepali are the subjects students most often assume will look after themselves.
They are in your own language or close to it. The chapters read easily. Nothing in them is as visibly hard as a physics numerical. So they get the leftover hour on Sunday, and then the results come back and the marks are lower than the effort in maths - which had four times the attention.
The reason is that these are the subjects where the content is easy and the answer is hard. Understanding a chapter about local government takes twenty minutes. Producing eight identifiable points about it under time pressure, with the right terminology, in the structure a marker is looking for, is a different task entirely, and it is the one being examined.
The Marks Are in the Answer, Not the Knowing
Start with the thing that changes how you revise.
In a numerical subject, if you know how to do it you get the mark. In a content subject, knowing is the entry requirement and the mark is decided by what you put on the page. Two students with identical knowledge routinely score several grades apart, and the difference is entirely in the writing.
What separates them, in order:
Number of distinct points. An eight-mark question wants around eight things a marker can find and tick. A student who writes three excellent points in flowing prose loses five marks that were available.
Findability. The points have to be visible. A dense paragraph containing eight ideas scores worse than a numbered list containing the same eight, because a marker awarding marks has to locate them, and marks that cannot be located are not awarded.
Correct terminology. These subjects mark specific words. "The group that makes the laws" and the actual term are not equally credited.
Answering the verb. "Describe" wants features. "Explain" wants causes. "Give your opinion, with reasons" wants a position and reasons, and an answer that only describes gets a fraction of it.
So revision for these subjects should be at least half practice at producing answers, and it is almost always entirely reading.
Reading the Chapter Is Not Revising the Chapter
The specific trap in content subjects is that reading works well enough to be convincing.
You read the chapter on Nepal's physical features. It makes sense. Every sentence is familiar by the second pass, and familiarity is a strong feeling that has almost nothing to do with whether you can produce the content on a blank page. Recognition and recall are different capacities and only one of them is examined.
The fix is the same in every subject and it is unpopular: read the section once, close the book, and write down what you remember. Then check. That is active recall, and in content subjects it is worth several times what re-reading is, because the whole examination is a recall task.
What you will find the first time you try it is that you can produce three of the seven points you had just read. That gap is the entire reason a chapter you "know" produces a mediocre answer, and it is invisible until you close the book.
Do the second pass a day later, not an hour later. Retrieving after a gap is much more effective than retrieving immediately, because the forgetting curve has begun to bite and the effort of recovery is what builds durability. Immediate re-testing feels better and teaches less.
For every chapter, keep one sheet with the chapter's questions on the left and nothing on the right. Revision is answering them from memory in the blank column, in points, then checking. Twenty minutes of that beats an hour of re-reading, and it is the closest thing to the exam you can do at your desk.
Structure Every Long Answer the Same Way
The single biggest improvement available in these subjects is a template, memorised, that you write into every time. It removes the deciding from the exam hall.
For a "describe" or "explain" question:
- One opening sentence that answers the question directly. Not a preamble - the answer.
- The points, one per line or numbered, each with a supporting detail. Number them to match the marks.
- One closing line only if the question asked for a conclusion.
For an "opinion" or "do you agree" question:
- State your position in the first sentence. Markers are looking for it and a hedged opening loses marks.
- Two or three reasons, each with an example.
- One acknowledgement of the other side. This is what distinguishes a strong answer from an adequate one in almost every scheme.
- Restate the position in one line.
For a "compare" question: compare point by point, not one thing then the other thing. "X does this, whereas Y does that" on each dimension. Two separate descriptions is the most common way a compare question loses half its marks.
Write the templates on a card and use them until they are automatic. In the hall you should not be deciding on a structure; you should be filling one in. That is where the time saved goes into content.
Remembering Content That Is Genuinely Long
These subjects have a volume problem: lists, sequences, dates, provisions, features. Reading them repeatedly does not work. There are techniques that do.
Group before you memorise. Seven separate items are hard; three groups of two or three are much easier. This is chunking, and it works because working memory holds a small number of units rather than a small amount of information - so making each unit bigger is free capacity. Find the natural grouping in a list of nine features and you have three things to remember.
Use a mnemonic for anything that is a fixed list in a fixed order. First letters, a sentence, anything. It feels like a trick and it is a legitimate memory tool with a long history. Make your own rather than borrowing one - the making is part of why it sticks.
Use the method of loci for long ordered sequences. Place each item somewhere along a route you know well - your walk to school, the rooms of your house - and walk it mentally. This is the oldest memory technique on record and it is unreasonably effective for ordered lists, which is exactly what these subjects are full of.
Connect facts to something rather than memorising them bare. A date attached to a cause and a consequence is far more retrievable than a date alone, because elaborative encoding gives retrieval more than one route in. "1951" is fragile; "1951, and it happened because, and what followed was" is robust.
Timelines and maps, drawn from memory. For history and geography content, drawing the timeline or sketching the map from memory and then checking is worth more per minute than reading either.
Nepali: Two Different Papers in One
Nepali gets treated as one subject and behaves as two, which is why students who are fluent still lose marks.
The literature half
Know the text, not the summary. Questions on a prescribed text want specifics - what a character did, the sequence of events, the particular line. A general impression of the story produces a general answer and general answers score in the middle.
Learn to write about a text, not to retell it. A question asking about a theme wants analysis with evidence from the text. Retelling the plot is the most common answer and it is not what was asked. State the point, quote or cite the evidence, explain how it supports the point.
Character and theme questions are predictable. For every prescribed text, prepare the main characters, the central themes, and two or three pieces of evidence for each. That preparation covers most of what can be asked.
The language half
Grammar in Nepali is examined as transformation and correction, and it is learned by doing exercises, not by reading rules. A rule you can state and cannot apply earns nothing.
Writing tasks have formats with required parts, exactly as in the English paper - a letter, an application, an essay, a report each has components that carry marks. Learn the checklists cold. This is the cheapest set of marks in the subject and it is thrown away every year.
Handwriting and script matter here more than anywhere else. Devanagari has letter pairs that become ambiguous when written quickly, and an ambiguous word in a language answer is a word the marker cannot credit. If your hand degrades in the third hour, practise writing for three hours - it is a physical skill.
Read Nepali outside the syllabus. Newspapers, essays, anything. You cannot produce sentence patterns you have never encountered, and the students who write well in Nepali are almost always the ones who read it for pleasure. This is the one piece of advice here that keeps paying after the exam.
Social Studies: Understand the System, Not the Sentences
Social Studies rewards understanding how something works, because a question can approach the same content from any angle.
Learn the structure, then the details. How local government is organised, what each level does, how they relate. Once the structure is in place, the details attach to it. Memorised details with no structure fall apart when a question comes at them sideways.
Know the "why" for everything. Why is this provision there? Why did that policy change? Cause-and-effect questions are common and cannot be answered from a list of facts.
Current context is fair game and is what distinguishes the top answers. An answer that connects a syllabus concept to something real earns more than one that recites the concept. This does not require following the news obsessively; it requires one example per major topic, prepared in advance.
Maps and diagrams are marks. Physical features, resources, administrative divisions. Draw them from memory. Free international material will not help here - the scope is defined by your own curriculum, which the Curriculum Development Centre publishes, and getting the scope right matters more in this subject than in almost any other.
Read the education system content carefully if it appears - questions about the system you are inside are common and students answer them from assumption rather than from the chapter.
Practise With Real Papers, and Mark Yourself Harshly
These subjects reward past-paper practice more than any others, because the failure is in the answering.
Write full answers, not plans. The skill being trained is producing eight findable points in the time available. A plan does not train it.
Time them. A ten-mark question has a budget of around fifteen minutes on a typical paper. Writing a beautiful answer in twenty-five minutes means something else got nothing.
Count your points against the marks. After writing, count them. If you wrote four for an eight-mark question, that is the finding, and it is more useful than any content revision.
Get one marked by a teacher. One properly marked long answer teaches more than ten self-marked ones, because the gap between what you thought you wrote and what a marker saw is the whole problem. If nobody is available, ask an AI tutor what a marker would take marks off for - but ask that, not "improve this", because a rewritten answer teaches you nothing you can reproduce in the hall.
And practise the essay formats to the word count. Exceeding a stated limit loses marks however good the content is, and discovering your natural length in the exam is an avoidable mistake.
When the Question Is One You Did Not Prepare
It will happen, and there is a method rather than a panic.
Find the syllabus concept it is standing on. A question phrased around something unfamiliar is almost always testing a concept you have studied, approached from a new direction. Ask what topic it belongs to before deciding you cannot answer it. In content subjects, the number of concepts is finite and the number of phrasings is not.
Write the structure first, then fill it. Even with limited content, an answer with a direct opening line, three attempted points and a closing line scores meaningfully. A blank scores nothing and a wandering paragraph scores close to nothing.
Use what you do know from adjacent chapters. These subjects are connected in ways maths is not. A question on a policy you have not memorised can often be answered from what you know about how the system it belongs to works.
Reason from principles rather than guessing facts. If you cannot recall the specific provision, write what such a provision is for and what it would be expected to achieve. Partial credit for correct reasoning is real, and it is more likely than credit for an invented fact.
Do not leave content questions blank, ever. Unlike a numerical, where a wrong method earns nothing, a partially informed content answer usually earns something. The expected value of an honest attempt is always above zero.
A Four-Week Plan for Both Subjects
Week 1 - build the question sheets. For every chapter, one sheet: the chapter's own questions plus every past-paper question you can find on it, in the left column, blank on the right. This is a dull week and it is the infrastructure everything else runs on.
Week 2 - retrieval, a day apart. Work the sheets by writing answers from memory in points. Check, mark the gaps, and come back to the same sheet two days later rather than immediately. Add the timelines and maps drawn from memory.
Week 3 - answer structure. Memorise the templates. Then write six full timed long answers - two describe, two explain, two opinion - and count your points against the marks each time. Get at least one marked by a teacher.
Week 4 - the volume content and the writing formats. Chunk the long lists, build the mnemonics, walk the loci route for anything ordered. For Nepali, the format checklists memorised cold and one piece of each writing task produced to the word count.
Then keep the question sheets in rotation. Twenty minutes a day of closed-book answering, spread across weeks, is what distributed practice is for, and it is why these subjects reward a small daily habit far more than a long Sunday session.
Where the Marks Go Missing
The recurring losses, in order of cost.
Reading chapters instead of writing answers. The subject is examined by production and revised by consumption.
Too few points for the marks. Four points on an eight-mark question is the commonest single failure and it is arithmetic, not knowledge.
Points buried in prose. Written, present, not found, not awarded.
Retelling instead of analysing in literature questions.
Format marks in the writing section. The cheapest marks on the Nepali paper, thrown away by improvising a letter's layout.
Exceeding a stated word limit. A good summary over the limit loses marks that a shorter, weaker one keeps.
Ambiguous script in the third hour. In a language paper especially, a word the marker cannot resolve is a word that earns nothing.
The Short Version
Social Studies and Nepali are not easy subjects that you happen to under-revise. They are subjects where knowing the content is the entry requirement and the marks are decided by the answer - the number of findable points, the terminology, and the structure.
So stop reading chapters. Close the book and write from memory, a day later rather than an hour later, and count what you produced against what was there. Memorise a template for each question type so the hall is filling in a structure rather than inventing one.
Chunk long lists into groups, build a mnemonic for anything with a fixed order, and attach every date to a cause and a consequence so it has more than one route back.
Then write full timed answers on past papers and count your points against the marks. Four points on an eight-mark question is the most common reason a student who knew the chapter scored in the middle of the class.