PassProfessor’s Linear Algebra program is part of the expanding PassProfessor catalog. The program is organized chapter by chapter for chapter-based simulation exams matched to your textbook’s organization. Join the notify list and we will email you the day it opens — then use this guide to plan your studying.
This program is part of the expanding PassProfessor catalog — join the notify list and we’ll email you when enrollment opens.
This guide presents the standard scope of the undergraduate linear algebra course as general disciplinary background: solving systems of equations by elimination, matrix algebra and inverses, determinants, vector spaces and subspaces, span, independence, basis and dimension, rank, linear transformations and their matrices, eigenvalues, eigenvectors and diagonalization, and inner products with orthogonality and least squares. It explains who takes the course, how its chapters build on each other, how mathematics courses are honestly graded (your syllabus is the only authority), and a study method built for a course where computation and proof develop together. PassProfessor’s companion program organizes original practice chapter by chapter against this scope.
Linear Algebra, in one paragraph
Linear Algebra is the undergraduate course built around one problem — solving systems of linear equations — and the structures that problem generates: matrices and their algebra, determinants, vector spaces and subspaces, span, independence, basis and dimension, rank, linear transformations, eigenvalues and eigenvectors with diagonalization, and orthogonality with projection and least squares. It is the first course for many students where computation and proof carry equal weight, and its concepts are the working language of engineering, computer science, data analysis, and the later mathematics curriculum.
Who takes linear algebra, and why?
Linear algebra sits on more degree plans than almost any other mathematics course:
- Mathematics, engineering, and physics majors take it as a required core course — the first course in their programs built around structures and proofs as much as around calculation.
- Computer science and data science students take it because graphics, machine learning, optimization, and data analysis are all written in vectors and matrices; this course is where that notation becomes usable.
- Economics and quantitative social science students meet it where their programs turn mathematical — systems, optimization, and multivariate models all presuppose it.
- Students preparing for graduate and professional study take it as foundational preparation whose concepts recur in advanced coursework long after the course itself ends.
Whichever path you are on, the scope of the course itself is broadly the same — the discipline has a settled core, and undergraduate courses teach that core in a recognizable order. Instructors differ in emphasis, depth, and exam style; they rarely differ in what the subject is.
The course, piece by piece
The standard linear algebra course moves through five broad blocks. The table below describes each block’s content as general disciplinary knowledge — this is the background every student in the course is expected to build, whatever textbook their course assigns.
| Course block | What it covers |
|---|---|
| Systems and matrices | Solving linear systems by elimination, matrix algebra, inverses, and determinants. The computational engine room: every later definition is first met here as a procedure on matrices. |
| Vector spaces | Vector spaces and subspaces, span and independence, basis, dimension, and coordinates. The course’s abstract turn — the same structures, now described without coordinates. |
| Rank and transformations | Row, column, and null spaces, rank, and linear transformations with their matrices. Structure measured and mapped: what a matrix preserves, collapses, and reaches. |
| Eigenvalues | Eigenvalues, eigenvectors, eigenspaces, and diagonalization with its applications. The course’s second summit — finding the directions in which a transformation acts simply. |
| Orthogonality | Inner products, orthogonal sets, Gram-Schmidt, projection, least squares, and symmetric matrices in outline. Geometry repaid: distance and best approximation inside the algebra. |
Three structural facts about the course shape how you should study it:
- Computation and theory are the same subject here. Each procedure — elimination, expansion, Gram-Schmidt — arrives with a theorem stating when it works and what its output means. Exams grade both registers, often in one item: compute the object, then say what its value tells you about the matrix.
- Rank is the course’s recurring measurement. Solvability of systems, independence of lists, invertibility, dimension of solution sets, one-to-one and onto behavior — all are rank statements in different clothing. Students who learn to translate questions into rank questions find the second half of the course markedly simpler.
- Definitions are working tools, not vocabulary. Subspace, basis, and eigenvector are tests to run, not phrases to recite. The standard exam move is an object that almost satisfies a definition; the points go to students who check the failing condition precisely rather than argue from resemblance.
How linear algebra courses are graded — honestly
There is no official answer here, and any page that gives you one is inventing it. Linear Algebra has no administering body, no published exam format, no passing standard, and no retake policy, because it is not a licensing exam. Your instructor sets the number and format of chapter tests and unit exams, whether homework, projects, or participation are graded separately, and what the final covers. Your course syllabus is the only authority for all of that, and the first study act of the semester is reading it closely: note the exam dates, the chapters each exam covers, and whether the final is cumulative. What this guide can give you is the discipline’s standard scope (above), a chapter-by-chapter walk through the course (below), and a study method built for a cumulative, chapter-organized course — the things that transfer between instructors.
One more honesty note, because students ask: the components differ, but their logic does not. Homework and chapter quizzes reward steady contact with the material while it is current. Unit exams reward integration across a block of chapters. A cumulative final rewards retention — the willingness to keep early chapters retrievable after the course has moved on. A study plan that only ever serves the next test therefore mis-serves the final by design, which is why the planning section below treats retention as a standing task rather than an end-of-term emergency.
A note on editions, ordering, and local variation
No two sections of this course are identical, and the differences are worth naming precisely so you can discount them correctly. Textbooks divide the same scope differently: a topic that is one chapter in one book is two in another, and closing topics migrate between the end of this course and the start of the next in the sequence. Instructors reorder within that freedom — some front-load the chapters their later coursework needs most, some follow their book exactly. Assessment differs more than content does: one section’s midterm is another’s two unit exams, and project, participation, and homework weights are local decisions all the way down.
What does not vary is the discipline’s core: the scope summarized in this guide, the order in which its ideas depend on each other, and the points where students predictably lose marks. Those are properties of the subject, not of any syllabus. Plan against your own syllabus for dates and weights; plan against this guide for scope, sequence, and trouble spots — and when the two seem to disagree, your syllabus governs your grade while the guide still describes your subject.
How to study for linear algebra
- *Read the syllabus like a contract.* Note exam dates, the chapters each exam covers, homework and proof weight, and whether the final is cumulative — in this course the final is cumulative in effect even when not in name, because the theory chapters restate the computational ones.
- *Compute by hand until the mechanics disappear.* Row reduction, inverses, characteristic polynomials: work them slowly and in full until errors stop. Procedural fluency frees exam time for the theory items that actually distinguish grades.
- *Keep a theorem notebook.* One entry per major result: the statement, the hypotheses, one example, and one near-miss that fails a hypothesis. The invertible matrix theorem and the diagonalization test are learned this way, not by rereading.
- *Translate every new concept back to systems.* Ask of each abstraction what it says about solving equations — span about reachability, null space about free variables, rank about information. The translation habit keeps the abstract chapters anchored to the course’s opening problem.
- *Drill your weakest chapter, not your favorite.* Chapter results expose the leak — commonly rank and nullity or the basis definitions. Theory gaps compound in this course, so the weakest chapter is always the highest-value hour.
- *Simulate before each graded test.* Sit a timed set over exactly the chapters the test covers, mixing computation and short theory items the way your instructor’s exams mix them. Pacing across both registers is a trainable skill.
Planning the semester, block by block
A chapter-organized course has a rhythm, and planning in its rhythm beats planning against the calendar alone. The course moves through five broad blocks, from systems of linear equations and Gaussian elimination to symmetric matrices and quadratic forms, in outline, and each block changes what good studying looks like.
In the opening block, the work is foundation: new vocabulary, new notation or procedure, and the first habits of the discipline. Study time here should be weighted toward accuracy over speed — slow, checked work that builds the representations everything else will stand on. Errors made carelessly in the first block are the most expensive in the course, because later chapters inherit them silently.
In the middle blocks, the work is integration. Chapters begin to cite each other: techniques combine, distinctions multiply, and exam items stop announcing which chapter they belong to. This is where mixed practice earns its keep — sets that draw from several chapters at once, so the skill being trained is recognizing which tool a problem wants, not merely executing the tool of the week. It is also where most students discover their real weak chapter, usually an early one whose gap only becomes visible when a later chapter spends it.
In the closing block, the work is consolidation under time. The material is mostly familiar; the challenge is retrieval speed and endurance across the full scope. Timed, cumulative practice is the right instrument here, and the correct measure of readiness is a cold solve — a problem answered correctly days after its chapter was last reviewed, without a warm-up set first.
Across all three phases, one habit separates the students who finish strong: a standing review rotation. A short weekly pass over earlier chapters — a handful of items from each, answered cold — keeps the early scope retrievable at trivial cost, and converts the final from a re-learning project into a review. The rotation is deliberately boring. That is what makes it work: it does not depend on motivation, deadlines, or panic, and it compounds quietly from the second week of the term onward.
The chapters, one by one
Systems of Linear Equations and Gaussian Elimination. The course opens with its central problem: solving several equations in several unknowns. Augmented matrices, row operations, echelon and reduced echelon forms, and the language of consistency — no solution, exactly one, or infinitely many — set up every method and definition that follows in the semester.
Matrices and Matrix Operations. Matrices as objects with their own algebra: addition, scalar multiplication, matrix multiplication and its noncommutative character, transposes, and partitioned forms in outline. Fluency here is mechanical and non-negotiable, because transformations, systems, and data tables all get written in this notation from now on.
Matrix Inverses and Elementary Matrices. Invertibility defined, tested, and computed: the inverse algorithm by row reduction, elementary matrices as single row operations in matrix form, and the invertible matrix theorem gathering the course’s equivalent conditions into one statement students will cite for the rest of the term.
Determinants and Their Properties. The determinant as a single number summarizing a square matrix: cofactor expansion, row-operation effects, the product rule, and invertibility detected by a nonzero value. Computation matters, but the graded skill is using determinant properties to reason before expanding anything by hand.
Vector Spaces and Subspaces. The course turns abstract: vector space axioms, subspaces and their tests, and examples beyond arrows — polynomials, functions, and matrices themselves. Students learn to verify structure rather than compute answers, and this chapter’s proof style governs how the remaining theory chapters are examined.
Span and Linear Independence. Two questions about any list of vectors: what can it build, and is any member redundant? Spanning sets, independence tests, and the dependence relation connect back to solution structure — free variables and dependence turn out to be the same phenomenon described twice.
Basis, Dimension, and Coordinates. A basis is an efficient spanning set, dimension counts it, and coordinates translate vectors between bases. The chapter converts earlier tests into a measurement system for spaces, and its representation ideas prepare the matrix of a transformation two chapters later.
Row, Column, and Null Spaces and Rank. Four spaces attached to a matrix, three fundamental ones in most courses: row, column, and null spaces, with rank and nullity summing to the number of columns. This is the course’s bookkeeping summit — nearly every earlier concept is counted, compared, or located here.
Linear Transformations and Their Matrices. Functions between vector spaces that preserve the operations: kernel and range, one-to-one and onto in rank language, and the matrix of a transformation relative to chosen bases. Composition becomes multiplication, and geometry — projection, reflection, rotation, shear — gives the algebra visible content.
Eigenvalues and Eigenvectors. The vectors a transformation merely stretches: eigenvalues, eigenvectors, and eigenspaces, found through the characteristic equation. The chapter reframes earlier skills — determinants, null spaces, polynomial factoring — as one search, and it motivates everything the course does with diagonal forms afterward.
Diagonalization and Its Applications. When a matrix has enough independent eigenvectors it becomes diagonal in the right basis, and powers, exponentials, and dynamical systems turn easy. Similar matrices, the diagonalization test, and defective cases mark both the payoff and the honest limits of the eigenvalue method.
Inner Products and Orthogonality. Geometry enters the algebra: inner products, length, distance, and angle in general spaces, orthogonal and orthonormal sets, and orthogonal complements. Orthogonality converts messy projection problems into independent one-dimensional ones, which is why the next chapter’s constructions work as cleanly as they do.
Gram-Schmidt, Orthogonal Projection, and Least Squares. Building orthonormal bases from arbitrary ones, projecting vectors onto subspaces, and solving impossible systems as well as possible: least-squares solutions and the normal equations. Data fitting gives this chapter its applied weight — the best approximation is a projection, stated in the course’s own language.
Symmetric Matrices and Quadratic Forms in Outline. In outline in most courses: symmetric matrices diagonalize orthogonally, their eigenvalues are real, and quadratic forms classify by definiteness with applications to optimization and conic sections. The chapter closes the course by joining its two threads — eigenvalues and orthogonality — in a single theorem.
Where students lose points
Row reduction errors carried silently forward
One arithmetic slip in an early pivot corrupts the rank, the solution set, and every conclusion drawn from them. The repair is checking discipline: verify a candidate solution in the original system, and sanity-check rank against the number of unknowns before interpreting anything.
Definitions recited instead of applied
Students state what a subspace is and then accept a set that fails closure because it looks like a plane. Apply definitions as checklists — zero vector present, closure under addition, closure under scaling — in that order, every time, until the checklist is reflex.
Matrix multiplication treated as commutative
Products are composed in the wrong order, inverses are cancelled from the wrong side, and transformation compositions come out reversed. Read products right to left as function composition, and never cancel a matrix — multiply by an inverse, on the correct side, or not at all.
Eigenvalue computation without eigenspace discipline
Characteristic polynomials get factored and the work stops, as if eigenvalues alone answer the question. Each eigenvalue must be followed into its null space: the eigenvectors, their independence, and their count are what diagonalization and most exam rubrics actually grade.
Frequently asked questions
What is linear algebra actually about? Systems of linear equations and the structures built to understand them: vectors, matrices, spaces, and transformations. Its recurring questions are whether a system can be solved, what all its solutions look like, and in which directions a transformation acts simply — asked first computationally, then in general.
Is linear algebra hard? Its computation is modest; its difficulty is conceptual layering. Definitions stack — basis needs independence, diagonalization needs eigenspaces — and the course moves between calculating objects and proving facts about them. Students who keep a theorem notebook and compute regularly find it demanding but orderly.
How much proof does the course involve? It varies by institution and section: most undergraduate courses mix computation with definition-based reasoning and short proofs, and many students meet genuine proof expectations here for the first time. Your syllabus and your instructor’s past-exam style, where shared, are the honest guide.
Why do computer science students take linear algebra? Because the field’s core objects are linear-algebraic: graphics transforms are matrices, data sets are matrices, recommendation and learning systems diagonalize and project them, and search ranking began as an eigenvector problem. The course supplies the working notation for all of it.
What is a vector space, in plain terms? Any collection of objects that can be added and scaled while obeying the familiar arithmetic rules — arrows, lists of numbers, polynomials, functions, matrices. The abstraction pays because one theorem proved for all such collections replaces separate arguments for each.
What is the difference between a basis and a spanning set? A spanning set reaches everything but may carry redundancy; a basis spans with no redundancy — every vector has exactly one coordinate description. Dimension is then well defined: every basis of the same space has the same number of elements.
What are eigenvalues used for? They find the directions a transformation preserves, which simplifies powers of matrices, systems of differential equations, vibration and stability analysis, data compression by projection, and ranking problems. Diagonalization — the chapter built on them — is the course’s most applied theorem.
Is the determinant still important? Yes, though modern courses demote its computation: the determinant survives as a test — a matrix is invertible exactly when its determinant is nonzero — and as the engine inside the characteristic equation that produces eigenvalues. Properties matter more than large expansions.
How is this course graded? By your instructor’s syllabus: typically chapter tests or midterms mixing computation and theory, homework that often carries real weight, and a cumulative final. There is no external standard or passing score — the syllabus is the only authority, and this guide invents none.
How should I study for linear algebra? In both registers at once: hand computation until procedures are automatic, and a theorem notebook that pairs every result with an example and a near-miss. Then timed, mixed practice before each test, because exams interleave the registers without warning.
What comes after linear algebra? Differential equations spends its systems chapters; numerical analysis, graphics, optimization, machine learning, and abstract algebra each build on a different block. The course is a junction — which is why so many degree plans route through it.
Do I need a computer or software for the course? Some sections use computational tools for large examples; others stay entirely by hand at small sizes where structure is visible. Whether tools are used — and whether they are permitted on exams — is your instructor’s decision, stated in the syllabus.
What is least squares doing in an algebra course? It is orthogonal projection applied to data: a system with no exact solution gets the closest solution instead, found by projecting onto the column space. It is the course’s clearest bridge from pure structure to statistical practice.
Using this guide with the practice program
This guide and the companion Linear Algebra practice program are built to be used together, in a specific order. The guide comes first: read the block map and the chapter walk for the part of the course you are entering, so the scope and the trouble spots are familiar before you practice against them. Forewarned is a real effect in a cumulative course — knowing where students lose points in a chapter changes what you notice while your instructor teaches it.
Then practice in the program’s own rhythm. Drill the chapter you are currently in while it is current, in short sets, and let the chapter-level results tell you the truth about the chapter while repair is still cheap. When a graded test approaches, switch instruments: a timed simulation over exactly the chapters the test covers, taken in one sitting, under the same constraints your syllabus describes. Afterward, return to this guide’s trouble-spot section for the chapters the simulation exposed — the diagnosis there is written for exactly that moment.
Before midterms and finals, widen the lens. Work the integrated material that combines chapters, because cumulative exams are written that way, and re-read the structural facts at the top of this guide: they describe how the course’s parts connect, which is what a final actually examines. If a chapter’s results stay weak after honest drilling, treat that as information about sequence, not effort — in a course built like this one, a stubborn chapter is usually standing on a cracked earlier one, and the repair belongs there.
A last word on what neither this guide nor the program can do. Neither can read your syllabus, attend your lectures, or know your instructor’s emphases — the local facts that decide real grades. Use both as they are designed to be used: as the discipline-level map and rehearsal space that make your own course’s local facts easier to master, week by week, while the semester is still young enough for the work to compound.
Disclaimer
PassProfessor is an independent exam-preparation and study provider, based in Salt Lake City, UT and operating entirely online. This guide is general course background and study guidance only: it is not affiliated with, endorsed by, or sponsored by any textbook publisher, textbook author, college, university, or instructor, and it reproduces no textbook content. Your course syllabus is the authority for your course’s schedule, formats, and grading.