Getting Access to Strang's Linear Algebra Course Materials
Gilbert Strang's MIT 18.06 course is one of the most widely studied linear algebra resources available online. The textbook Introduction to Linear Algebra pairs with free video lectures on MIT OpenCourseWare. Students and self-learners regularly search for Introduction To Linear Algebra Strang Solutions to check their work on homework problems. Here is how the materials actually work and where to find them. The course page sits at mit.edu/ocw and contains the full set of lecture videos, notes, assignments, and exams. The textbook is published by Wellesley-Cambridge Press and covers everything from matrix multiplication to singular value decomposition. Problem sets run roughly five per chapter with solutions scattered across different years of the course schedule. Strang publishes two types of answer material. The first is the back-of-the-book answers, which give final results for selected odd-numbered problems. The second is the detailed solution manuals for instructor use, which walk through every step of even the most tedious row reduction sequences. The student-facing document is usually titled something like "Solutions to Homework Problems" and drops onto the OCW page each semester.
One thing beginners consistently miss is that the OCW solutions are not always aligned with the current edition of the textbook. I ran into this in 2019 when I was working through Chapter 3 on subspaces and the posted solution referenced equation numbers from the third edition while I was using the fifth. The math was identical but the problem numbering shifted by about twenty questions between editions. I just worked through the first problem in each section manually to map the correspondence, which took me about ten minutes and saved me from copying the wrong answer set.
Numerical Precision and Row Reduction
When you are checking your Gaussian elimination work against Strang's solutions, the decimal rounding in the manual can be misleading. The solutions sometimes show intermediate values rounded to two or three decimal places while the final answer retains more precision. If your calculator output differs in the third decimal place, do not immediately assume your method is wrong. I once spent an hour convinced I had made an arithmetic error in a 4x4 elimination until I traced back through Strang's own intermediate steps and found he had rounded 0.33333 to 0.33 halfway through a column operation. The propagated error made my answer look wrong when it was not. A practical workaround is to use exact fractions during your own row reduction whenever the numbers stay manageable. Once you reach reduced row echelon form, convert to decimals only at the end. This also makes it easier to spot when Strang has chosen a particular pivot strategy that differs from yours. His solutions sometimes use a different row swap order than what you would naturally pick, which changes the appearance of intermediate matrices but not the final subspace answers.
Get the Full Details
Common Pitfalls When Using These Solutions
The biggest issue people run into is that Strang's approach emphasizes conceptual understanding over mechanical computation. His solutions often explain why a pivot position matters or what a free variable represents rather than showing a grind-through of arithmetic. If you are studying for a course that grades heavily on showing row reduction steps, reading his solution directly may not match what your professor expects. I had a student once who turned in Strang's elegant explanation of column space as a span of pivot columns and got half credit because the grader wanted to see the actual elimination tableau written out. Another issue is that some online sources host unofficial solution PDFs with errors. The MIT OCW page is the only authoritative source. Third-party sites sometimes mislabel solutions for the wrong chapter or carry typographical errors that propagate through copy-paste study sessions. I once saw a solution site list the eigenvalues of a symmetric matrix as complex numbers, which is structurally impossible and should have been an immediate red flag.
What the Course Covers and What It Skips
The syllabus moves from basic matrix operations through determinants, eigenvalues, positive definite matrices, and Fourier transforms. It does not cover computational linear algebra tools like numpy or MATLAB in any depth. If you need code implementations alongside the theory, you will need to supplement with a separate resource. The course also assumes comfort with high school algebra and basic calculus. People who struggle with summation notation or function composition tend to hit a wall around Chapter 4. The lecture videos are approximately one hour each and correspond to single class periods. They work best when you watch them after attempting the problem set questions at least once. Strang repeats key ideas across lectures in ways that reinforce understanding but also mean watching a video passively without doing the problems is largely a waste of time. The material clicks when you are pencil-in-hand and the solution manual becomes a verification tool rather than a substitute for effort. The most useful feature is probably the pace. He deliberately slows down on topics like null spaces and rank that students routinely misunderstand. If you find yourself stuck on why the dimension of the null space equals the number of free variables, the relevant lectures contain multiple passes through the same argument from different angles. That redundancy is intentional and pays off.
Accessing the Textbook
The textbook is available through Wellesley-Cambridge Press at wellesleycambridge.com and through major booksellers. Strang also posts draft chapters on his personal website at math.mit.edu/~gs. The online drafts match the current print edition closely but may include errata corrections that have not yet made it into bookstores. If you are reading digitally, the drafts are the version to use. Beyond the official materials, there are a handful of resources worth knowing about. The YouTube channel "MIT OpenCourseWare" hosts the full video archive. There is also a companion book called Linear Algebra and Its Applications by Strang that contains more applied examples and additional practice problems. For students who want computational exposure, the free textbook Elementary Linear Algebra by Keith Conrad provides proofs that fill gaps Strang intentionally leaves open. I have also found the problem collections from other MIT courses, particularly 18.02 and 18.03, useful for extra practice on related topics. The boundary between linear algebra and differential equations blurs quickly in Strang's later chapters, and additional problems from 18.03 on eigenvector methods for systems of ODEs help cement the connection.
