Grasple is unique in providing support for Linear Algebra.

In this guide you can find what is currently possible and what is not possible yet.

We will start by showing different questions and how you can enter those in Grasple.
The main things to keep in mind:

• You have to enter the correct answer

• There needs to be one unique solution to the question. If there are infinite possible solutions (ie: provide a basis), please contact support to see if we can support your case.

### The answer is a matrix

Examples of questions that ask for a matrix include: providing the reduced echelon form of a given matrix, provide the inverse of matrix or perform some matrix calculation. The answer in this case is a matrix

In Grasple, select the Math question type, click the squares in the math-bar, and fill in the correct answer. The answer given by the student will be compared with the correct answer as an exact match, meaning that all the numbers have to be the same and in the same position in the matrix.
This means that if the correct answer contains pi, entering 3.1415 will not be correct.

### The answer is a vector

We consider a vector to be a "n x 1" matrix, so this is the same as adding a matrix.
A column vector and row vector are two different answers. If you want both of them to be marked as correct, provide both of them as correct answer by adding multiple correct answers. Lean how to add extra answers for math exercises here.

When a question asks for say the determinant of a matrix, the answer is simply one number

Depending on the question, you can use the Numeric or Math type.

• Use Numeric type in case you want to allow numeric approximations. (ie: 2.892)

• Use Math type in case the number has to be exact (ie: 3/4 would be correct and 0.749 would not be correct)

Note

You can also add subquestions, meaning that you can first ask the learner to provide one of the eigenvalues and then one of the eigenvectors.

What is not possible for now is linking the subquestions, ie: provide one of the eigenvalues and then only marking correct the corresponding eigenvalue.