10.4.3 Tensor Change Matrix Entry Meaning

Understanding how tensor change matrices transform entries between bases in multilinear algebra.

Tensor Change Matrix Entry Meaning is the collection of equivalent ways to interpret a single entry (A^j_{\ i}) of a change-of-basis matrix, covering its reading as an expansion coefficient, as a dual basis evaluation, as a partial derivative in the coordinate setting, and as a row-versus-column distinction within the matrix as a whole.

Reading an Entry by Row and by Column

Column Meaning: Which Target Vector

Reading down a fixed column (i) of (A) answers the question of how the single target basis vector (e'_i) is built from every source basis vector in turn, with each entry in that column supplying the coefficient of one particular source basis vector.

e_{i}' = A_{i}^{1} e_{1} + A_{i}^{2} e_{2} + \dots

Row Meaning: Contribution to Every Target Vector

Reading across a fixed row (j) of (A) answers the complementary question of how much the single source basis vector (e_j) contributes to each target basis vector in turn, showing the same set of numbers organized from the opposite perspective.

Entry Meaning as a Dual Basis Evaluation

Extracting an Entry Through the Dual Basis

Any single entry of (A) can be extracted directly by applying the source dual basis covector (e^j) to the target basis vector (e'_i), giving a compact, coordinate-free way to state what any individual entry means.

A_{i}^{j} = e^{j} (e_{i}')

This reading emphasizes that an entry of (A) is nothing more than an ordinary tensor component, specifically the (j)-th component of the vector (e'_i) relative to the source basis, obtained by the same dual basis extraction technique used for any vector.

Entry Meaning as a Partial Derivative

The Jacobian Interpretation

In the coordinate transformation setting, each entry of the transformation matrix is interpreted as a partial derivative, expressing the rate at which one new coordinate changes per unit change in one old coordinate, holding the other old coordinates fixed.

A_{i}^{j} = \frac{\partial x^{j}'}{\partial x^{i}}

This meaning generalizes the expansion-coefficient reading to the case of a position-dependent basis, where the entry is no longer a fixed number but a function that must be evaluated at the specific point of interest.

Entry Meaning as a Scaling Factor

Diagonal Entries as Direct Rescaling Factors

When the change of basis is purely a rescaling along fixed directions, with no mixing between them, the diagonal entries of (A) take on the simplified meaning of direct scaling factors, each one describing how much longer or shorter the corresponding basis direction has become.

A_{i}^{i} = λ_{i} (no summation, diagonal case)

Off-Diagonal Entries as Mixing Factors

Off-diagonal entries, by contrast, carry the meaning of mixing or shearing: a nonzero off-diagonal entry indicates that the target basis vector in that column has a component along a source basis direction other than its own principal one.

Visual Illustration

Why Multiple Equivalent Readings Are Useful

Having several equivalent ways to interpret a single matrix entry, as a coefficient, a dual basis value, a derivative, or a scaling factor, allows a practitioner to choose whichever reading is most natural for the problem at hand, whether working with abstract vector spaces, coordinate charts, or simple geometric rescalings. Recognizing that these readings are all descriptions of the same underlying number is what prevents them from being mistaken for competing or inconsistent definitions.