2D shapes have measurable attributes: number of sides, angles, pairs of parallel sides, symmetry. A rectangle has 4 right angles, 2 pairs of parallel sides, and 2 lines of symmetry. Comparing shapes by attributes deepens geometric reasoning.
In 2nd grade you learned to name 2D shapes by sight — triangles, squares, rectangles, hexagons. Now you're going a level deeper: instead of just naming a shape, you're describing it by its attributes, which are measurable or definable features that a shape either has or doesn't have. This shift from naming to analyzing is the heart of geometric reasoning.
The key attributes to examine are: the number of sides (and their relative lengths), the number and type of angles (especially whether any are right angles, which are exactly 90°), whether any sides are parallel (lines that would never cross if extended), and whether the shape has lines of symmetry (fold lines where the two halves match). A triangle has 3 sides and 3 angles. A quadrilateral has 4 sides and 4 angles. But within quadrilaterals, the specific combination of attributes is what distinguishes a square from a rectangle from a rhombus from a trapezoid.
Consider rectangles and squares. Both have 4 right angles and 2 pairs of parallel sides. What makes a square special is that all 4 sides are equal in length. So a square is a special kind of rectangle — every square satisfies all the rectangle attributes plus one more. This is why in geometry, the categories overlap rather than being mutually exclusive: a square is always a rectangle, but a rectangle is not always a square. Thinking about attributes helps you see these relationships instead of treating every shape as a completely separate category.
Symmetry is also an attribute you can check systematically. A rectangle (non-square) has exactly 2 lines of symmetry — the horizontal midline and the vertical midline. A square has 4. An equilateral triangle has 3. A scalene triangle has 0. Checking symmetry by asking "could I fold this so both halves match?" gives you a precise test that works for any shape, familiar or unfamiliar. The ability to classify by attributes — not just name — is what lets you handle shapes you've never seen before.