Light

Angle of incidence and refraction are measured from the normal (perpendicular to the surface), never from the surface. A ray at 50° to the surface has $i = 90° - 50° = 40°$. Feeding the surface angle into $n = \sin i / \sin r$ gives a wrong $n$ — often below 1, impossible for glass. Fix: draw the dashed normal first, then measure to it.

Two traps. (1) Stating only one condition: full marks need BOTH — light goes from a denser to a less dense medium AND $i$ exceeds the critical angle $C$. (2) Believing a refracted ray survives past $C$: at TIR all light reflects back and there is no refracted ray. "The angle is large" alone scores partial marks.

The splitting of white light by a prism is dispersion. The common wrong answers are refraction (that is the mechanism, not the name) and diffraction (the spreading of waves through a gap or round an obstacle — a different effect). Learn the term: dispersion = white light separated into a spectrum by a prism.

"A measure of how much a material bends light" earns nothing. The mark scheme needs a two-element definition: $n = \dfrac{\sin i}{\sin r}$ — the ratio of $\sin$ of the angle of incidence (in air) to $\sin$ of the angle of refraction (in the medium). Also accepted: $n = c/v$ (speed in vacuum ÷ speed in the medium). Both forms give $n$ as a pure number, no unit.

A plane mirror image is upright, not upside-down. It is laterally inverted — left and right are swapped (which is why AMBULANCE is written reversed on the front of the vehicle). Writing "inverted" alone implies upside-down, which is wrong for a plane mirror. The image is also virtual, the same size, and as far behind the mirror as the object is in front.

A prism has non-parallel sides, so the emergent ray is not parallel to the incident ray and the colours disperse. A rectangular block has parallel sides: the ray bends at each surface but the emergent ray leaves parallel to the incident ray (just shifted sideways), with no net dispersion. Do not draw a prism's output like a block's.