Yes, you have to use the propagation speed of the cable. Wikipedia lists the value for CAT5 cable: 5.3ns/m. In general, you have to know the exact type of cable used (i.e. CAT5, CAT6, CAT6A). Then you'll be able to find the propagation speed, which is generally some large fraction of the speed of light, but it can be significantly lower (values of 60% speed of light are quite typical for coax cables, for example). Wikipedia has an article about this.
The propagation speed for optical cables seems to generally be on the order of 5µs/km, according to m2optics.
If your 0.7ms value came from a CAT5 ethernet cable, that cable would be 132km long (which then of course means that it can't be CAT5 but must rather be the optical cable). Using the value for the optical cable, we'd get 140km instead.
However, you have to keep in mind that this is for the cable alone. It doesn't include the latency generated by the transmitter, receiver, and (if present) other network hops on the route. You also have to distinguish between round-trip-time and one-way latency.