Stats help - uni project

Hi all, im currently at uni and need help doing some statistical analysis.
So what my project is primary based about is around animals, we score their weight on a scale of 1-9 (called body condition score) where 9 would be obese and 1 would be super skinny. My project hypothesises that owners dont know what their pets BCS is and once appropriate training has been given, they can more closely guess the correct BCS. The 'correct' body condition score is one thought of by the vet.

Because my hypothesis is based around being closer, its essentially their second guess is closer to the vets actual guess. I have collected all the data (refer to the graph) and have no way about statistically creating graphs and or why certain graphs are used. I am also going of the basis that say if an owner guessed 4 and after training they guessed 6 when the vet guessed 5 then they are no closer then the first time as they are still one out if that makes sense? I know my data is two categorical variables and i have been suggested using a t-test? Any help will be appreciated!

I also have graphpad prism for complicated calculations, but all i essentially want to show (on a graph) that post training is closer to actual BCS and pretty much want to know if it is significant (p<0.05) or not