first one is A i believe, pretty easy if you look at it.
It says 
"key outcome from teh study was that the mode of testimony did not have a meaningful impact on the jury outcomes. Nor was there any coinsistent impact on the jurors' perceptions or discisions from the victim's presentation style"If this was the key outcome, then it means that they were 
'investigating  whether the mode of victim testimony...." Hence A
- If this doesn't make sense to you, think of it this scenario that I made up:
" I conduct an experiment, in which i am investigating whether aggravated bees sting more on average than normal bees. 
- I seperate the aggravated and normal bees. Then, I put a dummy in each one for 10 minutes. 
- After 10 minutes, i take the dummies out and count the number of stings in each one. 
- 
The outcome of the experiment shows that the aggravated bees stung more on average, than the normal bees. "^ from this, If i took out the first part, about what i am investigating, could you not deduce from my outcome ("Aggravated bees stung more...") what I was investigating?  Sorry if i confused you more 

________________
2nd question can be deduced when u have the first question. As the end of the first question's answer states that "affects the verdict passed by a jury".
With that, you can immediately rule out B and D as they have practically nothing to do with the purpose of investigation. You are then left with A and C. 
To figure out the best answer from here, use the table. If it were "A", why would they provide us with the other crimes? And how can you conclude that 64% of verdicts are passed as guilty 'is too few' - that would be a personal opinion, right ? 
Therefore the best option is C.
_________________
3rd question, key here is the word 
least. Age can be a factor can't it? Is a 5 year old likely to respond in the same way as a 35 year old? Of course not, so A would obviously have an impact on results (And we can rule it out).
B - This would also have a big impact. If you have a group of 30 potential jurors, and select the most unsympathetic ones, then your result will differ obviously, - ruled out.
C - This confused me, I had to look at medentry's sample answer which was adaqueate. (kinda silly to get confused here, but it seems so obvious that this answer is stupid lol). 
D - Increasing participants would increase the accuracy, but i was consued as to why it would least effect the validity. (If it increases validity, it still has an effect on the validity of results, correct?) To be honest, i would have juggled between C and D.