I wouldn’t be at all surprised if it’s right though. The population split is pretty equal in terms of Protestant to Catholic and we all know what Catholics want. Then you have to consider the amount of people from all faiths who depend on the south for work whether that be that they work in the south or their business/the business they work for depends on the south. To some people simply being able to pay their mortgage might be enough to sway their vote. It’s finely balanced in the north. Regardless of the source of the story I wouldn’t be so quick to dismiss it.
And it's not the source I'm questioning Tony - the question is bad and the sample size is so small as to make the results irrelevent. I know a little about surveys from my degree. Let me try to explain.
1) If you want to draw conclusions you need to ensure that your question is clear, has no room for noise and doesn't lead the responder to answer in a specific way. The question fails that test. If asks if people want to lose the Good Friday agreement and citizens' rights as well - you cannot conclude from the answers that most want to join the republic.
2) In order to draw conclusions, you need a sample size big enough with respect to the population such that your hypothesis can be supported with any confidence. I won't go into the details of the calculation, but there is a mathematical formula to assess the confidence you can have in the result (confidence is measured as a percentage).
Let me give you an example. If I give you a bag with 100 balls inside and you draw 3 at random and they are all black, how confident would you feel saying that most of the balls in the bag are black? Now let's use the proportions from this survey: 2000 people in a population of 1.8 million. That's the equivalent of giving you a bag with 1,000 balls and you drawing one (that's 1!) and see it is black. Even if you were confident drawing 3 from 100 (and I'm not) - surely you can see that the result is irrelevant on that basis?