The confidence interval is NOT "6.4% - 10.8% 19 times out of 20". That makes no sense. What is the definition of "Confidence Interval"? It's a terrible question. I can see why the author worded it that way, but no, just NO. Making it simpler or surreptitiously providing information should not require making it wrong. A fence that should not be crossed. If you are using a 95% confidence level, then the the Confidence Interval IS "6.4%-10.8%". It won't be different the next time you calculate it.