Answer – 95 percent
As stipulated by standard convention, a researcher must have at least a 95 percent assurance that the difference which was observed by her between experimental conditions was not merely coincidental in order to justifiably conclude that the difference she observed has statistical significance.