In a Data for Progress poll of 822 likely voters in Iowa, 45% said they planned to vote for incumbent Republican senator Joni Ernst in Iowa's Senate race, and 44% said they planned to vote for her Democratic challenger, Theresa Greenfield. This kind of result is difficult to characterize because two things are true: (a) more people said they planned to vote for Ernst than Greenfield, but (b) based on the amount of sampling variation you'd expect in 822-person samples, it's very plausible that in the larger population the percentage of voters who plan to vote for Greenfield is larger. Because more people said they planned to vote for Ernst, some people characterize this as "Ernst leads by 1 point"; because the difference is within the margin of error and so it's plausible that either candidate's share of thepopulationvote could be larger, other people characterize this as a "statistical tie."

1. Which characterization do you think is more accurate, "Ernst leads by 1 point" or "statistical tie"?
2. Explain why.