I only

Evaluate the three statements. Statement I is true, since the smallest population for New York over the period was about 10 million, while the largest population for Wisconsin was less than half of that-about 4.6 million. Statement I must be a part of the answer; choices 2 and 3 can be eliminated.

Statement II is a little harder. The percent increase is the ratio of the amount of increase to the original whole. So there are two things we're interested in. Looking at the solid line for the two periods, we see that the amount of change between 1960 and 1970 was about the same as the amount of change between 1920 and 1930. What does that tell us about the percent increase? Well, the population in 1960 was much greater than the population in 1920, so the original whole in 1960 was greater than the original whole in 1920 . So we have the same approximate amount of increase for the two periods, but since we started at a smaller number in 1920 than in 1960, the percent increase for the earlier period must be greater than the percent increase for the later period. In other words, statement II is not true. At this point, we can eliminate everything except choices 1 and 5.

Evaluating statement III depends on understanding what the density figures mean. If we look at the change in the density figures, we can determine the largest increase in population per square mile for the period discussed. But does that translate to the largest population growth? Not necessarily. We need information about the areas of the states in order to answer the question. Texas, for instance, is a much larger state than New York, so it's possible that the smaller amount of increase per square mile in Texas would translate to a larger overall increase, because there are more square miles to be considered. So statement III is not inferable based on the information given. Statement I was the only acceptable inference.