The key to this argument is: The proportion of women among students enrolled in higher education programs.
That is, we are comparing the proportion of women to men at colleges. The evidence used, however, is the percent of 20 -21 yrs. old females attending college over two time periods, 1959 and 1981. Even if more 20-21 yrs. old females are attending college in proportion to the overall population of females, that fact alone does not speak to the proportion of women to men in college. To illustrate: let's say in 1959, 20% of those attending college were women. In 1981 that number could fall to only 10%. This is not inconsistent with the increase in attendance amongst 20-21 yrs. old females. While their ranks may have swelled, perhaps there has been a drop off in the percentage of 18-19 or 25 and over.
Thus to evaluate the argument, we need to know how many men were enrolled in college. Therefore the answer is (D).
Hope that helps clear any confusion

.
Thank you for your response, but I still dont get it. Let us use your illustration above, but this time let it be for men. in 1959, 80% of those attending college were men. in 1981 that number could have increased to 90% (as females reduced to 10%). That increase could be among men of ANY age group. It could still be possible that the NUMBER of men between 20 and 21 could well have reduced and the numbers in other age groups, say 25 above could have increased. (D) does not help us hence. Answer E is more general; it helps us evaluate the WHOLE pool of men as against one age group which could have either increased or decreased. Am i missing something here?