TO: Students in CJ 161-003 Spring 03
FROM: R. B. Taylor
DATE: 3/21/03
RE: Comments on paper 2

You may find on your paper various numbers - they refer to the numbered list below. The list represents some of the most common ways that papers went awry.

  1. When you say the median is the difference between the first and third quartile, you need to be more specific: it is the range, as you proceed from lower to higher scores,  between where the first quartile ends and where the third quartile ends, for example.
  2. When talking about the explanation for the gender difference, a stronger paper, rather than just listing a number of possible arguments, would try to just develop one.
  3. Your paper makes the presumption that the male and female respondents are reporting about the same households. Not so. In each household there is just one respondent. That may be a male or it may be a female. The males and females are reporting about DIFFERENT households, not the same ones. There is no implication that the men and women are married.
  4. The paper would have been stronger if you had been a little more detailed about the differences in the patterns of the extreme cases.
  5. Paper would have been stronger had you supplied a clear description of who the respondents were. Make it clear that we have restricted the group here to just gunowning households.
  6. The introduction makes reference to the three different surveys reported on by Ludwig et al, which is good, because it shows you did that reading. But the paper fails to clarify that the specific data here came from only one of those studies, the NSPOF.
  7. It appears there is some misunderstanding about the median.
  8. Paper needs to better explain the IQR.
  9. It is unwarranted to conclude that the gender differences reported link to a differential willingness to participate in the survey based on gender; the survey procedures control for differences in availability of various persons in the household.