January 16, 2003 communication from John Logan to the David Vogel, Senior Editor, Journal Sentinel News Department:

Dear Dave,

Your newspaper has done good reporting on the census results, but recent stories on racial segregation by Bruce Murphy are an exception to this. In one story he falsely states that I told him I approve of the UMW study methodology. In fact, he never told me there was such a study, and I have never before heard of a methodology like theirs. I certainly don't think it is a reasonable approach to comparing metro areas around the country. Another story is specifically about my work. I have to object to his headline conclusion that the Mumford Center chooses to disregard segregation in regions with smaller shares of black residents. We published and provided comparisons among all metro areas in the U.S. It is only in selecting the "largest 50" for specific tables in a report that we omitted the other 280. And our choice to list the 50 with the largest black populations for a table is, in my opinion, much more defensible than if we had chosen to list the 50 with the largest total population. I told him this, and he disregarded it.

I spoke at length with Mr. Murphy. I directed him to web pages where we provide not only the index of dissimilarity, but also exposure and isolation indices, information on overall racial composition, information on neighborhood quality, etc. I told him specifically that I consider the most salient question is whether whites and blacks live in equal neighborhoods, not whether they are integrated into the same neighborhoods. I also told him that I consider it a mistake for researchers to rely on any single measure. Yet he chose to characterize our work as a one-dimensional reliance on the index of dissimilarity.

This is just plain false, and he knew it.

Mr. Murphy also asserts that the UWM study is groundbreaking in using census blocks rather than tracts as the unit of analysis. In fact, I pointed him to the place on our webpage where we have reproduced block-level indices calculated by the Associated Press. I told him, correctly, that these indices are generally a few points higher than tract-level indices, but that they generally rank places in the same way. (In the city of Milwaukee, the tract index is 69.0, the block index is 74.6.) In his story he asserts that they rank places very differently. But what he compares there is actually a tract-based Index of Dissimilarity with an entirely different block-based index created by UWM researchers. Perhaps he didn't notice this slip in his logic, but that's bad reporting, and my impression is that the text came out that way because he has an axe to grind.

Dave, I am writing to you because you are a senior editor in the news department, and you probably have enough of an overview to make a reasonable evaluation of these stories and my reaction to them. Yesterday I sent a very short Letter to the Editor about the series, not complaining about the reporting but simply stating the most basic facts about segregation in Milwaukee in a way that would tend to correct Mr. Murphy's conclusions. It wasn't published today, and I wonder if you would be kind enough to check into its fate.

I do feel that there is a need to correct the record in some way. If you think there is a more appropriate person for me to contact, please let me know.