Dear CTPP-News people: Some friends and I had a dinner party yesterday, and 3 of 5 received the Census 2000 long form. Of course, I didn't get one, so I have "long form envy". Luckily for our region the press is picking up good stories about "long form envy" instead of the libertarian cr*p about invasion of privacy. And, yes, my friends had concerns about the level of research (in hours?!) needed to provide accurate answers about utility costs.... In Ron Tweedie's example, it may prove useful if the Census Bureau in coordination with the State Data Center could provide the local newspaper editor the reasons and rationale for collecting this long form data. If congestion and poverty are issues in your metropolitan area, then household income is an absolutely critical data need. And I do believe there are Census Bureau documents citing the chapter and verse of the US Code where each census data item is required. (It's somewhere on the Bureau's WWW site.) My concern right now is the interpretation by census respondents to the term "last week" (as used in our journey-to-work questions.) We have traditionally (incorrectly?) assumed that the term "last week" refers to the "census reference week" or the last week of March in 1970, 1980, 1990 and 2000. Well, the Census Bureau mailed forms to all (or most) americans the week of March 13-17. So, is "last week" to be interpreted as the week before they received the form (March 6-10); the week before they fill out the form (either March 13-17, or this week -- March 20-24, or next week -- March 27-31)!!!! The answer is, probably yes, to all of the above. Yikes! People are (and should be?) mailing in their long forms *RIGHT NOW* and this "last week" ambiguity will be interesting. My thinking is that people will substitute an "unusual last week" with a "usual last week" if they want to provide "relevant" data to the Census Bureau. In my own situation, I was in Phoenix last week on vacation. This week I'm at home in Oakland. If I had a census long form, my inclination would be to delay answering the census until *next* week, when my "last week" commute pattern is more typical of my regular and usual commute pattern. My friend showed me her long form packet, and there were no apparent instructions to clarify what is "last week" or even *WHEN AM I SUPPOSED TO MAIL THIS BACK?* Given this ambiguity in terms of what is "last week," my recommendation to the USDOT, Census Bureau, State DOTs and MPOs is to start compilations of events of interest that may influence journey-to-work patterns in *all* of March 2000: 1. Weather conditions (daily clippings from the newspapers, and should report on natural events including quakes, floods, snowstorms, precipitation levels, sunny days, etc.); 2. Labor conditions (transit union strikes or slowdowns; other major labor events); 3. Gas Prices (we're now at $1.90 per gallon in the Bay Area, and rising! Also, surveys of other transportation costs such as tolls, fares, parking costs would be ideal to conduct this month and next.); 4. Road and Transit Network Status (important if new or expanded facilities are opened in early 2000). That's about it. Anybody else have ideas on things we should be monitoring for understanding "last week"? cheers, Chuck Purvis, MTC