– Stories on states and counties beginning to top their response rates from 2000, on hurdles to participation and  Census jobs going unfilled.
– Maps tracking participation and an interactive where you can search rates by state, county or locality.
– A neat interactive explaining how Congress is reapportioned after every Census.

More to come as Census workers fan out to households that didn’t respond. But, of course, this is nothing compared to the deluge we’ll face next spring when Census turns on the firehose of the actual data.

Through late April, the Census Bureau is updating the data each weekday. They’ve launched their own interactive map and offer the data in CSV or double-pipe-delimited format (a new one for us). We didn’t want to duplicate the bureau’s map, but we did want to offer something Census isn’t: the ability to quickly find and rank geographies.

Here’s more on how it came together:

Our database team is keen on moving more of our apps away from Flash, so I let this small-scale project serve as a learning exercise. While Paul focused on using SAS to parse the text file and convert geography names to AP style, I built a SQL Server stored procedure to extract the relevant geographies into a small table. (We keep our production-environment data sets as small as possible for best performance.)

The SQL was the easy part, as was a basic HTTP handler in ASP.NET/C# to pass and return calls from the page to the DB. Not much sweat.

Then the learning fun happened. I needed HTML/CSS for the table, search buttons, drop downs and wrapper. And I wanted to “appropriate” JavaScript my coworkers Brad Heath and Jack Gillum had written for similar pages.

The CSS was a bear. I thought I could just modify a presentation Jack had built around NCAA finances, but soon I realized the interactivity and display were different. And our design folks requested style changes. So, I ended up using about 50% of the code from Jack’s project and engineered the rest. The centerpiece is a table with Div elements for the bars displaying the percent of forms mailed back. When the handler draws the table HTML, the bar’s Div width in pixels gets set based on the value in the field. Jack suggested the idea, and it was a discovery for me.

While doing that, I also was enjoying new adventures in JavaScript. I’d never worked in it before, and the code looks a lot like C# but has enough syntax differences to keep me on my toes. After many visits to Google and sites like this, my buttons, drop downs and display worked fine. I ended up using a snippet of JQuery as well to set the default values in my drop downs.

In all, I spent about a day on the DB back end and handler plus a week on the HTML/CSS/JavaScript. It’s a small-scale presentation, but the learning benefits were more than worth the sweat.

Pew staff recorded the sessions and is sharing some of the material on its Census site. Here are some of my notes for those who couldn’t make it:

‘Conducting Census 2010′
Presented with the Washington Statistical Society and D.C.-American Association of Public Opinion Research

Robert Groves,  director, U.S. Census Bureau
Constance Citro, director, Committee on National Statistics
Jeffrey Passel, senior demographer, Pew Research Center
Joseph Salvo, director, Population Division, NYC Dept. of City Planning
Scott Keeter, director of survey research, Pew Research Center

Groves led with a broad description of the planning and operational aspects of the count:

• Forms will be delivered March 1 to April 6. Census day is April 1.
• The form is one of the shortest — 10 questions. There is no long form; that’s been replaced by the American Community Survey.
• Households will get an advance letter, a reminder card and a replacement questionnaire if they don’t return the first one.
• For its communications campaign, the Census Bureau “identified levers that could be pulled” by assessing response data from prior counts. It also built a tract-level database and mapped out churches, businesses and other organizations it could enlist for support.
• The first national Census ad debuted during the Golden Globes.
• In summer 2009, Census workers with GPS units marked 98% of addresses.
• The Census Bureau has prepared 120 million questionnaire packets. Groves said they “burned out” the country’s printing capacity.
• 13 million questionnaires are in Spanish. Forms are available in six languages — English, Spanish, Chinese, Korean, Vietnamese and Russian — and there are guides in 59 languages.
• The Bureau is doing daily research on the public’s awareness and attitudes toward the Census.
• And, finally, two Muppets characters — Count von Count and Rosita — will assist with outreach to kids, which the Bureau recognizes as important to raising awareness in immigrant households.

Salvo discussed New York City’s efforts to make sure all its addresses are represented:

• The Master Address File is the foundation of the Census.
• Current updates are done through the U.S. Postal Service, field efforts in rural areas, discovery via other surveys and a full canvass.
• New York City submitted 196,000 additions to the address list. The Census Bureau accepted 121,000, and the city submitted 36,000 for appeal.
• To illustrate the difficulty in identifying households, Salvo showed photos of mailboxes in building foyers pasted with multiple names.

Citro had some pointed comments about Census glitches and offered suggestions for planning:

• The bureau “had a harrowing 2008 after hand-held (GPS) problems blew up out of control.”
• The Census coverage measurement program “will not do as good a job as it could because the independent interviewing is starting at a late date, which increases the likelihood of errors in recalling Census day residence.” Some background is here.
• Planning now for Census 2020 is “crucial.”
• She recommended ways to cut costs, including greater use of administrative records, getting a contractor now to design hand-helds, and use of Internet for responses.

Passel’s portion was titled “Will Success Spoil the Census Bureau?” and focused on the intersection of the decennial counts and the annual ACS:

• Census 2000 was a success, with a very low undercount. Just 0.12% of the population was missed.
• Whether Census can match or improve on that is a question.
• He cautioned that the ACS needs to be regarded as “Census-like” data, not a complete count. Its population and race totals come from estimates, not counts.
• He recommended that Census weight ACS 2010 data to Census 2010, not Census 2000, and delay releasing it if necessary to do that.

‘Covering Census 2010′
Presented with Investigative Reporters & Editors

Ron Nixon, projects reporter, The New York Times
Paul Overberg, database editor, USA TODAY
D’Vera Cohn, senior writer, Pew Research Center

Nixon began with suggestions on topics reporters might want to explore:

• Cost: Census is spending $14.7 billion on the effort and in 2009 asked for $2.1 billion, he said.
• Some dress rehearsals were canceled for budget reasons
• Among the potential problem areas: foreclosed homes, fear over immigration status, continuing problems in rural areas in terms of identifying place names, scams, etc.
• Some volunteer organizations say they’re being vetted closely by Census because of the “Acorn effect.”
• Examine efforts of state and local governments.
• Examine advertising spending on local radio, TV stations.

Overberg focused on “What’s Different and What’s Not” about Census 2010 vs. 2000:

• Sampling: There’s no discussion this time of adjusting the count via sampling as happened in the run-up to Census 2000.
• There’s lots of discussion on the attention being paid to media. The Bureau is identifying the trust points and institutions in each neighborhood, noting why various people are more likely to respond and how it can appeal to the issues they care about.
• The form is short — one page per person and the shortest mail Census form since that started in 1970.
• Precision: The bureau used GPS units to map addresses.
• The recession: Big worry that foreclosures have doubled-up households and whether people will note that on questionnaires. Some may feel embarrassed that they’re living in their brother’s basement, for example.
• Targeting: Census has greater ability to target people in terms of media and message.
• Polyglot: Last time, the form was in one language. This time, it’s in six languages — English, Spanish, Chinese, Korean, Vietnamese and Russian.
• Power shifts: We can already predict which states will gain or lose seats in the House of Representatives. Iowa, Pennsylvania and New York will be among those losing a seat. Texas will gain three or four. California may not gain for the first time. In state houses, we may see upstate/downstate shifts in places such as New York and Virginia.

Cohn, Nixon and Overberg wrapped by noting sources for journalists:

• U.S. Government Accountability Office
• Inspector General reports
• Minutes of Census Advisory Committees meetings
• Office of Management and Budget reports
• The Census Project
• Census Bureau Director Robert Groves’ blog

Cohen said Pew or IRE would make the audio and PowerPoint presentations available at some point. Audio of Robert Groves’ presentation is already posted on Pew’s site.

But how did Solace fare against the rest of the Bond canon, which stretches back to 1963′s Dr. No? The answer depends on whether you adjust for inflation.

We all know that the price of a loaf of bread isn’t what it used to be. The cost of consumer goods tends to rise each year, except during downturns or various calamities. So, taking inflation (or deflation) into account is the only way to  meaningfully compare dollar amounts over time.

There are plenty of apps just for this. The Bureau of Labor Statistics offers one basic calculator, and there’s another at this site. They’re fine for a quick check, but I’d rather do my own calculations. A web app might not have the latest data. And if you’re adjusting more than a couple of amounts, using a spreadsheet will save time. Here’s an exercise from Bond-land:

Let’s compare the $168.4 million Quantum of Solace earned in 2008 to the $35.4 million earned by 1973′s Live and Let Die. In 2008 dollars, which movie earned more?

Finding out is fairly straightforward:

1. Get the index that measures the cost of goods and services. In the U.S., the most widely cited is the Consumer Price Index for All Urban Consumers, aka the CPI-U. The Bureau of Labor Statistics’ main page on the CPI is here, and you should spend some time on the FAQ. The index values are in this table, which goes back to 1913. It also has the annual average and the yearly percent change in both the annual average and the December values.

2. For this exercise, we’ll use the average CPI-U values for 1973 (44.4) and 2008 (215.303). Because the CPI-U includes relatively volatile items such as food and fuel, I tend to avoid the monthly values with their fluctuations. If you want to adjust to the most recent data available, consider averaging the most recent six months of CPI-U values.

3. Multiply the 1973 film’s gross times the 2008 CPI-U average. Divide that product by the 1973 CPI-U average. Here’s the setup:

2008_value = (1973_amount * 2008_CPI-U) / 1973_CPI-U

With the numbers plugged in, we have:

2008_value = ($35,377,836 * 215.303) / 44.4

Always remember that operations in parenthesis happen first. Do the math, and we get:

2008_value = $171,553,023

So, though Quantum of Solace’s $168.4 million was hefty, after adjusting for inflation it’s about even with what Live and Let Die’s original box office is worth today.

So, the 2008 version of the ACS was a good warm up. It saw the Census Bureau add questions on health insurance and the number of times a person had been married, among other tweaks. The release also came in the midst of the country’s economic woes, so journalists everywhere were on the lookout for how the data would reflect the recession.

We are fortunate at USA TODAY to have Paul Overberg on our team, one of the smartest Census analysts in the business. We began planning early and scrambled through a busy weekend to put together four stories, an interactive state map and assorted print graphics. Here’s what we did:

Immigrant population dipped last year, Census says
Driving habits alter during recession, Census reports
Housing is getting even less affordable
Census: 76% marry just once; new count for same-sex couples
State-by-state maps on foriegn born, other indicators