Sometimes I enjoy seeing architecture and other human artifacts from the river's vantage point as much as I like experiencing nature. Garbage and pollution are major exceptions.

The Mississippi River offers a great mix of nature and civilization as it flows through Minneapolis toward Saint Paul: bald eagles flying near skyscrapers, blue herons fishing from the dams, geese wading under bridges, and trees clinging to public beaches.

My friend, Chris Desjardins, a veteran of the Mississippi River Challenge, guided us down the urban route from Boom Island Park to Hidden Falls Park. The photo at right shows us paddling past the Education Sciences Building where our offices are located.

We hoped to make it to the backwaters near the Minnesota River confluence, but it took too long to navigate the locks with so many boats on the river for Labor Day. I plan on putting in at Hidden Falls for my next urban trip.

We camped the night before at Jay Cooke State Park, dropped our canoe off at the put-in on state highway 23 in Minnesota, drove our car to the take-out on county highway W in Wisconsin, and then took a taxi back to the put-in.

The river was running at about 300 cubic feet per second (cfs), which is about the minimum for an enjoyable trip and only occurs after significant rainfall. The Nemadji carries a large amount of clay sediment, making the water red and obscuring "widow-maker" rocks just below the surface.

We saw no signs of human life along the river, except for few deer stands and a couple pieces of trash. We did, however, see several mergansers, two blue herons, a hawk, and several beaver chews.

I highly recommend a trip down the north fork of the Nemadji, if you can time the river level and are willing to arrange a shuttle.

I wrote to Laura, and she graciously reviewed my results. She attributed the standard error discrepancies to a change in IRT score variances. Specifically, the ECLS-K math and reading scores have been re-scaled since the original data set was published, resulting in larger variances in the currently published version. She supplied me with her version of the original public-use data set.

The updated results with the original data set are nearly identical to her published results. I feel confident now about using ECLS-B jackknife replicate weights for SEM in , and I learned a lot from attempting the replication.

The ECLS-B data and other large microdata sets represent the population, offer good statistical power, and provide comprehensive measures, making them suitable for structural equation modeling. However, those advantages are often achieved through stratified cluster sampling, which nests participants within primary sampling units in order to ensure adequate representation of strata and hold down data collection costs. Moreover, individuals representing small groups in the population are oversampled, which requires analytically re-weighting those cases downward to reflect population proportions but not down-weighting sample sizes in the standard error calculations. Calculating standard errors under complex sampling conditions is not straightforward compared to simple random sampling.

Is it possible to fit structural equation models of complex sample data in ? Several statistical software programs, including the survey package, can perform standard analyses (e.g., means, generalized linear models) in a manner appropriate for complex sample data. However, hardly any programs offer the ability to fit structural equation models to such data. Using some guidance offered by John Fox, author of the sem package, and an excellent article by Laura Stapleton, I decided to give it a try with R.

Stapleton used data from the ECLS Kindergarten Cohort and two commercial statistical software packages to demonstrate a structural equation model that applies sampling weights and accounts for multistage sampling. Because the ECLS-K is publicly available and is free, I was able to attempt her jackknife example. As hoped, the replication yielded parameter estimates that were comparable to Stapelton's, as well as standard errors that were larger than the naive standard errors. However, my jackknife standard errors were consistently larger than Stapleton's. I don't yet know why they were so large, but it will be good practice for me to find out. It will also be good practice to replicate her example of bootstrapping standard errors. I welcome any feedback about this approach.

School level results go a long way towards providing local-level information, but what about early learning outcomes in between school locations, where children live? As shown in the map at right, school-level results can be used to interpolate outcomes between schools. Interpolation is the process of using known values at a set of locations to estimate unknown values at different locations. When estimating an unknown value, nearby known values are typically given more weight (i.e., made to be more influential) than distant known values. Interpolation may be thought of as a three-dimensional version of locally weighted scatterplot smoothing (LOWESS). LOWESS yields a line, while spatial interpolation yields a surface, both lacking an explicit functional form.

Click on the image to see a larger map created with the help of Finley and Banerjee's Multilevel B-spline Approximation (MBA) package.

Microsoft's support page attributes the problem to corrupt registry files. Their solution requires a Windows XP startup disk, but Lenovo does not provide such a CD when you buy one of their computers. Lenovo instead supplies its Rescue and Recovery software, which allows the owner to restore the computer to its factory state or to a back up point if the owner has taken initiative to archive their operating system configuration. I regularly archive my files onto DVDs, but I had not archived my system. I didn't want to restore my computer to its factory state because re-installing all of my software would take a very long time.

I successfully recovered my computer without the Windows XP startup disk, so there's hope for those of you with running XP on a Lenovo (or another brand that provides a pre-installed rescue option in lieu of a startup CD). Here's how, with a second, functioning computer nearby:

• I read Microsoft's poorly written blue screen of death support page. The Guided Help option was no help at all... it kept stalling out on the second computer.
• I read this great entry posted by subtle on http://www.geekstogo.com/.
• I installed Avira NTFS4DOS Personal on the second computer and used it to create a DOS boot disk. (Note that if you do not have Lenovo Rescue and Recovery, then you can use the boot disk and DOS commands to follow the steps below, but it may be more difficult because the file names will appear truncated in DOS.)
• I used Lenovo Rescue and Recovery to copy the system, software, sam, security, and default files from c:\windows\repair\ to a USB flash drive.
• I also copied recent registry snapshots to the flash drive. Those files are located in c:\system volume information\_restore{***}\RP##\snapshot\, where *** is a bunch of seemingly random letters and numbers and ## are numbers corresponding to the order of recent snapshots made. In DOS, the location will appear as c:\system~1\_restor~1\rp##\snapshot\. The registry files corresponding with system, software, sam, security, and default are named _REGISTRY_MACHINE_SYSTEM, _REGISTRY_USER_SYSTEM, _REGISTRY_MACHINE_SAM, _REGISTRY_MACHINE_SECURITY, and _REGISTRY_USER_.DEFAULT. The DOS dir command will list the dates when the registry files were last saved, which can help you decide which snapshot number to use (i.e., how many days back to revert).
• I used DOS commands to overwrite the files named system, software, sam, security, and default in c:\windows\system32\config\ with those from c:\windows\repair\. (You may want to back up the config files first.) Then I re-started the computer and performed repairs with chkdsk and Rescue and Recovery, which resulted in another blue screen of death failure:
  System error: Lsass.exe
When trying to update a password the return status indicates that the value provided as the current password is not correct.
  I was happy to see that message because it occured much later in the boot up process than the original failure. In other words, I could tell I was making progress.
• I renamed the snapshot registry files on the flash drive from _REGISTRY_MACHINE_SYSTEM and so forth to their respective config names: system, software, sam, security, and default.
• Returning to DOS, I overwrote (again) those files located in c:\windows\system32\config\ with the just-renamed registry snapshots, thereby restoring the computer's registry files to a point from a few days earlier.

It took many, many hours, but I was able to recover my computer using the steps above. I'm thankful for the DOS I learned in high school and for the tips and shareware provided freely on the internet. I could have avoided a lot of headache if I had just used Rescue and Recovery to perodically archive my system configuration. I will from now on.

I paddled the Apostle Islands National Lakeshore last week with my good friends, Darren and Jesse. What an amazing place! Several local paddlers recommended that I visit the Apostle Islands, but I had no idea how incredible it would be.

On the first morning we rented sea kayaks and took a safety course from a local outfitter. After a shuttle to to Little Sandy Bay, we paddled out into Lake Superior. Paddling on the largest lake in the world (by surface area) was intimidating at first. Thankfully, we had low winds and sunny weather to ease our fears.

Our first destination was Sand Island, which is notable for its sea caves, lighthouse, and old-growth trees. The calm weather allowed us to spend a lot of time exploring the caves. The lighthouse was picturesque from the water as we rounded the point. Our campsite was in Lighthouse Bay on a secluded, tropical-looking beach. After setting up camp, we hiked up the beach for a visit to the lighthouse and the old-growth stand.

On the second day we paddled about 10 miles from Sand to Oak, stopping at York for lunch. We faced relatively high winds and choppy waves at first, but the weather eventually calmed and we strategically paddled on the leeward sides of the islands. We covered the 10 miles quickly--much faster than hiking the same distance--and arrived at Oak for a two-night stay. We chose to take a layover day at Oak because it offered inland hiking, including a trail up the highest point in the islands.

On our first night at Oak, we shared the bay with two sailboats seeking refuge from southwesterly winds. The crew from one of the sailboats came ashore in their dinghy to check out the trailhead originating from our campsite. They were very nice folks from the Twin Cities, and they invited us out to see their sailboat. I took them up on their offer and paddled the solo kayak out to their boat to watch the sunset.

We woke up to beautiful weather again on the third day. We explored the beach, took turns paddling the solo kayak out to some rock formations on the northeastern corner of Oak, and hiked about two miles to the high overlook. From above, Lake Superior looked even bigger--much too big for kayaks. Another highlight from day three... practicing my kayak roll that I learned while attending Maryville College.

We packed up early on the fourth and last day and headed back to the outfitter where our car was parked. Our next stop was Duluth and Fitger's Brewhouse for food and a celebratory beer. My friends really liked western region of Lake Supeior, as well as Minneapolis. I hope they'll return for another adventure soon.

"it is usually not possible to randomly assign advertisers to treatment groups due to contractual obligations and/or their willingness to be 'experimental units' for a service for which they are paying. In such cases, we ... use statistical methods that try to tease out causal inferences. Propensity score matching, inverse propensity weighting, and double robust estimates are some of the methods established in social and biological sciences currently in use at Google when randomization is not possible."

That approach mirrors best practices in quantitative evaluation. Randomized field trials are considered the gold standard for judging the degree to which a program or its components cause a desired outcome; when random assignment is not feasible, quasi-experiments provide a valuable alternative. Evaluation researchers rarely have as much control over conditions as Google's "engineers." Consequently, evaluators must rely more on quasi-experiments to "tease out causal inferences." Another key difference is that no matter how enormous a program data set may seem and no matter how many parameters a client might want an evaluator to estimate, those amounts will never reach the terabytes of data or the millions of parameter estimates that Pregibon describes as commonplace in life of a Google statistician.

By the way, my master's paper involved applying inverse propensity weighting to account for self selection into a local public school district. Does that mean a career as a Google statistician is in my future?

Try it for yourself. You can click on the image below to explore Minnesota's 3rd grade math proficiency results with Google Earth. After you've gotten your fill of zooming around the map, try the following:

• Click on a district to activate a pop-up window containing the district's name and results.
• Click on the "+" sign next to "2007 Minnesota Comprehensive Assessments (MCA-II)" under the Places sidebar. The district results will expand downward, showing results in tabular format with map links. Double-click a district name in the sidebar to zoom to that location.
• Use the Search -> Find Business section of the sidebar to find an after school tutoring program in a district of your choice.

Please leave a comment if you have any feedback about this approach.

I spent some of my break finishing proposals for the American Evaluation Association conference and a survey instrument for the North Star STEM Alliance. Nevertheless, we experienced a lot in just a few days:

• skied Bogus Basin,
• took a dip in the Pine Flats hot springs,
• visited the World Center for Birds of Prey,
• hiked around Camel's Back Park,
• picnicked along the Boise River, and
• watched Gran Torino at the cinema.

I highly recommend a visit to Idaho, if you've never been.

My presentation, "A spatial analysis of school district performance in Minnesota, demonstrating spatially enabled evaluation," consolidated some of my earlier posts to this blog and summarized the paper that I am scheduled to present at the American Educational Research Association (AERA) meeting in San Diego.

By looking to other social science disciplines that have become spatially enabled, I identified some ways that applied educational researchers can make better use of geographic mapping and spatial statistics. For example, we could enhance survey data and minimize respondent burden by spatially referencing the primary data and joining secondary data from the Census Bureau. Extending the promising uses, I spatially analyzed school district performance in Minnesota in 2007 and found the following.

• Neighboring school districts influenced performance to a small degree in limited instances (i.e., reading proficiency of third graders in poverty).
• Several school districts were found to have outperformed their neighbors, including Randolph, Medford, Orono, and Royalton.
• Six clusters of adjacent, correlated school districts exhibited low reading proficiency, suggesting they should collaborate to improve preschool and early elementary reading efforts.

You can click on the image to access the full presentation:

Rokia and her band exceeded our expectations. They packed the house and had everyone dancing. Rokia's singing was forceful at times and sultry at others, such as when she sang Gershwin's "The Man I Love" to honor Billie Holiday. I looked around the audience and saw many jaws drop when she showed off her vocal prowess, not to mention her dance moves. The band was tight and having a lot of fun. As a Kentucky boy, I especially liked the banjo-like riffs from the Malian guitar. Their one encore song lasted about twenty minutes, segued in and out of Fela Kuti's "Lady," and featured a solo by each member of the band. Shane, Amy, and I went to Palmer's afterwards, where we sat around the bonfire and swapped concert reviews with other folks who had been there.

Don't miss her when she comes to your town!

Below are some video clips from last night's concert. In the first clip, watch Rokia get rid of her mic stand so she can dance freely. She plays guitar in the second clip.