Interpolating proficiency rates

One of my research interests has been the geographic distribution of early student achievement at the school district-level. District-level outcomes apply to large areas that are often out of alignment with neighborhoods, cities, and counties where local information could benefit parents and local governments making early education decisions. Additionally, I suspect that learning outcomes vary spatially according to a stochastic process influenced only partially by district boundaries, which in the context of multilevel modeling might appear as autocorrelation among within-group errors or as spillovers resulting from the modifiable areal unit problem. Therefore, I have begun examining sub-district geographic distributions.

Minnesota_math_proficiency_surface.gif

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.

Posted in Praxes | Comments Off on Interpolating proficiency rates

Computer recovery

My computer crashed a few days ago. I own a Lenovo T60 running Windows XP. The crash happened very suddenly, right in the middle of working on an AERA proposal. My battery gauge needs calibrating, so I probably just ran out of power. I plugged in, restarted, and was greeted by a blue screen saying,
Stop: c0000218 {Registry File Failure}
The registry cannot load the hive (file):
SystemRootSystem32ConfigSOFTWARE
or its log or alternate.

This type of failure has been appropriately dubbed the “blue screen of death.”

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:windowsrepair 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~1rp##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:windowssystem32config with those from c:windowsrepair. (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:windowssystem32config 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.

Posted in Personal | Comments Off on Computer recovery

Apostle Islands adventure

IMG_0216.jpg

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.

P1010050.JPG

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.

IMG_0325.jpg

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.

IMG_0276.jpg

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.

P1010083_2.JPG

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.

P1010057.JPG

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.

Posted in Personal | Comments Off on Apostle Islands adventure

Quasi-experiments at Google: Evaluation insights

Google’s statisticians routinely use randomized experiments to improve their products (and profit), but did you know they also conduct quasi-experiments when random assignment is not feasible? I receive the American Statistical Association’s (ASA) membership magazine called Amstat News. Daryl Pregibon, a Google statistician (or “engineer” as they are called internally), was invited to write about the company’s statistical practices in the May issue. He writes that Google users can be randomly assigned to treatment conditions, but

“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?

Posted in Praxes | Comments Off on Quasi-experiments at Google: Evaluation insights

Boundary Waters Canoe Area Wilderness (BWCAW) maps in Google Earth

It has been a while since I’ve posted any fun, canoe-related content. So I created a set of Google Earth KML maps of the BWCAW. I was inspired to post the map set after I saw the UMN Borchert Map Library post a link to an interactive BWCAW map. Canoeists, enjoy.

Boundary Waters Canoe Area Wilderness (BWCAW) BWCAW.png
Source of geographic information: Superior National Forest GIS downloads (SHP files created by USFS June 2006)
Source of functions for converting from SHP to KML: maptools and rgdal

Posted in Personal | Tagged , | Comments Off on Boundary Waters Canoe Area Wilderness (BWCAW) maps in Google Earth