Only one indicator.

The FIRST rating the Lancaster ISD failed for fiscal year ending 2005 asked the question:

Did The Comparisons Of PEIMS Data To Like Information In Annual Financial Report Result In An Aggregate Variance Of Less Than 4 Percent Of Expenditures Per Fund Type ?

This is also termed the "Data Quality Measure". The PEIMS data is stuff like daily average attendance, teacher-student ratios, -- general nuts-and-bolts data about how many whos are where doing which whats. If the sum of all the numbers in all the columns equals the sum of all the numbers in every row, then the "variance" would be zero. Generally, in an imperfect world, the numbers do NOT add up, exactly. But within 4% or so, they should.

Only 14 districts out of over one thousand failed this test in fiscal year 2005. One district that failed was our neighbor Irving. The Irving ISD, which like Lancaster also earned a "Superior Achievement" rating by failing only this one indicator, posted an aggregate variance of 13.2%.

The North Forest ISD earned an "Above Standard" financial rating .. which is down one notch from "Superior". (This lower ranking was because they failed two additional indicators.) They failed the PEIMS data quality question with a variance of 10.9%

Quinlan ISD faced a "Substandard" Rating, with 6 negative indicators, one of which was PEIMS data quality variance of 8.7%.

Valentine ISD eked out a marginal "Standard" Rating with 5 negative indicators, and PEIMS data quality variance of just under 5%.

Thirteen, eleven, almost nine, not quite five ... anybody care to guess what the Wilmer Hutchins district reported as a variance in fiscal 2005? How about 100% ? That's amazing. Every data element they reported was in variance. 100% variance. Can it get any worse?

Well, the LANCASTER ISD with its "Superior Achievement" ranking managed to fail the PEIMS data quality indicator with a 121% variance.

The Lancaster ISD had more variance than they had data.

And this is the second FIRST ranking year, the second fiscal year, in a row where the PEIMS data quality has been at unaccepable variance. In fiscal year 2004 the variance was 107%.

Now, in 2003 -- the last fiscal year before Dr Larry Lewis arrived, the variance was only 0.49%. That's like about one-tenth the variance allowed by FIRST standards.

The same staffer, Mr Greg Long, has held the post of PEIMS expert since before 2003 and now since Financial Officer Eugene Smith has come and gone. One might think that Greg Long would know, by now, how to keep his data quality within tolerance. If, that is, he were allowed to.

But for the last two years, while Greg Long has fallen lower and lower on the Lancaster Fianance Office pecking order, the PEIMS data quality has gotten worse and worse.

Hey, Mr Long, if you're reading, would you care to comment?