Data: Don’t Believe It

A few weeks ago, the Sacramento Bee newspaper, the only major newspaper in the Sacramento, California, area, ran an article about school superintendent pay.

The article compared the pay of superintendents relative to the number of students in the district. If one were to read the article, and let it be at that, then one would be led to believe that the cost per student in terms of superintendent pay was the highest in a small, two school, rural district. The cost for a superintendent in a large, urban, district was said to be much smaller.

The article was well written. The conclusions were also absolutely wrong. It was a comparison that, on the surface looked legitimate, but it was an apples to alligators comparison. A little bit of research would have revealed that the superintendent of the small, rural district, had no assistants, just the normal office staff: bookkeepers, secretary, financial officer. A look at the large urban district revealed that the superintendent had many assistant superintendents (human resources, curriculum, IT, facilities, maintenance, community resources, and so on). There were 65 different assistants of one kind or another under the superintendent.

When one compares the actual cost of the district offices, the small district is a bargain on a per pupil basis. The large urban district is far more costly per pupil. It just depends on what one counts, and the reason for counting in the first place. Bias comes to mind.

The drop out rate of students across the US is said to be anywhere from 30 to 50 percent, or more. These figures have gained traction among the charter school and choice business people. The figures are not accurate and it doesn’t really take much work to find out that they aren’t. What is true is that school districts across the US don’t have a really good method of figuring out where their students go when they leave.

In an April 19 story, Ed Bott of ZDNet, wrote about the reported number of pc’s that are infected with malware. “Nearly half of personal computers in the U.S. are compromised by malware.” This statement can be found on the internet and in various newspapers and magazines. Bott calls this statement “…an outright fabrication. It is not true. It is not even remotely accurate, based on objective data.” The actual number of infected pc’s, those that have Windows auto update turned on, or who practice normal internet security is around 1%-2%. That’s a far cry from 50%.

Bott simply took the time to track down the information that turned into the 50% number. It turns out that it was a usurped from a statement from Panda ActiveScan regarding people who already thought their pc’s had some sort of malware on them.

The information was quickly misquoted and put out as a fact on the internet, where it picked up a life as a “Fact”.

Pre-school is good for all children, and will keep them out of jail, and in good jobs. It is easy to find this one out there too. The reality is that this was taken from a Rand Corp. study regarding pre-school and the effects it had on a very specific group of children. It was not, and is not, applicable to “all children”.

It doesn’t really take much to look a bit further when reading various claims about just about anything. In just a few minutes, most of the information that is written or quoted, can be verified, or shown to be absolute hog wash. The Internet can be a pretty good research tool when used by inquiring minds.

The constant mantra of the Eduskeptic is always “assume nothing, verify everything.” The education establishment, one would hope, would be leading the charge when it comes to verification of various theories. Assume nothing, verify everything.