Bias

One of the most common comments I receive on my posts is that my information is not reliable because it is from a source that the commenter believes has a liberal media bias. I assume their belief is that data or studies produced by liberal leaning sources cannot be impartial. Their solution; I should use conservative leaning sources.

The thing I find most amusing is how many of these people make this claim and then supply absolutely no information of their own to refute the "facts" in my sources. If my sources are so inaccurate then it should be easy to link an article that proves their inaccuracies.

Media bias is a lazy argument that proves nothing.

Regardless, I thought it would be a good idea to further examine a conservative leaning article that I read in creating my recent post on education.

The Heritage foundation put together a report titled "Assessing the Compensation of Public School Teachers". In their summary statement they ask the question "Do teachers currently receive the proper level of compensation?" followed by the answer "Standard analytical approaches to this question compare teacher salaries to the salaries of similarly educated and experienced private-sector workers, and then add the value of employer contributions toward fringe benefits. These simple comparisons would indicate that public-school teachers are under compensated." and further on stating that "public-school teachers receive salaries that are 19.3 percent lower than non-teachers who have the same observable skills."

So their initial conclusion is in line with other research on the subject except this is a conservative think tank and just like all of my liberal sources the Heritage foundation has an agenda and they must massage the data until it fits their narrative. Which is why the report follows that up with "However, comparing teachers to non-teachers presents special challenges not accounted for in the existing literature."

Luckily for us the Heritage foundation is smarter than all of the other organizations who have studied this topic and will set the record straight. They identify 6 previously unaccounted for categories and oddly enough every single one of them supports their eventual conclusion. I think this would qualify as the definition of bias when you discount the data that disagrees with you and in uncovering the "truth" every metric you decide should also be considered just happens to support your conclusion. A conclusion which took public school teachers from being 19.3% under compensated to "52 percent greater than fair market levels".

Having said that I will give them the benefit of the doubt and assume that there are truly only 6 different metrics that were not previously accounted for and every one of them shows that public school teachers are over compensated.

Unfortunately it doesn't take long to find issues with their claims and the data sets used to uncover the truth.

Most concerning is their claim that public teachers job security leads to them being over compensated. I don't take issue with the belief that job security has a value but to determine job security they use unemployment. First, they only use unemployment numbers up to 2010. Thanks to the stimulus package schools were given extra money over the past couple a years to retain teachers. Those benefits are now ending and so too is the job security that public teachers have enjoyed.

While the 2010 and 2011 unemployment numbers by profession are not available yet, the number of teachers employed is. Over the last 15 months the number of public school teachers has decreased by over 169,000. Accountants, which the Heritage Foundation report references as having a higher unemployment number and subsequently lower job security, have actually seen an increase of over 50,000 jobs in that same time frame.

In addition to this cherry picking of data, it should also be noted that 52% of teachers hold a Master's degree while only 21% of accountants currently have a Master's Degree. Data shows that the higher the level of education, the lower the unemployment rate. Perhaps teachers aren't filing for unemployment because they are able to find other jobs instead. If that is the case then using unemployment as a measure of job security is either intentionally misleading or willfully ignorant.

Of course using only the most advantageous data sets is not the only trick they use. They also frequently use the phrase "authors calculations based on data from the US Census" yet the data they supply the reader does not correlate with their conclusion. In one instance they use a chart to show that public school teachers are overpaid by 9.8% using one measure yet the numbers provided only show an advantage of 1.7%. I guess we are to assume the author has included other data that he felt is not worth providing the reader and since no link or reference is given to the specific data set used the reader must rely on the work of the author.

Other manipulations include is routinely making claims without supporting data, drawing definitive conclusions from extraordinarily small data sets, discounting all other studies on the matter as flawed, using select facts that support their conclusion while ignoring others that don't, making the claim that "no comprehensive data set exists" and then manufacturing a data set that just so happens to prove their point, and perhaps most egregious, listing an article in the reference and claiming it supports your data when the linked article actually says the opposite.

After reading the “study” I choose to ignore their findings not because they are a conservative leaning source but because when I analyze the data there are too many flaws to consider it a reliable source.

Maybe that is what you think of my sources too but using obtuse arguments like media bias as your only support doesn’t prove my sources are wrong or add anything to the debate.