Many public school experts and corporate legislators have debated the cause of low student achievement in our public schools for years, even decades, but have failed to agree on the primary factors affecting student test scores, etc... Several "experts" (including me) believe that low grades in any particular high school class is directly correlated to the ratio of girls to boys in a classroom, and point to anecdotal data in order to confirm their suspicions. For instance, the only 'F' I ever received in high school was in Accounting class. I blame it on the fact that there were 27 girls in the class, but only 3 boys - a ratio of 9 girls for every boy. While this ratio may be very good for a cattle rancher, it's not good in an accounting class. Rick Dorman, another boy in the class, made a slightly higher grade, so the ratio of girls to boys can't be blamed entirely as the cause for low "accounting" grades. (Disclaimer: this is a partially satirical view and partially scientific view of low student achievement in public schools, but 100% true.)
Many corporate lawmakers who consider themselves public school experts, believe low student achievement is directly related to "bad teaching", if not the primary cause. This theory is questionable, because they cite no supporting data which backs this claim. It has about the same validity to say "the ratio of girls to boys is the cause of low student achievement among boys", as it does to say "bad teachers cause low student achievement".
The only way to provide even partial validity in determining the cause of low student achievement in public schools is to do a little scientific analysis of existing data. We are lucky, as Education Week annually provides public school data from all fifty states and the District of Columbia, in it's 2017 Quality Counts State Report Card. We must only provide an analysis of the data, in order to determine links between factors:
Oklahoma public school performance grades:Oklahoma finished 47th among all 50 states and DC with an overall score of 68.3 out of 100 points and a grade of 'D+'. While Education Week provided data in three categories and eight subcategories of state grades, we'll only analyze three subcategory grades for indications of correlation and possible causality. Oklahoma received an 'F' (52.8 points) in the status of K-12 Achievement (measures the state's current student achievement in reading and math performance, graduation rates and results of advanced placement exams. Oklahoma earned an 'F' (43.4 points) for spending on public schools, which surprises none - and is one of the lowest 'F's ever recorded in the history of Quality Counts. The sub-category Spending include measures of average state-wide per student legislator spending and the share of state resources spent on K-12 education. The third sub-category examined, but not analyzed, is Accountability for quality of The Teaching Profession - which Education Week provided no data in 2017, so the data from 2012 is: Oklahoma received a 'B-' (82.4 points). Aside from the subcategories Achievement Equity (88.2) and funding equity (88.2) of 2017, the Quality of the teaching profession for Oklahoma schools had the highest point total at 82.4. This Oklahoma data point alone may eliminate the validity of "corporate legislators' claim that Oklahoma's poor teachers are the root cause of low student achievement, as Oklahoma ranked 3rd among all 50 states and D.C. in the Accountability for Quality of the teaching profession subcategory. We will analyze only two subcategories, school spending and status of student achievement, by correlating the states' point totals in each category to one another. The degree of correlation between the two data sets may be determined by utilizing the following correlative formula: r= the sum of XY - (the sum of X times the sum of Y) divided by the number of data points (ndp); divided by the square root of [the sum of X squared - (the sum of X) squared divided by the ndp] [the sum of Y squared - (the sum of Y) squared divided by the ndp]...
Now that we're all thoroughly confused, I'd like to apologize to my O.U Statistics professor - Dr. Osborn; my Doctoral Chair - Dr. Maiden; and the Dean of the O.U. Education Department - Dr. Garn, for limiting this analysis to only two variables and utilizing a very simplified correlative analysis. An in-depth analysis could have been provided, but I don't have the appropriate computer app. program, and my time is limited. Also, algebraic expressions, symbols, and terminology have been limited or omitted to prevent overload of corporate and crony brains... for an overloaded corporate mind is akin to the last string of Christmas lights which blows a fuse.. or resembles a crack head's mind on meth (bacon frying). I'd also like to apologize before publishing this column - to all readers who are offended by the truth of assigned responsibility and indications of causal relationships, as we know that some will be (namely corporate lawmakers and their cronies). OK, lets plug in the values and calculate r for state spending and student achievement for all states, using the computational formula as a whole: r = (199,773 - 195,150) divided by the square root of 78,613,942 = 4,623 divided by 8,866 = .52...
(Warning: Governor Fallin and corporate legislators in the House and Senate (and their corporate crony friends) should avoid staring directly at the formula as it could cause temporary blindness... or at the very least, brain function could improve..)
According to Cohen's (1988) Guidelines for Correlation Coefficients, a Large Correlation between variables is .50 to 1.00, so r = .52 is a medium/large correlation. (Although it won't damage their mental processes for corporate lawmakers to read the last sentence, it won't help them either, as it could just as well be in a foreign language.) The bottom line is that legislative public school spending is highly correlated (.52) with the status of student achievement. We can also analyze r squared (.27), by determining that 27% of the differences in state student achievement can be predicted by state legislator education spending. Holy Cow! Corporate state legislators blamed low student achievement on poor teaching. Like we said earlier, a correlative procedure could be run, analyzing State Teacher Training, of which Oklahoma scored 82 points in 2012, and Status of Student Achievement - but no correlation at all would probably be determined.
On a final note: Just because two variables are highly correlated does not mean that one is causing the other. There could be several variables at play which affect Student Achievement, such as the ratio of girls to boys (9 to 1) in the Accounting Class I mentioned earlier. But across all 50 states, in a scientific analysis comparing State Education Spending with Student Achievement, we can safely say that the amount of state spending on public schools causes low or high student achievement... but what other school factors may play a role in student achievement? Oklahoma Corporate Lawmakers believe that "Reducing Administrative Costs" is the factor that influences Student Achievement 'more so' than their own state spending for public schools. Many public school experts believe this "hammering of school administrators" is only political rhetoric, so we'll put their belief to the test and correlate State School Administrative Costs with Student Achievement to determine if corporate legislators have a valid argument for public school consolidation, or if it's only empty political rhetoric (lies) to say that only school consolidation will improve student achievement.
Since it's now post Christmas, I'm once again not apologizing to corporate legislators and their friends for conducting the above research and publishing the results. They probably don't understand it anyway. So, not sorry...
No comments:
Post a Comment