A Lesson Regarding Causation, Correlation, and Costs in Education
Correlation: a relation existing between phenomena or things or between mathematical or statistical variables which tend to vary, be associated, or occur together in a way not expected on the basis of chance alone. Causation: the act or process of causing; the act or agency which produces an effect. Cost: the outlay or expenditure (as of effort or sacrifice) made to achieve an object-Merriam Webster Dictionary
These definitions are important because of how easy it is to confuse correlation with causality. Note that variables or events can show a spurious causal relationship to each other because of a third factor that is acting on each of them. For example, smoke is seen rising from a cannon, a second or two later a cannonball hits a fortress wall. The events are related but both were caused by the explosion and burning of powder.
Recent comments of several school officials from around Allegheny County on the subject of funding full-day kindergarten reveal a serious lack of understanding of the difference in correlation and causation. Officials have extolled the advantages of full-day kindergarten with comments including “those students are more likely to continue their education, have jobs, and own their own homes” and “…the likelihood of them being institutionalized [in prison] increases fourfold if they don’t get a good early education up to grade three”. Such statements are made easily and gratuitously as if there must be a direct causal link between a good education up to grade three and success in later life. Worse, they are made with the full expectation that no one can possibly object to them.
Obviously this view involves a tremendous discounting and ignoring of other important causal intervening factors that influence rates of home ownership, employment, and incarceration. Moreover, it ignores the well-established fact that, in far too many schools, achievement scores in third grade fail utterly to predict the high school 11th grade performance. Just as there is no guarantee that good scores in early grades will translate to good scores later in high school, it is not written in stone that children in schools without a major kindergarten to third grade spending effort-as advocated by educrats-are doomed to failure in later years. This is especially true if these children have a school and home environment and culture that is supportive and maintains discipline and encouragement.
Millions of people have owned homes, built businesses and communities, etc. without full-day kindergarten-or any kindergarten for that matter-and some without even finishing high school. Family, societal expectations, and peer pressure are far more important to continued academic progress, or the lack of progress, and success in life beyond school than all the spending school districts can come up with.
To say there is an emphasis on early childhood education (the percentage of kindergarten pupils in a full-day program rose from 25.1% in 1979 to 63% by 2000, according to the Education Commission of the States) and all the good it can do and all the harm that can come from its absence is an understatement if there ever was one. But the proof of the wisdom or irrelevance of this emphasis is found in the impact it has had. For example, have SAT scores improved during the period? In a word, no. Have graduation rates improved? Are incarceration levels lower? So what is the educrats response to the absence of evidence that all this effort has made much difference? That’s easily predicted. We need even more resources. Or, the situation would be much worse if not for the increased spending-a hypothetical assertion that cannot be proved. In this case, not only has there been no positive causal relationship, there is not even a correlation. But advocates will not be deterred.
So why, one may reasonably ask, are local officials singing the praises of full-day kindergarten now? It is because there is a good chance that a major funding source could be on the chopping block in the coming fiscal year. The state’s Accountability Block Grant (ABG) started in the 2004-05 school year and provided $165.9 million statewide to pay for full-day kindergarten in 2009-10. Twenty five of Allegheny County’s 43 school districts utilized ABG funding to provide full-day kindergarten for some or all of their kindergarten pupils, according to data from the Pennsylvania Partnership for Children.
If the ABG is eliminated, many of the full-day kindergarten programs may end or districts will have to reallocate other dollars to keep them going. The East Allegheny School District has already voted to go from full-day to half-day. Other districts are thinking about scaling back but want to see what actually comes out in the final state budget. One district, McKeesport Area, will use other sources of funding to keep full-day kindergarten in place.
One publication from the Pennsylvania Department of Education demonstrates the correlation/causality confusion. It noted that “the students who attended full-day kindergarten programs in 2004-05 took the third grade PSSA in spring of 2008. Seventy-seven percent of 3rd grade students were proficient or advanced in reading in 2008 compared to sixty-nine percent of 3rd grade students in 2006″. Clearly there are questions about these findings. How many of the 2006 test takers were graduates of full-day kindergarten? Was the performance of half-day kindergarten graduates evaluated separately? Was there more emphasis placed on prepping for the PSSA in school years following 2004-05? Did the PSSA test get easier?
Let’s use two north suburban districts to do a quick comparison of the question posed above. In 2009-10, North Hills used $364,000 in ABG funding to serve 316 students in full-day kindergarten; adjacent Hampton had 166 students in half-day kindergarten with no ABG funding. In the 2006-07 school year (when last year’s third graders would have been kindergartners), North Hills had full-day kindergarten with ABG funding and Hampton only offered half-day, so the Department of Education’s theory would hold that North Hills should have outscored Hampton on the PSSA. In fact, the opposite happened as the percentage of students scoring “advanced” or “proficient” on the reading exam was 93.3 percent in Hampton and 89.3 percent in North Hills. Granted, this is one pair of schools. But there is no gainsaying the impressive scores posted by Hampton third graders. The proponents of full-day kindergarten need to find examples of schools that did as well as Hampton-focusing especially on those getting ABG money.
Finally, are school officials cognizant of the concept of opportunity costs? One advocate for early childhood education stated about the cuts and the decisions that must follow that “…I wish they didn’t have to make these tough choices”. Sorry to say but that is how scarce resources are allocated since taxpayer dollars do not come from a bottomless pit. That’s true even with the ABG money itself, as it is permitted to be applied to uses like support for struggling students, teacher improvement, or research based programs.
Every dollar spent on any one aspect of education is a dollar that cannot be spent on anything else. Thus, it is incumbent for government officials and school boards to ensure that every dollar expended yields the highest possible payoff in terms of student learning. That requires a correct understanding of the true value and contribution of each program. In turn, that means officials must not be deluded or misled by “experts” who do not know or care about the tremendous difference between correlation and causality.
In reaction to the possible education cuts in the state budget, one local school official noted “anything that is nonacademic [is up for cutting]” and another school head said “specials would be gone and our primary focus would be reading and math”. This comment recognizes there are distinct and important differences in the educational impact of the multitude of items schools spend money on. What an important realization of a basic principle. If adopted wholesale by officials in their decision making it could help us get to a “thorough and efficient system of public education to serve the needs of the Commonwealth” as prescribed in the state Constitution.