Towards a technography of Title 1A
I had a doctoral student named Rob Rust who wrote a great dissertation on surveillance capitalism and educational technology. In the part where he tells readers what to do to push back on this tendency, he recommends something called technography, which is where you treat technical codes and machines like agents and do ethnography on them. Here's how he describes it:
The technographer’s role is to ask what algorithms are suggestive of in the same way that the ethnographer “seeks to understand culture primarily through the meanings attached to the world by people” (p. 60). Technographic inquiry, according to Bucher (2018), does not require expertise but encourages “a readiness to engage in unknown knowns, seeing the black box not as an epistemological obstacle but as a playful challenge that can be described in some ways (but not all)” (p. 61). Technographic inquiry ultimately has the potential to answer questions like “what is the world view of an algorithm”, “how does it work”, and “who does it work for” (p. 61). Most importantly, technographic inquiry eliminates instrumental, deterministic, and substantive assumptions about these technologies and imparts awareness to the user and the agency to gaze back at the algorithm in a more involved manner. (Rust 111-112)
I've been working on a paper with friend and comrade and colleague Kate Cairns applying this approach to education policy, specifically school funding formulas. We think you can use technography to get involved with stuff like the Title 1A formulas--federal school finance policy that distributes money to poor schools--in the way Bucher recommends, to study their world view, how they work, and who they work for. What follows is a sketch from that paper. (Comments welcome!)
Rather than approach the formulas instrumentally and deterministically we hope to counter-interpellate their particularly esoteric sort of technical-complexity and gaze back at them to examine the social and political world they create in education finance.
Indeed, the Title 1A formulas are the agency of the agency of federal school financing, in the sense that, presuming that the formulas have a sort of agency unto themselves, we see that the agency of the Department of Education—the implementer of Title 1A itself—only has its agency through the formulas, which calculate the amounts that districts and local education agencies receive. They are recognized as algorithms in a recent analysis of them (IES 2019).
To do the technography on these algorithims however, we have to first understand the policy “coding” within which the formulas live. There is a formal architecture out of which they grow and within which they continue to exist.
The Big Grandfather and the Formula Child
The world of educational policy is a dry, arid, and spare environment. Reading the documents containing the lines of legal code which determine the distribution of billions upon billions of dollars can feel like occupying a one-dimensional universe whose ecology is composed exclusively of letters, statements, and legal-juridical arrangements thereof. There are technical statements stipulating criteria for what shall be, sections organizing these statements, and an impressively ornate but nonetheless desiccated cascade of subsections. Each of these features of a policy can refer to one another. It is in this desert of government desserts in which we find what is known colloquially as Title 1.
At this point, we can observe that when we hear the proper name “Title 1” it is typically, though not exclusively, a reference to this first part of the first title of ESEA. Part B for instance is called “Student Reading Skills Improvement Grants” which has multiple subparts on literacy. Part C is about education and immigration, Part D covers at-risk youth, Part E covers assessment, Part F is on school reform and on and on. But this insight is not trivial. But as we will see, even this specification does little to disambiguate exactly how the monies will travel. For Part A itself is a complex and undulating sector of the desert region we are describing, with its own towering landmarks and smaller though nonetheless important escarps within which its denizens, the formulas, live their lives. To see these formulas in their natural habitat and describe their culture we must traverse these terrains and landscapes.
In the beginning, there was the basic grant, or as we have named him, the Big Grandfather. In brief: the Basic Grant in Title 1A is calculated using a state’s average per pupil expenditure and the number of poor children in the state. The federal government pays 40 cents for every dollar the state spends to educate the number of poor kids in their care. This amount is indexed to the national average, creating thresholds for minimum and maximum basic grants, to make sure each state does not get too much less or too much more than the national average per pupil expenditure. There are also hold harmless provisions built into the grant to prevent districts from receiving less in a given than they received in previous year. Since each of the three other formulas depend on the basic grant calculation, we devote more time and space to him for analysis.
Thus the Big Grandfather’s body and mind are woven with these key phenotypical and genotypical aspects, which we might call a duality: the multiplication of each state’s average per pupil expenditure poor students by .4 and thus, perhaps most importantly as we will see, creating a tiny but essential character—the prince of this strange land—the formula child. To fully understand the Big Grandfather’s life, to see him with fullest clarity and how the formula child emerges from his body much like the other formulas, and yet emerges as the essential character in all of them, we must go step by arduous step through the dried brush and whipping sands of the Title 1A region’s slice of the policy desert.
As detailed previously, the ESEA contains numbered titles, which contain lettered parts, within which we find numbered chapters. We find one such chapter called Chapter 70 , or “Strengthening and Improvement of Elementary and Secondary Schools.” In these numbered chapters there are numbered subchapters. Underneath chapter 70, the first subchapter is called “Improving the Academic Achievement of the Disadvantaged.” Within that subchapter, we find a number of lettered sections and numbered sub-sections. Whereas the chapters and titles are the landscape of their native lands, these sections and sub-sections are the dried wood, mortar, and thatched roof of the longhouse where the Basic Grant and his tribesmen reside.
In the dried-out stacks of one-dimensional numbers and letters constituting this longhouse, through the dehydrated thickets we have had to cut through with our machetes, a sign comes into view on the front of the structure: Section 6333 of Chapter 70, it says, as well as “Basic Grants to Local Educational Agencies.” The sign is affixed over the door and through the door of Section 6333, Chapter 70, we can see the outlines of bodies busy at work. Chief and largest among them we can see the Big Grandfather, the Basic Grant, in his two-partedness, pouring over spreadsheets reporting state average expenditures and jotting down new numbers, multiplied by .4, guided by thresholds. These formulas are as intensely focused as they are aloof and uncaring about our apparent intrusions into their work. They do not notice we are here. They may feel a sense of surprise to be come upon in this way by flesh and blood humans, but they permit us to look at them and their longhouse very closely and do not trouble us. After all, we are their creators and they serve our children.
To understand the Big Grandfather, we must examine the materials of his sub-section of the longhouse, proudly marked 6333(a). Subsection 6333(a) is a law that stipulates the number of children that make a district eligible for the Basic Grant. Sitting at the Big Grandfather’s feet as he assiduously and perhaps obliviously combs through his numbers, we see a little one, a small entity, not quite a formula, not quite a word, not quite a letter or a statement. Rather this little entity is like a position, a person-shaped hole, a ghostly form emanating from the Big Grandfather’s substance but having its own being. In 6333(a) we read that “the number of children counted” is essential to the Basic Grant formula.
Almost like an incantation of some technocratic religion, the being of this little entity is conjured by an utterance referring to the materials comprising the longhouse. The number of formula children is equal to “the number of children counted under subsection (c).” Thus we must shift our attention from the Big Grandfather slightly to his left, to subsection 6333(c), which is called “Children to be counted” and has two numbered sub-subsections and three lettered sub-sub-subsections.
The first of these, 6333(c)(1), defines categories of children to be counted through sub-sub-subsection 6333(c)(1)(A): “the number of children aged 5 to 17, inclusive, in the school district of the local educational agency from families below the poverty level as determined under paragraph (2)”. (We might imagine a policy where the age range drops to include those 0 years of age, thus including those children pre-kindergarten and perhaps providing resources for childcare and daycare at the federal level. But the range stays at 5.)
The following paragraph lays out a complex criterion to determine this number of children, which results from calculations made by the Department of Commerce. The details for the determination of the number of formula children continue on in a nested tangle of criteria, including that the local education agency must have 10 or more children and more than 2 percent of the total school-age population within the agency’s jurisdiction, which are further elucidated in 6333(c)(1)(B) and 6333(c)(1)(C).
Again, we might imagine (and policymakers did indeed push for, see below) a policy more directed towards urban areas with higher absolute numbers of children in poverty, raising the minimum number of children to 1,000. But a district can qualify with only 10. The final point that bears examination is that the poverty line is decided by the Secretary of Education and the Commerce Secretary using numbers produced by the Census and the Bureau of Labor Statistics, specifically the Consumer Price Index for All Urban Consumers, part of the Small Area Income and Poverty Estimates program of the U.S. Census Bureau (Snyder, Dinkes, Sonnenberg, & Cornman 2019, p. xi). These students are children who part of families receiving Temporary Assistance for Needy Families (TANF). There are further stipulations for what these leaders should do if and when they conclude the poverty line data is inappropriate or inadequate, which is another avenue cabinet members have to shift the scope and depth of the policy.
At this point however, we have not yet arrived at the formula itself in the language of the law. We have seen the stipulations for how to calculate a key component of the formula, which is the number of formula children. Formula children are all those below the poverty line (according to the Department of Commerce, Census, and Bureau of Labor Statistics) within the delimitations set out above: those between 5 and 17, within a local educational agency’s jurisdiction, and in districts with 10 or more poor children. To get the amount of the Basic Grant, this number of total formula children is then multiplied by a percentage of the state’s per-pupil expenditure, or how much that state spends on each student’s public education, within a window of the national average per pupil expenditure. The exact wording is as follows:
40 percent of the average per-pupil expenditure in the State, except that the amount determined under this subparagraph shall not be less than 32 percent, or more than 48 percent, of the average per-pupil expenditure in the United States.
We begin to get a sense the formula’s distribution in this paragraph, which tracks who receives funds:
The total of grants for such counties shall be allocated to such a local educational agency, which local educational agency shall distribute to schools in each county within such agency a share of the local educational agency's total grant that is no less than the county's share of the population counts used to calculate the local educational agency's grant.
A recent analysis of the four formulas helps clarify this language and provides concrete numbers from 2015. The Big Grandfather takes up 45 percent of the space in the longhouse: “Basic grants accounted for approximately $6.4 billion of the total Title I funds in FY 2015, or about 45 percent of the $14.3 billion allocated.” (xiv) When it comes to the areas that the basic grant serves, they are mostly remote rural and suburban areas: “Allocations [of the basic grant] were highest in remote rural areas ($583) and small suburban areas ($563) and lowest in midsize cities ($532) and fringe rural areas ($534).” (xiv) This makes sense give the Basic Grant’s focus on relative percentage of formula children rather than absolute numbers. The relative number of poor children is highest in places with small numbers of people, many of whom are poor. This issue lead to the creation of the next grant.
The Congressional Research Service history tells us the story, which we can summarize. After a few years of the program, in the late 1960s, policymakers and researchers noticed that “the 20 wealthiest counties in the nation...were allocated almost twice as much as the 10 poorest counties” (14). Some lawmakers called the formula “patently foolish” because it “reflected a ‘political decision’ to spread the funds as broadly as possible, rather than concentrating them in the areas of greatest need, to build a ‘powerful lobby for the continuation and expansion of the program” (14). While there had been a program proposed for “special grants to urban and rural schools which provided additional funding to areas where there were high concentrations of disadvantaged children,” (20) it was not until an analysis of the basic grant in 1978 that researchers demonstrated that “per pupil Title 1 expenditures were higher in predominantly urban and suburban states and the inclusion of AFDC children in the formula provided the largest benefits to northeastern states” (28).
The 2019 analysis we have just cited actually finds that the Basic Grant still has this problem. There are two ways this problem manifests in the formulas. First, there are areas with low population but relatively high concentrations of poverty that still need more resources. Second, there are areas with much larger populations who have bigger absolute numbers of formula children. Thus it was that the Concentration Grant and Targeted Grant were born.