Self-fulfilling prophecies in “high level” schools

Folks in Japan are very sensitive to the issue of high schools, to the point where it is a sort of mild taboo to ask someone which one they went to if they have reason to think you know the academic rankings of high schools in their hometown.  It’s something like common sense here that which high school you went to shows how smart and successful you are, much like universities around the world, to the point that you actually write your HS alma mater on your resumé here in addition to your university and graduate school.

High schools make some sense as gauges of academic ability – they’re not technically mandatory (although 95% enroll) and you need to take a test to get into them.  High-ranked ones do indeed yield vastly improved chances at getting into college (see below).  Parents of teenagers can be heard talking about the quality of education at this or that high school in the city, planning which HS entrance exams to take (in my area, you can only take one public HS entrance exam, encouraging people to aim a bit low on the principle of better safe than sorry). If there is a private HS, people will consider it sort of a get-out-of-jail-free card for university entrance exams (or as a reader pointed out to me after this post went up, they will in my area and maybe not yours), as enrolment at private HS, JHS or elementary practically guarantees a spot in its affiliated university, the highest step in a so-called “escalator school”.  This raises an issue of how matriculation at those universities can mean anything if students effectively tested in at age 12, but that’s one for another day.

(Ok, so if kids are basically guaranteed admission to a university after they get into its affiliated HS, JHS, elementary school or even preschool, how can these institutions still be status-bearing?  They might offer good programs but the main thing that justifies their prestige is the difficulty of their entrance exams, and in the case of escalator schools everyone knows a significant portion of the undergrad population was sorted in before even hitting puberty.  I know I’m basically affirming academic credentialism by complaining about this, but a little consistency isn’t too much to ask.  I’ve known kids who tested into these escalators at the higher steps and they always hated the ones who got in on the ground floor, who they felt were lazy and entitled from years of no competitive pressure.  Again, not that competition ought to be the point of education, but when the system is premised on the battle to beat the statistical average of an entire country of test-takers, things like this along with AO and “recommendation” admissions threaten the validity of entire enterprise on its own terms.)

What puzzled me for a long time was how wide the gap in academic strength is from one HS to the next, when they don’t differ by economic class of the areas they serve (since students can test in from any area) and nominally have to follow the general national curriculum.  I have a theory on this I already kind of gave away in the title but which I’ll spell out in a bit more detail below.

The mystery

Academic strength is not merely a matter of perception – I got the 進路 shinro “career pathways” statistics from the public HSs my former students went to, and the differences I found make it hard to believe they serve the same community and exist less than 10 miles from one another.  This graph shows the matriculation rates for graduates from 6 public HSs in raw numbers.

Screen Shot 2016-09-06 at 10.52.05 AM.png
Left to right in order of prestige. Senmon = technical school; Rōnin = taking a year off to study and give college entrance exams another try next year (apparently Fuji students don’t accept going to their fallback school); Shūshoku = employment.
And this graph gives the same data as a percentage of graduates:

Screen Shot 2016-09-06 at 10.52.08 AM.png
Note: 2-year colleges in Japan are not generally analogous to junior or community colleges in the US,.  In general, 2-year colleges in Japan are finishing schools for women (over 90% of 2-year college students are women) and like 4-year universities are status-bearing institutions, higher-ranked than technical schools.
Tuition at all 6 schools according to this site is between 190,000 and 260,000 yen for the first year, or about $1800 to $2500.  Public high schools are not free, but considering that many families will spend 3 times that much on juku fees, they’re not that expensive either.

A lot of public HS websites have the odor of the work of near-retirement school staff taking weekend computer lessons.  This image comes from the Fujinomiya Kita web site.  Not sure if I would want prospective students to see this first.  Fuji High School’s website looks like a beginner’s web design project from the late 90s.  For some reason every website has its school song but none have faculty listings.
Clearly, HS rank means something.  What I’m not sure about is whether the rates of success from going to these schools come from what people tend to assume they do.

The solution?

First, let me address the things that I don’t think are reasons that such large gaps in perception and achievement exist. Teachers vary in quality within and between schools, and there may be reason to assume the highest-ranked schools would be more selective when it comes to hiring staff. But in the end each school needs roughly the same number of teachers; and aside from private schools each has roughly the same budget for staff and the same pay scale. Therefore there are logical limits on how much we can expect the quality of staff, facilities, and materials to vary between public HSs – all the ones I’ve seen look the same, more or less like American prisons.  If the part of the brick-and-mortar school, its materials, and its teachers plays in giving students a good or bad education is as limited as it appears to be, what accounts for the wide difference in their outcomes and reputations?  Surely the gap between Fuji and Fugakukan above must be due to something besides the aspects of the school that are affected by their budgets.

(There is a different set of limitations on the effects of teaching staff and economics at JHS – JHS teachers are rotated throughout their district from school to school every few years, meaning one JHS can’t hope to build up a reputation for strong teachers.  On the other hand, JHS but not HS student populations are determined by dividing the city into zones for each school, and therefore a JHS simply being in an affluent neighborhood can result in more privileged students on average, as with most HSs in the US.)

In judging the strength of a school we tend to focus on the pieces of the school experience which are static – the location, the buildings, the programs, and the teachers.  What I believe is much more influential is simply that the demographics of the communities of students in them at any given time are so precisely selected, by the school and by themselves.  We ought to look at schools and their rates of success more in terms of the characteristics of the incoming students that form their classes, as either Communities of Purpose or Communities of Practice (depending on how expert we believe students are in navigating the business of being a student), and less as immutable institutions that confer some of their attributes onto the people who pass through them.

Whose approval do students seek the most during their teenage years?  Whose opinions do they care about and who do they try to impress?  To take a slightly left-field example, do children speak with the accents of the teachers and parents, or that of their peer group?  I know I only sporadically was aware of my teachers’ opinions of me during my teen years, and generally that was of concern mainly because it affected how my peers saw me.  Ditto for grades and most other things.  A teacher supplies some of the things that teenagers use to establish an identity with their cohorts, but the ultimate gatekeepers of your self-image as a teenager are your peers, not your teachers and certainly not your tests.  When examining the educational experience of teenagers at this or that school, it makes sense to think first of the kind of person that is their usual classmate.  In that sense the fine slicing of JHS graduate populations into incoming HS populations affects their eventual educational and career outcomes not by sending them to the appropriate institutions for their academic level but by placing them with peers of the same level, high or low, in effect reproducing the levels of success already evidenced by their admittance to that school.

So in a nutshell, my theory is that HSs produce the outcomes they do, successful or not, mainly because their students are similar in motivation and ability and progress through their years of schooling as part of a cohort of like-minded peers, and not due to any attributes of the schools themselves.  This isn’t necessarily a new phenomenon – people generally think something like this about colleges here, which often don’t even pretend to be about anything but confirming the results of earlier sorting mechanisms – but I haven’t heard it expressed about HSs before, and I’m not saying (unlike in the case of universities, which are basically ability-sorted gap years) that HS students don’t learn anything.  I’m just saying they don’t learn because of anything essential about the schools that they go to.

The circumstantial evidence

The ideal way to prove this would be to quantify all the permanent aspects of the school as an institution and predict a difference in educational outcomes based on those, and then compare that to the observed outcomes shown above, also quantified with values assigned to each university or other institution, including companies, that students go on to after graduation. The problem is that it would be hard to agree on a quantifiable measure of teacher quality, quality of textbooks, and conduciveness of other factors of the school environment to learning – and that schools, if they collect this information (my wife insists that teachers are not evaluated by students, and I haven’t been able to find anything analogous to don’t seem to make this information public.  As I mentioned, most schools don’t even list their staff on their public websites, and the school evaluation information provided under the name 自己評価 jikohyouka “self-evaluation” covers different categories with different questions, wordings and criteria at each school (see Fuji and Fujinomiya Kita) making teasing out institutional aspects of the educational experience at HSs difficult to say the least.

The way I’d like to prove my theory is by having the student bodies of Fuji Higashi (a high-ranked school) and Fugakukan (a low-ranked one) switch for one year, swapping facilities, materials and teachers and see how their experience changes.  If my theory is right, the Fuji Higashi kids will buttress each others’ efforts and achieve high success rates while the Fugakukan kids will quickly settle into the established patterns because of the norms of their group, even in a new environment.  Or we could have one kid who tested into Fugakukan go to Fuji Higashi and vice versa without them or anyone else knowing – again, if I’m right, the effects of the Community of Practice will push them toward the mean for their group regardless of how much “innate intelligence” they possess.  Compensation for the human rights violation this would constitute might come in the form of automatic admittance to any public university in the prefecture.

You could even raze Fuji Higashi to the ground and as long as its student population had another place to go together, you’d expect the promise of their high HS entrance exam scores to be realized in university entrance exam scores.  Now of course no one will run an experiment like these, and my theory is just a pat unprovable explanation for why HSs see such wide gaps in educational outcomes.  There is reason to believe even attempting to test my theory would thwart its effects.

Self-fulfilling prophecies are sensitive – if we start swapping lucky Fugakukan entrants into Fuji Higashi and this becomes known, it will damage its academic image, leading fewer top-level students to aspire to go there and bringing down its required score for admittance.  This will lead its allotted thin slice of JHS graduate demographics to slide downward from almost all kids intent on going to university to something more like Fujinomiya Kita, where less than half do.  In fact any change to any of the 6 schools above that can be perceived to portend or be a result of falling academic standards could produce the same chain reaction.  The perception of each school being the kind of school it is drives students of that kind to attend it and produce the outcomes it does, and a change in the perception, even a groundless one, will yield evidence to justify it within a few years.

Because in my view it is the fine slicing of demographics of inbound students rather than anything permanent about the schools that produces their differing results, maybe Japan has it right by spending the bare minimum on those permanent aspects of schools.  Schools looking like prisons, students dressed for uniformity rather than comfort forced to sit in un-climate-controlled classrooms, and overworked and undertrained teachers overseeing classes of 40 make sense if your theory of education rules those out as significant factors in producing educational outcomes.  Rather than schools and teachers lifting students up by providing knowledge and the means to learn, students lift each other up (or bring each other down) by providing each other with motivation.  In this light it is only logical to outsource the business of actually learning to the students (who spend huge amounts of money and time on cram schools, tutoring and self-study) while making public education purely a venue for students to form and maintain communities.


The merits of this student sorting system depend on the perception that testing has merits, that it gauges something of relevance. On the other hand, if the public ever gets around to critically examining the tests that result in this fine slicing and discovers (if they are as badly written as university entrance exams) that the rubric by which students are divided actually has little to do with their ability to learn or skills they will need after graduation, the system will lose its perception of validity and therefore its function.  The coherence of the cohorts that make up HS classes will be threatened by the shattering of the impression that those students and only those students deserve to be there.

To me, a test which requires students to have attended cram schools just to know the material covered is already invalid, testing the ability of families to afford cram schools rather than the learning ability of their children.  According to the blog of a cram school in the Fuji area:



There are questions [on the HS entrance exams] which exceed the level of school textbooks.

For entrance examinations, you need to develop applied skills.

For the moment, the public apparently buys into the legitimacy of HS entrance examinations that measure dedication to the exams themselves and is willing to shoulder the additional burden that places on families in order for their children to succeed.  Ironically, in my experience working-class families whose economic burden for cram school is a proportionally larger part of their income are much less likely to question the merits of this system, instead blaming themselves for not devoting themselves more to academics in their youth or their parents for not giving them opportunities to study (a phrase I’ve heard is 親が行かせてくれなかった oya ga ikasete kurenakatta, “My parents didn’t let me go [to cram school]”).  These parents redouble their efforts to succeed vicariously through their own children, forgoing vacations to send them to cram schools for 夏期講習 kakikoushuu “summer session”, and contributing to a society-wide ratcheting up of the statistical mean scores that determine school placement in the first place, meaning no one child gains an advantage from the proliferation of after-school cramming.  If you accept either that these exams are valid or that you individually can’t do anything about them, you certainly don’t want to be the only parent whose child doesn’t have the extra help.

It might do Japan some good to let the institutions do a little more of the work where learning is concerned – the shifting of responsibility for academic outcomes from peer pressure + cram schools to the work of professionals and resources within the public school system would mean a lot more available time and money for families of teenagers.  On the other hand, school costs might rise as more teachers handle smaller classes and districts buy computers that run on something other than Windows XP. It would also mean entirely new expectations of the culture of the classroom, where students cannot rely on 40 students from the same decile to motivate them to keep up.  There would undoubtably still be some type of community in the classroom, just not one so tightly oriented around shared academic values and abilities.

Step one would be the legal banning of afterschool lessons in excess of 3 hours a week.  Step two would be making high school mandatory and enacting zoning in rather than testing in (the funding mechanism ought to remain equal – you don’t want the equivalent of Beverly Hills HS and East Valley HS in Japan).  Step three would be hiring coaches for afterschool clubs so teachers have time for professional development.

Or just to keep schools looking just enough to pass for educational institutions and letting the emergent success of self-selecting student bodies do the work of actually educating.

4 thoughts on “Self-fulfilling prophecies in “high level” schools

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s