Facebook and privacy

Oops, this blog post isn’t about Facebook at all!  Here goes:

Records and interviews show that colleges are building vast repositories of data on prospective students — scanning test scores, Zip codes, high school transcripts, academic interests, Web browsing histories, ethnic backgrounds and household incomes for clues about which students would make the best candidates for admission. At many schools, this data is used to give students a score from 1 to 100, which determines how much attention colleges pay them in the recruiting process.

Admissions consulting companies charge schools tens of thousands of dollars a year to collect and analyze the data of millions of students. In emails reviewed by The Post, employees of Louisville-based Capture Higher Ed urged school administrators to hand over all data they felt comfortable sharing.

“We love data, so the more the merrier,” one of Capture’s consultants wrote in a 2017 email to the admissions director at UW-Stout.

The more the merrier!  And did you know that The New York Times will sell subscriber data about you?

Here is the full article, via the excellent Samir Varma.

Comments

Web browsing histories?

I can envision my startup where I sell you the right web browsing history to get the most points in whatever metric =)

I'm curious about that one, too. It's hard to come up with a legal (or even non-evil(!)) purpose for that information by a state entity like UW-Stout.

Even more interesting is that basically most ways to get a web browsing history (browser side, that is) are not legal.

On the other hand, your ISP is in a perfect position to know your web browsing history. And to profit from it, of course.

My ISP...or Google (Doubleclick, etc), FB, or the originators of any of the dozens of tracking cookies on my computer. That fight was lost years ago. And so far, the ISP's probably are the least evil ones.

The noteworthy part here is that state governments aren't supposed to monitoring what you are saying, what you reading, to whom you are saying it, etc.

'or Google (Doubleclick, etc), FB, or the originators of any of the dozens of tracking cookies on my computer.'

It is easy enough to turn off cookies and javascript. And images, which are also used for tracking. And flash, which almost seems designed to defeat all the previous methods.

But you are right, that fight was lost decades ago.

Respond

Add Comment

Colleges must do studies of which applicants are most likely to donate decades later to the endowment fund. Have any of these models ever been revealed to the public?

I suspect these data help explain why Harvard admissions loves white minor sport jocks so much.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

"most ways to get a web browsing history (browser side, that is) are not legal."

I just installed a new browser. Now it seems that all (free or otherwise) add-ons apps want the right to the browsing history. I declined them all. Some of them claimed installed base of multi-millions.

Respond

Add Comment

Respond

Add Comment

A VPN costs $5 per month. You can vicarioiusly log in from any country around the world with, for example, ExpressVPN. It won’t fool the NSA nor the GCHQ, but I think it’s sufficient for keeping your web history away from your ISP and colleges. In my experience you can do anything, including video, with little disruption, via a VPN, with the exception of Skype.

For extra special protection, use Tor Browser, which you can tunnel through the VPN, i.e., use when the VPN is on. There are some uses that are impractical with Tor’s bandwidth, or where it’s prohibited, notably Google.

'A VPN costs $5 per month.'

And already identifies you as a higher value monitoring target, though nothing like being a Tor node. And of course, you are trusting your VPN to not be passing along or selling your browsing history data.

But sure, there are ways to tighten things up in comparison to a normal user, and no reason to dismiss such things out of hand. Even simple things, like using a non-standard browser and OS (SeaMonkey and Linux is a fine combination) is a step up - and one that very few people will ever make.

Respond

Add Comment

I think the point is not hiding what you do, but projecting a positive image.

Similar to double bookkeeping, one part of the job is hiding the real numbers, while another part of the job is promoting the cooked numbers.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

We do the same thing to decide who to date, so what's the big deal?

Really! I admit I'm not in the dating scene and haven't been for 32 years but I always made my decision by talking to the person.

Respond

Add Comment

You looked at Mark Antonhy's browsing history? What did you find there that you liked?

Drug fueled orgies, one assumes.

Well, at least that is what the series Rome suggested, in tasteful detail.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

There have been a number of comparisons offered here in the comments, between the Chinese state panopticon and the American multi-vendor version.

Some were over the top, but the fundamentals seem pretty straightforward.

Without privacy law, privacy is well and truly dead.

Privacy law is worse than useless. All it does is makes it impossible and bureaucratic to do the most basic functions, all the while making it legal for government to gather everything they want.

A simple beating of the applicable college administrators in a back alley somewhere would solve the problem within a week.

It is a very shallow complaint that an entire category, an entire galaxy, of solutions must be wrong, because some are.

And yet we see it every time.

Respond

Add Comment

By the way, for actual expertise I would outsource to:

https://www.eff.org/issues/privacy

What would this hypothetical law say

From my reading of that page, there shouldn't be just one. That would create nightmarish complexity.

But perhaps there could be a number of simpler laws, limiting data retention, for instance.

That’s extremely vague.

Do you have a more fleshed out idea of legislation for this.

Actually, no. The nice thing about outsourcing is that you can leave it to domain experts and lawyers to address.

The Campaign against Established Knowledge and Why it Matters

Ideally of course we'd all be able to read those expert proposals at a good "explainer" website.

So you have a vague non-proposal which would necessitate a massive regulatory apparatus costing businesses potentially hundreds of billions in compliance costs.

But there’s no there there, so we’ll never know. There’s nothing to debate pro or con. Just a drive by link and a snide meta comment and another link to Amazon.

Why?

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Sounds like yet another advantage for all the little club-attending careerists who game the system already.

Respond

Add Comment

'this blog post isn’t about Facebook at all'

It is about the data privacy nightmare that is modern America, right?

'... employees of Louisville-based Capture Higher Ed urged school administrators to hand over all data they felt comfortable sharing'

This would have been so illegal in West Germany even before the GDPR. As for East Germany, well, the DDR was unconcerned about who felt comfortable when sharing data.

'And did you know that The New York Times will sell subscriber data about you?'

Um, didn't everyone know that subscriber data - and not just from the NYT - has been sold for decades, long before the World Wide Web existed? Next, will we learn that direct mail companies buy data from a variety of sources?

Respond

Add Comment

As higher education enrollment continues to fall, universities are going to have to work harder to reach prospective students. One way to do it is through better targeted marketing. So go long on companies like Capture, I guess is what I'm saying. Facebook can still eat a fart, though.

Respond

Add Comment

I suppose if it were possible for me to hold the people who play the ridiculous game that is college admissions in lower regard maybe I would?

As for privacy - I’ve long given up on it. It’s liberating in a way to assume everyone knows the worst about me and dare them to do something about it.

Respond

Add Comment

well, also need a sophisticated system compiling detailed personal data on all college professors, teaching staff, and administrative staff.

Student/parent consumers need good data to choose which college services are worth the heavy time & money investment.

Respond

Add Comment

"Oops, this blog post isn’t about Facebook at all!"

Why did you bring Facebook into it then? The article is raising awareness about privacy problems in college admissions, which is perfectly consistent with Facebook being a problem too.

I hope it isn't as facile as "see, everybody does it, so it's fine."

Respond

Add Comment

"Why did you bring Facebook into it then?"

Analogy, between obvious and not so obvious personal data exhibitionism.

So why is the analogy marked with the troll-flag "oops!" when we know that even tweets don't work that way? I come here for Straussian insights and all I hear is dog whistle.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

It is written: "Come out of her, my people, that ye be not partakers of her sins, and that ye receive not of her plagues."

Respond

Add Comment

'“We love data, so the more the merrier,” one of Capture’s consultants wrote in a 2017 email to the admissions director at UW-Stout'

Well, of course a large quantity of low-quality data is better than a lesser quantity of high-quality data. Isn't it? Umm, well, is it?

The large quantity may be better if you're just looking for a quickly disqualifying "gotcha," but perhaps not so much if you're actually interested in evaluating the candidate?

Respond

Add Comment

Nobody is more proficient at using digital data than the Republican Party: https://www.nytimes.com/2019/10/16/opinion/trump-digital-campaign.html In large part the Republicans use Facebook data. Do read the link. It's not about typical data mining, but identifying where one is (where one is says a lot more about one than I realized.)

Respond

Add Comment

They actually called their company "Capture Higher Ed"?? Christ, reality's war on satire knows no bounds...

Respond

Add Comment

Still waiting for a single brave college to step up and evaluate students based on transcripts, scores, activities and written recommendations -- without accessing the student's name, address, ethnic identity, gender identity, social media and family wealth/connections. It would be fair. It might begin to ameliorate what is starting to smell like rank corruption and the sale of credentials.

What we have instead are admissions departments staffed by people who know little and who are influenced by calls from college administrators, college fundraisers, alumni departments, college coaches and, effectively, references to various the social registers and lists of national and international political luminaries..

After years of sending big money (for me) to my almae matres, I am giving less and only to funds that are 100 percent devoted to scholarships. Even that makes me feel guilty because it enables colleges to keep bumping up tuitions beyond already-unconscionable levels.

Instead of waiting (you may wait for long, I'm afraid), you can move to one of many other countries. In France, for example every college or "Grand École" evaluates students based on transcripts and scores only (often scores only, because transcripts are thought to be too school-dependent), without accessing activities, written recommendations, the student's name, address, ethnic identity, gender identity, social media and family wealth/connections.

However, from someone who has made the move in theater direction, I would not recommend that without serious thinking. The way American colleges choose their college is awful, but they have other advantages. 15 years after, I am still wondering if it was a right move, and do not have a clear answer.

"in theater direction" -> "in the other direction"

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

One of the reasons we don't care about privacy is that we bundle together all sorts of stuff and call it "private data". Test scores, zip codes, ethnic background, etc., are usually data people don't give a crap about. Yes, please share information you could find out about me anyway. Even "browse history" is misleading here (yes, I am fine with companies grabbing my history as long as it is not attached to any particular logged in user).
At the end of the day, the problem is that we have people who don't know a thing about tech trying to regulate tech.

Respond

Add Comment

If only we could set up an auction market for admissions, like Google AdWords, to match price charged by colleges to student interest (which is probably what they want out of the browser history).

As a parent of one in college and one a year away from applications, I'd appreciate that greatly.

Respond

Add Comment

WaPo, where Bolsheviks use capitalism to fund the overthrow of capitalism.

Respond

Add Comment

Respond

Add Comment