Prediction Accountability

The technology buzzword standard for prediction appears to be Netflix and Amazon. Everyone wants to get to where they make recommendations customers will buy. But are these predictions any good?

Out of the slew of emails you get from Amazon, how percentage do you actually buy? How many do you sneer at it and hit delete in disgust that they could get it that wrong? For me, the latter is more common than the former. Certainly it is not from a lack of data, I buy more off that site than I do all bricks and mortar stores excepting groceries combined. (And that makes me re-think how I buy groceries.) Maybe Amazon has too much data that confuses it mixed with correct data. I look things I have no interest in buying such as someone mentioned having problems with a product. Though I have to question Amazon recommending I buy the camera I bought from them a couple months prior.

Netflix really is not any better. Their top 10 recommendations change weekly for me. In my current top 10, one was already rated 5 stars. Another four were already in my queue. The remaining five predicted I would like them between about 3.0 and 3.3 stars. That is out of five. There are 27 items in my queue with higher predictions than these.

Before I start tracking these predictions to gauge how effectiveness, do I even really care? Am I going to stop consuming from companies that overstate their claims? Or should I close my ears when clueless people spout the prediction buzzword? Not really. No. Guess that is what I am left doing.

I think the standard comes not from them being any good. Instead decision makers are aware of them, so they understand wanting to emulate them.

Privacy and Technology

Isaac Asimov has an interesting pre-World-Wide Web quote, “The advance of civilization is nothing but an exercise in the limiting of privacy.” Janov Pelorat in Foundation’s Edge (1982). Think about the word “civilization”. The root, civil, means to treat others well. In one ideal world, everyone would treat everyone else well for no reason. In hunter-gather societies, groups were small because groups were split when the group grew too large. At these small sizes, human abilities to track trustworthiness worked. Farming attached people to the land, making laws and people to enforce the laws necessary.

Privacy is good for the individual. Eliminating privacy is good for the state. Both claim their point of view is good for society. Both are right. A state with too much ability to see into the lives of  individuals will eventually abuse that power to mistreat its citizens. A state with too little ability to see into the lives of individuals will be too blind to protect citizens from mistreatment by criminals. During times of war, people want the state to protect them and the refrain, “Innocent people have nothing to hide,” gets resurrected. During times of peace, people want the state to leave them alone. It has everything to do with trust. When people no longer trust each other, they turn to the state. When people can trust each other, they stop trusting the state.

Ah. Oversimplification

As technology improves, we gain access to tools which allow us to do more with less effort. With information technology, this means we can gather more and accurate information. At the same time, it means less privacy for us.

Part of my work is to provide evidence to deans, department heads, and instructors about student online behavior. Students would be surprised at how much I can find about what they did in our system. Of course, the campus administrations would like us to be able to know everything about what the students and instructors are doing. To get the same in the bricks-and-mortar parts of campus, cameras and microphone would record and store all the audio and video for every classroom and office.

Failure to Improve Online Retention

A former coworker, Cat Finnegan, worked on retention of online students. Recent articles about online students having higher drop out rates than face-to-face caught my eye. Especially the rationales. The recent articles all are about the work done at KSU (whose online class system we host, well one of them: Blackboard) which will be published in the International Journal of Management in Education as “The Impact of Student Retention Strategies: An Empirical Study”.

Online courses cover the same material as traditional classes. The tuition costs are the same and they’re are on the same semester system as bricks-and-mortar classes. But some online students struggle because they can’t keep with the material, get distracted by work or family or miss interacting with professors and other students.

Faculty use different strategies to combat this problem — calling students at home, sending e-mails, even asking students to sign contracts pledging to stay on top of assignments. Campbell and five other professors at Kennesaw State’s Coles College of Business wondered whether these methods work and tested them during the spring 2009 semester.

I am eagerly awaiting publication of this paper. I’d like to understand what actually was tried and failed rather than depend on these summaries.

The former coworker’s paper references a paper called “Seven Principles For Good Practice in Undergraduate Education” by Arthur W. Chickering and Zelda F. Gamson which addresses what are the things which help students. This seems like exactly the kind of thing Campbell et al should identify how their faculty members are addressing each in their online classes.

  • Encourage student-faculty contact,
  • Encourage cooperation among students,
  • Encourage active learning,
  • Give prompt feedback,
  • Emphasize time on task,
  • Communicate high expectations, and
  • Respect diverse talents and ways of learning.

I’m curious if some equivalent was used, how they identified the struggles of the students, and how they determined to try these methods to solve the issue.

Georgia Gwinnett College, another school in our system provided faculty smartphones so students can call. While KSU had the faculty call the students, these seem like similar approaches to improve retention.

Plus Size Soccer Jerseys

Some things are easy to buy online. A few times in the past month I’ve gone looking for a soccer jersey for any of the many football (soccer to my fellow Americans) team I follow. Three futile hours later, I am considering changing which football I follow since stores can carry appropriately sized apparel for it. I’m someone who wears a fairly rare to find size of 3XLT. (Or 4XL when tall is not an option. Or sometimes 3XL for teeshirts, but that usually means exposed back when I sit.)

It annoys me to find a design I like for a reasonable price only to discover there is no size available to me. In bricks and mortar stores, it means never returning until I hear they have a “Big and Tall” section hidden somewhere not obvious.

Here is a place where the Long Tail falls down. According to it, online stores, with their enormous warehouses can better afford to carry a more broad selection of less frequently selling items. They give us more choice. Therefore, it means I ought to find more choice online. For things not in my size, this is true. There is tons of choice. The same stores in a mall who carry clothing in the right sizes seem perfectly capable of offering a wider selection. Yet, an online store like Amazon can’t make it easy for me to find clothing that fits?

My main beef with Worldsoccershop.com is the lack of product in a size I can wear. (The one jersey would make me a Chelsea supporter.) They do get a couple things right.

  1. Quality search: I can put 4xl in the search and get back items with a size of 4xl. All these sites have a search. However, for many sites, size doesn’t appear to be a relevant word. The term “4XL” lands items with “XL” in the name. Useless!
  2. Narrow results by size: Brand, price, and seller are options Amazon offers for narrowing the search to more useful options. How is size not important enough to include? Useless!

Ultimately, I guess not enough people my size have enough interest in soccer jerseys. They end up American football or basketball or baseball fans which have clothes large enough for me. Maybe I should switch sports allegiances? It would help my political allegiances.