Monitoring Nowadays

The longish piece on data analytics in the workplace is interesting but to me it's also frightening.  I thought it makes for a good counterpart to our discussion of monitoring in the Shapiro-Stiglitz model.

Cashing it in a little closer to home

For those who were there today I appreciate your commitment and effort.  You may not know this unless you have an older sibling who also went to the U of I, but we used to have classes scheduled during the week of Thanksgiving - through and including Wednesday of that week.  Of course, attendance would drop.  If you were teaching on Mon - Wed - Fri, as we used to do things, attendance on Monday would be lower than normal and attendance on Wednesday would be quite low indeed.  So they changed the Academic Calendar to give the full week off.  Today's attendance gives some evidence that there is an "endgame problem" even now, an example of what happens in repeated Prisoner's Dilemma with a known last period.  All I can add is that I'm glad our class doesn't meet on Friday.

Now that we have the model of repeated Prisoner's Dilemma to describe both what the cooperative solution looks like as well as what the solution looks like when everyone cheats, I thought you might be interested in considering college education from that perspective.

The following quote is taken from the piece cited below.  It is about something called The Disengagement Compact.  It is not uplifting to read about.  But ask yourself whether it is descriptive of some (perhaps many) of the courses you've taken.

What We’re Learning About Student Engagement from NSSE: Benchmarks for Effective Educational Practices. By: Kuh, George D.. Change , March/April 2003, Vol. 35 Issue 2, p24-32, 9p;

And this brings us to the unseemly bargain, what I call the "disengagement
compact": "I'll leave you alone if you leave me alone." That is, I won't make
you work too hard (read a lot, write a lot) so that I won't have to grade as
many papers or explain why you are not performing well. The existence of this
bargain is suggested by the fact that at a relatively low level of effort, many
students get decent grades--B's and sometimes better. There seems to be a
breakdown of shared responsibility for learning--on the part of faculty members
who allow students to get by with far less than maximal effort, and on the part
of students who are not taking full advantage of the resources institutions
provide.

I will add to this that in my observation many students seem to feel they've been tasked to memorize the instructor's lectures and to more or less confirm that expectation many instructors write exams that reward such memorization.  So this is how the Disengagement Compact may play out - a feigning of learning with little of substance actually happening.  To the extent it is widespread, it suggests massive cashing it in.

I don't want to end on a downer right before the holiday, so let me try to tie learning in a different way to being successful on the job.  The meta skill that all good students should acquire is the ability to satisfy their own curiosity to a substantial degree.  Different students might come up with alternative ways to do this.  The more extroverted might do it by networking with people and feel comfortable going to the right source who can then steer them to the information and answers the student is looking for.  The more introverted might do it via their own self-initiated investigation. Both approaches have merit and probably people need to learn how to do it both ways.

In class on Tuesday, we talked about an intern or a new employee taking the initiative and going beyond the work that was assigned.  I brought up that there is an issue about how to identify appropriate work - work that the worker has the competence to do well and work that the company will value when it has been completed.  So there is a puzzle in the identification and its exactly that sort of puzzle that should make the intern curios.  The meta skill described in the previous paragraph fits this situation to a tee.  It is not something that is acquired overnight.  There should be a lot of practice with it.  College is a good time for that.

Have a wonderful holiday and I'll see you back in class on Tuesday December 3rd.

The Last Excel Homework on the Shapiro-Stiglitz Model Is Now Available

You can find it here.  Apart from the economics, one wrinkle in it is the use of Comments (little red triangles in the upper right corner of some cells).  To access the comment simply mouse over the cell with the red triangle without clicking on it.

This homework is due Wednesday December 4.  If you have questions on it please post as a comment to this post.

Something cutsie and fun

It occurred to me after class that I didn't spend time on how one might create a bonding experience for others.  Managers sometimes need to to this for their staff, so it is an important thing to consider and think through.

This is one example I was involved in.  It is a seven minute video done in summer 2007 for the Educause Learning Technology Leadership Program, which that year was held in Madison Wisconsin.

Here is a little setup of what is in the video.  Heather Stewart of NYU and I were leading a session (there were 50 attendees from campuses around the country) on "presenting data."  I had not been involved the previous year with the LTL program, but Heather was and did a similar session then.  It didn't go well so she wanted to try something new. We discussed what that might be over several months.  Finally in one call we each shared our experiences of presenting in front of the Provost and how frustrating that was.  So we hit on an idea of a video to show such a presentation - one full of mistakes.  The ideas was to get the audience to enjoy the video for itself and then after to deconstruct it, find the errors, and have the audience offer up better ways to go about things.

In the video, Heather plays the learning technologist who will be making the presentation.  She is very good and "emotes" well.  I play her boss - the Campus CIO.  I'm something of a jerk in this movie and kind of throw her under the bus rather than prepare her properly for the presentation.

There are two others in the video.  The Provost is played by Perry Hanson, who was then the CIO at Brandeis (he has since retired).  He calls himself James Dean in the movie.  He ad-libbed that.  The other is Kathy Christoph who plays the Provost's budget officer (CFO).  Kathy was the head of Academic Technology at UW-Madison at the time and hosted the program.  (She too is retired now).

The conference started on a Monday afternoon and there was a banquet that evening.  We shot the video after the banquet to use in a session on Tuesday - so there is a just-in-time aspect to it and it is certainly a low budget thing.  Heather and I let Kathy know about it a week or two ahead of time because we needed Kathy to provide logistics support to make it work.  We only told Perry about it, however, at lunch on Monday just before the Institute started.  He was delighted to help and went with the flow.

There is a little bit of hiss in the video but otherwise it plays reasonably well on my PC using Windows Media Player.  I just tried out the link and it works that way.  If you right click on the link and launch the application requested it should play.

mms://winstreamer.doit.wisc.edu/ltde/provost512K_Stream.wmv

This is the link to the Mac version.  I don't know if it is live or not, but you can give it a try. 

rtsp://qtstreamer.doit.wisc.edu/ltde/provost512K_Stream.mov

The audience really liked the video (perhaps because it is so campy) and the session went really well.  Indeed, even though Heather ended her participation in the LTL program after that year, they continued to use the video for the next two years, where Malcom Brown and I lead the session.  Malcolm was fine doing that even though he didn't appear in the video.  It worked pretty well in those later sessions too.

One last comment - I know nothing about video production.  But it was my idea to draw the clock on the white board and make that part of the movie.  

Enjoy.

Grades Have Been Uploaded in Moodle

I uploaded the Midterm 2 grades into Moodle as well as for Excel Homeworks 5 and 6.  There is also a histogram of the exam scores in Moodle.

I have no yet updated the blog post tracking.  I started to poop out doing the clerical work and better to not make errors.  So I will do it tomorrow before class.

Also, I haven't decided about another bonus.  I will leave that for another day.

There ultimately will be only one more Excel homework and two more blog posts.  That will give us 10 blog posts, which was my goal at the outset, but I originally planned for 10 Excel homeworks.  So we are clearly falling short of that and some grade adjustment needs to be made for that to get my scale to work properly.

I will return the exams tomorrow in class and we will briefly review the text before getting into our discussion about reputations.

Reminder - Blog Post Due Tomorrow

Here is the prompt:

The topic is personal reputations and their role in influencing behavior. Describe some domain where you have a strong reputation with others (it could be with friends, it could be with your family, it could be at some place you worked, etc). Then discuss how your reputation developed. Consider what you do to keep your reputation intact or enhance it further. Finally, reflect on whether there are occasions where you'd like to stray from the behavior suggested by your reputation and what you do on those occasions. Have you ever "cashed it in" by which I mean you abandon your reputation altogether in favor of some immediate gain?

Sorry about the technology snafus today

Below is the graph from the Excel embedded in the PowerPoint.  It has the IC constraint explicit.  It is  the dashed upward sloping curve in black that is to the left of the main diagonal.  To satisfy the constraint you have to be on that curve or to the left of it.

Note that the IC constraint goes right through the cross point of the two indifference curves.   That means you don't really need all three curves to determine the optimal contract that induces high effort.  You only need the high effort indifference curve and one of the other two curves to determine the solution.

Add caption

Link to Test from Last Year and Candidate Essay Questions for Exam

Last year's midterm 2.

Candidate Questions

A.  This question concerns procurement.  Discus why in government agencies procurement is subject to  regulation.  What is the regulation aimed at preventing?  What does it try to accomplish?  Explain why during the bidding process the lowest price bidder may not be selected, although the process is competitive.  Discuss the various goals the agency conducting the procurement may have going into the process.

B.  This question concerns conflict in organizations and, in particular, Model 1 and Model 2 of Schon and Argyris.  Explain how a manager can without intending to do so cause the staff under him to be angry with his management approach.  What is it that causes their anger?  What are the consequences for the unit when the manager behaves this way?  Explain the Schon and Argyris recommendation about how the manager might diffuse the tension.  Why does their recommendation have a chance to be effective?

C.  This question concerns an agent who while performing the same function serves two different principals.  What are some examples where this situation emerges?  What problems are created that don't exist in the case where there is a single principal?  Explain what "capture" means and how it arises in this context.  What might be done to discourage capture?

Review of Tuesday's class session

It is unusual for me to do this so here is a brief explanation why.  The content of Tuesday's class is fair game for the essay part of the next midterm.  Some of it, both the triangle principal-agent and the stuff about procurement, you won't find in either textbook.  So for those who missed you need to get some exposure to that stuff.

I am doing this from recall of our class discussion.  That may be imperfect.  If you read this and were in class yesterday and you find something important is omitted below, please make a comment to this post that includes those ideas.  Also note that the order of things as I present them may not completely jive with the order in class yesterday.  Such are the pitfalls in doing this entirely from memory.

* * * * *

The Triangle Principal-Agent Problem

I took this by borrowing from the economics of regulation, where there is something called the capture theory.   This theory was posed by Chicago School economists as a different critique of regulation.  It said that regulation can be ineffective, even if it is desirable, because the regulator can become captured by the industry it is supposed to regulate.  As an example, I suggested that perhaps Congress has been captured by Wall Street.

The example was meant to illustrate that capture happens mainly by means of bribes.  In the case of Congress this would be via substantial campaign contribution or other perqs given by industry to members of Congress.  We also discussed the "revolving door" problem where regulators get offered plush jobs by the industry after they step down from government,  though we also noted that in some cases such hiring is efficient because the person has expertise that is otherwise scarce.

We then qualified the circumstances where the triangle problem arises.  These cases don't fit all the criteria needed to generate the triangle problem.

(1)  There is an employee who has a boss and the boss has a boss, so the employee is indirectly under the boss's boss and may sometimes take orders from the person two rungs above.  In this case it is possible that the orders from on high conflict with the orders from directly above.  The example we gave is where a TA may say some things that contradict what the instructor has said.  Who is the student to follow?  This is not a triangle problem because it can be resolved by realignment happening between the instructor and the TA.

(2)  There can be triangles but in non-work situations.  Love triangles make for good fiction.  Triangles within a family where the parents disagree about what the child should do are a more relevant example.  The triangle is real enough but we wouldn't consider this a situation with a principal and an agent.

(3) An employee tends to a customer's needs, which is just what the employer wants the employee to be doing.  There is no conflict here so no triangle problem.

The triangle emerges when the employee tends to the customer's needs in a way that goes beyond what the employer wants.  In this case we might say the employee is captured by the customer.  Whether capture happens or not, the employee is tugged in different directions, one suggested by the employer, the other suggested by the customer.  I said this feeling of being tugged in different directions is the reality in many work situations, even if the setup is not exactly as it is here.  The reason this is worth discussing is because it is so typical in many lines of work.

We then discussed how capture might happen in this case since it typically won't happen via bribes. We gave three different possibilities:

(a) the employee becomes empathetic for the customer's plight and provides help for the customer as it is the decent thing to do, even if it is not the profit maximizing thing to do.

(b) the employee gets bullied by the customer and rather than cause a to-do acquiesces to the customer's demand because that is the path of least resistance.

(c) the employee gets enjoyment (think of it as a consumption benefit) from serving the client.  So doing this sort of work is for the experience itself.  It is capture when the employer doesn't see how the benefit from that experience can be internalized for the organization.

Procurement

I gave a little setup where the RFP for a major piece of software, perhaps costing several million dollars and still kind of new to the marketplace, had just concluded.  The winning bidder had been determined.  This company was not the market leader and was chosen in large part because it was thought the company would be more attentive to the customer than the market leader would.

I said that one of the big issues was to determine where the preferences of the buyer and seller were aligned and where they were in conflict.  The procurement process would encourage the aligned part to work well and for the non-aligned part there needed to be other mitigation by the buyer to make things work okay.

We started off with the question of whether you (thinking as the buyer) want a long term contract or a short term contract, noting that you are risk averse and are trying to use the process as a way to reduce risk.

One student responded that you want a short term contract because if the product proves a lemon then you can go with another seller after that.  This is certainly true in a retail setting, but it is not the right way to think about a major procurement.  Once you buy from a company there is substantial lock-in, meaning the costs at the back end, getting staff trained on the software, and having the thing up and running are quite large, so you don't want to incur that again.  Further, even if there are problems with the software, users adjust to it.  That is costly for them.  They don't want to have to go through the adjustment process again. So the procurement is a commitment to that particular product, at least for a while.  The question is then are there other things that can be done to reduce risk given that you are committed to the product.

A long term contract, even one that allows for some price increases in the maintenance portion (this allows the buyer to get upgrades in the software) and the service portion (this allows the buyer to get help when there are problems) offers the benefit that out year expenditure on the software can be predicted.  Knowing what you will spend up front is a big way to reduce risk.

We then discussed whether you want to get bargain basement prices or if you're willing to pay more.  I argued that what you want to avoid, if you can, is the company going out of business or it getting taken over by the market leader.  Either of those outcomes will likely mean that your software will reach end of life and those dreaded switching costs will have to be incurred.  At bargain basement prices the company is not making money, so going bankrupt is a possibility.  I didn't say this in class, but at too high a price, the company starts to look like a cash cow and that encourages a takeover possibility.  So what you want is some middle ground.

I then made a brief departure and talked about a management practice called "No Surprises" which means if you have bad news get it out when you learn about it; don't wait till the last moment when it will get out on its own.  The reason for this is that users will have time to adjust to the news.  Users will not be happy at first and as the messenger you might bear the brunt of their ire.  But they will get over it.  Getting the bad news out early is a way to maintain trust.

In Higher Ed and in good government, you might expect No Surprises to be the norm.  (The bad example set by the cover up after the Watergate break in giving all the justification that is needed.)   In the business world, perhaps not.  We talked about the case where the vendor (the winner of the RFP) has software that is buggy, perhaps in a way that the client won't discover for some time yet.  The vendor wants to use the client as a referral for potential customers who are considering buying the product.  If it is known that the software is buggy, that might curtail such sales.  So the vendor has some incentive to hold back that information, even if the client would benefit having it for the client's planning.

We discussed what mitigation a client might take to deal with this concealment of information.  Somebody suggested talking with other vendors.  That may have some benefit, but is likely not to tell you what is really going on with the software.  The competitors won't have detailed knowledge of those issues.  We then discussed the possibilities of forming a buyer consortium for the purpose of sharing information about use experience.  I didn't say this in class, but if the use experience suggests a common complaint, then the group as a whole can act in concert and bring the complaint together.  There is more power with the group than with each buyer acting individually.  If the vendor is to put significant resources into resolving the problem, it is more likely to happen this way.

We concluded this part by discussing that the product will evolve over time and determining what the buyer's preferences are in regard to that.  We said that the buyer wants to influence product development in a way where the buyer benefits most.  That means that the when the buyer prioritizes feature changes it wants the vendor to seriously consider those preferences.  We said that while not all changes can necessarily be done in one product development cycle, the vendor actually wants the buyer (and like minded other buyers) to participate in the product development so the product evolution is in accord with what buyers want to see.  This helps to make the clientele loyal and enhance customer satisfaction.  So it is win-win for the seller to involve the buyer deeply into product development.

B&D Chapters 7 and 13 - how to really motivate employees to produce at their utmost

I started this part out with a priming exercise.  I asked students about activities they enjoyed.  I reminded the class that earlier in the semester somebody said they liked to cook.  Many students concurred.  I also said that one student had mentioned enjoying going to avant-garde films at the Art Theater.  A few other students indicated a similar preference.

Then students suggested other activities they enjoyed.  A partial list includes tennis, working out, reading, and driving cars fast (at a raceway).

I then asked whether these activities commanded their full concentration.  There was some agreement that reading a good book and watching an avant-garde film can be completely absorbing.

Because I knew that one of the students was a pilot, I asked whether piloting an airplane was completely absorbing.  He responded that it was like driving a car.  I then I said that driving is completely absorbing when you take driver's ed, because you are still learning how to control the car and that takes concentration.  When you have mastered that, mostly driving becomes an autonomous process, meaning that your mind can be elsewhere and the driving takes care of itself.  You do have to monitor more closely when the roads are slick or when there has been an accident that you need to navigate around.

I asked: for these enjoyable activities would students just do them on their own or did they need some encouragement to do them?   One student mentioned that for working out sometimes it is hard to initiate.  Afterwards he is happy to have done that, but even with knowing that ahead of time, sometimes he wouldn't go to work out.

I responded that much of this is about habit formation.  If you have a routine to do things then you will initiate with them because you have developed the habit to do so.  If you don't yet have the habit formed and must choose to do the activity or not, sometimes initiation is difficult.

At this point I specifically mentioned immersion into the activity.  Somebody then responded that video games are an activity that produces total immersion.  We discussed video games for a bit after that.

Then I said we were ready to discuss Bolman and Deal.  I wrote three words on the board: motivation, intrinsic, and extrinsic.  Intrinsic motivation comes from inside the person.  Bolman and Deal argue that an employee is much more productive when the employee has intrinsic motivation for doing the work.  Extrinsic motivation comes from outside the person.  Economic incentives are a kind of extrinsic motivation.  Put a different way, the issue is whether it is the work itself that is important to the person or if it is what the person earns that is of primary importance.  In some sense, this is like the essay question from the first midterm.  Bolman and Deal come down very much for the third alternative - provide intrinsic motivation for the the employee via giving the employee a kind of ownership in the organization. Chapters 7 and 13 can then be seen as how to's for getting that done.

I said there are really two different things to be done to encourage intrinsic motivation.  One is to get rid of distractions that impede being able to concentrate on the work.  The other is to empower the worker to take ownership.

I then gave a personal view - my own powers of concentration have diminished with age.  A big source of distraction are the aches and pains I have, primarily from arthritis.  I asked the class to come up with other types of distraction that exist at work.

One student said that social networking a la Facebook can be a distraction and there could be a rule at work that employees can't access such sites with their computers.  I responded that B&D would be against this approach.  Rules of this sort are treating employees like children.  B&D say that itself is a distraction and it is better to not have such rules and leave these things to the discretion of the employees.

Another student suggest that conflict in the workplace can be a distraction.  I agreed but tried to clarify this a bit.  Conflict of a personal nature really should be nipped in the bud and is to be avoided.  Conflict about how to get the work should be done is healthy and to be encouraged.  Such disagreements can lead to a better understanding of what the work is about and produce a new synthesis on how people should approach their jobs.

We then talked about pay.  I asked when pay can be a distraction.  Students agreed that happens when people feel they are underpaid.  In fact, B&D argue that pay should be generous to avoid this issue.  I argued that it is a little bit harder than this.  An employee who is part of a highly successful project that become visible to various competitors might very well find job offers coming in because these rival companies will want to see a similar sort of project happen at their own firm.  At this point what had been very generous pay no longer is.  So some reconsideration of what the employee earns must happen if the employee is not to go elsewhere. Until that concludes the employee may again feel underpaid.  The issue happens periodically.  In between pay is out of the employee's mind and the employee can concentrate on the work.  When pay becomes prominent you can think of that as a way for the employer to re-initiate the employee into the work.

We turned to how to empower employees.  B&D argue for democracy in the workplace.  This implies a flat org structure rather than hierarchy.  It also means that employees can on their own change how work gets done, with an eye to making the organization more productive.  A student pointed out that this can lead to coordination issues, when all employees are so empowered.  Indeed it can.  I responded that there are trade-offs here.  But B&D argue that most American companies err on the side of top-down approaches.  They may give lip service to allowing employees to contribute their ideas, but in practice don't encourage it.  Note the parallel in this B&D argument with them also saying (in Chapter 8) that Argyris and Schon's Model 1 proliferates among upper level executives.  A democratic workplace really demands Model 2 to be in place.

We then talked about executive pay and executive function in this context.  B&D argue that very high multiples of executive pay to pay for the lowest paid employee are inconsistent with a democratic approach to the workplace. They say that such multiples are as high as 400 for some American companies.  They recommend much lower multiples, say no more than 10.  They also recommend that the executives be treated just like every other employee - no executive dining room, required to make a new pot of coffee when they've had the last cup from the previous pot, fly coach instead of first class, etc.  In other words, in a democratic workplace executives don't take some of their compensation in the form of job perqs.

I gave a different sort of argument for keeping multiples reasonably low.  Employees want to know that the executives are in it for the long term and are not trying to take as much as they can out of the company in the near term with the intent to gut the company (lead it to bankruptcy or to a takeover of some sort).  If everyone is in it for the long haul then employees are willing to contribute their part, but otherwise not.

We concluded with a brief discussion of the symbolic approach and the ideas in chapter 13.   To get the class into the right mindset, I asked whether any of them had been in stage productions while in high school.  The aura that happens in such a production is what B&D have in mind in chapter 13.  I asked about whether students had some disagreement in the rehearsals leading up to the production.   Students responded there were some.  But they were able to get past these, because people were willing to surrender themselves to the larger goal - making the production a success.

We then talked a bit about friendships forming in such a setting.  And then that partly to blow off steam and partly because that is what you do with friends, joking around happens and other type of play becomes part and parcel of doing the work.  B&D stress certain rituals become a way to encourage making the group really tight.  They also say that the group will come up with a language just meant for members to describe the group activities.  In other words, it is not just work.  The experience needs to be amplified to make it like a club that people want to join.  Clubs have rites of passage.  They exist to make make club members closer to one another and to respect each other more.

That concludes the review.  If you have questions or comments about it please post them here.

Because attendance was low in class today...

I'm posting an announcement I made there.  This regards our upcoming exam to be held a week from Thursday.

On Thursday morning before class I will post
(1)  Last year's midterm 2 and
(2) Three candidate essay questions.  One of those will be on this year's midterm.

Our review session is a week from today.

Also, as a quick summary of what we covered today we discussed:
(a)  The blog posts and the underlying issues with the triangle principal-agent problem.
(b)  More about procurement.
(c)  B&D chapter 7 and 13 on how to really motivate employees to be highly productive.

I view this stuff as the most fun part of the course.  I'm sorry that many of you missed.  I may write up a long post  to highlight the points we discussed.  I don't really want to encourage people to miss class, but some of this may be on the exam, so you should be prepared for that.