Sep

28

Testing in the schools starts in a few weeks. We’ve been writing a computer game to teach math. Here is a look at some of the scenes.

artwork from the game

The poster is thanks to one of the awesome members of our graphics team, Justin Flores. You can find him at Jflo Productions.

Of course, we will be using statistics on the back end to see what parts of it students play the most and correlating that to student outcomes.

Sep

25

Las Vegas at duskLet’s say that all you knew about probability were some basic rules. Would it be enough to convince you that gambling is a bad bet? I think so.  Let’s begin with these four:

Probability of something not happening = 1 – the probability of that something happening

P(Not X) = 1 – P(X)

Since the total of all probabilities = 100%  , that is, there will be SOME result, and the probabilities of something not happening and the same something not happening are mutually exclusive, the probability of something NOT happening must be 100% minus the probability of it happening. (This relates to the next rule).

The probability of either of two mutually exclusive events is the sum of their individual probabilities

P(A or B) = P(A) + P(B)

Notice that the key phrase there is mutually exclusive. Here is where people often go astray.

There was the famous case of the gambler who bet that a six would come up at least once in four throws of a die. A person misapplying this rule would say the probability is

2/3 or 4/6 that this would happen.

That is, your odds are 1 out of 6 possibilities on the first throw, the second throw, the third throw and the fourth throw.

These are not mutually exclusive probabilities. A person can throw a six the first time and the third time. So, no, this rule won’t work.

For independent events, the probability of any combination of them is the product of the individual probabilities.

That is,

P( A and B) = P(A) * P(B)

Let’s take our gambler again, written about in a free and highly recommended book on probability by Grinstead and Snell.

So, the probability of NOT getting a six is 5/6 – since there are 5 sides on a die that are NOT a 6.

The probability of not getting a 6 four times in a row is

5/6 * 5/6 * 5/6 * 5/6  = .482

Since  48.2% of the time you won’t get a 6, that means 51.8% of the time you will. The odds are in your favor to bet on getting a 6.

So … why did the gambler lose money? This, my friends, brings into consideration,

The Law of Large Numbers

The best description of the Law of Large Numbers I have read comes from wikipedia

“It follows from the law of large numbers that the empirical probability of success in a series of Bernoulli trials will converge to the theoretical probability.”

A Bernoulli trial is an event that can have one of two possible outcomes, and which outcome is obtained is determined at random. Sounds like throwing a die and getting either a 6 or not a 6.

So, over the long run, the larger and larger number of trials you have, the closer you will get to the theoretical distribution, in this case, getting a six 48.2% of the time.

Incidentally, for all the trashing academics do of wikipedia, I have found their sections on statistics to be extremely accurate and well-written. Whoever the people are who write it, kudos to them.

So, if we had an infinite number of trials, we would always end up with 51.8% of the time a 6 coming up on four throws of a die.

Here are the results from someone who bet 100 times on a binary outcome, with a distribution of 10,000 trials. That is, assume you went to Las Vegas for 10,000 days in a row and every day you made 100 bets.

Graph showing distribution of 10,000 trials

Two take-away points:

  1. The house always wins because the casino makes a LARGE number of bets. If you count every time someone pulls a slot machine or bets on 21, it is hundreds of thousands of bets a day. If they are taking the equivalent of that 51.2% bet, over a large number of trials, they will always win. If the odds are 50%, as above, they are going to be centered around 50. If, as on most bets people make, the odds are more around 52% , in the house’s favor, the distribution is going to be centered around 52%. More than half of the time, the house wins.
  2. People will say things like, “Yes, but what if you are the one on the end before the house starts to win? What if you are that one person out of the 10,000 where the die comes up with a 6 only 30 times out of 100? You’d win big, wouldn’t you? That’s possible, isn’t it? Yes, it is possible, but it’s not the way to bet. The odds are not zero, but they ARE against you.

So, you now you know why statisticians are no fun in Vegas. In fact, I was at a conference not long ago where someone at the casino told me,

“We hate you people (statisticians). Everyone always makes less money on the statistics conferences than any other event because they understand the odds, so they very seldom gamble.”

Sep

23

How (not) to be annoying

September 23, 2012 | 2 Comments

Okay, young people, the time has come for us to have a serious discussion.

DiscussionWhen I was young, I was very annoying. Now that I am old, I am frequently annoyed, proving thereby that God has a sense of humor. I think pretty often the people who are annoying me are not doing so deliberately. In fact, they often want something from me, whether it be a job, a grade, an internship, assistance with a technical problem or a consulting contract, so it would be counter to their own self-interest to deliberately annoy me. And yet ….

So, here, as a private service to me and a public service to anyone else like me, are Dr. AnnMaria’s instructions on how to be annoying. If you’d like not to be annoying, just cut out everything below. If you’d LIKE to be annoying, on the other hand, feel free to carry on as you are, and have a nice life, just don’t have it around me.

seal swimming

  1. Be simultaneously vague and demanding. “This computer is totally screwed up. I’ve never seen it do this before. I need you to fix it!” Do you know why when you call technical support they ask you a litany of questions beginning with what kind of operating system you have and whether the computer is turned on? Because specific solutions can only be given to specific problems. What exactly does, “I’ve never seen it do this” mean? Does it mean your monitor turned into a seal and swam away? I certainly have never seen THAT. If you have a problem, be very specific about what you have done and how the obtained results differ from what you want, e.g., “I was using the tabs in the jQuery-ui library and it worked fine in Safari as far as when I clicked on the tab, a new page came up, as expected, but when I tried the same page in Firefox, when I clicked on the tab, nothing happened.”
  2. Be helpless. Whether it is an employee,  a student or just a random acquaintance, you will be less annoying if you at least make some effort to fix the problem. Try Google. I have heard that you can find lots of things on the Internet. You know how I solved my jquery ui problem? I read a book.  Before you ask someone else to fix your problem, unless you PERSONALLY are paying that person WELL, you should probably try the simplest things – turn the computer off and on, make sure you’re connected to the internet, use Google, look in the manual, check the settings on your computer, read the textbook – you get the idea. Of course, as far as clients, I bill by the hour. If you want to call every fifteen minutes and ask what time it is, I can have someone in my office sit by a clock, take your calls and tell you the time. Unfortunately, when the bill comes in, your boss may have something to say about it.
  3. Tell everyone to use Google.  The Rocket Scientist walked by and suggested this. He mentioned those people on forums who will answer every question with, “Why don’t you just use Google?” and then post a link saying, “I found this in eight seconds.” This post is often followed by the original poster responding, “Yes, I read that link and it was completely unhelpful.” If you don’t have any good advice, just shut up.
  4. Answer every question, whether you know the answer or not. If you don’t have any good advice, just shut up. One of the MOST annoying things to me is people who post answers on a forum who are just WRONG. If you don’t know, don’t say anything.
  5. Convince yourself that criticizing what someone else made is just as valuable a contribution as making something, because “You made it better”. This is probably my GREATEST annoyance from young people. I write programs that run. They do stuff. Little guys move across the screen, standard errors of estimates are produced, or whatever. We write other programs for clients to their specifications. I write grant proposals that start with me sitting down at a word processor and opening a vein; 225 pages later we have a needs assessment, literature review, design, preliminary data, budgets, person-loading charts. It takes me about 200 hours over two months and three bottles of Chardonnay. If you come in on it during the last week and point out that I left out a semi-colon on page 38, have square bullet points on page 67 but round ones on page 115 or make a suggestion for the color scheme for the daily, weekly and monthly reports produced by our program, that is a contribution and it is appreciated. It is NOT, however, an equal contribution. Without you, our proposal might not have gotten as high of a score, our program might not have produced web pages that are as visually appealing. Without me (or whoever wrote the thing), the proposal or web pages would not exist. There is an enormous difference between making something from nothing and making something better.
  6. Think you’re stupid. I doubt anyone thinks you’re stupid. *I* certainly don’t think you’re stupid. Pretty much all of the above I have done in spades (except the helpless thing). Probably most older people, when you do something really annoying cringe not at your stupidity, but at the memory of themselves doing the exact same things many time.

So, the good news is that no matter how immature and annoying you are, the likelihood is great that eventually you will mature out of it.

Of course, then, you’ll be old.

And have to put up with annoying young people.

How depressing is that? Oh well, there’s always Chardonnay.

Sep

21

I’ve been fairly positive about SAS On-demand but I think the keyword in that sentence is “fairly”. One of the major advantages of SAS over R (which has no formal tech support) and SPSS (which, before it was bought by IBM you could wait 45 minutes on hold for tech support) is that if you were stuck there was somewhere to call.

The first problem is that SAS tech support is open 9 a.m. to 8 p.m. Eastern time which means that by 5 p.m. west coast time it is closed. This may not be a problem for business support as many business people do go home around 5 – I never have personally, but I have witnessed this.

However, for students, particularly many of the students who are likely in graduate courses using SAS On-Demand, they are probably working later than 5 pm, many of them because they have jobs that go until 5 pm. This is also true of many of the faculty members who teach those courses.

One of my pressing problems is that I always have a couple of students who say they cannot log into SAS On-demand. One of them had her laptop in class and it appeared to be installed correctly but whenever she put in her username and password it just came back up and asked for her credentials. I just received an email from a second student who was having the same problem. I forwarded her email to SAS Technical Support.

I am particularly at a loss because I have installed SAS On-demand on six computers to test it, ranging from boot camp on a macbook to a really crummy old desktop running Windows. I did not have a problem with any of them. Yet, every semester, there will be three or four students who DO have a problem getting SAS downloaded, installed and logging in.

Just in case you might be one of those people, here are a few tips:

You try logging in with your credentials, it looks like it is working and then the login screen comes back up. 

Could it be that you are using your email to log in? That won’t work. Even though you use your email to log in to your SAS profile, and even though your password is the same as for your SAS profile, you do NOT use your email to log in to SAS On-Demand. You have a username. Use that instead of your email and it may solve your problem.

Okay, you know that, but you have forgotten your stupid username. And SAS Technical Support is closed. Fume.

Go log in at the Control Center. When you login you will see a page that says, “Your SAS server id is … “. Make sure that is the id you are using to login.

Do you even HAVE  a SAS profile?

I had assumed that every student who had SAS installed must have a SAS profile because otherwise how could they have downloaded it? Very helpful person from SAS Technical Support told me that often they see students who just copied over the .exe file from one of their classmates and just assumed that their username was first initial plus last name like their friend. That will install but you won’t have a SAS username and password for SAS On-Demand. Also, your SAS username is NOT always your first initial plus your last name. Imagine how many guys in the country there are named Andy Lee. They can’t all be alee. What about the Alice Lees and Angela Lees?

You are sure you have a SAS profile, the correct username and password and it STILL doesn’t work.

Yet another possibility is mentioned in this SAS Technical Note http://support.sas.com/kb/20/939.html which suggests deleting the ConfigurationV43oda.xml file. I haven’t tried this one because the above have always solved my problems.

Two other suggestions from Derek (sp?) from SAS Technical Support were that:

  1. Students / faculty may be logging in from their “day job” where the company firewall prevents access. If, as in the case for my class,  some of the students could access SAS On-Demand from the classroom and others could not, it is clearly not a firewall problem because in that case no one could access it.
  2. There may be a permissions problem. Uninstall SAS On-Demand and instead of double-clicking on the .exe file you downloaded to run it, RIGHT-CLICK on that file and select RUN AS ADMINISTRATOR .

When you try to go to the Control Center, for some bizarre reason your browser keeps taking you to SAS On-Demand for Professionals and you get a message saying your email is already associated with  a SAS On-Demand for Academics account

I have had this happen on two different computers. I can speculate on the reason, but here is how I got out of it – I closed the browser I was using and opened a different browser, e.g., closed Firefox and tried again with Chrome and it worked. I am guessing it has something to do with the wrong information getting cached when I was looking at SAS On-Demand for Professionals. That’s just a guess though, because I don’t remember actually doing that. Not that not remembering it doesn’t mean I didn’t do it – who remembers all the various web pages they have looked at? Anyway, closing the browser and trying a different browser works.

Chris, from SAS Technical Support, suggested also trying going to http://support.sas.com/ondemand/ and clicking at the bottom where it says Go to the Control Center.

So, here are a few late-night fixes. Please feel free to add your own in the comments because this is a major issue to me. When students have difficulty getting the software to install, it kind of gets the course off to the wrong start. It makes me long for the good old days of campus computer labs before the Bring Your Own Device cost-cutting gurus took over.

 

 

Sep

19

My long-time business partner, Dr. Erich Longie, is a very experienced pool player. Today, we were discussing a business decision where someone had to choose between two potential employees. One was exceptionally talented but not particularly ethical – nothing terrible, mind you, like stealing out of the cash register, or sexually harassing the interns – but just not great. You know the type – padding expense accounts, always rounding up on invoices so that an hour and fifteen minutes became two hours. This employee, let’s call him Bob1, also had the passive type of lack of ethics. That is, not only did he DO things, but there were also the things he DIDN’T do, like never suggesting anything unless he was being paid specifically to come up with new ideas, never working five minutes over what was in the contract. Bob1 was smart. I would say, too smart for his own good. For example, on the point of never making any suggestions, I once overheard him advise a younger person,

“Why would you suggest anything to a client that was not in your contract? I get paid for my brains, to come up with new concepts. I’m not giving anything up for free. That’s just bad business.”

Bob1 is really smart, with a degree from a top school and a talented developer, not just intelligent but able to come up with new ways of solving problems and even new opportunities.

Then there was the second person, let’s call him Bob2. He is no dummy but an objective observer would have to conclude that he isn’t as good at the work as Bob1. The work Bob2 does is above-average to very, very good, whereas Bob1 is outstanding. On the other hand, Bob2 is outstanding in the ethics field. He’s the kind of guy you could leave your wallet on the desk and yell from the next room,

“I don’t know how much is in there, but just take what I owe you and leave the rest.”

… and you’d just know that Bob2 wouldn’t take any more than you owed him right down to the penny. He is always coming up with new ideas, some of which aren’t the best, but some are really good. He’s the person who starts when he says he’ll start a project and never misses a deadline.

 

pool cue and cue ball

Erich said that making the choice between the two is just a matter of how much honesty and ethics matter. He gave this analogy:

There was a pool player I knew who was a great shot. He also drank too much. If we were doing a short tournament, say one day or one evening, we might pick him because he was a great shot and he would come through for us. BUT … if it was a two-day tournament, we wouldn’t want him on our team because he let us down too many times. Maybe he would show up on the second day and play great, or maybe he would be too drunk. If it was a situation where we needed to rely on somebody, we would pick someone else.

 

If I was doing a short-term project where I needed a top guy, I’d pick Bob1. If I was looking for someone to take on as a full-time employee, I’d pick Bob2. The question is, “How much do his lack of ethics off-set how talented he is?” I think that’s the key here. For a short period, I can see how he’d be worth it. In the long run, you don’t want to always have to be looking over your shoulder, watching out to make sure he is doing his job, not cutting corners, not holding back on the work because he resents that you aren’t paying him enough, appreciating his great talent enough.

While I tended to think that employers would go for one or the other, choosing either talent or reliability, Erich’s point was a very insightful one. Sometimes being the smartest, most talented person in the room matters. And sometimes other things matter more.

 

Sep

11

I have a quad-core i5 with 12GB RAM on my desk that has an ethernet connection as well as a back-up wireless connection option. I also have a Macbook pro for business travel.

These are what I use for development and other work.

I also have three increasingly POS (as in piece of shit, not as in point of sale) computers. One runs Windows 7 and Ubuntu (dual boot) on a really old HP desktop I got for $100 very used. A second is a re-furb laptop I bought for $300 with barely enough RAM to run Windows 7 Home Premium. The third is even older and is in the closet at the moment.

These are what I use for testing.

Why? Did you ever call tech support and have the person on the other end of the line say,

“Well, I don’t know, it works fine on my computer.”

Ronda choking TravisHow much did you want to choke them just like this? The reason it works on THEIR computer is that their computer doesn’t have an old processor, has enough memory, has current software and the latest operating system. Unless you have $1,000 or more to update your system to be like theirs, knowing that doesn’t help you at all, does it?

We learned early on that things that work on our shiny new computers with the latest operating systems and the coolest browsers and the updated everything often did NOT work on the computers out in the field used by those poor saps working for pointy-haired bosses who didn’t see any point in wasting money on equipment that their employees would use every day.

Going into schools and colleges, we were often surprised to find that those pretty monitors on top of the desks were connected to obsolete hardware under the desk. The computer labs looked great but looks are deceiving.  As for the software, we had assumed that schools would have the latest, because they were teaching students for the future, right? Wrong. Often budgets for both software updates and the personnel to install those updates had been axed long ago.

We don’t try to develop for everyone. We don’t see a point in developing for Windows XP, Mac OS 9 or IE6 a product that is going to be out in a year or so. We are developing to run on the earliest version of Mac OS X, Windows 7 and IE9. Looking to the future, we see a huge expansion in the schools of iPads replacing computers in many areas,  so we chose HTML5 over Flash.

That being said, we are assuming that many people will have slow connections, old computers and outdated software. So, after a few false starts we learned to develop for the crappiest 20% of the market in terms of hardware and software that we expect by the time we go into production. The last time I updated the laptop there were 66 updates/ fixes. I do this deliberately. I also test anything that runs on the Internet in the corner by the window. This is as far as I can get from the router and still be in the building. I’ll see which of our two networks has the weakest signal at the moment and connect to that one.

You know the old 80-20 principle, that 20% of your customers will have 80% of your problems? Well, we try to develop to meet the needs of that 20% and usually the other 80% are taken care of at the same time.

 

 

Sep

10

So, here we are with a new semester and my second try at using SAS On-Demand with SAS Enterprise Guide in my Advanced Quantitative Data Analysis class. I am toying with the idea of possibly offering JMP as an option as well but leaning towards no since it would mean twice as much explanation of software in class time that is already limited for the amount of information to be learned. For a start, I am asking students to get in groups of two or three at least one of whom has Windows Vista or Windows 7. Students who have  Mac can use boot camp to get Windows installed.

Still, with Apple making more and more in-roads into the corporate sector and more universities with “bring your own device” policies, hence a substantial number of Mac users, the SAS Windows only stance is making it hard to justify as a teaching tool compared to Stata, R, SPSS or JMP, all of which have Mac native versions.

Why continue to use it then? Several reasons:

  1. It has a huge installed corporate base, and thus, offers a marketable career skill for students only a year or two from graduating, as the students in this class.
  2. Since the students will all be working in groups, it only requires 30-50% of the students have Windows, which is pretty much a sure bet.
  3. SAS is useful for analysis of large data sets, and much of the open data the students will use is downloadable from the census, ICPSR or other sites as SAS data sets.
  4. The active user groups (at least in the So Cal region) offer valuable networking opportunities for students.
  5. Here is a huge reason though – I downloaded this on to the crummiest computer I own to test it. I bought a used laptop piece of crap for $300 just for such situations. Now, at a total cost for hardware and software, I have a laptop running SAS. Given the cost of SAS when I was in graduate school, this is just amazing.

So much for the why, here is the how:

First of all, students should NOT try to do this right before class. It’s really easy to get SAS On-Demand but it does take a few steps that may have a few minutes (or more) in between.

Secondly, note, there is an EXTREMELY detailed step by step breakdown of how to install the software from step 0 here

http://support.sas.com/ondemand/manuals/enterprise_guide_student.pdf

However, it does leave out a couple keys right up front, like that SAS On-Demand doesn’t work on a Mac except using boot camp. So, you might want to read this blog post first.

Go to the student guide, freely downloadable online.

http://support.sas.com/ondemand/manuals/StudentManual.pdf

Start on Chapter 2 (page 5)  and read pages 5-8. Feel free to read anything else you would like, but it is not strictly necessary to get started.

If you don’t meet the system requirements – stop.

If you do, read on down to page 8.

Then, go to:

https://support.sas.com/ctx3/sodareg/index.html

Either click on the link NEW USER on the bottom left of the screen (since there is a great probability you ARE a new user). If not, log in.

When you select NEW USER you will be asked to fill out your name, email, etc. All of it should be pretty obvious except maybe “AFFILIATION WITH SAS” select Customer/User from the drop-down menu.

Once your profile has been created you need to go to your email to activate it. That just takes seconds. Once THAT is done, you should be on a page that gives you the option to of Continuing Registration with SAS On-Demand for Academics.  (That link is near the bottom of the page). Click that.

There are choices of Instructor, Student or Academic Researcher.

When selecting STUDENT, you will be prompted to select your country, etc. It should be pretty obvious.

Once you have selected your university, you will need to ACQUIRE A LICENSE.  (As with everything else, there is a link to click). You will then be prompted for the course. This is the only part that might be tricky if your university offers multiple sections or different courses that use SAS On-Demand with different types of software – JMP, Enterprise Guide, Enterprise Miner. Hopefully, most students know what course and section they are enrolled in.

You will have a username suggested. You can accept that or select a new one as long as it is available.

Note that you will be prompted to log in again and you should use your EMAIL, not that username you just got.

Okay …. NOW go to the installation guide

http://support.sas.com/ondemand/manuals/enterprise_guide_student.pdf

This is very step by step and you are probably past several of the steps. Pat yourself on the back and feel intelligent.

One little glitch you might see is – nothing.

Other times, something starts to install – some .NET stuff SAS uses – and it will automatically install it if you don’t have it on your computer, then – nothing.

I don’t know why this always happens to me but SAS seems to have “pop-down” windows instead of pop-up windows. Minimize your browser window and you will probably see a window hanging out behind it asking if you want to install SAS On-demand. Just click NEXT.

SAS On-Demand for Academics should install. You should see an icon on your desktop. Click on it. Open it up. Login with your username (remember your username) and your password. The password is the same one as for your SAS profile.

 

Sep

6

I am at the Western Users of SAS Software conference this week and just like the Joint Statistical Meetings, SAS Global Forum and SPSS Directions, there is about as much diversity here as at the Republican National Convention.

I brought this up here and at two other meetings I attended. Each time at least some of the people I talked to dismissed me with,

“Well, isn’t it obvious why? Asian students work harder. It’s their culture.”

This, I believe is a perceived politically correct way of calling African-American and Latino students lazy. After all, they work “less hard”, correct?

It is NOT obvious to me. I drive by the strawberry fields pretty regularly and see lots of people picking in the fields. They don’t look like they have a lazy culture to me. Admittedly not a random sample, but I do know some pretty damn hard working African-Americans – my friend who works 80 hours a week for LAPD, my students who work an 8-hour day and then attend six hours of class in the evening.

Inspired by the hilarious Baratunde Thurston’s #negrospotting at the GOP convention, I decided to engage in my own exercise at WUSS. The total for the first day was five.

Hispanic spotting is a little trickier because Hispanics can be of any race and don’t necessarily have a Hispanic last name. Based on the last name, people I knew to be Hispanic and counting myself, I came up with a total of seven.

I am going to ignore the insulting, repeated suggestion that it is because members of these minority groups are lazy or “don’t have a natural aptitude for math”.

I don’t KNOW the reason why we see so many fewer people from these underrepresented groups. I don’t know why I “made it”. I have a hypothesis about both.

The first hypothesis is that Hispanic and African-American students are subjected to a constant barrage of “you can’t make it” and an almost complete drought of mentors. I see almost no efforts of outreach to urban schools. I spoke to over a dozen classes at three middle schools in Los Angeles last year about a career as a statistician, and to an urban school in another state via Skype. I was asked to speak a fifth time but I was out of town and unable to do it. Whether I meet with students in sixth grade or doctoral programs, I am usually the first person in their life who has said,

“You should really consider a career in statistics. It’s interesting, it pays well and you could do it. “

On the contrary, I think in both covert and overt ways students are discouraged from kindergarten on.

My second hypothesis, regarding why I have been successful is that I am a stone-cold bitch. When I have been discouraged from taking a course in Calculus, specializing in statistics, taking more advanced courses in statistics outside of my department, taking more courses in statistics, publishing in refereed journals, applying for tenured positions, my reaction has always been a silent (or sometimes not so silent)

“Fuck you! I can do it and I will!”

This attitude is what probably feminists in the 1950s and 1960s needed and perhaps why they were frequently characterized as angry. No wonder they were angry. Not everybody has that will, self-esteem, arrogance or whatever you want to call it to believe they can succeed when they are constantly being told they can’t.

I was also lucky. I had a few people in my life – my mother, my brother, my grandmother, my aunts, and later a few professors, who told me that I could succeed and did encourage me. Believe, me, though, those folks were by far the minority and except for one nun in the sixth grade, I did not meet any of those people outside of my family until my senior year of college.

So, what can you do about it, if you care to do anything? Five suggestions

  1. NOTICE. The next time you are at an event, notice if there are no Latinos in the building except for the waiters and housekeeping staff.
  2. SPEAK. Go to your local middle school or high school about a career as a statistician.
  3. MENTOR. Encourage ALL of your students to present, major in statistics, learn programming, do an internship, work on a project.
  4. SPEAK UP. I was hesitant at first to say anything because I really like the hard-working people who put on the WUSS conference, and the SAS Global Forum and JSM people seem pretty sincere also. If no one says anything, though, no one will see this as a problem, and I *DO* see a problem when I live in a state that is 45% Hispanic and African-American that there is almost no one from these two groups in the profession I have chosen.
  5. REACH OUT. Really reach out. If we are sincerely concerned, as Maura Stokes said in her keynote, that we will need 160,000 new statisticians then maybe we should look into fields like social science or nursing or social work where students DO have to take a course or two in statistics and research methods and learn some programming and recruit from there.

If we don’t do these things then it is a sign that we don’t really want more students in statistics unless they look , talk and think just like us. Well, none of them look and talk like me. That’s my point.

Sep

3

teenager and toddler in deep discussionMany years ago, when I was a young MBA student, we were all advised to find mentors who would help our careers.  However, I found that many people I was interested in were not particularly interested in me. That was disappointing and irritating.  After all, I believed I was a smart, hard-working person and I even had some degrees and awards to support that belief.

Now that I am older and occasionally wiser, I have gained a little perspective in why young people don’t always find the mentors they believe they deserve.

It’s nothing personal.

  1. They don’t have time for you. Anyone who has obtained a fair degree of success has a lot of demands on his or her time. Your potential mentor has proposals to review, budgets to approve, meetings in six states and three continents. You are interested in this person because of the knowledge and connections they have.  Meeting with anyone, including you, cuts into the time available for applying that knowledge and strengthening those connections.
  2. Someone else beat you to it. I’m very fortunate to know people who are both very successful and very empathetic. These people take on a limited number of graduate students, interns, formal or informal mentees. Each of those junior professionals gets a certain share of that person’s time. Just like number 1, when that allotment is used up, they refuse to take an interest in anyone else because they just can’t.
  3. Some people really are jerks. This is the minority, in my experience, but there are those with the attitude, “I got mine and screw everyone else.” They won’t mentor you unless they see something it for them, say, you have a family member who is president of General Motors.

It is personal.

  1. You don’t have the right school, personal or family connections to help you get hooked up with a mentor. There are people who, if they ask me a favor, I have to say, “Yes.” If one of them calls me, I will make time to help you, find a project for you.  The fewer people like that you know, the harder it is for you to find a mentor.
  2. You have almost nothing in common with your want-to-be mentor. Maybe you are both passionately interested in Widget Design. Other than that, they are worried about managing their investments so they can both retire and send their children to good prep schools. They have problems with their spouse of twenty years wanting to move to the mountains and start an apple orchard. You are worried about how to pay your electricity bill and getting laid on Friday night.
  3. They feel uncomfortable around you. You are a different gender, religion, race and generation. They wear designer suits to the office. You come to work every day in clothes they wouldn’t wear to the beach, swear like a sailor who just hit his thumb with a hammer.
  4. You are a jerk. (This was often my problem.) Maybe you were the smartest person in your high school. So, even though you were a bit of an arrogant prick, people cut you slack because you were the smartest person in the school. Then you go to college, and on to a good graduate program. Guess what? Probably half the people in your graduate program were the smartest kid in high school, and if you are a pompous ass, no one needs to put up with you.

I once overheard a manager give terrific advice to a mutual friend of ours,

“No matter how brilliant you are, there comes a point when it is not worth the pain in the ass of putting up with you, even if you are the best person in the world in your field. They’ll just get the second-best person in the field. More likely, they will just get someone equally good who is not a pain in the ass.”

What can you do, to increase your odds of finding a good mentor?

It is too late for you to be born into a different family, but there are some steps you can take.

 1. Meet lots of people. Yes, networking. Go to conferences. Go to meet-ups. Accept the fact that 90% of the people you meet will fit in one of the first three categories above. However, the more of those people you meet, the more you increase your odds that you will meet someone who has just had a student graduate, intern get a new job or in other way have some time open up for you.

2. Follow-up WITHOUT being crazy-stalker about it. Send someone you met an email telling him or her how much you liked their presentation/ product/ demonstration/ conversation. Don’t take it personally that they often won’t write you back. (See #1 through 3 above). Whether they write you back or not send another email a few weeks or months later if something comes up.

“I read your book on widget design and I noticed that you did not mention widget type #123Grape. I was wondering if that was because you saw some flaw in it or did it just happen to come out on the market after your book went to press?”

3. Try to fit in. This doesn’t mean don’t wear your St. Jude medal to work or change your name from Tanisha to Tammy.  It does mean try not to swear (I tell myself this every time I open my mouth in public), try to conform to the dress code for your organization. If everyone reads the New Yorker, attends a certain conference, reads a certain book, check it out. The key word here is “try”. If you absolutely hate the New Yorker or think Widget World blows, you don’t have to do it again. When the subject of Widget World comes up, you can honestly say that you thought the keynote by the inventor of the widget was awesome. You don’t have to add that you thought the rest of the presenters were a bunch of douche bags. I never learned to golf and the number of times I have worked 9-5 this year can be counted on two hands. However, I own a closet full of suits and read thousands of pages a year of everything from the New York Times to Statistical Medicine.

4. Be a person who saves your mentor time. I’ll be honest with you – most interns and new employees take up more of my time than they save. Many conversations with junior people outside my office take up time that I won’t get back. If you suggest something you can do to save me time, and then you actually do it and don’t screw it up, I will definitely remember you.

“I noticed you had some graphics in your widget design book. I was a graphics art minor and I thought those could be re-done in photoshop like ….”

I just used that example because I suck at art and I love the editor for my book coming out this fall because she has magically made all the art stuff happen.

5. Don’t be too impressed with yourself. Many young people I have had work for me over the years have an unwarranted attitude of superiority if they know something I don’t. Believe me, if I don’t know every option in Final Cut Pro it is not because I am unable to figure it out and if I ask you what is the capital of South Dakota it isn’t because I can’t find Wikipedia on the Internet. It’s because I’m busy doing the things that enable the company to pay your salary.  They often felt that the office schedule should rotate around their needs and whims – thought we should meet at 7 a.m so they could leave in time to watch the Dodgers game and wear shorts to that meeting with the client from the Church of Old Conservative People because it is hot out. Um, no.

On the other hand, if you have done all of those things, don’t be too hard on yourself and don’t think that all of those people you were hoping to impress think you are an idiot. I can’t speak for anyone else, but I certainly don’t think that.  In fact, when I have some young person who does all of the above, I often think to myself,

“This is how God punishes people who were complete self-important smart asses when they were young. He makes them become successful and then spend the rest of their lives meeting people just like they were.”

Don’t ask me how I know this.

Blogroll

WP Themes