Feb

23

5 Basics of Consulting Success: Part 1

Filed Under 30 consulting tips, Dr. De Mars General Life Ramblings | Leave a Comment 

Last week, I mentioned that successful consultants have five categories of skills; communication, testing, statistics, programming and generalist.

COMMUNICATION

Communication is the number one most important skill. All five are necessary to some extent, but a terrific communicator with mediocre statistical analysis skills will get more business than a stellar statistician that can’t communicate. Communication is a lot more than explaining results to clients or making small talk at meet ups.

Documentation

Communication includes documentation, both in your code and internal documents such as codebooks or an internal wiki. It includes letting clients know what you’re going to do, what it’s going to cost, what that cost includes, what were your results and what those results mean. If you’re good at communicating with clients, colleagues and your future self, you’re half-way to success.

An example of the critical nature of communication can be found in the following retraction:

The identified programming error was in a file used for preparation of the analytic data sets for statistical analysis and occurred while the variable referring to the study “arm” (ie, group) assignment was recoded. The purpose of the recoding was to change the randomization assignment variable format of “1, 2” to a binary format of “0, 1.” However, the assignment was made incorrectly and resulted in a reversed coding of the study groups.”

Aboumatar and Wise (2019, p. 1417)

Because of this incorrect coding, the reported results were the exact opposite of what actually occurred.

Document coding!

Here is an example from a current research project where the CES-D depression scale was used, which requires several items to be reverse-coded before scoring.

In the HTML file where the user enters data that’s written to the database there is this comment:

    <h5 >I felt that I was just as good as other kids.</h5>
    <! –– This is reverse-coded. Don’t you dare change it. ––>
<div class=”row mb-3″>
    <button id=”cesd4_1″ data-src=”3″ class=”cesd4 btn btn-light shadow-box col-5 my-3 mx-auto”>Not at all</button>

 In the original file to read in the data to SAS, there is a comment:

*** NOTE: CESD IS ALREADY REVERSE-CODED. DOES NOT NEED CODING!;

FILENAME REFFILE2 ‘/home/directory3/data_analysis_examples/crossroads/cesd.xlsx’;

In the internal wiki, there is this note:

Tables in Acme Project Database

CESD – Center for Epidemiologic Studies Depression Scale – NOTE: The data are reverse coded at data entry. There is no need to reverse code these. There are 25 columns in this table; ID, username, session number, questions 1 through 20 of the CESD scale, the CESD total which is the sum of the 20 questions, named item21 for some odd reason, and a time stamp.

Document everything! Document how are items coded, how subscales or totals are computed.

This may seem like overkill, but how many retractions could be prevented by this level of documentation? If you are a consultant, it’s probable that at some point someone else will be looking at these data, or that you may be called back a year later to do a longitudinal analysis. Your colleagues and future you will thank you. A year or two from now, I don’t want to be looking at this data set and wondering if I need to reverse-code those items or if it was already done. I want to KNOW!

I deeply suspect that there are more erroneous results published due to incorrect coding of data than to incorrect analyses. After all, the peer reviewers, editors and readers see how you analyzed your data. No one sees how you coded it but you and, possibly, the person who has your position after you.

Feb

17

The one skill a statistical consultant must have

Filed Under Dr. De Mars General Life Ramblings | Leave a Comment 

A few weeks ago, I ended my post with “there is one thing a statistical consultant absolutely must have and promised to say what that is in the next post. Maria and I had just picked up our rental car at the Detroit airport when she turned to me and asked:

So, what is the one thing a statistical consultant has to have?

I told her,

“I have absolutely no idea what I was thinking last month!”

In my defense, I have been in five states and 22 cities in the past 21 days. Maria says it is only 16 because I was in Minneapolis, Fargo and Denver twice each. She also says I can’t count Denver, Chicago or San Francisco since I only changed planes there. Poo!

In Long Beach thinking about statistical consultants

Now that I am back in Los Angeles and my brain has unfrozen I think there are actually five things you must have but one of these is the most important. In my not at all humble opinion, though, you need ALL FIVE.

The actually five skills a statistical consultant must have

Man playing drum in preparation for me saying what are the 4 skills statistical consultants must have
Drum roll, please
  1. COMMUNICATION – This is the number one most important skill. If you don’t have the rest, you’ll still suck and be unemployed but a terrific communicator with mediocre statistical analysis skills will get more business. I don’t just mean shaking hands and small talk at conferences, either. Communication includes documentation, both in your code and in codebooks, an internal wiki, etc. It includes letting clients know what you’re going to do, what it’s going to cost, what that cost includes, what were your result and what those results mean. If you’re good at communicating with clients, colleagues and your future self, you’re half-way to success.
  2. TESTING – I’ve ranted on this blog a lot about testing because it is one of the areas where people often seem to fall short early in their careers. I got a lot of hate for this post when I said I don’t hire self-taught developers because there are things they don’t teach themselves adequately, like testing.
  3. Statistics – Well, duh. Props to the person in the Chronicle of Higher Education forum whose signature read, “Being able to find SPSS in the start menu does not qualify you to run a multinomial logistic regression.” Your clients may not know what power, quasi-complete separation or multicollinearity mean in interpreting an analysis. They do trust that you understand whatever is necessary to be understood for the work. Don’t let them down.
  4. Programming – when I was a graduate student Very Important Professors had lowly peon graduate students and programmers to write their code for them. All of those people had started their careers using punched cards, (honest!) it was that long ago. All of the statistical consultants I know do programming, or can code their own analyses if necessary. Even if you aren’t doing it all yourself – I’m certainly not these days – you need to know enough to review the code your minions wrote or help said minions when they get stuck. Sometimes, it’s just quicker to do it by yourself than explain to someone else, especially if you need to fix a bug in a code that a client is waiting on.
  5. Be a generalist – I’ll have more to say about this in future posts. In brief, even the consultants I know who are well-known specialists in one language know and use others. If you think your career is going to be you sitting on a mountain or in penthouse office, pontificating to others about sums of squares, the computation of Wilks’ lambda or options for PROC GLMSELECT , you are going to be sadly disappointed. On the other hand, if you do know of a job like that, I would consider taking it for a sufficiently large quantity of money.

Jan

19

The first things a statistical consultant needs to know

Filed Under 30 consulting tips | 3 Comments 

I’ll be speaking about being a statistical consultant at SAS Global Forum in D.C. in March/ April. While I will be talking a little bit about factor analysis, repeated measures ANOVA and logistic regression, that is the end of my talk. The first things a statistical consultant should know don’t have much to do with statistics.

A consultant has paying clients.

In History of Psychology (it was a required course, don’t judge me) one of my fellow students chose to give her presentation as a one-woman play, with herself as Sigmund Freud. “Dr. Freud” began his meeting with a patient discussing his fee. In fact, Freud did not accept charity patients. He charged everyone. There’s a winning trivial pursuit fact for you.*

Why am I starting with telling you this? Because I have had plenty of graduate students whose goal is “to be a consultant” but they seem to think their biggest problem when they start out is going to be whether they should do propensity score matching using the nearest neighbor or caliper method.

Here are the biggest problems you’ll face:

  • Getting your first clients
  • Getting paid
  • Getting your data into shape
  • Communicating results to your clients.

Let’s start with getting clients. I can think of four ways to do this; referrals, as part of a consulting company, through your online presence and through an organization. I’ve done three of them. First, and most effective, I think, is through referrals. I got my first two clients when professors who did consulting on the side recommended me. I do this myself. If someone can’t afford my fees or I am just booked at the moment, I will refer potential clients to either students, former students or other professionals I know who are getting started as a consultant. It’s not competing with my business. I am never going to work for $30 an hour again and if that’s all that’s in your budget, I understand. If all you need is someone to do a bunch of frequency distributions and a chi-square for you, you don’t need me, although I’m happy to do it as a part of a larger contract.

Lesson number one: Don’t be a jerk.

Referrals mean I’m using my own reputation to help you get a job and so I’m going to refer students who are good statisticians and who I think will be respectful and honest with the client. Don’t underestimate the latter half of that statement.

Lesson number two: It helps if you really love data analysis.

I’d be the first to say that I’m a much nicer person now than when I was in graduate school. Yes, it took me a while to learn lesson one, I am embarrassed to say. However, I really did love statistics and if any of my fellow students had trouble, I was the first person they asked and I was really happy to help. When those students later became superintendents of schools or principal investigators of grants, they thought of me and became some of my earliest clients. Some of my professors also became clients, although those were after I’d had several years of experience.

Lesson number three: Don’t think you are smarter than your clients.

A young relative, who has a Ph.D. In math asked me, “No offense but isn’t what you do relatively easy, like anyone who understood statistics could do it? Why are you so in demand?”
Corollary to this lesson: If you find yourself saying, “No offense” just stop talking right then.

One reason a lot of want-to-be consultants go bankrupt or have to find another line of work is they do think they are smarter than their clients. This manifests itself in a lot of ways so we’ll return to it later, but one way is that they charge much more than the work is worth.

How do you know how much your work is worth?

Lesson number four: Ask yourself, if I had twice as many grants/ contracts as I could do and I was paying someone to do this work, what would I be willing to pay?

That’s a good place to start.

I’ve met a lot of people over the years who charged much more than me and bragged to me about it. In the long run, though, I’m sure I made a lot more money. Clients talk. They find out that you are charging them three times as much as their friend down the block is getting charged by their consultant. You may think you’re getting away with it, but you won’t. You may get paid on those first few contracts but you’ll have a very hard time getting work in the future.

Lesson number five: Know multiple languages, multiple packages

I’ve had discussions with colleagues on whether it is better to be a generalist or a specialist.

I have had a few jobs where they just needed propensity score matching or just a repeated measures ANOVA but those have been the small minority over the past 30 years.

I would argue that even those who consider themselves specialists actually have a wide range of skills. Maybe they are only an expert in SAS but that includes data manipulation, macros, IML and most statistical procedures.

In my case, I would not claim to be the world’s greatest authority on anything but if you need data entry forms created in JavaScript/HTML/CSS, a database back end with PHP and MySQL, your data read into SAS, cleaned and analyzed in a logistic regression, I can do it all from end to end. No, I’m not equally good at all of those. It’s been so long since I used Python, that I’d have to look everything up all over again.

I’ve used SPSS, STATA, JMP and Statistica, depending on what the client wanted. I think I might have even had a couple of clients using RapidMiner. For the last few years, though, the only packages I’ve used have been SAS and Excel. Why Excel? Because that’s what the clients were familiar with and wanted to use and it worked for their purposes. (See lesson three.)

I was really surprised to read Bob Muenchen saying SPSS surpassed R and SAS in scholarly publications. Almost no one I know uses SPSS any more, but, of course, my personal acquaintances are hardly a random sample. I suppose it depends on the field you are in.

I have never used R.

Some people think this is a political statement about being a renegade. Others think it’s because I’m too old to learn new things or in subservience to corporate overlords or some other interesting explanation. (The Invisible Developer, who has been reading over my shoulder, says he never got past C, much less D through Q.)

Since I fairly often get asked why not, I will tell you the real reasons, which is a complete digression but this is my blog so there.

  1. In my spare time that I don’t have, I teach Multivariate Statistics at a university that uses SAS. Since I’m using SAS in my class anyway and need real life data for examples, when a client has licenses for multiple packages and doesn’t care what I use (almost always the case), I use SAS.
  2. About the time that R was taking off, my company was also taking off in a different direction. The Invisible Developer and I own the majority of 7 Generation Games which is an application of a lot of the research done by The Julia Group. When we started developing math games, we needed to learn Unity, C#, PHP, SQL, JavaScript, HTML/CSS. We also needed to analyze the data to assess test reliability, efficacy, etc. I called the analysis piece and told The Invisible Developer I was interested in all of it so I’d do whatever was left. He was really interested in 3D game programming so he did the Unity/C# part. I did everything else. Then, after a few years, I moved to Chile, where the language I most had to improve was my Spanish.
Games in Spanish, English and Lakota

It worked out for me. We have a dozen games available from 7 Generation Games and now we’re coming out with a new line on decision-making.

I mention all this because I want to emphasize there isn’t a single path to succeeding as a consultant. There isn’t a specific language or package you have to learn. There is one thing you absolutely must have, though, and that’s the next post.

* (See Warner, S. L. Sigmund Freud and Money. (1989) Journal of the American Academy of Psychoanalysis. Winter;17(4):609-22)

Jan

6

From PHPMyAdmin to SAS Studio for lazy people

Filed Under Dr. De Mars General Life Ramblings | Leave a Comment 

I need to get a data set into SAS for a course I’m teaching in March. Students like real data and some kind folks were willing to allow their de-identified data to be used. Win-win.

How did I get this data? In a SAS data set with a handy code book? Oh, very funny!

A bunch of people laughing
This blog will pause while we all laugh at your naivete

I received a login to PHPMyAdmin where the data which were definitely not created for my personal convenience reside.

First, I downloaded the data as CSV for Excel. This gave me a file where everything was like this.

re_apply;”consumer_id”;”email”;”counselor”;”gender”;”date_of_birth”;”age”;”primary_disability”;”secondary_disability”;”education”;”member”;”tribe”;”district”;”job_when_entered”;”if_job_earnings”;”earnings_type_before”;”referral”;”other_refer”;”application_date”;”assessment_date”;”eligibility_date”;”ipe_date”;”notify_rights”;”vocational_goal”;”state_vr”;”status”;”closure_date”;”status_type”;”employment_date”;”type_employ”;”start_job_earnings”;”post_job_earnings”;”earnings_type_after”;”intermediate_goal”;”semesters”;”int_completed”;”intermediate_date”;”last_contact_date”;”comment”

Yes, one, long line with everything in quotes and every column separated by a semi-colon. These are the column names but all of the data are in this exact same format as well.

Of course, you COULD upload this file and read it into SAS but that would take time and effort.

OR you can download a regular CSV or ODS spreadsheet file, just pick one of the other options, and then all your data would be in nice columns but you have no header row. Of course, that’s pretty easy to write an input statement in SAS. You just need to type in a hundred or variable names and be sure to have the format correctly specified. Not hard but probably take you more than a minute.

You COULD theoretically download the data as an SQL file and use SQL connect according to some smart and less lazy people on SAS Communities.

And, as some of the posters noted on that forum, not everyone has access to SQL connect.

Or, you could be a completely lazy person like me and fix it all in about 12 clicks.

Here is how:

  1. Download the file as CSV for Excel
  2. Copy and paste the first line, the header, into Word
  3. Replace all the quotes with nothing, using Replace from the EDIT menu
  4. Under the TABLE command select “CONVERT TEXT TO TABLE”. For “Separate text at”, click on Other and put in a semi-colon.
  5. Now you have all of your filenames in nice columns as a table, copy that.
  6. Download the file again as CSV
  7. Insert a row and paste the filenames you copied in Step 5
  8. Save that file as an Excel file
  9. Upload it into SAS Studio
  10. Under TASKS and UTILITIES, select IMPORT DATA, drag your file you uploaded to the window and click on the little running guy.
Converting from text to table
Click 8 or So
Tasks and utilities
All that’s left is to drag the file

Dec

1

Why use SAS as a statistical consultant?

Filed Under 30 consulting tips | Leave a Comment 

Anyone who uses SAS (or doesn’t) probably has their own reasons. I have a few but a major one is the ease of importing just about any type of data.

Mo’ clients, Mo’ problems

There are multiple types of consultants. I’m the type who is, literally, all over the map. I’ve been in five countries this year and I think 11 states plus the District of Columbia, but I might have left off a couple. I said 9 in a post on a different blog where I occasionally write about my life and judo, but then I remembered I’d been in Texas for SAS Global Forum where I gave a talk on biostatistics and also in New Mexico speaking on transition from school to work for tribal youth with disabilities.

What that means is that I work with a wide range of organizations and their data is not all in the same format.

If you work with a wide range of clients, ease of data import matters

If you’re a consultant who works consistently with one client, data formats may not be your biggest issue. You probably wrote a program to read in that data, no matter what messy format it was in and you’re good to go. In my case, though, every dataset, every project is different.

All the data, all the time

In the previous post, I mentioned reading in the IPEDS data, which is a relatively small public data set (around 7,000 x 60). Fantastically, that came with a SAS program so all I needed to do was upload the raw data file and change the INFILE statement.

Proc import does not a consultant make

Maybe when you were a student you imported your data sets by a PROC IMPORT step. This isn’t terrible. You should use this procedure when you can. However, you’re going to need to go several steps further.

Even worse, if you’ve been getting your data by simply using the LIBNAME statement your professor provided you or doing some pointy-clicky thing with SAS Studio or Enterprise Guide (or SPSS) you have a lot to learn.

Every year, I have graduate students who tell me they are going to become consultants. More often than not, I shake my head and think,

“You have no idea what you are getting into.”

– Me

If you are going to be working as a statistical consultant for a variety of clients, far more than PROC LOGISTIC or PROC GLIMMIX, your time is going to be spent in the DATA step.

It’s not just a matter of data formatting or missing data, but of creating the data you need that isn’t there. What do I mean by that? Ha ha, that is a future blog post that I may write next time I’m on a plane somewhere and have a spare moment. Probably tomorrow.

Oct

25

7 Tips to Not Getting Screwed by Your Data

Filed Under 30 consulting tips, Dr. De Mars General Life Ramblings | Leave a Comment 

First of all, I want to draw your attention to this retraction in the Journal of the American Medical Association and mad props to Drs. Aboumatar and Wise and John Hopkins for doing the right thing in publicly retracting it.

For the TL; DR crowd

Someone who is probably now unemployed miscoded the study groups in this randomized clinical trial of self-management of Chronic Obstructive Pulmonary Disease. What does that mean? In this case, it meant that the reported results were the exact opposite of what was really observed because the treatment groups were coded incorrectly. Also, read the seven tips at the end of this post.

When I talk about statistical analysis, I focus 80% or more of my time and attention on the basics of knowing your data, cleaning your data and examining your data some more. To some, mostly younger, statisticians, that is not the sexy stuff. Why am I not talking about neural nets or generalized linear mixed models? Don’t I know that improving your prediction by .3% can result in millions of dollars in profit for a corporation that has 38 million customers?

What I know is that problems like the one in that JAMA article occur more often than we like to admit.

Recently, a student sent thesis results and then the next day sent an email saying, “Oops, I meant to use the DESCENDING option in PROC LOGISTIC but I didn’t, so the results are the exact opposite of what I said.”

A couple of years ago, I did an analysis with a depression scale for which the standardized coding is 0 to 3, but the application had used 1 to 4. The first analysis showed that every single person in the sample was clinically depressed. Fortunately, I caught this before it was published. Even when I re-analyzed the data with the correct scoring the mean score was extremely high. This was not a random sample of the population, but rather, children with a family member addicted to methamphetamine. The original (incorrect) analysis wasn’t in the opposite direction but it did somewhat overstate the problem.

Several years before that, I worked for a client who had a previous consultant with no knowledge of their particular field but who was a very good programmer. In reviewing some of that person’s code to understand the data and how it had been scored, I found that NONE of the items that should have been reverse-coded had been. The consultant had simply taken the sum of all of the items. This research had been published, by the way. I mentioned this to the client and suggested that a retraction was in order. That retraction never happened and I never worked for that client again.

My Six Tips for Saving Your Ass

  • Learn to code. I don’t mean you need to be the greatest SAS/ R/ Python whatever guru in the world but you should be able to read through the code someone else wrote and understand it. This means you should be able to read an IF-THEN statement, a loop re-coding all the items in an array and the statistical procedures used in your analysis.
  • Understand that the DESCENDING option in PROC LOGISTIC means that the probability modeled is reversed. So, by default, PROC LOGISTIC models the probability of response levels with lower Ordered Value, and if you have death (coded 0= lived, 1= died) as the dependent, the procedure is predicted who lived. If you use the DESCENDING option, it’s going to predict who died.
  • Know how many people should be in each group; control, experimental condition 1, experimental condition 2. Do a PROC FREQ and see if it matches what you expect.
  • Know the range for each item in your analysis and do a PROC MEANS with mean, minimum, maximum and standard deviation. Even if you have 500 or 600 variables it shouldn’t take you all that long to scan through that many lines and see if anything is out of range.
  • Know which items should have been reverse-coded and check if that was done.
  • Compute reliabilities for each scale in an analysis. While the reliability would not have been changed in the depression example where 1 was added to every response, it would have picked up those cases where the variables were not re-coded by showing very low reliabilities.

A seventh, extra bonus tip

If you can’t understand the code that someone has written, not because you are a moron (can’t help you there), but because they are one of those people who never write comments in code, don’t believe in documentation and write code that includes an unnecessary number of macro variables, user-written macros and overly complicated solutions, fire their sorry ass and hire someone less pompous. I’m not saying you shouldn’t have macros or that because a person uses a DATA step and you prefer PROC SQL you should get rid of them. What I am saying is if you ask a person what decisions they made in writing that code and what was the reason for, say, using a generalized linear model instead of a general linear model, they should be able to tell you.

Sep

30

Don’t believe your numbers too much: IPEDS example

Filed Under 30 consulting tips | Leave a Comment 

The famous statistician, F.N. (for Florence Nightingale) David was a professor at UC Riverside, where I earned my doctorate. My advisor told this story about her:

We were on this dissertation committee – I forget if it was for biology or what, back then, this was a small campus so if you were in statistics you could end up on any committee. So, he gets to the end of his defense, and F.N. David pulls the cigar out of her mouth and says,

“Young man, you believe your numbers far too much.”

The point Dr. Eyman was trying to make to me was that even if you have done every single computation perfectly …

“The government are very keen on amassing statistics. They collect them, add them, raise them to the nth power, take the cube root and prepare wonderful diagrams. But you must never forget that every one of these figures comes in the first instance from the village watchman, who just puts down what he damn pleases.”

– Josiah Stamp

What is a conscientious statistical consultant to do?

Start with getting to know your data better than God knows the Bible. Let’s start with analyzing secondary data, for example, IPEDS, that has already been collected. I’ll talk about collecting your own data later. Let me just put in a plug for doing it electronically if possible. Also, make sure your data entry staff know which is the intervention and which is the control group. (You think I’m kidding but I’m not.)

Secondary data analysis: Read the documentation!

You think that is obvious, do you? IPEDS is the Integrated Postsecondary Education Data System, collected by the National Center for Education Statistics. It is my favorite type of data set and the type you almost never get. It includes pretty much the entire population of interest.

If you don’t know these things, you don’t know your data:

  • Is it a sample or the entire population?
  • If it’s a sample, what proportion of the population was sampled and how? Randomly? Stratified random?
  • Does the data set have sampling weights? What is the variable name for those weights (You’re going to use them, aren’t you? Please say yes.)
  • How were the data recorded?

This isn’t all you need to know. We’ll talk about specific variables next.

One reason I like IPEDS is that you can be pretty sure everyone reported data because it’s mandatory for any institution who gets federal financial aid. It also includes the U.S. service academies, which are about the only post-secondary institutions who don’t. It also gives you a SAS program for reading the data after you upload it. There are also SPSS and STATA programs.

Another thing I liked about IPEDS is it is, inside and out, one of the best documented data sets I’ve seen. I’d recommend it as an example of how to do things if you are going to be creating data sets for secondary analysis yourself. Don’t get used to it, though, because most of what you’ll find in your career is far worse than this. Here is just a simple example from one data set.

*** Created:    October 2, 2018                                ***;
 *** Modify the path below to point to your data file.        ***;
 ***                                                          ***;
 *** The specified subdirectory was not created on            ***;
 *** your computer. You will need to do this.                 ***;

If you want to analyze it using SAS Studio, now you know that once you’ve uploaded the data, you do need to change the INFILE statement. If you don’t know the full path, ctrl-click (Mac) or right-click (Windows) on the data file and select PROPERTIES

Select Properties to get the path to your file

Change the INFILE statement to what you see in the path, so now it looks like this

infile '/home/your_directory/IPEDS/hd2017.csv' delimiter=',' DSD MISSOVER firstobs=2 lrecl=32736;

You won’t necessarily have the delimiter, etc. It depends on your file. Okay, run it, you have data. Awesome!

When I run frequencies for the IPEDS data, I get 7,153 institutions but the IPEDS methodology report says there are 6,642. What the hell? Looking through the data, I find that 287 institutions were closed in either 2017 or prior. Another 38 were combined with another institution or not to be include for some unspecified reason “out of scope”. There were 41 that were “not primarily post-secondary institutions”, so I dropped those also. Since I’m only interested in individual, active institutions for the research I’m doing, I’m dropping those.

There were 88 institutions that were new in 2017 or had their Title IV (financial aid) eligibility restored. After debating back and forth, I decided to drop those, too. My interest is in developing a baseline of enrollment and retention, which these new institutions will only have for one year.

My point is that I’ve gotten one of the best data sets you could ever find and 7% of the data is inappropriate for my purpose. Does it matter as long as 93% of the data are correct? Well, I definitely think that my results would be less accurate.

My second point is that there is not anything “wrong” with the IPEDS data. I can imagine plenty of circumstances in which one would want to have the data on closed institutions.

These may seem like details, but I am pretty convinced that if you are not a “detail person” you are never going to make it in the long run as a statistical consultants. These details add up fast.

One last thought – if 7% of the data needed to be tossed out before we even got started, and this is an extremely well-funded, well-designed data set, what do you think the average secondary analysis is going to be like?

Sep

16

30 Things I Learned in 30 Years as a Statistical Consultant – Part 1 of lots

Filed Under Dr. De Mars General Life Ramblings, statistics | Leave a Comment 

Never fear, I’m not going to post all 30 things in this post. This is a series. A LONG series. Get excited.

I was invited to speak at SAS Global Forum next year and it occurred to me after thinking about it for 14.2 seconds that there are plenty of people at SAS and elsewhere that are more likely to have new statistics named after them than me.

While I can code mixed models, path analysis and factor analysis without much trouble, I’d be the first to admit that there are plenty of new procedures and ideas I see every year that I never really master. I mean to, I really do, but then I get back to the office and attacked by work. So, the person to introduce you to every facet of the bleeding edge, nope, that’s probably not me, either.

If you think this is where I experience impostor syndrome and say “I couldn’t possibly have anything worth saying”, we have obviously never met.

I’m the old person on the left. The youngest of many daughters is on the right.

Okay, there’s the most current picture of me, so now you sort of know who I am. I figured I better post a current one because I had not updated my LinkedIn photo in so long that I connected with someone who said,

“Oh, I have met your mom.”

And I had to reply,

“No, you have met me. My mom is 86 years old and retired to Florida, as federal law requires. Florida state motto: Your grandparents live here.”

So, when do you get to these 30 things?

Now. I decided to divide everything I learned into four categories.

  1. Getting clients
  2. Getting data into shape
  3. Getting answers
  4. Getting people to understand you.

I picked four because if I had five or six categories, people would expect there to be an even number of points in each because 30 divides evenly by five and six. See? I am good at math.

The money part: Getting clients

First, decide what kind of statistical consultant that you want to be.

Are you a specialist or a generalist?

You can be like my friend, Kim Lebouton, who specializes in SAS administration for the automotive industry and seems intent on keeping with the same clients until she or they die, whichever comes first. I linked to her twitter because she is too cool to have a web page.

You could be like Jon Peltier of Peltier Tech and specialize in Excel. Basically, if there is anything Jon doesn’t know about Excel, it’s not worth knowing. Personally, I feel as if most things about Excel are not worth knowing, which is why I’m not that kind of consultant.

I do love that the Microsoft Store carries our games for Windows, though, so woohoo for Microsoft.

Canoe the rapids and learn fractions, with your kids or by yourself because maturity is overrated

I’m the kind of statistician that doesn’t have a time zone.

A few years ago, I was at a conference when people were trying to coordinate their schedule for an online meeting. They were saying what time zone they were in and someone asked me,

“You’re on Pacific Time, right?”

My friend interrupted and said,

“She doesn’t have a time zone.”

It’s true. I was on Central Time last week, in North Dakota. I’m in California this week. Next week, I’m back on Central Time in Minnesota and South Dakota. The following week, I’m on Eastern Time in Boston.

In the winter here (which was summer there), I was in Chile. During the spring here (which was fall there), I was in Australia, and I’m in the U.S. now.

BUT HOW DO YOU FIND CLIENTS?

This is probably the question I get the most and I have an odd answer.

Get really good at something and the clients will find you.

Jon’s really good with Excel. Kim is superb at SAS administration. What am I good at? I’d say I am excellent at taking something that a client may only be vaguely aware is a statistical problem and solving it from beginning to end, in a way that makes sense to them.

If you try mansplaining me in the comments that what I do is called applied statistics, I will find where you live and slap you upside the head. I teach at National University in the Department of Applied Engineering. It’s in the fucking department name. I KNOW.

In response to the question in stats.stackexchange regarding the difference between mathematical statistics and applied statistics, there was this answer:

Mathematical statistics is concerned about statistical problems, while applied statistics about using statistics for solving other problems.

– Random person I don’t know on the Internet

Mathematical statistics often involves simulated (that is, fake) data, and nearly always uses data that is cleaned of data entry errors – in other words, not very representative of real life.

If you ask me, and even if you don’t , many data scientists act as if data issues can be fixed by having big enough data. This always seems to me similar to those startups who are losing money on every sale but aren’t worried because they are going to make it up on volume. Since data is key, let’s talk about that in the next post.

But wait! How do you get those first clients?

There is never a surplus of excellence – unless maybe you are an English professor, but they’re not reading this blog.

Network.

Let your professors know that you are interested in consulting. I got my first consulting contracts by referrals from professors who had more work than they could do. Similarly, I have referred several potential clients to students and junior professionals either because I was too busy, not interested or they could not afford my rates.

Go to conferences

I’ve had clients referred by other consultants who met me at a conference and a particular contract was not in their area of expertise but they thought it might be in mine. Similarly, I’ve referred clients to other people because I don’t really do that thing but maybe this person will be available.

Most jobs come by word of mouth

There is an evaluation consultant organization. I don’t know who the hell belongs to it. Much of the work that I do, someone’s job is on the line. That is, if they can’t demonstrate results, they may lose their funding and everyone in the building loses their job. In almost all of it, at some point the project director or manager or whoever is going to go present these results to a federal agency, tribal council or upper management, trusting that everything they say is true because I said so.

In that type of high stakes situation, they’re not going to get someone from an ad on Craig’s list. If that sounds like bad news, the good news is that after you have been around for a while and done good work, the jobs come to you.

Since a big difference between mathematical statisticians and applied statisticians is the messiness of the data, I’m going to address that in the next few posts. Expect more swearing. Because data.

Jul

16

Who’s a 10X Engineer?

Filed Under Dr. De Mars General Life Ramblings, Technology | Leave a Comment 

A twitter storm erupted recently in response to one person’s thread about how to find a 10x engineer . Since I started programming FORTRAN with punched cards back in 1974, was an industrial engineer in the 1980s and now run a software company, I’ve worked with a few people, rightly or wrongly considered to fall into that category. So, I thought I’d weigh in on the original author’s points.

10X Engineers hate meetings

There are only two types of software developers who don’t dislike meetings. New developers don’t mind meetings too much because they have a lot of questions like “who do I talk to if I need access to this repository” or “What version of Unity was used to develop this game I’m supposed to update?” They also have specific questions about why the sound function they wrote is not working and Bob, who wrote similar functions for another game is sitting right there. Another type of developer actually likes meetings because he is complete shit at his job and it gives him an excuse not to be expected to do it.

Every other engineer I have ever met either dislikes meetings or actively hates them. The ones you think don’t dislike meetings are just pretending.

We have a 10-minute meeting every morning at 7 Generation Games. People not in the office drop in online. Everyone complains about it but we do it anyway. Why? Because, for example, I can find out that Adekola actually finished the teacher reports for Making Camp Premium before he left and see an example. Then, I can tell the people in marketing to include that in their discussions with schools. I can also tell one of the developers to take that code and modify it for Tribu Matemática , the Spanish version of Making Camp. In 3 minutes, everyone knows what the reports look like, that they are available and who is working on the next one. This leaves seven minutes for José to ask Bob about the sound function.

10X Engineers have irregular hours and work when other people aren’t around

I can’t think of any software developers who work better when other people are around. Writing code for anything complex requires having a mental model in your head of at least the part you are writing and, hopefully, some of the larger project in which it is used.

I’ve worked with a few people who were hit it out of the park better than anyone else. One definitely was a late night person and preferred to get to work when he got there. However, when crunch time came, he could work 8am to 10pm and code all that time if he had to do it. He wasn’t going to like it, though, who would?

On the other hand, about half of the really top engineers I know – both software and hardware – choose to work 9 to 5, even when telecommuting. The main reason they give is that those hours allow them to spend time with their children or spouse. Contrary to popular belief, the 10x engineers I’ve known tended to be married, although they did seem to get married a little older than the average.

10X Engineers know every line of code that has gone into production

This is just nonsense. I remember when SAS was rewritten in C (yes, I am that old) and hearing that it was something like 3,000,000 lines of code. I am assuming the author meant that these 10x engineers know every line of code WRITTEN BY THEM that has gone into production.

I don’t believe that, either, assuming what he means is that they can recall it immediately and say,

“Yes, in that function beginning on line 683, I pause the audio that’s playing, change the source file to the audio for the next scene, change the image file for the image for the next scene, increment the counter by one and restart the audio”.

If what he means is that they kind of recognize it like that person you met at a conference two years ago and are trying to remember their name, I might faintly agree.

We wrote Spirit Lake: The Game in 2012-2014. NO ONE who worked on that game remembers all of the code in it. I can say this because it was all done by me and The Invisible Developer and he is as good as you’ll ever find.

Here is an experience I share with every software engineer I have ever met, including the very best ones. I look at code and think,

“Who wrote this crap? Please don’t let it be me three years ago.”

10x engineers laptop screen background color is typically black (they always change defaults). Their keyboard keys such as i, f, x are usually worn out faster than of a, s, and e (email senders).

They always change the defaults part is true. One thing for sure all of the best engineers I ever met had in common is they like to mess with things. I only knew two people who had black backgrounds – ever. When I have time I’ll have to post about pseudo-10x engineers. Anyway, neither of those guys are anything special unless weirdness is a category.

Most of the best people I know have either pictures of their family or their favorite activity, like soccer or hiking, or a vacation photo as a background. Usually the e key gets worn out first because it is the most common letter in the English language. People usually name directories, datasets and variables something comprehensible.

My kids and one of my kid's dog. What real 10x engineers laptop backgrounds look like
Their laptop backgrounds look like this, except with their own kids, not my kids, because that would be weird

Is there anything true about a 10x engineer?

Since my 10x merit badge hasn’t come in the mail yet, I don’t have time to address all 10 points from the original thread. There were two points he made that were consistent with my experience.

Most really good engineers aren’t really good interviewers

I could only speculate about why that is true, so I will leave it as that is what I’ve observed. Maybe it’s because they are uncomfortable with exaggeration or with being asked to prove their competence.

10x engineers rarely job hunt

I have found this totally to be true and it makes sense. If you have someone that good in your organization and your management is not made up of complete morons, they are doing all they can to hang on to their best people. Usually, unless they work for morons, people that good are hard to hire away, too, because their current company is doing its best to keep them.

How would I find a 10x engineer?

I wouldn’t, because we are a small company and we can’t afford to pay what someone like that is worth. On rare occasions, we have been super lucky to be able to catch someone great for a short term contract that they just wanted to take for personal reasons.

We find good people and we develop them to be at the top of their field. I think the best way to identify a good software developer in an interview is take a look at their code. Ask them to bring something to the interview and explain how they solved particular problems in the code. Ask why they made the choices they did. If it is a project they know well and are proud of, you’ll get a lot of information. If they say, “I don’t know” a lot, that’s a bad sign. I’ve also found that people who typically “don’t interview well” forget about the interview part, focus on the project and become interested in telling you all about it.

Oh , and for all those people on twitter who said, “I wish you all got as exercised about diversity and inclusion as you do about 10x engineers “

Well, I am way ahead of you, sister. I have a lot to say about women in tech and over on our 7 Generation Games blog, too.

Jul

4

Stupid Mistakes with SAS I have made

Filed Under Dr. De Mars General Life Ramblings | Leave a Comment 

One thing I like about our company a lot is none of our developers fit the stereotype of the 10x coder asshole. Don’t get me wrong, we have more than our share of people who are absolutely great at their jobs. What we don’t have is the arrogant attitude of:

What do you mean you don’t know how to integrate 3-D scenes from Unity with web pages? I knew how to do that when I was in the third grade!

First of all, I bet most of those people are liars and if you found their third-grade teacher she would say,

Oh, little Larry? I vaguely remember him. Wasn’t the brightest crayon in the box, now was he?

The fact is that all programmers make mistakes, including really dumb ones. As we get older, we may catch these before we hit the commit or run button, but, then again, maybe not.

Why is this code not working?

data fixdata ;
    set fix1;
   
    *** FIXES RECORDS WITH WRONG USERNAME ;
username = trim(username);
    if username = "" then delete;
    pos = index(username,'-') + 1;
    username2 = substr(username,pos);
    if upcase(index(username,'test)' > 0 then delete ;

Three mistakes with SAS in the code

I made all three of these mistakes lately (though not all the same day).

The first two SAS will catch for you if you read your log. Also, I cheated you here by not including the color coding that you’ll see in the SAS editor just to make it harder for you.

First of all, I have an unmatched parenthesis.

if upcase(index(username,'test)' ) > 0 then delete ;

This still doesn’t work because I have the closing quote in the wrong place. 

if upcase(index(username,'test')) > 0 then delete ;

The hardest errors to find are when your code is running but still wrong

Now my log shows no errors but it’s still not working. I still have more users in the file than I should. The hardest kind of errors to find are logic errors. SAS will usually find the coding ones for you.

What I want to do is delete any usernames that have the word “test” in them, whether written as TEST, Test or test – or any other weird combination people might come up with, like TestAM or TestGarbanzoBeans12.

The problem here is that I used the UPCASE function after I had already searched for the value of “test” in the username. The INDEX function returns a number, which is the position at which the first character of the first occurrence of a string occurs.

NOTE: Numeric values have been converted to character values at the places given by: (Line):(Column). 74:12

Here is what I really need to do

if index(upcase(username),'TEST') > 0 then delete ;

I needed to change a few things. First, I need to put the UPCASE function inside the arguments passed to the INDEX function so that the first thing that happens is that username is set to upper case. Next, I need to change the string it’s matching to “TEST” since I set the username to uppercase. Now, finally, I pass that to an INDEX function and see if the string exists. If so, I delete the record.

This is a relatively simple bug to fix when presented like this. However, when it is hidden in hundreds of lines of code and all I know is that the number of records doesn’t match what I should have, it’s not so obvious, particularly when there are no errors showing in your SAS log. Did the username2 throw you off? Well, imagine hundreds more lines like that.

My points, and I do have more than one …

  1. Everyone makes mistakes, no matter how experienced they are.
  2. A bug fix may seem obvious after you have found it or if it’s been pointed out to you, but buried in hundreds or thousands of lines of code, it’s not so easy. An argument for modular development.
  3. READ THE NOTES! If you are deluded into thinking that if there is nothing in red in your SAS log all is good, get over that notion now.
  4. The enhanced editor is your friend. The incorrect quotation mark you could have picked up in the editor if you noticed that the closing parenthesis was in purple. In this particular case, it wasn’t so obvious because it was only off by one character. However, if you see a whole bunch of purple, meaning it is quoted, or green, for comments, it can be a tip right away that something is off.

One last thing – the SAS Enhanced Editor with color coding – I remember when it was brought up as a new, improved version of SAS and I laughed.

“Really? That’s what you’ve got? Color-coding? I mean, I’m sure it might help someone their first week on the job but we are all professionals here .. ha ha ha.”

Yeah, that was stupid, too. That enhanced editor has helped me catch so many bugs in code as I was writing it!

Random fact about me – I host a podcast

Yes, yes I do. It’s called More than Ordinary. This week’s co-host is Drew Kim who co-founded and runs an esports company with his mom. (Yes, you read that right.) Next week will be the continuation of a discussion of all things judo with Jason Harai of Ippon Dojo, in Washington. Basically, it’s all about having guests who are doing something out of the ordinary.

Next Page →