A review of tips for teaching applied statistics from Statistical Sage

Hello All,

Another semester is about to start. To help people who are new to Statistical Sage or even for those of you who visit here often, I thought it helpful to review some key prior postings on the teaching of applied statistics. I have included tips along with the links to the posts from Statistical Sage.

1. A great class has to start with a great syllabus.  https://statisticalsage.wordpress.com/2011/08/16/backward-design-and-syllabus-formation/
2. Even for those of you who don’t want to teach applied statistics, you can still do a great job and even have fun doing it. https://statisticalsage.wordpress.com/2012/01/09/so-you-dont-want-to-teach-stats/
3. Great ideas can come from anywhere … so keep on the look out. (and please share them with us when you have them). https://statisticalsage.wordpress.com/2012/01/20/great-ideas-can-come-from-anywhere/   https://statisticalsage.wordpress.com/2012/01/27/using-current-research-to-help-students-understand-concpets-in-applied-statistics/ 
4. Don’t forget to apply the concepts of Cognitive Development to the classroom. https://statisticalsage.wordpress.com/2012/02/11/applying-the-science-of-cognitive-development-to-the-classroom/
5. Talk with (or read) others who have experience in teaching applied statistics; they can serve as a wealth of knowledge and help you to minimize errors and maximize learning (both yours and your students). https://statisticalsage.wordpress.com/2011/02/12/bonnie-and-hals-five-tips-to-teaching-statistics/
6. Applied statistics is more than just calculations. It is important that you get students thinking about statistics. https://statisticalsage.wordpress.com/2011/03/08/more-than-calculations-guiding-students-to-thinking-with-statistics/
7. Make sure to start that first class out, right! https://statisticalsage.wordpress.com/2011/01/29/first-day-of-class-starting-off-right/
8. Don’t forget to have fun teaching!!! https://statisticalsage.wordpress.com/2012/02/01/having-fun-teaching-applied-statistics/

I hope that everyone has a very productive fall semester, and that students’ excuses are few and class attendance is high!

Bonnie

Advertisement

EPA Presentation

Hello All,

I’m just back from Cambridge, MA where the Eastern Psychological Association held it’s annual conference. Though I had to pop in and out  quicker than I would have preferred (due to family constraints), I enjoyed reconnecting with friend and colleagues and meeting a few new ones. There are so many great reasons to make time to attend a conference like EPA, though few offer the kind of dedication to teaching and high quality research, at such an affordable price as EPA. (Yearly membership for EPA is $45, and includes the cost of the conference).

As for the past few years (8 — believe it or not), I was fortunate to be able to participate in the CUPP symposium (another great and inexpensive organization) that is dedicated to quality psychology curriculum. As in years past, this symposium was organized by Susan Nolan (her webinar on teaching statistics is just around the corner https://statisticalsage.wordpress.com/2011/02/21/webinars-with-nolan-and-heinzen/) and Janine Buckner, enthusiastic and dedicated faculty members from Seton Hall.

The topic of the symposium was putting students on the path for Life Long Learning. Working with Irina Khusid and Jyh-hann Chang, from ESU, we provided a brief overview of starting students off with the best foundation for Life Long Learning, especially in statistics.

Here is the link to the PowerPoint presentation. In it, includes several links to articles, surveys, and prior blogs.  In truth, much of it has been discussed in some detail on these pages. EPA symp 2011

As with all posts and links, I look forward to your comments.

Another Thought on Diversity

Bonnie’s recent post on diversity of skill sets, knowledge base, and other student attributes in statistics classes made me think of diversity as the raison d’être for descriptive and inferential statistics.

Students beginning a statistics class often say something to the effect “I want to be a counselor, why do I need to take statistics? One of my favorite answers to this question is to ask students to imagine a world where every person is a clone of me. Everyone looks like me, acts like me, in fact, is identical to me in every way, physically, cognitively, and emotionally. Of course this scenario leads to gasps of horror, especially from the women in the class. With a little class discussion, however, the realization suddenly grows that in this world all  behaviors would be normal because there would be no variability among people. There would be no standard deviation for any measure we might obtain. Hence, no counselors would be needed. No one would be handsome,  beautiful, intelligent, arrogant, energetic, or helpful (I’m not implying that any of these adjectives actually describe me). No behaviors would be abnormal, criminal, empathic, altruistic, selfish, or whatever. Such concepts imply a diversity in physical appearance, intellectual functioning, behavioral actions toward others, and so forth. If we want to know anything about such a population, we need measure only one member of the population. A measure taken on one person would describe all other people.  Descriptive statistics wouldn’t be needed in this world.

Students soon realize that diversity is the reason that statistics is a necessary discipline to understand and explain our world. When there is diversity no single term adequately describes everyone. Thus we have had to develop statistics that describe “typical” and the spread around the typical.

A similar discussion can lead to an understanding for the need for statistical hypothesis testing. Think how easy it would be to decide if an independent variable has an effect on behavior if every person’s behavior were identical. If we introduce the independent variable with one person, and it changes that person’s behavior from the state prior to its introduction, then we know it is effective. And it will have the same effect for everyone.

A world without variability wouldn’t require statistics, but it wouldn’t be much fun to live in either.

First Day of Class — starting off right

How do students comes to us on the first day of class? Yes, I can just about hear you mumbling …

(1) They wonder why they even have to take this class … after all they are a [non-quantitative] major. Why does [psychology, sociology, business, education, etc] need statistics?

(2) They may have had really bad math experiences in the past leading to (a) math anxiety (b) poor math attitudes including a low self efficacy and/or (c) weak math skills.

(3) They have heard lots and lots of stories as to how hard or useless or manipulative statistics can be. We have all heard the quote … and so have they … “There are lies, … , and statistics!”

But the first thing I want to let you know is … instructors of applied statistics may be over estimating the negative thinking of their students. Mills (2004) http://findarticles.com/p/articles/mi_m0FCR/is_3_38/ai_n6249218/?tag=content;col1 found that, in general, students attitudes towards statistics tended to be more positive then negative. Now, granted, that could simply be the case at Mill’s school. Yet, it is worth challenging … what do our students really think? How is this impacting their academic success? Is there anything we can do to get them from less adaptive to more adaptive ways of approaching this class?

If you would like to specifically see what attitudes your students have, the Mills (2004) article includes an abstract with a validated and reliable measure of  Survey of Attitudes Toward Statistics (SATS). 

Though I do find that there are some positive attitudes regarding statistics in the students I see, the majority of students are still coming in with negative attitudes and ways of thinking about the class. Almost importantly, my first day of class activities do as much to help the students with positive attitudes as it does for the negative ones.  

In short … I try to start of the first day of my semester differently than any other class students may have experienced.

I often start by asking students to describe what a typical first day of class look like. That is right, I don’t even introduce myself or the class. I walk in and say, “So … what does the first day of class typically look like.” After a few blank stares, and me having to repeat the question. The students start talking.  There are many types of descriptions … for which I then define variability. We often talk about what is the most typical? Students respond by raising their hands (of course, I have to instruct this) High means strongly agree, low means mildly agree or disagree. Moderately high hands mean that you moderately agree. Sitting on one’s hands means you completely disagree (thus everyone has to do something, everyone has to actively participate. The student with his hands on the desk is ask to participate.)

Yes, as we discuss “typical,” I will often tell students my name, hand out the syllabus and tell them to put it away, and briefly describe things like … there will be four exams, as they state that professors tell them about how the class will be graded. So, the students end up getting the “typical” information from me in an atypical manner.

 Then, I ask students’ their personal preferences of the activities they listed, again  having them to respond with their hands.  This activity really helps to focus on the concept of individual differences, which I define. As individual differences is a precursor to sampling error, which will follow weeks from the first day of class, students are already beginning to conceptualize this class.

I then define statistics as a tool to help us to answer questions by making sense of variability, taking into account individual differences of the subjects we are working with.

Yes, I start the class by forcing students to respond verbally and with voting using their hands … From the first minutes of class, students realize that they have to be engaged in this class … there are no other options. From the first minutes of class, they are beginning to understand two of the most critical concepts … variability and individual differences!

So, why start off the class so differently than most? Because, in the event students are not excited about taking an Applied Statistics class, may be down right afraid of this class, and/or think the class is unnecessary, right from the start, they start to recognize this class is unlike any others.   By starting off so differently, I can challenge their preconceived notions regarding classes in general, and statistics in particular.

At this point, I tend to talk to them about what they had heard or how do they feel about statistics, thus making them face their preconceived notions right away. Again, we conduct another hand poll … and I define for them the concepts of data and sample.

This opening exercise involves kinesthetic action, thus, forces student engagement. Moreover, it takes the abstract concept of variability, individual differences, data, and sampling and puts it into a context of something all of your students can understand, the first day of class. It provides students for context in which statistics functions.

I follow this first day of activities up with Assignments and Exercises 1.1, 1.2, 1.3, & 1.4 all requiring students to think about their behavior and attitude for class. Instructors who are using Kiess and Green (2010) as their statistics textbook can access those assignments at http://www.pearsonhighered.com/educator/product/Statistical-Concepts-for-the-Behavioral-Sciences/9780205626243.page.

Will one day be enough? Of course not … but it’s always the first step!

As always, I would love to hear from others regarding their first day activities!

Green, B. A., & Sandry, J. D. (2010) Assignments and Exercises for Students for Statistical Concepts for the Behavioral Sciences, 4/E . Boston: Pearson.

Mills, J. D. (2004). Students’ attitudes toward statistics: implications for the future. College Student Journal. FindArticles.com. 29 Jan, 2011. http://findarticles.com/p/articles/mi_m0FCR/is_3_38/ai_n6249218/

A statistics teacher’s New Year’s Resolution

Hello All, and welcome to the New Year with Statistical Sage. 2010 was our first year. We began modestly in June. By the end of July, every day two or three people would stumble upon our web site. By November, that number was well over 20 every day, with some days attracting 40 or more visits. That level of intensity remained even as we approached the end of semester, much to my surprise. So, it will be exciting to see what 2011 has in store of us.

I suppose my first New Year’s resolution is that the Sages and I will continue to provide our experience regarding the teaching of statistics, integrating information from both the literature and our years in the classroom. Of course, resolution are probabaly suppose to require more effort and not be so much fun, right? I guess that makes this not much of a resolution.

Now, exercising as a new year’s resolution … that would be a challenge for me. Yet, with each passing year and recognizing the importance that good cardiovascular health has on one’s cognitive function and how exercise helps increase a person’s energy, energy that is needed in the class room … I suppose I must list this as a resolution.

I am hoping that through our discussions on these pages that I will be able to formulate two articles suitable for publication during 2011.

Finally, 2011 is the year I will be applying Mathematica demonstrations to my teaching! My goal for the spring semester is to identify and use at least 5 Mathematica demonstrations and make them available to students electronically via Desire 2 Learn (ESU’s course delivery system du jour). I also hope to make them available to everyone at Statistical Sage as well. For Fall 2011, it is my goal to identify Mathematica demonstrations for all of the concepts I teach.

I do hope those of you who are reading Statistical Sage resolve to visit us again and again throughout the year. I also hope many of you begin to actively participate in our discussions. There is something to be said for having an opportunity to discuss teaching with others who teach the same courses. So, please, make 2011 the year you join us!

Of course, I do wish you and your families* all a healthy*, happy*, and prosperous* 2011!

* Feel free to operationally define these variables in any manner you desire!

Intro to Statistics – Day 1

This is my first statistics blog, in fact, my first blog ever.  So I want to start at a very basic level in dealing with the general question of student motivation in statistics classes.  When I was a young professor in the Psychology Department, in spite of loving to teach statistics, I was the only one willing to argue that a statistics course should not be required of all psychology majors.  I was aware that most majors did not want to take statistics, and thought maybe they were right.  Statistics can be viewed as a technical subject matter needed by researchers or those who need to read and evaluate the research literature.  My view was that if a student did not follow a career in psychology, as many would not, then taking statistics courses would be wasteful and unpleasant.  The appropriate time for such technical learning was graduate school.  My own experience supported this view.  I believe I was the only student in my graduate program who had not had a statistics course as an undergraduate, yet I stood at the top of my first year graduate statistics courses and did not feel I had been at a disadvantage.  (Of course, the quality of undergraduate statistics courses then was not what it is today.)

But the rest of the department agreed unanimously that there should be a required statistics course for undergraduate majors.  Over the years I taught that course I realized that I thought of statistics not just as a technical exercise for professionals, but as a way of dealing with more general issues, and eventually I tried to communicate that to my students in hopes that it would make the subject matter more meaningful and so more interesting.

So I told the students at the beginning of the course that they probably thought of this statistics course as a narrowly conceived and pretty much useless requirement.  On the contrary, I argued, this is probably the broadest course they would ever participate in, since it dealt with one of the fundamental characteristics of the entire universe: random chance.  Contrary to what television CSIs tell us: that there are “no coincidences,” in the real world there are an uncountable number of coincidences.  As time goes by an infinite stream of events flows with it; some of these events are related, some cause others, but it is not easy to detect these relationships (the “signal”) in the vast real-world flow of all the other random events (the “noise”).  Science is a powerful tool developed to discover and confirm relationships, especially causal, among events.  But the role of chance is ever present.  One implication is that none of our measurements of anything, psychological (e.g., measures of intelligence, anxiety, etc.) or physical (e.g., length of a snake jaw bone, reaction time to a signal) is perfect.  All measurements are contaminated with “error.”  (Here I would give some examples:  (1) Early recognition of measurement error by human observers in the field of astronomy: different observers could not agree on the timing of celestial events.  (2) The data that provided “proof” of Einstein’s theory by confirmation of his prediction that light should bend in gravitational fields (i.e., as it passed near the sun) did not precisely fit the numerical prediction made by the theory).  If we cannot trust the accuracy of our observations and measurements, how can we ever learn anything?  Believe it or not, I would tell them, this statistics course attempts to deal with this fundamental randomness of the universe.  We will find ways to measure the amount of error in our measurements, and by doing so, try to draw conclusions about, in particular, cause and effect relationships.  This process will necessarily involve issues of “experimental design” (the name of the course, at least for some years, was “Experimental Design and Statistical Analysis”).  You might even be able to apply your new understanding of randomness in the universe to your everyday lives.

And that was my attempt to interest students in the content of the statistics course.  Did it work?  For one (!) student at least, a very bright student who later became one of my Apprentice Teachers in statistics; he mentioned later that in that first lecture he had thought, “Aha, this sounds promising.”

Statistical Concepts are at the center of this blog

Probably like many of you, I really enjoy attending conferences like APS or EPA. One of my favorite conferences, and I don’t even know if they still have it, was the Armadillo Conference held in Texas. That was a neat bunch of individuals.

One of the commonalities I notice when I attend conferences like EPA is that though there is a core group of individuals who seem very interested in keeping their skills honed in the teaching of statistics, others seem to almost want to dismiss students ability to even master this material. I am sorry to say I used to be one of those dismissive faculty, I just didn’t know it. I remember once saying to a person who was teaching statistics, very matter of fact, that students recalled very little of that information when they hit research methods, a course that followed and had statistics as a prerequisite. She was devastated, and set to work with intensity to  find a way to get her students to master the material not just for the short-term, but for the long-term. This interaction had a lasting impact on me, as I started to wonder what biases for student learning I may have been harboring. And could my implicit beliefs that students really weren’t capable of learning statistics well impacting how I was teaching?  

During my last EPA visit in Brooklyn (March 2010), I attended a great workshop on hand-on activities to use with the teaching of statistics. Every one of the presenters, however, said the same thing, which I will paraphrase. If we want students to master the material of statistics and truly commit it to long-term memory, we have to approach it from a conceptual perspective. (Conceptual vs. computation approaches to teaching statistics will be a blog for a future date). I would say that I firmly believe in this statement as well. However, each person, one after the other stated … there are NO undergraduate statistics books that take a conceptual approach while maintaining academic rigor. One after the other, as I heard them utter these words, I was left wondering … what do I do, do I tell them about Statistical Concepts for the Behavioral Sciences, 4/e? If I do, won’t it seem like I’m “hawking my wares?” So I sat quietly. As we get this blog going, and we use the month of July for introductions and laying the ground work, it seems reasonable for me to tell you, if you are looking for a place to develop your teaching skills in statistics, this is the right place. If you think we are going to encourage you to “dumb down” your work or decrease the expectations you hold for your students, it is probably the wrong place. If, however, you want to know the secret to helping students to learn the material, mastering it for long-term use (the key is grounding all that you do in the concepts of statistics) this blog will help you to do just that. We’ll point you to important resources, too, including http://www.pearsonhighered.com/product?ISBN=0205626246 the web site for you to take a closer look at a textbook dedicated to assuring students, even reluctant ones with weak math backgrounds, become successful students of statistics. We welcome comments from anyone who is using Statistical Concepts for the Behavioral Sciences, 4/e.

Meet the Sages — from Bonnie’s Perspective

Welcome to Statistical Sage! Let me tell you about the format of this blog. I, Bonnie Green, will be writing weekly on various topics related to the teaching of statistics in social science, business, education, and health. I will be joined by four “sages,” individuals who possess far more impressive credentials and experience in teaching applied statistics than me. Though they will each tell you more about themselves, I’ll tell you a bit as well.

There is Harold Kiess, I call him “Hal.” I first “met” Hal as I was seeking the right statistics book to use with my students. I wanted a book that was accurate (and as you know that’s not always easy to find), I wanted a book that hit on critical points, and I wanted a book that could serve as a good resource for both the students and for me. Out of the countless books I reviewed, I selected Statistical Concepts for the Behavioral Sciences, 3^rd Edition. I read Hal’s words again and again and again before ever meeting him. When we finally connected via phone, it was like reconnecting with an old friend. As you will see in other places on the blog, I joined Hal in co-authoring the fourth edition of Statistical Concepts for the Behavioral Sciences. So over the course of the last few years, though we have only seen each other three times, we have spent hours talking about and corresponding about the teaching of statistics.

There is also Martin Richter, “Marty.” I’ll save my first conversation with Marty for another time … let’s just say, it was memorable. When I speak of Marty to others who don’t know him , I just refer to him as my “statistics mentor.” Next to my academic advisor and husband, Marty was the only other person who was with me at every step of graduate school, often telling me how I could do more and better. He is a brilliant statistics teacher. However, even more important than teaching me statistics, Marty is the one who convinced me that I could write. I, like many of the students we may face in our classes, was certain, due to comments made  by my high school English teacher, that I would never be a good writer. My behavior (or lack there-of) guaranteed that I would never get any better. I almost didn’t apply to grad school given my beliefs about my writing skills (I had an entity view on writing and an incremental view on math and science — most of your students will have the opposite pattern – we’ll talk more of this in a latter blog). Marty, without citing any research, actually convinced me with practice and effort I could learn how to write.

 

Each week (or so) one of them will be adding to the on going conversation of teaching applied statistics. As I have told them, they are free to remark on any issue relating to the teaching of statistics, so it should prove to be interesting. During this first month, I have asked them to introduce themselves to you. Thus, expect more in depth conversations to begin next month.

I encourage you to share this site with others whom you know are teaching statistics. Among the three of us, we have over 80 years of teaching experience and thousands of students who have been successful in our classrooms. We have formally study how students learn, and we have even failed … a lot. But from those mistakes, we have learned. It is this knowledge that we hope to impart to you.

I encourage you to post questions or comments you may have, as like students’ questions when we teach, I am sure your comments or questions will serve to shape our conversations.

Hello All!

Welcome to the newly formed Statistical Sage Blog, dedicated to focusing on teaching applied statistics at the college level. During this first month, I will be joined by four sages, each of whom will introduce themselves to you. We will begin in August talking about teaching statistics to college students, right about the time most of us will be gearing back up for another great year of teaching!

Happy thinking,

Bonnie