CCSS and Community College Math Programs

We may need a complete redesign of the developmental math program in US two-year colleges.

My campus currently uses a placement test (Mathematics Diagnostic Test Project) to determine if students are ready for transfer level courses (math for elementary school teachers, stats, trig, precalculus, calculus)  or what remedial course (arithmetic, prealgebra, elementary algebra, intermediate algebra) they should take.

But the Common Core State Standards for mathematics will have high school students studying mathematics organized in a fashion that does not align with our existing math courses.

California is one of the 45 states that have formally adopted the CCSS for mathematics, and I am on a recently appointed state committee whose charge is to align California’s math standards (a.k.a. the California Framework) with the CCSS.

One of the main reasons that I applied to be on the Mathematics Curriculum Framework and Evaluation Criteria Committee (MCFCC) was to better familiarize myself with what is to be taught in California's K-12 schools.  (Another reason was to lose myself in abbreviations:  SBE for State Board of Education, CDE for California Department of Education, IQC for Instructional Quality Commission, the body that forwarded my name to the SBE for approval to serve on the MFCC to align the CF with the CCSS.)

The CCSS specify a consensus of what math is required for students to be college or career ready.  The standards are grouped into six conceptual categories:  Number and Quantity, Algebra, Functions, Modeling, Geometry, and Statistics and Probability.  (There are separately eight standards for mathematical practice that go across all grade levels.)

The CCSS differ significantly from what is typically required for graduation in most American high schools today.  For example, the treatment of statistics and probability includes not only descriptive statistics but also conditional probability, inference, decisions based on probability, and rules of  probability. 

The CCSS include not only right-triangle trigonometry but also trig functions of a real variable, to be used in modeling periodic behavior.  Thus trig spans the geometry, algebra, and function categories.

The CCSS gives math standards for high school without specifying courses or order of topics.  But evidently the introduction of functions includes an emphasis on (linear and) exponential functions with domains restricted to a subset of the integers--sequences are explicitly studied as functions.

California community colleges do not require a high school diploma for admission.  A student who masters the first CCSS high school math course will already have compared exponential functions with linear functions and solved equations both algebraically and graphically. The student will have had explicit instruction on descriptive statistics.  The student may have worked with constructions and transformations in the plane and proved simple geometric theorems algebraically but not yet worked with polynomials (and specifically not with quadratic functions or quadratic equations).

How will our placement system advise this student?

One of the recommendations  of California's Student SuccessTask Force is for better alignment between high school and college curricula.  With the CCSS adopted across states, it looks as if most community colleges will need to make adjustments to their way of placing and educating their math students.

Joint Mathematics Meetings 2012

About 7000 mathematicians gathered in Boston for the 2012 Joint Mathematics Meeting January 4-7, 2012.

My first event was the Mathematics Digital Library advisory board meeting.  One of MathDL director Lang Moore's items was that the MathDL's  Course Communities  now include Developmental Math. These resources are from the Knowledge Exchange Networks spearheaded by Tom Carey of U. Toronto, San Diego State U., and the Carnegie Foundation for the Advancement of Teaching.

Haynes Miller (MIT) reported on  MIT's online resource for faculty collaboration.   First developed as an in-house tool to help facilitate writing across the curriculum, the resource is now being opened up for faculty collaborations throughout the US.

The CRAFTY (Curriculum Renewal Across the First Two Years) committee sponsored a contributed papers session on preparing students for calculus, and the College Board/MAA Committee on Mutual Concerns sponsored a panel session  on promoting student success in calculus.  Alison Ahlgren of U. Illinois and Marilyn Carlson of Arizona State University painted vastly different pictures.

Alison believes the UI has found a working solution using ALEKS (Assessent and LEarning in Knowledge Spaces) as both an assessment and placement tool.  All UI students receive ALEKS assessment, and ALEKS placement scores are strictly enforced--even students who pass the UI precalculus course must earn the appropriate ALEKS score to be eligible to enroll in calculus.

Alison has data from many thousands of students on what ALEKS tests.  Analysis of what topics were or were not mastered by successful vs unsuccessful students informs UI about what precalculus topics should receive greater of less emphasis.  For example, UI students all appear to have mastered polynomials, but the great majority of pre-calculus students have little mastery of exponential functions and logarithms.

In contrast, Marilyn argues that ALEKS and other currently popular assessment systems do not measure conceptual understanding, and hence student success with ALEKS does not necessarily facilitate mastery of what we really want our students to learn.  Marilyn led the development of ASU's Precalculus Concept Assessment.

The Committee for Two-Year Colleges sponsored a panel session on math for non-STEM students.  Panelists Bernadine Chuck Fong and Jane Muhich of the Carnegie Foundation spoke about the philosophical and research-based underpinnings of the Statway™ and Quantway™ projects, while Larry Gray of U. Minnesota spoke on his reaction to the actual Statway™ lessons which he has reviewed.  The audience included Mary Parker of Austin Community College and Katherine Yoshiwara of Los Angeles Pierce College, both faculty teaching  Statway™ this year, and both speaking in favor of the project but indicating that there is still much work to be done.

A joyful conspiracy

Uri Treisman's Joyful Conspiracy from CarnegieViews on Vimeo.

The Carnegie Foundation for the Advancement of Teaching is organizing a “joyful conspiracy” to help community colleges provide pathways to success for students who initially are placed in developmental mathematics courses.  The Statway will bring non-STEM students from the level of elementary algebra up to and through a transfer-level statistics course in one year.

The Statway 2010 Summer Institute brought teams from 19 community college campuses to the Stanford University campus July 25-30 to meet, share with, and learn from each other and from Carnegie Foundation leaders and consultants.  

We practiced the protocol for presenting, critiquing, and giving feedback on the lessons we will be piloting in the coming year.  Each lesson will involve students working on a rich task with clearly defined learning goals.  A key assumption of Statway is that statistics can provide a context for students to learn to think and reason quantitatively.  The necessary algebraic skills will be embedded within the lesson, rather than holding center stage.

Another core part of the instructional experience is that having students struggle with problems is desirable.  This student engagement, even when students do not discover or invent the necessary mathematics on their own, can be crucial to preparing the students for making sense of the central topic of the lesson.

Statway lesson protocol

Thursday afternoon and Friday morning (July 8 – 9, 2010) Pierce College math faculty Vic LaForest, Bob Martinez, Kathy Yoshiwara, and I were in a “fishbowl” as part of the development of the Statway project.

In the coming year, faculty teams from 19 community college campuses will take materials (developed by the Carnegie Foundation for Advancement of Teaching) and create, test, analyze effectiveness of, and give feedback on statistics lessons. The purpose of our two-day experience was to test out a protocol developed for carrying out this process.

The lead facilitator was Bill Saunders, formerly of Pearson Learning Teams. He was joined by UCLA research colleagues Jim Stigler and Karen McGivven, Kris Bishop of the Dana Center (UT Austen), and Alicia Grunow of Carnegie.

We were not allowed to see the proto-lesson until we met Thursday. After a brief introduction to the protocol and the Statway lesson approaches, we four faculty members spent much of the afternoon working among ourselves deciding how we could best implement that lesson, while a video camera and the observers watched on.

We were expected to have the lesson design completed before our 5:30 pm adjournment Thursday. Bob was chosen to deliver the lesson at the start of the Friday session, and Karen volunteered to acquire the materials needed for our modified lesson. Kathy and I agreed to put together and email some of the materials Bob would need for his handouts.

Bob was working until 2:00 am putting together the materials.

Our Pierce College deans Jacquinita Rose and Crystal Kiekel were attending as guests. But when only 3 students showed up from the 8 students that had been recruited for Friday morning, we put Crystal and Alicia to work, enlisting them to act as students for Bob's lesson.

Bob did a great job running the lesson. After the students departed with their $20 iTunes gift cards, we continued the lesson protocol with the debriefing of how the lesson went, analyzing the student work, and writing feedback on the lesson.

The researchers were pleased with how everything went, and promised we would not be seeing the videos on YouTube.

Variance in values from prediction by regression

In a section about linear regression in Understanding Statistics in the Behavioral Sciences (7th), Robert Pagano, Thomson, 2004, we find the following equation on page 119.

After an explanation of the notation, we find,

We could then construct for each score.  If we squared each and summed over all the scores, we would obtain

Obtaining this second equation from the first seems remarkable, but the textbook offered no insights on how one could see this.

Here's one possibility.

If we think of vectors , , and , then the second equation above is the assertion that

This equation is true whenever the vectors and are orthogonal.

But is precisely the orthogonal projection of onto the space spanned by and (see my earlier blog), so is in that space and is in the orthogonal complement.

Statway: A pathway from developmental math through statistics

Los Angeles Pierce College has been invited to be one of sixteen community colleges to participate in the Carnegie Foundation's Statway project.

The key goal of the project is to provide a pathway for developmental math students to progress successfully from elementary algebra to completion of a transferable statistics course, all in one year.

The Statway project has already collaborated with AMS, ASA, MAA, AMATYC, NADE, NACME (National Action Council for Minorities in Engineering), and CAUSE (Consortium for the Advancement of Undergraduate Statistics Education).  Selected faculty from the professional mathematics societies make up the Carnegie Committee on Statistics Learning Outcomes, which has been working on identifying the core concepts, topics, and learning outcomes for transfer-level statistics.  The CCSLO is also identifying the developmental math learning outcomes needed to prepare students for learning statistics.

But in addition to redesigning the content and pathway to statistics, Statway will incorporate a student engagement component--roughly survival skills for a college student.

The Correlation Coeffiicent as cosine theta

Mathematicians define the dot product between vectors and as

On the other hand, the alternate geometric definition for the dot product popular with physicists is

So

And statisticians define Pearson's correlation coefficient r so that

Thus if we set and , then .

The idea is to think not of n ordered pairs (x₁, y₁), (x₂, y₂), ..., (x_n, y_n), but rather to think of two vectors in n-dimensional space. When the vectors are pointing in the same direction, the angle between them is zero and the correlation coefficient is cos 0 = 1. When the vectors point in opposite directions, the correlation coefficient is the cosine of a straight angle, r = -1. And when the vectors are orthogonal, the correlation coefficient is the cosine of a right angle, r = 0.

The only tricky part is that the two n-dimensional vectors are not the vectors and  , the vectors containing all the and respectively.  Instead, the necessary two n-dimensional vectors are the and defined above.

And nicely, the least-squares regression line for the data is y = mx + b, where and .  (Notice that the variance , so m can also be written as  .


One typically derives the least-squares regression line by finding m and b that minimize .  But one can alternatively use the n-dimensional vector point of view, where the coefficients m and b correspond to the solution of the vector equation .  The vector is the vector of all 1's and the vector   is the orthogonal projection of the vector  onto the space spanned by and .