Student Learning Outcomes: A rEvolution

Let me start with the present, with the way I currently measure performance in MS 150 Statistics. In Statistics students are graded based on attendance, homework, quizzes, tests, a midterm examination, and a final examination.

In the present incarnation of MS 150 Statistics, attendance and homework are worth one point. For the sake of keeping this example reasonably simple, set aside the attendance, homework, and quiz points. Let the tests be 8 points each, the midterm be 14 points, and the final is 20 points. This means that there are 50 points possible in the whole term.

Each point on a quiz or test represents a single question that measures a student learning outcome specified in the outline.

A student's grade can then be calculated by dividing the total points attained by the total possible.

Name GradePercSumT1T2MxFx
AndreciaA100%50881420
BetalynB80%40661117
CassioC70%35661013
DumandD60%3055812
FailynF50%2544710

As one member of faculty once noted, "If a brain surgeon got a 94% on his brain anatomy examination, wouldn't you wonder if you were their patient what was in the 6% they missed?" In the example above Dumand has passed the course, but has apparently not mastered something on the order of 40% of the outcomes.

Suppose I decide to document specific outcomes attained as seen further below. To incorporate this into my existing system of points I award 5 points per outcome attained. The assigning of 5 points per outcome allows me to continue to use a point based system, necessary because I actually have attendance and homework to account for as well. Note that for simplicity I have only three outcomes, in reality there would be a bank of outcomes. The three outcomes are calculation of sample size, mean, and standard deviation. See the outline in progress to see the actual outcomes.

Name GradePercSumT1T2MxFxCountMeanStDev
AndreciaA100%65881420555
BetalynB85%55661117555
CassioC77%50661013555
DumandD62%4055812550
FailynF54%3544710550

While I am not doing the above in the Statistics course, the above is exactly what I am doing with my PE 101j course. Students get 3 points for attending class ready to run with 3 tennis balls, 2 attendance points if they are dressed for running but lack balls, and 1 point if they show up not dressed to run. There are no written quizzes nor tests. Each outcome is assigned 5 points, so my grade book looks something like the following:

Name GradePercSum8/208/228/278/29Jog 20 minJuggle 10 minJoggle 20m
AndreciaF59%163323005

As most of the outcomes are not scheduled to be attained until late in the term, most students would have low percentages of the total possible points. I partially correct for this by "floating" the total possible points in the term, using until late in the term the highest sum for any student as the total possible.

Physical Education with its clean and clear psychomotor domain outcomes is where I first realized that points do not make any real sense. Either a student has mastered the specified outcomes or not. Simplify the course to the three outcomes used above:

If a student manages to do all three, then they have mastered the course. Those outcomes are the learning goals for the course. Either a student can do all three or not. Early on I saw that the course would have to be pass/fail due to the student learning outcomes. What measure would distinguish an A from a C? I could not award A's because a more fit student at term start could go beyond the course requirements. Hence the course is pass/fail.

Since the course is pass/fail with all the desired outcomes spelled out in the outline, does it particularly matter the exact path a student takes to mastery of the outcomes? In other words, my grade book should simply be a record of outcomes met:

NameCompetencyJog 20 minJuggle 10 minJoggle 20m
AndreciaPassYesYesYes
ComingalongaleenFailYesYesNo
WhupsterFailNoNoNo

In this grade book attendance, other than as a separate policy, has no direct impact on competency. And Comingalongaleen may yet pass the course if the above grade book is a midterm grade book. Hence the suggestion by Brent to use terminology such as "Competent" and "Not yet competent".

The student who misses two classes and the student who misses four classes can both pass if they accomplish the outcomes. If someone has five absences in the Tuesday Thursday section but accomplishes all of the outcomes, are they any less competent than the two absences student? Six absences and competent in all outcomes? Eight? What if the student achieves mastery of all outcomes by midterm? Can they quit attending the course and receive their pass? They attained all that was specified for them to obtain.

But attendance is important in a PE course, no? Part of PE is working together as a group, participation. So if I want students to "do" attendance, then I have to measure it with a student learning outcome. I am now in the process of writing an outcome that reads, "Students will attend n-m classes where n is the number of classes in the term and m is the number of absences allowed for that type of section (6 MWF and 4 TuTh)."

Yet the same outcomes accomplished approach can be done in Statistics. A "cross-tabulation" table can be made to track each student's accomplishment of each student learning outcome.

NameCompetencyn-m classes attendedCalculate sample sizeCalculate meanCalculate standard deviation
AndreciaCompetentYesYesYesYes
ComingalongaleenNot yet competentYesYesYesNo
WhupsterCannot become competentNoNoNoNo

Note the last student is barred from achieving competency by the n-m outcome: excessive absences. However, if they were to eventually accomplish all of the other outcomes, why should they fail based on noncompliance with one single outcome? Does a student have to master every single outcome? Or are there core outcomes/competencies and peripheral outcomes/competencies? Again, PE makes this possibility the clearest. In order to run and juggle a juggling pattern called an inside cascade is used.

I also require students to learn the shower pattern while standing still, but non-mastery does not impact on core joggling ability. All my years of running and juggling and I can't run and juggle using a shower pattern. The difficulty is timing issues and the difficulty of de-synchronizing one's arms and legs. So if my student's cannot perform the shower pattern there is no serious impact on their ability to accomplish the core outcomes surrounding joggling.

Whether or not there are core and peripheral outcomes, one other facet of this outcome oriented world has not been touched upon. Since a student is being required to show mastery of all outcomes, nonattainment of an outcome means reteaching the outcome. This does not necessarily mean bogging down in a class because of a few members of the class.

There seem to be a couple systems for handling nonaccomplishment. One system is to allow the student a single second chance at performance. Another system is design a spiral curriculum where old outcomes come back up over and over. I believe there are core and peripheral outcomes. For statistics being able to calculate the mean and standard deviation are core competencies, being able to run a hypothesis test using the t-test is likely a peripheral competency as some statisticians believe confidence intervals should be used and not hypothesis tests.

Students calculate the mean on multiple quizzes, each test, and the midterm. During the next class after a quiz, test, or midterm, I go over the material again. In effect, I reteach on an ongoing basis. Statistics lends itself well to doing this.

I have yet to think about running Ethnobotany under a competent/not competent outcome orientation. I need to clarify my own thinking as to what students should know, do, or value, to paraphrase Sue. I have an outline with student learning outcomes, but the outline is not yet ready to use in a cross-tabulated manner.

I am also left thinking about enrichment and opportunities of the moment. Sometimes I walk into IS 201 and relate a new development in the world of computer technology. I might even put the new information on a quiz or test. Yet it is not in the outline because it did not exist when the outline was written. In a cross-tabulation gradebook with only thumbs up or thumbs down assessment of previously determined outcomes there is no way to add an outcome on the "fly." And if one does add an outcome on the fly, then one has to take the outline to the chair for reapproval through curriculum committee.

Courses like ethnobotany do not have one single core approach or established required knowledge sets. This is a young field that is still developing. Outlines would likely change from instructor to instructor. I know some people are certain that if worked at hard enough a single outline could be generated consisting of core skills. But the longer and harder I look at ethnobotany the less that seems likely. Outlines in some courses will change with changes in the instructor.

This does mean that for articulation and other purposes a database of old outlines will have to be maintained for reference.

I also wonder about a course such as one I took here in 1992. I took John Haglelgam's Compact of Free Association. The course was fascinating. I was riveted from day one to the end of the course. The course was primarily a lecture course, and we were being lectured by someone who had been there first person. Almost everything I know about modern Micronesian political history and the Compact comes from that course. Evaluations were by essays and involved reflection on the part of the students. The course is a real life story told chronologically. There is no single daily outcome to be achieved, yet the course is cram packed with solid information. How a course like this maps into specific outcomes I am not certain. I remain concerned about a potential loss of richness - the course is a 48 hour story that surely varies in structure from term to term.

As an institution we are about embark on a bold journey to go where we have not gone before. My own thinking is that a whole bunch of people who have worked with many institutions in many settings over many years realize that this is the way to go. I also gather that WASC is coming to the table last with this move towards measurable student learning outcomes with institutional accountability for documenting that the outcomes were met by students as individuals. My understanding is that other regional accrediting commissions are farther down this road that we are. I often wonder whether grade inflation was one factor that initially drove this movement. Grades were or have lost meaning in many schools.

I have concerns and misgivings, and I know the road will be difficult at times, but I also went through the following experience. Unprompted, a student came to me talking about this wonderful class they were in. The student really liked the class because they knew how they were doing each and every day, and they knew what skills they were trying to master on any given day. They noted that they were never sure how they were doing in their other classes, and were never certain what the skill or knowledge set was that they were supposed to absorb. The wonderful class is one operating on a system where the students have folders with the course outcomes in their class folder. As I understand it, when the students shows competency in an outcome they get that outcome marked as competent. They can see their progress and the road ahead.

I hope to rewrite my own course outlines, although I am now wrestling with what skills are core and what are peripheral. Recognizing that grades are probably not about to go away except in courses such as PE, I am hoping that peripheral competencies can be used to sort out the A's from the C's.

All of this is strictly my own opinion. Pardon me if I have misquoted, or misinterpreted. I apologize in advance for my errors, for I know they are many.