Network News

X My Profile
View More Activity

Posted at 9:00 AM ET, 04/28/2010

How to measure student growth

By Valerie Strauss

Teacher evaluation has emerged as one of the most controversial issues in school reform today. My guest today to discuss evaluation models is Lisa Guisbond. She is a policy analyst for the National Center for Fair and Open Testing, known as FairTest, a Boston-based organization that aims to improve standardized testing practices and evaluations of students, teachers and schools.

By Lisa Guisbond
Growth models are pitched as a thoroughly modern, scientific way to answer the age-old parental question: “What did you learn in school this year, my dear?”

It seems reasonable enough: Test students at the beginning and the end of the year and measure the “growth” in test scores to see if student achievement is improving. Given the narrow and limited nature of the data used to measure this growth, however, there’s often less than meets the eye in the results. Fortunately, there are better, if little known or understood, ways of both measuring and stimulating real growth in learning.

The Florida legislature recently tried to climb on the growth model bandwagon, passing Senate Bill 6, which would have spawned a whole new series of tests to both measure student growth and evaluate teachers with the results. Floridians weren’t buying that bill’s expensive, simplistic and heavy-handed approach, however. Once Florida Gov. Crist got an earful from tens of thousands of his constituents, with 22 opposed for every one in favor, he did the right thing and vetoed the bill.

Nonetheless, there’s widespread acknowledgment that we need better ways of measuring student growth. No Child Left Behind’s approach of comparing one year’s fourth graders to the next, for example, makes little sense.

Teachers and many parents understand that a class’s makeup can vary widely from year to year. If 4th grade test scores leap from one year to the next, it may appear that the teacher has suddenly gone from lackluster to superior, but what really happened was two or three high-achievers moved into the neighborhood, or a child with severe academic challenges moved out.

Lost in the shuffle is the trajectory of individual fourth graders as they tackle fifth grade work. It means a lot if a child goes from being unable to read to being an enthusiastic reader and confident member of the class, even if test scores are not yet “proficient,” but NCLB doesn’t recognize this as an achievement.

So measuring growth would be a big step forward. The problem with most current systems is they measure growth by using standardized test scores in a few academic subjects, usually math and reading, which are not a very accurate or comprehensive way to check on overall student progress.

They simply leave out too much that matters, including other academic subjects, like social studies and science, electives, as well as an array of skills and capacities we expect students to be developing, such as critical thinking, problem solving, creativity, collaboration, communication.

Measuring student growth without relying solely on narrow standardized tests involves looking at multiple measures of student learning, such as essay exams, portfolios of students’ work in various subjects, and group projects that require analysis, investigation, experimentation, cooperation, and written, oral, or graphic presentation of findings.

An individual student’s learning growth in one year or over the course of several years could be assessed by creating a record of student achievement that includes representative examples of student work.

Standardized test scores could be one piece of evidence in such a record, as well as examples of essay exams, group projects, and so on.

Such a record would offer concrete evidence of a student’s growth in writing, for example, showing the improvement from a weak essay early in the year to a strong, organized, grammatical and compelling piece of writing later on. A DVD could be included to document the growth in a student’s ability to speak in front of the class, as could be examples of group projects from earlier and later in a student’s career, showing improvement in the ability to communicate and collaborate. (More details are in this report from the Forum on Educational Accountability.)

There are several important payoffs from using multiple measures.

First, measuring a broad array of subjects and skills encourages teachers to teach them all and try to do it well. It’s well known that NCLB’s narrow focus on reading and math test scores meant that too many students, especially poor students, ended up with little in their school day other than preparation to take tests in math and reading. How deadly can you get?

If we moved to using these richer measures in all of our schools, it could mean that not just upper middle class and affluent kids at schools like Sidwell Friends School, where President Obama's children attend, but poor public school kids too would be taught and expected to demonstrate the ability to think, express original ideas, problem solve, work in teams, do research papers and other kinds of projects requiring critical thinking.

Second, measuring and promoting this kind of learning would accomplish Obama’s stated goal of helping make every kid college- and career-ready in ways that test prep simply doesn’t.

College professors have probably always complained that too few students are ready to tackle the demands of college-level work. Employers too have long complained that they can’t find qualified workers to fill their openings.

But more and more college instructors are noticing that kids nurtured on test prep are simply unprepared for the kind of reading, thinking, writing and inquiry demanded at the college level. It’s left up to them to teach freshmen how to think.

Third, by using these kinds of high-quality assessments, we’d be following in the footsteps and reaping the same benefits as high-achieving nations like Finland and Singapore. We might be able to stop incessantly worrying that foreign students will eat our children’s lunch (or is it have our children for lunch?).

The slightly more complicated part is taking this array of multiple measures information and using it systematically to determine growth in a way that can be used to hold schools accountable.

Next time, we will introduce how that can be done.


Follow my blog all day, every day by bookmarking And for admissions advice, college news and links to campus papers, please check out our new Higher Education page at Bookmark it!

By Valerie Strauss  | April 28, 2010; 9:00 AM ET
Categories:  Guest Bloggers, Lisa Guisbond, No Child Left Behind, Race to the Top, Standardized Tests, Teachers  | Tags:  Race to the Top, growth models, how to evaluate teachers, nclb, no child left behind, schools and growth models, standardized tests, teacher assessment, teacher evaluation, teachers  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: New study: 1 in 3 college students transfer
Next: Social media addiction: Worse than you think


This article is right on, and triggers a couple of random thoughts:

"Summer School", an early Mark Harmon comedy film about a PE teacher that is forced to teach failing students English in the summer,has a great finale regarding
some kernals of truth in student motivation and progress.

Ms. Guisbond, other researchers and teachers KNOW that there are many ways of assessing students, but until we get away from our factory made, cookie-cutter models of testing, we will short-change students and schools.

I think fear of change and the dollars involved in other assessments is holding us back.

Posted by: PLMichaelsArtist-at-Large | April 28, 2010 9:43 AM | Report abuse

You hit the nail on the head.

I find it ironic that Sidwell Friends and other high end private schools aren't even close to basing everything on Standardized Test Scores.

Posted by: celestun100 | April 28, 2010 10:10 AM | Report abuse

Recently I looked up Sidwell Friends and other exclusive, private schools. Almost without exception their sites show children engaged in higher-level learning: looking through microscopes, reading, participating in field trips, etc. Contrast this with urban public schools that tout higher test scores and even athletic victories.

Of course a teacher can be evaluated by looking at student growth. However, it takes a real person to go into each classroom in the fall, determine the child's level of achievement and then assess again at the end of the year. This person would take into account the fact that a child missed one third of the school year or another child started out not knowing the alphabet at the age of ten.

There IS no single test to evaluate both child and teacher. Right now people want a quick and easy way to determine who is effective and who is not. Well, there isn't any way to do this at the present time.

Posted by: Linda/RetiredTeacher | April 28, 2010 11:52 AM | Report abuse

I suggest that our administration look to Sidwell Friends as a model of education. All schools will do what they do and all schools will have their class sizes and resources. Let's close the gap for real. Then I am all for linking teachers evaluations to student growth.

Posted by: celestun100 | April 28, 2010 12:00 PM | Report abuse

Typical response from Fair Test. "...looking at multiple measures of student learning, such as essay exams, portfolios of students’ work in various subjects, and group projects that require analysis, investigation, experimentation, cooperation, and written, oral, or graphic presentation of findings."

Lisa, if you think folks are gaming the system or outright cheating on the state multiple choice tests, you ain't seen nothing yet. The essays, projects, reports, portfolios, etc., will be compromised till the cows come home.

How in creativity will anyone ever be able to determine who actually did the work on one of these subjective assessments and beyond that, how will they be impartially graded? Will the work be done, in fact, by the student or a friend, a classmate, the teacher, a parent, a sibling, a classroom aid, a guidance counselor, an administrator, a neighbor, a relative, etc., etc.? NO ONE WILL EVER KNOW USING THESE KINDS OF ASSESSMENTS.

Again, this is so typical of Fair Test to offer these kinds of alternative assessments, anything to avoid reality.

Posted by: phoss1 | April 28, 2010 3:41 PM | Report abuse

To phoss1: Have you ever taught? Every teacher I know who knew their students could tell when outside work was not their own - BECAUSE they monitored their students' work while they were in class. One of a teacher's most difficult tasks can be having to confront a parent who has been a little too enthusiastic in 'helping' their child.

And there are other difficulties with extracurricular "help", but it's also true that most students who do well benefit from genuine help, i.e. guidance and clarification from other people outside the school. The sad truth for many disadvantaged students is that there isn't a caring person to help them when they
get to a stuck point on their homework.

Finally,just because other assessment methods have inherent difficulties does not mean they shouldn't be used. It does probably mean that those doing the assessments will have to do so carefully, thoroughly and with adequate time.

Posted by: PLMichaelsArtist-at-Large | April 28, 2010 4:40 PM | Report abuse

I think there are concerns about using portfolio assessments because we all know how many parents spend tons of time helping their kids with projects, portfolios, take-home exams, term papers, etc. This raises the question of whether these products really show what students can do.

When FairTest proposes alternative assessments, we are primarily talking about using classroom-based activities and individual work products. That way, it should be clear that these are authentic measures of what a student can do, not something produced with a lot of help, or entirely by someone else.

Plenty of people share the concern about "subjective" grading, but many also recognize that some Olympic events, for example, are judged by "subjective" human beings in a manner that most people find fair and valid.

Posted by: guisbond | April 28, 2010 6:44 PM | Report abuse

" should be clear that these are authentic measures of what a student can do, not something produced with a lot of help, or entirely by someone else." Whenever I hear "authentic" associated with assessments or education of the "whole child" I get very suspicious of their origins. Sounds too NEA for me. Not sure they can be considered authentic when the kid sitting next to them "helps" them at every turn. How much good does this really do the student in question?

And as I raised the issue above, how do these projects, reports, portfolios, etc., then get fairly/impartially graded by the teacher? Next to impossible. And they'll be no Soviet judges to artificially/politically raise the scores of neighboring bloc countries either.

And to PLMichaels, 34 years as a Massachusetts public school teacher (elementary classroom teacher), not an artist at-large; not that there's anything wrong with that.

Posted by: phoss1 | April 28, 2010 7:12 PM | Report abuse

So Phoss1,
What is the purpose of assessment? If we listened to your argument, students that attend Sidwell Friends school aren't any smarter than kids in disadvantaged schools because, obviously, their teachers subjectively grade their projects etc.
I understand assessment to be the process of determining where a student is at according to a learning goal or understanding, and then moving them towards that standard. No test is perfect and all tests are biased.
But phoss1, do you really think that a multiple choice test is the best way to determine if a child knows how to read complex text, analyze it, evaluate it, and then apply it to other life areas?
Can you imagine doing this in any other profession? I want a doctor who can perform an operation and not just know how to take a multiple choice test that names the body parts.
I want a builder who can build a house and not just answer multiple choice tests on what goes where etc.
If these people can perform these tasks, they will be able to also answer the lower level questions a multiple choice test would ask.
No doubt you've heard the analogy of comparing this high stake testing culture to putting someone on a scale. Continuing to weigh them won't change their weight. Our education woes are due to a system that has yet to figure out how to exercise, and eat right. Using other assessment forms will do this for us.

Posted by: tutucker | April 29, 2010 12:35 AM | Report abuse

what a waste. another cottage industry. a competent principal "knows" a competent teacher. he/she knows (should) all the variables that need to be considered. late blooming, motivational and learning disabilities are only the tip of the iceberg as to how "UNSCIENTIFIC" coupling student achievement to effective teaching practices is. consider, failing all students who do not reach adequate growth. in areas that have done this adequate growth has risen substantially. in my experience, 20 years, high achieving elementary, the students are increasingly enabled and fearless of any accountability. if you are going to look at japan and other high achieving foreign countries, look at their homogeneous nature and their expectations and accountability toward their students by their culture, parents and the educational system. you can lead a horse to water yet not make them drink. should teachers be penalized for this?

Posted by: agra09 | April 29, 2010 3:49 AM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company