Over the past five years, teachers in the St. Charles, Illinois, Community Unit School District 303 have written and refined hands on science units for the elementary classrooms. At the same time, we have been developing more authentic approaches to assessing the learning and growth of K 12 students. As a science teacher who has been serving on the curriculum writing team while also researching authentic assessments, I would like to share our experiences and the lessons we've learned.
Our first and most important lesson was that students need several ways to demonstrate their learning. This became clear to me when I taught a unit on electricity at the Wild Rose School. One 5th grader responded with correct and appropriate answers on every written test and assignment. But when an assessment required her to construct different electrical circuits and solve problems that came up, she was unable to demonstrate a clear understanding.
Another student had the opposite problem. He had difficulty expressing his learning through written exams but could easily construct and solve problems of electrical circuitry. Thus, if I had used paper and pencil tests exclusively, he would have failed science. Instead, he ended up tutoring the classmate who was strong only on the written assessments.
Four Authentic Approaches
- An assessment that requires students to individually sketch their answer to the question on levers (fig. 1). This mirrors the National Assessment of Educational Progress science tests, which incorporates questions that ask students to draw their answers.
- An assessment that requires students to work in groups to construct and explain how three pulley arrangements work. Once the group constructs and practices using the pulleys, they are asked questions about the mechanics of the pulleys. Group members respond out loud.
- A take—home exercise that requires each student to connect with the world outside the classroom the concepts of inclined planes, wedges, and screws. Students must describe examples of these simple machines and explain how they work or might be modified (fig. 2).
- A complex performance in which students may either design and construct a complex machine that performs some task (fig. 3) or, alternatively, take apart a compound machine and show how the simple machines inside make it work. In my class, students built contraptions ranging from a banana peeler and egg cracker to a mechanical snowball thrower. They also disassembled everything from wind up clocks and toy cranes to record players and bicycles.
Figure 1. Lever Assessment for 5th Grade Science Class
Your parents moved a bookcase from one room to another. Your favorite book is not on any of the shelves and you can't find it anywhere else in the house. You suspect that when the bookcase was moved, the book may have fallen underneath it. Because you need it right away, you do not have time to empty all the shelves so the bookcase will be light enough to move. No one else is available to help you. You find a strong board and a couple of bricks. How can you use them to check under the bookcase for your lost book? Draw a picture clearly showing how you would try to get at the book. Include yourself in the picture.
Figure 2. Simple Machines Assessment for 5th Grade Science (Inclined Plane, Wedge, Screw)
Directions: To show what you have learned about simple machines, answer the questons below about the inclined plane, wedge, and screw.
Kind of Simple Machine: Inclined Plane
Example: What does it do? How does it help us do work? What could you change on it? How would it work differently?
Figure 3. Section of the Simple Machines Final Project
Choose one of the following three options. Remember that every simple machine helps us do work easier by multiplying the force or changing the speed and possibly the direction of the force. If necessary, you may work with a classmate. There are lots of resources for ideas or information. One wonderful book is The Way Things Work.
1. Go-Nowhere Machine: Design a go-nowhere machine on paper. Remember that it is a compound machine made of several simple machines. (See the Rube Goldberg cartoons game for examples.)
Get some feedback on your design from other students and your teacher. The modify your design to make it more workable.
Go ahead and gather the materials needed and construct it! You will demonstrate the machine to us. Be ready to explain how it makes work easier.
By intentionally planning a variety of ways for students to display or explain what they had learned throughout the unit, we gleaned a more accurate picture of individual and group learning. No one authentic assessment provided a complete picture of what each student had learned. Some were richer and more complex, but each provided a piece or pieces of information needed to give that student feedback and to inform further instruction.
Further, by using these four types of assessments, we gave all students a fair chance to show what they had learned. This fits in with Perrone's (1991) suggestion that teachers use observations, performances, and test like procedures for student assessment.
But Is There Time?
Once we designed these assessments, their practical application confronted us with a number of challenges. First was the amount of time it took teachers and curriculum specialists to develop assessments that were relevant to a specific school district culture and community. Of course educators can simply purchase curriculums and assessments. But authentic assessments should be tailored to the specific context in which they are used (Wiggins 1993).
Our district gave teachers opportunities during the summer to work with specialists in devising the assessments. The district also gave release time during the school year so that teachers could gather feedback on assessment implementation. The district even offered classes for credit on performance assessments.
We quickly discovered, too, that performance type assessments take more classroom time to implement. Would it be wise for us to take up our time this way? As we struggled with this issue, we came to view assessments as educational as well as summative (Glaser 1988). We concluded that the best way to collect data was to design assessment measures that resembled actual learning tasks.
Are We Skilled?
Another challenge we faced was the skills needed for complex authentic assessments on the part of teachers as well as students. Although some schools focus on the experiences and skills that both will need, others completely ignore teacher preparation, and some ignore both.
In our case, the mechanical skills needed to construct these compound machines varied among students. To prepare all the students, we used numerous hands on activities both in the classroom and as homework. To stretch their thinking, we showed them many models of compound machines, such as the Mousetrap game and Rube Goldberg's humorous drawings.
As for teacher preparation, we wanted our staff to be better versed in a variety of teaching styles, thinking this would equip them to use a variety of assessment approaches. Some teachers needed help with cooperative groups; others needed more practice with mechanical objects and tools.
Consistent and Reliable?
One final challenge was the reliability of the assessors. As Raizen and colleagues (1989) have noted, most teachers struggle to interpret assessments consistently. Because our teachers had varying degrees of experience with authentic assessment, it was difficult to compare the results among classes. By way of solution we are developing clear standards for some key assessments. We have involved teachers in establishing these standards and they are interpreting them collaboratively, as well as engaging in some self and peer assessment.
As Wiggins (1993) has suggested, we are also allowing students to have a say in the process and some practice in interpreting assessment tools. Figure 4 shows an assessment tool that my 5th grade class helped me create. As a result of helping me determine what was of most worth in our culminating assessment, my students found it easier to assess their own work and the work of their peers. They also had a clearer idea of what was expected of them.
Figure 4. Rubric to Assess Final Project on Simple Machines
Pulleys, Planes, and Student Performance - table
Above Standard | At Standard | Below Standard | |
|---|---|---|---|
| Use of Simple Machines | Invloves more than 4 kinds of simple machines. | Involves 3–4 kinds of simple machines. | Involves fewer than 3 kinds of simple machines. |
| Understanding of Simple Machines | Can clearly and in detail explain how it works. Can clearly and in detail explain how it helps us do work. | Can generally explain how it works. Can generally explain how it helps us do work. | Has difficulty explaining how it works. Has difficulty explaining how it helps us do work. |
| Quality of Project | Very dependable and almost always works. Display or video or construction is of superb quality (outstanding appearance, durable materials, clear labels, well organized). | Fairly dependable and usually works. Display or video or construction is of good quality (god appearance, adequate materials, decent labels, not whipped together at last minute). | Seldom or barely works. Display or video or construction is of poor quality (sloppy appearance, inadequate materials, unclear or missing labels, looks like it was put together at last minute). |
Bring It Home to Parents
Many parents are unfamiliar with the assessments schools use in content areas. To engage and inform parents, we invited them to come and view the final projects. We even asked them to help us assess projects using the assessment tool the class designed (fig. 4). This forced students to articulate what they had done and to explain why and how the machine worked. By talking about the assessment tool, many parents came to accept the absence of a letter grade for each performance.
Although each student was responsible for constructing his or her own compound machine, some needed a parent's or sibling's help to videotape their finished product because it was too large to bring to school or connected to some structure outside of school. (One girl, for example, designed a Rube Goldberg like pet feeder that consisted of a string and pulley tied to a rocking chair and food supply. When the chair tipped as one was rocking, it would pull the string, setting in motion a process that ended with food being poured into a bowl.)
By involving some parents in videotaping projects, we elicited a more favorable parent response to our performance assessments. We also encouraged parents to teach their children to use tools safely, which restrained them from getting too involved and making the projects their own.
Tie It to Curriculum
As a school district, we have learned that the assessments must match the curriculum. If they do not, a dual system of teaching may result one for standardized tests and one for good instruction (Darling Hammond & Wise 1985).
Based on my review of 15 of our 5th grade science classes, we concluded that when teachers use authentic assessments, student performance on standardized, multiple choice science tests does not necessarily improve. Student performance did improve significantly on tests requiring students to draw or construct their responses, much like one of our performance assessments. When schools or school districts invest time and money in developing and implementing performance assessment, they might not see significantly higher achievement on standardized multiple choice tests.
In spite of the time, effort, and difficulty we faced and continue to wrestle with, we believe that the assessments were worth the trouble. But performance assessments are only valuable in the context of sound educational practice. I am convinced of this when I see students actively engaged in relevant issues or problems, seeking and creating solutions with teachers, and with parents learning and working alongside them.
