Assessment

Why is assessment important?

In an era of shrinking budgets and data-driven decisions, assessment has become an important tool by which information about students’ knowledge, skills and abilities can be gathered to inform educational policy and decisions at a federal, state, district, school and classroom level. Although the purposes and strategies of the assessment may differ at each of these levels, the overall need for accountability and desire to modify instruction based on identified individual or group needs is present in some form for all levels of assessment.

There are many large-scale assessments that are used to assess student achievement both nationally (National Assessment of Educational Progress [NAEP], Trends in International Mathematics and Science Study [TIMSS], Program for International Student Assessment [PISA] and ACT) and statewide (Minnesota Comprehensive Assessments [MCA] in Science). However, in this document we will focus on assessments that are within the teachers control and are best suited to guide instructional practices and student accountability at the classroom level.

“Both assessment and evaluation are based on the judgment of an experienced, thoughtful human being—an expert. Machines don’t assess, papers don’t assess, tests don’t assess. Humans assess. And what better person is there to assess the progress and development of his or her students than the classroom teacher?”

—Larry Malone (Hart, 1994)

What is assessment?

Assessment is a systematic process to gather information about students’ knowledge, skills and abilities. While there are many different types of assessments used in the classroom, all assessment types can be categorized as diagnostic, formative or summative depending on the intended use and structure of the assessment. It is by using a variety of assessment types that are tightly integrated with curricular instruction that valid and reliable inferences can be made about a student’s knowledge, skills and abilities.

Relation between Instruction and Assessment (Gronlund, 2006)

Instruction Assessment

Diagnostic assessments are intended to be given before instruction begins to gauge a student’s prior knowledge and skills so that instruction can be adjusted to meet the needs of the student. In a classroom, this might resemble a pre-test that would be given before a unit to check for background knowledge and misconceptions or a more in-depth assessment to understand learning deficiencies that may inhibit learning outcomes. The key to using diagnostic assessments involves using the data to guide and adjust curricular material and instruction before it has begun to facilitate building on the student’s knowledge, and to address the misconceptions that are brought into the classroom. Diagnostic assessments are low stakes for the student and not graded.

Formative assessments are used “for” learning and should occur regularly and frequently throughout the instructional process. Formative assessments are low stakes for the student and not formally graded or used to determine a student’s grade. In fact, almost all interaction with students about their work could be a form of formative assessment. Because these interactions can help guide instruction and learning, it is often difficult to separate the relationship between teaching and formative assessments (Bybee, 2002).

Some of the questions that guide the formation and use of formative assessments in the classroom are:

• Where are you trying to go?

• Where are you now?

• How can you get there?

• How will you know when you get there?

Formative assessments can either be student-centered or teacher-centered depending on who takes responsibility for using the information to guide the next steps.

In student-centered formative assessments, the student uses the feedback from the assessment to take action to close the gap between the intended educational outcome and their current level of understanding. It also helps students learn how to learn and apply the expected educational outcomes during the course of doing their work, not just at the end. This is one way that assessment, self-assessment in this case, can become an integral part of the learning process (Atkin, Black & Coffey, 2001). In order for this to happen, quality feedback that is specific and understandable to the student must be given so that the student can use this information to take action. This type of formative assessment relies on student-directed learning using the feedback from the assessment to inform them of their strengths and weaknesses in the learning objectives.

Teacher-centered formative assessments are used to evaluate instructional effectiveness. In this case, information is gathered by the teacher and is used by the teacher to guide instructional activities and possible remediation to facilitate student progress towards specific learning goals. The use of common formative assessments developed and used by a team of teachers can benefit both the teacher and students alike. Common formative assessments can be used to identify individual student understanding and evaluate teaching strategies that are most effective in helping students acquire the intended knowledge skills and abilities (Dufour, 2006).

Summative assessments are a more structured, formal evaluation usually given at the end of a chapter or unit, following instruction. The results of these assessments are typically recorded and are high stakes for students. Inferences that are made about student achievement based on assessment scores depend on the validity and reliability of the assessment. When making inferences, these two attributes of assessment are important and should be considered early in the development of the assessments (Gronlund, 2006). Developing and delivering common summative assessments can also provide data that can be used to identify instructional practices and curricula that enhance learning.

“When the cook tastes the soup, that’s formative; when the guests taste the soup, that’s summative.” Robert Stakes

Planning & Instruction

How do I intentionally plan for and use assessment?

A comprehensive assessment system takes into account the multiple assessment needs and purposes within a course or lesson and incorporates a variety of techniques to get at these needs in the most efficient manner. Planning for instruction and assessment should be simultaneous and encompass the diagnostic and formative assessments that will be a part of instruction as well as the summative assessments which provide information on the resulting student learning (Enger & Yager, 2001).

Plan to use a variety of assessment techniques and question types to minimize assessment bias which results from individual learning styles. The choice of assessment type will depend on whether the purpose is diagnostic, formative or summative, which skills or knowledge are being assessed, and the amount of time available for creation, delivery and evaluation of the assessment. The choice of question type is influenced by several factors, including which type is the most direct and efficient measure of the learning objective.

Summary of the Relative Merits of Different Question Types modified from (Gronlund, 2006)

In order to be effective, the assessment must measure what was intended (Gronlund, 2006). Question and task alignment to the intended learning objectives is a critical step in creating valid and reliable assessments. When making decisions about alignment, determine what knowledge or skills are needed to answer the question and evaluate whether those knowledge and skills reflect either a portion of or the entire learning objective. Questions or tasks that reflect a larger portion of the objective are more strongly aligned, however if the goal is to modify instruction, it can be important to evaluate student understanding of specific parts that comprise the whole objective. In addition to its importance in validity and reliability, an alignment task done by collaborating with colleagues will often have the additional benefit of generating deeper discussions of that learning objective, including depth of instruction, appropriate vocabulary and expected cognitive range.

Assessments should include questions or tasks that assess multiple cognitive levels within an objective. If the assessment only includes questions at a single cognitive level, the instrument is not evaluating the full range of student skills and understanding for that objective. The expected range of instruction and learning for a particular objective will influence the appropriate cognitive range for the assessment.

In all assessment planning, a collaborative process to developing and making decisions is an important step in building a comprehensive assessment system supported at the school or district level. The creation and evaluation of assessment tools provides a broader range of perspectives and increases the connections across all instruction at the school or district level by including colleagues in the planning. It can also generate efficiencies in the amount of time needed to create and refine the assessment while also drawing on a larger pool of student results to make inferences about student learning.

TALK: Reflection & Discussion

What will we assess?

• What learning objectives do we currently assess in our course, school or district?

• Are there weak areas in current assessment results where we would like more feedback on student learning?

• What are the learning objectives that we want the assessment to focus on?

How will we assess?

• Which type of assessment (diagnostic, formative or summative) is most appropriate to the situation and to measure the desired outcomes?

• What is the timeline and schedule to create the instrument and who will be involved?

• How often will we assess students?

What will become of the results?

• How will the student responses be evaluated and what feedback will be provided to students?

• How will we analyze results and evaluate necessary instructional changes?

• What is our plan to be able to modify instruction based on the results?

DO: Action Steps

• Evaluate assessments that are already in use and look for ways to increase the diversity of cognitive levels and item types present in these existing assessments. This could be done by writing additional questions, making modifications to existing questions and tasks or adopting questions from other sources. See the additional resources section for references on Page Keeley’s Uncovering Student Ideas in Science series and Science Formative Assessment: 75 Practical Strategies for Linking Assessment, Instruction and Learning which include some formative assessment strategies and questions.

• Avoid introducing test bias that may impede some students. Avoid unnecessary vocabulary when developing assessments. Use the simplest possible sentence structure and review the directions, question and any stimuli or graphics for clarity. Students may struggle with certain question types due to the structure of the question, not the content. These difficult types of questions include completion items (The elephant is ____.) and negatively worded questions (Which is NOT the color of an elephant?).

• When writing open-ended questions, it is important to provide clear directions and expectations. Rubrics should be developed at the same time as the question and used to score responses consistently. Rubrics can and should be revised if a student’s response is correct, but unanticipated in the rubric’s structure. Use open-ended questions if there are multiple ways a student could express understanding or there are multiple correct answers. Open-ended questions are best for measuring abilities and skills requiring higher cognitive thinking and not for measuring recall of knowledge. If there is only one correct answer or way to respond, a multiple choice or other closed type of question is more efficient (Enger & Yager, 2001).

• Evaluate assessment questions or tasks for the following characteristics that can provide clues to the correct answer (Gronlund, 2006):

• Verbal associations which occur when the correct answer is the only option that reflects the same vocabulary as the question. It can also occur when the vocabulary or structure of the correct answer mimics the exact wording provided in instruction.

• Cueing occurs when material in the assessment provides answers to a different task or question. This material may take the form of background information, parts of other tasks or other questions

• The use of the words “sometimes”, “always”, or “never” in the question as well as answer options such as “none of the above” or “all of the above” are often clues to test savvy students, and correct answers may not reflect true understanding.

• Budget time for evaluation of the data. A study done with 70 school districts showed that best practice districts provided elementary teachers 3.1 hours a week to review and analyze student data compared with 1.9 hours a week in other districts. Other findings of this research showed best-practice districts also spent professional development time focused on how to use student data to help teachers make instructional changes (Maxwell, 2008).

References & Resources

Additional resources:

Education Northwest (previously The Northwest Regional Educational Laboratory) has put large sections of its training kit, Improving Classroom Assessment: A Toolkit for Professional Developers online at http://educationnorthwest.org/resource/700

The Exploratorium’s Institute for Inquiry has outlines and workshop materials for formative assessment professional development is online at http://www.exploratorium.edu/ifi/workshops/assessing/

Assessment question sources and formative assessment probes:

• Keeley, P. et al. (2005, 2007, 2008 and 2009). Uncovering Student Ideas in Science, Volumes 1-4. Arlington, VA: NSTA Press.

• Keeley, P. (2008). Science Formative Assessment: 75 Practical Strategies for Linking Assessment, Instruction, and Learning. Thousand Oaks, CA: Corwin Press.

• National Assessment of Educational Progress (NAEP). (2011). NAEP Questions Tool. U.S. Department of Education. Institute of Education Sciences. National Center for Education Statistics. Washington D.C. http://nces.ed.gov/nationsreportcard/itmrlsx/default.aspx

• Sadler, P.M., H.P. Coyle, N. Cook-Smith, J.L. Miller, J. Murray, & A. Trenga Rumpf. (2011). Misconceptions-Oriented Standards-Based Assessment Resouces for Teachers (MOSART). Funded by the National Science Foundation and located at http://www.cfa.harvard.edu/smgphp/mosart/index.html

Works Cited

Atkin, J. Myron., Black, Paul., Coffey, Janet. (2001). Classroom Assessment and the National Science Education Standards. Washington, D.C.: National Academy Press

Bybee, R. (2002). Learning Science and the Science of Learning. Arlington: NSTA Press.

Enger, S. K., & Yager, R. E. (2001). Assessing Student Understanding in Science: A Standards-based K-12 Handbook. Thousand Oaks, CA: Corwin Press, Inc.

Gronlund, N. E. (2006). Assessment of Student Achievement. Boston: Allyn and Bacon.

Hart, D. (1994). Authentic Assessment: A Handbook for Educators. Menlo Park, CA: Addison-Wesley.

Maxwell, L. A. (2008, July 22). 70 Districts Compare Practices on Collecting, Analyzing Data. Retrieved 2008, from Education Week.

Dufour, R. D., R. Dufour, R. Eaker (2006). Professional Learning Communities at Work. Bloomington, IN: Solution Tree.