Académique Documents
Professionnel Documents
Culture Documents
Thomas R. Lord
Donald P. French
Linda W. Crow
Arlington, Virginia
Copyright 2009 NSTA. All rights reserved. For more information, go to www.nsta.org/permissions.
Claire Reinburg, Director
Jennifer Horak, Managing Editor
Judy Cusick, Senior Editor
Andrew Cocke, Associate Editor
Betty Smith, Associate Editor
College science teachers guide to assessment / Thomas Lord, Donald French, and Linda Crow, editors.
p. cm.
Includes index.
ISBN 978-1-933531-11-3
1. ScienceStudy and teaching (Higher)Methodology. 2. ScienceStudy and teaching (Higher)Evaluation.
3. Academic achievementEvaluation. 4. Educational tests and measurements. I. Lord, Thomas R. II. French,
Donald P., 1952- III. Crow, Linda W.
Q181.C535 2009
507.11dc22
2008041750
NSTA is committed to publishing material that promotes the best in inquiry-based science education. However, conditions
of actual use may vary, and the safety procedures and practices described in this book are intended to serve only as a guide.
Additional precautionary measures may be required. NSTA and the authors do not warrant or represent that the procedures
and practices in this book meet any safety code or standard of federal, state, or local regulations. NSTA and the authors
disclaim any liability for personal injury or damage to property arising out of or relating to the use of this book, including
any of the recommendations, instructions, or materials contained therein.
Permissions
You may photocopy, print, or e-mail up to five copies of an NSTA book chapter for personal use only; this does
not include display or promotional use. Elementary, middle, and high school teachers only may reproduce a
single NSTA book chapter for classroom- or noncommercial, professional-development use only. For permission
to photocopy or use material electronically from this NSTA Press book, please contact the Copyright Clearance
Center (CCC) (www.copyright.com; 978-750-8400). Please access www.nsta.org/permissions for further information
about NSTAs rights and permissions policies.
Copyright 2009 NSTA. All rights reserved. For more information, go to www.nsta.org/permissions.
Table of Contents
Contributors ix
Preface xi
Section 1. General Assessment Topics
Chapter 1: Survey Instrument Validation: 3
The First Commandment of Educational Research
Cynthia Cudaback
B. Alternative Assessments
Chapter 26: Practices That Jeopardize Bona Fide Student Assessment 137
Thomas R. Lord
Index 155
W
elcome to the sixth in the series of the value and importance of assessment in a col-
SCST monographs created by the lege science setting.
members of the Executive Board The monograph examines assessment issues from
of the Society for College Science several different viewpoints and is broken into several
Teachers in cooperation with the National Science chapters. The first section deals with general assess-
Teachers Association. This document covers an ment topics such as validation of survey instruments
extremely important and often controversial topic, and creating a culture for faculty-owned assessment.
that of evaluating the value of students and pro- The second section concerns traditional and alter-
fessorial works. The jointly sponsored monograph native forms of assessment in both science and the
has been three years in the making with the initial science education classroom. The third section pres-
agreement with the National Science Teachers ents a series of how-to assessment practices that have
Association taking place in the fall of 2006. been successfully utilized in the field. Finally, the
Each submission in this monograph was re- fourth section provides a series of tips to enhance
viewed by at least two members on the Editorial assessment in the college science classroom.
Board with the published authors responding to The editors would like to thank all the contribu-
reviewers critiques and providing the final proof- tors to the monograph. The quality of the initia-
ing of their own entry. Articles were selected on tive is indicative of the time and energy they put
the quality of the writing and their contribution to into this work.
Acknowledgments
The editors wish to thank Holly Travis for her tireless effort and dedication to the construction of this document.
Thanks also to Ellen Yerger and Tom Melvin for their help in making this monograph a success.
help students be well prepared by using higher- 1. Knowledge: the ability to remember/recall
order multiple-choice questions for assessment of previously learned material.
course material starting in the freshman year. Examples of behavioral verbs: list, name, identify,
Assessment must match ones teaching style define, show
inquiry teaching must be followed by assessment Sample learning objectives in science: know com-
techniques that match the inquiry method of mon terms, know specific facts, know basic
Section teaching. If one follows the learning cycle (5E or procedures and methods
2
other similar models), assessment is encountered
throughout the teaching and learning continuum, 2. Comprehension (understanding): the
and that assessment must be related to the phase ability to grasp the meaning of material, and to
of the cycle (exploration, extension, etc.). Cer- explain or restate ideas.
tainly higher-order questions capture the essence Examples of behavioral verbs: chart, compare,
of exploration and extension much better than contrast, interpret, demonstrate
lower-order questions do. Sample learning objectives in science: understand
The use of higher-order questions does not facts and principles, interpret charts and
mean an end to using lower-order questions. graphs, demonstrate laboratory methods and
Rather, we are referring to a shift from the t procedures
8090% lower-order questions typically found in
college science exams toward a balance between 3. Application: the ability to use learned mate-
lower- and higher-order questions. The goal of rial in new situations.
undergraduate science instruction should be Examples of behavioral verbs: construct, manipu-
critical thinking rather that memorization. Many late, calculate, illustrate, solve
students come to the university with the assump- Sample learning objectives in science: apply con-
tion that science is just a lot of memorization, cepts and principles to new situations, ap-
and college instructors often need to work hard ply theories to practical situations, construct
to destroy that myth. However, that myth is often graphs and charts
kept alive by the choice of questions used on the
exams. If they favor knowledge-style questions, 4. Analysis: the ability to separate material into
then students will continue to believe that science component parts and show relationships between
is mostly about memorization rather than about the parts.
inquiry and analysis. Examples of behavioral verbs: classify, categorize,
organize, deduce, distinguish
Sample learning objectives in science: distinguish be-
Understanding Blooms tween facts and inferences, evaluate the relevan-
cy of data, recognize unstated assumptions
Taxonomy
Most college instructors are familiar, on some 5. Synthesis: the ability to put together sepa-
level, with Blooms taxonomy of learning (Bloom rate ideas to form a new whole or establish new
et al. 1956). Much has been written about the relationships.
use of Blooms taxonomy in the construction of Examples of behavioral verbs: hypothesize, cre-
exam questions, but few instructors take to heart ate, design, construct, plan
the need to use all of the levels instead of just the Sample learning objectives in science: propose a plan
first two in constructing examination questions. for an experiment, formulate a new scheme
Here is a quick review of Blooms taxonomy as it for classifying, integrate multiple areas of
relates to the teaching of college science. learning into a plan to solve a problem
other than the one the doctor wanted the shy away from and may distort the question as a
drug to have valid assessment tool. Likewise, use all of the above
b. is the main effect of a drug and none of the above sparingly.
c. additionally benefits the individual When all exam questions have been con-
structed, check each one to see where it falls in
Distracters (incorrect answers) must be incor- Blooms hierarchy. Construct a simple table such
Section rect yet plausible. If a recognizable key word ap- as that shown in Table 1 to see the distribution of
2
pears in the correct answer, it should appear in questions. If the questions are disproportionately
some of the distracters as well. Be sure to check distributed, then rewrite enough questions to bal-
will the answers help to distinguish why a student ance the exam between lower-order and higher-
got it wrong? This is an important part of assess- order questions.
ment that is often overlooked by instructors, but is
a critical part of helping students to learn.
Avoid microscopic distinctions between an- Examples of Multiple-Choice
swers, unless this is a significant objective of the
course. Be sure to stagger the correct responses in
Questions at Each Level
their order (use all answer positions as equally as The following examples illustrate the construction
possible). Limit the number of optionsmost au- of multiple-choice questions that fit the higher
thors agree that 45 answers is plenty, and there is levels of Blooms taxonomy. For most an expla-
no assessment advantage in using more than five. nation is included describing why it fits where it
Use all, always, never, none, etc., rarely. These are does, and what a student needs to know to be able
answers that students have been programmed to to answer the question correctly.
a.
b.
c.
d.
e.
f.
Total
3. During an otherwise normal pregnancy, a dents knowledge of the brain is used to estab-
woman begins to experience light-headedness lish a new relationship beyond those studied in
and a decline in energy levels near the end of class. All of the answers are logical outcomes
the first trimester. Which of the following is the of brain dysfunction and help the instructor to
most likely cause of her symptoms? pinpoint the misunderstandings that students
a. lack of B vitamins due to poor diet have.)
Section b. decline of blood pH due to overuse of
2
muscles 2. A neighbor found some mammal bones on the
c. decrease in blood pressure due to ex- grounda keeled sternum and an ulna, where
panding fetal circulation the olecranon occupies 30% of the length of
d. decline in estrogen levels due to ovarian the bone. These bones most likely came from
shutdown what type of mammal?
(All of these answers involve factors that could a. flyer
cause tiredness in a woman. To determine the b. climber
most likely cause in this scenario, a student c. runner
needs to understand the basic mechanics of d. digger
pregnancy and the biochemical changes that e. swimmer
occur during it. The answer given shows a stu- (Here a student must understand the various
dents ability to carefully analyze the situation bones and what their functions are. The stu-
and determine causality.) dent must then formulate a relationship be-
tween the type and formation of the bones and
4. The seeds of various plants vary in size from a the activity that it would promote in a mam-
fraction of a millimeter to several centimeters. mal.)
The most critical factor controlling the size
seed a plant produces is 3. What would be the most logical result of mix-
a. size of the maternal flower ing X and Y, both solubilized in distilled H2O
b. projected size of the animal pollinator at room temperature?
c. quantity of the abiotic pollinator a. precipitation of a solid
d. length of predicted dormancy b. a change in color of the liquid
e. method of distribution of seed c. a rapid rise in temperature
d. a rapid decrease in temperature
Synthesis Questions (To answer correctly, a student must put to-
1. Domoic acid, produced by diatoms, has been gether knowledge of X and Y as compounds
found to bind to hippocampal glutamate re- with knowledge of their dissociations and the
ceptors. If a person were to accidentally con- reactions of the individual components. All of
sume a lot of shellfish contaminated with this the answers reflect outcomes that the student
organism, what effect might be expected? has previously experienced when two com-
a. blindness pounds are mixed.)
b. deafness
c. amnesia Evaluation Questions
d. aphasia 1. Your fitness regimen involves jogging on the
e. rigidity school track 23 miles per day with a friend.
(To answer this question, one must understand On a particular day, about 15 minutes into
the role of the hippocampus and the role of your jog, your friend suddenly pulls up and
glutamate in this area of the brain. Here a stu- falls down, grasping her right calf in pain.
1. Assessments need to be frequent and scaf- 2. Assessments and exams are not always the
folded for legitimate success. Working adults same thing. Assessments come in many
dont want only one or two assessments of forms. In the assessment report I file each
their progress during the semester. They want year to the divisional office to show that my
to know exactly how well they are performing nonscience majors course is worthy of the ti-
in the class at each assignment. They want tle of science CORE(which means it meets
4
Section feedback, personal and directed. These are the criteria of inquiry, shows the processes
practical folks! Not surprisingly, they want and limitations of scientific thought, and
good grades, but they also want real success. analyzes data), no fewer than 12 different as-
They dont mind being challenged, especially sessment techniques are listed.
after theyve been successful. Successful gate- Assessments include the exams, concept
way instructors know that a 16-week semes- maps (two varieties), Vee diagrams, labora-
ter might have seven or eight large exams. tory reports, and capstone projects. All of
The first exam should be the least challeng- these give the instructor information about
ing. Each subsequent exam should be more the students learning and their mastery of
challenging. Students want their efforts to content. Alternative assessments that are real,
show. They want to believe theyre progress- targeted to the content can be more revealing
ing, that hard work pays off. than an exam. Anxiety plays a role in exam
The experts might suggest that assess- taking, but a student has control over a proj-
ments be formative, giving frequent feedback ect. Presentations or projects that allow for
toward the mastery of content. Classroom as- research, sharing of ideas, and collaboration
sessment techniques (CATS) are usually for- are valid assessments. These include contex-
mative assessments, and might include the tual, problem-, case-, or performance-based
quick think pair share or the one-minute assessments.
paper (Angelo and Cross 1993). Students 3. Embedded assessment across multiple sections
want definite feedback on all formative as- has advantages and disadvantages. Commu-
sessments. They want to know that youve sin- nity colleges are notorious for having large
cerely read each and every one. Formal grades numbers of adjunct professors. Mine is not
arent required, just your attention and con- an exception. Subsequently, as the full-time
structive remarks in some format. professor responsible for reporting on multi-
Summative assessments are aligned with ple course offeringseven when I am not the
the evaluation of content mastery or the instructor of recordmy job becomes very
completion of instruction. Many community challenging. Embedded questions on each
college students need state or national bench- exam allow for a logistically simple method
marks (standards) for their instruction in their for tracking all sections of a single course.
trade or vocational courses. My students see Instructors simply provide me detail of the
the state and national benchmarks for science embedded questions after each exam. Thats
education (Roseman and Koppal 2006), and the advantage. The disadvantage is that I have
they know that like other professionals, they no idea why the students miss the embedded
need to meet those standards. Unlike their questions on particular topics. The variables
certification exams for careers, my class is only are too numerous. I have a vague idea within
the beginning of their journey toward scien- my own classes, as I can monitor absences
tific literacy. Benchmarks and standards are or recall the day in class when the topic was
a goal for attainment with the expected out- discussed. There is no information from the
come of lifelong scientific literacy. classes I didnt teach. The number is a cold sta-
ten. A set of poorly written lab reports takes aging remarks on their progress (e.g., Nice
about twice as long. Each report needs care- improvement on this multiple-choice section;
fully worded constructive feedback. Rubrics keep working!) Statistical analysis is given
on written assignments (and oral for presen- on the entire class performance, and the class
tations) are given in advance, and students are discusses improvement strategies for the next
expected to follow the same criteria for excel- exam. Besides the exams, alternative assess-
4
Section lence in writing as they would in a composi- ments are a near daily occurrence.
tion class. Even on exams, common mistakes 11. Ideally, assessments inform teaching, and self-
are explained. The feedback on the exam is assessments can even inform the learner. Assess-
another opportunity to teach. ment is not only about evaluating the learning
9. Owning and expressing your expectations for process. It should change the teaching process.
their success is crucial. Students rise to the Each assessment should inform the instructor
challenge of high expectations. When given as to needed changes in pedagogy, presentation
rationale for a challenge, they accept. They or missing fundamentals for conceptual un-
will even accept the frustrations of disequilib- derstanding. Self-assessments can provide the
rium if they understand the rationale. Explicit learner with great potential to change.
reasons for content expectations are essential.
Community college students will accept the
because its on the test but are likely to ask The view from the gate as I encourage
you why its on the test. They want a more students to consider the science disciplines is
practical reason for learning the content. Ide- generally positive. Together, the students and
ally, the reason is tied to a potential career, or I investigate, listen to each other, and plan our
an everyday application. journey together. Although assessment reports
10. Assessments need to be varied, perceived as are needed for multiple levels (divisional, pro-
fair and attainable, and evaluated both objec- gram, departmental) there is enough consis-
tively and subjectively (Mintzes, Wandersee, tency across the requirements that only the
and Novak, 1999). The brain loves novelty. perspective changes. Not only have I learned
It fatigues when offered routine. With 12 how to assess my typical community college
different types of assessments throughout students so that I know what they are learning,
the semester, fatigue is more physical than Ive also learned how to teach better science
cerebral! Each exam has a variety of ques- through our shared assessments.
tion types. Students create or correct con-
cept maps; they evaluate true and false state- References
ments, correcting the false. They also write Angelo, T. A., and K. P. Cross. 1993. Classroom
brief answers and traditional multiple para- assessment technique: A handbook for college teachers.
graph essays. Each exam also has multiple- 2nd ed. San Francisco, CA: Jossey Bass
choice questions and paragraph comple- Publishers.
tions. Students analyze their exam results at
the conclusion of each unit and write goals Mintzes, J., J. Wandersee, and J. Novak, eds.
to improve weak performance areas. Stu- 1999. Assessing science understanding: A human
dents know how exams are evaluated. They constructivist view. San Francisco, CA:
know that each section is evaluated indepen- Academic Press.
dently without my knowledge of the test au-
thor. They also know when the sections of Roseman, J. E., and M. Koppal. 2006. Ensuring
the exam are totaled, I often write encour- that college graduates are science literate:
155
Copyright 2009 NSTA. All rights reserved. For more information, go to www.nsta.org/permissions.
assessment, defined, 7 Engagement, student, 9597
building, obstacles to, 8 Epistemological commitments underlying
challenges, 8 activities of science, 79
Commission on Future of Higher Education, 8 Evaluation, Blooms taxonomy, 37, 51
communication, 10 Evaluation questions to assess higher-order
Faculty Learning Communities, 11 thinking, 4041
faculty time, 8 Exam analysis, instructor perspective, 109110
foundations, 8 Exam corrections, analysis, student perspective,
isolation, 8 107108
No Child Left Behind Act, 8 Expert-teaching others in class, as alternative
planning, 910 assessment, 5354
results, 1213 Explaining to child, as alternative assessment, 53
skepticism, 8
values F
clarification of, 9 Face validity, defined, 34
structure reflecting, 1011 Faculty Learning Communities, 11
varying involvement, 8 Faculty-owned assessment, culture of, 713
visible, vocal leadership, 1112 American Association for Higher Education, 9
assessment, defined, 7
D challenges, 8
DAT. See Graduate entrance exams Commission on Future of Higher Education, 8
Data-oral alternative assessment, 53 communication, 10
Design/build/produce as alternative assessment, Faculty Learning Communities, 11
52 faculty time, 8
Diagnostic learning log, 9597, 96 foundations, 8
Distracters, multiple-choice assessments, 46, 46 isolation, 8
DLL. See Diagnostic Learning Log No Child Left Behind Act, 8
Documenting learning outcomes, multiple planning, 910
measures, 9 results, 1213
Donations Please project, 91 skepticism, 8
values
E clarification of, 9
e-mail structure reflecting, 1011
formative assessments using, 8788 varying involvement, 8
sending grades via, 153 visible, vocal leadership, 1112
e-portfolios. See Electronic portfolios Faculty time as obstacle to building campuswide
Edmiston, Dr. Judith, 109 assessment program, 8
Electronic portfolios, college science course Failure prediction, assessment as tool for, 2526
assessment, 5966 FCI. See Force Concept Inventory
assessing, 6163 Field-testing multiple-choice assessments, 47
checklist for, 62 5-E lesson plan development, science content
defined, 60 learning with preservice teachers, 8184
National Coalition for Electronic Portfolio FLCs. See Faculty Learning Communities
Research, 60 FMCI. See Force and Motion Concept
qualitative scoring, 63 Evaluation
rubric for, 63 Force and Motion Concept Evaluation, 2122
Embedded assessment, advantages, Force Concept Inventory, 18, 22, 25
disadvantages, 124125 Foundations of culture of faculty-owned
V
Validation of survey instruments, 36
construct validity, 4
definitions, 34
listening to students, 45
open-ended written questions, 4
principles of survey instrument development, 4
Validity, defined, 34
Values, clarification in building culture of faculty-