TLC's Selection of Instructional Topics:
Assessing Student Learning and Class Climate
There are three parts to this section:
Testing and grading fairly are vital. The assessment instruments chosen for a particular course must be in line with the stated goals, the content, and teaching/learning strategies. For example, the syllabus in Course A states that the course is meant to foster critical thinking. However, lectures and an emphasis on facts versus process dominate the class. Exams test rote memorization and use multiple-choice questions based on minor material instead of broader concepts. It is obvious that there are major inconsistencies in Course A. What was taught was not tested; what was envisioned was not created.
Course B also wants to foster critical thinking. Here, however, discussion of issues and a high degree of student interaction and input characterize much of the class hour. Essay questions asking for higher-order thinking combined with journals and term papers are the basis for assessing student learning. In Course B goals, strategies, and assessments are in agreement.
There are, of course, a multitude of ways to assess student learning. The Teaching and Learning Center has materials which outline different assessment strategies. The TLC can also help an instructor research alternative means for getting classroom feedback.
Careful planning prior to creating, distributing, and grading exams is well worth the effort and time. It is essential to make the nature of the exam explicit to oneself and the students. It is also critical to follow through on that design. In order to help this process, ask:
- What is to be tested? -- Exactly what knowledge, skills, and/or values are being tested? Are students expected to know certain facts in some rote manner, analyze or synthesize new material, manipulate certain variables, take a position on some issue?
- How can this best be accomplished? -- What kinds of questions (multiple choice, true/false, essay) are most appropriate? Perhaps an assessment tool other than an exam is more appropriate.
- Is the exam congruent with the course goals, materials covered , and activities? -- Is the material being tested consistent with what has transpired in the course? While it is one thing to test the ability to deal with new material using the information and skills acquired in the class, it is quite another to emphasize different information or skills on the exam than what was emphasized in class.
- What kinds of answers are expected and why? -- On an essay exam, writing out a sample answer helps in both the design of the question and the grading. Are the expected answers reasonable in light of course activities?
- How will the points be distributed? -- Do the point values match the effort expected in answering the questions? On an essay test, are there points for different sections of the answer or for recognition of different key concepts that might appear? Is there partial credit? Are the points on a continuous scale or in categories such as superior, satisfactory, and poor? What are the criteria?
While multiple-choice tests are meant to be objective, they are often perceived of as tricky, picky, unclear or convoluted. Here are some guidelines for designing multiple-choice questions:
- present a single, definite statement to be answered by one of the several given choices;
- keep student reading efforts to a minimum and eliminate irrelevant material;
- ensure that only one response is considered best or correct by experts in the field;
- avoid negative statements (if negatives must be used, capitalize, underscore or otherwise highlight the negative term);
- use plausible or attractive alternatives as incorrect response choices; and
- avoid giving clues to the incorrect option (e.g., grammatical clues or using "always" or "never").
In order to encourage critical thinking and to facilitate awarding partial credit, some instructors allow students to write a short rationale detailing why they chose a particular answer for a subset of the multiple-choice questions.
Some other suggestions for maximizing the effect of multiple-choice exams include:
- modeling sample questions in class and working through them;
- requesting a statistical output of student answers, reviewing these, and considering reasons behind common errors in terms of question validity;
- giving explanations on the more difficult questions that the class misses;
- using part of the next class period to discuss the exam in general viewing the exam as not only an assessment tool, but a way to review and reinforce concepts; and
- being open to feedback regarding the questions in order to modify the questions in the future.
Essay questions have advantages in that they are easy to prepare and can assess higher-order thinking; they also have disadvantages in that they cover a narrower range of content, take more time to score, and can be difficult to grade consistently.
Here are some suggestions to consider when constructing essay questions:
- use key words to specify the thinking process expected (for example, "define," " explain," "predict," " outline," " propose," or "critique");
- state the requirements clearly and avoid global, ambiguous questions which could result in a hodgepodge of answers;
- give sufficient direction to enable a well-informed student to respond in the manner expected;
- indicate the length of the response expected and perhaps an estimate of the time allocated; and
- indicate the relative weight of each question.
Some considerations for evaluating and learning from essay questions include:
- structuring the question so there is little room for subjective interpretation (unless that's the purpose of the question);
- making the grading criteria explicit;
- having specific responses in mind at least in terms of key terms, ideas, or procedures and for constructing and using a scoring key;
- breaking each answer into sections with point values to assist both grading and the posting and discussion of the results;.
- offsetting "halo effects" by concealing students' names when scoring, shuffling papers after scoring each question, and grading one question at a time (perhaps doing a quick review of all the answers for that question first to get a sense of the student responses);
- posting sample responses and the breakdown of point values within questions in order to serve as an additional teaching/learning tool, and to answer student questions regarding the exam.
In addition to giving a set number of questions on an exam, one might consider the following options:
- optional questions for extra credit;
- a choice of questions to answer;
- giving out questions in advance from which some subset will be tested;
- having the students write their own question and answer it.
For a checklist of strategies for test writing, see "Designing Classroom Tests: A List" in Teaching at UK, Vol 1, No. 3.
Enid Waldhart (Communications, UK) wrote, "Getting Ready: Some Preparation Tips" (in Teaching at UK, Vol. 3, No. 1), in which she describes effective strategies for requiring and evaluating student oral presentations.
Although grading is a difficult and confusing necessity, grading policies are rarely discussed by faculty at most campuses. There are both very practical and more theoretical issues associated with grading.
On the practical level, a good grading system should:
- accurately reflect differences in student performance;
- be clear to students so that they can chart their progress; and
- be fair (e.g., based on many different kinds of assignments over a semester vs. one big test).
Different competencies should have criteria of evaluation particular to each requirement. For example, Derek Lane (Communications, UK) wrote "Strategies for Effective Communication Assessment" in Teaching at UK, Vol. 3, No. 1.
The syllabus needs to present a clear picture of the grading system, including the amount of credit per assignment, how the final grades will be calculated, grade equivalents for final scores, and perhaps a rationale for the grading policy.
On a more theoretical level, there are many questions that might be addressed concerning the entire range of grading issues. While a detailed discussion of alternate grading procedures is beyond the scope of this handbook, some of the issues can at least be raised. Should grades reflect achievement relative to others in the same class (norm-referenced) or absolute achievement level according to a set of standards (criterion-referenced system)? Is the assumption made that student performance will follow a normal distribution? What's the purpose of grading? Ranking students? Recognizing attainment of some level of performance?
There is a whole range of questions concerning the components to be included in the definition of a "grade." Should grades include components such as attendance, participation, attitude, diligence, etc.? Will such practices makes grades even more ambiguous than they already are? How can such factors be assessed accurately and consistently? Should effort and attitude be recognized as well as knowledge and skills attained? Should grades report achievement or growth?
There are a number of ways to get feedback about teaching effectiveness, student learning, and classroom climate. Formative feedback can help an instructor reflect upon and develop his or her teaching skills. The Teaching and Learning Center can help an instructor gather formative feedback by:
- Videotaping a class, the TLC staff can also help review the video with the instructor;
- Classroom Observation, conducted after a pre-class meeting with the instructor during which time the specific items and activities to be observed are determined;
- Small-Group Instructional Diagnosis, held with students who, in the absence of the instructor and working in small groups, brainstorm responses regarding what aspects of the course to date have helped them learn, what has hindered their learning, and what suggestions they have for changes;
- The Teaching Analysis by Students (TABS) questionnaire includes three different perspectives on teaching effectiveness cross-analyzed by computer: students record their perspective of the course, the instructor details his or her perspective, and the instructor imagines how students view the course; and
- Incorporating the excellent resource book by Angelo and Cross, Classroom Assessment Techniques (1993) which has a collection of 50 ways to collect ongoing feedback regarding course-related knowledge and skills ( prior knowledge, recall, understanding, critical thinking, creative thinking, and problem-solving), learner attitudes, values, and self-awareness (students' awareness of their attitudes and values, students' self-awareness as learners, course-related study skills), and learner reactions to instruction (reactions to teaching, class activities, assignments).
Every instructor can survey students at various times during the semester. See the TLC abstracts of books about using a Myers-Briggs Type Indicator in the college classroom. Easy, yet effective ways of getting feedback include asking students to respond anonymously to the following sort of prompts:
- If I am to achieve my goals in this course,
I need to
do more ... (describe),
do less ...,
continue to ...
- I would like the instructor to:
continue to ...,
do more ...,
do less ....
- What has helped me learn in this class?
What has made learning difficult?
What suggestions do you have for changes?
Posted July 1, 1997
http://www.uky.edu/UndergraduateStudies/tlc/topic/assessing.html
Back to Main Menu
Back to Topics to Explore
Hotlinks to Teaching-related Websites
Research Request Form
Teaching Packets
Practitioners