Performance Assessment 3 страница

WRITING ASSESSMENT

Name

Evaluation

PREWRmNG/WRITING

1. The topic is stated clearly.

2. The details fit the topic.

REVISING

3. The details are in an order that makes sense.

4. The final copy is well organized.

EDITING

5. Every sentence has a subject and a predicate (naming and telling parts).

6. Pronouns are used correctly.

7. The spelling, punctuation, and capitalization are correct.

POSTWRITING

8. The final copy is neat, clean, and
easy to read.

□ □

□ □

□ □

□ □

Excellent Satisfactory

□ □

□ □

□ □

□ □

Needs Improvement

□ □

□ □

□ □

□ □

FIGURE 7.2 Textbook-Provided Performance Assessment: Writing [From D. Alvermann, C. Bridge, B Schmidt, L. Searfoss, & P. Winogard, Language Assessments Attitude/Appreciation Assessments, Grade 2, p. 16. Copyright © 1989 D. C. Heath & Co. Reprinted by permission.]

a separate, controlled exercise: (1) the frequency with which the performance of interest naturally occurs in the classroom and (2) the seriousness of the decision to be made (Stiggins, 1987).

If the performance of interest occurs infrequently in normal classroom activity, it may be more efficient to structure an exercise or situation in which pupils must perform the desired behaviors. For example, pupils rarely have the opportunity in the normal flow of classroom activities to give a planned five-minute speech on a

FIGURE 7.3 Textbook-Provided Performance Assessment: Listening and Speaking [From D. Alvermann, C. Bridge, B. Schmidt, L. Searfoss, & P. Winogard, Language Assessments Attitude/ Appreciation Assessments, Grade 2, p. 13. Copyright © 1989 D. C. Heath & Co. Reprinted by permission.]

topic, so the teacher should not wait for the behavior to occur nat­urally in the classroom. Setting up an exercise in which each pupil must develop and give a five-minute speech would be more effi­cient. Oral reading, on the other hand, occurs frequently enough in elementary school classrooms that pupils' performance can be ob­served as part of the normal flow of classroom activities.

Name

Observational Category: Effort	Often	Sometimes	Never
The child:
1. tries to improve reading abilities and habits (Examples: pays attention during reading group; corrects own work voluntarily; follows directions; responds to suggestions; identifies work needing attention; etc.)	□	□	□
2. does not give up without trying (Examples: tries to read independently; maintains attention on a given task; rereads a sentence to figure out a word; persists in trying to solve own difficulties; uses the dictionary or other references to find information; etc.)	□	□	□
3. asks for help when necessary (Examples: asks for clarification of directions; asks questions to clarify task; asks for help in pronunciation; asks about meaning of words; etc.)	□	□	□
4. helps others when appropriate (Examples: works constructively with a partner; participates fully in collaborative/ cooperative group efforts; helps peers with directions; etc.)	□	□	□

FIGURE 7.4 Textbook-Provided Performance Assessment: Effort [From D. Alvermann, C. Bridge, B. Schmidt, L. Searfoss, & P. Winogard, Language Assessments Attitude/Appreciation Assessments, Grade 2, p. 23. Copyright © 1989 D. C. Heath & Co. Reprinted by permission.]

PERFORMANCE ASSESSMENT 279

The importance of the decision that is to result from the perfor­mance assessment should also influence the context in which obser­vation takes place. In general, the more important the decision to be made, the more structured the assessment environment should be. Grade-to-grade promotions, high school graduation, or course grades are all very important decisions. If performance assessment contributes to such decisions, the evidence should be gathered un­der structured, formal circumstances so that every pupil has a fair and equal chance to exhibit his or her performance. The impor­tance of these decisions calls for making sure that the assessment is similar and familiar to all pupils. When the decisions are only for internal classroom use, conditions can be less stringent as long as each pupil has an opportunity to perform.

Another consideration in setting up the observations is the amount of information that is needed to make the desired decision. Will one observation of each pupil's performance suffice? Does the teacher need more than one observation to confidently judge the pupil's performance? Regardless of whether a teacher uses a per­formance or a paper-and-pencil assessment, the evidence obtained from one assessment describes a single example of a pupil's perfor­mance. For a variety of reasons such as illness, home problems, or other distractions, pupil performance at a single time may not pro­vide a true indication of achievement. To be certain that one has an accurate indication of what a pupil can and cannot do, multiple ob­servations are useful. If the different observations produce similar performance, a teacher can have confidence in the evidence and use it in decision making. If different observations contradict one another, more information should be obtained.

Recall that, in the discussion of informal observations in Chapter Two, we emphasized that a single, spontaneously occurring event may not provide an accurate representation of a pupil's true char­acter, ability, attitude, or performance. More than one instance of the observed behavior was necessary before a teacher could justify making judgments about the pupil. The same general principle ap­plies in formal, structured performance assessment. If you want to be confident that you have assessed the pupil's performance accu­rately, more than one observation of the performance is important. The actual number of observations needed varies with the impor­tance of the decision to be made, the amount of time it takes to complete a single observation, and the teacher s sense of whether or not enough samples of a pupil's performance have been gathered.

Scoring or Rating Performance

The final step in performance assessment is to rate or score the pu­pil's performance. As in previous steps, the nature of the decision to

be made influences the type of scoring or rating system used. Scor­ing a performance assessment is similar to scoring an essay ques­tion: it can be either holistic or analytic. In situations when the de­cision to be made is a general one such as group placement, selection, or grading, holistic scoring is most useful. To make such decisions, a teacher basically seeks to rank individuals in terms of a single overall score, so holistic scoring of performance is appropri­ate. If, on the other hand, the purpose of the assessment is to diag­nose pupil difficulties or certify pupil mastery of each individual performance criterion, then analytic scoring in the form of a sepa­rate score or rating on each performance criterion is appropriate. In either case, the performance criteria dictate the scoring or rating approach.

In most classroom performance assessments, the teacher is the observer and scorer. In situations where an important decision is to be made, additional observer-scorers may be added. It is common for performance assessments in diving, swimming, music, debate, and art competitions to have more than one observer-scorer in or­der to make scoring fairer. If more than one observer-scorer is used, they will need to be trained in the meaning and application of the performance criteria in order for them to provide fair and ac­curate judgments.

A number of options exist for recording observations of pupil performance. Three of the most common will be discussed here: anecdotal records, checklists, and rating scales.

Anecdotal Records Anecdotal records are written accounts of events and behaviors the teacher has observed in the classroom. Only those observations that have special significance in describing or under­standing a pupil are included in an anecdotal record (Cartwright and Cartwright, 1984; Gronlund, 1985). Figure 7.5 shows an example of an anecdotal record. Notice that it provides information about the learner, date of observation, name of teacher observing, and a factual description of the event.

Most teachers have difficulty identifying particular events or be­haviors that merit inclusion in an anecdotal record. What is signif­icant and important in the life of a pupil is not always apparent at the time an event or behavior occurs. How is a teacher to select, from the hundreds of observations made each day, the ones that might be important enough to write down? It may take many ob­servations over many days to recognize which events are really sig­nificant. Moreover, anecdotal records are time-consuming to pre­pare and should be written up soon after the event or behavior is observed so that the description will be accurate. This is not always

FIGURE 7.5 Anecdotal Record for Lynn Gregory

possible. For these reasons, anecdotal records are rarely used by teachers. This does not mean that teachers do not observe and judge classroom events; we know that they do that. It simply means that they seldom write down descriptions of these events.

An alternative to anecdotal records is personal record cards that some teachers maintain for each pupil in the class. The records con­tain background information and summaries of observations and perceptions the teacher has had about the pupil during the school year. The information on the personal record cards is kept for the teacher only, not for the school record File, not for other teachers in the building, not for the benefit of substitute teachers. They are the teacher's personal record of pupil characteristics that are pertinent to the classroom teacher's purposes. A few of the teachers who maintained these records never brought them to school. The few teachers we surveyed who used these records destroyed them at the end of the school year when their pupils moved on to another teacher. Figures 7.6 and 7.7 are hypothetical example? of what a personal record form might look like for two different pupils.

The teachers we interviewed started their personal record cards during the first month of school and added to them every month or so. The entries were brief and included general teacher perceptions of pupils (both academic and affective characteristics), changes (good and bad) in performance over time, and strengths and weak­nesses. The cards also contained background information about the pupil, including address, parents' home and work telephone num­bers, date of birth, parents' marital status and custody arrangement, and medication or physical needs. Notice that the entire record of a pupil's year in the class is brief and concentrates only on those events that are significant to the teacher.

For example, observations such as the following might be found in a teacher's personal record card: "Bill has been slow and unable" to finish his work in language"; "Rebecca cannot do seatwork for more than a few minutes before she is distracted from her task"; "Since he came back from vacation, Sal has worked very hard"; "Os­car refuses to interact with any of the boys during recess"; "Diana often comes to class without her homework completed"; "Rudolfo has matured greatly in the last half of the year"; "Rolf has been un­able to concentrate on his work since his parents separated."

These record cards are helpful to teachers in many ways. They provide a quick, written refresher on the progress, problems, and activities of a pupil during the year. They help the teacher prepare for parent-teacher conferences by highlighting classroom behav­iors, attitudes, and performances that the teacher may wish to dis­cuss with the parents. They also record the teacher's perceptions of parent-teacher conferences. They help the teacher provide input in special education assessment meetings, in recommending a suitable teacher for the next academic year, and in recommending place­ment in a particular track or ability level in middle or high school. Finally, they form the basis for the written comments about pupil attitudes and work behaviors that appear on report cards.

It takes a certain discipline to maintain these pupil records dur­ing the school year, but they are worth the effort because they make the teacher observe and think about pupils, afford a basis for meet­ings with parents, and provide insights that can help identify appro­priate pupil placements. More than anything else, however, writing down perceptions in the form shown in Figures 7.6 and 7.7 helps the teacher maintain a private, permanent record that will not be ignored, misrepresented, or forgotten in the press of classroom ac­tivities and responsibilities. It is emphasized that the record cards are for the teacher's personal use only and should be destroyed when the school year is finished.

Anecdotal records and personal record cards are the most qual­itative of all performance assessments because they provide only the written descriptions and perceptions of the classroom teacher. No score or rating is produced from these descriptions; there are only the descriptions themselves, which are interpreted qualitatively, not quantitatively, when used to help make decisions.

Checklists A checklist is a written list of the performance criteria associated with a particular performance or product. As a perfor­mance is observed or a product judged, the scorer considers each criterion and determines whether the performance or the product fulfills that criterion. If it does, a checkmark is placed next to that criterion indicating that it was observed; if it does not, the checkmark is omitted.

The performance criteria shown in Figure 7.2 could serve as the basis for an oral presentation checklist. Essentially, once the perfor­mance criteria are identified and written down, the teacher simply places a check beside each performance criterion the pupil displays in his or her presentation. Figure 7.8 shows a completed checklist for one pupil, Rick Gray, who gave an oral presentation.

The advantages of a checklist are many. When performance cri­teria are written down as in Figure 7.8, they help the teacher focus on the important aspects of the performance. Checklists provide a detailed record of the pupil's performance, one that can and should be shown to the pupil to help him or her see where improvement is needed. Rick Gray's teacher could sit down with him after his pre­sentation and point out the things he did well, as well as those he needs to work on before his next oral presentation. Because it fo­cuses specifically on behavioral aspects of a performance, a checklist provides diagnostic information. The checklist can be used many times. In fact, repeatedly using the same checklist is an easy way of providing information about a pupil's improvement over time.

There are, however, disadvantages associated with the use of checklists. One important disadvantage is that a checklist gives the teacher only two options for each performance criterion: present or absent. A checklist is like a series of true-false questions, except the options are whether a behavior is present or absent in the perfor­mance. Just as in the true-false item, the checklist provides no mid­dle ground. Suppose that Rick Gray stood straight and faced the audience most of the time during his oral presentation—or paced his words evenly except in one brief part of the speech, when he spoke too quickly and ran his words together. How should his teacher treat these performance criteria on the checklist? Should Rick receive a check because he did them most of the time, or should he not receive a check because his performance was slightly flawed? Sometimes this is not an easy choice. A checklist provides no middle ground for judging performance; it forces teachers to make a present or absent decision for each performance criterion, which they do mainly by comparing the observed performance to an implicit standard of satisfactory performance.

A second disadvantage of checklists is the difficulty of summarizing a pupil's performance in a single score or rating. We saw how useful checklists can be for diagnosing pupils' strengths and weak­nesses. But what if a teacher wants to summarize performance across a number of criteria to arrive at a single, holistic score, as when grading pupils? It is possible to summarize performance across criteria on a checklist, but to do this, a teacher must have a rule or process to apply. There are many rules or processes that can be used, but we shall consider two simple ones that might apply to Rick Gray's performance.

The simplest way to summarize Rick's performance into a single score would be to translate the performance criteria he demon­strated into percentage terms. Thus, if there were a total of thirteen different performance criteria that the teacher wanted pupils to demonstrate during their oral presentations and Rick did nine of them during his presentation, this would translate into a score of 69 percent (9/13 =.69). Thus, Rick demonstrated 69 percent of the desired performance criteria. (In Chapter Eight, we will discuss the way scores like Rick's 69 percent are turned into grades.)

A second way to summarize performance would be for the teacher to set up a standard for rating performance. Suppose Rick's teacher set up the following standard:

Excellent 12 or 13 performance criteria shown
Good 9 to 11 performance criteria shown

Fair 5 to 8 performance criteria shown

Poor 5 or fewer performance criteria shown

Establishing a standard allows the teacher to summarize a pupil's performance by rating it on a scale that goes from excellent to poor. The same standard would be used to summarize each pupil's per­formance. Rick performed nine of the thirteen criteria. Comparing Rick's score of nine to the teacher's standard indicates that his per­formance should be classified as "good." Many such standards might be set up by a teacher. The standards can be made quite com­plicated by considering not only the number of performance crite­ria a pupil manifested, but also the areas in which they were mani­fested (e.g., physical expression, vocal expression, and verbal expression). It is advisable, however, to keep summarizing rules or standards as simple as possible.

Figures 7.9 and 7.10 show examples of checklists for observing a pupil's cooperation in groups and returning a tennis serve. While multiple observations of pupil performance increase confidence in the findings, some behaviors are easier to observe than others. Note that determining cooperation in groups requires considerable time to observe the pupil on all performance criteria, while returning a tennis serve can be observed many times in a short period.

Name___________________________

______ Reports to group project area on time

______ Starts work without being told to

______ Participates in group discussion

______ Volunteers for group activities or assignments

______ Contributes ideas and suggestions

______ Does his or her share of the work

______ Listens to others' ideas

______ Does not make fun of others' ideas

______ Gives others a chance to talk

______ Accepts majority rule without rancor

______ Completes assigned tasks on time

FIGURE 7.9 Checklist for Observing Cooperation in Group Activities

Name___________________________

______ Knees bent

______ Back straight

______ Watches ball come off server's racket

______ Follows ball over net

______ Moves to ball beginning backswing

______ Hits ball while it is rising off court

______ Hits ball before it is parallel to body

______ Follows through on stroke

______ Racket head is above the wrist

______ Watches flight of return

______ Assumes position to return next volley

FIGURE 7.10 Checklist for Return of Tennis Serve

Checklists provide an easy way of combining observation and judgment. They provide an easy-to-interpret record of pupil per­formance that can be used for diagnosis and for assessing many dif­ferent kinds of school performances and products. However, they do not provide a middle ground for judging performance on each criterion, and they must be summarized in order to be used for grading or selection.

Rating Scales Rating scales are similar to checklists except that rating scales allow the observer to judge performance along a con­tinuum from adequate to inadequate, rather than as a present-absent dichotomy. Like checklists, rating scales are based on a set of performance criteria, but while a checklist gives the observer two categories for judging, a rating scale gives more than two.

Look back at Figures 7.2 through 7.4, which are examples of rat­ing scales. Each performance criterion is followed by three possible ratings. The continua from "Excellent" to "Needs improvement" and from "often" to "never" are the scales that are used to rate per­formance. Notice that if the scales were omitted from these assess­ment instruments and teachers were asked only to make a decision about whether each performance criterion was present or absent, the instruments would become checklists. Rating scales allow the teacher to make finer discriminations of pupils' performance than checklists. They overcome the dichotomy limitation in checklists.

Three common types of scales are used to rate pupils' perfor­mances and products: numerical, graphic, and descriptive. Each of these types contains more than two rating points, and each de­scribes a continuum of performance from "good" to "poor" or "al­ways" to "never." Figure 7.11 shows an example of each of these scales as applied to two specific performance criteria associated with giving an oral presentation.

Each type of scale requires the teacher to observe performance and compare it to a standard or norm. In numerical scales, a num­ber stands for a point on the rating scale. Graphic scales require the rater to mark a position on a line, guided by points on the scale un­derneath the line. Descriptive rating scales require the rater to mark a position on a line defined by descriptions of actual perfor­mance.

Regardless of the type of rating scale one chooses, a couple of general rules will help make their use easier. First, limit the number of rating categories. There is a natural tendency to think that the greater the number of rating categories to choose from, the better the scale. In practice, this is not the case. Few observers can make consistent discriminations in pupil performance across more than five or six rating categories. Raters cannot consistently differentiate between a rating of 8 and a rating of 9 on a numerical scale or a "sometimes" and a "not very often" on a graphic scale. Stick to four or five well-defined and distinct points on rating scales. "Excellent, above average, average, below average, and unsatisfactory" or "al­ways, frequently, sometimes, seldom, and never" are good exam­ples of useful rating scales.

Second, as much as possible, use the same rating scale to rate each performance criterion. This is not always possible in descrip­tive rating scales, where the descriptions may vary with each perfor­mance criterion. However, for numerical and graphic scales, it is best to select a single format for the rating scale and to use that for­mat for all performance criteria. Requiring the observer to change rating scales frequently while rating a performance and using too many rating categories distract the rater's attention from the per­formance with a corresponding increase in rating inaccuracy.

Figure 7.12 shows Sarah Jackson's completed rating form for an oral presentation. Compare the checklist in Figure 7.8 to the rating scale in Figure 7.12 to see the finer distinctions on the rating scale.

While rating scales provide more categories for assessing a pupil's performance, thereby providing detailed diagnostic information, the multiple rating categories complicate the process of summing across performance criteria to arrive at a single score for a pupil. With a checklist, summing across performance criteria simply in­volves giving credit for criteria checked and no credit for criteria not checked. With a rating scale one cannot do this because perfor­mance is judged in terms of degree, not presence or absence. A teacher must treat ratings of "always," "usually," "seldom," and "never" differently from each other. Otherwise, there is no point to having the different rating categories.

Summarizing across performance criteria in a rating scale is ac­complished in two main ways: numerical summarization and de­scriptive summarization. Numerical summarization produces a score for the pupil's performance, while descriptive summarization provides a general descriptive rating of the pupil's performance (e.g., excellent, good, average, fair, poor). Descriptive ratings pro­vide more information about the pupil's level of performance than a numerical score, but descriptive criteria are harder to state and apply than numerical scoring.

The most straightforward and recommended type of numerical scoring assigns a point value to each category in the rating scale and sums across the performance criteria. For example, consider Sarah Jackson's ratings in Figure 7.12. To obtain a summary score of Sarah's performance, one can assign 4 points to a rating of "always," 3 points to a rating of "usually," two points to a rating of "seldom," and 1 point to a rating of "never." The numbers 4, 3, 2, and 1 were selected because there are four possible ratings for each perfor­mance criterion and because "always" is the most desirable response and "never" the least desirable. Thus, high scores will indicate good performance. Notice that, for convenience and consistency, the numbers to be circled on the rating form correspond to the selected point value of the ratings.

Sarah's total score, 39, can be determined by adding the circled numbers. The highest possible score on the rating scale is 52; if a pupil were rated "always" on each performance criterion, the pu­pil's total score would be 52 (4 points X 13 performance criteria). Thus, Sarah got a score of 39 out of a possible 52 points.

A second way to summarize a pupil's performance on a rating scale involves constructing a summary scale that contains descrip­tions of different levels of performance on the behavior of inter­est. Here is a descriptive scale that a teacher developed for sum­marizing performance on the oral presentation rating scale. The descriptive scale is made up of brief, written descriptions of dif­ferent levels of pupil performance. A scale that just has "excel­lent," "good," "fair," and "poor" without the accompanying be­havioral descriptions would be inadequate. The behavioral descriptions are needed to correspond to the performance the teacher has been observing.

Excellent Pupil consistently faces audience, stands straight, and

maintains eye contact; voice projects well and clearly; pacing and tone variation appropriate; well-organized, points logically and completely presented; brief summary at end.

Good Pupil usually faces audience, stands straight, and

makes eye contact; voice projection good, but pace and clarity vary during talk; well-organized but repetitive; occasional poor choice of words and incomplete sum­mary.

Fair Pupil fidgety; some eye contact and facial expression

Дата добавления: 2015-11-04