Interpretations of informally observed behaviors—especially those interpretations related to attitudes, values, or personality traits—tend to be subjective. Behaviors occur fast and fleetingly in the classroom. There is not a long time to reflect on the meaning and interpretation of such behaviors before the next behavior assaults one's consciousness. Many behaviors happen at once. There is a need for closure regarding the pupils and their characteristics. Under these circumstances subjectivity is likely, just as it is likely that interpretations made in the heat of the moment may be erroneous.
4. Informal assessment is descriptive. It has been noted that part of the assessment process is the synthesis of information to provide a description of pupils. In informal assessment, the information collected and the descriptions arrived at tend to be narrative, not numerical. Teachers tend to describe pupils not in terms of numbers, but rather in terms of words—words like "motivated," "creative," "high ability," "self-confident," "lazy"—which serve to summarize the informal observations they have made and to label pupils. Often, these labels ignore and hide the specific observations that they are derived from. Teachers observe so many things about pupils that they must simplify the information so that it is useful to them. In order to simplify, tbey construct a model of the pupil by synthesizing information into a few general categories or clusters (Gordon, 1987; Simon, 1957). Names such as "motivated," "bright," "creative," "troublemaker," or "isolate" are assigned to these behavior clusters in order to provide a description of the pupil.
5. Informal assessment is covert. The results of informal assessments are rarely written down. Teachers carry their perceptions of pupils in their memory, preferring not to commit them to paper because they often include affective judgments about pupils which the teacher may regard as private. Also, since the perceptions are based largely on informal observation, there is little hard evidence to support the perception if the teacher is challenged on it. The absence of a written record of sizing up perceptions protects both the teacher and the pupil. However, when prompted, teachers can provide rich and detailed descriptions of their pupils. It is almost as if the process of informal observation and assessment is carried on without conscious effort by most teachers. It is a natural activity that characterizes all social settings; it is part of the natural flow of pupil-teacher interaction. Fleeting perceptions based on informal observations somehow come together in teachers' minds to produce assessments of their pupils.
In summary, informal assessment, like all assessment, involves four steps: observation, interpretation, synthesis, and description. The process of informal assessment is cyclical: new observations are considered in terms of prior interpretations, and the interpretations are either strengthened or revised. When sufficient evidence supports an interpretation, the pupil's behavior is given a name and the pupil is described and perceived in terms of that name. Informal assessment carried out in the classroom has five characteristics that influence its nature and create potential problems in securing appropriate assessment information. Informal assessment is naturalistic, occurring as part of the natural flow of classroom activities. It is idiosyncratic, influenced by each teacher's perspective and based on observations of different pupils under different conditions. It is subjective, being based on interpretations made by the classroom teacher alone. It is descriptive, summarizing perceptions into a portrait of a pupil that is qualitative rather than quantitative. It is covert, residing in the teacher's head, not in a formal, written record.
QUALITY OF INFORMAL ASSESSMENTS
The rich perceptions teachers form about their pupils from sizing up assessment, and the naturalistic, almost unnoticed manner in which most assessment evidence is gathered, should not distract us from examining the quality of sizing up assessments in a critical light (Evertson and Green, 1986). Teachers employ informal assessments to gather information that will help them represent the reality of pupil characteristics so they can form a classroom society and carry out instruction. It is appropriate to ask how well informal assessments permit them to do these things. It is also appropriate to
identify the potential pitfalls that can reduce the quality of data gathered through informal assessment.
The two main criteria of good assessment information are validity and reliability. Validity is concerned with the collection of appropriate evidence—i.e., evidence that will permit teachers to make the generalizations about pupil behavior they wish to make. "Does the evidence gathered represent or include information about the characteristic I wish to judge?" is a question concerned with the validity of assessment evidence. Reliability pertains to the collection of sufficient quantities of evidence to be certain that the behavior of the pupil that has been observed represents the pupil's typical performance. "Does the evidence gathered represent typical pupil performance?" is a question concerned with the reliability of assessment evidence. Validity and reliability work hand in hand to ensure that the perceptions formed from sizing up assessment are appropriate and trustworthy so that good decisions can be made about pupils.
Table 2.2 shows a number of factors that conspire to threaten the validity and reliability of informal sizing up assessments. For the most part, these factors are a result of the informal assessment process itself and the subjective, idiosyncratic, naturalistic context in which these assessments are made. Consider the helter-skelter na-
TABLE 2.2 THREATS TO THE VALIDITY AND RELIABILITY OF SIZING
_____________ UP ASSESSMENTS ___________________________________________
Validity Threats
I. Observer bias: prevents the teacher from making an objective
assessment of the pupil.
A. Prior information from school grapevine, siblings, or non-
classroom experiences
B. First impressions which influence subsequent impressions
C. Personal theories about variables that define pupils'
characteristics (e.g., girls can't do math; athletes have no interest
in serious academic pursuits, etc.)
II. Logical errors: teachers judge pupils based on the wrong
characteristics (e.g., observe attention and judge learning; observe
clothes and judge ability)
Reliability Threats
I. Inadequate behavior sampling: too few observations prevent teacher from learning about pupils' typical behavior and characteristics.
A. Basing decisions about a pupil on a single piece of information
B. Observing behaviors in one setting (e.g., the playground) and
assuming behavior will be the same in another setting (e.g., the
classroom)
ture of most classrooms and the relatively short time available to form initial perceptions of pupils, and one can see the potential for assessments that are based on information that is not valid or reliable. In this section we shall consider a few of the main problems that diminish the validity and reliability, and therefore the usefulness, of informal sizing up assessments for decision making.
Problems That Affect Validity
There are two main problems that occur during informal assessment that diminish the validity of the information gathered. These problems involve, respectively, observer bias and logical error. Observer bias occurs when prior information or belief leads to labeling a pupil prematurely and incorrectly. Logical error occurs when the wrong indicators are used to assess a pupil's status and characteristics.
Observer Bias The term "observer bias" pertains to situations in which a teacher's preknowledge, first impressions, or personal biases interfere with the teacher's ability to make a fair and objective assessment of a pupil. All of us have some biases; we prefer some things over other things and some kinds of people over others. We have beliefs, interests, ideas, and expectations that differentiate us from other people. These facts are givens, and normally do not create problems for us. However, during informal classroom assessment a real and potentially harmful problem is that the teacher's likes, dislikes, beliefs, and biases will interfere with the ability to make fair pupil assessments, thereby leading to invalid assessments.
The following statements or ones very like them are heard often in schools. Consider them. "He comes from the Oldtown section, and kids from there inevitably are poor learners and discipline problems." "Everyone knows that members of that group have no interest in school." "Look at how she's dressed! How can parents who care about their kid let her out of the house looking like that." "Girls just are not as good in math and the sciences as boys are." "Parents who don't show up at the school have no interest and expect little in the way of school performance from their children." "The school grapevine says she is a brilliant student." "I've taught all the children in that family and I expect the latest to be about as good as his older brothers and sisters." "He's just another dumb jock." "I had him in study hall last year and he couldn't sit still." "I heard from Ms. Pheasant, her teacher last year, that she is lazy, defiant, and demanding." Statements like these represent potential sources of bias in efforts to size up a pupil or group of pupils accurately at the start of the school year. They can lead to prejudgments being made before a pupil has had an opportunity to show his or her true behavior and character.
There are three main sources of biasing information. The first is prior information a teacher has about a pupil before the teacher has met the pupil. Information passed through the school grapevine or the performance of prior siblings can influence and bias a teacher. Second, first impressions tend to influence subsequent impressions. If the teacher bases judgments of a pupil on how the pupil is dressed on the first day of school or behaved in study hall last year, the teacher may unconsciously let this impression dictate interpretations of subsequent observations. Third, teachers' personal theories (Gordon, 1987) can produce biased perceptions of a pupil. A personal theory is a generalization derived from experience.
Let's go back to some of the statements above: "This pupil is from Oldtown, and kids from Oldtown are poor learners and discipline problems"; "Girls are no good in math"; "Everyone knows that members of that group have no interest in school"; "He's just another dumb jock." When a teacher thinks such thoughts, he or she is using an implicit theory of what people are like and what factors underlie certain behaviors. The extent to which the implicit theory is true or not true for a given pupil strongly influences the validity of assessment and teachers should be particularly sensitive to gender, race, cultural, and social class biases. Prior knowledge, first impressions, and implicit theories may all exert a biasing influence on a teacher's perceptions of a pupil and thereby diminish the validity of the teacher's assessment of that pupil.
Does this mean that all prior knowledge about a pupil, all first impressions, and all personal theories are wrong? Of course not. Information from the school grapevine and other teachers is often correct. While first impressions may bias subsequent impressions, they are not always wrong. And there is usually some truth in personal theories. Many kids from Oldtown are behavior problems and do poorly academically. Many families do have a consistently poor record in school, and the failure of many parents to participate in school activities is a sign of low interest and lack of concern with their children's school performance. It is because there is some truth to these observations that they become accepted. But what is true of some or even most information that comes from prior knowledge, first impressions, or personal theories is not true of all of it, and this is a key point. Individuals have to be judged individually.
Pupils change and mature over time. The way a pupil behaves in one setting is not necessarily the way the pupil will behave in a different setting. Each pupil is an individual regardless of the section of town he comes from or the group to which she belongs, and each one deserves to be judged as an individual. Each pupil needs to be judged on his or her own merits and not be tarnished by biases before being afforded a chance to show what he or she is really like. This is a teacher's ethical responsibility.
First impressions and perceptions are important to most teachers in carrying out pupil assessment. Since people perceive what they expect to perceive, information a teacher has about a pupil before the pupil enters the classroom for the first time can bias the initial observations and perceptions of that pupil. It is very important, therefore, that initial impressions be as valid as possible; teachers should not let prior knowledge, first impressions, or personal theories bias pupil assessments. Invalid information will lead to invalid judgments and poor decisions.
Many teachers recognize the danger of personal bias invalidating the initial, informal assessments of pupils that they make.
I don't like to hear anything about a student's behavior from his/her past instructor. Every teacher is different, just like every student is different. A student may have a negative experience with one teacher, but a positive experience with another teacher. I prefer to make my own decision about every child.
It is so easy to be influenced by what other teachers tell you. Many times I have known a child without ever having met him—and he has had to pay the consequences of my ignorance.
I remember the time I stereotyped three of my female students as "valley girls"—not too bright and mainly superficial—on the first day of class. This assessment came about due to their physical appearances and their shallow contributions in discussion. Yet when it came time for formal assessment these three individuals ranked the highest in the class.
From the class list, one does notice "familiar" names. These could be the troublemakers, intellectuals, or the siblings of previous students. Information about the names comes from various sources. It could be personal relations with a family, teachers' room conversations, town gossip, or just plain exposure to students in other classes. It's easy to be aware of these things, but I make it a rule to never let this become my first impression. A list is just that—a list. Talk and/or gossip can be helpful at times, but for the most part I find it can become a hindrance and even discriminatory. Here the child is at a disadvantage because he or she has not had a chance to show his or her character. Instead an outsider with an unfair advantage can blacken the child's reputation before any interaction begins. In my mind it's more important to show the students who you are, as a teacher, your demands. Also try to learn things about them, then sit back, observe his or her performance and behavior. That's when my assessment begins.
Each of these teachers recognizes that their perceptions of pupils can be influenced by external factors that bias their assessments.
Logical Errors To let prior information bias initial impressions of pupils is one error. To select the wrong indicators to judge a pupil's characteristics is a different error. Observers often make a logical error by assuming that the behavior they observe provides information about the pupil characteristic they wish to describe. For example, suppose we hear someone say about a teacher, "He is warm and friendly in his interactions with pupils. They must learn a lot from him." We very likely are hearing a logical error. Warmth and friendliness with pupils may contribute to learning, but by themselves they are certainly no guarantee that pupils will learn. The person who makes this statement is assuming that two observed characteristics, warmth and friendliness, provide information about a third, unobserved characteristic, pupil learning. If the assumption is incorrect, the observer has made a logical error by linking warmth and friendliness to pupil learning. Logical errors lead to incorrect interpretations and invalid assessment.
Earlier in this chapter a number of comments from teachers were presented which indicated the behaviors or characteristics they looked for in the first few days of school to help them size up their pupils. These teachers reported that they judged pupils' academic and personality characteristics by observing the pupils' body language, dress, posture, attendance, volunteering of answers, and so on. From these observations teachers made decisions about the pupils' abilities, personalities, self-concept, motivation, and home background. The important validity question to ask is whether body language and dress are likely to provide useful information about ability, personality, self-concept, and the rest.
Suppose that during the first few days of school a pupil volunteers answers to many questions during group lessons. Does this observation tell you some things about that pupil? Certainly it does; it tells you that the pupil is willing to volunteer and speak out in class. Does that necessarily mean that the pupil has a great interest in the subject matter, or is motivated, or has a strong self-concept? Not necessarily; one needs more information about the pupil to decide about these things. It is tempting to read a great deal into a single observation, especially at the start of the year when the teacher is in a hurry to determine the characteristics of each pupil. There are many things the teacher would like to know about the pupils. How convenient it would be to observe the pupil raising a hand and read into that observation a whole series of interpretations about motivation, attention span, interest in the subject, self-concept, and leadership. Maybe all the inferences would prove to be correct, but it is wrong and dangerous not to recognize the difference between information that is directly observed and interpretations of that information. When observation of one characteristic is used to infer about other, unobserved characteristics, the potential for logical errors—and therefore invalid assessment—is great.
A fifth grade teacher described Katie's first day in school this way: "I knew right away that Katie was cooperative and a hard worker. She was the first child to complete her 'Summer Vacation' essay, and instead of wasting her free time, she offered to get the dictionary and help the other children with their spelling."
Katie's teacher observed her finishing her essay before her classmates, getting the dictionary, and going around the room seeking to help her classmates with their spelling. This is what she observed. The teacher interpreted this observation to mean that Katie was cooperative and a hard worker. Is that what Katie's behavior showed? Maybe. Or could it be that Katie is a busybody who wants to know what everyone is doing? Or maybe Katie is a showoff who likes to be the center of attention? Or maybe Katie rushed haphazardly through her work because she is bored easily. Consider two teachers who taught a pupil who did not participate much in either class. For one teacher the pupil was "aloof," but for the other teacher, who observed identical behaviors, the pupil was "not alert." Clearly the teacher who interpreted the behaviors to indicate "aloofness" will respond to the pupil differently than the teacher who interpreted the same behaviors to indicate lack of "alertness."
The names teachers give to describe what they observe represent interpretations on the teachers' part. Teachers do not actually see direct evidence of the characteristics that they label as "motivation," "intelligence," "leadership," "self-confidence," "aggressiveness," "anxiety," "shyness," "intolerance," "trustworthiness," and the like. Teachers observe a pupil behaving in some way, and they interpret what the behavior signifies and give the behavior a name. In most cases, it is the name given to the behavior that attaches to the pupil, not the specific behavior that prompted the name. Teachers remember that a pupil is a bully, self-confident, aggressive, aloof, motivated, or shy, not the specific observations that prompted them to label the pupil in that way.
The naturalistic, idiosyncratic, subjective nature of informal observation and the fast-paced, hectic classroom setting in which it takes place conspire to produce logical errors in assessments. Occasions on which a teacher observes a behavior but misinterprets what the behavior is indicating about a pupil are common in the classroom, especially when the teacher is rushing to size up pupils to organize the classroom society. When logical errors are made and pupils are mislabeled because the wrong behavior was observed or an incorrect interpretation of the behavior was made, the validity of the assessment is poor and the subsequent decisions and actions that are based on that assessment are compromised. As noted earlier, the lack of valid assessment in these circumstances can affect the pupil greatly during the entire school year. It is, therefore, very important that informal observation be carried out in a way that permits valid interpretations of pupil characteristics.
Problems That Affect Reliability
While validity pertains to collecting the appropriate kinds of information about the pupil characteristics assessed, reliability has to do with the collection of sufficient information to be sure that the behaviors observed represent typical student behavior. Whether formal or informal, teachers' assessments are based on samples of their pupils' behavior. These samples are used to make generalizations about the pupils' typical behavior. This process is analogous to political polling, in which a sample of people is asked how they will vote in the next election and the responses of the sample are used to infer the likely voting pattern of the entire electorate. Similarly, teachers observe samples of a pupil's behavior and use those samples to form perceptions of the pupil's likely behavior pattern. The issue in both political polling and teacher assessment is how well the observed sample represents the overall pattern. This is the concern in the reliability of assessment information. Reliable information represents consistent and stable characteristics of pupil behavior.
The nature of informal assessments in classrooms creates problems that influence the reliability of the information that is collected. When pupils are sized up early in the school year, reliability problems are particularly acute. The spontaneity of naturalistic observation limits what teachers see and what pupils are willing to show. The time available to observe pupils is often brief, since attention has to be turned to other pupils and classroom activities. The lack of time for prolonged observation reduces the likelihood of being able to perceive stable and typical behavior patterns. The fewer the opportunities to observe a pupil, the more difficult it is to obtain assessment information that is typical of the pupil's behavior. In the terms used above, the sample of behavior that is observed under these circumstances may not be a reliable indicator of a pupil's typical behavior.
Many of the teachers we interviewed recognized this problem, particularly as it applies to informal observations at the start of the school year.
First impressions are so important. They can either make or break a child. It all depends on how much opportunity a particular teacher gives to a student to prove himself before passing a judgment. Unfortunately, so many children get stuck with the short end of the stick early in the year— and pay for it.
The first three days are very difficult. The students will not even present their "normal" classroom behaviors to you in the first three days. They are somewhat intimidated and uncomfortable; they don't know you. Even kids who are badly behaved in the first three days, they're just feeling you out, they're testing, trying to see how far they can get.
It's easy to misinterpret some deviant behavior as a big problem when in fact it might be the uncomfortableness of a new situation. A child's behavior in the first few days of class is not always the typical behavior of that child.
Carol breaks up with her boyfriend a week before the beginning of school, leaving her depressed and unmotivated. Does her English teacher know the reason for Carol's behavior? Is her assessment of Carol after one day of school correct? Josh had been a below average student his first two years of high school. He had never fulfilled his potential before, but now is looking to go to college. He is finally motivated to work hard. Can he overcome the teachers' lounge grapevine and prove himself to his new teachers? The teacher's view of the pupil plays a large role in the pupil's view of him or herself and, consequently, in the pupil's academic achievement. If Carol's teacher views her as a disinterested student based on the first week of school, then, even though Carol normally is a very energetic and motivated student, there is a strong possibility that Carol will respond to being treated as disinterested by being disinterested. Josh, by the same token, has been labeled as a poor student and will most likely have to overcome his reputation by extraordinarily hard work.
Simply because John hit Stephan in the nose this morning out in the schoolyard doesn't necessarily mean that he will be the class bully. It may be that he has had a very bad day and Stephan said something to throw him over the edge. You must give each child half a chance to show you what he or she is made of.
The implication of these comments is that teachers must be sure they observe sufficient samples of pupils' behavior before they solidify their perceptions of pupils and use these perceptions for decision making. There are times, such as the start of the school year, when pupils' behavior may not be indicative of their typical behavior. They may be nervous, uncomfortable with a new teacher, not yet settled after summer vacation, intimidated by the teacher or new classmates—or any of a range of other factors that may distort their behavior until they feel comfortable in the new surroundings.
One's first impression of a pupil represents a quick social evaluation that permits one to interact with the pupil. It takes the pupil as he or she is perceived to be at that moment and uses that perception to begin social contact. The behavior that produces a perception may or may not represent the pupil's typical behavior, especially when such perceptions are based on one fleeting observation during the first few days of school. What is typical behavior cannot be determined by observing a pupil just once, especially at a time when the pupil may feel uncomfortable in new surroundings. One would be on shaky ground if one let the opinion of a single voter represent the opinions of all voters. Similarly, one is on shaky ground when one tries to use a single observation to represent the pupil's typical behavior. If the assessment evidence gathered through informal observation of pupil behavior is not typical of the pupil's normal behavior, errors in perceptions can lead to errors in the decisions made about the pupil.
IMPROVING INFORMAL ASSESSMENTS
Informal observation is an important assessment technique used by all teachers at all levels of education. We learn a great deal about our pupils by observing them informally in the naturalistic setting of the school and classroom. We use the information to form perceptions which help us make many decisions about pupils. However, because of its nature and because of the way it is used in classrooms, the process of informal assessment creates many potential problems that can diminish the validity and reliability of the information gathered. The preceding sections discussed the problems of logical error and observer bias, which diminish the validity of observational data, and the problem of limited sampling of behavior, which diminishes the reliability of observational data. In this section, strategies to improve the quality of informal assessment evidence will be discussed. The application of the strategies outlined in Table 2.3 should improve the quality of informal observational data and reduce the number of decision errors that occur.
Дата добавления: 2015-11-04; просмотров: 35 | Нарушение авторских прав
| <== предыдущая лекция | | | следующая лекция ==> |