Showing papers in "American Educational Research Journal in 1972"

Patterns of Mental Abilities: Ethnic, Socioeconomic, and Sex Differences

Abstract: Past research has shown that males and females exhibit characteristic patterns of mental abilities; however, comparatively few studies have considered the patterns of mental abilities that distinguish different ethnic and socioeconomic groups (Anastasi, 1958). One such study, conducted by Lesser, Fifer, and Clark (1965) found characteristic patterns of mental abilities for first grade children from different ethnic backgrounds (Chinese, Jewish, Negro, and Puerto Rican). The ethnic patterns differed in shape and average level. Socioeconomic status (SES) was not related to the shape of the patterns, but those of higher SES tended to have higher mean scores on the mental ability tests. A replication study (Stodolsky and Lesser, 1967) confirmed these results on a sample of Negro and Chinese first graders; however, no common ethnic pattern emerged for samples of lower- and middle-class Irish-Catholics.

A National Experiment in Curriculum Evaluation

Abstract: Although the need for true experiments on broadly defined populations has long been recognized, there are very few local experiments and no national experiments in curriculum research. For example, among 46 government-sponsored course development projects in science and mathematics, a few relied on teacher reports and classroom visits for evaluation, but only four used true experiments in their evaluation strategies (Welch, 1969). The purpose of the present paper is to report the feasibility of a national educational experiment and to present the summative findings regarding Project Physics, a physics course for high school students.

On the Power of Statistical Tests in the "American Educational Research Journal"

Abstract: It is almost universally accepted by educational researchers that the power (probability of rejecting Ho when Ho is false, that is, 1-0-) of a statistical test is important and should be substantial. What is not universally accepted or known is that the power can and should be calculated and reported for every standard statistical test. The power of statistical tests already conducted in educational research is equally unknown. It is the purpose of this paper to report the level of power for recent statistical tests reported in the AERJ and to propose alternative reporting schemes relative to hypothesis testing to include power and effect size as well as the traditional a. Cohen (1962, 1969), Tversky and Kahnman (1971), Overall (1969) and others argue quite strongly that explicit computation of power relative to reasonable hypotheses should be made before any study is completed and subsequently reported. Tversky and Kahnman (1971) suggest three reasons why this computation is important: (1) Such computations can lead the researcher to the conclusion that there is no point in running the study unless the sample size is materially increased; (2) The computation is essential to the interpretation of negative results, that is, failures to reject the null hypothesis; and (3) Computed power gives the researcher an indication of the level of the probability of a valid rejection of the null hypothesis.

Open Education: An Operational Definition and Validation in Great Britain and United States'

Abstract: "Open Education," "the British Infant School," "the Developmental Classroom," "the Leicestershire Plan," "the Integrated Day"these phrases refer to an educational movement that began in Great Britain and is growing rapidly and generating a great deal of interest among administrators, teachers, and parents in the United States (Silberman, 1970). However, there has been very little research and evaluation on Open Education, aside from testimonials by exponents and reporters. Recently Educational Development Center, a United States Office of Education-sponsored educational laboratory, commissioned TDR (Training-Development-Research) Associates, Inc. to review available Open Education literature, analyze the concept into its component parts, verify the analysis with prominent Open Educators, and develop instruments for classroom measurement of its properties (Walberg and Thomas, 1971). The present paper presents a validation of the instruments in United States Open and Traditional classes and British Open classes.

Self Concept and Socio-Economic Status in Elementary School Children.

Abstract: In an earlier study which was part of a four year longitudinal investigation conducted under a large, federally funded Title III project,' it was found that children of low socio-economic status scored significantly higher than children of middle socio-economic status (Clark and Trowbridge, 1971; Trowbridge 1969, 1970a, 1970b) on a self concept instrument known as the Self Esteem Inventory (Coopersmith, 1967). The number of investigations of the relationship between self concept and socio-economic status (SES) has been increasing, but the relationship remains undetermined. Findings appear to be in conflict. A number of persons have postulated that children of low socio-economic status do actually reflect the negative image society holds of them (Ausubel and Ausubel, 1963; Erickson, 1963; Witty, 1967). Some investigations appear to support this thesis (Deutsch, 1960; Long and Henderson, 1968; Wylie, 1963), while others suggest no significant differences in self concept of children of different SES (Coleman, 1966; McDaniel, 1967; Scott, 1969). Studies by Clark and Trowbridge, 1971; Green and Rohwer, 1971; Soares and Soares, 1969, 1970; Trowbridge, 1969, 1970a, 1970b and Zirkel and Moses, 1971, indicate that by some criteria the self concept of low SES children may be even more positive than that of middle class children.

Learning Concepts from Definitions

Abstract: Most of the time in school we expect students to learn concepts from being told about them. However, very few studies have investigated the learning of concepts from definitions and explanations (Johnson & Stratton, 1966). In most of the research on so-called "concept learning" people are expected to induce concepts, such as two green borders, from cards displaying geometric figures. It is obvious that if the subject were told such a concept in advance he would sort out the cards perfectly. Hence, this research is of doubtful relevance to the concept teaching that goes on in schools (though see Carroll, 1964). The first purpose of the experiment described in this paper was simply to explore to what extent people can acquire concepts from exposure to definitions. A person has acquired a concept if he can identify instances of the concept and discriminate them from noninstances. The substantive language of the instruction and the test questions must be different. Otherwise, the person may be able to answer by "rote," that is, on the basis of surface orthographic or phonological information, without having comprehended the concept (see Anderson, 1972). Recent research indicates that procedures which induce semantic encoding strongly facilitate associative learning. Bobrow and Bower (1969) instructed subjects in one group to compose a sensible

Classroom Observation Schemes: Where are the Errors?

Abstract: Much of the confusion about the notion of reliability was cleared with the publication of Technical Recommendations for Psychological Tests and Diagnostic Techniques (American Psychological Association, 1954) and Technical Recommendations for Achievement Tests (American Educational Research Association, 1955) The subsequent publication of Standards for Educational and Psychological Tests and Manuals (American Psychological Association, 1966), while it introduced considerable changes in the conception of validity, redefined reliability in essentially the same terms The term "reliability" was clarified through the recognition of its several meanings and, consequently, that " 'reliability coefficient' is a generic term referring to various types of evidence; each type of evidence suggesting a different meaning" (APA, 1966, p 25) Although the explicit definition of a series of alternative conceptions of reliability clarified the issue it did not resolve a number of important questions of interpretation For example, there is no clear guide for the selection of an appropriate reliability coefficient nor is there any resolution of the problem of interpreting differences in the values of the various coefficients for the same test

Relations Between Language Attitudes and Teacher Expectancy

Abstract: Considerable research evidence has accumulated in recent years which provides information on the relationships between language or dialect features and persons' attitudes toward, and expectations of, speakers of a particular language or dialect. In one well-known study (Lambert, Hodgson, Gardner, and Fillenbaum, 1960) listeners evaluated the speech of French and English speakers on such traits as good looks, intelligence, dependability, and self-confidence, unaware that the speech samples were provided by perfectly coordinate bilingual speakers. It was found that English and French listeners evaluated the English samples more favorably than the French samples provided by the same speakers. Similar research has since been conducted on Jewish-accented English speech (Anisfeld, Bogo, and Lambert, 1962) and upon selected American dialects (Tucker and Lambert, 1969) in which it was found that the standard or unaccented speech samples were rated more favorably by all listener subgroups. More linguistically oriented research in this area has been conducted by Labov (1966) who selected five phonological variables which were known to be socially stratified in New York speech, and then presented samples of speech containing these variables to listeners who were asked to rate the speaker according to his probable occupation. Occupations ranged from

Comparison Behavior in Elementary School Children.

Abstract: In the social psychology literature of recent years considerable research has been concerned with "social comparison processes." In much of this work a core theoretical assumption is that when the individual is uncertain about how to evaluate himself or part of the external world, he turns to relevant others to gain greater certainty (Latane, 1966). Implications of this line of thinking and research would seem to be of immediate relevance to classroom situations. The writer

The Effect of Letter-Name Knowledge on Learning to Read.

Abstract: There is considerable interest at the present time in teaching children to name letters to the alphabet in the belief that letter-name knowledge facilitates learning to read. This interest is manifest at kindergarten and first grade levels where instruction in naming letters is commonly given as part of the reading readiness program. Sesame Street, the television program for preschoolers, also includes instruction in letter naming. Durrell and Murphy (1964, p. 143) claim that: "Most letter-names contain their sounds, and this assists the child in relating the phoneme in the spoken word to its form in print. Children who know letter-names learn words more readily... ." Belief that letter-name knowledge facilitates learning to read has a longer history than most would suspect. The purpose of this paper is to explore the origin of this belief and to test the validity of the assumption. Huey (1908, p. 265) wrote: "The alphabetic method, used almost universally in Greece and Rome, and in European countries generally until well into the nineteenth century, and which was nearly universal in America until about 1870, is now chiefly of historical interest." In the alphabet method of reading instruction, as practiced in Europe and this continent, the child learned to name letters before

A Factorial Study of Tutoring Versus Classroom Instruction

Abstract: Tutoring is one of the oldest variables in educational theory. Long before Emile, possibly before Plato, it was hypothesized to be superior to other instructional methods and class sizes. Despite this venerable history, tutoring as an educational variable has largely escaped empirical scrutiny in this century, due perhaps to its seeming impracticality. Despite practical constraints, however, supplementary or remedial tutoring is being increasingly employed within contemporary school systems with severely academically deficient children. Furthermore, the practicality of individual instruction has been increased in recent years by the advent of increased leisure time coupled with the realization that teacher training and experience do not appear to affect the amount children learn (Popham, 1971; Moody and Bausell, 1971). If teacher training and experience are not requisite for successful teaching, then it would appear that non-professional manpower might be tapped for tutoring services. Several programs, in fact, have been instituted in which non-professional tutoring plus classroom instruction have been compared to classroom instruction alone (e.g. Ellson, Harris, and Barber, 1968; Tannenbaum, 1968). Tutoring plus classroom instruction generally leads to increased (reading) achievement, although not always (Niedermeyer and Ellis, 1970). Unfortunately, since total instructional time is not controlled, this line of research does little to answer the theoretically and practically important question: does tutoring result in greater student achievement than does classroom instruction?

Some Additional Factors in Student Evaluation of Courses

Abstract: A variety of forces are leading to increased use of course evaluation questionnaires by students. Results of such evaluation inevitably influence the way in which teaching will be done. Whether it be to secure a promotion or simply to improve the quality of his teaching, the typical instructor will modify his teaching to get a high rating. To some extent, the results obtained in such evaluations depend on the particular questions asked. It is important, therefore, that a sufficient variety of questions be asked about the course (or instructor) to give one an adequate basis for rating courses and instructors, and ultimately for modifying the quality of teaching. The traditional type of course evaluation questionnaire has focused attention on the instructor's behavior and content of the course. A number of distinguishable factors have been identified with this type of questionnaire (see, e.g. Caffrey, 1969; Isaacson et al., 1964; Meredith, 1969; Spencer and Aleamoni, undated). Predictably, the factors identified in these studies have centered around the nature of the course (e.g. structure, interest, difficulty) and the instructor's behavior (e.g. rapport, preparation). The purpose of the present study was to change the focus of attention in course evaluation from the course-instructor complex to the student's estimate of self-development as a result of the course. Or, in more current terminology, the focus was changed from the process of education to student estimates of outcomes. More specifically, an

Interaction between College Effects and Students' Aptitudes

Abstract: The effects of college environments on student achievement on three GRE area achievement tests, when the initial ability of students was controlled, were studied. A wide variety of environmental measures for 95 colleges were included. The method computed within-school regression lines, then colleges were clustered on regression lines, using a taxonomic technique, and multiple group discriminant functions using college descriptive variables were computed. It was found that in all three achievement areas one group or cluster of colleges could be identified as being more effective than the remaining groups of colleges. In the humanities areas the most effective group was characterized by a higher proportion of faculty with doctorates, greater selectivity, and a larger budget. In social sciences and the natural sciences, emphasis on subject matter area as measured by the proportion of majors in the appropriate fields of study was the only variable which demonstrated some discrimination between the most and least effective groups.

Comparisons of Letter Name and Letter Sound Training as Transfer Variables

Abstract: What factors influence the acquisition of a reading vocabulary? Instructional variables which produce positive transfer to word recognition are not well understood, although they are widely manipulated in various reading programs For example, programs based on a language experience approach assume that knowledge of word meanings facilitates word recognition In contrast, certain basal reader programs appropriate early instructional time to learning letter names, while still other programs with a phonics base emphasize letter sounds as facilitators of later word recognition The use of letter names as a transfer variable has strong correlational support Dykstra (1970) reported that the best single predictor of reading achievement is knowledge of letter names (r = 45-56) On the other hand, Samuels (1970) and Johnson (1970) in experimental studies failed to demonstrate that training in letter names facilitated either word recognition or end of year reading achievement Nor is the evidence for transfer effects of instruction in letter sounds particularly strong This variable has been isolated experimen-

Estimating the Effects of Intervention Into a Non-stationary Time-Series:

Abstract: Observation of a variable Z at several equally spaced points in time yields the observations z1, z2,..., ZN. Suppose that an intervention (T) is made at some point in time before time N into the process presumed to be controlling Z. The time-series is said to be interrupted at a point in time, say nl less than N: z1, ..., zn, T, zn + 1, ..., ZN. When the intervention is made for purposes of studying its effects, whether they be a change in level or direction, on the time-series, the pardigm is that of what has come to be called in educational research and the social

Generalized Rules for Calculating the Magnitude Of an Effect in Factorial and Repeated Measures ANOVA Designs

Abstract: Many present day researchers are using the phrases "approaching significance," "significant," "highly significant," and "dramatically significant" in an attempt to describe the magnitude, and therefore the inferred importance, of effects in comparative experiments. The use of a particular phrase is dependent on the relative difference between the standard alpha probability value, .05 or .01, and the obtained or "looked up" probability value associated with the observed t or F for an effect. For an example, an F ratio with a reported p < .10 is reported as "approaching significance" whereas an F with a reported p < .0001 is reported as "dramatically significant." Such reporting of obtained probability values has received severe criticism (Hays, 1963; Cohen, 1969; Bakan, 1966) indicating that the size of the obtained t or F, together with its corresponding p value, depends directly on the size of the sample and does not necessarily reflect the magnitude or importance of the treatment effect.

Effects of Competition as a Motivational Technique in the Classroom

Abstract: The use of competition with speed-related tasks in an educational setting has repeatedly been shown to increase performance (Chapman and Feder, 1917; Clayton, 1964; Clifford, 1971; Hurlock, 1927; Maller, 1929) On the other hand, Clifford, Cleary and Walster (1971) found that a competitive treatment had no effect on the performance of Ss taking a power-oriented test (ie, a power-test or complex problemsolving test on which a time limit is imposed) On the basis of two related experiments, Shaw (1958) likewise concluded that competition more easily affected performance on mechanical or skill-oriented tasks than on complex, problem-solving tasks These studies imply an inverse relationship between the complexity of the task and performance-increase due to competitive motivation Support for this inverse relationship is also provided by theorists who have examined the more general relationship between task-performance and motivation (Ahlstrom, 1957; Guilford, 1954; Lowell, 1952) At the same time, there is evidence that the relationship between motivation and performance is curvilinear (Atkinson and O'Conner, 1966; Bruner, Matter and Papanek, 1955; Yerkes and Dodson, 1908) The effect of educational competition is likely to be a function of the

Some Effects of Frequent Quizzes on Inferential Thinking

Abstract: The idea that methods of study and the type of information focused upon in preparation for examinations is influenced by the student's expectations of the form and content of the test is not new. Meyer (1935), for example, reported that students list different methods of study used in preparation for multiple choice tests than for essay tests. He also reported that performance on the two types of tests appeared to be a function of the type of test expected (Meyer, 1934). Indeed, pre-tests have been listed as a contaminating variable in some experimental designs because they may induce a set to study relevant material (Campbell and Stanley, 1963). Anastasi, however, points out that any distinction between test behavior and underlying skill is inconsistent with test theory, that learning that takes place because of a test is still learning and may be applied in non-test situations (1949). Recently efforts have been made to harness "testing effects" for use as instructional devices. Rothkopf (1970) suggests that test-like events interspersed in prose may reinforce specific study habits so that these test questions act as a tool in manipulating study behavior. He has demonstrated that placing questions after relevant passages of prose maintains reading time in successive passages and improves recall of information not directly treated by experimental questions (Rothkopf, 1966). In addition, Rothkopf and Bisbicos (1967) found that students consistently quizzed on names and measures recalled a different set of names and measures from the text better than students who had been quizzed on common and technical terms. They also found that attention to the respective

The Heritability of Jensen's Level I and II and Divergent Thinking

Abstract: Arthur Jensen's seminal article (1969) has rekindled interest in the subject of heritability: the proportion of variance of a manifested trait due to genetic variation. Summarizing the literature on the heritability of whatever common factor is measured by the conventional IQ tests, Jensen concludes that 80% of the variation in IQ is concomitant with variation in genetic composition. A good deal of the rekindled interest created by the article centers around the very nature of intelligence. Few scholars today still assert that intelligence is unitary in nature and instead assert the presence of a number of separate factors in intelligence, and several have attempted to isolate those factors to assess separate heritabilities even before the publication of the Jensen article. Jensen focused his review of heritability on the underlying common factor in intelligence tests, Spearman's 'g,' yet there are other factors of intelligence with heritability that differ from heritability of 'g.' Jensen himself hypothesizes two levels of learning ability to explain differences in performance on tests of intelligence, learning and scholastic achievement. The first, Level I, is associative learning, in which there is little transformation of the input, manifested by digit memory. The second level, Level II, involves self-initiated elaboration and transformation of the input, before it becomes an output response. Level II is best measured by "intelligence tests with a low cultural loading and a high loading on 'g' such as Raven's Progressive Matrices" (Jensen, 1969, p. 111). He suggests that short term memory, his Level I, is a necessary but not a sufficient condition for high intelligence,

Essay Grades: An Interaction Between Graders' Handwriting Clarity and the Neatness of Examination Papers

Abstract: Various empirical studies have demonstrated the limitations of essay examinations as compared with objective tests. More than 50 years ago, studies by Starch and Elliott (1912, 1913a, 1913b) revealed that the grades assigned to essay papers are highly unreliable. More recently, research has shown that good grades are assigned to essay papers for reasons other than the quality of the ideas set forth by the writer. In a study by Scannell and Marshall (1966), graders were found to be influenced by the grammatical accuracy of essays--even though they were instructed to evaluate the essays solely on the basis of content. Further evidence (Marshall and Powers, 1969) has indicated that students who have neat, clear handwriting are likely to receive higher grades than students of equal ability who have messy handwriting. The present study was undertaken to determine whether all graders are influenced by the factor of "essay neatness." More specifically, this research endeavor was conducted to ascertain whether an interaction exists between the writing clarity of graders and the neatness of the essay papers which they grade. It was hypothesized that (a) graders who have neat, clear handwriting will unintentionally lower their evaluation

On Teaching Disadvantaged Preschoolers to Read: A Successful Experiment

Abstract: A number of experiments have indicated that the rate at which children learn to read may be accelerated via token reinforcement with material backups (Hamblin, Buckholdt, Ferritor, Kozloff, and Blackwell, 1971; Pfeiffer, 1969; Staats and Staats, 1963; Wolf, Giles, and Hall, 1968). In other experiments the rate of learning to read has been accelerated by using more able pupils who are succeeding at reading to help those who are achieving poorly (Hamblin et al., 1971; Myers, Travers, and Sanford, 1965; Riesman, 1965). The purpose of this experiment is to explore these possibilities further by assessing the independent and combined effects of token reinforcement and peer tutoring in accelerating learning to read with what has been traditionally a particularly slow group of learners, i.e., disadvantaged black and white preschoolers of the inner city.

Prophecy Effects and Tutorial Instruction for the Disadvantaged Child.

Abstract: In what has become one of the most widely-publicized educational experiments of the decade, Rosenthal and Jacobson (1968) found support for the premise that one person's prophecy of another's intellectual performance can come to determine that other's intellectual performance. Their results indicated that a randomly chosen group of children, of whom teachers were led to expect marked intellectual growth, achieved significantly greater IQ gains than did control subjects. But reviews by Thorndike (1968) and Snow (1969) have raised serious questions about the legitimacy of the inferences drawn from that study. These critiques focus on two main points. First, analyses in the now famous experiment were performed on IQs derived from raw scores obtained with the TOGA (Tests of General Ability) which was chosen partly because it would probably be unfamiliar to teachers. Thorndike (1968) and Snow (1969) point up the inadequacy of the TOGA IQ norms for the youngest children tested-those for whom the reported prophecy effect was shown most clearly. Second, the reviewers call attention to the fact that the difference in mean gain between experimental and control groups was essentially zero for all grades except the first two. A methodological issue which has not yet been elaborated, however, is that all testing in the Rosenthal and Jacobson experiment This research was supported by a grant to the first author from the Center for Research and Advanced Studies at San Jose State College. The authors thank Mr. James Henman for his assistance in collecting the data for the study. 2 Requests for reprints should be sent to Robert J. Pellegrini, Department of Psychology, San Jose State College, San Jose, California 95114.

Effects of Pattern Drill on the Phonology, Syntax, and Reading Achievement of Rural Appalachian Children

Abstract: Recently, increased attention has been directed toward the influence of nonstandard dialects on school performance (Baratz, 1968; Garvey and McFarlane, 1970; Loban, 1966, 1968; Stewart, 1964). Some educators and linguists contend that possession of a nonstandard dialect has direct effects on school learning. Stewart (1964) and Labov (1970), for example, maintain that a nonstandard dialect constitutes a source of interference for a child entering the public schools where standard English constitutes the vehicle of at least adult-child communication. Not only may communication problems be expected, but as has been suggested (Baratz, 1968; Loban, 1968; Shuy, 1968; Stewart, 1967), learning to read may be hindered by specific dialect features which deviate clearly and systematically from standard English. Even many of the instruments employed by educators to assess student progress require some degree of facility in the use of standard English (Garvey & McFarlane, 1970). Research focusing on dialect is confronted by sensitive political, social, and historical factors from which spring several complex issues. Discrimination is one such issue. Speakers of nonstandard dialects often encounter discrimination in their contacts with society at large. Educators in confronting this issue sometimes fail to recognize it for

Assessing the Importance of Independent Variables in Nonlinear Causal Laws

Abstract: A multivariate causal law is one in which the expected value of a dependent variable Y is expressed as a function of several independent variables X,, X2,..., Xn. In the behavioral sciences, such laws are usually constructed by least-squares techniques, particularly by multiple-regression techniques. When variables are measured on scales which are at least interval, the best-known work in the behavioral sciences on constructing such laws in the linear case is by Blalock (1964); a brief description and further references are given by Darlington (1968, pp. 166-168). The nonlinear case is more complex; two important but difficult papers are by Box and Hunter (1957) and Draper and Hunter (1967). Once a causal law is constructed by these techniques, the problem often arises of stating the "importance" of each X variable in the law. This paper attempts to present a general solution to this problem, which applies to both linear and nonlinear causal laws. When applied to the linear case, the solution is equivalent to that described by Blalock and others, which is associated with the term "path analysis." Thus the principal contribution of the new solution is in its application to the nonlinear case.

Behavior Modification and Operant Psychology: An Anthropological Critique

Abstract: Operant behaviorism is, in today's United States, a major branch of academic and of applied psychology Yet I shall argue that its cross-cultural application (by Anglos, whether to minority enclaves or overseas) is fallacious Incidental to or excluded from this discussion are its application to (1) subhuman animals, where I gladly concede its efficacy; (2) humans handicapped in or related to the brain/the emotions: and (3) Anglos (Yankees), because such programs tend to be designed by the same (Anglo) ethnicity The human unsuitability of behaviorism has been noted at least since the pre-World War I criticisms of early behaviorist John Watson Such objections have, however, been leveled from philosophical or humanistic bases (e g, Chomsky 1959) The behaviorist paradigm has received scant attention from the discipline most concerned with that essence of humanity, culture-namely, cultural anthropology Hence, as a member of that discipline who has been encountering increasingly more of behavioristic craft, I believe the time is over-ripe to indicate some problems both in its theory and in its application

Grading Standards: The Relation of Changes in Average Student Ability to the Average Grades Awarded:

Abstract: The question of academic standards is a persistent theme in rhetoric about higher education. On the one hand we hear that yesterday's college student could not compete academically with his peers of today and, on the other, that "standards" are slipping badly. Constantly rising admissions standards, according to one view, are resulting in college freshman classes that are brighter than previous freshman classes and will cause many able students to flunk out. According to another view, the floodgates have been opened to masses of unqualified students who will make an "A" meaningless. In particular, this is the position taken by many critics of the "open admissions" concept. They feel that lower admissions standards will produce lower levels of achievement. In this discussion it seems sensible to look at the college grades being awarded to these students. One would expect higher grades to be awarded to brighter students.2 Thus, we should examine the stability or

Statistical Descriptors of School Classes

Abstract: Schools are designed for the grouping of pupils into classes sized to fit into classrooms, and are staffed and equipped for the teaching of pupils in classes. That ancient invention of civilization, the school class, has enough of a future to warrant consideration of statistical tools adequate for the tasks of describing, predicting, and assessing the results of teaching pupils in classes. Despite the current emphasis on the individual pupil as the learner for whom learning experiences are provided, it continues to be a class which is entrusted to a teacher, and her opportunities and accomplishments are defined in terms of the spectrum of individualities she finds in her class and the spectrum of results her teaching produces across all those individuals. Artificial it may be, but the class is a legitimate concern to the teacher and her supervisors. The correct standard for evaluating teaching is a direct application of the Utilitarian principle, "the greatest good for the greatest number." In our populous, industrial democracy this ethic of quantitative humanism is the best bedrock for a philosophy of education. (The parable of the shepherd who left the 99 sheep to search for the one stray only makes sense if he left the flock in good hands or secure circumstances.) Although the teacher should treat each child as the individual he is, she should distribute her resources over all the children effectively. She has to plan for the class in the knowledge that her program will be evaluated by the changes it produces in the class. Enthusiasts for individualization of instruction are sometimes too ready to reject statistical description of the class as inconsistent with their approach. They may not recognize that the statistical way of thinking about a class of pupils strongly encourages attention to each of the pupils as a source of measurements, but with a democratic sweep of attention over all pupils simultaneously. The irony is that the enthusiast who rejects the notion of "class" is more likely to talk in terms of a

Ecological Correlates of Ambience in the Learning Environment

Abstract: Two series of studies of social environments in high school physics demonstrated the validity of student perceptions of their classes in predicting cognitive, affective, and behavioral learning (see reviews by Rosenshine, 1970, and Walberg and Ahlgren, 1970). Recently Anderson, Walberg, and Welch (1969), using a true experimental design, showed that the environments of classes randomly assigned to different physics courses differed significantly. Also, Walberg and Ahlgren (1970), in correlational studies, found that the environment in high school can be predicted from cognitive and noncognitive pretests, student biographical characteristics, teacher experience, class size, and other variables. In a quasi-experiment, Walberg (1970) found that, whereas elementary school children bussed to suburban schools around Boston did not gain more than their siblings (who remained in city schools) on arithmetic and reading achievement, there were significant differences between the two groups on perceptions of their classes; for example, bussed children in grades 3 through 6 rated their classes more satisfying than their siblings did. The present study is an attempt to