Search

Editor's Note: I am delighted that EEPA is publishing such an extremely relevant article in the field of educational evaluation, for as Professor Travers observed, it deals with one of the most basic problems in all of evaluation. However, evaluators have not yet recognized the problem as a problem. Present literature avoids discussion of two core problems in the area of evaluation: (1) What is a sound basis for classifying behavior? and (2) What is the nature of the value judgments through which such behaviors are judged to be worthy? In both of these areas, the evaluation specialist is virtually no more sophisticated than the ordinary citizen. This article demonstrates that a sophisticated classification system must result from thought at a sophisticated scientific level, and not from the kind of committee deliberations from which taxonomies of behavior have been derived.

The preferred model of perception presented by writers in the audiovisual field has been one in which information reception can occur simultaneously through different sense modalities. Experimental studies of the last decade, notably those of Broadbent (1958), have cast considerable doubt on the validity of such a model. The data have generally supported a model in which the perceptual system is represented as a single channel, except for the case in which the amount of information arriving is below the information capacity of the system. The single channel can receive information from a number of distinct information sources that may be auditory, visual, or tied to other sense modalities. The data on which the model is based also lead to the conclusion that, whenever a change is made in the source or original information channel from which the perceptual system receives information, time is lost from the time available for either perception or learning. In other words, switching attention from one source of information to another occupies time that cannot be used for other psychological purposes. The research reported here is an attempt to estimate the time lost on a single switch between the auditory and visual channels and to determine whether such losses can appreciably depress learning in an extended task involving information channel switching. In 1954, Broadbent conducted an experiment on immediate

Kale and Grosslight (1955) report a study in which the task was to write the correct Russian words when presented with their English equivalents. The presentation modes were the visual-print, the visualpictorial, the visual-pictorial with motion (the object representing the word was shown in motion), the audio and the kinesthetic (the S's vocalization of the learning materials), and their combinations. It was found that the presentation mode did make a difference in the amount of Russian vocabulary learned, with the visual-pictorial-with-motion mode being superior to the visual-print (p < .05) and to the complex combination of the visual-print, audio, kinesthetic, and visual-pictorial modes (p < .01). The audio mode was the least effective.

The use of audiovisual teaching materials involves the presentation of information through both the audio and visual channels. The use of colored embellishment in the visual channel has been assumed to have differential effects, depending upon the type of information that was to be transmitted. However, it is common for color and other fonns of embellishments to be used when these do not have a function with respect to information transmission other than that of attracting attention or producing aesthetic effect (realor supposed). This experiment is designed to study the effect on learning of colored embellishments of the visual materials involved in a task to be learned. The task involves both visual and auditory transmission of information, but only the visual component is presented, both with and without

DRAWTINGS HAVE become increasingly popular as an aid in clin ical diagnosis. * It has been claimed that they can be used todetectpath ology and to formulate hypotheses about the presence of disturbances in interpersonal relationships and distortions in the self concept. As an adaptation of this technique to individuals in education, Travers and Rab inowitz devised a drawing test in which their subjects were asked to draw a teacher with a class. The hypothesis was that such drawings would reveal the individual's concept of the role of the teacher. It was further hypothesized that distortions in the depiction of the teacher and the students in the class, the amount of tension and emotional disturb ance in the drawings (determined by ratings) would be related to disturb ed behavior in teacher training.

The relative effectiveness of two reinforcing contingencies referred to as verbal and physical reinforcements were compared as a function of age and sex. 20 males and 20 females at each of four age levels represented prekindergarten and Grades 3, 6, and 9. A highly significant difference was found between the reinforcing conditions, physical reinforcement being more effective than verbal at all age levels. These results had been predicted in the two oldest age groups, however, hypotheses made for the two youngest groups were not supported. Verbal reinforcement in combination with the physical reinforcement resulted in no greater response than did the physical reinforcement alone. Verbal reinforcement was significantly more effective among prekindergarten females than among third grade females, while males at both age levels were similarly affected by verbal reinforcement.

THERE HAVE been a number of investigaj tions in the past few years designed to investigate the effects of varying the speed of presentation of meaningful material upon listener compre hension. There have also been studies which have investigated the differences between the auditory, visual, or audiovisual modes of presentation. There have, however, been few studies which have investigated the interrelationships between the speed and the mode of presentation. Goldstein (2) presented connected meaningful material at several rates ranging from 100 to 322 words per minute (wpm) both auditorally and visually. He found an almost linear loss in comprehension as the speed was increased. He also found that for his subjects who ranged in age from 18 to 65 years the auditory presentation resulted in higher levels of comprehension than the visual presenta tion at the lower speeds but that as the speed was increased, comprehension by the visual mode of presentation gradually became equal to that of the auditory presentation. Goldstein varied the rate of presentation by training an expert speaker to deliver the ma terial at the various rates used in the study. This j material was recorded and minor adjustments in speed were made by varying the playback speed of the recordings. Although Goldstein claimed a high level of intelligibility of the auditory ma terial, there is a possibility that some loss of in telligibility resulted from the speaker's increased rate. Goldstein presented the visual material with motion picture equipment. The rate of presenta tion could be very accurately controlled so that it matched exactly the rate of the auditory pres entation. Goldstein's conclusion that there seemed to be no "optimum rate" at which the ma terial should be presented was based upon a rather straightforward interpretation of the linear rela tionship between the rate of presentation and comprehension. If some sort of efficiency index had been calculated, an optimum level might have been determined, but this cannot be computed from the data given. Fairbanks, Guttman, and Miron (4) did compute an efficiency index based upon the number of comprehension test items answered correctly per unit time. Their study, however, was concerned only with the auditory mode of presentation. The optimum rate for the auditory presentation was reported as about 280 wpm. This rate was twice as fast as the original recording and is considerably faster than every day speech. In a later study by Fairbanks, Gutt man and Miron (5) subjects were presented with the material twice instead of just once. The re sults clearly indicate that the double presenta tion at the rate of 282 wpm results in higher comprehension scores than a single presentation at the slower rate of 141 wpm (p

Ss were given a series of learning trials with stimulus materials of differing degrees of meaningfulness and redundancy using three modes of presentation. All Ss received one list of the stimulus materials through an auditory presentation, one list visually and one list audio-visually. Forty-eight Ss learned lists of nonsense syllables, 48 learned lists of common words, and 48 learned lists of words with constraint. One-fourth of each group received the stimuli at the rate of one each 4 sec., one-fourth at the rate of one each 2 sec., one-fourth at a 1-sec. rate, and one-fourth at one stimulus each .6 sec. There were no significant differences between the visual and audio-visual modes of presentation across all three types of stimulus material, although at the shorter stimulus exposure times a slight trend was observed for the audio-visual mode to be inferior to the visual. When the stimulus material was nonsense syllables, the auditory mode of presentation produced learning inferior to the other two ( p < .001), but the visual and audio-visual modes were similar.

The purpose of this paper is to review tech niques for evaluating the outcomes of the teach ing of English primarily at the high-school and j college freshman level. While many measur| ing techniques have been developed in this area during the last thirty years, they have often | been published in journals that are not widely j available and In at least one notable case (27) | have remained unpublished. The following j pages review these new evaluation techniques and discuss some of their merits and limita tions.

Professor Travers is one of the country's leading educational researchers. He has served in teaching and research capacities at Columbia University, Ohio State Uni versity, the University of Michigan, and the University of Utah. He was chairman of the Department of Educational Psychology at the University of Utah for four years, and in 1962 became director of the Bureau of Educational Research. Dr. Travers is well known among educators for his con tributions to research as well as for his books dealing with testing,Educational Measurement; research methodology, An Introduction to Educational Research; and learning,Essentials of Learning: An Over view for Teachers. Mr. Reid, one of Dr. Travers' collabo rators, is a research associate in the Bureau of Educational Research at the University of Utah. Dr. Van Wagenen now serves in the ca pacity of assistant professor of education at Arizona State University; his principal interests are in the area of school learning.

This study explores the concepts that elementary school children have of their teachers and some of the conditions that influence these concepts. The technique of exploration involves the use of the drawings that children make of their classes and teachers collected under standardized conditions. The data are consistent with the hypothesis that boys are more rejecting of their teachers than are girls, and that boys show no increase in rejection of their teacher as age increases, but that girls show an increasing rejection. Finally, the degree to which girls show a negative attitude towards their teachers depends to some extent on the characteristics of the teacher.

This investigation examined the effect of sense modality switching ofinformation at various positions of a serial learning task. Different numbers of flashes of light and different numbers of “blips” ofa pure tone were used as signalsin a four-position serial learning task. The design of the study involved a switch in sense modality at each one of the four possible positions in the series. Thus, it was possible to determine if the transmission of information through one sense modality makes it progressively more difficult or more time consuming to switch to another source of information transmitted through another modality. The results generally supported the contention that, as information is received through a particular modality, there is a build-up of the disruption involved in switching sense modality.

Pupils learned a German vocabulary task either working in pairs or working alone. Those working in pairs could either assume a teacher role, a pupil role, or alternate these roles. The time taken to undertake the task was measured and achievement was appraised by means of an achievement test. Thus it was possible to estimate the amount of learning per unit of time. The latter is a measure of efficiency of learning. Those who worked alone tended to learn more per unit of time than those who worked in pairs. Subjects showed an increase in efficiency from the fourth to the sixth grade.

Previous research has indicated that, when a serial learning task involves switching sense modality, learning is depressed. The switching involves a period of about 200 msec., which is not a part of the time available for learning. In a serial learning task, with nonsense syllables, sense modality was switched from the auditory to the visual to the auditory or the reverse, but only one switch was involved in the 7-syllable list. The prediction was that a depression would be produced in the learning of the syllable following the switch and that the syllable involved in the switch would be influenced both by the von Restorff effect and by the reduction in time available for learning as a result of switching. At the faster of two speeds some evidence was consistent with predictions.

Each of 34 Ss was presented with a list of 12 paired associates which were arranged according to high-low or low-high stimulus and response meaningfulness and also in a simultaneous or sequential time relationship. Meaningfulness level on the stimulus side of the dyad rather than on the response side was found to be more crucial for learning, and significantly more learning occurred also when the dyads were presented in the simultaneous condition. The findings were discussed in terms of both association theory and the differences between the present procedure and the conventional anticipation method.

Three rote-learning tasks were administered to 180 children. Reinforcement was given either on right responses or on wrong responses and consisted either of the words “right” or “wrong,” a 400-cps tone, or an electric shock. Each task was given either in a long or a short form and was always run to 12 trials. No significant effect was found for reinforcement mode. Significantly more learning occurred when the feedback was contingent on a wrong response than on a right response.

The purpose of this investigation was to measure the minimum interval for recognition under conditions of high and low discrimination redundancy. The minimum interval for recognition is the briefest interval between two stimuli at which the first stimulus can be processed. Two sets of four digits were used. Digits were selected so that one set exhibited high discrimination redundancy and the other low discrimination redundancy. Ten Ss viewed digits with high discrimination redundancy, and 10 Ss viewed digits with low discrimination redundancy. The minimum interval for recognition for Ss tested with high discrimination redundancy was nearly twice as large as the minimum interval for recognition of the digits with low discrimination redundancy.