Tài liệu The Use of Hyper-Reference and Conventional Dictionaries (Ronald Aust, Mary Jane Kelley, Warren Roby)

The Use of Hyper-Reference and Conventional Dictionaries [ ] Ronald Aust Mary Jane Kelley Warren Roby A hyper-reference is an online electronic aid that provides immediate access to adjunct information with a direct-return path to the target information. Eighty undergraduateforeign language learners participated in a comparison of hyper-reference and conventional paper dictionary use on the measures of consultation frequency, study time, efficiency, and comprehension. Hyper-reference users consulted over two times as many definitions as conventional dictionary users. Analyses of efficiency (consults per minute)found a higher consultation rate for hyper-r~ference users than for conventional dictionary users. The study also compared bilingual (Spanish/English) and monolingual (Spanish~simplified Spanish) dictionary use. Bilingual dictionary users consulted 25% more definitions than did monolingual dictionary users. Bilingual dictionary users completed reading in 20% less time than monolingual dictionary users. Efficiency was also higher during bilingual dictionary use than during monolingual dictionary use. Differences in comprehension were not significant. Directions for further research and development concerning electronic text and hyper-references are offered. ETR&D, Vol. 41, NO. 4, pp. 63-73 ISSN 1042-1629 [] A little over a decade ago, personal computing launched a revolution in the way humans produce and consume information. Word processing exemplifies the production aspect of this revolution, as it has transcended the traditional typewriter by offering superior capabilities for managing information, facilitating revisions (Leonardi, 1987), producing professional print quality that increases pride of authorship (Baer, 1988; Kahn, 1987), and easing access to writing tools such as a thesaurus or spelling checker (Kasnic & Stefano, 1987). With the production aspects secured, developers are now seeking to revolutionize information consumption by advancing key technologies involving information storage, computer portability, networking, intelligent interfaces, a n d screen fidelity. As these technical advances mature and as the economical and ecological advantages for conserving paper are realized, the widespread reading of electronic text may initiate the next major transition in the information revolution. Electronic references that aid in understanding text-- namely dictionaries and glossaries-represent one of the most rapidly growing forms of electronic text (Weissman, 1988; Wooldridge, 1991). Early proponents of automated dictionaries (Fox, Bebel, & Parker, 1980; Miller, 1980) recognized that the ability ETR&D,VoL 41, No. 4 to immediately consult a reference and then return directly to the main reading passage has implications for learning. After all, readers are distracted from the target information when they locate a paper reference (which is often across the room), flip through pages, and struggle with reading the fine print. Users who would otherwise forgo these lengthy paper searches may use electronic references frequently because the searches take place in a matter of seconds and involve less diversion from the study text. As was true of their paperbased forbears, there are many design issues for electronic texts and online references. HYPERTEXT AND HYPER-REFERENCES The general term "hypertext" describes an array of emerging technologies for accessing, organizing, and relating electronic-based information. Ted Nelson (1981,1982) first used this term to describe a hypothetical electronic notebook for immediately consulting elaborations on any word or phrase by simply pointing to or "clicking" information the user wishes to learn more about. Advances in computer hardware and in software, such as HyperCard, protocols for intelligent communication devices (Hill & Yamada, 1993), and pen-based portable computers, are making hypertext accessible to a much broader range of developers, who in turn are fostering innovative strategies for applying hypertext to instruction (Campbell & Goodman, 1987; Halasz, 1987; Smith & Weiss, 1988). Some recent hypertext innovations capitalize on the mass storage capacity of videodiscs and CD-ROM to integrate media in enhanced i m p l e m e n t a t i o n s k n o w n as h y p e r m e d i a (Kinzie & Berdel, 1990). A full-fledged hypermedia environment gives users unlimited access to diverse multimedia knowledge bases with unrestricted cross-referencing, Using a hypermedia reference, a learner might click on the word "whistle" and have the option of hearing either the standard pronunciation or one of several whistle sounds. In another hypermedia scenario, a user who clicks on the word "throw" in the text of a story can view a full-motion video clip depicting several throwing techniques, then click on a picture of a baseball to see how it is manufactured, and so on. These systems are very flexible, but hypertext systems that are completely open can create what Kerr (1986) described as "wayfinding" problems that cause users to become lost in hyperspace. For this study, we chose a hypertext design that uses an electronic dictionary because the searching advantages of hypertext adaptations are especially evident with large modular r e f e r e n c e s (Locatis, L e t o u r n e a u , & Banvard, 1989; Raymond & Tompa, 1988). The dictionary is also a good reference to study because it is "the most widespread single language improvement device" (Brumfit, 1985, p. v.), yet people find it "arduous" to consult (Keller, 1987, p. 17). The design we chose is a simple, bidirectional hypertext implementation that limits potential wayfinding confusion. Our electronic book looks similar to an open paperback, with two pages shown on each computer screen. The user navigates through the pages in a linear fashion by clicking on small arrows at the bottom of the screen. A user who wishes to consult the dictionary clicks on a word in the text and the definition for the word appears immediately in a window on the opposing page of the electronic book. After reading the definition, the user clicks on the "close" button on the definition window and the underlying text reappears (see Figure 1). We describe this electronic dictionary as a "hyper-reference" implementation, meaning an electronic reference aid that offers immediate access to supportive information with a clear and direct return path to the target information. Although hyper-references do not offer the unrestricted mobility envisioned both by early innovators (Bush, 1945; Engelbart & English, 1968; Nelson, 1981) and more recent hypertext proponents (McKnight, Dillon, & Richardson, 1991; Rada, 1991), their ability to limit wayfinding problems may outweigh alleged advantages of complex hypertext when the intent is to provide immediate ancillary support. Moreover, when using hypertext systems that have weak link indicators, "most 65 HYPER-REFERENCES FIGURE 1 [ ] Electronic Book with Hyper-Reference i Gonzdlez "Nunca m e d o y por vencida" Comenz6 su lucha temprano, Los obstaculos no h a n podido con ella, es m~s, la h a n motivado a laborar con mayor ahinco. Primero fue el estudio, luego largas horas de trabajo -de ir aprendiendo y mejorando- h a s t a lograr el objetivo: instalar u n a pequefia industria. Sin embargo, al poco tiempo u n incendio destruy6 todos sus sue~os y meses delesfqerzo]y dedicaci6n. Pero e s t e ~ r i m e r gran tropiezo en su vida empresarial e~erzo ® esfuerzo m. effort, exertion; [valor) courage, bravery; (~nimo) spirit, heart. In the above depiction of the electronic book, the user has clicked on the word "esfuerzo* and the definitionhas appeared in the window on the opposing page. Thus, the selected word remains visiblein the context of the storywhile the user reads the definition.The button in the upper rlght-hand corner isused to close the definitionwindow. Thisfigure shows the bilingual(English)version.A second verSion that provided monolingual (SimplifiedSpanish) definitionswas also used in the study. students tend to start from a single organizing node, examine links from that node and return immediately to the same node" (Welsh, Murphy, Duffy, & Goodrum, 1993, p. 20). FOREIGN LANGUAGE DICTIONARIES We chose foreign language as the content discipline for investigating the use of references because foreign language study typically involves considerable dictionary usage, thereby creating a rich "look-up" environment. Another reason for selecting this content was the need for additional research concerning dictionary use during foreign language learning. Although self-report questionnaires offer in- sights concerning the use of dictionaries (B6joint, 1981; Coviello, 1987; Galisson, 1983; Tomaszczyk, 1979), several scholars (Crystal, 1986; Hartmann, 1983, 1987; Ilson, 1985; Lantolf, Labarca, & den Tuinder, 1985; Zgusta, 1975) have called for empirical evidence to support opinions concerning when and how learners make effective use of dictionaries. Dictionaries designed to support foreign language learning fall into two primary categories: bilingual dictionaries, which provide the reference word in the target language and definitions in the readers' native language (e.g., Spanish/English), and monolingual dictionaries, which present both the reference word and simplified definitions in the target language (e.g., Spanish/simplified-Spanish). Foreign language learners prefer bilingual dic- 66 ETR&D, Vol.41, No. 4 tionaries (Atkins, 1985; Benoussan, Sim, & Weiss, 1984). This preference possibly stems from the desire to gain near-term understanding of the material they are immediately reading, as opposed to the longer-term desire of mastering a foreign language. However, many language educators (Snell-Hornby, 1984; Yorkey, 1970) believe that bilingual dictionaries are counterproductive because they cultivate the erroneous assumption that there is a one-to-one correspondence between the words of the two languages. Bland, Nobitt, Armington, and Gay (1990) refer to this problem as the "naive lexical hypothesis." Moreover, these critics contend that bilingual dictionaries encourage learners to engage in surface translating rather than reading and thinking in the target language. tionaries, we expected that the combined reading and consultation time would be less when consulting a bilingual dictionary than when consulting a monolingual dictionary. M~HOD Design A 2 x 2 design was used in comparing reference media (hyper-referenceand conventional paper) and dictionary language (bilingual and monolingual). Comparisons of reference media were made on four dependent measures: frequency of consultations, reading time, efficiency, and comprehension. Subjects RESEARCHASSUMPTIONS This study investigated the computer's potential to alter the way information is consumed by expediting the use of adjunct aids while reading electronic text. Because hyper-references offer immediate access to supportive information, we expected that the definitions would be consulted more frequently and that the combined reading and consultation time would be less when consulting a hyper-reference than when consulting a conventional dictionary. The ease of accessing a hyper-reference and returning to the target text should also benefit comprehension by enabling learners to focus on the reading passage. Consequently, we anticipated that comprehension would be greater when reading and consulting a hyperreference than when reading the same passage while consulting a conventional dictionary. Another interest of this study was to compare bilingual and monolingual dictionaries across both the hyper-reference and conventional paper conditions on frequency of use and overall reading time. The preference for bilingual dictionaries reported in previous studies was expected to prevail, with the bilingual dictionary being consulted more frequently than the monolingual dictionary. Because definitions in bilingual dictionaries tend to be briefer and require less effort to process than definitions in monolingual dic- Of the 80 undergraduates who volunteered to participate in the study, 27 were male and 53 were female. At the time of the study, all subjects were enrolled in sections of a fifth-semester university Spanish course. They reported an average of 3.3 years of Spanish coursework before attending the University. The mean number of university semester hours completed was 67, and the subjects' mean GPAwas 3.16 for all university coursework. Subjects also reported that they had completed an average of 11 hours of university Spanish coursework, with a mean GPA of 3.44. Materials The text used in all conditions was a replica of the 420-word article entitled "Nunca me doy por vencida" that appeared in the November 21,1989, issue of Rumbo, a Costa Rican magazine. A professor of Spanish judged the article to be moderately difficult for students in their fifth semester of Spanish. Results from a cloze version of the article, completed by 16 fifth-semester students during the pilot phase, supported the professor's judgment. The mean restoration rate on the cloze analysis (42.3%) approached the range (44-53%) that serves as a practical guide in determining whether a passage is above frustration level while being 67 HYPER-REFERENCES challenging enough for instructional use (HaskeU, 1975). We prepared replicas of the article for four conditions: A. an electronic article with a bilingual hyperreference dictionary based on the American Heritage Larousse Spanish Dictionary B. an electronic article with a monolingual hyper-reference dictionary based on the Diccionario Larousse del espafiol moderno C. a paper article with a bilingual paper dictionary D. a paper article with a monolingual paper dictionary. For the two hyper-reference conditions (A and B), we, transcribed the Rumbo article to a five-page prototypic electronic book using HyperCard software. Readers navigated through this electronic article by clicking forward and back buttons at the bottom of each page. They could consult definitions for any word in the electronic book, and after clicking on a word, the definition window immediately appeared on the page opposite from the selected word. This opposing-page feature allowed learners to see the word in its context while reading the definition (see Figure 1). Definitions for words in condition Awere transcribed from the bilingual dictionary used in condition C. Definitions for condition B were from the monolingual dictionary used in condition D. Parallel forms of the electronic book were produced for the conventional dictionary conditions (C and D). To maintain format and size consistency across all conditions, paper copies were made from screen files of the electronic book. Subjects in conditions C and D received these paper versions of the Rumbo article and paper dictionaries. Individual orientation sessions preceded both media conditions. Subjects who received the electronic text completed a computer tutorial that provided practice in using the mouse, navigating through the electronic book, consulting the hyper-reference, and reviewing the guidelines for the study. Subjects in the conventional conditions received a written orientation to the study and instructions to underline all words consulted in the paper dictionary directly on the passage. Measures The consultation measure was the raw number of words the subjects looked up in the dictionary. The computer automatically recorded each word consulted in the hyper-reference conditions. For the paper conditions, the consultation measure was the number of words the subjects underlined in the passage. Study time was the number of minutes that transpired between the time the subjects completed the orienting phase and indicated they were ready to begin reading and the time when they reported that they had completed reading and were ready for testing. Thus, study time included both the time involved in reading the text and the time taken to consult the dictionary. We used a proposition recall protocol to measure comprehension (Bernhardt, 1983). Before testing, two evaluators independently reviewed the article using the identification scheme refined by Deese (1984), and they agreed that the text contained 65 propositions. A proposition was defined as a unit of meaning expressed in the form of a simple declarative sentence. For example, the sentence "My fat cat is lazy" can be described as having three propositions: the cat is lazy, the cat is mine, and the cat is fat. After testing, the two evaluators reported the number of propositions that they found in each of the 80 written recalls. A correlation coefficient of .96 between the two evaluators' proposition ratings indicated a high degree of inter-rater reliability (Moore, 1983). The reported comprehension score was the mean of the two evaluators' ratings. Procedures The 80 students who volunteered to participate in the study were randomly assigned to one of the four treatment conditions. Before beginning each individual testing session, the subject participated in orientation session that lasted about ten minutes. Subjects in the computer conditions (A and B) received a corn- 68 ETR&D,Vol,41, No. 4 puter tutorial explaining how to use the online dictionary. Subjects in the paper condition saw an example of the underlining procedure. All subjects read the following statement: You will have up to 30 minutes to read the fivepage Spanish passage, so you will have plenty of time to look up as many words as you wish. After reading the Spanish passage, you will be asked to write everything you can remember about the story. Your score will be based entirely on the contents of your description and will not be influenced by the number of words you looked up or by the amount of time you took to read the stor~ At the conclusion of the orientation phase, the observer responded to any questions and then advised the subjects that they should begin reading. Study time began when the subjects started reading the story. After a subject finished reading the story, the observer recorded the end of study time and gave the subject a response sheet. Instructions on the response sheet explained that subjects should write everything they could remember about the story. Lee (1986) found that comparable subjects can recall significantly more when they write in their native language than when they write in the target language. Thus, we directed subjects to write their recalls in English. Observers recorded results of an open and informal interview with ten subjects after they had handed in their written recalls. The questions asked were designed to provide general insights into the subjects' opinions about the usefulness of the dictionaries and the difficulty level of the passage. RESULTS A two-way A_NOVA was used to assess the effects of the two independent variables, reference media and definition language, on the four dependent variables. Means for all dependent variables are given in Table 1. Analysis of consultation frequency revealed that the hyper-reference group consulted over two times as many definitions (M = 28.3) as those w h o used conventional dictionaries (M = 13.1), F(1, 76) = 26.96, M S e = 170.28, p < .001. Bilingual dictionary users made over 25% more consultations (M = 23.6) than those w h o had access to the monolingual version (M = 17.7), F(1, 76) = 4.09, M S e = 170.28, p < .05. The interaction between reference media and definition language on consultations was not significant, F(1, 76) = 1.88, M S e = 170.28, p = .17. The analysis of efficiency (consultations per minute) confirmed that subjects who used the hyper-reference consulted more references per minute (M = 1.49) than those who used conventional dictionaries (M = .62), F(1, 76) = 90.1, M S e = .00004, p < .001. Efficiency was also TABLE 1 [ ] Mean Outcomes by Reference Media and Definition Language DEPENDENT MEASURES Consultations a M Hyper-reference (computer) Bilingual Monolingual Total 33.20 23.30 28.25 SD Study Timeb M 19.82 17.46 14.75 20.13 17.96 18.80 Efficiency c SD M SD 6.21 9.71 8.16 1.84 1.15 1.49 0.75 0.49 0.62 Conventional (paper) Bilingual 14.05 6.68 19.73 9.47 Monolingual 12.15 5.11 26.90 14.51 Total 13.10 5.95 23.31 12.63 Note: n = 20 in each treatment cell (total n = 80). aConsultafions:Raw number of words consultedin the dictionary. bStudyTune:Numberof minutes spent readingand consultingthe dictionary. CEfficiency: Numberof consultationsper minute. dComprehension:Numberof propositionsrecalledon the written posttest. Comprehension d M SD 0.63 0.44 0.64 11.80 10.10 10.95 6.62 6.55 6.56 0.25 0.17 0.25 13.10 12.20 12.65 7.58 6.84 7.14 HYI~R-I~FERENCE$ 69 higher during bilingual dictionary use (M = 1.30) than during monolingual dictionary use (M -- .82), F(1, 76) = 27.38, M S e = .00004, p < .001. There was a significant interaction between reference media and definition language on efficiency, F(1, 76) = 5.48, MSe = .00004, p < .05. Post hoc Scheff~ analysis was used to compare the mean efficiency for the four groups: (A) hyper-reference/bilingual, (B) hyperreference/monolingual, (C) conventional dictionary/bilingual, (D) conventional dictionary/monolingual. Analysis at the .05 level revealed that group A was significantly different from all other groups. Group B was significantly different from group C and from group D. Group C was not significantly different from group D. Although not extreme, this ordinal interaction (see Figure 2) shows that the efficiency gains of hyper-reference over paper were more pronounced with the bilingual dictionary (M = 1.84 compared to M = .75) than with the monolingual dictionary (M = 1.15 compared to M = .49). The difference in consultation efficiency might be attributed primarily to differences in consultation frequency. However, it should also be noted that, although the difference was not significant, the mean study time for the hyper-reference users (M = 18.80 min.) was nearly 20% less than the mean study time of the conventional dictionary users (M = 23.31 rain.), F(1, 76) = 3.77, MSe = 108.26, p = .O56. Subjects who used the bilingual versions spent about 20% less time reading and consulting the dictionary (M = 18.59 min.) than those who used the monolingual versions (M = 23.52 min.), F(1, 76) = 4.48, MSe = 108.26, p < .05. The interaction between reference media and definition language on study time was not significant, F(1, 76) = .94, M S e = 108.26, p = .34. The mean number of propositions recalled (comprehension) by the groups w h o used the hyper-reference dictionaries (M = 10.95) was not significantly different from those w h o used paper dictionaries (M = 12.65), F(1, 76) = 1.21, M S e = 47.74, p = .28. The difference on comprehension was also not significant when users of bilingual dictionaries (M = 12.45) were compared to users of monolingual dictionaries (M = 11.15), F(1, 76) = 1.21, M S e = 47.74, p = .40. The interaction between reference media and definition language on comprehension was not significant, F(1,76) = .07, M S e = 47.74, p = .80. A post hoc analysis revealed a significant positive correlation, r = .44, p < .01, between subjects' comprehension and overall GPA, as well as between subjects' comprehension and GPA in university Spanish courses, r = .35, p < .01. Comprehension measures usually correlate with GPA. Consequently, these cor- FIGURE 2 [ ] Interaction between Dictionary Language and Reference Media on Efficiency 1.90 171 ........................... I I /" 11,11/ i o7°11 0.19 0.00 I O m ~ I 70 ETR&D,Vol.41,No.4 relations supported our notion that the proposition recall measure was valid for this study. We also conducted post hoc analyses to consider the possibility that a higher consultation rate by low-ability students may mask comprehension differences between hyperreference and paper media. However, the correlations w e r e not significant b e t w e e n consultation frequency and overall GPA (r = -.033), or between consultation frequency and GPA in Spanish courses (r = .067). Further post hoc analysis did not find significant differences in comprehensionbetwee~ the reference media when either overall GPA or GPA in Spanish courses was used as a covariate. The exit interviews indicated that the difficulty level of the passage was appropriate. The most common concern voiced during these interviews came from those who received monolingual dictionaries. For example, one subject under the monolingual conventional dictionary condition explained, "It was not like the dictionary I'm used to. It wasn't much help." When asked which type dictionary she used, she pulled from her purse a worn copy of the same dictionary that was used in the bilingual conditions. When subjects who were familiar with the use of computers were asked what they thought of the hyper-reference dictionary, their responses were generally enthusiastic. For example, one subject responded, "It's great! It makes looking up words a lot easier. Where can I get one?" One of the subjects in the monolingual hyper-reference condition also commented that he wished he could have clicked on the words in the definitions for further explanation. In essence, he was asking for a more fully implemented hyper-reference. DISCUSSION AND RECOMMENDATIONS The Use of Hyper-References Results of this study indicate that readers consuit hyper-references much more frequently than comparable paper references. Because hyper-references offer more efficient access, they appear to lower the "consultation trigger point," thereby increasing the learners' appetite for elaboration. For example, one hyperreference user made 79 unique consultations in the 420-word article, whereas the highest number of consultations in the paper condition was 31. In most cases, easy access to references should benefit students by encouraging them to become actively involved in learning, but it is possible that some learners will spend an excessive amount of time repeatedly looking up relatively familiar information. Findings from this study did not suggest that hyper-references adversely affect study time. Instead, using hyper-references in rich "lookup" environments may reduce the overall study time. Even with the short (420-word) reading passage used in this research, the mean study time of the hyper-reference group was about 20% less than that of subjects who used the paper dictionaries---a difference that approached statistical significance (p = .056). Researchers who wish to investigate this notion further could increase sensitivity to study time by incorporating challenging reading materials that are longer than the passage used in this study. This study did not find significant differences in comprehension. However, it would be premature to conclude that the use of hyperreferences does not benefit learning. More conventional comprehension measures, such as multiple-choice questions or sentence-completion tasks, may demonstrate different learning advantages of hyper-references. We used a short passage and time scope for this study. Further research could manipulate the parameters of text type (i.e., genre) and length, and measure effects of hyper-reference use over longer periods. Investigators might also consider the influence of hyper-references on outcomes such as incidental vocabulary acquisition and the development of inquiry skills. References in Foreign Language Learning The finding that foreign language readers consuit bilingual dictionaries 25% more often than monolingual dictionaries supported previous contentions concerning user preferences 71 HYPER-REFERENCES (Atkins, 1985; Benoussan et al., 1984). A more novel finding is that, even though bilingual dictionary users looked up more words, they completed the reading task in 20% less time than the subjects who used the monolingual dictionary. This indicates that bilingual dictionaries expedite reading during foreign language learning to a greater extent than monolingual dictionaries; however, questions regarding their relative contribution to learning remain. Comprehension differences between bilingual and monolingual dictionary use were not significant. Given the scope of content used in this study, we caution against concluding that the language used in reference does not influence learning. The notion that greater improvements in vocabulary ability will result from monolingual dictionary use seems reasonable because the cognitive tasks are more directly associated with understanding the foreign language than when the learner uses a bilingual dictionary and cycles from one language to another. However, the propositionrecall measure used in this study is not sensitive to incidental vocabulary acquisition because the subjects' responses are written in English. Further research comparing the influence of monolingual and bilingual dictionary use should compare conditions over extended periods with diverse reading passages on both comprehension and vocabulary acquisition. The significant interaction between definition language and reference media found that efficiency gains of hyper-references are more pronounced with bilingual dictionaries than with monolingual dictionaries. Possibly, this is because the bilingual hyper-reference combination is closest to an automatic translation device. One argument is that this combined "mechanical" and native-language advantage could benefit foreign language learners as they strive to read increasingly more difficult material and are slowly weaned from reliance on the reference. On the other hand, some foreign language scholars might view the bilingual hyper-referenceas an "automated crutch" that supports the naive lexical hypothesis and thus presents an even stronger barrier to foreign language immersion than the bilingual paper dictionary (Bland et al., 1990). FurtherResearch and Development Answers to questions regarding hyper-reference design issues also offer promise for advancing learning from electronic text. For example, are designs that allow users to consuit words within definitions more effective than the direct-return approach used in this study? What will be the trade-off between open access and wayfinding confusion? Do some learners benefit more than others from designs that offer media augmentations? When does media augmentation, including word pronunciation or graphical representations, promote learning, and when is it distracting? Do learners benefit from annotational capabilities that include glosses supplied by editors or possibly the more personalized "metanotes" (Wolfe, 1990, p. 223) that are similar to the handwritten note which readers write for themselves in margins? Alan Kay (1991), a noted developer of simulation software, anticipates that within the next ten years, educational software will begin to incorporate "flexible agents" that tailor learning environments to each learner and situation. Such flexible agents could be used in constructing personalized adjunct aids according to the context of the target information as well as the preferences and learning styles of the users. How much of this processing should be driven by machine and how much should be left to the learner? Jonassen (1991) favors incorporating the diversity and open-access notions of hypermedia within existing instructional design strategies to develop models that are more compatible with cognitive views of learning. Clearly, more discussion and research are needed before adopting such models. For example, the use of a hyper-reference augmented with sound may improve performance on certain pronunciation measures but raise controversy regarding the homogenization of dialects. Or, the bells, whistles, and freedom of choice in a completely open hypermedia environment may prove effective in promoting general inquiry abilities, but may interfere with instruction that focuses on the near-term understanding of specific information. As Jonassen and others (Locatis et al., 1989) have suggested, instruc- 72 tional models designed to integrate hypermedia must accommodate the diverse preferences of instructional developers, a range of instructional tasks and learning situations, as well as differences in learning styles. The role of computers in this study underlines their benefit for research and education. They are routinely used in cognitive psychology experiments because they provide unobtrusive, reliable, and manageable means of data collection (Gagn~, 1985). Lexicographers who wish to base their practice on empirical grounds (Hartmann, 1987; Hatherall, 1984) also find these compelling reasons for using c o m p u t e r s . Foreign language researchers (Bland et al., 1990) recognize that online data collection o p e n s a diagnostic w i n d o w to aspects of the reading process that were previously not readily accessible. By incorporating hyper-references within a local computer network, instructors can easily track student performance over the course of a semester on such variables as study time and the frequency of consultations. The resulting data give instructors additional insights for adjusting instruction to both the individual and group needs of their students. A u t o m a t e d data collection on hyper-references will have increased potential when educational institutions become linked b y the proposed broadband national networks (Gore, 1991). Wide-area databases could then compile data on such variables as the most c o m m o n l y looked-up words, which texts prompted the greatest number of consultations, and the percentage of consultations by part of speech. These data would assist educators in teaching reading and vocabulary more effectively, aid lexicographers in improving references, and open the door to a range of research possibilities for addressing questions such as w h y the same word is consulted repeatedly in one passage but not in another. Hypertext and its derivatives have been touted as revolutionary tools that will dramatically change the w a y humans consume informarion. Results from this study indicate that at least those hypertext features that support direct and immediate access to relevant information improve the use and efficiency of adjunct aids for reading. More research and development are needed to investigate how and if ETR&D,Vol. 41, No. 4 these mechanical advantages influence learning. Even with supportive findings, the advantages of electronic text m a y be slow in reaching m a n y readers because, as Hartley (1987) foretold, those who appreciate the advantages are already beginning to use electronic text as their primary means of communication. [] Ronald Aust is an Associate Professor of Curriculum and Instruction at the University of Kansas. Mary Jane Kelley is an Assistant Professor of Spanish in the Modem Languages Department at Ohio University. Warren Roby is an Assistant Professor of Foreign Languages at Washington State University. The authors thank Paul Markham and Phil McKnight of the University of Kansas Department of Curriculum and Instruction, and the ETR&D reviewers for their thoughtful critiques of the initial manuscript. REFERENCES Atkins, B. T. (1985). Monolingual and bilingual learner's dictionaries. In R. Ilson (Ed.), Dictionaries, lexicography, and language learning (pp. 15-24). Oxford: Pergamon Press. Baer, V. H. (1988). Computers as composition tools: A case study of attitudes. Journal of Computer-Based Instruction, 15, 144-148. B~joint, H. (1981). The foreign student's use of monolingual English dictionaries: A study of language needs and reference skills. Applied Linguistics, 2, 325-336. Benoussan, M., Sim, D., & Weiss, R. (1984). The effect of dictionary usage on EFL test performance compared with student and teacher attitudes and expectations. Reading in a Foreign Language, 2, 262-276. Bernhardt, E. B. (1983). Testing foreign language reading comprehension: The immediate recall protocol. Die Unterrichtspraxis, 16, 27-33. Bland, S. K., Nobitt, J. S., Armington, S., & Gay, G. (1990). The naive lexical hypothesis: Evidence from computer-assisted language learning. Modern Language Journal, 74, 440-450. Brumfit, C. J. (1985). Preface. In R. Ilson (Ed.), Dictionaries, lexicography and language learning (p .v). Oxford: Pergamon Press. Bush, V. (1945, July). As we may think. Atlantic Monthly, 101-108. Campbell, B., & Goodman, J. (1988). HAM: Generalpurpose hypertext abstract machine. Communications of the Association of Computing Machinery, 31, 856--861. CovieUo, G. (1987). I1dizionario oggi. Due gruppi di studenti messi a confronto su un "oggetto" molto discusso. Rassenga Italiana di Linguistica AppIicata, 19, 109-129. HYPER-REFERENCES Crystal, D. (1986). The ideal dictionary, lexicographer and user. In R. Ilson (Ed.), Lexicography: An emerging international profession (pp. 72-81). London: Manchester University Press. Deese, J. (1984). Thought into speech: The psychology of a language. Englewood Cliffs, NJ: Prentice Hall. Engelbart, D. C., & English, W. K. (1968). A research center for augmenting human intellect. AFIPS Proceedings, 33, 395-410. Fox, M. S., Bebel, D. J., & Parker, A. C. (1980). The automated dictionary. Computer, 13, 35-48. Gagnd, E. (1985). The cognitive psychology of school learning. Boston: Little, Brown. Galisson, R. (1983). Image et usage du dictionnaire chez des dtudiants (en langue) de niveau avancd. l~tudes de Linguistique Appliqug,s, 59, 5-88. Gore, A. (1991, September) Infrastructure for the global village. Scienti~ American, pp. 150-I53. Halasz, 1=. (1987). Reflections on notecards: Seven issues for the next generation of hypermedia systems. Communications of the Association of Computing Machinery, 31, 836-852. Hartley, J. (1987). Designing electronic text The role of print-based research. Educational Technology Research & Development, 35(1), 3-17. Hartmann, R. R. K. (1983). The bilingual learner's dictionary and its uses. Multilingua, 2, 195-201. Hartmann, R. R. K. (1987). Dictionaries of English: The user's perspective. In R. W. Bailey (Ed.), Dictionaries of English: Prospects for the record of our language (pp. 121-135). Ann Arbor: University of Michigan Press. Haskell, J. F. (1975). Putting cloze into the classroom. English Record, 27, 83-90. Hatherall, G. (1984). Studying dictionary use: Some findings and proposals. In R. R. K. Hartmann (Ed.), LEXeter '83 Proceedings (pp. 183-189). Ttibingen: Max Niemeyer Verlag. Hill, G. C., & Yamada, K. (1993, February 8). Five electronic giants hope general magic will turn the trick. The Wall Street Journal, pp. A1, 4. Ilson, R. (1985). Introduction. In R. Ilson (Ed.), Dictionaries, lexicography, and language learning (pp. 1-6). Oxford: Pergamon Press. Jonassen, D. H. (1991). Hypertext as instructional design. Educational Technology Research & Development, 39(1), 83--92. Kahn, J. (1987). Learning to write with a new tool: Young children and word processing. The Computing Teacher, 14,11-12. Kasnic, M. J., & Stefano, S. (1987). Computers and the language arts: More than a spelling checker. The Computing Teacher, •5(2), 31-32. Kay, A. C. (1991, September) Computers, networks and education. Scientific American, pp. 138-148. Keller, H. H. (1987). Pedagogical wishes for a machine dictionary: An example from Russian. Modem Language Journal, 71,12-17. Kerr, S. (1986). Learning to use electronic text: An agenda for research on typography, graphics and interpersonal navigation. Information Design Journal, 4(3), 206-211. 73 Kinzie, M. B., & Berdel, R. M. (1990). Design and use of hypermedia systems. Educational Technology Research & Development, 38(3), 61-68. Lantolf, J. P., Labarca, A., & den Tuinder, J. (1985). Strategies for accessing bilingual dictionaries: A question of regulation. Hispania, 68, 858-864. Lee, J. F. (1986). On the use of the recall task to measure L2 reading comprehension. Studies in Second Language Acquisition, 8, 201-212. Leoi~ardi, E. (1987). The micro in the English classroom: Shifting the emphasis from processing words to processing ideas. Educational Technology, 27, 45-47. Locatis, C., Letoumeau, G., & Banvard, T. (1989). Hypermedia and instruction. Educational Technology Research & Development, 37(4), 65-77. McKnight, C., Dillon, A., & Richardson, J. (1991). Hypertext in context. Cambridge, England: Cambridge University Press. Miller, G. A. (1980). Automated dictionaries. In A. Melmed (Ed.), Automated dictionaries: Reading and writing. Report of a conference on educational uses of word processors with dictionaries (pp. 25--29). Washington, DC: National Institute of Education. Moore, G. W. (1983). Developing and evaluating educational research. Boston: Little, Brown+ Nelson, T. H. (1981). Literary machines. Swarthmore, PA: Author. Nelson, T. H. (1982, March). A new home for the mind. Datamation, pp. 169-180. Rada, R. (1991). Hypertext: From text to expert. London: McGraw-Hill. Raymond, D. R., & Tompa, E W. (1988). Hypertext and the Oxford English Dictionary. Communications of the Association of Computing Machinery, 31(7), 871-879. Smith, J. B., & Weiss, S. E (1988). Hypertext: An overview. Communications of the Association of Computing Machinery, 31(7), 816--819. Snell-Hornby, M. (1984). The bilingual dictionary: Help or hindrance? In R. R. K. Hartmann (Ed.), LEXeter '83 Proceedings (pp. 274-281). Ttibingen: Max Niemeyer Verlag. Tomaszczyk, J. (1979). Dictionaries: Users and uses. Glottodidactica, 12, 103-119. Weissman, R. F. E. (1988). From the personal computer to the scholar's workstation. Academic Computing, 3, 10-14; 30--41. Welsh, T. M., Murphy, K. P., Duffy, T. M.. & Goodrum, D. A. Accessing elaborations on core information in a hypermedia environment. Educational Technology Research & Development, 41(2), 19-34. Wolfe, R. (1990). Hypertextual perspective on educational computer conferencing. In L. I-Iarasim (Ed.), Online education: Perspectives on a new environment (pp. 215-228). New York, Prager. Wooldridge, R. (1991). List of electronic dictionaries. ACH Newsletter, 13, 1, 4. Yorker; R. C. (1970). Electronic dictionaries in CALL. Computer Assisted Language Learning, 1, 95-109. Zgusta, L. (1975). Linguistics and bilingual dictionaries. Studies in Language Learning, 1, 95-109. Do You Know The Value of Your Audiovisual Media Collection? The Association for Educational Communications and Technology announces the publication of a new book for a new field, Appraising Audiovisual Media, written by Dr. Steve Johnson, one of only three full-time professional audiovisual media appraisers working in the U.S. APPRAISING AUDIOVISUAL MEDIA A Guide for Attorneys, Trust Officers, Insurance Professionals, and Archivists in Appraising Films, Video, Photographs, Recordings, and Other Audiovisual Assets By Who Uses Media Appraisals? media educators, universities with media collections, photographers and fdmakers, museums, archivists, donors, artists and producers, attorneys, trust officers, insurance professionals and others. What are Audiovisual Media Assets? Items include motion pictures, records, video productions, slides, cameras, optical disc recordings, projectors, audio scripts, animation eels, magic lantern slides, media books and periodicals, cinemabilia and much more! Dr. Steve Johnson Anyone who needs to determine or defend the value of audiovisual media and related rights: • L e a r n H o w to Find and W o r k With an Appraiser • P r o t e c t Y o u r M e d i a • U n d e r s t a n d the Appraisal Process • Over 20 Case Studies Bibliography, index. Published January 1993. ISBN 0-89240-068-4. 136 pages hardcover. $34.95, nonmember, $27.95 AECT member and bookstores. Add $3 shipping & handling to all orders. MasterCard & VISA aeeepted. COPYRIGHT INFORMATION SERVICES 1025 Vermont Ave., NW, Suite 820 Washington, DC 20005-3547 Phone (202) 347-7834 Fax (202) 347-7839