[316]

- 2 -

Applied Linguistics

Prepared by Karen L. Smith

How Does Varying Pre-listening Activities Affect Second Language Listening Comprehension?

Jane E. Berne

University of North Dakota

Abstract: Despite the increasing importance given to pre-listening activities in second language (L2) listening instruction, no empirical research to date has compared the relative effectiveness of different pre-listening activities or examined whether the relative effectiveness of different pre-listening activities varies as a function of multiple exposures to the listening passage. A comparison of the listening comprehension performance of second language learners of Spanish who completed three different pre-listening activities after one and two exposures to the listening passage shows that scores for subjects completing the question preview activity were significantly higher than scores for subjects completing the filler activity and that additional exposure to the listening passage improves listening comprehension performance significantly, irrespective of pre-listening treatment. These findings have implications for pedagogy.

Key Words: classroom-based research, foreign language learning, listening comprehension assessment, listening comprehension instruction, pre-listening activities, multiple readings

Introduction

     Despite the increasing emphasis on pre-listening activities in second language (L2) listening instruction, no empirical research has been found to date which compares the effects of different types of pre-listening activities on L2 listening comprehension performance. This question regarding the relative effectiveness of different pre-listening activities was prompted by research into the effects of different pre-reading activities on reading comprehension performance. Hudson (1982) examined the relative effectiveness of different pre-reading activities on the reading comprehension performance of three levels of adult learners of English as a Second Language. Subjects completed one of three pre-reading activities: 1) the PRE activity which consisted of looking at visuals, answering questions based on those visuals, and then making predictions regarding passage content; 2) the VOC activity which consisted of reading a list of vocabulary items from the passage whose definitions were then provided orally; and 3) the RT activity which consisted of reading the passage, completing a multiple-choice comprehension test, and then repeating both procedures. Hudson found that at the beginning level, the PRE activity was significantly more effective than either the VOC or the RT activity in facilitating comprehension. At the intermediate level, both the PRE and the VOC activity were significantly more effective than the RT activity in facilitating comprehension. At the advanced level, there were no significant differences between any of the pre-reading activities, but there was a tendency for the RT activity to be more effective than either of the other two activities. Based on these results, Hudson concluded that different types of pre-reading activities were most effective at different levels of L2 proficiency.

     Stahl (1983) compared the effects of two different types of vocabulary training methods on the L1 reading comprehension performance of fifth graders with average reading ability. The first vocabulary training method, the definitional method, emphasized learning the definitions of particular words. The second vocabulary method, the mixed method, combined learning the definitions of particular words and learning the meanings those words have in different contexts. A third activity, serving as the [317] control, consisted of completing and discussing exercises from a supplementary comprehension skills book, but involved no vocabulary instruction.

     Results indicated that subjects who received either form of vocabulary training scored significantly higher than subjects in the control group on sentence cloze and sentence anomaly tests. This was also true for two of three groups of subjects on a multiple-choice test of passage comprehension and a multiple-choice synonym test. Regarding the relative effectiveness of the definitional and mixed methods of vocabulary training, subjects who received the mixed method of vocabulary training scored higher than those receiving the definitional method on the sentence cloze and sentence anomaly tests. Subjects receiving the mixed method of vocabulary training also tended to score higher than subjects receiving the definitional method on the multiple-choice synonym and passage comprehension tests; however, these differences were not significant. Based on these findings, Stahl concluded that there was qualified support for providing vocabulary training before reading and for using a mixed method of teaching vocabulary rather than a definitional method.

     Taglieber, Johnson, and Yarbrough (1988), investigated the relative effectiveness of different types of pre-reading activities on the reading comprehension performance of intermediate-level adult learners of English as a Foreign Language in Brazil. For each of four passages, each of the following pre-reading activities was completed by separate groups of subjects: 1) a pictorial context activity in which subjects described the content of three pictures, tried to make connections between them, and then made predictions regarding passage content; 2) a vocabulary pre-teaching activity which consisted of presenting vocabulary words to the subjects in meaningful, but unrelated sentences on the blackboard and then having them predict the meaning of the words; and 3) a pre-questioning activity which consisted of providing subjects with a one sentence oral summary of the passage content and then having them generate a list of questions that they thought the passage might answer. A fourth group of subjects, serving as a control group, did not complete any activity prior to reading.

     Results revealed that subjects completing the pictorial context, vocabulary pre-teaching, and pre-questioning activities scored significantly higher than subjects in the control group. In addition, subjects completing the pictorial context and pre-questioning activities scored significantly higher than subjects completing the vocabulary pre-teaching activity. However, these findings applied only to scores on a test consisting of multiple-choice items. There were no significant differences between scores for any of the four pre-reading treatments on a test consisting of open-ended items.

     Based on these findings, Taglieber, Johnson, and Yarbrough concluded that pre-reading activities facilitate reading comprehension performance and that pre-reading activities based on providing pictorial context and on pre-questioning facilitate reading comprehension more than pre-reading activities which teach vocabulary. They attributed the differences between subjects' performance on the open-ended and multiple-choice tests in part to the fact that the three experimental activities they employed in this study may have facilitated the recall of details. They argue that this enhanced recall of details may have allowed subjects in the experimental groups to outscore subjects in the control group on the more detail-oriented multiple-choice test, but did not affect their performance on the more globally-oriented open-ended test.

     Given the somewhat equivocal nature of these results, the aforementioned lack of data on the relative effects of pre-listening activities, and the differences between L2 reading and listening comprehension that have been emerging in the research (e.g., Bernhardt and James 1987, Lund 1991, Mecartty 1993), it is difficult to make empirically-based recommendations regarding the use of pre-listening activities in the L2 context. In an effort to help fill this gap, the [318] present study compared the effects of different pre-listening activities on the listening comprehension performance of adult learners of Spanish as a Foreign Language.

     Two activities served as experimental treatments: 1) a question preview activity and 2) a vocabulary preview activity. The question preview activity consisted of allowing subjects to study the questions and possible responses used to assess comprehension of the passage prior to listening. The vocabulary preview activity consisted of allowing subjects to study a list of ten key words from the passage and their English equivalents prior to listening. A third group of subjects, serving as a control group, completed a filler activity unrelated to the listening passage. The decision to employ a question preview activity was based on the following recommendation from Alice Omaggio: «Consider providing the questions, completions, or other type of comprehension check before [italics original] students hear the passage. This gives students an idea of passage content, thus serving as an advanced organizer and providing 'schema' for comprehension» (1986: 143). The decision to employ a vocabulary preview activity reflected research which has shown that lexical ignorance is the main obstacle to listening comprehension (Kelly 1991).

     Another reason for choosing these two activities is that each would seem to foster a different means of processing passage content, thereby providing an interesting contrast. Question preview activities, by leading listeners to focus on the passage as a whole rather than individual words or structures, encourage a more top-down approach to processing passage content. In contrast, vocabulary preview activities, by leading listeners to focus on individual words rather than the passage as a whole, encourage a more bottom-up approach to processing passage content.

     A second question that this research sought to address was what effect would multiple exposures to the listening passage have on subjects' listening comprehension performance. Two studies have shed light on the question of whether or not L2 listening comprehension performance varies as a result of multiple exposures to the listening passage. Lund (1991) found that, for adult learners of German as a Foreign Language, listening comprehension performance, as measured by propositions and lexical items recalled, improved after a second opportunity to listen to the passage. Results also indicated that this improvement was greater for third-semester learners than it was for first- and second-semester learners.

     Cervantes and Gainer (1992) found that, for adult learners of English as a Second Language, scores on a partial dictation test taken after hearing a listening passage with a high degree of subordination were higher when the passage was repeated than when it was heard only once. In addition, scores obtained when a simplified version of the passage, characterized by a lower degree of subordination, was heard once did not differ significantly from scores obtained when the syntactically complex version was heard twice. These results indicate that repetition facilitates listening comprehension. Furthermore, since there was no difference in the scores of groups hearing the syntactically simplified version once and those hearing the syntactically complex version twice, these results also suggest that syntactic modifications intended to simplify a text may be unnecessary if other types of modifications, such as repetition, are available. The present study sought to expand this line of research by investigating what impact multiple exposures to the listening passage would have on the relative effectiveness of the three pre-listening activities.

Research Questions and Hypotheses

     The study addressed the following specific research questions:

     With respect to the first research question, the hypothesis was that subjects completing both the question and vocabulary preview activities would receive significantly higher scores than subjects completing the filler activity. Based on the results of Taglieber, Johnson, and Yarbrough (1988) and on the commonly accepted assumption that top-down processing is more efficient than bottom-up processing, it was further hypothesized that subjects completing the question preview activity would receive significantly higher scores than subjects completing the vocabulary preview activity. In terms of the second and third research questions, based on the results of previous research (e.g., Lund 1991, Cervantes and Gainer 1992), the hypothesis was that scores would improve after the second exposure to the listening passage, irrespective of pre-listening activity completed.

Subjects

     A total of 62 learners of Spanish at the university level served as subjects. All were native speakers of English and were enrolled in one of four sections of the third-semester course of a four-semester basic language sequence at a medium-sized university in the north central United States. The course in which the subjects were enrolled is a comprehension-based course. The language of instruction is Spanish and classroom activities emphasize comprehension, communication, and interaction. There is some explicit instruction of grammar in class, but most grammar practice is carried out in homework assignments.

     In class, learners in this course regularly participate in a variety of communicative activities which directly involved listening such as Total Physical Response, interviews, surveys, choosing vocabulary items based oral definitions, drawing pictures based on oral descriptions, and marking maps based on oral instructions. From time to time, they also listen to taped passages which accompany the text and complete the corresponding exercises. Additionally, the mid-term exam, the final exam, and each of five bi-weekly quizzes given in this course all have a section which explicitly tests listening comprehension. The listening sections of these exams and quizzes consist of prose passages or definitions read by the instructor. The prose passages relate to particular chapter topics and incorporate vocabulary from the text to a greater or lesser degree. In some cases, learners answer multiple-choice questions while in others, they indicate whether or not statements related to the passage are true or false and then correct the false statements. The definitions are Spanish definitions of vocabulary items from the chapter or chapters being tested. Learners match up the definitions with the corresponding vocabulary items listed on their test papers.

     The in-class listening activities listed above not only provide an informal means for assessing listening comprehension skills using materials related to particular chapters in the text; they also provide opportunities for developing and practicing listening skills or strategies. The listening sections on the tests and quizzes allow the instructor to assess knowledge of particular course material, usually vocabulary items, through the aural mode.

     As a result of the emphasis placed on listening comprehension in the course in which they were enrolled, the subjects who participated in this study were relatively experienced L2 listeners of Spanish. Moreover, they were familiar with the types of activities and tasks employed in this study. However, despite their experience with listening in Spanish and their familiarity with the format of the experimental materials, there appeared to be little correlation between subjects' listening comprehension performance in the classroom context and their performance in the experimental context. Scores on classroom listening comprehension measures were generally quite [320] high whereas scores on the experimental tasks were generally quite low.

     This discrepancy may be the result of the particular passage. It was longer than the passages the subjects usually heard in class, it involved unfamiliar vocabulary, and subjects had no prior exposure to the topic in their text. Furthermore, subjects were accustomed to having listening passages read by their instructor and outside of the tapes which accompanied the text, they had little exposure to other speakers of Spanish. Finally, the structure of the experimental passage was more complex (e.g., more compound sentences, more variety of tenses) than the passages used in class. All of these factors related to the experimental passage may have led to the apparent discrepancy between subjects' listening comprehension performance in classroom and experimental contexts.

Materials

     The materials prepared for this study included a listening passage, a vocabulary preview activity, and a multiple-choice comprehension test. The listening passage consisted of an 862 word expository lecturette in Spanish on the emergence of videotaped résumés. To create the listening passage, the researcher prepared a script based on an article published in the Colombian weekly magazine Semana. Once the script was completed, a female native bilingual speaker of Spanish and Catalan from Barcelona, Spain recorded it on videotape. The researcher made the decision to use videotape as the mode of presentation on the basis of research which has shown that the use of appropriate videos may enhance L2 listening comprehension performance (e.g., Rubin 1990, Secules, Herron, and Tomasello 1992).⁽³⁵⁾

     The decision to use a prepared lecture as the listening passage was the result of several factors. First of all, subjects would have been be familiar with the basic format of a lecture from their other classes and could have made use of that knowledge to help them comprehend the passage. Second, the prepared lecture was similar to the types of «teacher talk» which the subjects had experienced in their Spanish class. Thirdly, subjects had experience with only a limited variety of listening passages: instructor monologues, live conversations or discussions with classmates, and taped conversations. Fourth, due time constraints and limited access to such materials, subjects had little exposure to authentic listening passages (i.e., listening passages involving native speakers and intended for a native speaking audience). A prepared lecturette was thus consistent with what the subjects had done in class, thereby giving them a point of departure for completing the experimental task.

     In addition to being consistent with what the subjects had done in class, the use of a prepared lecturette is consistent with many of the listening comprehension materials currently available. Authentic listening passages are becoming more common; however, prepared prose passages and dialogues still predominate, particularly in standardized tests of listening comprehension. Similarly, much of the material to which listeners are exposed through broadcast media is in fact prepared prose similar to the experimental passage. News anchors read from prepared sheets or the Tele-Prompt-R, feature reporters present previously prepared material, hosts and performers on various types of programs often read from cue cards, interviewers read prepared questions, and actors generally follow prepared scripts. Moreover, with the exception of live broadcasts, much of the material presented via the broadcast media is rehearsed and/or edited before airing, Hence, a prepared lecturette presented on videotape is also consistent with what appears on the broadcast media, at least to some extent.

     Another consideration in choosing a prepared prose passage was the fact the use of an unfamiliar type of passage or an authentic passage may have increased the difficulty of the task and the anxiety level of the subjects. Both could have had a negative effect on subjects' performance. Furthermore, subjects' lack of familiarity with certain [321] types of passages and with authentic passages may have introduced an additional variable into the study. Questions could have been raised as to the extent to which subjects' performance was affected by their lack of familiarity with nature of the listening passage. The use of a type of passage with which the subjects were familiar eliminated this as a possible factor in subjects' performance.

     The vocabulary preview activity, presented in Appendix A, consisted of a list of ten words or expressions taken from the passage and their English equivalents. In an effort to provide some context, the sentence in which the word or expression first occurred in the passage, or a close approximation thereof, appeared below each entry. Following procedures that Taglieber, Johnson, and Yarbrough (1988) employed when developing their vocabulary pre-teaching activity, the researcher chose the words and expressions included in the vocabulary preview activity according to two criteria. First, the word or expression had to be important to the overall comprehension of the passage and secondly, the word or expression had to be unfamiliar to the subjects. An experienced instructor of Spanish then reviewed the researcher's choices and proposed appropriate changes.

     The comprehension test, shown in Appendix B, consisted of 10 short, incomplete statements and a set of three options for completing each statement. To avoid confounding subjects' listening comprehension performance with their L2 reading ability, the subjects' native language, English, served as the language of assessment on the comprehension test. The researcher chose a multiple-choice format as the assessment task because it is still a common and very familiar means of measuring comprehension, both on classroom and standardized tests of listening comprehension. In addition, such multiple-choice tests of passage content have been used in previous L2 listening research (e.g., Shohamy and Inbar 1991).⁽³⁶⁾

Data Collection Procedures

     Prior to testing, the researcher prepared three different sets of experimental packets. Each set of experimental packets consisted of written instructions and materials pertaining to one of the three pre-listening activities, a copy of the comprehension test, a sheet of paper with instructions for completing a written recall, and a second copy of the comprehension test⁽³⁷⁾. On the day of testing, the researcher randomly distributed the experimental packets to the subjects in four classes so that each of the three pre-listening activities was completed by approximately one-third of the subjects in each class. The researcher chose to randomly distribute the pre-listening activities within classes rather than across classes in an effort to control for any differences between the individual classes which may have affected the results.

     Once the packets had been distributed, subjects had three minutes to complete the activity described in the instructions they received. Subjects completing the question preview activity studied the questions and possible responses which made up the comprehension test. Subjects completing the vocabulary preview activity studied the prepared vocabulary list. Subjects completing the filler activity wrote out the numbers from one to fifty in Spanish. The question preview activity followed procedures used in previous L2 listening comprehension research which employed similar question preview activities (e.g., Buck 1991, Shohamy and Inbar 1991). The vocabulary preview activity was a variation of the vocabulary pre-teaching activity that Taglieber, Johnson, and Yarbrough (1988) employed, but differed from it in terms of how it was implemented. In the present study, the chosen vocabulary items appeared along with their L1 (English) equivalents and the subjects studied them silently while in the Taglieber, Johnson, and Yarbrough study, the L1 (Portuguese) equivalents of the words did not appear and the subjects read the sentences aloud, predicted the meanings of the words, and then [322] discussed their predictions.

     There are some limitations associated with the vocabulary preview activity employed in this study. First, an activity where vocabulary items are studied silently is not representative of vocabulary activities typically carried out in language classrooms. Second, an activity where the learners do not make predictions regarding unfamiliar vocabulary but simply study the definitions provided may not encourage the engagement of background knowledge. Finally, an activity where learners read the vocabulary items but do not hear them may put learners with an auditory learning style at a disadvantage.

     Despite these limitations, the researcher chose to employ the vocabulary preview activity, mainly for reasons of logistics. Since all subjects in a particular class were tested at the same time and in the same room, it was not possible to employ pre-listening activities which involved oral presentation and participation. Additionally, employing similar procedures for both the question and vocabulary preview activities reduced the possibility that differences between them could be attributed to differences in procedure or mode of presentation.

     Logistical concerns also entered into the decision to use a filler activity. Since the control and experimental groups were not separated, the researcher determined that a filler activity was necessary to keep subjects in the control group focused while subjects in the two experimental groups completed their assigned activities. The decision to have subjects write out the numbers from one to fifty in Spanish was based in part on previous psycholinguistic research which used writing numbers as a filler or distractor activity (e.g., Bock and Brewer 1974; Brewer 1975). Furthermore, since there were few references to numbers in the listening passage, the researcher thought that writing out numbers would not influence comprehension of the passage, either directly or indirectly.

     After completing their assigned pre-listening activity, subjects watched the videotaped lecture. The researcher directed the subjects not to take notes, but rather to listen and concentrate on remembering as much information as possible. When the tape was finished playing, subjects turned to the first copy of the comprehension test and answered the questions. Subjects then completed a free written recall of the passage content in English. Completion of both tasks required approximately 15 minutes. Once this portion of the testing was completed, subjects watched the videotaped lecture a second time and then re-took the comprehension test. Administration of the complete series of experimental tasks lasted approximately 40 minutes.

Data Analysis Procedures

     Subjects' scores on each of the two comprehension tests consisted of the number of correct responses out of a possible 10 correct responses. The researcher then submitted the raw scores to a three X two Analysis of Variance (ANOVA) with repeated measures. The ANOVA procedure allows researchers to compare the mean scores of three or more groups of subjects, or cells, on some measure of performance. Subjects are divided among the cells according to certain characteristics. These characteristics are the variables manipulated by the researcher and are called independent variables. Each independent variable may consist of two or more categories, or levels. Independent variables may be either between-subject which means that data from a particular subject relative to that variable appear once in the analysis, or within-subject which means that more than one sample of data from a particular subject relative to that variable appear in the analysis. The measure of performance is called the dependent variable. A repeated-measures ANOVA compares mean scores from the same subjects on several different measures or on the same measure administered at different times.

     In this study pre-listening treatment and exposure to the passage, as measured by subjects' scores on the two comprehension [323] tests, served as independent variables. Pre-listening treatment served as a between-subjects variable while exposure, the repeated-measure, served as a within-subjects variable. Listening comprehension performance, as measured by the subjects' average scores across the two comprehension tests, served as the dependent variable. In order to determine significance, the researcher chose an alpha level of .05.

Results

     Means and ANOVA summary tables for this analysis appear in Tables One and Two respectively. Results indicate that when the scores were averaged across the two comprehension tests, there were main effects for pre-listening activity and for exposure; however, there was no interaction between pre-listening activity and exposure.

     In order to examine the source of the main effect for pre-listening activity, the researcher conducted a post-hoc Games-Howell test. Results of this test, displayed in Table Three, revealed that there was a significant difference between the average scores for subjects completing the question preview and filler activities. A post-hoc contrast confirmed this finding (p =.0227). No other comparisons reached significance; however, a post-hoc contrast revealed that the difference between the average scores for subjects completing the question and vocabulary preview activities approached significance (p=.0543).

Table Two
ANOVA Summary Table for Combined Comprehension Scores.
	Source	df	Sum of Sq	Mean Square	F-Value	P-Value
	Prelist	2	21.897	10.949	3.203	.0478 *
	Subject (Group)	59	201.692	3.419
	Scores	1	35.175	35.175	30.8	.0001 *
	Scores X Prelist	2	3.137	1.568	1.375	0.2608
	Scores X Subject (Group)	59	67.290	1.141
Dependent: List Perf
p .05
Prelist = pre-listening activity, Scores = scores on comprehension tests 1 and 2, List Perf = average scores of both comprehension tests

Table Three
Games-Howell Test for Pre-listening Activity.
		Vs.	Diff	Crit. Diff
	Fill	V Prev	.129	0.859
		Q Prev	.933	.794 S
	V Prev	Q Prev	.804	0.845
S = significantly different at this level; p .05
Q Prev = question preview activity, V Prev = vocabulary preview activity, Fill = filler activity

[324] Regarding exposure, post-hoc contrasts revealed that scores on the second comprehension test were significantly higher than scores on the first comprehension test for all three pre-listening activities (the question preview activity: p =.0137; the vocabulary preview activity: p =.0001, the filler activity: p =.0117). The lack of an interaction suggests that the effects of the additional exposure to the listening passage were relatively consistent across the three pre-listening activities.

Discussion

     With respect to the first research question, these results suggest that listening comprehension performance of foreign language learners does vary as a function of pre-listening activity. Despite this affirmative response, the data only partially support the hypothesis that subjects who completed both the question and vocabulary preview activities would receive significantly higher scores than subjects who completed the filler activity. Only subjects who completed the question preview activity received significantly higher scores than subjects who completed the filler activity. Likewise, the data only partially support the hypothesis that scores would differ significantly across the two experimental groups. Subjects who completed the question preview activity received higher scores than subjects who completed the vocabulary preview activity, but this difference was not significant.

     The fact that scores for subjects completing the question preview activity were higher than scores for subjects completing the filler activity is not surprising. Previewing the comprehension questions before listening allows the learners to see what information they will be responsible for, which then allows them to focus their attention on the appropriate portions of the passage. In contrast, the fact that scores for subjects completing the vocabulary preview activity did not differ significantly from scores for subjects completing the filler activity was comsomewhat unexpected given the results of Stahl (1983), Taglieber, Johnson, and Yarbrough (1988), an Kelly (1991). This finding suggests that, despite the apparent importance of lexical knowledge to listening comprehension, studying a vocabulary list prior to listening may not be a particularly effective means of improving listening comprehension performance.

     There are several possible reasons for the apparent failure of the vocabulary preview activity to facilitate listening comprehension. One reason may lie in the nature of the activity itself. As Carol Herron points out, studying vocabulary lists prior to listening is not a strategy that listeners would employ in real life. Specifically, she states that one does not drill vocabulary before turning on the television in order to better understand the program (1994;196). Her recommendations to provide learners with some type of brief summary before listening. She argues that this reflects typical behavior as people do read short synopses of television programs and movies before viewing them. Future research comparing pre-listening activities should include some type of summary-based pre-listening activity. [325]

     In addition, the fact that the vocabulary preview activity was entirely passive may have reduced its effectiveness as a pre-listening activity. As noted previously, the vocabulary preview activity consisted of having the subjects read the words and their English equivalents. It involved no engagement of background knowledge on the part of the subjects nor were they encouraged to develop any hypotheses about the passage they were about to view. It could be argued that subjects who completed the question preview activity, as a result of seeing the questions and possible responses, were better able to develop hypotheses or schemata relating to passage content and thus receive higher scores. Perhaps a more effective vocabulary-oriented pre-listening activity would entail actively engaging learners by having them brainstorm possible English equivalents of the items on the vocabulary list. Learners could also match up a list of vocabulary items with their target language definitions. Both activities would encourage learners to use their previous knowledge of both the topic and the target language as well as contextual cues to predict the meanings of the words.

     Another reason for the apparent failure of the vocabulary preview activity may be that the researcher chose the vocabulary words for the activity based on certain assumptions about which words the learners did not know and which of these were the most important to understanding the passage. It is possible that the researcher's choices did not include the vocabulary items that would have been most helpful to the subjects. Vocabulary preview activities may be more successful if they somehow allow learners to indicate the vocabulary items with which they are unfamiliar.

     One final reason for the reduced effectiveness of the vocabulary preview activity may be that subjects in this group were so focused on listening for the individual vocabulary items that they became distracted and were unable to attend to the passage content.

[326] In essence, they may have lost the forest for the trees. This may be the most logical explanation for the fact that subjects in the vocabulary preview group received the lowest scores on the first comprehension test.

     Regarding the second research question, the results suggest that the listening comprehension performance of foreign language learners varies as a function of multiple exposures to the listening passage. Scores for all three groups of subjects improved significantly as a result of viewing the passage a second time. While this finding is not surprising, the mean scores and the graph in Figure One reveal an interesting fact. On average, scores for subjects who completed the vocabulary preview activity improved nearly twice as much as those for the other two groups of subjects (1.53 pts. vs. .82 pts. for the question preview activity and .86 pts. for the filler activity).

     This finding may lend further credence to the argument that previewing vocabulary items distracts listeners from attending to the content of the passage. It is possible that upon viewing the passage a second time, subjects who completed the vocabulary preview activity were able to shift their attention to the passage content, thereby allowing them to overcome the negative effects of being distracted by focusing on vocabulary items when viewing the passage the first time.

     In terms of the third research question, the results suggest that the relative effectiveness of certain pre-listening activities does not vary as a function of multiple exposures to the listening passage. As noted previously, the lack of an interaction between pre-listening activity and exposure indicates that the effects of the additional exposure were statistically equal across all three pre-listening activities.

Conclusion

     While these findings are subject to several limitations, they suggest two possible conclusions. First, it appears that overall, pre-listening activities which involve previewing comprehension questions facilitate listening comprehension while pre-listening activities which involve passively previewing vocabulary do not facilitate listening comprehension and under certain circumstances, may even impede listening comprehension. This would imply that instructors are better off employing pre-listening activities which encourage learners to focus on the overall content of the passage rather than on individual elements of the passage. Second, it appears that the most effective means of improving listening comprehension performance is through additional exposure to the passage. This recommendation is not only supported by the data, but is also consistent with listening guidelines such as those prepared by ACTFL and the Interagency Language Roundtable. Both sets of guidelines indicate that at low and low-intermediate levels of listening proficiency, repetition may be necessary. Furthermore, this recommendation is consistent with authentic listening behavior. Repetition, restatement, and paraphrasing are all important factors in the negotiation of meaning, even among native speakers of a language. In the language learning context, instructors encourage learners to ask speakers to repeat, restate, or paraphrase if they miss something or if they don't understand. Indeed, such appeals for additional opportunities to hear the information constitute a highly effective strategy for improving listening comprehension.

     In terms of implications, perhaps the time has come to reevaluate procedures for listening comprehension assessment, particularly those employed on standardized tests of listening comprehension, so that they incorporate opportunities for additional exposure to the passage and negotiation of meaning. By incorporating these natural aspects of listening behavior into listening comprehension assessment procedures, we can perhaps achieve fairer and more accurate measures of L2 listening comprehension performance.

     One question this study did not address [327] is whether or not the relative effectiveness of different pre-listening activities varies as a function of listening proficiency⁽³⁸⁾. The findings of Hudson (1982) and those of several studies which incorporated L2 listening proficiency as a variable (i.e., Mueller 1980, Murphy 1987, O'Malley, Chamot, and Küpper 1989, Chiang and Dunkel 1992, and Vandergrift 1993) point to the possibility that different types of pre-listening activities might prove to be most effective at different levels of L2 listening proficiency. Another question this research did not address is whether or not there is a significant relationship between pre-listening activity, multiple exposures to the listening passage, and listening proficiency.

     These are just two of many other questions related to the selection and implementation of pre-listening activities that L2 listening comprehension research has yet to investigate. It is only by expanding the limited corpus of research involving pre-listening activities that we will be able to learn more about which pre-listening activities might be most effective for particular groups of learners.⁽³⁹⁾

WORKS CITED

     American Council on the Teaching of Foreign Languages. ACTFL Provisional Proficiency Guidelines. Hastings-on-Hudson: ACTFL Materials Center, 1985.

     Berne, Jane Ellen. The Effects of Text Type, Assessment Task, and Target Language Experience on Foreign Language Learners' Performance on Listening Comprehension Tests. Diss. U of Illinois at Urbana-Champaign, 1992. Ann Arbor: UMI, 1993. DA9236396.

     Bernhardt, Elizabeth B., and Charles J. James. «The teaching and Testing of Comprehension in Foreign Language Learning». Proficiency, Policy, and Professionalism in Foreign Language Education. Ed. Diane W. Birckbichler. Lincolnwood: National Textbook Company, 1987. 65-81.

     Bock, J. Kathryn, and William F. Brewer. «Reconstructive Recall in Sentences with Alternative Surface Structures». Journal of Educational Psychology 103 (1974):837-43.

     Brewer, William F. «Memory for Ideas: Synonym Substitution». Memory and Cognition 3 (1975): 458-64.

     Buck, Gary. «The Testing of Listening Comprehension: An Introspective Study». Language Testing 8 (1991): 67-91.

     Cervantes, Raul, and Glenn Gainer. «The Effects of Syntactic Simplification and Repetition on Listening Comprehension». TESOL Quarterly 26 (1992): 767-70.

     Chiang, Chung Shing, and Patricia Dunkel. «The Effect of Speech Modification, Prior Knowledge, and Listening Proficiency on EFL Lecture Learning».TESOL Quarterly 26 (1992): 345-74.

     Herron, Carol. «An Investigation of the Effectiveness of Using an Advanced Organizer to Introduce Video in the Foreign Language Classroom». Modern Language Journal 78 (t994): 190-98.

     Hudson, Thom. «The Effects of Induced Schemata on the 'Short Circuit' in L2 Reading: Non-decoding Factors in L2 Reading Performance». Language Learning 32 (1982):1-31.

     Interagency Language Roundtable. Interagency Language Roundtable Skill Level Descriptions. Princeton: Educational Testing Service, 1982.

     Kelly, Peter. «Lexical Ignorance: The Main Obstacle to Listening Comprehension with Advanced Foreign Language Learners.» IRAL 29 (1991):135-49.

     Lund, Randall J. «A Comparison of Second Language Listening and Reading Comprehension». Modern Language Journal 75 (1991): 196-204.

     Mecartty, Frances. «Lexical and Grammatical Knowledge in Reading and Listening Comprehension by Foreign Language Learners of Spanish». Diss. U of Illinois at Urbana-Champaign, 1993.

     Mueller, Gunther A. (1980). «Visual Contextual Cued and Listening Comprehension: An Experiment». Modern Language Journal 64 (1980): 335-40.

     Murphy, John Merton. «The Listening Strategies of English as a Second Language College Students». [328] Research and Teaching in Developmental Education 4 (1987): 27-46.

     Omaggio, Alice C. Teaching Language in Context. Proficiency-oriented Instruction. Boston: Heinle and Heinle, 1986.

      O'Malley, J. Michael, Anna Uhl Chamot, and Lisa Küpper. «Listening Comprehension Strategies in Second Language Acquisition». Applied Linguistics 10 (1989): 418-37.

     Rubin, Joan. «Improving Foreign Language Listening Comprehension». Georgetown University Round Table on Languages and Linguistics 1990: Linguistics, Language Teaching, and Language Acquisition: The Interdependence of Theory, Practice, and Research. Ed. James A. Alatis. Washington, DC: Georgetown UP, 1990. 309-16.

     Secules, Teresa, Carol Herron, and Michael Tomasello. «The Effect of Video Context on Foreign Language Learning». Modern Language Journal 76 (1992): 480-90.

     Shohamy, Elana, and Ofra Inbar. «Validation of Listening Comprehension Tests: The Effect of Text and Question Type». Language Testing 8 (1991): 23-40.

     Stahl, Steven. «Differential Word Knowledge and Reading Comprehension. «Journal of Reading Behavior 15.4 (1983): 33-50.

     Taglieber, Loni K., Linda L. Johnson, and Donald B. Yarbrough. «Effects of Pre-reading Activities on EFL Reading by Brazilian College Students». TESOL Quarterly 22 (1988): 455-72.

     Vandergrift, Laurens. «The Comprehension Strategies of Second Language (French) Listeners». Session on Research on L2 Listening. The Third Conference on the Relationship between Second Language Acquisition and Foreign Language Learning. West Lafayette, 27 February, 1993.

Appendix A

Vocabulary Preview Activity

Appendix B

Multiple-Choice Comprehension Task

     Circle the response which best completes [329] each statement. Some of the distractors are quite plausible, but when answering, think back to what was actually stated in the passage you just watched.

1. Videotapes __________ interviews as a means of evaluating job candidates.

2. Given the short length of the typical videotape, job candidates should concentrate on __________.

3. For prospective employers, one specific advantage of using videotapes is that __________.

4. Videotapes allow the job candidates __________ the evaluation process.

5. Videotapes do not provide prospective employers any information regarding the candidates' __________.

6. For job candidates, one specific disadvantage of using videotape is that __________.

7. Ensuring a high quality videotape often demands __________.

8. The job candidates should __________ when determining the content of the videotape.

9. For evaluating job candidates seeking executive positions, videotapes are __________.

10. According to the speaker, __________ must make the final decision regarding whether or not a videotape is an appropriate means of evaluating candidates for a particular position.

Media/Computers

Prepared by Richard A. Raschio, Mark D. Larsen

Reviews

     Isabel Allende: The Woman's Voice in Latin American Literature is a thoughtful, generally well-constructed, high quality video. Bettina Gray serves both as narrator and interviewer, leading the viewer through the historical background necessary to understand Allende's political views, exile, and, most importantly, her narrative work. Allende, however, is the real star, who takes center stage from the opening moments of this 56 minute video. Her speech, body language, and presence energize the scene and capture the full attention of the informed viewer. Despite the brutality of much of her subject matter, Allende's optimism and enthusiasm for life are infectious. It is a special delight to hear her talk about her relationship with the characters she has created.

     Allende's own experiences, those of the people closest to her, and the general political history of her continent inform her perspective on war and violence. She seems to have a personal, universal philosophy that is at once profoundly sacred and refreshingly practical. She refers explicitly to the inherent goodness or evil in all human actions and believes that individuals are accountable for them. One can imagine Allende herself practicing «random acts of kindness and senseless acts of beauty» on a regular basis.

     There are provocative resonances of the narratives in her speech, and the video is especially effective when Allende discusses her personal experience of writing. She is most eloquent when discussing the power of language and literature, the «universal matter» which has the ability to effect change. Allende's conviction that she writes under a socio-political, moral and artistic imperative is so intense that the viewer is convinced that this is a woman who has no choice but to write.

     Viewers receive a wealth of information about the author, although she freely admits to inventing many of her memories and much of her past. While we get a real sense of her devotion to her family, we are left disappointed regarding details of her personal life in her years of exile before settling in the United States.

     Informed viewers will probably appreciate this video the most. It is clear that the interviewer is very familiar with Allende's works and viewers who have not read the narratives, particularly Eva Luna, might feel at a loss. Time has addressed some of the questions that remain only partially answered in the video, e.g., how Allende adapted to life in California, and how this has affected her creative output. Followers of Allende's work will note, for example, that she wrote her latest novel, The Infinite Plan, in California with a setting in the United States. It is worth remarking that the video's title is somewhat misleading. While the title alone might draw the attention of Women's Studies specialists, Isabel Allende: The Woman's Voice in Latin American Literature is not really a piece about women's issues and voices, although they do constitute one of many threads woven into the discussion. In fact, these topics are not even touched upon until the interview is well underway. The video is thus not the best vehicle for communicating Allende's widely-regarded feminist perspective, although it is certainly a supportive example of her posture. Bettina Gray ends the video with a more appropriate title -«A Conversation with Isabel Allende».

     While the video ends on a high note, with Allende communicating her advice to anyone who loves to do anything («Just do it!»), viewers are left with the distinct impression that the interview itself had not really [331] ended. The abrupt termination seemed, unfortunately, to be more the result of poor editing and a concern for running time than it did for preserving the integrity and coherence of the piece.

     This film is certainly worthwhile for students of Latin American narrative in general and to fans of Allende in particular. Since the interview is in English, it is not of particular use in the second language classroom. Allende's comments on women's writing and issues, which comprise no more than five minutes of the video, might also be of interest to Women's Studies specialists who would not need to rely on this material as a primary source.

     For colleagues who enjoy varying their teaching tools and try to involve as many of their students' senses as possible in the learning experience, Español Level Ia may be a wish come true. A component of The Rosetta Stone Language Library, Español Level Ia is a multimedia package with full documentation and an attractive format.

     To install and run the program, users need a color monitor, a Macintosh with a hard disk drive, and a CD-ROM player. If the Macintosh runs System 6.0, it needs at least 2 megabytes of random-access memory (RAM); for System 7 and beyond, 4 megabytes are the minimum requirements. To use the program's recording feature, one also needs a microphone. (For those interested in the technical background of the software, the programmers developed it with Macromedia's Director and Director Player for use with Windows).

     The full package includes detailed instructions for installing and running the software, a user's guide, a textbook, and a CD-ROM disc. The spiral-bound textbook is available in both Spanish and English versions, and contains primarily the script of the disc's contents: 92 chapters divided into eight units. Additionally, the textbook includes a practice section for each chapter and a glossary. The documentation is impressively complete, easy to follow, and adds to the professional feel of the package.

     Español Level Ia provides practice in a variety of delivery formats which incorporate illustrations, text, and audio, and organizes vocabulary and structures equivalent to the first year of instruction in discrete learning chunks. The first lesson, for example, includes only the following nouns: niño, niña, perro, gato, pelota, caballo, avioneta, hombre, mujer, the software then presents the nouns in phrases using encima de, y, debajo de, and con. Audio prompts or written instructions guide users to click on a photograph or graphic that represents the appropriate word or phrase.

     Although presentation of the material is sequential, users are free to explore topics according to their personal needs and interests. Verb structures include the present progressive and the preterit, imperfect and perfect tenses of the indicative. Vocabulary varies, from the basics to the lexicon of several professional fields. Practice of idiomatic expressions focuses on typical problems for English-speaking learners.

     The package provides only limited feedback: a few subdued bars of a melody and either a happy face (for correct responses), or one crossed with a diagonal bar in a circle (for incorrect ones). Scoring is unrestrictive, i.e., the software only displays scores for information and the user can continue in the lesson regardless of performance. One might question whether such minimal reinforcement is pedagogically sufficient. On the other hand, audio prompts and reproductions of user-recorded sound are crisp and clear, and thus useful for improving aural and oral skills.

     The Rosetta Stone Language Library's approach helps the learner build confidence by providing plenty of opportunities for practice and a variety of techniques in an individualized mode. The product description [332] from Fairfield Language Technologies indicates that the user does not require previous exposure to the foreign language to benefit from the lessons. However, the scope and sequence of the contents also build on a learner's existing skills and helps expand previously acquired abilities. Furthermore, the screen presentations are superior motivators for students who may need encouragement. Due to the series' approach, Español Level Ia can be helpful for review and independent study, and the learning activities in this package definitely transcend typical off-the-shelf reference programs for tourists and hurried business people.

     Español Level Ia could not replace teachers as facilitators of learning or as the best models to emulate, but it can enhance their instruction with a variety of resources that they could only gather and incorporate into their lessons with inordinate amounts of time. It can therefore be a useful alternative to remedial instruction courses or a potentially excellent supplement to traditional delivery techniques.

     Thanks to its accuracy, completeness and ease of use, Español Level Ia is a welcome, practical application of current research in multimedia-based instruction. Indeed, the only weakness I found was one typographical error in the table of contents!

     Para entrar textos en la tan pregonada «supercarretera informacional» llamada World Wide Web (W3), éstos tienen que estar en un formato estándar que pueda ser leído por Mosaico, la interfaz más común del Web. El formato, llamado HTML (Hyper Text Markup Language), consiste en códigos que señalan elementos estructurales de un documento (párrafos, títulos, nexos con otros documentos, imágenes) y tipográficos (negrilla, itálica, marcas diacríticas) que de otro modo pueden perderse en la transmisión electrónica de datos. HoTMetaL es un programa «editor» que permite crear documentos ya habilitados para la distribución vía el Web. Entre otros usos, los profesores pueden aprovechar este medio para comunicar asignaciones, ejercicios e información a sus estudiantes.

     HoTMetaL es el primero de este tipo de producto, que promete proliferar en el próximo futuro como un componente de los tratamientos de texto más populares. Otro, llamado Internet Assistant, para distribución sin costo a usuarios registrados de Microsoft Word 6.0, fue anunciado en Info World el 28 noviembre último (29).

     Hay algo paradójico y desconcertante en el concepto de «editar» textos en la pantalla para casi imposibilitar su lectura. Tal es la tarea que lleva a cabo HoTMetaL, pero es sólo temporal la transformación, el paso necesario para poder pasar un texto, conservando toda su forma original, al Web, donde Mosaico lo exhibe en la pantalla de una forma perfectamente normal y legible.

     Los códigos de HTML proceden de SGMI, (Standard General Markup Language), que a su vez tuvo antecedentes en los códigos de identificación que ya publicó en 1987 la Universidad de Chicago en su manual Chicago Guide to Preparing Electronic Manuscripts for Authors and Publishers, como los que se dan a continuación:

     Es fácil olvidar el segundo código de los que exigen pareja, y los códigos son muchos, por lo cual es aconsejable acudir a un programa editor que los inserta y que verifica la corrección del formato HTML para que el documento no sea rechazado por Mosaico.

     La figura 1 enseña la pantalla en HoTMetaL para un documento sencillo creado con el programa que consiste en identificación, título, dos encabezamientos y varios párrafos con texto que contiene caracteres con marcas diacríticas. Para facilitar la creación de documentos, el programa incluye plantillas que proporcionan estructuras básicas dentro de las cuales se inserta el texto. Se utilizó una de ellas al crear este ejemplo.

Figura 1

     Lo ideal sería que los códigos de HTML se prestaran a otros usos más corrientes, como la comunicación de texto ASCII y el envío de mensajes por correo electrónico, pero comparados a los códigos tan sencillos creados por Larsen (reproducidos en esta sección), resultan muy torpes, en particular para la representación de marcas diacríticas. En fin, fuera del mundo de Mosaico, por ahora HTML no parece un vehículo adecuado para otros tipos de aplicaciones.

     El manual invita sugerencias (7) para futuras revisiones, y algunas son muy necesarias. La documentación podrá satisfacer a programadores y técnicos, pero está hecha a la inversa de las expectativas del usuario, cuyo deseo -y lo dice su designación como «usuario»- es usar el programa. El manual detalla más que nada lo que hace el programa y no lo que hace el que lo maneja. El que no conoce un programa se beneficia más de documentación organizada a base de tareas que de la descripción del logical y menúes. Definitivamente, debe haber un tutorial.

     El lector llega a saber primero que el logical sirve para editar materiales para W3 (9), pero hay que entender que para los efectos de este programa «editar» significa «crear» y no editar documentos ya hechos. Dada la sofisticación y comodidad de los tratamientos de texto, me parece mucho más probable que una persona prefiera editar documentos ya creados con su programa favorito y luego emplear HoTMetaL para convertirlos a HTML. El manual apenas reconoce esta posibilidad y simplemente alude de paso en el breve apartado que menciona la capacidad de importar archivos «not in HoTMetaL at all» a través de un filtro (31). Menciona una compañía que distribuye tales filtros (sin mayor información) y que se incluye uno llamado «TIDY» que supuestamente sirve en el 90% de los casos para documentos llamados de «legado» (no creados con esta versión de HoTMetaL). Es decir, el programa parece apto para crear nuevos documentos pero no para transformar texto ya existente.

     Esta poca atención al uso del logical para textos existentes conduce a mayores defectos como la omisión de instrucciones de qué tipo de texto es capaz de ser importado. Resulta que es ASCII, pero el usuario que no tiene considerable experiencia en la informática no tiene modo de saberlo.

     No hubo problema en importar un pequeño archivo en español en ASCII a través del filtro TIDY, pero encontré espacios en el lugar de las letras extranjeras de ASCII [334] alto, enumeradas de 182 a 256. Al entrar vocales acentuadas desde el teclado, se reproducen en la pantalla los códigos correspondientes en cuadritos -«aacute» y «ntilde», por ejemplo- pero queda después un espacio que hay que cerrar. Otros caracteres, como las vocales que en portugués llevan tilde, se consiguen por «Markup/ Insert character entity», que presenta una tabla de 90 caracteres de los llamados ISO 8859-1. Los puntos de interrogación y de exclamación inversos aparecen en la pantalla como números de ASCII (#191 y #161). La figura 2 enseña los resultados de la importación después de reemplazar los espacios con los caracteres especiales del español.

     La conversión de un texto importado trajo varios problemas no previstos en el manual. Al querer corregir un error, no encontré manera de sacar el número equivocado sino acudir a Edit/Cut o bloquear el área primero y suprimirla. Para consignar el fin de línea para cada entrada ya existente hubo que bloquear y luego insertar el código P (párrafo) para cada nueva línea por medio de las siguientes selecciones de menú: Alt M(arkup)/S(urround)/P(aragraph), y «Enter».

     Las instrucciones tampoco aclaran que básicamente el programa es capaz de dos métodos distintos de edición: (1) escoger primero el elemento adecuado del menú para insertar el par de códigos correspondientes y luego entrar el texto o (2) escribir el texto primero y rodearlo después con los códigos.

     Para crear un documento se puede usar una de las plantillas o el usuario puede escoger «New» e insertar el código HTML para empezar un documento de su propio diseño. Si el documento no concuerda con las reglas, si falta el segundo código de una pareja, por ejemplo, aparecerá después un mensaje y el programa impide guardarlo.

     El programa se ha diseñado con cuidado, con detalles muy útiles. Se puede esconder los códigos en la pantalla, para repasar el texto en una forma que se puede leer sin estorbos. También hay funciones para deshacer una operación ya ejecutada («undo»), verificar ortografía (por ahora sólo en inglés) y ver la estructura jerárquica entera de un documento.

     Un profesor podría usar el programa para crear documentos breves para colocar en el Web para uso de sus estudiantes, pero para textos largos, sería más cómodo y rápido crearlos con tratamiento de texto y pasarlos a HTML con un logical más sofisticado, tipo traductor, que pudiera convertirlos con poca intervención del autor. Tales programas están aún en vías de desarrollo, pero su precio promete ser el doble del de HoTMetaL, por lo cual este logical ofrece una buena pero limitada alternativa.

Indice de Hispania convertido con HoTMetaL (recuperado con tratamiento de textos)

Pedagogy: Colleges and Universities

Prepared by Dolly Young

Raising General Awareness of Language Learning Strategies: A Little Bit Goes a Long Way

Jeffra Flaitz, Carine Feyten, Sallie Fox, and Keya Mukherjee

University of South Florida

Abstract: While researchers and teachers have observed the use of language learning to produce a positive effect on student achievement, it is still unclear how best to proceed in training students in the effective use of these strategies due to such constraints as insufficient room in the curriculum and lack of teacher training in the area of language learning strategies. In this study, an experimental group of 130 students of college-level Spanish I and II received Metacognitive Awareness Raising (MAR) -a single 50-minute session which, by means of brainstorming and jigsaw activities, dynamically involved students in developing a general, overarching awareness of language learning strategies rather than exposing them to the selection, presentation, and practice of a set of specific strategies over a period of weeks or months. Final course grades were compared across control and experimental groups, revealing significantly higher scores for experimental group subjects.

Key Words: strategy training, language learning strategies, strategies, SLA, awareness raising, Spanish, metacognition

Introduction

     The purpose of this study was to investigate the effect of heightening learners' general awareness of language learning strategies on student achievement in beginning and intermediate college-level Spanish courses. The authors distinguish here between the process of raising general awareness and that of strategy training which involves a greater commitment of time and focused attention on specific strategies.

     Much research on language learning strategies has established their role in making language learning more efficient and effective (Chamot 1993; Cohen 1990; O'Malley and Chamot 1990; Oxford 1990, 1993; Oxford and Crookall 1989; Rubin 1975; Wenden and Rubin 1987). Oxford (1990: 1) defines learning strategies as «steps taken by students to enhance their own learning». According to Oxford, students in the process of language learning can make particular use of strategies as «tools for active, self-directed involvement». The natural outcome of this research has been a proliferation of studies examining the effect of strategy training on student performance (Brown et al. 1983; Chamot and Kupper 1989; Hosenfeld et al. 1981; O'Malley et al. 1985; Oxford 1989, 1990; Oxford et al. 1990; Pressley et al. 1990; Pressley and Harris 1990; Russo and Stewner-Manzanares 1985; Weinstein and Mayer 1986; Weinstein et al. 1988).

     Yet, in the present pedagogical context of crowded curricula, high teacher-student ratios, and the demands of a communicative approach to foreign language teaching, teachers often perceive strategy training of any kind as an extra burden (Chamot 1993). Moreover, many teachers are uncomfortable insofar as they themselves have never been trained in the use of language learning strategies. Nevertheless, if learning strategies are central to language learning, if there is a possibility that awareness raising does lead to increased performance, and if teachers are indeed pressed for time or if they lack the training to help their students develop facility with strategies, then it is imperative that researchers investigate the effect of heightening learners' general, overarching awareness of language learning strategies through less time-consuming and more manageable alternatives. [338]

     The mandate for strategy training poses another kind of problem, namely that while there are indications that strategy training in the language learning classroom can lead to greater achievement, by and large the research community has not as yet proven that there is a consistent positive effect. For example, in a study by O'Malley et al. (1985), learners benefited from strategy training in the areas of speaking and, to some extent, in listening, but only the Hispanic students in the sample gained in the area of vocabulary enhancement. Often research design flaws are cited as critical factors in producing contradictory or inconsistent results. Typical problems, according to Oxford (1993), include strategy training that is too brief, not well integrated into the pedagogical task, too easy or too difficult or premature (i.e. poor pretraining assessment of strategy use among learners). Oxford also argues that most strategy training efforts are limited in their focus on the cognitive components (involving such mental operations as reasoning, analyzing, problem solving, practicing, etc.) and metacognitive components (including planning, self-monitoring, and self-evaluation). As a result, researchers have paid little attention to the development of the affective and social domains of strategy training. Affective strategies are focused inward and encourage students to consider their motivations, emotions, and attitudes about their learning (Oxford 1990: 140). Social strategies, on the other hand, involve students in communication through asking questions, cooperating with peers, and developing cultural awareness (Oxford 1990: 145). Despite the limited research in these areas, there remains a strong sense, though not borne out empirically by all studies, that strategy training has the potential to play a critical role in language teaching and learning.

     In order to develop that potential, however, researchers must arrive at an effective means by which to deliver strategy training. Should it be short-or long-term? Blind or informed? Detailed or general? Most strategy training projects have focused on short-term delivery since long-term projects are often plagued by practical impediments. In turn, such studies often produce tenuous or marginal results, leading researchers to suggest that the duration of the training is at fault, and recommending sessions that continue for up to and even over a year.

     As concerns the explicitness of the training, Wenden (1986) cited evidence which suggests that students benefit from conscious awareness of the purpose, nature, and significance of strategy training. She pointed out the superiority of «informed» versus «blind» training. In informed training facilitators emphasize the need, usefulness, and anticipated benefits of a given strategy over the means by which it is to be implemented. In blind training, on the other hand, facilitators embed strategies into tasks and materials without informing students how to use, apply, or transfer them. Therefore, while blind training can promote success on individual tasks with which the strategy is associated, research (Brown et al. 1983) reveals that learners are unable to generalize beyond the set task and they eventually abandon the strategy. It appears, then, that learners benefit from developing an appreciation of the overarching effects of strategy training. Scholars in a number of other disciplines such as counseling psychology, communication science, and education (Curtis 1986; Devine 1978; Feyten 1991; Nichols and Stevens 1957; Rankin 1930) have also at least anecdotally recognized the positive influence produced by awareness raising in relation to specific problems or issues. Indeed, the twelve-step programs used to treat addiction and other psychological and physical disorders operate on the premise that awareness is the first step toward recovery.

     The problem with detailed, comprehensive training, in which learners are exposed to the full range of strategies, is that not all learners will benefit from each and every strategy, and they end up being overwhelmed and thus frustrated by overchoice (Chamot 1993). Moreover, individual learners tend to prefer certain strategies over others, making the task of accommodating an entire classroom of students an unwieldy [339] task for the teacher. The solution may not lie in paring down the selection of strategies or in proceeding through the tedium of matching an individual learner with a tailored set of strategies. Given the lack of opportunity to provide full-scale strategy training in our less-than-ideal classrooms, it may be useful to raise students' general awareness of language learning strategies, thus improving their attitudes, motivation, and beliefs about language learning, all of which fall within the socio-affective domain of learning strategies so often neglected by researchers (Oxford 1990, 1993).

     The authors of this article have specifically focused their efforts on examining the effect of metacognitive awareness raising (MAR) of language learning strategies on student achievement in first and second semester college-level Spanish courses (Spanish I and II). MAR is defined as the process of heightening learners' general awareness of language learning strategies through the administration of a one-time, fifty-minute session which includes interaction with the material, active involvement of students, use of higher order thinking skills, and accommodation of students' social and affective needs. Keeping this approach in mind, the following research questions were critical to the study:

Method

     Sample. The sample for the present study represented a subset of a population of students enrolled in Spanish classes at the University of South Florida. Twelve sections of Spanish were randomly selected out of a total of 24 sections, with roughly equal sections of Spanish I and Spanish II being represented in the sample of 229 students. These twelve sections were subsequently randomly divided into six experimental and six control groups (Table I).

     All subjects provided information on six demographic characteristics, including gender, age, academic level, native language, previous language study, and non-academic language study. Table I reveals similar distribution of students in the experimental and control groups along the afore-mentioned characteristics. Moreover, since the university imposes a language requirement for graduation, the sample is representative of the university student population at large. However, due to the high demand for admission into beginning language classes, the program cannot accommodate all students and therefore gives priority to seniors and juniors, hence the predominance of upperclassmen in the study.

     Table I also indicates that 62.7% (N=83) of the students in the sample (N=229) had previously studied Spanish. This reflects the distribution of subjects into Spanish I and Spanish II courses. Researchers expected an examination of the achievement of students in each of these levels to provide insight into the possibility of differential performance between the two. In other words, would the metacognitive awareness raising sessions prove inconsequential to students with previous successful (in the sense that they had advanced to Spanish II) language learning experience?

     Materials. The present study made use of a wide variety of materials (see appendices) during the different stages. All subjects received a demographic questionnaire, and their final course grades were tabulated at the end of the semester. Only the control groups received a placebo questionnaire relating to myths about language learning. This questionnaire was a Likert-scaled instrument attempting to ascertain students' [340] beliefs about and attitudes toward language learning.