Web-based Foreign Language Reading: Affective and Productive Outcomes*

Kerstin Lück
University of California, Davis

ABSTRACT
This study aimed to investigate whether pedagogically guided web-based reading can improve skimming and scanning significantly (i.e., increased productive outcomes) and whether it can enhance student participation and motivation (i.e., increased affective outcomes). Forty-six students enrolled in two German 3 classes at the high school level were selected. The participants read texts about German culture and politics. Qualitative and quantitative findings indicated that the students in the treatment group (Group A) increased their skimming and scanning performance significantly compared to the stu¬dents in the control group (Group B). Furthermore, the students in Group A were able to foster their reading skills in such a way that they were able to use it with linear materials as well. Besides skimming and scanning, the students also increased their participation and motivation. Although the participants in Group B made some progress in the fol¬lowing research period, their achievement remained significantly lower than that of the participants in the treatment group. These findings not only make clear that pedagogi¬cally guided web-based instruction has a positive impact on foreign language reading skills, participation, and motivation, but they also support the argument that a variety of methods and approaches should be used and that traditional approaches alone should not dominate in foreign language instruction.
KEYWORDS
German, Web-based Reading, Skimming and Scanning, Student Participation and Motivation

INTRODUCTION
Most L2 expert educators and researchers agree that computer-assisted language learning (CALL) can enhance the acquisition of foreign language skills. The use of the web for pedagog¬ically guided authentic foreign language reading is especially widely accepted (Levine, Ferenz, & Reves, 2000; Brandl, 2002). Khan (1997), Hancock (1999), and Gambrell (2005) argue that the web is revolutionizing the acquisition of L2 by giving learners access to an unlimited database of authentic materials. Current documents, papers, virtual books, and even new discoveries from around the world can be accessed. All this corresponds well to the ideal of a foreign language student as researcher and self-directed learner (Lemke, 1998; Calderon-Young, 1999; Kramsch, A’Ness, & Lam, 2000; Bussière, 2004).
In order to take advantage of these sources of authentic materials, databases, and ideas from the foreign culture, students have to use effective reading strategies, such as skimming and scanning, to comprehend L2 texts, or, as Kramsch (1993) points out,
*This article is dedicated to the Fulbright Teacher Program and to L2 teacher Jodie Depke.
CALICO Journal, 25 (2), p-p 305-325. © 2008 CALICO Journal
CALICO Journal, 25 (2) Web-based L2 Reading: Affective and Productive Outcomes
306
everyday texts of information require readers to adopt the communicative reading strategies of native speakers: Skim and scan for desired information, capitalize on the natural redundancy of a text and get clues from its context, recognize authorial intention and act upon it. (pp. 177-178)
However, educational practice shows that L2 students often approach paper-based linear texts with ineffective reading strategies, despite their theoretical knowledge about valu¬able techniques such as skimming and scanning. On the other hand, the nature of nonlinear texts and effective web-based instruction encourages skimming and scanning (Nielsen, 2000). Therefore, it is the aim of this study to investigate whether pedagogically guided web-based reading can improve skimming and scanning significantly and whether it can enhance student participation and motivation.
Web-based Reading
Hypertext pioneers like Vannevar Bush (1945, 1967) have already predicted that the us¬age of information technology would soon redefine the nature of text and reading. In an article in the Atlantic Monthly in 1945, Bush prognosticated that electronic text would make it possible to escape the monotonous linear approach of traditional reading. Those electronic texts would allow readers to explore a universe of discourse. Other scholars referred to an electronic hypertext that would interconnect all of the world’s literature, so that researchers, teachers, and students could have immediate access to any of the world’s stored texts in dif¬ferent languages. These thoughts and ideas are becoming commonplace. In most developed countries, the majority of reading on politics, economy, education, and social life takes place on the computer (Castells 1996, 1998; Castells & Catterall, 2001). Although reading on the computer has not replaced books or readers, it is crucial to point out that the computer has already joined them as a major form of literary activity (Warschauer, 1999; Sutherland-Smith, 2002; Blanchard, McLain, & Bartshe, 2005).
However, educators of information technology still have divergent opinions as to whether reading on the screen should be promoted or rejected (Warschauer, 1999). Scholars such as Davis and Lyman-Hager (1997), Sutherland-Smith (2002), Horning (2002), McNabb, Hassel, and Steiner (2002), and Martin (2003) support web-based reading. They argue that electronic texts incorporate broader and more flexible ways to present and access informa¬tion. Furthermore, Spiro, Feltovich, Jacobson, and Coulson (1995) found that cross-linked hy-pertext can promote a degree of cognitive flexibility in a field that would otherwise take many years of traditional reading to acquire. In the view of its academic proponents, web-based instruction also facilitates dynamic approaches to reading and fosters critical thinking as well as problem-solving skills (Levine, Ferenz, & Reves, 2002; Horning, 2002; Thompson, Martin, Richards, & Branson, 2003). Students do not just have access to well structured knowledge, but they also encounter content on the web that is ill structured and complex where they have to abstract to higher levels of thinking or may discover multiple ways of evaluating a text. Students further acquire rich semantic networks of information and the ability to structure and restructure new and prior knowledge in activities designed to anticipate the demands of variable and changing situations encountered in the online environment.
Other experts, for instance Talbott (1995) and Campbell (1998), see web-based read¬ing as a major disadvantage in language education, with readers surfing through appealing but often irrelevant materials, never pausing long enough to study specific topics intensively. Campbell (1998) argues that web-based reading might encourage ‘hyperthinking.’ Whereas rational thinkers approach texts in a logical and systematic way, hyperactive thinkers are only 307 CALICO Journal, 25 (2) Kerstin Lück
“devoted to stimuli that are novel and eye-catching rather than thought-provoking” (p. 24); they have poor productive and receptive communication skills, a lack of metacognitive abili-ties and they are unable to handle multiple sources of information.
Although some ideas brought forth by Campbell are too single sided and to a certain extent even inconsistent, it should be taken into account that some of the arguments con¬cerning the use of web-based reading in the classroom as an alternative to traditional instruc¬tion are similar to those discussed by other experts. Birkerts (1994), for example, argues that the web could destroy a learner’s ability for deep, reflective reading.
There are also some other problems associated with reading activities on the web. One of the major points of criticism of the use of the web for reading and research purposes lies in the fact that some of the materials on the web are superficial or unreliable (Cerf, 2003; Ricketts & Zakrzewski, 2005). Therefore, Cerf argues that “[w]e truly must think about what we see and hear. We must evaluate and select. We must choose our guides“ (p. 7). Spiro et al. (1995) also point out that it is crucial not “to connect everything with everything else” (p. 96) and we have to provide guidance to the students (see also Burke, 2000; Sutherland-Smith, 2002; Forbes, 2004; Krymes, 2005).
Web-based reading differs substantially from traditional classroom instruction. In the traditional classroom, teacher activities often exceed student participation. The teacher large-ly determines the use of the class time and typically focuses almost all student attention on the textbook (Relan & Gilliani, 1997; Brandl, 2002). In contrast, activities on the web can require a tremendous amount of student activity, and the teacher often just guides students in their efforts, offers feedback to them, and helps them to find solutions. Group projects in which students work in research teams are encouraged. Thus, web-based reading encour-ages collaboration and serves as a platform for the expression and contribution of cognitive meanings and understandings (Trollope, 1995; Osuna & Meskill, 1998; Kramsch et al., 2000). Thus, the web promotes student-centered interaction. Finnemann (1996) and Brandl (2002) refer to this student-centered nature of the web but also discuss its teacher-centered use in the L2 classroom. Students can appreciate the open-ended system inherent in the web, its virtual cultures, and its nondiscriminatory environment (Finnemann, 1996; Khan, 1997; Owston, 1997; Owston & Wideman, 2001). In addition, instructors can take advantage of the vast library of available materials and information on the web for use for the L2 classroom (Finnemann, 1996; Brandl, 2002). How to take advantage of those materials and how to inte¬grate them into teaching is well explained in Warschauer´s (1997a) guidelines for web-based learning activities. The goals of web-based reading activities should be carefully considered, and online reading activities should be integrated into the course rather than adding them on top of other classroom activities in a disconnected fashion. Furthermore, instructors should not underestimate the complexity of web-based reading and should provide necessary sup¬port (Warschauer, 1997a).
Skimming and Scanning
Skimming involves searching for main ideas in the foreign language by reading, for example, the first and last paragraphs and noting summaries, conclusions, and suggestions by the au-thor. It is important to teach beginning L2 students this reading technique because otherwise they react by shifting their reading speeds to a very low level, processing material word by word. To master a text in a foreign language, students must thoroughly understand the major ideas and concepts presented.
Scanning is a very effective technique to locate specific information in a text. Scan¬CALICO Journal, 25 (2) Web-based L2 Reading: Affective and Productive Outcomes 308
ning involves running ‘one’s eyes down the page,’ looking for important facts or key aspects. Scanning can also be an aid in locating new terms introduced in a text on a webpage. Unless the students understand new vocabulary and structures, it is impossible to follow the author’s reasoning.
Some scholars such as Nielsen (2000) and Morkes (1997) argue that skimming and scanning are most natural to the web, and webpages are often designed accordingly. How¬ever, it can be also argued that this is not always the case because skimming and scanning has more to do with the genre of a text and with the reading tasks and aims than the medium in which the text is presented. There is little point in skimming and scanning Goethe’s Faust, even if it is placed on the web. However, when it is a question of specific texts on culture, politics, or economic affairs, which are usually packed with many facts and also unimportant details, the use of skimming and scanning becomes crucial.
Nevertheless, Nielsen’s (2000) findings shed some light on the issue of the natural¬ness of skimming and scanning on the web. First, it should be noted that we rarely find dra¬mas and entire novels on the web, but the web does contain a lot of information on German culture, politics, and so forth. Thus, there is a much better selection on the web than in the textbooks or other books available for use in the classroom when it comes to materials that lend themselves to the development of scanning and skimming skills. Second, it is easier to find materials related to a given topic on the web (e.g., by means of links or search engines), allowing users to read a webpage written in a simpler version of the target language before working on a more challenging webpage. Third, a lot of young people spend more of their leisure time reading materials on the web than reading books. Working on reading techniques in an environment that is similar to that of their leisure time activities can be an advantage. In addition, the availability of authentic materials and choice of texts make learning activities more interesting (Burke, 2001; Blanchard, McLain, & Bartshe, 2005).
The Role of Motivation and Participation
Many researchers argue that there is a need to reexamine the relationship of motivation and web-based reading (Davis & Lyman-Hager, 1997; Kramarski & Feldman, 2000) because most of the previous studies are outdated or they are too general (Song, 2000; Chen & McGrath, 2003). Furthermore, it is also evident that the importance of affective factors with respect to computer-assisted reading has been underestimated in the past (Keller, 1999; Rosalia, 2002). However, web-based foreign language learning involves a variety of psychological, social, and cognitive factors which have a significant impact on student learning and achievement (Warschauer, 1996). Accordingly, there is a need to examine “specific aspects of computer-assisted language learning and student motivation” (Warschauer 1996, p. 29) and their inter-relationships. It is one of the goals of this study to extend previous research and investigate the nature of participation and motivation in connection to skimming and scanning through correlational and experimental approaches.
Statement of the Problem
The major research question of whether web-based reading promotes effective strategies such as skimming and scanning has not been previously investigated. In this study, all students read linear (i.e., paper-based) texts in the course textbook and a reader, and the participants in the treatment group also read some web-based texts. In addition, all students received in¬struction in the use of reading strategies. Even though the students received reading strategy 309 CALICO Journal, 25 (2) Kerstin Lück
instruction, the majority of them continued to try to translate the linear (paper-based) texts word by word. They often forgot what they had read in the beginning and did not have time to finish even half of the assignment. Therefore, the level of reading comprehension was low, concepts and general meaning were not always understood, and detailed questions were not answered in many cases.
It is the aim of this study to investigate the effect of web-based reading combined with an effective instruction on students’ skimming and scanning skills. It is hypothesized that foreign language students who work with linear (paper-based texts) and nonlinear texts (web-based texts) will score significantly higher on measures of skimming and scanning than those who work with linear texts only. It is further hypothesized that there will be significant positive relationships among skimming, scanning, participation, and motivation for beginning foreign language students who work with linear and nonlinear texts and those who work with linear texts only.
1. How do students who work with both linear and nonlinear texts react to the nonlinearity of texts on the web?
2. a) Do students focus more on general meaning or unnecessary details when reading nonlinear or linear texts?
b) Do they use skimming and scanning more often in nonlinear or in linear texts?
c) What is their opinion about the level of difficulty of German materials on the web?
3. What influences does web-based reading have on students’ participation and motivation, especially their motivation to read German materials during leisure time?
4. Is there a significant difference in skimming and scanning between the stu-dents who work with both linear and nonlinear texts and those who work with linear texts only in the research-in-progress and end-of-research pe-riod?
5. Are there correlations between: (a) skimming and participation, (b) skim-ming and motivation, (c) scanning and participation, and (d) scanning and motivation in students who work with nonlinear and linear texts and those who work with linear texts only at the end-of-research period?
PARTICIPANTS, METHODS, AND PROCEDURES
In connection with the qualitative and quantitative orientation of this semester long project, a variety of research methods was used to collect data. The major focus of these data collec-tions was the observation and study of the participants who read both linear and nonlinear texts (Group A) and those who read linear texts only (Group B) with a focus on skimming and scanning.
The participants in this study were 46 students enrolled in two German 3 classes at the high school level. The students in one class were assigned to Group A, and the students in the other class were assigned to Group B. Prerequisite to this German 3 class was German 2 or an equivalent German language background. All participants had at least basic computer skills, CALICO Journal, 25 (2) Web-based L2 Reading: Affective and Productive Outcomes 310
which were measured prior to this study (see background questionnaire in Appendix A). The participants read German fairy tales and texts about German culture and politics. They had a choice of several texts in these three genres.
The students in Group A worked with nonlinear and linear texts. They used their text¬book, Komm mit, and websites such as Logo-Kindernachrichten, Spiegel-Online with the ru¬brics Politik, Kultur, and Wissenschaft (related to culture, politics, and human relations only) (http://www.spiegel.de), Das Politische System der Bundesrepublik Deutschland (http://www.hkbu.edu.hk/~europe/polshyp/run.htm), and Regionalpolitikseite der Europäischen Kommis¬sion (http://ec.europa.eu/regional_policy/index_de.htm).
The students in Group B read linear texts only. They used their textbook, Komm mit, and a reader. The students in each group always worked in pairs. Both classes had the same instructor. Finally, all students had a similar achievement level in German, and there were no significant differences between the groups prior to this study.
In the preresearch period, background information was collected on the participants’ basic language skills, reading performance, computer experience, and level of motivation via a paper-based questionnaire. The background questionnaire was distributed to all participants who were requested to complete and return it prior to taking part in the research project. In addition, the participants’ latest L2 reading achievement in skimming and scanning and their level of motivation and participation during the performance of the reading task were mea-sured and analyzed. The students’ reading performance and participation was graded by the teacher and measured in scores from 100 to 0 (i.e., 100 to 90 = A, 89 to 80 = B, etc.). Stu-dent motivation was rated by the participants themselves and also measured in scores from 100 to 0 (i.e., 100 to 90 = very high, 89 to 80 = high, etc.).
Data on student reading performance and participation were also obtained through direct observation and classroom recording (all students were recorded during each ses¬sion), several interviews, questionnaires, and reading tests regarding their ability to skim and scan texts during the research-in-progress and the end-of-research period. The students were recorded (audio and video) while reading linear texts in the textbook or reader. Atten¬tion was paid to their reading performance and affective outcomes. All participants worked in pairs and used the “pair-talk-aloud” strategy (one group consisted of three participants). The pair talk-aloud protocol differs from the think-aloud protocol (Krymes, 2005) because “it is derived from a discussion rather than from the subjects being asked to verbalize their thoughts” (Trollope, 1995, p. 15). In addition to the video recording of the reading approaches (i.e., skimming and scanning), a microphone with video sound mixing was used to facilitate simultaneous recording of the participants’ talk onto the videotape. Altogether, the recording procedures allowed the researcher to collect substantial amounts of data from the talk-aloud protocols that reflected the students’ reactions.
While the students in Group A were engaged in the reading of nonlinear texts on the web, the researcher used video-capturing software to record the texts they were reading. It was therefore possible to observe what the participants had displayed on their computer screens and to take note of their reaction to the texts in a hypertext environment. The video¬taped pairs of participants were asked to use the “pair-talk-aloud” strategy to describe what they were doing at the particular moment and what their overall task was.
Student interviews and a questionnaire were used at the end of the project to collect supplementary data. The interview questions and questionnaire items focused on the follow-ing aspects:311 CALICO Journal, 25 (2) Kerstin Lück
1. self-evaluation (i.e., level of participation, motivation, skimming, and scan-ning);
2. students’ focus on general meaning of texts or on unnecessary details in the texts; and
3. (Group A only) students’ skimming and scanning of German texts on the web versus their approach to linear texts in the textbook/reader, their opin¬ion on the level of difficulty of German texts on the web versus those in the textbook/reader linear texts, their preference for linear or nonlinear texts.
During the research-in-progress and the end-of-research period L2 reading perfor¬mance was tested again. Participation and motivation data were also analyzed. In addition to introspective data, posttask data were also collected in order to implement additional fea-tures. Mertens (2005) argues that posttask retrospection is very important because it serves as a means of further investigating and clarifying crucial findings. The posttask research was completed after finishing the research project (i.e., during the last week of the semester).
DATA ANALYSIS
All interviews, observation notes, talk-aloud protocols, and recordings were transcribed and analyzed through discourse analysis. The questionnaire data were used to supplement the qualitative findings obtained during the classroom observations in all research periods. Quali-tative and quantitative data were always evaluated according to the next higher attainment levels. This means that an increase in the use of a particular reading technique reflects not merely progress with respect to previous results but also an increase with respect to higher academic demands set for the research-in-progress and end-of-research periods.
Information from the questionnaire items and interviews (all research periods) was analyzed in a descriptive way. The pretask, research-in-progress, and end-of-research data regarding the participants’ achievement were analyzed by Chi-square and Fisher’s Exact Test. Both tests were used to compare the achievement levels between the groups in order to ex¬plore any differences between Group A and B.
The end-of-research period data concerning the relationship of the variables skim¬ming-motivation, skimming-participation, scanning-motivation, and so on within each re¬search group were computed through correlation and regression analysis.
FINDINGS
Prior to this research study, the participants in both groups tended to focus more on unnec-essary details than on the general meaning of given texts. There were no significant differ-ences between these groups (Fisher’s two-sided test =.722; see Appendix B). The majority of the students had difficulties in acquiring the overall gist and context of their assignments. However, in the research-in-progress period the students in Group A already started to focus more on the general meaning when reading materials on the web. Classroom observations and tests supported this fact. During the research-in-progress period 15 out of 23 students in Group A were able to comprehend the general meaning of texts compared to 9 out of 23 students in Group B. However, there still was no significant difference between the groups CALICO Journal, 25 (2) Web-based L2 Reading: Affective and Productive Outcomes 312
(Chi-square = 3.14, p = .076, Fisher’s two-sided test = .139; see Appendix B). However, during the end-of-research period, Chi-square analysis showed a significant difference in the performance between the groups (Chi-square = 5.25, p = .022). Fisher’s Exact Test showed that the students in Group A outperformed those in Group B (i.e., left-sided Pr <= F = .996; right-sided Pr >= = F .024 with a Table Probability P = .020).
The semester long observation of Group A revealed that, when reading web-based materials, they tended to focus more on general meaning (skimming) than on unnecessary details and on finding particular information (scanning). Throughout this study the results of this group with regards to skimming and scanning increased to a higher degree than those of the students in Group B. In the end-of-research period there was a significant difference between the groups: Chi-square = 11.82 (p = .037) for skimming and Chi-square = 12.13 (p = .033) for scanning (see Appendix C). Seventeen students in Group A and 9 students in Group B achieved excellent, very good, and good results in skimming. The only students with poor results were those in Group B. With respect to scanning, 19 students in group A achieved excellent, very good, and good results compared to 11 students in group B. Furthermore, 7 students in Group B had scanning scores slightly below average or even very poor results, but no one in Group A had a scanning performance below average.
The positive test results in Group A were also supported by interviews with the par¬ticipants. When asked about reading materials on the web during the research-in-progress period, 15 students in Group A indicated that they were using skimming and scanning more effectively and to a larger extent. Only 3 students stated that they used skimming and scan¬ning more often in linear texts, and 5 students noted no difference at all (see chart in Appen¬dix D). Expressions coded as “skimming and scanning” included processes such as getting the gist, figuring out the overall meaning, getting the general idea, and finding particular informa¬tion.
Interviews with participants in Group A lend credence to these findings.
Student 1: I definitely use skimming and scanning more often … . I also … mm … focus more on general meaning than on details and on finding particular information that is relevant to a question.
Student 2: When I …, for example, … read something about the Green Party in German, there were always a lot of things I didn’t understand but … I got the gist of it on the web because of statistics and visuals on the internet … and helpful links … .
Student 3: I am kind of used to it … I often skim and scan … on the internet.
These sample statements show that students indeed skim and scan more often for in-formation in nonlinear texts. Most of the students in Group A also indicated that they always approached web assignments with a higher level of encouragement than other assignments. They found web exercises as an addition to other, more traditional exercises more motivating, and they developed a very open attitude towards German as a foreign language.
It was also interesting to find out that the students in Group A generally preferred to do their skimming tasks first (see transcript extract below). They affirmed that it was very important to them to understand the gist of a text at first and then seek specific details and key phrases by scanning.313 CALICO Journal, 25 (2) Kerstin Lück
Sample Transcript (Group A Students)
(The tasks preceding task 4 focused on scanning. Task 4 focused on skimming.)
Task 4
Transcript Video commentary
E(3): We should … mmm … wir machen task four zuerst …
‘… mmm … we do task four first …
F(4): ist gut , wenn man den gist kennt!
‘Yes, it is good if you know the gist.’
E(5): Hmm,… die Grünen sind durchaus eine wichtige Partei im Landtag und Bundestag … und sie setzen sich nicht nur für eine saubere Umwelt ein …
‘Hmm, … the Green Party is an important party in the Landtag and Bundestag … and they support not only a clean environment …’ -E is operating the mouse.
F(6): … ja, sie setzen sich für demokratische Politik ein. In der Partei … mhm … Look! … da sind Demokraten, Fe¬ministen und Umweltrechtler … mhm …
‘… yes, they support democratic politics. In the party … mhm … Look! … there are democrats, feminists, and environmentalists … mhm …’ -F is pointing at the screen.
…
F(8): Ach so! … Sieh mal … eine wichtige politische Auf¬gabe der Grünen ist neben der Umwelt die Beschäf¬tigungs-politik und Aktionen gegen Korruption und Mißwirtschaft.
‘I see! … Look here … a very important task of the Greens, besides the environment, are issues of em¬ployment and actions against corruption and misman¬agement.’ -F is taking the mouse and is pointing at key aspects.
E(9): Richtig … “und die Grünen fordern den europaweiten Eintritt in die Sozial-und Umweltunion.”
‘Right … ‘and the Green Party wants a Europe wide membership into a social- and environmental union’.’ -E is operating the mouse.
F(10): und sie fordern eine Progression in der wirtschaftlichen Integration Europas.
‘and they want a progression of the economic integra¬tion of Europe.’
…