Running head: COMPUTER ASSISTED READING
Improving Children’sReading Comprehension and Use of Strategies
through Computer-based Strategies Training
Yao-Ting Sung1, Jung-Sheng Huang2, and Kuo-En Chang3
National Taiwan Normal University, Taipei, Taiwan
1Yao-Ting Sung, Ph. D.
Address: Department of Educational Psychology and Counseling, National Taiwan
Normal University, 162, Sec. 1, Hoping E. Rd. Taipei, Taiwan.
Telephone: 011886-2-23952445 ext 509 E-Mail:[email protected]
2Jung-Sheng Huang
Address: Department of Educational Psychology and Counseling, National Taiwan
Normal University, 162, Sec. 1, Hoping E. Rd. Taipei, Taiwan.
Telephone: 011886-2-23622841 ext 52 E-Mail: [email protected]
3Kuo-En Chang, Ph. D.
Corresponding Author
Address: Department of Information and Computer Education , National Taiwan Normal University, 162, Sec. 1, Hoping E.
Rd. Taipei, Taiwan.
Telephone: 011886-2-23622841 ext 18 E-Mail: [email protected]
Paper submitted to: Journal of Experimental Education
09/25/2005
Improving Children’sReading Comprehension and Use of Strategies
through Computer-based Strategies Training
Abstract
In this study the
Attention-Selection-Organization-Inte gration-Monitoring (ASOIM) model, revised from Mayer’s(1996)SOI model of text comprehension, was used as a foundation to design a user-friendly, multi-strategy based Computer Assisted Strategies
Teaching and Learning Environment (CASTLE). CASTLE aims to enhance learners’abilitiesofusing reading strategies and text comprehension. The effects of CASTLE on students with different reading ability were
empirically evaluated. 130
sixth-graders took part in an 11-week computer-based reading strategies course. The results show that
CASTLE helpsenhancestudents’use of strategies and text comprehension at all ability levels.
Improving Children’sReading Comprehension and Use of Strategies
through Computer-based Strategies Training
Introduction
For enhancing reading ability one of the methods most often recommended by researchers is reading strategies
instruction. Reading strategies instruction is found to be highly effective both in dealing with problems on the vocabulary and sentence levels (Fischer, 2003;
Sindelar,Monda,& O’Shea,1990)and on higher-level issues such as text
comprehension (Alfassi, 2004; Brown, Pressley, Van Meter, & Schuder, 1996; De Corte, Verschaffel, & De Van, 2001;
Johnson-Glenberg, 2000; Van Keer, 2004).
In recent years researchers have tried to deduce principles of reading strategies from empirical studies (Dole, Duffy, Roehler, & Pearson, 1991; Mastropieri &
Scruggs, 1997; Pressley, Johnson, Symons, McGoldric, & Kurita, 1989). Some of the more widely recommended approaches are determining the main messages (e.g., summarization), using text enhancements (e.g., illustrations, text structure
representations, mental images), question and answer drills (e.g., self-questioning), meta-cognition (e.g., comprehension monitoring). Paris and Paris (2001) postulated that there are several important characteristics in successful reading strategy instruction: (1) using multiple strategies; (2) presenting clear messages for application, such as why, when and how to use the strategies; (3) making students attribute their learning achievements to the correct use of strategies; (4) applying strategies in peer interaction; (5) incorporating cognitive strategies into a more wide-ranging domain of self-regulated learning; (6) embedding strategies in the reading
activities in daily life to provide more opportunities for application and practice.
In fact, considering the
characteristics of successful strategy instructions described by Paris and Paris (2001), one can find that it is rather difficult for teachers to create those conditions in the classroom environment.
Take multiple-strategy instruction for example. It takes a significant amount of training for most teachers to become familiar with using the strategies and being able to teach them to their students (Pressley & El-Dinary, 1997; Pressley, Duke, & Boling, 2004; Duffy, 1993a).
Further, preparing teaching materials for instruction simply takes too much time and energy (Pressley, Goodchild, Fleet, Zajchowski, & Evans, 1989). Furthermore, there are complexities involved in
providing students with the opportunities to apply the strategies in daily life.
Perhaps the difficulties above are the
major reasons why strategies instruction is seldom implemented in a classroom (Pressley, 2000; Van Keer, 2004).
Therefore, it is essential to create a proper environment for teaching assistance which relieves teachers of their burden in
implementing strategy instruction and increasesstudents’opportunitiesand willingness to use reading strategies outside the classroom.
The difficulties of implementing reading strategies instruction in the classroom may be reduced with the assistance of information technology.
Computer-assisted instruction has been very popular during the last two decades, and scholars agree on the feasibility of applying computers in reading instruction under appropriate design (Labbo &
Reinking, 1999; Leu, et al. 1998;
Lungberg, 1995; Reinking, Labbo, &
Mckenna, 2000). There are generally several advantages to the use of computers
in reading instruction (Lynch, Fawcett, &
Nicolson, 2000; Mathes, Torgesen, &
Allor, 2001). Firstly, computers can provide immediate individual feedback based on thestudent’slearning condition.
Secondly, learning with computers allows students to control the pace of learning by themselves. Thirdly, properly arranged courses may be operated independently with computers, thus relieving teachers of some of the burden and giving students more opportunities to learn independently.
Fourthly, through presentations using different media, it may strengthen students’motivation to read.
It is worth noting that, although the computer-assisted reading has had some impressive results (See Blok, Oostadam, Otter, & Overmaat, 2002; McArthur, Ferretti, Okolo, & Cavalier, 2001, for reviews), there are some limitations. On the level of reading ability, most studies on computer-assisted reading have dealt
with reading issues on a more fundamental level, such as word
recognition (e.g., Frederiksen, Warren, &
Rosenberry, 1985; Wise, 1992; van Daal
& Reitsma, 2000) or phonological
awareness (e. g., Farmer, Klein, & Bryson, 1992; Mathes et al., 2001;van Aarle & van den Bercken, 1999). There are relatively few studies of assisting with higher-level text comprehension through computer technology. Currently, the majority of efforts on computer-assisted
comprehension focus on the effects of text enhancement (such as indexes for difficult vocabulary or background information for text, Chang, Sung, & Chen, 2001; Leong, 1995; MacAthur & Haynes, 1995;
Reinking, 1988) or visual or auditory aids ( such as illustrations, images or speech, Elkind, Cohen, & Murray, 1993; Montali
& Lewandowski, 1996) on enhancing text comprehension. Those studies usually concerned with manipulating the assisting
messages themselves, and dealt with the improvementoflearners’reading capabilities to a very limited degree. Their practical implications are still in debate (McArthur et al., 2001).
Computer-assisted reading strategies training is a feasible approach to
enhancing students’capabilitiesofdealing with text. Regarding the training of
strategies for text comprehension, there have been some studies of using computer technology to conduct strategy instruction (Caverly, 1998; Chang, Sung, & Chen, 2002; Kaniel, Licht, & Peled, 2000). But the strategies are chosen in a fragmented fashion, and lack integration. For example, Caverly et al. used software for strategies of gathering, arranging and presenting materials, while Chang et al. used the concept mapping strategy. Because currently, much emphasis is placed on using multiple-strategy instruction in the classroom situation (Paris & Paris, 2001;
Pressley et al., 1989), we suggest it is important to find out how to design proper multiple strategies with computer
technology to facilitate text
comprehension ability. Considering that there are a limited number of computer environments that are directly aimed at the learner’sreading ability and areequipped with a set of comprehension strategies, the first purpose of this study is to design a reading instruction system with multiple strategies based on the components in the text comprehension process.
Another current limitation in the study of computer-assisted reading is that most studies conducted on elementary school children focus on students with below-average learning ability (McArthur et al., 2001). In contrast, there are fewer studies using students of average or above-average ability as participants.
Some studies on strategy instruction indicated that children of lower learning
ability benefited less than those with average or above-average ability (Garner, 1987). Nevertheless, some other
researchers found that students with poorer ability still managed to benefit from strategy instruction (Brown et al., 1996; De Corte et al., 2001; Duffy, 1993b;
Palinscar & Brown, 1984). There are currently few studies related to the issue of whether strategy instruction in a computer environment produces different instruction effects in children of different ability. The second purpose of this study is to compare the benefits for children with different abilities from computer-assisted reading strategies.
Method
Participants
The participants in this study are sixth-grade students in four classes from an elementary school of Taoyuan County, Taiwan. The majority of the children were from middle-class families. The age of the
students range from 12 to 13, with an average of 12 years 3 months. Each class is randomly assigned into the experiment group and the control group. There are 65 students in the experimental group (35 boys and 30 girls) and 65 in the control group (34 boys and 31 girls). The students are given the Reading Comprehension Screening Test (Ko, 1999), and the students whose scores are higher than the average of all sixth-grade students go into the high reading ability group, while those with scores lower than average go into the low reading ability group. The
experimental group has 31 high-ability and 34 low-ability students. The control group has 35 high-ability and 30
low-ability students.
Design
This study employed a quasi-experimental design. The
independent variables in the experiment included group (experiment or control)
and reading ability (high or low). The dependent variables included a reading comprehension score for the narrative, a reading comprehension score for the exposition, and a score of using reading strategies, which was further divided into five categories, self-questioning, error detection, inference blank-filling, summarization, and prior knowledge integration. The covariate was the student’sscoresin languageartsin the previous school year.
Experimental Tool
This study designed a Computer Assisted Strategies Teaching and Learning Environment (CASTLE) as the major tool for implementing the strategies instruction.
To find a sound base for designing strategies for CASTLE, we examined various models of reading comprehension (e.g., Just & Carpenter, 1987; Kintsch, 1998; Mayer, 1996). We think that the Mayer’s(1996)
Selection-Organization-Integration (SOI) model for text comprehension is a good framework for choosing and designing reading strategies. Selection means that in reading an article, one needs to know which parts of the content are more important and which ones are less important. Organization means that one has to re-organize the selected messages in the short-term memory, and form a coherent, interconnected overall concept.
Integration means linking the knowledge organized in short-term memory with related old knowledge in long-term memory. However, studies pointed out thatthereader’sattention and whetherhe knows to monitor his own comprehension are important factors influencing
comprehension (Dole et al., 1991). The SOI model does not sufficiently reflect the components of attention and monitoring so thisstudy expanded Mayer’sSOI model by adding the components of
attention and monitoring. Also, the
ASOIM model ( Figure 1) was used as the framework for selecting and designing comprehension strategies. These strategies then form the basis for designing a
multiple-strategy reading assistance system which we designated CASTLE.
Examples of the comprehension strategies chosen in this study according to the ASOIM model are shown in Figure 3. The following explains the detailed design methods.
Strategies for attention.
Self-questioning and error detection strategies are used to help readers concentrate on texts. Self-questioning means that the learner asks himself questions while reading an article, such as messages in the article regarding the who (characters), what (events), where
(settings), when (time) and why (reasons), and then answers his own questions. For expository texts, questions regarding what,
how, why, or relationships are more emphasized. Anderson (Andre &
Anderson, 1978-1979; Anderson, 1978) thinks that by using that strategy, students have to constantly focus their attention on the material they are reading. A number of studies (Clark, Deshler, Schumaker, Alley,
& Warner, 1984; Short & Ryan, 1984) also find that the strategy helps enhance comprehension. Error detection means that the reader should find some explicitly or implicitly contradictory messages in the content of the article he is reading
(Markman, 1979). Because in the process of error detection the reader must
concentrate to find any contradictions, this study uses it as one of the strategies for concentration.
Strategies for selection. We use strategies of concept map blank-filling and highlighting to train the readers to select important messages. Novak (1984) thinks that when teaching materials are
presented through concept maps, students can easily find the interrelation of
concepts with the results that concepts in the teaching materials become clearer and more specific. To avoid the difficulty of producing concept maps from scratch, this study uses the concept map blank-filling strategy by Chang, Sung and Chen (2001), asking students to select corresponding important messages to supplement the missing parts in the concept maps. And the highlighting strategy is used by researchers to confirm key concepts and enhance comprehension (Shaughnessy &
Baker, 1988; Wade & Trathen, 1989).
Strategies for organization. This study uses the strategies of concept map correction and inference blank-filling to train readers how to organize the
messages in an article. The concept map correction method works as follows. After the student has finished reading an article, he/she is given a complete concept map in
which some concepts are wrong. The student needs to select the correct concept from a list of concepts provided by the system and fill them in. Because the student must understand the organization and relationship of the main concepts in the entire concept map before he can judge which concept nodes are correct or incorrect, the student has to organize and link the concepts in the entire article. This procedure may help students organize the ideas in the text (Chang, Sung, & Chen, 2002). Inference blank filling requires the reader to resort to his own ability to inference in order to supply the missing important messages in the context of the article. The inference blank-filling strategy used in this study operates as follows. Leave out some blanks in an article; the reader needs to make
judgments according to the context of the article, choose the correct words or phrases from a pull-down menu provided
by the system, and then fill in the blanks.
Strategies for integration. This study uses proposition-combining and
summarization strategies to train readers in integrating knowledge. Proposition combining is derived from the
sentence-combining strategy (Fusaro, 1993; Neville & Searls, 1991). The difference is that combining sentences forms paragraphs, while combining propositions uses concepts (arguments) as units and combine them into propositions.
The strategy operates as follows.
Important concepts are picked out from an article, and then mixed with new concepts related or unrelated to the old propositions that are not found in the present article (the concepts are chosen from textbooks previously used by the reader). Then the system provides relation links (predicates) for the reader to use to integrate the concepts. Summarization is a strategy preferred by many researchers. Hidi and
Anderson (1986) point out that a summary is a brief narrative of condensed
information to represent the original article, which can help the student comprehend the reading material and organize the article he is reading. In fact, it is not easy to re-organize the important concepts in an article without a fair amount of prior knowledge as foundation.
Therefore, summary to a certain degree should involve integrating new and old knowledge (Brown & Palincsar, 1984).
This study uses a method based on Brown and Day (1983) and Dole, Duffy, Roehler, and Pearson (1991), which ask the reader to select important content from an article, delete the trivial and redundant details, and combine the remaining important sentences.
Strategies for monitoring. The purpose of the monitoring strategy in this study is to train students to understand how well they are executing the strategies
in the four processes described above. By doing this,wetry to fosterthereader’s awareness in his own comprehension and employment of strategies. The method used in this study is based on the informed strategies for learning by Paris et al.
(1984), which asks the reader to remind himself and evaluate his state of his learning after completing each strategy learning session through the verification of sentences related to the use of
strategies.
The functions of CASTLE. Students log on to the CASTLE through a browser on the local computer, and retrieves data of the teaching materials from a database on the remote server. CASTLE uses an agent to guide the user. The agent gives voice instructions to provide guidance on three major procedures: the work procedure, the user’s interface, and feedback. The guidance on the work procedure is presented to the user on
system startup.Forguidance on the user’s interface instructions are given to the users regarding the operations of the system interface. For guidance on feedback users are given proper feedback according to the results after any strategies are completed. Figure 2 shows how the agent and the toolbox work.
In CASTLE the workflow for reading strategy instructions are designed according to the steps of reading strategy instruction recommended by Taylor, Harris, and Pearson (1995): 1. Explain what is in the strategy (what). 2. Explain why the strategy is important (why). 3.
Demonstrate how to use the strategy (how). 4. Explain in what situation this strategy may be used (when). 5. Guide the students to practice using actual articles. 6.
Ask the students to practice independently through reading articles. The steps CASTLE executes are: on startup, CASTLE communicates with the user
through text and voice instruction by the agent to inform the user about the meaning of reading strategy. Then the system provides an example, in which the agent operates and demonstrates the instruction procedures for a reading strategy. To make the user familiar with that reading strategy, he should choose another article to read, and actually use the strategy. After the operation is completed, CASTLE provides the user with feedback, and then the user verifies the processes of using a strategy. The four processes described above form a complete cycle; after reading one article, the reader needs to go through all the processes before he can go to the next article. As example we explain the process using the instruction on error detection strategy.
1. Explanation of reading strategy and practice on examples: CASTLE informs the users about the meaning of the
reading strategy through text and voice narration by the Agent, and asks the user to practice on an example. The user needs to practice successfully on the example before he is allowed to choose an article to read (Figure. 3a).
2. Reading the article and using the strategy: After the reader has read the
article, he can use the mouse to highlight the sentences in the article, and then press the“underscore”button to transform the highlighted sentences into blue characters to signify the semantic contradictions. The reader can also use the mouse to highlight the words orsentences hedoesn’t know and then pressthe“voice” button to ask the agent to pronounce or read aloud.
article, he can use the mouse to highlight the sentences in the article, and then press the“underscore”button to transform the highlighted sentences into blue characters to signify the semantic contradictions. The reader can also use the mouse to highlight the words orsentences hedoesn’t know and then pressthe“voice” button to ask the agent to pronounce or read aloud.