A technique for tracking the reading rate to identify the e-book reading behaviors and comprehension outcomes of elementary school students

Abstract

Tracking individual reading behaviors is a difficult task, as is carrying out real-time recording and analysis throughout the reading process, but these aims are worth pursuing. In this study, the reading rate is adopted as an indicator to identify different reading behaviors and comprehension outcomes. A reading rate tracking technique is thus developed with an Interactive E-book Learning System (IELS), and this study examines whether the reading rate detected by this system can properly reflect the actual reading behaviors and comprehension outcomes of users. A total of 500 quantitative records were collected from the reading profiles of 43 fifth-grade students, separated into two groups for oral and silent reading behaviors, and then analyzed to reveal the reading rates accounting for specific reading behaviors and comprehension outcomes. Our findings indicate that the reading rate can accurately reflect students’ reading behaviors. In addition, a somewhat positive correlation between reading rates and comprehension outcomes was only found in the silent reading group. By using the technique presented in this work, the reading behaviors related to e-books can be easily interpreted based on the reading rate over time throughout the reading process, something that is much more difficult when using printed books. Based on the results of this work, it is anticipated that educators can get better insights into students’ behaviors with regard to reading, so as to deliver more personalized and effective instruction. Introduction Liu (2005) investigated how people’s reading behaviors had changed over the previous decade, with the focal concern being the rise of screen-based reading. This is important, for an entire generation has grown up with new technology and is likely to have different behaviors when using digital media for reading. Numerous e-book platforms and reading devices have been developed in recent years, and many studies (eg, de Jong & Bus, 2004; Grimshaw, Dungworth, McKnight & Morris, 2007; Korat & Shamir, 2007, 2008) have revealed that e-books can provide individualized, on-demand multimedia materials that seem to be able to promote learning efficiency. Many researchers (eg, Bierman, Ortega & Rupp-Serrano, 2010; Woody, Daniel & Baker, 2010) have examined how the students use e-books in academic contexts. Liu (2005) stated that more studies on the reading behaviors that occur in a digital environment are needed for students, and further empirical research that identifies these behaviors from the reading process is crucial before more extensive use of e-books is made in educational contexts. Woody et al (2010) British Journal of Educational Technology Vol 46 No 4 2015 864–876 doi:10.1111/bjet.12182 © 2014 British Educational Research Association suggested that the designers of e-books should consider how to develop them for optimal use, and thus they need to further explore the various user experiences that occur when reading e-books. This is especially true with regard to children, who seem to prefer a screen-based reading and are generally very capable of using digital devices (Huang, Liang, Su & Chen, 2012; Liang & Huang, 2014; Liu, 2005). Many studies (Korat, 2010; Korat & Shamir, 2007, 2008; Segal-Drori, Korat, Shamir & Klein, 2010; Shamir, Korat & Barbi, 2008) have revealed that all the aspects of children’s emergent literacy, including vocabulary, word recognition and phonological awareness, can be improved by reading e-books. Therefore, more efforts should be made to explore children’s use of e-books (Salmerón & García, 2011), including the related reading traits, reading preferences and cognitive processes, especially in educational contexts. In the printed book reading context, tracking every student’s reading behaviors through the whole reading process is very difficult. However, with the aid of information and communication technology, teachers are now able to examine individual reading behaviors in a convenient and timely manner. For example, several studies (eg, Erickson et al, 2011; Jukka, 2010; Marisa, 2011; van Gog & Scheiter, 2010) use an eye-tracking technique to record the reader’s gazing location and the fixation duration to examine the acquisition process of information from texts. Moreover, Cole et al (2011) investigated the cognitive processes that occurred in several reading tasks with Practitioner Notes What is already known about this topic • Many studies have revealed that the e-books can provide individualized, on-demand multimedia materials, which seems be able to promote learning efficiency. • Tracking individual reading behaviors is a difficult task, and carrying out the real-time recording and analysis throughout the reading process is even more challenging. • Reading rate is a useful indicator to assess students’ reading performance. What this paper adds • With the reading rate tracking technique embedded in the Interactive E-book Learning System (IELS), this study successfully tracked, recorded, and analyzed the individual reading rates as well as the corresponding reading statuses. • By using the reading rate tracking technique presented in this work, the reading behaviors related to e-books can be easily interpreted based on the reading rate and the corresponding reading status over time, which provide more authentic details of the reading behaviors that occur throughout the reading process than is possible when using printed books. Implications for practice and/or policy • In practice, the individual reading profiles collected by the system can be used to verify whether a student’s reading rate is normal, helping the teacher to identify any weaknesses in the reading of e-books. Furthermore, based on the results of this work, it is anticipated that educators can get better insights into students’ behaviors with regard to reading, so as to deliver more personalized and effective instruction. • Finally, the current study demonstrated a reading rate tracking technique that can provide a clearer view of the details of the actual reading behavior than is possible when using a printed book. Based on this technique, ineffective or inefficient reading practices can be uncovered, and this is of value for future research on the use of e-books in an academic context. Tracking reading rate of e-book reading 865 © 2014 British Educational Research Association this technique to model the users’ search behaviors. Unfortunately, this technique cannot be universally applied in ordinary classroom practice, and there also remains a lack of empirical evidence in the literature with regard to reading behaviors and comprehension outcomes in the overall e-book reading process. Little is thus known about the process of reading e-books, and formative evaluations of this process are especially scarce. An alternative approach is to use an e-book reading platform along with efficient tracking, recording and analyzing functions to gather details of reading behaviors. This approach can provide a low cost and relatively unobtrusive way to collect personalized reading profiles, enabling educators to obtain more insights into students’ reading behaviors, so as to provide appropriate guidance throughout the whole reading process (Huang et al, 2012). Many researchers (Carver, 1977, 1983, 1990; Duggan & Payne, 2009; Dyson & Haselgrove, 2000; Fraser, 2007; McLay, 2007; Rasinski, 1999, 2000) stated that the reading rate (measured as the words read per minute, or wpm) is a useful indicator to assess students’ reading performance. An excessively slow and disfluent reading rate generally leads to poor comprehension outcomes. By contrast, readers with good word recognition accuracy tend to have a faster reading rate and better comprehension outcomes (Rasinski, 2000). In addition, the reading rate is also a significant factor in classroom for teachers’ perceptions of their students’ degree of proficiency in reading (Rasinski, 2000). It should thus be possible to use the reading rate as an indicator to investigate the related reading behaviors, and thus to derive a formative evaluation of the overall reading process in this context. Our previous work (Huang et al, 2012) developed a tailor-made e-book learning system for elementary school students, namely the Interactive E-book Learning System (IELS), in which a reading rate tracking technique was embedded that aims at collecting students’ reading rate profiles for depicting details of reading actions, as seen in their exterior reading behaviors. This system can be applied to gain more insights into students’ reading processes and comprehension outcomes when using e-books, so as to support observations of different reading behaviors and comprehension outcomes by more fundamental investigations in this field, such as identifying the specific reading rate patterns (Liang & Huang, 2014) or the gender differences in the reading of e-books (Huang, Liang & Chiu, 2013; Liang & Huang, 2014). Background and related work The reading rate, reading behaviors and comprehension outcomes McLay (2007) argued that measuring the reading rate is difficult, and the results may be puzzling, as they do not always represent how fast the words are being processed or how well they are understood by the readers. Nevertheless, several researchers (Joshi & Aaron, 2000; Liang & Huang, 2014; Rasinski, 2000) have found that the reading rate is positively correlated with comprehension outcomes. In other words, the reading rate can indicate to what extent information is retained in readers’ working and short-term memories at various levels of comprehension (Liang & Huang, 2014; Nation & Cocksey, 2009). It is thus important to explore the pedagogical value of reading rates in educational practice. As children still have to learn how to decode many words, their emergent literacy is generally associated with oral reading (Nation & Cocksey, 2009), which affects their reading rate, comprehension, motivation and self-confidence. However, skilled readers are usually able to adjust their reading rate to maintain the desired level of comprehension (Harris & Sipay, 1990), and thus the reading rate is a useful indicator of reading behavior, as stated by many researchers (Carver, 1977, 1983, 1990; Duggan & Payne, 2009; Dyson & Haselgrove, 2000; Fraser, 2007; McLay, 2007; Rasinski, 1999, 2000). To identify a wide range of reading behaviors, this study surveyed the literature to define the generalized reading rates associated with various reading behaviors. 866 British Journal of Educational Technology Vol 46 No 4 2015 © 2014 British Educational Research Association Based the findings of Fraser (2007), that the rates of Chinese reading ranged from 61 to 984 wpm, this study assumed that an individual can read texts at a rate belonging to any of the above-mentioned behaviors that are grouped into the category of “on-reading,” as shown in Table 1. On the other hand, Carver suggested that it would be impossible for an individual to read a passage at 1000 wpm or more (Carver, 1977, 1984), and thus if the rate is greater than or equal to 1000 wpm, this study categorized this situation as representing “off-reading.” Under the off-reading category, the related reading behaviors, such as flipping pages or glimpsing the text, were also reported by researchers (Carver, 1977, 1984; Harris & Sipay, 1990). The second column in Table 1 gives the likely range of reading rates with respect to the onor off-reading categories. The third column lists the well-regulated reading behaviors, which are the exterior reading actions or manifest behaviors reported in the literature for the related reading rates classified in the first column. Identifying the reading behaviors with the reading rate tracking technique As shown in Figure 1, the reading rate tracking technique embedded in IELS was used to track the students’ reading rate throughout the reading process. When students log into the system, an e-book shelf is shown on the screen (Figure 1a), and then users can select any e-book from the shelf to read. Two reading modes, full and normal screen, are provided for different reading interfaces, as shown in Figure 1c and d. In order to uncover the students’ e-book reading behavior, this study developed a reading rate tracking technique, including the tracking, recording and analyzing functions embedded in the IELS. As shown in Figure 2, when the student starts reading an e-book, the system automatically activates the tracking function. Whenever the student turns to the next or previous page (step I in Figure 2), the system will instantly record this information into their individual reading profile via the recording function. The system then continues to track the student’s reading rate until it is turned off. Table 1: The list of reading rates and statuses associated with reading behaviors in the literature Reading category Reading rate (wpm) Reading behavior Study On-reading 0–1000 Excessively slow Inefficient reading Disfluent Labored Inexpressive Unenthusiastic rendering (Harris & Sipay, 1990; Rasinski, 2000; Walczyk, Marsiglia, Bryan & Naquin, 2001) Sustained attention In-depth reading Oral reading* Concentrated reading Annotation (highlight) Silent reading* Keyword spotting One-time reading Reading selectively Browsing and scanning Nonlinear reading (Carver, 1977, 1990; Duggan & Payne, 2009; Fraser, 2007; Gillett & Temple, 1986; Harris & Sipay, 1990; Liu, 2005; Liu & Huang, 2008; Rasinski, 1999; Reader & Payne, 2007; Reading, 2012; Stroud & Henderson, 1943) Off-reading ≥1000 Flip page Glance and glimpse (Carver, 1977, 1984; Harris & Sipay, 1990) “*” represents the reading behaviors examined in this study. Tracking reading rate of e-book reading 867 © 2014 British Educational Research Association E-books Reading rates

DOI: 10.1111/bjet.12182

5 Figures and Tables

Cite this paper

@article{Huang2015ATF, title={A technique for tracking the reading rate to identify the e-book reading behaviors and comprehension outcomes of elementary school students}, author={Yueh-Min Huang and Tsung-Ho Liang}, journal={BJET}, year={2015}, volume={46}, pages={864-876} }