The Effects of Self vs. Group selection on Engagement in a Graded Reading Activity:An Exploratory Study - -

The Effects of Self ^-  vs. Group ^- selection on Engagement  in a Graded Reading Activity : An Exploratory Study

Kenneth Schmidt and Chein

^-

Man Lee

Abstract : This exploratory, 10

^-

week study employed online questionnaires to gauge Japa- nese university students’ sense of engagement as they participated in selection, reading and discussion stages of a graded reading activity. Specifically, we examined the effects of self

^-

vs. 

group

^-

selection of graded readers

─

done in alternate weeks throughout the study. Results revealed similarly moderate levels of reported engagement, with no statistically significant dif- ferences and small to negligible effect sizes at each stage of the activity. However, an explor- atory examination of the results beyond strict statistical constraints hints at slight advantages for reading of group

^-

selected readers in the effort domain, and for discussion of group

^-

selected readers in terms of engagement, competency and success. Further work may be merited in these areas. In sum, students responded positively to both approaches, and lacking strong evidence of differential engagement, instructors working with similar activities in similar contexts can feel free to vary their use of self

^-

or group

^-

selection depending on practical constraints and/or the type of activities they want to try. This study was conducted as part of the 2018 Quantita- tive Research Training Project, a professional development program for language teachers in Ja- pan looking to gain knowledge and experience in quantitative research methods.

Keywords : ELT, engagement, extensive reading, graded reading, language activities

Introduction

Extensive reading (ER) and associated activities can have a tremendous impact on foreign language learning and attitudes toward language learning (Day & Bamford, 1998). Engagement is a critical fac- tor in learning and academic success (Wang & Degol, 2014). A logical question is how ER

^-

related ac- tivities can be designed to facilitate engagement and potentially yield enhanced learning outcomes. 

This exploratory study examines the effect of task design on student engagement in an extensive read- ing activity, specifically addressing this research question : Does self

^-

or group

^-

selection of readers con- tribute to greater engagement in the selection, reading and discussion stages of an extensive reading activi- ty ?

The authors undertook this study in parallel with instructors at 20 universities in Japan participating

in the 2018 Quantitative Research Training Project, a professional development program for language

teachers in Japan looking to gain knowledge and experience in quantitative research methods (Sholdt,

2018, 2019). Led by Gregory Sholdt at Konan University, the project was supported by a Grant

^-

in

^-

Aid for Scientific Research ─ “Development of a Second Generation Research Training Program for Language Teachers” (JSPS KAKENHI Grant No. 16K02920). Participating instructors joined togeth- er in training sessions and online forums while conducting similar, but separate, research projects at their respective universities. This is a report on the study we conducted at Tohoku Fukushi Univer- sity.

Background and literature review Extensive Reading

Extensive reading (ER) in the EFL (English as a foreign language) context describes an on

^-

going activity in which learners read large amounts of English text sufficiently matched to their level of com- petency as to allow them to read relatively fluently ─ quickly and with good comprehension ─ as they focus on meaning ─ the story, scene, situation, etc. being described (Waring & McLean, 2015). This is in contrast to the translating, wrestling with linguistic features and consulting of dictionaries learn- ers might do in puzzling out more difficult texts.

ER texts are typically graded readers ─ novels, short story collections and nonfiction works written at various levels (thus the term “graded”) ─ allowing learners of varying advancement to access lev- el

^-

appropriate reading material. These are most commonly accessed in book form (Day & Bamford, 1998) or online through various electronic library services (e.g., Xreading.com) (Milliner & Cote, 2014).

Numerous studies have documented the utility of ER in facilitating language acquisition and fluency development in a range of ways (Beglar & Hunt, 2014 ; Jeon & Day, 2016 ; Krashen, 2007 ; McLean

& Rouault, 2017 ; Nakanishi, 2015 ; Waring, 2014), and many researchers consider it, along with ex- tensive listening (EL) (Renandya & Jacobs, 2016), as cornerstones of effective foreign language pedagogy. As Maley (2008) states : “The only reliable way to learn a language is through this mas- sive and repeated exposure to it in context : precisely what ER provides” (p. 147).

As indicated above, “extensive” reading implies a large quantity of reading (Waring & McLean,

2015). What quantities are needed to yield meaningful language gains ? For vocabulary learning,

Nation and Wang (1999) concluded that at least one graded reader a week at a learner’s current level

would be the minimum needed for sufficient repetition of exposure to unfamiliar words. Waring

(2013), cited in Waring and McLean (2015), found that even two to three readers a week might not

suffice. Nishizawa, Yoshioka, and Fukada (2010) identified a 300,000 word threshold for meaningful

fluency gains. For this reason, we prefer the term graded reading (GR) when learners are reading

graded readers, but not in quantities that could be called extensive (Schmidt, 2007). The study re-

ported on here, in which students generally read one book per week, sits on the border of GR and ER.

While ER has taken hold in Japan, as evidenced by 23 ER

^-

oriented presentations at the 2019 Japan Association for Language Teaching International Conference in Nagoya, it is still far from a given in EFL curricula around the country, and far less so in most countries where EFL is taught.

So, ER is the most efficient way of promoting foreign language learning. It is therefore ironic that most programmes of instruction make little or no room for it. . . On the one hand we have a proven resource for promoting language learning. On the other, a massive indifference or even resistance to it. (Maley, 2008, p. 149)

The need for massive amounts of comprehended input is a given in language learning theory (Ellis, 2005 ; Krashen, 2004 ; Yamashita, 2013), but the fact that so few educators even suggest ER/EL to their students underscores the need for further research on the efficacy of ER and its implementation in language curricula. This study is a small part of that ongoing effort to present evidence and options for instructors considering ER/EL opportunities for their own students.

Engagement

In considering effective teaching and learning of any kind, a central concept is engagement. Stu- dent engagement in learning activities has been shown to be a key factor in learning and academic suc- cess at all levels ─ primary (Goss & Sonnemann, 2017), secondary (Lee, 2014 ; Wang & Eccles, 2011) and tertiary (Kuh, Cruce, Shoup, Kinzie, & Gonyea, 2008). Wang and Degol (2014) state :

When students are engaged with learning, they can focus attention and energy on mastering the task, persist when difficulties arise, build supportive relationships with adults and peers, and con- nect to their school. Therefore, student engagement is critical for successful learning. (p. 137)

Engagement is a multilevel construct, including involvement with 1) the school community, 2) the classroom or subject domain, and 3) specific learning activities (Wang & Degol, 2014). Engagement is also multi

^-

dimensional, including behavioral (e.g., effort, participation), emotional (e.g., interest, enjoy- ment) and cognitive (e.g., depth of processing, quality of thinking) factors (Shernof et al., 2017 ; Wang

& Degol, 2014).

Shernof et al. (2017) characterize engagement within the framework of flow theory (Csikszentmih-

alyi, 1998 ; Csikszentmihalyi & Rathunde, 1993), but distinguish classroom engagement from flow in

this sense :

“. . . flow is typically conceptualized as a fairly dichotomous state (one is in the heightened state of flow, or not). In contrast, motivation and engagement in classrooms is conceptualized to be fair- ly continuous. Relatively high engagement may be expected to lead to better performance on a test or in a course ; but doing well does not require the “heightened” engagement or motivation of flow.” (p. 4)

The “continuous” nature of engagement suggests the possibility of assessing degree of engagement in a task or process. Attempts to measure engagement often involve operationalizing the construct with statements or questions drawn from these dimensions (e.g., “I enjoyed the task.”) and having stu- dents respond to these during or after the activities or events being investigated (Fredricks & McCols- key, 2012). This was the approach taken in this study.

Engagement in ER

This positive relationship between engagement and learning has been shown to apply to specific skills, like reading (Guthrie, Klauda, & Ho, 2013 ; Lee, 2014 ; Ng & Bartlett, 2017) and to foreign lan- guage learning (Dincer, Yes¸ilyurt, Noels, & Vargas Lascano, 2019 ; Egbert, 2003 ; S¸ahin

^-

Kızıl, 2014 ; Yu et al., 2019 ; Stroud, 2013).

Within EFL, researchers have also been exploring how to enhance engagement in ER and ER

^-

relat- ed activities (Amelsvoort, 2017 ; Milliner & Cote, 2015 ; Minkowitsch, 2013 ; Yamashita, 2013 ; Yo- shida, 2017). One innovation involves the role of literature circles. Self

^-

selection of readers for lev- el, interest and enjoyment is a basic principal in many successful reading programs (Day & Bamford, 1998), and in 38 of 44 programs surveyed by Day (2015) students individually select their own reading materials. However, numerous educators extol the virtues of literature/reading circles in creating community and fostering engagement in reading and in

^-

class discussion (Baharuddin Marji, Rafik

^-

Ga- lea, & Mei Yuit, 2015 ; Furr, 2007 ; Maher, 2015 ; Shelton

^-

Strong, 2012). These schemes typically involve groups of students discussing a reader they chose together or were assigned by the instructor. 

Does discussion of a shared book offer significant advantages over students introducing and discussing self

^-

selected books they have read ?

This presents an interesting contrast and opportunity to investigate engagement in ER

^-

related activities. Individually selecting graded readers would seem to offer advantages in dimensions of en- gagement such as control, interest and enjoyment, especially during reading. Group

^-

selection could offer its own advantages, particularly in the discussion stage.

We thus conducted a study to address this research question : Does self

^-

or group

^-

selection of readers

contribute to greater engagement in the selection, reading and discussion stages of a graded reading activity ?

Method Participants

Participants were 32 second

^-

year students majoring in rehabilitation (22), nursing (8), education (1) and psychology (1) at Tohoku Fukushi University. They were enrolled in two sections of a required, general

^-

education English course taught by Schmidt and meeting once

^-

a

^-

week for 90 minutes. The year

^-

long course was content

^-

based, examining Japanese culture and other cultures in relation to sev- eral topical areas (e.g., body language, possessions, dating, special occasions, food, customs), with a supplemental graded reading component. While students in the course generally deal successfully with materials in the Common European Framework of Reference (CEFR) A2 range, no more specific proficiency data were available for individual students, and English level was not included as a variable in the data analysis. A total of 53 students were actually enrolled in the two classes. All signed in- formed consent forms and participated in the activities involved in the study, which ran through the bulk of the first, 15

^-

week semester. However, students frequently neglected to submit weekly, online questionnaires, and only 32 submitted sufficient questionnaires to be included in the data analysis.

Design of study

The study

^-

related activities were introduced in the syllabus as a normal part of the course. In pre- vious years, graded reading had been employed as a supplemental activity, with students reading at least six books and 24,000 words during each 15

^-

week semester and occasionally discussing the books they were reading in class. For purposes of the study, the graded reading and discussion component was expanded to a 15

^-

20 minute part of each, weekly class. Also, rather than reading paperback books from the school’s grader reader library, students read and listened to readers online, using the Xreading.com virtual library. This allowed multiple students to concurrently read the same book, and made for easy collection of data on words read, reading speed, comprehension quiz scores, etc.

Only seven students had previously read more than three graded readers, and none had experience

with the Xreading virtual library, so we began with a two

^-

week, pre

^-

study orientation and practice

period. Over this period, the rationale for graded, extensive reading was explained and students par-

ticipated in all study

^-

related activities, including selection and reading of books on Xreading, taking

Xreading comprehension quizzes, completing online, Google Forms questionnaires, and preparing for

and participating in in

^-

class discussion. During this time, students were encouraged to sample a

number of readers at different levels and identify their “comfort zone” ─ levels at which they occa-

sionally met a word they didn’t know, but this didn’t interfere significantly with their understanding or

enjoyment of the content. Studies indicate that 95

^-

98% of words appearing in a text should be known

for reasonable comprehension (Hu & Nation, 2000 ; Laufer, 1989 ; Schmitt, Jiang, & Grabe, 2011) fa- cilitating enjoyment/engagement and implicit learning from context, a major benefit of graded/exten- sive reading (Hunt & Beglar, 2005 ; Liu & Nation, 1985).

The study then commenced and continued for ten weeks, divided into five, two

^-

week sets. Stu- dents were required to read at least one reader prior to each class. In the first class of each two

^-

week set, students counted

^-

off to form groups of 3

^-

4 in which they introduced and discussed their self

^-

se- lected readers for 10

^-

15 minutes. They then chose a common, group selected reader ─ at a level comfortable for all ─ to read out

^-

of

^-

class and discuss together the following week. Over the span of the study, students thus worked with five different groups and read a total of ten readers over that span

─ five self

^-

selected and five group

^-

selected. 

Each week, there were several stages and associated tasks : Selection stage

  • Choose a book to read on Xreading ─ alternately done individually or with a group   • Fill out the online “Selecting a Book” engagement questionnaire (see Appendix 1).

Reading stage

  • Read the selected book on Xreading, (listening while reading if desired).

  • Take the five

^-

item Xreading comprehension quiz. A three out of five score was required to re- ceive credit for reading the book.

  • Fill out the “Reading the Book” engagement questionnaire (see Appendix 1).

  • Complete a 4

^-

6 sentence “summary and response” report (see Appendix 2) to facilitate discus- sion in the next class.

Discussion stage

  • Join in a 10

^-

15 minute discussion on the book(s) read over the previous week. Students were provided with a “Book Talk” sheet (see Appendix 3) with suggested discussion questions and space for key

^-

word notes to motivate listening and facilitate peer

^-

learning. They were encour- aged to conduct their discussions in English, but comments and clarifications were allowed, and often done, in Japanese.

• Fill out the online “Discussing the Book” engagement questionnaire (see Appendix 1).

Instruments

Three engagement questionnaires were thus completed each week, one immediately following each

stage of the activity. In studying links between student engagement, student practices and student

performance, Shernof et al. (2017) employed an Experience Sampling Method (ESM) (Csikszentmih-

alyi, 1998) in which students completed online surveys at intervals as they participated in class, re-

sponding to questions about their practices and perceptions of the activity and emotional and cognitive state at the time. Our use of questionnaires to assess dimensions of engagement was loosely based on this scheme, although employing a subset of dimensions we felt most relevant to this graded read- ing activity. Each questionnaire (see Appendix 1) included two statements related to each of four di- mensions of engagement ─interest, enjoyment, concentration, effort─ and one statement on “overall”

engagement. The selection questionnaire also included two statements related to a fifth dimension

─ control. 

Students rated their level of agreement with each statement (e.g., “I enjoyed reading the story.”) on a six

^-

point Likert scale (Table 1).

A mean value for each pair of responses was calculated to yield a single value for that dimension from each questionnaire. Including the single item for “overall” engagement, each questionnaire thus yielded five engagement values ─ the four engagement dimensions plus “overall” ─ along with a fifth value for control in the selection stage. (Combining item pairs (multiple items) to yield single values is a common approach to increasing reliability (Diamantopoulos, Sarstedt, Fuchs, Wilczynski, & Kaiser, 2012 ; Gliem & Gliem, 2003 ; Sarstedt & Wilczynski, 2009), but see related note in Weaknesses, below). 

As mentioned above, of the 53 enrolled students, only 32 completed all three questionnaires for at least two self

^-

selected readers and two group

^-

selected readers, so only these subjects and their four sets of data were included in the analysis. 

Results

Composite scores (across the two weeks) for each selection method were calculated for each en- gagement dimension (interest, enjoyment, concentration, effort, overall) at each stage (selection, read- ing, discussion) for each student (Table 2). The scores for the two selection methods were then com- pared for each stage using 16 paired

^-

samples t

^-

tests (Table 3). An alpha (significance criterion) of .00156 was set after a Bonferroni correction was applied to avoid spurious “false positive” (Type 1) er- rors resulting from running multiple comparisons (α = .05 ÷ 16 (t

^-

tests) ÷ 2 (2

^-

tailed test) =

Table 1. Six

^-

point Likert scale used for all online questionnaire items.

1 2 3 4 5 6

Strongly

disagree Moderately

disagree Mildly

disagree Mildly

agree Moderately

agree Strongly

agree

はまらない全く当て 少し当て

はまらない やや当て

はまらない やや当て

はまる 少し当て

はまる とても当て はまる

.00156). t

^-

tests were run using a confidence interval of 0.996875 (1 − (.05 ÷ 16)). 

Mean values for all dimensions of reported engagement in all three stages were in the “moderate”

range (5 out of 6 on the 1

^-

6 Likert scale), with no statistically significant differences at the strict, α = .0167 level, and only small (0.2<d<0.5) to negligible (d<0.2) effect sizes (Cohen, 1988).

Preliminary discussion

The lack of significant differences in the selection and reading stages was somewhat surprising, as we expected students to find greater enjoyment and satisfaction in self

^-

selecting and reading graded readers appropriate to their own levels and interests ─ a cornerstone of many extensive reading pro- grams (Day, 2015). However, their similarly moderate levels of engagement under the group selec- tion condition may indicate another, social aspect of engagement. Looking for a good book together

Table 2.  Descriptive statistics for engagement scores in each dimension by reader selection method and activity stage (N=32)

Activity Stage Engagement Dimension

Selection Method

Self Group

Mean SD SE Mean SD SE

Selection

Interest 5.03 0.718 0.127 5.08 0.544 0.096

Enjoyment 4.93 0.736 0.130 5.08 0.587 0.104

Concentration 5.06 0.778 0.137 5.02 0.607 0.107

Effort 5.10 0.790 0.140 5.06 0.496 0.088

Control* 5.11 0.675 0.119 4.73 0.835 0.148

Overall engagement 4.75 1.047 0.185 4.77 0.695 0.123

Reading

Interest 5.07 0.773 0.137 4.99 0.839 0.148

Enjoyment 4.92 0.869 0.154 4.89 1.028 0.182

Concentration 4.73 0.957 0.169 4.55 1.110 0.196

Effort* 4.89 0.657 0.116 5.09 0.751 0.133

Overall engagement 4.92 1.048 0.185 5.05 1.110 0.196

Discussion

Interest 5.17 0.661 0.117 5.29 0.565 0.100

Enjoyment 5.19 0.619 0.109 5.23 0.672 0.119

Concentration 5.34 0.662 0.117 5.40 0.595 0.105

Effort 5.08 0.771 0.136 5.20 0.706 0.125

Overall engagement* 4.91 0.856 0.151 5.09 0.865 0.153

  *Referenced in Exploration section, below.

and negotiating a choice may have had its own rewards, and a sense of responsibility to comprehend and be ready to discuss jointly chosen books may have contributed to focus/concentration while read- ing.

The slightly stronger (though not statistically significant) reported engagement in discussion of group

^-

selected readers was surprising, as well. In

^-

class discussion of self

^-

selected readers seemed, from the researchers’ point of view, to flow more easily ─ with fewer pauses and silences ─ as stu- dents took turns sharing about their readers. On the other hand, students may have had some diffi- culty understanding each other’s English comments on unfamiliar, self

^-

selected readers, and felt en- gaged as they shared views on mutually understood stories.

It should be noted that no efforts were made to employ well

^-

established methods to enhance discus- sion of group

^-

selected readers, such as assigning of discussion roles, etc. (Mark Furr, 2007). Employ- ment of these strategies might have tipped the scales further in favor of group

^-

selection. On the other hand, while not prototypical discussions, there are many strategies for engaging interaction based on individually selected readers (Bamford & Day, 2004 ; Dubravcic, 1996).

Table 3.  Results of t

^-

test comparisons of engagement scores in each dimension by reader selection method for each activity stage

Activity

Stage Engagement Dimension t df p Mean diff.

(Self

^-

Group) SD diff. SE diff.

99.6875%

Confidence Interval Cohen’s

d Lower Upper

Selection

Interest −0.399 31 0.693 −0.047 0.664 0.117 −0.423 0.329 −0.071 Enjoyment

¹

−1.351 31 0.187 −0.148 0.622 0.110 −0.501 0.204 −0.239 Concentration 0.333 31 0.741 0.039 0.663 0.117 −0.337 0.415 0.059

Effort 0.316 31 0.754 0.039 0.699 0.124 −0.357 0.435 0.056

Control* 2.041 31 0.050 0.375 1.040 0.184 −0.214 0.964 0.361

Overall engagement −0.103 31 0.918 −0.016 0.857 0.151 −0.501 0.470 −0.018

Reading

Interest 0.573 31 0.571 0.078 0.771 0.136 −0.359 0.515 0.101

Enjoyment 0.208 31 0.837 0.031 0.851 0.151 −0.451 0.514 0.037 Concentration 1.401 31 0.171 0.172 0.694 0.123 −0.221 0.565 0.248 Effort* −2.07 31 0.047 −0.203 0.555 0.098 −0.518 0.111 −0.366 Overall engagement −0.903 31 0.373 −0.125 0.783 0.138 −0.569 0.319 −0.160

Discussion

Interest −0.995 31 0.327 −0.117 0.666 0.118 −0.494 0.260 −0.176

Enjoyment −0.494 31 0.625 −0.047 0.537 0.095 −0.351 0.257 −0.087

Concentration −0.839 31 0.408 −0.055 0.369 0.065 −0.264 0.154 −0.148

Effort −1.269 31 0.214 −0.125 0.557 0.099 −0.441 0.191 −0.224

Overall engagement* −1.531 31 0.136 −0.188 0.693 0.122 −0.580 0.205 −0.271

  *Referenced in Exploration section, below.

Preliminary conclusion

Students reported finding all stages of the activity ─ selection, reading and discussion ─ moderate- ly engaging (5/6 on a 1

^-

6 Likert scale) under both conditions ─ self

^-

and group

^-

selection. This was encouraging, indicating that this type of graded reader activity has promise as a useful format moving forward. The lack of statistically significant differences in student response to self

^-

selection vs. 

group

^-

selection of graded readers, along with the minimal effect sizes, indicate that students found the variations similarly engaging. This suggests that instructors employing similar graded reading activi- ties with similar groups of students might base their self

^-

vs. group

^-

selection choices on consider- ations other than student engagement ─ for example, employing both methods to provide a stimulat- ing variety of activity and discussion format, or being satisfied with self

^-

selection if multiple copies of readers are unavailable.

Exploration

As detailed above, the lack of significant differences and small effect sizes for reported engagement under the self

^-

vs. group

^-

selection conditions is, in itself, a meaningful result. However, as this is an exploratory study, a cautious look (see Tables 3 & 4) beyond the constraints of strict statistical signifi- cance

¹

to examine some of the differences showing relatively large effect sizes may reveal possibilities for further research. 

Selection stage

In the selection stage, the difference in the control dimension shows a small effect in favor of self

^-

selection. This makes intuitive sense, as we would expect students to feel more control in self

^-

selecting readers than when choosing them as a group. It is actually a bit surprising that the differ- ence was not greater. Apparently, even when choosing with a group, students felt meaningfully involved in the decision

^-

making process. 

1 

Wholesale application of the Bonferroni correction

─

used to avoid fishing for or harvesting of serendipitous differences and responsible for the strict α = .00156 criterion used above

─

is a matter of debate in statistics, with many arguing against its use when the results of each test are considered individually (as in this study) and in exploratory studies in which results will be used to form hypotheses for further investigation (the purpose of this “Exploration” section) (Armstrong, 2014 ; Perneger, 1998). 

As such, relaxation of the strict α criterion seems justified here.

Reading stage

In the reading stage, differential engagement in the effort dimension shows a small effect in favor of group

^-

selection. This, again, would be consistent with students feeling a responsibility to read atten- tively in order to competently discuss shared readers.

Discussion stage

Finally, at the discussion stage, overall engagement shows a small effect in favor of group

^-

selection. 

A slight tendency for students to feel more involved and take part more frequently when discussing a mutually understood reader would be understandable. We will return to this point below. 

These hints suggest that more fine

^-

grained investigation, particularly on effort while reading, and en- gagement during discussion could yield greater insight into student response and options for optimizing these activities. 

Further exploration : Success and challenge vs. skill

Besides items on engagement, the weekly questionnaires also contained a pair of items focused on success and four items on challenge vs. skill (i.e., “Are my skills sufficient to meet the challenge of the task ?”) (see Appendix 1). Means were calculated to convert the two success items and four challenge vs. skill items into single values, respectively. These were not included in the initial analysis of en- gagement, because rather than being aspects of engagement, success was seen as a result of engage- ment, and a reasonable balance of challenge vs. skill was seen as a condition for engagement. Howev- er, many researchers see skill (or competency), engagement and success as operating in a feedback loop. Feeling competent encourages one to engage. Engagement leads to success and further rein- forcement of feelings of competence, which contributes to further engagement, and so on (Reschly, 2010 ; Sharrock & Rubenstein, 2019). 

Student responses regarding success and challenge vs. skill (Table 4) do yield some insight into the engagement results for the discussion stage, in particular. 

No significant differences were seen in reports of success or challenge vs. skill for the selection and

reading stages, and effect sizes were negligible. However, the difference in reported success in the

discussion stage showed a small effect in favor of group selection. And feelings of competency (chal-

lenge vs. skill) for the discussion task were higher for the group

^-

selection condition, with a medium ef-

fect size. It is difficult to account for these differences, but they are consistent with the weak trend

toward greater engagement in discussion of group

^-

selected readers, and may reflect the type of posi-

tive feedback loop involving feelings of competency, success and engagement mentioned above. 

Some insight is gained by separating the items used to calculate the combined challenge v. skill val- ues and testing these individually. Of the four items, most of the difference in this dimension was ac- counted for by these two (see Table 5) :

 16 : “It was difficult to express my ideas in English.”

 19 : “It was difficult to understand what my group members said.”

These were “reverse” items, so scores were inverted (score = 7−score) before calculating means, making the items equivalent to “It was not difficult to express my ideas in English” and “It was not dif- ficult to understand what my group members said.” Results for these items in isolation showed small and medium effect sizes, respectively.

Thus, there was a slight, but not statistically significant tendency to struggle more in explaining or discussing self

^-

selected readers, and a clearer trend toward easier understanding of one another when discussing group

^-

selected readers. This ease of understanding helps account for the higher challenge vs. skill* scores in the discussion stage under the group

^-

selection condition, and may have contributed

Table 4. Comparison of “success” and “challenge v. skill” scores by reader selection method

Activity

Stage Dimension

Selection Method

t df p Mean Diff.

(Self

^-

Group) Cohen ’ s

Self Group d

Mean SD Mean SD

Selection Success 4.98 0.798 4.98 0.600 0.059 31 0.954 0.008 0.010

Challenge vs. Skill 3.81 0.830 3.74 0.534 0.558 31 0.581 0.074 0.099

Reading Success 5.30 0.706 5.32 0.576

−0.230 31 0.82 −0.023 −0.041

Challenge vs. Skill 4.52 0.876 4.47 0.848 0.326 31 0.746 0.047 0.058

Discussion Success* 4.31 0.835 4.53 0.897

−1.720 31 0.095 −0.219 −0.305

Chall. vs. Skill* 3.86 0.739 4.24 0.742

−2.83 31 0.008 −0.379 −0.500   *Referenced below.

Table 5. Comparison of “challenge v. skill” scores by reader selection method

Activity Stage Item

Selection Method

t df p Mean Diff.

(Self

^-

Group) Cohen ’ s Self Group d

Mean SD Mean SD

Discussion

It was (not) difficult to ex-

press my ideas in English. 2.08 0.853 2.45 1.370

−1.822 31 0.078 −0.375 −0.322

It was (not) difficult to un-

derstand what my group

members said. 4.61 0.998 5.25 0.730

−3.073 31 0.004 −0.641 −0.543

to the kind of virtuous cycle supporting engagement mentioned above. 

Weaknesses of the study

This was an exploratory study conducted in parallel with instructors at 20 universities in Japan. 

The main purpose of the program was to help participating instructors gain an understanding of basic concepts and procedures in quantitative research and identify possibilities for further study. As such, some shortcuts were taken to simplify the study design and accomplish the training and research in a reasonable amount of time. 

One such shortcut involved the development of the paired questionnaire items for each engagement dimension. Ideally, these would have been trialed, with a measure─such as Cronbach’s Alpha (Taber, 2018) ─ applied to determine how well responses to each pair correlated (i.e., indicating how well the concepts dovetailed in the minds of students). Poorly correlated items would then be modified and tested again, all prior to beginning the study proper. This step was omitted to focus on a smaller number of quantitative concepts and keep the project within workable time constraints. 

Another issue is the considerable debate on the use of ordinal, Likert scale values in the calculation of means, and their subsequent use in parametric tests, such as a t

^-

test (Kostoulas, 2020 ; Sullivan &

Artino, 2013). An examination of the distribution characteristics of the Likert results would have been prudent before further statistical analyses, but again, this was not deemed practical given the scope and time constraints of the project. 

Given the lack of existing data (TOEIC scores, etc.) on individual student level and practical (time, cost, complexity) barriers to implementing reliable measures, English proficiency was also not ac- counted for in the study. Xreading comprehension quiz scores and self

^-

reports on Graded Reading Records (see Appendix 2) indicated that participants had little or no trouble finding readers at suitable levels. On the other hand, proficiency data might have helped reveal, for example, relatively greater engagement among advanced readers reading higher level books with more complex, interesting plots, or that students with stronger speaking and listening skills found participation in the group discussion sessions easier and thus more engaging. The lack of published research testing these suppositions suggests they could be fertile topics for future study.

Another potential weakness is that the discussion tasks for self

^-

vs. group

^-

selection conditions were

clearly not parallel. Discussion of self

^-

selected readers revolved around students’ taking turns de-

scribing and responding to the books they had read. Discussion of group

^-

selected readers involved

sharing answers to discussion

^-

starter questions and comparing contrasting their responses to the

book. Although the different nature of the discussions may have influenced, student response, these

differences seem unavoidably tied to the treatments themselves. 

Despite the original 10

^-

week design of the experiment, only 32 of 53 participating students submit- ted online questionnaires for all three stages in at least two self

^-

selection weeks and two group

^-

selec- tion weeks. While this did not prevent analysis or invalidate the results, a more robust data set would have been desirable, and we had hoped most participants would submit a full set of questionnaires for four weeks under each condition (eight weeks total). While the instructor provided links to the ques- tionnaires in class and online through the school LMS, a more convenient approach, such as a wallet

^-

sized card containing QR codes for all three questionnaires might have been helpful. Some students also reported that the three weekly 20

^-

item questionnaires, typically taking 2

^-

3 minutes to complete, were burdensome. This may have affected students’ motivation to submit the questionnaires. Rols- tad, Adler, and Rydén (2011) citing Turner, Quittner, Parasuraman, Kallich, and Cleeland (2007) also mention that repeatedly responding to similar questionnaires can contribute to “response burden.”

Over the ten weeks of this study, students were asked to respond to a total of 30 similar question- naires, so this burden may have weighed on response rate and potentially contributed to “straight

^-

line response,” decreasing reliability (although no analysis of this has been done). It may be advisable in a future study to focus on a particular stage of an activity to reduce this burden.

A related issue was that while reading, completing quizzes, filling out summary/response forms and participating in discussions were all explicitly included in the grading rubric for the course (with very close to 100% completion by students), questionnaire submission was not. This left submission up to the good will of students, and participation fell accordingly. While it would be tempting to also require submission of the questionnaires as part of the course rubric, this would bring research ethics con- cerns, and lacking justification on educational grounds, would not be advisable. 

Conclusion

Again, the two conditions ─ self

^-

selection vs. group

^-

selection of graded readers ─ yielded similarly moderate levels of reported engagement, but no statistically significant differences and small to negli- gible effect sizes at each stage of the activity. However, an exploratory examination of the results be- yond strict statistical constraints hints at slight advantages for reading of group

^-

selected readers in the effort domain, and for discussion of group

^-

selected readers in terms of engagement, competency and success. More focused examination of engagement during reading and/or reading

^-

related discussion stages, possibly involving different approaches to discussion, seems warranted.

In sum, subjects responded positively to both approaches, and lacking strong evidence of differential

engagement, instructors working with similar activities in similar contexts can feel free to vary their

use of self

^-

or group

^-

selection depending on practical constraints and/or the type of activities they

want to try.

Acknowledgements

We would like to extend deep thanks to Gregory Sholdt at Konan University for organizing and lead- ing the 2018 Quantitative Research Training Project. His instruction, kind advice and tireless support made this research possible. We would also like to thank the other participants in the project for their contributions and encouragement throughout the process. The project was supported by a Grant

^-

in

^-

Aid for Scientific Research ─ “Development of a Second Generation Research Training Program for Language Teachers” (JSPS KAKENHI Grant No. 16K02920) ─ for which we are also thankful.

Appendix 1 : Items from the three weekly online questionnaires ─ Selecting, Read- ing, Group Discussion

Students received print and electronic versions of a “Weekly Questionnaires Guide” with explana- tion and links (QR code and URL) for each of the three weekly questionnaires. On the actual ques- tionnaires, items were ordered following the Item # on the right. R

^-

designated items (e.g., Q13R) were worded negatively, and scores were reversed (R score = 7−score) to yield results consistent with the positively worded items. For example, for item Q13R (Interest b.) below (I felt bored while looking for a good reader), a response of 1(strongly disagree) was converted to a 6, indicating high inter- est and thus consistent with responses to item Q5 (Interest a. It was interesting to browse the readers while making a choice). 

1 2 3 4 5 6

Strongly

disagree Moderately

disagree Mildly

disagree Mildly

agree Moderately

agree Strongly

agree

はまらない全く当て 少し当て

はまらない やや当て

はまらない やや当て

はまる 少し当て

はまる とても当て はまる

Selecting the Graded Reader Item #

Interest

a. It was interesting to browse the readers while making a choice. Q5 次に読む本を選ぶために幾つかの本に目を通すことは面白かった。

b. I felt bored while looking for a good reader. Q13R

本を探すことはつまらなかった。

Enjoyment

a. I enjoyed the process of selecting this reader. Q19

本を選ぶ過程が楽しかった。

b. I felt irritated during the selection process. Q8R

本を選ぶ過程にいらいらした。

Concentration

a I was focused on finding a good reader during the selection time. Q6 本を選ぶことに集中できた。

b. I found my mind wandering while choosing the reader. Q11R

本を選んでいるときに違うことを考えてしまった。

Effort

a. I tried hard to find a good reader. Q12

いい本を選ぶよう努力した。

b. I was mostly interested in finishing the selection process as quickly as possible. Q15R できるだけ早く終われるように本を適当に選んだ。

Control

a. I feel I had little control over the choice of the reader. Q7R 本の選択に関し，自分の意思がほとんど反映できていないと感じた。

b. My preferences were an important part of the selection process. Q16 自分の好み通りに選ぶことができた。

Success

a. I successfully completed the task of selecting this reader. Q17 私はうまくこの本を選ぶ作業を完了した。

b. I am looking forward to reading this story. Q9

この本を読むのを楽しみにしている。

Challenge vs. Skill

a. Selecting the graded reader was a challenging task. Q10R

本を選ぶのが難しかった。

b. I felt I had sufficient English ability to successfully select the reader Q18 本をスムーズに選ぶ英語力が自分にはあると感じた。

c. It was easy to find a reader that I wanted to read Q14

読みたい本を探すのは簡単だった。

d. I had trouble understanding the English descriptions of the stories Q20R

本についての英語の説明を理解することは難しかった。

Engagement

a. I felt engaged in the task of selecting a reader. Q21

本を選ぶことに没頭した。

Reading the Graded Reader Item #

Interest

a. I felt bored while reading the story. Q18

この本を読むのは退屈だった。

b. The plot of the story was interesting. Q6

話の筋は面白かった。

Enjoyment

a. I enjoyed reading the story.  Q14 読んでいて，楽しかった。

b. I became immersed in the story while reading. Q8

読みながら，段々話に入り込んだ。

Concentration

a. My mind was wandering while I was reading Q15R

読みながら，ボーっとする時もあった。

b. While I was reading, I stayed focused on the task Q11

読んでいる間ずっと話に集中した。

Effort

a. I did my best to finish the reader by the deadline. Q16

締め切りまでに読み終わろうと頑張った。

b. I put a lot of effort into this assignment Q19

今回の課題にかなり努力した。

Success

a. I did everything that I was assigned to do Q12

課題のタスクをすべて完了した。

b. I feel good that I could read a whole book in English. Q7 英語 1 冊の本を読み終わったことにたいして満足感がある。

Challenge vs. Skill

a. The language was easy to understand.  Q20 この本の英語は分かりやすかった。

b. I could read at a fast, steady pace.  Q10

一定の速度で読むことができた。

c. The plot of the story was difficult to follow.  Q9R 話の筋は分かりにくかった。

d. I often stopped for unknown words.  Q17R 知らない単語を調べるために何度も止まった。

Engagement

a. I felt engaged in the reading activity.  Q13 読むことに没頭した。

Group Discussion Item #

Interest

a. My group’s discussion was interesting. Q7

私のグループのディスカッションは面白かった。

b. I felt bored during the discussion. Q17R

ディスカッションは退屈だった。

Enjoyable

a. Overall, I enjoyed discussing the story. Q15

全体として本の内容のディスカッションは楽しかった。

b. It was fun to hear what other students in my group thought.  Q11 色んなグループメンバーの意見を聞いて楽しかった。

Concentration

a. I was focused on understanding what all of my group members were trying to say. Q9 他のメンバーの話している内容を集中して聞き取ろうとした。

b. My mind was wandering during our discussion. Q12R

ディスカッション中はあまり集中できなかった。

Effort

a. I did my best to express my opinion about the reader. Q13 本に関しての自分の意見を発言しようと努力した。

b. I tried hard to contribute to the discussion. Q6

ディスカッションに貢献しようと頑張った。

Success

a. I was an active participant in the discussion. Q14

私はディスカッションに積極的に参加することができた。

b. I brought some good ideas into our discussion. Q18

私はディスカッション中，良い意見を発言できた。

Challenge vs. Skill

a. It was difficult to express my ideas in English. Q16R

ディスカッション中，英語で発言することは難しかった。

b. It was difficult to understand what my group members were saying.  Q19R 他人の話している内容をあまり理解できなかった。

c. I had sufficient English ability to discuss the book with my group.  Q8 他のメンバーと本の内容に関して英語で十分にディスカッションできた。

d. I had trouble finding opportunities to add my ideas to the discussion. Q10R ディスカッション中に中々自分の意見を言い出せなかった。

Engagement

a. I felt engaged in the discussion activity. Q20

ディスカッションに没頭した。

Appendix 2 : Summary & response report ─ Graded Reading Record

After reading each book and taking an online comprehension quiz, students filled out this brief sum- mary and response form to help prepare for the discussion in the following class meeting. The form below contains a slightly edited example offered to the students at the beginning of the study. 

Graded Reading Record Name : Haruka Suzuki SN : 17HA555

Title Publisher

（出版社）

Xreading

Level Date Completed

（月/日）

Reading

（時間

time :

分）

1

^-

Quite Easy 2

^-

My level 3

^-

Too difficult

1

^-

Good 2

^-

So

^-

so 3

^-

Poor

Next Door to Love Cambridge 4 5/29 1 : 58 2 1

Summary (1

^-

2 sentences). Tell another student about the book. What’s it about ? What happens ? Stella falls in love with her new neighbor, Tony, and his daughter, Daisy. When Tony’s ex

^-

wife moves to a new town, Stella and Tony decide to move there also, to be near Daisy.

Response : (3

^-

4 sentences) Tell another student your response to the book (feelings, opinions, etc.) Did you like it ? Do you recommend it ? Why/Why not ? What did you think about while reading ? What experiences/memories did it bring to mind ? Any comments ?

I really like this book. I thought that it is difficult to love someone. I was surprised that Tony and Stella moved to Scotland, but I was glad they can always see Tony s daughter.

Comments, vocabulary, questions, etc.

Appendix 3 : Book talk (discussion) sheets ─ Self

^-

selection, Group

^-

selection

These forms were used to help support students through the discussion stage and facilitate reader selection and questionnaire completion following.

Schmidt & Lee. p. 17

A

Appppeennddiixx 33:: B Booookk ttaallkk ((ddiissccuussssiioonn)) sshheeeettss -- SSeellff--sseelleeccttiioonn,, G Grroouupp--sseelleeccttiioonn

These forms were used to help support students through the discussion stage and facilitate reader selection and questionnaire completion following.

Bo ok Ta lk – G roup Selec tion #1 Name: _____ _________ ___ SN: _______ Date: __/__

Group _____ members:

• _ __ ___ __ ___ ____ ___ ___ ____ • __ __ __ _____ ___ ____ ___ ___ _ • _ __ ___ __ ___ ____ ___ ___ ____ • __ __ __ _____ ___ ____ ___ ___ _ • _ __ ___ __ ___ ____ ___ ___ ____ • __ __ __ _____ ___ ____ ___ ___ _

Bo ok Ti tle: _ __ __ ___ __ ___ __ ___ _____ _____ ____ ____ _____ ____ _

Discussion Questions:●How did you feel about the book? How did you like it? Why?●What was your favorite character? Why? ●Do you have any other thoughts or comments about the book? Discussion:

Ta lk w ith your gr oup ab out th e bo ok you re ad . Ma ke a co nve rsa tio n. Yo u can us e the qu est ion s abov e, b ut you c an a lso a sk any q ue stio ns yo u lik e, m ake c omm en ts ab ou t t he b ook , te ll sto rie s, e tc. You wil l sp ea k fo r ab ou t 1 0 m inu tes w ith yo ur gr ou p. Try t o spe ak in En glis h as muc h as possib le. Y ou w ill h av e so m e ti m e to spea k in Ja pa nese afte r yo ur Eng lish disc ussi on .

●

Ta ke som e ke yw ord not es o n the d isc uss ion. You do n’t ne ed to wri te ev eryt hing —j ust some ke ywo rds th at will hel p you re m embe r y our c onve rsa tion. ___ ___ ___ __ ___ ___ ____ ___ __ ___ ___ ____ ___ __ __ ____ ____ ___ ___ __ ___ ___ ____ ___ __ ___ ___ ____ ___ __ _ ___ ___ ___ __ ___ ___ ____ ___ __ ___ ___ ____ ___ __ __ ____ ____ ___ ___ __ ___ ___ ____ ___ __ ___ ___ ____ ___ __ _ ___ ___ ___ __ ___ ___ ____ ___ __ ___ ___ ____ ___ __ __ ____ ____ ___ ___ __ ___ ___ ____ ___ __ ___ ___ ____ ___ __ _ ___ ___ ___ __ ___ ___ ____ ___ __ ___ ___ ____ ___ __ __ ____ ____ ___ ___ __ ___ ___ ____ ___ __ ___ ___ ____ ___ __ _

DDiissccuussssiinngg tthhee BBooookk アアンンケケーートト:

Fill th is ou t a fte r di scussin g the book wit h yo ur gro up.

-------------------------------------------------------------------------- Next week’s assignment:Individual Selection #2: 5/30-6/06

Th is wee k, you wi ll fr ee ly c hoose a book to read . T he Xr ead in g syst em wil l

not

a llo w y ou to choo se th e sa me bo ok as a class m ate, so E V ER Y ON E w ill r ead a D IF FE RE N T b oo k. A s alwa ys, fil l o ut th e Se lec ting an d Bo ok

アンケート

, Rea ding a B oo k アン

ケート,

an d you r GRR and PRL. Also ta ke th e qu iz a nd ra te th e b ook o n Xr ea ding. Ne xt we ek , you wil l te ll a ne w g roup a bout th is bo ok. You wi ll t alk ab out these qu est io ns /top ic s:

●What’s the book about?●How did you like it? Why?●What was the most interesting thing (event, character, scene, idea) in this book?

Be read y to disc us s thes e to pics with you r next grou p. N ex t w eek yo u w ill a lso ch oo se a ne w bo ok to re ad w ith yo ur g rou p ( G rou p Selection). Please prepare by finding 1-2 new book s to sugge st to the gro up

「この本を読みましょう。」.

Le ve ls 2-4 mig ht be OK.

1. ___________________________________________________________________________

2. ___________________________________________________________________________ After Discussion

After Selection

After Reading

References

Amelsvoort, M.V. (2017). Extensive reading onboarding : Program design for increasing engagement. 

Juntendo Journal of Global Studies, 2, 98

^-

106. 

Armstrong, R.A. (2014). When to use the Bonferroni correction. Ophthalmic and Physiological Optics, 34(5), 502

^-

508. https : //doi.org/10.1111/opo.12131

Baharuddin Marji, A., Rafik

^-

Galea, S., & Mei Yuit, C. (2015). Evaluating the effects of extensive reading circles on ESL learners’ oral communicative competence. Malaysian Journal of Languages and Linguistics, 4(1), 75

^-

92. https://doi.org/10.24200/mjll.vol4iss1pp75

^-

92

Bamford, J., & Day, R. (Eds.) (2004). Extensive reading activities for language teaching. Cambridge : Cambridge Unversity Press.

Beglar, D., & Hunt, A. (2014). Pleasure reading and reading rate gains. Reading in a Foreign Language, 26(1), 29

^-

48.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, N.J. : L. Erlbaum Asso- ciates.

Csikszentmihalyi, M. (1998). Finding flow : The psychology of engagement with everyday life. New York : Basic Books.

Csikszentmihalyi, M., & Rathunde, K. (1993). The measurement of flow in everyday life : Toward a the- ory of emergent motivation. In Nebraska Symposium on Motivation, 1992 : Developmental perspec- tives on motivation (pp. 57

^-

97). Lincoln, NE, US : University of Nebraska Press.

Day, R., & Bamford, J. (1998). Extensive reading in the second language classroom. Cambridge : Cam- bridge University Press.

Day, R. (2015). Extending extensive reading. Reading in a Foreign Language, 27(2), 294

^-

301. Re- trieved from https://files.eric.ed.gov/fulltext/EJ1078441.pdf

Diamantopoulos, A., Sarstedt, M., Fuchs, C., Wilczynski, P., & Kaiser, S. (2012). Guidelines for choosing between multi

^-

item and single

^-

item scales for construct measurement : a predictive validity perspective. Journal of the Academy of Marketing Science, 40(3), 434

^-

449. https://doi.org/10.1007/

s11747

^-

011

^-

0300

^-

3

Dincer, A., Yes¸ilyurt, S., Noels, K.A., & Vargas Lascano, D.I. (2019). Self

^-

determination and classroom engagement of EFL learners : A mixed

^-

methods study of the self

^-

system model of motivational development. SAGE Open, 9(2). https://doi.org/10.1177/2158244019853913

Dubravcic, J. (1996). Telling the truth about extensive reading. The Language Teacher, 20(12), 52. 

Egbert, J. (2003). A study of flow theory in the foreign language classroom. The Modern Language Jour- nal, 87(4), 499

^-

518. https://doi.org/10.1111/1540

^-

4781.00204

Ellis, R. (2005). Principles of instructed language learning. System, 33(2), 209

^-

224. https://doi.

org/10.1016/j.system.2004.12.006

Fredricks, J.A., & McColskey, W. (2012). The measurement of student engagement : A comparative anal- ysis of various methods and student self

^-

report instruments. In S.L. Christenson, A.L. Reschly, &

C. Wylie (Eds.), Handbook of research on student engagement (pp. 763

^-

782). Boston, MA : Springer US. https://doi.org/10.1007/978

^-

1

^-

4614

^-

2018

^-

7_37

Furr, M. (2007). Oxford Booklworms Club reading circles : Teacher’s handbook. Oxford, Oxford Universi- ty Press.

Furr, M. (2007). Reading circles : Moving great stories from the periphery of the language classroom to its centre. The Language Teacher, 31(5), 15

^-

18. 

Gliem, J.A., & Gliem, R.R. (2003). Calculating, interpreting, and reporting Cronbach’s Alpha reliability coef-

ficient for Likert

^-

type scales. Paper presented at the 2003 Midwest Research to Practice Confer-

ence in Adult, Continuing, and Community Education. Retrieved from https://scholarworks.iupui.

edu/bitstream/handle/1805/344/Gliem+&+Gliem.pdf?sequence=1

Goss, P. & J. Sonnemann (2017). Engaging students : Creating classrooms that improve learning. Victo- ria, Australia, Grattan Institute. Retrieved from https://apo.org.au/node/72749

Guthrie, J.T., Klauda, S.L., & Ho, A.N. (2013). Modeling the relationships among reading instruction, mo- tivation, engagement, and achievement for adolescents. Reading Research Quarterly, 48(1), 9

^-

26. 

https://doi.org/10.1002/rrq.035

Hu, M., & Nation, P. (2000). Unknown vocabulary density and reading comprehension. Reading in a Foreign Language, 13. 

Hunt, A., & Beglar, D. (2005). A framework for developing EFL reading vocabulary. Reading in a For- eign Language, 17(1), 23

^-

59.

Jeon, E.Y., & Day, R.R. (2016). The effectiveness of ER on reading proficiency : A meta

^-

analysis. Read- ing in a Foreign Language, 28(2), 246

^-

265. Retrieved from http://nflrc.hawaii.edu/rfl/October2016/

articles/jeon.pdf

Kostoulas, A. (2020). On Likert scales, ordinal data and mean values. Retrieved October 22, 2019 from https://achilleaskostoulas.com/2013/02/13/on

^-

likert

^-

scales

^-

ordinal

^-

data

^-

and

^-

mean

^-

values/#Likert Krashen, S.D. (2004). The power of reading : Insights from the research (2nd ed.). Libraries Unlimited.

Krashen, S.D. (2007). Extensive reading in English as a foreign language by adolescents and young adults : A meta

^-

analysis. International Journal of Foreign Language Teaching, 3(2), 23

^-

29. Re- trieved from http://ijflt.com/

Kuh, G.D., Cruce, T.M., Shoup, R., Kinzie, J., & Gonyea, R.M. (2008). Unmasking the efects of student engagement on first

^-

year college grades and persistence. The Journal of Higher Education, 79(5), 540

^-

563. https://doi.org/10.1080/00221546.2008.11772116

Laufer, B. (1989). What percentage of text

^-

lexis is essential for comprehension ? In C. Lauren & M. 

Nordman (Eds.), Special language : From humans thinking to thinking machines (pp. 316

^-

323). 

Clevedon : Multilingual Matters.

Lee, J.

^-

S. (2014). The relationship between student engagement and academic performance : Is it a myth or reality ? The Journal of Educational Research, 107(3), 177

^-

185. https://doi.org/10.1017/

S0958344010000273

Liu, N., & Nation, I.S.P. (1985). Factors affecting guessing vocabulary in context. RELC Journal, 16(1), 33

^-

42. https://doi.org/10.1177/003368828501600103

Maher, K.M. (2015). EFL literature circles : Collaboratively acquiring language and meaning. The Lan- guage Teacher, 39(4), 9

^-

12. Retrieved from http://jalt

^-

publications.org/files/pdf/the_language_

teacher/39.4tlt.pdf#page=11

Maley, A. (2008). Extensive reading : Maid in waiting. In B. Tomlinson (Ed.), English language learning materials : A critical review (pp. 133

^-

156). London : Continuum.

McLean, S., & Rouault, G. (2017). The effectiveness and efficiency of extensive reading at developing reading rates. System, 70, 92

^-

106. https://doi.org/10.1016/j.system.2017.09.003

Milliner, B., & Cote, T. (2014). Effective extensive reading management with Xreading. The Language Teacher, 38(6), 32

^-

33. Retrieved from https://www.researchgate.net/publication/283333275_Effective_extensive_

reading_management_with_Xreading

Milliner, B., & Cote, T. (2015). One year of extensive reading on mobile devices : engagement and impres- sions. Paper presented at the 2015 EUROCALL Conference, Padova, Italy. http://dx.doi.

org/10.14705/rpnet.2015.000366

Minkowitsch, M. (2013). Motivation and engagement in an extensive reading project. Mag. phil.(Diplo- marbeit). University of Vienna, Vienna. Retrieved from https://othes.univie.ac.at/30417/

Nakanishi, T. (2015). A meta

^-

analysis of extensive reading research. TESOL Quarterly, 49(1), 6

^-

37. 

https://doi.org/10.1002/tesq.157

Nation, P., & Wang, K.M.

^-

t. (1999). Graded readers and vocabulary. Reading in a Foreign Language, 12(2), 355

^-

380. Retrieved from http://nflrc.hawaii.edu/rfl/PastIssues/rfl122nation.pdf

Ng, C., & Bartlett, B. (2017). Improving reading and reading engagement in the 21st century : International research and innovation. Singapore : Springer.

Nishizawa, H., Yoshioka, T., & Fukada, M. (2010). The impact of a 4

^-

year extensive reading program. Paper presented at the JALT2009 Conference Proceedings, Tokyo. Retrieved from http://jalt

^-

publications.

org/archive/proceedings/2009/E035.pdf

Perneger, T.V. (1998). What’s wrong with Bonferroni adjustments ? BMJ (Clinical research ed.), 316(7139), 1236

^-

1238. https://doi.org/10.1136/bmj.316.7139.1236

Renandya, W.A., & Jacobs, G.M. (2016). Extensive reading and listening in the L2 classroom. In W.A.

Renandya & P. Handoyo (Eds.), English language teaching today (pp. 97

^-

110). New York : Rout- ledge.

Reschly, A.L. (2010). Reading and school completion : Critical connections and Matthew effects. Read- ing & Writing Quarterly, 26(1), 67

^-

90. https://doi.org/10.1080/10573560903397023

Rolstad, S., Adler, J., & Rydén, A. (2011). Response burden and questionnaire length : Is shorter better ? A review and meta

^-

analysis. Value in Health, 14(8), 1101

^-

1108. https://doi.org/10.1016/j.

jval.2011.06.003

S¸ahin

^-

Kızıl, A. (2014). Blended instruction for EFL learners : Engagement, learning and course satisfaction. JALT CALL Journal, 10, 227

^-

240. Retrieved from https://www.researchgate.net/

publication/269927692_Blended_instruction_for_EFL_learners_Engagement_learning_and_

course_satisfaction

Sarstedt, M., & Wilczynski, P. (2009). More for Less ? A Comparison of Single

^-

item and Multi

^-

item Measures. Die Betriebswirtschaft, 69, 211

^-

227. Retrieved from https://www.researchgate.net/

profile/Marko_Sarstedt/publication/281306739_More_for_Less_A_Comparison_of_Single

^-

item_

and_Multi

^-

item_Measures/links/5623546608ae70315b5a72e6/More

^-

for

^-

Less

^-

A

^-

Comparison

^-

of

^-

Single

^-

item

^-

and

^-

Multi

^-

item

^-

Measures.pdf

Schmidt, K. (2007). Five factors to consider in implementing a university extensive reading program. 

The Language Teacher, 31(5), 11

^-

14. 

Schmitt, N., Jiang, X., & Grabe, W. (2011). The Percentage of Words Known in a Text and Reading Comprehension. The Modern Language Journal, 95, 26

^-

43. https://doi.org/10.1111/j.1540

^-

4781.2011.01146.x

Sharrock, D., & Rubenstein, R. (2019). Chapter 11

─

Student

^-

Centered Practices for Student Mathe- matical Agency and Engagement. In J.A. Fredricks, A.L. Reschly, & S.L. Christenson (Eds.), Handbook of student engagement interventions (pp. 151

^-

168). Academic Press.

Shelton

^-

Strong, S. (2012). Literature Circles in ELT. ELT Journal, 66(2), 214

^-

223. https://doi.

org/10.1093/elt/ccr049

Shernof, D.J., Ruzek, E.A., Sannella, A.J., Schorr, R.Y., Sanchez

^-

Wall, L., & Bressler, D.M. (2017). Stu- dent engagement as a general factor of classroom experience : Associations with student practices and educational outcomes in a university gateway course. Frontiers in Psychology, 8(994). https://

doi.org/10.3389/fpsyg.2017.00994

Sholdt, G. (2018). The 2018 Quantitative Research Training Project : First stage progress report. Ex- tensive Reading in Japan, 11(1), 14

^-

17. 

Sholdt, G. (2019). Sharing experiences from the 2018 Quantitative Research Training Project. Extensive Reading in Japan, 12(2), 3

^-

5. 

Stroud, R. (2013). Increasing and maintaining student engagement during TBL. Asian EFL Journal, 59,

28

^-

57. Retrieved from https://www.researchgate.net/publication/321359893_Increasing_and_