 

 

“ '33” LIBRARY
Unrversity

 

 

This is to certify that the
dissertation entitled

INDIVIDUAL DIFFERENCES AND TEXT GENRE
IN L2 FRENCH READING COMPREHENSION

presented by

JULIE A. FOSS

has been accepted towards fulfillment
of the requirements for the

PhD. degree in French, Language & Literature

 

 

 

!

 

 

$71190; gill/a W
Baird/x. 3:) m, ,3

Date

MSU is an Afﬁrmative Action/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K‘IProlecctS-PreslcIRCIDaIeDue.indd

INDIVIDUAL DIFFERENCES AND TEXT GENRE
IN L2 FRENCH READING COMPREHENSION

By

Julie A. Foss

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
French, Language and Literature

2009

C0pyright by
JULIE A. FOSS
2009

ABSTRACT

INDIVIDUAL DIFFERENCES AND TEXT GENRE
IN L2 FRENCH READING COMPREHENSION

By
Julie A. Foss

Reading in another language (L2) is a complex, multidimensional process
dependent upon both reader-based and text-based factors. The purpose of this study was
to investigate the roles of reader-based individual difference variables and of the text-
based variable of genre in reading comprehension in French. The sample included 153
adult learners enrolled in beginning, intermediate and advanced level university French
courses.

Structural equation modeling results provided support for a model in which L1
reading ability and L2 proﬁciency positively inﬂuenced L2 reading (recall)
comprehension, metacognitive knowledge positively inﬂuenced L1 reading ability and
L2 proﬁciency, L2 contact positively inﬂuenced L2 proﬁciency, motivation positively
inﬂuenced L2 proﬁciency, L2 contact and study habits (motivational intensity), and
anxiety positively inﬂuenced study habits. This model accounted for 73% of variance in
L2 reading comprehension.

A comparison of models for narrative and expository comprehension showed that
L2 proﬁciency contributed signiﬁcantly less to comprehension of the narrative genre than
to the expository genre. Furthermore, recall comprehension scores for narrative texts
were signiﬁcantly higher than scores for expository texts. Factorial analysis of variance
(AN OVA) showed a signiﬁcant main effect of genre on comprehension, as well as a

signiﬁcant main effect for text. A signiﬁcant interaction of genre and text with learner

level was also present. While differences in comprehension scores by genre were
signiﬁcant at all three levels of French study, genre effects were more robust among
intermediate and advanced learners than among beginners, and different patterns of
comprehension for each text existed between the advanced learners and the intermediate
and beginning learners.

An additional aim of the study was to assess readers’ affective reactions to
reading each genre in the L2 and to reading the experimental passages. Factorial AN OVA
showed signiﬁcantly greater levels of enjoyment, greater comprehensibility and less
anxiety for the narrative genre in general, as well as for the narrative experimental texts.
Whereas no signiﬁcant differences in perceptions of the utility of the narrative and
expository genres in general existed, participants found the two expository experimental
texts more useful to read than the narrative passages. Reactions to each genre and to the
experimental texts varied signiﬁcantly by learner level, with advanced learners reacting
more positively to both genres as well as to the experimental texts than beginning and
intermediate students. A signiﬁcant interaction between all categories of affective
reactions and text, however, revealed considerable variability in reactions to each of the
four texts. Participant comments suggested that factors such as personal engagement with
the topic, text organization, salience of details, and word and sentence-level features also

affected their reading experience.

ACKNOWLEDGMENTS

I am grateful to a number of people who helped me to complete this project.
Above all, I would like to acknowledge the assistance of my dissertation directors,
Professors Anne Violin-Wigent and Diana Pulido. Professor Violin-Wigent has always
been supportive of my interest in second language acquisition, and her continued
guidance has been essential to my completion of what she aptly characterizes as the
“marathon” of doctoral studies. She has also been a valued and inspiring mentor as a
French teacher. Professor Pulido’s expertise and enthusiasm fostered my interest in
second language reading, and her work in the ﬁeld has been very inspirational to me. Her
knowledge and advice have been indispensable in every aspect of my research. I am
honored to have had the opportunity to work with a scholar of her caliber. The
encouragement and kindness of both of my directors has also been of great support to me
throughout this process.

I would also like to thank Professor Deidre Dawson, who in addition to serving on
my dissertation committee has provided me with unique opportunities to interact with
French students outside the classroom. Working with her has enriched my graduate
experience immensely.

Many thanks are due to Professor David Z. (Zach) Hambrick, who gave
generously of his time to help me with the statistical analyses used in this dissertation.
Without his help, I cannot imagine having completed it successfully, and I am indebted to

him for having taught me so much in such a short period of time.

I am also grateful to Professor Ehsan Ahmed for reading preliminary versions of
this dissertation, as well as for his advice and support throughout my graduate career.

I would also like to acknowledge the assistance of my colleagues and the faculty
in the Department of French, Classics, and Italian, in particular Sandhya Shanker,
Coordinator for French, for their unﬂagging support of my research and help with
participant recruitment. I am also very grateful to the many MSU students who
participated in this research.

Finally, I cannot imagine having arrived at this point in my life or my academic
career without the love and support of my family. My parents, John and Janis Foss, and
my sister, Jennifer, have been my constant champions. It is overwhelming to know that
people whom I love and admire so much are so proud of me. My husband, Eric Mueller,
deserves the biggest thanks I can muster for his love, patience, good humor, and support
in every way throughout my graduate career. I cannot thank him enough for making it

possible for me to do something that makes me so happy.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................ xi
LIST OF FIGURES ....................................................................... xiii
CHAPTER 1: INTRODUCTION ......................................................... 1
1.1 L2 reading ...................................................................... 1
1.2 Purpose of this study .......................................................... 3
1.3 Summary of research questions ............................................. 4
1.4 Signiﬁcance of this study ..................................................... 5
1.5 Deﬁnition of terms ............................................................ 7
1.6 Overview of the chapters ..................................................... 8
CHAPTER 2: LITERATURE REVIEW ................................................ 10
2.1 Individual differences ......................................................... 10
2.1.]. L1 reading ability and L2 proﬁciency .......................... 11
2.1.2 Metacognitive knowledge ......................................... 14
2.1.3 L2 exposure ......................................................... 16
2.1.3.1 Language exposure and L2 development ............ 16
2.1.3.2 Language exposure and L2 reading development... 17
2.1.4 Motivation ........................................................... 20
2.1.4.1 Motivation and L2 learning ........................... 20
2.1.4.2 Motivation and L2 reading ............................ 23
2.1.5 Anxiety .............................................................. 27
2.1.6 Summary ............................................................. 30
2.2 Text genre ....................................................................... 32
2.2.1 Narrative and expository genres ................................. 33
2.2.2 Text genre and L1 reading ........................................ 34
2.2.3 Text genre and L2 reading ........................................ 37
2.2.4 Summary ............................................................ 41
2.3 Conclusion ...................................................................... 42
CHAPTER 3: METHODOLOGY ....................................................... 44
3.1 The pilot study ................................................................ 44
3.1.1 Experimental design .............................................. 44
3.1.2 Research questions ................................................ 44
3.1.3 Hypotheses .......................................................... 45
3.1.4 Method .............................................................. 46
3.1.4.1 Participants ............................................... 46
3.1.4.2 Materials ................................................. 46
3.1.4.3 Measures ................................................. 50
3.1.4.3.] Predictor variables ........................... 50
3.1.4.3.1.1 Text genre ......................... 50

vii

3.1.4.3.l.2 Motivation for French study. . .. SO

3.1.4.3.2 Outcome variable ............................ 51
3.1.4.3.3 Pilot instruments ............................. 53
3.1.4.3.3.1 Demographic questionnaire..... 54
3.1.4.3.3.2 L2 proﬁciency test ............... 54
3.1.4.333 Reading activity questionnaire. 55
3.1.4.3.3.4 Familiarity questionnaire ........ 56
3.1.4.3.3.5 Text reactions questionnaire. . .. 56
3.1.5 Procedure ............................................................ 56
3.1.6 Scoring ............................................................... 57
3.1.6.1 Motivation ................................................ 57
3.1.6.2 Reading comprehension ................................ 58
3.1.7 Results ............................................................... 58
3.1.7.1 Descriptive statistics .................................... 58
3.1.7.1.] Motivation .................................... 58
3.1.7.1.2 Reading comprehension ..................... 60
3.1.7.2 Data analysis ............................................. 61
3.1.7.2.] Research Question 1: Text genre and L2
reading comprehension ................................ 61
3.1.7.2.2 Research Questions 2 and 3: Motivation
and L2 reading comprehension ........................ 62
3.1.7.3 Pilot instruments ........................................ 64
3.1.7.3.] L2 proﬁciency test ........................... 65
3.1.7.3.2 Reading activity questionnaire ............. 66
3.1.7.3.3 Familiarity questionnaire ................... 68
3.1.7.3.4 Text reactions questionnaire ............... 70
3.1.8 Discussion and implications for the present study ............. 77
3.1.8.1 Text genre and motivation ............................. 77
3.1.8.2 L2 proﬁciency .......................................... 80
3.1.8.3 Reading activity ........................................ 81
3.1.8.4 Topic and genre familiarity ........................... 81
3.1.8.5 Text reactions ........................................... 81
3.2 The present study .............................................................. 82
3.2.1 Experimental design ................................................ 82
3.2.2 Research questions ................................................. 83
3.2.3 Hypotheses .......................................................... 84
3.2.4 Method ............................................................... 87
3.2.4.1 Participants ............................................... 87
3.2.4.2 Materials ................................................. 89
3.2.4.3 Measures ................................................. 92
3.2.4.3.1 Predictor variables ........................... 92
3.2.4.311 Text genre ......................... 92
3.2.4312 L1 reading ability ................. 92
3.2.4.3.1.3 L2 proﬁciency ..................... 92
3.2.4.314 Metacognitive knowledge ....... 93
3.2.4.3.1.5 L2 exposure ....................... 94

viii

3.2.4.316 Motivation and anxiety .......... 95

3.2.4.3.2 Outcome variables ........................... 96
32.4.3.2] Reading comprehension ......... 96
3.2.4.322 Genre reactions ................... 96
3.2.4.323 Text reactions ..................... 97
3.2.5 Procedure ............................................................ 98
3.2.6 Scoring .............................................................. 98
3.2.6.1 Predictor variables ...................................... 98
3.2.611 L1 reading ability ............................ 98
3.2.612 L2 proﬁciency ................................ 99
3.2.613 Metacognitive knowledge .................. 100
3.2.6.1.4 L2 exposure ................................... 100
3.2.615 Motivation and anxiety ..................... 101
3.2.616. Learner level ................................. 101
3.2.6.2 Outcome variables ...................................... 102
3.2.6.2.1 Reading comprehension ..................... 102
3.2.6.2.2 Genre reactions .............................. 102
3.2.6.2.3 Text reactions ................................. 102
CHAPTER 4: RESULTS .................................................................. 104
4.1 Data screening ................................................................. 104
4.2 Descriptive statistics and correlations ....................................... 105
4.2.1 Predictor variables .................................................. 105
4.2.1.1 L2 proﬁciency ........................................... 105
4.2.1.2 Ll reading ability ....................................... 107
4.2.1.3 Metacognitive knowledge .............................. 107
4.2.1.4 L2 exposure .............................................. 108
4.2.1.5 Motivation and anxiety ................................. 110
4.2.2 Outcome variables ................................................. 117
4.2.2.1 Reading comprehension ................................ 117
4.2.2.2 Genre reactions .......................................... 118
4.2.2.3 Text reactions ............................................ 119
4.3 Research Question 1: Individual differences and L2 reading
comprehension ..................................................................... 123
4.3.1 Structural equation modeling ............................ 123
4.3.1.1 Graphic representation of SEM models. . 126
4.3.1.2 Conﬁrmatory factor analyses ................. 127
4.3.1.2.]. Predictor variables .................. 127
4.3.1.2.2. Outcome variable .................. 131
4.3.1.3 Structural model ................................ 131
4.4 Research Question 2: Text genre and L2 reading comprehension ...... 136
4.4.1 Research Questions 2a and 2b: Genre, level and reading
comprehension ............................................................ 136
4.4.2 Research Question 20: Genre and individual differences. . 142
4.5 Research Question 3: Genre and text reactions ........................... 145
4.5.1 Research Question 3a: Genre reactions ......................... 145

ix

4.5.2 Research Question 3b: Text reactions ...........................

4.6 Summary of results ............................................................
CHAPTER 5: DISCUSSION AND IMPLICATIONS .................................
5.1 Research Question 1: Individual differences and L2 reading
comprehension .....................................................................
5.1.1 L1 reading ability and L2 proﬁciency ...........................
5.1.2 Metacognitive knowledge .........................................
5.1.3 L2 exposure ..........................................................
5.1.4 L2 study habits ......................................................
5.1.5 Motivation ...........................................................
5.1.6 Anxiety ...............................................................
5.2 Research Question 2: Text genre and L2 reading comprehension ......
5.2.1 Research Question 2a: Genre effects on L2 reading
comprehension ............................................................
5.2.2 Research Question 2b: Genre and level in L2 reading
comprehension ............................................................
5.2.3 Research Question 2c: Genre and individual differences.
5.3 Research Question 3: Genre and text reactions ...........................
5.4 Methodological implications ................................................
5.5 Pedagogical implications .....................................................
5.6 Limitations .....................................................................
5.7 Directions for future research ................................................
5.8 Conclusion .....................................................................
Appendix A Texts and English translations ......................................
Appendix B Propositional analyses of texts ......................................
Appendix C Motivation for French study questionnaire, pilot study.... . . . . ..
Appendix D Reading activity in French, pilot study ............................
Appendix E Familiarity questionnaire, pilot study ..............................
Appendix F Text reactions questionnaire, pilot study ..........................
Appendix G Mctacognition questionnaire .......................................
Appendix H Demographic and language contact questionnaire ...............
Appendix 1 Motivation and anxiety for L2 reading and study
questionnaire ..........................................................
Appendix J Genre reactions questionnaire .......................................
Appendix K Text reactions questionnaire ........................................
Appendix L Inter-item correlations for motivation and anxiety for L2
study ....................................................................
Appendix M Inter-item correlations for motivation and anxiety for L2
reading ..................................................................
Appendix N Text comments ........................................................
REFERENCES ..............................................................................

150
152

156

156
157
157
159
159
160
161
162

163

164
165
166
169
171
175
177
177

1 79
1 88
200
202
206
207
208
21 1

217
220
221

224

236

Table 1
Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table 10
Table 11
Table 12

Table 13

Table 14

Table 15

Table 16

LIST OF TABLES

Text characteristics, pilot study .............................................. 49
Descriptive statistics and correlations for motivation, pilot study ...... 59

Descriptive statistics for reading comprehension, pilot study
(propositions recalled) ......................................................... 60

Analysis of variance for effect of genre and text on reading
comprehension, pilot study ................................................... 62

Analysis of variance for effect of genre and text on reading
comprehension with motivational orientations as covariates, pilot

study ............................................................................. 64
Descriptive statistics and correlations for French proﬁciency,

pilot study ....................................................................... 66
Descriptive statistics for reading activity in French, pilot study (hours

per week) ........................................................................ 68
Descriptive statistics for topic and genre familiarity, pilot study ....... 70
Descriptive statistics for difﬁculty and enjoyment ratings, pilot

study ............................................................................. 71
Text reaction comments, pilot study ........................................ 74
Participant background characteristics ...................................... 88
Text characteristics ............................................................ 91

Descriptive statistics and correlations for French proﬁciency and

vocabulary size ................................................................ 106
Descriptive statistics and correlations for metacognitive

knowledge ..................................................................... 107
Descriptive statistics and correlations for French exposure (hours per
month) ........................................................................... 108
Descriptive statistics and correlations for motivation and anxiety 115
subscales ........................................................................

xi

Table 17 Descriptive statistics for reading comprehension (propositions
recalled) ......................................................................... 117

Table 18 Descriptive statistics, correlations and reliability for genre
reactions ........................................................................ 119

Table 19 Descriptive statistics, correlations and reliability for text
reactions ........................................................................ 120

Table 20 Model ﬁt statistics and suggested acceptable values ..................... 126

Table 21 Initially speciﬁed measurement model: latent and observed
variables ........................................................................ 128

Table 22 Analysis of variance for effects of genre, text and level on reading
comprehension ................................................................. 137

Table 23 Analysis of variance for effects of learner level on genre reactions. . .. 146

Table 24 Means and post hoc results (Bonferroni) for effects of learner level
on genre reactions .............................................................. 149

Table 25 Analysis of variance of effect of genre and text on text reactions ...... 151

xii

Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7

Figure 8

Figure 9

LIST OF FIGURES

Measurement model for predictor variables ................................ 130
Measurement model for outcome variable ................................. 131
Initial structural model (all hypothesized paths) ........................... 133
Final structural model (signiﬁcant paths only) ............................. 135
Reading comprehension scores by genre and learner level ............... 139
Reading comprehension scores by genre and text ......................... 139
Reading comprehension scores by genre, level and text ................. 141

Structural model for narrative text comprehension (signiﬁcant paths
only) ........................................................................... 143

Structural model for expository text comprehension (signiﬁcant paths
only) ........................................................................... 144

xiii

CHAPTER 1: INTRODUCTION

1.1 L2 reading

Reading is one of the most useful skills for foreign languagel students, whose
opportunities to read in the target language (L2) are typically more plentiful than their
opportunities to interact with speakers of that language. Reading provides language
learners with input that facilitates the development of language proﬁciency and cultural
competence. In addition, reading in another language promotes lifelong language
learning, as L2 reading skills tend to be more stable and better retained than productive
skills like speaking and writing (Bernhardt, 1991). Yet L2 reading is an extraordinarily
complex process dependent upon an array of reader-based, text-based and contextual
factors that are far more varied than those implicated in native language (L1) reading. As
a result, as useful and important as they may be, L2 reading skills are not always easily
acquired.

Reading has long been characterized as an interactive process in which readers
use various types of knowledge and abilities to construct meaning from written input.
Reading in another language is a still more complicated process in that it draws on
knowledge and skills developed in both L1 and L2 experiences. In addition to depending
on knowledge of L2 language structures and vocabulary, success in reading in another
language is also determined in part by native language reading ability (Bernhardt, 2005;
Bernhardt & Kamil, 1995; Bossers, 1991; Brisbois; 1995, Carrel], 1991; Lee & Schallert,

1997). Metacognitive knowledge, or what readers know about reading, developed during

 

' I use foreign language study to denote language study occurring outside of a community where
the target language is spoken (e.g., native English speakers learning French in the United States).
This is in contrast to second language study, which occurs in the target language environment
(e.g., native English speakers learning French in France).

both L1 and L2 experiences has also been shown to inﬂuence L2 reading (Gelderen,
Schoonen, Glopper, Hulstijn, Simis, Snellings, & Stevenson, 2004; Schoonen, Hulstijn,
& Bossers, 1998). Other experiential factors, such as the amount of contact with the
target language (Ellis, 2002), particularly the amount of exposure to print (Constantino,
Lee, Cho, & Krashen, 1997; Elley, 1991; Elley & Mangubhai, 1983; Kim & Krashen,
1997) may also predict reading L2 comprehension outcomes.

Readers in any language, however, bring more to the task than knowledge, ability
and experience. L1 research (Baker & Wigﬁeld, 1999; Gottfried, 1990; Guthrie,
Wigﬁeld, Metsala, & Cox, 1999) has provided evidence that motivational (those related
to reasons for choosing to read or not to read) and affective (those related to emotions)
differences between readers may also affect reading outcomes. Empirical studies have
found strong relationships between motivation, anxiety and various measures of L2
learning, and recent research suggests that these types of non-linguistic variables may
also contribute to L2 reading (Argarnon & Abu-Rabia, 2002; Brantrneier, 2005; Kondo-
Brown; 2006; Mori, 2004; Sellers, 2000; Yamashita, 2004).

Reading also requires an interaction between these myriad reader-based factors
and a particular text, whose structure and content affect how the reader’s knowledge is
deployed and how he or she reacts affectively to the text. L1 reading research has shown
that text type, or genre, is an important text-based predictor of reading comprehension.
Though L1 reading studies (Graesser, Hauft-Smith, Cohen, & Pyles, 1980; Kintsch &
Young, 1984; Kozminsky, 1977; Petms, Norgaard, Olson, & Tabor, 1989; Tun, 1989;
Zabrucky & Ratner, 1992) have provided ample evidence that narrative, or story-based,

texts are easier to recall and understand than expository, or informational, texts, the

effects of these text genres on reading comprehension have rarely been the focus of L2
reading studies. Yet despite the fact that empirical evidence has not demonstrated that
narrative texts are any less accessible to L2 readers, expository, cultural-informational
texts receive far more attention in the foreign language curriculum than narrative, literary
texts (Frantzen, 2002; Paesani, 2004). The exclusion of narrative texts ﬁ'om lower-
division L2 reading curricula may reﬂect instructor beliefs that less-proﬁcient learners
lack the linguistic skills necessary to interpret them (Lee, 1986). Informational,
expository readings may also be seen as more representative of texts that students may
encounter in “real world” language use. Yet narrative texts also serve a variety of real
world pin-poses, including reﬂecting readers’ shared experiences, providing opportunities
to interpret varied forms of discourse, and helping students to better understand the
human condition (Grasesser, Golding, & Long, 1984; Jurasek & Jurasek, 1991; Paesani,

2004).

1.2 Purpose of this study

Reading in a foreign language is thus a multidimensional process involving the
interaction of reader-based knowledge and skills, experiences and affect with text-based
features. Previous L2 reading research has largely investigated these factors separately in
what Koda (2005) deems “single-focus” studies. As Koda points out, however, many
abilities and knowledge somces key to reading comprehension depend on other
component skills, and their contributions therefore cannot be accounted for adequately in
isolation. For this reason, the present study uses a multi-componential approach in order

to identify not only reader and text-based variables that contribute to reading

comprehension, but also interrelationships between these variables and their relative
contributions.

The present study also seeks to determine the relative difficulty of narrative and
expository text genres in L2 reading, as well as the contribution of individual differences
to comprehension of each genre. Though L1 studies have consistently shown that
narrative texts are easier to understand and recall than informational texts, the few L2
studies examining these text genres have reached different conclusions about their
relative difﬁculty. The question of how individual reader differences contribute to
comprehension of each genre in the L2 has not been thoroughly investigated.

Finally, this study examines readers’ affective reactions to the narrative and

expository genres as well as to reading particular texts.

1.3 Summary of research questions

The research questions posed by this study relate to the contributions of individual
reader differences to L2 reading comprehension and to comprehension of the narrative
and expository genres, the effects of these genres on L2 reading comprehension, and the
affective reactions of readers to each genre and to the experimental texts. These
questions, which will be presented in detail in Chapter 3, can be summarized as follows:

1) What are the contributions of readers’ L2 proﬁciency, Ll reading ability,
metacognitive knowledge, L2 exposure, motivation and anxiety to L2 reading
comprehension? What are the relationships between these individual differences in a

model of L2 reading comprehension?

2) Is there a difference in L2 reading comprehension due to text genre (narrative
and expository)? Are there differences in comprehension of each genre by learner level
(beginning, intermediate, advanced)? Are there differences in the contributions of
individual difference variables to comprehension of each genre?

3) What are learners’ affective reactions to reading each genre in the L2? Do they
prefer reading narrative or expository texts in the L2? Are there differences in these
reactions by learner level? What are learners’ affective reactions to reading the texts in

this study? Do these reactions differ by genre, or only by text?

1.4 Signiﬁcance of this study

The ﬁndings of this study are expected to contribute to L2 reading research and
instruction in several ways. First, identifying the relative contributions of individual
differences to reading comprehension, and the interrelationships between these
differences, will help account for previously unexplained variance in comprehension.
Bernhardt’s (2005) review of L2 reading studies concluded that collectively, existing
research accounts for only about 50% of L2 reading comprehension variance, which
includes variance explained by L1 reading ability and L2 knowledge. By examining other
types of reader differences, including those in metacognitive knowledge, language
exposure, motivation and anxiety, the present study should account for a greater
percentage of this variance and should contribute to a more complete model of individual
differences in L2 reading. Expected contributions of L2 proﬁciency, metacogrritive
knowledge and language exposure to L2 reading should also provide support for

interactive theories of second language reading (Carrell, Devine, & Eskey, 1998). These

theories explain L2 reading as an interaction of linguistic decoding processes, which are
dependent on L2 proﬁciency, with processes drawing on readers’ prior knowledge, which
may include prior linguistic knowledge (L2 proﬁciency) and knowledge of texts and
reading (metacognitive knowledge) in addition to general world knowledge.

Beyond L2 reading, the present study will provide additional insights into the role
of individual diﬂ‘erences in complex cognitive processes. The expected contributions of
language exposure within a model of L2 reading comprehension will also provide support
for connectionist theories of second language acquisition, which propose that language
learning is experience-based and is promoted by frequent contact with the L2.

This study’s analysis of individual reader differences will identify knowledge,
abilities and characteristics - or combinations thereof - key to successful L2 reading. This
is an important ﬁrst step in developing pedagogical interventions to help less successful
readers, and in identifying readers who may beneﬁt from such interventions.

This study also seeks to ﬁll the gap in existing L2 text genre research, which has
not examined the relative difﬁculty of narrative and expository texts for L2 readers at
various levels of instruction or the contribution of these genres to reading comprehension.

Results related to comprehension of each text genre should have implications for
reading curriculum design, including identifying the types of texts that can be read most
successfully at beginning, intermediate and advanced levels, and identifying the level of

instructional support required for each text genre.

1.5 Deﬁnition of terms

Individual dzﬁizrences in L2 and psychological research refers to stable, enduring
personal characteristics common to all people, but that differ in degree between people.

L2 proﬁciency is deﬁned as target language knowledge and skills, including
knowledge of grammar and vocabulary and listening and reading skills.

Metacognitive knowledge includes knowledge about particular learning tasks, and
knowledge that allows for monitoring and regulation of these tasks.

Intrinsic motivation refers to reasons for engaging in an activity that are related to
personal pleasure or interests.

Extrinsic motivation refers to instrumental reasons for engaging in an activity that
are not related to intrinsic interest (e.g., to further one’s career).

Amotivation is a motivational orientation where learners do not value an activity,
believe their behavior has no relationship with outcomes related to that activity, and
believe that engaging in the activity is a waste of time.

Narrative refers to a genre of text broadly deﬁned as “stories”. Narrative texts
typically contain characters, episodes or sequences of events, temporal and spatial
placements of events, and temporal or causal relationships between events. Affective
engagement of the reader is one of the principal communicative intents of narrative texts.

Expository refers to informational texts. Expository texts typically convey
information in a static, non-chronological manner. The primary communicative purpose

of expository texts is to inform.

Semantic propositions are units of meaning that make up texts. Each proposition
consists of a predicate and its arguments. According to Kintsch (1998), a reader’s mental
representation of a text is a network of interrelated propositions, known as a textbase.

Structural equation modeling (SEM) is a multivariate statistical analysis approach
that permits relationships between multiple, interrelated variables to be tested. SEM
analyses include latent variables, or theoretical constructs that cannot be directly
observed (e.g., motivation, anxiety, etc.). Instead, latent variables are assessed using
measurable indicators, also known as observed variables. SEM techniques include
conﬁrmatory factor analysis, in which a measurement model that tests whether the latent
variables are satisfactorily measured by their observed indicators is constructed. Then,
structural modeling tests relationships between predictor (exogenous) and outcome

(endogenous) variables, as well as interrelationships between endogenous variables.

1.6 Overview of the chapters

In Chapter 2, I will review previous research that guided the research questions,
hypothesizes and design of the pilot and main experimental studies. To begin, I will
discuss studies on the role of individual differences in L2 reading, including differences
in L1 reading ability, L2 proﬁciency, metacognitive knowledge, L2 exposure, motivation
and anxiety. Next, I will deﬁne narrative and expository texts, and discuss both L1 and
L2 studies that have investigated reading comprehension of both genres.

Chapter 3’s focus is the methodology of the main experimental study. However,
because this methodology was adapted ﬂour a pilot study, I will begin by describing the

pilot study’s research questions, hypotheses, methodology and results, followed by a

discussion of their implications for the main study. Then, I will present the research
questions, hypotheses and methodology of the main experiment, including data scoring
procedures.

Results of the main experimental study are presented in Chapter 4. I will begin
with a discussion of data screening procedures. I will then present descriptive statistics
and correlations for the data. This will be followed by a description of the main statistical
analysis procedure used in the study, structural equation modeling, and ﬁnally the results
of the statistical analyses.

In Chapter 5, I will discuss theoretical, methodological and pedagogical
implications of the results before addressing the study’s limitations, exploring possible

avenues for future research, and presenting general concluding remarks.

CHAPTER 2: LITERATURE REVIEW

In this chapter, I will deﬁne individual dizferences and discuss existing L2 reading
research on the individual difference variables that are the subject of this study. These
include L2 proﬁciency, L1 reading ability, metacognitive knowledge, L2 exposure,
motivation and anxiety. I will also deﬁne the two text genres under investigation in this
study - narrative and expository - and discuss L1 and L2 studies examining the

relationship of text genre to reading outcomes.

2.1 Individual differences

L2 individual difference research has long sought to determine what
characteristics “good” language learners possess (MacIntyrc & Noels, 1994). Broadly
deﬁned, individual diﬁrences include anything that identiﬁes a person as a distinct and
unique human being (Ddrnyei, 2005), however, it is evident that not all of the
innumerable possible differences between individuals are relevant to language learning
(e.g., hair color, religious beliefs, etc.). L2 individual difference research has therefore
concentrated on a subset of relatively broad factors that discriminate between individuals
whose relationships with learning outcomes have been demonstrated in both L1 and L2
contexts. Among these are differences in native language ability, L2 proﬁciency,

metacognitive knowledge, L2 exposure, and affective differences including motivation

and anxiety.

10

2.1.1. LI reading ability and L2 proﬁciency

Questions about relationships between native and target languages are central to
second language acquisition research (Brown, 2007; Ellis, 2008; Gass & Selinker, 2008;
Larsen-Freeman & Long, 1991). Clearly, L2 skills such as reading depend on some
knowledge of the target language. However, since reading in any language involves a
similar set of component skills (for example, word recognition, sentence parsing,
constructing meaning from text content and making inferences beyond the content of the
text, among others) and since most L2 learners have attained some level of reading
competence in their native language before starting to read in the target language, it is
reasonable to expect that the ability to read in one language contributes to reading in
another. Alderson (1984) was among the ﬁrst to wonder whether foreign language
reading is “a reading problem or a language problem” (p. 1); that is, whether the source
of L2 reading difﬁculties is inadequate native language reading skills or insufficient
target language knowledge. After reviewing relevant research, Alderson concluded that
both L2 knowledge and L1 reading skills have important effects on L2 reading
comprehension, though L2 knowledge is the primary contributor. Subsequent empirical
studies have provided strong evidence supporting this conclusion.

Carrell (1991) investigated L1 reading and L2 knowledge effects with two
different groups, adult Spanish Ll-ESL students (n = 45) and adult English Ll-Spanish
L2 students (n = 75). L1 and L2 reading ability were measured by performance on
multiple-choice tests after participants read L1 and L2 expository texts, and L2
knowledge was operationalized by level of instruction. Though Carrell’s reported results

do not identify the individual contributions of L1 reading ability and L2 knowledge,

11

together these variables accounted for 35% of reading comprehension variance among
the native Spanish group and for 53% of the variance in comprehension scores among the
native English group.

Other studies identifying the individual contributions of these variables have
yielded relatively consistent results. Though both are important predictors of L2 reading
comprehension, the contribution of L2 proﬁciency (operationalized variously by learner
level, grammar and vocabulary knowledge, or proﬁciency tests including measures of
grammatical knowledge and reading skills) is more substantial than that of L1 reading
ability.

Using similar methodology and data analysis procedures as Carrel] (1991),
Bossers (1991) examined these variables with intermediate and advanced Turkish Ll-
Dutch L2 learners (n = 50). As measures of L1 and L2 reading ability, participants read
two expository passages each in Turkish and Dutch and answered 16 multiple choice
comprehension questions per text. Knowledge of L2 grammar and vocabulary as
measured by a standardized test battery explained 54% of L2 reading comprehension
variance, and L1 reading ability explained an additional 19%.

In Brisbois’ (1995) study of beginning and upper-level English Ll-French L2
students (n = 131), level of L2 instruction and performance on an L2 vocabulary
translation test accounted for 57% of L2 recall comprehension variance. L1 reading
ability, assessed by an English recall comprehension task, accounted for 7% of variance.

Bernhardt and Kamil (1995) examined L1 and L2 effects on comprehension with
native English-speaking learners of Spanish (n = 187) at beginning, intermediate and

advanced levels of instruction. L1 reading ability was assessed with two standardized

12

reading tests (Nelson-Denny and the Adult Basic Learning Examination, or ABLE), L2
reading comprehension was measured by the Spanish version of the ABLE, and level of
instruction served as an indicator of L2 knowledge. Multiple regression analyses showed
that 30-3 8% of comprehension variance was accounted for by L2 knowledge and 10-16%
was accounted for by native language reading ability.

Lee and Schallert (1997) found a similar pattern of contributions to reading
comprehension among Korean adolescents (n = 809) learning English. L2 proﬁciency as
measured by vocabulary and grammaticality judgment tests accounted for 57% of
variance in scores on standardized L2 reading tests, and scores on standardized tests of
L1 reading explained 30% of L2 reading variance.

F ecteau’s (1999) investigation of the effects of these two variables found, to the
contrary, that L1 reading ability is a more powerful predictor of L2 reading
comprehension than L2 knowledge. However, a methodological anomaly may explain
these divergent results. Intermediate English L1 -French L2 university students (n = 42)
read two narrative texts, one in English and one in French. Text comprehension in both
languages was assessed by both multiple choice and recall measures. While multiple
regression analyses did not ﬁnd a signiﬁcant contribution of either L1 reading ability or
L2 proﬁciency to multiple-choice reading scores, possibly due to a lack of variance in
these scores, together these variables explained 48% of variance in L2 recall
comprehension. Yet in this analysis, only the contribution of L1 reading ability was
statistically signiﬁcant. The lack of signiﬁcant contribution of L2 proﬁciency, however,
appears to have resulted ﬁom the fact that an administrative problem prevented over half

of the participants (24 of 42) from taking the test of language proﬁciency.

l3

Though research on the contributions of L1 reading and L2 proﬁciency has by
and large produced consistent ﬁndings (signiﬁcant contributions of both L1 reading
ability and L2 knowledge to L2 reading, with a greater contribution of L2 knowledge), it
has by no means accounted for all the variance in L2 reading comprehension. Most
studies investigating the effects of L1 reading ability and L2 knowledge without
including other individual reader differences have accounted for only slightly more than
halfof all variance in L2 comprehension. It is clear that other factors also contribute to
the L2 reading process, and must be accounted for in any research seeking to explain
reading comprehension variance.

2.1.2 Metacognitive knowledge

A limitation of studies that use Ll reading as a predictor of L2 reading is a
consequence of the multifaceted nature of the reading process. Schoonen et a1. (1998)
propose that native language reading ability includes both language-speciﬁc knowledge,
which is less likely to transfer to L2 reading, and general knowledge of reading, which is
more likely to transfer. This general knowledge of reading thus must be accounted for in
research examining L1 effects on L2 reading. The researchers identiﬁed the main
component of this general knowledge base as metacognitive knowledge. Broadly deﬁned,
metacognitive knowledge includes “cognition about cognition” (Garner, 1994, p. 716), or
what learners know about learning. This includes both learners’ knowledge about
particular learning tasks, and knowledge that allows them to monitor and regulate their
progress (Wenden, 1999). Schoonen et al.’s (1998) deﬁnition underscores this twofold
nature of metacognitive knowledge, which includes “knowledge about one’s cognition

(‘knowing that’) and about the regulation of that cognition (‘knowing how’)” (p. 74). In

14

their study of 685 Dutch LI-English L2 schoolchildren in grades 6, 8 and 10, the
researchers found that self-reported knowledge about texts, or “knowing that”, and
knowing about reading strategies, or “knowing how”, were more signiﬁcant predictors of
both L1 and L2 standardized reading test scores than other types of “meta” knowledge,
such as knowledge of reading goals and self-assessed reading ability.

Schoonen et al. (1998) also conﬁrmed that metacognition functions cross-
linguistically, contributing to both native and target language reading comprehension.
They found no signiﬁcant difference between the contributions of metacognitive
knowledge to L1 and L2 reading comprehension scores among Grade 8 and 10 students
(Grade 6 participants in the study did not take the L2 reading comprehension test). After
metacognitive knowledge was partialled out, the amount of shared variance between L1
and L2 reading dropped ﬁom 38% to less than 1%, indicating that metacognition explains
in large part the relationship between native language and target language reading. A
follow-up study of Dutch schoolchildren validated the ﬁnding that metacognitive
knowledge is activated in reading in both languages. In a model comparing components
of L1 and L2 reading, Gelderen et al. (2004) found that metacognitive knowledge made
signiﬁcant unique contributions to both L1 and L2 reading among these participants (n =
397), as measured by scores on multiple choice tests. Adding metacognitive knowledge
to a model of L2 reading that also included Ll reading ability and L2 knowledge also
allowed the researchers to account for more L2 reading comprehension variance (83%)
than previous studies that did not include metacognition as a variable.

Further analyses by Gelderen et al. (2004) revealed complex interrelationships

between L1 and L2 knowledge and ability and metacognitive knowledge. When L1

15

reading ability was added to a model of L2 reading that included L2 grammar and
vocabulary knowledge, metacognitive knowledge, and word and sentence recognition
speed, metacognition was no longer a signiﬁcant predictor of L2 reading comprehension.
Instead, Ll reading ability made the largest contribution to L2 reading performance, with
L2 vocabulary knowledge its only other signiﬁcant predictor. The researchers explained
the diminished contribution of metacognition in this model by its strong relationship with
native language reading ability, resulting in the “takeover” of metacognition’s predictive
role by Ll reading ability. Otherwise stated, metacognition may contribute to target
language reading comprehension indirectly through its contribution to native language
reading ability. These results underscore the importance not only of including measures
of metacognition in L2 reading research, but also of using research designs and data
analysis procedures that allow examination of the interrelationship and moderating
effects of these individual difference variables on comprehension.
2.1.3 L2 exposure

2.1.3.1 Language exposure and L2 development

Language learning is widely understood as a process of hypothesis formation and
testing based on available language input (Koda, 2005). Learners use L2 input
encountered while reading, writing, speaking and listening to make connections between
linguistic forms and their meanings, and to determine the probability that these forrn-
meaning patterns will occur. As learners are exposed to more language, the processing of
this language, or the “mapping” and retrieval of these patterns, becomes automatic.
Under this connectionist view, language learning is thus experience-based and facilitated

by frequent language contact. Ellis (2002) maintains that ﬁ'equency of exposure to

16

language plays an important role in developing and retrieving these patterns, or in other
words, “(language) learning accords to the power law of practice” (p. 144).
2.1.3.2 Language exposure and L2 reading development

The contributions of written language contact to reading comprehension have
been observed in both L1 and L2 contexts. Stanovich’s (1986) analysis of L1 studies
concluded that reading practice outside of the classroom is one of the most important
factors differentiating good and poor native-language readers. Numerous L2 studies have
revealed strong relationships between print exposure and gains in various measures of L2
literacy. Elley and Mangubhai (1983) found that Fijian schoolchildren who participated
in an extensive reading, or “book ﬂood”, program where they read for 20-30 minutes a
day had signiﬁcantly greater gains in L2 reading comprehension and grammar knowledge
than a control group that followed a standard English curriculum containing very little
reading. Follow-up testing after one year showed that these gains persisted over time.

Elley’s (1991) review of subsequent reading programs, also mainly with Paciﬁc
island schoolchildren, noted that increased exposure to print resulted in rapid growth in a
wide-range of measures of L2 development. Students were exposed to reading materials
in these programs through “shared-book” approaches, where reading and discussion of
books served as the basis of classroom L2 instruction, or through extensive reading or
reading aloud. In all cases, groups who participated in these reading programs
outperformed control groups who received form-focused instruction in reading and
listening comprehension, vocabulary and grammar knowledge, and writing and speaking

skills.

17

Among adult ESL students, L2 free reading amount has also predicted a more
general measure of language skill, TOEFL scores. Constantino et al. (1997)
operationalized reading amount as the frequency with which students reported reading
English newspapers, engaging in free reading, reading English books for fun, and the
number of English books they reported reading before taking the TOEFL. Though free
reading amount was a signiﬁcant predictor of TOEFL scores using multiple regression
analysis, the researchers argue that free reading amount is potentially a stronger predictor
than their results indicate, as participants reported engaging in very little free reading in
English.

Qualitative studies of ESL students have revealed that exposure to high-interest
reading materials may also facilitate comprehension by encouraging readers to focus on
text meaning instead of on decoding individual words. Kim and Krashen (1997) found
that when adult Korean L1 -English L2 readers were provided with an interesting novel in
English, they “just read on” and “didn’t care about words [they] couldn’t understand”
because reading the story was “so much fun” (p. 28).

The importance of language exposure to literacy development also extends
beyond the exposrn'e to the printed word. Oral language exposure appears to have similar
beneﬁts in reading comprehension. The strong relationship between oral language and
reading development has been well documented in L1 studies, where the ability to
pronounce written words has consistently predicted reading success among children
(Bowers, Golden, Kennedy, & Young, 1994; Share & Stanovich, 1995; Wagner,
Torgeson, & Rashotte, 1994). Koda (2005), in fact, concludes that the ability to decode

oral language phonologically may be the key competency in reading acquisition in all

18

languages. Given this relationship, it is not surprising that measures of proﬁciency related
to oral language have predicted L2 reading success. In a study of Spanish Ll-English L2
schoolchildren, Proctor, August, Carlo, and Snow (2005) found that listening
comprehension skills and oral vocabulary knowledge were more powerful predictors of
L2 reading comprehension than written word decoding. In the same way that exposure to
print contributes to reading development, exposure to oral language should facilitate
comprehension of spoken language, which in turn will contribute to reading
development.

University students may also have substantial contact with the target language
through study abroad experiences. Of more than 223,000 American students who studied
abroad in the 2005-2006 academic year - an increase of 150% over the last decade - more
than 75% studied in countries whose primary language is not English (Chin & Bandari,
2007). Using a version of the Language Contact Proﬁle (Freed, Dewey, Segalowitz &
Halter, 2004), an instrument that asks students to report the amount of time they spend
engaging in various types of L2 exposure, as well as interviews and observations, Dewey
(2004) found that students studying abroad in Japan (n = 15) spent signiﬁcantly more
time reading and interacting with native speakers than students enrolled in a Japanese
immersion program in the United States (n = 15). However, this increased L2 contact did
not translate into signiﬁcant differences in gains on reading comprehension measures
between the two groups, likely because of variability in both the amount of contact and
the reading scores of the study abroad group. Study abroad, nonetheless, exposed

students to signiﬁcantly more target language, and among a larger sample size a

19

signiﬁcant relationship between this increased language contact and reading outcomes
would be expected. '

Carrel] (1991) in fact proposed that the superior performance of Spanish-speaking
ESL students studying in the US. on multiple-choice reading tests compared to that of
native English-speaking students studying Spanish in the US. resulted from differences
in learning context and language exposure. Learners studying in a second language
setting were immersed in the target language both inside and outside of the classroom,
whereas those studying in a foreign language setting had limited, mainly classroom-based
L2 exposure.

2.1.4 Motivation

2.1.4. 1 Motivation and L2 learning

A substantial amount of research suggests that individual affective differences,
including motivation, may predict language learning outcomes. Ddrnyei (2005) maintains
that motivation “provides the primary impetus to initiate L2 learning and later the driving
force to sustain the long and often tedious learning process” (p. 65). Though the term
motivation is often used to designate a single, monolithic entity, L2 motivation research
has provided much evidence that it is a complex and multidimensional construct. Social
psychologists Gardner and Lambert (1959, 1972). were the ﬁrst to identify motivational
components related to language learning. In exploring the relationship between
motivational orientations, or classes of reasons for learning another language, and L2
outcomes, they identiﬁed two possible learner orientations: integrative and instrumental.
An integrative motivational orientation reﬂects a desire to interact with and be accepted

by the target language community, whereas an instrumental motivatiorml orientation

20

indicates a practical or extrinsic purpose for language learning (e.g., in order to get a
good job). These orientations became the basis of Gardner’s (1985) socioeducational
model of second language acquisition, which conceptualized motivation as the interaction
of learner orientations with attitudes towards the language and the learning situation.
While Gardner’s early studies of Canadian English-speaking learners of French showed
that integrative orientations were stronger predictors of achievement (Gardner & Lambert
1972; Gardner 1985), subsequent studies have shown that instrumental orientations can
also inﬂuence learning. deyei (1990) found that Hungarian native-speaking adults’

(n = 134) instrumental reasons for learning English (for example, for professional
advancement) were strongly correlated with their desire to achieve intermediate levels of
language proﬁciency. Gardner and Maclntyre (1991) examined the effects of both types
of motivation on French vocabulary learning among English L1 college students (n = 92).
The researchers created an instrumental motivation condition for half the participants by
offering to pay them $10 if they missed no more than two answers on the sixth trial of a
computerized English-French paired-associates vocabulary test, a test procedure that
allowed students to review their answers and study the correct word pairs during each
trial. Students in the instrumental motivation condition had signiﬁcantly higher
vocabulary test scores than those who were not promised the incentive. Participants’
levels of integrative motivation were also assessed with a self-report questionnaire, and
similar results were obtained for high versus low integratively motivated students. Both
types of motivation also had considerable effects on participants’ behavior in learning the

word pairs. Students with high integrative motivation and those in the instrumental

21

motivation condition spent signiﬁcantly more time viewing the English words, reviewing
their French answers and studying the English-French pairs.

Some scholars, however, have questioned the emphasis that Gardner’s
socioeducational model places on the integrative-instrumental distinction, especially in
foreign language contexts where opportunities for interaction with target language
communities are rare. deyei (1990) noted that foreign language learners often have not
had enough contact with the target language community to form attitudes about it,
therefore they may not be as inﬂuenced by a desire for contact and integration as learners
in second language environments. In light of such considerations, Noels and associates
reconceptualized motivation using a cognitive rather than socioeducational approach.
Using the theoretical framework proposed by Deci and Ryan’s (1985) self-determination
theory, Noels, Clement, and Pelletier (1999) and Noels, Pelletier, Clement, and Vallerand
(2000) identiﬁed three alternative orientations towards language study among English-
speaking learners of French: intrinsic orientation, or reasons related to one’s personal
pleasure or interests, extrinsic orientation, or reasons instrumental to a consequence
unrelated to inherent interest (e.g., to get a job), and arnotivation, an orientation where
learners do not value language study, believe their behavior has no effect on language
learning outcomes, and think that they are wasting their time studying a language. In a
subsequent study Noels, Clement, and Pelletier (2001) proposed that integrative
motivation might be subsumed into this three-orientation model. In analyzing self-report
questionnaire responses by F rench-Canadian learners of English, they found that what
Gardner (1985) had identiﬁed as integrative reasons for language study (e.g. “I love

the. . .culture and the language is beautiful”) (p. 52) correlated highly with intrinsic

22

motivation. Qualitative research also suggests that in certain contexts integrativeness may
not be distinguishable ﬁom other motivational orientations. After analyzing Indonesian
schoolchildren’s (n = 219) responses to open-ended questionnaire items and interview
questions about their reasons for learning English, Lamb (2004) concluded that
“integrative and instrumental orientation are difﬁcult to distinguish as separate concepts.
Meeting with westerners, using computers, understanding pop songs, studying or
travelling abroad, pursuing a desirable career—all these aspirations are associated with
each other and with Englis ” (pp. 14-15).
2.1.4.2 Motivation and L2 reading

A range of studies have found strong relationships between motivation and
various measures of L2 development, including vocabulary learning (Gardner, 1985;
Gardner, & Maclntyre, 1991), listening comprehension (V andergriﬁ, 2005), writing skills
(T remblay & Gardner, 1995), grades or evaluations by teachers (Clement, DGrnyei &
Noels, 1994; Clement & Kruidenier, 1985; Noels, et al., 2001; Samimy & Tabuse, 1992;
Wen, 1997), and overall L2 proﬁciency (Gardner & Lambert, 1972; Gardner, Tremblay,
& Masgoret, 1997). These results would suggest that motivation could have similar
effects on L2 reading. L1 reading studies provide persuasive evidence of a link between
reading performance and motivation. In both cross-sectional and longitudinal studies of
children, Gottﬁied (1990) established that intrinsic academic motivation was a signiﬁcant
predictor of reading and other academic success as measured by standardized tests and
teacher ratings. Baker and Wigﬁeld (1999) found moderate correlations between both
intrinsic and extrinsic motivations and 5th and 6'” grade girls’ scores on standardized

reading tests. Some L1 research suggests that as readers mature, the relationship between

23

motivation and reading outcomes is strengthened. Whereas Guthrie et al. (1999) reported
no signiﬁcant relationship between elementary and middle-school students’ motivation
and reading achievement, a second study of high-school students revealed strong
correlations between general reading motivation, a construct reﬂecting both intrinsic and
extrinsic motivational orientations, and performance on a reading test.

Before examining its effects on reading outcomes, L2 reading motivation research
ﬁrst sought to identify the components of motivation to read in another language.
Componential studies have conﬁrmed that intrinsic, extrinsic, and amotivation
orientations also exist for L2 reading. Using factor analysis of questionnaire results, Mori
(2002) identiﬁed intrinsic and extrinsic value as two main subcomponents of motivation
to read in a foreign language among Japanese L1 learners of English (n = 447). Two
other components, readers’ expectancy of success and their assessment of the importance
of reading, also emerged. Kondo-Brown’s (2006) research with English L1 students of
Japanese (n = 43) identiﬁed four related components of L2 reading motivation based on
students’ responses on self-report questionnaires: intrinsic orientation, extrinsic
orientation, amotivation and self-perception, which was analogous to the readers’
expectancy of success factor described by Mori. Mori’s follow-up study (2004) with
Japanese learners of English (n = 100) provided evidence that componential differences
between L2 reading motivation and motivation to study L2 exist. Though intrinsic value
emerged as a factor for both types of motivation, extrinsic motivation was associated with
motivation for L2 study but not L2 reading. Learners’ perception of the importance of

success was a factor in both reading motivation and general L2 study motivation.

24

Though not abundant, L2 research has begun to examine these motivational
factors in relation to reading outcomes, including reading amount and reading
comprehension. Studies examining motivation’s effects on reading amount have not
produced consistent results. In the same study described above, Mori (2004) found little
relationship between reading amount and motivation among participants in a Japanese
Ll-English L2 extensive reading program. Only students’ motivational intensity for L2
study as measured by reported study habits and their intrinsic interest in the stories that
they read predicted the amount read, but these two contributing factors were negatively
correlated to reading amount; those who did not perceive themselves as diligent English
students or who found the stories boring read less. In other words, lack of interest and
lack of motivational intensity caused students to read less, but intrinsic interest in the
reading material and high motivational intensity did not motivate students to read more.
Other intrinsic and extrinsic motivational factors were not signiﬁcant predictors of
reading amount. Yamashita (2004) found that attitudes towards reading such as comfort
and self-perception of reading ability as measured by a self-report questionnaire
contributed to the average number of pages of extensive reading done per week by
J apanese-speaking learners of English, but extrinsic motivational factors such as readers’
perception of the value of reading did not affect reading amount.

To date, only a handful of studies have examined the contributions of
motivational factors to L2 reading comprehension, and they have obtained dissimilar
results. Brantrneier’s (2005) research with advanced college-level English Ll-Spanish L2
Students (n = 88) provides evidence that intrinsic factors contribute to reading

comprehension. Brantrneier found that self-report questionnaire items assessing L2

25

reading enjoyment had a signiﬁcant effect on recall comprehension of an authentic
narrative passage in Spanish. Though no relationship between enjoyment and a second
comprehension measure, multiple-choice test scores, was found, this likely resulted from
the lack of variance in these scores. In a subsequent study with a similar participant group
(n = 104) using the same narrative text, Brantrneier (2006) found that the intrinsic factors
of personal interest in the text and engagement with the text, also assessed through self-
report questionnaires, predicted scores on sentence completion and multiple-choice
reading tests, but not recall comprehension. The absence of a relationship between recall
comprehension and these factors, however, may have resulted from students’ relatively
poor performance on the recall measure, which the researcher attributes to the length of
the reading passage (1218 words). In contrast, Kondo—Brown (2006) found no signiﬁcant
correlations between advanced English Ll-Japanese L2 (n = 43) students’ intrinsic
motivation for L2 reading or study and either of two measures of reading comprehension,
a multiple-choice reading test and a test of kanji knowledge. Extrinsic motivational
orientations also failed to correlate with reading comprehension, but amotivation for
reading Japanese had a signiﬁcant negative relationship with both comprehension
measures.

The failure of these studies to ﬁnd consistent, direct effects of intrinsic and
extrinsic motivational factors on reading comprehension may stem from the fact that they
investigated motivation in isolation ﬁom other cognitive, experiential and behavioral
individual differences. Domyei and Korrnos (2000) argue that while motivation is

strongly related to achievement, its effects on learning outcomes may be indirect:

26

...from a theoretical point of view, the relationship between motivation and
achievement is not straightforward. Motivation as a psychological term is used to
refer to the antecedent of action rather than achievement. It is true that motivated
learners will demonstrate more effort and persistence in their task behaviour,
which in turn can lead to increased achievement. However, this relationship is
indirect, because achievement is also inﬂuenced by a host of other factors, most
notably the learners’ ability, learning opportunities and the instructional quality of
the learning tasks. (pp. 281-282)
Thus, motivation may operate indirectly on L2 reading comprehension by inﬂuencing
other behaviors and experiences. Research discussed in previous sections of this chapter
has shown that variables such as L1 reading ability, L2 knowledge, metacognition, and
language exposure have important and often direct effects on L2 reading. The
relationship between motivation and reading may be moderated by some of these other
variables. For example, motivation may prompt students to gain more L2 exposure by
reading, watching ﬁlms or listening to music in the target language, to take advantage of
opportunities to communicate in the L2, to improve their L2 knowledge through study
and practice, and to use comprehension strategies while reading the target language.
2.1.5 Anxiety
The notion that anxiety plays an inhibitive role in language acquisition owes
much to the inﬂuential works of Stephen Krashen (1982, 1985). Krashen argued that
language is acquired though exposure to comprehensible L2 input, but that a high
“affective ﬁlter” prevents learners from integrating this input into their developing

language systems. Anxiety was identiﬁed as a key component of this affective ﬁlter.

27

Though Krashen’s hypotheses have been the subject of criticism and debate (Gregg,
1987; Lee & VanPatten, 2003; McLaughlin, 1987), the idea that anxiety acts as a
roadblock to language learning remains intuitively appealing. Perhaps this is because
most people who have studied another language can recall at least one (if not many)
difﬁcult, anxiety-ﬁlled moment in their L2 learning experiences. Horwitz and Young
(1991) in fact maintain that anxiety is prevalent among language learners, producing
debilitating effects among about half of those who engage in L2 study.

Gardner and Maclntyre (1993) deﬁne language anxiety as fear or apprehension
experienced when a learner is expected to perform in the target language. Horwitz,
Horwitz and Cope (1986) argue that language anxiety is distinct from general anxiety and
other types of performance anxiety because it is directly related to performing in the L2.
Subsequent research reviewed by Horwitz (2001) has conﬁrmed that foreign language
anxiety is an independent construct, correlated only weakly with general anxiety and
other types of learning anxiety, such as test anxiety. A large body of research has
demonstrated strong and consistent negative relationships between language anxiety and
various types of L2 achievement, including course grades (Aida, 1994; Horwitz, 2001;
Saito & Samimy, 1996), proﬁciency test scores (Ganschow, Sparks, Anderson, Javorshy,
Skinner, & Patton, 1994; Gardner, Lalonde, Moorcroft, & Evers, 1987; Gardner &
Maclntyre, 1993; Yamashiro & McLaughlin, 2000), listening comprehension (Elkhafaiﬁ,
2005), and oral and written production (Young, 1986).

Some researchers, nonetheless, have argued that “helpful” language anxiety may
exist. Scovel (1978) proposed that anxiety facilitates language learning by increasing

students’ attentiveness. In a study of adults (n = 855) in intensive language courses,

28

Ehrman and Oxford (1995) found evidence of both helpﬁrl and harmful language anxiety.
Self-reported anxiety about speaking in class, as measured by questionnaire items taken
ﬁ'om the Foreign Language Classroom Anxiety Scale (F LCAS) (Horwitz, et al., 1986)
was highly and positively correlated with measures of both speaking and reading
proﬁciency, however, other negatively worded questionnaire items on anxiety (e.g., “It
upsets me when others do better than I”) (p. 79) were signiﬁcantly and negatively
correlated with both measures of proﬁciency. The researchers noted that anxiety may
motivate students to study more, which on its face would seem helpful to language
learning. However, the researchers further observed that this increased study, or
“overstudying”, is often counterproductive, citing the case of a woman whose anxiety
about making mistakes in her beginning Spanish class pushed her to study eight hours a
day without any corresponding improvement in performance.

Other research has pointed to differences in L2 anxiety depending on the language
task. Speaking is widely considered the most anxiety-provoking L2 activity (Horwitz et
al., 1986) since it involves “face-threatening” risks: speakers must reveal themselves and
their (possibly deﬁcient) language skills, and subject themselves to immediate (and
possibly negative) evaluation by their interlocutor. Receptive language skills such as
reading appear to inspire less anxiety than speaking. Kim (1998, cited in Horwitz, 2001)
found that Korean EFL students enrolled in a reading class had signiﬁcantly lower levels
of language anxiety as measured by the F LCAS questionnaire than those enrolled in a
conversation class. Using both the FLCAS and an adapted version of this questionnaire

measuring reading anxiety, Saito, Horwitz, and Garza (1999) similarly found that

29

students of French, Japanese and Russian experienced less reading-related anxiety than
general foreign language anxiety.

Nonetheless, anxiety cannot be ignored as a negative predictor of reading
outcomes. Argamon and Abu-Rabia (2002) found signiﬁcant negative correlations
between foreign language learning anxiety as measured by a subset of questions ﬁom the
FLCAS among adolescent Hebrew L1 learners of English (n = 68) and scores on a true-
false reading comprehension test. Using written recalls to assess comprehension, Sellers
(2000) reported a similar signiﬁcant negative relationship between reading
comprehension and reading anxiety questionnaire items among English Ll-Spanish L2
students (n = 89). This correlational research, however, cannot be taken as evidence that
anxiety is a proximate cause of poor L2 reading performance. It is quite probable that,
like motivation, anxiety has indirect effects on reading comprehension by contributing to
related behaviors. Horwitz et al. (1986), for example, observed that anxious language
students may avoid studying, skip class or avoid participating in class in an effort to
relieve their anxiety. Thus, it would not be unexpected that language anxiety negatively
affects reading comprehension by decreasing language exposure and detening learners
ﬁ'orn taking advantage of opportunities to develop L2 knowledge and proﬁciency.

2.1.6 Summary

Research has provided evidence of relationships between individual differences in
L1 reading ability, L2 proﬁciency, metacognition, L2 exposure, motivation and anxiety
and L2 reading outcomes. In studies investigating the relative contributions of L1 reading
ability and L2 proﬁciency, the latter has quite consistently accounted for more reading

comprehension variance than Ll reading ability among cross-sections of L2 learners.

30

However, adding metacognition as a predictor variable may help account for additional
L2 reading variance, and may change the relative contributions of L1 reading and L2
knowledge to L2 reading comprehension.

Exposure to the target language is another important contributing factor to
language and literacy development. Frequent contact with L2 input promotes connections
between language forms and meanings, and promotes automaticity of retrieval of these
form-meaning patterns. Since both written and oral language exposure have been shown
to make sizeable contributions to reading development, it is important to measure
exposure to these types of language in the many forms in which language learners may
encounter them.

Existing research provides support for a three-factor model for motivation,
comprised of intrinsic motivation, extrinsic motivation, and amotivation. Motivation and
these component orientations have important effects on L2 development, including
general language proﬁciency and speciﬁc language skills such as listening and writing.
The contributions of these motivational orientations to L2 reading comprehension,
however, have not been convincingly established in the studies that have investigated this
question. Whereas Brantrneier (2005, 2006) found that intrinsic factors such as
enjoyment may positively inﬂuence reading comprehension, Kondo-Brown (2006) found
that only amotivation, not intrinsic or extrinsic motivation, had signiﬁcant predictive
effects on L2 reading comprehension. However, since it is posited that motivation has
indirect effects on achievement by inﬂuencing other behaviors and experiences (Domyei
& Kormos, 2000), the lack of consistent, direct relationships in these studies cannot be

taken as evidence that motivation is not implicated in L2 reading. Evidence of

31

relationships between other motivational factors, including motivational intensity and
self—perceived ability, and reading amount also exist (Mori, 2004; Yamashita, 2004),
suggesting that motivation indeed may shape behaviors that contribute to reading
performance instead of directly causing that performance.

Although there is some evidence that certain types of anxiety may play a
facilitative role in language learning, the preponderance of language anxiety research has
revealed signiﬁcant, negative relationships with L2 learning outcomes. Similar negative
relationships between language anxiety and reading comprehension have also been
observed, though the correlational nature of these relationships precludes drawing
deﬁnitive conclusions about anxiety’s direct effects on reading comprehension. Instead,
language anxiety may operate indirectly on reading comprehension by discouraging

learners ﬁom L2 contact or study.

2.2 Text genre

In the previous section, I have presented evidence that characteristics particular to
individual readers make important contributions to L2 reading. Because reading involves
an interaction between reader and text, logically, characteristics of the particular text that
is read should also play an important role in L2 reading processes. Though, like people,
no two texts are exactly the same, texts can be classiﬁed into different genres based on
common features or communicative intents. In this section, I will provide deﬁnitions of
two of these genres, narrative and expository, and discuss both L1 and L2 research on

genre’s role in reading.

32

2. 2. 1 Narrative and expository genres

Although there is no prevailing consensus on a precise deﬁnition of narrative text,
an imprecise deﬁnition might be that narrative texts are “stories”. Graesser et al. (1991)
note that whether they relate actual or ﬁctitious events, narratives are composed of story-
like elements, such as characters, temporal and spatial placements (“A long time ago in a
galaxy far, far away. . .”), and episodes or sequences of events. Larsen (1984) maintains
that a recounting of these events is the hallmark of narrative texts. In addition to being
situated in a particular place and time, events in narratives typically unfold over some
period of time. These chronological events also share causal relationships. For example,
narratives usually describe complications or problems that the main character(s)
encounters (e.g., Little Red Riding Hood discovers a big, bad wolf in her grandmother’s
bed), and subsequent events are related to the resolution of the complications. Graesser,
et al. (1991) also observe that narratives are intended to arouse affective reactions in the
reader. Narratives are typically written to entertain the reader in some way, and do so by
manipulating readers’ emotions such as curiosity, surprise, amusement or suspense.

Expository texts, on the other hand, are intended to inform. Instead of recounting
an episodic story, expository texts use communication strategies such as orientation and
clariﬁcation to convey information to the reader (Larsen, 1984). Whereas narratives
describe events that develop over time, expository texts often communicate information
in a static, non-chronological fashion. Unlike narratives, expository texts are usually not
intended to engage readers effectively. Though expository texts certainly may elicit
affective reactions in some readers, arousing an emotional response is typically not

among the author’s primary communicative purposes.

33

Weaver and Kintsch (1991) thus summarize the differences between the two
genres as such: “. . .the main thrust of expository texts is to communicate information so
that the reader might learn something. The main focus of a narrative text is to tell a story
so that the reader may be entertained.” (p. 230). However, they also concede that the
distinction between expository and narrative texts is not absolute. Narrative texts such as
fables, for example, have a clear didactic purpose. Expository texts like newspaper or
magazine articles may contain some narrative elements (for example, describing an
episode or a sequence of events), and may be entertaining as well as informative.

It is also important to note that these two genres represent only a general level of
text classiﬁcation. Within each text genre, a variety of subgenres exist. Novels, short
stories and plays can all be categorized as narratives, for example, while the expository
genre includes texts such as essays, news reports, or journal articles.

2. 2.2 Text genre and LI reading

L1 studies have overwhelmingly shown that narratives are easier to comprehend
and recall than expository texts. These results have been consistent for both child and
adult readers. Zabrucky and Ramer’s (1992) study of 6‘“ grade children found that
students recalled signiﬁcantly more from reading narrative texts than expository texts.
The researchers further observed that reading expository passages required signiﬁcantly
more effort, as measured by “lookbacks” or re-readings of portions of the text, and longer
reading times than the narrative passages. In two studies with college students (Kintsch &
Young, 1984; Kozminsky, 1977), participants read one narrative and two types of
expository texts, a description and a report. Students recalled signiﬁcantly more from the

narrative texts (on average, 30-33% of all text propositiOns) than either the descriptions

34

(20-26%) or the reports (12-20%). Graesser et al. (1980) also found signiﬁcant
differences between narrative and expository text comprehension by college students
using a recall measure. 62% of narrative propositions were recalled, compared to only
37% of expository propositions. The researchers had asked other students to judge the
narrativity of each passage using Likert-scale ratings, where the least narrative passages
described “static information” and the most narrative described “active information with
events unfolding over time”. When degree of narrativity was used as a predictor of recall
comprehension in a multiple regression analysis, it explained a hefty amount of variance
(84%) in these scores.

Superior outcomes for narrative texts have also been observed in studies
comparing reading comprehension between groups of young adults and senior citizens.
Tun (1989) found that both college students and older adults (aged 65-80) comprehended
signiﬁcantly more ﬁom narrative texts as assessed by both a recall measure and a
multiple-choice test. Young adults recalled 51% of propositions ﬁom narrative texts and
scored 92% on the narrative multiple-choice test, compared to 41% recall and 88%
multiple-choice scores for expository texts. Older participants recalled 31% of narrative
propositions and answered 93% of multiple-choice questions correctly; their expository
comprehension scores were 19% for recall and 73% for multiple choice. Petros, et al.
(1989) reported a similar trend in a study where texts of each genre were presented orally
instead of in writing. Both young adults and seniors recalled signiﬁcantly more
propositions from narrative texts (72% and 60%, respectively) than from expository texts

(57% and 39%).

35

Some of the unique features that distinguish narratives from expository texts also
seem to explain their superior comprehensibility. As described in section 2.2.1, narratives
are characterized by both temporal and causal cohesiveness. Certain theorists propose
that this type of coherence facilitates the creation of a mental model of the text, the
retrieval of related information ﬁ'om memory, and the integration of this information into
the mental text model. According to Kintsch (1988, 1998), texts are composed of a series
of propositions, each consisting of a predicate and one or more arguments. Reading is a
process of connecting the text’s propositions into a semantic network called a textbase, or
the reader’s mental representation of the text’s meaning. This textbase includes a
microstructure, representing the surface form of the text (i.e., actual words and phrases),
and a macrostructure, representing relationships between individual propositions
contained in the text, or what is more commonly described as the “gist” of the text.
Kintsch (1998) argues that the ability to form this macrostructure is the key to reading
comprehension: “for comprehension and memory, the gist of the text —- expressed
formally by the macrostructure — is usually what matters most” (p. 67). Kintsch and
Young (1984) maintain that narrative text structures promote macrostructure
development, and therefore comprehension and memory of the text, because of their
cohesiveness. The salient temporal and causal cues contained in narratives also help
readers to efﬁciently retrieve both textual information and prior knowledge (sometimes
called schemata) from memory and to integrate it into their mental representation of the
text.

The familiarity of narrative structure also appears to play an important role in

comprehension of this genre. Koda (2005) notes that whereas formal training may be

36

required to recognize the structure of exposition and argumentation, narrative story
structures are familiar even to young children who can not yet read or write. Graesser et
al. (1991) propose that the event sequences that comprise narrative texts are so familiar to
us because they also characterize our personal, everyday experiences. Otherwise stated,
we live stories, and we see familiar reﬂections of our own stories in narrative texts.

2. 2. 3 Text genre and L2 reading

Given the overwhelming evidence of narratives’ superior comprehensibility in L1
settings, it would be expected that similar results would be obtained in L2 contexts.
However, very little research examining the relationship of the narrative and expository
genres to reading comprehension has been conducted with L2 learners, and the few L2
studies that have reported comprehension scores for narrative and expository texts have
reached divergent conclusions.

Bullock and Lantolf (1987) used an excerpt ﬁom a narrative short story and an
expository text on insects to compare English native and normative speaker performance
on cloze tests for each genre. Although one group of normative speakers, students taking
a year-long advanced English class to prepare them for graduate school (n = 9), scored
much higher on the narrative cloze test (75%) than on the expository test (51%), another
group of normative speaker graduate students (n = 11) had virtually identical scores on
both measures (narrative 97%, expository 94%). However, no statistical comparison of
these scores was performed to determine whether these differences were signiﬁcant.

In Bensoussan’s 1990 study, translation of L2 texts into the L1 was used to
measure reading comprehension of narrative and expository texts. Advanced university

EFL learners (n = 105) read an expository text in English about paranoia and then

37

at

I

b."

\5

translated it into Hebrew, their native language. A separate group of advanced EFL
learners (n = 60) read a narrative excerpt ﬁ'om an F. Scott Fitzgerald story in English and
translated it into their native language (Hebrew or Arabic). The researcher analyzed each
text according to discourse aspects including both local (word and sentence-level) and
global level (units of meaning involving more than one sentence) structures; these
analyses were then compared to the participants’ translations. Bensoussan found
signiﬁcantly higher rnistranslation rates for the narrative than the expository text (335 and
190 mistranslations, respectively) across nearly all discourse categories. The narrative
text translations included more mishanslations of propositional content, vocabulary and
expressions, pronoun agreement and communicative function (e.g., translating what was
a neutral statement by the author as a positive or negative opinion). The only discourse
category that proved more problematic for expository texts was the local-level feature of
grammatical cohesion, or the translation of linking and transition words that connected
ideas (e.g, moreover, therefore, because).

In contrast with these ﬁndings but consistent with Ll research, DuBravac and
Dalle (2002) found greater miscomprehension of expository texts. Instead of a translation
task, the researchers used student—generated questions to assess comprehension.
Intermediate-level English L1 -French L2 students (n = 47) read one of two narratives and
one of two expository texts. Expository texts were taken ﬁ'om French daily newspapers,
and narrative texts were excerpted from contemporary literary works. Both the researcher
and the students’ instructor judged all texts comparable in subject matter, difﬁculty,
phrase length and syntactic and lexical complexity. Each text was divided into ﬁve

sections; upon reaching the end of each section, participants were instructed to formulate

38

a question whose goal was to assess how well another person reading the text would have
understood that section. The number of questions indicating that the students had
misunderstood parts the text was signiﬁcantly higher for expository texts: 24%, compared
to a 6% miscomprehension rate for narrative texts.

Methodological differences between these two studies may explain their
incompatible ﬁndings. First, because different groups of students read each text type in
Bensoussan’s (1990) study, the results may have been inﬂuenced by any number of
individual reader differences within these two groups. Furthermore, her results showing
greater difficulty in comprehending the narrative text were likely affected by differences
between the two texts. While Dubravac and Dalle (2002) made attempts to control for
topic familiarity, difficulty and complexity of grammar and vocabulary between the
narrative and expository texts, Bensoussan (1990) notes that in her study the texts
representing each genre were not necessarily equivalent. She observes that the expository
text used everyday language and did not require specialized knowledge to understand,
whereas the narrative text contained difficult vocabulary and idiomatic language as well
as potentially unfamiliar social and cultural references. Finally, the difference in tasks
(translation versus question formation) may also have accounted for differences in
results. Although it may true, as Bensoussan (1990) explains, that translation “is closely
linked to the reading process; only those text elements that have been read and
comprehended can be adequately translated” (p. 51), it can also be argued that reading
and translation are two distinct skills. Indeed, the author concedes that the translation task
itself may have been the source of some errors. DuBravac and Dalle’s (2002) question

formation task also has pitfalls. The authors acknowledge that because participants were

39

not required to supply the answers to questions they had formulated, in certain cases it
could not be veriﬁed that the questions they generated demonstrated that they had indeed
understood the text.

These L2 genre studies also limited their investigation to one level of learner,
either intermediate (Dubravac & Dalle, 2002) or advanced (Bensoussan, 1990; Bullock &
Lantolf, 1987). Other reading research, however, suggests that learners’ level of language
proﬁciency may determine how well they are able to exploit the features of different text
structures. Horiba (1990), for example, found differential effects of proﬁciency level on
the amount of attention paid to local and global text features. A think-aloud protocol was
used to compare the mental reading processes between these proﬁciency groups. After
reading each sentence of a narrative story, participants (9 native Japanese speakers, 11
nonnative advanced Japanese speakers) reported what they were thinking about when
reading. Quantitative and qualitative analyses of their comments revealed that nonnative
speakers attended signiﬁcantly more to local (word and sentence-level) text features,
whereas native speakers concentrated on global text comprehension. Normative speakers
ﬁequently made comments about their efforts to decode words and phrases, whereas
native speakers made no comments about the linguistic features of the text. Instead,
native speakers commented signiﬁcantly more about making inferences and connecting
the text to their general knowledge. Chen and Donin (1997) reported similar results in a
study of Chinese Ll-English L2 university students (n = 36). The researchers
operationalized lower-level text processing as the amount of time taken to read each of
two expository texts in English, as longer reading times are associated with additional

mental effort required to decode words and sentences. Learners at low-intermediate and

40

intermediate levels spent signiﬁcantly more time attending to these local-level text
features than high-intermediate and high proﬁciency learners. These results suggest that
lower-level learners may be less able to take advantage of genre-level text cues that
facilitate comprehension, such as the salient temporal and causal relationships between
events in narrative texts.
2. 2. 4 Summary

Narrative texts engage readers in unparalleled ways. Narratives share much in
common with readers’ everyday experiences: both involve characters, settings in time
and place, the passage of time, and actions and their consequences. These types of texts
entertain readers and play on a wide spectrum of their emotions. Narratives also appear to
engage readers cognitively in different ways than other genres. The cohesiveness and
familiarity inherent to the narrative genre provide cues that facilitate the mental
construction of text meaning, the retrieval of related knowledge from memory, and the
integration of this knowledge into the representation of the text’s meaning. Ll studies
have provided a wealth of evidence supporting the facilitating role of the narrative genre
in reading comprehension. However, these results have not yet been replicated in L2
settings, where the few studies examining narrative and expository text comprehension
have obtained conﬂicting results. The lack of consistency in results as well as
methodology in these studies justiﬁes further investigation of the relationship between
text genre and L2 reading comprehension. Furthermore, it is important to examine this
relationship among a cross-section of L2 learners, as there is some evidence of the
existence of a language proﬁciency threshold below which readers may not effectively

take advantage of genre cues.

41

2.3 Conclusion

Research has identiﬁed an array of linguistic and non-linguistic individual
differences that contribute to reading comprehension in another language. A wide range
of studies has established that individual differences in target language proﬁciency and
native language reading ability have direct and robust effects on L2 reading
comprehension. Metacognitive knowledge, or knowledge about reading and texts,
appears to contribute to both L1 and L2 reading skills, making it a necessary component
of any model seeking to explain L2 reading comprehension variance. Individual
differences in exposure to both written and oral language also make signiﬁcant
contributions to L2 reading, as well as to the development of other language skills.

An abundance of research has reported signiﬁcant relationships between non-
linguistic variables, including motivation and anxiety, and L2 learning. However, direct
predictive effects of these affective variables on L2 reading comprehension have not been
convincingly demonstrated. Instead, it is likely that motivation and anxiety affect reading
outcomes by shaping behaviors, for example, by encouraging or discouraging learners to
read in the target language or to engage in other activities that put them in contact with
the L2.

Since reading involves an interaction between reader and text, however, it is not
sufﬁcient to examine only these reader-based differences. Text-based factors such as
genre also make important contributions to the reading process. Research in native
language contexts has conclusively established that narrative, story-based texts are more
easily comprehended than expository, informational texts. However, these results have

yet to be convincingly reproduced in L2 settings. Though some research suggests that

42

genre-level cues that may facilitate comprehension are only accessible to learners at
higher proﬁciency levels, the question of the interaction of genre and learner level in L2

reading comprehension also remains open.

43

CHAPTER 3: METHODOLOGY

The methodology for the experimental study to be described in this chapter was
adapted ﬁom methods used in a pilot study conducted at Michigan State University in
Spring 2006. I will therefore begin with a discussion of the pilot study’s methodology, its
results and their implications for the main experimental study. This will be followed by a

description of the methodology used in the present study.

3.1 The pilot study
3.1.1 Experimental design
The pilot study examined the effects of text genre and motivation for L2 study on
reading comprehension in French. Intermediate-level French students completed self-
report questionnaires assessing their motivations for learning French. Participants read
two narrative and two expository tests and performed ﬁee written recalls in English of
everything they remembered from each text. The experiment used a repeated measures
design; all research participants read the same texts and performed the same recall tasks
for each passage. Written recalls were scored based on a propositional analysis of each
text.
3.1.2 Research questions
The pilot study was limited to investigating the relationships between text genre
and individual differences in motivation for language study. Questionnaires and tests
measuring other individual differences, including L2 proﬁciency and L2 print exposure,
were also piloted in this study. However, statistical analyses were limited to answering

the following research questions:

44

1) Does text genre affect L2 (French) reading comprehension? From which genre
do students comprehend more?

2) Does motivation for language learning affect L2 reading comprehension? If
so, which motivational orientations (intrinsic, extrinsic, amotivation) contribute most?

3) Are effects of text genre moderated by motivational orientations?

3.1.3 Hypotheses
The hypothesized answers to these research questions were:

1) In light of consistent supporting evidence from L1 reading research (discussed
in section 2.2.2), it was expected that text genre would have a signiﬁcant main effect on
recall comprehension, and that participants would recall signiﬁcantly more ﬁorn
narratives than ﬁ'om expository texts.

2) It was expected that all motivational orientations (intrinsic, extrinsic,
amotivation), would affect reading comprehension, but that intrinsic motivation would
contribute most. Ll reading research has found relationships between both intrinsic and
extrinsic orientations and reading outcomes, with some research suggesting that intrinsic
motivation is a stronger predictor (Gottﬁied, 1990). L2 research has provided additional
evidence that intrinsic motivation predicts reading comprehension (Brantrneier, 2005).
Correlational research also indicates that negative relationships between amotivation and
L2 reading comprehension exist (Kondo-Brown, 2006).

3) No signiﬁcant interactions between motivational orientations and text genre
were expected given the absence of research demonstrating that relationships exist

between learner motivations and comprehension of particular types of texts.

45

3.1.4 Method
3.1.4.1 Participants
F ifty-seven adult learners of French were recruited ﬁom intermediate (third and
fourth semester) university French classes. The sample included 43 females and 14 males
ranging in age from 18-59 years (M = 20.6). Participants received 15 points extra credit
for participating in all study sessions. All participants were native English speakers.
3.1.4.2 Materials

The reading passages used in the pilot study included two narrative literary texts
and two expository informational texts. All passages were excerpted from longer
passages found in contemporary (published within the previous 10 years) intermediate-
level university textbooks or readers. The texts and their English translations are found in

Appendix A.

Narrative Text 1, Les trois motocyclistes (The Three Motorcyclists), a short story
by Marie Cardinal, is a recollection of a camping trip where the author and her
companions beﬁiended three American bikers. Narrative Text 2, L ’wufde Pciques (The
Easter Egg) from a short story by Henri Crespi, describes the role of an Easter tradition
in the evolution of a couple’s relationship. Expository Text 1, Le café dans la vie des
étudiants (The Café in Student Life) from _I_._a condition étudiante by Catherine
Va“abregue discusses the importance of cafés to French university students’ social
development Expository Text 2, La vague bouddhiste (The Buddhist Wave) is an excerpt
fI‘Om an article from the magazine L’Express about the increasing popularity of

Buddhism in France.

46

These reading passages were taken from contemporary intermediate-level
textbooks in order to approximate materials that students would be likely to encounter in
their French studies. These texts appeared to be “authentic” using Young’s (1999)
deﬁnition that authentic materials are unsimpliﬁed and are written for a native L2
population. The fact that the passages contained glosses and notes explaining lexical and
grammatical items (which were not included in the versions of the texts used in the study)
suggested that the texts had not been simpliﬁed, and their attribution to French
publications or authors other than the textbook author was taken as evidence that they
were written for a F rench-speaking audience. The majority of texts contained in the
intermediate-level textbooks and readers that were reviewed also appeared to be
authentic. Empirical evidence that learners at all levels of instruction are capable of
comprehending authentic texts exists. Allen, Bernhardt, Berry, and Demel’s (1988) cross-
sectional study of high-school learners, for example, showed that even ﬁrst-year students
of French, Spanish and German were capable of comprehending different types of
authentic expository texts. DuBravac and Dalle (2002) found that more advanced (fourth-
semester) French students could comprehend both authentic narrative and expository
texts. Young (1999), in fact, showed that intermediate students of Spanish actually
comprehended more from authentic texts of both genres than from texts whose
vocabulary and grammar had been simpliﬁed. Because they seemed representative of
reading materials used in the L2 classroom and because they are comprehensible for
learners from the beginning level forward, authentic texts were therefore judged

appropriate for use in the present study.

47

Efforts were made to select texts comparable in length (261-287 words) and in
lexical and syntactic complexity. Some differences in syntactic features existed between
texts, as it was not possible to ﬁnd authentic texts with exact correspondences between all
structural and grammatical characteristics without manipulation that would have rendered
them less than authentic. The expository texts, for example, tended to contain more words
per sentence, whereas the narrative texts contained more object pronouns and reﬂexive
constructions. An analysis of lexical and grammatical features of the texts is presented in

Table 1.

48

Table 1

Text characteristics, pilot study

 

Number of words
1K words a

Function
Content

2K words b

C

3K

Words not on list
Words per sentence
Object pronouns
Reﬂexive verbs
Relative pronouns

Semantic propositions d

 

 

Narrative Expository
N1 N2 M E1 E2 M
287 261 274 269 261 265
.77 .84 .81 .81 .74 .78
.51 .54 .53 .51 .49 .50
.24 .28 .26 .31 .25 .28
.06 .02 .04 .10 .08 .09
.01 .01 .01 .02 .04 .03
.15 .14 .15 .07 .15 .11
9.6 13.2 11.4 12.6 16.6 14.6
3 2.5 0 0 0
8 6 2 3 2.5
6 5 5.5 8 l 4.5
94 82 88 84 92 88

 

Note. N1:The Three Motorcyclists, N2: The Easter Egg, E1: The Café in Student Life, E2:
The Buddhist Wave. 3 Words appearing on the Lexical Frequency Proﬁle’s (Goodfellow,

Jones & Lamy, 2002) list of the one thousand most frequently

occurring words in French;

analysis performed using Web VP en francais (Cobb, 2006). Words 1001-2000. 6
Words 2001-3000. d Listed in Appendix B.

Because topic familiarity may affect reading comprehension (Allen et al., 1998;

Barry & Lazarte, 1998; Carrell, 1987; Chen & Donin, 1997), texts dealing with topics

that are likely somewhat familiar to intermediate French students were chosen (camping,

hiding Easter eggs, French student life, religion in France). These texts were also selected

because they contained few unfamiliar culture-speciﬁc references that might impede

49

comprehension. Participants’ actual familiarity with the text topics was assessed using a
questionnaire administered after the reading and recall tasks were completed.
3.1.4.3 Measures

3. I . 4. 3. I Predictor variables.

3.1.4.311 Text genre.

Participants read two texts representing each of the genres under investigation:
narrative (The Three Motorcyclists, The Easter Egg) and expository (The Cafe’ in Student
Life, The Buddhist Wave).

3.1.4.312 Motivation for French study.

Participants completed a 28-item questionnaire to assess their motivations for
studying French. The questionnaire was adapted from Vallerand, Pelletier, Blais, Brierc,
Senécal, and Vallieres’ (1992) Academic Motivation Scale/College Version (AMS), an
English version of an instrument originally used in motivation studies among French-
speaking college students (V allerand, Blais, Briere, & Pelletier, 1989). When tested with
English L1 college students (n = 217), the AMS had satisfactory reliability (mean
Cronbach’s alpha score for the intrinsic, extrinsic and amotivation subscales = .81) and
stability of scores over a one-month period (mean test-retest correlation = .79)

(V allerand, Pelletier, Blais, Briére, Seneca], & Valliéres, 1993). An adapted version of
the questionnaire has also been used to assess motivation of college students to study
another language. Noels, et al. (2000) reported the same internal reliability (mean a =
.81) when the questionnaire was used to measure English L1 students’ motivation to

study French.

50

In this pilot study, questions referring to college in general (e.g., “Because I think
a college education will help me better prepare for the career I have chosen”) were
modiﬁed similarly to Noels at al. (2000) to refer speciﬁcally to language study (e.g.,
“Because I think studying French will help me better prepare for the career I have
chosen”). The questionnaire included 12 items each assessing intrinsic motivation (e.g.,
“Because I experience pleasure and satisfaction while learning new things”) and extrinsic
motivation (e.g., “Because studying French will help me get a high-paying job”), and
forn' items measuring amotivation (e.g., “Honestly, I don’t know; I really feel I am
wasting my time studying French”). Students were asked to rate each item on a 7-point
Likert scale, where a response of 1 indicated that the item did not correspond at all for
their reasons for studying French and a response of 7 indicated it corresponded exactly. A
copy of the questionnaire is found in Appendix C.

3.1.4. 3.2 Outcome variable

Reading comprehension was assessed using free written recalls. After reading
each text, participants wrote everything that they remembered about the text in English
without looking back at it. According to Koda (2005), “[t]here is general agreement that
ﬁ‘ee recall is the most straightforward procedure for assessing the outcome of reader-text
interaction” (p. 236). This method is widely accepted in L2 reading research for several
reasons. Most importantly, recalling information ﬁom a text does not inﬂuence the
reader’s understanding of it. Other measures of comprehension, such as multiple-choice
or short-answer tests, may provide readers with cues about text meaning, prompting them
to make interpretations based on the question’s content instead of their own

understanding of the text. When used to assess comprehension of short passages such as

51

the ones used in this study, it may also be difficult to construct a sufficient number of
questions per passage to observe variability that achieves statistical signiﬁcance.
Illustrations of this difﬁculty can be found in some recent L2 reading studies, where
signiﬁcant effects of individual reader differences on recall comprehension scores were
noted but a lack of variance in multiple-choice test scores prevented duplication of these
results with this measure (Brantrneier, 2005; Fecteau, 1999).

Participants were asked to complete the recalls in English instead of French in
order to avoid any L2 proﬁciency effects on the recall task. Lee (1986a) found that
English Ll-Spanish L2 students at four different semester levels recalled signiﬁcantly
more when they wrote recalls in English instead of Spanish. He concluded that
“reproducing a passage in the target language imposes a ceiling due to the learners’
limited L2 production abilities; when learners are tested in their own language, this
ceiling disappears” (Lee, 1986, p. 185).

Each text was translated into English and broken down into propositions using a
system of propositional analysis adapted from Kintsch (1998). The propositional
breakdowns of the texts are found in Appendix B. Propositions consist of a predicate (a
verb, adverb, conjunction or other relational term) and its argument(s). The following
excerpt ﬁ'om Narrative Text 1 (The Three Motorcyclists) will be used to illustrate
propositional analysis:

Original French text: Gre'goire a pris son banjo.

English translation: Gregory picked up his banjo.

In this example, Gregory and banjo are arguments of the predicate picked up. The content

of this proposition is represented in the analysis as follows:

52

pick Up [Gregorx banjo}

This type of propositional analysis represents the textbase, or the underlying
semantic content of the text, instead of the text’s surface form. Chen and Donin (1997)
and Pulido (2004) note that propositional analysis’ validity as an assessment tool is well-
documented; it receives more support in relevant psychological literature than multiple-
choice tests, idea units, or methods assessing understanding of a text’s surface structure.

Participants’ written recalls were scored based on the number of propositions that
were recalled using a system adopted from Pulido (2004). Propositions containing a verb
as their relational term were worth two points. Those containing an adverb, conjunction
or other relational term were worth one point. Full credit was awarded for recalling the
entire proposition (e.g., predicate and all arguments), half credit for recalling part of the
proposition (e.g., predicate and/or some of the arguments), and no points were awarded if
no part of the proposition was mentioned. Because the texts were translated ﬁ'om French
to English in order to score recalls that participants would write in English, several
possible translations of words from the text were oﬂen possible. All were accepted as
full-credit responses. In the example cited above (Gregoire a pris son banjo/Gregory
picked up his banjo), for instance, both Grégoire and Gregory were judged as acceptable,
as were contextually-appropriate synonyms of the French verb prendre (originally
translated as pick up), such as get, get out, or take out.

3. 1. 4.3.3 Pilot instruments.

Though not directly related to this study’s research questions, the instruments

described in this section were also piloted in the study.

53

3.1.4.3.3.l Demographic questionnaire.

Participants completed a survey of demographic information including age, sex,
and previous French study experiences. Students were also asked about their major,
whether they were studying French to complete a degree requirement, whether they
would be studying French without any degree requirements, and the highest level of
French they intended to complete.

3.1.4.3.3.2 L2 proﬁciency test.

A 39-item pencil and paper test was created by adapting MSU’s computerized
French placement testz. Students who have previously taken French and who intend to
continue their language studies at MSU are required to take the placement test, typically
during ﬁeshman orientation, so that they may enroll in an appropriate level class. The
placement test is geared towards beginning and intermediate learners; students who miss
no more than one question are permitted to enroll in advanced level classes. This
instrument includes items testing listening (n = 6) and reading (n = 6) skills as well as
knowledge of grammar (n = 12) and vocabulary (n = 15). The listening and reading
sections consist of one passage followed by multiple-choice comprehension questions.
Grammar questions are also multiple-choice; each item asks students to choose the
grammatically correct response to complete a sentence. Vocabulary knowledge is
assessed by providing students with a description in French and asking them to choose
the word that does not ﬁt the description ﬁom a list of choices.

The pilot study proﬁciency test was constructed by selecting each of the 39 items

from a pool of four possible questions that are randomly generated when students take the

 

2 Because permission to adapt and use this proprietary test was granted on the condition that it not be
reprinted, a copy of the instrument is not provided here.

54

computerized version of the placement test. This random generation of items resulted in a
low probability (one in four) that participants who had previously taken the MSU
placement test would have encountered any of the particular items appearing on the pilot
study version of the test. It was also considered unlikely that participants who had
previously taken the placement test would have remembered individual items from it, as
the most recent date that any normally would have completed the test was nine months
prior to the pilot study. Moreover, it was expected that a signiﬁcant proportion of these
intermediate-level participants had begun their college French study at the beginning
level, so that they would have taken the placement test at an even earlier date if at all.
3.1.4.333 Reading activity questionnaire.

Exposure to French in print was assessed using an l8-question survey adapted
from Guthrie, McGough and Wigﬁeld’s (1994) Reading Activity Inventory, a measure of
reading amount ﬁrst used in L1 studies of elementary school students. Survey questions
were changed to ask about reading in French instead of L1 reading. The questionnaire
was divided into sections on reading for school and reading for pleasure. Students were
asked to report whether they had read certain types of French texts in the past week,
including French textbooks, literature or ﬁction, non-ﬁction, magazines, newspapers, and
online materials, and if so to identify the materials read. They were also asked how often
they read particular types of texts (almost every day, about once a week, about once a
month, or almost never) and for how many hours per day, week or month they typically

read these materials. This questionnaire is reproduced in Appendix D.

55

3.1.4.3.3.4 F arniliarity questionnaire.

Students were asked how familiar they had been with the topics of each of the
four reading passages before they had read them. They provided ratings on a 7-point
Likert scale where a response of 1 indicated the topic was not at all familiar and 7
indicated it was completely familiar. Students were also asked to rate the familiarity of
the two text genres that were the focus of this study (short stories and informational
texts). Items asking for familiarity ratings of other topics and genres were also included
to gather information for possible future research. The questionnaire is presented in
Appendix E.

3.1.4.3.3.5 Text reactions questionnaire.

Further information about participants’ experiences reading each of the four texts
was collected using a survey including both quantitative and qualitative items. First,
students were asked to rate the difficulty and their enjoyment of the four texts using a 7-
point Likert scale. Next, participants were asked to provide comments explaining their
difﬁculty and enjoyment ratings. This post-reading survey is found in Appendix F.

3.1.5 Procedure

Data was gathered in two sessions conducted one week apart. In the ﬁrst session
participants completed the demographic, Motivation for French Study, and Reading
Activity in French questionnaires, and the French proﬁciency test. During the second
session participants read the two narrative and two expository texts and completed free
written recalls in English. A repeated measures design was used; all participants read the
same four passages. In order to control for text presentation order effects, the texts were

counterbalanced. Eight different versions of the reading booklet were distributed where

56

each text was presented ﬁrst, second, third or fourth in two of the versions. In each
version the location of each text in relation to the others was also varied.

After reading each text, participants were instructed to write in English everything
they remembered about each text in as much detail as they remembered. No time limit
was placed on either the reading or recall tasks. Because the recall task was intended to
elicit the representation of text meaning that readers had processed and stored in long-
term memory, not information that remained in working memory after reading,
participants completed a short math task between the reading and recall to clear working
memory. Five simple math problems whose answers were numbers from 1-20 were
written in digits, and participants were asked to write the answer to each problem out in
French words (e.g., “2 + 3 = cinq”). Finally, after ﬁnishing the four written recalls,
participants completed the Familiarity and Text Reactions questionnaires. After all
participants had completed both experimental sessions, they were informed of the study’s
purpose via electronic mail.

3.1.6 Scoring

3.1.6.1 Motivation

Overall motivation to study French was calculated by tabulating a score for the 28
7-point questions on the Motivation for French Study questionnaire. Because questions
assessing amotivation were negatively worded (e.g., “Honestly, I don’t know; I really feel
I am wasting my time studying French”), scores for these items were reversed for the
purposes of calculating the global motivation score. For example, a score of 7, indicating
complete agreement with an item describing amotivation, was converted to a score of 1,

which would reﬂect complete disagreement with items reﬂecting a positive form of

57

motivation. Scores were also compiled for the three motivational orientations under
investigation from the items identiﬁed by Vallerand et al. (1992) that assessed each of
these constructs.

3.1.6.2 Reading comprehension

Participants’ comprehension scores were computed for each text by calculating
the percentage of propositions recalled. In order to examine variance in comprehension
scores by genre, a collapsed mean score was calculated for the two narrative texts and the
two expository texts.

3. I . 7 Results
3.1. 7.1 Descriptive statistics

3.1. 7.1.] Motivation.

Before tabulating descriptive statistics for motivation, Cronbach’s alpha, a
coefficient of reliability, was computed to test the internal consistency of the motivation
questionnaire. Cronbach’s alpha measures how well a set of items measures an
underlying construct by calculating average intercorrelations between items as a function
of the number of items. According to Nunnally (1978), Cronbach’s alpha levels of .70 or
higher are acceptable for research instruments. Internal consistency of the motivation
questionnaire was quite high overall (or = .97), as was consistency of items in the intrinsic
(or = .95), extrinsic (or = .95) and amotivation (or =.82) subscales. These values indicate
that the instrument was a reliable measure of motivation and its three component
orientations.

Descriptive statistics for motivation are displayed in Table 2.

58

Table 2

Descriptive statistics and correlations for motivation, pilot stuay

 

 

Mal? SD 1 2 3
Overall motivation 3.84 1.25 (.97)
1. Intrinsic motivation 3.56 1.08 (.95)
2. Extrinsic motivation 3.36 1.00 .84 (.95)
3. Amotivation 1.87 1.00 -.62 -.52 (.82)

 

Note. Values in parentheses are estimates of internal consistency (Cronbach’s or).
a Minimum possible = 1 (reason does not correspond at all); maximum possible = 7

(reason corresponds exactly). b n = 57.

The mean score for overall motivation (3.84) was just slightly above the midpoint
on the 7-point Likert scale, indicating that on the whole participants were moderately
motivated to study French. Means for intrinsic (3.56) and extrinsic (3.36) orientations
were quite similar to each other. Scores for amotivation were considerably lower than the
other motivation scores. The average amotivation response (M = 1.87) indicated that
items describing amotivation corresponded little if at all to participants’ reasons for
studying French. A ﬁ'equency analysis of questionnaire items also revealed that
amotivation items elicited stronger disagreement than did items in the intrinsic or
extrinsic subscales. 65% of participants responded that amotivation reasons did not
correspond at all to their reasons for L2 study. This contrasted with approximately 20%
of respondents who reported that intrinsic and extrinsic orientations did not reﬂect their

reasons for language study.

59

3.1. 7.1.2 Reading comprehension
Descriptive statistics for reading comprehension are found in Table 3.
Table 3

Descriptive statistics for reading comprehension, pilot study @ropositions recalled)

 

 

 

 

 

 

M 3 SD min max

Narrative

N1 .18 .10 .04 .49

N2 .19 .1 1 .01 .49

Both narrative .19 .10 .03 .49
Expository

E1 .09 .04 .02 .18

E2 .17 .10 .02 .47

Both expository .13 .06 .02 .33
All texts .16 .08 .03 .41

 

Note. N1 :The Three Motorcyclists, N2: The Easter Egg, E1: The Café in Student Life,
E2: The Budcﬁrist Wave. a n = 57.

Overall, participants recalled 16% of all propositions from the texts. Collapsed
means show superior recall for narratives (19%) over expository texts (13%). Participants
also comprehended more ﬁom each narrative passage than from either of the expository
texts. Nearly identical percentages of propositions were recalled ﬁom both narrative
passages (1 8-19%); however, participants recalled nearly twice the percentage of

propositions ﬁ'om Expository Text 2 (17%) than ﬁom Expository Text 1 (9%).

60

3.1. 7.2 Data analysis

3.1. 7.2.1 Research Question 1: Text genre and L2 reading comprehension.

Research Question I asked whether participants would comprehend more ﬁom
narrative or expository texts, and if text genre affects L2 reading comprehension. It was
hypothesized that signiﬁcantly more would be recalled from the narrative texts, and that
genre would have a signiﬁcant main effect on L2 reading comprehension.

Collapsed means reported in Table 3 showed that participants did in fact
comprehend more from the narrative passages. To determine whether this difference was
statistically signiﬁcant, a paired sample t-test was conducted. The t-test is used to identify
statistically signiﬁcant differences between means. Results conﬁrmed that participants
recalled signiﬁcantly more from the narrative passages [t = 6.778, df = 56, p = .000].

The effect of text genre on reading comprehension was assessed using repeated
measures factorial ANOVA. This test measures the statistical signiﬁcance of
relationships between more than one categorical predictor variable and one continuous
outcome variable, and is used in experimental designs where all subjects perform the
same tasks (in contrast to experiments where different groups of participants perform
different tasks or receive different experimental treatments). A 2 x 2 AN OVA was
conducted with genre and text as ﬁxed factors, each with two levels. For genre these
levels were narrative and expository; within these genres the two levels were Text 1 and
Text 2. ANOVA results are presented in Table 4. In this analysis, both genre [F = (1, 56)
= 46.38, p = .000] and text [F = (1, 56) = 22.82, p = .000] had signiﬁcant main effects on
reading comprehension. In other words, signiﬁcant differences in reading performance

existed depending on the genre and also on the text read. The size of genre’s effect on

61

reading comprehension as determined by its Pearson’s product-moment correlation
coefficient (r) was quite large3 (r = .67). The effect of text on reading comprehension was
smaller than that of genre, but still quite large (r = .54). However, a significant interaction
between genre and text was also found [F = (1, 56) = 15.52, p = .000], conﬁrming that
comprehension scores for Expository Text 1 and Expository Text 2 were signiﬁcantly
different. This effect size was moderately large (r = .47).

Table 4

Analysis of variance for effect of genre and text on reading comprehension, pilot study

 

 

Source df ss MS F p
Genre 1 .17 .17 46.38 .00
Text 1 .10 .10 22.82 .00
Genre x Text 1 .08 .08 15.52 .00
Error 56 .30 .01

 

3. 1. 7.2.2 Research Questions 2 and 3: Motivation and L2 reading comprehension.

Research Questions 2 and 3 asked about the relationship of motivational
orientations to reading comprehension as well as to differences in reading comprehension
by genre. More speciﬁcally, Research Question 2 asked if motivation for language
learning affects L2 reading comprehension, and if so, which motivational orientations
have the greatest effect. It was expected that all motivational orientations would affect L2
reading comprehension, with intrinsic motivation making the greatest contribution.
Research Question 3 asked whether motivational orientations moderated the effects of

genre on reading comprehension, or in other words, whether reading comprehension

 

3 Cohen (1992) proposes that r values of .10, .30 and .50 represent small, moderate, and large effect sizes,
respectively.

62

scores for either genre varied according to the reader’s levels of intrinsic motivation,
extrinsic motivation or amotivation. No signiﬁcant interactions between motivational
orientations and text genre were expected.

Another iteration of the factorial ANOVA described in section 3.1.7.2.2 was
performed to answer these questions. This test was identical in design to the previously
described AN OVA (2 x 2 with genre and text as ﬁxed factors), however, it added
intrinsic motivation, extrinsic motivation and amotivation as covariates in the proposed
model. This test therefore would account for effects of these motivational orientations on
reading comprehension and would identify any interactions between motivational
orientations and genre or text. The results of the analysis are presented in Table 5. None
of the motivational orientations had signiﬁcant effects on reading comprehension, and no
signiﬁcant interactions between motivation and genre or motivation and text were
present. However, in contrast to the ﬁrst analysis of variance that found signiﬁcant main
effects for genre and text, in the analysis that controlled for motivational orientations
these effects became non-signiﬁcant [genre F = (1, 53) = 2.80, p = .10;
text F = (1, 53) = .33, p = .57]. The interaction between genre and text, however,

remained signiﬁcant [F= (1, 53) 4.75,p = .03].

63

Table 5
Analysis of variance for eﬂect of genre and text on reading comprehension with

motivational orientations as covariates, pilot study

 

 

Source df SS MS F p
Genre 1 .01 .01 2.80 .10
Genre x Intrinsic 1 .00 .00 .30 .58
Genre x Extrinsic 1 .00 .00 .24 .62
Genre x Amotivation 1 .00 .00 .24 .63
Text 1 .00 .00 .33 .57
Text x Intrinsic 1 .00 .00 .28 .60
Text x Extrinsic 1 .00 .00 .03 .87
Text x Amotivation 1 .00 .00 .04 .84
Genre x Text 1 .03 .03 4.75 .03
Genre x Text x Intrinsic 1 .00 .00 .57 .45
Genre x Text x Extrinsic 1 .02 .02 .29 .10
Genre x Text x Amotivation 1 .01 .01 2.73 .11
Error 53 .28 .015

 

3.1. 7.3 Pilot instruments
Although not directly related to the pilot study research questions, the following
instruments were also tested for possible use or adaptation in the planned follow-up

study.

3.1. 7.3.1 L2 proﬁciency test.

Descriptive statistics of the proﬁciency test scores are summarized in Table 6.
Internal consistency of the 39 test items (a = .65) indicated that the instrument was a
reasonably reliable measure of French proﬁciency. The mean proﬁciency test score was
60%, with scores ranging between 33% and 87%. Participants scored highest on average
on the listening section (81%). Vocabulary and reading scores were quite similar,
averaging 64% and 61% respectively. Performance was weaker on the grammar section,
with a mean score of 45%. With the exception of this grammar section, ceiling effects
were observed in all sections of the test. Frequency analyses showed that nearly 40% (n =
22) of participants attained perfect scores on the listening section, 7% (n = 4) scored
100% on the reading section, and one of the 57 participants scored perfectly on the
vocabulary section. A ﬂoor effect was also observed in the reading section, where two of

57 participants answered all questions incorrectly.

65

Table 6

Descriptive statistics and correlations for French proﬁciency, pilot study

 

 

Ma SD 1 2 3 4
Total score .60 .12 (.65)
1. Listening .81 .19 (.37)
2. Vocabulary .64 .17 .19 (.47)
3. Reading .61 .24 .14 .26 (.55)
4. Grammar .45 .17 .24 .38* .13 (.29)

 

Note. Values in parentheses are estimates of internal consistency (Cronbach’s a). Mean
values reﬂect proportion correct. Correlations marked with an asterisk C“) are statistically

signiﬁcant at the .01 level. a n = 5 7.

3. 1. 7.3.2 Reading activity questionnaire.

This 18-item survey asked participants to report the amount that they read in
French for both academic purposes and for pleasure. Because the questionnaire allowed
students to indicate the number of hours per day, week or month they read each type of
material, reported horns were converted to hours per week by dividing hours per month
by 4 or multiplying hours per day by 7 (e. g., 2 hours per month was converted to .5 hours
per week; 1 hour per day was converted to 7 hours per week). Reading activity results are
displayed in Table 7. On average, students reported reading 5.75 hours per week in
French. The preponderance of this reading (5.15 hours, or 90%) was for school. Just
under 70% of this academic reading was ﬁom textbooks. An additional 14% was reading
nonﬁction books, and about 17% was reading newspapers, magazines or online materials.
Reading literature or ﬁction accounted for less than 1% of all school-related reading. In
contrast, students who read for pleasure reported spending 17% of their reading time

reading ﬁction books and only 5% reading nonﬁction books. Online reading constituted

66

the bulk (47%) of reported reading for pleasure. Magazines and newspapers accounted
for an additional 13% of this pleasure reading time. Finally, students reported reading
other materials for pleasure 18% of the time. Overall, however, students reported
spending very little time - just over half an hour per week - reading materials unrelated to
French class. A frequency analysis further showed that the majority of these
intermediate-level students (67%) reported spending no time at all reading in French for

their own interest or pleasure.

67

Table 7

Descriptive statistics for reading activity in French, pilot study (hours per week)

 

 

 

 

 

 

M3 SD min max

Academic reading 5.15 5.74 .00 24.00

Textbook 3.53 2.81 .00 10.50

Other (magazine, newspaper, online) .85 3.32 .00 21.00
Nonﬁction book .73 1.77 .00 7.00
Literature or ﬁction book .04 .27 .00 2.00

Reading for pleasure .60 1.81 .00 12.00
Online .28 1.00 .00 7.00

Other .11 .42 .00 3.00
Fiction book .10 .51 .00 3.75
Magazine or newspaper .08 .20 .00 1.00
Nonﬁction book .03 .10 .00 .50

Total reading 5.75 5.74 .00 24.00

 

n =57.

3.1. 7. 3. 3 Familiarity questionnaire.

After reading and completing written recalls for all four reading passages,
participants were asked to rate their pre-reading familiarity with the text topics on a 7-
point Likert scale. Students were also asked to rate the familiarity of the two text genres
that were the focus of this study, short stories (narratives) and informational (expository)
texts. Familiarity results are reported in Table 8.

Participants reported similar levels of familiarity with both Narrative Text 1’s (M
= 2.00) and Narrative Text 2’s (M = 2.18) topics. These 7-point scale ratings indicated

that the narrative topics had not been very familiar to them before they read the texts. A

68

paired-sample t-test conﬁrmed that the familiarity ratings for the two narrative texts were
not signiﬁcantly different (p = .000). Participants were more familiar with the expository
text topics than with the narrative topics, differences that were statistically signiﬁcant

(N l-El p = 000; Nl-E2 p = .001; N2-E1 p = .000, N2-E2 p = .02). Participants also
reported being signiﬁcantly more familiar with Expository Text 1’s topic (cafés in French
student life) than that of Expository Text 2 (Buddhism in France) (p = .01). The mean
familiarity ratings for the expository texts (E1 M = 3.35; E2 M = 2.74) indicated that
these topics had been a little familiar to participants before reading the texts.

Turning to genre familiarity, scores for the expository genres (M = 5.25) were
slightly higher than those of the narrative genre (M = 4.96). However, this difference was
not statistically signiﬁcant, indicating that participants were equally familiar with both
genres. The mean ratings for both genres combined (M = 5.11) indicated that the

narrative and expository genres were both very familiar to them.

69

Table 8

Descriptive statistics for topic and genre familiarity, pilot study

 

 

 

 

 

Ma SD min max

Topics

N1: The Three Motorcyclists 2.00 1.51 1 6

N2: The Easter Egg 2.18 1.71 1 7

E1: The Cafe' in Student Life 3.35 1.73 l 7

E2: The Buddhist Wave 2.74 1.48 1 6
Genres

Narratives (short stories) 4.96 1.41 2 7

Expository (informational) texts 5.25 1.43 2 7

 

Note. Minimum possible score = 1 (not at all familiar), maximum = 7 (completely
familiar). a n = 57.

3. 1. 7.3.4 Text reactions questionnaire.

A second post-reading questionnaire asked participants to rate the difﬁculty of the
four reading passages as well as their level of enjoyment of each text. These ratings are
reported in Table 9.

Collapsed results show that overall participants found the narrative texts more
difﬁcult (M = 3.80) than the expository texts (M = 3.22), a difference that was
statistically signiﬁcant (p = .004) as measured by paired-sample t-test. These ratings
indicated that narrative texts were “somewhat difficult” and expository texts were “a little
difﬁcult”. Mean difﬁculty scores by text showed no signiﬁcant differences in difﬁculty
ratings for the two narratives (p = .53) or the two expository texts (p = .06).

No signiﬁcant differences existed between the reported enjoyment of narrative

and expository texts (p = .09). Mean enjoyment responses (narrative M = 4.11; expository

70

M = 4.43) indicated that overall participants neither liked the texts representing both
genres, though not intensely. Though there were no signiﬁcant differences in enjoyment
between the two narrative texts (p = .20), participants reported enjoying Expository Text
2 signiﬁcantly more than Expository Text 1 (p = .004).

Table 9

Descriptive statistics for difﬁculty and enjoyment ratings, pilot stuay

 

 

 

 

 

 

 

Difficulty a Enjoyment b
M” SD min max M SD min max

Narrative texts 3.80 1.08 1 6 4.11 1.19 1.5 7

N] 3.72 1.31 1 6 3.95 1.33 1 7

N2 3.88 1.54 1 7 4.28 1.72 1 7
Expository texts 3.22 .94 1 5 4.43 1.02 2 6.5

E1 3.42 1.28 1 6 4.09 1.33 1 7

E2 3.02 1.17 1 6 4.77 1.34 2 7

 

Note. a Minimum possible = 1 (not at all difﬁcult); maximum possible = 7 (extremely
difﬁcult). b Minimum possible = 1 (disliked completely); maximum possible = 7 (liked a

lot). err = 57.
Participants were also asked to provide comments explaining their difficulty and
enjoyment ratings. Eight different categories were identiﬁed for these responses:
1) Vocabulary. The text’s vocabulary was difficult or easy, known or unknown.
2) Grammar. Students knew or did not know the grammatical structures

contained in the text.

71

3) Comprehension: Students’ ratings were based on whether they understood or

4)

5)

6)

did not understand the text, without providing additional explanation of what
made the text comprehensible or not.

Complexity. Students found the text simple or complex for reasons other than
its lexical or grammatical content (e. g., “There was a lot of action and it didn’t
seem to ﬂow together”).

Genre. The text’s genre was identiﬁed as the source of diﬁiculty/ease or
enjoyment/lack of enjoyment (e. g., “Easier because it was a story”).

Familiarity. Students were or were not familiar with the text topic.

7) Aﬂect. Ratings were based on a positive or negative affective reaction to the

8)

text (“I thought it was entertaining”; “The topic was not so interesting”).
Task A few students identiﬁed aspects of the experimental task as helpful or
detrimental to their reading and text recall. Several of these comments had to
do with the presentation order of the texts. For example, one participant noted

that for the ﬁrst text, “I didn’t know what to expect so I didn’t read carefully”.

In many cases students provided more than one reason justifying their responses; all
reasons were tallied. The number of comments in each category was tabulated per text,
then by genre, and ﬁnally for all texts. Comments were also classiﬁed based on whether
they reﬂected positively or negatively on students’ perceptions of difﬁculty and
enjoyment of the texts. For example, “I knew most of the words and could ﬁgure out
meanings of unfamiliar words ﬁorn the context of the sentence” was counted as a positive
comment on difficulty in the vocabulary category, while, “[t]here were a variety of words

I didn’t know” was a negative comment. “I liked reading this text because I like. . .news

72

 

information articles” was considered a positive comment related to enjoyment in the
genre category, while “. . .it was not in story form which made it very dull” was counted
as a negative.

The results of these analyses of the comments are presented in Table 10. For both
the narrative and expository texts, perceptions of difﬁculty were overwhelmingly
attributed to text vocabulary. In contrast, only a few comments (less than 3%) connected
grammatical knowledge to text difﬁculty. Overall, comments revealed that not knowing
the relevant vocabulary made a text difﬁcult more than knowing it made the text easier. A
greater proportion of comments indicating vocabulary difﬁculty were recorded for the
narrative (about 51%) than for the expository texts (35%). The next largest number of
comments for all the texts (about 20%) expressed difﬁculty as a lack of comprehension
without attributing the difﬁculty to any particular feature of the passage. Participants had
differing perceptions of the roles of topic familiarity and genre in the difficulty of the
narrative and expository genres. Only about 5% of comments mentioned topic familiarity
in relation to difﬁculty for the narratives, while nearly 20% of difﬁculty comments for
the expository texts indicated that topic familiarity played a role. However, all of the
participants who mentioned genre in relation to difficulty for the narrative texts found
that this genre facilitated comprehension (e. g., “It was easier than some others because it
was a story”) while the perception of the expository genre’s effect on text difficulty was

split between positive and negative.

73

 

 

 

 

 

 

 

m v S N v Nﬁ NN ~ ~ on on n _ ma Nm

2 a an 3 mm co 02 3x8 =<
m N m m N a m m _ N. ﬂ _ 2 m 0 Ne N~

o n on a ON NN em 50m
N N m n v w v o S N o _N n

n v 3 o a 3 N Nm
m o N E _ w m m c _ c _N m

m m ON m _ _ _ _ N 5
o N m m o N a v 2 E v ~ on 0N

m N w n 3 mm 2. 50m
o 0N v N o m o N o a N o N S

v N o m w m _ cm NZ
o o _ ~ o v m N 2 m N _ cm 2

_ o N a m 3 ca _Z
- + - + + - + - + - + - + .2355

x3... Hoob< ham—mash 9:80 bmonEoU .conaneU 358.80 E3ﬁoo>

 

ASE. ENE .étmsﬁec «8.283 NHQ

S 033.

74

 

 

 

 

 

 

 

 

Nv m3 N. 3 N N on N o v
m 9: m N S on _ _ mesa =<
5N mm b m H m _ S _ _ o m
_ om NN N_ N o 53
a Nm N m N m o o N o m
o 3 N N. 2 o Nm
2 ”N m S o ﬂ _ _ v o o
_ an m _ m m g o E
m _ on o m E _ am 2 o ﬂ
N we m 9 Nm m 50m
n Nm o m o w o S e o o
_ on m a N _ NZ
w 3 o o 3 # NN n o ~
_ N o _ ﬂ N v _Z
- + - + + + - + + + ceoéowcm
x3... 60% bis—Ban 2:00 bus—980 .eonoEEoU 388.20 $39a8>

 

 

@035qu 0— 073%

75

Comments explaining participants’ enjoyment ratings, not surprisingly, centered
on affective reactions to the texts. A greater percentage of affect-related comments were
provided for the expository texts (53%) than the narratives (44%). Many more of the
affective comments for the narratives passages, however, were positive (77%) than were
comments on the expository texts (51%). Within the expository genre, Expository Text 2
inspired a greater proportion of positive affective comments (78%) than Expository Text
1 (54%). Differences were not as pronounced between the two narrative texts (69% and
82% for Texts 1 and 2, respectively). Turning to other, non-affective factors, 27% of
comments mentioned comprehensibility as an important component of text enjoyment or
lack thereof. About 10% implicated genre in enjoyment of the texts, with different
patterns of genre’s role emerging for the narrative and expository texts. All but one of the
comments (95%) stated that the narrative genre played a positive role in text enjoyment,
whereas more than 30% of comments for the expository texts indicated that the
expository genre had negatively affected their enjoyment of the text. A large portion of
these negative comments, however, were made in response to Expository Text 1; genre-
related comments for Expository Text 2 were largely positive (71%). Nearly 90% of
comments on topic familiarity’s role in text enjoyment were made for the expository
passages, and for these texts, comments reﬂected a mix of positive and negative
familiarity effects on enjoyment of the texts. Interestingly, although some participants
viewed familiarity as an aid to comprehension that therefore improved their enjoyment of
the expository texts (or, similarly, reduced their enjoyment when they didn’t know
enough about the subject), several others found that too much familiarity with an

expository text’s topic made it less enjoyable to read (e.g.; “I liked this the least because

76

I’ve heard it all before”; “This was a bunch of information that I mostly knew”). In the
few topic familiarity comments for the narrative texts, however, both familiarity and lack
of familiarity were viewed as positive (e. g.; (“I liked this story the best because I was
familiar with the Easter tradition”; “Cool, because I didn’t know about this tradition”). In
contrast with the comments on text difﬁculty, vocabulary knowledge was rarely
mentioned as a factor in text enjoyment, and grammatical knowledge was not mentioned
at all.

3.1.8 Discussion and implications for the present study
3.1.8.1 Text genre and motivation

The pilot study research questions asked about the relative comprehensibility of
L2 narrative and expository texts, as well as the effects of text genre on L2 reading
comprehension. They also inquired about the contributions of motivation and its
component orientations to L2 reading comprehension and the interactions of motivational
orientations with any text genre effects.

Results showing superior L2 reading comprehension for narrative texts provide
some preliminary evidence conﬁrming L1 ﬁndings that narratives are more
comprehensible than expository texts, at least among intermediate learners. The
interaction between genre and text present in both analyses of variance, however, calls
for further consideration of these results. Though collapsed means revealed signiﬁcant
differences between comprehension of the two narratives and two expository texts,
comparisons of means of individual texts showed that signiﬁcant differences in
comprehension existed only between the two narrative texts and Expository Text 1.

Statistically, participants comprehended as much from Expository Text 2 as from both of

77

the narratives. This ﬁnding prompted a post hoc examination of this text, which revealed
that it contained some narrative elements that appear to have aided comprehension. The
passage, The Buddhist Wave, uses a description of the daily routine of one practitioner of
Buddhist meditation to introduce information on the increasing popularity of Buddhism
in France. This four-sentence portion of the text shares many of the characteristics of
narrative texts identiﬁed by Graesser et a1. (1991) and Larsen (1984) that may facilitate
comprehension and recall: it introduces a character (Francois) and it recounts a sequence
of events (he gets up, showers, dresses, and meditates) situated in a speciﬁc place and
time (every morning in his Parisian apartment). Participants in fact recalled signiﬁcantly
more propositions in this narrative-like section (24%) than in the remainder of the text
(13%). This analysis of Expository Text 2 seems to provide further evidence that
narrative text structures aid comprehension in the L2. However, results on the relative
comprehensibility of the two genres would have been more convincing absent the
interaction between genre and text. It is expected that replacing Expository Text 2 in the
main experimental study with a text that does not contain narrative-like elements will
eliminate this interaction.

Factorial AN OVA results also suggest that when individual reader differences are
accounted for, genre may play less of a role in L2 reading comprehension. Though a
signiﬁcant main effect for genre was found when genre and text were the only predictor
variables included in an analysis of variance, genre’s effect became non-signiﬁcant when
learners’ motivational orientations were added. This suggests that reader-based affective
differences may moderate text genre effects, and that contributions of individual

differences to comprehension of each genre may be different. The effects of genre in

78

relation to individual reader differences, however, merits further investigation with a
greater array of individual difference variables.

The lack of signiﬁcant relationships between motivation for L2 study and reading
outcomes suggests that motivation to study a language is not a proximate, determining
factor in overall reading success or in success reading either genre. The absence of
relationships between intrinsic motivation, extrinsic motivation or amotivation and L2
reading comprehension differs from ﬁndings of certain other reading studies. In L1
contexts, for example, relationships between both intrinsic and extrinsic motivation and
reading comprehension have been observed (Gottfried, 1990; Guthrie et al., 1998;
Guthrie et al., 1999). Some evidence of relationships between intrinsic motivation
(Brantrneier, 2005) and amotivation (Kondo-Brown, 2006) and L2 reading also exists.
One explanation for the difference between these results and those of the pilot study may
relate to differences in the type of motivation that was measured. In contrast to this pilot
study, the above-referenced research by and large has focused on motivation to read
instead of general motivation to study a language. Though previous research has
identiﬁed similarities in the components comprising motivation for L2 study and
motivation for L2 reading (Kondo-Brown, 2006; Mori, 2002), there is also some
evidence that motivation for L2 study and L2 reading are two distinct constructs (Mori,
2004). Learners may study a foreign language for reasons unrelated to those that motivate
them to read (or not to read) in that language, and the latter might be better predictors of
reading outcomes.

Participant comments also provided some qualitative evidence that the motivation

questionnaire used in the pilot study did not tap the full range of reasons why students

79

study another language. One participant, who clearly did not ﬁnd the reasons provided in
the questionnaire adequate to describe her motivation for studying French, was prompted
to write “[h]ow about I want to be able to communicate in France & I think its (sic)
pretty” in the questionnaire margin. Because the questionnaire was adapted from an
instrument originally used to assess general academic motivation and not motivation for
language study, it in fact did not contain questions about these sorts of communicative
and esthetic reasons that may induce students to study another language. Adding items to
the questionnaire that measure such motivations may yield different results.

Though the analyses performed in the pilot study sought to ﬁnd a direct
relationship between motivation for L2 study and reading comprehension, it has been
suggested that the relationship between motivation and language learning outcomes is
indirect (D6rnyei & Kormos, 2000). It is therefore possible that motivation for language
study moderates other variables that may in turn affect reading comprehension.
Motivation may, for example, drive students to seek out opportunities to use or be
immersed in the L2, to expend time and effort studying, or to use learning strategies to
improve their performance. It will be important, therefore, to include variables such as
language proﬁciency, language contact and metacognition in the statistical analyses of
the main experimental study, and to use statistical analysis methods that allow any of
motivation’s possible indirect effects to be assessed.
3.1.8.2 L2 proﬁciency

Analyses of internal consistency showed that the French test used in the pilot
study was reasonable reliable measure of French proﬁciency. However, ceiling effects

observed in three of the four sections of the test conﬁrmed that this instrument is not of

80

sufﬁcient difficulty to test the language proﬁciency of students beyond the intermediate
level. The difficulty of this test will therefore need to be increased when used with
advanced learners in the present study.
3.1.8.3 Reading activity

Results of the reading activity questionnaire showed that the majority of students’
target language reading is from textbooks and is related to their French coursework.
About two-thirds of participants reported never reading in French for non—academic
reasons, and those who do read for pleasure reported spending very little time doing so.
These results underscore the importance of measuring other, non-print sources of L2
contact in the present study.
3.1.8.4 Topic and genre familiarity

Though participants reported being signiﬁcantly more familiar with the
expository text topics than those of the narrative texts, comprehension scores for the
expository texts were lower. Thus, topic familiarity did not appear to predict reading
comprehension in the pilot study. Topic familiarity is therefore not expected to be
function as a predictor of reading comprehension in the present study. The lack of
reported differences in students’ familiarity with the two genres under investigation in
this study also helps rule out any familiarity effects on comprehension scores for each
genre.
3.1.8.5 Text reactions

Somewhat paradoxically, even though participants comprehended more overall
ﬁom the narrative texts, they perceived these texts as more difficult than the expository

passages. Yet students who mentioned genre in relation to text difﬁculty in their

81

comments all believed that the narrative genre facilitated comprehension. Instead,
comments attributed difﬁculty largely to the vocabulary contained in the texts. The
perception of narratives as more difﬁcult than the expository texts might be further
explained by students’ minimal exposure to narratives relative to expository texts.
Responses on the Reading Activity questionnaire indicated that less than 1% of the time
students devoted to reading for school was spent reading literary texts. When reading for
pleasure, students also reported reading narratives only about 16% of the time. Students’
perceptions of difficulty of the narrative texts read in the pilot study, then, might be
attributed in part to their lack of experience with narratives in the target language. These
experiential effects on students’ perceptions of each genre are expected to be replicated in
the present study.

No signiﬁcant differences were reported in students’ enjoyment of the narrative
and expository texts, though their enjoyment responses were rather neutral. Student
comments, however, indicated more positive affective responses to the narrative texts

than the expository texts, a result that is also expected in the present study.

3.2 The present study
3. 2. 1 Experimental design
The present study sought to expand upon the pilot study by examining the effects
of text genre and of a wider range of individual difference variables on reading
comprehension in French. It investigated these effects among a cross-section of learners
at three different levels of French study (beginning, intermediate and advanced). A design

similar to the pilot study was used. Participants completed questionnaires and other

82

measm'es of the individual difference variables under investigation (L2 proﬁciency, L1
reading ability, metacognition, L2 exposure, motivation and anxiety), then read two
narrative and two expository tests and performed free written recalls in English of
everything they remembered from each text. As in the pilot study, a repeated measures
design was used in which all participants read the same texts and completed the same
recall tasks. Participants’ written recalls were again scored based on propositional
analyses of the texts.
3. 2.2 Research questions

The research questions in the present study deal primarily with the contributions
of individual differences and text genre to reading comprehension in French. These
questions are presented in detail below:

1a) What are the contributions of readers’ L2 proﬁciency, L1 reading ability,
metacognitive knowledge, L2 exposure, motivation and anxiety to L2 reading
comprehension?

1b) What are the relationships between these individual differences in a model of
L2 reading comprehension?

2a) Is there a difference in L2 reading comprehension due to text genre (narrative
and expository)?

2b) Are there differences in comprehension of each genre by learner level
(beginning, intermediate, advanced)?

2c) Are there differences in the contributions of individual difference variables to

comprehension of each genre?

83

3a) What are learners’ affective reactions to reading each genre in the L2? Do
they prefer reading narrative or expository texts in the L2? Are there differences in these
reactions by learner level?

3b) What are learners’ affective reactions to reading the texts in this study? Do
these reactions differ by genre, or only by text?

3. 2. 3 Hypotheses

The following hypotheses are proposed in response to the research questions:

1a) Consistent with numerous studies demonstrating substantial contributions of
L1 reading ability and L2 knowledge to L2 reading comprehension (Bernhardt & Kamil,
1995; Bossers, 1991; Brisbois, 1995; Carrell, 1991; Gelderen et al., 2004; Lee &
Schallert, 1997), it is expected that both of these variables will also make direct
contributions in the present study.

1b) The following indirect effects of individual difference variables are expected:

i) Metacognition is hypothesized to have an indirect effect on L2 reading
comprehension by contributing to L1 reading ability. Though relationships between
metacognitive knowledge and L2 reading comprehension have been observed in some
research (Schoonen et al., 1998), Gelderen et al.’s 2004 study provided evidence that
metacognition’s contribution is moderated by L1 reading ability when both of these
variables are taken into accormt.

ii) Given the large corpus of research showing positive relationships between
exposure to the target language and development of L2 knowledge and skills (see

discussion in section 2.1.2.1), it is expected that L2 exposure will contribute to L2

proﬁciency.

84

iii) Consistent with Domyei and Kormos’ (2000) observations, it is hypothesized
that motivation will make an indirect contribution to L2 reading comprehension.
Motivation is expected to contribute directly to students’ L2 exposure and in turn to their
L2 proﬁciency. Motivation is also expected to contribute to L2 study habits, a measure of
motivational intensity (Mori, 2004).

iv) Students’ L2 study habits are expected to contribute indirectly to reading
comprehension through positive effects on L2 proﬁciency.

v) Anxiety is expected to have indirect negative effects on L2 reading by
inhibiting L2 exposure and L2 proﬁciency.

23) Based on results from L1 text genre research as well as ﬁ'om the pilot study, it
is expected that students will comprehend signiﬁcantly more from the narrative passages
than ﬁom the expository texts.

2b) Learners at all levels of French instruction are expected to comprehend more
from the narratives than from the expository texts. However, given research suggestive of
a linguistic threshold for effects of global-level text features such as genre (Chen &
Donin, 1997; Horiba, 1990), differences in comprehension of the two genres are expected
to be greater among advanced learners than either intermediate or beginning learners, and
greater among intermediate level learners than beginners.

2c) In light of L1 evidence of narrative text structures’ facilitative role in reading
comprehension, it is expected that the contributions of L1 reading ability and L2
proﬁciency to narrative text comprehension will be less than their contributions to
expository text comprehension. In other words, language-related knowledge and skills

will be less important in narrative than in expository comprehension.

85

3a) Given that one of the primary purposes of narrative texts is to engage readers
affectively, it is expected that participants of all levels will have greater positive intrinsic
affective reactions (enjoyment, interest) to reading narratives than to expository texts in
the L2, and will express a preference for reading narratives over expository texts. In light
of expository texts’ primarily instructive purpose, it is expected that learners will believe
that expository texts have a greater extrinsic utility value than narratives. Because
students enrolled in upper-level French classes likely have more experience reading in
general in the L2 and reading narrative L2 texts in particular, however, it is expected that
advanced students will report signiﬁcantly greater more positive affective reactions
(enjoyment, utility, comprehensibility) and less anxiety related to both genres than
intermediate and beginning-level students, and that these differences will be more
pronounced for the narrative than the expository genre.

3b) It is also expected that students will report greater levels of intrinsic interest in
and enjoyment of the narrative texts they read in the study for similar reasons as those
described above; narrative texts are written with the intention of engaging readers
affectively. Though students reported enjoying the narrative texts more in the pilot study,
these results were not statistically signiﬁcant. In the present study, it is expected that the
larger and more diverse sample and a change in the scale of questionnaire items from 7 to
6 points (intended to eliminate neutral responses) will result in signiﬁcantly greater levels
of reported interest in and enjoyment of the narrative texts. Students are also expected to
ﬁnd the narrative texts more comprehensible and less anxiety provoking because of the
positive affective reactions they inspire. In contrast, given the informative, utilitarian

nature of expository texts, it is expected that students will place a greater extrinsic utility

86

value on the expository passages. However, in light of the varied reactions to each text
reported in the pilot study, it is expected that reactions will vary not only by genre, but
also by text.

3. 2. 4 Method
3. 2. 4. 1 Participants

153 adult learners of French were recruited from beginning (n = 26), intermediate
(n = 91) and advanced (n = 36) undergraduate French classes at Michigan State
University in Spring 2007. The sample was comprised of 111 females and 42 males
ranging from 18-28 years old (M = 20.0). All participants were native English speakers.
Since many of the same materials had been used in the pilot study and its participants had
been informed of the study’s purpose upon its completion, students who had participated
in the pilot were not permitted to participate in this experiment. Beginning and
intermediate-level students and the majority of the advanced-level students received extra
credit equivalent to approximately 1% of their ﬁnal grade for participating in the study.
Ten advanced-level students who were enrolled in courses in which extra credit was not
offered were paid $15 for their participation.

A summary of participants’ background characteristics as obtained from the
demographic questionnaire (described in more detail in section 3.2.4.3.1.5) is presented
in Table 11. On average, students had been studying French for just over 5 years, with the
number of years of French study ranging from 3.4 for beginning to 7.1 for advanced. Less
than 20% of beginning and intermediate students had studied abroad, compared with over
half of advanced students. Over three-quarters of the sample reported that they were

taking French to meet degree requirements; this included over half of beginners, about

87

three-quarters of intermediate students, and over 90% of advanced students. Of those

taking French to meet degree requirements, about one-third of both beginners and

intermediate students reported that they would not be taking French if it were not

required. In contrast, no advanced students, of which nearly 60% were French majors or

minors, reported that they would not take French is if were not required for their degree.

Table 11

Participant background characteristics

 

Years of French study

Study Abroad (%)

Meeting Degree Req. (%)
Would not take
French if not

required (%)

French major/minor
(%)

 

 

 

Total Beginning Intermediate Advanced
(n =153) (n = 26) (n = 91) (n = 36)
M SD M SD M SD M SD
5.4 2.5 3.4 2.2 5.3 1.9 7.1 2.8

28.1 15.8 19.8 58.3

75.2 53.8 74.7 91.7

23.6 30.8 30.8 0.0
17.6 0.0 6.6 58.3

 

Beginning-level participants were enrolled in a second-semester French class

(French 102). F irst-semester students were not included in the sample because the ﬁrst-

semester French course was not offered in the semester during which data was collected.

Intermediate students were taking a third or fourth semester class (French 201 or 202),

and advanced students were enrolled in classes at the ﬁlth semester level and above (300

and 400-level classes). The focus of these advanced-level classes was varied, including

88

grammar and composition, conversation and civilization, linguistics and literature. 42%
of advanced students were taking at least one literature class at the time of the study. A
majority of advanced students (64%) was enrolled in more than one French class.

Students at each level of instruction had varied opportunities for reading, and for
reading each genre of text, within the curriculum. At the beginning level, students read
one expository cultural note in French for each of the six chapters covered in their
textbook, plus a longer reading selection. Five of six of these reading selections were
expository texts or informational realia such as television listings or recipes; only one of
these selections was a narrative text (a Jacques Prévert poem). Students at the
intermediate level read two to three expository cultural notes for each of the ﬁve chapters
covered per semester. At the time this study was conducted, ﬁrst-semester intermediate
students also read a total of ﬁve short narrative stories over the course of the semester.
Advanced students had the greatest opportunities to read in French. Although over half of
advanced students were enrolled in classes that did not focus speciﬁcally on reading
literature, their textbooks were entirely in French and they read a variety of supplemental
materials in French, including informational texts, songs, poems and stories.
3. 2. 4. 2 Materials

Three of the four passages used in the pilot study were retained in the present
study, including Narrative Text 1, Les trois motocyclistes (The Three Motorcyclists),
Narrative Text 2, L ’ceufde Pa‘ques (The Easter Egg) and Expository Text 1, Le caﬁé dans
la vie des étudiants (The Café in Student Life). The second pilot expository text, La vague
bouddhiste (The Buddhist Wave) was excluded because it contained a narrative-like

sequence that appeared to have facilitated its comprehension in the pilot study. Sport:

89

l’homme sans Iimites (Sports: Man without Limits), an excerpt from an article about the
use of technology in competitive sports ﬁ'om the magazine L’Express (also excerpted
from an intermediate-level textbook), was chosen as a replacement. These passages and
their English translations are reproduced in Appendix A. It was expected that students
would be somewhat familiar with the text topic. This text was also similar in length and
in lexical and syntactic complexity to the other passages. A comparison of these features
is presented in Table 12. One notable difference between the replacement expository text
and the other reading passages was that it contained more semantic propositions.
Nevertheless, this text was chosen despite this difference because of the dozens of
passages reviewed for potential use in the study, this text was the most similar to the
others in terms of cultural and topic familiarity, lexical and grammatical features, length,

and source (intermediate-level textbooks).

90

Table 12

 

 

 

Text characteristics
Narrative Expository
N1 N2 M E1 E2 M

Number of words 287 261 274 269 272 271
1K words a .77 .84 .81 .81 .74 .78

Function .51 .54 .53 .50 .43 .47

Content .24 .28 .26 .31 .31 .31
2K words b .06 .02 .04 .10 .09 .10
3K c .01 .01 .01 .02 .02 .02
Words not on list .15 .14 .15 .07 .15 .11
Words per sentence 9.6 13.2 11.4 12.6 10.5 11.6
Object pronouns 3 2.5 0 1 .5
Reﬂexive verbs 8 6 2 2 2
Relative pronouns 6 5 5.5 8 0 4
Semantic propositions 94 82 88 84 121 102.5

 

Note. N1:The Three Motorcyclists, N2: The Easter Egg, E1: The Cafe' in Student Life, E2:

Sports: Man without Limits. 8 Words appearing on the Lexical Frequency Proﬁle’s
(Goodfellow, Jones & Lamy, 2002) list of the one thousand most frequently occurring

words in French; analysis performed using Web VP en ﬁancais (Cobb, 2006). b Words
1001-2000. ° Words 2001-3000. d Listed in Appendix B.

91

3. 2. 4.3 Measures

3. 2. 4. 3. I Predictor variables.

3.2.4.311 Text genre.

As in the pilot, the present study examined the narrative and expository genres
using two reading passages per genre (narrative: The Three Motorcyclists, The Easter
Egg; expository: The Café in Student Life, Sports: Man Without Limits).

3.2.4.312 L1 reading ability.

Participants’ ACT reading and/or SAT verbal scores were obtained as measures
of native language reading ability. The ACT reading test requires students to read four
passages and answer related comprehension questions. The SAT verbal test, which was
revised and renamed “critical reading” in Spring 2005, assesses reading comprehension at
both the sentence and paragraph levels. Prior to Spring 2005, the test also included items
requiring students to complete analogies.

3.2.4.313 L2 proﬁciency.

The instrument used to assess L2 proﬁciency in the pilot study was a 39-item
version of MSU’s French placement test. The test included sections measuring grammar
and vocabulary knowledge and listening and reading skills. Because ceiling effects were
observed in three out of four sections of the test when taken by intermediate level
students, it was deemed necessary to add some more difficult questions before
administering it to a sample that included advanced-level students. The test was therefore
supplemented with questions from the 2003 College Board French Advanced Placement
(AP) test, an exam intended for advanced learners. An additional listening passage and

ﬁve multiple-choice comprehension questions from the AP exam were added to the test.

92

Five grammar questions were added asking students to ﬁll in blanks to complete
contextualized statements with a demonstrative pronoun, relative pronoun or preposition.
In addition, the reading section on the pilot study test was replaced with a reading
passage from the AP exam and seven multiple-choice comprehension questions. After
these revisions, the test included a total of 50 items: 11 listening, 7 reading, 16 grammar
and 15 vocabulary.

A separate instrument to supplement the vocabulary section of the test was also
administered. X_Lex: the Swansea Vocabulary Levels Test v2.05 is a 120-item lexical
decision test assessing the percentage of French words participants know in ﬁve different
frequency bands ranging from 0-5000 words. The test was administered in pencil and
paper format. Students were presented with a list including both French words and
pseudowords and asked to indicate whether or not they knew what each word meant. The
pseudowords, which respect French phonological and orthographic rules, are included in
this test to control for guessing. The number of correct identiﬁcations of French words
and rejections of pseudowords is adjusted by the number of incorrect acceptances of
pseudowords to infer the number of words that the test taker actually knows (Mochida &
Harrington, 2006).

3.2.4.314 Metacognitive knowledge.

A 55-item questionnaire was adapted from the reading and text metacognition
questionnaires used in Project NELSON, a longitudinal study of reading and writing
among Dutch Ll-English L2 schoolchildren. (Gelderen et al., 2003, 2004; Schoonen,
Gelderen, Glopper, Hulstijn, Snellings, Simis, & Stevenson, 2002). The instrument

consists of statements about text characteristics and reading strategies; participants read

93

each statement and indicated whether or not they agreed with it. Some statements apply
to reading or texts in general (e.g., “To be able to understand a text properly, you
sometimes need to know things that are not said in the text”) while others relate
speciﬁcally to reading or texts in French (e.g., “If you don’t understand a word in a
French text, it is useful to try and guess its meaning from the surrounding words and
sentences”). The questionnaire is reproduced in Appendix G.

3.2.4.3.1.5 L2 exposure.

The pilot study measured only one dimension of students’ exposure to French:
contact with reading material. Given the small amount of time that students reported
spending reading in French and the nmnerous other contexts in which students may come
in contact with the L2, the pilot study instrument was replaced with a more
comprehensive questionnaire to measure the ﬁ'equency and amount of students’ exposure
to French from a greater variety of sources. The instrument adopted for the present study
was the pre-test section of Freed et al.’s Language Contact Proﬁle (LCP) (2004). The
LCP was originally developed for use in research involving students enrolled in L2
immersion or study abroad programs. It consists of a pretest assessing students’ language
experiences prior to embarking on the program and a posttest asking about the amount
and type of L2 contact they had during the program. This post-test was judged
inapplicable to the present study where the majority of participants had not participated in
study abroad programs.

The questionnaire used in the present study (Appendix H) included two sections.
The ﬁrst section asked students for demographic data (e.g., age, sex) and information on

previous and intended language study as reported in section 3.2.4.1. The second part of

94

the questionnaire asked students to indicate how often (daily, weekly, monthly or never)
and how much time they spent doing activities where they were exposed to French. These
included communicating with French speakers, watching French ﬁlms or TV, listening to
French music, and browsing French websites in addition to reading various types of
materials in French.

3.2.4.3. 1.6 Motivation and anxiety.

Results obtained in the pilot study prompted the replacement of the motivation
questionnaire in the present study. The pilot questionnaire seemed to have failed to tap
the full spectrum of students’ reasons for studying French. It was also suspected that the
lack of relationship between motivation and L2 reading comprehension in the pilot study
could have resulted from the fact that the questionnaire measured motivation only for
language study in general instead of motivation speciﬁc to L2 reading. Moreover, upon
further consideration of research on motivation and affect in L2 reading (discussed in
section 2.1.3.2), I believed that it would be important to add questions assessing other
affective reactions that may be related to motivation for L2 reading and study. Both
Kondo-Brown (2006) and Mori (2002), for example, found that students’ perception of
themselves as readers was a component of L2 reading motivation. Mori (2002) also
identiﬁed readers’ beliefs about the importance of reading as key to reading motivation.
In a follow-up study, Mori (2004) concluded that motivational intensity as measured by
students’ self-reported study habits might also have relationships with reading motivation
and outcomes. Yamashita’s (2004) research pointed to comfort and anxiety as additional

potential motivational factors in L2 reading.

95

In light of these considerations, a new questionnaire was constructed consisting of
separate sections on motivation for and anxiety related to reading in French (38 items)
and studying French (40 items). This instrument is presented in Appendix 1. Items in both
sections were adapted from Mori (2004), Kondo-Brown (2006) and Yamashita (2004).
The questionnaire included items assessing intrinsic motivation, extrinsic motivation and
amotivation as well as students’ perceptions of their L2 reading and general L2 ability,
their perceptions of the importance of L2 reading and L2 study, their L2 motivational
intensity as measured by study habits, and their levels of comfort and anxiety in relation
to L2 study and L2 reading. The scale used in the questionnaire was also changed from a
7-point to a 6-point Likert scale in order to prevent neutral responses.

3. 2. 4.3.2 Outcome variables.

3.2.4.3.2.1 Reading comprehension.

Free written recalls in English were again used to measure L2 reading
comprehension. The recall protocols were scored using propositional analyses of English
translations of the four texts as described in section 3.1.4.3.2.1; full, half and no credit
responses were possible depending on whether participants recalled all, part or none of
each proposition. Propositional analyses of the texts are found in Appendix B.

3.2.4.3.2.2 Genre reactions.

A 25-item questionnaire, found in Appendix J, was developed to measure
students’ affective reactions to reading narrative and expository texts in French. Because
students may not have been familiar with the meaning of the terms “narrative” and
“expository”, the items used the wording “story” to refer to narratives and “informational

texts” to refer to the expository genre. Questionnaire instructions further deﬁned stories

96

as “(excerpts from) short stories, novels, plays, etc.” and informational texts as
“(excerpts from) news articles, cultural readings, etc.” The questionnaire included items
to assess attitudes such as enjoyment of and interest in the genre (e.g., “I enjoy reading
stories/informational texts”), beliefs about each genre’s utility (e. g., “I think it is useful to
read stories/informational texts”), their perceptions of the genre’s comprehensibility (e.g.,
“I think I understand a lot when I read stories/informational texts in French”), anxiety
when reading each genre, and overall genre preference when reading in the L2.
3.2.4.3.2.3 Text reactions.

In the pilot study, a post-reading questionnaire was administered to assess
students’ reactions to reading the four mst passages using both quantitative and
qualitative measures. Participants rated the difficulty and their enjoyment of each of the
four reading passages on a 7-point Likert scale, then were asked to write a comment for
each text explaining their rating. An analysis of these comments informed the
development of a new questionnaire for the present study assessing reactions to each text.
This questionnaire asked students to rate 18 statements per text on a 6-point Likert scale
(a change from the 7-point scale in an effort intended to prevent the overwhelmingly
neutral responses provided in the pilot study) in order to elicit more detailed information
on why students enjoyed or did not enjoy each text or had positive or negative
experiences when reading it. These questions were intended to tap the main constructs
described in student comments in the pilot study as well as other affective dimensions
measured in the present study. These factors included enjoyment and interest in the text
(e.g., “I enjoyed reading this text”), utility value of the text (e.g., “It was useful to read

this text”), text comprehensibility (e.g., “This text was difficult”, “I think I understood

97

nearly all of the text”) and anxiety. Students were also asked to rate their familiarity with
the text’s topic. Finally, participants were provided space to write any comments about
their experience reading each text. This questionnaire is found in Appendix K.
3. 2. 5 Procedure

A similar procedure as described in section 3.1.5 was used in the present study. In
the ﬁrst of two experimental sessions, participants completed the predictor variable
measures (L2 proﬁciency tests, metacognitive knowledge questionnaire, L2 contact
questionnaire, motivation questionnaire) and the genre reactions questionnaire. In a
second session one week later, all participants read the same two narrative and two
expository passages, which were counterbalanced, and completed free written recalls in
English without looking back at the texts. Students were instructed to write everything
they remembered in as much detail as they remembered about each passage. There was
no time limit on the reading or recall tasks. As in the pilot study, students were asked to
complete an intervening math task in French after reading each text and before beginning
the recall protocol. Participants then completed the Text Reactions questionnaire after
performing all four reading and recall tasks. After all participants had completed the
study they were provided with a description of the study’s purpose by electronic mail.

3. 2. 6 Scoring

3. 2. 6. 1 Predictor variables

3.2. 6.1.] L1 reading ability.

All participants consented to release their ACT reading and/or SAT verbal

scores as a measure of L1 reading ability. Scores were obtained from student records

housed at Michigan State University’s Office of the Registrar. Because ACT scores were

98

available for the majority of participants (139 of 153), and only SAT scores were
available for very few (n = 8), these SAT scores were converted to an equivalent of the
ACT score for data analysis purposes. A multiplier was calculated based on the scores of
the 45 participants who had taken both tests; each participant’s ACT score was divided
by his/her SAT score, a mean was calculated, and this number (.047) was then multiplied
by the SAT scores for these eight participants.l Neither ACT nor SAT scores were
available for six of the participants. These missing scores were replaced with the mean
ACT score for the sample (27.64).

3. 2. 6. 1.2 L2 proﬁciency.

Questions on the 50-item French proﬁciency test were equally weighted at one
point each. A percentage correct for the overall test and for each of its four component
sections (Listening, Grammar, Vocabulary and Reading) was calculated by dividing the
number of correct responses by the total number of questions.

Participants also completed a paper and pencil version of X__Lex: the Swansea
Vocabulary Levels Test v2.05 to assess their French vocabulary size; students indicated
whether they knew the meaning of each of 120 French words and pseudowords by
selecting a “yes’ or “no” box. Each student’s responses were then entered into the
computerized version of the test, which used the number of correctly identiﬁed words and

correctly rejected pseudowords to estimate how many of the 5000 most frequently used

 

lAlthough the College Board, administrator of both tests, publishes a concordance of ACT and SAT scores,
the most recent comparisons available at the time the data was analyzed were based on scores that were
between 12-14 years old. The College Board ﬁrrther advises that such comparisons “[are] dependent upon
the sample used to establish the relationship between two sets of scores. Other available SAT-ACT tables
use different samples of colleges and students than this table, resulting in slightly different equivalent
scores. For this reason, the best concordance table is one that is established for and used by a speciﬁc
institution” (College Board). Given that Michigan State University does not maintain its own concordance
table (Mike Cook, personal communication, November 6, 2007), calculating a multiplier based on the
present study’s sample was thus judged the best method to determine equivalent scores.

99

words in French that each student knew. This score was then converted to a percentage
by dividing the raw score by 5000.

3. 2. 6. 1.3 Metacognitive knowledge.

The metacognitive knowledge questionnaire was comprised of 25 items assessing
knowledge about texts and 30 items assessing knowledge about reading. Each item
consisted of a statement with which participants agreed or disagreed. Correct responses
were awarded one point and incorrect responses received zero points. A percentage score
was then calculated for all 55 items as well as for each of the two component sections.

3. 2. 6. 1.4 L2 exposure.

Participants completed the pre-test section of the Language Contact Proﬁle, an
instrument assessing their amount of exposure to French from various sources. Students
indicated how often they engage in a range L2 activities and how much time they spend
doing each activity; they were permitted to provide these time increments in hours per
day, week or month. Because the majority of participants provided responses based on
amount of contact per month, for data analysis purposes all responses were converted to
hours per month by multiplying hours per day by 30 or hours per week by 4. This number
of hours of French exposure per month was further broken down into subscores for
different types of contact assessed by the questionnaire, including Speaking, Reading,
Media (e.g., television, ﬁlms, and Internet), and Other contact. The amount of students’
exposure to French through study abroad experiences was represented by scores of 0 (no
study abroad experience), 1 (one semester or less), 2 (two semesters) or 3 (more than two

semesters).

100

3.2. 6.1.5 Motivation and anxiety.

Two questionnaires measured participants’ motivation and anxiety to French
study and to reading in French. Each questionnaire contained items assessing intrinsic
and extrinsic motivation and amotivation as well as affective reactions to each activity,
including perceptions of its importance, self-perceived ability, and anxiety. Questions
about participants’ L2 study habits were also included in the French study portion of the
questionnaire as a measure of motivational intensity. Scores were compiled for each of
these constructs by summing the 1-6 point Likert-scale responses for the relevant
questionnaire items. Scores for negatively worded items were reversed for the purposes
of this calculation. While preparing data for scoring, a review of the motivation and
affective reactions questionnaire prompted the elimination of two items. First, item 22 on
the L2 Study portion of the questionnaire was removed, as it was an inadvertent duplicate
of item 6. Item 8 (“I don’t like learning French, but I think learning French is important”)
was also eliminated. Though originally intended as part of the Amotivation subscale, it
was considered ambiguous because it also appeared to measure students’ perception of
the importance of learning French, a separate construct under investigation in this study.

3. 2. 6. I . 6. Learner level.

Students reported whether they were enrolled in beginning (100-level),
intermediate (200-leve1) or advanced-level (300 or 400-level) French classes in the
semester during which data was gathered. Beginning-level students were enrolled in
second-semester French, intermediate students were taking third or fourth-semester

classes, and advanced students were enrolled in ﬁﬁh-semester or higher courses.

101

3. 2. 6.2 Outcome variables

3. 2. 6.2.1 Reading comprehension.

As in the pilot study, reading comprehension scores were tabulated using a
scoring protocol based on propositional analyses of the four reading passages. This
process is described in detail in section 3.1.4.321. A percentage score was calculated for
each text by dividing the ntunber of propositions each participant recalled by the total
number of propositions contained in the text. Collapsed mean comprehension scores were
then calculated for the two narrative and two expository texts.

3. 2. 6. 2.2 Genre reactions

A 25-item questionnaire assessed participants’ attitudes and affective reactions to
reading narrative and expository texts in French, including their enjoyment of and interest
in the genre, their perceptions of each genre’s utility, their perceptions of the
comprehensibility of each genre, their levels of anxiety when reading each genre, and
their overall genre preference when reading in the L2. The mean 6-point Likert scale
responses for each of these affective factors were compiled, aﬁer reversing scores for any
negatively worded items.

3. 2. 6.2.3 Text reactions

For each of the four reading passages, participants answered 18 Likert-scale
questions inquiring about their reactions to reading the text. These included their interest
in and enjoyment of the passage, their perception of its utility, their perception of the
text’s comprehensibility, their anxiety level when reading the text, and their familiarity

with the text topic. Mean scores were tabulated for each of these constructs. Participants

102

were also provided space to write any additional comments about their experience

reading each text; these comments were all transcribed.

103

 

CHAPTER 4: RESULTS

This chapter will report the results of the main experimental study. After
discussing data screening procedures, I will report descriptive statistics and correlations
for the data. I will then describe the principal statistical analysis technique employed in
the study, structural equation modeling, and present the results of the statistical analyses

in response to the research questions.

4.1 Data screening

The two principal statistical methods used in the present study, structural equation
modeling and analysis of variance, are parametric tests based on assumptions of normally
distributed data. Data was therefore screened for normality by examining the histogram
and skewness and kurtosis values for each variablez. A frequency analysis was also used
to screen data for outliers, which were identiﬁed as values greater than 4 standard
deviation units from the mean. Overall, data was normally distributed (i.e., skewness and
kurtosis coefﬁcients less than an absolute value of 2) with the exception of French
exposure, which was positively skewed. The variables of Speaking, Reading, Media and
Other French contact contained some outliers, including several implausibly large values.
When the responses of the four participants who reported these large amounts of French
exposure were examined individually, it was found that all had recently participated in
study abroad experiences. In two cases, the participants made notes on the questionnaire

that their responses were based on the amount of French contact they had had during

 

2 Unless otherwise speciﬁed, statistical analyses were performed using SPSS statistical software, version
1 5.0.

104

these experiences. Given that these responses appeared not to reﬂect participants’ L2
exposure outside of this one-time study abroad experience and that their study abroad
experiences would be accounted for as a separate L2 exposure variable, these outliers
were replaced. Because variability in French exposure was expected by learner level, the
replacement values were the mean scores for these L2 exposure variables as calculated
for participants at same level of French study.

Even after these outlying values were replaced, the data for the French exposure
variables was still non-normally distributed with the majority of values clustered at or
near zero. The coefﬁcients of skewness and kurtosis for contact involving reading,
speaking, and media ranged from 2.63 to 5.83 and 6.73 to 42.12, respectively. A log 10
transformation was performed to correct this distribution.

Data was also screened for missing values, which were found for several items on
the metacognition and motivation questionnaires. Since variability in metacognitive
knowledge and motivation was also expected by learner level (beginning, intermediate,
advanced), a mean score for each level was calculated for each item that contained
missing values, and these values were replaced with the mean score corresponding to the

participant’s level of French study.

4.2 Descriptive statistics and correlations
4.2.1 Predictor variables
4. 2. 1. 1 L2 proﬁciency
Descriptive statistics for the measures of French proﬁciency are presented in

Table 13. Estimates of internal consistency (or = .81) indicate that overall the 50-itcm test

105

had acceptable reliability, though reliability estimates of items within its four component
sections (Listening, Grammar, Vocabulary, Reading) were lower (as = .43 - .62).
Moderate correlations (r = .42 - .56) were observed between these variables, suggesting
that they are related but independent constructs. The mean overall score on this
proﬁciency test was 56%, with the best performance observed in the Vocabulary and
Listening sections (68% and 60%, respectively). Reading (54%) and Grammar scores
(43%) were somewhat weaker.

Table 13

Descriptive statistics and correlations for French proﬁciency and vocabulary size

 

 

M3 SD 1 2 3 4 5
1. Listening .60 .16 (.43)
2. Grammar .43 .17 .56 (.62)
3. Vocabulary .68 .17 .45 .51 (.54)
4. Reading .54 .24 .50 .53 .43 (.52)

5. Vocabulary Size .47 .13 .43 .58 .50 .45 (-)

 

Note. Values in parentheses are estimates of internal consistency (Cronbach’s 0t). Dashes
(-) indicate that internal consistency estimates could not be calculated. Mean values
reﬂect proportion correct. All correlations are statistically signiﬁcant at the .01 level.

a n = 153.

Vocabulary size was measured using a separate instrument, the 120-itcm lexical
decision test. The average score on this test indicated that participants knew 47%, or
about 2350, of the 5000 most common words in French. Though internal consistency

estimates could not be calculated for items on this computerized test, previous research

examining the reliability of this lexical decision test has reported alpha levels averaging

106

.76 (Mochida & Harrington, 2006), indicating good reliability. Vocabulary size scores
correlated similarly with the other measures of French proﬁciency (r = .43 - .58).
4.2.1.2 LI reading ability

Ll reading ability score as measured by ACT Reading scores (or their equivalent,
as described in section 4.1.1.2) averaged 27.6 out of 36 possible points (SD = 4.6).
Though he does not specify an estimate speciﬁcally for the Reading section, Passow
(1995) notes that internal consistency estimates for the ACT English, Reading, and Math
composite tests range from .88 to .93, indicating very good reliability.
4.2.1.3 Metacognitive knowledge

Descriptive statistics and correlations for metacognitive knowledge are displayed
in Table 14.
Table 14

Descriptive statistics and correlations for metacognitive knowledge

 

 

M3 SD 1 2
1. Texts .60 .16 (.62)
2. Reading .43 .17 .24 (.71)

 

Note. Values in parentheses are estimates of internal consistency (Cronbach’s a). Mean
values reﬂect proportion correct. Correlation is statistically signiﬁcant at the .01 level.

3n = 153.

Internal consistency estimates suggest that the questionnaire is an acceptably
reliable measure of metacognitive knowledge (or = .65), and that its component subscales
are acceptably reliable measures of knowledge about texts (or = .62) and about reading

(or = .71 ). Participants’ mean scores for overall metacognitive knowledge were slightly

107

above the midpoint (56%), demonstrating greater knowledge of texts and text
organization (60%) than of reading strategies (43%).
4.2.1.4 L2 exposure

L2 exposure descriptive statistics and correlations are presented in Table 15.
Table 15

Descriptive statistics and correlations for French exposure (hours per month)

 

 

Ma so 1 2 3 4 5
1. Speaking 5.53 2.63 (.35)
2. Reading 1.38 3.06 .34 (.51)
3. Media 1.34 3.51 .39 .88 (.77)
4. Other 6.03 21.53 .38 .37 .27 (-)
5. Study) abroad .29 .49 .43 .34 .33 .17* (-)

time
Note. Values in parentheses are estimates of internal consistency (Cronbach’s 0t). Dashes
(-) indicate that internal consistency estimates could not be calculated. Mean values
reﬂect proportion correct. All correlations are statistically signiﬁcant at the .01 level
except as marked with an asterisk (*p < .05).

an = 153. b 0 = no study abroad experience, 1 = 1 semester or less, 2 = 2 semesters.

Internal reliability estimates of .77 indicate that the language contact
questionnaire had acceptable reliability, though these estimates were lower for the
Speaking and Reading subscale items (or = .35 - .51), likely a function of the small
number of items for each subscale. Internal consistency estimates could not be calculated
for Other or Study Abroad exposure because each was measured by only one item. Most

types of L2 exposure were moderately correlated with each other (r = .27-.43), though

108

correlations were stronger (r = .88) between Reading and Media and weaker (r = .17)
between Study Abroad and Other exposure.

On average, participants reported just over 14 hours per month, equivalent to
roughly 45 minutes a day, of outside-of-class exposure to French. The preponderance of
this contact, about 6 hours per month, was reported in an open-ended item that asked
students to specify any “other activities” they did using French. A review of individual
responses indicated that this contact ranged ﬁom participating in extracurricular French
clubs and workshops to teaching or tutoring others in French, using French while gaming,
doing French homework, and “self—talk” or rehearsal of French expressions. Though
Other contact constituted the bulk of hours of exposure to French, it should be noted that
only about one-quarter of participants reported engaging in these activities. Speaking
contact also represented a large part of reported French exposure: about 5.5 hours per
month, or roughly 40% of total L2 contact. Students reported nearly equal amounts of
contact with French reading materials and media, though these amounts were quite small:
less than 1.5 hours per month, or approximately 3 minutes per day. Frequency analyses
revealed that more students had contact with French through speaking (85%) than either
reading (52.9%) or media (39.2%).

The mean study abroad time for the sample was slightly under one-third of a
semester. Just under 30% of participants reported exposure to French through study

abroad experiences. No participants reported studying abroad for more than two

semesters.

109

4.2.1.5 Motivation and anxiety

Inter-item correlations for the motivation and anxiety for French study and
reading in French are presented in Appendices L and M3, respectively. A review of these
correlational analyses resulted in the removal of some questionnaire items. First, item 15
(“I am learning French because I would like to get good grades”) was eliminated ﬁ'om the
L2 Study/Extrinsic subscale because it was negatively correlated with the other items on
this scale, and only weakly correlated with most other questionnaire items. Because this
item was worded similarly to an item on the L2 Reading/Amotivation subscale (“I am
learning to read in French merely because I would like to get good grades”), further
analyses were performed to determine whether the item was a better ﬁt on the L2
Study/Amotivation scale. However, the item was only weakly correlated (r = .13 - .21)
with other Amotivation items and its inclusion would have reduced internal consistency
of that subscale from .74 to .74, so it was instead eliminated. Next, item 30 (“Even when
reading materials are dull or uninteresting, I always ﬁnish the assignments”) was
removed. Though intended as part of the L2 Reading/Intrinsic subscale, it was not
signiﬁcantly correlated with any of the other subscale items and was very weakly
correlated with most other questionnaire items. Finally, ﬁve L2 Study Habits items
(10, 21, 25, 28 and 30) that were only weakly correlated with the other subscale items
(r = .00 -. 35) as well as with other questionnaire items were eliminated. These items
described students’ in-class behavior (e.g., “I actively participate in French classes”) and
whether they scheduled a regular time for French study outside of class (e.g., “I work on
my French assignments according to a prcplanned schedule”). The six items remaining

on the L2 Study Habits subscale, in contrast, captured students’ motivational intensity as

 

3 These tables are presented as an appendix instead of a table because of their length.

110

measured by their persistence in completing homework assignments (e.g., “Even when
homework in French is tiresome, I will work hard on it”) and the amount of time they
spent studying French (e.g., “I think I study French a lot”). The elimination of the items
also improved internal consistency estimates for the Study Habits subscale from .78 to
.80. Within the remaining six Study Habits items, a pattern of stronger intercorrelations
was observed within the items related to persistence in completing homework

(average r = .45) and within those assessing the amount of time devoted to L2 study
(average r = .56) than between the two sets of items, indicating that these two sets of
items likely comprised two distinct factors.

These inter-item correlational analyses also prompted the reclassiﬁcation of the
Comfort and Importance items on the questionnaire to different subscales. Only one item
assessed students’ comfort with L2 study and reading (“Even if I cannot understand
F rench/what I read in French completely, it doesn’t bother me”). Given that it was the
only item measuring this construct, it seemed to be related to an absence of anxiety, it
correlated well and the most strongly with the other Anxiety items (r = .40—.50), and it ﬁt
reliably into these subscales (a = .77), it was reclassiﬁed as a measure of Anxiety after
reversing its score. It was also noted that the items measuring students’ perceptions of the
importance of L2 study and reading were more highly correlated with the items on the
Extrinsic subscales (r = .78 -.89) than they were with each other (r = .61 - .76) or with
items in other subscales. Associating the Importance items (e.g., “Studying/Reading in
French is important because it will be conducive to my general education”) with Extrinsic
motivation items was judged to be a theoretically sound change, since these questions

seemed to tap a construct ﬁtting with accepted deﬁnitions of extrinsic motivation (i.e.,

lll

external reasons unrelated to intrinsic interest). Furthermore, reclassifying the Importance
to the Extrinsic subscales improved Extrinsic subscale internal consistency estimates for
the L2 study portion of the questionnaire from .89 to .92, and for the L2 reading items
from .86 to .89.

Table 16 presents descriptive statistics, correlations and reliability estimates for
the motivation and affect subscales after the changes described above were made.
Internal consistency estimates were quite high (a = .77-.94), indicative of good reliability
of the questionnaire subscales. Some similar correlation patterns were observed within
both the L2 study and L2 reading sections of the questionnaire. First, there were
consistently strong correlations between the Intrinsic, Extrinsic and Amotivation
subscales within each section, with absolute correlation coefﬁcients ranging from
.59 - .79. Relationships between Intrinsic and Extrinsic motivation were strongest
(r = .78 - .79). Amotivation was negatively related to both, and somewhat more weakly
related to Extrinsic motivation (r = -.60 and -.62) than Intrinsic (r = -.72). Self-Perceived
Ability for L2 study was also strongly correlated with Intrinsic, Extrinsic (positively) and
Amotivation (negatively), with absolute correlation coefficients of .50 - .65. While self-
pereeived L2 reading ability also had a strong positive relationship with Intrinsic
motivation to read in French (r = .63), its correlations with Extrinsic and Amotivation to
read were moderate (r = .39 and -.39, respectively). Anxiety, on the other hand, was only
weakly related to Intrinsic, Extrinsic and Amotivation. Correlations between Anxiety and
Intrinsic motivation for L2 study and L2 reading were -.29 and -.24, respectively.
Anxiety had a positive relationship with Amotivation, though the relationship was weaker

for L2 reading (r = .15) than for L2 study (r = .26). Correlations of Anxiety with

112

Extrinsic motivation for both L2 study and L2 reading were negative, though not
statistically signiﬁcant. Anxiety’s strongest relationship for both L2 study and L2 reading
was an inverse one with Self-Perceived Ability, with correlations of -.45 and -.3 7,
respectively.

L2 Study Habits were positively and strongly correlated with Intrinsic and
Extrinsic motivation for L2 study ( r = ..48 and .51, respectively) and strongly negatively
related to Amotivation (r = -.42). L2 Study Habits also had a weaker but statistically
signiﬁcant relationship with Self-Perceived Ability for L2 study (r = .27). L2 Anxiety
was not signiﬁcantly correlated with Study Habits. Correlation patterns between L2
Study Habits and motivation and affective measures related to L2 reading were quite
similar. In fact, slightly stronger relationships were observed between Study Habits and
Intrinsic and Extrinsic motivation to read (r = .52) than Intrinsic and Extrinsic motivation
to study French. Study Habits’ relationship with self-perceived L2 reading ability was
also somewhat stronger (r = .37) than with self-perceived L2 ability.

Correlations between the subscales in the L2 study and L2 reading sections of the
questionnaire were also uniformly strong, with coefficients ranging from .76 between the
Amotivation subscales to .84 between the subscales for Extrinsic motivation for L2 study
and L2 reading.

Turning to the descriptive statistics, participants reported relatively high levels of
extrinsic and intrinsic motivation for studying French (M = 4.91 and 4.71, respectively,
out of a maximum of 6 points), and low amotivation (M = 2.15). Responses also
indicated that students felt studying French is important (M = 4.87). Participants’

motivational intensity for L2 study as measured by study habits was moderate

113

(M = 3.92), as was their assessment of their L2 ability (M = 3.90). Students also reported
only moderate levels of anxiety related to French study (M = 3.34).

Similar patterns were observed in scores for motivation and affective reactions to
L2 reading. Extrinsic and intrinsic motivation scores were relatively high
(M = 4.00 and 3.80, respectivelY), though not as high as corresponding scores for
motivation to study French. Amotivation scores for L2 reading were low (M = 2.24), but
slightly higher than amotivation for L2 study. Scores also indicated that participants
believed reading in French was important (M = 4.98); these scores were even slightly
higher than scores for perceptions of the importance of studying French. Mean scores for
self-perceived L2 reading ability were also moderate (M = 3.06), but were nearly one
scale tmit lower than scores for self-perceived general L2 ability. While L2 reading-
related anxiety was moderate (M = 3.63), it was somewhat higher than anxiety related to

L2 study.

114

 

 

 

83 8. R. a..- a. we. 2: gm 352 came
:5 3.. ea. 2.- 8.- e? a? badge
$3 on- am. 8. x: 8m been escapee-:8 .e
95 8.. Na.- 2: 3m deceased... .m
3.3 we a: one sameness nausea .m
93 w _ ._ E. Susana swans .c
@333
2 e m a m N a an as:

 

83935. Novena: BS» eczexpteﬁ aneéScSeL-aee use mutate:- matghamQ

3 038-

115

.3888 ”03339.8 nee-mob o u ﬂame-om Engage ”an we teamed-So we: meow :8va m n 01:32. 825:3 a .mm_ H 2 a
.56— 8. 0% com N. as new _o>o_ mo.
05 8m 3. “a “Enos-“Emu bﬁoumuﬁm v.8 mucus—280 .«e Peugeoch museum—So 3885 me 83:58 can meanness-am E mes—Sr $82

 

:5 an- 2. 8.- 3..- 3. ma. $.- :. 8. S.- :._ Sm 53.:
$3 an- an. 8. R. a..- 5.. .3..- :.. mm. 8; 85 base Banana-meme.
63 an- 9.- 2.- am. .9..- ea. 8.. Q.. a. an newness-a. .e

83 an. mm. :.- me. an- em. m. m 2 m2. ceases: annex-m .m

Geog Nm. oNr hm. am... on. an. no; owﬁ douguoe ORE-55.5

 

382 .S

S e m a e m a m N c a. an:

 

 

Bags—co ©— 0353—.

116

4. 2. 2 Outcome variables
4. 2. 2. 1 Reading comprehension
Descriptive statistics for text comprehension are displayed in Table 17.
Table 17

Descriptive statistics for reading comprehension arropositions recalled)

 

 

 

 

 

 

M a SD min max

Narrative

N1 .22 .12 .02 .54

N2 .22 .14 .02 .67

Both narrative .22 .12 .02 .61
Expository

E1 .12 .07 .01 .40

E2 .09 .06 .00 .35

Both expository .10 .06 .01 .38
All texts .16 .08 .02 .43

 

Note. N1 :The Three Motorcyclists, N2: The Easter Egg, E1: The Cafe' in Student Life,
E2: Sports: Man Without Limits. Means reﬂect proportion of propositions recalled.

an=153.

On the whole, participants recalled 16% of text propositions. Collapsed means
show more was recalled ﬁ'orn the narrative texts (22%) than the expository passages
(10%), and more was also recalled ﬁ'om the two narratives than ﬁom either expository
text. The same percentages of propositions were recalled from both narrative passages
(22%). Differences in recalls between the two expository texts, however, existed, with

more recalled from Text 1 (12%) than Text 2 (9%).

117

4. 2. 2.2 Genre reactions

Descriptive statistics, correlations and internal consistency estimates for attitudes
about the narrative and expository text genres are displayed in Table 18. Reliability
estimates ranging from .76 - .86 indicated that scales were reliable measures of genre
interest/enjoyment, utility, and comprehensibility. Reliability estimates for the anxiety
items for both genres were lower (.58), presumably because this subscale contained only
two items. All subscales were signiﬁcantly correlated, with absolute correlation
eoeﬂicients ranging from .39 - .75.

Students reported higher interest and enjoyment for narratives in general (mean
score of 4.34 on a 6-point scale) than for expository texts (M = 3.83). Ratings of the two
genres’ utility were nearly identical (4.50 for narratives, 4.51 for expository texts).
Students believed that narratives were somewhat more comprehensible (M = 3.56) than
expository texts (3.29), and also reported experiencing less anxiety when they read
narrative texts (M = 3.10) than expository passages (3.24).

One additional item on the questionnaire asked students about their L2 genre
preference (“If I had a choice, I would rather read stories than informational texts in
French”). Mean scores (4.29) indicated that overall, participants agreed with this

statement.

118

Table 18

Descriptive statistics, correlations and reliability for genre reactions

 

Math SD 1 2 3 4

 

Narrative

 

1. Enjoyment/interest 4.34 .98 (.86)

 

2. Utility 4.50 1.07 .75 (.89)
3. Comprehensibility/ 3 .56 1.01 .55 .51 (.76)
Ability
4. Anxiety 3.10 1.13 -.45 -.33 -.73 (.58)
Expository

 

1. Enjoyment/interest 3 .83 1.1 l (.85)

2. Utility 4.51 1.13 .69 (.86)
3. Comprehensibility/ 3.29 1.05 .58 .53 (.78)
Ability
4. Anxiety 3.24 1.15 -.43 -.39 -.73 (.58)

 

Note. Values in parentheses are estimates of internal consistency (Cronbach’s a). All
correlations are signiﬁcant at the .01 level.

an = 153. b Minimum possible = 1 (reason does not correspond at all); maximum
possible = 6 (reason corresponds exactly).
4. 2. 2. 3 Text reactions

Correlations and reliability estimates for reactions to the experimental texts are
presented in Table 19. Internal consistency estimates for the four text reactions
questionnaires (or = .79 - .86) and their component subscales were unifonnly high (a =
.72 - .93) with the exception of the anxiety subscales, which were somewhat lower (a =
.62-.77). Reliability estimates were not calculated for topic familiarity, as only one

question measured this construct. Reactions to all four texts exhibited a similar pattern of

119

correlations. Enjoyment and interest were fairly highly correlated with perceptions of the
utility of the text (r = .55 - .67). Comprehensibility ratings were also correlated with
enjoyment of and interest in the text, though the relationships were somewhat weaker (r =
.45 - .65). Anxiety had robust negative correlations with perceptions of comprehensibility
(r = -.68 — 82). In contrast, students’ assessment of the comprehensibility of the text was
fairly weakly related to the text’s utility value (r = .17 - .34). Finally, topic familiarity had
very weak correlations with the other text reactions (r = .07 - .33), several of which failed
to achieve statistical signiﬁcance. F arniliarity with a text’s topic also did not correlate
signiﬁcantly with comprehension of the text (r = .05 - .14).

Table 19

Descriptive statistics, correlations and reliability for text reactions

 

 

 

Ma SD 1 2 3 4

Narrative
N1

1. Enjoyment/Interest 4.08 .96 (.87)

2. Utility 3.57 1.02 .63 (.75)

3. Comprehensibility 4.07 1.07 .45 .17 (.90)

4. Anxiety 2.73 1.10 -.48 -.18 -.79 (.64)

5. Topic Familiarity 3.36 1.42 .15 .11 .22 -.21
N2

1. Enjoyment/Interest 4.23 1.16 (.92)

2. Utility 3.70 1.10 .67 (.77)

3. Comprehensibility 3.93 1.23 .65 .34 (.93)

4. Anxiety 2.72 1.22 -.63 -.30 -.82 (.77)

5. Topic Familiarity 3.15 1.51 .33 .26 .28 -.25

120

Table 19 continued

 

Both
1. Enjoyment/Interest
2. Utility
3. Comprehensibility
4. Anxiety
5. Topic Familiarity

 

Expository

 

E1
1. Enjoyment/Interest
2. Utility
3. Comprehensibility
4. Anxiety
5. Topic Familiarity
E2
1. Enjoyment/Interest
2. Utility
3. Comprehensibility
4. Anxiety
5. Topic Familiarity
Both
1. Enjoyment/Interest
2. Utility
3. Comprehensibility
4. Anxiety
5. Topic Familiarity

4.16
3.63
4.00
2.72
3.26

3.77
3.98
3.76
2.93
3.87

3.22
3.67
3.12
3.29
2.90

3.49
3.82
3.44
3.11
3.39

SD

.96
.98
1.06
1.09
1.25

1.11
.98
1.06
1.08
1.31

1.06

1.13
1.09
1.19

.96
.91
1.00
1.00
1.03

.68
.56
-.56
.31

(.90)
.55
.49
-.45
.26

(.89)
.67
.43
-.38
.36

.63
.45
-.41
.28

.25
-.23
.18

(.77)
.22
-.25
.07

(.72)
.24
-.20
.14

.23
-.22
.08

3 4
-.83
.23 -.22
(.89)
-.68 (.62)
.31 -.21
(.91)
-.73 (.62)
.32 -.22
-.72
-.30 -.21

 

Note. Values in parentheses are estimates of internal consistency (Cronbach’s a). All
correlations are signiﬁcant at the .01 level.

a

possible = 6 (reason corresponds exactly).

121

n = 153. Minimum possible = 1 (reason does not correspond at all); maximum

Enjoyment and interest ratings for both narrative texts (M = 4.08 — 4.23) were
higher than those for the expository passages (M = 3.22 — 3.77). Utility ratings were
similar for all four texts, though collapsed mean scores for the two expository texts (M =
3.82) were somewhat higher than those of the narratives (M = 3.63). Comprehensibility
ratings were lower for the expository texts, with a mean making of 3.44 compared to 4.00
for the narrative passages. However, Expository Text 2 was rated as more difficult (M =
3.12) than Expository Text 1 (M = 3.76). Anxiety scores were somewhat lower for the
narratives than the expository texts (collapsed Ms = 2.72 and 3.11, respectively), though
again there were greater differences in these ratings between the two expository texts
than the two narratives; Text 1 provoked less anxiety (M = 2.93) than Text 2 (M = 3.29).
Finally, collapsed topic familiarity ratings for the two narrative (M = 3.26) and two
expository texts (M = 3.3 9) were quite similar, though ratings were higher for Narrative
Text 1 (M = 3.36) and Expository Text 1 (M = 3.87) than for Narrative Text 2 (M = 3.15)
and Expository Text 2 (M = 2.90).

Students’ comments about the four reading passages are found in Appendix N.
Both narrative texts inspired more comments (Text 1, n = 30; Text 2, n = 27) than the
expository passages (Text 1, n = 20; Text 2, n = 15). Comments for all texts included
mentions of whether students understood the text, and often, to what they attributed the
passage’s comprehensibility. When these attributions were made, they most often
mentioned vocabulary. Genre was also mentioned in relation to comprehension. For the
narrative texts, three comments implicated genre in text comprehensibility (e.g., “It was

easier to remember because it was story-like”). Two comments for Expository Text 1

122

mentioned that the expository genre aided comprehension, while the two genre comments
for Expository Text 2 were split (“Another one I understood better than the stories”,
“There was just more to think about than for the stories”). Several comments mentioned
familiarity as a factor in comprehensibility (e.g., “Easier, familiar objects, experiences”),
and others mentioned other features speciﬁc to the text (e.g., “Very clear details”).

Other comments indicated that students liked or disliked a text, and sometimes
provided reasons for this reaction. Items mentioned in relation to these comments about
enjoyment included genre (e.g., “I liked this excerpt the best b/c it was more of an actual
story”) and topic (e.g., “The Olympics are always interesting to me”), including personal
connections to the topic (e.g., “I also had personal memories of sitting in European cafes
meeting people. . .”). Some students also reacted speciﬁcally to or asked questions about
the text content or message (e.g., “I knew cafes were important in France, but I didn’t
realize that they had their own personalities”; “. . .were the three motorcyclist (sic) part of
the 9 campers? “, “It made me think of the gatorade (sic) commercials with the athletes

on the treadmills, and they are monitoring their performance”).

4.3 Research Question 1: Individual differences and L2 reading comprehension
4. 3. 1 Structural equation modeling
Research Questions 1a and lb asked about the contributions of individual
difference variables to L2 reading comprehension and the interrelationships between
these variables. Structural equation modeling (SEM) was judged the most appropriate
statistical method to examine these relationships. SEM is a multivariate statistical

analysis approach used to interpret relationships between multiple, interrelated variables.

123

SEM methods have several advantages over other related approaches, such as multiple
regression analysis. First, SEM allows the examination of complex interrelationships
between variables; it is the only analysis method that permits all relationships between
multidimensional variables to be tested completely and simultaneously (U llman &
Bentler, 2004). SEM also allows the use of latent variables, or theoretical constructs that
cannot be observed directly, but instead must be assessed by a number of indicators,
called observed variables. Furthermore, because it estimates and removes measurement
errer from its analysis of relationships between these variables, SEM allows for more
accurate measurements of their effects on one another. For these reasons, SEM
approaches are increasingly used in multivariate L2 individual difference research (Csizer
& Demyei , 2005; Gelderen et al, 2004; Holmes & Omar, 2006; In’nami, 2006; Lee,
2005; Proctor, et al., 2005; Pulido & Hambrick, 2008; Schoonen, et al., 1998; Woodrow,
2006; Yashima, Zenuk-Neshide, & Shimizu, 2004).

Modeling with SEM involves two core techniques, conﬁrmatory factor analysis
and structural modeling. Conﬁrmatory factor analysis is used to test hypothesized
relationships between latent variables and observed variables, or in other words, the
extent to which the constructs under investigation are represented by the measures used.
The model speciﬁed during the conﬁrmatory factor analysis phase is known as a
measurement model. Then, in the structural model, hypothesized relationships between
these latent variables are speciﬁed and tested. Throughout both of these processes, model
ﬁt is evaluated and models may be respeciﬁed to improve ﬁt. As Kline notes (2005),

model respeciﬁcation and reevaluation is often necessary, and is appropriate provided

124

that changes to the model are guided by appropriate theoretical principles and the
researCher’s hypotheses.
The following statistics, obtained through analyses performed using AMOS
statistical analysis software version 7.0.0, will be used to describe model ﬁts in this
section. A summary of these model ﬁt statistics and acceptable values is presented in
Table 20.
Chi-square (X2). The X2 statistic reﬂects whether signiﬁcant differences exist
between the covariance matrix estimated by the model and the observed (sample)
covariance matrix (U llrnan & Bentler, 2004). Kline (2005) describes X2 as a “badness-of-
ﬁt” index; higher values indicate worse ﬁt of the model to the data. Although X2 values
are customarily reported in all SEM results, because they may be sensitive to sample and
correlation sizes, they should not be relied upon as a sole ﬁt statistic (Bollen & Long,

1 993; Kline, 2005).

Comparative ﬁt index (CFI) and Normed ﬁt index (NFI). CF I and NFI measure
the improvement in ﬁt of the proposed model over a baseline model that assumes zero
covariances among observed variables (Kline, 2005). Both indices are sensitive to
missDeﬁcifications in relationships between latent variables and observed variables. CFI
and NFI values greater than .90 indicate acceptable ﬁt of the proposed model (Kline,
2005). Although both indices are routinely reported in SEM results, Kline proposes that
CPI is the better index of ﬁt because NFI can be overly sensitive to sample size, resulting
in lower Values with smaller samples. As the sample size in the present study (n = 153) is

at the low end of the ISO-case minimum recommended by Kline (2005), CFI values will
re imel'ln‘eted as the best measure of model ﬁt.

125

Root mean squared error of approximation (RMSEA). The RMSEA statistic
reﬂects differences between observed and predicted covariances, and is sensitive to
misspeciﬁcations of relationships between latent variables. It is generally accepted that
RMSEA values of .06 or below indicate “good” ﬁt between the proposed model and the
observed data, though values as high as .08 may be considered acceptable (Byme, 2001 ).
Table 20

Model ﬁt statistics and suggested acceptable values

 

 

 

Model ﬁt statistic Acceptable Notes
Values
X2 p < .05 Sensitive to sample, correlation size
NFI > .90 Sensitive to sample size
CFI > .90
RMSEA < .08

 

 

Note. From Bollen & Long, 1993; Byme, 2001; Kline, 2005, Ullrnan & Bentler, 2004.

4.3.1.1 Graphic representation of SEM models
Throughout this section, SEM models will be depicted graphically using output
from AMOS 7.0.0. In these diagrams latent variables are represented by ovals, and
observed variables by rectangles. A residual error term, representing random
measurement error, is associated with each observed variable and is depicted by a circle
labeled “e”. Single-headed arrows point from each latent variable to its observed
variables and from each error term to its observed variable. Factor loadings are printed

(57“ t0 the paths ﬁ'om latent to observed variables. These loadings are standardized

126

regression coefficients; when squared, they represent the percent of variance in the
obserVed variable that is explained by the factor. Factor loadings of .40 or greater are
generally considered acceptable in conﬁrmatory factor analysis (Raubenheimer, 2004),
while loadings of .70 and above provide very strong evidence that the observed variables
are represented by the associated latent variable (Garson, 2008). Covariance or
correlational relationships between variables are represented in SEM models using
curved two-headed arrows. Coefficients for these correlational relationships are printed
next to these paths.

Structural models add single-headed arrows to this diagram, indicating paths from
predictor (exogenous) to effect (endogenous) latent variables, as well as the related path
coefﬁcients. Each endogenous variable is also associated with a disturbance, a construct

analogous to residual error, representing unspeciﬁed causes of the variable. Disturbances
are represented in the diagram by circles labeled “(1”.
4. 3.1.2 Conﬁrmatory factor analyses

In a ﬁrst step, conﬁrmatory factor analyses were performed to verify that the
0903111106 under investigation had been satisfactorily measured. Separate analyses were
carried Ollt for the independent and dependent variables.

4. 3. I . 2.1. Predictor variables.

In the analysis of the predictor variables, a measurement model including the
latent variables and their observed variables (indicators) described in Table 21 was
initially Speciﬁed. With the exception of L1 Reading Ability, which had been assessed
using only one measure (ACT Reading scores or equivalent), all factors included in this

with] mOde] were latent variables composed of two or more indicators.

127

Table 2 1

Initially SPeCiﬁed measurement model: latent and observed variables

 

 

Latent Observed
I—Z proﬁciency Listening, Reading, Grammar, Vocabulary, Vocabulary
Size
(None) L1 Reading Ability
Metacognitive knowledge Texts, Reading
L2 exposure Speaking, Reading, Media, Study Abroad, Other
Motivation L2 Study/Intrinsic, L2 Study/Extrinsic, L2
Study/Amotivation, L2 Reading/Intrinsic, L2
Reading/Extrinsic, L2 Reading Amotivation
Anxiety L2 Study Anxiety, L2 Reading Anxiety
Self-perceived ability L2 Study Ability, L2 Reading Ability
Study Habits Homework, Study Time

 

 

This model, however, resulted in an inadmissible solution, with negative variance

occurring in the error term for Anxiety and Self-Perceived Ability. Kline (2005) notes

that this type of inadmissible solution can occur in measurement models containing

factors With only two indicators when smaller samples (150 cases or fewer) are involved.

In this initial model both these latent variables were comprised of only two indicators,

and the satirple size in the present study (n = 153) is just on the edge of this 150 case

threshold. High intercorrelations between the indicators may also contribute in an

inadmissible solution, since observed variables that are closely related statistically may

128

not in fact measure two distinct constructs. In light of these considerations and the rather
strong correlations observed between the two Anxiety indicators (L2 Study Anxiety and
L2 Reading Anxiety; r = .78) as well as the two Self-Perceived Ability indicators (L2 and
L2 Reading; r = .78), the model was revised to include Anxiety and Self-Perceived
Ability as observed variables comprised of a collapsed mean score for their component
items. While models containing Anxiety as an observed variable resulted in admissible
solutions, any models containing Self-Perceived Ability as either an observed or latent
variable continued to reach inadmissible solutions. I consequently decided to remove this
variable from all analyses.

A new model was then speciﬁed eliminating Self-Perceived Ability and including
Anxiety as an observed instead of a latent variable. Fit indices for this revised model,
however, were outside the acceptable range [1:2 (190) = 413.33 (p = .00), CFI = .88, NFI
=30, RMSEA = .09]. Modiﬁcation indices provided evidence that this was because the

six observed variables comprising Motivation were not distinct factors. Instead, a three-
factor solution for Motivation, consisting of collapsed mean scores for L2 Study and L2
Reading Intrinsic, Extrinsic and Amotivation, was indicated. This was consistent with
correlation patterns described in section 4.2.1.5 above, where strong correlations
(r = .76 - .84) were noted between the Intrinsic, Extrinsic and Amotivation subscales for
L2 Study and L2 Reading. After revising the Motivation variable as described, model ﬁt
imProved substantially, with indices indicating acceptable ﬁt: 1'2 (130) = 220.54 (p = .00),
CFI = 93, NFI =.85, RMSEA = .07. Figure 1 represents this model and the loadings for
each factor. Loadings ranged from .38 to 1.00 and were all statistically signiﬁcant

[1’ < ~01), indicating adequate measurement of the latent variables by their indicators.

129

I

ILIWI‘GG

 

 

 

 

 

 

 

 

 

 

ITReading I. ,,
[Mn/7 T .81 A .
W464 ajI L2 Proﬁcrency #, I, , ,:
F, x .- 77,7777 57, II \\\
VocabularyIA' I .26 \_,
FVocabula ; .70 e77 a.) I. ,- , I
“757129 I I L1 Reading 1 , , I \\
I Ability :43 :1- I
I I‘m 43 , 7,," I“ '51 .46 I 59
I Texts \ . ~ \
‘7 a MIgItIacqggiﬁve 4 I u
I I ,/ /I I I
I Reading «57 °‘”° 93 ’ :.01 I I
";n . . I I.62‘ I
I Speaking I‘ 45 : 01 I I I II
‘~\¥‘ I- I I I I
‘T Tv'ﬁ I I I “ I
L Reading-Ii” I.o7 'I I I
' . I I I I
i ‘ I 5*. A I I ‘ I I I
Media 4 W 4 L2 Exposure I I I .36‘:
I ' 1'18 ‘I I I'.15 I III ‘I 36
I m * i * ’ I II I‘ I III I II II I"
I I .02 II II I
I\ Other —‘I‘ .38 51 I I I
k " I -.19I’
I313” I‘, - -ﬁi ,I ’ II If I
Extrinsu; ‘ *7 7 ~ Motivation I II
I; :I ‘ S 17;
«.58 I
k5: L2$tudy Habits‘ II
Edy\ Tim e+~l I III
-83 * I.1o
a- ﬁ , 'I‘I. ,
Anxiety 3 '1 g
I a:

Figure 1. Measurement model for predictor variables.

130

4. 3. 1.2. 2. Outcome variable.

A measurement model for Reading Comprehension was speciﬁed and tested,

consisting of the observed variables of recall comprehension scores for Narrative Text 1,

Narrative Text 2, Expository Text 1 and Expository Text 2. Model ﬁt was excellent

[x2 (2) = .226 (ns), CFI = 1.00, NFI = .999, RMSEA = .00], and factor loadings were

high, ranging from .68 to .88. The model is presented in Figure 2.

 

 

 

Reading
Comprehension
.83 .\ REM/x” .68
./ .88 I'76 .\
4M hm I__ i I I ‘1
' I . I I
NT1 I NT2 I ; ET1 I ' ET2

 

i
I
I

 

 

FWI L-.- r__; L IT“) __ 4

Figure 2. Measurement model for outcome variable.

4.3.1.3 Structural model

In the second step, a structural model was created to test the relationships among

variables. Initially, paths were speciﬁed between all predictor variables and the outcome

variable to determine which had direct, signiﬁcant effects on reading comprehension.

Paths were also speciﬁed between the predictor variables reﬂecting the hypotheses

presented in section 3.2.3. To summarize these hypotheses, L1 Reading Ability and L2

froﬁciency were expected to have direct, positive effects on Reading Comprehension;

131

Mctaeognition was hypothesized to have positive effects on L1 Reading Ability; L2
Exposure was expected to contribute to L2 Proﬁciency; positive effects of Motivation
were expected on L2 Exposure, Study Habits and L2 Proﬁciency; L2 Study Habits were
hypothesized to have positive effects on L2 Proﬁciency; Anxiety was predicted to have

negative effects on L2 Exposure and L2 Proﬁciency.

The initially speciﬁed structural model is presented in Figure 3'. Overall model ﬁt

was within acceptable parameters: x2 (214) = 326.68 (p = .00), CFI = .94, NFI = .84,
RMSEA = .06. In this model, consistent with hypotheses, L2 Proﬁciency had positive,
signiﬁcant effects on Reading Comprehension (.74). Effects of L1 Reading Ability were
also signiﬁcant and positive, though weaker (.15). Direct effects of the other predictor
variables on Reading Comprehension were small and non-signiﬁcant. Metacognitive
Knowledge’s contributions to Reading Comprehension were mediated as expected
through effects on L1 Reading Ability (.53), but also through L2 Proﬁciency (.38). L2

EXPosure, as predicted, contributed signiﬁcantly (.39) to L2 Proﬁciency. Motivation also

contributed directly (.40) to L2 Proﬁciency, with even larger contributions to L2

Exvosure (.50) and L2 Study Habits (.68). Contrary to expectations, L2 Study Habits had

no Signiﬁcant effects on any other factors. Its contribution to L2 Proﬁciency, in fact, was
negative, though quite weak (-.05). Finally, though Anxiety had an expected negative
effect on L2 Proﬁciency (-.12), this effect was non-signiﬁcant. Anxiety’s negative effect
on L2 ExDosure was also very weak (-.04). Instead, Anxiety had a signiﬁcant positive

effect on L2 Study Habits (24)-

/\

' Indicators that were depicted in graphic representations of the measurement models have been removed
50m the representations of structural models for ease of presentation.

132

.aaaa nausea; =3 .88. 3385 .33 .m seem

so.

 

 

we,“
.3/
A, 232.. spawn. Y1

//NI\\

. . l.)
.. .. \\ rl/l.
. a /
\ \e x 4.,
x ..

x E O _
x . r a.
x x r; t..

, v . \t

.\
a . x
, \
. , x
. s. .\ x
t \
. \ .
\ \\
s .. t\t\
\ K»
n .. . .\ a
\ . s ,

 

 

\.
x. t . x,
\ t . . // I;
\
4‘ x. ,. B.

l... \..
.\

A. coacogoasoo Wt/ttt 3.
,E /// ..///

mated F 2. x; 552
, 7 9.68m 3

v/
as

t \\
\
\

 

 

 

 

,
x
(V
x
/
om. ..
I

t x. ..
..t .t ., \ \
\.\. .

t.\,..\ // .s ./ C

3...... .. 2,, 3:03:99... N.— Ti|ttt4 ,.
. . mm. ,,

 

133

In a ﬁnal step, the non-signiﬁcant paths in the structural model were removed.

This ﬁnal structural model is presented in Figure 4. The removal of non-signiﬁcant paths
resulted in changes in regression weights for relationships between L1 Reading Ability,
L2 Proﬁciency, Metacognitive Knowledge, and Reading Comprehension. L2
Proficiency’s effect on Reading Comprehension increased slightly (.79), but the
contribution of L1 Reading Ability increased more (.19). Together, these variables
accounted for 67% of variance in Reading Comprehension. Metacognitive Knowledge’s
contribution to L1 Reading Ability remained the same, but its effect on L2 Proﬁciency
increased from .39 to .43. Regression coefficients for other paths between variables in the

model changed little or remained constant. Fit indices for this ﬁnal structural model were

identical to the previous model and again indicated acceptable ﬁt: x2 (214) = 326.68

(p = -00),c1=1 = .94, NFI = .84, RMSEA = .06.

134

/"‘
I, \.

> . .
‘

.........tco_mcocoEEoo ,

f
I
I
l/

Ytttt

2.

.

38cm 3 v

or.

a W\tt/// 8.
. 9.5a: %Bw N4 V.

\

a
x
_
a

r/
1 I
.l/ t
u t’.

.28 Has sewage 388 559% was .s semi

 

.\

\ /

3:20:21 N._ {t
mm.

 

/

 

_ 5.2 It
f 958m 5 T

1,,
_. .

 

,/

A, C

 

c2523:

.l t, 05398 N. ,V.

 

VN.

 

 

 

. \ttIt/II
,
V
/ . U
_
..
I . ...
..(t.

 

.tttt\\\.\l\VI|/-/,/ .
emecioex
. s

,, ozgcmooﬁez
//t\\4

‘J _—

135

4.4 Research Question 2: Text genre and L2 reading comprehension
4. 4. I Research Questions 20 and 2b: Genre, level and reading comprehension

Research Question 2a asked about differences in L2 reading comprehension by
text genre. It was expected that signiﬁcantly more would be recalled from the narratives
than the expository texts, and that genre would have a signiﬁcant main effect on reading
comprehension. Research Question 2b inquired about differences in comprehension of
each genre by learner level (beginning, intermediate, advanced). It was hypothesized that
differences in comprehension of the two genres (higher comprehension scores for
narrative texts) would be greatest among advanced learners, and greater among
intermediate level learners than beginners.

Collapsed mean scores (displayed in Table 17) showed that participants
comprehended more from the narrative texts, a difference that was statistically signiﬁcant
as determined by paired-sample t-test (p = .00). To determine the effects of genre on
reading comprehension scores and the effects of learner level on comprehension of each
genre, a 2 x 2 factorial ANOVA was performed with Level as a between-subjects factor,
and Genre and Text as within-subjects factors, each with two levels (Narrative and
Expository; Text 1 and Text 2). A Bonferroni adjustment for multiple comparisons was
applied to this analysis. Results, reported in Table 22, showed signiﬁcant main effects for
Level [F (2, 150) = 20.33, p = .00], Genre [F (l, 150) = 219.51, p = .00], and Text
[F (1, 150) = 8.93, p = .00]. Genre’s effect size very large (r = .77), and was more than
twice that of Level (r = .35) and three times that of Text (r = .24). These results
conﬁrmed that there were signiﬁcant differences in reading comprehension scores by

genre and by learner level, and also by text that was read. Students of all levels

136

comprehended signiﬁcantly more from the narrative passages. Not surprisingly, students
enrolled in advanced classes outperformed intermediate and beginning-level students, and
intermediate students outperformed beginners.

Table 22

Analysis of variance for eﬂects of genre, text and level on reading comprehension

 

 

Source df SS MS F p
Between Subjects
Level 2 .86 .43 20.33 .00
Error 150 3.18 .02
Within Subjects
Genre 1 1.32 1.32 219.51 .00
Genre x Level 2 .15 .07 12.05 .00
Error (Genre) 150 .90 .01
Text 1 03 .03 8.93 .00
Text x Level 2 .00 .00 .45 .64
Error (Text) 150 .44 .00
Genre x Text 1 .04 .04 11.20 .00
Genre x Text x Level 2 .05 .03 8.22 .00
Error (Genre x Text) 150 .47 .00

 

Signiﬁcant interactions between Genre and Level [F (2, 150) = 12.05, p = .00],
Genre and Text [F (1, 150) = 11.20, p = .00], and Genre, Text and Level [F (2, 150) =
8.22, p = .00] were also observed. All effect sizes were moderate, ranging from .23 to .27.
As hypothesized, differences in comprehension scores between the narrative and
expository texts were greatest for advanced-level students, and greater for intermediate
students than beginners. Figure 5 illustrates the interaction between Genre and Level.

Advanced learners recalled 15% more narrative (29%) than expository propositions

137

(14%). This difference was somewhat less pronounced for intermediate students, who
recalled 21% of narrative propositions and 10% of expository propositions, a difference
of 11%. Though statistically signiﬁcant, the differences between narrative (12%) and
expository (7%) scores for beginners were the least marked (5%).

The interaction between Genre and Text, represented in Figure 6, pointed to a
different pattern in comprehension scores within each genre. Estimated marginal means
for this interaction showed that while comprehension scores for the two narrative texts
were identical (22%), scores for the two expository texts were not; fewer propositions

were recalled from Expository Text 2 (9%) than from Text 1 (12%).

138

 

 

 

 

 

0.30 - ‘
0.25 - \.,
\O
020 - . ~ \~.
\
\ \
x o. . .
\ \ \ \ Begmnmg
0'15 ‘ \ .° Intermediate "' """
Advan "" ' "' ' "' '
0.10 d
0.05 d
T I
Narrative Expository
Figure 5. Reading comprehension scores by genre and learner level.
0.21 "
0.18 "
Text 1
0-15 " Text 2 ------
0.12 "

 

 

 

Narrative Exprository

Figure 6. Reading comprehension scores by genre and text.

139

The interaction between Genre, Text and Level as depicted in Figure 7 shows a
differential pattern of scores between the two texts by genre among beginning,
intermediate and advanced students. Within the narrative genre, both beginners and
intermediate readers comprehended somewhat more from Text 1 than from Text 2
(beginning 2% more; intermediate, 1% more). Advanced students, on the other hand,
comprehended 4% more from Text 2 than Text 1. Within the expository genre, all groups
comprehended more from Text 1 than Text 2. There was no signiﬁcant interaction,
however, between Text and Level, indicating that there was no difference in scores by

learner level between all the four texts taken together.

140

 

 

 

 

 

 

0.35 -+
0.30 ~ ,.——- - ' A
. ’
o ' ’
r Beginning
0.25 " Intermediate
. __________ -. Advanced
0.20 ‘
0.15 "
F
a
0.10 "
I r
Text 1 Text 2
Narrative
0.18 “
'~.
\ O
0.15 °\.
°\. Beginning
0.13 ‘ °\ Intermediate
. ~ ' ‘ Advanced
awe ‘~~‘
‘ '0
0.08 ‘ F
a
r I
Text 1 Text 2
Expository

Figure 7. Reading comprehension scores by genre, level and text.

141

4. 4. 2 Research Question 20: Genre and individual did'krences

Research question 2c asked whether individual differences contributed differently
to comprehension of the two genres. It was expected that the contributions of L1 reading
ability and L2 proﬁciency would be greater for expository texts than for the narrative
passages. To test this hypothesis, two separate structural models were speciﬁed and
tested, one with narrative comprehension and one with expository comprehension as the
outcome variable. Regression coefﬁcients for paths within the two models were then
compared for statistical signiﬁcance. These models and their signiﬁcant paths are
presented in Figures 8 and 9.

Fit statistics for both models were acceptable [narrative model, 1:2 (181) = 277.22
(p = .00), CFI = .94, NFI = .84, RMSEA = .06; expository model, 1:2 (181) = 295.23

(p = .00), CFI = .92, NFI = .83, RMSEA = .06]. Although both models showed
similar relationships between variables, differences in several path regression weights
were observed. As expected, the contributions of L1 Reading Ability and L2 Proﬁciency
to L2 Reading Comprehension were greater in the expository comprehension model than
in the model of narrative comprehension. However, only the differences in the
contributions of L2 Proﬁciency between the two models were statistically signiﬁcant (p <
.05) as determined by the critical ratio for differences between these regression weights
(CR = 2.29). Differences in path regression coefﬁcients were also noted in the
contributions of L2 Exposure to L2 proﬁciency and Motivation to L2 Proﬁciency, but

these differences were not statistically signiﬁcant.

142

Arno mﬁa Becamewmmv comaonoaﬁoo i3 3:83: new $38 gogm .m. 233k

 

1.. 5.5285. s

F I\\\/ 8. ..-.t /,
\. .«l ..J fl zo_xc<

 

a, gas... seam N. .it! . .

. seasoned S Tillie 2325 S ,V

x
w . a
h ..
z ..x a o
... I // Fm .,./ \\.
.. ,/ x
. /’{I‘\A II/ .\\.\1\\

 

 

NZ l7\I\I/,/~ /./////..
. coﬁcocanoo .L
me.

93252 _4 //

 

 

 

 

an A, u ., w 9. I 352 LTI mate—BE! xv
f/t. .. a 05301 w..._ ﬁZtcuOOEOE
1/ L #0. (\f
w

/

A “L / .
\ /\

II .\

/t\.

 

143

.315 2:3 camoﬁcwmmv :ommcosoaﬁoo 38H b86098 “8 $608 Razosbm a Emmi

e

 

and: spam N._

om.

No.

 

co_mco;anoo

 

 

 

bozmoaxw
Fm F

 

 

3:30:05 3

mm.

Nv.

 

5. a F 8.

 

3:52
968m 3

 

 

em.

.3:

 

cozmzuos. bo_xc<

 

 

 

Fm.

osmoaxm N1.

 
     

  

monotsocv.

ozgcmoomaos.
em.

144

.‘4’1

4.5 Research Question 3: Genre and text reactions
4. 5. I Research Question 3a: Genre reactions

Research Question 3a inquired about readers’ reactions to the narrative and
expository genres in the L2, their L2 genre preference, and whether these reactions
differed by learner level. It was expected that readers of all levels would report greater
levels of interest and enjoyment for the narrative genre and would express a preference
for reading narrative texts, but would believe expository texts have greater utility value.
Advanced students were expected to report greater levels of enjoyment, greater utility
value, greater comprehensibility and less anxiety related to both genres than intermediate
or beginning students, however, this difference between advanced and lower-level
students was expected to be more pronounced for the narrative genre.

As hypothesized, overall mean scores for interest and enjoyment were greater for
the narrative genre, a difference that was statistically signiﬁcant by paired-sample t-test
(p =.OO). Participants likewise found the narrative genre signiﬁcantly more
comprehensible (p = .00) and less anxiety provoking (p = .01) than expository texts.
Mean scores of 4.29 on a scale of 6 for Item 17 (“If I had a choice, I would rather read
Stories than informational texts in French”) also conﬁrmed the hypothesis that overall
students would prefer the narrative genre. Contrary to expectations, however, participants
did not associate a greater utility value with the expository genre; mean utility scores
we": nearly identical for the two genres (narrative M = 4.50; expository M = 4.51;

P = .90),
To analyze these results by learner level, a one-way between-subjects analysis of

Variance (AN OVA) compared genre reaction scores (Enj oyment/Interest, Utility,

145

 

Comprehensibility, Anxiety) and scores for Item 17 (Genre Preference) for the beginning,
intermediate and advanced groups. These results are presented in Table 23. There was a
signiﬁcant effect (p < .04) of learner level on all reactions to both the narrative and
expository genres. Within both genres, these effect sizes were small to moderate for

Enj oyment/Interest and Utility (r = .21-.24), moderate for Anxiety (r = .31-.35) and large
for Comprehensibility (r = .43-.44). Preference of the narrative genre, however, did not
vary by learner level (p = .93).
Table 23

Analysis of variance for eﬂects of learner level on genre reactions

 

 

 

Source df SS MS F p
Narrative

Enj oyment/Interest
Between 2 8.29 4.14 4.55 .01
Within 150 136.63 .911
Total 152 144.92

Utility
Between 2 9.90 4.95 4.50 .01
Within 150 165.08 1.10
Total 1 52 174.98

Comprehensibility
Between 2 28.86 14.43 17.10 .00
Within 150 126.60 .844
Total 152 155.46

Anxiety
Between 2 18.94 9.47 8.13 .00
Within 150 174.69 1.17
Total 152 193.63

146

Table 23 continued

 

 

 

 

 

 

 

Source df SS MS F p
Expository
Enj oyment/Interest
Between 2 7.87 3 .93 3 .29 .04
Within 150 179.09 1.19
Total 152 186.97
Utility
Between 2 1 1.23 5.62 4.62 .01
Within 150 182.25 1.215
Total 152 193.48
C omprehensibility
Between 2 31.48 15.74 17.46 .00
Within 150 135.20 .901
Total 152 166.68
Anxiety
Between 2 25.24 12.62 10.68 .00
Within 150 177.31 1.18
Total 1 52 202.55
Treference
T Between 2 .33 .16 .03 .93
Within 150 315.02 2.10
Total 152 3 15.35

To further analyze level effects, post hoc comparisons of genre reaction scores by
level Were performed using a Bonferroni adjustment for multiple comparisons. Results
are displayed in Table 24. Within the narrative genre, beginning and intermediate-level
learn€31‘s’ reactions were for the most part aligned, with one exception: beginners found

the narrative genre signiﬁcantly less comprehensible than intermediate students. Clear

147

differences in all categories of reactions to the narrative genre were observed between _
beginners and advanced students, as well as between intermediate and advanced students.
Advanced students reported greater levels of enjoyment/interest, perceptions of genre
utility and comprehensibility, and less anxiety related to reading narrative texts. Turning
to expository genre reactions, there were no differences between beginning and
interlnediate learners’ reactions to this genre, however, the same pattern of differences
emerged as observed in the narrative genre when comparing reactions of beginning and

internrediate students with those of advanced learners.

148

Table 24

Means and post hoc results (Bonferroni) for eﬂects of learner level on genre reactions

 

 

 

 

 

 

 

 

Narrative
Enjoyment/ Utility Comprehensibility Anxiety
Interest
M SD M SD M SD M SD
Beginning 4.17,), 1-07 4.19, 1-29 2.95 a 1-07 3.46a 1-17
Intermediate 422 a .97 4.41 a 1.04 3.47 b .90 3.24 a 1.04
Advanced 4.76 b .83 4.93 b .85 4.27 c .84 2.48 b 1.10
Expository
Enjoyment] Utility Comprehensibility Anxiety
Interest
M SD M SD M SD M SD
Beginning 3.84 a -93 4.23 a 1-14 2.69 a -92 3.65 a 1-11
Intermediate 3.67 a 1.18 4.40 a 1.18 3.16 a .87 3.41 a 1.09
Advanced 423 b .95 4.93 b .83 4_05 b .88 2.53 b 1.06
Preference
M SD
Beginning 4.19 a 1-34
Intermediate 432 a 1.48
Advanced 4.28 a 1.43

 

Note. Means with different subscripts differ signiﬁcantly vertically at p < .05.

149

it"

9

to.

“DU

PICS

flit

4. 5. 2 Research Question 3 b: Text reactions

Research Question 3b asked about students’ reactions to the four passages they
read in the study, and whether these reactions differed by genre. It was hypothesized that
students would report greater levels of enjoyment/interest, greater comprehensibility and
less anxiety related to the narrative texts that they read, but would ﬁnd that the expository
texts that they read had greater utility value. It was further expected that these reactions
would vary not only by genre, but also by text. Descriptive statistics for text reactions,
presented in section 4.2.2.3, were consistent with these hypothesized patterns. To test the
statistical signiﬁcance of these relationships, a 2 x 2 repeated measures AN OVA was
performed for each of the text reactions (Enjoyment/Interest, Utility, Comprehensibility,
Anxiety!) with Genre and Text as ﬁxed factors, each with two levels (Narrative and
Expository; Text, Text 1 and Text 2). Results of this analysis are presented in Table 25.
As hypothesized, Genre had a signiﬁcant main effect (p = .00) on all four categories of
text reaction scores, ranging from a small effect size on Anxiety (r = .16) to moderate
effect on Utility (r = .24) and large effects on Comprehensibility (r = .47) and Enjoyment
(r = .52). There was also a signiﬁcant main effect for Text (p < .04), which was small for
Utility (r = .17), moderate for Anxiety (r = .16) and Enjoyment, (r = .30) and large for
Comprehensibility (r = .51). A signiﬁcant interaction between Genre and Text (p = .00)
also existed, which was small for Utility (r = .14) and moderate for Anxiety (r = .31),
Comprehensibility (r = .3 7) and Enjoyment (r = .43). Thus, not only did reactions to the

texts vary by genre but they also varied by text, and within each genre a different pattern

 

' Though data on topic familiarity was also gathered on the Text Reactions questionnaire, it was not
included in the present analysis. Since each text dealt with a topic completely unrelated to those of the other
texts, genre was not expected to contribute to variance in topic familiarity. Topic familiarity was also not
judged relevant to other analyses in the present study because, as described in section 4.2.2.3, it did not
correlate with comprehension of the texts.

150

of reactions was observed between the two texts. Within the narrative genre, estimated

marginal means for this interaction showed that students enjoyed Text 2 more than Text 1

and believed it had greater utility value. However, they found Narrative Text 1 more

comprehensible and less anxiety provoking than Text 2, though anxiety ratings for the

two narrative texts were relatively low (M = 2.72). In the expository genre, in contrast,

reactions to Text 1 were consistently more positive; students reported greater levels of

enjoyment, greater utility value, greater comprehensibility and less anxiety.

Table 25

Analysis of variance of eﬂect of genre and text on text reactions

 

 

 

 

 

Source df SS MS F p
Enjoyment/Interest
Genre 1 67.09 67.09 56.41 .00
Error (Genre) 152 180.77 1.19
Text 1 6.40 6.40 15.21 .00
Error (Text) 152 63.93 .421
Genre x Text 1 18.33 18.33 33.65 .00
Error (Genre x Text) 152
Utility
Genre 1 5.47 5.47 9.03 .00
Error (Genre) 152 92.18 .61
Text 1 1.36 1.36 4.40 .04
Error (Text) 152 46.86 .31
Genre x Text 1 7.57 7.57 23.90 .00
Error (Genre x Text) 152 48.12 .317

15]

Table 25 continued

 

 

 

 

 

 

Source df SS MS F p
Comprehensibility
Genre 1 47.68 47.68 43.87 .00
Error (Genre) 152 165.18 1.09
Text 1 23.60 23.60 54.83 .00
Error (Text) 152 65.43 .43
Genre x Text 1 10.07 10.07 23.48 .00
Error (Genre x Text) 152
Anxiety
Genre 1 22.83 22.83 29.74 .00
Error (Genre) 152 116.68 .77
Text 1 4.62 4.62 13.13 .00
Error (Text) 152 53.53 .35
Genre x Text 1 5.65 5.65 16.38 .00
Error (Genre x Text) 152 52.42 .35
4.6 Summary of results

Research questions 1a and 1b asked about the contributions of individual
differences to L2 reading comprehension, as well as the interrelationships of these
variables. Structural equation modeling was used to examine these relationships. Within
the resulting model, L1 reading ability and L2 proﬁciency made signiﬁcant and direct
contributions to reading comprehension in French. L2 proﬁciency, however, was a much
greater predictor of L2 reading comprehension, accounting for more than 15 times more
variance than Ll reading ability. These two variables explained 67% of the variance in
L2 reading comprehension. Other individual differences contributed indirectly to L2

reading through direct contributions to other variables in the model. As hypothesized,

152

metacognitive knowledge contributed to L1 reading ability, but also made a signiﬁcant
though somewhat smaller contribution to L2 proﬁciency. L2 exposure contributed to L2
proﬁciency as expected. Motivation contributed to both L2 exposure and proﬁciency,
with its largest contribution to L2 study habits. L2 study habits, however, did not
signiﬁcantly predict L2 proﬁciency. Rather smprisingly, their relationship with L2
proﬁciency was negative, but statistically non-signiﬁcant. Anxiety did not make any
hypothesized contributions to variables in the model, though it did make an unexpected
signiﬁcant and positive contribution to L2 study habits. Together, all these individual
difference variables explained 73% of L2 reading comprehension variance.

A second set of research questions inquired about differences in reading
comprehension due to text genre (Research Question 2a), differences in the
comprehension of each genre by learner level (Research Question 2b), and the
contributions of individual differences to comprehension of each genre (Research
Question 2c). Participants comprehended signiﬁcantly more ﬁ'om the two narrative
Passages than from the expository texts, as expected. The effect of genre on reading
comprehension was signiﬁcant and sizeable. However, signiﬁcant interactions between
genre and text existed, indicating differences in comprehension by text within each genre.
Colllpl'chension scores were identical for the two narrative texts, but students recalled
Signiﬁcantly less from the second expository text than from the ﬁrst. Even so,
comprehension scores were signiﬁcantly higher for both of the narrative texts than for
either one of the expository passages. These differences in comprehension of each genre
Were formd among learners at all three levels of instruction (beginning, intermediate and

advanced), although the differences were more pronormced at the advanced level than

153

intermediate or beginning, and more pronounced among intermediate learners than
beginners. When comprehension of each genre was included in a model of individual
differences, L2 proﬁciency’s contribution to expository comprehension was signiﬁcantly
greater than its contribution to narrative comprehension.

A third set of research questions inquired about students’ affective reactions to
reading both genres in general and the experimental texts in particular. Research question
3a inquired about students’ affective reactions to reading the narrative and expository
genres in the L2, and whether these reactions vary by learner level. Signiﬁcant main
effects of genre on students’ reactions indicated greater enjoyment, greater perceptions of
comprehensibility and less anxiety related to the narrative than the expository genre in
the L2. Furthermore, students reported a preference for reading narrative texts. Somewhat
unexpectedly, however, students did not ﬁnd reading expository texts in the L2 more
useful than reading narratives even though the expository genre ‘3 primary
communicative pin-pose is informational. Reactions to each genre varied signiﬁcantly by
learner level, with advanced students reporting greater enjoyment, utility and
comprehensibility and the less anxiety related to both genres. Beginning and intermediate
students’ reactions to genre were statistically identical, with the exception of perceptions
of comprehensibility; beginners found narrative texts less comprehensible than
intermediate students did.

Research Question 3b asked about students’ reactions to the four passages they
read in the present study. The signiﬁcant effect of genre on their reactions to the texts
indicated that they enjoyed the two narrative texts more, believed they understood them

more, and experienced less anxiety when reading them. However, they found the two

154

experimental expository passages signiﬁcantly more useful to read. Within each genre,
nevertheless, a signiﬁcant main effect of text on students’ reactions indicated that these
reactions to the individual texts were quite varied. Participants’ comments suggested that
in addition to genre, other text characteristics, including knowledge and interest in the
topic, text organization, prominence of text details, vocabulary and grammar also

inﬂuenced their experience reading the four texts.

155

CHAPTER 5: DISCUSSION AND IMPLICATIONS

In this chapter I will discuss the results reported in Chapter 4 as they relate to the
research questions investigated in the present study. For each research question I will
summarize and discuss the results as well as their theoretical implications.
Methodological and pedagogical implications of the study will then be discussed. Finally,

I will discuss the study’s limitations and make recommendations for future research.

5.1 Research Question 1: Individual differences and L2 reading comprehension
The ﬁrst research questions inquired about the contributions of individual

differences to L2 reading comprehension and the interrelationships of these variables in a
model of L2 reading. As hypothesized, French proﬁciency and English reading ability
made substantial, direct contributions to reading comprehension in French. Together,
these variables explained two-thirds of reading comprehension variance. The predictive
power of L2 proﬁciency, however, was far greater, accounting for over 15 times more
variance than L1 reading ability. Other individual differences made indirect contributions
to L2 reading comprehension by contributing to other variables in the model.
Metacognitive knowledge contributed as expected to L1 reading ability, but also made a
signiﬁcant though somewhat smaller contribution to L2 proﬁciency. Exposure to French
also made an expected contribution to French proﬁciency. While motivation contributed
to both L2 exposure and proﬁciency, its largest contribution was to L2 study habits.
Nevertheless, these study habits did not predict French proﬁciency, and in fact had a

negative but statistically insigniﬁcant relationship with this variable. Though anxiety’s

156

relationships with L2 exposure and L2 proﬁciency were negative as expected, they were
not signiﬁcant. In fact, anxiety did not contribute signiﬁcantly to any variables in the
model except L2 study habits, where higher levels of anxiety were associated with more
time and effort dedicated to French study. The model of individual differences including
all these described variables accounted for 73% of variance in L2 reading
comprehension.
5.1.] L1 reading ability and L2 proﬁciency

The signiﬁcant contributions of both L1 reading ability and L2 proﬁciency to L2
reading comprehension in the present study provide additional evidence that reading in
another language is neither “a reading problem or a language problem” (Alderson, 1984,
p. 1). Rather, reading in another language is a complex, multidimensional process
involving both native and target language knowledge and skills. The relative
contributions of these two variables in the present study also substantiate previous
ﬁndings that target language proﬁciency, however, plays the much larger role in L2
reading (Bernhardt & Kamil, 1995; Bossers, 1991; Brisbois, 1995; Carrell, 1991; Lee &
Schallert, 1997). This result supports interactive theories of reading, which underline the
importance of language knowledge in word decoding, a necessary precondition for
constructing meaning from text and connecting text meaning with other knowledge
sources.

5.1.2 Metacognitive Imowledge

Metacognitive knowledge’s indirect contribution to L2 reading comprehension

through both native language reading ability and target language proﬁciency validates

previous ﬁndings on its role in the L2 reading process. That is, general knowledge about

157

texts and reading is implicated cross-linguistically in both L1 and L2 reading (Schoonen,
et al., 1998). Participants in the present study, who with few exceptions began their
French studies as adolescents or adults and who reported reading very infrequently in
French, had undoubtedly developed the bulk of their knowledge about texts and reading
strategies when reading in English. Yet, the metacognitive knowledge they had
developed through Ll experiences also contributed to their L2 knowledge and skills.

Given that metacognitive knowledge speciﬁc to texts and reading was measured
in the present study, this contribution to general L2 proﬁciency was not hypothesized.
However, it is not an entirely unexpected result given that language proﬁciency was
assessed using measures of not only L2 knowledge (vocabulary and grammar) but also
L2 skills, which included reading. Clearly, metacognition about texts and reading would
be implicated in the measures of reading skills included in the French proﬁciency test. It
is also possible, however, that certain types of strategic knowledge measured by the
reading metacognition questionnaire are related to strategies that foster the development
of other types of L2 knowledge. For example, the strategy of inferring word meaning
from context cues likely facilitates vocabulary knowledge as well as reading skills.

The supporting role played by metacognition also supports interactive views of
L2 reading, which maintain that both linguistic knowledge and other sources of
knowledge are implicated in the reading process. This was conﬁrmed in the present
study, where readers used both knowledge of French (L2 proﬁciency) and knowledge of

texts and reading to construct meaning from the reading passages.

158

5.1.3 L2 exposure
Exposure to French contributed to French proﬁciency in the present study. The
time that students spent in contact with French through speaking, reading, media, study
abroad and other experiences made positive, signiﬁcant contributions to their language
skills. This result supports connectionist theories of language acquisition, which maintain
that L2 exposure facilitates L2 development.

The indirect contribution of L2 exposure to L2 reading through language
proﬁciency also provides additional evidence that exposure to oral language improves
reading performance. In the present study, students’ extracurricular French contact was
overwhelmingly with oral language, yet even this contact contributed (albeit indirectly) to
reading comprehension scores.

5.1.4 L2 study habits

It was hypothesized that learners’ L2 study habits, a measure of motivational
intensity, would make positive contributions to L2 proﬁciency, and thus to L2 reading
comprehension. Study habits were measured using self-reports of time students studied
for French relative to other courses and their persistence in completing French
homework. However, study habits’ contributions to language proﬁciency were not
signiﬁcant, and quite unexpectedly, the relationship of these two variables was negative:
the more students reported studying French and doing homework, the poorer their actual
French performance.

One explanation for this counterintuitive result may lie in the fact that the
questions assessing study habits in the present study focused more on the quantity of time

students spent on French study than on its quality. Plant, Ericsson, Hill and Asberg

159

(2004) explain the absence of reliable relationships between study habits and academic
outcomes within the framework of “deliberate practice”, proposing that only highly
focused, deliberately planned study time promotes achievement. In a study examining the
effects of study time and study quality on college grade point average (GPA), the
researchers found that students who reported studying in distraction-free environments
had signiﬁcantly higher GPAs than those who did not, and that study environment was a
signiﬁcant predictor of GPA. Study time, in contrast, only predicted GPA when study
environment and previous academic achievement were controlled for, and often lower
achievers reported spending much more time studying than students with higher grades.
Future L2 research examining the predictive capacity of study habits, therefore, should
include measures of study quality as well as quantity.
5.1.5 Motivation

Motivation was conceptualized in the present study as a three-factor construct,
composed of intrinsic, extrinsic and amotivation orientations. This model is consistent
with the theoretical framework adopted in L2 research most notably by Noels et al.
(1999; 2000). Conﬁrmatory factor analyses provided support for this three-factor
motivational model.

Given Mori’s (2004) ﬁndings of componential differences between motivation for
L2 study and for L2 reading, it was initially expected that these two types of motivation
would emerge as separate factors in the motivation model. However, conﬁrmatory factor
analyses did not show that motivation for L2 study and for L2 reading were distinct
constructs. In other words, students’ reasons for studying French were not signiﬁcantly

different than their reasons for reading in French. This result may be explained by the fact

160

that most participants’ L2 reading experiences were related to their French classes.
Overall, students reported spending a negligible amount of time (less than 25 minutes per
week) reading in French outside of class, with nearly half reporting that they never do so.
It seems likely, then, that participants viewed reading in French as part and parcel of
studying French, and their motivations for engaging in both activities were therefore
indistinguishable. In contrast, Mori (2004) measured motivation to read as it related to a
speciﬁc, self-directed extensive reading activity that students performed outside of class.

Motivation contributed indirectly to L2 reading comprehension in the present
study through contributions to L2 exposure, and to a lesser extent, L2 proﬁciency.
Motivation’s largest contribution was to study habits, though as described in section 5.1.4
above, these habits did not predict either French proﬁciency or reading comprehension.
These ﬁndings support Demyei and Korlnos’ (2000) assertion that motivation “is the
antecedent of the action rather than achievement” (p. 281). Though motivation may not
be directly predictive of particular L2 outcomes, it plays a ﬁmdamental role in their
eventual attainment by triggering more proximally-related behaviors. In the present
study, these behaviors included engaging in activities involving the target language and
devoting time and effort to French study, though only the former contributed to target
language proﬁciency and in turn to L2 reading performance.

5.1.6 Anxiety

Contrary to expectations, anxiety was not a signiﬁcant predictor of either
language exposure or proﬁciency. Although non-signiﬁcant, relationships between
anxiety and both L2 exposure and L2 proﬁciency were negative: higher levels of anxiety

were associated with less contact with French and lower French proﬁciency. The

161

direction of this relationship is consistent with the preponderance of L2 anxiety research,
which has found that anxiety plays an inhibitive rather than facilitative role in the
development of L2 skills.

A possible explanation for the absence of any signiﬁcant contributions of anxiety
to L2 outcomes is suggested by a line of research conducted by Ganschow, Sparks and
associates (Ganschow & Sparks, 1996; Ganschow, et al., 1994; Sparks & Ganschow,
1991). These researchers dispute the notion that anxiety is the cause of poor language
performance, proposing instead that the reverse is true: poor language skills cause
learners to become anxious about using them. Although this hypothesis was not tested in
the present study, future research involving linguistic and affective variables may wish to
take this possibility into account.

Anxiety’s only signiﬁcant contribution in the present study was to L2 study
habits. Anxious students reported studying more, yet more study time did not predict
better performance, mirroring Horwitz, et al.’s (1986) observation that anxiety may

motivate unproductive “overstudying”.

5.2 Research Question 2: Text genre and L2 reading comprehension
The second set of research questions asked about the role of text genre in L2
reading comprehension. More speciﬁcally, they asked about differences in reading
comprehension due to genre (narrative or expository), whether differences in the
comprehension of each genre existed by learner level, and about the contributions of

individual differences to comprehension of each genre.

162

5. 2. I Research Question 20: Genre eﬁ’ects on L2 reading comprehension

As hypothesized, participants comprehended signiﬁcantly more from the two
narrative passages than from the expository texts, and genre had a signiﬁcant and very
large effect on reading comprehension. There was however, a signiﬁcant interaction
between genre and text, indicating differences in recall comprehension scores by text
within each genre. Students recalled almost exactly the same percentage of propositions
ﬁom both narrative texts, however, they recalled signiﬁcantly less from Expository Text
2 than from Expository Text 1. Nevertheless, comprehension scores were signiﬁcantly
higher for both narrative texts than for either of the expository passages.

The superior comprehension outcomes for narrative texts in both the pilot and
present studies provide evidence that as in one’s native language, narratives are more
comprehensible in the L2 than expository texts. These ﬁndings are consistent with L1
reading research, which has consistently pointed to an advantage for narrative texts. In
both studies, however, an interaction between genre and text was present, indicating
different comprehension outcomes for the two expository texts. In both cases these
differences are likely attributable to text characteristics that were not adequately
controlled for. In the pilot study, signiﬁcantly more was comprehended ﬁ'om Expository
Text 2 than Expository Text 1. A post hoc analysis revealed that this passage in fact
contained a narrative-like description from which signiﬁcantly more was recalled than
was from the remainder of the text. The passage was therefore replaced in the present
study with one that contained no such narrative elements. However, the replacement text,
although equivalent to the other texts in length and measures of lexical and grammatical

complexity, contained 22-3 2% more semantic propositions than the other texts. Since

163

there was simply more to remember from this text than the other passages, it is therefore
possible that participants’ failure to recall as much from this text is explained more by the
additional cognitive demands it imposed than by its genre.

These interactions, nonetheless, should not call the results on the relative
comprehensibility of the two genres into question. To the contrary, the fact that
signiﬁcantly more narrative-like propositions were recalled ﬁ'om Expository Text 2 in the
pilot study seems to strengthen the conclusion that narrative text structures aid
comprehension in the L2. The patterns of comprehension scores obtained for the
narrative texts and the remaining expository text in both studies (narrative scores roughly
50% higher than scores for the expository text) also suggest an advantage for narrative
texts. Moreover, the relative effect sizes of genre and text in the present study, with more
than a three times greater effect for genre than text, indicate that the genre of text that was
read was implicated much more in comprehension than the text that was read.

5. 2.2 Research Question 2b: Genre and level in L2 reading comprehension

It was further hypothesized that learners at all levels of instruction (beginning,
intermediate, advanced) would comprehend more from the narratives than the expository
passages. However, given evidence that L2 proﬁciency affects the amount of attention
readers can allocate to global text features (Chen & Donin, 1997; Horiba, 1990), it was
expected that differences in comprehension scores by genre would be greater among
advanced learners than either intermediate or beginning learners, and greater among
intermediate level learners than beginners. Results conﬁrmed these hypotheses, providing
evidence suggestive of a linguistic threshold for genre. That is, below a certain level of

language instruction, readers may be less able to successfully exploit the cues that

164

facilitate narrative comprehension. An alternative explanation of this result, however,
may be that the participants’ exposure to the narrative genre in French increased as they
progressed in their French studies. Advanced-level participants were enrolled in upper-
level content courses requiring regular and abundant reading in French, and nearly half
were taking classes focusing on the reading and interpretation of narrative, literary texts.
While some narrative readings were included in the intermediate-level curriculum at the
time the study was conducted, intermediate students had far less exposure to narratives
than advanced students. The beginning-level curriculum, on the other hand, provided
almost no opportunities for students to read narrative texts in French. Because exposure
to and familiarity with each genre in French were not controlled for in the present study,
deﬁnitive conclusions about a genre threshold cannot be drawn from these results.
5. 2. 3 Research Question 20: Genre and individual diﬂerences

A ﬁnal genre-related research question inquired about the contributions of
individual difference variables to comprehension of each genre of text. It was expected
that linguistic knowledge and skills (i.e., L1 reading ability and L2 proﬁciency) would
play a lesser role in narrative comprehension than in expository text comprehension. This
hypothesis was only partially conﬁrmed. Though both variables did make larger
contributions to expository text comprehension, this difference was only signiﬁcant for
the contribution of L2 proﬁciency. In other words, less French proﬁciency was needed to
comprehend the narrative genre than the expository genre. This result provides additional

evidence of the narrative genre’s facilitative role in text comprehension.

165

5.3 Research Question 3: Genre and text reactions

A third set of research questions examined participants’ affective reactions to
reading narrative and expository texts in the L2 in general, and their reactions to reading
each of the four experimental texts. Concerning genre, it was expected that learners at all
levels would report greater enjoyment of narrative texts and therefore would report an
overall preference for reading narratives, but would view the expository genre as more
useﬁrl. It was further hypothesized that participants would ﬁnd the narrative genre more
comprehensible and less anxiety-provoking. It was also expected that advanced learners
would have more positive affective reactions and less anxiety than lower-level learners
related to texts of both genres, but that this difference would be more pronounced for the
narrative genre. Students’ reactions to the four experimental passages were also expected
to vary by genre, with similar patterns of reactions as described above (greater
enjoyment, better comprehensibility, and less anxiety for the two narrative texts; greater
utility value for the two expository texts). However, reactions were also expected to vary
by text within each genre.

The hypotheses related to genre reactions were for the most part conﬁrmed.
Students of all levels reported that they enjoyed reading the narrative genre more than the
expository genre in French, and if they had a choice would rather read narrative texts.
These results seem reﬂective of one of the primary communicative intents of the
narrative genre, affective engagement of the reader, and point to a relationship between
readers’ affective engagement and preferences. As expected, students also found the
narrative genre more comprehensible than expository texts in the L2, which mirrored

actual reading comprehension outcomes by genre in the present study. Narrative texts

166

were also perceived as less anxiety-provoking than expository texts. This inverse
relationship between anxiety and perceptions of comprehensibility was expected given
previous research associating beliefs about reading difﬁculty with reader anxiety (Saito et
al., 1999). What was unexpected, however, is that students did not ﬁnd reading
expository texts signiﬁcantly more useful than reading narratives in the L2, despite the
fact that the expository genre is more closely associated with the practical purpose of
obtaining information. One possible interpretation of this result is that students associated
all reading in French with their French studies, an explanation supported by the relatively
small amount of time that students reported reading in French outside of class. Students
might therefore have considered reading texts of either genre useful and necessary for
completing assignments, meeting course objectives, and/or developing language skills.
Given that students reported enjoying the narrative genre more, it is also possible that
they associated greater levels of enjoyment with utility value. Students may believe that
being entertainedbyatextisauseful reason forreadingitanoﬁonthatafterall, isatthe
heart of reading for pleasure.

As hypothesized, genre reactions varied signiﬁcantly by learner level, with
advanced students reporting the greatest enjoyment, utility and comprehensibility and the
least anxiety for both genres. These results were expected given that advanced students
were enrolled in upper-level content courses that require extensive reading in French,
whereas beginning and intermediate students had little relative experience reading in their
French classes. As expected, positive perceptions of the narrative genre were more
pronounced for advanced students. Again, these perceptions were likely fostered by

advanced students’ experiences, since the majority of these students were enrolled in

167

classes focusing on literary interpretation. Lower-division students, on the other hand,
had less experience on which to base their assessment of this genre. These results suggest
that the beneﬁts of increased language exposure may not only be linguistic, but also
affective.

The only signiﬁcant difference between beginning and intermediate learners’
reactions to each genre was in perceptions of comprehensibility. Beginners believed
narrative texts were less comprehensible than intermediate-level students did, though
there was no difference between beginning and intermediate students’ perceived
comprehensibility of the expository genre. This result also appears to reﬂect beginners’
relative inexperience reading narrative texts, considering that at the time the study was
conducted the beginning-level curriculum included no narrative readings, but that the
intermediate-level curriculum included some narrative texts. In contrast, both beginning
and intermediate students were required to read expository, cultural-informational
passages in their French classes.

Hypotheses about students’ reactions to the four experimental reading passages
were conﬁrmed. Students reported enjoying the narrative texts more, understanding them
more, and experiencing less anxiety when reading them, but they believed the two
expository texts were more useful to read. With the exception of this judgment about the
texts’ utility, these results mirrored their reactions to each genre in general. Differences in
perceptions of utility for the expository genre in general and for the two expository
passages read in the study may be explained by differences in students’ purposes for L2
reading in general and for performing the reading tasks in the present study. As

previously discussed, participants’ French reading experiences and therefore their

168

exposure to different text genres were largely related to their French classes.
Consequently, students may have found all L2 reading useful regardless of text genre in
order to meet class requirements. In the present study the reading tasks were not tied to
such goals, and readers therefore may have perceived the expository texts that they read
as more related to the useful purpose of acquiring information.

Within each genre, however, there was much variability in students’ reactions to
the individual texts. This wide range of reactions was evidenced by the signiﬁcant effects
of text on all the affective dimensions under investigation, as well as the range of student
comments that each text elicited. Therefore, while genre may play a role in students’
reactions to a particular text, it is certainly not the sole factor determining their
perceptions. Participants’ comments suggested that other text characteristics including
personal engagement with the topic, text organization, salience of text details, and word

and sentence-level features also affected their reading experience.

5.4 Methodological implications

The present study sought to contribute to L2 reading and individual differences
research using several methodological innovations. First, questionnaires for assessing
motivation and affective reactions to L2 study and L2 reading were developed by
adapting items used in existing research instruments (Mori, 2004; Kondo-Brown, 2006;
Yamashita, 2004). Internal consistency estimates and conﬁrmatory factor analyses
indicated that these new questionnaires were reliable measures of the constructs under
investigation, including intrinsic and extrinsic motivation, amotivation, anxiety and

motivational intensity (L2 study habits). These questionnaires may be useful in ﬁrture

169

research on L2 motivation and affect, especially since they measure a broad range of
motivational and affective orientations related to L2 study and L2 reading. Instruments
assessing affective reactions to text genre and to reading particular texts were also
developed based on text-related comments solicited in the pilot study. Estimates of
internal consistency indicated that the subscales assessing enjoyment/interest, perceptions
of comprehensibility and of utility of texts and text genres were reliable measures of
these constructs. These instruments could contribute to future research on the effects of
affective reactions to texts and genres to reading comprehension outcomes.

In contrast to most L2 individual difference research that has involved students at
only one level of instruction, this study used a cross-sectional approach in order to
examine individual difference variables among learners ranging ﬁom beginning to
advanced levels. This approach should allow results to be generalized to more diverse
populations instead of to only one level of learner. This methodology also provided new
insights into L2 reading outcomes by text genre and in differences in affective reactions
to text genre by different learner levels.

This study also used a multi-componential approach in order to identify the
relative contributions of a variety of individual difference variables to L2 reading
comprehension. This approach resulted in a greater amount of L2 reading comprehension
variance (73%) accounted for than generally has been the case in single-focus studies
(Bernhardt, 2005). The multi-componential design, as well as the use of structural
equation modeling techniques, also allowed for both direct and indirect effects of

individual difference variables on reading comprehension variables to be identiﬁed.

170

Some methodological limitations of this study stem from sample size and the
requirements of structural equation modeling approaches. While 153 participants would
be considered large compared to samples in many other empirical L2 reading studies, this
number is on the low end of the sample size required for SEM techniques. This minimal
sample size appears to explain at least in part the failure of one of the affective variables
that was initially measured, self-perceived ability, to ﬁt into an admissible model of
individual diﬂerences. This may also have been explained by the fact that the
questionnaire items measuring self-perceived ability did not tap enough dimensions of
this construct to allow reliable assessment of this variable. Future studies using SEM
approaches should therefore include larger sample sizes and a greater number of observed

indicators to measure such latent variables.

5.5 Pedagogical implications

Although the present study was not intended to test particular pedagogical
interventions, its results nonetheless may have practical implications for L2 reading
instruction.

While this study provided evidence that a portion of students’ potential for L2
reading success is predetermined by their native language reading ability before they
even enter the classroom, ultimately it was found that their ability to read in another
language is much more dependent on their target language proﬁciency. This result should
be encouraging for language educators, who aim to help students develop L2 knowledge

and skills through language study.

171

The relationships revealed between metacognitive knowledge and L2 proﬁciency,
and in turn L2 reading outcomes, point to potential linguistic beneﬁts of helping students
develop knowledge of texts and reading strategies. This knowledge can be fostered
through pre-reading activities that focus on metacognitive awareness as well as explicit
metacognitive strategy training (Carrell, 1989; Carrell, 1998; Tang, 1992). The fact that
participants reported knowledge of less than half the reading strategies identiﬁed in the
metacognition questionnaire also suggests that a need exists for strategy awareness
training at the college level.

The present study also conﬁrmed the important facilitative role of target language
exposure in L2 learning and reading. Language educators should seek to exploit these
frequency effects by maximizing students’ opportunities for target language contact both
in and out of the classroom. Extensive reading programs, as described by Bamford and
Day (2004), and book ﬂood programs, as described in section 2.1.2.2, have been shown
to be effective means of increasing learners’ exposure to written language within the L2
curriculum. The fact that students of all levels reported spending so little time reading in
French (10 minutes per week for beginning and intermediate students, 45 minutes per
week for advanced) indicates a need for additional L2 print exposure among this
population. Providing incentives for students to seek other types of target language
experiences outside the classroom also seems key to improving both language and
reading performance.

The counterintuitive ﬁnding of a negative relationship between L2 study habits
and language outcomes has important implications for language students and educators.

First, students should be made aware that increasing study time may be less effective than

172

increasing study quality. It is also relatively common for language instructors to
recommend prescribed amounts of study time to their students, typically one to two hours
for every hour spent in class (Monmouth College; Seattle language Academy, University
of Wollongong). This type of “one size ﬁts all” prescription is clearly inadequate. This is
not to suggest, of course, that students be discouraged from spending time studying.
Rather, they should be encouraged to make the most productive possible use of study
time, and should receive instruction in the language study skills necessary to achieve this
goal. Haggstrom (1993) and McDevitt (1997) provide recommendations on effective
study skills and study-skill training for language students.

The role of individual affective differences in L2 learning and reading also should
not be ignored in the classroom. The present study shows that motivation plays a
substantial role in shaping behaviors that subsequently inﬂuence reading comprehension.
Instructors should attempt to create motivating reading environments in order to enhance
students’ reading performance. The inclusion of interesting, relevant and engaging
reading materials in the curriculum is an important component of such an environment.
Because pleasing all readers all of the time is a near-impossible task, educators should
consider introducing reading activities that take student preferences and choice into
account, such as free reading or extensive reading, in order to foster student motivation.

Since the majority of this study’s participants indicated they were studying French
to meet a degree requirement other than a French major or minor, it would be expected
that their primary motivation for language study would be extrinsic, and that high levels
of amotivation would be reported. Yet overall, participants reported similar levels of

intrinsic and extrinsic motivation for L2 study, as well as low levels of amotivation. This

173

motivational proﬁle of students who otherwise may not have chosen to study a language
should be encouraging to educators at institutions with language requirements.
Classifying students who are meeting a language requirement as “unmotivated” may
unwittingly discourage otherwise motivated learners from effort and subsequent
achievement, if not future foreign language study.

While similar motivational patterns were observed between motivation for L2
study and L2 reading, it should be noted that overall students were less motivated to read
in French than to study French. Making such students aware of the real-life purposes and
beneﬁts of reading in another language (developing linguistic and cultural competency,
developing critical thinking skills through text interpretation, and gaining insights into the
human condition, among others) and providing them with incentives to read within the
language curriculum may help foster their reading motivation.

Although anxiety did not have any signiﬁcant effects on language proﬁciency in
this study, it had the undesirable result of pushing students to study more without any
corresponding gains in French proﬁciency. Language instructors and learners alike
should be aware of this anxiety-related phenomenon, and instructors should promote
more productive uses of L2 study time, especially among anxious students.

The superior reading comprehension outcomes for narrative texts in the present
study should help refute beliefs that this genre is the most problematic for language
students, and provide support for including narratives in L2 reading curricula. This study
also provides evidence to challenge the claim that narrative, literary texts are
inappropriate for beginning-level learners. Although beginners were less able to take

advantage of the genre-level features that promote narrative comprehension than more

174

advanced learners, they nonetheless understood more ﬁ'om the narrative texts than from
the expository passages. Furthermore, L2 proﬁciency was a less important predictor of
comprehension of the narrative genre than the expository genre. These ﬁndings provide
support for including narrative reading activities at all levels of language instruction.
They further suggest that lower-division students may require additional instructional
support to help them exploit genre-level cues when reading both genres. Grabe and
Stoller (2002) provide a range of suggested activities to call attention to such text
features, including pre—reading activities focusing on genre, creating visual
representations of text structure, and teaching signal words common to the genre.
Readers’ affective reactions to each genre also have implications for L2 reading
curriculum and instruction. Mori’s 2004 study provides evidence that enjoyment and
interest in reading materials affects the amount that students will read; lack of interest and
enjoyment motivate students to read less. Selecting reading materials for the L2
classroom ﬁom genres and texts that engage students is therefore an important ﬁrst step
in encouraging them to read. In the present study, students of all levels clearly preferred
the narrative genre as well as the two experimental texts that represented this genre. This
provides an additional, compelling reason to include narrative texts in the L2 reading

curriculum at all levels of instruction.

5.6 Limitations

The present study is subject to several limitations. First, it cannot be ruled out that

the weaker recall comprehension scores for Expository Text 2 were inﬂuenced by the fact

175

that it contained a greater number of semantic propositions than the other three texts,
imposing an additional cognitive load that may have hindered text recall.

As in any experimental study, there are also limitations in generalizability to more
naturalistic settings. For example, the reading and recall tasks in the present study likely
do not represent classroom-based reading, where comprehension may be aided by
activities that connect the text with learners’ background knowledge or by text glosses.
Indeed, the relatively low percentage of propositions that participants recalled in both the
pilot and present studies may reﬂect this limitation. In this experiment, texts were also
chosen for their lack of unfamiliar topics or culture-speciﬁc references. In the L2
classroom, however, texts are often chosen for instructional purposes precisely because
they contain new information or unfamiliar cultural scenarios.

It is also important to note that the results related to motivation and affective
reactions to genre may have limited generalizability from this foreign language context to
second language settings. Motivational proﬁles for students studying in a target language
enviromnent likely differ from those studying another language in their home country.
Students who are immersed in the target culture are expected to have different reasons for
reading each genre of text than those who have little need to read in the L2 outside the
classroom, and their affective reactions to each genre would therefore be likely to differ.

Finally, although the present study accounted for a hefty amount of reading
comprehension variance (73 %), it leaves a substantial amount unexplained. This
additional variance may be explained by other cognitive and non-cognitive factors that

were not measured in the present study (see section 5.5 below).

176

5.5 Directions for future research

The present study opens several possible avenues for future inquiry. Concerning
individual differences, including additional variables in the model identiﬁed here may
account for remaining unexplained variance in L2 reading comprehension. These might
include cognitive factors such as working memory and intelligence, and non-cognitive
factors such as personality, willingness to communicate, and creativity (Domyei, 2005).
In light of evidence of relationships between affective reactions to reading and reading
outcomes (Brantrneier, 2005; 2006), future studies may also wish to examine the
predictive role of these variables in a model of reading comprehension. Longitudinal
studies of individual differences in reading comprehension may also be able to identify
how these variables contribute to changes in reading skills over time.

Future text genre research may wish to investigate reading outcomes for other
commonly read text genres or subgenres (for example, poems, plays, argumentative or
procedural texts). The predictive role of affective reactions to genre and text in L2
reading comprehension should also be explored, as well as the text-based factors
identiﬁed in participant comments, including personal engagement with the topic, text

organization, salience of text details.

5.6 Conclusion
L2 reading remains a complex, multidimensional process where both reader and
text-based variables interact. By providing additional insights into the relationships
between a variety of these reader-based differences and into the effects of the text-based

variable of genre, it is hoped that this study’s results contribute to the understanding of

177

the L2 reading process. It is further hoped that the ﬁndings related to text genre will
encourage the early and frequent inclusion of narrative texts in addition to expository

texts in the L2 curriculum.

178

Appendix A
Texts and English Translations

Narrative Text 1: Les trois motocyclistes

Au cours d’un été nous campions au bord d’un lac canadien. La nuit était tombée,
nous avions diné. Nous étions neuf en tout : six adolescents, Jean-Pierre, moi, Dorothe'e
qui avait douze ans. J’avais sommeil. Je les ai laisse's autour du feu et je suis allée dans la
tente.

Pendant que je me préparais a me coucher j’ai entendu une pétarde formidable.
Nous campions dans le creux d’une grande dune de sable qui descendait jusqu’a l’eau. Je
suis sortie et j’ai vu un spectacle incroyable : trois puissantes motocyclettes qui
absorbaient la pente raide de la dune dans des geysers de sable et un cataclysme de bruit.
La panique m’a prise. Je croyais que c’était la police qui venait faire éteindre notre feu,
ou Dieu sait quoi. Ce n’était pas la police mais trois jeunes hommes, dans les vingt-deux
ans, secs, habillés de cuir noir, avec de gros dessins colorés sur leurs blousons. Les
machines étaient magniﬁques, les ﬂarnmes faisaient briller leurs chromes par éclats, les
garcons étaient efﬁayants, dangereux, les yeux froids dans des visages bardés de casques
et de mentonniéres. J ’étais en retrait, je voyais la scene. Je m’attendais au pire. Jean-
Pierre avait fait un pas vers eux.

« Hello, good evening. »

Pas de réponse. Ils sont venus pres du feu. Tout le monde était debout. Cela a duré
un moment. Puis les enfants ont commencé a s’asseoir. Les trois motocyclistes aussi.

Gregoire a pris son banjo, Alain sa guitare. Ils se sont mis a gratter. Les trois

179

motocyclistes ont souri. On a passe des oranges. Alors a suivi une des soirées les plus
intéressantes que j’aie vécues ces dernieres années.
Narrative Text 1 Translation: The Three Motorcyclists

One summer we were camping on the shores of a Canadian lake. Night had fallen,
we had eaten dinner. There were nine of us in all: six teenagers, Jean-Pierre, me, and
Dorothée who was twelve. I was sleepy. I left them around the ﬁre and I went into the
tent.

While I was getting ready for bed I heard an enormous backﬁre. We were
camping at the top crevice of a big sand dune that sloped down to the water. I came out
and saw an unbelievable spectacle: three powerful motorcycles that were absorbing the
steep slope of the dune in geysers of sand and a cataclysm of noise. Panic seized me. I
thought it was the police who had come to make us extinguish our ﬁre, or God knows
what. It wasn’t the police but three young men, about twenty-two years old, dressed in
black leather, with huge designs colored on their jackets. The machines were
magniﬁcent, the ﬂames made their chrome shine in ﬂashes, the boys were ﬁightening,
dangerous, cold eyes in faces clad in helmets and chinstraps. I was in retreat, I was
watching the scene. I was expecting the worst. J can-Pierre stepped towards them.

“Hello, good evening.”

No answer. They came close to the ﬁre. Everyone was standing up. That lasted
for a moment. Then the children started to sit down. The three motorcyclists, too.
Gregoire picked up his banjo, Alain, his guitar. He started to strum. The three
motorcyclists smiled. We passed around oranges. And so followed one of the most

interesting nights I’ve experienced in these last few years.

180

Narrative Text 2: L 'aeufde Pa‘ques

Il s’appelait Siméon Farfelu. Sa femme s’appelait Nadine.

Ils se sont connus il y a dix-sept ans, exactement le 12 mai 1932. Le jour de
Paques, ils s’étaient trouvés sur un sentier dans les Bois de Boulogne. Il lui avait
demandé de lui tourner 1e dos et s’en était aller cacher un (euf en chocolat dans un
buisson. Elle l’avait cherché, trouvé et avait mordu 1e chocolat. A l’intérieur, un morceau
de papier sur lequel était écrit « je t’aime ». Elle s’était jetée a son cou et l’avait couvert
de baisers.

Depuis ce temps, chaque année, le jour de Péiques, Simeon Farfelu cachait un ocuf
dans la maison ou dans le petit jardin. Nadine, des le réveil, se mettait a chercher partout
et Simeon la suivait avec un regard amuse.

Quand elle avait enﬁn trouvé, comme la premiere fois, elle mordait hardiment
dans le chocolat pour savoir ce que l’oruf cachait. Le plus souvent, elle y avait trouvé un
bijou. A plusieurs occasions, elle avait trouvé des bouts de papier sur lesquels elle lisait :
« Bon pour une robe » ou « Bon pour une surprise. »

L’année précédant celle dont il est particulierement question dans cette histoire,
Nadine Farfelu avait peut-étre été un peu décue. Simeon, tres occupé par ses affaires, ne
s’était pas donné la peine de chercher une cachette diﬁicile. Nadine avait trouvé l’ceuf
apres cinq minutes dans la casserole or‘r elle devait faire bouillir le lait.

Elle avait mordu et elle avait vu. . .un billet de cent francs.

Narrative Text 2 Translation: The Easter Egg
His name was Simon Scatterbrained. His wife’s name was Nadine. They met

seventeen years ago, exactly on May 12, 1932. On Easter, they found themselves on a

181

path in the Bois de Boulogne. He asked her to turn her back and went to hide a chocolate
egg in a bush. She looked for it, found it, and bit into the chocolate. Inside was a piece of
paper on which was written “I love you.” She threw her arms around his neck and
covered it with kisses.

Since that time, every year on Easter, Simon Scatterbrained would hide an egg in
the house or in the little yard. As soon as she got up, Nadine would start looking for it
everywhere and Simon would follow her with an amused look.

When she had ﬁnally found it, like the ﬁrst time, she bit hard into the chocolate to
ﬁnd out what the egg was hiding. Most often, she found jewelry. Several times, she found
pieces of paper on which she read “Good for a dress” or “Good for a surprise”.

The year before the one that is the subject of this story, Nadine Scatterbrained had
been a little disappointed. Simon, very busy with his work, hadn’t gone to a lot of trouble
to ﬁnd a difficult hiding place. Nadine had found the egg in the pan she used to boil milk.

She bit into it and she saw. . .a one hundred ﬁanc note.

Expository Text I : Le café dans la vie des étudiants

Dans la vie sociale des étudiants francais, le café joue un role important. Le
temps important que tant d’étudiants passent au café, peut appara’r‘tre comme une
« perte de temps ». Il répond, a vrai dire, en grande partie, a tout un ensemble de besoins
que l’étudiant cherche a satisfaire, en dehors du temps consacré au travail. Aller au café,
ce n’est pas uniquement chercher un reﬁrge entre deux cours quand la bibliotheque est
pleine, ou éviter un déplacement lorsqu’on habite une chambre éloignée de la Faculté, ou
trop exigué, ou mal chauffée, ou encore coﬁteuse a chauffer. Il semble que le café

réponde avant tout a un besoin de contacts que les structures universitaires n’ont pas

182

satisfait jusqu’a maintenant. Le café est bien souvent 1e seul lieu or‘r fuir la solitude, ou
nouer des connaissances et faire des amis.

Les plus esseulés vont toujours au meme cafe’. Ils se sentent un peu chez eux, ils
prennent plaisir a retrouver les memes habitue's, a étre traités amicalement par les
garcons. A l’ége ou les relations arnicales ou arnoureuses commencent a prendre plus
d’irnportance que les relations familiales, les étudiants apprécient tout ce que peut leur
apporter l’ambiance d’un café farnilier. L’amitié, la camaraderie, ne peuvent pas se
trouver dans la famille. Au cafe, on n’est pas forcé de répondre a tous les appels. Ca
détend, on est entre jeunes, c’est mieux que d’étre en famille.

Au cafe, l’étudiant devient un étre sociable. Il goﬁte l’imprévu des rencontres,
qu’elles soient de peu d’irnportance ou marquantes.

Expository Text I Translation: The Café in Student Life

The café plays an important role in the social life of French students. The
important time that so many students spend in cafés may seem like a “waste of time”. In
truth, it responds in an important way to a variety of needs that the student is looking to
satisfy outside of time devoted to study. Going to a café is not only looking for a refuge
between classes when the library is full or avoiding the need to go back a room that’s far
away from the university, or cramped, or poorly heated, or expensive to heat. It seems
that the café responds above all to a need for contact that the university hasn’t satisﬁed
until now. The café is often the only place to ﬂee solitude, to strike up acquaintances and
make friends.

The loneliest ones always go to the same café. They feel at home, they take

pleasure in seeing the same regulars there, in being treated in a ﬁiendly way by the

183

waiters. At the age where friendships and romances are starting to become more
important than relationships with family, students appreciate everything that the
environment at a familiar cafe offers. Friendship and camaraderie can’t be found with
your family. At a cafe, you don’t always have to answer the phone. It’s relaxing, you’re
with other young people, it’s better than being with your family.

At the café, the student becomes a sociable being. He experiences random
encounters, be they of little importance or signiﬁcant.

Expository Text 2 (Pilot Study): La vague bouddhiste

Comme chaque matin depuis bientot trois ans, dans son studio parisien of: i1 vit
seul, Francois se leve a1’aube. Apres une douche rapide, il enﬁle un pantalon du jogging
pour faire zazen, la meditation assise silencieuse du bouddhisme zen. Il s’assied en
tailleur sur un zaﬁt, un coussin rond et épais face an mur. Le dos parfaitement droit, les
deux mains jointes en contact avec l’abdomen, les yeux mi-clos, il porte son attention sur
sa respiration, longue et profonde.

Francois fait partie des dizaines de milliers de Francais venus au bouddhisme par
la meditation. La plupart de ces nouveaux adeptes ont, comme lui, entre 30 et 45 ans, sont
des citadins célibataires ou vivent en concubinage. Comme Francois, ils ont fait des
études supérieures. Ils sont, eux aussi, cadres ou exercent une profession libérale. Depuis
peu, 1e bouddhisme commence a s’étendre a toutes les couches de la population. 11
devient l’un des phénoménes spirituels les plus étonnants de notre société, en cette ﬁn de
XXe siecle.

A la question « Quelle est la religion qui a votre preference ? » 2 millions de

Francais répondent : « Le bouddhisme. » Plus surprenant encore, la religion du Bouddha

184

vient au troisieme rang, loin derriére 1e catholicisme (68%), mais presque a l’égalite' avec
le protestantisme (6%). Dans un classement or‘r l’on peut citer trois religions par ordre
d’inte’rét décroissant, le bouddhisme obtient les suffrages de 6 millions de Francais. C’est
aussi la premiere fois que le bouddhisme apparait dans un sondage d’opinion sur les
croyances des F rancais.

Expository Text 2 Translation (Pilot Study): The Buddhist Wave

Like every morning for nearly three years, in his Parisian studio apartment where
he lives alone, Francois gets up at dawn. After a quick shower, he slips on some jogging
pants to do zazen, the silent seated mediation of Zen Buddhism. He sits cross-legged
facing the wall on a zafu, a round, thick cushion. His back perfectly straight, his two
hands joined over his abdomen, his eyes half closed, he focuses his attention on his
breathing, long and deep.

Francois is part of tens of thousands of French people who came to Buddhism
through meditation. Most of these new followers, like him, are between 30 and 45 years
old, single city-dwellers or living with a partner. Like Francois, they are college-
educated. Also, they are in management or are professional people. Recently, Buddhism
has started to spread to all levels of the population. It is becoming one of the most
stunning spiritual phenomena of our society at the end of the 20th century.

In response to the question “Which religion do you prefer?” 2 million French
people respond “Buddhism”. Even more surprising, the religion of Buddha comes in third
place, well behind Catholicism (68%), but almost equal to Protestantism (6%). In a

ranking of the three religions by decreasing order of interest, Buddhism receives the votes

185

of 6 million French people. This is also the ﬁrst time that Buddhism has appeared in a
public opinion poll about the beliefs of French people.
Expository Text 2 (Present Study): Sport: l’homrne sans limites

Plus vite, plus haut, plus fort. Depuis la ﬁn du XIXe siecle et la renovation des
Jeux olympiques par le baron Pierre de Couvertin, une loi régit 1e sport modeme :
toujours plus, toujours mieux. Grignoter des secondes, des centimetres. Le Moyen Age a
inventé l’horloge mécanique et ainsi discipline le temps. Aujourd’hui, c’est 1e syndrome
du record — i1 occupera encore tous les esprits pendant les 22° Jeux olympiques. Plus de
15 000 sportifs vont tenter de dépasser leurs limites. Quelques autres essaieront d’aller
au-dela des « limites naturelles ». Mais jusqu’ou iront les champions ? D’ailleurs, ces
limites, existent-elles vraiment ?

Le stade, la piscine ou le gymnase sont devenus de centres de recherche. Les
sportifs de haut niveau se gavent a la haute technologie. Avec un masque d’oxygene sur
le visage ou un casque-came’ra sur la téte. Et, sur le bord des pistes ou des bassins, on
teste. On mesure. Tout. Toujours. Le potentiel métabolique. La consommation maximale
d’oxygene. La vitesse d’impulsion. L’arnplitude de la foulée. Des escrimeurs répetent
leurs coups favoris sur l’Arrnivex, un panneau métallique hexagonal portant six petites
cibles circulaires munies d’une ampoule en leur centre : des que l’une s’allume, l’athlete
doit la toucher. Et un ordinateur calcule son temps de reaction.

L’évolution du materiel et de la technique est aussi directement liée a la haute
technologie. Aujourd’hui, 1e moindre geste sportif est décortiqué, decompose, analyse par
les ordinateurs des scientiﬁques. Ainsi, lorsqu’un nouveau geste, une nouvelle technique

apparaissent sur le stade, dans le gymnase ou la piscine, ils ne doivent rien au hasard.

186

Expository Text 2 Translation (Present Study): Sports: Man without limits

Faster, higher, stronger. Since the end of the 19th century and the renovation of
the Olympic games by Baron Pierre de Couvertin, one law governs modern sports:
always more, always better. Gnawing off seconds, centimeters. The Middle Ages
invented the mechanical clock and thus disciplined time. Today, it’s the syndrome of the
record - it will be on everyone’s mind during the 22'"l Olympic games. More than 15,000
athletes will attempt to surpass their limits. Some others will try to go beyond “natural
limits.” But how far will the champions go? Besides, do these limits really exist?

The stadium, the pool and the gym have become research centers. Athletes at the
high levels will stuff themselves with high technology. With an oxygen mask on their
faces and a helmet-camera on their heads. And, on the sidelines of the tracks or the pools,
others test. They measure. Everything. Always. Metabolic potential. Takeoff speed. The
length of the stride. Fencers repeat their favorite blows on Armivex, a hexagonal, metallic
shield carrying six small circular targets equipped with a bulb in their center: as soon as
one of them lights up, the athlete has to touch it. And a computer calculates his/her
reaction time.

The evolution of equipment and technique is also directly tied to high technology.
Today, the smallest sports movement is dissected, broken down, analyzed by scientists’
computers. Thus, when a new movement, a new technique appears in the stadium, in the

gym or in the pool, they leave nothing to chance.

187

rrlPDhWVHWHMI.

.Wththmvhpllri

Appendix B
Propositional Analyses of Texts

Narrative Text 1: The Three Motorcyclists

P1 2 camp [we]

P2 1 Canadian [lake]

P3 2 on-shores [P2]

P4 1 CON (one summer) {Pl}
P5 2 fall/is [night]

P6 2 eat [we, dinner]

P7 2 is [nine]

P8 1 CON (in all) [P7]

P9 2 is [6 teenagers]

P10 2 is [Jean-Pierre]

P11 2 is [me]

P12 2 is [Dorothee, 12]

P13 1 CON (and) [P11,P12]
P14 2 is [me, sleepy]

P15 2 leave [me, them]

P16 2 around-ﬁre [P15]

P17 2 go [me, tent]

P18 1 CON (and) [P15, P17]
P19 2 get ready [me, bed]
P20 1 CON (while) [P19]
P21 2 hear [me, P22]

P22 1 enormous/loud [noise]
P23 2 camp [we, P24]

P24 2 at top-crevice [P23]
P25 2 of big [dune]

P26 2 slope [dune]

P27 2 down to-water [P25]
P28 2 come out [me]

P29 2 see [me, P30]

P30 1 unbelievable [spectacle]
P31 1 CON (and) [P28, P29]
P32 2 absorb [P33, P35]

P33 1 three [motorcycles]
P34 1 powerful [motorcycles]
P35 1 steep [slope/dune]

P36 2 in-sand geysers [P32]
P37 2 in-noise cataclysm [P32]
P38 1 CON (and) [P36, P37]

188

 

l'lT-lrhlull-llllll,

P39
P40
P41
P42
P43
P44
P45
P46
P47
P48
P49
P50
P5 1
P52
P53
P54
P55
P56
P57
P58
P59
P60
P61
P62
P63
P64
P65
P66
P67
P68
P69
P70
P71
P72
P73
P74
P75
P76
P77
P78
P79
P80
P81

NHNNNNNNNNNNNNNNNN—‘NNNNNNNHNNHN—‘NNHNHNNNNNN

seize [partie, me]

think [me, P41]

is [police]

come [police]

make/cause [police, us]
extinguish/put out [us, ﬁre]
CON (in order to) [P42, P43]
know [God, what]

CON (or) [P42, P46]

is not [police]

is [men, three, young]
CON (but) [P49]

aged/is [men, 22]

CON (about) [P51]

is [men, dressed]

in-black leather [P54]
color [designs]

on-jackets [P55]

is [machines/motorcycles, magniﬁcent]
make/cause [ﬂames, P59]
shine [chrome]

in-ﬂashes [P59]

is [boys, ﬁ'ightening]

is [boys, dangerous]

CON (and) [P61, P62]

is [eyes, cold]

in-faces [P64]

is [faces, clad]
with-helmets [P66]
with-chinstraps [P67]

is [me, in retreat]

watch [me, scene]

expect [me, the worst]
step/walk [J can-Pierre]
towards-men [P72]

say [Jean-Pierre, "Hello, good evening"]
not answer [men] / is not [response/answer]
come [men]

close to-ﬁre [P76]

is [everyone, standing]

last [standing]

CON (for a moment) [P79]
start [children, sit down]

189

Wm“?

P82
P83
P84
P85
P86
P87
P88
P89
P90
P91
P92
P93
P94

r-‘NI-‘t-‘NNNNNNr—N—

163

CON (then) [P81]

sit [men/motorcyclists]

CON (too) [P83]

pick up [Gregoire, banjo]
pick up [Alain, guitar]

start [Alain/Gregoire, strum/play]
smile [men/motorcyclists]
pass around [we, oranges]
follow/happen [P91]
interesting [night]

CON [one of the most) [P91]
experience [me]

CON (last few years) [P93]

Narrative Text 2: The Easter Egg

P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
Pl 1
P12
P13
P14
P15
P16
P17
P18
P19
P20
P21
P22
P23
P24
P25
P26
P27
P28

NNNNNNHNNN—‘NHNNNNN—‘NNHr-rr—r—NNN

is [name, Simon Farfelu/Scatterbrained]
is [wife, name, Nadine]
meet [Simon Nadine]

CON (17 years) [P3]

CON (ago) [P3, P4]

CON (May 12, 1932) [P3]
CON (exactly) [P6]

ﬁnd/be [Simon Nadine]
on-path [P8]

CON (Easter) [P8]

in-Bois de Boulogne [P8]
ask [Simon Nadine, P13]
turn [Nadine, back]

go [Simon]

hide [Simon, P16]

chocolate [egg]

in-bush [P15]

CON (and) [Pl4, P15]

look [Nadine, egg]

ﬁnd [Nadine, egg]

bite [Nadine, chocolate/egg]
CON (and) (PZO, P21]

is [paper, written, "I love you"] / says [paper, "I love you"]
inside-egg [P23]

throw [Nadine, arms, Simon]
around-neck [P25]

cover [Nadine, neck, Simon]
with-kisses [P27]

190

 

P29
P30
P3 1
P32
P3 3
P34
P3 5
P36
P3 7
P3 8
P3 9
P40
P4 1
P42
P43
P44
P45
P46
P47
P48
P49
P50
P5 1
P52
P53
P54
P5 5
P56
P57
P5 8
P59
P60
P61
P62
P63
P64
P65
P66
P67
P68
P69
P70
P71
P72

r—Au—sNNNNu—Iu—INu—Nu—tNNr—INu—tu—tNr—Iu—INNu—tr—‘r—twNu—NNr—No—INv—‘r—u—‘r—INNNv—sr—I

CON (and) [P25, P27]

CON (since that time/then) [P31]
hide [Simon, egg]

in-house [P31]
in-garden/yard [P3 1]

little [P33]

CON (or) [P3 1, P33]

CON (every year) [P31]
CON (on Easter) [P31]

look [N adine, egg]

CON (everywhere) [P3 8]
wake up [Nadine]

CON (as soon as) [P40]
follow [Simon, Nadine]
with-amused [look] [P42]
CON (and) [P38, P42]

ﬁnd [Nadine, egg]

bite [Nadine,chocolate/egg]
CON (hard) [P46]

CON (when) [P45]

CON (ﬁnally) [P45]

ﬁnd out [Nadine, P49]
inside/hiding inside [egg]
CON (like) [P21/ﬁrst time]
CON (in order to) [P46, P50]
ﬁnd [Nadine, jewelry]

CON (often) [P54]

CON (most) [P55]

ﬁnd [Nadine, paper]

CON (several times) [P57]
read [Nadine]/said [paper] ["good for a dress"]
read [Nadine]/said [paper] ["good for a surprise"]
CON (or) [PS9, P60]

is [year, subject, story]

CON (before) [P62]

is [Nadine, P63]

little [disappointed]

CON (maybe/perhaps) [P64]
is [Simon, busy]

with-work [P67]

not go to trouble [Simon]
ﬁnd [Simon, P69]

difficult [hiding place]

CON (in order to) [P70]

191

 

P73
P74
P75
P76
P77
P78
P79
P80
P8 1
P82

Or—NNNv—r—NNNN

p—a
1.).)

ﬁnd [Nadine, egg]
in-pan/casserole dish [P73]

use [Nadine, pan]

boil [Nadine, milk]

CON (in order to) [P75]

CON (after/in ﬁve minutes) [P73]
bite [Nadine, chocolate/egg]
see/ﬁnd [Nadine, P79]

100 franc [note/bill]

CON (and) [P79, P80]

Expository Text I: The Café in Student Life

P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
Pl 1
P12
P13
P14
P15
P16
P17
P18
P19
P20
P21
P22
P23
P24
P25
P26
P27
P28
P29
P30
P31

HN—‘NHNNNNNI—‘NHNNNv-‘NNNNNNNr-‘NNNH—‘N

plays [café, P2]

important/large [role/part]
French [students]

in-social lives [P3]

spend [P6, P7]

so many [students]
important/large amount of [time]
in-cafes [P5]

seem like [P7, waste of time]
respond [P7, P11]
various/variety of [needs]
want/look to [student, satisfy, P11]
devote [student, study time/ studies]
outside [P13]

CON (in truth/opposition) [P9, P10]
go [student, to-café]

look [student, refuge/escape]
between-classes [Pl7]

CON (not only) [P16, P17]

is [library, full]

CON (when) [P17, P20]

avoid [student, need to go back]
to-room [P22]

is [room far away]
ﬁom-university [P24]

is [room, cramped/small]

CON (or) [P24, P26]

is [room, poorly heated]

CON (or) [P26, P28]

is [room, expensive to heat]
CON (or) [P28, P30]

192

 

P3 2
P3 3
P34
P3 5
P3 6
P3 7
P3 8
P39
P40
P4 1
P42
P43
P44
P45
P46
P47
P48
P49
P50
P5 1
P52
P53
P54
P55
P56
P57
P58
P59
P60
P61
P62
P63
P64
P65
P66
P67
P68
P69
P70
P71
P72
P73
P74
P75

NNNNNﬂ—‘NNNHNN-‘HNNHNH—‘NN—‘NNNNr—rF‘HN—‘NN—N—‘N—‘N—‘NN

seems [café, respond, P33] / respond [café, P33]
for-contact [need]

CON (above all) [P32]

not satisfy [university, P33]

CON (until now) [P35]

is [café, place]

CON (often) [P37]

ﬂee [student, solitude]

CON (only) [P37, P38]

strike up/make [student, acquaintainces]
make [student, friends]

CON (and) [P4], P42]

go [P45, to-P46]

loneliest [students]

same [café]

CON (always) [P44]

feel [loneliest students, at-home]

take pleasure [students, see, regulars]
take pleasure [students, P51]

treat [waiters, students, friendly way]
CON (and) [P49, P50]

become [ﬁiendships, more important]
become [romances, more important]
CON (and) [PS3, PS4]

CON (at an age where]) [PS3, PS4]

is [relationships, with-family, important]
CON (than) [PS3, PS7]

appreciate [students, everything]
offers/provides [environment, of-P61]
familiar [café]

CON (that) [PS9, P60]

cannot [student, ﬁnd, friendship]

cannot [student, ﬁnd, camaraderie]
CON (and) [P63, P64]

with-family [P63, P64]
not have to/not be obligated to [one/student, P68]
answer [one/student, calls/demands/phone]
CON (always) [P67]
CON (at cafe)
is [café, relaxing]
is [one/student, P73]
with-other young [people]
is [being at cafe', better]
be [student, with-family]

193

 

P76
P77
P78
P79
P80
P8 1
P82
P83
P84

OOr-‘NN—‘NNO-‘N—I

u—c
b)

CON (than) [P74, P75]
become [student, P78]
sociable [being]

at-café [P77]

experiences [student, P81]
random [encounters]

is [encounters, unimportant]
is [encounters, signiﬁcant]
CON (or) [P82, P83]

Expository Text 2 (Pilot Study): The Buddhist Wave

P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19
P20
P21

P22

P23
P24
P25
P26
P27
P28
P29
P30
P3 1
P32

1

NNN—N—I—NNNN-~NN~N-N~NH~NNN~HN

every [morning]

three years [P1]

CON (nearly) [P2]

CON (like) [P1, P2]

get up [Francois]

at-dawn [P5]

in-P8 [P5]

studio [apartment]

Parisian [P8]

live [F rancois, P8]

CON (alone) [P10]

shower [Francois]

quick [P12]

CON (after) [PlZ]
slips/puts on [Francois, P16]
jogging {Pants}

do [F rancois, zazen]

is [zazen, P19]

silent [meditation]

seated [meditation]

zen buddhist [meditation]
sit [Francois, cross-legged]
face [F rancois, wall]
on-zafu [P24]

is [zaﬁg cushion]

round [cushion]

thick [cushion]

is [back, Francois, straight]
CON (perfectly) [P28]

is (hands, Francois, joined/folded]
over-abdomen [P3 0]

is [eyes, Francois, half-closed]

194

 

P33
P34
P35
P36
P37
P38
P39
P40
P41
P42
P43
P44
P45
P46
P47
P48
P49
P50

P51
P52

P53
P54
P55
P56

P57
P58
P59
P60
P61

P62
P63
P64
P65
P66
P67
P68
P69
P70
P71
P72
P73
P74
P75

NNNNNNNN—NN-‘I—‘NNHN—‘N—‘HNNr-‘N—HNNNNNH—‘NNNNNHHHN

focus on/pay attention to [F rancois, breathing]
long [breathing]

deep [breathing]

CON (and) [P33, P34]

is [F rancois, part/one]

of-tens of thousands [P3 9]
of-French people [P40]

come [P39, to Buddhism]
through-meditation [P40]

most [P43]

new [followers]

aged [P43, 30-45 years]

is [P43, single]

is [P43, city-dweller] / live [P43, in city]
live [P43]

with-partner [P47]

CON (or) [P45, P48]

CON (like) [Francois, P44]

is [followers, college-educated]
CON (like) [Francois, P51]

is [followers, management]

is [followers, professional]
CON (or) [PS3, PS4]

CON (also) [PS3]

start [Buddhism, spread]

CON (recently) [P57]

to-levels [P57]

all [PS9]

of-population [P59]

become [Buddhism, P63, P64]
stunning [phenomena]

spiritual [phenomena]

most [P63]

of-society [P63]

CON (at end of 20th century) [P62]
respond [P69, Buddhism]

2 million [French people]
to-question [P68]

prefer [French people, religion]
ranks/comes in [Buddhism/religion of Buddha]
in-3rd place [P72]
behind-Catholicism [P72]

68% [Catholicism]

195

 

P76
P77
P78
P79
P80
P81
P82
P83
P84
P85
P86
P87
P88
P89
P90
P91
P92

WNNF‘NN—‘NNr—‘NNNNH—‘N-H

—i
A

CON (even more surprising) [P72]
is [Buddhism, equal]

CON (almost) [P80]

CON (but) [P77]

to-Protestantism [P77]

6% [Protestantism]

receive [Buddhism, P83, French people]
6 million [votes]

in-ranking [P82]

of-three religions [P84]
by-decreasing order [P84]
of-interest [86]

appear [Buddhism]

in-public opinion poll [P89]

CON (ﬁrst time) [P89]
about-beliefs [P90]

of-French people [P90]

Expository Text 2 (Present Study): Sports: Man without Limits

P 1
P2
P3
P4

P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19
P20

P21
P22

faster [athletes]
higher [athletes]
stronger [athletes]
renovate [P5, P6]

Baron [Pierre de Couvertin]
Olympic [games]

CON (since) [P4, P8]

CON (end of 19th century) [P4]
govern [P10, P11]

one [law]

modern [sports]

more {performance}

better [performance]

CON (always) [PlZ, P13]
gnaw off [athletes seconds]
gnaw of)” [ athletes, centimeters]
invent [Middle Ages, P18]
mechanical [clock]

CON (and) [P17, P20]
discipline [Middle Ages, time]
CON (thus) [P17, P20]

CON (today) [P23]

196

 

P23

P25
P26

P28
P29

P30
P3 1
P32
P33

P34
P3 5
P36
P37
P3 8
P39
P40
P41
P42
P43
P44
P45
P46
P47
P48
P49
P50
P5 1
P52
P53
P54
P55
P56
P57
P58
P59

P60
P61

h-II—I—tI—tt—il-dt—It—ih-bh—It—INI—It—‘NNN—iN—i—tNt—lh—ANNl—‘l—INN—Hh—I—iI—I—il—tNr—I

record [syndrome]

is [P23]

CON (on) [P26]

mind [everyone]

CON (during) [P28, P29]
olympic [games]

22nd [P28]

15,000 [athletes]

more than [P30]
attempt [P30]

surpass [P30, limits]
some [P35]

other [athletes]

try [P35, P3 7]

go beyond [P3 8]
natural [limits]

CON (but) [P40, P41]
go [champions]

CON (how far) [P40]
CON (besides) [P43]
exist [limits]

CON (really) [P43]
become [stadium, P49]
become [pool, P49]
become [gym, P49]
CON (and) [P45, P46, P47]
research [centers]
stuff [athletes, self]
CON (at) [P50, P52]
high [levels]

CON (with) [P50, P52]
high [technology]
CON (with) [P56]
oxygen [mask]

CON (on) [P5 8]
their/athletes [faces]
CON (or) [PS6, P60]

helmet [camera]
CON (on) [P62]

197

 

P62
P63
P64
P65
P66
P67
P68
P69
P70
P7 1
P72
P73
P74
P75

P76
P77

P78
P79
P80
P8 1
P82

P83
P84
P85
P86
P87
P88
P89

P90
P91
P92
P93
P94
P95
P96
P97
P98

P100

P101

r—Ar—ov—Ir—tr—INu—tNr—tNNr—th—tNr—tr—su—sNr—Io—tr—Ir—nr—Nr—su—ia—au—tu—ta—tr—sNNu—sNN—u—iu—s

their/athletes [heads]
CON (and) [P63, P65]
CON (on) [P66, P67]
sidelines [tracks]
sidelines [pools]

CON (or) [P66, P67]
test [we/they/others]
measure [they/we/others]
test [everything]

test [always]

metabolic [potential]
maximal [consumption]
oxygen [consumption]
takeoff [speed]

stride [length]

repeat [fencers, P7 8]
favorite [blows]

CON (on) [P77, P80]

is [Arrnivex]

hexagonal [shield]
metallic [shield]

carry [P80, P84]

six [targets]

small [targets]

circular [targets]

is [targets, equipped]
CON (with) [P87, P89]
at—center [bulb, P87]
CON (as soon as) [P91]
light up [bulb/one]

must touch [athlete, bulb]
CON (and) [P94]
calculate [computer, P95]
reaction [time]

is [tied, P97, P98]
equipment [evolution]
technique [evolution]
CON (and) [P97, P98]
CON (also) [P96, P102]
CON (directly) [P96, P102]

198

 

P120
P121

P102 2 to-high technology [P96]
P103 1 CON (today) [PlO4]
P104 2 dissect [PlO9, P105]
P105 1 smallest [gesture]
P106 1 sports [gesture]
P107 2 break down [PlO9, P105]
P108 2 analyze [PlO9, P105]
P109 1 scientist [computers]
P110 1 CON (thus) [P112]
P1 11 l CON (when) [P112]
P112 2 appear [P113]
P1 13 1 new [gesture/movement]
P114 2 appear [P113]
P1 15 1 new [technique]
P116 2 in-stadium [P112,P115]
P117 2 in-gym[P112,PllS]
P118 2 in-pool [P112,P115]
P119 1 CON (or) [P1 16, P117, P118]
2
2

160

leave [they/ scientists, nothing]

to-chance [Pl20]

 

Note. Adapted ﬁ'om Pulido (2004). P=Proposition. CON=sentence connective (e.g.,
causal, condition, purpose, concession, conjunction, and temporal). l=proposition worth
1 point. 2=proposition worth 2 points. Italicized words represent overlap of arguments

not explicitly stated in the text.

199

Appendix C
Motivation for French Study Questionnaire, Pilot Study

WHY DO YOU STUDY FRENCH?

Using the scale below, indicate to what extent each of the following items currently
corresponds to one of the reasons you study French. Circle only one number per

question.

Does not Corresponds a Corresponds Corresponds a Corresponds

 

correspond at little moderately lot exactly
all
1 2 3 4 5 6 7
WHY DO YOU STUDY FRENCH?

1. Because studying French will help me get a high-paying job.

2. Because I experience pleasure and satisfaction while learning new things.

3. Because I think studying French will help me better prepare for the career I have
chosen.

4. For the intense feelings I experience when I am communicating with others.

5. Honestly, I don’t know; I really feel I am wasting my time studying French.

6. For the pleasure I experience while surpassing myself in my studies.

7. To prove to myself that I am capable of accomplishing something.

8. In order to obtain a more prestigious job later on.

9. For the pleasure I experience when I discover new things never seen before.

10. Because eventually it will enable me to enter the job market in a ﬁeld I like.
11. For the pleasure I experience when I read interesting authors.

12. I once had good reasons for studying French; however, now I wonder whether I

should continue.
13. For the pleasure that I experience when I am surpassing myself in one of my

personal accomplishments.
14. Because of the fact that when I succeed in my French studies I feel important.

15. Because I want to have “the good life” later on.
16. For the pleasure that I experience in broadening my knowledge about subjects

which appeal to me.

17. Because it will help me make a better choice regarding my career orientation.
18. For the pleasure I experience when I feel completely absorbed by what certain
authors have written.

19. I can’t see why I am studying French and frankly, I couldn’t care less.

20. For the satisfaction I feel when I am in the process of accomplishing difficult
academic activities.

21. To show myself that I am an intelligent person.

22. In order to have a better salary later on.
23. Because studying French allows me to learn about many things that interest me.

200

 

24. Because I believe that studying French will improve my competence as a worker.
25. For the “high” feeling that I experience while learning about various interesting

subjects.
26. I don’t know; I can’t understand why I am studying French.

27. Because it allows me to experience a personal satisfaction in my quest for

excellence in my studies.
28. Because I want to show myself that I can succeed in my studies.

201

Appendix D

Reading Activity in French, Pilot Study

Directions: We are interested in knowing about your reading activities in French and
ﬁnding out how often you do them. Circle the answers to some questions, and write the

answers to the others.

I. Questions about school reading in French
Directions: In this section, think about reading in French that you do for school and for
homework. CIRCLE ONLY ONE ANSWER PER QUESTION.

1) Did you read a French textbook for school last week?

No ...... 1

Yes...2

If yes, write in the speciﬁc topic(s) you read about:

 

2) How often do you read a French textbook for school?

Almost never. . .1 About once a

month. . .2

How many hours /
month?

About once a
week. . . 3

Almost every day...4

How many hours /
week?

How many hours / day?

3) Did you read a book of literature or ﬁction in French last week for school?

No ...... 1

Yes...2

If yes, write in the title, author, or the speciﬁc topic(s) you read about:

 

4) How often do you read a book of literature or ﬁction in French for school?

About once a
month. . .2

Almost never. . .1

How many hours /
month?

About once a
week. . . 3

Almost every day. . .4

How many hours /
week?

How many hours / day?

5) Did you read a nonﬁction book in French for school last week?

202

No ...... l Yes...2

If yes, write in the title, author, or the speciﬁc topic(s) you read about:

 

6) How often do you read a nonﬁction book in French for school?

Almost never... 1 About once a About once a Almost every day. . .4
month. . .2 week...3

How many hours / How many hours / How many hours / day?
month? week?

7) Did you read any materials other than books in French for school last week
(magazines, newspapers, journals, online materials)?

No ...... 1 Yes...2

If yes, write in the title, author, or the speciﬁc topic(s) you read about:

 

8) How often do you read materials other than books in French for school?

Almost never. . .1 About once a About once a Almost every day. . .4
month. . .2 week. . .3

How many hours / How many hours / How many hours / day?
month? week?

11. Questions about reading for your own enjoyment in French
Directions: In this section, think about reading in French that you read for your own
interest that are not assigned for school or homework. CIRCLE ONLY ONE ANSWER

PER QUESTTON.
9) Did you read a ﬁction book in French last week for your own interest?

No ...... 1 Yes...2

If yes, write in the title, author, or the speciﬁc topic(s) you read about:

 

203

 

   

10) How often do you read ﬁction books in French for your own interest?

About once a
month. . .2

Almost never. . .1

How many hours /
month?

About once a
week. . . 3

Almost every day. . .4

How many hours / How many hours / day?

week?

1 1) Did you read a nonﬁction book in French last week for your own interest?

No ...... 1

Yes...2

If yes, write in the title, author, or the speciﬁc topic(s) you read about:

 

12) How often do you read ﬁction books in French for your own interest?

About once a
month. . .2

Almost never... 1

How many hours /
month?

About once 3 Almost every day. . .4

week...3

How many hours / How many hours / day?

week?

13) Did you read a magazine or newspaper in French last week for your own interest?

No ...... l

Yes...2

If yes, write in the title, author, or the speciﬁc topic(s) you read about:

 

14) How often do you read magazines or newspapers in French for your own interest?

About once a
month. . .2

Almost never... 1

How many hours /
month?

About once 3 Almost every day. . .4
week. . .3

How many hours / How many hours / day?
week?

15) Did you read any online materials in French (web pages, online articles, blogs, etc.)

last week for your own interest?

No ...... 1

Yes. . .2

204

If yes, write in the title, author, or the speciﬁc topic(s) you read about:

 

16) How often do you read online materials in French for your own interest?

Ah'nost never. . .1 About once a About once a Almost every day. . .4
month. . .2 week. ..3

How many hours / How many hours / How many hours / day?
month? week?

 

17) Did you read any other materials in French last week for your own interest that were
not mentioned?

No ...... l Yes...2

If yes, describe what you read:

 

18) How often do you read other materials in French for your own interest?

Almost never. . .1 About once a About once a Almost every day. . .4
month. . .2 week. . .3

How many hours / How many hours / How many hours / day?
month? week?

 

205

Appendix E
Familiarity Questionnaire, Pilot Study

Using the scale below, indicate to what extent each of the following items is familiar to
you. Circle only one number per question.

 

 

Not at all familiar A little familiar Somewhat Very familiar Completely
familiar familiar
1 2 3 4 5 6 7

 

How familiar are you with the following topics or situations? Circle only one number
per question.

1) Registering for classes at a university
2) Being your own boss

3) Building a skyscraper

4) Dining out in France

5) Camping

6) Christmas traditions

7) Religion in France

8) Life in a big city

9) Easter traditions

10) French student life

How familiar are you with the following types of texts? Circle only one number per
question.

1) Poems

2) Informational articles (magazine, newspaper, etc.)

3) Fairy tales

4) Editorials

5) Short stories

6) Plays

7) Novels

How familiar were you with the topics of the texts you read for this study BEFORE you
read the texts? Circle only one number per question.

1) Les trois motocyclistes

2) L ’ceufde quues

3) Le cafe' dans la vie des e'tudiants

4) La vague bouddhiste

206

Appendix F
Text Reactions Questionnaire (Pilot Study)

Using the scale below, indicate the level of difﬁculty of each text that you read. Circle
only one number per question.

 

 

 

Not at all A little difficult Somewhat Very difficult Extremely
difﬁcult difﬁcult difficult
1 2 3 4 5 6 7
1) Les trois motocyclistes

2) L ’arufde Pa‘ques

3) Le cafe' dans la vie des étudiants

4) La vague bouddhiste

Please comment on why each text was difﬁcult or not difﬁcult. PLEASE PROVIDE A
COMMENT FOR EACH TEXT - DO NOT LEAVE ANY SECTIONS BLANK.

 

1) Les trois motocyclistes :

 

2) L ’ceufde Pa‘ques :

 

3) Le cafe' dans la vie des étudiants :

 

 

4) La vague bouddhiste:

 

Using the scale below, indicate your much you liked reading each text. Circle only one
number per question.

 

Disliked Disliked Neither liked Liked somewhat Liked a lot
completely somewhat nor disliked

 

l 2 3 4 5 6 7

 

1) Les trois motocyclistes

2) L ’aeufde Paques

3) Le cafe' dans la vie des étudiants

4) La vague boudtﬂriste

Please comment on why you liked or did not like reading each text. PLEASE
PROVIDE A COMMENT FOR EACH TEXT — DO NOT LEAVE ANY SECTIONS
BLANK.

 

1) Les trois motocyclistes :

 

2) L ’aeufde quues :

 

3) Le cafe' dans la vie des étudiants :

 

 

4) La vague bouddhiste :

 

207

 

 

Appendix G
M‘etacognition Questionnaire

Questions about yourself and what you know about texts and reading
In this questionnaire, you will be asked to read a number of statements about texts: what they
look like, how they are organized and how you can read them. Most statements have to do with
reading in any language. If a statement has to do with reading in English or French, this will be
mentioned clearly in the statement. Read these statements carefully and then circle whether you
agree or disagree with each statement. Circle only one answer per statement.

Texts:

The only purpose of paragraphs is to make sure that there are not too .
many sentences in a row. agree disagree
Words that look the same in French and English always have the same a di
meaning. gree W
A sentence that is polite in English is also polite if you translate it d'
literally into French. “3“” “agree
Words like ﬁrst and second are used to show that a text contains several d'
things about the same topic. agree rsagree
Changing the order of information in a text can change the meaning of a di

the text as a whole. gree sagree
Commas are put in the same place in French sentences as in English agree di
sentences. sagree
To be able to understand a text properly, you sometimes need to know d'
things that me not said in the text. “8“” “‘9‘”
Words like ﬁa'thermore, besides and moreover are used to show that the d'
writer is adding some extra information. agree 1W
Words like he and she often refer to people the writer has not mentioned d'
before. agree rsagree
Not all parts of a text deal with the main idea of the text. agree disagree
All the sentences in a text are equally important for the main idea of the .

text. agree d'sagm
There is alwaysjust one possible word order in a sentence. agree disagree
A paragraph usually has more than one main idea. agree disagree
If you translate an English letter into French, you cannot translate the d'
closing of the letter literally. agree isagree
Words like but and however are followed by something which differs d'

from what has already been said. agree 1W
A report is a description of a situation or event that provides the reader .

with information. “3"” may“
The title often tells you what a text is about. agree disagree
Texts in French and in English are clearly organized differently. agree disagree
The new information contained in a sentence is usually put at the d'
beginning of the sentence. agree rsagree
The order of the information in a text is usually unimportant. agree disagree
Spoken language and written language are exactly the same. agree disagree
You can usually tell from the words in a letter whether the writer and the agree disagree

208

person who will receive the letter know each other well.

The way texts are divided into paragraphs is generally the same in .
English and French. agree disagree
The word order in a French sentence is often different from the word d'

order in an English sentence. agree rsagree
Texts sometimes differ in the way they are organized. agree disagree

 

 

Reading texts:

The following statements are about reading texts about a certain topic (e.g. about history or
about a hobby). The statements are NOT about reading in an English or French reading
comprehension test.

 

While you are reading, it is sensible to spend the most energy on

remembering the details of the text. agree disagree
When youreadachapterinatextbook, itis usefultoﬁrstreadall the di
headings in the chapter to get an idea of what the chapter is about. agree 538T”

Ifyou don’t understand the meaning of a word, it is useful to try and
guess the meaning from the surrounding words and sentences.

When you read in French, it is useful to concentrate more on the meaning d'
of the words than on understanding the ideas. agree 1538703
If you read something that is difﬁcult to understand, it is useful to try and

agree disagree

put it into your own words. agree disagree
If you read a text to ﬁnd a speciﬁc piece of information quickly (e.g.: a di
date), it is sensible to read the text thoroughly. agree W
While you are reading, it is useful to mentally summarize what you have .
read so far. agree disagree
If you don’t understand something in a text, it is sensible to go back to a disa
place in the text before the problem and reread ﬁom there. agree gree
Before you start reading, it is sensible to use the title and any pictures to disa
make guesses about what the text will be about. agree gree
If you want to understand the most important ideas in a text, it is .

agree disagree

important to look up every word you don’t know in a dictionary.
It is sometimes useful to skip difficult words in a text. agree disagree
If you don’t understand the meaning of a word in a French text, it is

useful to look it up in a French-English dictionary. agree disagree
While ou are readin it is sensible to kee on silent tin eve .
word. y g, P ly repea 8 tr agree disagree
Ifyou don’t understand a word, it is useful not to read on until you di
understand it. agree W
While you are reading, it is sensible to try to remember all the words in d'
the text. “gm 'sag' °°
Before you start reading, it is wise to think about why you are going to di
read the text. agree sagree
It is sensible to change the way you read depending on what you want to .
know about a text. agree disagree
It is sensible to ask yourself questions about what you are reading. agree disagree
To understand a text, it is useful to try and remember every detail of what di
you are reading. agree sagree
Ifyou want to get a general impression of what a text is about, it is di
sensible not to read everything. agree sagree
If you don’t understand a sentence, it is sensible to ask yourself questions a di
to ﬁnd out exactly what it is you don’t understand. gree W

209

 

When you are reading, it is sensible to use the organization of the text to
help you understand it.

If you don’t understand a sentence, it is useful to look up every word in a
dictionary.

It is a good idea to check whether you know the exact meaning of every
word in a text.

If you don’t understand a sentence, it is useful to reread that sentence.
When you are reading, it is useful to ask yourself every now and then
whether you understand what you are reading or not.

If you don’t understand something in a text, it is useful to read on a little
to see if it becomes clearer.

If you want to ﬁnd a speciﬁc piece of information quickly (e.g.: a name),
it is sensible to let your eyes dart over the text until you ﬁnd the
information you are looking for.

Ifyou don’t understand a word in a French text, it is useful to try and
guess its meaning from the surrounding words and sentences.

When you are reading a French text, it is sensible to concentrate on one
sentence at a time and to forget about the rest of the text for the moment.

210

agree

agree
agree
agree

agree

agree

agree

agree

disagree
disagree
disagree
disagree
disagree

disagree
disagree

disagree

disagree

Appendix H
Demographic and Language Contact Questionnaire

Part I: Background Information

 

1) Age: 2) Sex: F M
3) Is English your 4) What is your MSU PID (student number
native language? beginning with A)?

Y N

 

5) Please circle all French classes that you are currently taking at MSU :

300-level 400-1evel Other
102 201 202 Course number(s): Course number(s): Course
number(s):

 

 

If yes, when?

 

 

6) Have you traveled to 9
a French-s; . g Y N 3:31; ‘7 (circle on] one)
country for the purpose g ' y '
' 9
Of studying French. 1 semester 2 semesters more than 2
or less semesters

7) Other than experience mentioned in Question 5, have you ever lived in a situation
where you were exposed to a language other than your native language (e.g., by living
in a multilingual community; visiting a community for the purpose of study abroad or
work; exposure through family members, etc.)?

If yes, please give details below. If more than three, list on the last page of this
questionnaire.

 

From when to when

211

8) In the boxes below, rate your language ability in each of the languages that you
know. Use the following ratings:

0) Poor, 1) Good, 2) Very Good; 3) Native/nativelike
Language Listening Speaking Reading Writing Number of years of study

 

English

 

French

 

 

 

 

 

 

 

Other

 

9) Have you studied French in school in the past at each of the levels listed below? If
yes, for how long?

 

 

 

 

 

 

 

 

 

 

 

Circle one If “Yes”, how many years?
Elementary school No Yes How many years?
Middle $0th No Yes How many years?
High 8011001 No Yes How many years?
University/college No Yes How many years?
Other (please specify): No Yes How many years?

 

 

 

 

 

9) What year are you at MSU (circle only one)?
Freshman Sophomore Junior Senior Graduate Other

10) What is your major at MSU?

 

11) Does your major or program of study require you to study a foreign language?
Yes No

12) If you answered yes to Question 11, would you currently be enrolled in a French
class if it were NOT a degree requirement?

Yes No

212

 

 

13) Please indicate the highest level of French study you intend to complete (circle only
one):

more
than

at leastone one
at least one at least one 800-level 800-

FRN102 FRN 201 FRN 202 3031::el 403::el (graduate) level
class (grad

uate)

class

Part II: All of the Questions That Follow Refer to Your Use of French, Not Your Native
Langgge, Unless the Question Says Otherwise.

For each of the items below, please circle only one response and then under that response
write in the number of hours that you estimate that you personally spend doing each
activity.

The following sample response indicates that you communicated with native or ﬂuent
speakers about once a week for two hours:

---------------------------------------------------------------------------------------------------------------------
I

SAMPLE : a. On average, how often did you communicate with native or ﬂuent
speakers of French in French in the year prior to the start of this semester?

 

0) Almost 1) About once a month 2) About once a week 3) Almost every day

never
Hours per Hours per Hours per
month: week: 2 day:

 

 

 

 

0------—--------------------'

 

........................................................

14) On average, how often did you communicate with native or ﬂuent speakers of French
in French in the year prior to the start of this semester?

0) Almost 1) About once a month 2) About once a week 3) Almost every day
never

Hours per Hours per Hours per

month: week: day:

 

15) Prior to this semester, I tried to speak French to:
a. my instructor outside of class

213

0) Almost 1) About once a month 2) About once a week 3) Almost every day
never

Hours per Hours per Hours per

month: week: day:

 

 

 

b. ﬁiends who are native or ﬂuent speakers of French

0) Almost 1) About once a month 2) About once a week 3) Almost every day
never

 

 

 

Hours per Hours per Hours per
month: week: day:
c. classmates
0) 1) About once a month 2) About once a week 3) Almost every day
Almost
never Hours per Hours per Hours per
month: week: day:

 

 

 

d. strangers whom I thought could speak French

0) l) About once a month 2) About once a week 3) Almost every day
Almost
never Hours per Hours per Hours per

month: week: day:

 

 

 

e. a host family, if living in a F rench-speaking area

0) 1) About once a month 2) About once a week 3) Almost every day
Almost
never Hours per Hours per Hours per

month: week: day:

 

 

 

f. service personnel (e.g., bank clerk, cashier)

0) l) About once a month 2) About once a week 3) Almost every day
Almost
never Hours per Hours per Hours per

month: week: day:

 

 

 

214

16) For each of the items below, choose the response that corresponds to the amount of
time you estimate you spent doing each activity in French prior to this semester.

a. watching French language television

0) 1) About once a month 2) About once a week
Almost
never Hours per Hours per

month: week:

 

 

b. reading French language newspapers

0) 1) About once a month 2) About once a week
Almost
never Hours per Hours per

month: week:

 

 

c. reading novels in French

0) 1) About once a month 2) About once a week
Almost
never Hours per Hours per

month: week:

 

 

d. listening to songs in French

0) 1) About once a month 2) About once a week
Almost
never Hours per Hours per

month: week:

 

 

e. reading French language magazines

0) 1) About once a month 2) About once a week
Almost
never Hours per Hours per

month: week:

 

 

215

3) Almost every day

Hours per
day:

 

3) Almost every day

Hours per
day:

 

3) Almost every day

Horns per
day:

 

3) Almost every day

Hours per
day:

 

3) Almost every day

Hours per
day:

 

f. watching movies, videos or DVDs in French

0) l) About once a month 2) About once a week
Almost
never Hours per Hours per

month: week:

 

 

g. browsing French websites

0) 1) About once a month 2) About once a week
Almost
never Hours per Hours per

month: week:

 

 

3) Almost every day

Hours per
day:

 

3) Almost every day

Hours per
day:

 

17) List any other activities you commonly did using French prior to this semester:

Activity/Activities:
O) 1) About once a month 2) About once a week
Almost
never Hours per Hours per
month: week:

 

 

216

3) Almost every day

Hours per
day:

 

Appendix I
Motivation and Anxiety for L2 Reading and Study Questionnaire

Using the scale below, indicate your agreement with the following statements about
reading in French. Circle only one number per statement.

 

Strongly Moderately Somewhat Somewhat Moderatel Strongly

 

 

disagree disagree disagree agree y agree agree
1 2 3 4 5 6
1) I like reading in French.
2) It is fun to read in French.

3) I’m interested in reading in French.

4) Even if I were not required to read French, I would take a class where I had to read in
French anyway.

5) It is a waste of time to learn to read in French.

6) I am tired of reading in French.

7) It is difﬁcult to read in French.

8) It is boring to read in French.

9) I like reading French novels.

10) I like reading French newspapers and magazines.

11) Reading in French is important because it will broaden my view.

12) Reading in French is important because it will be conducive to my general
education.

13) Reading in French is important because it will make me a more knowledgeable
person.

14) I am learning to read in French merely because I would like to get good grades.
15) I am learning to read in French because I think it will help my future career.

16) I am learning to read in French because I would like to learn about opinions of
various people in the world.

17) I am learning to read in French because it will enable me to read French novels.
18) I am learning to read in French because it will enable me to read French newspapers
and magazines.

19) When I see people reading a French magazine or book, I think they are cool and
would like to be like them.

20) I am learning to read in French because I because I might work or study in a
French-speaking place in the future.

21) I think I am good at reading French.

22) Reading French is a challenge that I enjoy.

23) I get immersed in interesting stories even if they are in French.

24) By learning to read in French, I hope to enhance my ability to browse French
websites.

25) I would not voluntarily read in French unless it was required as homework or
assignment.

26) I tend to get deeply engaged when I read in French.

217

27) I do not have any desire to read French even if the content is interesting.
28) I am learning to read in French merely because it is required for graduation.
29) My grades for reading assignments in French have been excellent.

30) Even when reading materials are dull or uninteresting, I always ﬁnish the
assignments.

31) When I read in French, I feel anxious if I don’t know all the words.

32) I prefer to avoid reading in French as much as possible.

33) I feel anxious if I’m not sure whether I understood what I read in French.
34) Reading in French is my hobby.

35) Even if I cannot understand what I read in French completely, it doesn’t bother me.
36) I think I can read quickly in French.

37) I think my reading ability in French is advanced.

38) I think I read in French a lot.

Using the scale below, indicate your agreement with the following statements about
studying French. Circle only one number per statement.

 

Strongly Moderately Somewhat Somewhat Moderately Strongly

disagree disagree disggree agree agree agree
1 2 3 4 S 6

 

1) I enjoy learning French.

2) Learning French is like a hobby to me.

3) I’m interested in learning French.

4) French is one of the important school subjects.

5) Learning French is useful.

6) I try to study hard in French classes.

7) I would like to continue learning French in the ﬁrture.

8) I don’t like learning French, but I think learning French is important.

9) I lose interest in French classes.

10) I lose concentration in French classes.

11) Learning French is important because it will broaden my view.

12) Learning French is important because it will be conducive to my general education.
13) Learning French is important because it will make me a more knowledgeable
person.

14) I am learning French only because it is a required subject.

15) I am learning French because I would like to get good grades.

16) I am learning French because I would like to live or study overseas in the future.
17) I am learning French because it will help me when I travel overseas.

18) I am learning French because it will enable me to understand French novels and
movres.

19) I am learning French because it will help me in my future career.

20) I am learning French because I would like to communicate with French-speaking

people.
21) I study French according to a preplanned schedule.

218

22) I try to study hard in French classes.

23) Even when homework in French is tiresome, I will work hard on it

24) I procrastinate about my homework/assignments in French classes until right before
the due date.

25) I work on my French assignments according to a preplanned schedule.

26) Even if there were no homework, I would try to study French outside the class.
27) I work on my assignments just to the extent that I will not fail a French class.
28) I actively participate in French classes.

29) I spend more time studying for French than for other classes.

30) I take a chance in preparing for French tests by only studying what I think is most
likely to be asked.

31) I think I am good at French.

32) Studying French is a challenge that I enjoy.

33) I tend to get deeply engaged when I study French.

34) My grades in French have been excellent.

35) I prefer to avoid studying French as much as possible.

36) I feel anxious if I’m not sure whether I understand in French class.

37) Even if I cannot understand French completely, it doesn’t bother me.

38) I think my French ability is advanced.

39) I think I study French a lot.

40) I often feel anxious in French class.

219

Appendix J
Genre Reactions Questionnaire
The following statements have to do with reading different types of texts in French:

M refers to (excerpts from) short stories, novels, plays, etc.
“Informational texts” refers to (excerpts from) news articles, cultural readings, etc.

Circle only one number per statement.

 

Strongly Moderately Somewhat Somewhat Moderately Strongly
disagree disagree disagree agree agree agree
1 2 3 4 S 6

 

l) I think that stories are difficult to read in French.

2) I think that informational texts are difﬁcult to read in French.

3) I enjoy reading stories in French

4) I enjoy reading informational texts in French.

5) I think reading stories in French is interesting.

6) I think reading informational texts in French is interesting.

7) I think I learn a lot ﬁem reading stories in French.

8) I think I learn a lot from reading informational texts in French.

9) I think I am a good reader when I read stories in French.

10)I think I am a good reader when I read informational texts in French.
11)] think it is useful to read stories in French.

12)] think it is useful to read informational texts in French.

l3)I am comfortable reading stories in French.

14)I am comfortable reading informational texts in French.

lS)I am anxious when I read stories in French.

l6)I am anxious when I read informational texts in French.

17) If I had a choice, I would rather read stories than informational texts in French.
18)I think reading stories is boring in French.

19)I think reading informational texts is boring in French.

20)I think I understand a lot when I read stories in French.

21)I think I understand a lot when I read informational texts in French.
22)I think reading stories in French is worthwhile.

23)I think reading informational texts in French is worthwhile.

24)I tend to lose interest when I read stories in French.

25)I tend to lose interest when I read informational texts in French.

220

Appendix K
Text Reactions Questionnaire

Using the scale below, please answer the following questions about each of the four
texts that you read. Circle only one number per question.

 

Strongly Moderately Somewhat Somewhat Moderately Strongly

 

 

disagree disagree disagree agree agree agree
1 2 3 4 S 6
Les trois motocyclistes
1) This text was difficult.

2) The grammar in this text was difﬁcult.

3) The vocabulary in this text was difficult.

4) I was familiar with the topic of the text.

5) This text was too long.

6) This text was difﬁcult because it was a story.

7) I enjoyed reading this text.

8) The topic of the text was interesting.

9) I experienced positive feelings or emotions when I read this text.
10) I enjoyed reading this text because it was a story.
11) It was useful to read this text.

12) I was comfortable reading this text.

13) I think I understood nearly all of the text.

14) I was anxious when reading this text.

15) I was good at reading this text.

16) I learned something by reading this text.

17) It was a waste of time to read this text.

18) This text was boring.

Please provide any other comments you have about your experience reading Les trois
motocyclistes:

L ’aeufde Pa‘ques

 

Strongly Moderately Somewhat Somewhat Moderately Strongly

 

 

disagree disagree disagree agree agree _agree
1 2 3 4 5 6
1) This text was difﬁcult.

2) The grammar in this text was difficult.

3) The vocabulary in this text was difficult

4) I was familiar with the topic of the text.

5) This text was too long.

6) This text was difﬁcult because it was a story.

221

7) I enjoyed reading this text.

8) The topic of the text was interesting.

9) I experienced positive feelings or emotions when I read this text.
10) I enjoyed reading this text because it was a story.
1 1) It was useful to read this text.

12) I was comfortable reading this text.

13) I think I understood nearly all of the text.

14) I was anxious when reading this text.

15) I was good at reading this text.

16) I learned something by reading this text.

17) It was a waste of time to read this text.

18) This text was boring.

Please provide any other comments you have about your experience reading L ’aeufde
quues:

Le café dans la vie des étudiants

 

Strongly Moderately Somewhat Somewhat Moderately Strongly

 

disagree disagree disagree agree agree agree
1 2 3 4 S 6

 

I) This text was difﬁcult.

2) The grammar in this text was difﬁcult.

3) The vocabulary in this text was difﬁcult.

4) I was familiar with the topic of the text.

5) This text was too long.

6) This text was difficult because it was an informational text.

7) I enjoyed reading this text.

8) The topic of the text was interesting.

9) I experienced positive feelings or emotions when I read this text.
10) I enjoyed reading this text because it was an informational text.
11) It was useﬁrl to read this text.

12) I was comfortable reading this text.

13) I think I understood nearly all of the text.

14) I was anxious when reading this text.

15) I was good at reading this text.

16) I learned something by reading this text.

17) It was a waste oftime to read this text.

18) This text was boring.

Please provide any other comments you have about your experience reading Le café
dans la vie des étudiants:

222

Sport: l’homme sans limites

 

 

 

Strongly Moderately Somewhat Somewhat Moderately Strongly
disagree disagree disagree agree agree agree
1 2 3 4 S 6
1) This text was difficult.
2) The grammar in this text was difficult.
3) The vocabulary in this text was difficult.
4) I was familiar with the topic of the text.
5) This text was too long.
6) This text was difficult because it was an informational text.
7) I enjoyed reading this text.

8) The topic of the text was interesting.

9) I experienced positive feelings or emotions when I read this text.
10) I enjoyed reading this text because it was an informational text.
1 1) It was useﬁrl to read this text.

12) I was comfortable reading this text.

13) I think I understood nearly all of the text.

14) I was anxious when reading this text.

15) I was good at reading this text.

16) I learned something by reading this text.

17) It was a waste of time to read this text.

18) This text was boring.

Please provide any other comments you have about your experience reading Sport:

l’homme sans limites:

223

33 $3 $3 :3- 33 Q3 33 83 33 a3 33 .3353
be a he an; as a 3:82 382.3 means ea _ H2

 

 

 

333 :3 m _ 3- :3 83 333 83 $3 $3 $3 sees ea 22,8 5323 2385
3 0E 033:0 E? 3 03003 .Hoeocm @332 80 H 3H

:3 33- $3 33 e3 33 e3 93 e3 .8323 see
H :33 0E 33 :3» “H 00:33 3030an 93.82 83 H CH

23- X3 :3 333 9.3 :3 e3 33 .233 he a means 3333

.8 9»: 3 00H: 2:95 H 00933 305.5 38.82 83 H 3H

:3- 33. R3. 333- 83- R3- 33. than 383

30w 3 82 2:03 H 03:33 30:0an 3382 En H AmH

83 $3 $3 83 $3 $3 38: a 382.3 35.53 a

065.5me

53 $3 $3 £3 E3 382m 333 _ ea? Enema... menu some 22 c an

~33 33 33 R3 3.35 H as unease e a 3823 3:533 am

53 Q3 S3 .233

R3 :3 383 means a unease a; a

333 as a 332. e as a 3823 masses 3

30:05 $3532 50.30 H AH

amass

 

«H E at: 2 m mm mm Ht m N H

 

.455 NH 8H b0qu< Ea couguOSH 8H 205203.80 80:30::

3 338%..

224

 

 

 

 

 

 

 

333 Q3 33 23- S3- 33 :3- :3- 3.3- 33. $3 $3 .883 23832335. see e a:
03.08 HHHB 3 000003 00000003 3 3000.5 w0H080..H AmH
mod and 2.0. 2.0. 3.0 3.0. m H.o.. wmdt end- mmd mmd 00330000 _00000w b0 00 03000000
_ 3 =8 3 8383 050883 a 32.23 means H0
$0 8.0. H Hd- Hod- 3.0. and- Hmdt mad- de Hmd .303 ~20 000003
EB 3 3382 358383 a 3820 353.33 E
3.0- no.0- 8.0. $5. omd- det med- 00.0 3.0 3000.303 H00303 0088003 0350 000 3 3000.5 Av
000000003
v0.0 No.0 de 0nd Hmd 3.0 3.0.. 3.0. .0020 30005 E 000380 H33 0030 H 80
omd 3.0 cod 3.0. 50.0 mod- 00.0 K08 00383 0.03000
3 .30003080 30000.3 00580000 000000 H .3 0033 ﬁn
mud 2.0 H H.o wad mod- Hod .3020 30000.3 3
000500000 H 003833 0.80 80 05 .3 000083 H00.H H Gm
b03004.
$3 $3 $3 333- :3- .2880
00 30.00 00 3000i wag—V. H30; 00 00.8an H Gm
and mmd 0nd- Rd- 3020 3000i 0 HRH 80 HHHB
H 0030 000000 06 00 0003. $00000me00 H00 00 3.83 H Hum
33 $3- :3- .3833.
0030000 0 mm 3 000003 .300 30003 $30.80— 08 H A:
00.0- omd- 3030.30 3000.5 .3 0000003 08H H a
0033>308<
8.0 08003 m:§00%.30000m 3035 00003800000
3 00H: 2003 H 033 3000an w0H080H .03 H 8N
NH HH ow Hue. 0m mm Him I a on 8H

 

225

 

 

 

 

00.0 0:.0 00.0 00.0 0:.0 _00 :00 0:0 0:0 :0- :00 :00- 000020 0200
.80 0030 30000.3 8.3 mag: 0030 0.88 00000 H an
00.0 00.0. :00 0:0 00.0 0:.0 :00 3.0 00.0 00.0 0:.0 080% 0800 s 30002000 0308 0 :0
00.0 :00 00.0 00.0 :00 :0 :0 00.0 00.0 00.0 .0020 20 0:300 0820 0030
00 F 2003 H a3.830083 00 0003 0005 .3 00>mH Gm
00.0 00.0 0:0 0:0 0:.0 00.0 :0 00.0 :00 03023. 0800020
0 00 3308000 080%me 30000.3 30 00 383 H Gm
00.0 00.0 00.0 0:.0 00.0- :00 00.0 :00- .900 2.0 20 20000 £00 0:3 80020 082.0
5 30033083830803 30 0830 00050000008 H can
00.0 0:0 0:0 :00 00.0 00.0 :0 .0 00 0a: 0.83
:3 H 60800.30 3 3000.0: 03 3830083 0033 003m Ann
:00 0:.0 :0 000 0:0 000 030003. 085.0200 2000880 0008.0 0000 0 :0
:00 :0 000 0:0 000 .0330 00020 a 00000888 82 0 a:
00.0 0: .0 00.0 :00 080% 02.20 a 0.9: :03: e E 0 H0
0000: 0020
00.0 00.0 :0 080000 0 00:0 082... :8 0005 0 H0:
:0 0:.0 08085 :80 300 00:20 a 3080 :2 a:
00.0 .0800 a 080 pa 0 000 0 2:
0000.0. 020200.000
00 00 :0 00 :0 :0 0: 0 0: 0: : :_

 

226

._o>o_ 3. 05 .8.“ S. “a
was :33 mo. 05 8m 2. 3 EnumEmMm bﬁommuﬁm 0.3 macaw—9:00 6208 3.55.. 03 $1333 a 5m? 3an8 man: .982

 

3o 93 .8. a :85 %em _ x55 _ am
25 Q8? 8 9 395 use mm is _ E3 3.55;
baa 3 38» soap—m Bu 35¢ch E 83o a 83 H 8m

cm a

 

227

 

 

8d 85 :.o :.o 85 mod 85 ed 85 m; 98.82% 05 52m 93%
_ $338063: .5 :3 9a mEtBaE 3:32 5:3 =o>m 8m
85 a3 3o coo 35 So one 85 ad :85 5 Be _ a2? Baas 28% mm 9 Bo. _ Gm
Ed _2 23 86 3o _2 ”no mg :85
E 0.8 ~35 mm Go>o momuQ—m wﬂﬁmouoam 5 @8589: wow H AMN
Ed Rd 85 85 23 m3 m8 .3?» _ as ”@230 a a :08; 9553” am
95 as one and as and .283 05 a 238 325 Mo 953% 58¢ :32
e 9a 283 ~ 8:88 :8on a 82 3 maﬁa. :3 _ A2
#2 «2 $5 $5 o2 dogmas as eaamaoa :85 332 ea _ a:
86 85 ”no ”no .226: £82m 2:32 32 H a
who as $5 sis 685 a 32 3 9H _ 20:3 was
a 33» 2303 — .nocounm 30.— 3 @2368 «on 0.63 m .2 =o>m Av
25 8d :85 a 382 a 83.235 .5 a.
N3 6:5 5 Ba 3 5 mm a a
.58; a use? 85 _ 2
2835
cm 8 mm 2 S a V m N_

 

waged 3 Eu bomg 28 damage} 8m 333280 88285

2 598%?

228

$6 3.6 3.? mad- 36. had- :6. mad. and- :6- .noaﬂm E 32 3 wcton mm a G

 

 

 

 

«no m3- m2. 8.? $3. 33. on? 36. 85. 3823 3 332 mo 83 a“ _ 6
was. one- 33. N8- «8. 8.3 one. «3. 382,3 3 .32 3 BE 3 3.3% 333 a a 3 s
305305

and as $5 ”no 33 N2 86 .333» €3.33 8395 3 £33

m8 35:5 3 one: _ £8on E 32 3 mauso— mm Gum

m3 3% 3:. 3o 26 83 .233 2: 3 8% 33033823 3 3 3.33. s 383

233 H 388; _ 8:82 33.23 3 32 3 EEE :3 _ am

one as a... as 83 .303 32 B 3 ea vac; 33 38 pa 33 333

”icon .8 egg :82”— a 9:32 oaoon com :6; A3

33 as cg mod 38333 33 exams»: £83,,— 82 3

”a 2330 as a 0383 3823 3 Ba 3 3:332 :3 _ a:

”no .33 33 .305: £833 82 3

as 0380 as a 3:83 3833 3 32 3 maﬁa_ :3 H E

36 Ed .528 233 he 20:
__Egﬁf838338233323m3§_§:2

ommﬁuxm

85 .3313 3382333383 «3

 

 

229

m; 2.0. 8.0 3.? mad. and- Rd- Rd. 32. m3- 3333.? 32% 33 3 033.28
3 33 3 3:83 33:23 3 653 3 323 5

 

 

230

 

 

:6- 2d mod- omd- 33. Had- 2.0- mmd- omd. .303 m8
5395 HHHB «H 3303 Satan: ﬂ 3055 3H 933% 2 H

853.893

:8 38 :3 So- 33 mod 83 So ..3: 333 3.883 3 530338
:0:on :H 32 H 33H? c3203: 830 H .2 co>m Gm

ﬁne 26 cod :6 Ed Ed 35 £82m S 32

H 3:? H9833”? H .8633 25m 8: EH :3 323 32 H Amm

mad NHd de 26 Ed 86 .2203 05

:3 26qu r38 HHH .3on3 H8.“ H £32m 5 32 H 5:3 AH m

D253.

36 mmd 86 End 36 .oHnHmmonH mm :88 ma £255 5 332 £26 3 EH29 H Rm

:8 m2 5. 33 .8333» 3 3333

3 “H 8303 £208 no=2nH 3H 32 3 9332 3 H SN

83 ”we a; 333333 3 3338

2: HH :96 noa2nH 32 9 23% >3 03: 8: at H Ch

36 $6 35.9%ng .8 333238: 3 H233.—

33 “H 32:: n82nH E 32 £33329, 8: @303 H Ann

vmd 303% 3% 8w 8 25

2:03 H 8303 3208 no=2m 5 32 9 332 3 H a:

 

HH mm mm Hm Nm mm mm mm 3 a

was _u>o_ mo. 2: 8m 3. “a “5&6me b32623 v.3 $0322.80 6208 8:38 93 AL x383 5 5m? v3.38 2:3— 632

..22 8. 2: é a. a

 

 

 

Ed one as ”no one Nae ”do E a £55 a 32 _ “35 H :3
NB ..2 Rd NS 2d mg 305% mm 58¢ a E5“ 332 Exams H 5
one $5 m2 26 £5 dozens £23 825025523
”2 02 85 8.0 22.85
soon 96: coco:— E 858363 9:32 8m 33% >2 8N
owe a; ed :85 maﬁas Bowaaiaaza
8.? 8o ..ﬁﬁé a Re 9 558% a a 9
£5 Baoeomaom
a on an _N h 2 S

 

231

Appendix N

Text Comments

 

Narrative Text 1 (The Three Motorcyclists)

 

I remembered this one the best.

Of all the text (sic) I read in this handout, "Les trois motocyclistes" was the text I
understood the most.

Of the four texts, I found this one to be the easiest to read and understand.

The language seemed more simple than most texts which made it easier to follow.
This was the easiest to read, I knew a lot of the vocab and the storyline was pretty
easy to follow.

Very clear details.

It was easier to remember because if was story-like.

I liked it because it was a chain of events rather than a bunch of information.

I liked this excerpt the best b/c it was more of an actual story.

It was a cute story, I liked it.

It was a cute little story.

I really liked this story.

I found it interesting, but mostly because I go camping every year in Canada,
near/on dunes of Lake Huron.

I liked it. The way the author described the 3 motorcyclists when they were
coming toward the narrator was cool.

I liked it, I just don't know if I got it all, were the three motorcyclist (sic) part of
the 9 campers?

I found it hardest to remember names from the story and events that happened in
the middle paragraph.

In all honesty, I had no idea what 3 motorcyclists had to do with the 9 people
camping or what the narrator though when he/she saw them.

This I think was the hardest one for me unfamiliar w/ a good amount of the words.
I don't know, I didn't get all the vocab, so I'm sure I missed something. The
grammar wasn't hard to understand, though.

Some difﬁcult grammar/vocabulary made the point hard to understand.
Understood only parts of the story.

I didn't understand much of the text.

This one was very difﬁcult for me. I realize how poor my French literacy skills are.
I can catch words and phrases, but I can't comprehend many full sentences.
There were a few words I didn't understand.

I did not understand the part when the motorcycles came down the dune.

I didn‘t know much camping terminology the 3 people confused me a little.
This story from what I could read was really weird.

Wonder why part was in English.

I noticed F rench-Canadian"isms".

232

 

Narrative Text 2 (The Easter Egg)

 

This was the easiest one for me.

I found some of the details from this story easier to remember because there were
dates and numbers and direct quotes that I could remember. Plus I liked this story
the most!

This text was also easier to remember because it was a story

Cute story

It was a cute story!

Such a cute idea! Made me smile.

This was my favorite!

I liked the story and I think it's cute that every year the husband continues with the
tradition he started with hiding a chocolate eg for his wife.

Loved the story and am familiar with Easter traditions.

I enjoyed this text because I had personal connections to it. My brother and I hide
things on each others birthdays w/ surprises or notes that say "good for. . . "

I liked this story for the most part. I wish I could have understood more of it. :(

I want to know this story. I would like to hear it in English to understand it because
it sounded interesting and I think the point was lost on me.

Cute story but I might have misunderstood a few sentences, I think.

I don't think I really understood what it was about.

This text gave me the most problems in understanding what the story was about.
This was the hardest story for me to understand and I don't think I got a lot of it.

I had a harder time understanding this text. . .I'm not sure why.

The whole middle of the story is pretty much a blur because I couldn't understand
most of it. It was somewhat boring because I couldn't understand most of it.

Hard to tell what the main point of the story was about.

I didn't understand a lot of the vocabulary, so I was lost.

I don't really know what was going on here, because I don't know what cachait (or
any of its forms) means. I just couldn‘t follow this.

I can only gather tidbits. I only caught some vocab words in this one.

I don't understand some of the vocab.

Was nice to understand all of a story but do not know what Paques is

I wish I knew what Pﬁque (sic) was - no, wait I do it's caster (sic). Oops!

Sorry, forgot what a lot of the vocab meant

Eggs and chocolate were pretty much all I remember, however this was the ﬁrst
story so I don't think I read it ”right"/as critically as the others. Although of all the
4 stories I thought this was the most difﬁcult.

233

 

Expository Text 1 (The Cafe' in Student Life)

 

I thought this text was easier to read (possibly because it wasn't a story/it was more
interesting to me)

Out of all the texts this one was the easiest for me to understand. Perhaps because
it wasn't a story with a lot of details to try to remember I was better able to
understand the main point. Also, I think the vocabulary and grammar were at the
right level.

Understood basic concept, but unclear about some parts

This wasn't too hard, it's just hard to remember every detail.

I liked it but I have questions.

I really liked this text.

I could relate to this story.

Could relate to this story

It was a good text because college kids can relate to it somewhat.

I understood some of this and found it more intresting (sic) because of the culture-
base of the text.

I also had personal memories of sitting in European cafes meeting people and
studying so I enjoyed it because I recalled those fun times.

I understood this less, and remembered less than the others.

The hardest to understand because of the grammar/vocab I did not know many of
the phrases

The vocab was difﬁcult for me so I struggled w/ what was going on.

For me it was hard to think of speciﬁc details for this text because most of the
information seemed to be general ideas.

Seemed to have less detail/info than others.

I didn't know students in France spent so much time in cafes. But I guess American
students do spend a fair amount of time in coffee shops.

I knew cafes were important in France, but I didn't know that they had their own
personalities. They are kind of like bars over here in the US.

It's cool to think about kids in France @ a coffee shop like we do @ MSU. But
they are halfway around the world.

I assumed we were reading mostly about French students, however I think the
story is applicable to students in the US. as well.

 

Expository Text ZlS'ports: Man Without Limits)

 

Didn't get everything, but I think I got the point.

This story/article was interesting to me.

The Olympics are always interesting to me.

Another one I understood better than the stories. The link between the 19th century
and technology made it a little harder to understand the main point but again
vocabulary and grammar were ﬁne.

This was also a difficult one partially because it was informational and I lost
interest.

This text gave me some problems in the translation.

This was the most difﬁcult text.

234

This one was the hardest of all.

Did not understand very well.

Do not understand anything in this text at all.

This was the hardest for me to read and understand.

I tried to understand this more because I love the Olympics but it was still difﬁcult.
This one was hardest for me to remember (maybe because it was ﬁrst) but also
because it seemed like the ﬁrst paragraph talked about a different aspect than the
next 2 paragraphs. There was just more to think about than for the stories.

My least favorite of the four texts.

It made me think of the gatorade [sic] commercials with the athletes on the
treadmills, and they are monitoring their performance.

235

References

Aida, Y. (1994). Examination of Horwitz, Horwitz and Cope’s construct of foreign
language anxiety: The case of students of Japanese. The Modern Language
Journal, 78, 155-68.

Alderson, J.C., Clapham, C., & Steel, D. (1997). Metalinguistic knowledge, language
aptitude and language proﬁciency. Language Teaching Research, 1, 93-121.

Allen, E.D., Bernhardt, E.B., Berry, M.T., & Demel, M. (1988). Comprehension and text
genre: An analysis of secondary school foreign language readers. The Modern
Language Journal, 72, 163-72.

Argaman, O., & Abu-Rabia, S. (2002). The inﬂuence of language anxiety on English
reading and writing tasks among native Hebrew speakers. Language, Culture and
Curriculum, 15, 143-60.

Barnford, J ., & Day, RR. (2004) Extensive reading activities for language teachers.
Cambridge: Cambridge University Press.

Baker, L., & Wigﬁeld, A. (1999). Dimensions of children’s motivation for reading and
their relations to reading activity and reading achievement. Reading Research
Quarterly, 34, 452-77.

Barry, 8., & Lazarte, A. (1998). Evidence for mental models: How do prior knowledge,
syntactic complexity, and reading topic affect inference generation in a recall task
for non-native readers of Spanish? The Modern Language Journal, 82, 176-93.

Bensoussan, M. (1990). EFL reading seen through translation and discourse analysis:
Narrative vs. expository texts. English for Specific Purposes, 9, 49-66.

Bernhardt, E. (1991). Reading and development in second language: Theoretical,
empirical and classroom perspectives. Norwood, NJ: Ablex.

Bernhardt, E. (2005). Progress and procrastination in second language reading. Annual
Review of Applied Linguistics, 25, 133-50.

236

Bernhardt, E.B., & Kamil, ML. (1995). Interpreting relationships between L1 and L2
reading: Consolidating the linguistic threshold and the linguistic interdependence
hypotheses. Applied Linguistics, 1 6, 15-34.

Bollen, K.A., & Long, J .S. (1993). Testing structural equation models. Newbury Park,
CA: Sage.

Bossers, B. (1991). On thresholds, ceilings and short circuits: The relation between L1
reading, L2 reading and L2 knowledge. AILA Review, 8, 45-60.

Bowers, P., Golden, J ., Kennedy, A., & Young, A. (1994). Limits upon orthographic
knowledge due to processes indexed by naming speed. In V.W. Berninger (Ed),
The varieties of orthographic browledge 1: Theoretical and developmental issues
(pp. 173-218). Dordrecht, the Netherlands: Kluwer.

Brantmeier, C. (2005) Nonlinguistic variables in advanced second-language reading:
learners’ self-assessment and enjoyment. Foreign Language Annals, 38, 494-504.

Brantrneier, C. (2006). Toward a multicomponential model of interest and L2 reading:
Sources of interest, perceived situational interest, and comprehension. Reading in
a Foreign Language, 18, 89-115.

Brisbois, J .E. (1995). Connections between ﬁrst- and second-language reading. Journal
of Reading Behavior, 27, 565-84.

Brown, H.D. (2007). Principles of language teaching and learning (5th ed.). New York:
Pearson Education.

Bullock, B., & Lantolf, J .P. (1987). Cloze testing and text genre. Rassegna italiana di
linguistica applicata, 19, 95-104.

Byme, BM. (2001). Structural equation modeling with AMOS: Basic concepts,
applications, and programming. Mahwah, NJ: Lawrence Erlbaum Associates.

Carrell, P. (1987). Content and formal schemata in ESL reading. TESOL Quarterly, 21,
461-81.

237

Carrell, PL. (1989). Metacognitive awareness and second language reading. The Modern
Language Journal, 73, 121-34.

Carrell, PL. (1991). Second language reading: reading ability of language proﬁciency?
Applied Linguistics, 12, 159-79.

Carrell, PL. (1998). Can reading strategies be successfully taught? Australian Review of
Applied Linguistics, 21, 1-20.

Carrell, P., Devine, J ., & Eskey, D. (1988). Interactive approaches to second language
reading. Cambridge: Cambridge University Press.

Chen, Q., & Donin, J. (1997). Discourse processing of ﬁrst and second language biology
texts: Effects of language proﬁciency and domain-speciﬁc knowledge. Modern
Language Journal, 81, 209-227.

Chin, H.K., & Bandari, R. (2007) Open doors 2007: Report on international educational
exchange. New York: Institute of International Education.

Clement, R., Domyei, Z., & Noels, K. (1994). Motivation, self-conﬁdence and group
cohesion in the foreign language classroom. Language Learning, 44, 417-48.

Clement, R., & Kruidenier, 8.6. (1985). Aptitude, attitude and motivation in second
language proﬁciency: a test of Clement's model. Journal of Language and Social
Psychology, 4, 2 l -3 7.

Cobb, T. (2006) Web VP en ﬁ'ancais v.2.5. [Computer program]. Retrieved April 27,
2006, from http://www.lextutor.ca/vp/ﬁ'/

Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-9.

College Board. (n.d.) SAT -ACT Concordance Tables. Retrieved May 20, 2008 from
http://www.collegeboard.com/prod_downloads/highered/ra/sat/satACT_concorda

nce.pdf

238

Constantino, R., Lee, S., Cho, K., & Krashen S. (1997). Free voluntary reading as a
predictor of TOEFL scores. Applied Language Learning, 8, 111-18.

Csizér, K., & Domyei, Z. (2005). The internal structure of language learning motivation
and its relationship with language choice and learning effort. The Modern
Language Journal, 89, 19-36.

Deci, E. L., & Ryan, R M. (1985). Intrinsic motivation and self-determination in human
behavior. New York: Plenum.

Dewey, DP. (2004). A comparison of reading development by learners of Japanese in
intensive domestic immersion and study abroad contexts. Studies in Second
Language Acquisition, 26, 303-27.

Dornyei, Z. (1990). Conceptualizing motivation in foreign-language learning. Language
Learning, 40, 45-78.

Dbmyei, Z. (2005). The psychology of the language learner: Individual differences in
second language acquisition. Mahwah, NJ: Lawrence Erlbaum Associates.

DOmyei, Z., & Kormos, J. (2000). The role of individual and social variables in task
performance. Language Teaching Research, 4, 275-300.

DuBravac, S., & Dalle, M. (2002). Reader question formation as a tool for measuring
comprehension: narrative and expository textual inferences in a second language.
Journal in Research of Reading, 25, 217-31.

Ehrman, M.E., & Oxford, RL. (1995). Cognition plus: Correlates of language learning
success. The Modern Language Journal, 79, 67-89.

Elkhafaiﬁ, H. (2005). The effect of prelistening activities on listening comprehension in
Arabic learners. Foreign Language Annals, 38, 505- 13.

Ellis, NC. (2002) Frequency effects in language processing. Studies in Second Language
Acquisition, 24, 143-88.

239

Ellis, R. (2008). Study of second language acquisition. Oxford: Oxford University Press.

Elley, W.B. (1991). Acquiring literacy in a second language: the effect of book-based
programs. Language Learning, 41, 375-411.

Elley, W.B., & Mangubhai, F. (1983). The impact of reading on second language
learning. Reading Research Quarterly, 19, 53-67.

F ecteau, ML. (1999). First- and second-language reading comprehension of literary
texts. The Modern Language Journal, 83, 475-93.

Frantzen, D. (2002) Rethinking foreign language literature: Towards an integration of
literature and language at all levels. In V. Scott & H. Tucker (Eds) SLA and the
literature classroom: Fostering dialogues (pp. 109-30). Boston: Heinle.

Freed, B.F., Dewey, D. P., Segalowitz, N., & Halter, R. (2004). The language contact
proﬁle. Studies in Second Language Acquisition, 26, 349-56.

Ganschow, L., & Sparks, R. (1996). Anxiety about foreign language learning among high
school women. The Modern Language Journal, 80, 200-12.

Ganschow, L., Sparks, R.L., Anderson, R., Javorshy, J. Skinner, S., & Patton, J. (1994).
Differences in language performance among high-, average-, and low-anxious
college foreign language learners. The Modern Language Journal, 78, 41-55.

Gardner, RC. (1 985). Social psychology and second language learning: The role of
attitudes and motivation. London: Edward Arnold Publishers.

Gardner, R. C., Lalonde, R. N., Moorcroft, R., & Evers, F. T. (1987). Second language
attrition: The role of motivation and use. Journal of Language and Social
Psychology, 6, 29-47.

Gardner, R.C., & Lambert, W.B. (1959). Motivational variables in second language
acquisition. Canadian Journal of Psychology, 13, 266-72.

240

Gardner, R.C., & Lambert, W.B. (1972). Attitudes and motivation in second language
learning. Rowley, MA: Newbury House Publishers.

Gardner, R.C., & Maclntyre, PD. (1991). An instrumental motivation in language study:
Who says it isn’t effective? Studies in Second Language Acquisition, 13, 57-72.

Gardner, RC, & Maclntyre, PD. (1993). A student's contribution to second- language
learning Part II: Affective variables. Language Teaching, 26, 1-1 1.

Gardner, R.C., Tremblay, P., & Masgoret, A.M. (1997). Towards a full model of second
language learning: An empirical investigation. The Modern Language Journal,
81, 344-62.

Garner, R. (1994). Metacognition and executive control. In R.B. Ruddell, M.R Riddel &
H. Singer (Eds) Theoretical models and processes of reading (4th ed.) (715-32).
Newark, DE: International Reading Association.

Garson, GD. (2008). Factor analysis. Statnotes: Topics in Multivariate Analysis.
Retrieved May 30, 2008 from
http://www2.chass.ncsu.edu/garson/pa765/statnote.htm

Gass, S.M., & Selinker, L. Second language acquisition: An introductory course (3rd ed.).
New York: Routledge.

Gelderen, A. van, Schoonen, R., Glopper, K. de, Hulstijn, J ., Simis, A., Snellings, P., &
Stevenson, M. (2004). Linguistic knowledge, processing speed, and
metacognitive knowledge in ﬁrst- and second-language reading comprehension: a
componential analysis. Journal of Educational Psychology, 96, 19-30.

Goodfellow, R., Jones, G., & Lamy, MN. (2002). Assessing learners’ writing using
Lexical Frequency Proﬁle. ReCALL, 14, 129-42.

Gottﬁ'ied, A.E. (1990). Academic intrinsic motivation in young elementary school
children. Journal of Educational Psychology, 82, 525-38.

241

Grabe, W., & Stoller, F .L. (2002) Teaching and researching reading. New York: Pearson
Education.

Graesser, A., Golding, J .M., & Long, D.L. (1991). Narrative representation and
comprehension. In R Barr, M.L. Kamil, P. Mosenthal, & P.D. Pearson (Ed),
Handbook of reading research (V 01. 2, pp. 171-205). New York: Longman.

Graesser, A. C., Hauft-Smith, K., Cohen, A.D., & Pyles, LB. (1980). Advanced outlines,
familiarity, text genre and retention of prose. Journal of Experimental Education,
48, 209-20.

Gregg, K. (1984). Krashen’s monitor and Occam’s razor. Applied Linguistics, 5, 79—100.

Guthrie, J. T., McGough, K., & Wigﬁeld, A. (1994). Measuring reading activity: An
inventory (Instructional Resource No. 4). Athens, GA: National Reading Research
Center.

Guthrie, J .T., Wigﬁeld, A., Metsala, J.L., & Cox, KB. (1999). Motivational and
cognitive predictors of text comprehension and reading amount. Scientiﬁc Studies
of Reading, 3, 231-56.

Haggstrom, MA. (1993). Study strategies for the beginning foreign language classroom.
The Canadian Modern Language Review, 49, 236-54.

Holmes, R., & Omar, M. (2006). Adjunct language instruction: Modeling student
perceptions. [T L Review of Applied Linguistics, 151, 57-82.

Horiba, Y. (1990). Narrative comprehension processes: A study of native and non-native
readers of Japanese. Modern Language Journal, 74, 188-202.

Horwitz, E. (2001). Language anxiety and achievement. Annual Review of Applied
Linguistics, 21, 112-26.

Horwitz, E.K., Horwitz, M.B., & Cope, J. (1986). Foreign language classroom anxiety.
The Modern Language Journal, 70, 125-32.

242

Horwitz, E.K., & Young, DJ. (1991). Language anxiety: From theory and research to
classroom implications. Englewood Cliffs, NJ: Prentice Hall.

In’nami, Y. (2006). The effects of test anxiety on listening test performance. System, 24,
317-40.

Jurasek, B.S., & Jurasek, RT (1991). Building multiple proﬁciencies in new curricular
contexts. In J .K. Phillips (Ed.) Building bridges and making connections (pp. 89-
121). Burlington, VT: Northeast Conference Reports.

Kim, H., & Krashen, S. (1997). Why don’t language acquirers take advantage of the
power of reading? TESOL Journal, 6, 26-9.

Kintsch, W. (1988). The role of knowledge in discourse comprehension: A construction-
integration Model. Psychological Review, 95, 163-82.

Kintsch, W. (1998). Comprehension: A paradigm for cognition. Cambridge: Cambridge
University Press.

Kintsch, W., & Young, SR. (1984). Selective recall of decision- relevant information
from texts. Memory and Cognition, 12, 112—17.

Kline, RB. (2005). Principles and practice of structural equation modeling (2”l ed.).
New York: Guilford.

Koda, K. (2005). Insights into second language reading: A cross-linguistic approach.
Cambridge: Cambridge University Press.

Kondo-Brown, K. (2006) Affective variables and Japanese L2 reading ability. Reading in
a Foreign Language, 18, 55-71.

Kozminsky, E. (1977). Altering comprehension: The effects of biasing titles on text
comprehension. Memory and Cognition, 54, 482-90.

243

Krashen, SD. (1982). Principles and practice in second language acquisition. New
York: Pergarnon.

Krashen, SD. (1985). The input hypothesis. London: Longman.
Lamb, M. (2004). Integrative motivation in a globalizing world. system, 32, 3-19.

Larsen-Freeman, D., & Long, M. H. (1991). An introduction to second language
acquisition research. New York: Longman.

Larson, ML. (1984). Meaning-based translation: A guide to cross-language equivalence.
Lanham, MD: University Press of America.

Lee, J .F. (1986). Findings and implications of L2 reading research. Hispania, 69, 181-87.

Lee, J .F. (1986a). On the use of the recall task to measure L2 reading comprehension.
Studies in Second Language Acquisition, 8, 83-93.

Lee, J ., & Schallert, D.L. (1997) The relative contribution of L2 language proﬁciency and
L1 reading ability to L2 reading performance: A test of the threshold hypothesis
in an EFL context. TESOL Quarterly, 31, 713-39.

Lee, J. F., & VanPatten, B. (2003). Making communicative language teaching happen
(2"(1 ed.). New York: McGraw Hill.

Lee, S. (2005). Facilitating and inhibiting factors in English as a foreign language writing
performance: A model testing with structural equation modeling. Language
Learning, 55, 335-74.

Maclntyre, P., & Noels, K. (1994). Retrospective review article: The good language
learner. system, 22, 269-80.

McDevitt, B. (1997). Learner autonomy and the need for learner training. Language
Learning Journal, 16, 34-9.

244

McLaughlin, B. (1987). Theories of second-language learning. London: Edward Arnold.

Mochida, K., & Harrington, M. (2006). The yes/no test as a measure of receptive
vocabulary knowledge. Language Testing, 23, 73-98.

Monmouth College Ofﬁce of Student Affairs (n.d.). Study skills resources: Foreign
language. Retrieved September 30, 2008 ﬁ'om http://www.monm.edu/study-
skills/foreign-language.htm

Mori, S. (2002). Redeﬁning motivation to read in a foreign language. Reading in a
Foreign Language, 14, 91-110.

Mori, S. (2004) Signiﬁcant motivational predictors of the amount of reading by EFL
learners in Japan. RELC Journal, 35, 63-81.

Noels, K.A. (2001). New orientations in language learning motivation: Towards a model
of intrinsic, extrinsic and integrative orientations and motivation. In Z. Domyei &
R. Schmidt (Eds), Motivation and second language acquisition (pp. 43-68).
Honolulu: University of Hawaii Press.

Noels, K. A., Clement, R., & Pelletier, L. G. (1999). Perceptions of teachers'
communicative style and students' intrinsic and extrinsic motivation. The Modern
Language Journal, 83, 23-34.

Noels, K.A., Clement, R., & Pelletier LG. (2001). Intrinsic, extrinsic and integrative
orientations of French Canadian learners of English. Canadian Modern Language
Review, 5 7, 424-42.

Noels, K.A., Pelletier, L. G., Clement, R., & Vallerand, R]. (2000) Why are you learning
a second language? Motivation orientation and self-determination theory.
Language Learning, 50, 57-85.

Nunnally, J.C. (1978). Psychometric theory (2"d ed.). New York: McGraw Hill.

245

Passow, A. H. (1995). Review of the Enhanced ACT Assessment. In J. C. Conoley & J.
C. Impara (Eds), The twelﬂh mental measurements yearbook (pp. 355—5 8).
Lincoln, NE: University of Nebraska Press.

Paesani, K. (2004). Using literature to develop foreign language proﬁciency: toward an
interactive classroom. In C. Stivale (Ed), Modern French Literary Studies in the
Classroom: Pedagogical Strategies (pp. 13-25). New York: Modern Language
Association.

Petros T.V., Norgaard L., Olson, K., & Tabor, L. (1989). Effects of text genre and verbal
ability on adult age differences in sensitivity to text structure. Psychology and
Aging, 4, 247-50.

Plant, E.A., Ericsson, K.A., Hill, K., & Asberg, K. (2004). Why study time does not
predict grade point average across college students: implications of deliberate
practice for academic performance. Contemporary Educational Psychology, 30,
96-116.

Proctor, C.P., August, D., Carlo, M., & Snow, C. (2005). Native Spanish-speaking
children reading in English: toward a model of comprehension. Journal of
Educational Psychology, 97, 246-56.

Pulido, D. (2004). The relationship between text comprehension and second language
incidental vocabulary acquisition: A matter of t0pic familiarity? Language
Learning, 54, 469-523.

Pulido, D., & Hambrick, D1. (2008). The virtuous circle: Modeling individual
differences in L2 reading and vocabulary development. Reading in a Foreign
Language, 20, 164-90.

Raubenheirner, J. E. (2004). An item selection procedure to maximize scale reliability
and validity. South Aﬁican Journal of Industrial Psychology, 30, 59-64.

Saito, Y., Horwitz, E., & Garza, T. (1999) Foreign language reading anxiety. The Modern
Language Journal, 83, 202-1 8.

246

Saito, Y., & Samimy, K.K. (1996) Foreign language anxiety and language performance:
A study of learner anxiety in beginning, intermediate, and advanced-level college
students of Japanese. Foreign Language Annals 29, 239-51.

Samimy, K.K. ,& Tabuse, M. (1992). Affective variables and a less commonly taught
language: A study in beginning Japanese classes. Language Learning, 42, 377-98.

Schoonen, R., Gelderen, A. van, Glopper, K. de, Hulstijn, J ., Simis, A., Snellings, P., &
Stevenson, M. (2002). Linguistic knowledge, metacognitive knowledge and
retrieval speed in L1, L2 and EFL writing. In S. Ransdell & M. Barbier (Eds),
New directions for research in L2 writing (pp. 101-122). Dordrecht: Kluwer
Academic Publishers.

Schoonen, R., Hulstijn, J ., & Bossers, B. (1998). Metacognitive and language-speciﬁc
knowledge in native and foreign language reading comprehension: An empirical
study among Dutch students gades 6, 8 and 10. Language Learning, 48, 71-106.

Scovel, T. (1978). The effect of affect on foreign language learning: A review of the
anxiety research. Language Learning, 28, 129-42.

Seattle Language Academy (n.d.). Foreign language programs FA Q. Retrieved
September 30, 2008 ﬁ'om http://www.seattlelanguageacademy.org/ﬂp/ﬂp_faq.asp

Sellers, V. (2000). Anxiety and reading comprehension in Spanish as a foreign language.
Foreign Language Annals, 33, 512-21.

Share, D., & Stanovich, KB. (1995), Cognitive processes in early reading development:
Accommodating individual differences into a model of acquisition. In J .S.
Carlson (Ed.,), Issues in education: Contributions ﬁorn psychology (V 01. 1, pp. 1-
57). Greenwich, CT: JAI Press.

Sparks, R.L., & Ganschow, L. (1991). Foreign language leaming differences: Affective
or native language aptitude differences? The Modern Language Journal, 75, 3-16.

Stanovich, KB. (1986). Matthew effects in reading: some consequences of individual

differences in the acquisition of literacy. Reading Research Quarterly, 21 , 360-
407.

247

Tang, H.N., & Moore, D.W. (1992). Effects of cognitive and metacognitive pre-reading

activities on the reading comprehension of ESL learners. Educational psychology,
12, 315-331.

Tremblay, P.F., & Gardner, RC. (1995). Expanding the motivation construct in second
language learning. The Modern Language Journal, 79, 505-18.

Tun, RA. (1989). Age differences in processing expository and narrative text. Journal of
Gerontology, 44, 9-15.

Ullrnan, J .B., & Bentler, RM. (2004). Structural equation modeling. In M.A. Hardy & A.
Bryman (Eds) Handbook of data analysis (pp. 431-58). Thousand Oaks, CA:

Sage.

University of Wollongong School of English Literatures, Philosophy & Languages (n.d.).
Ten foreign language study tips. Retrieved September 30, 2008 from
http://www.uow.edu.au/arts/selpl/ml/UOW0 l 8798.html

Vallerand, R.J., Blais, M. R., Briere, N.M., & Pelletier, LG. (1989). Construction et
validation d’une échelle de motivation en education (EME). Canadian Journal of
Behavioral Sciences, 21, 323-49.

Vallerand, R.J., Pelletier, L.G., Blais, M.R., Briére, N.M., Senécal, C.B., & Vallieres,
ER (1992). The academic motivation scale: A measure of intrinsic, extrinsic and

amotivation in education. Educational and Psychological Measurement, 52,
1003-17.

Vallerand, R.J., Pelletier, L.G., Blais, M.R., Briere, N.M., Senécal, C.B., & Vallieres,
ER (1993). On the assessment of intrinsic, extrinsic and amotivation in
education: Evidence on the concurrent and construct validity of the academic
motivation scale. Educational and Psychological Measurement, 53, 159-72.

Vandergriﬁ, L. (1997). The comprehension strategies of second language (French)
listeners: A descriptive study. Foreign Language Annals, 30, 387-409.

248

Wagner, R.K., Torgeson, J .K., & Rashotte, CA. (1994). The development of reading-
related phonological processing abilities: New evidence of bidirectional causality
from a latent variable longitudinal study. Developmental Psycholog, 30, 73-87.

Weaver, CA. 111, & Kintsch, W. (1991) Expository text. In R. Barr, M.L. Kamil, P.
Mosenthal, & P.D. Pearson (Ed), Handbook of reading research (V 01. 2, pp. 230-
246). New York: Longman.

Wen, X. (1997). Motivation and language learning with students of Chinese. Foreign
Language Annals, 30, 235-51.

Wenden, AL. (1999). An introduction to metacognitive knowledge and beliefs in
language learning: beyond the basics. system, 2 7, 435-41.

Woodrow, LJ. (2006). A model of adaptive second language learning, The Modern
Language Journal, 90, 297-319.

X_Lex: the Swansea Vocabulary Levels Test. v2.05. Swansea: Lognostics. 2005.

Yamashiro, A., & McLaughlin, J. (2000). Relationships among attitudes, motivation,
anxiety and English language proﬁciency in Japanese college students. In S.
Cornwell & P. Robinson (Eds), Individual diarcrences in foreign language
learning: Proceedings of the symposium on intelligence, aptitude and motivation
(pp. 9-26). Tokyo: Aoyama Gakuin University.

Yamashita, J. (2004). Reading attitudes in L1 and L2, and their inﬂuence on extensive
reading in L2. Reading in a Foreign Language, 16, l-19.

Yashima, T., Zenuk-Nishide, L., & Shimizu, K. (2004). The inﬂuence of attitudes and
affect on willingness to communicate and second language communication.
Language Learning, 54, 119-52.

Yazzie-Mitz, E. (2007). Voices of students an engagement: A report on the 2006 high
school survey of student engagement. Retrieved October 24, 2008, from
http://ceep.indiana.edu/hssse/pdf/HSSSE_2006_Report.pdf

249

Young, D. J. (1986). The relationship between anxiety and foreign. language oral
proﬁciency ratings. Foreign Language Annals, 19, 439-45.

Young, D. J. (1999). Linguistic simpliﬁcation of SL reading material: Effective
instructional practice? Modern Language Journal, 83 , 350-66.

Zabrucky, K. & Ratner, H.H. (1992). Effects of passage type on comprehension
monitoring and recall in good and poor readers. Journal of Reading Behavior, 3,
373-91..

250

 
   

.0.

ll!!!

 
  

27

  

306

[illiﬁllﬁllln

   

llﬂlllljﬁlll

   

I

   

7m!