System 52 (2015) 14e25

Contents lists available at ScienceDirect

System
journal homepage: www.elsevier.com/locate/system

L2 multiple-documents comprehension: Exploring the

contributions of L1 reading ability and strategic processing
Mohammad Nabi Karimi*
Department of Foreign Languages, Kharazmi University, No 43, Mofatteh Street, Tehran, Iran

a r t i c l e i n f o a b s t r a c t

Article history: There is much prior L2 research estimating the contributions of strategic processing and L1
Received 9 September 2014 reading ability to L2 reading comprehension. This line of inquiry has, however, been
Received in revised form 26 April 2015 mainly followed in relation to single-text reading. While multiple-documents compre-
Accepted 29 April 2015
hension constitutes the bulk of what EFL/ESL students do in their current academic en-
Available online 22 May 2015
vironments, L2 research investigating the contributions of these two variables to this type
of comprehension in essentially lacking. Against this background, the present paper re-
Keywords:
ports on a study examining the relative contributions of strategic processing and L1
Multiple-documents comprehension
L1 reading ability
reading ability to L2 single-text vs. multiple-texts comprehension. To this end, 114 EFL
Strategic processing students took measures of L1 reading ability, L2 single-text and multiple-texts reading
L2 readers comprehension, as well as self-reported strategic processing. Data were analyzed using
regression analyses. The results indicated that while L1 reading ability and strategic pro-
cessing contributed significantly to L2 single-text reading, only strategic processing was
shown to contribute to L2 multiple-texts reading. Strategic processing was also shown to
account for a much larger proportion of variance in L2 multiple-texts comprehension than
L2 single-text comprehension.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

There is general agreement among researchers and educators involved in L2 education that reading is a critical skill for L2
learners because they often depend on their literacy input much more than oral input (Eskey, 2005). There is also little dispute
among researchers that reading is a highly complex process involving a multitude of variables, factors and componentsdboth
reader-based and language/text-baseddthat operate and interact simultaneously to help L2 learners in the process of con-
structing meanings from texts (Barry & Lazarte, 1998; Hammadou, 2000; McNeil, 2011; Nassaji, 2003; Phakiti, 2006; among
others). Thus, both as a highly required skill and as a multi-faceted complex process, reading comprehension has been the
focus of an extensive body of research in second language education which has yielded interesting results, has provided useful
insights and has offered practical guidelines as to the more effective conceptualization and instruction of reading. One vibrant
line of this body of research has been focused around identifying the contributions, to L2 reading, of variables thought to play
significant roles in it. Variables including L1 reading ability, strategy awareness and use, prior knowledge, and L2 proficiency
are among the ones which have been extensively studied as potential determinants of L2 reading ability.

* Corresponding author. Tel.: þ98 21 88329220 3; fax: þ98 21 88306651.

E-mail address: karimi_mn@yahoo.com.

http://dx.doi.org/10.1016/j.system.2015.04.019
0346-251X/© 2015 Elsevier Ltd. All rights reserved.
 M.N. Karimi / System 52 (2015) 14e25 15

However, almost all this research has been conducted in the context of single-document comprehension which is defined
as constructing meaning from a single text providing a source for answering a series of follow-up questions. Motivated
primarily by the rapid access to a multiplicity of primary and secondary information sources provided by computerized
information systems (Anmarkrud, Bråten, & Strømsø, 2014; Bråten & Strømsø, 2011), reading demands in the current aca-
demic contexts have far outgrown those required by single-document reading. The demands have moved towards “multiple-
documents literacy [which] concerns the ability to locate, evaluate, and use diverse sources of information for the purpose of
constructing and communicating an integrated, meaningful representation of a particular issue, subject, or situation” (Bråten
& Strømsø, 2010a, p. 635). Present-day knowledge societies are, in point of fact, characterized by ready access to information
sources, a web-like knowledge structure (Bråten & Strømsø, 2006a; Goldman, 2004) and accordingly the abundance of sit-
uations where students are routinely required to construct effective links across a variety of information sources and criss-
cross a single scientific topic or issue from a variety of, sometimes contrasting, perspectives presented across diverse sources
of information (Bråten & Strømsø, 2006a).
This, then, creates a new literacy context for students in general and EFL/ESL students, in particular. Within this new
context, the ecological validity and the real-world relevancy of the findings from the single-text-dominated line of research
are doubted and questioned (Strømsø & Bråten, 2002). The reason for this questioning lies in the fact “research in this area is
[assumed to be] somewhat out of step with the intertextual reality encountering most readers in present-day society” (Bråten
& Strømsø, 2011, p. 112) as this intertextual reality calls for a different set of conditions and requirements than those
investigated within the single-text paradigm of reading research. High-quality learning in this emerging context, in point of
fact, involves far more than simply constructing meaning from separate textual resources (Bråten & Strømsø, 2011).
It, thus, seems both interesting and necessary to investigate whether some of the findings within the single-text paradigm
of reading research continue to hold in this new context of reading. Put it differently, it needs to be investigated whether
developing the capital and competency to benefit from the emerging reading literacy required in the present-day academic
contexts with its intertextual nature can be based on the body of findings from the single-text paradigm of reading research.
Two of the major variables investigated extensively within the context of single-text reading have been readers' L1 reading
ability, conceptualized as understanding single texts, and their strategic processing. L1 reading ability was selected because it
has been proposed as a variable in the compensatory model of L2 reading (Bernhardt, 2005) and strategic processing was also
selected because it has been proposed as a component in the extended compensatory model of L2 reading (McNeil, 2012).
Various studies have been conducted aiming at determining the relative contributions of these variablesdin isolation or in
conjunction with other variablesdto L2 reading comprehension performance (Fecteau, 1999; van Gelderen, Schoonen, Stoel,
Glopper, & Hulstijn, 2007; Lee & Schallert, 1997; McNeil, 2011; Phakiti, 2003; Song, 2001; among others). However, to the best
of the present researcher's knowledge, no study has been conducted to explore the relative contributions of these two
variables to L2 multiple-documents comprehension. Therefore, the present study aims to explore the relative contributions of
L1 reading ability and strategic processing to L2 multiple-documents reading comprehension compared with L2 single-text
comprehension.

2. Relevant literature

2.1. L1 reading ability and L2 reading comprehension

As noted above, L1 reading ability is one of the variables extensively studied in relation to L2 reading performance. Most
research into this issue has been conducted in the context of Linguistic Interdependence (Cummins, 1979) or Linguistic Ceiling
(Clarke, 1980) Hypotheses. According to the former hypothesis, L1 literacy is assumed to provide an effective foundation for L2
literacy. This position postulates basic resemblances between L1 and L2 language skills and strategies. It further posits that
reading in L1 and L2 are interdependent (Jiang, 2011) and argues that reading ability in a second or foreign language has a
significant shared variance with reading ability in a first language (Bernhardt & Kamil, 1995). The latter hypothesis, however,
assumes a role for L2 proficiency in L2 reading performance and development. Accordingly, it postulates that readers need to
acquire a 'threshold' level of L2 proficiency before they are able to benefit from their L1 reading ability in their L2 reading
performance (Jiang, 2011). These two hypotheses have, then, motivated a number of studies investigating the contributions of
L1 reading ability to L2 reading performance, though mostly in conjunction with L2 proficiency.
Yamashita (2002), for example, examined the relative contributions of L1 reading ability and L2 language proficiency and
the mutual compensatory effects of these two variables on L2 reading comprehension performance. Participants of the study
were 241 Japanese university students who took tests of L2 proficiency, and L1 and L2 reading comprehension. Results of the
regressions analysis revealed that both variables strongly predicted L2 reading comprehension. Language proficiency was,
however, shown to be a stronger predictor of L2 reading comprehension than L1 reading ability. The results further showed a
mutual compensation between the two variables so that a small increase in L2 proficiency compensated for a significant
decline in L1 reading ability.
Pichette, Segalowitz, and Connors (2003) also investigated the relationship between L2 proficiency, L1 reading ability and
L2 reading comprehension. The study was conducted with a total of 52 L1-Serbo-Croatian learners of L2 French who took two
cloze tests of Serbo-Croatian and two cloze tests of French as measures of L1 and L2 reading ability, respectively, as well as a
measure of French knowledge. Results of the regression analysis with L2 reading ability as the criterion variable and
knowledge of L2 French and L1 reading ability as predictor variables revealed that together these two variables accounted for
16 M.N. Karimi / System 52 (2015) 14e25

44% of the variance in L2 reading comprehension. L2 proficiency was reported to be a stronger predictor explaining 53% of the
variance while L1 reading ability accounted for 18% of the variance.
In another similar study, Tsai, Ernst, and Talley (2010), as one part of their study, investigated the relative contributions of
L1 reading ability and second language proficiency to L2 reading compression ability. The study participants were a total of
222 Chinese-speaking EFL learners. They took measures of L1 and L2 reading comprehension, as well as L2 grammar and
vocabulary as indices of L2 proficiency. Correlations were calculated between the variables and the results suggested sig-
nificant correlations of L2 reading comprehension with L1 reading ability and L2 proficiency. L2 proficiency and L2 reading
comprehension were, however, more highly correlated.
In another study, Jiang (2011) investigated the relationship between L1 reading ability, L2 proficiency and L2 reading
comprehension. The study participants were 246 Chinese non-English-major college students. College admission exams were
used to measure L1 literacy and L2 proficiency while L2 reading comprehension was measured using the reading section of a
TOEFL test and a reading comprehension test constructed by the researcher. Results of the regression analyses revealed that
L2 proficiency explained around 27%e39% of the variance in L2 reading comprehension, while L1 literacy explained around 6%
of the variance.
Although much more research has been conducted on the relative contribution of L1 reading ability to L2 reading
comprehension (e.g. Bossers, 1991; Brisbois, 1995; Carrell, 1991; van Gelderen, et al., 2007; Lee & Schallert, 1997), the present
review has focused on some illustrative studies to ground the research reported in this paper. These studies highlight the role
of L1 reading ability as a predictor of L2 reading comprehension, although as a weaker predictor than L2 proficiency.

2.2. Strategic processing and L2 reading comprehension

Research on the link between strategic thinking/processing and reading comprehension has been a fixture of research on
L2 reading. Research on this issue has been either focused around identifying strategies often employed by more proficient vs.
less proficient readers (e.g. Brantmeier, 2005; Ikeda & Takeuchi, 2006; Sheorey & Mokhtari, 2001; Upton & Lee-Thompson,
2001; Yang, 2002) or concentrated on promoting L2 reading comprehension as a function of strategy instruction (e.g. Dhieb-
Henia, 2003; Dreyer & Nel, 2003; Salataci & Akyel, 2002; Zhang, 2008).
In addition to this research base, a number of studies have also investigated the contribution of strategic thinking/pro-
cessing to L2 reading comprehension. As an example, Schoonen, Hulstijn, and Bossers (1998) examined the relative contri-
butions of vocabulary knowledge, as a language-specific predictor, and general metacognitive knowledge to L1 and L2 reading
comprehension among a total of 416 students in grades 6, 8 and 10 in Netherlands. Strategy knowledge was measured
through a questionnaire covering four domains of metacognitive knowledge including (a) assessment of oneself as a reader,
(b) knowledge of reading goals and comprehension criteria, (c) knowledge of text characteristics and (d) knowledge of
reading strategies. Dutch and English reading comprehension were also measured through standardized multiple-choice
tests developed by Dutch National Institute for Educational Measurement. Vocabulary knowledge in Dutch and English
were measured in a multiple-choice and translation format, respectively. Results of the covariance structure analyses revealed
that L1 vocabulary knowledge accounted for 60% of variance in reading comprehension ability at grade 6, while at this level
metacognition explained only 4% of the variance. At grade 10, L1 vocabulary's role weakened, explaining 48% of reading
performance and metacognition played a larger role, accounting for 17% of reading comprehension scores. L2 vocabulary
knowledge accounted for 71% of variance in reading comprehension ability at grade 8, while metacognition accounted for 5%
of the variance. For reading comprehension ability at grade 10, L2 vocabulary explained 35% of variance in reading
comprehension while metacognition explained 25% of the variance. The results further showed L1 reading ability to be a
stronger predictor of L2 reading comprehension along with an increase in students' proficiency.
Phakiti (2003) also examined the relationship between EFL learners' cognitive and metacognitive strategy use and their
performance in reading comprehension tests. Data were collected from 384 Thai EFL learners through a questionnaire
assessing cognitive and metacognitive strategies, retrospective interviews and a multiple-choice English reading compre-
hension achievement test. Results of the study showed that cognitive and metacognitive strategies were positively correlated
with the participants' reading comprehension test performance, accounting for 15e22% of the variance in the reading test
scores. The results also revealed that the more successful readers used a greater number of metacognitive strategies than the
less successful readers.
In another study, van Gelderen et al. (2007) investigated the contributions of language-specific factors and metacognition
to L1 and L2 reading development. The participants were 389 L1-Dutch learners of English as a foreign language studying in
grades 8 through 10. Data were collected regarding students' word recognition, processing speed, knowledge of vocabulary
and grammar, as well as L1 and L2 reading comprehension. The participants' metacognitive knowledge was also measured
through a 54-item questionnaire covering reading and writing strategies and text characteristics. Using structural equation
modeling, the researchers analysed the data. The first structural model revealed that metacognitive knowledge accounted for
43% of the variance in L1 reading ability and 41% of the variance in L2 reading ability for students in grade 8. Metacognition
became a stronger predictor in subsequent years, accounting for up to 72% of the variance in L2 reading ability.
In a more recent study, McNeil (2011) investigated the combined and individual contributions of background knowledge
and comprehension strategies to L2 reading comprehension. In the study, comprehension strategy was operationalized as
self-questioning ability. Data were collected from 20 foreign students enrolled in an Intensive English Program at an American
university and were subjected to a couple of multiple regression analyses. The first regression model showed that the
 M.N. Karimi / System 52 (2015) 14e25 17

combination of the two variables investigated explained 56.7% of the variance in L2 reading comprehension. As background
knowledge was shown to explain only a small portion of the variance in L2 reading comprehension scores, it was excluded in
the second regression model. Self-questioning alone was shown to account for 56.3% of the variance in L2 reading
comprehension. Therefore, it was concluded that comprehension strategies, operationalized as self-questioning, was a far
stronger predictor of L2 reading comprehension than background knowledge.
As shown in the studies reported here and other studies not reviewed here (e.g. Phakiti, 2008; Taboada & Guthrie, 2006),
the general conclusion from this line of research points to the significance of strategic thinking/processing in predicting L2
reading comprehension.

2.3. Multiple-documents comprehension and strategic processing

Research on multiple-documents comprehension has been conducted mostly outside L2 reading contexts; therefore, the
role of L1 reading ability in this type of comprehension has not been investigated. However, there is a small body of research
investigating strategy use and effectiveness in multiple-documents comprehension (Anmarkrud, et al., 2014; Karimi, 2015;
Karimi & Shabani, 2013; Bråten & Strømsø, 2003; Kobayashi, 2009, 2010; Stadtler & Bromme, 2008; among others). The
general conclusions from this body of research point to the importance of strategic processing in multiple-documents
comprehension.
For example, Bråten and Strømsø (2003), as an early attempt at exploring strategy use in multiple-documents compre-
hension, examined the strategies university students used as they read multiple self-selected expository documents. The
participants were seven Norwegian law students who self-selected the documents as supporting materials for the textbooks
they covered for their civil law course. Specifically, the study aimed to examine the participant readers' text processing
strategies in their naturalistic academic reading contexts during one academic semester; therefore, the participants read the
self-selected documents for various purposes ranging from reading the texts to catch up with their course lectures to
reviewing the documents for examination purposes. Think-alouds as well as audio- and video-tapes of the reading sessions
were used for the purpose of data collection. Results of the study showed that the participant readers used elaboration and
monitoring strategies more frequently while reading multiple texts and that there were changes in the participants' strategic
processing of the documents according to changes in their perceptions of the nature of the reading tasks.
In another study, Kobayashi (2009) investigated how personal factorsdtopic knowledge, external strategy use and college
experiencedaffected the comprehension of intertextual relations among multiple controversial texts. The study participants
were 166 Japanese undergraduate students of education who read two controversial newspaper articles presenting con-
flicting views on the introduction of English into public elementary schools. The participants received these two articles in
conjunction with a sheet of paper and were informed that they were free to, if necessary, highlight or take notes of portions of
the texts. An intertextual relation task was used to measure the participants' comprehension of intertextual relations across
the documents. To measure the participants’ use of external strategies, their sheets of texts and notes papers were coded for
categories of external representations including highlighting, explanatory notes, personal ideas, summary notes, and inter-
textual notes. The results of the study revealed that out of the five categories of external representations they produced while
reading the documents, summary notes had direct and indirect effects on the comprehension of intertextual arguments.
Anmarkrud et al. (2014) also examined readers' strategic processing while reading multiple documents about “cell phone
signals and their associated health damages”. The study participants were 51 Norwegian undergraduate students who read
six documents about cell phones and their potential health risks. After reading the documents, the participants were required
to rate their trustworthiness and develop brief essays on the topic of the documents. Think-aloud methodology was used to
document the strategic thoughts readers used while reading the multiple documents. The analysis of the think-alouds
revealed that most participants were actively involved in strategic processing while reading the multiple documents.
Identifying and learning important information, monitoring and evaluating were the three categories of strategies most
participants readers used while dealing with the multiple documents.
In a more recent study conducted in an L2 context, Karimi and Shabani (2013) compared the differences in the strategic
processing of multiple documents between more successful vs. less successful readers of technical reading texts. The re-
searchers administered an Intertextual Inference Verification Task as a measure of reading comprehension of three technical
texts to 81 midwifery students. While reading the multiple texts, the readers were engaged in thinking aloud and reporting
their thoughts. Additionally, a questionnaire was given to the participants to assess their perceived use of reading strategies
while reading the texts. Comparisons of the think-aloud protocols of the 15 highest-scoring and the 15 lowest-scoring par-
ticipants and their responses to the items on the questionnaire revealed significant differences in the strategic processing of
the two groups.

3. The present study

As the review of the studies reported above shows, the role of strategic processing in multiple-documents has been
highlighted in the literature. However, as stated earlier, no study has been conducted on the relative contribution of strategic
processing in multiple-documents comprehension compared with single-text comprehension. The present study, therefore,
aims to investigate this issue. Specifically, the following research question is addressed in the present study:
18 M.N. Karimi / System 52 (2015) 14e25

What are the relative contributions of L1 reading ability, conceptualized as comprehending single texts, and strategic
processing to L2 single-text reading comprehension compared with L2 multiple-documents reading comprehension?

4. Method

4.1. Participants

The study participants were 114 upper-intermediate to advanced Iranian students learning English as a foreign language at
three private language institutes. They were made up of 75 female (65.79%) and 39 male (34.21%) students. The participants'
ages ranged from 17 to 28 with an average of 23.4 years and an SD of .84. All students had studied English for at least two years
in the institutes in addition to the English courses they had passed during high-school and university years. The rather broad
age range of the participants allowed for a broad spectrum of L1 reading abilities, a point suited to the purpose of the present
study.

4.2. Instruments

4.2.1. Intertextual inference verification task

Following the lead of earlier studies on multiple-documents comprehension (e.g. Bråten & Strømsø, 2010b; Bråten,
Strømsø, & Samuelstuen, 2008), an Intertextual Inference Verification Task was used to assess the participants' multiple-
texts comprehension. The task was based on four separate texts about 'Autism'. A 1562-word text was adapted from an
informational website. This text defined autism and discussed the common symptoms associated with the condition, the
common causes, diagnosis, treatment options, the role of inheritance in the condition, etc. The second document was a 788-
word text discussing the common causes, treatment, symptoms and Autism Spectrum Disorder. The third document was an
849-word popular science text describing causes of autism, signs associated with the condition, diagnosis, family care in
autism, etc. The fourth document was also a 1480-word popular science text which discussed the Autistic Spectrum Disorder,
the causes of the disease, and the most frequent characteristics of the people with the disease. The documents had a fair share
of overlapping information but they varied from the viewpoint of the approach to the topic, the terminologies used and the
different aspects of the condition they emphasized. This required the participant readers to integrate and synthesize ideas
across the documents to be able to respond to the questions. The documents were printed on separate sheets of paper and the
participant readers were free to read them in any order they preferred.
The task was comprised of 20 sentences constructed by integrating pieces of information across the documents to form an
either valid or invalid inference. The participants were asked to mark the sentence as valid if the idea expressed by it could be
inferred from integrating pieces of information across the multiple documents or invalid if the idea expressed by the sentence
could not be inferred by synthesizing ideas from the documents. For example, the following item is an example of a valid
inference which should be drawn by synthesizing bits of information from all the four texts:
In pervasive developmental disorder, which is a mild form of autism, the person may seem awkward in interacting with
others and delayed is some other basic skills but in classical ASD people may have mental retardation.
In comparison, the following item is a sample invalid inference which should be drawn by linking ideas across three of the
texts.
Although research has shown that both a child's surroundings and inheritance are influential in autism, the role of envi-
ronment is reported to be played down in presence of genetic disposition.
The participants were required to mark the sentences as right or wrong and their scores equaled the number of correct
discriminations they made. The task included 15 valid inferences and 5 invalid inferences. It should also be pointed out that, to
control for the effects of memory on the participants' performance, the documents were available while they answered the
items. The reliability of the measure with the present sample was calculated to be .78.

4.2.2. L2 intratextual inference verification task

Based on the fourth document which was discussed above, an Intratextual Inference Verification Task was developed. The
task was comprised of 10 statements constructed through combining bits of information across sentences in the text to form
an either valid or invalid inference. The participants marked the statement as valid if the idea expressed by it could be inferred
from synthesizing information bits across sentences in the text and invalid if the idea in the statement could not be inferred
by combining ideas across the sentences in the text. For example, the following item is an example of a valid inference:
Autism is a brain-related condition which may have common genetic links with schizophrenia and negatively affects suc-
cessful communication of attention-getting cues during interactions.
In comparison, the following item is a sample invalid inference:
That autism is described as a disorder on a wide spectrum indicates that people afflicted with the condition have a wide
range of similar characteristics.
 M.N. Karimi / System 52 (2015) 14e25 19

As in the intratextual inference verification task, the participants were required to mark the statements as either right or
wrong and their scores equaled the number of correct discriminations they made. The task included 6 valid inferences and 4
invalid inferences. The reliability of the measure with the present sample was calculated to be .82.

4.2.3. Self-reported reading strategies inventory

As the study was a large-scale one, the researcher used a follow-up task-specific self-reported strategy inventory to
investigate the strategies participant readers used while reading both the single and the multiple documents. The idea came
from Bråten and Strømsø (2011) who believe that self-report inventories are more relevant and useful compared with think-
alouds when the aim is to carry out large-scale investigations. Additionally, Samuelstuen and Bråten (2005) found the scores
on self-reported strategy inventories referring to recently completed reading tasks to better account for reading performance
than general self-report inventories eliciting general information on strategy use across tasks and situations. Therefore, items
on the Metacognitive Awareness of Reading Strategies Inventory (MARSI) developed and validated by Mokhtari and Reichard
(2002) were specifically tailored to the reading tasks used in the studydboth single-text reading and multiple-texts reading.
The inventory, in its original form, includes 30 items rated along a five-point Likert scale ranging from 1 indicating “I never or
almost never do this” to 5 indicating “I always or almost always do this”. The measure assesses strategies in three broad cat-
egories including Global Strategies, Problem-solving Strategies, and Support Strategies. In the tailored versions, items related
to “using reference materials”, “using tables, figures and pictures”, and “discussing the text content with others” were dis-
carded as they did not relate to the two reading tasks used in the present study. Therefore, there remained 27 items in the
tailored versions. A sample item tailored to the two reading tasks is presented below:
Original item: I summarize what I read to reflect on important information in the text.
Item tailored to the single-text reading task: While reading this text, I summarized what I read to reflect on important in-
formation in it.
Item tailored to the multiple-texts reading task: While reading these multiple texts, I summarized what I read to reflect on
important information in them.
The Cronbach Alpha reliability indices of the tailored versions were .77 and .79 for the single-text and multiple-texts
reading, respectively.

4.2.4. L1 reading comprehension measure

To assess the participants' L1 reading comprehension, a Persian reading comprehension test was developed. The test was
based on a 1044-word text about advances in treatment options for Attention Deficit Hyperactivity Disorder (ADHD). The text
discussed the increasing number of ADHD cases over the past 10 years and the treatment options for the condition including
the stimulant medications, non-stimulant medications and antidepressants. The test developed based on this text included 8
multiple-choice items with four alternatives. The items were scored as either correct or incorrect and each participant's score
equaled the number of correct responses to the items as one point was given to each correct answer and zero points to each
incorrect answer. The Cronbach Alpha reliability of the test with the study sample was calculated to be .83.

4.3. Procedures

As seen above, there were three measures of reading ability. To reduce test fatigue, the participants took these measures in
two separate testing sessions. In the first session, the L1 reading comprehension measure and the L2 intratextual inference
verification task were administered to the participants. Immediately, after the L2 intratextual inference verification task, the
participants responded to the items on the self-reported reading strategies inventory tailored to the single-text reading task.
The time allocated to each of these tasks was 30 min. In the second session, the participants took the intertextual inference
verification task along with the self-reported reading strategies inventory tailored to the multiple-texts reading compre-
hension task. The time allocated to this task was 90 min. The scores from the L1 reading ability measure, inference verification
tasks and the self-reported strategy inventories provided the data for the purpose of investigating the relative contributions
of L1 reading ability and strategic processing to the participants' multiple-texts reading comprehension compared with their
single-text reading comprehension which were analysed through multiple regression models.

5. Results

As stated earlier, the major aim of the study was to determine the relative contributions of L1 reading ability and strategic
processing to EFL students' single-text vs. multiple-texts comprehension. The scores from the L1 reading comprehension
measure, the L2 intratextual inference verification task and the self-reported strategy inventory tailored to single-text reading
provided the data for first regression model aimed at determining the contribution of the two predictor variables to L2 single-
text reading. Table 1 presents the means and standard deviations of the scores from these three measures:
As seen above, the mean score (out of 8) for the L1 Reading Comprehension Measure was 6.66 (SD ¼ .98); the mean score
(out of 10) for the L2 intratextual inference verification task was 6.06 (SD ¼ .99) and the mean score for the self-reported
strategy inventory was 52.50 (SD ¼ 11.31). To investigate the combined contribution of L1 reading ability and self-reported
strategic processing, the scores from the above measures were submitted to a multiple regression model. Prior to the
main regression model, preliminary analyses were conducted to ensure that the assumptions of regression (e.g. linearity
20 M.N. Karimi / System 52 (2015) 14e25

Table 1
Means and standard deviations of scores on the variables investigated in regression model 1.

Mean Std. deviation

L1 reading comprehension measure 6.66 .98
L2 intratextual inference verification task 6.06 .99
Single-text self-reported strategy 52.50 11.31

between the predictor and the dependent variables, absence of multicollinearity, outliers, normality, and absence of residual
autocorrelation) are met. Before presenting the regression model, the matrix of correlations among the variables in regression
model one is presented in Table 2.
As shown in Table 2, there are rather low correlations between L1 reading ability and the intratextual inference verification
task (r ¼ .36) and between single-text strategic processing and the intratextual inference verification task (r ¼ .32). Then, a
linear multiple regression analysis was conducted the results of which follow:
As seen in Table 3, the results of the first regression model indicate a significant effect for L1 reading ability and self-
reported strategic processing on L2 single-text reading comprehension (F ¼ 11.497, p < .05). The coefficient of determina-
tion of the regression model (R2) shows that the combination of these two variables predicts 15.7% of the variance in L2 single-
text reading comprehension which indicates that there are some other variables likely to predict variance in this type of
comprehension but are not included in the present regression model. As shown in the table, the Beta coefficient for self-
reported strategic processing is .311 and that for L1 reading ability is .212 which are both statistically significant at the .05
level (p < .05). Thus, it can be concluded that both of the predictor variables are crucial to L2 single-text reading compre-
hension and a higher share of each of these variables results in a better performance in L2 single-text reading comprehension.
The Beta-weight for strategic processing is, however, higher than that for L1 reading ability which indicates that strategic
processing is a stronger predictor of single-text reading comprehension than L1 reading ability.
The next regression model aimed at determining the relative contributions of L1 reading ability and self-reported strategic
processing to L2 multiple-documents comprehension. The scores from the intertextual inference verification task, the L1
reading comprehension measure and the self-reported strategy inventory tailored to multiple-texts reading provided the
data for this regression model. As in the first regression model, preliminary analyses were conducted to determine whether
the assumptions of regression are met. Before presenting the regression model, the means and standard deviations of the
scores used in the model (Table 4) as well as the matrix of correlations among the variables in this model are presented below:
(Table 5)
As shown in Table 5, there is a negligible correlation (r ¼ .06) between L1 reading ability and the intertextual inference
verification task but there is a moderate correlation (r ¼ .49) between multiple-texts-based self-reported strategic processing
and intertextual inference verification task.
Additionally, as reported in Table 4, the mean score (out of 8) for the L1 reading comprehension measure was 6.66
(SD ¼ .98); the mean score (out of 20) for the intertextual inference verification task was 8.94 (SD ¼ 1.40) and the mean score
for the multiple-texts self-reported strategy inventory was 62.76 (SD ¼ 11.95). A comparison of the means for the single-text
reading comprehension presented in Table 1 and the mean for multiple-texts comprehension presented here reveals the
more difficulty of the latter type of comprehension. Moreover, the Pearson-product moment correlation between the two
types of comprehension was found to be low (r ¼ .33, p < .05). A multiple regression analysis was then conducted the results
of which follow:
As reported in Table 6, the results of the regression model reveal a statistically significant effect for the combination of L1
reading ability and self-reported strategic processing on L2 multiple-text reading comprehension (F ¼ 19.377, p < .05). The R
Squared value of the regression model shows that the combination of these two variables accounts for 24.5% of the variance in
L2 multiple-text reading comprehension. The Beta coefficient for the two variables, moreover, indicated that strategic pro-
cessing explained a higher share of the variance in the model and was significant (t ¼ 6.175, p < .05) while the Beta coefficient
for L1 reading ability was shown to be insignificant (t ¼ 1.496, p ¼ .137). Thus, it can be concluded that only strategic pro-
cessing could significantly contribute to L2 multiple-texts reading comprehension. Because L1 reading ability was not shown
to be a significant predictor of L2 multiple-texts comprehension, it was excluded in the regression model and another
regression model was conducted with self-reported strategy scores as the only predictor variable. The results are reported in
Table 7 below:
As reported in Table 7, strategic processing, as the lone variable in the regression model, explained 23.7% of the variance in
multiple-text reading comprehension scores which was statistically significant (t ¼ 6.010, p < .05). Since the first regression
model with both L1 reading ability and strategic processing as the independent variables explained 24.5% of the variance in
multiple-texts comprehension and this latter model explained 23.7% of the variance, it becomes clear that L1 reading ability
explained only .8% of multiple-texts reading comprehension.

6. Discussion

This study examined the relative contributions of L1 reading ability and strategic processing to L2 single-text vs. multiple-
texts reading comprehension. The major finding of the study is that the combined and individual contributions of these two
 M.N. Karimi / System 52 (2015) 14e25 21

Table 2
Matrix of correlations among variables in regression model 1.

1 2 3
1. L1 reading Comprehension e .20 .36
2. Single-text self-reported strategy e e .32
3. L2 intratextual inference verification task e

Note: All correlations based on n ¼ 114.

Table 3
Multiple regression 1: single-text reading comprehension as a function of L1 reading ability and self-reported strategic processing.

(N ¼ 114)

Source df SS MS F R2 SE of estimate p
Regression 2 19.318 9.659 11.497 .157 .916 < .05
Residual 113 93.252 .840

Coefficients B SE Beta T p
Constant 3.172 .653 4.855 < .05
Strategic processing .028 .008 .311 2.508 < .05
L1 reading ability .215 .090 .212 2.388 < .05

Table 4
Means and standard deviations of scores on the variables investigated in regression model 2.

Mean Std. deviation

L1 reading comprehension measure 6.66 .98
Intertextual inference verification task 8.94 1.40
Multiple-texts self-reported strategy 62.76 11.95

Table 5
Matrix of correlations among variables in regression model 2.

1 2 3
1. L1 reading comprehension e .11 .06
2. Multiple-texts self-reported strategy e .49
3. L2 intertextual inference verification task e

Note: All correlations based on n¼114.

Table 6
Multiple regression 2: multiple-texts reading comprehension as a function of L1 reading ability and self-reported strategic processing.

(N ¼ 114)

Source df SS MS F R2 SE of Estimate p
Regression 2 57.368 28.684 19.377 .245 1.216 <.05
Residual 113 164.316 1.480

Coefficients B SE Beta T p
Constant 4.044 1.048 3.859 <.05
L1 reading ability .175 .117 .123 1.496 .137
Strategic processing .060 .010 .508 6.175 <.05

Table 7
Multiple regression 3: multiple-texts reading comprehension as a function of self-reported strategic processing.

(N ¼ 114)

Source df SS MS F R2 SE of Estimate p
Regression 1 54.054 54.054 36.116 .237 1.223 <.05
Residual 112 167.630 1.497

Coefficients B SE Beta T p
Constant 5.317 .615 8.649 <.05
Strategic processing .058 .010 .494 6.010 <.05
22 M.N. Karimi / System 52 (2015) 14e25

variables vary across the two types of reading comprehension. While L1 reading ability and strategic processing play sig-
nificant roles as predictors of L2 single-text reading ability, only strategic processing was shown to significantly contribute to
L2 multiple-texts reading comprehension. L1 reading ability, conceptualized as understanding single texts, failed to
contribute significantly to L2 multiple-documents comprehension. The results extend L2 research in this area by comparing
the contributions of these two variables to two types of comprehension as thus far, the bulk of research relevant in this area
has been focused around understanding single texts.
Additionally, estimating the contributions of L1 reading ability and strategic processing to single-text reading compre-
hension extend the current research in this area in two ways. First, that L1 reading ability accounted for 21.2% of the variance
in L2 single-text reading comprehension scores lends support to the assumption that L1 literacy explains 14%e21% of the
variance in L2 reading comprehension (Bernhardt, 2005; Bernhardt & Kamil, 1995). The interesting point about this finding is
that it comes from Farsi, a non-cognate counterpart of English, with a substantial orthographic distance between the two
languages. The results also lend partial support to the assumption that L1 reading ability accounts for a larger proportion of
the variance in L2 reading comprehension along with an increase in students' proficiency (Bernhardt & Kamil, 1995; Fecteau,
1999; van Gelderen et al., 2007; Pichette et al., 2003; Schoonen, et al., 1998). Although this assumption was not explicitly
tested in the present study as the contribution of L1 reading ability was not compared across proficiency levels, the conclusion
could be drawn from the participants' rather high e upper-intermediate to advancedelevels of proficiency.
That strategic processing also accounted for 31.1% of the variance in L2 reading comprehension scores renders support for
strategic knowledge/processing as a potential component in explaining the amount of unexplained variance in Bernhardt's
(2005) compensatory model of L2 reading comprehension (McNeil, 2012). Bernhardt considers her compensatory model of
reading to be composed of three core components including L1 literacy, L2 language knowledge and unexplained variance.
She proceeds to list reading strategies and background knowledge as other sources which could potentially explain the
unexplained variance in the model. The results also verify the findings of those studies which consider strategic knowledge/
processing as explaining a substantial variance of L2 reading (McNeil, 2011; van Gelderen et al., 2007; Phakiti, 2008) although
the proportion of variance it accounted four was not as high as those reported by some of the earlier research such as 56.3%
reported by McNeil (2011) or up to 72% reported by van Gelderen et al. (2007).
Additionally, the study, as stated earlier, points to the differential contributions of L1 reading ability and strategic pro-
cessing to single-text reading comprehension compared with multiple-texts reading comprehension. L1 reading ability,
conceptualized as understanding a single textdin tune with the prior research conducted in this aread failed to contribute to
L2 multiple-texts comprehension. In fact, less than 1% of the variance in multiple-texts reading was shown to be attributed to
L1 reading ability after its exclusion from the regression model. Moreover, while L1 reading literacy and strategic processing,
in conjunction, accounted for 15.7% of the variance in single-text reading, strategic processing alone could explain 23.7% of the
variance in multiple-texts comprehension with a Beta value of 49.4. The differential contributions of these two variables to
multiple-texts reading appear to indicate that it could be considered a qualitatively different reading literacy than single-text
understanding. Earlier research in this area has, in point of fact, shown that multiple-documents literacy is a more resource-
demanding literacy requiring the employment of more sophisticated higher-order processes and skills than understanding
separate textual resources (Bråten & Strømsø, 2006b; Kobayashi, 2010; Strømsø & Bråten, 2002). More intentional strategic
effort is required to construct a coherent mental representation across a series of documents (Bråten & Strømsø, 2011)
because this representation goes a step above the text-base and the situational representations (Kintsch, 1998) of a single text.
To fully comprehend a set of texts on the same topic, the text-base and the situational representation of each text should be
combined with those of other accompanying texts which results in what is called a “documents model” (Britt, Perfetti,
Sandak, & Rouet, 1999) which is a more “highly integrated situation model of the events described in the texts” (Bråten &
Strømsø, 2010b, p. 3). The substantial role of strategic processing in this process is best expressed in Bråten and Strømsø
(2011, p. 112):
…mental representations constructed from working with multiple texts can be regarded as authored by the readers
themselves, more than by the authors of the individual texts. Therefore, it stands to reason that when students try to
build an integrated understanding by reading multiple texts on a particular topic, more strategic effort is required than
when they try to understand a single text on the same topic.
The more difficult and the distinct nature of multiple-documents comprehension is also supported by the participants'
lower mean score in it and its low correlation with single-text comprehension which lends further support to the fact that the
two could be independent levels of comprehension. Therefore, it stands to reason to find variations in the contributions of L1
reading ability, in the sense thus far investigated in the relevant literaturedconceptualized as understanding single
textsdand strategic processing across L2 single-text and L2 multiple-texts comprehension.

7. Conclusion

The results of the present study indicate that two of the variables investigated as potential sources of variance in L2
reading comprehension research contribute differentially to multiple-documents comprehension. Given that this type of
reading literacy comprises the overwhelming bulk of what EFL/ESL students do in their present-day academic contexts, it
appears that the body of findings outlining the sources of variance in single-text reading may not truly reflect the nature of
this emerging reading literacy. Research on the potential sources of variance in L2 reading has treated reading mostly as a
 M.N. Karimi / System 52 (2015) 14e25 23

“global construct rather than delving into its multi-level representational architecture” (Alptekin & Erçetin, 2010, p. 206)1.
Results of the present study offer insights as to the fact that the variance in multiple-texts comprehension, being a higher-
level reading representation, appears to be differentially explained by the variables mostly investigated in relation to L2
reading which has, thus far, been conceptualized as understanding single texts. Therefore further research is required to
establish a body of findings to help effectively benefit from the emerging reading literacy in the current knowledge societies.

Acknowledgments

The author would like to express his special thanks to the participants in the study. Moreover, the author would like to
extend his appreciation to the three anonymous reviewers and the editors of System for their constructive comments on
earlier drafts of the paper.

Appendix A

Listed below are statements about what you might have done while reading the text/texts. After reading the each item,
indicate your level of agreement by circling one of the numbers which follow it.

1 ¼ I never or almost never did this.

2 ¼ I did this only occasionally.
3 ¼ I sometimes did this.
4 ¼ I often did this.
5 ¼ I always or almost always did this.
1. I had purposes in mind when I read the text./I had purposes in mind when I read the multiple texts.
2. I took notes while reading the text to help me understand what I read./I took notes while reading the multiple texts
to help me understand what I read.
3. While reading this text, I summarized what I read to reflect on important information in it./While reading these
multiple texts, I summarized what I read to reflect on important information in them.
4. I tried to get back on track when I lost concentration while reading the text./I tried to get back on track when I lost
concentration while reading the multiple texts.
5. I underlined or circled information in the text to help me remember it./I underlined or circled information in the
texts to help me remember it.
6. I used context clues to help me better understand what I was reading in the text./I used context clues to help me
better understand what I was reading in the multiple texts.
7. I paraphrased (restated ideas in my own words) to better understand what I read from the text./I paraphrased
(restated ideas in my own words) to better understand what I read from the multiple texts.
8. I guessed the meaning of unknown words by separating different parts of a word while reading the text./I guessed
the meaning of unknown words by separating different parts of a word while reading the texts.
9. I thought about what I know to help me understand what I read from the text./I thought about what I know to help
me understand what I read from the texts.
10. I previewed the text to see what it was about before reading it./I previewed the multiple texts to see what they were
about before reading them.
11. When the text became difficult, I read aloud to help me understand what I read./When the texts became difficult, I
read aloud to help me understand what I read.
12. While reading the text, I thought about whether the content fit my reading purpose./While reading the multiple
texts, I thought about whether the content fit my reading purpose.
13. While reading the text, I read slowly but carefully to be sure I understand what I was reading./While reading the
texts, I read slowly but carefully to be sure I understand what I was reading.
14. I skimmed the text first by noting characteristics like length and organization./I skimmed the texts first by noting
characteristics like length and organization.
15. While reading the text, I adjusted my reading speed according to what I was reading./While reading the text, I
adjusted my reading speed according to what I was reading.
16. While reading the text, I decided what to read closely and what to ignore./While reading the multiple texts, I
decided what to read closely and what to ignore.
17. When the text became difficult, I pay closer attention to what I was reading./When the texts became difficult, I pay
closer attention to what I was reading.
18. While reading the text, I stopped from time to time and think about what I was reading./While reading the multiple
texts, I stopped from time to time and think about what I was reading.

1
Alptekin and Erçetin state this in the context of research on working memory capacity and reading comprehension. However, I quoted it because I
found this to be also true about earlier research on the sources of variance in reading comprehension.
24 M.N. Karimi / System 52 (2015) 14e25

19. While reading the text, I tried to picture or visualize information to help remember what I was reading./While
reading the multiple texts, I tried to picture or visualize information to help remember what I was reading.
20. While reading the text, I used typological aids like boldface and italics to identify key information./While reading
the multiple texts, I used typological aids like boldface and italics to identify key information.
21. I critically analyzed and evaluated the information presented in the text./I critically analyzed and evaluated the
information presented in the multiple texts.
22. While reading the text, I went back and forth to find relationship among ideas in it./While reading the multiple
texts, I went back and forth to find relationship among ideas in them.
23. While reading the text, I checked my understanding when I came across conflicting information./While reading the
multiple texts, I checked my understanding when I came across conflicting information./
24. I tried to guess what the material was about when I read the text./I tried to guess what the material was about when
I read the multiple texts.
25. When the text became difficult, I reread to increase my understanding./When the texts became difficult, I reread to
increase my understanding.
26. While reading the text, I asked myself questions I liked to have answered in the text./While reading the multiple
texts, I asked myself questions I liked to have answered in the text.
27. While reading the text, I checked to see if my guesses about the text were right or wrong./While reading the
multiple texts, I checked to see if my guesses about the text were right or wrong.

References

Alptekin, C., & Erçetin, G. (2010). The role of L1 and L2 working memory in literal and inferential comprehension in L2 reading. Journal of Research in
Reading, 33, 206e219.
Anmarkrud, Ø., Bråten, I., & Strømsø, H. I. (2014). Multiple-documents literacy: strategic processing, source awareness, and argumentation when reading
multiple conflicting documents. Learning and Individual Differences, 30, 64e76.
Barry, S., & Lazarte, A. (1998). Evidence for mental models: how do prior knowledge, syntactic complexity, and reading topic affect inference generation in a
recall task for nonnative readers of Spanish? Modern Language Journal, 82, 176e193.
Bernhardt, E. B., & Kamil, M. L. (1995). Interpreting relationships between L1 and L2 reading: consolidating the linguistic threshold and the linguistic
independence hypothesis. Applied Linguistics, 16, 15e34.
Bernhardt, E. B. (2005). Progress and procrastination in second language reading. Annual Review of Applied Linguistics, 25, 133e150.
Bossers, B. (1991). On thresholds, ceilings, and short circuits: the relations between L1 reading, L2 knowledge and L1 knowledge. AILA Review, 8, 45e60.
Brantmeier, C. (2005). Effects of reader's knowledge, text type, and test type on L1 and L2 reading comprehension. The Modern Language Journal, 89, 37e53.
Bråten, I., & Strømsø, H. I. (2003). A longitudinal think-aloud study of spontaneous strategic processing during the reading of multiple expository texts.
Reading and Writing, 16, 195e218.
Bråten, I., & Strømsø, H. I. (2006a). Effects of personal epistemology on the understanding of multiple texts. Reading Psychology, 27, 457e484.
Bråten, I., & Strømsø, H. I. (2006b). Constructing meaning from multiple information sources as a function of personal epistemology: the role of text-
processing strategies. Information Design Journal, 14, 56e67.
Bråten, I., & Strømsø, H. I. (2010a). When law students read multiple documents about global warming: examining the role of topic-specific beliefs about
the nature of knowledge and knowing. Instructional Science, 38(6), 635e657.
Bråten, I., & Strømsø, H. I. (2010b). Effects of task instruction and personal epistemology on the understanding of multiple texts about climate change.
Discourse Processes, 47, 1e31.
Bråten, I., & Strømsø, H. I. (2011). Measuring strategic processing when students read multiple texts. Metacognition and Learning, 6, 111e130.
Bråten, I., Strømsø, H. I., & Samuelstuen, M. S. (2008). Are sophisticated students always better? The role of topic-specific personal epistemology in the
understanding of multiple expository texts. Contemporary Educational Psychology, 33, 814e840.
Brisbois, J. I. (1995). Connections between first and second language reading. Journal of Reading Behavior, 27, 565e584.
Britt, M. A., Perfetti, C. A., Sandak, R., & Rouet, J. F. (1999). Content integration and source separation in learning from multiple texts. In S. R. Goldman, A. C.
Graesser, & P. van den Broek (Eds.), Narrative comprehension, causality, and coherence: Essays in honor of Tom Trabasso (pp. 209e233). Mahwah, NJ:
Erlbaum.
Carrell, P. (1991). Second language reading: reading ability or language proficiency? Applied Linguistics, 12, 159e179.
Clarke, M. (1980). The short circuit hypothesis of ESL reading: or when language competence interferes with reading performance. Modern Language
Journal, 64, 203e209.
Cummins, J. (1979). Linguistic interdependence and the development of bilingual children. Review of Educational Research, 49, 222e251.
Dhieb-Henia, N. (2003). Evaluating the effectiveness of metacognitive strategy training for reading research articles in an ESP context. English for Specific
Purposes, 22, 387e417.
Dreyer, C., & Nel, C. (2003). Teaching reading strategies and reading comprehension within a technology-enhanced learning environment. System, 31,
349e365.
Eskey, D. E. (2005). Reading in a second language. In E. Hinkel (Ed.), Handbook of research in second language teaching and learning. Mahwah, NJ: Lawrence
Erlbaum Associates.
Fecteau, M. (1999). First- and second-language reading of literary texts. The Modern Language Journal, 83, 475e493.
van Gelderen, A., Schoonen, R., Stoel, R., Glopper, K., & Hulstijn, A. (2007). Development of adolescent reading comprehension in language 1 and language 2:
a longitudinal analysis of constituent components. Journal of Educational Psychology, 99, 477e491.
Goldman, S. R. (2004). Cognitive aspects of constructing meaning through and across multiple texts. In N. Shuart-Faris, & D. Bloome (Eds.), Uses of in-
tertextuality in classroom and educational research. Greenwich, CT: Information Age Publishing.
Hammadou, J. (2000). The impact of analogy and content knowledge on reading comprehension: what helps, what hurts. The Modern Language Journal, 84,
38e50.
Ikeda, M., & Takeuchi, O. (2006). Clarifying the differences in learning EFL reading strategies: an analysis of portfolios. System, 34, 384e398.
Jiang, X. (2011). The role of first language literacy and second language proficiency in second language reading comprehension. The Reading Matrix, 11(2),
177e190.
Karimi, M. N. (2015). EFL learners' multiple documents literacy: effects of a strategy-directed intervention program. Modern Language Journal, 99, 40e56.
Karimi, M. N., & Shabani, M. B. (2013). Comparing the strategic behavior of more successful vs. less successful readers of multiple technical reading texts.
Innovation in Language Learning and Teaching, 7, 125e138.
Kintsch, W. (1998). Comprehension: A paradigm for cognition. New York, NY: Cambridge University Press.
 M.N. Karimi / System 52 (2015) 14e25 25

Kobayashi, K. (2009). The influence of topic knowledge, external strategy use, and college experience on students' comprehension of controversial texts.
Learning and Individual Differences, 19, 130e134.
Kobayashi, K. (2010). Strategic use of multiple texts for the evaluation of arguments. Reading Psychology, 31, 121e149.
Lee, J., & Schallert, D. (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, 3, 713e739.
McNeil, L. (2011). Investigating the contributions of background knowledge and reading comprehension strategies to L2 reading comprehension: an
exploratory study. Reading and Writing, 24, 883e902.
McNeil, L. (2012). Extending the compensatory model of second language reading. System, 40, 64e76.
Mokhtari, K., & Reichard, C. (2002). Assessing students' metacognitive awareness of reading strategies. Journal of Educational Psychology, 94, 249e259.
Nassaji, H. (2003). Higher-Level and lower-Level text processing skills in advanced ESL reading comprehension. Modern Language Journal, 87, 261e276.
Phakiti, A. (2003). A closer look at the relationship of cognitive and metacognitive strategy use to EFL strategy reading comprehension test performance.
Language Learning, 53, 649e702.
Phakiti, A. (2006). Modeling cognitive and metacognitive strategies and their relationships to EFL reading test performance. Melbourne Papers in Language
Testing, 11(1), 53e95.
Phakiti, A. (2008). Construction validation of Bachman and Palmer's (1996) strategic competence model over time in EFL reading tests. Language Testing, 25,
237e272.
Pichette, F., Segalowitz, N., & Connors, K. (2003). Impact of maintaining L1 reading skills on L2 reading skill development in adults: evidence from speakers
of Serbo-Croatian learning French. The Modern Language Journal, 87, 391e403.
Salataci, R., & Akyel, A. (2002). Possible effects of strategy instruction on L1 and L2 reading. Reading in a Foreign Language, 14, 1e17.
Samuelstuen, M. S., & Bråten, I. (2005). Examining the validity of self-reports on scales measuring students' strategic processing. The British Journal of
Educational Psychology, 77, 351e378.
Schoonen, R., Hulstijn, J., & Bossers, B. (1998). Language-dependent and language-independent knowledge in native and foreign language reading
comprehension: an empirical study among Dutch students in grades 6, 8 and 10. Language Learning, 48, 71e106.
Sheorey, R., & Mokhtari, K. (2001). Differences in the metacognitive awareness of reading strategies among native and non-native readers. System, 29,
431e449.
Song, M. J. (2001). The role of L2 proficiency in L2 reading: the threshold model revisited. English Teaching, 56(1), 31e51.
Stadtler, M., & Bromme, R. (2008). Effects of the metacognitive computer-tool met.a.ware on the web search of laypersons. Computers in Human Behavior, 24,
716e737.
Strømsø, H. I., & Bråten, I. (2002). Norwegian law students' use of multiple sources while reading expository texts. Reading Research Quarterly, 37, 208e227.
Taboada, A., & Guthrie, J. T. (2006). Contributions of student questioning and prior knowledge to construction of knowledge from reading information text.
Journal of Literacy Research, 38, 1e35.
Tsai, Y., Ernst, C., & Talley, P. C. (2010). L1 and L2 strategy use in reading comprehension of Chinese EFL readers. Reading Psychology, 31, 1e29.
Upton, T. A., & Lee-Thompson, L. (2001). The role of the first language in second language reading. Studies in Second Language Acquisition, 23, 469e495.
Yamashita, J. (2002). Mutual compensation between L1 reading ability and L2 language proficiency in L2 reading comprehension. Journal of Research in
Reading, 25, 81e95.
Yang, Y. (2002). Reassessing readers' comprehension monitoring. Reading in a Foreign Language, 14, 18e42.
Zhang, L. J. (2008). Constructivist pedagogy in strategic reading instruction: exploring pathways to learner development in the English as a second language
(ESL) classroom. Instructional Science, 36, 89e116.