Snow 2016

Journal of Research on Educational Effectiveness
ISSN: 1934-5747 (Print) 1934-5739 (Online) Journal homepage: http://www.tandfonline.com/loi/uree20
Contributions of Academic Language, Perspective

Taking, and Complex Reasoning to Deep Reading
Comprehension
Maria LaRusso, Ha Yeon Kim, Robert Selman, Paola Uccelli, Theo Dawson,
Stephanie Jones, Suzanne Donovan & Catherine Snow
To cite this article: Maria LaRusso, Ha Yeon Kim, Robert Selman, Paola Uccelli, Theo Dawson,
Stephanie Jones, Suzanne Donovan & Catherine Snow (2016): Contributions of Academic
Language, Perspective Taking, and Complex Reasoning to Deep Reading Comprehension,
Journal of Research on Educational Effectiveness, DOI: 10.1080/19345747.2015.1116035
To link to this article: http://dx.doi.org/10.1080/19345747.2015.1116035
Accepted author version posted online: 06

Jan 2016.
Submit your article to this journal
Article views: 18
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=uree20
Download by: [Orta Dogu Teknik Universitesi] Date: 12 February 2016, At: 04:08
ACCEPTED MANUSCRIPT
CONTRIBUTIONS TO DEEP READING COMPREHENSION
Contributions of Academic Language, Perspective Taking, and Complex Reasoning to Deep
Reading Comprehension
Maria LaRusso1,2, Ha Yeon Kim1, Robert Selman1, Paola Uccelli1, Theo Dawson3, Stephanie
Jones1, Suzanne Donovan2 and Catherine Snow1

1
Harvard University, Cambridge, Massachusetts, USA
Downloaded by [Orta Dogu Teknik Universitesi] at 04:08 12 February 2016
2
Strategic Education Research Partnership, Washington, DC, USA
3
Lectica, Inc., Northampton, Massachusetts, USA
Address correspondence to Maria LaRusso, Harvard University, Larsen Hall 313A, 14 Appian
Way, Cambridge, MA 02138, USA. E-mail: maria_larusso@gse.harvard.edu
Abstract
Deep reading comprehension refers to the process required to succeed at tasks defined by the
Common Core State Literacy Standards, as well as to achieve proficiency on the more
challenging reading tasks in the Program for International Student Assessment (PISA)
framework. The purpose of this study was to test the hypothesis that three skill domains not
frequently attended to in instruction or in theories of reading comprehension -- academic
language, perspective taking, and complex reasoning -- predict outcomes on an assessment of
deep reading comprehension. The Global Integrated Scenario-based Assessment (GISA;
O'Reilly, Weeks, Sabatini, Steinberg & Halderman, 2014) is designed to reflect students' abilities
to evaluate texts, integrate information from an array of texts, and use textual evidence to
formulate a position, all features of deep reading comprehension. We tested the role of academic
language, perspective taking, and complex reasoning in explaining variance in end-of-year GISA
1
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
scores, controlling for beginning-of-year scores and student demographics. All three predictors
explained small, but significant, amounts of additional variance. We suggest that these three skill
domains deserve greater attention in theories of reading comprehension and in instruction.
Keywords
reading comprehension, academic language, perspective-taking
2
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Reading comprehension is undeniably the literacy challenge of the 21st century. A half-
century of systematic research has provided strong guidance to instruction in word reading
accuracy and fluency (NICHD, 2000). Whereas U.S. students perform well in international
comparisons at grade four (Martin & Mullis, 2013) when these technical reading skills are the
prime determinant of successful performance, they perform relatively poorly by high school
when comprehension tasks are more challenging (NCES, 2015). Unfortunately, research has
generated much less consensus about either the cognitive processes involved in or the best
instructional approaches to successful reading comprehension (RAND, 2000).
In a notable exception to this generalization, the simple view of reading (SVR) has
achieved widespread acceptance. The SVR (Gough & Tunmer, 1986) is, simply, that
comprehension is the product of decoding and oral language comprehension. In other words, any
text that a student can understand orally also can be understood through reading, if decoding of
words in that text does not form a barrier. This view had the important effect of directing
attention to oral language as a crucial predictor of comprehension outcomes. It has generated
dozens of studies, most of which have generally confirmed the predictions of the simple view,
across several populations, including: monolingual normally developing participants reading in
English (e.g., Savage, 2001) and other languages (e.g., Gentaz, Sprenger-Charolles, Theurel, &
Colé, 2013; Ho, Chow, Wong, Waye & Bishop, 2012), second language readers of English (e.g.,
Erdos, Genesee, Savage & Haigh, 2010, Farnia & Geva, 2013; Proctor, August, Carlo & Snow,
2006), and children with a variety of language and cognitive disabilities (e.g., Catts, Adlof &
Weismer, 2006; Palikara, Dockrell & Lindsay, 2011). These various studies have operationalized
the key variables, decoding, oral comprehension, and reading comprehension, in a variety of
3
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
ways, and most have been conducted with students reading at or below grade four level, a level
at which comprehension is assessed using fairly simple texts and relatively low-inference
comprehension items. In fact, decoding skill has been shown to explain a large percent of the
variance on many of the widely used comprehension assessments, especially for younger readers
(Cutting & Scarborough, 2006).
Despite its success, the SVR has been criticized for its failure to recognize the role of
broader cognitive abilities that are strongly related to comprehension, including memory and
word knowledge (Carrroll, 1993), IQ (e.g., Tiu, Thompson & Lewis, 2003), efficiency (Høien-
Tengesdal & Høien, 2012), and fluency (Silverman, Speece, Harring & Ritchey, 2013),
particularly in accounting for older students' reading (Macaruso & Shankweiler, 2010).
Furthermore, the SVR encounters limitations when extended to explaining performance on the
literacy tasks of adolescence, the comprehension outcomes defined as expected of all students by
the Common Core State Standards (CCSS; NGA, 2010), and the content-area-specific challenges
to comprehension described by those studying disciplinary literacy (Goldman, 2012; Goldman &
Snow, 2015; Lee, 2004; Shanahan, Shanahan & Misischia, 2011). It is these limitations that led
us to propose an enriched model of reading comprehension, a model to explain what we call
deep reading comprehension. The comprehension tasks encountered in secondary school, higher
education, and many employment settings require acquisition of knowledge, integration of newly
acquired with pre-existing conceptual structures, analysis and critique of texts and their sources,
and synthesis across multiple texts and sources. We hypothesize that success at these deep
reading tasks are, in turn, dependent on abilities in three domains that go well beyond decoding
and oral comprehension: academic language, perspective taking, and complex reasoning.
4
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
The purpose of the study presented here is to report the first findings testing this
hypothesis about the predictors of deep reading comprehension. The findings derived from a
study conducted under the Reading for Understanding initiative, in the context of a larger effort
to evaluate instructional interventions to promote deep comprehension skills. In this report,
though, we focus on examining relations among predictors and outcomes among students who
did not receive the intervention, rather than testing intervention effects. We start by exploring the
characteristics of deep reading comprehension and ways of assessing it, and then go on to review
prior research supporting the relationship of academic language, perspective taking, and complex
reasoning to comprehension.
Deep Reading Comprehension
Literacy tasks envisioned in the CCSS (NGA, 2010) include reading to collect textual
evidence in support of a claim, inferring word meanings from close reading of text, integrating
information about a topic from multiple texts, and reading to critique others' arguments (e.g., 6th-
8th grade literacy in history/social studies reading standards 1, 4, and 8 respectively). Such tasks
are considered good preparation for the college- and career-ready literacy challenges toward
which the Common Core standards build. These challenges also are reflected in the expectations
of the 2012 Program of International Student Assessment (PISA) Reading Framework, which
defines the highest level of proficiency as follows:
Tasks at this level typically require the reader to make multiple inferences, comparisons
and contrasts that are both detailed and precise. They require demonstration of a full and
detailed understanding of one or more texts and may involve integrating information
from more than one text. Tasks may require the reader to deal with unfamiliar ideas, in
5
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
the presence of prominent competing information, and to generate abstract categories for
interpretations. Reflect and evaluate tasks may require the reader to hypothesize about or
critically evaluate a complex text on an unfamiliar topic, taking into account multiple
criteria or perspectives, and applying sophisticated understandings from beyond the text.
A salient condition for access and retrieve tasks at this level is precision of analysis and
fine attention to detail that is inconspicuous in the texts. (OECD, 2013, p.79)
The difficulty of preparing students to be highly proficient at literacy tasks like those required by
PISA and the CCSS is clear from the finding that only 0.8% of U.S. high school students tested
by PISA perform at this level. So the questions that arise are what instructional approaches might
help students meet these challenges, and what skills those 0.8% possess that less skilled students
lack.
The default in comprehension instruction in the United State is teaching comprehension
strategies, the approach endorsed by the National Reading Panel (NICHD, 2000). At least before
the advent of professional development and curricula aligned with the CCSS, there was little
insistence in most U.S. schools that students grapple with critically evaluating complex texts on
unfamiliar topics, generating abstract categories for interpretation, or sifting through competing
information. Perhaps the primary locus for actual training in tasks like these in U.S. K-12
education has been Advanced Placement history and literature courses (Young & Leinhardt,
1998), courses to which only a small minority of students have access.
One reason that deep reading comprehension has received so little instructional attention may be
that standardized assessments of reading comprehension rarely measure deep comprehension
skills. Emphasizing deep comprehension may require features that have traditionally been
6
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
difficult to incorporate into relatively brief, group-administered, machine-scorable assessments,
such as the use of multiple texts and the option of open-ended responses with unconstrained
themes or formats (Sabatini, Petscher, O'Reilly, & Truckenmiller, 2015). Furthermore,
interpreting performance on deep reading comprehension assessments may require ancillary test
data, for example, an evaluation of respondents' background knowledge about the topic.
Fortunately for the program of research reported on here, novel and much more informative
assessments have been made possible by the expansion of digital capabilities. For example,
Educational Testing Service (ETS) has developed an innovative suite of computer-delivered
assessments called the Global Integrated Scenario-based Assessments (GISA). GISA is a
theoretically-based measure designed to reflect an updated understanding of the construct of
reading comprehension while taking advantage of the affordances of technological delivery
(O'Reilly & Sabatini, 2013; Sabatini & O'Reilly, 2013; Sabatini, O'Reilly, & Deane, 2013). In
the study reported here, we use the GISA as our outcome measure of deep reading
comprehension (see Methods for a fuller description of the test and its psychometric properties).
Predictors of Deep Reading Comprehension
Hypotheses about the specific skills that need to be added to the SVR to explain deep reading
comprehension derive from a variety of sources: an analysis of deep reading tasks, evidence
about what professional historians, scientists, and literary analysts do when reading, evidence
about factors that increase processing load while reading, and hints derived from the content of
effective programs of professional development and instruction for (post)adolescent readers. In
the brief review that follows, we suggest how these various sources contributed to generating the
hypothesis to be tested in this study.
7
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Academic language
Academic language is now widely cited as a literacy challenge for many students. Precise
definitions or conceptualizations of academic language have varied widely (Snow & Uccelli,
2009), and in the practice and professional development literature academic language is still
often primarily identified with use of “academic vocabulary,” itself a construct based on corpus
analysis (Coxhead, 2000) rather than theory. There is a long history of research within
sociolinguistics and discourse analysis documenting the ways in which written language differs
from oral and formal registers differ from informal, all of which are relevant to defining the
endpoints of the continuum from casual to academic language. Key features of academic
language, described by Halliday (1987, 1993), Schleppegrell (2001, 2007), Scarcella (2003), and
others, include reduction in use of subject pronouns and action verbs; increase in
nominalizations, passives, embedded relative clauses; and lexicalized discourse, stance and
epistemic markers. These features have been shown to increase processing burden during reading
(Carpenter & Just, 1989), in particular for readers who have not encountered them frequently.
Schleppegrell has built programs of professional development around the analysis of
academic language features in high-school content-area texts, to display to teachers constructions
and linguistic usages that might well be unfamiliar and puzzling to their students (e.g.,
Schleppegrell, Achugar & Oteiza, 2004; Schleppegrell & deOliveira, 2006). The close reading
practices encouraged by the CCSS teaching guidelines also may be useful in focusing student
attention on the academic language structures in their texts. Instruction in writing often provides
students with model texts and sentence starters that display how academic language is used. To
our knowledge, though, in none of these instructional approaches have improvements in
8
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
academic language skills been specifically related to changes in reading comprehension
outcomes.
Perspective taking
Particularly when reading literary narratives and history, a great challenge is to recognize
that different actors have different experiences of the same events. The ability to understand and
to navigate those varying perspectives, which we refer to as perspective taking, is critical to

understanding literary and historical conflicts, to attributing psychological causality, and to
evaluating actions. Chall‟s (1983) stage theory of reading development identified stage five,
which she associated with the high school years, as the stage of multiple perspectives, and a long
history of theorizing in literature studies promotes novel reading as a way to enhance
perspective-taking (Leverage, Mancing, Schweickert & William, 2011). The relevant empirical
literature, on the other hand, is scattered, in part because the construct of perspective-taking has
been variously operationalized as theory of mind (Pelletier, 2006), narrative empathy (Keen,
2007), social cognition (Shantz, 1982), or social information processing (Donahue, 2001).
Kessler and Donahue (2008) found that performance on a measure of students' social
information processing, in particular their ability to encode social cues, predicted processing of
character perspectives relevant to understanding a short story‟s surprise ending. Fewer than half
of seventh and eighth grade readers fully understood the surprise, a finding Kessler and Donahue
tentatively attribute to the difficulty of resolving the perspectives of three characters with
different representations of relevant information. Donahue (2014) hypothesizes that failure of
social information processing skills may explain the particular comprehension difficulties of
students with learning disabilities, and the unexpectedly good comprehension of second language
9
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
learners still struggling with reading proficiency (Pelletier, 2006). Gardner and Smith (1987)
defined the development of perspective taking within a Piagetian perspective, as increase in the
capacity to acknowledge reciprocal relationships; whereas they found that perspective taking
skills have no correlation with reading comprehension as measured by the California
Achievement Test, they did find perspective-taking skills to be significantly correlated with deep
reading comprehension as measured by responses to script-implicit (i.e., deep inferential)

questions about a narrative.
In processing narratives about what happens when one character deceives or
inadvertently misleads another, good readers have to keep track of what both characters know.
Younger readers/listeners often attribute what they themselves know about the location of
objects or contents of a container to characters who have not had access to that information in the
narrative (Biancarosa, 2006; Lynch & van den Broek, 2007). Biancarosa demonstrated that better
comprehenders slow down when reading sentences that imply characters' access to concealed
information, suggesting strongly that good readers keep better track of characters' perspectives
than poor readers. These rather simple challenges to perspective taking presage the difficulty
older readers have in understanding that there are multiple defensible positions on many
questions, that people hold those opinions for reasons that have to do with their own life
experiences, and that understanding another‟s point of view neither requires nor excludes
agreeing with it.
Complex reasoning
The third skill hypothesized to relate to deep reading comprehension has to do with the
complexity of students' epistemic thought--- their reasoning about inquiry, evidence, truth,
10
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
knowledge, reasoning, conflict, and deliberation. It has been characterized as the ability to think
effectively about complex issues that have no single correct answer, such as those involving
relationships among food sources, breeding seasons, climatic conditions, and risks of predation
that determine growth or decline of a population within a particular biome. Problems like these
involve multiple statistical and transactional connections among events and consequences.
Reading an explanation of the carbon cycle, or the decline of the Roman Empire, or the long
period of stagflation recently experienced by Japan requires understanding and evaluating claims
made with varying degrees of certainty about multiple, interlocking influences. Evidence that
tracking explanations of complex phenomena while reading is a severe challenge to students is
abundant in teacher reports, and in the workarounds content area teachers introduce -- lectures,
powerpoint presentations -- to convey basic information to students unable to process such
information by reading their textbooks.
Early research on children‟s and adolescents' reasoning focused on the changes, described
by Piaget, that occurred from concrete formal operational thinking, theorized to begin at about
age 11 or 12 and marked by the ability to use hypothetical and deductive reasoning, among other
abstract skills (Inhelder & Piaget, 1958). Later work undertaken by Carey (1985), Vosniadou
(2013) and diSessa (1988) has focused on the nature of the conceptual change that has to occur --
that is not so much on the reasoning process itself as the knowledge structures that students have
to reason with. Much of the relevant work uses a confusing mix of overlapping terms and
concepts, such as complex reasoning, scientific reasoning, argumentation, critical thinking and
scientific thinking, among others. Most of the research has focused on how these capacities
develop, with surprisingly few studies examining how reasoning skills relate to other
11
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
developmental outcomes. Nonetheless, it is clear that complex reasoning develops over the
course of the lifespan in a sequence of increasingly complex levels (Perry, 1970), and relates
moderately to academic and verbal ability (Wood, 1997).
However, research has shown that reasoning skills are often deficient or lacking for both
children and adults (e.g., Kuhn, 1991) and that classroom practices that promote argumentation
are rare (Newton, Driver, and Osborne, 1999). Studies in science education found that students
struggle with several aspects of reasoning and argumentation, such as finding adequate evidence
to support a claim, integrating contradictory evidence, and challenging the claims made by others
(Bell, 2004; Cavagnetto, Hand, & Norton-Meier, 2010; Evagorou, Jimenez-Aleixandre, &
Osborne, 2012; Jimenez-Aleixandre & Pereiro-Munoz, 2002; Sandoval & Milwood, 2005).
However, children‟s scientific reasoning can be effectively promoted by specific classroom
practices, such as inquiry-based instruction (e.g., Gerber, Cavallo & Marek, 2001; Kuhn, 1997)
and implementation of classroom strategies for teaching argumentation (Osborne, Erduran, &
Simon, 2004). While it seems obvious that students need reasoning skills to navigate texts
reporting complex relationships or phenomena (for example, in history, science or literature),
such as those frequently encountered in the middle grades and higher, there has been little
research examining the relationship between reasoning and reading comprehension. In the
context of science literacy, however, there has been research demonstrating that students have
insufficient skills to evaluate and infer meaning from media reports about science (Norris &
Phillips, 1994; Zimmerman, Bisanz, Bisanz, Klein, & Klein, 2001).
Purpose of the Study
12
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
This study is designed to test the hypothesis that academic language, perspective taking, and
complex reasoning predict outcomes on an assessment of deep reading comprehension.
We collected data on academic language, perspective taking, and complex reasoning skills as
well as on deep reading comprehension performance from students in grades four through seven
participating in a large-scale experimental evaluation of a novel curriculum. Utilizing data from
the control group, we start with analyses relating academic language and perspective taking to
deep reading comprehension, controlling for demographic variables and grade. We then present a
more exploratory model that includes the complex reasoning measure, which for technical
reasons is only available on a reduced portion of the sample.
Methods
This study utilizes data from an IES-funded evaluation of Word Generation (WG), a
research-based academic language program for middle school students designed to teach novel
vocabulary and literacy through language arts, math, science, and social studies classes. The WG
evaluation is a school-level experimental study that includes 24 schools randomized to treatment
and control conditions. The first cohort began in 2011 and participated for three years, and the
second cohort began in 2012, participating for two years. Data for the present study are from
control schools only (n = 12), in the second year of the study, after both cohorts were enrolled.
Participants
The participants in this study included 2933 students in with 124 classrooms (grades 4-7),
from diverse backgrounds reflecting the demographics of the urban and semi-urban communities
of the schools (51% female, 40% Black, 28% White, 3% Asian, 27% Latino, 1% Native
American/Pacific Islander, 1% Mixed Race/Other, 83% eligible for free/reduced price lunch, 8%
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
English language learners (ELLs), and 14% with special education classification (Table 1). The
study had “exempt status” and no consent process was required because assessments were
administered as part of standard educational practices to assess the effectiveness of school
curricula. Assessments were administered to all students present on testing days.
Measures
Core academic language skills

The Core Academic Language Skills-Instrument (CALS-I) is a group-administered
instrument designed to assess core academic language skills (CALS) in grades four to eight.
CALS refer to the constellation of high-utility language skills that correspond to linguistic
features prevalent in oral and written academic discourse across school content areas, but that are
infrequent in colloquial conversations (e.g., logical connectives, such as nevertheless,
consequently; structures that pack information densely, such as nominalizations or embedded
clauses; markers of organization in argumentative texts, such as first, on the other hand) (Uccelli,
Phillips Galloway, Barr, Meneses, & Dobbs, 2015). Each CALS-I form consists of a 50-minute
paper and pencil test that includes eight tasks: Connecting Ideas, Tracking Themes, Organizing
Texts, Breaking Words, Comprehending Sentences, Identifying Definitions, Interpreting
Epistemic Stance Markers, and Understanding Metalinguistic Vocabulary. Tasks assess students'
skills through a range of multiple-choice, matching, or short written responses.
The development of the CALS-I was based on an extensive literature review, followed by
an iterative design process that unfolded in the following sequence: a task design phase, and pre-
pilot study, a series of qualitative and quantitative pilot studies, an expert review panel, and a
norming phase (for more information, Uccelli, Barr, Dobbs, Phillips Galloway, Meneses, &
14
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Sánchez, 2014; Uccelli et al., 2015). Two forms of the CALS-I were used in this study: CALS-I-
Form 1 designed for grades four to six (α = .90, number of items = 49) and CALS-I-Form 2 for
for grades seven and eight (α = .86, number of items = 46). The items that were not scored
dichotomously as correct/incorrect were rescaled to be between 0 and 1 so all items were equally
weighted in estimating the total score. Using Rasch item response theory analysis, factor scores
were generated for the CALS-I using a vertically equated scale.

Social perspective taking
The Social Perspective Taking Acts Measure (SPTAM) is a scenario-based instrument
that analyzes a participant‟s written short answer responses to standard questions across social
situations that commonly occur in middle schools. The SPTAM assesses the way individuals, at
grade four and up, take multiple individuals' perspectives into account when responding to
questions about hypothetical social dilemmas. As operationalized in the SPTAM, social
perspective taking refers to the skills people can draw upon to “read” the social world, e.g.,
through print text, through social discourse, and in the navigation of complex social relationships
and civic participation. A coding manual is used to assess the expression of three performative
social skills (acts): the acknowledgement of those parties relevant to the situation described in the
scenario, the articulation of the thoughts and feelings of selected parties acknowledged therein,
and the positioning of those scenario-based actors, depending upon their roles, circumstances,
contexts, cultural background, and motivations. In this analysis, articulation and positioning
scores are used. A validation study of the SPTAM found evidence of good internal reliability
(alphas for the three scales, acknowledgment, articulation and positioning were .80, .83, and .70),
with construct validity confirmed by the findings that girls and older participants exhibited better
15
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
performance than boys and younger students, and that the SPTAM exhibited a negative
association with aggressive interpersonal strategies and small to moderate associations with
academic language and with basic reading skills (Diazgranados, Selman & Dionne, in press). In
the present study, we utilize only the articulation and positioning scales. Acknowledgment, while
also important, captures a more basic perspective taking competency while the articulation and
positioning scales reflect more advanced perspective-taking skills needed by early adolescents to
comprehend more complex texts.
Complex reasoning
Based on King and Kitchener‟s (1994) work, Lectica, Inc. built a set of developmental
assessments that measure the complexity of complex reasoning skills in children and adults,
referred to collectively as reflective judgment (RFJ) tasks. The version used in this study is the
RFJ001 which, like all forms of the RFJ, assesses the way individuals apply what they know
about inquiry, evidence, truth, knowledge, reasoning, conflict, and deliberation to the solution of
complex problems. This computer-administered assessment presents students with a scenario,
typically a complex but familiar conflict that has no correct answer. Students write responses to a
series of questions designed to elicit strategies for deciding on an answer to the conflict. Scores
on the RFJ001, like scores for all of Lectica‟s assessments (Dawson & Stein, 2008, 2011),
represent levels on Fischer‟s dynamic skill scale, a lifespan scale of increasing hierarchical
complexity (Fischer 1980, 2008; Fischer & Bidell, 2006). Responses are
scored with low-inference rubrics in which each choice is associated with a phase (on fourth of a
developmental level) on the skill scale. Previous research has shown that the RFJ is sensitive to
developmental differences, and reliably measures complex reasoning skills, that are
16
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
distinguishable from literacy skills (Dawson, 2014; Dawson & Stein, 2012). Based on the results
of a confirmatory Rasch item response analysis of 3,754 performances spanning nine phases,
Dawson (2014) has reported a person separation reliability of .91 with an estimated alpha of .94.
Deep reading comprehension
The Global Integrated Scenario-based Assessments (GISA), developed by ETS, are
computer-based assessments that use scenarios, technology, and reading strategies to motivate
students, to model skilled reading, and to help disentangle key areas for improvement. In a
scenario-based assessment design, students are given a plausible purpose for reading (e.g., decide
if a wind farm is a good idea for your community) and a collection of source materials (e.g.,
blog, website, e-mail, news article, textbook excerpt).
Example scenarios include having students imagine that they are preparing to lead a class
discussion or that they are part of a study group. Sources are written by a variety of authors that
include a range of perspectives and levels of credibility on the issue in question. As such,
students are expected to evaluate, integrate and synthesize the materials in order to make a
decision, or solve the problem outlined in their original purpose for reading. Because the sources
are thematic, and the materials are sequenced to build up students' understanding of the topic,
deeper questions can be asked of the student that not only taps their basic understanding (what
the text is about), but also their ability to apply what they read to different contexts, situations
and perspectives.
Deep comprehension is in part, facilitated by adding a social dimension to the
assessment. In a typical GISA assessment, there are a number of simulated authority figures
(e.g., teacher) and peers (e.g. students). The simulated teacher is designed to clarify expectations
17
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
and provide hints that encourage desired responses. Simulated peers also may serve a similar
function, but sometimes the test taker is asked to comment on the simulated peer‟s understanding
of the sources that may contain misconceptions or errors in reasoning. Adding the simulated peer
responses allows the test designer to more effectively target metacognitive and self-regulatory
behaviors that encourage deeper understanding and perspective taking. This strategic approach to
reading also is facilitated by adding a range of empirically supported reading strategies as tasks
(e.g., summarization, use of graphic organizers, questioning). The reading strategies serve to
encourage more thoughtful processing and model skilled reading.
GISA batteries cover a range of content areas, including science, social studies, and
English language arts. Presented with a specific purpose for reading, students are then provided
materials that are focused on a common topic and that include a variety of text types that
students regularly encounter (e.g., expository texts, fiction, e-mail, web pages, and blogs).
Students advance through the materials in a structured way that enables them to: produce
evidence of complex mental models of text content; organize what they read; and synthesize
what they have learned to satisfy their original purpose for reading.
Though it is a relatively new instrument and more nuanced evaluations are ongoing, the
GISA is appropriate for a wide range of ability levels, and can reliably measure a range of “21st
century” reading skills that go beyond those assessed in more traditional, low-inference
comprehension assessments (O'Reilly, Weeks, Sabatini, Halderman, & Steinberg, 2014; Sabatini,
O'Reilly, Halderman & Bruce, 2014a, b). ETS reported results from a study of middle school
students administered the organic farming version of the GISA that revealed adequate
psychometric properties (Sabatini, O'Reilly, Halderman & Bruce, 2014), including Internal
18
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
consistency (alpha) reliability (0.89) and split half reliability (.76), with each half of the test
showing adequate alpha reliability (α = 0.80 and α = 0.82 respectively). Test-retest reliability
was r (283) = .87 and there was no significant difference in mean scores. The GISA also
demonstrated a strong concurrent validity with more conventional reading comprehension tests,
as well as component reading subtests. For this study, the following versions of the GISA were
administered: Deserts (4th & 5th grade, Fall 2012), Satellites (4th & 5th grade, Spring 2013),
Organic Farming (6th & 7th grade, Fall 2012) and Wind Power (6th & 7th grade, Spring 2013).
Although the GISA is currently designed to produce a single score, the items and tasks in these
forms in principle measure a skills including evaluation of websites, distinguishing claims and
evidence, integrating information across multiple texts, interpreting graphs or cartoons,
questioning, predicting, summarizing and evaluating of summaries, paraphrasing, using graphic
organizers, analogical thinking, and learning. Scores are reported on a common, cross form scale
based on a large-scale study conducted by the ETS research team. Scaling used a concurrent,
multi-group approach, with a two-parameter logistic IRT model (2PL) for form pairs.
Analysis Plan
Preliminary analysis
Table 2 presents means, standard deviations, and correlation coefficients for the main
variables of interest. Spring GISA scores (our measure of deep reading comprehension) were on
average higher than, and were highly associated with, fall GISA scores, r = .72, p < .001. In
addition, spring GISA scores were associated with the hypothesized predictors of GISA,
including the two perspective taking scores---perspective articulation (r = .38, p < .001) and
positioning (r = .30, p < .001) as well as complex reasoning scores (r = .52, p < .001), and
19
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
academic language (r = .68, p < .001). Predictors of reading comprehension were positively
correlated with each other as well, r = .29-.65, p < .001.
Missing information
The initial analytic sample for this study included students who attended the participating
control schools during the 2012-2013 school year and took any of the assessments administered
in this school year (n = 2933). Among the participants fewer than 3% were missing demographic
information (race/ethnicity 3%; gender, free/reduced-price lunch eligibility, ELL status,
receiving special education 1%). The rates of missingness were larger for the student assessment
data, ranging from 12% to 25%. Specifically, GISA scores were not available from 17% of the
participants in both fall and spring; the perspective-taking scores (15%) and CALS (12%) had
slightly lower rates of missingness, whereas 25% of the complex reasoning scores were missing.
To maintain consistency across analytic models with respect to degrees of freedom and
allow comparison between models, only participants with complete data on the main predictor
variables (demographic characteristics, fall deep comprehension scores, perspective taking skills,
and academic language scores) were included in the primary models (n = 1965). Compared to the
full sample, this final sample was more likely to include white students (c2 = 4.38, p < .05),
compared to the full sample, and less likely to include students who were black (c2 = 6.61, p =
.01), mixed/other race (c2 = 6.49, p = .01), English language learners (c2 = 10.45, p = .001), and
eligible for special education (c2 = 35.98, p < .001).
Because of the large proportion of missing information, complex reasoning was explored
in a separate set of exploratory models, in which only the subsample with complete complex
reasoning scores (n = 1615) was included. Compared to the final sample above, this exploratory
20
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
sample was even more likely to be white (c2 = 34.98, p < .001) and Latino (c2 = 3.49, p > .05),
but less likely to include students who are black (c2 = 60.84, p < .001), eligible for free/reduced-
price lunch (c2 = 5.58, p < .05), or eligible for special education (c2 = 17.38, p < .001)
The high rate of missingness in assessments is not unexpected, given that 10-46% of
students switch schools within one academic year in the participating schools, which primarily
serve low income communities. In addition to high student mobility, there is an impact of student
absences on testing days.
Multilevel models
Research questions were tested using a series of multi-level models to address the nested
structure of the data (i.e., children nested in classrooms and schools), using STATA xtmixed
procedure. In all models, continuous variables were group mean centered within classrooms, as
the main interest of the current study is to examine relations between individual students'
academic-related skills and their reading comprehension performance. This approach allows
unbiased estimates of relations between variables at the individual level and produces the most
accurate estimates of the slope variance (Enders & Tofighi, 2007; Raudenbush & Bryk, 2002).
Full maximum likelihood estimation was used in order to handle missing data in the outcome
variable (Allison, 2012).
Because of the data limitation of the complex reasoning measure, we tested the research
question using two separate sets of models. The first set of models examined the role of
perspective taking skills and academic language with the full analytic sample (n = 1965). The
second analysis explored the additional role of complex reasoning with a subset of the larger
sample (n = 1615).
21
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Results
Prior to testing hypotheses, an unconditional (i.e., null) model (Table 3 Model 1) was
estimated, to determine the amount of variance within and between classrooms and between
schools in three-level models. The intraclass correlation coefficients (ICCs) of GISA scores were
.14 at the school level and .14 at the classroom level. This suggested the greatest variance in deep
reading comprehension scores was at the individual student level (.72), whereas the variance
between classrooms and schools was large enough to be worth examining (>.10, Lee, 2000), with
design effects >2 for both classroom (2.92) and school level (2.24). Therefore we used three-
level models for further analyses.
Role of Perspective Taking skills and Academic Language
Models presented in table 3 tested the role of the two perspective-taking skills
(articulation and positioning) and academic language in predicting deep comprehension. Model 2
including the student covariates produced findings similar to those for the full sample, with sixth
(coefficient = 26.38, p < .001) and seventh (coefficient = 28.50, p < .001) grade showing
significantly steeper growth across one academic year than fourth graders, and fifth graders
showing a trend toward steeper growth (coefficient = 12.99, p = .07). Students who were low
income (coefficient = -10.98, p < .01), ELLs (coefficient = -20.22, p < .001), and in special
education (coefficient = -25.06, p < .001) had slower growth in their deep comprehension skills
than their peers in the subsample, echoing the findings for the full sample. These student
characteristics explained 45% of variance on GISA scores at the individual level (c2 = 853.73, p
< .001).
The role of complex reasoning is tested in Model 3. The results suggest that fall scores
22
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
were a strong and positive predictor of spring deep comprehension scores (coefficient = 44.24, p
< .001). Inclusion of complex reasoning scores in the model explained an additional 2% of the
variance at the student level over and above the student covariates (versus Model 2, c2 = 43.17, p
< .001).
Finally, Model 5 examined whether complex reasoning contributes to explaining the
growth of deep reading comprehension over and above perspective taking and academic
language skills (see model 4, for the results with the subsample). The findings reveal that fall
complex reasoning independently and positively predicts deep reading comprehension
(coefficient = 18.01, p < .01), when perspective taking and academic language skills are
controlled for. Including complex reasoning scores in the model explained an additional .02% of
the variance compared to Model 4 (c2 = 6.94, p < .01). Overall, this final model including all
three predictors of reading comprehension (perspective taking, academic language, and complex
reasoning) and student covariates explained 52% of the total variance in deep comprehension at
the student level (c2 = 1012.84, p < .001).
Exploring the Role of Complex Reasoning
Due to the constraints on the available data, we examined the role of complex reasoning
in predicting GISA in a separate set of models (Table 4). Model 1 presents the null model for the
subsample with available complex reasoning scores. This subsample had slightly smaller ICCs
on deep reading comprehension, .11 at the school level and .12 at the classroom level.
Model 2 including the student covariates produced findings similar to those for the full
sample, with 6th (coefficient = 25.91, p < .001) and 7th (coefficient = 28.46, p < .001) grade
showing significantly steeper growth across one academic year than 4th graders, and 5th graders
23
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
showing a trend toward steeper growth (coefficient = 12.97, p = .07). Students who were low
income (coefficient = -11.09, p < .01), ELLs (coefficient = -20.35, p < .001), and in special
education (coefficient = -24.92, p < .001) had slower growth in their deep comprehension skills
than their peers in the subsample, echoing the findings for the full sample. These student
characteristics explained 45% of variance on GISA scores at the individual level (c2 = 845.85, p
< .001), as they did in model 2 for the full sample.

The role of complex reasoning is tested in Model 3. The results suggest that fall scores
were a strong and positive predictor of spring deep comprehension scores (coefficient = 44.70, p
< .001). Inclusion of complex reasoning scores in the model explained an additional 2% of the
variance at the student level over and above the student covariates (versus Model 2, c2 = 43.40, p
< .001).
Finally, Model 5 examined whether complex reasoning contributes to explaining the
growth of deep reading comprehension over and above perspective taking and academic
language skills (see model 4, for the results with the subsample). The findings reveal that fall
complex reasoning independently and positively predicts deep reading comprehension
(coefficient = 18.56, p < .01), when perspective taking and academic language skills are
controlled for. Including complex reasoning scores in the model explained an additional .02% of
the variance compared to Model 4 (c2 = 7.28, p < .01). Overall, this final model including all
three predictors of reading comprehension (perspective taking, academic language, and complex
reasoning) and student covariates explained 51% of the total variance in deep comprehension at
the student level (c2 = 1005.31, p < .001).
Discussion
24
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
The goal of the study presented here was to provide a first test of the hypothesis that deep
reading comprehension is co-determined by students' ability to understand academic language, to
take and understand social perspectives, and to engage in complex reasoning about challenging
problems. The findings support the credibility of the hypothesis. Academic language was the
strongest of our hypothesized predictors of deep comprehension, suggesting that this is an
important area of focus to prepare students for secondary school texts with their increasingly
unfamiliar and challenging language (e.g., Schleppegrell, Achugar & Oteiza, 2004;
Schleppegrell & deOliveira, 2006). Both perspective articulation and positioning were found to
have relationships with deep comprehension, although when academic language is added to the
model, only positioning is statistically significant, suggesting that, in addition to academic
language, it is particularly important for early adolescents to be able to interpret varied
perspectives in the context of a character‟s or author‟s roles, circumstances, cultural
backgrounds, etc. This is aligned with Kessler and Donahue (2008)‟s work suggesting that early
adolescents' comprehension can be limited by insufficient skills for processing characters'
multiple and sometimes conflicting positions.
Additional exploratory analyses from the present study indicate that complex reasoning is
also a significant predictor of deep comprehension. While previous research has shown complex
reasoning to be related to academic and verbal ability (Wood, 1997), we believe this is the first
study to demonstrate that complex reasoning predicts deep comprehension. While these analyses
were conducted with a reduced sample and are interpreted with caution, they do offer
preliminary evidence that complex reasoning may be important for the reading comprehension
tasks required of young adolescents. In addition, academic language and perspective positioning
25
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
continue to be significant predictors of deep comprehension in these exploratory analyses that
include complex reasoning.
Taken together these findings on the roles of academic language, perspective-taking and
complex reasoning in deep comprehension suggest that, for students in grades four through
seven, we need to consider other models of reading comprehension beyond the SVR. In
particular, as students enter the middle grades and need to understand increasingly complex
texts, decoding and oral language skill may be necessary but insufficient for reading
comprehension. Indeed the literature in support of the SVR has most often focused on younger
students and others have argued that the SVR neglects other skills, such as efficiency and
fluency, that may become more important for students as they reach the higher grades (Høien-
Tengesdal & Høien, 2012; Silverman, Speece, Harring & Ritchey, 2013; Macaruso &
Shankweiler, 2010). In addition, research on the SVR has utilized measures of reading
comprehension that assess a relatively basic level of reading comprehension, using relatively
simple texts and noninferential questions, while success in middle and high school requires
students to learn from more complex tests, across disciplines, and to interpret and analyze these
texts in more sophisticated ways. Our study is one of the first to employ a “deep comprehension”
assessment that incorporates a variety of textual sources and more challenging reading
comprehension tasks, such as evaluating and synthesizing materials and applying what has been
read to different contexts and perspectives. Thus, the results of this study demonstrate not only
that other reading related skills (e.g. academic language) are important in early adolescence, but
that such skills are important for the deep comprehension required of students in upper grade
classrooms
26
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
This is, of course, only the first step in the analytic program testing this hypothesis; future
analyses should examine whether these three components predict reading comprehension beyond
the contributions of decoding and oral language. In future work we will conduct similar analyses
for a subsample of participants for whom we have measures of general vocabulary and decoding,
as well as fluency, and more traditional assessments of reading comprehension, including state
ELA tests, allowing us to more directly compare the SVR with our model of deep reading
comprehension predicted by academic language, perspective taking and complex reasoning.
Our analysis of predictors of deep comprehension also included several key demographic
factors that deserve mention. After controlling fall deep reading comprehension scores, several
demographic variables predicted spring comprehension scores. Students who are eligible for free
or reduced lunch, English language learners, and/or have a special education classification scored
significantly lower on deep reading comprehension. While racial/ethnic differences on academic
outcomes including reading comprehension are common (National Center for Education
Statistics, 2013; Reardon, Valentino, & Shores, 2012), there was no relationship between
students' racial/ethnic backgrounds and GISA comprehension scores. This is true for other recent
studies using the GISA (O'Reilly, Sabatini, Halderman, Bruce, Weeks & Steinberg, 2015).
Lastly, there was a significant effect for grade, with students in higher grades scoring
significantly higher than 4th grade students on deep comprehension.
There are of course limitations to the analysis presented here. Missing data are a chronic
problem in school-based research, one that is exacerbated when the data derive from four distinct
assessments, administered in different sessions, in whole-class configurations, and in two cases
digitally in elderly computer labs. Though we made every effort to reschedule testing sessions
27
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
when absences were a significant problem, high school mobility rates and shifting class
schedules rendered success incomplete.
Nonetheless, we are able to report findings based on a substantial number of students.
Unfortunately, there is some suggestion in the pattern of missingness that the students whose
scores were omitted differed on demographic factors from those included in the final analytic
sample or subsample, a finding that requires on-going analysis. Fortunately, because the
predictors reported here were assessed in only one wave of data collection, in a study in which
six waves are available on some participants and four on most, we will have the opportunity in
future analyses to pursue these same questions with additional waves of data, modeling growth
trajectories over as many as three years for both predictors and the outcome.
Another limitation is the lack of classroom and school level predictors (e.g., aggregated
student demographics) despite sufficient classroom and school level variance in deep
comprehension to warrant multi-level modeling. We decided to not include such predictors in
this analysis because of the challenges of multicollinearity among classroom and school level
demographics and because our focus in this paper is on student level questions. Future analyses
will examine the context-level variation in deep comprehension.
Despite these limitations, these results provide preliminary evidence that academic
language, perspective taking, and complex reasoning predict deep reading comprehension. This
finding has important implications for theories of reading comprehension and for designing
interventions to improve comprehension outcomes. These three predictive factors have not
previously been systematically attended to in curricular design or instruction. Programmatic
efforts to manipulate these input factors, through curricular design and/or special interventions,
28
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
and to assess impacts on reading comprehension outcomes will provide the ultimate test of their
importance as components of the deep comprehension process.
The relevance of academic language, perspective taking, and complex reasoning to deep
comprehension suggests as well the importance of ensuring that teachers of literacy-heavy
content areas (history and science, as well as English language arts) understand these
phenomena. It is difficult to predict where students will struggle with the text or misinterpret
author‟s intent without some understanding of academic language and perspective taking. Texts
encountered in history are inherently perspectival, and science texts demand following extended
lines of reasoning explained verbally. Teachers' sensitivity to these challenges will improve
content area learning as well as literacy outcomes.
Practices widely promoted as helping students reach the expectations of the Common
Core State Standards include close reading and assigning more complex texts. Our findings
suggest these practices are unlikely by themselves to be helpful to students struggling with
academic language, perspective taking, and complex reasoning, and might in fact lead to
frustration and reduced motivation. A better understanding of the processes underlying deep
reading comprehension will, we hope, generate approaches to instruction that directly address the
linguistic and cognitive challenges students face.
Acknowledgments
The authors contributed to this manuscript in various ways. LaRusso and Jones directed the
research and advised on the analytic strategy. Kim conducted the analyses. Selman, Uccelli and
Dawson developed assessments for the key predictors, perspective-taking, academic language,
and complex reasoning respectively. The manuscript was prepared primarily by Snow, LaRusso
29
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
and Kim. Snow and Donovan were co-PIs. We would like to thank James Kim, Lowry Hemphill,
McCaila Ingold-Smith, Jen Winsor, Jill Joseph, Chris Barr, Emily Phillips Galloway, Silvia
Diazgranados and our many research assistants for their contributions to this research, as well as
the students, teachers, and administrators for their participation.
Funding
The research reported here was supported by the Institute of Education Sciences, U.S.
Department of Education, through Grant R305F100026 to the Strategic Educational Research
Partnership Institute and grant R305F100005 to Educational Testing Service as part of the
Reading for Understanding Research Initiative. The opinions expressed are those of the authors
and do not represent views of the Institute or the U.S. Department of Education.
30
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
References
Allison, P. D. (2012). Handling missing data by maximum likelihood. In SAS global forum (Vol.
213). Orlando, FL. Retrieved from: http://statisticalhorizons.com/wp-
content/uploads/MissingDataByML.pdf
Bell, P. (2004). Promoting students' argument construction and collaborative debate in the
science classroom. In M. Linn, E. Davis, & P. Bell (Eds.), Internet environments for science
education (pp. 115--143). Hillsdale, NJ: Lawrence Erlbaum.
Biancarosa, G. (2006). Reading minds: What sentence-reading rate tells us about comprehension
processing. Unpublished dissertation, Harvard University.
Carey, S. (1985). Conceptual change in childhood. Cambridge, MA: MIT Press.
Carpenter, P. & Just, M. (1989). The role of working memory in language comprehension. In D.
Klahr & K, Kotofsky (eds.), Complex information processing: The impact of Herbert A.
Simon (pp. 31-68). Hillsdale, NJ: Lawrence Erlbaum.
Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. New York:
Cambridge University Press.
Catts, H. W., Adlof, S. M., & Weismer, S. E. (2006). Language deficits in poor comprehenders:
A case for the simple view of reading. Journal of Speech, Language, and Hearing
Research, 49, 278-293. doi:1092-4388/06/4902-0278
Cavagnetto, A., Hand, B., & Norton Meier, L. (2010). The nature of elementary students' science
discourse in the context of the science writing heuristic approach. International Journal of
Science Education, 32, 427--449. doi: 10.1080/09500690802627277
Chall, J. S. (1983). Literacy: Trends and explanations. Educational Researcher, 12, 3-8.
31
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Coxhead, A. (2000). A new academic word list. TESOL Quarterly, 34, 213-238.
Cutting, L. E., & Scarborough, H. (2006). Prediction of reading comprehension: Relative
contributions of word recognition, language proficiency, and other cognitive skills can
depend on how comprehension is measured. Scientific Studies of Reading, 10, 277-299.
Dawson, T. (2014). A confirmatory Rasch analysis of the RFJ001. International Objective
Measurement Workshop. Philadelphia.

Dawson, T. L. and Z. Stein (2008). “Cycles of research and application in education: Learning
pathways for energy concepts.” Mind, Brain, & Education, 2, 90-103.
Dawson, T. L. and Z. Stein (2011). We are all learning here: Cycles of research and application
in adult development. In C. Hoare (ed.). The Oxford Handbook of reciprocal adult
development and learning (pp. 447-460) New York: Oxford University Press
Dawson, T. L. and Z. Stein (2012). Measuring the growth of reflective judgment with cognitive
developmental assessments. Annual meeting of the Northeastern Educational Research
Association. Rocky Hill, CT.
Diazgranados, S., Selman, R., & Dionne, M. (in press). Acts of social perspective taking: a
functional construct and the validation of a performance measure for early adolescents.
Social Development.
diSessa, A. A. (1988). Knowledge in Pieces. In G. Forman & P. B. Pufall (Eds.), Constructivism
in the Computer Age. Hillsdale, NJ: Lawrence Erlbaum.
Donahue, M. (2001). “Hanging with friends:” Making sense of research on peer discourse in
children with language and learning disabilities. In K. Butler & E. Silliman (Eds.),
32
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Speaking, reading, and writing in students with language learning disabilities (pp. 239-
258). Mahwah, NJ: Lawrence Erlbaum.
Donahue, M. (2014). Perspective-taking and reading comprehension of narratives: Lessons
learned from The Bean. In C,A. Stone, E. Silliman, B. Ehren & G. Wallach (Eds.),
Handbook of language and literacy development and disorders, 2nd edition (pp 323-338).
New York: Guilford Press.

Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel
models: A new look at an old issue. Psychological Methods, 12, 121--138.
http://doi.org/10.1037/1082-989X.12.2.121
Erdos, C., Genesee, F., Savage, R., & Haigh, C. A. (2010). Individual differences in second
language reading outcomes. International Journal of Bilingualism, 15, 3-25.
doi:10.1177/1367006910371022
Evagorou, M., Jime´nez-Aleixandre, M., & Osborne, J. (2012). „Should we kill the grey
squirrels?‟ A study exploring students' justifications and decision-making. International Journal
of Science Education, 34, 401--428. doi:10.1080/09500693.2011.619211
Farnia, F., & Geva, E. (2013). Growth and predictors of change in English language learners'
reading comprehension. Journal of Research in Reading, 36, 389-421.
doi:10.1111/jrir.12003
Fischer, K.W. (2008). Dynamic cycles of cognitive and brain development: measuring growth in
mind, brain, and education. In A.M. Battro, K.W. Fischer & P. Léna (Eds.), The
educated brain (pp. 127-150). Cambridge, U.K.: Cambridge University Press.
Fischer, K. W. (1980). A theory of cognitive development: The control and construction of
33
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
hierarchies of skills. Psychological Review, 87, 477-531. doi: 10.1037/0033-
295X.87.6.477
Fischer, K. W. and T. R. Bidell (2006). Dynamic development of action, thought, and emotion.
In W. Damon and R. M. Lerner (Eds.), Handbook of child psychology: Theoretical
models of human development, (pp. 313-399). New York, Wiley.
Gentaz, E., Sprenger-Charolles, L., Theurel, A., & Colé, P. (2013). Reading comprehension in a
large cohort of French first graders from low socio-economic status families: A 7-month
longitudinal study. PLoS ONE, 8(11): e78608. doi:10.1371/journal.pone.0078608
Gardner, M. & Smith, M. (1987). Does perspective taking ability contribute to reading
comprehension? Journal of Reading, 30, 333-336.
Gerber, B.L., A.M. Cavallo, & E.A. Marek. (2001). Relationships among informal learning
environments, teaching procedures and scientific reasoning ability. International Journal
of Science Education, 23, 535-549. doi: 10.1080/09500690116971
Goldman, S. (2012). Adolescent literacy: Learning and understanding content. Future of
Children, 22, 89-116.
Goldman, S. & Snow, C.E. (2015). Adolescent Literacy: Development and Instruction. In
Handbook on Reading (pp 463-478), A. Pollatsek & R. Treiman (eds.). New York:
Oxford University Press.
Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading, and reading disability. Remedial and
Special Education, 7, 6-10. doi:10.1177/074193258600700104
Halliday, M. A. K. (1987). Spoken and written modes of meaning. In R. Horowitz & S. J.
Samuels (Eds.), Comprehending oral and written language (pp. 55-82). San Diego, CA:
34
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Academic Press.
Halliday, M.A.K. (1993). Some grammatical problems in scientific English. In M. A. K.
Halliday & Martin, J. R., Writing science: Literacy and discursive power. London:
Falmer (Critical Perspectives on Literacy and Education); Pittsburgh: University of
Pittsburgh Press. (Pittsburgh Series in Composition, Literacy, and Culture).
Ho, C. S., Chow, B. W, Wong, S. W, Waye, M. M. Y., Bishop, D. V. M. (2012). The genetic and
environmental foundation of the simple view of reading in Chinese. PLoS ONE, 7(10):
e47872. doi:10.1371/journal.pone.0047872
Høien-Tengesdal, I., & Høien, T. (2012). The reading efficiency model: An extension of the
componential model of reading. Journal of Learning Disabilities, 45, 467-479.
doi:10.1177/0022219411432688
Inhelder, B., & Piaget, J. (1958). The growth of logical thinking from childhood to adolescence:
An essay on the construction of formal operational structures. London: Routledge & Kegan
Paul.
Jimenez-Aleixandre, M. P., & Pereiro-Munoz, C. (2002). Knowledge producers or knowledge
consumers? Argumentation and decision making about environmental management.
International Journal of Science Education, 24, 1171--1190. doi:10.1080/09500690210134857
Keen, S. (2007). Empathy and the novel. New York: Oxford University Press.
Kessler, M., & Donahue, M. (2008, November). Reading comprehension and social inferencing
in middle school students with learning disabilities. Research poster presented at the
American Speech-Language-Hearing Association Annual Convention, Chicago, IL.
King, P.M. & Kitchener, K.S. (1994). Developing reflective judgment: Understanding and
35
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
promoting intellectual growth and critical thinking in adolescents and adults. San
Francisco: Jossey-Bass.
Kuhn, D. (1991). The skills of argument. Cambridge, UK: Cambridge University Press.
Kuhn, D., Shaw, V., & Felton, M. (1997). Effects of dyadic interaction on argumentative
reasoning. Cognition and Instruction, 15, 287--315. doi:10.1207/s1532690xci1503_1
Lee, C. (2004). Literacy in the academic disciplines and the needs of adolescent struggling
readers. Voices in Urban Education, 3, 14-25.
Lee, V. E. (2000). Using hierarchical linear modeling to study social contexts: The case of
school effects. Educational Psychologist, 35, 125--141. doi:10.1207/S15326985EP3502_6
Leverage, P., Mancing, H., Schweickert, R., & William, J. M. (Eds.) (2011). Theory of mind and
literature. W. Lafayette, IN: Purdue University Press.
Lynch, J. & van den Broek, P. (2007). Understanding the glue of narrative structure: Children‟s
on- and off-line inferences about characters' goals. Cognitive Development, 22, 323-330.
doi:10.1016/j.cogdev.2007.02.002
Macaruso, P., & Shankweiler, D. (2010). Expanding the simple view of reading in accounting for
reading skills in community college students. Reading Psychology, 31, 454-471.
doi:10.1080/02702710903241363
Martin, M.O. & Mullis, I.V.S. (Eds.). (2013). TIMSS and PIRLS 2011: Relationships among
reading, mathematics, and science achievement at the fourth grade---Implications for
early learning. Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston
College.
MacGinitie, W., MacGinitie, R., Maria, K., & Dreyer, L. (2000). Gates-MacGinitie Reading
36
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Tests, Fourth Edition. New York: Riverside Publishing.
National Center for Educational Statistics (NCES). (2015). Selected findings from PISA 2012.
Retrieved August 4, 2015 from https://nces.ed.gov/surveys/pisa/pisa2012/pisa2012
highlights_1.asp
National Center for Education Statistics. (2013). The nation’s report card: Trends in academic
progress 2012. Washington, DC: Institute of Education Sciences, U.S. Department of

Education.
National Institute of Child Health and Human Development (NICHD). (2000). Report of the
National Reading Panel. Teaching children to read: An evidence-based assessment of
the scientific research literature on reading and its implications for reading instruction
(NIH Publication No. 00-4769). Washington, DC: U.S. Government Printing Office.
Retrieved from http://nationalreadingpanel.org.
National Governors Association (NGA). (2010). Common Core State Standards. Washington,
DC: National Governors Association Center for Best Practices, Council of Chief State
School Officers.
Newton, P., Driver, R., & Osborne, J. (1999). The place of argumentation in the pedagogy of
school science. International Journal of Science Education, 21, 553--576.
doi:10.1080/095006999290570
Norris, S. P., & Phillips, L. (1994). Interpreting pragmatic meaning when reading popular reports
of science. Journal of Research in Science Teaching, 31, 947-967.
doi:10.1002/tea.3660310909
37
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
O'Reilly, T., & Sabatini, J. (2013). Reading for understanding: How performance moderators
and scenarios impact assessment design (Research Report No. RR-13-31). Princeton, NJ:
Educational Testing Service.
O'Reilly, T., Sabatini, J., Halderman, L., Bruce, K., Weeks, J., & Steinberg, J. (2015, March).
Building theoretical and developmentally sensitive reading assessments for students in
3rd through 12th grade: implications for intervention, and potential changes in reading
proficiency. Paper presented at SREE. Washington, D. C.
O'Reilly T., Weeks, J., Sabatini, J., Halderman, L., & Steinberg, J. (2014). Designing reading
comprehension assessments for reading interventions: How a theoretically motivated
assessment can serve as an outcome measure. Educational Psychology Review, 26, 403-
424. doi:10.1007/s10648-014-9269-z
Organisation for Economic Co-operation and Development (OECD). (2013). PISA 2012
assessment and analytical framework: Mathematics, reading, science, problem solving
and financial literacy. OECD Publishing.
Palikara, O., Dockrell, J. E., & Lindsay, G. (2011). Patterns of change in the reading decoding
and comprehension performance of adolescents with specific language impairment.
Learning Disabilities: A Contemporary Journal, 9, 89-105.
Osborne, J., Erduran, S., & Simon, S. (2004). Enhancing the quality of argumentation in school
science. Journal of Research in Science Teaching, 41, 994--1020. doi:10.1002/tea.20035
Pelletier, J. (2006). Relations among theory of mind, metacognitive language, reading skills and
38
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
higher order story comprehension in L1 and L2 learners. In A. Antonietti, O.Liverta-
Sempio, & A.Marchetti (Eds.), Theory of mind and language in developmental contexts
(pp. 77-92). New York: Springer
Perry, W. G. (1970). Forms of intellectual and ethical development in the college years. New
York: Holt, Rinehart, & Winston.
Proctor, C.P., August, D., Carlo, M.S., & Snow, C.E. (2006). The intriguing role of Spanish
vocabulary knowledge in predicting English reading comprehension. Journal of
Educational Psychology, 98, 159-169. doi: 10.1037/0022-0663.98.1.159
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data
analysis methods (Vol. 1). Thousand Oaks, CA: Sage.
Reardon, S., Valentino, R. and Shores, K. (2012). Patterns of literacy among U.S. students,
Future of Children, 22(2), 17-37.
Sabatini, J., O'Reilly, T., Halderman, L. & Bruce, K. (2014a). Integrating scenario-based and
component reading skill measures to understand the reading behavior or struggling
readers. Learning Disabilities Research & Practice, 29, 36-43. doi: 10.1111/ldrp.12028
Sabatini, J., O'Reilly, T., Halderman, L. & Bruce, K. (2014b). Broadening the scope of reading
comprehension using scenario-based assessments: Preliminary findings and challenges.
International Journal Topics in Cognitive Psychology, 114, 693-723.
Sabatini, J., & O'Reilly, T. (2013). Rationale for a new generation of reading comprehension
assessments. In Miller, B., Cutting, L., & P. McCardle (Eds), Unraveling Reading
Comprehension: Behavioral, Neurobiological, and Genetic Components, (pp. 100-111).
Baltimore, MD: Brookes Publishing.
39
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Sabatini, J., O'Reilly, T., & Deane, P. (2013). Preliminary reading literacy assessment
framework: Foundation and rationale for assessment and system design. (Research
Report No. RR-13-30). Princeton, NJ: Educational Testing Service.
Sandoval, W. A., & Millwood, K. (2005). The quality of students' use of evidence in written
scientific explanations. Cognition and Instruction, 23, 23--55.
doi:10.1207/s1532690xci2301_2
Savage, R. (2001). The „simple view‟ of reading: Some evidence and possible implications.
Educational Psychology in Practice, 17, 17-33. doi:10.1080/02667360120039951
Scarcella, R. (2003). Academic English: A conceptual framework. The University of California
Linguistic Minority Research Institute. Technical Report 2003--1. Santa Barbara, CA:
University of California Linguistic Minority Research Institute
Schleppegrell, M. J. (2001). Linguistic features of the language of schooling. Linguistics and
Education, 14, 431--459. doi:10.1016/S0898-5898(01)00073-0
Schleppegrell, M. J. (2007). The linguistic challenges of mathematics teaching and learning: A
research review. Reading & Writing Quarterly: Overcoming Learning Difficulties, 23,
139--159. doi:10.1080/10573560601158461
Schleppegrell, M. J., Achugar, M., & Oteiza, T. (2004). The grammar of history: Enhancing
content-based instruction through a functional focus on language. In TESOL Quarterly:
A Journal for Teachers of English to Speakers of Other Languages and of Standard
English as a Second Dialect, 38, 67--93. doi: 10.2307/3588259
Schleppegrell, M. J., & de Oliveira, L. C. (2006). An integrated language and content approach
for history teachers. Journal of English for Academic Purposes, 5, 254-268.
40
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
doi:10.1016/j.jeap.2006.08.003
Shanahan, C., Shanahan, T., & Misischia, C. (2011). Analysis of expert readers in three
disciplines: History, mathematics, and chemistry. Journal of Literacy Research, 43, 393-
429. doi: 10.1177/108629611424071
Shantz, C. U. (1982). Thinking about people thinking about people. Contemporary Psychology,
27, 376--377.
Silverman, R. D., Speece, D. L., Harring, J. R., & Ritchey, K. D. (2013). Fluency has a role in
the simple view of reading. Scientific Studies of Reading, 17, 108-133.
doi:10.1080/10888438.2011.618153
Snow, C. & Uccelli, P. (2008). The challenge of academic language. In D. Olson & N. Torrance
(Eds.), The Cambridge Handbook of Literacy (pp. 112-133). New York: Cambridge
University Press.
Tiu, R. D., Thompson, L. A., & Lewis, B. A. (2003). The role of IQ in a component model of
reading. Journal of Learning Disabilities, 36, 424-436.
doi:10.1177/00222194030360050401
Uccelli, P., Phillips Galloway, E., Barr, C., Meneses, A., Sánchez, E., & Dobbs, C. (in press).
Beyond vocabulary: Core Academic Language Skills (CALS) that support text
comprehension. Reading Research Quarterly.
Uccelli, P., Barr, C. D., Dobbs, C. L., Phillips Galloway, E., Meneses, A., & Sánchez, E. (2014).
Core academic language skills (CALS): An expanded operational construct and a novel
instrument to chart school--‐relevant language proficiency in pre-adolescent and
adolescent learners. Applied Psycholinguistics, 36, 1077-1109.
41
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
doi:10.1017/S014271641400006X
Vosniadou, S. (2013). Conceptual change in learning and instruction: The framework theory
approach. In S. Vosniadou (Ed.), International handbook of research on conceptual
change (pp. 11-30). New York: Routledge.
Wood, P. K. (1997). A secondary analysis of claims regarding the Reflective Judgment
interview: Internal consistency, sequentiality and intra-individual differences in ill-

structured problem-solving. Higher education: Handbook of theory and research (pp.
243-312). New York: Agathon.
Young, K. M., & Leinhardt, G. (1998). Writing from primary documents: A way of knowing in
history. Written Communication, 15, 25-86. doi: 10.1177/0741088398015001002
Zimmerman, C., Bisanz, G. L., Bisanz, J., Klein, J. S., & Klein, P. (2001). Science at the
supermarket: A comparison of what appears in the popular press, experts' advice to
readers, and what students want to know. Public Understanding of Science, 10, 37-58.
doi: 10.1088/0963-6625/10/1/303
42
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Table 1 Sample characteristics (N = 1965)
Variable %
Female 51%
Black 40%
White 28%
Asian 3%
Latino 27%
Native American/Pacific Islander 1%
Mixed/Other 1%
Free/Reduced-price lunch 83%
English Language Learners 8%
Special education 14%
Grade 4 22%
Grade 5 23%
Grade 6 29%
Grade 7 26%
43
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Table 2 Means and standard deviations of the measures (N = 1965)
M SD 1 2 3 4 5 6
1. Spring GISA Score 976.87 69.21 -
2. Fall GISA Score 963.56 69.30 0.72 *** -
3. Fall perspective articulation 2.03 0.66 0.38 *** 0.37 *** -
4. Fall perspective positioning 0.61 0.41 0.30 *** 0.23 *** 0.29 *** -
5. Fall Academic Language 0.30 1.04 0.68 *** 0.67 *** 0.46 *** 0.36 -
6. Fall complex reasoning (subsample n
= 1610) 9.52 0.24 0.51 *** 0.50 *** 0.34 *** 0.26 *** 0.64 ***
***
p < .001
44
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Table 3 Predictors of spring GISA scores: Role of academic language and perspective taking (N =
1965)
Model 1 Model 2 Model 3 Model 4 Model 5
Variables coefficient SE coefficient SE coefficient SE coefficient SE coefficient SE
Intercept 967.12 *** 7.65 962.15 *** 8.53 962.26 *** 8.50 957.16 *** 8.65 957.55 *** 8.63
Grade 5 14.27 * 6.40 14.41 * 6.52 13.53 * 6.73 13.64 * 6.78

Grade 6 26.20 *** 6.13 26.26 *** 6.24 24.82 *** 6.44 24.86 *** 6.49
Grade 7 28.03 *** 6.22 27.83 *** 6.34 27.04 *** 6.55 26.96 *** 6.59
Female -0.54 2.25 -3.61 2.30 -0.85 2.15 -2.41 2.21
White 4.16 3.86 4.33 3.82 2.82 3.69 3.00 3.68
Asian 12.47 6.88 12.82 6.82 9.63 6.58 10.01 6.56
Latino 4.97 3.66 5.57 3.62 4.69 3.49 5.07 3.49
Native
American/Pacific
Islander -1.17 14.15 -1.79 14.01 1.08 13.50 0.36 13.46
Mixed/Other 18.66 11.13 18.48 11.02 16.00 10.63 16.10 10.59
Free/Reduced-
price lunch -11.93 *** 3.32 -11.53 *** 3.29 -8.36 ** 3.18 -8.39 ** 3.17
English
Language
Learners -22.64 *** 4.57 -20.67 *** 4.54 -14.30 *** 4.42 -13.83 ** 4.41
45
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Special
education -25.55 *** 3.50 -22.42 *** 3.52 -16.30 *** 3.43 -15.22 *** 3.44
Fall GISA scores 0.59 *** 0.02 0.56 *** 0.02 0.44 *** 0.02 0.43 *** 0.02
Fall perspective
articulation 6.82 *** 2.05 2.65 2.00
Fall perspective
positioning 12.40 *** 3.17 7.91 ** 3.07
Fall academic
language 20.66 *** 1.66 19.54 *** 1.71
Random effects Variance Variance Variance Variance Variance
School level 583.51 459.24 447.64 469.41 462.00
Classroom level 612.64 374.38 396.46 442.68 451.69
Student level 3654.04 2013.30 1971.81 1828.07 1816.22

*
p < .05;
**
p < .01;
***
p < .001
46
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Table 4 Predictors of spring GISA scores: Role of complex reasoning (N = 1615)
Model 1 Model 2 Model 3 Model 4 Model 5

Variables coefficie coefficie coefficie coefficie coefficie
SE SE SE SE SE
nt nt nt nt nt
969. ** 7. 961. ** 9.0 960. ** 9.0 956. ** 9.1 956. ** 9.1
Intercept
83 * 62 77 * 7 75 * 7 85 * 7 86 * 6
12.9 + 11.9 + 7.2 12.0 11.7 7.5
Grade 5 7.2 7.6
9 5 2 7 3 8
26.3 ** 6.7 26.1 ** 6.7 **
7.1 **
7.1
Grade 6 24.8 * 24.8 *
8 * 7 3 * 9 4 3
** ** ** **
6.8 27.6 6.8 27.7 7.2 27.4 7.2
Grade 7 28.5 *
6 2 * 8 9 * 4 8 * 3
- - 2.4 - 2.4 - + 2.4
Female 2.5
2.72 3.99 7 3.89 4 4.36 4
4.2 4.1 4.0 4.0
White 6.61 6.99 + 4.56 4.89
4 8 1 1
13.6 + 7.5 11.8 7.4 7.1 7.1
Asian 9.88 9.49
1 7 4 5 6 4
4.0 3.8 3.8
Latino 7.54 + 7.5 +
4 6.15 6.27
6 5 4
Native
American/P - 15. - 14. 14. - 14.
-1.1
acific 6.96 1 4.56 87 27 0.65 23
Islander
Mixed/Othe 14.1 11. 13.9 11. 12.3 11. 12.4 10.
r 2 67 5 48 3 01 2 99
Free/Reduc -
** 3.6 - ** 3.5 * 3.4 - * 3.4
ed-price 10.9 -7.1
3 9.66 8 4 6.88 4
lunch 8
English - ** - ** -
5.2 5.1 * 5.0 - +
Language 20.2 * 17.0 * 10.2 5
2 7 1 9.75
Learners 2 4 5
- ** - ** - ** - **
Special 3.9 3.9 3.8 3.8
25.0 * 22.1 * 14.1 * 13.8 *
education 9 5 6 6
6 9 8 6
** ** ** **
Fall GISA 0.0 0.0 0.0 0.0
0.6 * 0.54 * 0.43 * 0.42 *
scores 2 2 2 2
47
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Fall
2.1 2.1
perspective 1.43 1.22
9 8
articulation
Fall
** 3.3 * 3.3
perspective 8.69 8.49
5 4
positioning
Fall ** **
1.9 20.1 2.0
academic 21.9 * *
1 3 2
language
Fall **
44.2 6.6 18.0 ** 6.8
complex *
4 8 1 3
reasoning
Random
Variance Variance Variance Variance Variance
effects
School
560.59 456.62 459.47 460 460.17
level
Classroom
597.49 386.76 402.82 477.75 475.41
level
Student
3758.7 2066.31 2000.17 1831.51 1822.72
level
*
p < .05;
**
p < .01; ***p < .001
48
ACCEPTED MANUSCRIPT

Snow 2016

Uploaded by

Copyright:

Available Formats

Snow 2016

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Snow 2016

Uploaded by

Copyright:

Available Formats

Journal of Research on Educational Effectiveness

ISSN: 1934-5747 (Print) 1934-5739 (Online) Journal homepage: http://www.tandfonline.com/loi/uree20

Contributions of Academic Language, Perspective

To link to this article: http://dx.doi.org/10.1080/19345747.2015.1116035

Accepted author version posted online: 06

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

CONTRIBUTIONS TO DEEP READING COMPREHENSION

Contributions of Academic Language, Perspective Taking, and Complex Reasoning to Deep

Jones1, Suzanne Donovan2 and Catherine Snow1

Way, Cambridge, MA 02138, USA. E-mail: maria_larusso@gse.harvard.edu

frequently attended to in instruction or in theories of reading comprehension -- academic

language, perspective taking, and complex reasoning -- predict outcomes on an assessment of

deep reading comprehension. The Global Integrated Scenario-based Assessment (GISA;

domains deserve greater attention in theories of reading comprehension and in instruction.

reading comprehension, academic language, perspective-taking

instructional approaches to successful reading comprehension (RAND, 2000).

attention to oral language as a crucial predictor of comprehension outcomes. It has generated

across several populations, including: monolingual normally developing participants reading in

(Cutting & Scarborough, 2006).

us to propose an enriched model of reading comprehension, a model to explain what we call

to evaluate instructional interventions to promote deep comprehension skills. In this report,

Deep Reading Comprehension

defines the highest level of proficiency as follows:

The default in comprehension instruction in the United State is teaching comprehension

1998), courses to which only a small minority of students have access.

that standardized assessments of reading comprehension rarely measure deep comprehension

difficult to incorporate into relatively brief, group-administered, machine-scorable assessments,

themes or formats (Sabatini, Petscher, O'Reilly, & Truckenmiller, 2015). Furthermore,

Educational Testing Service (ETS) has developed an innovative suite of computer-delivered

assessments called the Global Integrated Scenario-based Assessments (GISA). GISA is a

theoretically-based measure designed to reflect an updated understanding of the construct of

reading comprehension while taking advantage of the affordances of technological delivery

Predictors of Deep Reading Comprehension

effective programs of professional development and instruction for (post)adolescent readers. In

hypothesis to be tested in this study.

Schleppegrell has built programs of professional development around the analysis of

academic language features in high-school content-area texts, to display to teachers constructions

our knowledge, though, in none of these instructional approaches have improvements in

academic language skills been specifically related to changes in reading comprehension

to navigate those varying perspectives, which we refer to as perspective taking, is critical to

understanding literary and historical conflicts, to attributing psychological causality, and to

history of theorizing in literature studies promotes novel reading as a way to enhance

different representations of relevant information. Donahue (2014) hypothesizes that failure of

skills have no correlation with reading comprehension as measured by the California

reading comprehension as measured by responses to script-implicit (i.e., deep inferential)

questions about a narrative.

In processing narratives about what happens when one character deceives or

agreeing with it.

tracking explanations of complex phenomena while reading is a severe challenge to students is

powerpoint presentations -- to convey basic information to students unable to process such

information by reading their textbooks.

moderately to academic and verbal ability (Wood, 1997).

However, children‟s scientific reasoning can be effectively promoted by specific classroom

reporting complex relationships or phenomena (for example, in history, science or literature),

Phillips, 1994; Zimmerman, Bisanz, Bisanz, Klein, & Klein, 2001).

Purpose of the Study

complex reasoning predict outcomes on an assessment of deep reading comprehension.