Content uploaded by Marcela Ruiz Zúñiga
Author content
All content in this area was uploaded by Marcela Ruiz Zúñiga on May 14, 2024
Content may be subject to copyright.
1
Reading Research Quarterly, 0(0)
pp. 1–25 | doi:10.1002/rrq.192
© 2017 International Literacy Association.
ABSTRACT
Although literacy achievement has improved in Chile, adolescents’ under-
performance in reading comprehension is still a serious concern. In English,
core academic- language skills (CALS) have been found to significantly predict
reading comprehension, even controlling for academic vocabulary knowledge.
CALS are high- utility language skills that support reading comprehension
across school content areas. Guided by an operational definition of Spanish
CALS (S- CALS), three goals drove this study: to develop two psychometrically
reliable tests, the S- CALS Instrument and the Spanish Academic Vocabulary
(S- AVoc) Test; to explore the dimensionality of core academic- language pro-
ficiencies, as measured by these two tests; and to examine the contribution
of core academic- language proficiencies to reading comprehension. A cross-
sectional sample of 810 Chilean students (grades 4–8) participated in four as-
sessments that measured standardized reading comprehension, word- reading
fluency, Spanish academic vocabulary, and S- CALS. Using classical test theory
and item response theory analyses, results yielded robust reliability evidence
for both instruments. Consistent with prior research, S- CALS and academic
vocabulary scores displayed upward trends in higher grades yet considerable
within- grade variability. Confirmatory factor analysis revealed that S- CALS
and S- AVoc were best conceptualized as part of a higher order construct,
the Spanish core academic- language and vocabulary skills (S- CALVS). The
aggregated S- CALVS scores predicted reading comprehension, beyond the
contribution of grade, school factors, and word- reading fluency. This study
advances our scientific understanding of CALS as relevant for adolescent lit-
eracy beyond the English language. The high- utility school- relevant language
and vocabulary skills offer promising tools to inform and evaluate innovative
reading comprehension interventions for Spanish- speaking adolescents.
Over the past decade, Chile’s educational system has made
major strides in student learning outcomes, but literacy
achievement among adolescents continues to be an area of
national concern. In the most recent PISA literacy assessments,
despite significant improvement over time, Chilean adolescents
scored, on average, only at the second- lowest performance level of a
6- point scale (Ministerio de Educación, 2011; OECD, 2014, 2016).
Although Chilean students performed, on average, significantly
higher than students from other Latin American countries, results
were far from optimal. Indeed, in PISA 2012, the Chilean literacy
mean (441) was 28 points higher than the Latin American mean but
55 points lower than the OECD average (Agencia de Calidad de la
Educación, 2014). Furthermore, serious socioeconomically based
inequalities are salient in the PISA literacy results, with Chilean
Alejandra Meneses
Pontificia Universidad Católica de Chile,
Santiago
Paola Uccelli
Harvard Graduate School of Education,
Cambridge, Massachusetts, USA
María Verónica Santelices
Pontificia Universidad Católica de Chile,
Santiago
Marcela Ruiz
Daniela Acevedo
Universidad Alberto Hurtado, Santiago,
Chile
Javiera Figueroa
Pontificia Universidad Católica de Chile,
Santiago
Academic Language as a Predictor
of Reading Comprehension in
Monolingual Spanish- Speaking Readers:
Evidence From Chilean Early Adolescents
2 | Reading Research Quarterly, 0(0)
students from low socioeconomic backgrounds per-
forming significantly worse than more privileged stu-
dents (Agencia de Calidad de la Educación, 2014). This
socioeconomic discrepancy has also been observed
across many Latin American countries (Flotts et al.,
2016). Collectively, national and international results
call for research to better understand the sources of dif-
ficulty for adolescents’ reading underperformance, with
the ultimate goal of more effectively supporting literacy
achievement for all adolescents in Chilean classrooms.
Overall, the limited research available has indicated
that by midadolescence, Spanish- speaking students typi-
cally face no major difficulty in decoding the words from a
text, yet they often struggle to comprehend and learn from
texts (De Mier, Amado, & Benítez, 2015; De Mier, Borzone,
& Cupani, 2012; García, Bustos, & Sánchez, 2015; Ripoll &
Aguado, 2014). Reading comprehension continues to be a
challenge for students and an urgent concern for educa-
tional researchers and practitioners in the region. In other
countries and languages, vocabulary knowledge and addi-
tional language skills have been shown to play an increas-
ingly important role in predicting reading comprehension
starting around fourth grade (Adlof, Perfetti, & Catts,
2011; Barr & Uccelli, 2016; Compton & Pearson, 2016;
Connor, 2016; Geva & Farnia, 2012; LaRusso etal., 2016;
Uccelli, Barr, etal., 2015; Uccelli & Meneses, 2015; Uccelli,
Phillips Galloway, Barr, Meneses, & Dobbs, 2015). In
Spanish, however, most studies have focused on decoding
and fluency during the early elementary years, with mini-
mal research on the role of language skills in predicting
adolescents’ text comprehension abilities (García et al.,
2015; Mata, Gallego, & Mieres, 2007; Morales, Verhoeven,
& van Leeuwe, 2011; Sánchez & García, 2009; Sánchez,
García, & Bustos, 2017; Sánchez, González, & García,
2002). The present study examines the contribution of aca-
demic vocabulary and additional high- utility cross-
disciplinary academic- language skills to reading compre-
hension in Chilean monolingual Spanish- speaking
adolescents (grades 4–8), controlling for grade, school fac-
tors, and word- reading fluency. Investigating academic
vocabulary and more general academic- language skills
has the potential to advance our scientific understanding
of predictors of reading comprehension in Spanish, with
the ultimate goal of informing the design of innovative
reading comprehension interventions.
The Contribution of Core
Academic- Language Skills
to Reading Comprehension:
Prior Research in English
Academic language, also called the language of school
texts or the language of science, refers to language forms
and functions used recurrently to support the expres-
sion and comprehension of school curricular content
and scientific learning. Academic language includes
diverse abstract vocabulary, complex structures, logical
connectives, and stance markers to support precise, con-
cise, logically organized, and ref lective communication
(Nagy & Townsend, 2012; Schleppegrell, 2001; Snow &
Uccelli, 2009). These resources, which differ systemati-
cally from those typically used and frequently practiced
in everyday interactions, have been shown to pose diffi-
culty for adolescents’ reading comprehension at school
(Fang & Schleppegrell, 2008; Fang, Schleppegrell, &
Cox, 2006; Hwang, Lawrence, Mo, & Snow, 2015;
LaRusso etal., 2016; Uccelli, Barr, etal., 2015; Uccelli &
Meneses, 2015).
Beyond its contribution to school reading, becom-
ing proficient in academic language is necessary to
master the complex communication skills identified as
crucial in today’s information- based society. In our cur-
rent world, complex communication skills (e.g., arguing
with precision, analyzing a complex text) are among the
most sought- after abilities in higher education and pro-
fessional circles (Levy & Murnane, 2013). Academic-
language skills are also crucial for accessing health,
political, and civic information, which is frequently
communicated through texts written for public dissem-
ination (LeVine, LeVine, Schnell- Anzola, Rowe, &
Dexter, 2012; Levinson, 2012; Shanahan & Shanahan,
2008).
In spite of the centrality of mastering academic lan-
guage to access educational and adult life opportuni-
ties, adolescents today seem not to be adequately or
equitably prepared for the language demands of school-
relevant text comprehension. Indeed, the hypothesis
that drives this study is that later reading comprehen-
sion challenges are in large part the result of adolescents’
struggles to access academic language. This hypothesis
is informed by two sources of research evidence:
psychological models of reading comprehension, which
highlight the importance of language to predict read-
ing, and ethnographic and linguistic studies, which
demonstrate how language learning is inseparable from
context. First, the influential simple view of reading
(SVR) model conceptualizes reading comprehension as
the product of decoding and oral language skills (Gough
& Tunmer, 1986; Hoover & Gough, 1990). By now,
decoding skills have been delineated (e.g., letter knowl-
edge, alphabetic principle, decoding) and extensively
studied in numerous languages, including Spanish
(Georgiou, Torppa, Manolitsis, Lyytinen, & Parrila,
2012; Ziegler etal., 2010; Ziegler & Goswami, 2005).
However, the second broad component of the SVR
equation, oral language, has remained imprecisely
defined (Adlof etal., 2011; Compton & Pearson, 2016;
Language and Reading Research Consortium, 2015;
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 3
LaRusso et al., 2016). In Spanish, language skills have
been only minimally investigated for adolescent readers
(García et al., 2015; Mata et al., 2007; Morales et al.,
2011; Sánchez et al., 2002, 2017; Sánchez & García,
2009). The second line of evidence comes from ethno-
graphic and functional linguistics research. These stud-
ies have revealed that students are enculturated at home
into the language of face- to- face interaction, which pre-
pares them well for colloquial interactions in their
respective communities. However, for many colloqui-
ally fluent students, participating successfully in oral
and written academic discourses tends to be challeng-
ing if they have not been granted ample opportunities
to be socialized into the particular ways of using lan-
guage for school reading and learning (Cummins, 2008;
Heath, 2012; Schleppegrell, 2004; Snow & Uccelli,
2009).
Informed by these prior findings, a new program of
research has proposed and investigated the core
academic- language skills (CALS) construct to examine
the contribution of language skills to reading compre-
hension during early adolescence (grades 4–8). The
CALS construct is defined as “a constellation of the
high- utility language skills that correspond to language
features that are prevalent in academic discourse across
school content areas and that are infrequent in collo-
quial conversations” (Uccelli, Phillips Galloway, etal.,
2015, p. 338). CALS research has offered promising
findings; yet, so far, it has been conducted exclusively
with U.S. English- speaking students.
As part of the English CALS research program, the
English CALS Instrument (CALS- I), an innovative,
theoretically grounded, and psychometrically robust
instrument, has been administered to more than 7,000
English- speaking students and validated through mul-
tiple studies (Uccelli, Barr, etal., 2015; Uccelli, Phillips
Galloway, etal., 2015). Guided by linguistic analyses of
experts’ academic texts, evidence from adolescent lan-
guage development, and descriptive evidence of the lan-
guage demands of U.S. school textbooks and class-
rooms, the CALS- I comprises eight tasks: unpacking
complex words, comprehending complex sentences,
connecting ideas, tracking participants, organizing
analytic texts, understanding writers’ viewpoints,
understanding metalinguistic terms, and identifying
academic register. The final selection of items in the
English CALS- I was informed by classical test theory,
item response theory, and theoretical considerations. A
number of cross- sectional and developmental studies
have offered robust evidence of reliability (α coefficients
of .9) and validity (assessed using various standardized
reading comprehension assessments as criteria). Factor
analyses support the unidimensionality of the CALS
construct (comparative fit index [CFI]= 0.93, Tucker–
Lewis index [TLI] = 0.92, root mean square error of
approximation [RMSEA] < 0.05). More important,
within- and across- grade individual variability in
English CALS has been found to be substantive and
predictive of reading comprehension, even after con-
trolling for sociodemographic factors, word- reading
fluency, and academic vocabulary knowledge (Barr &
Uccelli, 2016; Phillips Galloway, 2016; Uccelli, Barr,
et al., 2015; Uccelli, Phillips Galloway, et al., 2015).
Furthermore, in recent theoretical models seeking to
expand the SVR, CALS—alongside other sociocogni-
tive skills—have been identified as significant contribu-
tors to adolescents’ reading comprehension (LaRusso
etal., 2016).
A question that emerges from these promising find-
ings is whether the CALS construct would be relevant
for languages other than English and educational con-
texts other than the United States. In this study, we
extend this line of research to Spanish, with the ulti-
mate motivation of identifying potentially malleable
skills that can later inform research- based Spanish
reading comprehension instruction.
Core Academic- Language
Proficiency in Spanish:
A Still Unexplored Predictor
of Chilean Adolescents’
Reading Comprehension
In Chile, recent important research efforts have been
addressing early and elementary education (Bravo
Valdivieso, Villalón, & Orellana, 2004, 2006; Marchant,
Lucchini, & Cuadrado, 2007; Strasser, Larraín, & Lissi,
2013; Treviño, Toledo, & Gempp, 2013). Although these
have been important initiatives, extensive research has
shown that without support throughout the later years,
positive outcomes achieved during elementary school
are likely to fade, particularly for vulnerable popula-
tions (Biancarosa & Snow, 2006).
Although some reading comprehension research
has been conducted with Spanish- speaking early ado-
lescents in Latin America (Canet Juric, Urquijo,
Richard’s, & Burin, 2009; De Mier etal., 2012; García
et al., 2015; Morales et al., 2011; Thorne et al., 2013),
most has focused on early word decoding. Prior
research has identified code- based skills, such as
reading accuracy and word- reading fluency, as signifi-
cant predictors of early reading proficiency (Bravo-
Valdivieso & Escobar, 2014; Bravo Valdivieso et al.,
2004, 2006; Guardia, 2014). Reading fluency has also
been found to be positively associated with academic
achievement in fourth and sixth graders (Marchant
et al., 2007). Beyond basic code- level skills, advances
have been made in revealing the role of vocabulary
4 | Reading Research Quarterly, 0(0)
knowledge, comprehension monitoring, and theory of
mind, yet so far only with early narrative listening com-
prehension (Strasser & del Río, 2014; Strasser, del Río, &
Larraín, 2013; Strasser, Larraín, & Lissi, 2013). As dis-
cussed previously, research has suggested that during
the upper elementary and middle school years, decod-
ing skills no longer pose a major challenge to early ado-
lescents (García etal., 2015; Mata etal., 2007; Morales
etal., 2011; Sánchez & García, 2009). Given that Spanish
has a fairly transparent orthography, the early and rela-
tively easy mastery of code- level skills is not surprising
(Ziegler & Goswami, 2005). Yet, in light of the widely
documented later struggles with reading comprehen-
sion, these findings highlight the need to investigate
additional contributors to text comprehension, such as
language proficiency.
In this study, we explore for the first time the rele-
vance of the CALS construct for a language other than
English by examining cross- disciplinary Spanish aca-
demic vocabulary knowledge (S- AVoc) and additional
Spanish CALS (S- CALS) as predictors of reading com-
prehension in monolingual Spanish- speaking adoles-
cents. We opted for measuring academic vocabulary
independently because, in contrast to the extensive
research on the impact of academic vocabulary on
English reading comprehension (Lesaux, Crosson,
Kieffer, & Pierce, 2010; Mancilla- Martinez & Lesaux,
2010; Stahl & Nagy, 2006), this contribution has not
been investigated in Spanish- speaking adolescents.
This decision is supported by recent findings that
revealed two distinguishable language constructs
associated with reading development in English:
vocabulary knowledge and additional areas of lan-
guage knowledge, including morphology, syntax, and
oral comprehension (Connor, 2016). Moreover, one
illuminating finding replicated across studies in
English is that CALS and academic vocabulary knowl-
edge, despite their high correlation, independently
contribute to reading comprehension (Uccelli, Barr,
etal., 2015; Uccelli, Phillips Galloway, etal., 2015).
Thus, a subgoal of the present study was to examine
whether S- AVoc and S- CALS together—independently
or as an aggregated predictor—would explain reading
comprehension better than regression models that
include only one of these predictors. This study expands
our prior CALS research by investigating for the first
time the likely hypothesis that academic vocabulary
might be distinguishable from CALS yet still part of a
common underlying higher order construct of core
academic- language proficiencies. Given that no instru-
ments were available to measure either of these con-
structs, the first step involved proposing operational
definitions and developing theoretically and psycho-
metrically robust instruments to measure S- CALS and
S- AVo c.
The Current Study
In this study, we first proposed operational definitions
of S- CALS and S- AVoc and developed the Spanish
CALS Instrument (S- CALS- I; in Spanish, Evaluación de
Lenguaje Académico) and the Spanish Academic
Vocabulary Test (S- AVoc- T). Second, we used item- level
confirmatory factor analysis (CFA) to test whether the
constellation of S- CALS- I items functioned as a unidi-
mensional construct, as found for the English CALS;
then, we examined S- CALS- I and S- AVoc- T together to
explore whether these two sets of items might function
as a unidimensional construct or possibly as dimen-
sions of a common underlying higher order construct.
Finally, we tested our main hypothesis of whether, after
controlling for grade, school factors, and word- reading
fluency, academic vocabulary knowledge and S- CALS
would significantly contribute to reading comprehen-
sion in a socioeconomically diverse sample of fourth- to
eighth- grade students from urban Santiago, Chile.
Three research questions guided this study:
1. In grades 4–8, do Spanish CALS (S-CALS) or
academic vocabulary knowledge (S-AVoc) vary
within and across grades and schools?
2. In grades 4–8, based on students’ performances
in the S-CALS-I items, do the constellation of
CALS function as a unitary construct? Based on
students’ performances in both the S-CALS-I
items and the S-AVoc-T items, do the CALS and
vocabulary knowledge function as a unitary or
higher order construct?
3. Controlling for grade, school factors, and word-
reading fluency, do fourth- to eighth-grade stu-
dents’ S-CALS or academic vocabulary knowl-
edge, independently or as an ag gregated predictor,
contribute to explain the variability in reading
comprehension?
S- CALS and
Academic Vocabulary:
Operational Definitions
In this section, we introduce the proposed operational
definitions of S- CALS and S- AVoc and briefly describe
the development of each respective assessment.
Adopting the English CALS definition, the S- CALS
construct is defined as the array of high- utility Spanish
skills that correspond to linguistic and discourse
resources that are both prevalent in academic discourse
across school content areas and infrequent in colloquial
interactions (Uccelli, Barr, etal., 2015; Uccelli, Phillips
Galloway, etal., 2015). Guided by rigorous psychomet-
ric standards (Kane, 2016; Messick, 1989, 1995), the
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 5
development of the S- CALS- I for monolingual Spanish
speakers in grades 4–8 followed the assessment design
blueprint of the English CALS- I. Given the close iso-
morphism between Spanish and English for the skill
sets already included in the English CALS construct, all
tasks were hypothesized as relevant for measuring high-
utility cross- disciplinary academic Spanish skills. After
a review of relevant Spanish research and extensive
pilot testing of all tasks and item types, the original
eight CALS tasks were maintained with some adapta-
tions. As explained in the Methods section, the target
structures tested in the S- CALS- I were selected after
systematic consultation with available corpora of
Spanish academic discourse and a frequency analysis of
Chilean Spanish textbooks across content areas. Next,
we briefly review the available research evidence from
Spanish that supports the inclusion of each of the eight
skill sets/tasks in the S- CALS- I. (For descriptions of
S- CALS- I tasks and sample items, see Appendix A.)
Packing and Unpacking
Nominalizations
Studies conducted in English have demonstrated that
skills in composing and decomposing morphologically
derived words are positively associated with reading
comprehension (Kieffer & Lesaux, 2007, 2012; Nagy,
Berninger, & Abbott, 2006). In Spanish, morphologi-
cally derived words, especially nominalizations, also
pose difficulties for readers’ comprehension of academic
texts (Chamorro, Barletta, & Mizuno, 2013; Cinto, 2009;
Cuñarro, 2010; García Negroni, Hall, & Marín, 2005;
Mizuno & Moss, 2009). A nominalization is the gram-
matical transformation of a verb or adjective into a noun
(e.g., evaporar→ evaporación [evaporate→ evapora-
tion], complejo→ complejidad [complex→ complexity]).
Nominalizations are highly prevalent in Spanish aca-
demic writing and in middle school textbooks across
content areas (Chamorro et al., 2013; Colombi, 2000;
Moss, Barletta, Chamorro, & Mizuno, 2013). As in the
CALS- I, this task was inspired by Kieffer’s (2009) mor-
phological task, which was an adaptation of Carlisle’s
(2000) original measure. Preliminary analyses indicated
that including both morphological decomposition and
morphological derivation items improved the accuracy
of the test in differentiating students’ performances.
Thus, this S- CALS- I task includes both decomposition
items (also present in the CALS- I) and derivation items
(an item type present only in the S- CALS- I).
Organizing Compact
and Complex Sentences
Complex and compact syntactic structures that enable
writers to pack dense information for concise commu-
nication (e.g., embedded clauses, extended noun
phrases) appear frequently in academic texts both in
English (Fang etal., 2006; Fang & Schleppegrell, 2008;
Schleppegrell, 2001) and in Spanish (Battaner, Atienza,
López- Ferrero, & Pujol, 2009; Colombi, 2000; Soto &
Zenteno, 2004; Venegas, 2010). Consistent with research
conducted in English, monolingual Spanish readers’
syntactic skills have been found to be positively associ-
ated with reading comprehension (Mata et al., 2007;
Navarro & Rodríguez, 2014). In the English CALS- I, the
syntactic skills task was modeled after the widely used
Test for Reception of Grammar—version 2 (Bishop,
2003). Given that this test was not available in Spanish,
the syntactic skills task was modeled after Navarro and
Rodríguez’s (2011, 2014) syntax assessment, designed
and validated for Spanish- speaking middle schoolers.
Items in this S- CALS task test a variety of syntactic
structures prevalent in academic texts, such as sen-
tences with extended noun phrases and a single verb
predicate, complex sentences with embedded clauses,
passive sentences using se (e.g., “Se encontró un tesoro
que estaba enterrado en el jardín” [“A treasure that was
buried in the garden was found”]).
Connecting Ideas Logically
Logical connectives are central textual devices that sup-
port cohesion in informationa l school texts. Connectives
make explicit the intended conceptual relations between
ideas (Álvarez, 2001; Calsamiglia & Tusón, 2002;
Halliday & Hasan, 1976). Yet, if academic connectives
are unknown, these supposedly helpful signaling
devices instead become roadblocks for readers. Early
adolescents’ individual variability in understanding
connectives as cues into how ideas are linked in a text
has been shown to be considerable and positively asso-
ciated with text comprehension in both English (Ben-
Anath, 2005; Crosson, Lesaux, & Martiniello, 2008;
Maury & Teisserenc, 2005;) and Spanish (García etal.,
2015; Sánchez etal., 2002; Sánchez & García, 2009). In
addition to the sources consulted for the overall
S- CALS- I design, an online Spanish dictionary of dis-
course markers (Briz, Pons, & Portolés, 2008) informed
the selection of connectives to be tested. This S- CALS
task includes additive, contrastive, causal, and conces-
sive connectives characteristic of Spanish academic dis-
course and used frequently in middle school texts across
content areas (Álvarez, 2001; Errázuriz, 2012; Martín
Zorraquino & Montolío, 1998; Montolío, 2001).
Tracking Participants and Themes
Coreference chains link to the different expressions
used throughout a text to refer to the same participant,
event, concept, or process and are frequently used in
academic texts as a cohesion mechanism. A specific
type, conceptual anaphora, is especially frequent in
6 | Reading Research Quarterly, 0(0)
academic texts in English and Spanish (e.g., “Movable
printing technology was invented toward the end of the
15th century, and this invention…”). These noun
phrases or pronouns encapsulate an entire clause or
even complex sentences mentioned earlier in the text.
Not surprisingly, being unskilled at resolving concep-
tual anaphora makes text comprehension harder (Biber
& Conrad, 2009; Flowerdew, 2003). In Spanish, middle
school students’ skill in resolving conceptual anaphora
has been found to be positively related to reading com-
prehension (García et al., 2015; Sánchez et al., 2002;
Sánchez & García, 2009). Similar to the CALS- I, items
in this task were designed using authentic fragments
selected from middle school textbooks that were repre-
sentative of prevalent ways of tracking referents in
Spanish academic discourse (González, Cervera, &
Miralles, 1998).
Interpreting Writers’ Viewpoints
The capacity to understand markers that convey the
writer’s viewpoint or attitude is another skill that we
included as supportive of reading comprehension.
Among the multiple types of attitudes that writers
might convey, such as affectionate, deontic, or epistemic
(Berman, Ragnarsdóttir, & Strömqvist, 2002), we
focused only on epistemic markers, those that convey a
writer’s degree of certainty with respect to a specific
claim. Epistemic markers (e.g., ciertamente [certainly],
sin lugar a dudas [undoubtedly]) are common in English
(Hyland, 2009; Hyland & Tse, 2004) and Spanish aca-
demic texts (De Saeger, 2007; López Ferrero, 2001).
Given that, arguably, the main goals of academic dis-
course are to communicate, construct, or critique scien-
tific knowledge, qualifying the degree of certainty of
the facts, scientific findings, or interpretation provided
by the writer is central to fully grasp the meaning of
academic texts across disciplines.
Prior cross- sectional research has documented a
developmental progression in the use of epistemic
markers in Spanish- speaking students from the upper
elementary to the high school years (Berman, 2005;
Berman etal., 2002). Yet, no research has directly exam-
ined this skill in relation to reading comprehension in
Spanish. Despite this lack of research, we focused on
epistemic markers because of their central role and
prevalence in academic discourse across disciplines and
because, in many instances, text comprehension may
indeed depend on understanding these subtle markers
(e.g., “Scientists agree that it is extremely likely/unlikely
that humans are contributing to global warming”).
Understanding Metalinguistic Terms
Metalinguistic terms are words that refer to discourse
or thought processes related to argumentation and text
discussion, such as citar (to quote) and generalización
(generalization; Astington & Olson, 1990; Myhill &
Jones, 2015). By naming linguistic and cognitive pro-
cesses central to text discussion and argumentation in
academic contexts, metalinguistic vocabulary makes
these processes more visible for students. Despite less
research in this domain in Spanish (Camps, 2009;
Camps & Milian, 2000; Flórez et al., 2005; Milian &
Camps, 2006), we hypothesized that knowledge of these
terms would also capture individual variability relevant
to explain reading comprehension. Items in this task
are very different from those of conventional vocabu-
lary tests. Modeled after the English CALS- I items, in
this task, students are presented with two statements
and asked to select, from among four options, the cor-
rect metalinguistic term that qualifies or labels the sec-
ond statement in relation to the first (e.g., whether the
second statement is an example, a generalization, a
quote, or a contradiction).
Organizing Analytic Texts
Following the English CALS- I, we anticipated that
skill in organizing the components of analytic texts,
more specifically of argumentative texts (e.g., thesis,
argument, counterargument, rebuttal, conclusion),
would contribute to students’ reading comprehension.
Two lines of evidence supported this decision. First,
narrative structures are typically mastered by the
upper elementary years, but learning how to structure
analytic texts is a process that typically continues
throughout adolescence (Berman & Ravid, 2009).
Second, during the elementary school years, knowl-
edge of narrative structure has been shown to be pre-
dictive of narrative text comprehension (Cain &
Oakhill, 2006; Perfetti, Landi, & Oakhill, 2005).
Analogously, we anticipated that argumentative text
organization skills would impact reading comprehen-
sion throughout middle school. We focused on argu-
mentative texts because of their prominence in aca-
demic discourse, their pedagogical relevance, and
their somewhat predictable logical structure (Rex,
Thomas, & Engel, 2010; Reznitskaya, Anderson, &
Kuo, 2007). In English, narrative text comprehension
has shown to be, overall, easier than expository text
comprehension (McNamara, Ozuru, & Floyd, 2011).
In Spanish, García et al. (2015) showed that under-
standing expository texts’ macro- organization is posi-
tively implicated in upper elementary students’ read-
ing comprehension. Modeled after the English CALS- I,
in this task, students are presented with components
of an argumentative text (e.g., thesis, argument, con-
clusion) in random order and asked to sequence these
components according to a conventional argumenta-
tive text structure.
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 7
Identifying Academic Register
This domain captures the capacity to distinguish lan-
guage that is characteristic of informal face- to- face
communication from more academic language typi-
cally used in school tasks. Written definitions offer a
convenient minimal genre to test academic register
awareness. Definitions can be written on a continuum
from more colloquial to more academic dictionary-
like entries (Benelli, Belacchi, Gini, & Lucangeli, 2006;
Kurland & Snow, 1997). Moreover, academic defini-
tions, such as those found in the Oxford English
Dictionary or the Diccionario de la Lengua Española,
display language resources characteristic of the aca-
demic register, that is, lexical precision and syntactic
complexity. As we did in English, we hypothesized that
this domain would be relevant for reading comprehen-
sion, based on research on early register awareness and
studies linking definitional skills with later reading
comprehension (Benelli etal., 2006; Hernández, 2008;
Murillo- Rojas, 2014; Snow, 1990). In this task, students
are provided with three definitions of a well- known
word (e.g., paraguas [umbrella]) that have been system-
atically manipulated to present a range from more to
less lexically precise and syntactically concise. Students
are then asked to compare the three and select the
most academic, dictionary- like definition, one that
would be appropriate for an academic dictionary for
adults.
In addition to the tasks that correspond to the
English CALS- I, in this study, we also designed the
S- AVoc- T. Given the unavailability of a cross-
disciplinary academic vocabulary test in Spanish, this
step was necessary to accomplish our subgoal of test-
ing the potential independent contribution of aca-
demic vocabulary and CALS to reading comprehen-
sion. Cross- disciplinary academic vocabulary, also
called all- purpose academic vocabulary, is high- utility
content words (i.e., nouns, verbs, adjectives, adverbs)
found frequently in texts across content areas yet rare
in adolescents’ colloquial interactions (e.g., patron
[pattern], sistema [system]; Hwang et al., 2015; Zeno,
Ivens, Millard, & Duvvuri, 1995). The S- AVoc- T was
translated and adapted from the Word Generation
Academic Vocabulary Test of English academic words
(Hwang etal., 2015), which measures the knowledge of
cross- disciplinary content words by presenting the
target word underlined in a sentence and asking stu-
dents to select the correct synonym from among four
options.
The development of the S- CALS- I and the S- AVoc- T
was not a simple translation of the English CALS- I and
Word Generation Academic Vocabulary Test items for
mere linguistic equivalence. Instead, the functional and
cultural relevance and adequacy of each item was taken
into consideration in the initial design and throughout
the final testing phase (Peña, 2007).
Methods
Participants
As shown in Table1, a total of 810 students, similarly
distributed across grades 4–8, participated in this study.
This cross- sectional sample comprised slightly more
females (56%) than males. Participants attended one of
three schools with different socioeconomic levels in
urban Santiago, Chile. Data on socioeconomic status
(SES) and parents’ education were not available at the
student level, so we used information on mean family
income and parental levels of education available
through a national agency (Agencia de Calidad de la
Educación, 2013). The first was a public school serving
students from low socioeconomic backgrounds, whose
families had an average monthly salary of US$350 and
whose parents had attained a maximum of eight years
of education. The second was a publicly subsidized
school serving students from middle income communi-
ties, whose families had an average monthly salary of
US$750 and whose parents had attained between 11 and
12 years of education. The third was a private school
serving students from high socioeconomic back-
grounds, whose families had an average monthly salary
of US$2,000 and whose parents had attained 15 or more
years of education (Agencia de Calidad de la Educación,
TAB LE 1
Sociodemographic Characteristics of the Study Sample
(N = 810)
Characteristic nPercentage
Gender
Female 454 56.0
Male 356 44.0
Grade
4 163 20.1
5 173 21.4
6 165 20.4
7 155 19.1
8 154 19.0
Socioeconomic status of the school
High 265 32.7
Medium 281 34.7
Low 264 32.6
8 | Reading Research Quarterly, 0(0)
2013). The three schools were comparable in size, in
student–teacher ratio (mean [M] = 27 students per
classroom), and in their implementation of similar cur-
ricula required to follow national grade- specific stan-
dards. This sample reflects the reality of school segrega-
tion in the Chilean educational system, in which
students from similar SES backgrounds attend the same
type of school (Elacqua, 2012; Valenzuela, Bellei, & de
los Ríos, 2013).
Written consent to participate in the study was
obtained from all participants, their parents, and their
teachers. Students were tested at school as part of their
regular school day.
Measures
Four group- administered paper- and- pencil assess-
ments were given to participants.
Reading Comprehension
Reading comprehension was assessed using the
30- minute group- administered standardized reading
comprehension assessment Procesos de Lectura
(PROLEC- SE). This assessment covers the age range
from 10 to 16 years. Participants silently read two
expository texts and answer brief constructed- response
questions (n = 20) that assess literal comprehension
(e.g., “How do Australian aboriginals dress?”) and
inferential comprehension (e.g., “Why do Australian
aboriginal hunters follow a kangaroo for several days?”).
Following the standardized scoring rubric, items are
scored as correct (1 point) or incorrect (0 points), with a
maximum possible score of 20 points. The Cronbach’s
alpha for this measure has been reported as .85 (Cuetos
& Ramos, 1997; Ramos & Cuetos, 2011).
Word- Reading Fluency
In the rapid word segmentation task (Elosúa etal., 2012;
López- Escribano, Elosúa, Gómez- Veiga, & García-
Madruga, 2013), a group- administered word- reading
fluency task, participants are asked to identify words in
strings of letters that contain three words each with no
spaces in between. Students are asked to draw a line
between the words. For example, robleautosalón
(oakcarhall) should be parsed as roble/auto/salón
(oak/car/hall). The total score corresponds to the num-
ber of correct words identified in 90 seconds. This silent
word- reading fluency test was inspired by Bråten, Lie,
Andreassen, and Olaussen’s (1999) original test and val-
idated for Spanish readers in Spain. For its application
in this study, the task was minimally adapted, with very
minor changes required to be appropriate for speakers
of Latin American varieties of Spanish. Scores for
Chilean students are interpreted in relation to those
obtained for sixth graders from Spain (M=59.87, stan-
dard deviation [SD] = 12.41, minimum = 32, maxi-
mum=93; Elosúa etal., 2012).
S-CALS
The S- CALS- I is a 60- minute paper- and- pencil, group-
administered research instrument designed to measure
high- utility cross- disciplinary academic language skills
in grades 4–8. As described previously, this assessment
was inspired by the English CALS- I, an innovative, the-
oretically grounded, and psychometrically robust
instrument that has been validated through multiple
studies (Uccelli, Barr, et al., 2015; Uccelli, Phillips
Galloway, etal., 2015). The S- CALS- I entails 53 items
across eight tasks: packing and unpacking nominaliza-
tions, organizing compact and complex sentences, con-
necting ideas logically, tracking participants and themes,
interpreting writers’ viewpoints, understanding meta-
linguistic terms, organizing analytic texts, and identify-
ing academic register. Skills were assessed with a variety
of multiple- choice, sort list, and checkbox items, as well
as one- word written responses. All items were scored as
correct (0 points) or incorrect (1 point), except for the
organizing analytic text task (0–3 points), which was
rescaled to be the same weight of the other tasks. The
total possible score of the final instrument is 53. For ease
of interpretation, results are reported as percentage cor-
rect scores. (See Appendix A for detailed descriptions of
tasks and for item examples.) The Cronbach’s alpha for
this measure has been reported as .88.
The development of the S- CALS- I unfolded in three
phases. Phase 1 entailed three integrated reviews: a
comprehensive review of Spanish textual and develop-
mental linguistics studies to identify whether English
CALS skill sets were relevant for Spanish, an extensive
review of available corpora of Spanish academic texts
(i.e., Corpus de Referencia del Español Actual: Real
Academia Española, 2014; El Corpus del Español:
Davies, 2002) to guide the selection of targets to be
tested, and a systematic analysis of Chilean school text-
books. The two corpora enabled us to gather valuable
evidence of the prevalence of words/structures in
Spanish academic discourse. Words/structures selected
for the S- CALS- I ranged in frequency from 9% to 25%
in the corpora. Because these corpora do not focus on
school texts, we conducted the systematic analysis of
Chilean school textbooks to verify that the selected lan-
guage targets were indeed prevalent in school texts.
Science, language arts, and social studies textbooks
from grades 4–10 were analyzed to verify that a mini-
mum of 20 instances were found across these texts for
each target selected for testing.
In phase 2, a multidisciplinary team of linguists,
educators, and psychometricians engaged in an iterative
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 9
process of item design, pilot testing with small groups
of students and teachers, and item refinement.
Following the design criteria of the English CALS- I,
items were designed to reduce, as much as possible, the
demands of decoding, background knowledge, higher
order reasoning, and nontarget vocabulary knowledge
to more precisely measure the targeted S- CALS.
In phase 3, a pilot battery of the S- CALS- I with a
total of 123 items (more than double the items estimated
to be needed for the final form) was administered to
students in grades 4–8 in two separate sessions (65 min-
utes each). Guided by classical test theory, item response
theory, and theoretical considerations, a final set of 53
items was selected. (See Appendix B for S- CALS- I psy-
chometric information.)
S- AVoc
This group- administered, paper- and- pencil test assesses
students’ knowledge of cross- disciplinary academic
vocabulary (e.g., system, pattern) with 15 multiple-
choice items. For each item, the underlined target word
is embedded in a sentence, and students are asked to
choose the most appropriate synonym from among four
options. The distractors are always an unrelated word, a
phonological associate, and a general semantic associ-
ate. All items are scored as correct (0 points) or incor-
rect (1 point), with a total possible score of 15 points.
(See Appendix A for a detailed description of the test
and an item example.) The Cronbach’s alpha for this
measure has been reported as .80.
The development of the S- AVoc- T followed the same
three- phase design as for the S- CALS- I, starting with
the translation and adaptation of the English Word
Generation Academic Vocabulary Test (Hwang etal.,
2015). In the English version, the target words were
selected mainly from the Academic Word List (Coxhead,
2000), which includes vocabulary words found fre-
quently in academic English texts across content areas.
A similar list does not exist for Spanish. Yet, given that
approximately 70% of words in Coxhead’s Academic
Word List are English words with a corresponding
Spanish cognate (also frequent in academic Spanish),
the same list was used as the initial source. The preva-
lence of the selected words was examined in the Spanish
corpora and the Chilean textbooks to confirm their rel-
evance for Spanish. After an iterative process of pilot
testing, we conducted classical test theory and item
response theory analyses to inform the final selection of
items (see Appendix B).
Analytic Plan
Based on Kane’s (2016) and Messick’s (1989, 1995) psycho-
metric guidelines and the Standards for Educational and
Psychological Testing (American Educational Research
Association, American Psychological Association, &
National Council on Measurement in Education, 1999),
the development of the S- CALS- I and the S- AVoc- T
entailed examining the reliability evidence for each
instrument, the dimensionality of the internal structure
of the S- CALS, followed by that of the S- CALS- I and the
S- AVoc- T together, and the tests’ validity using a reading
comprehension assessment as criterion.
To examine within- and across- grade individual
variability, we generated descriptive statistics by grade
and gender for all four administered tests: S- CALS- I,
S- AVoc- T, word- reading fluency, and reading compre-
hension. Next, we used CFA to examine unidimen-
sional and higher order factor models using item- level
data from the full sample. Items were mapped onto each
task (dimension) based on the S- CALS- I test design
specifications. CFAs were first conducted to test a uni-
dimensional model with the S- CALS- I items only. Next,
we added the S- AVoc- T items to test two models: a uni-
dimensional model with S- CALS- I and S- AVoc- T items
and a second higher order model with eight S- CALS- I
dimensions (one per task) and S- AVoc- T as a ninth
dimension. CFA from the R software (R Development
Core Team, 2013) and, more specifically, the lavaan
package and the diagonally weighted least square esti-
mator were used. Nonresponse data at the item level
were treated as zero because in this type of timed test,
which covers a wide developmental range (grades 4–8),
nonresponse may be interpreted as the student having
an insufficient ability level to perform at the difficulty
level of the item. On average, there was 8.5% nonre-
sponse at the item level across the S- CALS- I and the
S- AVoc- T. Finally, regression analyses were conducted
to examine S- CALS- I and S- AVoc- T scores as predic-
tors of reading comprehension, controlling for school
factors, grade, and word- reading fluency.
Results
Variability in S- CALS and Academic
Vocabulary (Research Question 1)
Participants’ means and standards deviations for the
S- CALS, academic vocabulary, word- reading fluency,
and reading comprehension scores were generated by
grade. As can be seen in Table2, S- CALS- I scores dis-
played within- and across- grade variability, showing
that the instrument was sensitive to individual vari-
ability across the wide span range from fourth to eighth
grade. The mean percentage correct S- CALS- I score
for the full sample was .54 (SD = .23), with grade-
specific mean S- CALS- I scores increasing progres-
sively across higher grades. As expected, grade 4 stu-
dents displayed the lowest S- CALS- I mean (M=.39,
10 | Reading Research Quarterly, 0(0)
SD =.19), and grade 8 students displayed, on average,
the highest S- CALS- I performance (M=.68, SD =.19).
Despite the evident upward developmental trends,
substantive individual variability was detected within
each grade, with grade- specific standard deviations
ranging from .19 to .22. The across- grade upward
trends and the within- grade individual variability can
be observed in the S- CALS- I box plots by grade dis-
played in Figure1a.
Academic vocabulary knowledge (S- AVoc- T) scores
mirrored those of the S- CALS- I. The average percent-
age correct score for the full sample was .57 (SD = .24).
Fourth- grade students demonstrated the lowest level of
academic vocabulary knowledge (M= .43, SD = .19),
whereas eighth- grade students showed the strongest
average performance (M=.70, SD =.22). Similar to the
S- CALS- I results, the S- AVoc- T score box plots in
Figure 1b display an average upward progression in
higher grades yet also considerable individual variabil-
ity within each grade.
In addition, Figure 2 allows us to observe within-
and across- grade performances for each of the
S- CALS- I tasks. A similar upward trend in the higher
grades was apparent for each task.
When results of the S- CALS- I and of the S- AVoc- T
were disaggregated by gender, performances in each
test were found to be very similar across gender. In the
S- CALS- I, female students displayed a mean percentage
correct of .55 (SD =.22), whereas males’ mean was .54
(SD = .24). For S- AVoc- T, female students earned a
mean score of .57 (SD =.23), and males earned a mean
score of .58 (SD =.25).
For word- reading fluency, fourth- grade students
displayed, on average, the lowest performance (52.04
words correctly read in 90 seconds). Intriguingly, sixth-
grade students achieved the highest performance, on
average (80.12), which was only slightly better than
those of seventh- (77.14) and eighth- grade students
(73.46).
TAB LE 2
Descriptive Statistics for the Spanish Core Academic- Language Skills Instrument, the Spanish Academic
Vocabulary Test, Word- Reading Fluency, and Reading Comprehension Scores by Grade (N = 810)
Grade n
Spanish core
academic- language
skills Spanish academic
vocabulary Word- reading fluency Reading
comprehension
MSD MSD MSD MSD
4 163 .39 .19 .43 .19 52.04 28.74 .23 .20
5 173 .46 .21 .50 .22 61.60 31.33 .28 .22
6 165 .58 .22 .58 .24 80.12 51.07 .41 .23
7 155 .63 .22 .65 .23 77.14 37.25 .47 .23
8 154 .68 .19 .70 .22 73.46 34.64 .50 .25
Total 810 .54 .23 .57 .24 68.62 38.76 .38 .25
Note. Spanish core academic- language skills, Spanish academic vocabulary, and reading comprehension are displayed as percentage correct scores;
word- reading fluency is expressed as raw scores.
FIGURE 1
Box Plots by Grade (N = 810): (a) Spanish Core
Academic- Language Skills Instrument Scores and
(b) Spanish Academic Vocabulary Test Scores
(a)
(b)
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 11
In reading comprehension, the mean percentage
correct score for the full sample was only .38. As
expected, fourth- grade students showed the lowest
reading comprehension mean score (M=.23, SD =.20),
and eighth- grade students exhibited the highest mean
score (M=0.50, SD =.25). Still, eight graders, on aver-
age, responded to only half the items correctly, reveal-
ing considerable room for further improvement in
reading comprehension skills.
S- CALS Dimensionality
(Research Question 2)
Pairwise correlational analyses showed that S- CALS- I
task- specific scores were significantly and positively
correlated with one another and with the S- AVoc- T
scores (see Appendix B). Estimates ranged from moder-
ate to high, suggesting that tasks measured interrelated
yet distinguishable dimensions. To investigate the
dimensionality of the S- CALS construct, as measured
by the S- CALS- I, we examined the internal structure of
the S- CALS- I scores by fitting the items to a unitary
model. As displayed in Table3, the S- CALS- I unitary
model exhibited good fit indexes, suggesting similarly
to the research on English CALS that this is a unidi-
mensional construct.
Next, we examined the additional hypothesis that
S- CALS- I scores and S- AVoc- T scores might fit well
together as part of a unitary or higher order construct.
Hence, we first tested an item- level unidimensional
model with S- CALS- I and S- AVoc- T and then a higher
order model with eight S- CALS dimensions (one per
S- CALS- I task) and a ninth dimension for S- AVoc- T.
As shown in Table3, both the unidimensional and
the higher order CFA models exhibited good fit indexes.
Although the improvement of the CFI and TLI of the
higher order model may not seem substantial over those
of the unidimensional models, the RMSEA indicators
of the higher order model and the chi- square test statis-
tics revealed a slightly better fit of the higher order
model over the fit of the unidimensional models.
Despite the adequate model fit indicators of all three
models, results support the better fit of the higher order
model.
FIGURE 2
Mean Percentage Correct Scores for Each Skill of the Spanish Core Academic- Language Skills Instrument by Grade
(N = 810)
12 | Reading Research Quarterly, 0(0)
This higher order model revealed first that, aligned
with the design of the eight S- CALS- I tasks, all items
mapped perfectly into their corresponding dimension,
with each dimension representing one of the eight
S- CALS- I tasks. Second, the S- AVoc- T fit into a ninth
distinguishable dimension, yet still part of a common
underlying higher order factor, which we call the
Spanish core academic- language and vocabulary skills
(S- CALVS) construct.
In the subsequent regression analyses, first, as an
exploratory move, we entered the S- CALS- I and
S- AVoc- T scores as independent predictors just to
observe in more detail the independent contributions of
each to reading comprehension. However, in the final
regression model, informed by these CFA results, we
added the aggregated S- CALVS scores as the main
question predictor.
S- CALS as Predictors of Reading
Comprehension (Research Question 3)
To answer the last research question related to the crite-
rion validity of the S- CALS- I, we used regression analyses
to examine the relation between S- CALS- I and S- AVoc- T
scores and reading comprehension. The baseline model
included grade and a dummy variable for school SES/type
(i.e., high SES/private; medium SES/subsidized; low SES/
public) to control for differences by school. This model
also included word- reading fluency scores, entered later
to assess its independent contribution to reading compre-
hension, controlling for grade and school factors.
As expected, model 1 showed that students in higher
grades displayed higher reading comprehension scores,
controlling for school factors. In line with prior research
on Chilean adolescent readers (Agencia de Calidad de la
Educación, 2014; Ministerio de Educación, 2011), this first
model also showed that in this sample, students in schools
with higher SES had significantly higher reading compre-
hension scores, controlling for grade (see Table4). More
specifically, controlling for grade, students attending the
high- SES/private school or the medium- SES/subsidized
school performed significantly better than those attend-
ing the low- SES/public school. This model accounted for
38% of the variance in reading comprehension. Next, in
model 2, we added word- reading fluency, which also
emerged as a significant predictor, controlling for grade
and school factors, both of which remained significant.
Word- reading fluency accounted only for an additional
0.5% of the variance in reading comprehension.
Models 3a and 3b are offered next as a set of explor-
atory moves to assess the independent contribution of
each of our main question predictors (S- CALS- I and
S- AVoc- T) when entered in separate models. In model
3a, we added S- AVoc- T scores. Then, in model 3b, we
replaced those with S- CALS- I scores. Later, in our final
model 3c, we entered the aggregated S- CALVS. Note
that in these three models, the R- square change is
reported in relation to model 2.
Results from model 3a show that academic vocabu-
lary knowledge predicts reading comprehension, con-
trolling for grade, school SES/type, and word- reading
fluency, contributing to explain an additional 15% of
the variance. In model 3b, S- CALS- I scores are also
identified as a significant predictor of reading compre-
hension above and beyond the contribution of grade,
school factors, and word- reading fluency. Interestingly,
model 3b reveals that the addition of S- CALS results in
an R- square increase even larger than that of academic
vocabulary. Indeed, S- CALS account for an additional
24% of variance (in comparison with model 2), with
this model explaining 62% of the variance in reading
comprehension.
In our final model, guided by the CFA results, we
added the aggregated S- CALVS scores to model 2.
Model 3c shows that S- CALVS was a significant predic-
tor, over and above the contribution of grade, school
factors, and word- reading fluency. This final model
accounted for 63% of the variance in reading compre-
hension. Because the coefficients are all in the same
standardized units, it is possible to compare these coef-
ficients to assess the relative strength of each predictor.
As shown in model 3c, S- CALVS has the largest beta
TAB LE 3
Item- Level Confirmatory Factor Analysis Results: Spanish Core Academic- Language Skills Instrument (S- CALS- I)
and Spanish Academic Vocabulary Test (S- AVoc- T; N = 810)
Model χ2df χ2/df p
Comparative
fit index Tuc ker
–
Lewis
index
Root mean
square error of
approximation (90%
confidence interval)
Unidimensional (S- CALS- I only) 2,220.51 1,325 1.67 .000 .991 .990 .029 [.027, .031]
Unidimensional (S- CALS- I and
S- AVoc- T)
3,297.14 2,210 1.49 .000 .992 .992 .025 [.023, .026]
Nine higher order dimensions
(S- CALS- I and S- AVoc- T)
2,334.27 2,201 1.06 .024 .999 .999 .009 [.004, .012]
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 13
coefficient (0.79), indicating that, compared with all
other predictors in the model, S- CALVS made the
greatest contribution to reading comprehension. Results
indicate that a one standard deviation increase in
S- CALVS leads to a .79 standard deviation increase in
reading comprehension, with all other variables in the
model held constant. Deserving special attention is the
fact that when the aggregated S- CALVS scores are
added to the model, the impact of school factors
becomes nonsignificant. Consistent with our prior
findings in English, these results suggest that students’
cross- disciplinary academic- language performance, as
measured by the S- CALS- I plus the S- AVoc- T, may be
capturing precisely the language skills that are
unequally distributed across socioeconomic lines in
this sample. However, more research with specific
socioeconomic information at the student level is
needed to confirm these findings.
Discussion
This study, to our knowledge, is the first to examine the
contri bution of word- read ing f luency, cross- d isciplin ary
academic vocabulary knowledge, and CALS to reading
comprehension in a Latin American sample of mono-
lingual Spanish- speaking early adolescents. To measure
academic- language proficiencies, two theoretically
robust and psychometrically reliable instruments were
developed: the S- CALS- I and the S- AVoc- T. Findings
revealed that Spanish academic- language proficiencies
significantly predicted reading comprehension, above
and beyond grade, school factors, and word- reading
fluency, in a cross- sectional sample of 810 Chilean
monolingual Spanish- speaking students attending
grades 4–8 in urban Santiago.
Consistent with prior research (Uccelli, Barr, etal.,
2015; Uccelli, Phillips Galloway, et al., 2015), results
revealed, first, that the S- CALS- I and S- AVoc- T scores
were sensitive to individual variability within and
across grades. For both S- CALS and S- AVoc, a steady
progression in mean scores was evident in higher
grades, yet considerable individual variability was
salient in every grade, despite the wide range of grades
covered. Aligned with prior research on socioeconomi-
cally based discrepancies in students’ literacy in Chile
(Agencia de Calidad de la Educación, 2014; Ministerio
de Educación, 2011), results revealed significant differ-
ences by school in academic- language and reading pro-
ficiencies, such that, controlling for grade, students
attending the higher socioeconomic schools performed
significantly better than those attending the low- SES
school.
Second, expanding our prior research on English
CALS, this study used item- level CFA to investigate
theinternal structure of core academic- language profi-
ciencies, including for the first time both CALS and
cross- disciplinary academic vocabulary. Interestingly,
students’ performances on the S- CALS- I tasks and
the S- AVoc- T were best captured as dimensions of a
common underlying higher order construct, the
S- CALVS. This higher order factor model yielded nine
TAB LE 4
Series of Regression Models to Predict Standardized Reading Comprehension by Spanish Core Academic- Language
Skills Instrument and Spanish Academic Vocabulary Test Scores, Controlling for Grade, School Socioeconomic
Status (SES)/Type, and Word- Reading Fluency
Predictor Model 1 Model 2 Model 3a Model 3b Model 3c
Grade .37*** .35*** .17*** .06*.04
High- SES school .55*** .55*** .19*** .01 −.03
Medium- SES school .29*** .30*** .11*** .02 −.01
Word- reading fluency .09** .04 .00 .00
Academic vocabulary (Spanish Academic
Vocabulary Test)
.54***
Core academic language (Spanish Core
Academic- Language Skills Instrument)
.75***
Spanish core academic- language and
vocabulary skills
.79***
Observations 729 729 729 729 729
R2.38 .38 .53 .62 .63
ΔR2.005*.15*** .24*** .25***
Note. For each variable, standardized beta coefficients are reported. The R- square changes for models 3a, 3b, and 3c are reported in relation to model 2.
*p < .05. **p < .01. ***p < .0 01.
14 | Reading Research Quarterly, 0(0)
dimensions, eight of which corresponded exactly to the
different S- CALS- I tasks/skill sets and a ninth repre-
sented by the academic vocabulary items. It is worth
noting that the operational definition of CALS is com-
prehensive enough to encompass lexical, morphologi-
cal, syntactic, and discourse skills and knowledge. For
the sake of precision, we call this aggregated construct
S- CALVS to distinguish it from other CALS research
that has not included, or might not include in the future,
academic vocabulary knowledge as part of the CALS
construct or instrument. The inclusion of cross-
disciplinary academic vocabulary knowledge as a ninth
skill set aligns well with the proposed definition of the
CALS construct. Far from contradicting our prior
research, these results complement them and suggest
that exploring English cross- disciplinary academic
vocabulary and CALS as part of a hypothesized higher
order construct might be a promising step also.
Finally, the aggregated S- CALVS construct was
found to be a significant predictor of students’ reading
comprehension, above and beyond the contribution of
grade, school factors, and word- reading f luency. Adding
S- CALVS to the model with control predictors
explained a larger proportion of the variance in reading
comprehension than adding only S- AVoc. In this final
model, the R- square increases, albeit only minimally, in
relation to the model with S- CALS only. Given the well-
documented role of academic vocabulary knowledge on
reading comprehension, though, we interpret these
results as highlighting the importance of investigating
both S- CALS and S- AVoc, as opposed to only one of
these subconstructs, to better understand and, presum-
ably, more efficiently support Spanish- speaking adoles-
cents’ reading comprehension.
Theoretically, this study provides evidence to begin
to more precisely delineate the vaguely specified lan-
guage comprehension component of the SVR and other
models of reading comprehension for Spanish- speaking
adolescents (Hoover & Gough, 1990; Perfetti & Stafura,
2014). This is particularly important given the urgent
need of more effectively supporting adolescents’ literacy
learning across Latin American countries.
S- CALS as Promising to
Inform Research in Spanish
Reading Comprehension
Two sets of interrelated language- proficiency constructs
were operationalized and measured as potential con-
tributors to Spanish reading comprehension: cross-
disciplinary academic vocabulary and additional CALS.
Consistent with prior findings in English (Hwang etal.,
2015; Mancilla- Martinez & Lesaux, 2011; Uccelli,
Phillips Galloway, etal., 2015), marked individual dif-
ferences in cross- disciplinary academic vocabulary
knowledge were found to be an important source of dif-
ficulty for text comprehension in this sample. Beyond
vocabulary, participants’ CALS, often assumed to be
known by the upper elementary years, varied substan-
tially across individual participants and were, in turn,
predictive of reading comprehension. These results
indicate that the texts that Spanish- speaking early ado-
lescents read to learn at school are challenging not only
for the technical words (e.g., photosynthesis) or general-
purpose academic vocabulary (e.g., process) but also for
the morphological, syntactic, and discourse resources
(e.g., connectives, nominalizations, complex sentences,
text structures) used to precisely and concisely commu-
nicate and organize curricular content in written dis-
course. Paradoxically, the academic- language resources
meant to support precise and clear written communica-
tion are often unfamiliar to adolescent readers and,
consequently, function instead as major roadblocks for
accessing the meaning of texts.
Compared with word- reading fluency, cross-
disciplinary academic- language proficiencies emerge as
more critical predictors of reading comprehension dur-
ing early adolescence. It is important to note, however,
that word- reading fluency was a significant predictor
even in these later grades. That said, the impact of
word- reading fluency became not significant once
S- CALS, academic vocabulary, or the aggregated
S- CALVS were added to the models. These results can
be interpreted in light of the most current theoretical
models of reading comprehension in which knowledge
of word meanings (and presumably language knowl-
edge more broadly) is understood as the crucial inter-
face between word identification and comprehension
(Perfetti, 2007; Perfetti & Stafura, 2014). Research in
other languages has demonstrated that increasing
word- reading speed does not always lead to improved
comprehension and that school- relevant language
knowledge is what is often challenging for large propor-
tions of adolescent readers (Perfetti, 2007; Uccelli,
Phillips Galloway, etal., 2015).
In Spanish, only limited research has pointed to
language knowledge as an important component of
reading comprehension. Prior evidence from two stud-
ies with upper elementary students pointed to vocabu-
lary knowledge as a significant predictor of narrative
text comprehension (Morales etal., 2011; Thorne etal.,
2013). Morales and colleagues found that word decod-
ing, vocabulary knowledge, reading motivation, and
SES directly and positively predicted narrative text
comprehension in Peruvian fourth graders. Consistent
with our results, in this study, gender was not signifi-
cant, whereas word decoding and vocabulary were the
strongest predictors. Morales et al. found also that
child- level SES predicted reading comprehension
directly and also indirectly, especially through its
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 15
impact on vocabulary knowledge. Beyond vocabulary
knowledge, only minimal research has offered evidence
of the associations between adolescents’ language and
reading proficiencies. García and colleagues (2015)
showed that skills in resolving conceptual anaphora
and understanding of organizational signals (e.g.,
“Others thought…,” “Another group…”) positively con-
tributed to literacy performances in monolingual
Spanish- speaking middle school readers. These find-
ings further highlight the importance of conceptualiz-
ing core academic- language proficiency as a construct
that is more inclusive than just academic vocabulary
knowledge.
Minimal intervention research in Spanish has
focused on expanding adolescents’ school- relevant lan-
guage knowledge as a mechanism to improve reading
comprehension. Analogous to the findings in of English
vocabulary interventions in the United States (Deshler,
Palincsar, Biancarosa, & Nair, 2007; Elleman, Lindo,
Morphy, & Compton, 2009), a digital reading interven-
tion focused on improving fifth- grade readers’ vocabu-
lary knowledge and reading strategy use in Peru found
significant gains in participants’ knowledge of taught
vocabulary but failed to detect satisfactory gains in
comprehension of informational texts (Thorne et al.,
2013). Reading interventions might benefit from paying
attention to complementary components of academic
language beyond vocabulary. These skills seem to be
teachable (Jones etal., 2017) and, thus, may boost ado-
lescents’ reading comprehension across Spanish-
speaking countries.
Finally, one finding was particularly intriguing in
our analysis. In our baseline regression model to pre-
dict reading comprehension, the impact of school fac-
tors was significant, such that participants in higher
SES schools performed significantly better, even con-
trolling for grade and word- reading fluency. Yet, the
impact of school factors became not significant once
academic- language skills were entered as predictors.
Because schools differed along many characteristics in
this sample, it is not possible to conclude that school
variability was only or mostly limited to socioeconomic
differences. Differences in SES were the most salient
across schools given that they were comparable in size,
in student–teacher ratio, and in their implementation of
similar curricula required to follow national grade-
specific standards. Still, the three schools in this sample
differed in their management structures (i.e., public,
subsidized, private) and potentially in principals’ and
teachers’ professional preparation and/or practices.
That said, in a region affected by deep socioeconomic
disparities, this pattern of results, although intriguing,
is particularly noteworthy. Given prior research on the
impact of SES on vocabulary knowledge and reading
(Morales etal., 2011), it is not unlikely that S- CALS may
capture precisely the skills that are most unequally dis-
tributed across SES groups. Yet, drawing any definite
conclusion on this point is beyond the evidence pro-
vided by this study; future research with individual-
level socioeconomic data is needed to investigate the
role of individual- and school- level SES factors.
In sum, in a region dominated by a focus on early
reading and basic skills (Bravo- Valdivieso & Escobar,
2014; Bravo Valdivieso etal., 2004, 2006; Guardia, 2014;
Marchant etal., 2007), the present results bring to light
not only academic vocabulary knowledge but also the
understanding of additional core academic- language
resources as essential contributors and potential pres-
sure points to improve midadolescence reading com-
prehension instruction (Perfetti & Adlof, 2012) toward
excellence and equity.
Advancing Our Understanding
of the Dimensionality of Core
Academic- Language Proficiencies
This study offers initial evidence in support of a com-
mon underlying higher order construct that includes
both academic vocabulary knowledge and CALS. CFAs
suggest that the item- level data were best represented by
a nine- dimension higher order model. The higher order
model solution is theoretically insightful. First, the nine
dimensions are perfectly aligned with the skill sets pro-
posed in the S- CALS- I and with S- AVoc- T as a ninth
dimension. Second, results highlight that although dis-
tinguishable, all nine dimensions are part of a common
underlying construct. We interpret our CFA results as
coherent with a view of adolescent language learning
not as a process of learning isolated discrete skills but
instead as learning registers, that is, constellations of
language resources used together and recurrently in
particular contexts to achieve specific purposes
(Berman & Ravid, 2009; Biber & Conrad, 2009; Uccelli,
Barr, etal., 2015). This view implies that teaching iso-
lated language skills would not yield significant gains
in reading comprehension. To foster students’ text com-
prehension, morphologically complex words and aca-
demic connectives, for instance, would need to be
learned together while being put to meaningful use in
authentic school- relevant practices (e.g., discussing and
constructing meaningful academic texts). Whereas
some interventions focused on isolated skills have
shown disappointing results (Connor, 2016; Elleman
et al., 2009), only future intervention studies will be
able to test whether more comprehensive language
approaches will indeed yield significant gains.
Our results point to a common underlying factor
for cross- disciplinary skills. Academic- language profi-
ciency, however, encompasses a larger set of subcon-
structs, such as disciplinary language skills (e.g., the
16 | Reading Research Quarterly, 0(0)
language of history or biology), which are unlikely to be
captured by a single common factor. Results should not
be extrapolated to include language skills beyond the
CALVS construct, as operationalized in this study.
From a sociocultural pragmatics- based framework
that understands language learning as inseparable from
context, and reading comprehension as learned socio-
cultural practices, our results indicate that delineating
adolescents’ language skills on the basis of their rele-
vance for reading academic texts offers a promising
path toward understanding school- relevant language
and reading relations.
Toward a Cross- Linguistic
Model of Midadolescence
Reading Comprehension
Overall, the present results are consistent with those
found in prior CALS research in English (Uccelli, Barr,
etal., 2015) yet also show that adaptations and adjust-
ments made to the tasks to measure academic- language
skills in Spanish were needed and suitable for capturing
variability in midadolescent students.
Two lines of research are particularly relevant to
situate our findings in the larger cross- linguistic
research context. First, the minimal contribution of
word- reading fluency detected in this study is aligned
with cross- linguistic research documenting the earlier
mastery of code- based skills in languages with more
transparent orthographies, such as Spanish (compared
with more opaque orthographies, such as English).
Second, our results are consistent with research, from a
variety of languages, documenting language skills as
increasingly predictive of reading comprehension after
the early elementary years, once variability in basic
code- based skills (e.g., decoding, word- reading fluency)
decreases as readers become automatic fluent decoders
(Adlof et al., 2011; Geva & Farnia, 2012; Mancilla-
Martinez & Lesaux, 2011; Morales etal., 2011).
Current theoretical models of midadolescence read-
ing comprehension have expanded the SVR model to
include academic- language proficiency and other socio-
cognitive skills as critical additional contributors of text
comprehension during the midadolescent years (Adlof
etal., 2011; Farnia & Geva, 2013; LaRusso etal., 2016).
Beyond its multicomponential nature, reading compre-
hension is currently understood also as a sociocultur-
ally situated process influenced by the characteristics of
a specific text, reader, and activity affecting the nature
and outcomes of text comprehension (Kintsch, 2004;
RAND Reading Study Group, 2002). In these nuanced
and complex models of reading comprehension, how-
ever, the role of languages as distinct systems (e.g.,
Spanish, English, Mandarin) has not yet been explicitly
theorized. It is worth noting that in contrast to
extensive cross- linguistic investigations on the acquisi-
tion of code- based skills across languages and their
respective alphabetic orthographies, cross- linguistic
research focused on the language components of read-
ing comprehension has been minimal.
The CALS research raises new questions about the
role of academic- language proficiency across different
languages. A first set of questions concerns the rele-
vance and nature of the construct; for example, Is the
distance between colloquial and academic discourse
features in a language sufficient to motivate research on
the CALS construct? If so, does the array of skills
included in a language- specific CALS construct differ
across languages? Whereas the present results reveal a
common set of similar skill sets as relevant for text
comprehension in both Spanish and English, research
still needs to further investigate similarities and differ-
ences between these two languages. A second set of
more ambitious questions concerns the relative contri-
bution of CALS to reading comprehension across lan-
guages and their respective orthographies; for example,
How does the relative impact of code- based skills versus
language skills on reading comprehension vary within
and across languages? We may hypothesize, for
instance, that for more transparent orthographies (e.g.,
Spanish), the impact of academic- language skills might
be stronger during midadolescence than for more
opaque orthographies. Yet, in light of a multidimen-
sional and situated understanding of reading compre-
hension (RAND Reading Study Group, 2002), more
text, reader, and activity factors need to be taken into
account in examining the relative contribution of basic
skills and language skills to reading comprehension.
Relatedly, research conducted by Kintsch (2004),
McNamara and colleagues (2011), and O’Reilly and
McNamara (2007) revealed that readers’ characteristics
(e.g., background knowledge, comprehension skill) and
text factors (more vs. less cohesive texts) interactively
impact reading comprehension. As we have argued
before, academic- language proficiency can be added as
another reader factor to further enrich this line of research
(Uccelli, Phillips Galloway, etal., 2015). Furthermore, we
argue here for the consideration of language as different
systems (e.g., Spanish, English, Mandarin Chinese) as an
additional factor that cuts across reader and text. How
does the relative impact of language skills vary in the con-
text of both reader and text characteristics across lan-
guages? Research on readers with different characteristics
(e.g., high vs. low decoding) within the same language, as
well as across languages (e.g., with transparent vs. non-
transparent orthographies, with different language typol-
ogies or academic discourse traditions), has the potential
to significantly advance theory building and educational
practice not only with monolingual readers cross-
linguistically but also with biliterate readers.
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 17
Limitations and Areas
for Further Research
Certainly, the design of this study limits the conclusions
that we can draw in a number of ways. Larger, more
diverse, multisite, and longitudinal samples of Spanish
speakers are needed to investigate the generalizability
of these findings and to examine individual growth tra-
jectories beyond reporting developmental trends.
Furthermore, this study offers preliminary evidence to
motivate further research on the potential mediating
effect of academic language on the relation between SES
and reading comprehension. Given the prevalent
socioeconomic- based segregation by school in the
Chilean educational system (i.e., students with similar
SES attend the same type of school; Elacqua, 2012;
Valenzuela et al., 2013), future studies need to adopt a
sampling strategy that allows for the closer examina-
tion of SES (e.g., randomly selecting students from a
larger number of schools with the goal of disentangling
socioeconomic factors from other various factors) and
to collect detailed information at the family and school
levels. Furthermore, more comprehensive studies that
measure other important areas, such as background
knowledge, perspective taking, and motivation, need to
be conducted to more accurately investigate the relative
contribution of academic language to monolingual
Spanish- speaking adolescents’ reading comprehension
(LaRusso etal., 2016). More research will be also essen-
tial to understand the relation between the cross-
disciplinary academic- language skills tested here and
learning in a particular discipline, such as science, his-
tory, or mathematics. Research on the precursors of
academic language and on opportunities to learn
S- CALS in and outside of school throughout develop-
ment also needs to be conducted (Uccelli, Demir, Rowe,
Goldin- Meadow, & Levine, 2017).
Academic- language research will benefit also from
cross- linguistic studies. The proposed S- CALS opera-
tional definition and S- CALS instrument are particu-
larly promising in making visible to educators and
researchers this array of school- relevant language skills
to inform future research and practice. The CALS- I
offers an initial entry point for cross- linguistic research
examining English CALS and S- CALS across addi-
tional samples of monolinguals, as well as Spanish–
English bilingual populations. The question of whether
this construct is promising also for other languages
remains to be explored.
NOTES
This research was supported by the David Rockefeller Center for
Latin American Studies as part of the Harvard–Chile Innovation
Initiative, CONICYT 2013, and FONDECYT REGULAR 1150238.
First and foremost, we express our deep gratitude to the many stu-
dents and teachers who participated in this study and to the
administrators who supported this project. We express our special
gratitude also to Christopher Barr, of the University of Houston,
who generously offered his psychometric advice throughout the
S- CALS- I development and validation process. We are also thank-
ful to our colleagues Emily Phillips Galloway, Alejandra Venegas,
and Maximiliano Montenegro for their various suggestions on
instrument design and data analysis. Finally, Diego Torrealba and
Camila Barros deserve immense thanks for their data collection
and Ignacia Jorquera for her cooperation in item design.
REFERENCES
Adlof, S.M., Perfetti, C., & Catts, H. (2011). Developmental changes
in reading comprehension: Implications for assessment and
instruction. In S.J. Samuels & A.E. Farstrup (Eds.), What research
has to say about reading instruction (4th ed., pp. 186–214).
Newark, DE: International Reading Association.
Agencia de Calidad de la Educación. (2013). Metodología de con-
strucción de grupos socioeconómicos: Pruebas SIMCE 2012
[Methodology for the construction of socioeconomic groups:
2012 SIMCE tests]. Santiago, Chile: Departamento de Pruebas
Nacionales, División de Evaluación de Logros de Aprendizaje,
Agencia de Calidad de la Educación. Retrieved from http://www.
agenciaeducacion.cl/wp-content/uploads/2013/02/Metodologia-
de-Construccion-de-Grupos-Socioeconomicos-SIMCE-2012.pdf
Agencia de Calidad de la Educación. (2014). Informe nacional resul-
tados Chile PISA 2012 [National report results Chile PISA 2012].
Santiago, Chile: División de Estudios, Departamento de Estudios
Internacionales, Agencia de Calidad de la Educación. Retrieved
from http://archivos.agenciaeducacion.cl/Informe_Nacional_
Resultados_Chile_PISA_2012.pdf
Álvarez, G. (2001). Textos y discursos: Introducción a la lingüística
del texto [Texts and discourses: Introduction to text linguistics].
Concepción, Chile: Universidad de Concepción.
American Educational Research Association, American
Psychological Association, & National Council on Measurement
in Education. (1999). Standards for Educational and Psychological
Testing. Washington, DC: American Educational Research
Association.
Astington, J., & Olson, D. (1990). Metacognitive and metalinguistic
language: Learning to talk about thought. Applied Psychology,
39(1), 77–87. https://doi.org/10.1111/j.1464-0597.1990.tb01038.x
Barr, C.D., & Uccelli, P. (2016). CALS-I psychometric report.
Unpublished internal report prepared for the Catalyzing
Comprehension through Discussion and Debate research project,
Strategic Education Research Partnership, Washington, DC.
Battaner, P., Atienza, E., López-Ferrero, C., & Pujol, M. (2009).
Características lingüísticas y discursivas del texto académico
[Linguistic and discursive characteristics of academic text].
Textos: Didáctica de la Lengua y la Literatura, 50, 47– 67.
Ben-Anath, D. (2005). The role of connectives in text comprehen-
sion. Working Papers in TESOL & Applied Linguistics, 5(2), 1–27.
https://doi.org/10.7916/D8NP23XX
Benelli, B., Belacchi, C., Gini, G., & Lucangeli, D. (2006). ‘To define
means to say what you know about things’: The development
of definitional skills as metalinguistic acquisition. Journal of
Child Language, 33(1), 71–97. https://doi.org/10.1017/S030500090
50 07312
Berman, R. (2005). Introduction: Developing discourse stance in
different text types and languages. Journal of Pragmatics, 37(2),
105–124. https://doi.org/10.1016/S0378-2166(04)00189- 4
Berman, R., Ragnarsdóttir, H., & Strömqvist, S. (2002). Discourse
stance: Written and spoken language. Written Language and
Literacy, 5(2), 253–287. https://doi.org/10.1075/wll.5.2.06ber
Berman, R., & Ravid, D. (2009). Becoming a literate language user:
Oral and written text construction across adolescence. In
18 | Reading Research Quarterly, 0(0)
D.Olson & N. Torrance (Eds.), The Cambridge handbook of liter-
acy (pp. 92–111). New York, NY: Cambridge University Press.
Biancarosa, G., & Snow, C.E. (2006). Reading next: A vision of action
and research in middle and high school literacy (2nd ed.).
Washington, DC: Alliance for Excellent Education.
Biber, D., & Conrad, S. (2009). Register, genre, and style. Cambridge,
UK: Cambridge University Press.
Bishop, D.V. (2003). Test for Reception of Grammar—version 2
(TROG -2). Oxford, UK: Pearson.
Bråten, I., Lie, A., Andreassen, R., & Olaussen, B.S. (1999). Leisure
time reading and orthographic processes in word recognition
among Norwegian third- and fourth- grade students. Reading
and Writing, 11(1), 65– 8 8. ht t p s: //doi .o rg /10 .102 3/A:10 0 79 76521114
Bravo-Valdivieso, L., & Escobar, J.-P. (2014). How transparent is
Spanish orthography?/¿Cuán transparente es nuestra ortografía
castellana? Estudios de Psicología, 35(3), 442–449. https://doi.org/
10.108 0/02109395. 2014.965455
Bravo Valdivieso, L., Villalón, M., & Orellana, E. (2004). Los pro-
cesos cognitivos y el aprendizaje de la lectura inicial: Diferencias
cognitivas entre buenos lectores y lectores deficientes [The cogni-
tive processes and the learning of initial reading: Cognitive dif-
ferences between good and deficient readers]. Estudios
Pedagógicos, 30, 7–19. https://doi.org/10.4067/S0718-070520040
00100001
Bravo Valdivieso, L., Vil lalón, M., & Orellana, E. (2006). Diferencias
en la predictividad de la lectura entre primer año y cuarto año
básicos [Differences in the predictability of reading between first
grade and fourth grade]. Psykhe, 15(1), 3–11. https://doi.
org/10.4067/S0718-22282006000100001
Briz, A ., Pons, S., & Portolé s, J. (Eds.). (2008). Diccionario de partícu-
las discursivas del español [Dictionary of discursive particles of
Spanish]. Retrieved from http://www.dpde.es/
Cain, K., & Oakhill, J. (2006). Profiles of children with specific
reading c omprehension diff iculties. Briti sh Journal of Educational
Psychology, 76(4), 683–696. https://doi.org/10.1348/000709905x
67610
Calsamiglia, H., & Tusón, A. (2002). La textura discursiva [The dis-
cursive texture]. In H. Calsamiglia & A. Tusón (Eds.), Las cosas
del decir: Manual de análisis del discurso [The things of saying:
Handbook of discourse analysis] (pp. 217–250). Barcelona, Spain:
Ariel.
Camps, A. (2009). Actividad metalingüística y aprendizaje de la
gramática: Hacia un modelo de enseñanza basado en la actividad
reflexiva [Metalinguistic activity and learning grammar:
Towards a teaching model based on ref lexive activity]. Cultura y
Educación, 21(2), 199–213. https://doi.org/10.1174/113564009788
345853
Camps, A., & Milian, M. (2000). Metalinguistc activity in learning
to write: An introduction. In A. Camps & M. Milian (Eds.),
Metalingui stic activity in learning to write (pp. 1–29). Amsterda m,
The Netherlands: Amsterdam University Press.
Canet Juric, L., Urquijo, S., Richard’s, M.M., & Burin, D. (2009).
Predictores cognitivos de niveles de comprensión lectora
mediante análisis discriminante [Cognitive predictors of reading
comprehension levels using discriminant analysis]. International
Journal of Pyschological Research, 2(2), 99 –111.
Carlisle, J.F. (2000). Awareness of the structure and meaning of
morphologically complex words: Impact on reading. Reading and
Writing, 12(3), 169–190. https://doi.org/10.1023/A:100813192
6604
Chamorro, D., Barletta, N., & Mizuno, J. (2013). El lenguaje para
enseñar y aprender las Ciencias Naturales: Un caso de oportuni-
dades perdidas para la formación [The language for teaching and
learning natural science: A case of missed opportunities for citi-
zenship education]. Revista Signos, 46(81), 3–28. https://doi.
org/10.4067/S0718-09342013000100001
Cinto, M.T. (2009). La nominalización: Obstáculo para la compren-
sión lectora [Nominalization: Obstacle for reading comprehen-
sion]. Educación, Lenguaje y Sociedad, 6(6), 171–186.
Colombi, C. (2000). En vías del desarrollo del lenguaje académico
en español en hablantes nativos de español en los Estados Unidos
[In developing academic language in Spanish in native Spanish
speakers in the United States]. In A. Roca (Ed.), Research on
Spanish in the United States: Linguistic issues and challenges
(pp.296–309). Somerville, MA: Cascadilla.
Compton, D.L., & Pearson, P.D. (2016). Identifying robust varia-
tions associated with reading comprehension sk ill: The search for
pressure points. Journal of Research on Educational Effectiveness,
9(2), 223–231. https://doi.org/10.1080/19345747.2016.1149007
Connor, C.M.D. (2016). A lattice model of the development of read-
ing comprehension. Child Development Perspectives, 10(4),
269–274. ht tps: //doi.org/10.1111/cde p.12200
Coxhead, A. (2000). A new academic word list. TESOL Quarterly,
34(2), 213–238. https://doi.org/10.2307/3587951
Crosson, A.C., Lesaux, N.K., & Martiniello, M. (2008). Factors that
inf luence comprehension of connectives among language minor-
ity children from Spanish- speaking backgrounds. Applied
Psycholinguistics, 29(4), 603–625. https://doi.org/10.1017/S014271
6408080260
Cuetos, F., & Ramos, J.L. (1997). Batería de evaluación de los pro-
cesos lectores en el alumnado del tercer ciclo de educación pri-
maria y educación secundaria obligatoria (PROLEC-SE)
[Evaluation battery of reading processes in students in the third
cycle of primary education and compulsory secondary educa-
tion]. Madrid, Spain: TEA.
Cummins, J. (2008). BICS and CALP: Empirical and theoretical sta-
tus of the distinction. In N. Hornberger (Ed.), Encyclopedia of
language and education (2nd ed., pp. 487–499). New York, NY:
Springer.
Cuñarro, M. (2010). Las nominali zaciones: Reconocim iento, compren-
sión y estrategias lingüísticas para su definición [Nominalizations:
Recognition, understanding and linguistic strategies for their defi-
nition]. In V. Castel & Cubo de Severino L. (Eds.), La renovación de
la palabra en el bicente nario de la Argentina: Los colores de la mirada
lingüística [The renew al of the word in the bicente nnial of Argent ina:
The colors of the linguistic gaze] (pp. 413–420). Mendoza,
Argentina: Editorial FFyL.
Davies, M. (2002). El Corpus del Español [The Spanish Corpus].
Retrieved from http://www.corpusdelespanol.org/
De Mier, M., Amado, B., & Benítez, M.E. (2015). Dificultades en la
comprensión de textos expositivos en niños de los primeros gra-
dos de la escuela primaria [Difficulties in the comprehension of
expository texts in children in the first grades of primary
school]. Psykhe, 24(2), 1–13. https://doi.org/10.7764/psykhe.24.
2.708
De Mier, M., Borzone, A., & Cupani, M. (2012). La fluidez lectora en
los primeros grados: Relación entre habilidades de decodifi-
cación, características textuales y comprensión: Un estudio piloto
con niños hablantes de español [Reading fluency in the early
grades: Relation between decoding skills, textual characteristics
and comprehension: A pilot study with Spanish-speaking chil-
dren]. Revista Neuropsicología Latinoamericana, 4(1), 18–33.
https://doi.org/10.5579/rnl.2012.0079
De Saeger, B. (2007). Evidencialidad y modalidad epistémica en los
verbos de actitud proposicional en español [Evidentiality and
epistemic modality in the verbs of propositional attitude in
Spa nish]. Interlingüística, 17, 268–277.
Deshler, D.D., Palincsar, A.S., Biancarosa, G., & Nair, M. (2007).
Informed choices for struggling adolescent readers. Newark, DE:
International Reading Association.
Elacqua, G. (2012). The impact of school choice and public policy
on segregation: Evidence from Chile. International Journal of
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 19
Educational Development, 32(3), 444–453. https://doi.org/10.1016/
j.ijedudev.2011.08.003
Elleman, A.M., Lindo, E.J., Morphy, P., & Compton, D.L. (2009).
The impact of vocabulary instruction on passage- level compre-
hension of school- age children: A meta- analysis. Journal of
Research on Educational Effectiveness, 2(1), 1–44. https://doi.
org/10.1080/19345740802539200
Elosúa, M.-R., García-Madruga, J.-A., Gómez-Veiga, I., López-
Escribano, C., Pérez, E., & Orjales, I. (2012). Habilidades lectoras
y rendimiento académico en 3° y 6° de primaria: Aspectos evolu-
tivos y educativos [Reading skills and academic performance in
grades 3 and 6: Evolutionary and educational aspects]. Estudios
de Psicología, 33(2), 207–218. htt ps://doi.org/10.1174/021093912
800676411
Errázuriz, M.C. (2012). Análisis del uso de los marcadores discursi-
vos en argumentaciones escritas por estudiantes universitarios
[Analysis of the use of discursive markers in arguments written
by university students]. Perfiles Educativos, 34(1 36), 98–117.
Fang, Z., & Schleppegrell, M. (2008). Reading in secondary content
areas: A language-based pedagogy. Ann Arbor: University of
Michigan Press.
Fang, Z., Schleppegrell, M., & Cox, B. (2006). Understanding the
language demands of schooling: Nouns in academic registers.
Journal of Literacy Research, 38(3), 247–273. https://doi.
org/10.1207/s15548430jlr3803_1
Farnia, F., & Geva, E. (2013). Growth and predictors of change in
English language learners’ reading comprehension. Journal of
Research in Reading, 36(4), 389–421. https://doi.org/10.1111/
jrir.12003
Flórez, R., Torrado, M.C., Arévalo, I., Mesa, C., Mondragón, S., &
Pérez, C. (2005). Habilidades metalingüísticas, operaciones
metacognitivas y su relación con los niveles de competencia en
lectura y escritura: Un estudio exploratorio [Metalinguistic abili-
ties, metacognitive operations and their relation with the levels of
competence in reading and writing: An exploratory study].
Forma y Función, 18, 15–44.
Flotts, P., Manzi, J., Jiménez, D., Abarzúa, A., Cayuman, C., &
García, M.J. (2016). Informe de resultados TERCE: Logros de
aprendizaje [TERCE results report: Learning achievements].
Paris, France: Organización de las Naciones Unidas para la
Educación, la Ciencia y la Cultura; Santiago, Chile: Oficina
Regional de Educación para América Latina y el Caribe.
Flowerdew, L. (2003). A combined corpus and systemic- functional
analysis of the problem- solution pattern in a student and profes-
sional corpus of technical writing. TESOL Quarterly, 37(3),
489–511. https://doi.org/10.2307/3588401
García, J., Bustos, A., & Sánchez, E. (2015). The contribution of
knowledge about anaphors, organisational signals and refuta-
tions to reading comprehension. Journal of Research in Reading,
38(4), 405–427. https://doi.org/10.1111/1467-9817.12021
García Negroni, M., Hall, B., & Marín, M. (2005). Ambigüedad,
abstracción y polifonía del discurso académico: Interpretación de
las nominalizaciones [Ambiguity, abstraction and polyphony of
academic discourse: Interpretation of nominalizations]. Revista
Signos, 38(57), 49–60. https://doi.org/10.4067/S0718-0934200500
0100004
Georgiou, G.K., Torppa, M., Manolitsis, G., Lyytinen, H., & Parrila,
R. (2012). Longitudinal predictors of reading and spelling across
languages varying in orthographic consistency. Reading and
Writing, 25(2), 321–346. https://doi.org/10.1007/s11145- 010-
9271-x
Geva, E., & Farnia, F. (2012). Developmental changes in the nature of
language proficiency and reading fluency paint a more complex
view of reading comprehension in ELL and EL1. Reading and
Writing, 25(8), 1819–1845. https://doi.org/10.1007/s11145-011-
9333-8
González, J., Cervera, T., & Miralles, J.L. (1998). La adquisición de
las relaciones anafóricas en castellano: Clases de anáforas y el
efecto de la distancia [The acquisition of anaphoric relations in
Spanish: Classes of anaphora and the effect of distance]. Infancia
y Aprendizaje, 21(82), 21–44. https://doi.org/10.1174/0210370
98320784844
Gough, P.B., & Tunmer, W.E. (1986). Decoding, reading, and read-
ing disability. Remedial and Special Education, 7(1), 6–10. https://
doi.org/10.1177/074193258600700104
Guardia, P. (2014). Is consonant rhyme important when learning to
read in Spanish?/¿Es importante la rima consonante para apren-
der a leer en el español? Estudios de Psicología, 35(3), 567–583.
https://doi.org/10.1080/02109395.2014.965459
Halliday, M.A.K., & Hasan, R. (1976). Cohesion in English. London,
UK: Longman.
Heath, S.B. (2012). Words at work and play: Three decades in family
and community life. Cambridge, UK: Cambridge University
Press.
Hernández, M. (2008). El desarrollo de la competencia definicional:
Estudio con niños costarricenses [The development of defini-
tional competence: A study with Costa Rican children]. Revista
Internacional de Lingüística Iberoamericana, 6(1), 7–29.
Hoover, W.A., & Gough, P.B. (1990). The simple view of reading.
Reading and Writing, 2(2), 127–160. https://doi.org/10.1007/
BF0 0401799
Hwang, J.K ., Lawrence, J.F., Mo, E., & Snow, C.E. (2015). Differentia l
effects of a systematic vocabulary intervention on adolescent lan-
guage minority students with varying levels of English profi-
ciency. The International Journal of Bilingualism, 19(3), 314 –332 .
https://doi.org/10.1177/1367006914521698
Hyland, K. (2009). Academic discourse. London, UK: Continuum.
Hyland, K., & Tse, P. (2004). Metadiscourse in academic writing: A
reappraisal. Applied Linguistics, 25(2), 156–177. https://doi.
org/10.1093/applin/25.2.156
Jones, S.M., LaRusso, M., Kim, J., Kim, H.K., Selman, R., Uccelli, P.,
… Snow, C.E. (2017). Experimental effects of Word Generation on
vocabulary, academic language, perspective taking, and reading
comprehension in high poverty schools. Manuscript submitted for
publication.
Kane, M. (2016). Explicating validity. Assessment in Education:
Principles, Policy & Practice, 23(2), 198–211. https://doi.org/10.10
80/0969594x.2015.1060192
Kieffer, M.J. (2009). The development of morphological awareness
and vocabulary knowledge in adolescent language minority learn-
ers and their classmates (Unpublished doctoral dissertation).
Harvard University, Cambridge, MA.
Kieffer, M.J., & Lesaux, N.K. (2007). Breaking down words to build
meaning: Morphology, vocabulary, and reading comprehension
in the urban classroom. The Reading Teacher, 61(2), 134–144.
https://doi.or g/10 .1598/ RT.61.2. 3
Kieffer, M.J., & Lesaux, N.K. (2012). Development of morphological
awareness and vocabulary knowledge in Spanish- speaking lan-
guage minority learners: A parallel process latent growth curve
model. Applied Psycholinguistics, 33(1), 23–54. https://doi.
org/10.1017/S0142716411000099
Kintsch, W. (2004). The construction-integration model of text
comprehension and its implications for instruction. In R.B.
Ruddell & N.J. Unrau (Eds.), Theoretical models and processes of
reading (5th ed., pp. 1270–1328). Newark, DE: International
Reading Association.
Kurland, B.F., & Snow, C.E. (1997). Longitudinal measurement of
growth in definitional skill. Journal of Child Language, 24(3),
603–625. https://doi.org/10.1017/S0305000997003243
Language and Reading Research Consortium. (2015). Learning to
read: Should we keep things simple? Reading Research Quarterly,
50(2), 151–169. https://doi.org/10.1002/rrq.99
20 | Reading Research Quarterly, 0(0)
LaRusso, M., Kim, H.Y., Selman, R., Uccelli, P., Dawson, T., Jones,
S., … Snow, C. (2016). Contributions of academic language, per-
spective taking, and complex reasoning to deep reading compre-
hension. Journal of Research on Educational Effectiveness, 9(2),
201–222. https://doi.org/10.1080/19345747.2015.1116035
Lesaux, N.K., Crosson, A.C., Kieffer, M.J., & Pierce, M. (2010).
Uneven profiles: Language minority learners’ word reading,
vocabulary, and read ing comprehension skil ls. Journal of Applied
Developmental Psychology, 31(6), 475–483. https://doi.org/10.
1016/j.appde v.2010.09.00 4
LeVine, R.A., LeVine, S., Schnell-Anzola, B., Rowe, M., & Dexter, E.
(2012). Literacy and mothering: How women’s schooling changes
the lives of the world’s children. New York, NY: Oxford University
Press.
Levinson, M. (2012). No citizen left behind. Cambridge, MA:
Harvard University Press.
Levy, F., & Murnane, R.J. (2013). Dancing with robots: Human skills for
computerized work. Washington, DC: Third Way. Retrieved from
http://www.thirdway.org/report/dancing-with-robots-human-
skills-for-computerized-work
López-Escribano, C., Elosúa, M., Gómez-Veiga, I., & García-
Madruga, J. (2013). A predictive study of reading comprehension
in third- grade Spanish students. Psicothema, 25(2), 199–205.
https://doi.org/10.7334/psicothema2012.175
López Ferrero, C. (2001). Personas del disc urso y despersonalizac ión
[Persons of the discourse and depersonalization]. In G. Vásquez
(Ed.), Actividades para la escritura académica [Activities for aca-
demic writing] (pp. 57–66). Madrid, Spain: Edinumen.
Mancilla-Martinez, J., & Lesaux, N.K. (2010). Predictors of reading
comprehension for struggling readers: The case of Spanish speak-
ing language minority learners. Journal of Educational
Psychology, 102(3), 701–711. ht tps://doi.org/10.1037/a0019135
Mancilla-Martinez, J., & Lesaux, N.K. (2011). The gap between
Spanish speakers’ word reading and word knowledge: A longitu-
dinal study. Child Development, 82(5), 1544–1560. https://doi.
org/10.1111/j.1467-8624 .2011.01633.x
Marchant, T., Lucchini, G., & Cuadrado, B. (2007). ¿Por qué leer
bien es importante? Asociación del dominio lector con otros
aprendizajes [Why is reading well so important? Association
between reading mastery and other learning processes].
Psykhe, 16(2), 3–16. https://doi.org/10.4067/S0718-2228200700
0200001
Martín Zorraquino, M.A., & Montolío, M. (1998). Los marcadores
del discurso: Teoria y analisis [Discourse markers: Theory and
analysis]. Madrid, Spain: Arco/Libros.
Mata, S., Gallego, J., & Mieres, C. (2007). Habilidades lingüísticas y
comprensión lectora: Una investigación empírica [Linguistic
skills and reading comprehension: An empirical research].
Bordón: Revista de Pedagogía, 59(1), 153 –166 .
Maury, P., & Teisserenc, A. (2005). The role of connectives in science
text comprehension and memory. Language and Cognitive
Processes, 20(3), 489–512. https://doi.org/10.1080/016909604
44000151
McNamara, D., Ozuru, Y., & Floyd, R. (2011). Comprehension chal-
lenges in the fourth grade: The roles of text cohesion, text genre
and readers’ prior knowledge. International Electronic Journal of
Elementary Education, 4(1), 229–257.
Messick, S. (1989). Validity. In R.L. Linn (Ed.), Educational mea-
surement (3rd ed., pp. 13–103). New York, NY: American Council
on Education & Macmillan.
Messick, S. (1995). Standards of validity and the validity of stan-
dards in performance assessment. Educational Measurement:
Issues a nd Practice, 14(4), 5– 8. ht tps ://do i.or g/10 .1111/j.1 745-3 992 .
1995.tb00881.x
Milian, M., & Camps, A. (2006). El razonamiento metalingüístico
en el marco de las secuencias didácticas de gramática (SDG)
[Metalin guistic reas oning in the fr amework of learnin g sequences
of grammar]. In A. Camps (Ed.), Diálogo e investigación en las
aulas: Investigaciones en didáctica de la lengua [Dialogue and
research in the classroom: Research in the language method] (pp.
25–53). Barcelona, Spain: Graó.
Ministerio de Educación. (2011). Resultados PISA 2009 Chile:
Competencias de los estudiantes chilenos de 15 años en lectura,
matemáticas y ciencias [Results PISA 2009 Chile: Skills of
15-year-old Chilean students in reading, math and science].
Santiago, Chile.
Mizuno, J., & Moss, G. (2009). La ecología: Una noticia de moda: El
registro periodístico y la transitividad en un texto de ciencias
naturales [Ecology: A fashion news: The journalistic record and
the transitivity in a text of natural sciences]. In M. Shiro, P.
Bentivoglio, & F.D. Erlich (Eds.), Haciendo discurso: Homenaje a
Adriana Bolívar [Making discourse: Homage to Adriana Bolivar]
(pp. 647–660). Caracas, Venezuela: Chirimi.
Montolío, M. (2001). Conectores de la lengua escrita:
Contraargumentativos, consecutivos, aditivos y organizadores de
la información [Connectives of written language: Counterargu-
mentatives, consecutives, additives and organizers of informa-
tion]. Barcelona, Spain: Ariel.
Morales, S., Verhoeven, L., & van Leeuwe, T. (2011). Socio- cultural
variation in reading comprehension development among fifth
graders in Peru. Reading and Writing, 24(8), 951–969. https://doi.
org/10.1007/s11145- 010-9242-2
Moss, G., Barletta, N., Chamorro, D., & Mizuno, J. (2013). La metá-
fora gramatical en los textos escolares de Ciencias Sociales en
español [Grammatical metaphor in social sciences textbooks in
Spa nish]. Onomázein, 28, 88 –104. https://doi.org/10.7764/
onomazein.28.3
Murillo-Rojas, M. (2014). Competencia definicional de los escolares
costar ricenses [Defin itional competence in C osta Rican]. Káñina,
38(3), 127–152. https://doi.org/10.15517/rk.v38i3.16459
Myhill, D., & Jones, S. (2015). Conceptualizing metalinguistic
understanding in writing/Conceptualización de la competencia
metalingüística en la escritura. Cultura y Educación, 27(4), 839 –
867. https://doi.org/10.1080/11356405.2015.1089387
Nagy, W., Berninger, V.W., & Abbott, R.D. (2006). Contributions of
morphology beyond phonology to literacy outcomes of upper
elementary and middle- school students. Journal of Educational
Psychology, 98(1), 134–147. https://doi.org/10.1037/0022-0663.
98.1.134
Nagy, W., & Townsend, D. (2012). Words as tools: Learning aca-
demic vocabulary as language acquisition. Reading Research
Quarterly, 47(1), 91–108. htt ps://doi.org/10.1002/RRQ.011
Navarro, J., & Rodríguez, I. (2011). Prueba de evaluación de la con-
ciencia sintáctica (CS-ORD) [Assessment of syntactic awareness].
Unpublished manuscript.
Navarro, J., & Rodríguez, I. (2014). Evaluación de la conciencia sin-
táctica: Efectos de la verosimilitud en la resolución de tareas y en
su relación con la comprensión de oraciones [Assessment of syn-
tactic awareness: Effects of plausibility in solving tasks and in its
relation to sentence comprehension]. Revista Signos, 47(84), 64–
90. https://doi.org/10.4067/S0718-09342014000100004
OECD. (2014). PISA 2012 results: What students know and can do:
Vol. 1. Student performance in mathematics, reading and science
(Rev. ed.). Paris, France: Author.
OECD. (2016). PISA 2015 results: Vol. 1. Excellence and equity in
education. Paris, France: Author.
O’Reilly, T., & McNamara, D.S. (2007). Reversing the reverse cohe-
sion effect: Good text can be better for strategic, high- knowledge
readers. Discourse Processes, 43(2), 121–152.
Peña, E. (2007). Lost in translation: Methodological considerations
in cross- cultural research. Child Development, 78(4), 1255–1264.
http s://doi.or g/10.1111/j.14 67-8624. 2007.0106 4.x
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 21
Perfetti, C.A. (2007). Reading ability: Lexical quality to comprehen-
sion. Scientific Studies of Reading, 11(4), 357–383. https://doi.
org/10.1080/10888430701530730
Perfetti, C.A., & Ad lof, S.M. (2012). Reading comprehension: A con-
ceptual framework from word meaning to text meaning. In J.
Sabatini & E. Albro (Eds.), Assessing reading in the 21st century:
Aligning and applying advances in the reading and measurement
sciences (pp. 3–20). Lanham, MD: Rowman & Littlefield.
Perfetti, C.A., Landi, N., & Oakhill, J. (2005). The acquistion of reading
comprehension ski ll. In M. Snowling & C. Hulme (Eds.), The science
of reading: A handbook (pp. 227–247). Oxford, UK: Blackwell.
Perfetti, C., & Stafura, J. (2014). Word knowledge in a theory of
reading comprehension. Scientific Studies of Reading, 18(1), 22 –
37. https://doi.org/10.1080/10888438.2013.827687
Phillips Galloway, E. (2016). The development of core academic
language and reading comprehension in preadolescent and adoles-
cent learners (Doc toral dissertation). Retrieved from https://dash.
harvard.edu/handle/1/27112709
Ramos, J.L., & Cuetos, F. (2011). Evaluación de los Procesos Lectores:
PROLEC-SE Manual [Assessment of readers’ processes:
PROLEC-SE manual]. Madrid, Spain: TEA.
RAND Reading Study Group. (2002). Reading for understanding.
Toward an R&D program in reading comprehension. Santa
Monica, RAND.
R Development Core Team. (2013). The R project for statistical com-
puting. Vienna, Austria: R Foundation. Retrieved from https://
www.r-project.org/
Real Academia Española. (2014). Banco de datos (CREA), Corpus de
Referencia del Español Actual [Database (CREA), Reference
Corpus of Current Spanish]. Retrieved from http://ww w.rae.es/
Rex, L.A., Thomas, E.E., & Engel, S. (2010). Applying Toulmin:
Teaching logical reasoning and argumentative writing. English
Journal, 99(6), 56– 62.
Reznitskaya, A., Anderson, R.C., & Kuo, L.J. (2007). Teaching and
learning argumentation. The Elementary School Journal, 107(5),
449–472. https://doi.org/10.1086/518623
Ripoll, J.C., & Aguado, G. (2014). Reading comprehension improve-
ment for Spanish students: A meta- analysis/La mejora de la com-
prensión lectora en español: Un meta- análisis. Revista de
Psicodidáctica, 19(1), 27–44. https://doi.org/10.1387/RevPsicodid
act.9 001
Sánchez, E., & García, J. (2009). The relat ion of knowledge of textua l
integration devices to expository text comprehension under dif-
ferent assessment conditions. Reading and Writing, 22(9), 1081–
1108 . https://doi.org/10.1007/s11145-00 8-9145 -7
Sánchez, E., García, J.R., & Bustos, A. (2017). Does rhetorical com-
petence moderate the effect of rhetorical devices on the compre-
hension of expository texts beyond genera l comprehension skills?
Reading and Writing, 30(3), 439–462. https://doi.org/10.1007/
s11145-016-968 4-2
Sánchez, E., González , A., & García, J. (2002). Competencia retórica:
Una propuesta para interpretar las dificultades de comprensión
[Rhetoric competence: A proposal to interpret comprehension
difficulties]. Psicothema, 14(1), 77–85.
Schleppegrell, M.J. (2001). Linguistic features of the language of
schooling. Linguistics and Education, 12(4), 431–459. https://doi.
org/10.1016/S0898-5898(01)00073-0
Schleppegrell, M.J. (2004). The language of schooling: A functional
linguistic perspective. London, UK: Erlbaum.
Shanahan, T., & Shanahan, C. (2008). Teaching disciplinary literacy
to adolescents: Rethinking content- area literacy. Harvard
Educational Review, 78(1), 40–59. https://doi.org/10.17763/
haer.78.1.v6244 4321p602101
Snow, C. (1990). The development of definitional skill. Journal
of Child Language, 17(3), 697–710. https://doi.org/10.1017/
S0305000900010953
Snow, C., & Uccelli, P. (2009). The challenge of academic language.
In D.R. Olson & N. Torrance (Eds.), The Cambridge handbook of
literacy (pp. 112–133). New York, NY: Cambridge University
Press.
Soto, G., & Zenteno, C. (2004). Los sintagmas nominales en textos
científicos escritos en español [Noun phrases in scientific texts
written in Spanish]. Estudios de Lingüística Universidad de
Alicante, 18, 275–292. https://doi.org/10.14198/ELUA2004.18.14
Stahl, S.A., & Nagy, W.E. (2006). Teaching word meanings. Mahwah,
NJ: Erlbaum.
Strasser, K., & del Río, F. (2014). The role of comprehension moni-
toring, theory of mind, and vocabulary depth in predicting story
comprehension and recall of kindergarten children. Reading
Research Quarterly, 49(2), 169–187. https://doi.org/10.1002/rrq.68
Strasser, K., del Río, F., & Larraín, A. (2013). Profundidad y ampli-
tud del vocabulario: ¿Cuál es su rol en la comprensión de historias
en la edad pre- escolar?/Vocabulary depth and breadth: Their role
in preschoolers’ story comprehension. Estudios de Psicología,
34(2), 221–225. htt ps://doi.org /10.1174/021093913806751401
Strasser, K., Larraín, A., & Lissi, M.R. (2013). Effects of storybook
reading style on comprehension: The role of word elaboration
and coherence questions. Early Education and Development,
24(5), 616–639. https://doi.org/10.1080/10409289.2012.715570
Thorne, C., Morla, K., Uccelli, P., Nakano, T., Mauchi, B., Landeo,
L., … Huerta, R. (2013). Efecto de una plata forma virtua l en com-
prensión de lectura y vocabulario: Una alternativa para mejorar
las capacidades lectoras en primaria [Effect of a virtual platform
on reading comprehension and vocabulary: An alternative to
improve reading skills in primary school]. Revista de Psicología,
31(1), 3–35.
Treviño, E., Toledo, G., & Gempp, R. (2013). Calidad de la educación
parvularia: Las prácticas de clase y el camino a la mejora
[Preschool education quality: Teacher practices and the path to
improvement]. Pensamiento Educativo: Revista de Investigación
Educacional Latinoamericana, 50(1), 40–62. https://doi.org/10.
7764/PEL.50.1.2013.4
Uccelli, P., Barr, C., Dobbs, C., Phillips Galloway, E., Meneses, A., &
Sánchez, E. (2015). Core academic language skills: An expanded
operational construct and a novel instrument to chart school-
relevant language proficiency in preadolescent and adolescent
learners. Applied Psycholinguistics, 36(5), 1077–1109. https://doi.
org/10.1017/s014271641400006x
Uccelli, P., Demir, Ö.E., Rowe, M., Goldin-Meadow, S., & Levine, S.
(2017). Children’s early decontextualized talk predicts academic
language proficiency in mid-adolescence. Manuscript submitted
for publication.
Uccelli, P., & Meneses, A . (2015). Habilidades de lenguaje académico
y su asociación con la comprensión de lectura en la escuela pri-
maria y media: Un nuevo constructo operacional [Academic lan-
guage skills and their association with reading comprehension in
primary and middle school: A new operational construct].
Miríada Hispánica, 10, 179–206.
Uccelli, P., Phillips Galloway, E., Barr, C., Meneses, A., & Dobbs, C.
(2015). Beyond vocabulary: Exploring cross- disciplinary
academic- language proficiency and its association with reading
comprehension. Reading Research Quarterly, 50(3), 337–356 .
https://doi.org/10.1002/rrq.104
Valenzuela, J.P., Bellei, C., & de los Ríos, D. (2013). Socioeconomic
school segregation in a market- oriented educational system: The
case of Chile. Journal of Education Policy, 19(2), 217–241. htt ps://
doi.org/10.1080/02680939.2013.806995
Venegas, R. (2010). Multi-d imensional analysis of an ac ademic corpus
in Spanish. In G. Parodi (Ed.), Academic and professional discourse
genres in Spanish (pp. 101–119). Philadelphia, PA: John Benjamins.
Wilson, M.R. (2005). Constructing measures: An item response the-
ory approach. Mahwah, NJ: Erlbaum.
22 | Reading Research Quarterly, 0(0)
Zeno, S.M., Ivens, S.H., Millard, R.T., & Duvvuri, R. (1995). The
educator’s word frequency guide. Brewster, NY: Touchstone.
Ziegler, J., Bertrand, D., Tóth, D., Csépe, V., Reis, A., Faísca, L., …
Biomert, L . (2010). Orthog raphic depth and its impact on universal
predictors of re ading: A cross- la nguage invest igation. Psychological
Science, 21(4), 551–559. https://doi.org/10.1177/0956797610363406
Ziegler, J., & Goswami, U. (2005). Reading acquisition, developmen-
tal dyslexia, and skilled reading across languages: A psycholin-
guistic grain size theory. Psychological Bulletin, 131(1), 3–29.
https://doi.org/10.1037/0033-2909.131.1.3
Submitted August 23, 2016
Final revision received March 16, 2017
Accepted April 6, 2017
ALEJANDRA MENESES (corresponding author) is an
associate professor at the Facultad de Educación at the
Pontificia Universidad Católica de Chile, Santiago; e-mail
amenesea@uc.cl.
PAOLA UCCELLI is an associate professor of education at the
Harvard Graduate School of Education, Cambridge,
Massachusetts, USA; e-mail paola_uccelli@gse.harvard.edu.
MARÍA VERÓNICA SANTELICES is an associate professor at
the Facultad de Educación at the Pontificia Universidad
Católica de Chile, Santiago; e-mail vsanteli@uc.cl.
MARCELA RUIZ is an assistant professor at the Facultad de
Filosofía y Humanidades at Universidad Alberto Hurtado,
Santiago, Chile; e-mail maruiz@uahurtado.cl.
DANIELA ACEVEDO is an instructor at the Facultad de
Filosofía y Humanidades at Universidad Alberto Hurtado,
Santiago, Chile; e-mail dacevedo@uahurtado.cl.
JAVIERA FIGUEROA is a doctoral candidate at the Facultad
de Educación at the Pontificia Universidad Católica de Chile,
Santiago; e-mail jfigueroami@uc.cl.
APPENDIX A
S-CALS-I
Tas k Skill measured Description of item Sample item
Packing and
unpacking
nominalizations
Morphologically
composing and
decomposing
nominalizations
The administrator reads
items in which students
are asked to decompose
a given nominalization
into a verb or
adjective or derive a
nominalization from a
given verb or adjective.
1. Refutación. El científico logró ___ la hipótesis. (Refutation. The
scientist made ___ the hypothesis.)
2. Obtener. La ___ de energía es importante para vivir. (Get. The
___ of energy is important to live.)
Organizing
compact
and complex
sentences
Organizing
compact
and complex
sentences
Students read a
disordered set of
words and are asked to
organize the sentence
(cues are provided for
first and last words).
1. formación un proceso La rocas de es lento. (formation a process
The rocks of is slow.)
Connecting ideas
logically
Understanding
school- relevant
connectives
Students are asked to
select the appropriate
connective to link two
sentences, or to select
the best continuation for
a fragment that includes
a connective.
1. Una araña tiene ocho patas; ___ una mariposa posee seis patas.
(A spider has eight legs; ___ a butterfly has six legs.)
a. de tal modo que (so that)
b. entonces (then)
c. en cambio (instead)
d. por consiguiente (therefore)
2. En el paseo de curso llovió todo el dia. No obstante, (On the
course ride it rained all day. However,)
a. fue una experiencia divertida para todos los estudiantes. (it
was a fun experience for all students.)
b. fue un día con gran cantidad de truenos y relámpagos. (it was
a day with a lot of thunder and lightning.)
c. fue un paseo aburrido para muchos de los estudiantes. (it was
a boring ride for many of the students.)
(continued)
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 23
Tas k Skill measured Description of item Sample item
Tracking
participants and
themes
Resolving
conceptual
anaphora and
other types
of anaphora
prevalent in
school texts
Students are asked to
select among three
options the correct
antecedent for the
underlined anaphor.
Nuestro planeta está hecho de muchas capas de roca. La capa
más profunda está tan caliente que una parte de la roca se
derrite. Los volcanes se forman cuando la roca derretida rompe
las capas superiores para salir a la superficie. Este fenómeno ha
sido ampliamente estudiado. (Our planet is made of many layers
of rock. The deepest layer is so hot that part of the rock melts.
Volcanoes are formed when molten rock breaks through the upper
layers to surface. This phenomenon has been widely studied.)
En este pasaje, “este fenómeno” se refiere a: (In this passage,
“this phenomenon” refers to:)
a. nuestro planeta está hecho de muchas capas de roca. (our planet
is made up of many layers of rock.)
b. la roca derretida rompe las capas superiores. (the melted rock
breaks the upper layers.)
c. la capa más profunda está tan caliente que la roca se derrite.
(the deepest layer is so hot that the rock melts.)
Interpreting
writers’
viewpoints
Understanding
epistemic
markers that
signal the
writer’s degree
of certainty
toward an
advanced claim
After presenting
a scenario, the
administrator reads
a series of scientists’
statements that include
epistemic markers.
On the basis of each
statement, students are
asked to decide how
certain the scientist is
about his or her claim.
¿Cree este científico que la roca viene del espacio? (Does this
scientist believe that rock comes from space?)
Sí (Yes) Quizás sí
(Maybe yes) Quizás no
(Maybe no) No
(No)
Es indudable
que la roca
viene del
espacio. (There
is no doubt
that the rock
comes from
space.)
Understanding
metalinguistic
terms
Understanding
school- relevant
metalinguistic
terms that refer
to cognitive
or discourse
processes
Students are asked
to select among four
options the appropriate
metalinguistic word for
an academic discussion.
Discussion
El recreo es el momento perfecto para hacer amigos. (Recess is
the perfect time to make friends.)
Item
Susana dice: “Un estudio reportó que el 92% de los estudiantes
hizo nuevos amigos durante el recreo.” La respuesta de Susana
presenta… (Susana says, “One study reported that 92% of the
students made new friends during recess.” Susana’s answer
presents…)
a. una hipótesis (a hypothesis)
b. una opinión (an opinion)
c. una definición (a definition)
d. una evidencia (an evidence)
Organizing
analytic texts
Organizing an
argumentative
text
Students are asked to
order fragments of an
argumentative text
presented in random
order. Discourse markers
(e.g., por último [by
last], en conclusión [in
conclusion], finalmente
[finally], por una parte
[on one side], por otra
parte [on the other
side]) and conventional
argumentative structure
were used.
Mejores almuerzos en el colegio (Better Lunches at School)
___ En segundo lugar, la comida del colegio debe variar porque a
los estudiantes les gusta elegir su comida. (Second, school food
should vary because students like to choose their food.)
___ En conclusión, los almuerzos del colegio serían mejores
si hubiera comida más sana, variada y en mayor cantidad. (In
conclusion, school lunches would be better if there were healthier,
more varied food and more.)
___ En primer lugar, necesitan comida más saludable para estar
sanos y así estudiar mejor. (First, they need healthier food to stay
healthy and so study better.)
___ Algunos piensan que los estudiantes deberían recibir mejores
almuerzos en el colegio. (Some think that students should receive
better lunches at school.)
___ Por último, muchos estudiantes necesitan más comida porque
están creciendo. (Finally, many students need more food because
they are growing.)
(continued)
S-CALS-I (continued)
24 | Reading Research Quarterly, 0(0)
APPENDIX B
Psychometric Information of S- CALS- I
and S- AVoc- T and Pairwise Correlations
S-AVOC-T
Tas k Skill measured Description of item Sample item
Understanding academic
vocabulary
Understanding school- relevant
cross- disciplinary academic
vocabulary
Students are asked to select,
among four options, the
correct synonym for an
underlined academic word
used in a sentence context.
Los estudiantes interpretaron
los resultados de manera
diferente. (The students
interpreted the results
differently.)
a. entendieron (understood)
b. incorporaron (incorporated)
c. señalaron (pointed)
d. crearon (created)
Psychometric Information
Reliability evidence for S- CALS- I was robust with a
Cronbach’s alpha index of .88. To select the items, item
difficulty was estimated (between −2.88 and 1.94). Of
the 53 items, 48 exhibited infit statistics within the
0.75and 1.33 range suggested by Wilson (2005). The
remaining five items showed infit statistics only slightly
below or above the cutoff score, not representing a rele-
vant deviation from the expected item functioning.
Additionally, biserial correlations (between .17 and .65)
and proportion correct (between .24 and .85) were cal-
culated to provide adequate S- CALS information from
grades 4–8.
For S- AVoc- T, items included in the final version ex-
hibited omission rates of between 3% and 8.1%, biserial
correlations between .3 and .58, item difficulty between
−1.53 and 1.32, and proportion correct between .32 and
.82. All 15 items exhibited infit statistics within the 0.75
to 1.33 range suggested by Wilson (2005).
Tas k Skill measured Description of item Sample item
Identifying
academic
register
Identifying
academic
definitions
Students are asked to
select the most academic
definition among a set of
three options.
Paraguas (Umbrella)
a. Un paraguas es un objeto que sirve de protección para la lluvia,
fabricado con tela impermeable. (An umbrella is an object that
serves as protection for rain, made with waterproof fabric.)
b. Un paraguas es algo que te tapa de la lluvia con una tela que se
pega sobre unos alambres doblados. (An umbrella is something
that gets you out of the rain with a cloth that sticks over bent
wires.)
c. Un paraguas es para que uno no se moje con la lluvia y está
hecho de tela o plástico que te protege. (An umbrella is so that
one does not get wet with rain and is made of cloth or plastic
that protects you.)
S-CALS-I (continued)
Academic Language as a Predictor of Reading Comprehension in Monolingual Spanish-Speaking Readers | 25
Pairwise Correlations for S- CALS- I Tasks and S- AVoc- T
Tas k 1 2 3 4 5 6 7 8 9
1. Packing and unpacking nominalizations
—
2. Organizing compact and complex sentences .69***
—
3. Connecting ideas logically .67*** .60***
—
4. Tracking participants and themes .58*** .57*** .60***
—
5. Interpreting writers’ viewpoints .49*** .44*** .52*** .47***
—
6. Understanding metalinguistic terms .61*** .56*** .63*** .53*** .57***
—
7. Organizing analytic texts .62*** .57*** .64*** .59*** .48*** .58***
—
8. Identifying academic register .50*** .48*** .53*** .47*** .45*** .49*** .53***
—
9. Understanding academic vocabulary .71*** .64*** .68*** .66*** .54*** .66*** .67*** .54***
—
*** p < .001.
