Academic language is a kind of social language for the purpose of schooling. It is central to disciplinary learning, thinking, and communication. This study examined adolescents’ use of academic language in informational writing, a genre highly valued in school, workplace, and society. Ninety-three seventh and ninth grade students from a U.S. public school were asked to write a science report based on a “wordless” picture book about a familiar class of animals called crocodylia. The student writing corpus was coded for presence of a constellation of academic language features. Statistical analyses of these data showed that (a) the adolescents made limited use of academic language features in their writing, (b) there were no significant differences between the two grade levels in academic language use, (c) there was a significant relationship between reading ability and academic language use, and (d) academic language use was a significant predictor of writing quality. These findings highlight both the importance of and the need for more explicit attention to academic language in secondary literacy instruction.
Academic language is, broadly speaking, a kind of social language for the purpose of doing school work. It is the language through which students build content knowledge, develop advanced literacy, acquire disciplinary habits of mind, and are assessed in their learning. This language has lexical, grammatical, and discursive features that are distinct from those that characterize the everyday language students use to socialize with peer friends and family members. Because of this, academic language tends to be less familiar and more challenging to students, who are expected to learn content along with the language through which this content is presented. This article reports on a study that examined adolescents’ use of academic language in informational writing, a type of nonfiction whose primary purpose is to classify, describe, explain, analyze, or argue (Derewianka, 1990). Informational writing is a genre highly valued in school, workplace, and society. As such, it is emphasized in both state and national standards. The U.S. Common Core State Standards (www.corestandards.org), for example, recommends that students be given a steadily increasing dose of informational text in their literacy diets over their K-12 experience. They are expected to not only interact with an increasingly higher percentage of informational text (50% for 4th graders, 55% for 8th graders, and 70% for 12 graders) but also develop proficiency in reading and writing these texts. Like any genre, informational text is constructed with lexical, grammatical, and discursive resources that are functional for making it the type of text it is. Understanding what these academic language resources are and how they are used to instantiate the genre is key to developing proficiency in informational writing. This study describes the linguistic resources that adolescents use in constructing and communicating their understanding about the natural world. Before describing the study, we review relevant research literature on the conception of academic language and discuss the role of academic language in literacy development and disciplinary learning in K-12 contexts.
Defining and describing academic language
Academic language is commonly understood to be a set of registers for the purpose of teaching, learning, and assessment in school. It has certain features that do not regularly occur in the discourses used in everyday spontaneous social interactions among close friends and family members. Three lines of scholarship have contributed to the conception of academic language and the identification of its features. The first line of scholarship is the sociolinguistic work of the 1970s and 1980s (e.g., Bernstein, 1971/2003; Hasan, 1989/2009; Heath, 1983). This line of research demonstrates that people from different social strata are oriented towards different ways of saying and meaning and that this difference in coding orientation impacts children’s academic performance in school. More specifically, middle class children come to school with a set of linguistic skills (e.g., labeling and naming features, providing narratives on items out of immediate physical context) and literacy experiences (e.g., extensive exposure to stories and imaginative talks) that teachers value and are reinforced in school but that working class children have yet to acquire. This semantic variation prepares children differentially for school success.
The distinction between school-based, or academic, language and home-based, or everyday, language was brought to the forefront of educators’ conscious attention in the second line of scholarship. The work of Jim Cummins, a professor of bilingual education in Canada, deserves special mention here. In discussing assessment issues involving bilingual learners, Cummins (1984) proposed a distinction between what he called “Basic Interpersonal Communication Skills” (BICS) and “Cognitive Academic Language Proficiency” (CALP), with the former referring to the language used for everyday spontaneous social interactions and the latter to the language needed for school learning. He noted that because the time needed for developing CALP, or academic language, is significantly longer (5–7 years) than that for developing BICS, or everyday language (6–24 months), educators should be careful not to conflate the two registers when designing language assessment tasks.
Building on Cummins’ work, Bailey and Heritage (2008) divided academic language into School Navigational Language (SNL) and Curriculum Content Language (CCL). SNL refers to the language through which students communicate with peers and teachers in school, and CCL refers to the language used in the process of teaching and learning content materials. The two varieties differ in their purpose, degree of formality, context of use, teacher expectations, and grade level expectations. A similar distinction was made by Bunch (2014), who introduced the notions of “language of display” and “language of ideas” to show that there is more than one style of language use in the classroom. Language of display refers to “the evolving oral and written texts students develop, either individually or as a group, to present to particular academic audiences”, and language of ideas consists of “the use of any and all linguistic resources students bring to bear on the engagement in and completion of an academic task” (p. 74). The distinction, while not perfect, serves to highlight the fact that children use language in different ways for different audiences and purposes as they engage productively with different academic tasks (e.g., discussing a novel, writing a lab report) in classroom settings. Nagy and Townsend (2012) offered a more focused definition, referring to academic language as “the specialized language, both oral and written, of academic settings that facilitates communication and thinking about disciplinary content” (p. 92). They noted that academic language is used to convey educational and disciplinary knowledge, value and reasoning in ways not commonly found in students’ everyday lifeworld. Academic language can be discipline-specific or cross disciplinary, with the former denoting the language that encapsulates core concepts and key relationships unique to specific disciplines and the latter to the language prevalent across disciplines.
Although a general distinction is now widely accepted between the language that is functional for doing school work (i.e., academic language) and the language used primarily for daily social interactions (i.e., everyday language), the boundary between the two registers appears to be less clear to many. In fact, some even questioned the existence of academic language as a legitimate linguistic entity. O’Connor and Michaels (2015), for example, asserted that there is “no definitive” structural or functional boundary between academic language and everyday language (p. 304). Other scholars have been able to describe the features of academic language, however. This strand of research suggests that although the line between academic language and every language can be blurry at times, there is a constellation of lexical, grammatical, and discursive features that contributes to a syndrome commonly recognized as academic language. For example, the landmark study by Biber and Gray (2016) revealed that academic language is characterized by an extensive use of embedded phrasal structures (e.g., expanded noun phrases with pre- and post-modifiers), whereas everyday language relies heavily on clausal features (e.g., finite dependent clauses). According to Bailey (2007), the key difference between academic language and everyday language resides not so much in their linguistic complexity or cognitive demands, but in the co-patterning of certain lexical, grammatical, and discursive features. Lexically, academic language makes greater use of general academic vocabulary and specialized terminology. Grammatically, passive voice constructions, relative clauses, conditional clauses, and long noun phrases are more often associated with academic language. Discursively, academic language involves the use of school-valued organizational patterns and participation structures associated with genre/discipline-specific practices (e.g., writing a report about a science experiment) and general academic discourse practices (e.g., making evidence-based arguments). While many of these linguistic features can indeed be found in everyday language, they tend to co-occur with greater frequency and higher concentration in academic discourses.
Drawing on the insights from corpus studies and discourse analytic studies of pedagogical texts, Schleppegrell (2001) identified a set of linguistic features that characterizes the language used in academic, particularly reading and writing, tasks. Specifically, she showed that in contrast to everyday spoken discourse, school-based texts tend to make more concentrated use of specific, technical vocabulary (e.g., aldehydes); dense noun phrases with pre- and post-modifiers (e.g., this effect of Earth’s rotation on the direction of winds and currents); declarative sentences with embedded clauses (e.g., Aldehydes are organic compounds in which the carbonyl group is attached to a carbon atom at the end of a carbon-atom chain.); logical links indicated through the use of nouns, verbs, prepositions, clauses, and conjunctions (e.g., Withno jaguars to eat them, the number of anteaters might increase. Alarmed by the fire and the destruction it caused, people in Ohio began a massive campaign to clean up the Cuyahoga.); and a tightly knit texture indicated through nominal, verbal and adverbial expressions (e.g., During combustion, the carbon and hydrogen combine with oxygen in the air to form carbon dioxide and water.This processreleases energy in the forms of heat and light.). These features help realize “contexts of information display, authoritativeness, and high degrees of structure” (Schleppegrell, 2004, p. 74) and are functional for creating the sort of texts that are valued in school and that students are expected to read and write for the purpose of schooling.
Academic language features are not equally distributed across all texts, however. Instead, they configure, in particular ways, with other lexicogrammatical (or semiotic) resources to meet the needs of particular disciplines, purposes, tasks, and audiences. This means that academic language is not a unitary or stable construct. It manifests differently across different contexts, as each task, genre, discipline, or grade level has its own expected ways of using language. Biber and Gray (2016) found that there is considerable internal lexicogrammatical variation across disciplines and registers, with scientific writing, for example, exhibiting extreme reliance on phrasal structure and writing in the humanities relying more heavily on clausal modification. As Hyland (2004) explained, texts within the same academic discipline tend to display similar “preferences for particular argument forms, lexical choices and discourse structures” because they adopt similar approaches and address “similar questions about a similarly conceived external world” (p. 177). Schleppegrell (2004) showed that different school-based genres—personal (e.g., recount, narrative), factual (e.g., procedural, report), and analytical (e.g., account, explanation, exposition)—marshal different constellations of lexical and grammatical resources and that the demand for more academic register features increases as students move from personal to factual and analytical genres. From a historical perspective, academic language has also undergone a dramatic shift over the past 300 years from the elaborated clausal style (i.e., sentence consisting of multiple clauses) to the compressed phrasal style (i.e., sentence consisting of long nominal phrases linked by a verb) (Biber & Gray, 2016). The variation in the way language is used across academic settings, disciplines, tasks, and contexts has led Fang (2016) to conceive of academic language as the braiding of three language strands—everyday language, which construes commonsense knowledge; written language, which construes educational forms of knowledge; and metaphoric language, which construes disciplinary knowledge—with certain strand(s) more heavily present in some texts than in others. In this sense, then, academic language is perhaps better referred to as “academic languages” or “academic registers”, for there are as many forms and functions of academic language as there are academic texts, oral or written.
Academic language, literacy development, and disciplinary learning
Academic language is the language in which school subjects are codified, transmitted, evaluated, and renovated. It is the language used to accomplish increasingly challenging discursive tasks in school teaching, learning, and assessment. As such, academic language is widely recognized as crucial to literacy development, disciplinary learning, and ultimately academic success in the traditional institutions of schooling. Heller and Morek (2015) suggested that academic language plays three roles in schooling: communicative, epistemic, and socio-symbolic. First, academic language is a medium of knowledge transmission. It is an indispensible resource for construing, presenting, and communicating educational and disciplinary knowledge in academic contexts. As students advance through grade levels, they are expected to engage with progressively more complex and specialized knowledge. This engagement calls for new ways of using language. Specifically, students must develop, in the words of Christie (2002), “the capacity to deploy language in ways that abstract away from immediate, lived experience, to build instead truths, abstractions, generalizations, and arguments about areas of life of various kinds” (p. 66). Second, academic language is a tool for thinking and intellectual development. It promotes systematic thinking and scientific reasoning. It enables execution of higher order cognitive operations such as abstraction, generalization, interpretation, and argument, allowing students to reconstrue their experience with the social, natural, and imagined worlds. Third, academic language is a ticket and a visiting card. As a school-privileged form of discourse, academic language represents cultural capital, acting as a gate-keeper to educational and disciplinary knowledge and ideology. For this reason, academic language has been critiqued as a “discourse of appropriateness” that privileges, reinforces, legitimates, and naturalizes dominant White perspectives on the standard linguistic and cultural practices of racialized/minoritized communities (Flores & Rosa, 2015). Control over academic language gives students power and agency in literacy development, knowledge building, identity construction, social positioning, and career advancement. Without access to academic language, students are at a distinct disadvantage in school learning, social positioning, and career development, further exacerbating existing educational inequality and social injustice. The literacy and learning problems experienced by many students in school have been attributed in large part to their struggle with academic language (Schleppegrell, 2004; Snow & Uccelli, 2009). A lack of familiarity with and proficiency in academic language limits access to content, impedes thinking and reasoning, handicaps learning and communication, undermines a sense of self-efficacy and personal empowerment, reduces opportunities for effective classroom participation, and increases the risk for academic failure and school drop out. As Francis, Rivera, Lesaux, Kieffer, and Rivera (2006) observed, “Mastery of academic language is arguably the single most important determinant of academic success for individual students” (p. 7).
Given the importance of academic language to disciplinary learning and school success, it is not surprising that it has received considerable attention among educational researchers. Much of this research, however, focuses on academic vocabulary, especially as it relates to the literacy and content learning outcomes of students for whom English is not the primary language (see DiCerbo, Anstrom, Baker, & Rivera, 2014; Nagy & Townsend, 2012 for partial reviews of this body of work). More recent studies have shown that richly explicit instruction in academic vocabulary produces gains in both word knowledge and text comprehension. For example, Lesaux, Kieffer, Kelley, and Harris (2014) conducted a randomized field trial to test an academic vocabulary intervention designed to improve the language and literacy skills of linguistically diverse sixth-grade students in 14 urban middle schools. The researchers reported that the 20-week classroom-based intervention improved students’ vocabulary knowledge, morphological awareness skills, comprehension of expository texts, and performance on a standardized measure of written language skills. They further noted that the effects were generally larger for students whose primary home language is not English and for those students who began the intervention with underdeveloped vocabulary knowledge. Similarly results were obtained by Vadsy, Sanders and Herrera (2015), who conducted a multi-cohort cluster randomized trial to examine the effects of classroom instruction of 140 Tier Two vocabulary words on reading comprehension of 1232 fourth and fifth grade students in 24 schools.
Moving beyond vocabulary, Uccelli, Galloway, Barr, Meneses, and Dobbs (2015) explored the relationship between cross-disciplinary academic language proficiency and reading comprehension among a linguistically and socioeconmically diverse cross-sectional sample of 218 students in grades 4–6. Students’ cross-disciplinary academic language proficiency was assessed using an empirically validated instrument that measures 8 core academic language skills—connecting ideas, tracking themes, organizing texts, breaking words, comprehending sentences, identifying definitions, interpreting epistemic stance markers, and understanding metalinguistic vocabulary. The researchers found that academic language proficiency is a significant predictor of students’ reading comprehension, above and beyond the contributions of academic vocabulary knowledge, word reading fluency, and sociodemographic characteristics. They also reported that in general, students in lower grade levels, from lower SES environments, or with an English Learner (EL) designation scored significantly lower than students in higher grade levels, from higher SES environments, or without an EL designation, respectively. Similar findings were replicated with monolingual Spanish-speaking students (e.g., Meneses et al., 2017).
Focusing on writing, Galloway and Ucceli (2015) examined the relationship between lexicogrammatical skills for concisely and precisely defining a familiar concept and discourse-organizational skills for structuring a cohesive paragraph, two facets of productive academic language skills that were hypothesized to support expository writing in grades 4 through 8. The researchers reported that older writers with more years of schooling produced written responses that displayed a higher degree of lexical precision and morpho-syntactic complexity, as well as greater skill in the production of academic micro-genres (concept definition, paragraph structuring), than did younger writers. Moreover, the magnitude of difference in lexicogrammatical skill between students in grades 4–5 and those in grades 6–8 was greater than that in micro-genre discourse organization skill between the two groups.
Employing a more extended writing task that was completed within 25 min, Uccelli, Dobbs, and Scott (2012) examined lexical and grammatical features of 51 persuasive essays written by high school students in response to a prompt similar to those found in college admission tests. They found that the frequency of organizational markers and epistemic hedges significantly predicted the writing quality of the persuasive essays, above and beyond the contribution of length, lexical density, lexical diversity, and syntactic complexity. Similarly, Figueroa, Meneses, and Chandia (2018) explored the relationship between academic language skills and the quality of written arguments and explanations composed by 8th grade students in Chile. The writing samples, completed in two 30-min sessions, were elicited using a fictional situation that described the use of tablets in the classroom. They reported that the students’ academic language proficiency, as measured by a Spanish version of Uccelli et al.’s (2015) Core Academic Language Skills Instrument, was a significant predictor of the quality of their writing in both genres, explaining 27% of the variance in arguments and 34% of the variance in explanations.
To summarize, this brief review of K-12 research literature suggests that academic language is a significant factor that positively impacts students’ reading/writing performance and their academic achievement. However, the bulk of the research focuses on academic vocabulary and its impact on reading comprehension, with particular attention to English Language Learners (ELL). Much less attention has been paid to other features of academic language and to the role of academic language in non-ELL, adolescent students’ writing development. As Snow and Uccelli (2009, p. 113) pointed out,
Ironically, although academic language skills are widely cited as the obstacle to achievement for struggling readers in general, much of the empirical research on academic language has been done by those studying English Language Learners (ELLs). In other words, learning ‘academic English’ is recognized as a challenging task for second-language speakers of English, but the challenges faced by native speakers in learning the rules, the structures, and the content of academic English have received much less attention.
The present study builds on and extends prior research to address these gaps by examining seventh-grade (early adolescence) and ninth-grade (mid-adolescence) students’ use of academic language in science report, a type of informational writing that is highly prized in the schooling contexts and whose primary purpose is to classify, describe, and attribute. It is guided by four research questions: (1) What features of academic language are present in the adolescents’ informational writing? (2) Are there differences between grade levels in the way these features are used? (3) Is proficiency in academic language use associated with students’ reading ability? and (4) Does academic language use predict the quality of students’ writing? Answers to these questions will help us uncover the linguistic resources adolescents deploy in constructing an important genre of schooling, gauge adolescents’ academic language proficiency, better understand how academic language develops among adolescents, gain insight into the relationship of academic language proficiency to reading ability and writing quality, and derive important implications for literacy instruction across the disciplines.
The study was conducted in a K-12 public school in a southeastern state of the United States. Participants for the study were 93 students from two intact classes each in seventh and ninth grades. The sociodemographic information about the participants (ages 12–15) is presented in Table 1.
During one of their language arts class periods, participants were asked to write an informational text in response to the following directions read aloud by their teacher:
Boys and girls, today we are going to write a science report about one of our state’s most unique animals, alligators. Before you get started, I would like you to pretend that you were a scientist and nonfiction book author. Now, imagine that you have been studying alligators at Lake Alice [a local swamp] for months and months. During that time you have taken photos of alligators and written notes about them. Pretend you have taken those pictures and arranged them onto the pages of this “book”. Take a few minutes to look through the pictures in this book. Now, your job is to report what you learned about alligators during these months of observation in the swamp. Based on the photos you’ve taken, write as much as you can about alligators from a scientist’s view, so that other students in your grade who have not seen and don’t know anything about alligators can get to know them. You have as much time as you need. You do not need to worry about spelling.
The “book” referenced in the prompt is a “wordless” picture book on crocodylia by Blake (1996), where all the words—including title, author, table of contents, captions, and index—had been removed for the purpose of this task. The original version of the book contains both text and photos, presenting rich factual information about crocodylia with respect to their origin, physical characteristics, living habitat, hunting habits, reproduction, and protection. The 24-page book was chosen for two main reasons. First, it presents a topic that was of sufficient familiarity and interest to the students. As residents in a city (and state) where alligators (and, to a lesser extent, crocodiles) are in abundance, participants had seen, read about, and discussed the reptile in both their in-school and out-of-school experiences. In other words, they knew quite a bit about the animal through both formal schooling and informal life experiences. Second, the book has many format-based features typical of an informational text: It has a table of content and an index; every page contains at least one colorful photo depicting an aspect of crocodylia; these photos do not have human characters in them and are neither temporally sequenced nor goal oriented. Thus, as a stimulus for the writing task, the book was likely to elicit an informational text (factual writing) rather than a story (personal writing).
The writing task resembled other writing assignments students had been doing in school, where teachers regularly gave prompts—some akin to real-life situations and others more contrived—in order to prepare students for high stakes writing assessments in the state. The task was administered within 2 weeks toward the end of the Fall semester. Students were provided sufficient pages of blank lined paper for writing. The task direction was read in a way similar to what would normally be done in a regular class prompt writing session. Besides reading the instruction, the teacher gave no further hints about the task. All students completed their writing within 45 to 60 min.
Data analysis proceeded in three phases. In phase one, each student’s informational writing was coded for evidence of academic language features (ALFs). For the purpose of this study, academic language was conceptualized broadly as the language used in the schooling contexts, where students are usually expected to “display knowledge authoritatively in highly structured texts” in writing tasks (Schleppegrell, 2004, p. 74). Drawing on the work of Biber and Gray (2016), Schleppegrell (2001, 2004), Christie and Derewianka (2008), and others (e.g., Fang, 2006, 2012; Perera, 1984; Ravid & Berman, 2010), we identified, in Table 2, eleven (11) lexicogrammatical features as characteristic of academic language.
Two researchers who were knowledgeable about academic language and trained in systemic functional linguistics each analyzed the texts independently and then cross-checked with each other. The agreement rate was close to 90%, with disparities resulting mainly from either coder overlooking certain features. Disagreements were resolved through discussion in light of each textual context and relevant prior research studies. Specifically, the following steps were observed in analyzing each text: (1) divide text into non-embedded clauses, (2) identify academic language features in each clause, (3) count the number of occurrences for each academic language feature in the text, and (4) calculate the total ALF score by adding up the number of occurrences for all 11 academic language features.
In phase two, each student’s text was rated holistically for its writing quality. A six-traits rubric, widely adopted in both classroom and high stakes writing assessments, was used in the rating. Specifically, each text was rated on a scale of 1–6 in each of the following 6 dimensions of writing—ideas/content, organization, conventions, voice, sentence fluency, word choice. This means that each text could receive a maximum score of 36 and a minimum score of 6. The rating was done by two former classroom teachers—one elementary school literacy teacher and the other secondary school science teacher—who were neither involved in nor familiar with the above linguistic analysis. Inter-rater reliability, measured by intraclass correlation coefficient, is 0.69, which is moderate and considered good enough (Cicchetti, 1994). The two raters’ scores were averaged to yield a holistic score indicative of the overall quality of writing.
In phase three, statistical analyses were conducted on the numerical data generated from phases one and two in order to answer the four research questions for the study. Because descriptive statistics revealed that the ALF data (frequency counts) were positively skewed, they were square root transformed for all parametric statistical tests (ANOVA, Pearson Product-Moment, Regression). Non-parametric tests (Kruskal–Wallis, Welch, Mann–Whitney) were also performed with the raw (untransformed) data to ensure robust evidence for interpretation. For all significance tests, the alpha level was set at 0.05.
To examine what features of academic language were used by the students, we generated descriptive statistics (i.e., means and standard deviations) for the 11 academic language features identified earlier, both individually and as a whole, by grade level (see Table 3). It is clear from the table that all 11 features of academic language were present in the students’ writing, with some features used much more frequently than were other features. Specifically, general academic vocabulary (e.g., characteristics, intriguing), expanded noun phrases (e.g., these fierce animals with cat-like reflexes; very powerful jaws that can bite a human’s arm off), specialized terminology (e.g., hides, reflexes), and epistemic hedges (e.g., Alligatorsusuallylive in freshwater; Alligatorstend tobe thicker and round.) were used with more frequency than were features such as nominalization (e.g., difference, protection) and passive voice (e.g., …alligators are raised to be killed for food and their skin). Other features were used sparingly, including non-finite clauses (e.g., Using its massive jaw and over 200 teeth, it can bite down with 2000 lbs of pressure.), juicy sentences (e.g., In every body of water lies an alligator or crocodile.), non-restrictive relative clause (e.g., Their skin has been made out of purses and shoes,whichcould make this animal in danger.), logical metaphors (e.g., Alligators and crocodiles might look coolwiththeir sharp teeth and tough hides. When alligators go after their prey they move extremely fastfortheir size.), and appositives (e.g., Alligators and crocodiles,the modern day dinosaurs, they’ve been around for millions of year.).
Moreover, the patterns of academic language use appear to be fairly consistent across the two grade levels. There are striking similarities in both the types and frequencies of academic language features used by seventh and ninth graders. In fact, one-way ANOVA did not clearly indicate a significant difference between seventh and ninth grades in the overall ALFs (F [1, 91] = 3.33, p = 0.07). The observed power for the test is 0.44, which is rather weak. Kruskal–Wallis Test (χ2 = 2.84, p = 0.09) and the Welch procedure (Fasymp = 2.45, p = 0.12) likewise indicated no significant grade-level effect on the use of ALFs. Further analysis of individual ALFs using Mann–Whitney Tests showed that ninth graders made significantly more use of non-finite clauses (U = 853.5, p = 0.04) and passive voice (U = 808.5, p = 0.03) than did their seventh grade peers. On the other hand, seventh graders made significantly more use of epistemic hedges (U = 791.5, p = 0.02) than did ninth graders.
To determine if academic language use is associated with students’ reading ability, we first divided students into three groups based on their reading achievement levels as determined by a high stakes reading test mandated by the state. The state assigned each student a level (1 to 5) based on their performance on the test. Students assigned levels 1 and 2 (i.e., low performing) did not meet the state expectation for reading achievement and required remediation in school. Students assigned level 3 (i.e., average performing) met the state expectation for reading achievement. Students assigned levels 4 and 5 (i.e., high performing) exceeded the state expectation for reading achievement. The students’ scores on the mandatory state reading test and their overall academic language features by reading level are presented in Table 4.
The relationship between reading test scores and academic language use is graphically represented in Fig. 1. Pearson Product-Moment Correlation analysis revealed significant—and moderately strong—positive association between reading ability and academic language use for seventh grade (r = 0.50, p < 0.001), ninth grade (r = 0.39, p < 0.001), and the entire sample (r = 0.46, p < 0.001). Further analysis using one-way ANOVA indicated significant differences in academic language use among all students with different reading achievement levels (F [2, 90] = 11.14, p < 0.001), regardless of grade level. About 20% of the ALF use is due to differences in students’ reading ability, and the observed power (0.99) is strong. Post-hoc pairwise comparisons (Tukey) revealed that students whose reading achievement exceeded the state expectation (levels 4 & 5) significantly outscored those who fell short of the state expectation (p < 0.001) or met the state expectation (p < 0.001). Alternative tests using Kruskal–Wallis Test (χ2 = 18.89, p < 0.001) and the Welch procedure (Fasymp = 11.23, p < 0.001) also indicated a significant effect of reading ability on the students’ use of ALFs.
ANOVA also showed significant differences among seventh grade students with different reading achievement levels (F [2, 45] = 10.64, p < 0.001). About 32% of the seventh graders’ ALF use is due to differences in their reading ability, and the observed power (0.99) is strong. Post-hoc pairwise comparisons (Tukey) revealed that seventh graders whose reading achievement exceeded the state expectation made significantly more use of academic language features than did those meeting the state expectation (p < 0.001) or falling short of the state expectation (p < 0.001). This finding is further confirmed by the results from the Kruskal–Wallis Test (χ2 = 15.68, p < 0.001) and the Welch procedure (Fasymp = 10.62, p < 0.001).
For ninth grade, ANOVA did not indicate significant differences in ALF use among students of different reading achievement levels (F [2, 42] = 2.82, p = 0.07). Additional analysis using the Kruskal–Wallis test (χ2 = 4.44, p = 0.11) and the Welch procedure (Fasymp = 2.00, p = 0.16) confirmed a lack of significant reading ability effect on the ninth graders’ ALF use.
Finally, we conducted correlation analysis to see if there is any significant relationship between academic language use and writing quality, as measured by holistic scores (see Table 4). One-way ANOVA showed no significant difference between seventh grade and ninth grade in the overall quality of writing (F [1, 91] = 3.05, p = 0.08). The observed power of the test is rather weak at 0.41. Likewise, the Kruskal–Willis test (χ2 = 2.42, p = 0.12) and the Welch procedure (Fasymp = 3.104, p = 0.08) did not detect any significant difference between the two grade levels.
Pearson Product-Moment correlation analysis showed a significant positive association (with moderate to high strength) between ALF use and writing quality for seventh grade (r = 0.72, p < 0.01), ninth grade (r = 0.41, p < 0.01), and the entire sample (r = 0.61, p < 0.01). Simple linear regression analysis showed that ALF use is a significant predictor of writing quality for seventh grade (F [1, 46] = 44.48, p < 0.001), ninth grade (F [1,43] = 5.57, p = 0.02), and the entire sample (F [1, 91] = 46.77, p < 0.001), accounting for, respectively, close to half (r2 = 0.48), slightly over one-tenth (r2 = 12%), and one-third (r2 = 0.33) of the total variance in writing quality. These outcomes were further confirmed by the finding that the slopes for the tests were significantly different from 0 (p < 0.001) and that the confidence interval around the unstandardized slope did not include 0.
Our study explored adolescents’ use of academic language in informational writing.
It found that the seventh and ninth graders in the study made limited use of ALFs in their writing and that there was considerable variation within each grade level but no significant differences between the two grade levels in ALF use. Additionally, ALF use is significantly associated with reading ability. That is, students with higher reading achievement levels tended to use more ALFs than did those with lower reading achievement levels. Moreover, academic language is a reliable predictor of writing quality. Texts with more ALFs tended to receive higher scores for their holistic quality than did those with fewer ALFs.
It is not surprising that the students in our study drew on a somewhat narrow set of ALFs in their informational writing. Instead, they relied quite extensively on the lexicogrammatical resources of everyday language, which are often not as valued in academic writing. These resources include colloquial expressions (e.g., I may be wrong but I’mpretty surethat there are alligators in Australia. This amphibian is highly dangerous butvery coolto study.), interrogative or imperative sentences (e.g., How long have alligators been on earth? Don’t mess with them.), first or second personal pronouns (e.g., Ifyoulook at them they can giveyouquite a scare.), reference to writer’s mental process (e.g., I thinkthat they have webbed feet to help them swim better.), discourse fillers (e.g., Alligators are also incredibly old,like, back to the cretaceous period old.), sentences starting with connective adjuncts (e.g., Butremember these fierce animals with cat-like reflexes have a dangerous personality.Also, they may not look very fast on land.), pronouns with unclear or inconsistent references (e.g., Theywatch over their nest, making sure no one comes to botherherbabies.), amplificatory noun phrase tag (e.g., They are really preciousthose alligators.), and recapitulatory pronouns (e.g., Alligatorstheyare very popular in Florida.). The two sample texts in Table 5 provide a snapshot of the kinds of language resources the students in the sample deployed in their writing.
Similar findings have been reported in other studies of adolescents’ informational writing. For example, Schleppegrell (1998) found that in writing a description of a picture of an animal or a plant in a science magazine, seventh and eighth graders made many lexical and grammatical choices that are characteristic of everyday language but often problematic for meeting school expectations. It is likely that the students in our study were limited in their academic language repertoire such that they had to rely more heavily on their everyday language, a resource that is more plentiful and familiar to them, for making meaning. It is also likely that our study’s task contributed to the students’ heavy use of everyday language. Although the students were explicitly asked to write as a “nonfiction book author” and “from a scientist’s view”, the task prompt used words like “imagine” and “pretend” that might have led them to write in a more storytelling-like manner, for which everyday language would be appropriate. Additionally, the students might have been influenced by the language patterns in an increasingly popular type of informational books, called narrative informational texts (e.g., The Magic School Bus books written by Joanna Cole and illustrated by Bruce Degen), that presents factual information in a narrative style. The intended audience for the writing, grade-level peers, may have also encouraged the students to draw on the more interpersonal and interactive lexicogrammatical resources available in everyday language.
Of the 11 ALFs present in the students’ writing, general academic vocabulary, specialized terminology, expanded noun phrase, and epistemic hedge were used most frequently. Nominalization and passive voice were also used, but to a much less extent. Nonfinite clause, non-restrictive relative clause, appositive, logical metaphor, and juicy sentence were used only sparingly. Given the students’ familiarity with the topic and their cumulative years of schooling, it makes sense that specialized terminology, general academic vocabulary, and expanded noun phrase were the most frequently employed features. Developmental factors may be at work here as well. Research on children’s language development has shown an increasing degree of structural complexity over the school years. This is evidenced in the greater use of longer and more complex noun phrases, passive voice, and modals as children move from elementary to middle and high school (see, for example, Logan, 1976; Perera, 1984; Myhill, 2009; Berman & Ravid, 2009). According to Christie and Derewianka (2008), a landmark volume on writing development from early childhood to late adolescence, expanded noun phrase, non-finite clause, and passive voice emerge during late childhood and early adolescence (ages 9–12) and become more common in mid- (ages 13–15 years) to late (ages 16–18 years) adolescence. They noted that nominalization and logical metaphor are not typical until late adolescence. They also suggested that attitudinal resources such as modal adverbs (e.g., likely, surely, undoubtedly) and modal verbs (e.g., could, may, would) are not common (especially in science) until mid-adolescence. The students in our study (ages 12–15) were able to use hedges such as tend to, probably, and likely to moderate their knowledge claims; however, they also used a strikingly heavy dose of boosters such as really, so, very, and extremely to strengthen their statements, claims and evaluations, which is more common in everyday language but uncharacteristic of informational writing in academic and disciplinary contexts.
Despite the general trend toward greater structural complexity reported in the language development literature, no significant difference was detected in the overall ALF use between seventh and ninth grades. Two features that ninth graders used significantly more than did seventh graders are non-finite clause and passive voice, though both features were used with rather low frequency. On the other hand, seventh graders used significantly more epistemic hedges than did ninth graders. These findings suggest that the development of academic language, like the development of language and literacy in general (e.g., Christie & Derewianka 2012; Ferguson & Slobin, 1973; Perera, 1984), may be nonlinear and feature-specific, with some features registering more robust growth than do others during a particular period of growth and maturation. The development may also be dependent on context and experience. That is, when students have ample opportunities to engage with academic language through exposure and/or instruction, they are more likely to notice, internalize, and appropriate its features. In this connection, it is important to point out that current literacy instruction in language arts and other content areas typically does not pay due respect to academic language outside of academic vocabulary and beyond the ELL contexts (Applebee & Langer, 2006). It is also worth pointing out that as students advance in schooling, they generally become less engaged with academic reading (Ivey & Broaddus, 2001), meaning that they have less exposure to informational texts and academic language. There is no reason to suspect, based on our work in the participating school, that the literacy experience of the participants in our study was markedly different from that reported in the general research literature. This potential lack of engagement with academic texts may explain the limited use of ALFs, a lack of significant growth across grades, and non-significant differences in holistic quality of writing between the two grade levels. On the other hand, given the complexity of academic language and the non-linear nature of language development in general, it may not be unreasonable to expect less than robust growth or even a temporary dip in academic language over a period of merely 2 years. Similarly, the considerable within-grade variation in academic language use is also likely due to developmental and experiential factors.
Another notable finding of the study is that reading ability is significantly related to academic language use. That is, students with higher reading achievement levels used more ALFs than did those with lower reading achievement levels. This finding is consistent with prior research (e.g., Uccelli et al., 2015), which found that academic language proficiency is a significant predictor of reading comprehension. Together, these studies suggest a robust reciprocal relationship between academic language proficiency and reading ability. That is, students with higher reading ability know more about academic language and students who use more (and by implication know more about) academic language read better. It is plausible that students who have higher reading proficiency read more for academic purposes, thus having more exposure to academic language; and familiarity with academic language in turn allows students to divest cognitive resources from linguistic processing to building a situation model (i.e., comprehension) during reading. It is somewhat puzzling, however, that the strength of the relationship appears to be stronger in seventh grade than in ninth grade. One possible explanation for this is that as students grow older, they tend to interact more with social media and spend less time on reading extended school-based texts for academic purposes, which in effect reduces their encounter with academic language. These interpretations are, of course, conjectures that require empirical validation.
The significant relationship between academic language use and holistic quality of student writing found in our study is not unexpected. Previous research (e.g., Figueroa et al., 2018; Galloway & Ucceli, 2015; Uccelli et al., 2012) has also reported significant positive relationship between certain academic language skills (e.g., use of linguistic markers for abstraction, precision, organization, and stance) and the overall quality of academic writing (e.g., definition, persuasive essay). Like previous studies, our study found that academic language use is a significant predictor of writing quality, accounting for a substantial amount of the total variance in holistic quality. It is conceivable that students with higher academic language proficiency have more exposure to models of academic writing through their interaction with academic texts. This proficiency in turn, as Galloway and Ucceli (2015) surmised, enables students to channel their cognitive resources to generating ideas, sharpening focus, and facilitating discursive flow during composition.
Given the robust relationship of academic language use to reading proficiency and writing quality, it is important that students develop a range of lexicogrammatical resources that are functional for constructing academic genres with different degrees of abstraction, density, objectivity, impersonality, and authoritativeness. Teachers can play an active role in supporting this development. However, current efforts to improve students’ reading comprehension and writing proficiency have tended to background the role of academic language (other than academic vocabulary) in reading and writing development, with the consequence that students often lack the linguistic tools for performing academic tasks in ways that meet the school expectations (Schleppegrell, 2004). One powerful way to help students expand their academic language repertoire is to engage them in reading and analyzing academic texts of different types, explaining to them why the author made the particular lexicogrammatical choices and how a different set of linguistic options may have altered meaning, changed style, or impacted discursive flow in ways that would enhance or impede communication (Fang, 2016; Fang & Schleppegrell, 2008). In this work, caution needs be exercised against providing lists of academic language features for students to memorize or use unreflectively. Instead, attention to authenticity, function, context and meaning is of paramount importance. As Langer (2011) emphasized, academic language needs to be taught “through first-hand disciplinary experiences, as language and thought-in-use, in content–area classes” (p. 4), where students have ample opportunities to read, write and talk about topics of significance as they engage in meaningful disciplinary inquiries. At the same time, it is worth bearing in mind that there is no intrinsic merit to academic language features; they are, to borrow from British writing researcher Debra Myhill (2009), “simply linguistic possibilities available to the writer as tools for shaping text” (p. 405). In other words, academic language is a network of choices that enables its users to mean in ways expected of them in school. It does not in itself make a text more grammatical or more communicative, but simply more the kind of text teachers expect, schools value, and experts favor.
Our study suffers from several limitations that must be mentioned. Because the participants in the study were predominantly White and from middle class backgrounds, we were not able to meaningfully explore the impact of sociodemographic factors on academic language use. Future studies can use a larger sample size with participants from socioculturally and linguistically diverse backgrounds to investigate whether and how academic language use varies among students from different socioeconomic, racial, or minoritized backgrounds; how academic language develops across different phases (or grade levels) of schooling; and how academic language proficiency relates to disciplinary reading and writing abilities. In addition, our study used a “wordless” picture as a stimulus to elicit writing. While the task is not uncommon in experimental or quasi-experimental research and has the advantage of maximizing comparability across grade levels, it is nonetheless somewhat artificial and may not call for deployment of the full range of academic language features identified in this study. There is thus a chance that the task does not afford all students the opportunity to fully demonstrate their academic language repertoire. Future research can examine a wider range of academic language features (beyond those used in the study) in multiple writing samples across multiple academic genres collected from authentic units of classroom instruction. These studies will provide a clearer picture of diverse students’ academic language repertoire and its developmental trajectories.
Finally, given the importance of academic language to literacy development and disciplinary learning, it is also imperative to examine effective ways of promoting academic language development in disciplinary contexts, addressing such questions as (a) what is the role of metalanguage in supporting students’ academic language development? (b) how can diverse students’ existing linguistic capital be leveraged to support the development of academic language repertoire and academic identity in service of disciplinary learning and communication? (c) how can a focus on linguistic markers of disciplinary literacy help students foster a sense of belonging within a discipline and boost their confidence in disciplinary exploration? and (d) how can teacher capacity be effectively enhanced with respect to their knowledge about language and their classroom practices involving academic language. Answers to these and other questions will yield more robust pedagogical approaches and evidence-based strategies for supporting academic language development and disciplinary learning at the same time.
Informational writing is a key genre of schooling that is also highly valued in workplace and society. Developing control over a range of linguistic resources for creating informational texts should thus be a high priority in literacy instruction, especially given the convergent findings from our research and other related but differently designed studies. Myhill (2009) argued that writing is a process of design that involves “moving along the trajectories from speech patterns to writing patterns, from declaration to elaboration, and from translation and transformation” (p. 412). Navigating this process requires that students develop a repertoire of linguistic resources that includes, but is not limited to, the academic language features described in our study, for the process cannot be set in motion without the very linguistic tools that are functional for construing markers of advanced literacy such as generalization, abstraction, interpretation, density, elaboration, objectivity, authority, hierarchy, and flow. This development work matters to all students and should be the responsibility of teachers in all subject areas.
Applebee, A. N., & Langer, J. A. (2006). The state of writing instruction in America’s schools: What existing data tell us. Albany, NY: Center on English Learning and Achievement, State University of New York at Albany.
Bailey, A. (2007). The language demands of school: Putting academic English to the test. New Haven: Yale University Press.
Bailey, A. L., & Heritage, H. M. (2008). Formative assessment for literacy, grades K-6: Building reading and academic language skills across the curriculum. Thousand Oaks, CA: Corwin Press.
Berman, R. A., & Ravid, B. (2009). Becoming a literate language user: Oral and written text construction across adolescence. In D. R. Olson & N. Torrance (Eds.), Cambridge handbook of literacy (pp. 92–111). Cambridge: Cambridge University Press.
Bernstein, B. (1971/2003). Class, codes and control: Vol. 1—Theoretical studies towards a sociology of language. London: Routledge.
Biber, D., & Gray, B. (2016). Grammatical complexity in academic English: Linguistic change in writing. Cambridge, UK: Cambridge University Press.
Blake, M. (1996). Crocodiles. Bothell, WA: The Wright Group.
Bunch, G. (2014). The language of ideas and the language of display: Reconceptualizing “academic language” in linguistically diverse classrooms. International Multilingual Research Journal,8(1), 70–86.
Christie, F. (2002). The development of abstraction in adolescence in subject English. In M. Schleppegrell & C. Colombi (Eds.), Developing advanced literacy in first and second language (pp. 45–66). Mahwah, NJ: Erlbaum.
Christie, F., & Derewianka, B. (2008). School discourse: Learning to write across the years of schooling. London: Continuum.
Cicchetti, D. V. (1994). Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment of instruments in psychology. Psychological Assessment,6(4), 284–290.
Cummins, J. (1984). Bilingual education and special education: Issues in assessment and pedagogy. San Diego: College Hill.
Derewianka, B. (1990). Exploring how texts work. Maryborough, VC: Primary English Teaching Association.
DiCerbo, P. A., Anstrom, K. A., Baker, L. L., & Rivera, C. (2014). A review of the literature on teaching academic English to English language learners. Review of Educational Research,84(3), 446–482.
Fang, Z. (2006). The language demands of science reading in middle school. International Journal of Science Education,28(5), 491–520.
Fang, Z. (2012). The challenges of reading disciplinary texts. In T. Jetton & C. Shanahan (Eds.), Adolescent literacy in the academic disciplines: General principles and practical strategies (pp. 34–68). New York: Guilford Press.
Fang, Z. (2016). Academic language and subject area learning. In K. Hinchman & D. Appleman (Eds.), Adolescent literacy handbook of practice-based research (pp. 323–340). New York: Guilford.
Fang, Z., & Schleppegrell, M. J. (2008). Reading in secondary content areas: A language-based pedagogy. Ann Arbor, MI: University of Michigan Press.
Ferguson, C., & Slobin, D. (1973). Studies of child language development. New York: Holt, Rinehart & Winston.
Figueroa, J., Meneses, A., & Chandia, E. (2018). Academic language and the quality of written arguments and explanations of Chilean 8th graders. Reading and Writing,31, 703–723.
Flores, N., & Rosa, J. (2015). Undoing appropriateness: Raciolinguistic ideologies and language diversity in education. Harvard Educational Review,85(2), 149–171.
Francis, D. J., Rivera, M., Lesaux, N., Kieffer, M., & Rivera, H. (2006). Research-based recommendations for instruction and academic interventions. Portsmouth, NH: Center on Instruction.
Galloway, E. P., & Ucceli, P. (2015). Modeling the relationship between lexico-grammatical and discourse organization skills in middle grade writers: Insights into later productive language skills that support academic writing. Reading and Writing,28, 797–828.
Hasan, R. (1989). Language and socialization: home and school. In L. Gerot, J. Oldenburg, & T. van Leeuwen (Eds.), Language in the processes of socialization. Sydney: Macquarie University. Reprinted in J. Webster (Ed.) (2009). Semantic variation: Meaning in society and in sociolinguistics—The collected works of Ruqaiya Hasan (Vol. 2, pp. 119–179). London: Equinox.
Heath, S. B. (1983). Ways with words: Language, life, and work in communities and classrooms. Cambridge, UK: Cambridge University Press.
Heller, V., & Morek, M. (2015). Academic discourse as situated practice: An introduction. Linguistics and Education,31, 174–186.
Hyland, K. (2004). Disciplinary discourses: Social interactions in academic writing. Ann Arbor, MI: University of Michigan Press.
Ivey, G., & Broaddus, K. (2001). “Just plain reading”: A survey of what makes students want to read in middle school classrooms. Reading Research Quarterly,36(4), 350–377.
Langer, J. (2011). Envisioning knowledge: Building literacy in the academic disciplines. New York: Teachers College Press.
Lesaux, N. K., Kieffer, M., Kelley, J., & Harris, J. R. (2014). Effects of academic vocabulary instruction for linguistically diverse adolescents: Evidence from a randomized field trial. American Educational Research Journal,51(6), 1159–1194.
Logan, W. (1976). Language development: Kindergarten through grade twelve. Urbana, IL: National Council of Teachers of English.
Meneses, A., Uccelli, P., Santelices, M. V., Ruiz, M., Acevedo, D., & Figueroa, J. (2017). Academic language as a predictor of reading comprehension in monolingual Spanish-speaking readers: Evidence from Chilean early adolescents. Reading Research Quarterly,53(2), 223–247.
Myhill, D. (2009). Becoming a designer: Trajectories of linguistic development. In R. Beard, D. Myhill, J. Riley, & M. Nystrand (Eds.), The SAGE handbook of writing development (pp. 402–414). New York: Sage.
Nagy, W., & Townsend, J. (2012). Words as tools: Learning academic vocabulary as language acquisition. Reading Research Quarterly,47(1), 91–108.
O’Connor, C., & Michaels, S. (2015). How’d you figure that …OUT?: What can micro-analysis of discourse tell us about fostering academic language? Linguistics and Education,31, 304–310.
Perera, K. (1984). Children’s writing and reading: Analysing classroom language. London: Basil Blackwell.
Ravid, D., & Berman, R. A. (2010). Developing noun phrase complexity at school age: A text-embedded cross-linguistic analysis. First Language,30(1), 3–26.
Schleppegrell, M. J. (1998). Grammar as resource: Writing a description. Research in the Teaching of English,32, 182–211.
Schleppegrell, M. J. (2001). Linguistic features of the language of schooling. Linguistics and Education,12, 431–459.
Schleppegrell, M. J. (2004). The language of schooling: A functional linguistics perspective. Mahwah, NJ: Erlbaum.
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: Cambridge University Press.
Uccelli, P., Dobbs, C. L., & Scott, J. (2012). Mastering academic language: Organization and stance in the persuasive writing of high school students. Written Communication,30(1), 36–62.
Uccelli, P., Galloway, E. P., Barr, C. D., Meneses, A., & Dobbs, C. L. (2015). Beyond vocabulary: Exploring cross-disciplinary academic-language proficiency and its association with reading comprehension. Reading Research Quarterly,50(3), 337–356.
Vadasy, P. F., Sanders, E. A., & Herrera, B. L. (2015). Efficacy of rich vocabulary instruction in fourth and fifth-grade classrooms. Journal of Research on Educational Effectiveness,8, 325–365.
The preparation of this paper was supported in part by the National Research Center for Foreign Language Education (MOE Key Research Institute of Humanities and Social Sciences at Universities), Beijing Foreign Studies University. We thank Valerie Gresser and Erin Mistry for their assistance with data analysis.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Fang, Z., Park, J. Adolescents’ use of academic language in informational writing. Read Writ 33, 97–119 (2020). https://doi.org/10.1007/s11145-019-09937-8
- Academic language
- Informational writing
- Literacy instruction
- Disciplinary learning
- Adolescent literacy
- Language and literacy development