Academic Language Skills and Reading Comprehension

July 18, 2014, The IRIS Center

Paola Uccelli, Associate Professor of Education at the Harvard Graduate School of Education, overviews her research into the domains of academic language skills that can help students to improve their reading comprehension (time: 10:46).

Question asked of Paola Uccelli in this audio:

Do you mind telling us about your research into the different domains of academic language skills?

Paola Uccelli, EdD
Paola Uccelli, EdD
Associate Professor of Education
Harvard Graduate School of Education

Transcript: Paola Uccelli, EdD

Narrator: This interview is brought to you by the IRIS Center, a national center funded by the Office of Special Education Programs.

Paola Uccelli, Associate Professor of Education at the Harvard Graduate School of Education, overviews her research into the domains of academic language skills that can help students to improve their reading comprehension (time: 10:46).

Narrator: Dr. Uccelli, do you mind telling us about your research into the different domains of academic language skills?

Paola Uccelli: So we’re working with this new operational construct of academic language that we label Core Academic Language Skills, or CALS. They refer to a constellation of high-utility language skills that support reading comprehension across school content areas. So far, we have conducted empirical studies, and we have been able to identify seven domains of this construct that seem relevant to reading comprehension. We do not see these domains as exhaustive but as critical pieces that we know contribute to reading comprehension but certainly and probably not the only ones. So the first domain is what we call academic register awareness, which refers to skills in identifying academic register. So basically we present students with more academic forms of discourse in comparison to more colloquial alternatives, are they able to distinguish those? Are they even implicitly aware that more precise, more concise, more explicitly connected language is what you would expect for a school task versus a more colloquial form of language? Our research has shown that kids tend to be aware of power dynamics of academic language and of being more polite, or sounding smarter when you use academic language, but much less conscious of the function of properties that we would want to emphasize at school using precise, concise, explicitly connected language to learn and to understand content. So being aware of these expectations of academic language and the distinctions of more colloquial versus more academic uses of language is the first domain that we that we assess and that we have collected data on.

The second domain is what we call organizing analytic text, which means literally skill in organizing school-relevant types of texts. And what happens when a reader approaches a text is that it is very helpful to have an organizing frame, sort of global map or template, with the expected components and expected sequence of that text. So for narratives, for example, we typically would expect a setting, a sequence of events, and a resolution at the end. And we know through a lot of research that children around the age of ten have mastered the structure of narratives. But yet around grade four, students are expected to read and write many different types of texts: explanations, argumentative essays, definitions. So skills in how these, what we call analytic non-narrative texts, are organized, developed throughout the middle school and high school years, and they seem to contribute to reading comprehension as well. So this second domain refers to these organizing analytic texts at a more global level.

The third domain that we have identified is unpacking dense information, which refers to skill in understanding more phonologically complex words and complex sentences. And what we emphasize in this domain is that beyond the meaning of words, understanding morphologically complex words and complex sentences, in which those words are used, also seem to contribute to reading comprehension. What we know is that school texts, academic texts in general are populated by morphologically complex words. There are many nominalizations in academic texts which means instead of saying, “The chemicals contaminated the rivers,” the text would say “The contamination of rivers.” So from contaminate to contamination, we have now a nominalization and beyond the meaning of the word, understanding that those two words are related and that you can understand contamination in light of what you know about contaminate, seems to contribute to reading comprehension. And the same with complex syntax. So complex morphology and complex syntax we view as skills to packing dense information. And when a text displays complex syntactic fractures, the challenge is not only in the meanings in the words but in how those words are related and what are the hierarchal relationships to understand the meaning.

The fourth domain that we have identified is connecting ideas logically and this domain refers to skill in understanding school-relevant connectives and discourse markers. What we know is that there are many connectives and discourse markers that are very frequent in school text but are very rarely used in oral conversations: consequently, nevertheless, on the one hand, on the other hand. These connectives, these discourse markers, are signals that are used in a text to help a reader understand the relationships among ideas. Yet, these connectives are only helpful if the reader knows them. And from our research, we see enormous individual variability in knowledge of connectives, so that has an impact that seems to be associated with reading comprehension as well.

The fifth domain that we have identified in our research is what we call tracking participants and themes. And that refers to skills in solving chains of reference. In other words, texts usually discuss a set of participants or a set of themes extensively. Therefore, texts need to refer to these participants or these themes repeatedly. And for our readers to track who or what the text is talking about as it unfolds can actually prove very challenging. In science, for example, it is very common to have a noun that refers to a full sentence. If I give you an example: “Water evaporates at higher temperatures. This phenomenon…” Well, now the word phenomenon is not referring to one thing; it’s referring to an entire sentence, and this proves challenging for kids, and its a skill that develops throughout the middle school and high school years. And also in history texts, for example, it’s not unusual to have very different expressions to refer to the same participants, right? So within a text, you might have the king, the tyrant, this leader, and kids need to track that all these expressions are referring to the same participant. Just tracking that and helping students track that, proves enormously helpful and the teachers we have worked with in history find this supportive strategy to actually promote historical understanding in their classes.

So the final two domains are more exploratory in our work. Domain number six is what we call understanding meta-linguistic vocabulary. And this refers to skills in understanding a particular subset of vocabulary words and those that refer to language or thinking processes and that support text-based discussion and argumentation—so, for example, generalization, counter-argument—and what we think is that understanding these words does not only expand vocabulary in general but actually they offer a label to highlight a thinking or a discourse process that makes it more visible for students and helps them in understanding what discussing and having an argument and thinking processes are.

The final domain is interpreting viewpoints and that refers to skills in understanding markers of stance. And we are particularly interested in markers of degree of certainty in relation to a particular statement. So, basically, when we have a claim, a text, or a writer, we’ll often evaluate that claim. So writers or texts might say, “It is possible that pollution causes global warming.” Or they might say, “Certainly, pollution causes global warming.” That ability of readers to understand what the stance in relation to a claim is seems to be also another skill that we’re exploring as associated to reading comprehension. Particularly, when we think about deep reading comprehension and comparing different texts, for example, paying attention to what is the stance of the text or the writer seems to be particularly critical.

So we see these seven domains that we have now empirical data on as an initial map but not an exhaustive list. But our research already shows that this combination of skills are highly predictive of reading comprehension in upper-elementary and middle school students.

Narrator: Thank you for listening to this IRIS Center interview. For more information about the IRIS Center and its resources, visit us at [].

Print Friendly, PDF & Email