The inference-making of elementary emergent multilinguals: Access and opportunities for learning

Abstract

Inference-making is integral to reading comprehension, defined as information ‘retrieved or generated during reading to fill in information left implicit in a text’. However, there are few studies regarding the inferencing of young emergent multilinguals that account for multilingualism and culture, attend to the learning processes influenced by classroom instruction and interaction, and utilize qualitative methods to explore the multifaceted process of inferencing. This study explores classroom instances in which two second-grade emergent multilinguals encountered difficulties with inference-making, specifically when their first language or background knowledge was not accessed to support inference-making. Framing their difficulties as missed opportunities for learning, this study considers first language integration and multilingual pedagogies as ways to enrich learning. Findings revealed that inference-making was highly contextual, differing from text-to-text. Learners were able to infer in many instances, but difficulties arose when learners desired to articulate their inferences with specific words. Additionally, students encountered difficulties with the content language and background knowledge necessary to access an expository text. Implications of this study include supporting inference-making with students’ first language and culturally sustaining pedagogies in order to provide meaningful opportunities to engage in higher-order thinking surrounding texts.

Keywords

Reading comprehension reading young children English as an additional language (EA)literacy development primary school

Making inferences is an integral part of the reading comprehension process, as well as a general skill (Kendeou, 2015; Kendeou et al., 2008; McNamara, 2007; Oakhill and Cain, 2007, 2012). Because of the relationship between inference-making and general reading comprehension, much research has documented the inference-making of young children during reading (e.g. Perfetti et al., 2005; Yuill and Oakhill, 1991). However, prior research has largely focused on monolingual learners. As such, there are missed opportunities to understand the complexities of reading comprehension and inference-making in an additional language, and to further explore how the classroom environment and instructional choices can limit emergent multilinguals'¹ opportunities to learn (Escamilla, 2015). Thus, prior studies on inference-making have not fully (a) accounted for students' multilingualism and cultural backgrounds, (b) attended to the processes of learning to inference through classroom instruction and interaction, or (c) utilized qualitative approaches to explore the contextual and individual dimensions of inferencing.

Given this gap in research, this study analysed instructional segments when four second-grade emergent multilinguals were not able to access their first language and/or connect to background knowledge as they made inferences during reading. By employing the term ‘access’, students’ difficulties with inference-making are not positioned as individual deficits, but rather recognized as instances that take place in a particular social context with affordances and constraints (Luke et al., 2011). As such, the present study draws on sociocultural perspectives of reading comprehension to expand current research on inference-making, further positing that social interactions, context and students’ experiences influence their understanding of texts (Aukerman, 2013; Street, 1993). By positioning enquiry in this way, the present study argues that these difficulties demonstrated missed opportunities to learn (Escamilla, 2015), as well as missed opportunities to integrate students’ languages and cultures as they developed their reading skills. Specifically, this study was guided by the following enquiry: ‘What difficulties arise when students are not able to access their first language or background knowledge to support inference-making?’

Background

This study expands cognitive conceptualizations of inference-making by considering sociocultural aspects of making inferences during reading in English-as-an-additional-language classrooms. An inference, in relation to reading comprehension, is ‘information that is retrieved from memory or generated during reading to fill in information that is not in a text’ (Elbro and Buch-Iversen, 2013; Kendeou et al., 2016: 63). In other words, in order to make an inference about a text, learners need to integrate prior knowledge or experience with textual clues to draw a conclusion that goes beyond what is explicitly stated in the text. A sociocultural expansion of this definition considers the multitude of experiences and interactions that inform students’ sensemaking processes, which may not align with the teacher's assumptions or other students’ experiences, and thus influence the ways students infer from texts (Aukerman, 2013; Luke et al., 2011; Street, 1993). For example, students may read a text that says, ‘The family had many things to do to prepare for their new puppy.’ Some students may reasonably infer that the family might buy a dog bed, food, a lead as per the student's own prior experiences, familiarity of owning pets in their culture, and evidence on the page. However, the experiences and prior knowledge of students unfamiliar with getting a new pet may cause unexpected inferences or difficulty. To support an extended perspective of inference-making, the following sections frame reading comprehension as a social and cultural endeavour and then further discuss the literature on inference-making and young children.

Theoretical framework: Access and positioning reading comprehension in social contexts

This study draws on sociocultural perspectives of learning to explore the access students had to their first language and background knowledge when inference-making was difficult. The term ‘access’ in this study derives from the notion of ‘opportunity to learn’ (Boals et al., 2018; Escamilla, 2015), or the consideration that instructional and social elements may impact on students’ access to content and learning in school. Escamilla (2015) explained that educators and schools may not provide holistic and/or appropriate learning opportunities for emergent bilinguals, such as extended time to talk about books, access to academic language and integration of students’ first languages. With the notions of access and ‘opportunities to learn’, reading comprehension is recognized as a meaning-making process that is continuously afforded and constrained by institutional, community, school and social relationships.

Some scholars consider comprehension a ‘cultural phenomenon’, in that understanding texts often requires a broader knowledge of possible ‘origins, motivations, and consequences of a text's meaning’, which ultimately connects to personal experience and knowledge (Bunch et al., 2014: 541; Luke et al., 2011: 160). Luke et al. (2011) proposed reading comprehension as a social and intellectual practice, and acknowledged the complexity of many layers of sociocultural aspects of reading, including the ways it can open doors for people:

We read in ways constrained and defined, enabled and afforded by contexts, then we read and make meaning not only through the reader/text interaction and cognitive processes described in traditional reading research, but also through entry into institutional contexts and social fields of exchange where texts are used. (p.161)

This quote suggests the ideological and political tensions that surround reading, recognizing that reading is entwined with the sociocultural contexts in which readers are situated.

The process of understanding texts in social environments is an important venture in of itself (Aukerman, 2013; Purcell-Gates, 2007; Street, 1993; Vygotsky, 1978). Aukerman (2013) emphasizes the importance of how students figure out texts, rather than what they produce. As such, this study is also broadly framed by a ‘comprehension-as-sensemaking’ (Aukerman, 2013) approach to consider the processes of sensemaking in the classroom and the choices teachers and researchers make to provide a generative sensemaking environment for students. Consequently, taking up sociocultural perspectives of reading comprehension requires critique of current practices and frameworks in the service of culturally and linguistically sustaining approaches that draw on all students’ linguistic and cultural resources as they make sense of texts (Escamilla, 2015; Fairbanks et al., 2017; Helman and Stai, 2017).

Context, culture and interaction in literacy learning

Through sociocultural and critical frameworks, research has demonstrated the importance of integrating prior knowledge, meaningful opportunities for interaction, and explicit use of cross-cultural knowledge to assist understanding and support emergent multilinguals’ general language development and reading comprehension (Aukerman et al., 2017; Carrell, 1983; Goldenberg, 2013; Lee, 2016; Martin-Beltrán et al., 2017; Soto Huerta, 2012; Swain et al., 2002). Martin-Beltrán et al. (2017) demonstrated how peer mediation, utilizing home language, and extended discussion can influence elementary emergent multilinguals’ meaning-making during reading. Although students were of different ages in a peer buddy programme, this study demonstrated how partners negotiated meaning by using their home language and discussing personal experiences. The researchers articulated that, through mediated support, peer discussion expanded the possibilities and processes of meaning-making by building a ‘Zone of Relevance’, or by making the text relevant to each other's lives. Aukerman et al. (2017) similarly analysed the collaborative nature of meaning-making during small-group reading amongst second-grade emergent multilinguals, describing how students interacted and extended each other's ideas. Notably, Aukerman et al. (2017) acknowledged the difficulty in relying on early language learners’ verbal utterances to understand their textual ideas, citing the need to offer opportunities and support for first language integration. A plethora of literature provides important insights into how context, culture and interaction influence literacy learning, yet few studies specifically examine the impact of these elements on young emergent multilinguals and their inference-making. As such, further research is necessary to determine the unique contributions emergent multilingual's linguistic and cultural backgrounds bring to this work.

Inference-making of young children

Inference-making is both a general skill and a central component of reading comprehension (Kendeou, 2015). Young children make inferences well before they learn to read (e.g. while listening to stories or watching a television show), and these inferences grow in complexity over time as they gain more knowledge and experience of the world (Kendeou et al., 2007, 2008, 2009, 2016). Additionally, some research has explored how general inference ability can influence reading comprehension and inferencing ability during reading (Kendeou et al., 2008). However, the parameters of this study limit the focus to studies related to how students make inferences and make sense from texts. In order to make an inference while reading, a learner must be able to decode text, access the content and vocabulary, understand text structure, connect ideas across the text, and integrate prior experience or background knowledge (Kendeou et al., 2014; Melby-Lervåg and Lervåg, 2014; Snow, 2002). It is also important to note that inference-making is a covert process, and thus students’ products of inference-making have been commonly assessed as students are reading (via think alouds, questions), and after reading (post-reading questions). As such, inference-making ‘during reading’ in this study encompasses the oral production of inferences students made during reading discussions.

As inference-making skill is related to skills in broader reading comprehension (Kendeou et al., 2014), it is necessary to mention the variety of skills that contribute to students' understanding of texts. One area in research has been concerned with how vocabulary and oral language development impact on reading comprehension for both monolinguals and multilinguals. Once students become proficient in decoding skills, oral language development and vocabulary can be attributed to comprehension success (e.g. Kendeou and Van den Broek, 2005; Proctor et al., 2005). For example, Proctor et al. (2005) explored the impact of decoding skill and oral language skill on reading L2 comprehension. In their study of 135 Latinx fourth graders, they assessed English oral language skills (listening comprehension and vocabulary), English decoding ability and English reading comprehension. After decoding skills proved to be adequate amongst many of the students, oral language skills were most significant for reading comprehension. Specifically, Proctor and colleagues found both a distal and proximal relationship between vocabulary and reading comprehension, meaning that vocabulary scores directly correlated with reading comprehension, and also indirectly correlated with reading comprehension via listening comprehension scores. Despite their findings, they noted the limitations in collecting English-only measures of literacy learning, recognizing students' first language as a unique variable. As a result, Proctor et al. called for more literacy research that considers all of the bilingual abilities of learners.

Children who are challenged by general reading comprehension are also challenged by the skill of inference-making during reading (Bowyer-Crane and Snowling, 2005). However, the specifics of the relationship between reading comprehension and inference-making have been articulated with less consistency across the literature, with researchers applying terminology such as causality (Cain and Oakhill, 1999), or simply explaining inference-making as an integral component of reading comprehension (Kendeou, 2015; Kendeou et al., 2016). The current study does not explore causality or quantify specific relationships but brings further attention to the nuances of the inference-making of young emergent multilinguals, who have a variety of language skills and cultural experiences that may influence their inference-making. Yet, studies with English monolinguals may inspire further exploration and replication of inference-making research with emergent multilinguals. For example, Cain and Oakhill (1999) found that 7–8-year-old children's difficulties with inference-making led to difficulties with comprehension, rather than vice versa, illustrating the significance of inference-making skill in relation to general reading comprehension. To select children, Cain and Oakhill first assessed children on their reading comprehension. The children read a series of short texts aloud, graded in difficulty, and were asked questions about the text. Then, the researchers created two groups, ‘skilled comprehenders’ (n = 24) and a group of ‘less-skilled comprehenders’ (n = 29). To explore sources of comprehension difficulty, the ‘less-skilled’ comprehenders read aloud short texts (at appropriate difficulty levels) and were asked questions about the reading. With the ‘less-skilled’ comprehenders, Cain and Oakhill considered and addressed potential background knowledge discrepancies, with opportunities to reread the texts. One of the significant findings from this study was that ‘less-skilled’ comprehenders had difficulties taking up strategies, such as drawing connections, to make inferences. Additional research has similarly revealed how young children can often make simple inferences about events and feelings (Kendeou et al., 2008; Van den Broek, 1997), but need explicit support during reading as inferences often do not happen spontaneously (Brown, 1977; McKoon and Ratcliff, 1992; Oakhill and Cain, 2007; Trabasso and Bouchard, 2002). In elementary settings today, explicit support manifests in the teaching of reading strategies and skills via guided reading, a small-group reading practice in which students read and are guided in using comprehension strategies by a teacher (Fountas and Pinnell, 1996).

With the many variables that intersect during text comprehension, it can be difficult to determine what causes difficulty during students' sensemaking. However, with regard to inference-making, Kendeou (2015) claimed that the biggest variance in inference-making ability is the reader's background knowledge, as children generally have difficulties making inferences when encountering difficult or unfamiliar content. Developing readers also often do not know when it is appropriate to draw inferences and may have difficulties using and integrating their prior knowledge (Cain et al., 2001; Oakhill and Cain, 2007; Yuill and Oakhill, 1991). Prior work with emergent multilinguals and English monolingual learners highlights the significance of background knowledge for reading comprehension, as learners may have differing experiences with text content (e.g. Ebe, 2010, 2012; García, 1991; Jimenez and Gamez, 1996). Elbro and Buch-Iversen (2013) explored the significance of background knowledge on inference-making with sixth-grade students. They examined students' inference-making and overall reading comprehension across 16 classes (236 students of similar reading abilities) after teaching experimental groups how to activate background knowledge while reading expository texts. Although the researchers did not account for important contextual factors, such as vocabulary and bilingual learners (9.7% of their participant population), they found that students in the experimental group (n = 151) had more success with inference-making when they were taught how to incorporate background knowledge. Despite the literature on activating prior knowledge and building background during reading, few studies have considered younger emergent multilinguals' linguistic and cultural backgrounds in connection to their opportunities to make inferences during reading, and fewer studies have taken up frameworks that integrate languages and cultures to explicitly support inference-making. As such, this study explored the difficulties that arose when students were not able to access their first language or background knowledge to support their inference-making during reading.

Methods

Setting and participants

The setting for this study was Leah Shepard-Carey's English as an additional language classroom in a metropolitan Midwestern city in the USA during January and February 2017. The school was composed of students from many cultures and language backgrounds. The students involved in the study were four second-grade students learning English as an additional language, who were seen daily in small groups for reading and language support. Students fit the district criterion of ‘approaching’ or ‘below’ grade-level readers and beginning/intermediate language learners, according to quarterly assessments and the WIDA ACCESS test (for additional descriptive information about the participants see Appendix 1). The participants included two boys, Ahmed and José, and two girls, Naima and Mariana (pseudonyms), who were eight years old by the end of the study. The students had similar beginner/intermediate English language proficiencies. Two students identified Somali (Naima and Ahmed) as their home language and two students identified Spanish (José and Mariana) as their home language. Despite some differences in students’ English language skills, they generally read at a mid-first grade level with somewhat complex storylines, grade-level content and varying sentence complexity. Students were proficient decoders but continued to develop in their reading of challenging blends, digraphs and vowel/consonant combinations.

Data collection and materials

Data for this particular study come from a larger project that took place over the course of eight weeks, during January and February 2017. Instructional segments from 11 lessons, from 23 January 2017 to 16 February 2017, were chosen for analysis, amounting to 5.5 hours, or 11 days of 30-minute instructional periods. Leah chose the 11 lessons for analysis because they were the most consistent in routine. Inferencing lessons prior to the guided reading lessons included mini-lessons, shared reading and whole group reading, which were seemingly beneficial, but students did not individually read the texts and have significant talk time with every text presented. During the 11 guided reading lessons, the four students read, discussed and completed activities surrounding three levelled texts from the school reading curriculum resources (see Table 1). These texts came from Reading A–Z and a Fountas and Pinnell reading curriculum. Texts were chosen based upon students' instructional reading level (Fountas and Pinnell, 1996), students' interests and experiences, and students' content knowledge. The texts chosen for the study's guided reading instructional period included two realistic fiction texts and one non-fiction text, which were aligned with instructional units and goals in the home classrooms at the time.

Table 1.

Small-group reading texts and descriptions.

Text	Descriptions
The Gecko that Came to School (Fountas & Pinnell, Level H)	Realistic fiction. Includes characters (including main character) of colour in the illustrations and storyline. A young girl thinks it would be a good idea to bring her pet gecko to school. The gecko causes trouble and chaos around the school.
Brother Messy, Brother Neat (Reading A–Z, Level H)	Realistic fiction. Includes characters of colour (main characters) in the illustrations and storyline. Two brothers, who disagree often and have different personalities, get a new baby sister, so they have to learn how to share a room.
Ships and Boats (Reading A–Z, Level H)	Non-fiction. This text informs the reader about ships and boats. The text explains and compares different characteristics of ships and boats. The text includes illustrations and photos of ships and boats.

The data presented in the transcripts below were derived from audio data from the 11 lessons, specifically relating to discussions during and after reading of the texts. As a result of district permission policies for video, the lessons were audio-recorded on a computer and/or mobile device. Due to the active participation by Leah, field notes were recorded during and immediately after the lesson, amounting to about 20 pages of typed notes. Lessons were transcribed from audio within a day or two after each lesson and compared to field notes, texts and artefacts from the lesson during transcription (Erickson, 2006). Leah typed memos after transcription. Artefacts included any graphic organizers, diagrams, illustrations or written products used by the students during instruction.

Instructional planning and support

Prior to the guided reading lessons, the students were introduced to the concept and language of inference-making via guided practice with pictures and shared texts for seven days (see example of picture and lesson dialogue in Appendix 2). The lessons in this set of data utilized a modified guided reading approach (see Avalos et al., 2007), during which text-specific vocabulary and language forms were taught before reading, with additional time spent making personal connections and building background. As both researcher and teacher, Leah planned lessons in advance, but modified them as needed. Typically, each text required three to four 30-minute lessons in order to address the language, content, multiple reads of a text, and after-reading instructional activities.

English learners need access to both the content and language in texts in order to comprehend them (August et al., 2005; Bunch, 2014). Hence, the first day with a text was spent building background knowledge, learning vocabulary and previewing the text with routine scaffolds such as pictures and graphic organizers (i.e. mind maps, Venn diagrams etc.; see Appendices 4 and 5 for examples of dialogue and instructional artefacts). The second and third days included reading the text together and discussing the text during reading, followed by an independent reading of the text. The multiple reads of a text also provided the opportunity to resolve issues with decoding through explicit support and hearing the text read aloud (e.g. sounding out a word together). The final day with a text included a brief post-reading discussion and a writing response (writing responses were not analysed because students completed writing in pairs). During and after reading, students made inferences with teacher questioning (i.e. ‘How does the character feel?’; ‘Why do you think Tess (character) said “This Gecko is the best pet ever?”’; ‘Why did they (people) need to use boats in the past?’). Most questions were planned ahead of time, but additional questioning took place for clarification and formative assessment. The guided reading approach, which had been in place for the entire school year, also provided a routine with which students were familiar, and thus students were seemingly comfortable with the structure and discussion practices.

Analysis

Analysis began with transcription and descriptive and analytic memoing of each lesson (Saldaña, 2015). Then, cycles of initial and axial coding (see codebook in Table 4 in Appendix 1) were employed (Miles et al., 2013; Saldaña, 2015). The initial coding stage involved identifying inferences and difficulties with inference-making. These initial codes were developed based on classroom data and prior literature, then merged and modified throughout two stages of coding (Saldaña, 2015). In the second-round, axial coding was employed to identify patterns in specific types of inferences and difficulties with inference-making. During each coding cycle, transcripts were compared with field notes and audio to verify or modify codes.

Intra-rater reliability was established using the code and recode method (Saldaña, 2015) by Leah. Inter-rater reliability can be contested in qualitative research that is independently carried out, as it is possible that we can train others to see our data as we wish (Armstrong et al., 1997; Morse, 1997). Hence, all transcripts from the 11 lessons were coded and recoded twice by Leah, who was the sole data collector, throughout each phase to ensure reliability, with 90% agreement. In the instances where inconsistencies occurred, Leah further consulted field notes and audio recordings, which resolved most discrepancies. Additionally, informal consultation with colleagues (graduate students and professors) regarding inconsistencies during the analysis process provided further insights and closer analysis to specific moments (Saldaña, 2015). Triangulation of data occurred across transcripts, audio data, memos and fieldnotes (Miles et al., 2013). Though comparison of transcripts and audio-recordings to memos provided the most consistency and reliability, fieldnotes added contextual information that was not necessarily captured in the transcripts or recordings (e.g. a student who was distracted that day). Coding provided initial insights into the difficulties that were encountered, but the codes did not suffice in providing detail surrounding the unit of analysis. Thus, descriptive analysis and additional memoing of selected moments were helpful in elucidating the findings.

Findings

In this section, the research question, ‘What difficulties arise when students are not able to access their first language or background knowledge to support inference-making?’ is addressed. Again, taking up the term ‘access’ reflects the need for further instructional support and acknowledges the classroom context as a space that privileges English (English-dominant school setting) and certain cultural norms. There were two main findings. First, some inference-making difficulties connected to the need to articulate specific vocabulary related to texts. Second, inference-making difficulties with the non-fiction text potentially resulted from students’ access to background knowledge related to the content. The findings are presented with descriptions of students’ general inference-making, identifying difficulties in inference-making, and instructional segments during which students needed access to language and background knowledge in order to make inferences.

According to the definition of inference-making by Kendeou et al. (2016), inferences were identified as instances in which a student ‘filled in’ information that was not explicitly stated in the text. In this study, all students made inferences with prompting and support, usually in response to teacher questioning and, at times, with peer support. Throughout the 11 lessons, there were seven codes within two larger categories related to the inferencing of individual students (see Table 4 in Appendix 1): (a) the types of inferences students made (five types, n = 68) related to the text and (b) the difficulties students had with inferences (two types of difficulties, n = 15). Students encountered difficulties with inference-making in 18% (15 out of 83) of inferencing moments (moments in which an inference was made or attempted), specifically related to background knowledge or in the language required to make the inference. Table 2 shows the inference-making amongst specific students and student absences (if applicable).

Table 2.

Students' inferences and difficulties with inference-making.

	Students
	Mariana	Jose	Naima	Ahmed	Total
Book 1
Inferences	7	6	10	5	28
Difficulties	2 *L	1 *L	0	1 *L	4
Absences	0	0	0	0	0
Book 2
Inferences	5	6	5	3	19
Difficulties	1 *L	1 *L	0	0	2
Absences	0	0	0	2	0
Book 3
Inferences	4	6	6	5	21
Difficulties	4 *BK	2 L 2 BK	1 *BK	0	9
Absences	1	1	0	0	2
Totals
Inferences	16	18	21	13	68
Difficulties	7	6	1	1	15
Absences	1	1	0	2	4

*L: difficulty with access to language; *BK: difficulty with access to background knowledge.

Table 2 shows discrepancies in the number of inferences each student made, most likely the result of a combination of absences, student characteristics and the dynamics in the classroom. As the teacher, Leah attempted to provide opportunities for every child to participate through prompting and questioning. But, it is difficult to make any claims as to why some students made more inferences than others or had more difficulties with inference-making than others.

Identifying difficulties during inference-making

Analysis revealed that students had two types of difficulties with inference-making (n = 15), accessing the language needed to express inferences (n = 8) and accessing the background knowledge (n = 7) required to make inferences. These difficulties were viewed as missed opportunities to draw on learners’ first languages and cultures. Most of the language-related difficulties took place during the first two books, which were realistic fiction. The difficulties with background knowledge took place with the non-fiction text. Instances were initially identified by Leah as ‘difficulties’ when (a) the inference was not fully articulated or misunderstood, (b) the student had difficulties articulating the inference because of vocabulary, (c) the inferences did not make sense given the text evidence, or (d) the student explicitly stated their confusion.

An example of an utterance coded ‘access to language’ included a moment when Mariana attempted to infer and describe her justification (i.e. explaining an inference with evidence or prior knowledge) of the feelings of the teacher, a character in one of the texts.

Example A: Access to language

1 Mariana: I think Ms. Peck will get mad because the principal told to all the teachers. [long pause and tries to figure out what to say] What it called?

2 Naima: To not bring?

3 Mariana: To not br: pets are not loud (allowed). //Even I have another inference!//

In this moment, Mariana expressed difficulty in articulating her thoughts with a pause and then explicitly asked for support. Difficulties with access to background knowledge were identified when students had difficulty producing a response that made sense with the context of the text or had difficulty drawing upon prior knowledge. The following example comes from a moment when the students were asked to infer about the ways small boats help people complete jobs or tasks.

Example B: Access to background knowledge

1 Leah: So you all are right that they use them (boats) to take them places(.) but what types of jobs did small boats help them do?

2 Mariana: Push the water.

3 Leah: Push the water, okay.

4 Naima: To carry people.

5 Leah: To carry people?

6 Naima: And to take them somewhere else.

7 José: To keep people safe, and to go to the restaurant.

The repetition and developing details in these responses led Leah to believe that students did not have access to the content or experience they needed to make inferences. In this moment, Leah expected Mariana and José to draw on prior knowledge or experience with boats. The subsequent sections extrapolate on the main findings with instructional segments to illustrate how difficulties played out in the classroom context.

Instructional segments: Students' access to language

In the data, difficulties while articulating inferences seemed to derive from access to language at the word level and describing inferences with details. At times, the challenges students encountered did not detract from the flow or understanding of the utterance. Other times, the difficulty stopped the process and required questioning and support. Most often, this difficulty became salient when students would repeat, approximate or get stuck on a word. Across all three texts, Naima did not present any overt difficulties expressing her inferences or accessing the language needed to make an inference aloud. However, it would be impudent to say she did not have difficulties relating to language at all because of the limitations present in the group setting and Leah's inability to access Naima's internal processing. In this section, Mariana's, Ahmed's and José's challenges regarding accessing the language during their inferencing are presented separately.

In excerpt one (also in the example), Mariana inferred in response to a question about the teacher's reaction to the student who brings her pet to school (See transcription conventions in Appendix 3).

Excerpt 1: 23 January 2017

1 Mariana: I think Ms. Peck will get mad because the principal told to all the teachers. [long pause and tries to figure out what to say] What it called?

2 Naima: To not bring?

3 Mariana: To not br: pets are not loud (allowed). //Even I have another inference!//

Mariana paused in line 1, and explicitly asked for language support, perhaps desiring to express a certain word or phrase to justify the inference. Naima tried to help her in line 2 and Mariana remembered the word ‘allowed’ in line 3. In this instance, Mariana illustrated that she had prior knowledge connected to the inference because she elaborated upon what a principal might hypothetically say to teachers, but one phrase or word prevented her from articulating it immediately. Notably, she also communicated that she had another inference, implying that she understood the concept.

In excerpt two, Mariana's difficulty was marked by pauses, which again seemed to result from the desire to articulate a specific word. In the following interaction, she inferred why the characters, two brothers, needed rules to share a room.

Excerpt 2: Rules, 6 February 2017

1 Leah: Why do they need rules?

2 Mariana: I think to not be organized, and to not be organized(.)To be(.) when their little (.) okay [shows frustration by sighing]

3 Leah: Do you want to take a minute and think? I think they need rules because?

4 Mariana: I think they need rules because when their sister grows up she'll be organized.

Indicated by the pauses and sigh in line 2, she was frustrated. She focused on the word ‘organized’ and struggled to express her thoughts. In line 4, after a chance to think, Mariana came back to the word ‘organized’. She made a linguistically coherent inference, but it was difficult to understand without further context. At this point, Mariana's level of frustration and Leah's need for further context highlighted a missed opportunity to draw on her first language for further learning.

Ahmed was absent twice throughout the data collection during small group lessons and had a quieter disposition, thus the data probably do not represent a holistic representation of his inference-making. In a conversation related to book 1, students were asked to infer about feelings of the teacher on a particular page. Ahmed inferred and used the word ‘jealous’, which did not make sense with the context.

Excerpt 3: The Teacher's Feelings, 24 January 2017

1 Leah: I love how you're making an inference about how the teacher is feeling right now. How do you think she's feeling? [directed towards the rest of the class]

[students share almost simultaneously]

2 Mariana: Happy, sad

3 Naima: =Excited

4 José: =Jealous

5 Leah: What do you think Ahmed? How is she feeling?

6 Ahmed: I think she is feeling, //jealous.//

7 Mariana: //Surprised!//

8 Leah: Jealous? What tells you she is feeling jealous?

[long pause]

9 Leah: Does it look like she wants a gecko in her classroom?

10 Students: No

11 Leah: So jealous would mean that I want what you have.

12 Naima: If she would be jealous she would be like [probably makes a facial expression]

13 Leah: Maybe she would make that face. Do you think she's jealous because she wants the gecko? How 1 do you think she's feeling Ahmed? I want to give you another chance

14 Ahmed: Surprised

Prior to line 1, Naima made an inference about the teacher's reaction to the pet gecko running around the building. Students were asked to share their thoughts, and José and Ahmed both said ‘jealous’, indicating a misunderstanding. Ahmed was asked to explain further (line 8), and there was a long pause, signalling that he may not have been sure how to articulate his response. As such, Leah attempted to prompt the learners and Naima interjected with a facial expression. In line 14, Ahmed provided the word ‘surprised’ without further explanation when he was asked to make an inference again. The pause in response and quick agreement with another student in Ahmed's utterances implied a potential need for further targeted vocabulary support related to describing feelings.

José's difficulties with access to language during inference-making were consistently related to a desire for particular vocabulary that would help him describe his inference. During discussions about the first book, he did not have any significant difficulties, but responded with brief inferences related to characters' feelings. At times, he approximated or created words to help him describe his inferences. For example, he described the main character Annie as feeling ‘giggly’ after Mariana used the word ‘giggly’, taking up her word to describe what he thought was a humorous event.

In excerpt four, José inferred about how two brothers felt about trying to solve their problems in sharing a room.

Excerpt 4: Solving a problem, 3 February 2017

1 Leah: How do you think they feel about working it out, Jose? Pablo is the messy one and Nico is the neat one.

2 José: Nico feels different, and Pablo feels mad.

José continued to describe how the characters felt by using the pictures in the book. But, in line 2, his use of the word ‘different’ to describe how Nico felt is not specific, signalling that he may have needed support in describing Nico's feelings. This type of utterance was not surprising, as José and other students’ descriptions of feelings tended to repeat the same words such as mad, sad, frustrated and happy.

In excerpt five, the class discussed the non-fiction text about ships and boats and decided that a smaller, paddle boat could be faster than a larger, frame boat (a boat in which several people rowed the boat). José showed interest in the topic and was asked why smaller boats might go faster than a larger boat.

Excerpt 5: Boats 1, 15 February 2017

1 Leah: Why do you think a frame boat goes slower than a (smaller) faster boat?

2 José: I mean, it goes fast.

[encourages Jose to pause and think about answer]

4 José: It is fast.

5 Leah: You think a frame boat goes fast? Why do you infer that?

6 José: Because there's like a wheel and a car thingy that you (.) I forgot what you

7 Leah: =Motor?

8 José: Yeah, it has it inside and there's like, then you paddle, then you turn it on, then they drive it.

In the excerpt, José seemed confused at first, and he was encouraged by Leah to pause in line 2 to think. This utterance was coded both for a difficulty with background knowledge and language because the inference required that he understood that large frame boats did not have motors and were powered by people. The photos provided some evidence, but he needed more context. José also used the phrase ‘wheel and car thingy’ to help him describe his justification. He also communicated the desire to express a particular word by saying ‘I forgot what you’ in line 6. Then, José communicated that he forgot the word for what he was describing. After the word ‘motor’ was suggested, he agreed with the suggested word and described the process of turning on and driving a boat. Despite needing support with background knowledge, his desire to express a specific word also paused his inference-making in this moment.

Instructional segments: Students' access to background knowledge

Over the course of the lessons surrounding the three texts, access to background knowledge became most salient with the third text, a non-fiction text about ships and boats. In fact, all of the seven difficulties with background knowledge during inference-making derived from discussions about the third text. Non-fiction texts often require knowledge that goes beyond what children may experience in their daily lives (Best et al., 2008; Elbro and Buch-Iversen, 2013; Kendeou et al., 2016), which seemed to be the case with the third text. The third text required that students could compare and contrast features of large ships and small boats, in addition to inferring about the purposes and characteristics of these boats. The class spent most of the first lesson building background knowledge (see an example of a diagram in Appendix 4), previewing the text and discussing prior experiences with boats. Students needed further support in being able to infer and make connections with this text as their inferences required significant prompting and included brief and repetitive utterances. In the following excerpts, students' inference-making instances are presented together because the moments come from the same class period.

In excerpt six, the class discussed the text after reading about characteristics of small boats.

Excerpt 6: Boats 2, 13 February 2017

1 Leah: What did they use small boats for, what do you think? What's your inference? ‘I think.’ Mariana, what did they use small boats for?

2 Mariana: For the water(.) To push the boat. To go sell your boat.

3 Leah: Okay.

4 Mariana: I don't know. [frustrated]

5 Leah: No, I'm just thinking. I think a lot of answers could be right. Keep sharing. Jose, what do you think they used small boats for in the past?

6 José: I think they used small boats for in the past, to go somewhere like far away, that's why they used ships or boats, or boats that look like logs?

7 Leah: Okay, what did they use small boats for in the past? Naima. ‘I think.’

8 Naima: I think they used small boats for the past, so they could go through swamps. [referring to a picture of a canoe in a swamp in the book]

9 Leah: So, you all are right that they use them to take them places. But, what types of jobs did small boats help them (people in the past) do?

10 Mariana: Push the water.

11 Leah: Push the water, okay.

12 Naima: To carry people.

13 Leah: To carry people.

14 Naima: And to take them somewhere else.

15 José: To keep people safe, and to go to the restaurant.

In this excerpt, students needed to draw upon prior knowledge and/or evidence from the book in order to infer how people used small boats in the past. In line 2, Mariana's response was marked by a pause and short sentences without further explanation, signalling a need for further support. Although her utterance may have resulted from limited access to her first language or a need for further language support, she said ‘I don't know’ in line 4 with frustration and repeated the idea in line 10 after hearing other students’ ideas.

José, in line 6, made an inference that made sense given the evidence in the text and built upon discussions during previous lessons. Later, in line 15, after Naima and Mariana contributed, José responded to the prompt with the idea that boats keep people safe and that people might go to restaurants on boats. This inference probably came from real prior experience or knowledge, but it was not prompted further. Hence, given the textual evidence and the topic of boats in the past, he needed access to content knowledge in order to make an inference connected to the text.

A few turns after excerpt six, Leah provided an example from a recent social studies unit on local communities of Native Americans, as the students learned that Native Americans used small boats to fish and collect wild rice. The class continued the conversation in excerpt seven.

Excerpt 7: Boats 3, 13 February 2017

1 Mariana: I know it's somewhere, where they have [movement]. If you want to have a date with someone, people do this. Pretend this is a boat, people do this.

2 Leah: It's called a gondola.

3 Mariana: A gondola, okay and then people kiss because (.)

[students laugh and overlap in speech. We talk about gondolas for a few minutes. Leah showed a picture of a gondola on the computer]

4 Leah: I want to go back to my question. Don't you think they use boats for (other) jobs?

5 Students: Yeah

6 Leah: What do you think they used it for?

7 Mariana: To push it (.) The water (.) To drive.

8 Leah: Well that's part of it, to travel. We said that already. Okay, Mariana? I'm thinking of something else. Great ideas. Jose?

9 José: Maybe they do the jobs, because like boats and ships and the log thingys, they used to do jobs, and they used to like carry people. And they used to give them money and that's like a job.

10 Naima: They used boats like for their job like for taxis.

In line 1, Mariana drew upon prior knowledge or an experience with details in her description about people going on a date in a boat. Mariana's initial response in line 1 contrasted with the discourse when she had difficulty accessing the background knowledge necessary to infer, which usually consisted of brief responses with pauses and frustration. The goal of this part of the discussion was to guide students to infer that boats assisted with transporting goods, fishing, or getting food. In lines 4 and 6, students repeated previous responses, so they were guided to continue to think about the jobs of small boats. In line 7, Mariana had some difficulty in expressing her thoughts and repeated her previous response, ‘to push it (.) The water.’ Mariana's difficulty in accessing the information she needed to make an inference about the topic was signalled by the brevity of and pauses in her utterances.

A few days later the students and Leah continued to discuss the characteristics of boats in the text. Excerpt eight presents an instance in which we discussed why larger sail boats might have more sails, which required background knowledge relating to basic properties of movement and weight and sailboats. Students knew what a sail was and had read about how when wind hits the sail, it pushes the boat. The text prior to Leah's questioning in this instance was, ‘Larger ships needed more sails. Sailing ships carried people around the world.’ In this instance, all students probably needed further language support, but Naima seemed to have difficulty connecting to the topic.

Excerpt 8: Ships, 15 February 2017

1 Leah: Why did larger boats need more sails? I think

2 Ahmed: I think they needed more sails because they had a lot of people that wanted to go places

3 José: I think because they are not big, they have to be big to carry a lot of people so they can go somewhere

4 Leah: Okay, yep. Naima? I think.

5 Naima: I think they need to do more sails because (.)they needed more (.) the people that wanted to go on the ship wanted to get (.) money.

6 Leah: Don't think about the people on the ship unless you're thinking about how many people are on the ship. Think about why the boat itself needed more sails. Ahmed kind of talked about it.

7 Naima: Because the boat can't go on its own.

8 Leah: Do you think a bigger boat (.) Well (.) Ahmed were you going to say something?

9 Ahmed: A bigger boat would help to get everybody aboard.

In excerpt 8, Ahmed inferred that larger boats needed more sails because they carry more people (line 1). José offers his response in line 2, preceded by Leah prompting Naima to share. Naima had an idea but could not connect to the topic and evidence at hand. The pictures showed big boats with several sails, thus the expected answers should have related to the heaviness of the boat or how difficult it would be to move the boat. Instead, Naima spoke about people who wanted to get money (line 5), and eventually responded with the idea that the boat needed sails because ‘it couldn't go on its own’ (line 7). In this particular moment, Leah's knowledge of Naima's tendency to respond frequently with details pointed to a disconnect between the topic and Naima's prior knowledge. It is likely that she had some difficulty in accessing the language needed to explain her thinking. Yet, Naima's ‘off-topic’ response was mediated by Leah pointing out Ahmed's thinking, to which Naima provided a more relevant response.

Discussion and implications

This study addressed the difficulties students encountered when they were not able to access their first language and background knowledge to support inference-making with the texts. Rather than approaching students' developing skills as deficits, the current study argues that difficulties with inferences may have related to students' opportunities to access the language they need to express inferences and relevant background knowledge. The key findings of this study show that (a) difficulties with inference-making seemed to relate to students' access to specific vocabulary in order to express an inference, which may have been mediated with multilingual strategies (Martin-Beltrán, 2014), and (b) students' difficulties with inference-making during the non-fiction text were seemingly related to access to background knowledge, but also complicated by the specific content vocabulary necessary to access the text. It should also be noted that the data do not illustrate all of the ways in which students were engaged and dynamic in their learning. However, the data show students' several assets and skills. Students desired to participate and learn and were strategic (e.g. José's approximation and coinage of words (Dörnyei and Scott, 1997)). Mariana could identify her inference-making at times, and all students drew upon prior experiences or knowledge (e.g. Mariana's description of a gondola). Additionally, students drew upon textual evidence (e.g. Naima referring to pictures in a book) in their inference-making.

Generally, the language and content in each of the texts differed, which most likely afforded and constrained inferencing opportunities. It is worth noting again that the structure of the reading group included significant time devoted to vocabulary, previewing the book and reading together, which are strategies commonly used with English learners (August and Shanahan, 2006; Carlo et al., 2004). However, research shows that fictional texts tend to facilitate text-to-self connections and inference-making from personal experience, whereas non-fiction texts usually require knowledge beyond daily life (Cain and Oakhill, 1999; Elbro and Buch-Iversen, 2013; Sipe, 2000). The realistic fiction texts included school and home vocabulary, which were somewhat relatable to the experiences and knowledge of the students. The non-fiction text proved to be much more challenging, seemingly in part because of the content-specific language and background knowledge required to make these inferences (Peregoy and Boyle, 2000).

Access to language

Issues with accessing language became salient when students wanted to contribute an original experience, draw on prior knowledge, or needed additional information to make an inference. As Proctor et al. (2005) found, oral language skills and vocabulary play a significant role in reading comprehension, which potentially played a role in this study as well, especially in students' ability to articulate their thinking. Engaging in higher-order reading skills, inclusive of inference-making, requires that learners synthesize describe, summarize and connect to texts in their own words (Perfetti et al., 2005), and communicating ideas may present additional language demands when students desire to use language beyond the text. Thus, there were missed opportunities for students to express themselves, therefore limiting opportunities to learn with this particular content (Escamilla, 2015).

In the examples, Mariana was frustrated in her inference-making when she was not able to access words but desired to articulate a specific word or phrase. She asked for assistance once, but otherwise her difficulties in accessing language during inference-making were evident through her pauses and the teacher prompting and questioning. Mariana's frustration also brings to light the potential impact of subtractive schooling contexts on English learners as early as elementary school (Valenzuela, 2010), as she was not further supported in drawing on her skills in both languages. Although Ahmed, José and Naima seemed to have less difficulties overall, opportunities for multilingual meaning-making were missed and may have enhanced opportunities for inference-making and discussing the story with others.

Access to background knowledge and non-fiction texts

Familiarity with informational (non-fiction) text is important in early childhood (Yopp and Yopp, 2006), especially with the pressures on students and teachers to meet the expectations of Common Core standards in literacy. The instructional segments demonstrated that inferencing with an informational text was challenging for learners. Comprehending informational texts involves a set of specific skills, further necessitating prior knowledge and/or experience with the topic, access to the content language, and knowledge of informational text structure (Mantzicopoulos and Patrick, 2011). Furthermore, Kendeou (2015) explains that background knowledge causes the most variance in inference-making amongst learners. Yet, few studies show how these difficulties play out in the classroom with learners who are learning additional languages. Data from Table 2 and the brevity of student discourse in transcripts with the non-fiction text discussions demonstrate that perhaps students had more access to the language and content of the realistic fiction books (books one and two).

Findings showed that students needed more support in being able to infer about the jobs and purposes of boats in the third text. Because of the metropolitan location of the students' homes and school, it is likely that the students did not have much experience with bigger ships or boats used for commercial purposes. There were distinctions between moments in which students could draw on prior knowledge and times when students had difficulty explaining their inferences. In excerpt 6, Mariana was able to explain prior knowledge of a gondola and then, a few turns later, she had difficulty providing an additional response to the question, which included brief, single-word responses. The data suggest consideration of additional strategies to facilitate inference-making and students' opportunities to learn (Escamilla, 2015), such as including familiar and culturally relevant content, providing engaging texts, activating background knowledge, and integrating first language as students are developing higher-order reading skills (Bowyer-Crane and Snowling, 2005; Cummins, 2015; De Jong and Harper, 2005; Ebe, 2010, 2012; Snow et al., 1998). Although previous research illustrates the importance of story structure and attention to textual elements (Perfetti et al., 2005), further research should interrogate the characteristics of inference-making across genres with younger English learners in order to make claims about inferencing with fiction and non-fiction texts.

Implications for research and practice

The data represent a small sample of instances when students were not able to access their first language and background knowledge to support inference-making, but findings provide implications for both research and practice. First, research has called for examining culturally sustaining approaches in relation to reading comprehension (Fairbanks et al., 2017), as few recent studies have investigated how creating multilingual learning environments and using culturally sustaining texts affect higher-order reading comprehension with younger emergent multilinguals. Future research should explore how culturally and linguistically sustaining approaches (see García & Wei, 2014; Machado, 2017; Paris, 2012; Paris and Alim, 2014 for further reading on culturally and linguistically sustaining frameworks and pedagogies) and materials influence younger multilinguals' higher-order reading comprehension. Future studies should also consider district curriculum, materials, resources and professional development as affordances and constraints in doing this work.

Second, examining the classroom discussions surrounding inference-making in this study provided insights into the complexities of learning how to read and infer in group settings. The discourse in lessons illustrated students' learning but also the potential impact of instructional choices (Rex and Green, 2008). Inference-making itself is a very contextual process, because it requires synthesis of textual evidence and prior knowledge or experience to draw conclusions beyond the text (Kendeou et al., 2016). Thus, prior research and results from this study illuminate the need for further research on inference-making, with continued consideration of instructional affordances and constraints and students' backgrounds. With the breadth of research that has documented the characteristics of inference-making of young English-speaking monolinguals, further research from cognitive and sociocultural perspectives should account for multilingualism to determine whether results from prior studies also apply to language learners.

On reflection, this study approached a widely studied cognitive process through a qualitative lens, focusing on how emergent multilinguals inferred during specific instructional segments in a classroom setting. This approach makes transparent the affordances and constraints involved in studying small-group learning, such as instruction, time and social dynamics. However, by situating reading comprehension as a sociocultural process, we can examine difficulties with inference-making in relation to students' access to first language and background knowledge. By analysing and positioning students' difficulties in this way, it highlights how educators and researchers can reflect on instruction and their classroom context. Furthermore, by reconceptualizing difficulties in terms of students' access to content, educators can seek to provide further learning opportunities that truly engage emergent multilinguals, not only in skills-based English instruction, but with culturally and linguistically sustaining materials and approaches that further encourage student dialogue.

Limitations

Limitations of this study necessitate further research on the topic of inference-making with emergent multilinguals. There was not sufficient space to acknowledge the myriad ways students' inference-making was afforded and constrained by classroom, instructional and social elements. Though difficulties with inference-making were coded as difficulties with access to language or with background knowledge, this study makes no claims for the individual cognitive processes of the students, recognizing how social interactions, individual development and instruction are entangled in learning. Hence, further work that investigates instructional interventions and teacher talk could provide important data on this topic. The study was both afforded and constrained by the small-group design of the study. As shown in Table 2, there was a significant range in the number of inferences and difficulties made by each of the students, thus this article does not give equal space to each student's inference-making. Due to social relationships and individual characteristics, some students responded more than others. This could be mediated by creating increased opportunities for partner talk, which would allow for more talk time for individual students. Despite Leah's bilingualism and close relationships with the students and families as their teacher, she identifies as White and middle-class, which differs from the labels that have been ascribed to the students, as well as their lived experiences in school. Analysis of their understanding was limited by Leah's own sociocultural lens and the missed potential that would have been gained by doing this work in the students' first language. Methodologically, the focus was on depth of analysis across and within events versus breadth. However, longitudinal work, which involves deeper researcher, student and school relationships, is necessary in order to better understand the contexts and backgrounds of learners in relation to their literacy learning and comprehension.

Footnotes

Acknowledgements

I would like to thank the anonymous reviewers and the editors of Journal of Early Childhood Literacy for their thoughtful and helpful feedback. I am also very grateful for the support of my advisors and peers who have spent much time reviewing and talking through my research with me.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Note

References

Armstrong

Gosling

Weinman

Marteau

, et al.(1997) The place of inter-rater reliability in qualitative research: An empirical study. Sociology 31(3): 597–606.

August

Carlo

Dressler

, et al.(2005) The critical role of vocabulary development for English language learners. Learning Disabilities Research and Practice 20(1): 50–57.

August

Shanahan

(2006) Developing Literacy in Second-Language Learners: Report of the National Literacy Panel on Language Minority Children and Youth, Mahwah, NJ: Lawrence Erlbaum.

Aukerman

(2013) Rereading comprehension pedagogies: Toward a dialogic teaching ethic that honors student sensemaking. Dialogic Pedagogy: An International Online Journal 1 DOI: 10.5195/dpj.2013.9.

Aukerman

Schuldt

Aiello

, et al.(2017) What meaning-making means among us: The intercomprehending of emergent bilinguals in small-group text discussions. Harvard Educational Review 87(4): 482–511.

Avalos

Plasencia

Chavez

, et al.(2007) Modified guided reading: Gateway to English as a second language and literacy learning. The Reading Teacher 61(4): 318–329.

Best

Floyd

Mcnamara

(2008) Differential competencies contributing to children's comprehension of narrative and expository texts. Reading Psychology 29(2): 137–164.

Boals

Castro

Willner

(2018) Moving beyond assumptions of cultural neutrality to improve accessibility and opportunity to learn for English language learners. In:. Handbook of Accessible Instruction and Testing Practices, Berlin: Springer, pp. 119–134.

Bowyer-Crane

Snowling

(2005) Assessing children's inference generation: What do tests of reading comprehension measure?. British Journal of Educational Psychology 75(2): 189–201.

10.

Brown

(1977) Knowing When, Where, and How to Remember: A Problem of Metacognition, Arlington, TX: Eric Reports.

11.

Bunch

(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.

12.

Bunch

Walqui

Pearson

(2014) Complex text and new common standards in the United States: Pedagogical implications for English learners. TESOL Quarterly 48(3): 533–559.

13.

Cain

Oakhill

(1999) Inference making ability and its relation to comprehension failure in young children. Reading and Writing 11(5–6): 489–503.

14.

Cain

Oakhill

Barnes

, et al.(2001) Comprehension skill, inference-making ability, and their relation to knowledge. Memory & Cognition 29(6): 850–859.

15.

Carlo

August

Mclaughlin

, et al.(2004) Closing the gap: Addressing the vocabulary needs of English-language learners in bilingual and mainstream classrooms. Reading Research Quarterly 39(2): 188–215.

16.

Carrell

(1983) Some issues in studying the role of schemata, or background knowledge, in second language comprehension. Reading in a Foreign Language 1(2): 81–92.

17.

Cummins

(2015) Intercultural education and academic achievement: A framework for school-based policies in multilingual schools. Intercultural Education 26(6): 455–468.

18.

De Jong

Harper

(2005) Preparing mainstream teachers for English-language learners: Is being a good teacher good enough?. Teacher Education Quarterly 32(2): 101–124.

19.

Dörnyei

Scott

(1997) Communication strategies in a second language: Definitions and taxonomies. Language Learning 47(1): 173–210.

20.

Ebe

(2010) Culturally relevant texts and reading assessment for English language learners. Reading Horizons 50(3): 193–210.

21.

Ebe

(2012) Supporting the reading development of middle school English language learners through culturally relevant texts. Reading & Writing Quarterly 28(2): 179–198.

22.

Elbro

Buch-Iversen

(2013) Activation of background knowledge for inference making: Effects on reading comprehension. Scientific Studies of Reading 17(6): 435–452.

23.

Erickson

(2006) Definition and analysis of data from videotape: Some research procedures and their rationales. In: Green

Camilli

Elmore

(eds) Handbook of Complementary Methods in Education Research, Mahwah, NJ: Erlbaum, pp. 177–192.

24.

Escamilla

(2015) Schooling begins before adolescence: The case of Manuel and limited opportunities to learn. In: Molle

Sato

Boals

(eds) Multilingual Learners and Academic Literacies: Sociocultural Contexts of Literacy Development in Adolescents, New York, NY: Routledge, pp. 210–227.

25.

Fairbanks

Cooper

Webb

, et al.(2017) Reading comprehension research and the shift toward culturally sustaining pedagogy. In: Israel

Duffy

(eds) Handbook of Research on Reading Comprehension, New York, NY: Guilford Press, pp. 459–478.

26.

Fountas

Pinnell

(1996) Guided Reading: Good First Teaching for All Children, Portsmouth, NH: Heinemann.

27.

García

(2009) Emergent bilinguals and TESOL: What's in a name?. TESOL Quarterly 43(2): 322–326.

28.

García GE (1991) Factors influencing the English reading test performance of Spanish-speaking Hispanic children. Reading Research Quarterly 26(4): 371–392.

29.

García

Wei

(2014) Translanguaging and education. In: García

Wei

(eds) Translanguaging: Language, Bilingualism and Education, Basingstoke: Palgrave Macmillan, pp. 63–77.

30.

Goldenberg

(2013) Unlocking the research on English learners: What we know-and don't yet know-about effective instruction. American Educator 37(2): 4–38.

31.

Graesser AC, Singer M, and Trabasso T (1994) Constructing inferences during narrative text comprehension. Psychological Review 101(3): 371–395.

32.

Helman

Stai

(2017) Public policy in an era of changing literacies: A focus on reading comprehension. In: Israel

(ed) Handbook of Research on Reading Comprehension, New York, NY: The Guilford Press, pp. 626–648.

33.

Jimenez

Gamez

(1996) Literature-based cognitive strategy instruction for middle school Latina/o students. Journal of Adolescent & Adult Literacy 40(2): 84–91.

34.

Kendeou

(2015) A general inference skill. In: O'Brien

Cook

Lorch

(eds) Inferences during Reading, Cambridge, MA: Cambridge University Press, pp. 160–181.

35.

Kendeou

Bohn-Gettler

White

, et al.(2008) Children's inference generation across different media. Journal of Research in Reading 31(3): 259–272.

36.

Kendeou

McMaster

Christ

(2016) Reading comprehension: Core components and processes. Policy Insights from the Behavioral and Brain Sciences 3(1): 62–69.

37.

Kendeou

Van Broek

White

, et al.(2007) Comprehension in preschool and early elementary children: Skill development and strategy interventions. In: McNamara

(ed) Reading Comprehension Strategies: Theories, Interventions, and Technologies, Mahwah, NJ: Erlbaum, pp. 27–45.

38.

Kendeou

Van den Broek

(2005) The effects of readers' misconceptions on comprehension of scientific text. Journal of Educational Psychology 97(2): 235–245.

39.

Kendeou

Van den Broek

Helder

, et al.(2014) A cognitive view of reading comprehension: Implications for reading difficulties. Learning Disabilities Research & Practice 29(1): 10–16.

40.

Kendeou

Van den Broek

White

, et al.(2009) Predicting reading comprehension in early elementary school: The independent contributions of oral language and decoding skills. Journal of Educational Psychology 101(4): 765–778.

41.

Kispal A (2008) Effective teaching of inference skills for reading. Cheshire, UK: National Foundation for Educational Research, Department of Education (Division of Children, School and Families).

42.

Lee

(2016) Letting the story out: Drawing on children's life stories and identities to help them read beyond and enhance their comprehension. Journal of Language, Identity & Education 15(6): 389–403.

43.

Luke

Woods

Dooley

(2011) Comprehension as social and intellectual practice: Rebuilding curriculum in low socioeconomic and cultural minority schools. Theory into Practice 50(2): 157–164.

44.

Machado

(2017) Fostering and sustaining diverse literacy practices in the early childhood classroom: Reviewing the literature in three areas. Literacy Research: Theory, Method, and Practice 66(1): 309–324.

45.

McKoon

Ratcliff

(1992) Inference during reading. Psychological Review 99(3): 440–466.

46.

McNamara

(2007) Reading Comprehension Strategies: Theories, Interventions, and Technologies, Mahwah, NJ: Erlbaum.

47.

Mantzicopoulos

Patrick

(2011) Reading picture books and learning science: Engaging young children with informational text. Theory into Practice 50(4): 269–276.

48.

Martin-Beltrán

(2014) “What do you want to say?” How adolescents use translanguaging to expand learning opportunities. International Multilingual Research Journal 8(3): 208–230.

49.

Martin-Beltrán

Daniel

Peercy

, et al.(2017) Developing a zone of relevance: Emergent bilinguals' use of social, linguistic, and cognitive support in peer-led literacy discussions. International Multilingual Research Journal 11(3): 152–166.

50.

Melby-Lervåg

Lervåg

(2014) Reading comprehension and its underlying components in second-language learners: A meta-analysis of studies comparing first and second-language learners. Psychological Bulletin 140(2): 409–433.

51.

Miles

Huberman

Saldana

(2013) Qualitative Data Analysis, London: SAGE.

52.

Morse

(1997) “Perfectly Healthy, but Dead”: The myth of inter-rater reliability. Qualitative Health Research 7(4): 445–447.

53.

Oakhill

Cain

(2007) Issues of causality in children's reading comprehension. In: McNamara

(ed) Reading Comprehension Strategies: Theories, Interventions and Technologies, Mahwah, NJ: Erlbaum, pp. 47–71.

54.

Oakhill

Cain

(2012) The precursors of reading ability in young readers: Evidence from a four-year longitudinal study. Scientific Studies of Reading 16(2): 91–121.

55.

Paris

(2012) Culturally sustaining pedagogy: A needed change in stance, terminology, and practice. Educational Researcher 41(3): 93–97.

56.

Paris

Alim

(2014) What are we seeking to sustain through culturally sustaining pedagogy? A loving critique forward. Harvard Educational Review 84(1): 85–100.

57.

Peregoy

Boyle

(2000) English learners reading English: What we know, what we need to know. Theory into Practice 39(4): 237–247.

58.

Perfetti

Landi

Oakhill

(2005) The acquisition of reading comprehension skill. In: Snowling

Hulme

(eds) The Science of Reading: A Handbook, Malden, MA: Blackwell Publishing, pp. 227–247.

59.

Proctor

Carlo

August

, et al.(2005) Native Spanish-speaking children reading in English: Toward a model of comprehension. Journal of Educational Psychology 97(2): 246.

60.

Purcell-Gates

(2007) Complicating the complex. In: Purcell-Gates

(ed) Cultural Practices of Literacy: Case Studies of Language, Literacy, Social Practice, and Power, Mahwah, NJ: Erlbaum, pp. 1–22.

61.

Rex

Green

(2008) Classroom discourse and interaction: Reading across the traditions. In: Spolsky

Hult

(eds) The Handbook of Educational Linguistics, Malden MA: Blackwell Publishing, pp. 571–584.

62.

Saldaña

(2015) The Coding Manual for Qualitative Researchers, Thousand Oaks, CA: SAGE.

63.

Sipe

(2000) The construction of literary understanding by first and second graders in oral response to picture storybook read-alouds. Reading Research Quarterly 35(2): 252–275.

64.

Snow

(2002) Reading for Understanding: Toward an R&D Program in Reading Comprehension, Santa Monica, CA: Rand Corporation.

65.

Snow

Burns

Griffin

(1998) Preventing Reading Difficulties in Young Children, Washington, DC: National Research Council, National Academy Press.

66.

Soto Huerta

(2012) Guiding biliteracy development: Appropriating cross-linguistic and conceptual knowledge to sustain second-language reading comprehension. Bilingual Research Journal 35(2): 179–196.

67.

Street

(1993) Cross-Cultural Approaches to Literacy. Vol. 23, Cambridge: Cambridge University Press.

68.

Swain

Brooks

Tocalli-Beller

(2002) Peer-peer dialogue as a means of second language learning. Annual Review of Applied Linguistics 22: 171–185.

69.

Trabasso

Bouchard

(2002) Teaching readers how to comprehend text strategically. In: Block

Pressley

(eds) Comprehension Instruction: Research-based Best Practices, New York, NY: Guilford Press, pp. 176–200.

70.

Valenzuela

(2010) Subtractive Schooling: US-Mexican Youth and the Politics of Caring, Albany, NY: SUNY Press.

71.

Van den Broek

(1997) Discovering the cement of the universe: The development of event comprehension from childhood to adulthood. In: Van den Broek

Bauer

Bourg

(eds) Developmental Spans in Event Comprehension and Representation: Bridging Fictional and Actual Events, Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 321–342.

72.

Vygotsky

(1978) Mind and Society: The Development of Higher Mental Processes, Cambridge, MA: Harvard Press.

73.

Yopp

(2006) Informational texts as read-alouds at school and home. Journal of Literacy Research 38(1): 37–51.

74.

Yuill

Oakhill

(1991) Children's Problems in Text Comprehension: An Experimental Investigation, Cambridge: Cambridge University Press.