View in EDS HTML Full Text PDF Full Text

The Role of Hang-Out Ties in Reading Comprehension and Small Group Interactions for Multilingual Students

Saved in:
Bibliographic Details
Title: The Role of Hang-Out Ties in Reading Comprehension and Small Group Interactions for Multilingual Students
Language: English
Authors: Andrew Weaver (ORCID 0000-0002-2278-0517), Michael J. Kieffer (ORCID 0000-0001-8000-7832), C. Patrick Proctor, Elise Cappella, Xiaoying Wu, Shaelyn Cavanaugh, Sasha Karbachinskiy
Source: Grantee Submission. 2025 60.
Peer Reviewed: Y
Page Count: 14
Publication Date: 2025
Sponsoring Agency: Institute of Education Sciences (ED)
Contract Number: R305B140037
R305A200069
Document Type: Journal Articles
Reports - Research
Education Level: Elementary Education
Grade 4
Intermediate Grades
Grade 5
Middle Schools
Descriptors: Reading Comprehension, Reading Instruction, Small Group Instruction, Interpersonal Relationship, Elementary School Students, Multilingualism, Grade 4, Grade 5
DOI: 10.1002/rrq.70053
Abstract: Small groups are commonly used in elementary reading classrooms, but relatively little research has examined the social factors that support small group reading instruction. This study tests whether the number of hang-out ties (i.e., the number of ties between students who like to hang out together) in small groups predicts reading comprehension among multilingual children in Grades 4 and 5, and whether interactions differ in small groups with high and low numbers of hang-out ties. We leverage data from a randomized controlled trial to examine whether the number of group hang-out ties predicts the later reading comprehension scores of participating multilingual students (N = 143; 46% female; 57% Asian American) using autoregressive path analysis. Furthermore, we explore differences in the patterns of interaction for groups with low (e.g., 0 to 2 ties) and high (e.g., 3 to 6) numbers of hang-out ties to see whether these differences might help explain the predictiveness of group hang-out ties on reading comprehension. Results indicated that the number of group hang-out ties negatively predicted later reading comprehension. Exploratory analyses of interaction patterns using observation data revealed that teachers asked more open-ended questions and elicited more student responses in groups with low numbers of hang-out ties. These findings suggest that small-group instruction may be most effective when students can learn from peers they know less well, as these other peers may offer novel linguistic assets, and that teachers may foster more group discussion when peers do not have existing hang-out ties.
Abstractor: As Provided
IES Funded: Yes
Entry Date: 2025
Accession Number: ED676548
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwHoTScfYlJ0l0mltM4yOo4yAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDH8XdSRl8xn4uXbCSwIBEICBm33HZizTfl6cVVSFkHwnsHv4uwZeGbpvOjpU3XVKl4ff0FgrHypNYIBavpG-PGGHrdB51zwsLTDATbaCISw-g2tWD_1Q1BOtfID1zCpbg2ejo2iOCmWR5tyLRo5IJQ4jNExZ-2qBS6newpC-okai0fUQX8WzSNbK-sRxJpx4obsLgDLdDYY7bgKh-rM9qzXpBIOML6L98rarBMfr
Text:
  Availability: 1
  Value: <anid>AN0188874259;[nrnu]01oct.25;2025Oct28.05:49;v2.2.500</anid> <title id="AN0188874259-1">The Role of Hang‐Out Ties in Reading Comprehension and Small Group Interactions for Multilingual Students </title> <p>Small groups are commonly used in elementary reading classrooms, but relatively little research has examined the social factors that support small group reading instruction. This study tests whether the number of hang‐out ties (i.e., the number of ties between students who like to hang out together) in small groups predicts reading comprehension among multilingual children in Grades 4 and 5, and whether interactions differ in small groups with high and low numbers of hang‐out ties. We leverage data from a randomized controlled trial to examine whether the number of group hang‐out ties predicts the later reading comprehension scores of participating multilingual students (N = 143; 46% female; 57% Asian American) using autoregressive path analysis. Furthermore, we explore differences in the patterns of interaction for groups with low (e.g., 0 to 2 ties) and high (e.g., 3 to 6) numbers of hang‐out ties to see whether these differences might help explain the predictiveness of group hang‐out ties on reading comprehension. Results indicated that the number of group hang‐out ties negatively predicted later reading comprehension. Exploratory analyses of interaction patterns using observation data revealed that teachers asked more open‐ended questions and elicited more student responses in groups with low numbers of hang‐out ties. These findings suggest that small‐group instruction may be most effective when students can learn from peers they know less well, as these other peers may offer novel linguistic assets, and that teachers may foster more group discussion when peers do not have existing hang‐out ties.</p> <p>Social factors like relationships between students are underexamined in research on small group literacy instruction. We find that the number of hang‐out ties between multilingual students in small groups negatively predicts their later reading comprehension.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/NRNU/01oct25/rrq70053-toc-0001.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="rrq70053-toc-0001.jpg" title="." /> </p> <p></p> <p>Multilingual students are a growing population in the United States (Takanishi and Menestrel [<reflink idref="bib50" id="ref1">50</reflink>]) who bring their diverse linguistic repertoires to the classroom when learning language and literacy. Many contextual factors shape multilingual students' learning. Ecological models of language development together with research on youth social capital suggest that social relationships are one contextual factor that shapes language and literacy learning, as these relationships allow peers to share their linguistic resources (Carhill‐Poza and Kurata [<reflink idref="bib8" id="ref2">8</reflink>]; Kramsch [<reflink idref="bib30" id="ref3">30</reflink>]; Raffo and Reeves [<reflink idref="bib46" id="ref4">46</reflink>]; van Lier [<reflink idref="bib52" id="ref5">52</reflink>]). Theories of literacy development that stress the importance of children's sociocultural context suggest that social relationships matter for how students learn to comprehend texts (RAND Reading Study Group [<reflink idref="bib47" id="ref6">47</reflink>]). In the elementary classroom, small groups are commonly used for delivering tailored instruction targeting reading skills (Foorman and Torgesen [<reflink idref="bib18" id="ref7">18</reflink>]; Hall and Burns [<reflink idref="bib25" id="ref8">25</reflink>]). While research on small group instruction has long examined important questions about whether to group students with similar or different skill proficiency levels (e.g., Lou et al. [<reflink idref="bib33" id="ref9">33</reflink>]), research has only recently examined how contextual factors like social relationships between students might shape small group instruction (e.g., Chen et al. [<reflink idref="bib9" id="ref10">9</reflink>]). Furthermore, to our knowledge, research has not examined how groups with different levels of existing social relationships among students differ in small group learning and interactions.</p> <p>Thus, research is needed to determine whether social relationships between multilingual students predict small group learning, and to explore what aspects of small group interactions might explain this predictive role. In the current study, we follow social network research (e.g., Borgatti et al. [<reflink idref="bib4" id="ref11">4</reflink>]) in calling the different social relationships students have with each other "social ties," and examine these ties between students who like to hang out together ("hang‐out ties," a form of friendship, e.g., Cappella et al. [<reflink idref="bib6" id="ref12">6</reflink>]). Hang‐out ties are a tie between a student and other students that the focal student likes to be around and have fun with (e.g., Hwang et al. [<reflink idref="bib26" id="ref13">26</reflink>]), rather than just groups of students identified as spending time together. These ties contrast with other social ties like classwork advice‐seeking ties, or ties between a focal student and other students identified as experts to seek out for advice on classwork (e.g., Cooc and Kim [<reflink idref="bib12" id="ref14">12</reflink>]). Empirical work suggests that forms of friendship like hang‐out ties positively predict academic achievement (Wentzel et al. [<reflink idref="bib53" id="ref15">53</reflink>]), indicating that they may be particularly important to examine in the context of small group instruction. Yet if students participate in small groups with students with whom they share fewer social ties, they may obtain access to new linguistic resources, indicating that small groups with more existing hang‐out ties may negatively predict reading comprehension.</p> <p>If hang‐out ties do matter within small groups, especially small groups focused on language and literacy learning, it is likely that they would shape the interactions between students in small groups. In this study, we use the term interaction to encompass aspects of communication between members of a classroom (i.e., students and teachers), including discourse more narrowly as the language spoken between members of a classroom. Research on small groups and literacy development has found the most productive approaches allow for students to hold the floor for extended periods and feature teachers prompting with open‐ended questions (Soter et al. [<reflink idref="bib49" id="ref16">49</reflink>]). Thus, small group interactions may differ between groups with low and high amounts of hang‐out ties, which may explain the predictive role of small group hang‐out ties in reading comprehension.</p> <p>The current study examines whether the number of hang‐out ties in small‐group literacy instruction predicts later reading comprehension scores for participating multilingual elementary students. Further, this study explores whether and how groups with low and high numbers of hang‐out ties differ in their interactions during small group instruction to see whether these differences help to explain the predictive role of hang‐out ties. We use data from a randomized controlled trial examining the effects of grouping students identified as English learners (Kieffer et al. [<reflink idref="bib28" id="ref17">28</reflink>]) in homogeneous versus heterogeneous groups, with a sample of participating multilingual students as the sample of interest (<emph>N</emph> = 143). In this study, we focus on "multilingual students," defined here as a student whose parent or guardian reported using a language other than English at home. Using autoregressive path models, we test whether the number of hang‐out ties significantly and meaningfully predicts later reading comprehension while controlling for prior performance and demographic covariates. Next, we examine differences between groups with low and high levels of hang‐out ties across various dimensions of teacher and student discursive interactions, using observation data from the implementation of the small groups. This exploratory analysis attempts to unpack the potential predictive role of hang‐out ties in small group reading comprehension learning.</p> <hd id="AN0188874259-3">Theoretical Framing: Ecological Models, Social Capital, and Reading</hd> <p>This study draws on the intersection of the ecological model of language (Kramsch [<reflink idref="bib30" id="ref18">30</reflink>]; van Lier [<reflink idref="bib52" id="ref19">52</reflink>]), social capital theory (Bourdieu [<reflink idref="bib5" id="ref20">5</reflink>]; Carhill‐Poza and Kurata [<reflink idref="bib8" id="ref21">8</reflink>]), and multidimensional theories of reading comprehension (Duke and Cartwright [<reflink idref="bib15" id="ref22">15</reflink>], [<reflink idref="bib16" id="ref23">16</reflink>]; RAND Reading Study Group [<reflink idref="bib47" id="ref24">47</reflink>]) to understand why social ties during small group instruction might relate to the reading comprehension of multilingual students. Language development is an interactive process embedded in social contexts and is not just a matter of internal cognition (Kramsch [<reflink idref="bib30" id="ref25">30</reflink>]).</p> <hd id="AN0188874259-4">Ecological Models</hd> <p>The ecological approach to language learning was developed to move past a narrow focus on linguistic input and instead examine how every context offers multiple layers of affordances for language learning (van Lier [<reflink idref="bib52" id="ref26">52</reflink>]). Affordances are best understood as a relationship between a learner and their environment, which either creates opportunities for or inhibits action. When students in a classroom perceive the linguistic affordances of the context, they engage in linguistic action that shapes their language development (Carhill‐Poza and Kurata [<reflink idref="bib8" id="ref27">8</reflink>]; van Lier [<reflink idref="bib52" id="ref28">52</reflink>]). In other words, the relationship between a student and their peers and teacher within their ecological context structures their opportunities for language learning.</p> <p>The ecological view of language learning has implications both for understanding why small group instruction supports language and literacy learning, and why existing hang‐out ties might predict the language and literacy learning of small groups. When teachers assign small groups for instruction, they create a new layer of affordances for language learning, but one that hinges on existing social ties or the lack thereof.</p> <hd id="AN0188874259-5">Social Capital Theory</hd> <p>Related to the ecological model of language development, social capital theory informs this conception of the linguistic affordances offered by peer relationships for multilingual students in small groups. Social capital theory describes how membership in certain networks (e.g., club membership) affords actual and potential resources through social bonds (Bourdieu [<reflink idref="bib5" id="ref29">5</reflink>]). In education, social capital theory provides a case for examining social ties between students, as these membership networks (i.e., students in the same class or school) afford potential social capital given the diversity of life experiences students bring to the classroom.</p> <p>Researchers studying multilingual students have considered the linguistic repertoires of each student to be forms of social capital that can be leveraged by other students to bolster their own language and literacy learning (Carhill‐Poza and Kurata [<reflink idref="bib8" id="ref30">8</reflink>]). Social capital theorists focusing on education have argued that there is a need to distinguish between accessible (but unused) social capital versus actively used social capital (Lee [<reflink idref="bib31" id="ref31">31</reflink>]). Thus, in small groups there is an exchange of linguistic social capital via student interaction, which might change depending on existing social ties between students. Small groups with students who like to hang out together may be more participatory and more willing to exchange their ideas and linguistic social capital. Yet it could also be the case that small groups without these existing ties benefit students more by mobilizing previously untapped social capital between peers who do not hang out together. In other words, while it seems like hang‐out ties should positively support group participation and learning outcomes, they may in fact negatively predict reading comprehension, to the extent that they may prevent students from interacting with new peers and accessing new forms of language.</p> <hd id="AN0188874259-6">Multidimensional Reading Comprehension</hd> <p>This study also draws on multidimensional theories of reading comprehension to link multilingual students' social ties with their literacy development (Duke and Cartwright [<reflink idref="bib15" id="ref32">15</reflink>], [<reflink idref="bib16" id="ref33">16</reflink>]; RAND Reading Study Group [<reflink idref="bib47" id="ref34">47</reflink>]). Classic reading comprehension models center on individual cognitive skills like word reading and language comprehension (e.g., Gough and Tunmer [<reflink idref="bib23" id="ref35">23</reflink>]). Additionally, language comprehension has been found to be a powerful predictor of reading comprehension for multilingual students reading in a second language (e.g., August and Shanahan [<reflink idref="bib2" id="ref36">2</reflink>]; Droop and Verhoeven [<reflink idref="bib14" id="ref37">14</reflink>]; Proctor et al. [<reflink idref="bib42" id="ref38">42</reflink>]). Newer theories embed these cognitive skills within a wider sociocultural context (Duke and Cartwright [<reflink idref="bib15" id="ref39">15</reflink>], [<reflink idref="bib16" id="ref40">16</reflink>]; RAND Reading Study Group [<reflink idref="bib47" id="ref41">47</reflink>]). In these newer conceptions, successful reading still requires important cognitive skills but also depends on social factors including the activity context. For example, the Deploying Reading in Varied Environments (Duke and Cartwright [<reflink idref="bib15" id="ref42">15</reflink>]) model of reading provides support for the study of social factors and their relations with literacy development. This model takes driving as a metaphor to explain how various individual and social processes are coordinated to engage in successful reading. This model compares reading conditions to weather conditions, which can shape the driving experience much as a distracted and noisy environment can influence the reading experience. Importantly, just as these pieces of the driving process are not discrete but interact, so too does the reading process feature interactions among these various individual and social processes. Thus, social processes like peer relationships may be one factor that shapes literacy development.</p> <hd id="AN0188874259-7">Why Hang‐Out Ties Might Matter for Small Group Learning</hd> <p>Small groups are a commonly used and recommended practice for advancing the literacy development of all students and particularly multilingual students in elementary school (Baker et al. [<reflink idref="bib3" id="ref43">3</reflink>]; Foorman and Torgesen [<reflink idref="bib18" id="ref44">18</reflink>]). Meta‐analytic evidence on small‐group reading interventions finds positive overall effects for their use (Hall and Burns [<reflink idref="bib25" id="ref45">25</reflink>]), and small group literacy curricula designed for multilingual students have been found to improve their reading comprehension (Baker et al. [<reflink idref="bib3" id="ref46">3</reflink>]). In addition to the design of the instruction itself, teachers face many contextual decisions when structuring small groups. One decision covered by a wealth of existing research is whether to form groups homogeneously or heterogeneously based on students' skill proficiencies, with a small overall effect size favoring homogeneous groups in prior meta‐analytic work (Lou et al. [<reflink idref="bib33" id="ref47">33</reflink>]). However, many other contextual decisions remain to be examined by researchers.</p> <p>One such contextual decision is whether to group students who like to hang out with each other. One recent study did examine the effects of friendships in a small group as a moderator for shyness in group participation (Chen et al. [<reflink idref="bib9" id="ref48">9</reflink>]). Using exponential random graph models, this study found that shy students are less likely to speak in group discussions, but that friendships in the small group moderated the effect, such that there were no longer differences. Such a finding suggests that hang‐out ties might positively predict small groups by encouraging multilingual students to engage in rich discussions with one another. However, this work did not explicitly examine whether the number of social ties at the group level positively or negatively influenced small group learning. Prior qualitative work indicates that small groups are more successful when teachers allow students to openly share their ideas and provide productive social norms that encourage intellectual collaboration (Kraatz et al. [<reflink idref="bib29" id="ref49">29</reflink>]). In groups with many hang‐out ties, it is possible that teachers feel they need to exert more control over the groups and allow for less intellectual collaboration that supports small group learning. Thus, while it seems like hang‐out ties may hypothetically positively predict small group learning, there are reasons to believe that this relationship may be negative.</p> <p>Outside of work on small groups, research on the role of friendships in learning and the role of social ties in multilingual students' academic development suggests hang‐out ties might positively influence learning (Carhill‐Poza [<reflink idref="bib7" id="ref50">7</reflink>]; Chow et al. [<reflink idref="bib10" id="ref51">10</reflink>]; Hwang et al. [<reflink idref="bib26" id="ref52">26</reflink>]; Johnson et al. [<reflink idref="bib27" id="ref53">27</reflink>]; Wentzel et al. [<reflink idref="bib53" id="ref54">53</reflink>]). A meta‐analysis found small to moderate effect sizes for studies examining the link between self‐reported number of friendships and academic outcomes, providing evidence of a positive relationship between hang‐out ties and reading comprehension (Wentzel et al. [<reflink idref="bib53" id="ref55">53</reflink>]). However, only 4 of the 22 studies examined included reading or language‐specific outcomes, and the meta‐analysis examined friendship rather than hang‐out ties. A recent study, which used multilevel modeling to account for the clustering of students in classrooms, found that the language skills of children in Kindergarten were associated with their number of friendships, providing further evidence of a link between language skills and social ties (Chow et al. [<reflink idref="bib10" id="ref56">10</reflink>]). Other work has examined the importance of social ties for multilingual students in secondary education and after‐school settings (Carhill‐Poza [<reflink idref="bib7" id="ref57">7</reflink>]; Hwang et al. [<reflink idref="bib26" id="ref58">26</reflink>]; Johnson et al. [<reflink idref="bib27" id="ref59">27</reflink>]). This research has broadly found that social ties do support the academic development of multilingual students, but it has not looked at small groups in particular. Together, these studies indicate that social ties do matter for the academic development of multilingual students, but with much more research needed on whether hang‐out ties predict small group learning, and if this relationship is positive or negative.</p> <hd id="AN0188874259-8">Hang‐Out Ties and Small Group Interactions</hd> <p>If social ties within small groups do predict the reading comprehension of participating students, it is likely that this predictive association would happen as a function of the discursive interactions within the small group. Reading researchers have argued for the importance of classroom interactions in supporting reading comprehension development (Nystrand [<reflink idref="bib40" id="ref60">40</reflink>]). Small groups provide an important opportunity for peers to interact with one another and express their ideas while jointly making meaning of a text. Teachers may take on a more authoritative role in small groups and lead discussions; or they may use more open‐ended questions and gradually cede control to students so they can exchange ideas (Murphy et al. [<reflink idref="bib37" id="ref61">37</reflink>]). If groups with different levels of hang‐out ties do differ in their reading comprehension learning, it is likely that the patterns of interactions of those groups differ as well. A meta‐analysis of studies examining classroom interactions found that teachers' uses of open‐ended discussion questions allow for students to give more elaborated responses (Soter et al. [<reflink idref="bib49" id="ref62">49</reflink>]). Based on the results of this meta‐analysis and other research, Murphy et al. ([<reflink idref="bib36" id="ref63">36</reflink>]) developed the Quality Talk framework for leading class discussions in a way that allows teachers and students to share control and for students to engage in critical‐analytic thinking with texts. If groups have high levels of hang‐out ties, they may also have students engaging in more off‐topic talk, leading to teachers taking a more authoritative role in leading discussions. Thus, groups with low numbers of hang‐out ties may have more student‐to‐student discussion.</p> <p>Teachers may use particular evidence‐based strategies to support multilingual students, especially those identified as English learners. Indeed, Goldenberg et al. ([<reflink idref="bib22" id="ref64">22</reflink>]) developed an observation protocol to identify the use of these strategies when providing whole‐class instruction. These strategies include modeling fluent speech when reading aloud and using more common words to clarify. If teachers do use these strategies with students, it should lead to more productive conversations in small groups that foster reading comprehension development (Goldenberg [<reflink idref="bib21" id="ref65">21</reflink>]). Yet, if teachers over‐employ these strategies at the expense of fostering discussion, then students may learn less. Teachers may employ these strategies when they feel they need to exert more control over the small groups and guide discussions, which could happen when students have many existing hang‐out ties in small groups. Thus, exploring differences in interactions by the number of hang‐out ties in the group could help explain why the number of hang‐out ties either positively or negatively predicts reading comprehension.</p> <hd id="AN0188874259-9">Current Study</hd> <p>This study examines whether the number of hang‐out ties among multilingual students participating in small groups focused on language and literacy development predicts reading comprehension, and whether there are observed differences in indicators of discursive interactions between groups with high and low numbers of hang‐out ties. The sample includes multilingual students (<emph>N</emph> = 143) who participated in small groups (small group <emph>N</emph> = 40) with their teachers (teacher <emph>N</emph> = 19). This study includes fall and spring assessment data as well as observation data assessing indicators of discursive interactions. Using autoregressive path models controlling for prior reading comprehension and demographic covariates, we test whether the number of hang‐out ties within a small group predicts the outcome reading comprehension measure. Autoregressive models more rigorously test whether group hang‐out ties predict reading comprehension development as opposed to just predicting levels of reading comprehension at posttest. Then, to descriptively unpack why one group might have higher outcomes, we look at groups with low and high numbers of hang‐out ties and examine whether there are differences in the patterns of interaction for these two types of groups. This paper addresses the following research questions:</p> <p>In the context of a small group literacy curriculum implementation for multilingual students,</p> <p></p> <ulist> <item> Does the number of hang‐out ties predict reading comprehension for a given group?</item> <p></p> <item> Are there differences in the patterns of discursive interactions for groups with low and high amounts of hang‐out ties?</item> </ulist> <hd id="AN0188874259-10">Method</hd> <p></p> <hd id="AN0188874259-11">Participants</hd> <p>This study leverages data from a randomized controlled trial (RCT) examining the effectiveness of grouping students identified as English learners in heterogeneous small groups (i.e., groups of English learner and non‐English learner students) compared to homogeneous small groups (i.e., groups of only English learner students) using a language and literacy curriculum designed for multilingual classrooms (Kieffer et al. [<reflink idref="bib28" id="ref66">28</reflink>]). In contrast to the RCT's focus on English learners, this study examines a wider sample of participating multilingual students in Grades 4 and 5 to answer new and distinct research questions. This sample of multilingual students' parents or guardians reported speaking a language other than English at home to at least some extent. The sample for this paper were multilingual students (<emph>N</emph> = 143) who participated in small groups (small group <emph>N</emph> = 40) with their teachers (teacher <emph>N</emph> = 19). This sample was balanced by gender (46% female) and was predominantly Asian American (57%) and Hispanic/Latino (26%). In the sample, 59% were Chinese‐English bilingual, 24% were Spanish‐English bilingual, while the rest (less than 5% of the sample for each group) spoke Arabic, Bengali, Punjabi, Urdu, or Korean.</p> <hd id="AN0188874259-12">Study Context</hd> <p>Students in the study were participating in a small group language and literacy curriculum designed for multilingual students, called Cultivating Linguistic Awareness for Voice and Equity in Schools (CLAVES; Proctor et al. [<reflink idref="bib43" id="ref67">43</reflink>]). The curriculum is guided by four principles for multilingual literacy instruction: (<reflink idref="bib1" id="ref68">1</reflink>) language and metalinguistic awareness; (<reflink idref="bib2" id="ref69">2</reflink>) dialogic approaches to engage students; (<reflink idref="bib3" id="ref70">3</reflink>) multimodal texts and scaffolds to support comprehension and expression; and (<reflink idref="bib4" id="ref71">4</reflink>) multilingual perspectives (Proctor et al. [<reflink idref="bib44" id="ref72">44</reflink>]). Prior research has found statistically significant and practically meaningful effects of the curriculum on academic language, reading comprehension, and argumentative writing (Proctor et al. [<reflink idref="bib43" id="ref73">43</reflink>], [<reflink idref="bib44" id="ref74">44</reflink>]).</p> <p>CLAVES is particularly relevant for the questions we pose in the current study because of the nature of its instruction. Specifically, all instruction for CLAVES was delivered in small groups of four students. Per the principles described above, the primary goal of CLAVES was to promote language‐based reading instruction using dialogic approaches that promote student‐to‐student discussion about big ideas in text and the functions and roles of language in text. As a result, the CLAVES curriculum provides a context where group hang‐out ties may be particularly salient, as students may differ in their willingness to engage in group discussion depending on the number of hang‐out ties in a small group. Thus, a question emerged for the research team: Do within‐group social relationships like hang‐out ties predict the reading comprehension of students receiving the CLAVES curriculum?</p> <p>In the current study, the curriculum consisted of two units with three lesson cycles in each unit. The first two lesson cycles of a given unit had a focal text and related guiding question (e.g., "Should wolves be reintroduced to national parks?"), while the final lesson cycle consisted of a writing activity focused on a unit‐wide question (e.g., "How are humans and plants and animals interdependent?"). In the first two lesson cycles, lessons include three types: (<reflink idref="bib1" id="ref75">1</reflink>) focal text reading, (<reflink idref="bib2" id="ref76">2</reflink>) lessons with supplementary multimodal text and extended work on different linguistic structures (i.e., vocabulary, morphology, and syntax), and (<reflink idref="bib3" id="ref77">3</reflink>) small group discussions where students discussed the lesson cycle's focal question. This paper leverages observations of the dialogic reasoning lessons for the second research question, as these are the lessons that most feature student talk and would be most sensitive to differences in interactions related to the number of group hang‐out ties.</p> <p>Interested schools and teachers were recruited in the summer of 2022. Students were recruited in the fall of 2022 to participate in the RCT study by sending consent forms home at the beginning of the year. Students in the RCT's sample were either English learner students who were not at the first two levels of English proficiency on a state assessment, or non‐English learner students. Thus, groups were either homogeneous (i.e., four English learner students) or heterogeneous (i.e., two English learner students and two non‐English learner students).</p> <p>The RCT and the current study differ in their purposes, research questions, key variables, and analytic samples. The RCT examined whether homogeneous or heterogeneous groups were more effective for English learners' reading comprehension, argumentative writing, and language comprehension (Kieffer et al. [<reflink idref="bib28" id="ref78">28</reflink>]). The RCT further asked whether there are moderating effects of English language skills on the effects of homogeneous and heterogeneous groups for English learners, and differences in instruction between the two types of groups. The RCT did not include group‐level hang‐out ties as a primary or secondary question and does not control for them in any of the analytic models. By contrast, this paper asks if group‐hang out ties predict reading comprehension, and whether there are differences in groups with low and high amounts of hang‐out ties. While the RCT focused on the differences between these homogeneous and heterogeneous groups for students identified as English learners (RCT analytic sample <emph>N</emph> = 84), the current study focuses instead on the sample of multilingual students, regardless of their English learner status and of grouping, who participated in the small groups (<emph>N</emph> = 143). The RCT did not include group hang‐out ties as a primary or secondary question and did not control for them in any analytic model. Thus, the RCT and this paper have distinct purposes, research questions, key variables, and analytic samples.</p> <p>Consented students then provided assent and completed a pretest battery of language and literacy assessments. Based on demographic background and pretest data, English learner students were randomized to participate in the small groups in either homogeneous or heterogeneous small groups. In the RCT, there was no main effect of grouping on reading comprehension (Kieffer et al. [<reflink idref="bib28" id="ref79">28</reflink>]), indicating that grouping should not confound results for the first research question. During implementation, research assistants completed an observational protocol to evaluate group interactions within the small groups on three occasions during one of the first two lesson cycles of a given unit (see below). These three observations captured the three types of lessons in a lesson cycle detailed above. At the end of implementation, students completed a posttest battery of language and literacy assessments. This study uses pretest, posttest, and observation data.</p> <hd id="AN0188874259-13">Measures</hd> <p>Measures for the study come from the pretest battery, the small‐group observation data collected during study implementation, and the posttest battery.</p> <hd id="AN0188874259-14">Reading Comprehension (Pretest and Posttest)</hd> <p>Reading comprehension was measured using the <emph>ReadBasix</emph> assessment program developed by researchers at the Educational Testing Service (Sabatini et al. [<reflink idref="bib48" id="ref80">48</reflink>]). <emph>ReadBasix</emph> is a web‐administered reading skills components battery designed with vertical scales for children in Grades 3–12. The reading comprehension subtest of <emph>ReadBasix</emph> presents an informational text with literal and inferential multiple‐choice questions. There are 4 passages with 7 questions for each passage, and students have 7 min and 30 s for each passage. The technical manual for <emph>ReadBasix</emph> reports an acceptable IRT marginal reliability of 0.753 for Grade 4 and 0.674 for Grade 5 of the norming sample (Sabatini et al. [<reflink idref="bib48" id="ref81">48</reflink>]).</p> <hd id="AN0188874259-15">Social Networks (Pretest)</hd> <p>Student hang‐out networks were ascertained at pretest using a social network survey. See Supporting Information for a demonstration version of this survey. Students saw a roster of consented students from their class and were asked to nominate peers who they "like to hang out with." In spoken directions, we instructed students to "think of the kids who you like to be around and have fun with" and that these ties could "be your friends or other kids you like to spend time with." Thus, these ties were similar to affiliative hang‐out groups (e.g., Neal [<reflink idref="bib39" id="ref82">39</reflink>]), but self‐reported. We were only allowed to list consented students on the class roster, per the partner district's IRB. Then, they nominated additional peers who were not listed on the consented roster by only writing the first letter of their first name. Next, for each student nominated, students indicated what languages they used when talking with that person with possible answers including "All English," "Mostly English," "Equal English and another language," "Mostly another language," and "All another language." For this study, we calculated the total number of hang‐out ties within a small group, which could range from 0 (no connections between students) to 6 (all possible connections between students). Both students did not have to nominate the tie for it to be considered.</p> <hd id="AN0188874259-16">Patterns of Discursive Interactions</hd> <p>During the implementation of CLAVES, we observed each small group on three occasions, with each observation capturing one of the three lesson types detailed above. We developed an observation protocol to evaluate the patterns of discursive interaction during implementation. The protocol draws from an existing observation scale focused on instruction for multilingual learners (Goldenberg et al. [<reflink idref="bib22" id="ref83">22</reflink>]) and a framework for conducting classroom observations (Murphy et al. [<reflink idref="bib36" id="ref84">36</reflink>]) with the aim of capturing teacher and student moves that are rooted in the principles of CLAVES as well as past research. Six research assistants were trained to use the protocol and were required to achieve 80% interrater reliability with key video examples from a prior study on the curriculum before conducting live observations in the classroom, in alignment with common research practices (e.g., Wilhelm et al. [<reflink idref="bib54" id="ref85">54</reflink>]). We also conducted a mid‐implementation check to safeguard against rater drift; again, all research assistants achieved 80% interrater reliability. We were prevented from having multiple observers in the classroom (to estimate real‐time interrater reliability) due to COVID‐related IRB restrictions from our district partner. We did not randomly assign raters to observe, but nor did we assign them systematically based on any characteristics of the group. Raters were instead assigned based on logistics and availability, a potential limitation. Results from confirmatory factor analyses supported six domains: (<reflink idref="bib1" id="ref86">1</reflink>) strategies to adapt instruction for multilingual students, (<reflink idref="bib2" id="ref87">2</reflink>) opportunities for interaction and language production in English, (<reflink idref="bib3" id="ref88">3</reflink>) primary language support, (<reflink idref="bib4" id="ref89">4</reflink>) thinking questions, (<reflink idref="bib5" id="ref90">5</reflink>) connection questions, and (<reflink idref="bib6" id="ref91">6</reflink>) student responses. Each model initially included all indicators and the prominence rating, but indicators were removed to improve fit. Two factors ("Strategies to adapt instruction for emergent multilingual students," and "thinking questions") did not achieve a benchmark of adequate for at least two indices examined. Rather than exclude specific dimensions, and thereby reduce construct coverage, we present results for these two factors with caution. See the Supporting Information for more information on the development and validity of this observation protocol, as well as a demonstration version of the protocol.</p> <hd id="AN0188874259-17">Controls</hd> <p>Models presented in this paper control for grade level, gender, home language background, and English learner status. We sent a background questionnaire to parents and guardians during the consent process that asked for grade level, gender, and home language background. For English learner status, we sent a brief survey to teachers asking them to identify which of their students were currently identified as English learners.</p> <hd id="AN0188874259-18">Analytic Plan</hd> <p></p> <hd id="AN0188874259-19">Missing Data</hd> <p>There were 7 (5%) students missing on reading comprehension, 5 students (3.5%) missing home language information, and 4 (3%) students missing gender information. These students were all included in the analytic sample because they completed at least some of the target measures. We used full information maximum likelihood in Mplus, as this approach is appropriate for data missing at random and operates as well as other suggested approaches (Enders and Bandalos [<reflink idref="bib17" id="ref92">17</reflink>]).</p> <hd id="AN0188874259-20">Model Building Process for First Research Question</hd> <p>We conducted an autoregressive path analysis using Mplus 8 (Muthén and Muthén [<reflink idref="bib38" id="ref93">38</reflink>]) to address the first research question about the extent to which the number of hang‐out ties in small groups predicts reading comprehension. Each model used the maximum likelihood estimator with robust standard errors, which is robust to issues of non‐normality and non‐independence of observations (Muthén and Muthén [<reflink idref="bib38" id="ref94">38</reflink>]). Each model also used clustered standard errors to account for the nesting of the students within groups (McNeish et al. [<reflink idref="bib34" id="ref95">34</reflink>]). Initial data cleaning was completed using R (R Core Team [<reflink idref="bib45" id="ref96">45</reflink>]) and RStudio (Posit team [<reflink idref="bib41" id="ref97">41</reflink>]).</p> <p>All models were autoregressive and predicted the posttest reading comprehension measure using pretest measures collected prior to the implementation of the small‐group curriculum. The baseline control model only included paths from the autoregressor and the demographic controls. Figure 1 displays this hypothesized path model. The final model included the path for the group number of hang‐out ties in addition to the paths for the autoregressor and the demographic controls. Figure 2 displays this hypothesized path model, with <emph>ρ</emph><subs>1</subs> the path of interest. We evaluated <emph>ρ</emph><subs>1</subs> in two ways. First, we conducted a likelihood ratio test (LRT) comparing the final model to the baseline control model. For this comparison, we used the Satorra‐Bentler chi‐squared difference test with robust maximum likelihood estimation. If we found that the second step model with <emph>ρ</emph><subs>1</subs> fitted significantly better than the first step, this provided evidence that this path was a meaningful predictor of spring reading comprehension. Next, we evaluated the significance test associated with the <emph>z</emph>‐statistic for <emph>ρ</emph><subs>1</subs>, as well as the magnitude of the standardized coefficient, to see if it was a significant and meaningful predictor of spring reading comprehension.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/NRNU/01oct25/rrq70053-fig-0001.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="rrq70053-fig-0001.jpg" title="1 Hypothesized baseline control model with autoregressor and demographic covariates predicting reading comprehension." /> </p> <p></p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/NRNU/01oct25/rrq70053-fig-0002.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="rrq70053-fig-0002.jpg" title="2 Hypothesized final model with hang‐out ties predicting spring reading comprehension, controlling for autoregressor and demographic covariates." /> </p> <p></p> <hd id="AN0188874259-23">Descriptive Analysis for Second Research Question</hd> <p>To address the second research question, we estimated differences in effect sizes between high and low groups for the six factors from the observation protocol. For these analyses We defined a "low hang‐out ties" group as having 0 to 2 total ties (group <emph>N</emph> = 22), and "high" as having 3 to 6 (group <emph>N</emph> = 18). We used this dichotomous difference of high and low groups because of the ease of between‐group effect size interpretation. We did not conduct statistical significance tests because of statistical power concerns with only 40 groups. For these comparisons we only looked at discussion lessons, given these lessons would feature more student talk and thus are more sensitive to differences in group number of hang‐out ties. We did examine whether observations of the same groups across lesson types could be combined as a composite or factor, but these different observations did not hang together. We calculated Hedges' <emph>g</emph> effect size differences between groups with low and high numbers of hang‐out ties. We used Hedges' <emph>g</emph> to correct for small sample bias and interpreted the magnitude of these effect sizes according to common benchmarks (Cohen [<reflink idref="bib11" id="ref98">11</reflink>]). In the RCT, there were differences in the observation factors by grouping type (heterogeneous or homogeneous; Kieffer et al. [<reflink idref="bib28" id="ref99">28</reflink>]). However, we conducted a chi‐square test that revealed no significant association between grouping type and the high/low hang‐out tie groups (<emph>χ</emph><sups>2</sups> = 0.21, df = 1, <emph>p</emph> = 0.65), indicating that grouping should not confound the results for this paper.</p> <hd id="AN0188874259-24">Results</hd> <p></p> <hd id="AN0188874259-25">Sample Descriptives</hd> <p>Table 1 displays the sample descriptives on the pretest and posttest variables used. For group hang‐out ties, the mean number of ties in a group was 2.3 hang‐out ties, while one hang‐out tie was the mode (30%). Table 2 displays the correlations between fall reading comprehension, spring reading comprehension, and the group number of hang‐out ties. Group hang‐out ties were not significantly correlated with either fall or spring reading comprehension (both <emph>p</emph>s > 0.05).</p> <p>1 TABLE Sample descriptives (N  = 143).</p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left">Continuous variables</th><th align="center"><italic>M</italic> (SD)</th></tr></thead><tbody valign="top"><tr><td align="left">Fall reading comprehension (Z‐score)</td><td align="center">0.47 (0.97)</td></tr><tr><td align="left">Spring reading comprehension (Z‐score)</td><td align="center">0.55 (1.03)</td></tr><tr><td align="left">Group hang‐out ties</td><td align="center">2.34 (1.5)</td></tr></tbody></table> </ephtml> </p> <p></p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left">Demographic covariates</th><th align="center"><italic>N</italic> (%)</th></tr></thead><tbody valign="top"><tr><td align="left">Female</td><td align="center">66 (46%)</td></tr><tr><td align="left">Grade 5</td><td align="center">56 (39%)</td></tr><tr><td align="left">English learner</td><td align="center">102 (71%)</td></tr><tr><td align="left">Home language: Spanish</td><td align="center">35 (25%)</td></tr></tbody></table> </ephtml> </p> <p>2 TABLE Sample correlations (N  = 143).</p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left" /><th align="center">Fall reading comprehension</th><th align="center">Spring reading comprehension</th><th align="center">Group hang‐out ties</th></tr></thead><tbody valign="top"><tr><td align="left">Fall reading comprehension</td><td align="center">—</td><td align="center" /><td align="center" /></tr><tr><td align="left">Spring reading comprehension</td><td align="center">0.75<xref ref-type="fn" rid="tfn1" /></td><td align="center">—</td><td align="center" /></tr><tr><td align="left">Group hang‐out ties</td><td align="center">−0.03</td><td align="center">−0.16</td><td align="center">—</td></tr></tbody></table> </ephtml> </p> <p>1 *** <emph>p</emph> < 0.001.</p> <hd id="AN0188874259-26">Do Small Group Hang‐Out Ties Predict Reading Comprehension?</hd> <p>Autoregressive path analyses revealed that the number of group hang‐out ties negatively predicted posttest reading comprehension. Table 3 displays the standardized path coefficients of all predictors and covariates, as well as the fit indices for these models. In the final model with paths for the demographic covariates and the autoregressor, the group number of hang‐out ties was a significant and negative predictor of spring reading comprehension (standardized path coefficient = −0.136; <emph>p</emph> = 0.041), with a medium effect size according to common benchmarks (Cohen [<reflink idref="bib11" id="ref100">11</reflink>]). Figure 3 displays this path model with standardized path coefficients. Finally, the Sartorra–Bentler chi‐squared test results comparing this final model with the baseline control model, which removed the key path from group hang‐out ties to spring reading comprehension, revealed that including the path significantly improved the model fit (Sartorra–Bentler <emph>χ</emph><sups>2</sups> = 4.69; df = 1; <emph>p</emph> = 0.023).</p> <p>3 TABLE Standardized path coefficients and fit indices for main models.</p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left">Predictor</th><th align="center">Standardized path coefficients</th></tr><tr><th align="center">Baseline control model</th><th align="center">Final model</th></tr></thead><tbody valign="top"><tr><td align="left">Group hang‐out ties</td><td align="center">—</td><td align="center">−0.136<xref ref-type="fn" rid="tfn2" /></td></tr><tr><td align="left">Fall reading comprehension</td><td align="center">0.708<xref ref-type="fn" rid="tfn3" /></td><td align="center">0.713<xref ref-type="fn" rid="tfn3" /></td></tr><tr><td align="left">English learner</td><td align="center">−0.057</td><td align="center">−0.057</td></tr><tr><td align="left">Home language: Spanish</td><td align="center">−0.122<xref ref-type="fn" rid="tfn2" /></td><td align="center">−0.104<xref ref-type="fn" rid="tfn2" /></td></tr><tr><td align="left">Female</td><td align="center">0.02</td><td align="center">0.01</td></tr><tr><td align="left">Grade 5</td><td align="center">−0.014</td><td align="center">−0.03</td></tr><tr><td align="left">Likelihood ratio test</td><td align="center">—</td><td align="center">4.69<xref ref-type="fn" rid="tfn2" /></td></tr><tr><td align="left">df</td><td align="center">—</td><td align="center">1</td></tr></tbody></table> </ephtml> </p> <ulist> <item>2 * <emph>p</emph> < 0.05.</item> <item>3 *** <emph>p</emph> < 0.001.</item> </ulist> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/NRNU/01oct25/rrq70053-fig-0003.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="rrq70053-fig-0003.jpg" title="3 Standardized path coefficients for the main model with hang‐out ties predicting spring reading comprehension, controlling for autoregressor and demographic covariates. *p < 0.05. ***p < 0.001." /> </p> <p></p> <p>We conducted two robustness checks by considering alternate ways of specifying hang‐out ties. Results for these robustness checks were substantively similar to the main model and are detailed in the Supporting Information (see Appendix Table S7).</p> <hd id="AN0188874259-28">Differences in Interaction Patterns for High and Low Hang‐Out Ties Groups</hd> <p>Exploratory analyses revealed that groups with high numbers of hang‐out ties had higher factor scores for factors addressing teachers' practices, while groups with low hang‐out ties had higher factor scores for teachers' questions and student responses. Figure 4 displays the Hedges' <emph>g</emph> effect sizes for these analyses, with negative effect sizes (the blue bars) favoring high hang‐out tie groups and positive effect sizes (the yellow bars) favoring low hang‐out tie groups.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/NRNU/01oct25/rrq70053-fig-0004.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="rrq70053-fig-0004.jpg" title="4 Bar graph display of Hedges' g effect sizes comparing high and low hang‐out tie groups on observation factor score." /> </p> <p></p> <p>For the first three factors, the effect sizes favor the high hang‐out tie groups, with effect sizes ranging from small to medium (Hedges' <emph>g</emph> = −0.21 to −0.47; Cohen [<reflink idref="bib11" id="ref101">11</reflink>]). These factors include indicators more focused on strategies employed by teachers for instruction of multilingual students ("Teacher uses strategies to adapt instruction for multilingual students"), for encouraging students to use the language targets of the CLAVES lessons ("The teacher/lesson provides opportunities for interaction that encourage student language production in English"), and for incorporating the primary language(s) of students into the lesson ("The teacher/lesson provides primary language [Spanish, Mandarin, etc.] support during lesson"). By contrast, the effect sizes for the final three factors favored the low hang‐out tie groups, with small effect sizes (Hedges' <emph>g</emph> = 0.11 to 0.2). These factors include indicators that are more focused on the discussion of the lessons, with indicators including the types of thinking questions asked by teachers, the types of connection questions asked by teachers, and the responses that students give, including instances where students engage in at least three consecutive turns at talk. Thus, the results indicate that in groups with lower number of hang‐out ties, students engaged in more discussion with one another, and were asked more questions that probed their thinking by teachers. These results shed light on the negative effect of group hang‐out ties on reading comprehension, as the low hang‐out tie groups may have more discussion among peers that leads to enhanced reading comprehension development.</p> <hd id="AN0188874259-30">Discussion</hd> <p>This study provides the first evidence that social ties between multilingual students predict literacy learning during small group reading instruction. We found that the number of hang‐out ties among 4th and 5th grade multilingual students within small groups negatively predicted reading comprehension in autoregressive path models that also controlled for demographic characteristics. Furthermore, descriptive analyses of observational data from the small groups during discussion‐based lessons found that groups with low numbers of hang‐out ties had more discussion questions and responses from students, while groups with hang‐out ties had more use of teacher strategies to lead the small groups. These findings suggest that small groups where students interact with peers they do not normally hang out with allow them to exchange previously untapped social capital, and that teachers may exert more control and allow for fewer student responses in groups with more existing hang‐out ties. Together, these findings indicate that social ties are one important aspect of the small group context that can sometimes shape literacy learning in unexpected ways.</p> <hd id="AN0188874259-31">Benefits of Peers Without Existing Ties</hd> <p>The first major finding of this study suggests that multilingual students benefit from instructional settings that allow them to interact with new peers. In contrast to our hypothesis, we found that the number of hang‐out ties among multilingual students in small groups negatively predicted reading comprehension, controlling for prior performance and demographic characteristics. This finding builds on the theoretically argued importance of social ties in ecological models of language learning (Kramsch [<reflink idref="bib30" id="ref102">30</reflink>]; van Lier [<reflink idref="bib52" id="ref103">52</reflink>]). We had hypothesized that the mobilization of existing social ties in small groups would allow multilingual students to engage in discussions and learn more than groups with a lower number of hang‐out ties more successfully. Instead, it appears that groups with lower numbers of hang‐out ties may allow students to mobilize untapped forms of linguistic social capital (Raffo and Reeves [<reflink idref="bib46" id="ref104">46</reflink>]).</p> <p>These results suggest that when students interact with peers they do not normally hang out with, they might exchange new ideas that allow them to learn more from small‐group instruction. Such a finding indicates that untapped social capital in the form of new social ties may allow students to find new ways of making meaning with texts that then translate to their overall reading comprehension level. This finding aligns with research on the importance of having many, diverse ties. The strength of weak ties theory argues that loose connections with different social networks can give access to new information (Alicea [<reflink idref="bib1" id="ref105">1</reflink>]; Granovetter [<reflink idref="bib24" id="ref106">24</reflink>]). Social capital theorists looking at peer networks often distinguish these weak ties as forms of "bridging" capital that provide new information across peers from different backgrounds (Lin [<reflink idref="bib32" id="ref107">32</reflink>]). Thus, in small groups with less existing hang‐out ties, it may be that building these weak ties through interaction allows students to gain access to new information, supporting their literacy learning.</p> <p>This finding slightly contrasts with work examining small groups that has found the number of friendships to be a protective factor for shy students, allowing them more comfort for participation (Chen et al. [<reflink idref="bib9" id="ref108">9</reflink>]). However, that research did not test whether friendships were a positive influence on academic outcomes of small groups, but rather found it moderated the negative effect of shyness on small group participation. While friendships may provide comfort for shy students, the current results suggest that hang‐out ties may negatively predict reading comprehension development from small groups and suggest that teachers should structure small groups to allow for students to interact with new peers. This finding also contrasts with meta‐analytic work that has found the number of friendships to positively predict academic achievement (Wentzel et al. [<reflink idref="bib53" id="ref109">53</reflink>]). However, that work did not examine hang‐out ties specifically and at the small group level. More research is needed to confirm this finding, but this preliminary evidence suggests that hang‐out ties are negatively associated with small group learning.</p> <p>These differences in findings with past research may also stem from the focus on a multilingual sample in the current study. The meta‐analysis did include several studies with racially and ethnically diverse samples, but did not look at language use in friendships (Wentzel et al. [<reflink idref="bib53" id="ref110">53</reflink>]). One of these studies used nationally representative data of Latino high school students and found a direct positive effect of the number of friendships on academic achievement (Delgado et al. [<reflink idref="bib13" id="ref111">13</reflink>]). However, this study did not examine how languages were used in these friendships. The younger students in the current study may not have opportunities to use their relationships for language learning, or may not demonstrate these positive effects until later years.</p> <p>This finding also adds to the existing literature on small group instruction that has focused on the components targeted or the question of ability grouping (Hall and Burns [<reflink idref="bib25" id="ref112">25</reflink>]; Lou et al. [<reflink idref="bib33" id="ref113">33</reflink>]). This study indicates that the number of hang‐out ties within small groups is a meaningful contextual factor that shapes small group instruction, in addition to previously identified factors like group size (Hall and Burns [<reflink idref="bib25" id="ref114">25</reflink>]). Future research should examine whether other types of social ties predict small group learning, and any other contextual factors that teachers can control when forming small groups.</p> <hd id="AN0188874259-32">Differences in Group Interactions Within High and Low Hang‐Out Groups</hd> <p>The findings from the second research question indicated that small groups with different numbers of hang‐out ties differ in the interaction patterns between students and with their teachers. We found that in small groups with high numbers of hang‐out ties (i.e., three or more ties), there were higher factor scores for factors tapping the teacher strategies employed. These factors included strategy use for multilingual instruction, use of the language targets featured in the curriculum, and strategies for promoting primary language support. These areas of the observation protocol were culled from work identifying strategies in English language arts instruction that support English learner students (Goldenberg et al. [<reflink idref="bib22" id="ref115">22</reflink>]). While these strategies certainly seem important for fostering language learning, they may not have been as useful for translating directly to improved reading comprehension. By contrast, in groups with low numbers of hang‐out ties (i.e., fewer than three ties), factor scores were higher for factors tapping the teacher's use of thinking questions, connection questions, and the responses given by students to their teacher and peers. Although analyses were exploratory, they suggest that teachers alter their behavior in groups with more hang‐out ties, possibly to exert more control over the groups.</p> <p>Recent research finds that student characteristics influence how much time their students receive small group reading instruction (Toste et al. [<reflink idref="bib51" id="ref116">51</reflink>]). Specifically, teachers in that study gave the most time in small groups to students with higher levels of both academic skills and problem behaviors. This work suggests that teachers factor student background into their decisions about small group instruction. The current study further indicates that teachers alter their behavior within small groups based on the group composition. Indeed, teachers may use more strategies in groups with high levels of hang‐out ties because they feel they maintain control over the small groups and redirect misbehavior. Qualitative work on small group instruction suggests that teachers give students less freedom to express their ideas and interact with one another when they do not have productive group norms in place (Kraatz et al. [<reflink idref="bib29" id="ref117">29</reflink>]). Further research is needed to better understand how groups with different levels of hang‐out ties differ, and how these differences relate to the outcomes of small group instruction.</p> <p>Our finding that groups with low hang‐out ties were characterized by more high‐level questions and student responses suggests these groups were providing interaction opportunities that support text comprehension. These three factors in the observation protocol (i.e., thinking questions, connection questions, and student responses) came from Murphy et al.'s ([<reflink idref="bib36" id="ref118">36</reflink>]) work developing the Quality Talk approach to small group discussions and coding of small group transcripts. In turn, this work was influenced by a meta‐analysis attempting to identify the most proximal features of high‐level class discussions around text comprehension (Soter et al. [<reflink idref="bib49" id="ref119">49</reflink>]). These aspects of group discussions are meant to highlight the high‐level text comprehension that is facilitated by extended turns‐at‐talk for students and interactions with one another. Indeed, a review of research on the connection between interactions and reading comprehension has argued that small groups where teachers probe with open‐ended questions while giving students extended periods to express their thinking and interact with one another most highly support the development of reading comprehension (Nystrand [<reflink idref="bib40" id="ref120">40</reflink>]). The current results indicate that these features of small group interactions are higher in groups with low levels of hang‐out ties, as teachers might feel they can more easily release control of the small groups to the students.</p> <hd id="AN0188874259-33">Limitations</hd> <p>Although this study does present a novel finding about student composition in literacy small groups, several limitations necessitate further research in this area. First, we were prevented from recording video observations of the small groups by the IRB of the partner district. Thus, we designed the observation protocol to gather data on the teacher and student practices within each small group. Although this protocol did produce valuable data on these constructs, it did not capture the actual discourse in the small groups. The group observation data was also susceptible to noise given the use of one observer; another limitation due to the IRB requirements of the partner district. Future research should collect full discourse data from recorded videos of the small groups. These videos would be useful for determining more specifically how the number of hang‐out ties predicts the discourse of the small groups, and in turn whether this predicts the outcomes of the curriculum.</p> <p>Furthermore, in this study we determined a hang‐out tie based on the report of one student, rather than requiring ties to be reciprocated by multiple students. This method contrasts with other methods that require multiple informants to identify hang‐out groups (e.g., Neal [<reflink idref="bib39" id="ref121">39</reflink>]), and means that some ties may be aspirational (Gifford‐Smith and Brownell [<reflink idref="bib20" id="ref122">20</reflink>]). However, others have argued that self‐reported ties are a meaningful perception (Furman [<reflink idref="bib19" id="ref123">19</reflink>]). Future research should examine the role of hang‐out ties in small group instruction using other methods of collecting network survey data. Additionally, research could examine the role of hang‐out ties between students with similar and different home‐language backgrounds. Such work could examine whether the number of cross‐linguistic ties and same‐language ties both negatively predict the outcomes of small groups, or if one type of tie has positive effects instead. Finally, our findings are limited by the group sample size (<emph>N</emph> = 40), and future research would benefit from examining a larger number of small groups to see if our findings replicate. This research could also test to see if groups of different sizes have a similarly negative relationship for hang‐out ties and literacy learning, and differences in group interactions.</p> <hd id="AN0188874259-34">Implications</hd> <p>This paper offers a concrete instructional implication as well as more general considerations for teachers and literacy researchers. Instructionally, these findings indicate that small group instruction may be more effective for students without existing hang‐out ties. Although some research has found positive effects of friendships in small groups (Chen et al. [<reflink idref="bib9" id="ref124">9</reflink>]), the current findings indicate that for multilingual students, there are benefits to structuring small groups so they have a chance to interact with new peers. From an ecological model of language learning perspective (Kramsch [<reflink idref="bib30" id="ref125">30</reflink>]; van Lier [<reflink idref="bib52" id="ref126">52</reflink>]), the current study suggests that layers of affordances are activated when students get the opportunity to interact with each other in a small group setting, meaning that new linguistic repertoires relevant for reading become available to students. In other words, our results indicate that teachers should consider the social relationships between students when creating small groups. In particular, given our findings about how group discourse differs based on high and low levels of group hang‐out ties, teachers should consider how they are facilitating discourse among students in small groups. Our findings indicate that when groups had low numbers of hang‐out ties, groups had more student‐to‐student interaction and more student responses. As a result, teachers could consider their goals for small groups when choosing students to group together. If teachers are trying to facilitate discourse among students, they should consider grouping students who do not normally work together.</p> <p>Another implication of this work is the need for causal research to examine the exact mechanism by which social ties influence literacy learning. The current results suggest that small group interactions do differ among groups with low and high amounts of hang‐out ties. Yet, given the correlational design, the current results do not indicate whether this difference is the mechanism by which social ties in small groups negatively predict reading comprehension. Furthermore, teacher practices in small groups may facilitate social tie formation among students, which could lead to reading comprehension development. However, this latter hypothesis is less supported by the current study, where social ties were surveyed prior to the implementation of the small group curriculum. More causal research on the relationship between social ties, small group interaction, and literacy learning is needed to understand the exact pathways.</p> <p>This study also suggests that reading researchers should further investigate social factors that shape small group discussion. Although some studies have examined how teacher moves can scaffold learning from small groups (Michener et al. [<reflink idref="bib35" id="ref127">35</reflink>]) and prior work has examined the question of whether to group students based on skill proficiency (Lou et al. [<reflink idref="bib33" id="ref128">33</reflink>]), relatively little work has addressed other social factors that might predict how students learn from small groups. The results of this study indicate that hang‐out ties between students are one social factor that predicts small group discussion. Other types of social ties, including students who like to work together in class, could be explored to better understand what types of social ties matter for small group learning. Further research experimentally manipulating the number of existing ties among students could help researchers better understand exactly how social ties impact small group instruction. Additional research could also examine how more broad social factors like gender, language background, race/ethnicity, and disability status impact social ties, and whether the impacts of these social factors on social ties also influence small group instruction.</p> <hd id="AN0188874259-35">Conclusion</hd> <p>Despite the widespread use of small‐group instruction in literacy classrooms, many questions remain about the contextual factors for this important instructional setting. The current study indicates that the number of hang‐out ties among multilingual students in small groups negatively predicts their reading comprehension. Exploratory analyses of observation data from the small groups indicated that groups with low numbers of hang‐out ties featured more teacher questions, more responses from students, and more student‐to‐student interaction. These results indicate that small groups with differing levels of hang‐out ties among students do differ in the interactions between participants and their teachers and suggest that these in turn may predict how much students learn from small groups. These findings also suggest that more research is needed on the social contexts of small group instruction, beyond the question of ability grouping, to better understand the best ways to design these settings for student learning.</p> <hd id="AN0188874259-36">Acknowledgments</hd> <p>The authors would like to thank Yoav Bergner and Sarah Beck for their feedback on this work. Additionally, thanks to Tyara Dabrio, Rachel Hodes, Audrey McMaster, Dee Perry, Aimee Salgado, Summer Wu, and Mayee Yeh for their essential contributions to data collection. Thanks also to Hsiaolin Hsieh for her help with assessment and study implementation. Finally, whole‐hearted thanks to the district's leadership, teachers, and students who made this work possible.</p> <hd id="AN0188874259-37">Disclosure</hd> <p>We complied with all APA ethical guidelines in the conduct of this research. The research reported here was supported by the Institute of Education Sciences, U.S. Department of Education, through grants R305B140037 and R305A200069 to New York University. The opinions expressed are those of the authors and do not represent views of the Institute or the U.S. Department of Education.</p> <hd id="AN0188874259-38">Conflicts of Interest</hd> <p>The authors declare no conflicts of interest.</p> <hd id="AN0188874259-39">Data Availability Statement</hd> <p>Research data are not shared.</p> <p>GRAPH: Data S1: rrq70053‐sup‐0001‐DataS1.docx.</p> <ref id="AN0188874259-40"> <title> Footnotes </title> <blist> <bibl id="bib1" idref="ref68" type="bt">1</bibl> <bibtext> Funding: The research reported here was supported by the Institute of Education Sciences, U.S. Department of Education, through grants R305B140037 and R305A200069 to New York University.</bibtext> </blist> </ref> <ref id="AN0188874259-41"> <title> References </title> <blist> <bibtext> Alicea, S. 2015. " Social Networks in Community Colleges: Influences on Social Capital, Academic Achievement, Employment Skills and Psychosocial Wellbeing [Ph.D., New York University]." In ProQuest Dissertations and Theses (1688760349). Dissertations & Theses @ New York University; ProQuest Dissertations & Theses Global. New York University.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref36" type="bt">2</bibl> <bibtext> August, D., and T. Shanahan. 2006. Developing Literacy in Second‐Language Learners: Report of the National Literacy Panel on Language‐Minority Children and Youth. Lawrence Erlbaum Associates Publishers. <ulink href="http://www.cal.org/projects/archive/nlpreports/executive%5fsummary.pdf">http://www.cal.org/projects/archive/nlpreports/executive%5fsummary.pdf</ulink>.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref43" type="bt">3</bibl> <bibtext> Baker, S., N. Lesaux, M. Jayanthi, et al. 2014. Teaching Academic Content and Literacy to English Learners in Elementary and Middle School (NCEE 2014‐4012). Institute of Educational Sciences. <ulink href="http://ies.ed.gov/ncee/wwc/publications%5freviews.aspx">http://ies.ed.gov/ncee/wwc/publications%5freviews.aspx</ulink>.</bibtext> </blist> <blist> <bibl id="bib4" idref="ref11" type="bt">4</bibl> <bibtext> Borgatti, S. P., M. G. Everett, and J. C. Johnson. 2013. Analyzing Social Networks. SAGE.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref20" type="bt">5</bibl> <bibtext> Bourdieu, P. 1986. " The Forms of Capital." In Handbook of Theory and Research for the Sociology of Education, edited by J. R. Richardson, 241 – 258. Greenwood.</bibtext> </blist> <blist> <bibl id="bib6" idref="ref12" type="bt">6</bibl> <bibtext> Cappella, E., H. Y. Kim, J. W. Neal, and D. R. Jackson. 2013. " Classroom Peer Relationships and Behavioral Engagement in Elementary School: The Role of Social Network Equity." American Journal of Community Psychology 52, no. 3 : 367 – 379. https://doi.org/10.1007/s10464‐013‐9603‐5.</bibtext> </blist> <blist> <bibl id="bib7" idref="ref50" type="bt">7</bibl> <bibtext> Carhill‐Poza, A. 2015. " Opportunities and outcomes: The role of peers in developing the oral academic English proficiency of adolescent English learners." Modern Language Journal 99, no. 4 : 678 – 695. https://doi.org/10.1111/modl.12271.</bibtext> </blist> <blist> <bibl id="bib8" idref="ref2" type="bt">8</bibl> <bibtext> Carhill‐Poza, A., and N. Kurata. 2021. " Social Network Analysis and Its Application in Applied Linguistics." In Social Networks in Language Learning and Language Teaching, edited by A. Carhill‐Poza and N. Kurata. Bloomsbury Academic. https://doi.org/10.5040/9781350114289.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref10" type="bt">9</bibl> <bibtext> Chen, J., T.‐J. Lin, L. H. Anderman, N. Paul, and S. Y. Ha. 2021. " The Role of Friendships in Shy Students' Dialogue Patterns During Small Group Discussions." Contemporary Educational Psychology 67 : 102021. https://doi.org/10.1016/j.cedpsych.2021.102021.</bibtext> </blist> <blist> <bibtext> Chow, J. C., M. D. Broda, K. L. Granger, B. T. Deering, and K. T. Dunn. 2021. " Language Skills and Friendships in Kindergarten Classrooms: A Social Network Analysis." School Psychology 37, no. 6 : 488 – 500. https://doi.org/10.1037/spq0000451.</bibtext> </blist> <blist> <bibtext> Cohen, J. 1992. " A Power Primer." Psychological Bulletin 112, no. 1 : 155 – 159.</bibtext> </blist> <blist> <bibtext> Cooc, N., and J. S. Kim. 2017. " Peer influence on children's reading skills: A social network analysis of elementary school classrooms." Journal of Educational Psychology 109, no. 5 : 727 – 740. https://doi.org/10.1037/edu0000166.</bibtext> </blist> <blist> <bibtext> Delgado, M. Y., A. V. Ettekal, S. D. Simpkins, and D. R. Schaefer. 2016. " How Do My Friends Matter? Examining Latino Adolescents' Friendships, School Belonging, and Academic Achievement." Journal of Youth and Adolescence 45, no. 6 : 1110 – 1125. https://doi.org/10.1007/s10964‐015‐0341‐x.</bibtext> </blist> <blist> <bibtext> Droop, M., and L. Verhoeven. 2003. " Language Proficiency and Reading Ability in First‐ and Second‐Language Learners." Reading Research Quarterly 38, no. 1 : 78 – 103. https://doi.org/10.1598/RRQ.38.1.4.</bibtext> </blist> <blist> <bibtext> Duke, N. K., and K. B. Cartwright. 2018. " The DRIVE Model of Reading: Deploying Reading in Varied Environments." In Theoretical Models and Processes of Literacy, edited by D. E. Alvermann, N. J. Unrau, M. Sailors, and R. B. Ruddell. Routledge.</bibtext> </blist> <blist> <bibtext> Duke, N. K., and K. B. Cartwright. 2021. " The Science of Reading Progresses: Communicating Advances Beyond the Simple View of Reading." Reading Research Quarterly 56, no. S1 : S25 – S44. https://doi.org/10.1002/rrq.411.</bibtext> </blist> <blist> <bibtext> Enders, C. K., and D. L. Bandalos. 2001. " The relative performance of full information maximum likelihood estimation for missing data in structural equation models." Structural Equation Modeling: A Multidisciplinary Journal 8, no. 3 : 430 – 457. https://doi.org/10.1207/S15328007SEM0803_5.</bibtext> </blist> <blist> <bibtext> Foorman, B. R., and J. Torgesen. 2001. " Critical Elements of Classroom and Small‐Group Instruction Promote Reading Success in All Children." Learning Disabilities Research & Practice 16, no. 4 : 203 – 212. https://doi.org/10.1111/0938‐8982.00020.</bibtext> </blist> <blist> <bibtext> Furman, W. 1996. " The Measurement of Friendship Perceptions: Conceptual and Methodological Issues." In The Company They Keep: Friendship in Childhood and Adolescence, edited by W. M. Bukowski, A. F. Newcomb, and W. W. Hartup, 41 – 65. Cambridge University Press.</bibtext> </blist> <blist> <bibtext> Gifford‐Smith, M. E., and C. A. Brownell. 2003. " Childhood Peer Relationships: Social Acceptance, Friendships, and Peer Networks." Journal of School Psychology 41, no. 4 : 235 – 284. https://doi.org/10.1016/S0022‐4405(03)00048‐7.</bibtext> </blist> <blist> <bibtext> Goldenberg, C. 2020. " Reading Wars, Reading Science, and English Learners." Reading Research Quarterly 55, no. S1 : 131 – 134. https://doi.org/10.1002/rrq.340.</bibtext> </blist> <blist> <bibtext> Goldenberg, C., R. Coleman, L. Reese, and C. Rodriguez‐Mojica. 2013. "Classroom Qualities for English Language Learners in Language Arts Instruction: Technical Report." https://scholarcommons.scu.edu/tepas/73.</bibtext> </blist> <blist> <bibtext> Gough, P. B., and W. E. Tunmer. 1986. " Decoding, Reading, and Reading Disability." Remedial and Special Education 7, no. 1 : 6 – 10. https://doi.org/10.1177/074193258600700104.</bibtext> </blist> <blist> <bibtext> Granovetter, M. S. 1973. " The Strength of Weak Ties." American Journal of Sociology 78, no. 6 : 1360 – 1380.</bibtext> </blist> <blist> <bibtext> Hall, M. S., and M. K. Burns. 2018. " Meta‐Analysis of Targeted Small‐Group Reading Interventions." Journal of School Psychology 66 : 54 – 66. https://doi.org/10.1016/j.jsp.2017.11.002.</bibtext> </blist> <blist> <bibtext> Hwang, S. H. J., E. Cappella, M. J. Kieffer, and M. Yates. 2022. " "Let's Hang Out!": Understanding Social Ties Among Linguistically Diverse Youth in Urban Afterschool Programs." Social Development 31, no. 1 : 52 – 68. https://doi.org/10.1111/sode.12531.</bibtext> </blist> <blist> <bibtext> Johnson, H. E., L. Molloy Elreda, A. K. Kibler, and V. A. Futch Ehrlich. 2020. " Creating Classroom Communities in Linguistically Diverse Settings: Teacher‐Directed, Classroom‐Level Factor Effects on Peer Dynamics." Journal of Early Adolescence 40, no. 8 : 1087 – 1120. https://doi.org/10.1177/0272431619891238.</bibtext> </blist> <blist> <bibtext> Kieffer, M. J., C. P. Proctor, A. Weaver, et al. 2025. " Effects of Heterogeneous Versus Homogeneous Grouping on English Learners' Language and Literacy Development: Evidence from a Randomized Controlled Trial." American Educational Research Journal.</bibtext> </blist> <blist> <bibtext> Kraatz, E., M. Nagpal, T.‐J. Lin, et al. 2020. " Teacher Scaffolding of Social and Intellectual Collaboration in Small Groups: A Comparative Case Study." Frontiers in Psychology 11 : 587058. https://doi.org/10.3389/fpsyg.2020.587058.</bibtext> </blist> <blist> <bibtext> Kramsch, C. J., ed. 2002. Language Acquisition and Language Socialization: Ecological Perspectives. Continuum.</bibtext> </blist> <blist> <bibtext> Lee, M. 2010. " Researching Social Capital in Education: Some Conceptual Considerations Relating to the Contribution of Network Analysis." British Journal of Sociology of Education 31, no. 6 : 779 – 792. https://doi.org/10.1080/01425692.2010.515111.</bibtext> </blist> <blist> <bibtext> Lin, N. 1999. " Building a Network Theory of Social Capital." Connect 22, no. 1 : 3 – 28.</bibtext> </blist> <blist> <bibtext> Lou, Y., P. C. Abrami, J. C. Spence, C. Poulsen, B. Chambers, and S. d'Apollonia. 1996. " Within‐Class Grouping: A Meta‐Analysis." Review of Educational Research 66, no. 4 : 423. https://doi.org/10.3102/00346543066004423.</bibtext> </blist> <blist> <bibtext> McNeish, D., L. M. Stapleton, and R. D. Silverman. 2017. " On the Unnecessary Ubiquity of Hierarchical Linear Modeling." Psychological Methods 22, no. 1 : 114 – 140. https://doi.org/10.1037/met0000078.</bibtext> </blist> <blist> <bibtext> Michener, C. J., C. P. Proctor, and R. D. Silverman. 2018. " Features of Instructional Talk Predictive of Reading Comprehension." Reading and Writing 31, no. 3 : 725 – 756. https://doi.org/10.1007/s11145‐017‐9807‐4.</bibtext> </blist> <blist> <bibtext> Murphy, P. K., C. M. Firetto, J. A. Greene, and A. M. Butler. 2017. Analyzing the Talk in Quality Talk Discussions: A Coding Manual. Pennsylvania State University. https://doi.org/10.18113/S1XW64.</bibtext> </blist> <blist> <bibtext> Murphy, P. K., J. A. Greene, C. M. Firetto, et al. 2018. " Quality Talk: Developing Students' Discourse to Promote High‐Level Comprehension." American Educational Research Journal 55, no. 5 : 1113 – 1160. https://doi.org/10.3102/0002831218771303.</bibtext> </blist> <blist> <bibtext> Muthén, L. K., and B. Muthén. 2017. Mplus User's Guide (Eighth Edition): Statistical Analysis With Latent Variables [Computer Software]. Muthén & Muthén.</bibtext> </blist> <blist> <bibtext> Neal, J. W. 2010. " Hanging out: Features of Urban Children's Peer Social Networks." Journal of Social and Personal Relationships 27, no. 7 : 982 – 1000. https://doi.org/10.1177/0265407510378124.</bibtext> </blist> <blist> <bibtext> Nystrand, M. 2006. " Research on the Role of Classroom Discourse as It Affects Reading Comprehension." Research in the Teaching of English 40, no. 4 : 392 – 412.</bibtext> </blist> <blist> <bibtext> Posit team. 2022. RStudio: Integrated Development Environment for R [Computer Software]. Posit Software, PBC. <ulink href="http://www.posit.co/">http://www.posit.co/</ulink>.</bibtext> </blist> <blist> <bibtext> Proctor, C. P., M. Carlo, D. August, and C. Snow. 2005. " Native Spanish‐Speaking Children Reading in English: Toward a Model of Comprehension." Journal of Educational Psychology 97, no. 2 : 246. https://doi.org/10.1037/0022‐0663.97.2.246.</bibtext> </blist> <blist> <bibtext> Proctor, C. P., R. D. Silverman, J. R. Harring, R. L. Jones, and A. M. Hartranft. 2020. " Teaching Bilingual Learners: Effects of a Language‐Based Reading Intervention on Academic Language and Reading Comprehension in Grades 4 and 5." Reading Research Quarterly 55, no. 1 : 95 – 122. https://doi.org/10.1002/rrq.258.</bibtext> </blist> <blist> <bibtext> Proctor, C. P., R. D. Silverman, and R. L. Jones. 2021. " Centering Language and Student Voice in Multilingual Literacy Instruction." Reading Teacher 75, no. 3 : 255 – 267. https://doi.org/10.1002/trtr.2051.</bibtext> </blist> <blist> <bibtext> R Core Team. 2021. R: A Language and Environment for Statistical Computing [Computer Software]. R Foundation for Statistical Computing. https://www.R‐project.org/.</bibtext> </blist> <blist> <bibtext> Raffo, C., and M. Reeves. 2000. " Youth Transitions and Social Exclusion: Developments in Social Capital Theory." Journal of Youth Studies 3, no. 2 : 147 – 166. https://doi.org/10.1080/713684372.</bibtext> </blist> <blist> <bibtext> RAND Reading Study Group. 2002. Reading for Understanding: Toward an R&D Program in Reading Comprehension. RAND.</bibtext> </blist> <blist> <bibtext> Sabatini, J., J. Weeks, T. O'Reilly, K. Bruce, J. Steinberg, and S.‐F. Chao. 2019. " SARA Reading Components Tests, RISE Forms: Technical Adequacy and Test Design, 3rd Edition." ETS Research Report Series 2019, no. 1 : 1 – 30. https://doi.org/10.1002/ets2.12269.</bibtext> </blist> <blist> <bibtext> Soter, A. O., I. A. Wilkinson, P. K. Murphy, L. Rudge, K. Reninger, and M. Edwards. 2008. " What the Discourse Tells Us: Talk and Indicators of High‐Level Comprehension." International Journal of Educational Research 47, no. 6 : 372 – 391. https://doi.org/10.1016/j.ijer.2009.01.001.</bibtext> </blist> <blist> <bibtext> Takanishi, R., and S. L. Menestrel, eds. 2017. Promoting the Educational Success of Children and Youth Learning English: Promising Futures. National Academies Press. https://doi.org/10.17226/24677.</bibtext> </blist> <blist> <bibtext> Toste, J. R., L. McLean, P. Peng, et al. 2023. " Do Teacher Perceptions of Students' Academic and Behavioral Skills Influence Time Spent in Small‐Group Reading Instruction? " Elementary School Journal 124, no. 2 : 245 – 269. https://doi.org/10.1086/727475.</bibtext> </blist> <blist> <bibtext> van Lier, L. 2004. The Ecology and Semiotics of Language Learning: A Sociocultural Perspective. Springer.</bibtext> </blist> <blist> <bibtext> Wentzel, K. R., S. Jablansky, and N. R. Scalise. 2018. " Do Friendships Afford Academic Benefits? A Meta‐Analytic Study." Educational Psychology Review 30, no. 4 : 1241 – 1267. https://doi.org/10.1007/s10648‐018‐9447‐5.</bibtext> </blist> <blist> <bibtext> Wilhelm, A. G., A. G. Rouse, and F. Jones. 2018. " Exploring Differences in Measurement and Reporting of Classroom Observation Inter‐Rater Reliability." Practical Assessment, Research & Evaluation 23, no. 1 : 4. https://doi.org/10.7275/at67‐md25.</bibtext> </blist> </ref> <aug> <p>By Andrew Weaver; Michael J. Kieffer; C. Patrick Proctor; Elise Cappella; Xiaoying Wu; Shaelyn Cavanaugh and Sasha Karbachinskiy</p> <p>Reported by Author; Author; Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib50" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib30" firstref="ref3"></nolink> <nolink nlid="nl3" bibid="bib46" firstref="ref4"></nolink> <nolink nlid="nl4" bibid="bib52" firstref="ref5"></nolink> <nolink nlid="nl5" bibid="bib47" firstref="ref6"></nolink> <nolink nlid="nl6" bibid="bib18" firstref="ref7"></nolink> <nolink nlid="nl7" bibid="bib25" firstref="ref8"></nolink> <nolink nlid="nl8" bibid="bib33" firstref="ref9"></nolink> <nolink nlid="nl9" bibid="bib26" firstref="ref13"></nolink> <nolink nlid="nl10" bibid="bib12" firstref="ref14"></nolink> <nolink nlid="nl11" bibid="bib53" firstref="ref15"></nolink> <nolink nlid="nl12" bibid="bib49" firstref="ref16"></nolink> <nolink nlid="nl13" bibid="bib28" firstref="ref17"></nolink> <nolink nlid="nl14" bibid="bib15" firstref="ref22"></nolink> <nolink nlid="nl15" bibid="bib16" firstref="ref23"></nolink> <nolink nlid="nl16" bibid="bib31" firstref="ref31"></nolink> <nolink nlid="nl17" bibid="bib23" firstref="ref35"></nolink> <nolink nlid="nl18" bibid="bib14" firstref="ref37"></nolink> <nolink nlid="nl19" bibid="bib42" firstref="ref38"></nolink> <nolink nlid="nl20" bibid="bib29" firstref="ref49"></nolink> <nolink nlid="nl21" bibid="bib10" firstref="ref51"></nolink> <nolink nlid="nl22" bibid="bib27" firstref="ref53"></nolink> <nolink nlid="nl23" bibid="bib40" firstref="ref60"></nolink> <nolink nlid="nl24" bibid="bib37" firstref="ref61"></nolink> <nolink nlid="nl25" bibid="bib36" firstref="ref63"></nolink> <nolink nlid="nl26" bibid="bib22" firstref="ref64"></nolink> <nolink nlid="nl27" bibid="bib21" firstref="ref65"></nolink> <nolink nlid="nl28" bibid="bib43" firstref="ref67"></nolink> <nolink nlid="nl29" bibid="bib44" firstref="ref72"></nolink> <nolink nlid="nl30" bibid="bib48" firstref="ref80"></nolink> <nolink nlid="nl31" bibid="bib39" firstref="ref82"></nolink> <nolink nlid="nl32" bibid="bib54" firstref="ref85"></nolink> <nolink nlid="nl33" bibid="bib17" firstref="ref92"></nolink> <nolink nlid="nl34" bibid="bib38" firstref="ref93"></nolink> <nolink nlid="nl35" bibid="bib34" firstref="ref95"></nolink> <nolink nlid="nl36" bibid="bib45" firstref="ref96"></nolink> <nolink nlid="nl37" bibid="bib41" firstref="ref97"></nolink> <nolink nlid="nl38" bibid="bib11" firstref="ref98"></nolink> <nolink nlid="nl39" bibid="bib24" firstref="ref106"></nolink> <nolink nlid="nl40" bibid="bib32" firstref="ref107"></nolink> <nolink nlid="nl41" bibid="bib13" firstref="ref111"></nolink> <nolink nlid="nl42" bibid="bib51" firstref="ref116"></nolink> <nolink nlid="nl43" bibid="bib20" firstref="ref122"></nolink> <nolink nlid="nl44" bibid="bib19" firstref="ref123"></nolink> <nolink nlid="nl45" bibid="bib35" firstref="ref127"></nolink>
Header DbId: eric
DbLabel: ERIC
An: ED676548
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: The Role of Hang-Out Ties in Reading Comprehension and Small Group Interactions for Multilingual Students
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Andrew+Weaver%22">Andrew Weaver</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0002-2278-0517">0000-0002-2278-0517</externalLink>)<br /><searchLink fieldCode="AR" term="%22Michael+J%2E+Kieffer%22">Michael J. Kieffer</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0001-8000-7832">0000-0001-8000-7832</externalLink>)<br /><searchLink fieldCode="AR" term="%22C%2E+Patrick+Proctor%22">C. Patrick Proctor</searchLink><br /><searchLink fieldCode="AR" term="%22Elise+Cappella%22">Elise Cappella</searchLink><br /><searchLink fieldCode="AR" term="%22Xiaoying+Wu%22">Xiaoying Wu</searchLink><br /><searchLink fieldCode="AR" term="%22Shaelyn+Cavanaugh%22">Shaelyn Cavanaugh</searchLink><br /><searchLink fieldCode="AR" term="%22Sasha+Karbachinskiy%22">Sasha Karbachinskiy</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Grantee+Submission%22"><i>Grantee Submission</i></searchLink>. 2025 60.
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 14
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2025
– Name: SourceSuprt
  Label: Sponsoring Agency
  Group: SrcSuprt
  Data: Institute of Education Sciences (ED)
– Name: NumberContract
  Label: Contract Number
  Group: NumCntrct
  Data: R305B140037<br />R305A200069
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Audience
  Label: Education Level
  Group: Audnce
  Data: <searchLink fieldCode="EL" term="%22Elementary+Education%22">Elementary Education</searchLink><br /><searchLink fieldCode="EL" term="%22Grade+4%22">Grade 4</searchLink><br /><searchLink fieldCode="EL" term="%22Intermediate+Grades%22">Intermediate Grades</searchLink><br /><searchLink fieldCode="EL" term="%22Grade+5%22">Grade 5</searchLink><br /><searchLink fieldCode="EL" term="%22Middle+Schools%22">Middle Schools</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Reading+Comprehension%22">Reading Comprehension</searchLink><br /><searchLink fieldCode="DE" term="%22Reading+Instruction%22">Reading Instruction</searchLink><br /><searchLink fieldCode="DE" term="%22Small+Group+Instruction%22">Small Group Instruction</searchLink><br /><searchLink fieldCode="DE" term="%22Interpersonal+Relationship%22">Interpersonal Relationship</searchLink><br /><searchLink fieldCode="DE" term="%22Elementary+School+Students%22">Elementary School Students</searchLink><br /><searchLink fieldCode="DE" term="%22Multilingualism%22">Multilingualism</searchLink><br /><searchLink fieldCode="DE" term="%22Grade+4%22">Grade 4</searchLink><br /><searchLink fieldCode="DE" term="%22Grade+5%22">Grade 5</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1002/rrq.70053
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Small groups are commonly used in elementary reading classrooms, but relatively little research has examined the social factors that support small group reading instruction. This study tests whether the number of hang-out ties (i.e., the number of ties between students who like to hang out together) in small groups predicts reading comprehension among multilingual children in Grades 4 and 5, and whether interactions differ in small groups with high and low numbers of hang-out ties. We leverage data from a randomized controlled trial to examine whether the number of group hang-out ties predicts the later reading comprehension scores of participating multilingual students (N = 143; 46% female; 57% Asian American) using autoregressive path analysis. Furthermore, we explore differences in the patterns of interaction for groups with low (e.g., 0 to 2 ties) and high (e.g., 3 to 6) numbers of hang-out ties to see whether these differences might help explain the predictiveness of group hang-out ties on reading comprehension. Results indicated that the number of group hang-out ties negatively predicted later reading comprehension. Exploratory analyses of interaction patterns using observation data revealed that teachers asked more open-ended questions and elicited more student responses in groups with low numbers of hang-out ties. These findings suggest that small-group instruction may be most effective when students can learn from peers they know less well, as these other peers may offer novel linguistic assets, and that teachers may foster more group discussion when peers do not have existing hang-out ties.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: CodeSource
  Label: IES Funded
  Group: SrcInfo
  Data: Yes
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2025
– Name: AN
  Label: Accession Number
  Group: ID
  Data: ED676548
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=ED676548
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1002/rrq.70053
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
    Subjects:
      – SubjectFull: Reading Comprehension
        Type: general
      – SubjectFull: Reading Instruction
        Type: general
      – SubjectFull: Small Group Instruction
        Type: general
      – SubjectFull: Interpersonal Relationship
        Type: general
      – SubjectFull: Elementary School Students
        Type: general
      – SubjectFull: Multilingualism
        Type: general
      – SubjectFull: Grade 4
        Type: general
      – SubjectFull: Grade 5
        Type: general
    Titles:
      – TitleFull: The Role of Hang-Out Ties in Reading Comprehension and Small Group Interactions for Multilingual Students
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Andrew Weaver
      – PersonEntity:
          Name:
            NameFull: Michael J. Kieffer
      – PersonEntity:
          Name:
            NameFull: C. Patrick Proctor
      – PersonEntity:
          Name:
            NameFull: Elise Cappella
      – PersonEntity:
          Name:
            NameFull: Xiaoying Wu
      – PersonEntity:
          Name:
            NameFull: Shaelyn Cavanaugh
      – PersonEntity:
          Name:
            NameFull: Sasha Karbachinskiy
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2025
          Numbering:
            – Type: volume
              Value: 60
          Titles:
            – TitleFull: Grantee Submission
              Type: main
ResultId 1