Success With Informational Text Comprehension: An Examination of Underlying Factors

Abstract

The purpose of this study was to develop a clearer understanding of the complex, interrelated factors that lead to successful informational text comprehension and to determine if or how these factors vary for higher and lower comprehenders. Participants (N = 177) were students in grades 3–5 who were predominantly African American (61%) and female (51%). Hierarchical linear regressions were used to investigate how decoding efficiency, vocabulary knowledge, prior knowledge, and intrinsic motivation influenced informational text comprehension. For the whole sample, the complete model with all predictors explained 62.5% of the variance in informational text comprehension. Each reading component explained unique variance, with vocabulary knowledge explaining the largest portion. For lower comprehenders, decoding efficiency accounted for a significant amount of variance in reading comprehension beyond age and grade but was not a significant predictor when all variables were included in the model. For higher comprehenders, vocabulary was a consistent predictor of informational text comprehension. The results reveal the importance of investigating informational text comprehension for different groups of readers and the value of multiple factors, especially vocabulary knowledge, for success with informational texts. 本研究旨在对导致成功阅读理解信息文本的复杂而相关联的因素开展一个更清晰的了解,并确定这些因素是否或如何因阅读者有较高或较低阅读理解能力而有所不同。研究参与者来自177名三至五年级小学生,他们主要是非裔美国学生(61%),其中女生占51%。研究者使用分层回归分析方法考查解码效率、词汇知识、先前知识和内在动机如何影响信息文本阅读理解。就整个样本而言,完整模型中所有的预测因子解释了62.5%的信息文本阅读理解中的方差。每个阅读组件因素解释了独特的方差,其中以词汇知识解释的部分为最大。就较低阅读理解能力的学生而言,解码效率能超越年龄和年级解释了阅读理解中一个可观的方差,但当模型包含所有变量时,解码效率却不是一个显著的预测因子。就较高阅读理解能力的学生而言,词汇知识是阅读理解信息文本的一个有一致性的预测因子。研究结果显示出为不同阅读群组考查其信息文本阅读理解能力的重要性,以及考查奠定成功阅读理解信息文本的多种因素(尤其是词汇知识)的价值。 El propósito de este estudio fue el de desarrollar un mejor entendimiento de los factores complejos e interrelacionados que llevan a una comprensión exitosa de textos informativos y determinar si o cómo dichos factores varían en los que muestran alta y baja comprensión. Los participantes (N = 177) eran estudiantes del tercer al quinto grado, predominantemente africano-americanos (61%) y hembras (51%). Se usaron análisis de regresión jerárquica para investigar cómo la eficiencia en descifrar, el conocimiento de vocabulario, el conocimiento anterior y la motivación intrínseca influyen la comprensión de textos informativos. Para el grupo completo, el modelo completo con todos los predictores explicó el 62.5% de las discrepancias en la comprensión de textos informativos. Cada componente de lectura mostró una discrepancia original, obteniendo el conocimiento de vocabulario la porción más grande. Para los de baja comprensión, la eficiencia al descifrar fue responsable por una cantidad fuerte de discrepancias en la comprensión lectora más allá de la edad y el grado pero no fue un predictor importante cuando se incluían todas las variables en el modelo. Para los de alta comprensión, el vocabulario resultó ser un predictor consistente de la comprensión de textos informativos. Los resultados muestran la importancia de investigar la comprensión de textos informativos de diferentes grupos de lectores y el valor de factores múltiples, especialmente el conocimiento de vocabulario, para la comprensión exitosa de textos informativos. الغرض من هذه الدراسة هوه تطوير فهم أوضح للعوامل المعقدة والمتداخلة التي تؤدي إلى نجاح فهم النص الإرشادي وتحديد ما إذا و كيف يمكن لهذه العوامل تختلف بين المستوعبين التقدمين وغير التقدمين. المشاركون (N = 177) طالبا وطالبة في صفوف الرحلة 3-5 ومعظمهم من الامريكيين من اصل افريقى (61٪) والإناث (51٪). استخدمت الانحدار الخطي الهرمية للتحقيق في كفاءة كيفية تحليل ومعرفة المفردات، والمعرفة السابقة، الدافع الجوهري أثرت في فهم النص الإرشادي. وبالنسبة للعينة الجمالية، فسر النموذج الكامل مع الأخذ بالإعتبارجميع المنبئات 62.5٪ من الفوارق في فهم النص الإرشادي. وأوضح كل مكون مقروء فارق مميز، شرحت معرفة المفردات الجزء الأكبر. بانسبة للمستوعبين غير التقدمين, شكلت كفاءة التحليل القدر الإكبر من الفوارق في القراءة والفهم فوق السن والمرحلة ولكن لم تكن مؤشرا كبيرا. إذا تم إدراج جميع المتغيرات في النموذج. بالنسبة للمستوعبين التقدمين كانت المفردات مؤشراً ثابتاً في فهم النص الإرشادي. النتائج أظهرت أهمية تحقيق الفهم النص الإرشادي لمجموعات مختلفة من القراء, وقيمة تعدد العوامل ، لا سيما معرفة المفردات، أدت للنجاح في النصوص الإرشادية. Цeль иccлeдoвaния: cфopмиpoвaть бoлee яcнoe пpeдcтaвлeниe o cлoжныx взaимocвязaнныx yмeнияx, кoтopыe oбecпeчивaют пoнимaниe инфopмaциoннoгo тeкcтa, и oпpeдeлить, в кaкoй cтeпeни пpиcyтcтвyют эти yмeния y читaтeлeй c xopoшим и cлaбым пoнимaниeм тeкcтa. B иccлeдoвaнии yчacтвoвaли yчeники 3-5 клaccoв (N = 177), пpeимyщecтвeннo aфpoaмepикaнцы (61%) и дeвoчки (51%). Иepapxичecкий peгpeccиoнный aнaлиз пoмoг oпpeдeлить, кaк влияют нa пoнимaниe инфopмaциoннoгo тeкcтa тaкиe фaктopы, кaк ycпeшнocть дeкoдиpoвaния тeкcтa, cлoвapный зaпac, paнee нaкoплeнныe знaния и внyтpeнняя мoтивaция. Для вceй выбopки пoлнaя мoдeль c yчeтoм вcex фaктopoв oбъяcнилa 62,5% paзличий в вocпpиятии тeкcтa. Кaждый oтдeльный кoмпoнeнт дaeт yникaльныe paзличия в peзyльтaтax, пpичeм cлoвapный зaпac oтвeчaeт зa бoльшyю иx чacть. Для yчaщиxcя co cлaбым пoнимaниeм пpoчитaннoгo – внe зaвиcимocти oт вoзpacтa – cyщecтвeнным явилocь yмeниe дeкoдиpoвaть тeкcт, нo и этoт фaктop нe пoзвoляeт c выcoкoй дoлeй нaдeжнocти пpeдcкaзaть peзyльтaт, кoгдa в мoдeль включaютcя вce пepeмeнныe. Для пpoгнoзиpoвaния peзyльтaтoв тex yчaщиxcя, кoтopыe xopoшo пoнимaют пpoчитaннoe, c нaибoльшeй тoчнocтью мoжнo opиeнтиpoвaтьcя нa иx cлoвapный зaпac. Peзyльтaты cвидeтeльcтвyют o тoм, чтo для ycпeшнoй paбoты c инфopмaциoнным тeкcтoм нeoбxoдимo иccлeдoвaть eгo вocпpиятиe paзличными гpyппaми читaтeлeй c yчeтoм иx cлoвapнoгo зaпaca и иныx знaчимыx фaктopoв. Cette étude avait pour but de clarifier notre connaissance des facteurs complexes et interdépendants qui conduisent à une bonne comprehension d'un texte informatif et à identifier si et comment ces facteurs diffèrent chez les bons et les mauvais compreneurs. Les participants (N = 177) étaient des élèves de la 3e à la 5e année avec une dominante d'Afro-américains (61%) et de filles (51%). Nous avons utilisé des analyses de regression hiérarchique pour voir comment le savoir décoder, la connaissance du vocabulaire, les connaissances antérieures et les motivations personnelles jouent dans la compréhension d'un texte informatif. Pour l'échantillon entier, le modèle complet avec tous les prédicteurs explique 62.5% de la variance de la compréhension d'un texte informatif. Chaque composante de la lecture explique une variance unique, mais c'est la connaissance du vocabulaire qui explique la plus grande partie. Chez les mauvais comprenants, le savoir décoder rend compte d'un montant significatif de la variance de la compréhension de la lecture, au-delà de l'âge et de la classe fréquentée, mais n'est pas un prédicteur significatif quand toutes les variables sont inclues dans le modèle. Pour les bons compreneurs, le vocabulaire est un prédicteur solide de la compréhension d'un texte informatif. Ces résultats montrent l'importance d'examiner la compréhension d'un texte informatif avec des groupes différents et l'intérêt de facteurs multiples, notamment la connaisance du vocabulaire, pour arriver à comprendre des textes informatifs. Many challenges confront researchers who study reading comprehension. Not only is reading comprehension a complex construct involving multiple processes that vary based on an individual's skills and purposes for reading (Duke, 2005; RAND Reading Study Group, 2002), but we also have growing evidence that the type of text matters (Best, Floyd, & McNamara, 2008; Eason, Goldberg, Young, Geist, & Cutting, 2012; McNamara, Ozuru, & Floyd, 2011). This latter point has significant implications, given a growing emphasis on informational text (National Governors Association Center for Best Practices & Council of Chief State School Officers, 2010). In consequence, it becomes critical that researchers and teachers understand the skills, knowledge, and abilities that elementary readers utilize to comprehend informational texts. Two theoretical models undergird the ideas and selection of variables in this study. First, the construction–integration model (Kintsch, 1988, 1998, 2004) highlights the complexity of comprehension by noting that reading is a resource-demanding cognitive process in which readers must purposely direct their attention and effort to problem solving as they construct meaning. To make sense of a text, a reader must read the actual words, understand their meaning within the context, and integrate the text with prior knowledge both to make inferences and to think critically about the text. Second, the interactive-compensatory model (Stanovich, 1980) explains how readers may depend on different processes or strategies to construct meaning from text. Specifically, readers may use different knowledge and processes when they encounter a problem while reading. Participants were 177 students in grades 3–5 from a Title I public elementary school in a medium-sized Southeastern U.S. school district (see Table 1). Students completed five assessments administered in two phases over the course of six weeks (see Table 2). In the first phase, participants were group-administered a reading comprehension assessment based on the National Assessment of Educational Progress fourth-grade reading test and a subtest of the Motivations for Reading Questionnaire (MRQ; Wigfield & Guthrie, 1995) in two 30-minute sessions. The reading comprehension assessment consisted of four informational and two narrative text passages with their corresponding multiple-choice questions. In the second phase, individual students were tested in 35–50-minute sessions and completed the Test of Word Reading Efficiency, second edition (TOWRE–2; Torgesen, Wagner, & Rashotte, 2012), the Test of Language Development (TOLD; Hammill & Newcomer, 2008), and the Woodcock–Johnson Tests of Achievement (WJ AKT; Woodcock, McGrew, & Mather, 2001). Three hierarchical multiple regressions were performed to determine which specific reading skills uniquely predicted informational text comprehension for all readers and then for those with higher and lower comprehension. The variables predicted to explain informational text comprehension were entered in five steps using a bottom-up approach. In step 1, age and grade were the independent control variables. In steps 2–5, the following independent variables were entered as separate steps in this order: decoding efficiency (TOWRE–2), vocabulary (TOLD), prior knowledge (WJ AKT), and intrinsic motivation (MRQ). To determine the extent to which these relationships differed for students with higher and lower comprehension, students were split into groups at the median based on their informational text comprehension scores. The independent variables examined explained 63% of the total variance, and all were unique predictors that explained additional variance in informational text comprehension when entered into the hierarchical regression model (see Table 3 for descriptive statistics and Table 4 for this model). Beyond the control variables of age and grade, decoding efficiency explained 23% of the variance in informational text comprehension. The addition of vocabulary knowledge explained 22% more variance. Although their predictive power was smaller, prior knowledge and intrinsic motivation each explained additional variance in the final two steps (1.2% and 1.1%, respectively). Overall, vocabulary knowledge demonstrated the largest influence on informational text comprehension (β = 0.026), and the influence of this variable was about twice as large as prior knowledge (β = 0.014) and decoding efficiency (β = 0.012). The split-sample analyses explained smaller amounts of variance than the model for the entire sample, and the selected reading components better predicted informational text comprehension for higher comprehenders than lower comprehenders (see Table 5 for descriptive statistics and Table 6 for this model). For lower comprehenders, the complete regression model explained 13% of the variance. Decoding efficiency was the only predictor that explained unique variance (4.2%), and this only occurred when the control variables—and no others—were entered into the model. For higher comprehenders, the complete regression model explained 43% of the variance. Vocabulary was the only predictor that explained unique variance (25%) when entered into the model after decoding efficiency and the control variables. When comparing the relative weights of the coefficients, decoding efficiency and intrinsic motivation were more important for lower comprehenders. The coefficient for the lower comprehenders compared with the higher comprehenders in decoding was 5 times larger, and the coefficient for intrinsic motivation was double. By contrast, vocabulary knowledge was a better predictor of comprehension for higher comprehenders than lower comprehenders. Current findings highlight the importance of moving beyond constrained skills (e.g., Paris, 2005) when we consider the requisite skills necessary for success with informational texts. Vocabulary, prior knowledge, and students' intrinsic motivation each contributed to variance. The importance of vocabulary knowledge, in particular, supports Hiebert and Cervetti's (2011) assertion that the content-based, complex, and unique vocabulary contained in informational texts make these texts more challenging to comprehend than fiction. Additionally, the results also point to the importance of examining these component skills for different types of readers because analyses of a whole group may mask key differences. Understanding specific types of readers proves important for both instruction and intervention (Buly & Valencia, 2002). Instruction for lower comprehenders of informational text may require an initial emphasis on decoding, whereas higher comprehenders may need instruction to emphasize and build vocabulary knowledge. Meghan D. Liebfreund received her PhD in literacy education from North Carolina State University. Her dissertation was supported by her committee members: Kristin Conradi (chair), Hiller Spires, Stephen Porter, and Samuel Miller (University of North Carolina at Greensboro). Liebfreund is an assistant professor in the Department of Elementary Education at Towson University, Maryland, USA, where she prepares undergraduate and graduate students to support the literacy success of diverse students; e-mail [email protected]. Her research focuses on better understanding and instructing students who experience difficulty with reading, especially in the area of comprehension.