View in EDS HTML Full Text PDF Full Text

Effects of Written and Auditory Language-Processing Skills on Written Passage Comprehension in Middle and High School Students

Saved in:
Bibliographic Details
Title: Effects of Written and Auditory Language-Processing Skills on Written Passage Comprehension in Middle and High School Students
Language: English
Authors: Caplan, David, Waters, Gloria, Bertram, Julia, Ostrowski, Adam, Michaud, Jennifer
Source: Reading Research Quarterly. Jan-Mar 2016 51(1):67-92.
Availability: Wiley-Blackwell. 350 Main Street, Malden, MA 02148. Tel: 800-835-6770; Tel: 781-388-8598; Fax: 781-388-8232; e-mail: cs-journals@wiley.com; Web site: http://www.wiley.com/WileyCDA
Peer Reviewed: Y
Page Count: 26
Publication Date: 2016
Sponsoring Agency: Institute of Education Sciences (ED)
Contract Number: R305A100261
Document Type: Journal Articles
Reports - Research
Education Level: Middle Schools
Secondary Education
Junior High Schools
High Schools
Descriptors: Middle School Students, High School Students, Language Skills, Language Processing, Reading Comprehension, Morphology (Languages), Syntax, Written Language, Oral Language, Auditory Perception, Factor Analysis, Accuracy, Reaction Time, Difficulty Level, Sentences, Structural Equation Models, Evidence
DOI: 10.1002/rrq.126/abstract
ISSN: 0034-0553
Abstract: The authors assessed 4,865 middle and high school students for the ability to recognize and understand written and spoken morphologically simple words, morphologically complex words, and the syntactic structure of sentences and for the ability to answer questions about facts presented in a written passage and to make inferences based on those facts. Factor analysis of combined accuracy and reaction time results for the tests of simple words, complex words, and sentences resulted in three factors in both the auditory and written modalities, reflecting recognition of forms of simple and complex words, understanding of simple and complex words, and recognition of the structure and understanding the meaning of sentences. Structural equation models showed direct effects of these factors in both modalities on written passage comprehension. The results provide evidence that skills in processing language in the written modality are separate from, although related to, skills in auditory language processing and are important determinants of comprehension of written passages in middle and high school.
Abstractor: As Provided
Number of References: 66
IES Funded: Yes
Entry Date: 2016
Accession Number: EJ1087345
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwEB2rShxcCzkSaRKuJyJyqoAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDIyaeY2BPN_EU1c9awIBEICBm_c6UGcjH3qd91bVwoKLbD_Mz3HJXHds8HluF-_K6v6oxVei4TFOMI46jflxsEpJ3wS6RyAq9XxDH3pSCAAAq0cx-E6H17HpFwse0HSE0bkzRpEe7lp9o8YeUsA021PweObloXmOUHd49L3CuzZkJfXdeE_00vQNPNWczyBk0zKBTyNo3-N2S3unA_N_WBvwyfWqm3VIJbPe0mEQ
Text:
  Availability: 1
  Value: <anid>AN0112047212;[nrnu]01jan.16;2025Apr04.08:25;v2.2.500</anid> <title id="AN0112047212-1">Effects of Written and Auditory Language-Processing Skills on Written Passage Comprehension in Middle and High School Students </title> <p>The authors assessed 4,865 middle and high school students for the ability to recognize and understand written and spoken morphologically simple words, morphologically complex words, and the syntactic structure of sentences and for the ability to answer questions about facts presented in a written passage and to make inferences based on those facts. Factor analysis of combined accuracy and reaction time results for the tests of simple words, complex words, and sentences resulted in three factors in both the auditory and written modalities, reflecting recognition of forms of simple and complex words, understanding of simple and complex words, and recognition of the structure and understanding the meaning of sentences. Structural equation models showed direct effects of these factors in both modalities on written passage comprehension. The results provide evidence that skills in processing language in the written modality are separate from, although related to, skills in auditory language processing and are important determinants of comprehension of written passages in middle and high school.</p> <p>This study explores the relation among skills in processing the lexical, morphological, and syntactic levels of language in the written and auditory modalities and written passage comprehension in middle and high school. The study addresses three theoretical questions:</p> <olist> <item> To what extent does language comprehension involve skills that recognize and assign meaning to specific levels of linguistic structure?</item> <item> Do these skills apply in different ways to written and auditory input?</item> <item> If separate skills in processing these levels of linguistic structure develop in the written and the auditory modality, to what extent does the ability to comprehend written passages depend on skills in the written modality as opposed to the auditory?</item> </olist> <p>On a practical level, the study has potential implications for skills that could be trained in middle and high school to achieve good reading abilities.</p> <p>The study was undertaken within a model of passage comprehension that is widely adopted in psycholinguistics (Kintsch, 1988). In this framework, understanding a passage begins with determining the meanings (propositional content) of the sentences it contains. These propositions are entered into a textbase that captures their basic relations to one another, such as their temporal order and causal connections. The reader supplements the textbase with his or her general knowledge about the world and about the subject of the passage and adds inferences based on combining information in the text-base with this knowledge. This leads to a final representation of the passage known as the situation model, so called because it includes information about the situation to which the text refers.</p> <p>The factors that lead to success in creating a situation model have been partially described. Knowledge of subject matter and vocabulary related to the topic of a passage (McCardle, Scarborough, & Catts, 2001; Nation & Snowling, 2004) and the ability to form inferences (Bowyer-Crane & Snowling, 2005; Cain, Oakhill, Barnes, & Bryant, 2001) are important determinants of the ability to understand a written text. Success in creating a situation model correlates with psychometrically measured intelligence (Ferrer et al., 2007), executive functions and speed of processing (Christopher et al., 2012), and metacognitive functions such as comprehension monitoring (Hogan, Sittner, Justice, & Cain, 2011).</p> <p>This study focuses on a different set of abilities: the largely unconscious, automatized skilled ability to recognize and understand simple words (lexical items), word structure (morphology), and sentence structure (syntax). These basic psycholinguistic operations are needed to understand the propositions in a passage and, therefore, are necessary to form the textbase. Work in experimental psycholinguistics has identified a large and complex set of basic psycholinguistic processes that recognize forms and access meaning at each of these levels of language (Traxler, 2012). None of the factors listed previously can substitute for these basic psycholinguistic abilities: A reader must recognize and understand the words and structure and understand the sentences in a text to extract the propositions in a passage regardless of his or her psychometrically measured intelligence, executive functions, speed of processing, ability to form inferences, or metacognitive functions.</p> <p>There is substantial evidence that the level of skill in these basic psycholinguistic abilities is related to successful comprehension of written language (Bryant, Nunes, & Barros, 2014; Cromley & Azevedo, 2007; Cromley, Snyder-Hogan, & Luciw-Dubas, 2010; Curtis, 1980; Curtis & Kruidenier, 2005; Cutting & Scarborough, 2006; Gough, Hoover, & Peterson, 1996; Hogan et al., 2011; Hoover & Gough, 1990; Joshi, Williams, & Wood, 1998; Keenan, Betjemann, & Olson, 2008; Mellard, Fall, & Woods, 2010; Nation, 2005; Vellutino, Tunmer, Jaccard, & Chen, 2007; Yuill & Oakhill, 1991). However, there are many unanswered questions about these skills and their relation to successful comprehension of written passages.</p> <p>One open question is, Does assigning linguistic structure and meaning constitute one integrated skilled ability, or does it consist of a set of separate operations that are related to the structure of language as described by linguists: recognition and comprehension of simple words, morphologically complex words, and sentences? These skills have been studied in isolation by psychologists but, to our knowledge, have never been examined in a single set of tests that have been analyzed to see if they show evidence of being separate skills.</p> <p>A second question is, To what extent do basic psycholinguistic skills differ in the auditory and written modalities? Readers could transcode written words and morphemes into the same phonological and lexical representations that are accessed from auditory input, and apply the basic psycholinguistic operations they already master in the auditory modality to these representations. In this model, all operations after word recognition (lexical access)--combining simple words and morphemes to form morphologically complex words, accessing the meanings and syntactic features of simple and morphologically complex words, and combining words into sentences and propositions--use the same operations regardless of whether language input is written or spoken. Using operations that apply to auditory input is the most parsimonious way for a reader to understand a written text in terms of the number of different operations that a language user must have available, and it may be the easiest. This would logically be especially true for developing readers, in whom basic psycholinguistic operations have already developed for spoken language and are just beginning to develop for written language.</p> <p>In addition, the abstract and highly specific nature of linguistic representations may encourage the use of a single set of skills in comprehension. The use of auditory language-processing skills may be the preferred way to comprehend language because infants hear spoken language before they encounter written language; because infants and young children hear auditory language in an affective and purposeful, interactive context (Tomasello, 2003) and generally encounter written language in an environment that does not have these features (Perfetti, 1987; Watson & Olson, 1987); because of a genetically determined preference to acquire and use spoken language; or for other reasons. Having a capacity to process linguistic representations in the auditory modality may restrict the development of skills in processing written language. The well-known simple model of reading (Gough & Tunmer, 1986) hypothesizes that this is how reading for comprehension takes place--that lexical access for written words leads to the activation of the phonological representation of words and that this information is then processed by the routines that apply to spoken language.</p> <p>However, there is also good reason to believe that written language comprehension differs from spoken language comprehension in more ways than simply recognizing simple words and morphemes. One reason for this view is that written and spoken language do not present exactly the same information to the comprehender (Chafe & Danielewitz, 1987; Halliday, 1987). Letters and words are often separated in orthography, whereas phonemes and words occur continuously and are subject to articulatory and acoustic overlap in speech. Punctuation marks and paragraph spacing mark sentence and discourse structure in written language, and intonation contours do so in a different way in spoken language. Different operations may develop in the two modalities to utilize these different cues. The evanescence of auditory input changes the need for short-term working memory in the two modalities. Even operations that have the same basic characteristics may differ in comprehension of written and spoken language. For instance, there is strong evidence that comprehenders make predictions about upcoming material based on how often words follow one another (i.e., surprisal models of building syntactic structure, parsing; Levy, 2008). Because words co-occur with different frequencies in written and spoken language, the predictions made during reading will differ from those made while listening. Finally, practice with reading could in and of itself result in skilled written language operations becoming increasingly automatized and separate from skilled spoken language processing.</p> <p>There is substantial evidence that written word recognition (i.e., lexical access, decoding) and listening comprehension are relatively independent skills (e.g., Catts, Gillispie, Leonard, Kail, & Miller, 2002; Catts, Hogan, & Fey, 2003; Hoover & Gough, 1990; Perfetti, Landi, & Oakhill, 2005; Savage, 2006; Savage & Wolforth, 2007). There is also evidence that skills in applying basic psycholinguistic operations beyond lexical access in the written modality are partially independent of these skills in the auditory modality (Bryant et al., 2014; Carlisle, 2010; Chafe, 1992; Chafe & Tannen, 1987; Danks & End, 1987; Horowitz, 2015; Horowitz & Samuels, 1987; Nagy, Berninger, Abbott, Vaughan, & Vermeulen, 2003; Nagy, Berninger, & Abbott, 2006; Townsend, Carrithers, & Bever, 1987) and that skills in applying basic psycholinguistic operations beyond lexical access develop in the written modality throughout middle and high school (Diakidoy, Stylianou, Karefillidou, & Papageorgiou, 2005; Sticht, Beck, Hauke, Kleinman, & James, 1974). The present study presents further data regarding the development of skills in both auditory and written language comprehension.</p> <p>A third question is, What is the relative importance of skills in processing language in the written and auditory modalities to comprehension of written passages? Although isolated abilities in assigning higher level structure and meaning in the written and auditory modalities to comprehension of written passages have been studied (see the previous discussion), to our knowledge, there has been no systematic study of the relation of skills in processing all levels of language in the two modalities to comprehension of written passages. The present study also examines this question.</p> <p>We studied these questions in middle and high school students, who often do not have adult-level reading skills and therefore may rely more on skills in processing spoken language to understand written passages than more practiced readers do. If we find evidence for separate basic psycholinguistic operations in written and auditory language processing in this population, it is likely that these operations remain separate in adult readers as well. In addition, focusing on this age group has potential practical implications. If we find that skills in applying basic psycholinguistic operations are important determinants of written passage comprehension, educational programs directed toward developing these skills might lead to better comprehension. If separate skills for processing written language play an important role in written comprehension in this age group, educational programs might target these skills in the written modality. Our study provides information relevant to these issues.</p> <hd id="AN0112047212-2">Methods</hd> <p>Participants</p> <p>We tested middle and high school students from an urban school district in Central New York State that consists of a pre-K (ages 1-4) school, five elementary schools (grades K-5), one middle school (grades 6-8), and a comprehensive high school (grades 9-12). The high school offers 23 courses for advanced placement, as well as a career and technical education program. The student population is 91% white, 4% black or African American, 2% American Indian or Alaska native, 2% Asian or native Hawaiian/other Pacific Islander, 1% Hispanic or Latino, and 1% multiracial (the percentages total over 100% because of rounding). Students come from a wide range of socioeconomic backgrounds. Approximately 29% of students are eligible for free or reduced-price lunch. Sixteen percent of students are on Individualized Educational Plans. The high school dropout rate is approximately 3%. There are very few students whose mother tongue is not English (1%). On the New York State Examinations in English Language Arts and Mathematics, 12% of students across the elementary, middle, and high school grades scored at level 1 (serious academic difficulty), 31% at level 2 (below standards and needs help to pass), 43% at level 3 (meets standards and with steady growth should pass), and 13% at level 4 (exceeds standards and moving toward high performance). The students are fairly typical of middle class urban schools. Unfortunately, we do not have reading level data for them.</p> <p>In 2011 and 2013, the entire middle and high school population was tested on the battery (2011 N = 1,836: grade 6 = 241, grade 7 = 270, grade 8 = 260, grade 9 = 291, grade 10 = 244, grade 11 = 284, grade 12 = 246; 2013 N = 1,642: grade 6 = 232, grade 7 = 254, grade 8 = 257, grade 9 = 236, grade 10 = 216, grade 11 = 243, grade 12 = 204). Half the students in 2013 were tested on the battery used in 2009 and 2011, and the other half were tested on a matched second version. The two versions showed similar effects for each test, and the results were combined in the analyses that we present here.</p> <p>Materials</p> <p>To study the questions posed previously requires a set of tests that assess the level of skill of individual students in performing basic psycholinguistic operations in the written and auditory modalities. To our knowledge, no existing assessment battery examines all levels of language with comparable materials in the two modalities. We therefore created a novel battery of tests that assessed sublexical, lexical, morphological, sentential, and discourse levels of language in the auditory and written modalities. The items in each test were also created to vary factors that have been shown in experimental studies to affect processing of these levels of language.</p> <p>The battery was a successor to a previous battery designed to assess language in people with aphasia (Caplan, 1992, 1993). The aphasia version of the battery was normed on 100 neurologically normal participants matched in age to the aphasic population tested. Portions of that battery have been used in various research studies (e.g., Caplan, Michaud, & Hufford, 2013; Caplan, Waters, DeDe, Michaud, & Reddy, 2007) and for clinical purposes at the Massachusetts General Hospital.</p> <p>We modified the aphasia battery for use with middle and high school students by changing some features of its presentational format and content. We created an entirely computer-administered battery and presented it on MacBook laptops running a proprietary program on a Java Runtime Environment 1.4-based platform. The program annotated all responses as correct or incorrect and measured the reaction time (RT) for each response. We changed all items on the battery to two-alternative forced-choice questions. We refined several factors that were incorporated into item selection to better vary and control psycholinguistic parameters in each test, and selected new items to test a younger population.</p> <p>We piloted the new battery in several ways before testing. Over a period of 2.5 years (2005-2008), we administered portions of the battery to approximately 2,000 students in middle and high schools in local school districts. Because of limited access to students, we could not adminster the entire battery. These studies led to modification of some materials. We also administered the second edition of the Test of Word Reading Efficiency (Torgesen, Wagner, & Rashotte, 2012) to 231 students and the Group Reading Assessment and Diagnostic Evaluation (Fugate, 2002) to 501 students who also performed the tests of sublexical and lexical processing on the battery, providing data relevant to the construct validity of the tests of decoding, word recognition, and word comprehension in the battery.</p> <p>We obtained a second, more complete set of pilot data by testing students in grades 6-10 in the participating school district in the spring of 2009 (total N = 1,356: grade 6 = 239, grade 7 = 288, grade 8 = 249, grade 9 = 308, grade 10 = 272). The results were used to investigate the internal consistency of the tests, the effects of the variables incorporated in each test, the effects of grade, and the factor structure of performance. Based on these results, we made further minor changes to items in some tests and then tested the entire student body in 2011 and 2013, as described previously.</p> <hd id="AN0112047212-3">The Assessment Battery</hd> <p>Within each language level, separate subtests assessed the ability to recognize the forms of language (e.g., that happiness is a word and lifement is not) and to understand the contribution of that level of language to meaning (e.g., that conditional is closer in meaning to dependent than to depends). Within each level, materials were chosen to assess specific parameters of language that linguistic, psycholinguistic, and educational studies have shown to be important in determining performance. The materials in the battery are outlined in Table 1 and subsequently described briefly, including a description of the parameters that were manipulated in creating the items in each test. A complete description of each test and its parameters is provided in Appendix A (available as supporting information for the online version of this article). We draw the reader's attention to the fact that there is no test 6.1</p> <hd id="AN0112047212-4">Sublexical and Lexical Levels</hd> <p>Processing of Sublexical and Lexical Form</p> <p>TEST 1 : DECODING (P SEUDOHOMOPHONE JUDGMENT). This test assessed the ability to convert graphemes to phonemes (one aspect of decoding) without requiring a spoken response. A written nonword was presented on each trial, and students were required to judge whether it would sound like a real English word if pronounced aloud. Forty pseudohomophones and 40 nonpseudohomophonic nonwords were created, based on moderate-to high-frequency words that were matched pairwise for length and frequency.2 Half of each type of stimulus consisted of sequences of high grapheme-phoneme correspondences (e.g., seet, stred), and the other half consisted of sequences of low grapheme-phoneme correspondences (e.g., rhole, yace). Pseudohomophones and nonpseudohomophonic nonwords, both with high and low grapheme-phoneme correspondences, were matched for bigram frequency and lexical neighborhood of the words from which they were derived (Glushko, 1979; Vitevitch & Luce, 1998).</p> <p>TEST 2: SIMPLE WORD RECOGNITION (LEXICAL DECISION: SIMPLE WORDS). This test assessed the ability to recognize written words. Participants saw 60 real words and 60 nonwords created by changing one letter in words matched to the target set for length, frequency, and number of neighbors. The nonwords were matched to the words for letter length, initial phoneme, and mean bigram frequency. Participants were required to indicate whether each letter string was a real English word. The word stimuli were based on Jared (2002) and consisted of an equal number of high-and low-frequency regular, regular inconsistent, and exception words. Half of the regular inconsistent words had more friends than enemies, and the other half had more enemies than friends (see Table 1).</p> <p>Word Comprehension</p> <p>TEST 3: WORD-PICTURE MATCHING: SIMPLE WORDS. On each trial, a written word was presented on the top of the computer screen, and two line drawings were presented on the bottom. Students were required to press a button on the computer corresponding to the side of the screen on which the correct picture was depicted. There were 40 target words, which were varied orthogonally in terms of lexical frequency (high, low) and spelling-sound consistency (consistent, inconsistent). The incorrect pictures (foils) were chosen to be closely semantically related to the target word.</p> <p>TEST 4: RELATEDNESS JUDGMENT: SIMPLE WORDS. On each trial, a printed word was presented at the top of the computer monitor, and two additional words were presented below. Students were required to indicate which of the bottom two words was most closely related to the word on top. The target stimuli consisted of 20 words with consistent spelling-sound correspondences and 20 words with inconsistent correspondences. Half of the stimuli of each type were high frequency, and the other half were low. The incorrect words (foils) were chosen to be closely semantically related to the target word (e.g., target: race; correct response: run; foil: walk).</p> <hd id="AN0112047212-5">Morphological Level</hd> <p>Recognition of Morphologically Complex Words</p> <p>TEST 5: COMPLEX WORD RECOGNITION (LEXICAL DECISION: MORPHOLOGICALLY COMPLEX WORDS). This test assessed the ability to recognize that written word roots and affixes can combine in different ways. Students saw real words made up of high-frequency stems (e.g., happy) and affixes (e.g., ness) that combine to form low-frequency morphologically complex words (e.g., happiness), and an equal number of stimuli that were illegal combinations of stems and affixes (e.g., lifement). The words consisted of 10 words with inflectional affixes (e.g., supposes), 10 with word-boundary derivational affixes (e.g., handful), and 10 with formative-boundary affixes (e.g., nationality). Ten nonwords were formed by combining a root with a derivational affix (e.g., lifement), 10 with an inflectional affix (e.g., camed), and 10 with a formative-boundary affix (e.g., powerity).</p> <p>Comprehension of Morphologically Complex Words</p> <p>TEST 7: WORD-PICTURE MATCHING: MORPHOLOGICALLY COMPLEX WORDS. On each trial, a written word was presented on the top of the computer screen, and two line drawings were presented on the bottom. Students were required to press a button on the computer corresponding to the side of the screen on which the correct picture was depicted. There were 40 target words: 20 with inflectional affixes (e.g., tripped) and 20 with derivational affixes (e.g., hopeful). The choice between the correct and incorrect picture depended on understanding the contribution made by the affix to the meaning of the stimulus word. For instance, with the target cleans, the correct picture depicted a person cleaning a room and the incorrect picture (the foil) depicted a person standing in a clean room looking satisfied (intended to represent cleaned). The lexical frequency of the words intended to be depicted in the two pictures on each trial, as well as of the items in the derivational and inflectional affix conditions, were matched.</p> <p>TEST 8: MORPHOLOGICAL FORM RELATEDNESS JUDGMENT. This task was based on part 3 of Mahony's (1994) morpheme substitution test. Students were presented with 20 pairs of words. Half of the words were derivationally related (e.g., help, helpful), and the other half were pseudorelated (e.g., should, shoulder). Students were required to indicate whether each pair of words were related to each other. Five classes of stimulus pairs were created by varying the phonological relationship between the base and derived words (neutral, stress shift/vowel change, consonant, vowel, silent letter). There were four morphologically related pairs and four unrelated pairs in each class. Derivationally related stimulus pairs were created using high-frequency base words to which the appropriate boundary type suffix was added to form a low-frequency stimulus. Pseudorelated stimulus pairs were created from high-frequency base words and low-frequency words that were orthographically similar to derived words.</p> <p>TEST 9: MEANING RELATEDNESS JUDGMENT: MORPHOLOGICALLY COMPLEX WORDS. On each trial, an affixed word was shown at the top of the computer screen along with two words below. The student was required to judge which of the two words was most related to the target. The choice of the correct response required understanding the contribution the affix made to the root; for example, for the target beginner, the two choices were student and study. There were 20 stimuli. The target and foil were matched as closely as possible in word frequency. In this subtest, no features of the affixed stimulus words were varied.</p> <hd id="AN0112047212-6">Sentence Level</hd> <p>Recognizing Syntactic Forms (Syntactic Awareness)</p> <p>TEST 10: GRAMMATICALITY JUDGMENT. This test assessed the ability to recognize that a sentence is grammatically well formed. On each trial, students were presented with a series of words and required to indicate whether they formed a grammatical or ungrammatical sentence in English. Ten examples of each of four different grammatical structures (passives, relative clauses, reflexives, sentences with prepositional phrase attachments) were presented. Half of the sentences were grammatical and the other half ungrammatical. The ungrammatical sentences were formed by deleting required words or morphemes (e.g., "The baby was wash by the man"), by introducing agreement errors (e.g., "The girl wash the baby"), or by inserting an extra noun phrase (e.g., "The baby who the boy carried the mother hugged the girl").</p> <p>Syntactically Based Comprehension</p> <p>TEST 11: SENTENCE-PICTURE MATCHING. This test assessed the ability to understand sentences using their syntactic structure. Students were shown a sentence with two pictures below it and required to choose the picture that corresponded to the sentence. The sentences tested knowledge of eight different grammatical structures (active, passive, cleft subject, cleft object, subject control, dative passive, subject relative, object relative), with five sentences of each type, for a total of 40 stimuli. The sentences were all reversible in English (i.e., sentences in which any noun could be the actor or the patient; e.g., "The boy was tickled by the girl"), and the foil picture depicted the reverse action of the target sentence (e.g., a picture of a boy tickling a girl), so students were required to process the sentences syntactically to choose the correct picture.</p> <hd id="AN0112047212-7">Discourse Level</hd> <p>Discourse Comprehension</p> <p>This subtest examined readers' ability to understand written passages in which the words were familiar and the topics unfamiliar. Students read and answered factual and inferential questions about sixteen 150-word passages. Four versions of each passage, matched for length and Lexile level, were created, in which macrostructural and microstructural complexity were orthogonally varied. Macrostructural complexity was achieved by reordering the sentences in the base (macrostructurally simple) version of each passage, such that topics in the passage were occasionally interrupted by material pertaining to other topics. Microstructural complexity was created by adding modifying information. An example passage and associated questions are presented in Table 1.</p> <p>To reduce the effect of prior knowledge on the ability to answer the questions, we created texts about topics that students are very unlikely to be familiar with (e.g., manrikis, ta moko, the McDonald Territory). To focus on students' ability to use language-processing skills to comprehend passages, we used words that students were likely to know, by creating four sets of passages at each of four Lexile levels (<reflink idref="bib790" id="ref1">790-890</reflink>, 890-940, 970-1010, 1000-1050), which are two years below the expected Lexile capacity of students in middle and high school.</p> <hd id="AN0112047212-8">Auditory Version</hd> <p>An auditory version of the battery was created. Test items were the same in both modalities. For subtests that required choices between a correct item and a foil, both of which were written words in the written version of the battery (tests 4 and 9), the target was presented only auditorily, and the correct response and the foil were presented in both written form and auditorily. For the morphological relatedness judgment test (test 8), the two stimuli were presented auditorily only. The auditory version of the written pseudohomophone judgment test does not test sublexical processing but rather is a second auditory lexical decision test. At the discourse level, half the passages were used in each modality, counterbalanced across participants.</p> <p>Procedures</p> <p>Testing was conducted in classrooms accommodating 30 students. Middle school students were tested in three 40-minute sessions, in which written tests 1-11, the corresponding auditory tests, and written and spoken discourse comprehension were tested (see the subsequent discussion). High school students were tested in two 60-minute sessions, one for the written tests and one for the auditory tests. Test sessions were separated by one week. In all sessions, tests of lexical processing preceded tests of morphological and sentence processing, and the order of tests within each level was counterbalanced across students. The order of written and auditory tests was counterbalanced across classes within grade. Students were told to respond as accurately and quickly as they could. Students were proctored by Boston University and retired school personnel.</p> <p>The stimulus items in the written modality were presented on the computer screen. The stimuli for the auditory modality were recorded by a male speaker in a soundproof booth and digitized at a sampling rate of 40 kHz with a 16-bit quantization by Praat (<ulink href="http://www.fon.hum.uva.nl/praat">www.fon.hum.uva.nl/praat</ulink>) or Sound Studio (felttip.com). The instructions for each test were digitized and presented both auditorily and in print. Students wore Logitech stereo headphones with adjustable headbands during the test to hear instructions and spoken stimuli.</p> <p>A series of practice items preceded each test. Students responded by pressing a designated key on the right side of the keyboard to indicate yes or to select the item on the right side of the screen and by pressing a key on the left side of the computer to indicate no or to select the item on the left side of the screen. In the written modality, the time from the presentation of the stimulus item until a student's response was timed in milliseconds. In the auditory modality, timing began at the offset of the stimulus item.</p> <p>For the test of passage comprehension, the sentences of the passage were presented one at a time and remained on the screen once they had been presented. When the entire passage had been read, 16 statements (eight factual, eight inferential) appeared one at a time. Students indicated whether each statement expressed information conveyed in the passage. The answer for half of the responses to each set of questions was true and for the other half false. The passage remained on the screen as the subject answered each statement. Processing time for reading each sentence in the passage and responding to each question was recorded. Accuracy of responses was also recorded. This presentation format emphasizes a student's ability to extract information from a passage, decreases the demands to remember that information, and corrects for rapid superficial reading by including RT in the measure of performance.</p> <hd id="AN0112047212-9">Results</hd> <p>We analyzed the data in many ways. We begin with an overview of our analyses as a guide for the reader and then present each type of analysis. We follow the widespread practice of presenting results of analyses in this section and reserving most discussion of what the results mean (their interpretation) for the Discussion section.</p> <p>We first examined basic psychometric properties of the data to determine whether they were reliable; subject to speed-accuracy trade-offs; and showed expected effects of grade, language parameters, and so forth. Once we were satisfied that the data were acceptable in these regards, we examined their structure in two ways. First, we performed factor analyses to see if performance on the tests of simple words, complex words, and sentences could be explained by a small number of hidden factors. We identified a small number of hidden factors that corresponded to skills in recognizing and understanding particular levels of language separately in the auditory and written modalities. Second, we performed structural equation modeling to see how the skills captured in these factors affected discourse comprehension. We found that effects of skills in processing both written and auditory language were important in predicting discourse comprehension.</p> <p>Basic Psychometric Analyses</p> <p>Accuracy and RT are shown in Tables 2 and 3, respectively. Data regarding construct validity, reliability, correlations between accuracy and RT, the effects of grade on performance, and the effects of the parameters within each test on performance are presented in Appendix B (available as supporting information for the online version of this article).</p> <p>To summarize these results, almost all correlations between performance on the tests of decoding, word recognition, and word comprehension in the battery and performance on the Test of Word Reading Efficiency word and nonword reading tests and the GRADE written comprehension test were significant. Split-half reliability (Pearson's r) was highly significant for all tests, and standardized Cronbach's a was above .75 for most tests. Accuracy increased and RT decreased with higher grades. There were expected effects of the parameters in all tests in both modalities of presentation in both middle and high school on both accuracy and RT. Correlations between accuracy and RT were generally nonsignificant, but there were several significant positive correlations, indicating that students sometimes traded speed for accuracy.</p> <p>The fact that accuracy and RT were largely uncorrelated measures of performance and that each showed expected effects of grade and linguistic parameters indicates that they reflect somewhat different aspects of the skills that we are trying to measure. The accuracy and RT measures also showed occasional trade-offs, which would affect analyses based on only one measure. We therefore combined accuracy and RT into a single measure. To do so, we first had to convert them, because they involve different units of measurement, into a common set of units. We accomplished this by transforming the accuracy and RT of each student to a z-score for accuracy or RT based on his or her grade. Because higher accuracy scores and lower RTs represent better performance, we considered higher accuracy and lower RT as producing more positive z-scores (i.e., we inverted the sign of the z-scores of the raw RTs). We then averaged the z-score for accuracy and the (inverted) z-score for RT, creating a measure in which higher accuracy and faster RT lead to a higher z-score. We called this combined measure the comprehensive z-score (COMP-Z).</p> <p>The results in Appendix B report reliability, the effects of grade on performance, and the effects of the parameters within each test on performance on COMP-Z. We also analyzed accuracy and RT separately for reliability, the effects of grade on performance, and the effects of the parameters within each test, with no significant changes in results. These analyses are available from the authors.</p> <p>Factor Analyses</p> <p>We used factor analysis to explore whether performance on the various tests that were administered is best described as due to each individual's capacity to perform each test or to a smaller number of capacities, each of which underlies performance on a number of tests. Factor analysis is a mathematical approach to determining the way a set of measures are related to one another. Factor analysis yields factors on which performance on different measures load. Factors are ordered according to how much variance they account for (lower numbers indicate the most variance). Factors that account for a significant amount of the variance on the measures are considered guides to the hidden structure of the measures. Because many factor analyses result in a first factor that explains a great deal of the variance, which is often considered as reflecting a general resource, factors are often rotated to identify the largest number of significant factors that can be related to aspects of a domain.</p> <p>There are two types of factor analysis: exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). EFA analyzes a data set without preconceived ideas about the number or nature of the factors that will result. The number of factors that are retained in an EFA is determined somewhat subjectively by the analyst's theoretical views and also by looking for a sharp drop in the amount of variance accounted for by later factors (this can be done by examining plots of the explained variance, known as Skree plots). CFA analyzes a data set postulating the number and sometimes the content of the factors in the analysis. These a priori constraints are derived from existing theories or from analysis of other data sets (including EFAs on similar data sets). There are established statistical goodness-of-fit criteria for determining whether a CFA fits the empirical data well.</p> <hd id="AN0112047212-10">EFAs</hd> <p>We performed EFAs with varimax rotation on the COMP-Z measures from the 2009 testing session. All students were analyzed. Analyses were done separately for middle school and high school. All analyses used all tests except the tests of discourse comprehension. These were omitted because the factors derived from the EFAs were the basis for creating CFAs of the 2011 and 2013 data, and we planned to use those CFAs to predict performance on the test of written discourse comprehension.</p> <p>The goal of the first set of EFAs was to determine whether performance on the auditory and written tests loaded on different factors, which would constitute evidence for separate skills in processing linguistic structures in the two modalities. Each EFA therefore analyzed the data from both the auditory and written tests. The results are shown in Table 4. With the exception of the tests of complex word-picture and sentence-picture matching in middle school, the tests presented in the auditory and written modalities loaded on separate factors. The second set of EFAs was designed to explore the hidden structure of skills in processing language (except the discourse level) in the auditory and written modalities. These analyses therefore analyzed the data from the auditory and written tests separately. The results are shown in Table 5. The four analyses yielded factors that had much in common and that grouped together tests that measured related psycholinguistic operations or involved the same tasks.</p> <p>In both middle and high school, in both modalities, the first factor contained significant loadings of the tests of decoding (in the written modality) and recognition of the forms of simple and morphologically complex words (a form recognition factor for simple and complex words). In middle school, in both modalities, a second factor contained significant loadings of the tests of comprehension of simple and morphologically complex words (a comprehension factor for simple and complex words), a third factor contained significant loadings of the tests of recognition of the form and comprehension of sentences (a sentence-processing factor), and a fourth factor contained significant loadings of the tests that required making judgments about how the meanings of simple and complex words were related (a word-meaning relatedness judgment factor). In high school, in the written modality, there was also a factor that contained significant loadings of the tests of comprehension of simple and morphologically complex words (a comprehension factor for simple and complex words) and a factor that contained significant loadings of the tests of recognition of the form and comprehension of sentences (a sentence-processing factor). In high school, in the auditory modality, the factor structure was more influenced by tasks than by language structures. A second factor contained significant loadings of the tests that required picture matching (a picture-matching factor) and a third factor that contained significant loadings of the test of complex word-meaning relatedness judgment and the test of judgment of well-formedness of sentences (a relatedness judgment factor).3</p> <p>These results provide preliminary answers to the first two questions raised in the introduction. The results show that processing language in the written and auditory modalities includes significantly separate skills in middle and high school and that each of these skills is divisible into separate abilities to process related aspects of language.</p> <hd id="AN0112047212-11">CFAs</hd> <p>We performed CFAs of the 2011 and 2013 COMP-Z measures based on the number of factors retained in the EFAs of the 2009 data. All of the students' results were used in these analyses. They were done separately for middle school and high school. The factors in the CFAs were used as the measurement model components of structural equation models (SEMs; see the subsequent discussion). For reasons of space, they are shown as part of the SEMs in Table 6 (see also Figures 1 and 2). They are also presented in Appendix B, along with the statistical measures of their goodness of fit.</p> <p>The best-fitting CFAs had slight changes from the factors in the EFAs. These CFAs were identical in the two testing sessions (2011 and 2013). In the auditory modality in middle school, the first factor (F1Aud) contained loadings of tests of lexical decision for simple and morphologically complex words (including the auditory version of the auditory pseudohomophone judgment test, which is a lexical decision test, as discussed previously) and was labeled Auditory Recognizing of Simple and Morphologically Complex Words. The second factor (F2Aud) contained loadings of word-picture matching and relatedness judgment for simple words and was labeled Auditory Understanding of Simple Words. The third factor (F3Aud) contained loadings of tests of sentence grammaticality judgment and sentence comprehension and was labeled Auditory Structuring and Understanding of Sentences.</p> <p>In the written modality in middle school, the first factor (F1Writ) contained loadings of tests of lexical decision for simple words and morphologically complex words and was labeled Written Recognizing of Forms of Simple and Morphologically Complex Words. The second factor (F2Writ) contained loadings of tests of word-picture matching for simple and morphologically complex words and relatedness judgment for simple words; it was labeled Written Understanding of Simple and Morphologically Complex Words. The third factor (F3Writ) contained loadings of tests of sentence grammaticality judgment and sentence comprehension and was labeled Written Structuring and Understanding of Sentences.</p> <p>The CFAs were identical for the auditory and auditory modalities in high school. In both modalities, the first factor (F1Aud, F1Writ) contained the same test loadings as in middle school. The second factor (F2Aud, F2Writ) contained the same test loadings as in the CFA of written tests in middle school and was labeled Auditory/Written Understanding of Simple and Morphologically Complex Words. The third factor (F3Aud, F3Writ) contained the tests of sentence processing, as in middle school, and also contained a loading of the test of meaning relatedness judgment of morphologically complex words; we labeled the third factor Auditory/Written Understanding of Morphologically Complex Words and Sentences.</p> <p>These results confirm the second finding on the EFAs, that language-processing skills in the written and auditory modalities are divisible into separate abilities to process related aspects of language.</p> <hd id="AN0112047212-12">SEMs</hd> <p>SEMs test models of how different factors influence one another. SEMs have two components: a measurement model component and a structural model component. The measurement model relates the factors in a SEM to the actual data. The measurement model component of a SEM often consists of factors in a CFA (called latent variables) because these factors capture the underlying, hidden structure of a domain and because they have appropriate statistical features, such as quantifiable error terms. The structural component of a SEM relates the factors (latent variables) to each other and to an outcome measure. It is possible to compare SEMs that specify different measurement and structural models using such measures such as the Akaike information criterion (AIC).</p> <p>We used SEMs to test different models of how skills at the lexical, morphological, and sentence levels of language in the auditory and written modalities were related to performance on the written discourse comprehension test. As mentioned previously, we used the factors of the CFAs as the measurement model component of the SEMs. We used performance (COMP-Z) on the written discourse comprehension test as the measure of comprehension of written passages (similar results were obtained when we used accuracy alone as the measure of written passage comprehension). We specified several structural models and compared them for goodness of fit. Figures 1 and 2 and Tables 6-8 show the SEMs with the best fits.</p> <p>All the structural models specified relations (paths) among the factors in the CFAs that reflect basic aspects of language processing. We always specified a path from the factor reflecting recognizing simple and morphologically complex words (F1) to the factors reflecting understanding simple and morphologically complex words (F2) and structuring and understanding sentences (F3), and a path from the factor reflecting understanding simple and morphologically complex words (F2) to the factor reflecting structuring and understanding sentences (F3). Based on the fact that language develops naturally and initially in the auditory-oral modality, and on empirical evidence that the development of written language comprehension depends on the development of auditory language comprehension, we also specified a path from each factor in the auditory modality to the corresponding factor in the written modality.</p> <p>In addition to the features that were constant in all the SEMs we ran, we also varied aspects of the models to test hypotheses about how language skills influenced one another and affected written discourse comprehension. We ran four models, each separately for middle and high school and separately for the 2011 and 2013 data (16 SEMs in all). Two models (models 1 and 2) contained paths from each auditory factor (F1Aud, F2Aud, F3Aud) to each written factor (F1Writ, F2Writ, F3Writ), as discussed previously. In two models (models 3 and 4), each written factor (F1Writ, F2Writ, F3Writ) also had a path to the corresponding auditory factor (F1Aud, F2Aud, F3Aud). In both cases, we ran one model that specified paths from each written factor to written discourse comprehension (models 1 and 3) and did not specify such paths from auditory factors, and one model that specified paths from both the auditory and written factors to written discourse comprehension (models 2 and 4).</p> <p>The AICs for the two models with paths from written factors to the corresponding auditory factors (models 3 and 4) were higher than those for models without these paths (models 1 and 2). We therefore did not consider models with paths from written factors to the corresponding auditory factors further. Table 7 shows the AIC values for the two models without these paths (models 1 and 2). For the middle school data in 2011, the AIC statistic showed that the most highly valued model only included paths from factors in the written modality, not the auditory modality, to written discourse comprehension (model 1). For the remaining three models, the AIC statistic showed that the most highly valued model specified paths from each factor in both the auditory and written modalities to written discourse comprehension (model 2). The diagram of the SEM in Figure 1 shows model 1, and the diagram of the SEM in Figure 2 shows model 2.</p> <p>The goodness-of-fit indexes for the best-fitting models were acceptable (see Table 7). There are two types of goodness-of-fit indexes: absolute and incremental. Absolute goodness-of-fit indexes determine how well an a priori model fits the sample data compared with no model at all, and incremental indexes compare the model with a baseline model. The absolute goodness-of-fit indexes were within recommended ranges. The chi-square p-value was below .05. The root mean square error of approximation was 0.07 or 0.08, as recommended by MacCallum, Browne, and Sugawara (1996) and Steiger (2007), and the standardized root mean square residual was 0.04 or 0.05, as recommended by Byrne (1998). Two incremental measures of goodness of fit were close to recommended values. The comparative fit index was 0.93 or 0.94, just below the recommended 0.95 (Fan, Thompson, & Wang, 1999; Hu & Bentler (1999) accept values > 0.9), and the Tucker-Lewis index ranged from 0.90 to 0.92 (the recommended value is ≥ 0.95; Hu & Bentler, 1999), although values as low as 0.8 have been said to be acceptable (Byrne, 1998).</p> <p>Table 8 shows the significant paths in the best-fitting models. All four best-performing models showed expected relations among the factors. In the auditory modality, in both middle and high school, the factor reflecting recognizing simple and morphologically complex words (F1Aud) had a direct positive effect on the factor reflecting comprehending simple and morphologically complex words (F2Aud), and the factor reflecting comprehending simple and morphologically complex words (F2Aud) had a direct positive effect on the factor reflecting structuring and understanding sentences (F3Aud). The factor reflecting comprehending simple and morphologically complex words (F2Aud) had a positive effect on the factor reflecting structuring and understanding sentences (F3Aud) in high school. In the written modality, the picture was similar. The factor reflecting recognizing simple and morphologically complex words (F1Writ) had direct positive effects on the factor reflecting comprehending simple and morphologically complex words (F2Writ) and the factor reflecting structuring and understanding sentences (F3) in both grade levels in both test sessions. The factor reflecting comprehending simple and morphologically complex words (F2Writ) had a direct positive effect on the factor reflecting structuring and understanding sentences (F3) except in middle school in 2011. Paths from auditory factors to the corresponding written factors (F1Aud→ F1Writ, F2Aud→ F2Writ, F2Aud→ F2Writ) were significantly positive in both middle and high school in 2011 and 2013 except for F3Aud→ F3Writ in middle school in 2013.</p> <p>With respect to the relation of skills in written and auditory language processing to written passage comprehension, as noted previously, for the 2011 test session in middle school, the AIC was lowest for the model that did not include paths from the auditory factors to written passage comprehension (model 1). In this model, the paths from the factor reflecting comprehending simple and morphologically complex words in the written modality (F2Writ) and from the factor reflecting structuring and comprehending sentences in the written modality (F3Writ) to written discourse comprehension were significant.</p> <p>For the remaining grade levels and test sessions, the AIC showed that the model that included paths from both auditory and written factors to written passage comprehension (model 2) was most highly valued, and paths from both auditory and written factors to written passage comprehension were significant. In these three models, the path from the factor reflecting recognizing auditory simple and morphologically complex words (F1Aud) to written discourse comprehension was significantly negative. The path from the factor reflecting auditory comprehension of simple and morphologically complex words (F2Aud) to written discourse comprehension was significantly positive in middle school in 2013. The path from the factor reflecting auditory structuring and understanding sentences (F3Aud) to written discourse comprehension was significantly positive in high school in both testing sessions. The path from the factor reflecting written recognition of simple and morphologically complex words (F1Writ) to written discourse comprehension was significant in middle school and in high school in 2013. The path from the factor reflecting written comprehension of simple and morphologically complex words (F2Writ) to written discourse comprehension was significant in high school in both testing sessions. The path from the factor reflecting auditory structuring and understanding sentences (F3Writ) to written discourse comprehension was significant in middle school in 2013 and in high school in 2011.</p> <p>These results provide an answer to the third question in the introduction to this article. They show that in middle and high school, the ability to process language in the written modality contributes significantly to comprehension of written passages in a direct manner, independent of the ability to process language in the auditory modality.</p> <hd id="AN0112047212-13">EFAs, CFAs, and SEMs of Accuracy and RT Data</hd> <p>We performed the EFAs, CFAs, and SEMs described previously separately on the accuracy and RT data. The results of the SEMs are presented in Appendix B; significant paths between factors and performance on written passage comprehension are highlighted for the reader's convenience. The results were very similar to those of the EFAs, CFAs, and SEMs performed on COMP-Z. There were some changes in how subtests loaded on factors. The most significant of these is that the best solutions for CFAs and SEMs of RT in middle school retained only two factors, not three. The goodness-of-fit measures of the analyses using accuracy or RT were not as good as those using COMP-Z. This would be expected if accuracy and RT both reflect students' abilities to process language but in somewhat different ways, and it provides an empirical reason to use COMP-Z in the analyses.</p> <p>With respect to the central question of the relation of skills in processing the lexical, morphological, and sentence levels of language in the written and auditory modalities on written passage comprehension, there were significant direct positive effects of performance on factors that reflected skills in processing written morphological and sentence-level structures on written discourse comprehension. In middle school, the factors that reflected accuracy of comprehension of written simple and morphologically complex words and accuracy of processing of written sentences predicted accuracy of written passage comprehension in 2011 and 2013, respectively, and the factor that reflected RT of processing of written sentences predicted RT of written passage comprehension in both 2011 and 2013. In high school, the factor that reflected accuracy of processing of written sentences predicted accuracy of written passage comprehension in both 2011 and 2013, and the factor that reflected RT of processing of written sentences predicted RT of written passage comprehension in both 2011 and 2013. The factor that reflected RT of recognizing written simple and morphologically complex words also predicted RT of written passage comprehension in 2013.</p> <p>There were also significant direct effects of processing auditory language on written passage comprehension. In middle school, accuracy of processing of auditory morphologically complex words and auditory sentences predicted accuracy of written passage comprehension in both 2011 and 2013, and RT of processing of auditory morphologically complex words and auditory sentences predicted RT of written passage comprehension in 2013. In high school, accuracy of processing of auditory sentences predicted accuracy of written passage comprehension in both 2011 and 2013, and RT of processing of auditory sentences predicted RT of written passage comprehension in 2013.</p> <p>The separate analyses of accuracy and RT showed that the negative effects of the ability to recognize auditory simple and morphologically complex words on written discourse comprehension seen in the analysis of COMP-Z arose in the accuracy data, not the RT data.</p> <p>These results are consistent with the conclusion drawn from the analysis of COMP-Z. They show that in middle and high school, whether measured in terms of accuracy or RT, skills in processing language in the written modality contribute significantly to comprehension of written passages in a direct manner, independent of skills in processing language in the auditory modality.</p> <hd id="AN0112047212-14">Discussion</hd> <p>We first discuss the three theoretical questions that we posed at the beginning of this article and then turn to practical implications of this study. We first asked if assigning linguistic structure and meaning constitute one integrated skilled ability or consist of a set of separate operations that are related to the structure of language as described by linguists. The results favor the second model. EFAs yielded factors that had loadings of tests of similar language operations or of tests that involved the same task. CFAs yielded a first factor reflecting recognition of simple and morphologically complex words, a second factor representing comprehension of simple and morphologically complex words, and a third factor representing structuring and understanding of sentences (in high school, one test of comprehension of morphologically complex words also loaded on this third factor). Skills in recognizing simple and morphologically complex words predicted skills in understanding those linguistic items, and skills in both recognizing and understanding simple and morphologically complex words predicted skills in structuring and understanding sentences. These results indicate that there are separate abilities for recognizing and understanding linguistic representations at the lexical, morphological, and sentence levels and that these skills are related to one another, as theories of language processing lead one to expect.</p> <p>Second, we asked to what extent these basic psycholinguistic skills differ in the auditory and written modalities. The EFAs showed that auditory and written tests loaded on separate factors. This indicates that there are separate processing systems for auditory and written language. The CFAs showed that skills develop to process similar levels of language in the two modalities.</p> <p>The results also document the existence of relations between skills in language processing in the two modalities. They indicate that each type of skill (recognizing simple and morphologically complex words, understanding simple and morphologically complex words, structuring and understanding sentences) in the auditory modality predicts the skill in processing the corresponding types of language representations in the written modality. The largest effect of skills in processing auditory language on processing written language was found in recognizing simple and complex words, with progressively less of an effect on comprehension of simple and complex words and processing sentences. There was only a small effect of auditory sentence processing on written sentence processing in high school. These results are consistent with the view that written vocabulary is significantly determined by spoken vocabulary and that skills in processing higher levels of written language, especially sentences, become independent of skills in processing the corresponding representations in the auditory modality.</p> <p>The third question was whether skills in recognizing and understanding linguistic elements and structures in the written modality have an independent effect on the ability to understand written passages. The results of the SEMs point to an independent role for these skills on this ability. In middle school in 2011, the best-fitting model only showed effects of skills in recognizing and understanding linguistic elements and structures in the written modality on passage comprehension; the model that also specified effects of skills in processing these representations in the auditory modality fit the data less well. The remaining three best-fitting models contained significant paths from factors reflecting skills in processing language in the auditory modality to passage comprehension, but there were also significant paths from factors reflecting skills in processing language in the written modality to comprehension of written discourse. We conclude that by middle school, the ability to process the lexical, morphological, and sentence levels of language in the written modality constitutes a set of skills that is partially separate from skills in processing these levels of language in the auditory modality, and that these skills are used to a significant extent in understanding written passages.</p> <p>The results provide evidence that the simple model of reading is too simple: Skills in comprehending language emerge in the written modality and play an important role in understanding written passages. This conclusion does not imply that skills in processing these levels of language in the auditory modality have no effect on written passage comprehension. Skills in understanding simple and complex words in the auditory modality (F2Aud) had positive effects in middle school in 2013, and skills in structuring and understanding sentences (F3Aud) had positive effects in middle and high school in 2013.</p> <p>Skills in recognizing auditory simple and complex words (F1Aud) had negative direct effects on written passage comprehension (path coefficients of -0.63, -0.25, and -0.37 for high school 2011, middle school 2013, and high school 2013, respectively). F1Aud had a positive indirect effect on written passage comprehension that was mediated through F1Writ, but the total direct and indirect effect of F1Aud on written passage comprehension was still negative (-0.45, -0.14, and -0.23 for high school 2011, middle school 2013, and high school 2013, respectively). Separate analyses of accuracy and RT showed that the negative relation between recognizing auditory simple and complex words and comprehending written discourse arose for the accuracy with which students performed these tests, not for the time they took to perform them.</p> <p>This unexpected result may be partially explained along the following lines. The tests of recognizing auditory simple and complex words (F1Aud) differ from the other tests in the battery in that they involve a functional ability that is not used in written passage comprehension: the ability to recognize simple and complex words from an auditory stimulus.4 Because the written passages were presented visually, perceptual identification of an auditory stimulus is not required--indeed, not even possible--in written passage comprehension. The other factors that included auditory tests--understanding simple and complex words in the auditory modality (F2Aud) and structuring and understanding sentences (F3Aud)--involve not only perceptual identification of simple language forms (words and morphemes) but also comprehension (accessing meaning from form) and a combination of those forms and meanings into higher level entities (morphologically complex words and sentences). These operations can be used in understanding written texts once written words are transcoded into phonological representations; the simple model of reading, which we are testing, claims that these are the operations that support reading comprehension. The fact that the operations that underlie F1Aud are not used in comprehension of written texts explains why the level of skill in these tasks does not predict written passage comprehension. It remains unclear, however, why the relation between accuracy of performance on F1Aud and written passage comprehension was negative. One speculation is that this might reflect reciprocal development of skills in auditory processing, including spoken word recognition, and visual processing. This result requires further study to understand.</p> <p>There are several limitations of this study that need to be mentioned and could be addressed by additional research. One is that the testing of the ability to process lexical, morphological, and sentence levels of language, although comprehensive, systematic, and informed by linguistic and psycholinguistic studies, was limited by standards applied in experimental psychology. We note that it is not possible to test middle and high school students for these abilities as extensively as psychologists test undergraduates in experimental studies, where a single experiment often focuses on a very specific effect, involves hundreds of trials, and takes 30-60 minutes to run. A battery that used enough tests and enough items on each test to thoroughly assess processing lexical, morphological, and sentence levels of language could easily involve days of testing. Although it is not feasible to obtain this degree of detail regarding these abilities in schools, replication of the results using a battery of similar length to ours that employed other tests of these abilities is feasible, and results of such a study would be valuable to see if the results reported here replicate when using different measures.</p> <p>A second limitation of this study is that the texts used here had several particular features. Four of these require discussion: the length of the passages, their subject matter, their vocabulary level, and the fact that they remained on the screen while students answered questions about them.</p> <p>The texts were very short (150 words). The length of the passages was chosen for reasons of practicality; students each read or heard four passages, and we had limited testing time. However, this feature might affect the relative importance of the skills that we tested compared with other skills. In particular, longer texts increase the extent to which readers maintain and retrieve information in short-term/working memory (Kintsch & van Dijk, 1978). This could reduce the percentage of variance in written discourse comprehension that is accounted for by skills in processing lexical, morphological, and sentence levels of language because more variance in comprehension is related to variance in short-term/ working memory in longer texts. However, even if that is the case, it would indicate that other skills (particularly short-term/working memory) are also needed to comprehend longer texts, not that the skills measured here are not important to understand longer texts.</p> <p>The same is true of the fact that the passages remained on display while the questions were presented; if the text disappeared when the questions appeared, the test would make more demands on memory. We elected to display the passages while the questions were presented to focus on what students could understand in a passage, not what they could remember. Measuring RT on the task captures the use of a strategy of skimming the text when it was originally presented and rereading it to answer the questions, which redistributes the time spent on the test (less time is spent on initial reading and more when the questions are presented).</p> <p>We created passages about unfamiliar subjects to not allow students' prior knowledge about topics to influence passage comprehension. The use of hard passages changes readers' approaches to comprehension (Rubin, Hafer, & Arata, 2000) but does not eliminate the difference between good and poor readers' comprehension of written texts (Horowitz & Samuels, 1985). It is likely that once they have developed, the skills measured here are used automatically to understand propositions in all written passages regardless of their familiarity. From a practical point of view, written texts that middle and high school students encounter deal with both partially familiar and relatively unfamiliar subjects, and the results here likely apply to a large number of materials that students have to have to read and understand.</p> <p>The passages contained words with Lexile levels two grades below students' grade level. We used a low Lexile vocabulary level to be able to assess morphological and sentence-level processing; if the students did not understand many words in the passages, this would directly affect their ability to answer questions about the passages. The use of familiar vocabulary might reduce the use of auditory language-processing skills. Having to recognize unfamiliar words would be expected to lead to more use of decoding, which makes phonological representations available and might lead to more use of skills in processing auditory language input. However, even if a reader sounds out unfamiliar words and this leads to using auditory language-processing routines, he or she is still likely to use skills in processing written language when dealing with familiar words.</p> <p>Overall, the role of basic psycholinguistic skills in the auditory and written modalities in comprehension may differ in different types of written texts. This requires more study. In addition, the skills studied here would be expected to have less of an overall effect on tasks that are more distant from immediate comprehension, such as the ability to produce a verbal or written description of the passage and to recall a passage several weeks or months after having read it. Although poor skills in processing the levels of language would be expected to be associated with poor performance on such tasks because they lead to limited initial comprehension, the beneficial effects of higher skill levels in language processing might be offset by lesser abilities in memory, verbal expression, and other cognitive functions. Another way to put this is to say that language comprehension is only one factor that affects retention and use of what has been understood. However, despite all of these caveats, the findings reported here are relevant to comprehension of a wide range of texts to which students in middle and high school are exposed.</p> <p>The results of this study are therefore potentially relevant to educational practice. The finding that skills in recognizing and understanding morphologically complex words and structuring and understanding sentences account for a significant amount of variance in understanding written texts suggests that improving these skills may lead to higher levels of literacy. The fact that there is a separate effect of skills in recognizing and understanding morphologically complex words and structuring and understanding sentences presented in the written modality suggests that this is true for skills in this modality and for skills in processing auditory language.</p> <p>However, more work needs to be done to know whether and how to leverage the findings here to achieve the goal of better academic performance. Other factors--some cognitive (e.g., memory) and others more personal (interest, motivational level)--account for much variance in school performance. That said, improving the ability to process language representations may have significant educational benefits. There may even be positive interactions between better language-processing skills and other factors that determine school performance. If students find it easier to understand the texts that they encounter, they may become more interested in the subject matter of these texts and more motivated to study these subjects.</p> <p>There are many important questions about how best to improve these skills. Some focus on the type of materials that could be used. Students could practice with items presented in isolation, as in the battery, or engage in linguistic analysis of sentences or in using morphology to determine the meaning of words. Others pertain to how materials are presented. Modern computer capabilities allow for many presentation formats, such as variable presentation speed of materials and subtitling of videos. Techniques for enhancing the skills assessed here are available (updated reports on these methods are presented on the What Works Clearinghouse website: ies.ed.gov/ncee/wwc). Another question is when to teach these skills: in early grades, as these skills are developing in the auditory modality, or in middle and high school? Yet another is how much instruction aimed at improving these skills should be part of the regular curriculum rather than occur in isolation (see Adams, 1994, for suggestions along these lines).</p> <p>Our research does not extend to these important practical matters. These questions about whether and how to invest time and resources into improving skills in auditory and written language processing will require more research to answer. The results here indicate that educators should be aware that reading comprehension requires more than efficient decoding, good auditory language-processing skills, and good higher order cognitive skills such as inferencing ability, executive functions, and metacognition. Educators should be aware that reading comprehension involves efficient skills in processing higher order language representations in both the written and auditory modalities. This indicates that educators should attempt to evaluate these skills in their students and consider training these skills to improve written passage comprehension and thereby possibly achieve better school performance.</p> <hd id="AN0112047212-15">NOTES</hd> <p>The research reported here was supported by a grant from the Institute for Educational Sciences (IES/DoE R305A100261) to Gloria Waters and David Caplan.</p> <p>1 The gap in numbering is due to us having developed another test that was dropped for various reasons. Many analyses had been done before dropping that test, and we maintained the original test numbers for consistency in our own records.</p> <ulist> <item>2 High-frequency words had frequencies of 100/million or higher, and low-frequency words had frequencies of 10/million or lower (Kucera & Francis, 1967).</item> <item>3 The test of relatedness of morphological form (test 8 in Table 1) did not load well on any factor. The loadings of the tests on the factors make it clear why this was the case. Test 8 involves judgments of both form and meaning, and the factors in the EFAs separate these dimensions.</item> <item>4 Note that the function reflected in F1Aud is not decoding (the ability to transform a written word into a phonological form) but the ability to recognize simple and complex words when they are presented auditorily.</item> </ulist> <hd id="AN0112047212-16">TABLE 1 Sample Stimulus Items From Each Task</hd> <p> <ephtml> <div class="table-size-normal table-border"><table border="1"> <tr> <td>Level of language</td> <td>Test</td> <td>Sample stimulus</td> <td>Stimulus variables manipulated</td> </tr> <tr> <td colspan="4">1. Sublexical elements</td> </tr> <tr> <td></td> <td rowspan="2">1. Decoding (pseudohomophone judgment): "Does this sound like a real word in English?"</td> <td>nues (yes)</td> <td>* Frequency of word derived from</td> </tr> <tr> <td></td> <td>spage (no)</td> <td>* Vowel vs. consonant change</td> </tr> <tr> <td colspan="4">2. Lexical items</td> </tr> <tr> <td rowspan="3">Lexical access</td> <td rowspan="3">2. Simple word recognition (lexical decision: simple words): "Is this a real word in English?"</td> <td>choice (yes)</td> <td>* Frequency</td> </tr> <tr> <td>bruth (no)</td> <td>* Spelling-sound regularity</td> </tr> <tr> <td></td> <td>* Spelling-sound consistency</td> </tr> <tr> <td rowspan="4">Lexical comprehension</td> <td rowspan="2">3. Word-picture matching (simple words): "Which of two pictures matches the printed word?"</td> <td rowspan="2">ship</td> <td>* Frequency</td> </tr> <tr> <td>* Spelling-sound consistency</td> </tr> <tr> <td rowspan="2">4. Relatedness judgment (simple words): "Which of two words is most related to the top word?"</td> <td>race</td> <td>* Frequency</td> </tr> <tr> <td>run walk</td> <td>* Spelling-sound consistency</td> </tr> <tr> <td colspan="4">3. Morphology</td> </tr> <tr> <td rowspan="2">Morphology access</td> <td rowspan="2">5. Complex word recognition (lexical decision: affixed words): "Is this a real word in English?"</td> <td>nationality (yes)</td> <td>* Word boundary vs. formative boundary</td> </tr> <tr> <td>lifement (no)</td> <td>* Derivational vs. inflectional</td> </tr> <tr> <td rowspan="6">Morphology comprehension</td> <td>7. Complex word-picture matching: "Which of two pictures matches the printed word?"</td> <td>cleans</td> <td>* Inflectional vs. derivational</td> </tr> <tr> <td rowspan="3">8. Morphological form relatedness judgment: "Are these two words related in meaning?"</td> <td>civil civility (yes)</td> <td>* Neutral vs. vowel change</td> </tr> <tr> <td>late lateral (no)</td> <td>* Consonant change</td> </tr> <tr> <td></td> <td>* Silent letter</td> </tr> <tr> <td rowspan="2">9. Meaning relatedness judgment (complex words): "Which of the bottom words is most related in meaning to the top word?"</td> <td>conditional</td> <td></td> </tr> <tr> <td>dependent depends</td> <td></td> </tr> <tr> <td colspan="4">4. Sentences</td> </tr> <tr> <td rowspan="2">Syntactic processing</td> <td rowspan="2">10. Grammaticality judgment: "Is this sentence grammatical in English?"</td> <td>The boy pushed the girl, (yes)</td> <td>* Syntactic complexity</td> </tr> <tr> <td>The boy was pushed the girl. (no)</td> <td></td> </tr> <tr> <td>Sentence comprehension</td> <td>11. Sentence-picture matching: "Which of two pictures matches the printed sentence?"</td> <td>The girl was tickled by the boy.</td> <td>* Syntactic complexity</td> </tr> <tr> <td colspan="4">5. Discourse</td> </tr> <tr> <td rowspan="3">Discourse comprehension</td> <td>Passage reading</td> <td></td> <td>* Easy vs. hard macrostructure</td> </tr> <tr> <td>Question answering</td> <td></td> <td>* Lexile level</td> </tr> <tr> <td></td> <td></td> <td>* Factual vs. inferential questions</td> </tr> <tr> <td colspan="4">Sample texts and comprehension questions</td> </tr> <tr> <td colspan="4">Easy macrostructure</td> </tr> <tr> <td colspan="4">"Ta moko is the permanent body and face marking of the Maori, the native people of New Zealand. It is different from tattoo in that the skin is carved with a chisel rather than punctured by a needle. This leaves the skin with grooves rather than a smooth surface. Moko on the face of the men served to identify the man based on his social status. The original chisels were made from bones which were fit with a handle. To carve deep marks in the face the chisels were struck by a mallet. Then the tools would be dipped into a colored skin dye made from vegetable caterpillar for the body color or burnt timbers for the face. Finally the dye was tapped into the skin. The process was very long and painful. Leaves were often placed over the cuts to speed the healing process."</td> </tr> <tr> <td colspan="4">Hard macrostructure</td> </tr> <tr> <td colspan="4">"Ta Moko leaves the skin with grooves rather than a smooth surface. It is different from tattoo in that the skin is carved with a chisel rather than punctured by a needle. Leaves were often placed over the cuts to speed the healing process. To carve deep marks in the face the chisels were struck by a mallet. This permanent body and face marking was performed by the Maori, the native people of New Zealand. Ta Moko on the face of the men served to identify the man based on his social status. The original chisels were made from bone which were fit with a handle. The tools would be dipped into a colored skin dye made from vegetable caterpillar for the body color or burnt timbers for the face. Finally the dye was tapped into the skin. The process was very long and painful."</td> </tr> <tr> <td colspan="4">Questions</td> </tr> <tr> <td rowspan="8">True</td> <td rowspan="4">Factual</td> <td colspan="2">"The Maori are from New Zealand."</td> </tr> <tr> <td colspan="2">"Different dyes were used for body and face."</td> </tr> <tr> <td colspan="2">"Ta Moko is permanent."</td> </tr> <tr> <td colspan="2">"AMoko marks the face and the body."</td> </tr> <tr> <td rowspan="4">Inference</td> <td colspan="2">"Ta Moko tools have since been made with material other than bone."</td> </tr> <tr> <td colspan="2">"Ta Moko was an important element of Maori Culture."</td> </tr> <tr> <td colspan="2">"Men with similar social status probably had similar Moko placement."</td> </tr> <tr> <td colspan="2">"Leaves were believed to have healing properties."</td> </tr> <tr> <td rowspan="8">False</td> <td rowspan="4">Factual</td> <td colspan="2">"Ta Moko is a tattoo."</td> </tr> <tr> <td colspan="2">"Ta Moko had no purpose other than to look pretty."</td> </tr> <tr> <td colspan="2">"Some Ta Moko dye is made from bone."</td> </tr> <tr> <td colspan="2">"Ta Moko leaves the skin smooth."</td> </tr> <tr> <td rowspan="4">Inference</td> <td colspan="2">"A person experiences no pain when receiving Ta Moko."</td> </tr> <tr> <td colspan="2">"Men and women are treated the same in Maori culture."</td> </tr> <tr> <td colspan="2">"A Maori male would be ashamed of his Ta Moko."</td> </tr> <tr> <td colspan="2">"The best way to describe the process of Ta Moko is 'irritating.'"</td> </tr> </table></div> </ephtml> </p> <p>Note. There is no test 6.</p> <hd id="AN0112047212-17">TABLE 2 Mean Accuracy (percentage correct) by School and Testing Session</hd> <p> <ephtml> <div class="table-size-normal table-border"><table border="1"> <tr> <td rowspan="3">Test</td> <td colspan="4">Session 1, 2009</td> <td colspan="4">Session 2, 2011</td> <td colspan="4">Session 3, 2013</td> </tr> <tr> <td colspan="2">Middle school</td> <td colspan="2">High school</td> <td colspan="2">Middle school</td> <td colspan="2">High school</td> <td colspan="2">Middle school</td> <td colspan="2">High school</td> </tr> <tr> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> </tr> <tr> <td colspan="13">Written</td> </tr> <tr> <td>Pseudohomophone judgment</td> <td>72</td> <td>14</td> <td>77</td> <td>15</td> <td>69</td> <td>12</td> <td>74</td> <td>12</td> <td>66</td> <td>13</td> <td>72</td> <td>13</td> </tr> <tr> <td>Lexical decision: SW</td> <td>84</td> <td>9</td> <td>87</td> <td>8</td> <td>85</td> <td>9</td> <td>89</td> <td>8</td> <td>81</td> <td>10</td> <td>85</td> <td>10</td> </tr> <tr> <td>Word-picture matching: SW</td> <td>88</td> <td>5</td> <td>89</td> <td>6</td> <td>89</td> <td>5</td> <td>90</td> <td>5</td> <td>90</td> <td>6</td> <td>91</td> <td>7</td> </tr> <tr> <td>Relatedness judgment: SW</td> <td>88</td> <td>8</td> <td>91</td> <td>6</td> <td>88</td> <td>8</td> <td>92</td> <td>6</td> <td>87</td> <td>9</td> <td>89</td> <td>10</td> </tr> <tr> <td>Lexical decision: MCW</td> <td>75</td> <td>11</td> <td>80</td> <td>9</td> <td>75</td> <td>11</td> <td>81</td> <td>9</td> <td>70</td> <td>11</td> <td>77</td> <td>11</td> </tr> <tr> <td>Word-picture matching: MCW</td> <td>82</td> <td>11</td> <td>87</td> <td>9</td> <td>83</td> <td>11</td> <td>90</td> <td>9</td> <td>82</td> <td>11</td> <td>87</td> <td>10</td> </tr> <tr> <td>Morphological relatedness judgment</td> <td>82</td> <td>13</td> <td>88</td> <td>11</td> <td>83</td> <td>13</td> <td>89</td> <td>9</td> <td>77</td> <td>14</td> <td>85</td> <td>13</td> </tr> <tr> <td>Relatedness judgment: MCW</td> <td>72</td> <td>14</td> <td>77</td> <td>13</td> <td>72</td> <td>14</td> <td>79</td> <td>14</td> <td>69</td> <td>14</td> <td>75</td> <td>15</td> </tr> <tr> <td>Grammaticality judgment</td> <td>74</td> <td>13</td> <td>77</td> <td>12</td> <td>73</td> <td>13</td> <td>79</td> <td>12</td> <td>69</td> <td>13</td> <td>75</td> <td>15</td> </tr> <tr> <td>Sentence-picture matching</td> <td>80</td> <td>12</td> <td>85</td> <td>10</td> <td>81</td> <td>11</td> <td>88</td> <td>10</td> <td>79</td> <td>13</td> <td>84</td> <td>15</td> </tr> <tr> <td>Written discourse</td> <td>70</td> <td>10</td> <td>73</td> <td>10</td> <td>72</td> <td>11</td> <td>76</td> <td>11</td> <td>67</td> <td>11</td> <td>69</td> <td>12</td> </tr> <tr> <td colspan="13">Auditory</td> </tr> <tr> <td>Pseudohomophone judgment</td> <td>90</td> <td>8</td> <td>91</td> <td>7</td> <td>86</td> <td>8</td> <td>88</td> <td>7</td> <td>83</td> <td>11</td> <td>86</td> <td>10</td> </tr> <tr> <td>Lexical decision: SW</td> <td>79</td> <td>8</td> <td>83</td> <td>7</td> <td>80</td> <td>8</td> <td>84</td> <td>8</td> <td>78</td> <td>10</td> <td>82</td> <td>11</td> </tr> <tr> <td>Word-picture matching: SW</td> <td>88</td> <td>5</td> <td>89</td> <td>5</td> <td>88</td> <td>6</td> <td>90</td> <td>6</td> <td>89</td> <td>7</td> <td>89</td> <td>10</td> </tr> <tr> <td>Relatedness judgment: SW</td> <td>87</td> <td>8</td> <td>90</td> <td>8</td> <td>86</td> <td>9</td> <td>91</td> <td>7</td> <td>84</td> <td>11</td> <td>87</td> <td>11</td> </tr> <tr> <td>Lexical decision: MCW</td> <td>76</td> <td>10</td> <td>81</td> <td>8</td> <td>76</td> <td>10</td> <td>82</td> <td>9</td> <td>71</td> <td>11</td> <td>78</td> <td>12</td> </tr> <tr> <td>Word-picture matching: MCW</td> <td>84</td> <td>11</td> <td>87</td> <td>9</td> <td>84</td> <td>11</td> <td>89</td> <td>10</td> <td>81</td> <td>12</td> <td>84</td> <td>14</td> </tr> <tr> <td>Morphological relatedness judgment</td> <td>82</td> <td>12</td> <td>87</td> <td>10</td> <td>82</td> <td>12</td> <td>88</td> <td>10</td> <td>77</td> <td>13</td> <td>83</td> <td>13</td> </tr> <tr> <td>Relatedness judgment: MCW</td> <td>71</td> <td>14</td> <td>75</td> <td>15</td> <td>70</td> <td>14</td> <td>77</td> <td>14</td> <td>69</td> <td>14</td> <td>75</td> <td>15</td> </tr> <tr> <td>Grammaticality judgment</td> <td>68</td> <td>9</td> <td>71</td> <td>10</td> <td>70</td> <td>10</td> <td>74</td> <td>11</td> <td>65</td> <td>11</td> <td>70</td> <td>13</td> </tr> <tr> <td>Sentence-picture matching</td> <td>83</td> <td>12</td> <td>87</td> <td>10</td> <td>83</td> <td>11</td> <td>88</td> <td>10</td> <td>81</td> <td>14</td> <td>83</td> <td>15</td> </tr> <tr> <td>Auditory discourse</td> <td>69</td> <td>9</td> <td>71</td> <td>10</td> <td>71</td> <td>10</td> <td>74</td> <td>10</td> <td>68</td> <td>10</td> <td>69</td> <td>11</td> </tr> </table></div> </ephtml> </p> <p>Note. M = mean; MCW = morphologically complex words; SD = standard deviation; SW = simple words.</p> <hd id="AN0112047212-18">TABLE 3 Mean Reaction Time (msec) by School and Testing Session</hd> <p> <ephtml> <div class="table-size-normal table-border"><table border="1"> <tr> <td rowspan="3">Test</td> <td colspan="4">Session 1, 2009</td> <td colspan="4">Session 2, 2011</td> <td colspan="4">Session 3, 2013</td> </tr> <tr> <td colspan="2">Middle school</td> <td colspan="2">High school</td> <td colspan="2">Middle school</td> <td colspan="2">High school</td> <td colspan="2">Middle school</td> <td colspan="2">High school</td> </tr> <tr> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> <td>M</td> <td>SD</td> </tr> <tr> <td colspan="13">Written</td> </tr> <tr> <td>Pseudohomophone judgment</td> <td>2,202</td> <td>704</td> <td>2,082</td> <td>645</td> <td>2,121</td> <td>656</td> <td>1,948</td> <td>582</td> <td>1,895</td> <td>622</td> <td>1,584</td> <td>532</td> </tr> <tr> <td>Lexical decision: SW</td> <td>1,280</td> <td>354</td> <td>1,189</td> <td>325</td> <td>1,301</td> <td>404</td> <td>1,118</td> <td>301</td> <td>1,212</td> <td>302</td> <td>983</td> <td>237</td> </tr> <tr> <td>Word-picture matching: SW</td> <td>2,361</td> <td>495</td> <td>2,082</td> <td>470</td> <td>2,408</td> <td>544</td> <td>2,041</td> <td>446</td> <td>2,256</td> <td>458</td> <td>1,847</td> <td>421</td> </tr> <tr> <td>Relatedness judgment: SW</td> <td>2,717</td> <td>609</td> <td>2,458</td> <td>548</td> <td>2,763</td> <td>652</td> <td>2,383</td> <td>539</td> <td>2,670</td> <td>620</td> <td>2,168</td> <td>555</td> </tr> <tr> <td>Lexical decision: MCW</td> <td>1,689</td> <td>495</td> <td>1,583</td> <td>471</td> <td>1,725</td> <td>519</td> <td>1,503</td> <td>477</td> <td>1,566</td> <td>472</td> <td>1,306</td> <td>396</td> </tr> <tr> <td>Word-picture matching: MCW</td> <td>3,181</td> <td>729</td> <td>2,874</td> <td>705</td> <td>3,258</td> <td>695</td> <td>2,766</td> <td>670</td> <td>2,911</td> <td>678</td> <td>2,319</td> <td>649</td> </tr> <tr> <td>Morphological relatedness judgment</td> <td>2,048</td> <td>588</td> <td>1,875</td> <td>514</td> <td>2,053</td> <td>551</td> <td>1,795</td> <td>485</td> <td>1,993</td> <td>550</td> <td>1,616</td> <td>442</td> </tr> <tr> <td>Relatedness judgment: MCW</td> <td>3,496</td> <td>977</td> <td>3,455</td> <td>966</td> <td>3,524</td> <td>924</td> <td>3,335</td> <td>954</td> <td>3,246</td> <td>973</td> <td>2,774</td> <td>970</td> </tr> <tr> <td>Grammaticality judgment</td> <td>4,165</td> <td>1,054</td> <td>4,281</td> <td>1,018</td> <td>4,105</td> <td>1,008</td> <td>3,992</td> <td>979</td> <td>3,742</td> <td>1,247</td> <td>3,387</td> <td>1,173</td> </tr> <tr> <td>Sentence-picture matching</td> <td>5,079</td> <td>866</td> <td>5,029</td> <td>799</td> <td>5,102</td> <td>808</td> <td>4,850</td> <td>881</td> <td>5,173</td> <td>1,056</td> <td>4,537</td> <td>1,252</td> </tr> <tr> <td>Written discourse</td> <td>4,443</td> <td>1,668</td> <td>3,956</td> <td>1,340</td> <td>4,500</td> <td>1,865</td> <td>3,726</td> <td>1,280</td> <td>4,414</td> <td>2,092</td> <td>3,133</td> <td>1,268</td> </tr> <tr> <td colspan="13">Auditory</td> </tr> <tr> <td>Pseudohomophone judgment</td> <td>567</td> <td>221</td> <td>516</td> <td>209</td> <td>646</td> <td>206</td> <td>545</td> <td>180</td> <td>610</td> <td>194</td> <td>477</td> <td>144</td> </tr> <tr> <td>Lexical decision: SW</td> <td>618</td> <td>212</td> <td>556</td> <td>192</td> <td>650</td> <td>210</td> <td>549</td> <td>167</td> <td>615</td> <td>196</td> <td>498</td> <td>153</td> </tr> <tr> <td>Word-picture matching: SW</td> <td>1,587</td> <td>352</td> <td>1,438</td> <td>344</td> <td>1,674</td> <td>411</td> <td>1,406</td> <td>366</td> <td>1,541</td> <td>392</td> <td>1,201</td> <td>316</td> </tr> <tr> <td>Relatedness judgment: SW</td> <td>2,755</td> <td>353</td> <td>2,665</td> <td>321</td> <td>2,810</td> <td>390</td> <td>2,641</td> <td>289</td> <td>2,648</td> <td>391</td> <td>2,417</td> <td>256</td> </tr> <tr> <td>Lexical decision: MCW</td> <td>677</td> <td>264</td> <td>585</td> <td>240</td> <td>711</td> <td>285</td> <td>594</td> <td>231</td> <td>677</td> <td>288</td> <td>537</td> <td>230</td> </tr> <tr> <td>Word-picture matching: MCW</td> <td>2,042</td> <td>543</td> <td>1,844</td> <td>523</td> <td>2,143</td> <td>586</td> <td>1,743</td> <td>507</td> <td>1,860</td> <td>569</td> <td>1,395</td> <td>464</td> </tr> <tr> <td>Morphological relatedness judgment</td> <td>735</td> <td>320</td> <td>641</td> <td>286</td> <td>778</td> <td>341</td> <td>644</td> <td>346</td> <td>742</td> <td>333</td> <td>561</td> <td>251</td> </tr> <tr> <td>Relatedness judgment: MCW</td> <td>3,036</td> <td>535</td> <td>3,057</td> <td>584</td> <td>3,115</td> <td>546</td> <td>3,011</td> <td>552</td> <td>3,003</td> <td>495</td> <td>2,792</td> <td>413</td> </tr> <tr> <td>Grammaticality judgment</td> <td>941</td> <td>396</td> <td>1,006</td> <td>507</td> <td>1,037</td> <td>505</td> <td>1,004</td> <td>457</td> <td>942</td> <td>443</td> <td>819</td> <td>436</td> </tr> <tr> <td>Sentence-picture matching</td> <td>2,889</td> <td>678</td> <td>2,693</td> <td>733</td> <td>2,940</td> <td>692</td> <td>2,583</td> <td>719</td> <td>2,811</td> <td>875</td> <td>2,221</td> <td>841</td> </tr> <tr> <td>Auditory discourse</td> <td>933</td> <td>366</td> <td>731</td> <td>297</td> <td>987</td> <td>472</td> <td>707</td> <td>254</td> <td>1,030</td> <td>578</td> <td>647</td> <td>282</td> </tr> </table></div> </ephtml> </p> <p>Note. M = mean; MCW = morphologically complex words; SD = standard deviation; SW = simple words.</p> <hd id="AN0112047212-19">TABLE 4 Exploratory Factor Analysis of the 2009 Testing Session: Written and Auditory Modalities Combined, Discourse Excluded, Composite Z Measure</hd> <p> <ephtml> <div class="table-size-normal table-border"><table border="1"> <tr> <td>Modality</td> <td>Test number</td> <td>Test</td> <td>Factor</td> <td colspan="2">Loading</td> </tr> <tr> <td colspan="6">Middle school</td> </tr> <tr> <td rowspan="10">Written</td> <td>Test01Z</td> <td>Decoding</td> <td>1</td> <td colspan="2">.51</td> </tr> <tr> <td>Test02Z</td> <td>Simple word recognition</td> <td>1</td> <td colspan="2">.71</td> </tr> <tr> <td>Test03Z</td> <td>Word-picture matching</td> <td>3</td> <td colspan="2">.70</td> </tr> <tr> <td>Test04Z</td> <td>Relatedness judgment</td> <td>1</td> <td colspan="2">.59</td> </tr> <tr> <td>Test05Z</td> <td>Complex word recognition</td> <td>1</td> <td colspan="2">.50</td> </tr> <tr> <td>Test07Z</td> <td>Complex word-picture matching</td> <td>6</td> <td colspan="2">.42</td> </tr> <tr> <td>Test08Z</td> <td>Root-suffix relationship judgment</td> <td>1</td> <td colspan="2">.82</td> </tr> <tr> <td>Test09Z</td> <td>Complex word meaning judgment</td> <td>1</td> <td colspan="2">.64</td> </tr> <tr> <td>Test10Z</td> <td>Syntax judgment</td> <td>1</td> <td colspan="2">.71</td> </tr> <tr> <td>Test1 1Z</td> <td>Sentence-picture matching</td> <td>4</td> <td colspan="2">.73</td> </tr> <tr> <td rowspan="10">Auditory</td> <td>Test13Z</td> <td>Decoding</td> <td>2</td> <td colspan="2">.49</td> </tr> <tr> <td>Test1 4Z</td> <td>Simple word recognition</td> <td>2</td> <td colspan="2">.47</td> </tr> <tr> <td>Test1 5Z</td> <td>Word-picture matching</td> <td>3</td> <td colspan="2">.81</td> </tr> <tr> <td>Test1 6Z</td> <td>Relatedness judgment</td> <td>8</td> <td colspan="2">.75</td> </tr> <tr> <td>Test1 7Z</td> <td>Complex word recognition</td> <td>2</td> <td colspan="2">.82</td> </tr> <tr> <td>Test1 8Z</td> <td>Complex word-picture matching</td> <td>6</td> <td colspan="2">.55</td> </tr> <tr> <td>Test1 9Z</td> <td>Root-suffix relationship judgment</td> <td>9</td> <td colspan="2">.42</td> </tr> <tr> <td>Test20Z</td> <td>Complex word meaning judgment</td> <td>7</td> <td colspan="2">.56</td> </tr> <tr> <td>Test21Z</td> <td>Syntax judgment</td> <td>5</td> <td colspan="2">.55</td> </tr> <tr> <td>Test22Z</td> <td>Sentence-picture matching</td> <td>4</td> <td colspan="2">.64</td> </tr> <tr> <td colspan="6">High school</td> </tr> <tr> <td rowspan="10">Written</td> <td>Test01Z</td> <td>Decoding</td> <td>1</td> <td colspan="2">.63</td> </tr> <tr> <td>Test02Z</td> <td>Simple word recognition</td> <td>1</td> <td colspan="2">.70</td> </tr> <tr> <td>Test03Z</td> <td>Word-picture matching</td> <td>3</td> <td colspan="2">.68</td> </tr> <tr> <td>Test04Z</td> <td>Relatedness judgment</td> <td>1</td> <td colspan="2">.78</td> </tr> <tr> <td>Test05Z</td> <td>Complex word recognition</td> <td>1</td> <td colspan="2">.48</td> </tr> <tr> <td>Test07Z</td> <td>Complex word-picture matching</td> <td>1</td> <td colspan="2">.80</td> </tr> <tr> <td>Test08Z</td> <td>Root-suffix relationship judgment</td> <td>1</td> <td colspan="2">.51</td> </tr> <tr> <td>Test09Z</td> <td>Complex word meaning judgment</td> <td>1</td> <td colspan="2">.68</td> </tr> <tr> <td>Test10Z</td> <td>Syntax judgment</td> <td>7</td> <td colspan="2">.67</td> </tr> <tr> <td>Test1 1Z</td> <td>Sentence-picture matching</td> <td>1</td> <td colspan="2">.62</td> </tr> <tr> <td rowspan="10">Auditory</td> <td>Test1 3Z</td> <td>Decoding</td> <td>2</td> <td colspan="2">.64</td> </tr> <tr> <td>Test1 4Z</td> <td>Simple word recognition</td> <td>2</td> <td colspan="2">.41</td> </tr> <tr> <td>Test1 5Z</td> <td>Word-picture matching</td> <td>3</td> <td colspan="2">.63</td> </tr> <tr> <td>Test16Z</td> <td>Relatedness judgment</td> <td>2</td> <td colspan="2">.78</td> </tr> <tr> <td>Test17Z</td> <td>Complex word recognition</td> <td>2</td> <td colspan="2">.80</td> </tr> <tr> <td>Test18Z</td> <td>Complex word-picture matching</td> <td>6</td> <td colspan="2">.45</td> </tr> <tr> <td>Test19Z</td> <td>Root-suffix relationship judgment</td> <td>2</td> <td colspan="2">.58</td> </tr> <tr> <td>Test20Z</td> <td>Complex word meaning judgment</td> <td>4</td> <td colspan="2">.59</td> </tr> <tr> <td>Test21Z</td> <td>Syntax judgment</td> <td>2</td> <td colspan="2">.32</td> </tr> <tr> <td>Test22Z</td> <td>Sentence-picture matching</td> <td>5</td> <td colspan="2">.76</td> </tr> <tr> <td>Factor</td> <td>Eigenvalue</td> <td>Variance explained</td> <td>Factor</td> <td>Eigenvalue</td> <td>Variance explained</td> </tr> <tr> <td colspan="3">Middle school</td> <td colspan="3">High school</td> </tr> <tr> <td>1</td> <td>3.30</td> <td>26.0%</td> <td>1</td> <td>3.91</td> <td>27.9%</td> </tr> <tr> <td>2</td> <td>2.16</td> <td>16.9%</td> <td>2</td> <td>3.21</td> <td>22.9%</td> </tr> <tr> <td>3</td> <td>1.52</td> <td>11.9%</td> <td>3</td> <td>1.99</td> <td>14.2%</td> </tr> <tr> <td>4</td> <td>1.33</td> <td>10.4%</td> <td>4</td> <td>1.52</td> <td>10.8%</td> </tr> <tr> <td>5</td> <td>1.08</td> <td>8.5%</td> <td>5</td> <td>1.06</td> <td>7.5%</td> </tr> <tr> <td>6</td> <td>0.89</td> <td>7.0%</td> <td>6</td> <td>0.89</td> <td>6.3%</td> </tr> <tr> <td>7</td> <td>0.85</td> <td>6.6%</td> <td>7</td> <td>0.83</td> <td>5.9%</td> </tr> <tr> <td>8</td> <td>0.83</td> <td>6.5%</td> <td>8</td> <td>0.29</td> <td>2.1%</td> </tr> <tr> <td>9</td> <td>0.51</td> <td>4.0%</td> <td>9</td> <td>0.17</td> <td>1.2%</td> </tr> <tr> <td>10</td> <td>0.26</td> <td>2.0%</td> <td>10</td> <td>0.15</td> <td>1.1%</td> </tr> </table></div> </ephtml> </p> <hd id="AN0112047212-20">TABLE 5 Exploratory Factor Analysis of the 2009 Testing Session: Written and Auditory Modalities Separate, Discourse Excluded, Composite Z Measure</hd> <p> <ephtml> <div class="table-size-normal table-border"><table border="1"> <tr> <td>Modality</td> <td>Test number</td> <td>Test</td> <td>Factor</td> <td colspan="2">Loading</td> </tr> <tr> <td colspan="6">Middle school</td> </tr> <tr> <td rowspan="9">Written</td> <td>Test01Z</td> <td>Decoding</td> <td>1</td> <td colspan="2">.46</td> </tr> <tr> <td>Test02Z</td> <td>Simple word recognition</td> <td>1</td> <td colspan="2">.73</td> </tr> <tr> <td>Test05Z</td> <td>Complex word recognition</td> <td>1</td> <td colspan="2">.74</td> </tr> <tr> <td>Test03Z</td> <td>Word-picture matching</td> <td>2</td> <td colspan="2">.72</td> </tr> <tr> <td>Test04Z</td> <td>Relatedness judgment</td> <td>2</td> <td colspan="2">.62</td> </tr> <tr> <td>Test07Z</td> <td>Complex word-pic matching</td> <td>2</td> <td colspan="2">.52</td> </tr> <tr> <td>Test10Z</td> <td>Syntax judgment</td> <td>3</td> <td colspan="2">.60</td> </tr> <tr> <td>Test11Z</td> <td>Sentence-pic matching</td> <td>3</td> <td colspan="2">.65</td> </tr> <tr> <td>Test09Z</td> <td>Complex word meaning judgment</td> <td>4</td> <td colspan="2">.44</td> </tr> <tr> <td rowspan="9">Auditory</td> <td>Test1 3Z</td> <td>Decoding</td> <td>1</td> <td colspan="2">.72</td> </tr> <tr> <td>Test14Z</td> <td>Simple word recognition</td> <td>1</td> <td colspan="2">.85</td> </tr> <tr> <td>Test1 7Z</td> <td>Complex word recognition</td> <td>1</td> <td colspan="2">.58</td> </tr> <tr> <td>Test21Z</td> <td>Syntax judgment</td> <td>2</td> <td colspan="2">.66</td> </tr> <tr> <td>Test22Z</td> <td>Sentence-picture matching</td> <td>2</td> <td colspan="2">.53</td> </tr> <tr> <td>Test1 5Z</td> <td>Word-picture matching</td> <td>3</td> <td colspan="2">.57</td> </tr> <tr> <td>Test1 8Z</td> <td>Complex word-picture matching</td> <td>3</td> <td colspan="2">.63</td> </tr> <tr> <td>Test1 6Z</td> <td>Relatedness judgment</td> <td>4</td> <td colspan="2">.64</td> </tr> <tr> <td>Test20Z</td> <td>Complex word meaning judgment</td> <td>4</td> <td colspan="2">.41</td> </tr> <tr> <td colspan="6">High school</td> </tr> <tr> <td rowspan="9">Written</td> <td>Test01Z</td> <td>Decoding</td> <td>1</td> <td colspan="2">.61</td> </tr> <tr> <td>Test02Z</td> <td>Simple word recognition</td> <td>1</td> <td colspan="2">.67</td> </tr> <tr> <td>Test05Z</td> <td>Complex word recognition</td> <td>1</td> <td colspan="2">.65</td> </tr> <tr> <td>Test09Z</td> <td>Complex word meaning judgment</td> <td>2</td> <td colspan="2">.46</td> </tr> <tr> <td>Test10Z</td> <td>Syntax judgment</td> <td>2</td> <td colspan="2">.73</td> </tr> <tr> <td>Test1 1Z</td> <td>Sentence-picture matching</td> <td>2</td> <td colspan="2">.65</td> </tr> <tr> <td>Test03Z</td> <td>Word-picture matching</td> <td>3</td> <td colspan="2">.71</td> </tr> <tr> <td>Test04Z</td> <td>Relatedness judgment</td> <td>3</td> <td colspan="2">.54</td> </tr> <tr> <td>Test07Z</td> <td>Complex word-picture matching</td> <td>3</td> <td colspan="2">.56</td> </tr> <tr> <td rowspan="9">Auditory</td> <td>Test1 3Z</td> <td>Decoding</td> <td>1</td> <td colspan="2">.78</td> </tr> <tr> <td>Test1 4Z</td> <td>Simple word recognition</td> <td>1</td> <td colspan="2">.82</td> </tr> <tr> <td>Test1 7Z</td> <td>Complex word recognition</td> <td>1</td> <td colspan="2">.62</td> </tr> <tr> <td>Test1 5Z</td> <td>Word-picture matching</td> <td>2</td> <td colspan="2">.62</td> </tr> <tr> <td>Test1 6Z</td> <td>Relatedness judgment</td> <td>2</td> <td colspan="2">.48</td> </tr> <tr> <td>Test1 8Z</td> <td>Complex word-picture matching</td> <td>2</td> <td colspan="2">.66</td> </tr> <tr> <td>Test22Z</td> <td>Sentence-picture matching</td> <td>2</td> <td colspan="2">.58</td> </tr> <tr> <td>Test20Z</td> <td>Complex word meaning judgment</td> <td>3</td> <td colspan="2">.67</td> </tr> <tr> <td>Test21Z</td> <td>Syntax judgment</td> <td>3</td> <td colspan="2">.57</td> </tr> <tr> <td>Factor</td> <td>Eigenvalue</td> <td>Variance explained</td> <td>Factor</td> <td>Eigenvalue</td> <td>Variance explained</td> </tr> <tr> <td colspan="3">Middle school,written modality</td> <td colspan="3">High school,written modality</td> </tr> <tr> <td>1</td> <td>1.82</td> <td>35.6%</td> <td>1</td> <td>1.83437901</td> <td>31.6%</td> </tr> <tr> <td>2</td> <td>1.57</td> <td>30.7%</td> <td>2</td> <td>1.74139814</td> <td>30.0%</td> </tr> <tr> <td>3</td> <td>1.27</td> <td>24.7%</td> <td>3</td> <td>1.59098751</td> <td>27.4%</td> </tr> <tr> <td>4</td> <td>0.46</td> <td>8.9%</td> <td>4</td> <td>0.64591448</td> <td>11.1%</td> </tr> <tr> <td colspan="3">Middle school,auditory modality</td> <td colspan="3">High school,auditory modality</td> </tr> <tr> <td>1</td> <td>1.83261887</td> <td>36.4%</td> <td>1</td> <td>2.05216591</td> <td>37.2%</td> </tr> <tr> <td>2</td> <td>1.20931717</td> <td>24.0%</td> <td>2</td> <td>1.80905525</td> <td>32.8%</td> </tr> <tr> <td>3</td> <td>1.18299177</td> <td>23.5%</td> <td>3</td> <td>1.65358209</td> <td>30.0%</td> </tr> <tr> <td>4</td> <td>0.81492384</td> <td>16.2%</td> <td></td> <td></td> <td></td> </tr> </table></div> </ephtml> </p> <hd id="AN0112047212-21">TABLE 6 Structural Equation Models: Measurement Model Components (Factors in Confirmatory Factor Analyses; all ps < .001)</hd> <p> <ephtml> <div class="table-size-normal table-border"><table border="1"> <tr> <td rowspan="2">Path</td> <td colspan="2">Middle school</td> <td colspan="2">High school</td> </tr> <tr> <td>2011</td> <td>2013</td> <td>2011</td> <td>2013</td> </tr> <tr> <td colspan="5">F1Aud</td> </tr> <tr> <td>→ Test13Z</td> <td>.69</td> <td>.73</td> <td>.75</td> <td>.78</td> </tr> <tr> <td>→ Test14Z</td> <td>.78</td> <td>.80</td> <td>.79</td> <td>.81</td> </tr> <tr> <td>→ Test17Z</td> <td>.84</td> <td>.78</td> <td>.84</td> <td>.85</td> </tr> <tr> <td colspan="5">F2Aud</td> </tr> <tr> <td>→ Test15Z</td> <td>.64</td> <td>.65</td> <td>.70</td> <td>.71</td> </tr> <tr> <td>→ Test16Z</td> <td>.67</td> <td>.74</td> <td>.78</td> <td>.74</td> </tr> <tr> <td>→ Test18Z</td> <td>--</td> <td>--</td> <td>.78</td> <td>.76</td> </tr> <tr> <td colspan="5">F3Aud</td> </tr> <tr> <td>→ Test20Z</td> <td>--</td> <td>--</td> <td>.68</td> <td>.67</td> </tr> <tr> <td>→ Test21Z</td> <td>.57</td> <td>.57</td> <td>.75</td> <td>.71</td> </tr> <tr> <td>→ Test22Z</td> <td>.78</td> <td>.69</td> <td>.76</td> <td>.74</td> </tr> <tr> <td colspan="5">F1Writ</td> </tr> <tr> <td>→ Test01Z</td> <td>.58</td> <td>.56</td> <td>.73</td> <td>.72</td> </tr> <tr> <td>→ Test02Z</td> <td>.80</td> <td>.76</td> <td>.85</td> <td>.85</td> </tr> <tr> <td>→ Test05Z</td> <td>.84</td> <td>.77</td> <td>.88</td> <td>.86</td> </tr> <tr> <td colspan="5">F2Writ</td> </tr> <tr> <td>→ Test03Z</td> <td>.62</td> <td>.56</td> <td>.69</td> <td>.66</td> </tr> <tr> <td>→ Test04Z</td> <td>.77</td> <td>.79</td> <td>.84</td> <td>.77</td> </tr> <tr> <td>→ Test07Z</td> <td>.75</td> <td>.72</td> <td>.81</td> <td>.76</td> </tr> <tr> <td colspan="5">F3Writ</td> </tr> <tr> <td>→ Test09Z</td> <td>--</td> <td>--</td> <td>.75</td> <td>.71</td> </tr> <tr> <td>→ Test10Z</td> <td>.66</td> <td>.61</td> <td>.74</td> <td>.73</td> </tr> <tr> <td>→ Test11Z</td> <td>.61</td> <td>.66</td> <td>.70</td> <td>.66</td> </tr> </table></div> </ephtml> </p> <p>Note. Test numbers as in Tables 4 and 5.</p> <hd id="AN0112047212-22">TABLE 7 Structural Equation Models: Goodness-of-Fit Indexes</hd> <p> <ephtml> <div class="table-size-normal table-border"><table border="1"> <tr> <td rowspan="2">Statistic</td> <td colspan="2">Middle school</td> <td colspan="2">High school</td> </tr> <tr> <td>2011</td> <td>2013</td> <td>2011</td> <td>2013</td> </tr> <tr> <td colspan="5">Model 1</td> </tr> <tr> <td>χ²</td> <td>264.02</td> <td>274.13</td> <td>897.55</td> <td>657.69</td> </tr> <tr> <td>df</td> <td>92</td> <td>92</td> <td>140</td> <td>140</td> </tr> <tr> <td>p</td> <td>.00</td> <td>.00</td> <td>.00</td> <td>.00</td> </tr> <tr> <td>Akaike information criterion</td> <td>352.02</td> <td>362.13</td> <td>997.55</td> <td>757.69</td> </tr> <tr> <td>RMSEA</td> <td>0.07</td> <td>0.07</td> <td>0.08</td> <td>0.07</td> </tr> <tr> <td>RMSEA upper 90% confidence limit</td> <td>0.08</td> <td>0.08</td> <td>0.08</td> <td>0.08</td> </tr> <tr> <td>Comparative fit index</td> <td>0.94</td> <td>0.92</td> <td>0.93</td> <td>0.93</td> </tr> <tr> <td>Tucker-Lewis Index</td> <td>0.92</td> <td>0.90</td> <td>0.91</td> <td>0.91</td> </tr> <tr> <td>Standardized root mean square residual</td> <td>0.048</td> <td>0.054</td> <td>0.047</td> <td>0.047</td> </tr> <tr> <td colspan="5">Model 2</td> </tr> <tr> <td>χ²</td> <td>262.13</td> <td>258.79</td> <td>889.32</td> <td>644.24</td> </tr> <tr> <td>df</td> <td>89</td> <td>89</td> <td>137</td> <td>137</td> </tr> <tr> <td>p</td> <td>.00</td> <td>.00</td> <td>.00</td> <td>.00</td> </tr> <tr> <td>Akaike information criterion</td> <td>356.13</td> <td>352.79</td> <td>995.32</td> <td>750.24</td> </tr> <tr> <td>RMSEA</td> <td>0.07</td> <td>0.07</td> <td>0.08</td> <td>0.07</td> </tr> <tr> <td>RMSEA upper 90% confidence limit</td> <td>0.08</td> <td>0.08</td> <td>0.08</td> <td>0.08</td> </tr> <tr> <td>Comparative fit index</td> <td>0.94</td> <td>0.93</td> <td>0.93</td> <td>0.93</td> </tr> <tr> <td>Tucker-Lewis Index</td> <td>0.92</td> <td>0.90</td> <td>0.91</td> <td>0.91</td> </tr> <tr> <td>Standardized root mean square residual</td> <td>0.047</td> <td>0.046</td> <td>0.051</td> <td>0.047</td> </tr> </table></div> </ephtml> </p> <hd id="AN0112047212-23">TABLE 8 Structural Equation Models: Structural Model Components in Best-Fitting Models</hd> <p> <ephtml> <div class="table-size-normal table-border"><table border="1"> <tr> <td rowspan="3">Path</td> <td colspan="2">Middle school</td> <td colspan="2">High school</td> </tr> <tr> <td>2011</td> <td>2013</td> <td>2011</td> <td>2013</td> </tr> <tr> <td>Model 1</td> <td>Model 2</td> <td>Model 1</td> <td>Model 2</td> </tr> <tr> <td>F1Aud → F1Writ</td> <td>.67* * *</td> <td>.59* * *</td> <td>.65* * *</td> <td>.53* * *</td> </tr> <tr> <td>F1Aud → F2Aud</td> <td>.77*</td> <td>.87* * *</td> <td>.80* * *</td> <td>.81* * *</td> </tr> <tr> <td>F2Aud → F2Writ</td> <td>.53* * *</td> <td>.45* * *</td> <td>.57* * *</td> <td>.50* * *</td> </tr> <tr> <td>F2Aud → F3Aud</td> <td>.60* * *</td> <td>.73*</td> <td>.39* * *</td> <td>.36*</td> </tr> <tr> <td>F1Writ → F2Writ</td> <td>.50* * *</td> <td>.48* * *</td> <td>.49* * *</td> <td>.49* * *</td> </tr> <tr> <td>F1Aud → F3Aud</td> <td>.16 (.22)</td> <td>.13 (.59)</td> <td>.56* * *</td> <td>.59* * *</td> </tr> <tr> <td>F3Aud → F3Writ</td> <td>.60* * *</td> <td>.13 (.21)</td> <td>.23* * *</td> <td>.13*</td> </tr> <tr> <td>F2Writ → F3Writ</td> <td>-.19 (.21)</td> <td>.08**</td> <td>.17**</td> <td>.18**</td> </tr> <tr> <td>F1Writ → F3Writ</td> <td>.58* * *</td> <td>.61* * *</td> <td>.57* * *</td> <td>.65* * *</td> </tr> <tr> <td>F1Aud → DiscWZ</td> <td>--</td> <td>-.63*</td> <td>-.25*</td> <td>-.37*</td> </tr> <tr> <td>F2Aud → DiscWZ</td> <td>--</td> <td>.52*</td> <td>.11 (.27)</td> <td>-.17 (.16)</td> </tr> <tr> <td>F3Aud → DiscWZ</td> <td>--</td> <td>.19 (.32)</td> <td>.23**</td> <td>.55* * *</td> </tr> <tr> <td>F1Writ → DiscWZ</td> <td>-.21 (.03)</td> <td>.25* * *</td> <td>.07 (.38)</td> <td>.18*</td> </tr> <tr> <td>F2Writ → DiscWZ</td> <td>.64* * *</td> <td>.14 (.23)</td> <td>.29*</td> <td>.37*</td> </tr> <tr> <td>F3Writ → DiscWZ</td> <td>.30* * *</td> <td>.28**</td> <td>.32* * *</td> <td>.06 (.57)</td> </tr> </table></div> </ephtml> </p> <p>Note. Model 1 = model with paths only from both written factors to written discourse comprehension. Model 2 = model with paths from both written and auditory factors to written discourse comprehension. Numbers in parentheses are p-values. *p < .05. "p < .01. * * *p < .001.</p> <p>FIGURE 1 Model 1: Structural Equation Model With Direct Paths Only From Written Factors to Written Passage Comprehension</p> <p>FIGURE 2 Model 2: Structural Equation Model With Direct Paths From Both Auditory and Written Factors to Written Passage Comprehension</p> <hd id="AN0112047212-24">REFERENCES</hd> <p>Adams, M.J. (1994). Beginning to read: Thinking and learning about print. Cambridge, MA: MIT Press.</p> <p>Bowyer-Crane, C., & Snowling, M.J. (2005). Assessing children's inference generation: What do tests of reading comprehension measure? British Journal of Educational Psychology, 75(<reflink idref="bib2" id="ref2">2</reflink>), 189-201. doi:10.1348/000709904X22674</p> <p>Bryant, P., Nunes, T., & Barros, R. (2014). The connection between children's learning about grapho-phonic and morphemic units spelling rules and their achievements at school. British Journal of Educational Psychology, 84(<reflink idref="bib2" id="ref3">2</reflink>), 211-225. doi:10.1111/bjep.12030</p> <p>Byrne, B.M. (1998). Structural equation modeling with LISREL, PRELIS and SIMPLIS: Basic concepts, applications and programming. Mahwah, NJ: Erlbaum.</p> <p>Cain, K., Oakhill, J.V., Barnes, M.A., & Bryant, P.E. (2001). Comprehension skill, inference-making ability, and their relation to knowledge. Memory & Cognition, 29(<reflink idref="bib6" id="ref4">6</reflink>), 850-859. doi:10.3758/BF03196414</p> <p>Caplan, D. (1992). Language: Structure, processing, and disorders. Cambridge, MA: MIT Press.</p> <p>Caplan, D. (1993). Toward a psycholinguistic approach to acquired neurogenic language disorders. American Journal of Speech-Language Pathology, 2(<reflink idref="bib3" id="ref5">3</reflink>), 59-83. doi:10.1044/1058-0360.0201.59</p> <p>Caplan, D., Michaud, J., & Hufford, R. (2013). Dissociations and associations of performance in syntactic comprehension in aphasia and their implications for the nature of aphasic deficits. Brain and Language, 127(<reflink idref="bib1" id="ref6">1</reflink>), 21-33. doi:10.1016/j.bandl.2013.07.007</p> <p>Caplan, D., Waters, G.S., DeDe, G., Michaud, J., & Reddy, A. (2007). A study of syntactic processing in aphasia I: Behavioral (psycholinguistic) aspects. Brain and Language, 101(<reflink idref="bib2" id="ref7">2</reflink>), 103-150. doi:10.1016/j.bandl.2006.06.225</p> <p>Carlisle, J.F. (2010). Effects of instruction in morphological awareness on literacy achievement: An integrative review. Reading Research Quarterly, 45(<reflink idref="bib4" id="ref8">4</reflink>), 464-487. doi:10.1598/RRQ.45.4.5</p> <p>Catts, H.W., Gillispie, M., Leonard, L.B., Kail, R.V., & Miller, C.A. (2002). The role of speed of processing, rapid naming, and phonological awareness in reading achievement. Journal of Learning Disabilities, 35(<reflink idref="bib6" id="ref9">6</reflink>), 510-525. doi:10.1177/00222194020350060301</p> <p>Catts, H.W., Hogan, T.P., & Fey, M.E. (2003). Subgrouping poor readers on the basis of individual differences in reading-related abilities. Journal of Learning Disabilities, 36(<reflink idref="bib2" id="ref10">2</reflink>), 151-164. doi:10.1177/002221940303600208</p> <p>Chafe, W. (1992). Writing vs. speech. In W. Bright (Ed.), Oxford international encyclopedia of linguistics (Vol. 4, pp. 257-259). New York, NY: Oxford University Press.</p> <p>Chafe, W., & Danielewitz, J. (1987). Properties of spoken and written language. In R. Horowitz & S.J. Samuels (Eds.), Comprehending oral and written language (pp. 83-113). New York, NY: Academic.</p> <p>Chafe, W., & Tannen, D. (1987). The relation between written and spoken language. Annual Review of Anthropology, 16, 383-407. doi:10.1146/annurev.an.16.100187.002123</p> <p>Christopher, M.E., Miyake, A., Keenan, J.M., Pennington, B., DeFries, J.C., Wadsworth, S.J., … Olson, R.K. (2012). Predicting word reading and comprehension with executive function and speed measures across development: A latent variable analysis. Journal of Experimental Psychology: General, 141(<reflink idref="bib3" id="ref11">3</reflink>), 470-488. doi:10.1037/a0027375</p> <p>Cromley, J.G., & Azevedo, R. (2007). Testing and refining the direct and inferential mediation model of reading comprehension. Journal of Educational Psychology, 99(<reflink idref="bib2" id="ref12">2</reflink>), 311-325. doi:10.1037/0022-0663.99.2.311</p> <p>Cromley, J.G., Snyder-Hogan, L.E., & Luciw-Dubas, U.A. (2010). Reading comprehension of scientific text: A domain-specific test of the direct and inferential mediation model of reading comprehension. Journal of Educational Psychology, 102(<reflink idref="bib3" id="ref13">3</reflink>), 687-700. doi:10.1037/a0019452</p> <p>Curtis, M. (1980). Development of components of reading skill. Journal of Educational Psychology, 72(<reflink idref="bib5" id="ref14">5</reflink>), 656-669. doi:10.1037/0022-0663.72.5.656</p> <p>Curtis, M., & Kruidenier, J. (2005). A summary of scientifically based research principles: Teaching adults to read. Washington, DC: National Institute for Literacy.</p> <p>Cutting, L.E., & Scarborough, H.S. (2006). Prediction of reading comprehension: Relative contributions of word recognition, language proficiency, and other cognitive skills can depend on how comprehension is measured. Scientific Studies of Reading, 10(<reflink idref="bib3" id="ref15">3</reflink>), 277-299. doi:10.1207/s1532799xssr1003_5</p> <p>Danks, J.H., & End, L.J. (1987). Processing strategies for reading and listening. In R. Horowitz & S.J. Samuels (Eds.), Comprehending oral and written language (pp. 271-294). New York, NY: Academic.</p> <p>Diakidoy, I.N., Stylianou, P., Karefillidou, C., & Papageorgiou, P. (2005). The relationship between listening and reading comprehension of different types of text at increasing grade levels. Reading Psychology, 26(<reflink idref="bib1" id="ref16">1</reflink>), 55-80. doi:10.1080/02702710590910584</p> <p>Fan, X., Thompson, B., & Wang, L. (1999). Effects of sample size, estimation methods, and model specification on structural equation modeling fit indexes. Structural Equation Modeling, 6(<reflink idref="bib1" id="ref17">1</reflink>), 56-83. doi:10.1080/10705519909540119</p> <p>Ferrer, E., McArdle, J.J., Shaywitz, B.A., Holahan, J.N., Marchione, K., & Shaywitz, S.E. (2007). Longitudinal models of developmental dynamics between reading and cognition from childhood to adolescence. Developmental Psychology, 43(<reflink idref="bib6" id="ref18">6</reflink>), 1460-1473. doi:10.1037/0012-1649.43.6.1460</p> <p>Fugate, M.H. (2002). Review of Group Reading Assessment and Diagnostic Evaluation. In B.S. Plake, J.C. Impara, & R.A. Spies (Eds.), The fifteenth mental measurements yearbook (pp. 425-429). Lincoln, NE: Buros Center for Testing.</p> <p>Glushko, R.J. (1979). The organization and activation of orthographic knowledge in reading aloud. Journal of Experimental Psychology: Human Perception and Performance, 5(<reflink idref="bib4" id="ref19">4</reflink>), 674-691. doi:10.1037/0096-1523.5.4.674</p> <p>Gough, P.B., Hoover, W.A., & Peterson, C.L. (1996). Some observations on a simple view of reading. In C. Cornoldi & J. Oakhill (Eds.), Reading comprehension difficulties: Processes and intervention (pp. 1-13). Mahwah, NJ: Erlbaum.</p> <p>Gough, P.B., & Tunmer, W.E. (1986). Decoding, reading and reading ability. Remedial and Special Education, 7(<reflink idref="bib1" id="ref20">1</reflink>), 6-10. doi:10.1177/074193258600700104</p> <p>Halliday, M.A.K. (1987). Spoken and written modes of meaning. In R. Horowitz & S.J. Samuels (Eds.), Comprehending oral and written language (pp. 55-82). New York, NY: Academic.</p> <p>Hogan, T.P., Sittner, M.S., Justice, L.M., & Cain, K. (2011). Increasing higher level language skills to improve reading comprehension. Focus on Exceptional Children, 44(<reflink idref="bib3" id="ref21">3</reflink>), 1-20.</p> <p>Hoover, W., & Gough, P.B. (1990). The simple view of reading. Reading and Writing, 2(<reflink idref="bib2" id="ref22">2</reflink>), 127-160.</p> <p>Horowitz, R. (2015). Oral language: The genesis and development of literacy for schooling and everyday life. In P.D. Pearson & E. Hiebert (Eds.), Research-based practices for teaching Common Core literacy (pp. 57-75). New York, NY: Teachers College Press.</p> <p>Horowitz, R., & Samuels, S.J. (1985). Reading and listening to expository texts. Journal of Reading Behavior, 17(<reflink idref="bib3" id="ref23">3</reflink>), 185-198. doi:10.1080/10862968509547539</p> <p>Horowitz, R., & Samuels, S.J. (1987). Comprehending oral and written language. New York, NY: Academic.</p> <p>Hu, L.T., & Bentler, P.M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(<reflink idref="bib1" id="ref24">1</reflink>), 1-55. doi:10.1080/10705519909540118</p> <p>Jared, D. (2002). Spelling-sound consistency and regularity effects in word naming. Journal of Memory and Language, 46(<reflink idref="bib4" id="ref25">4</reflink>), 723-750. doi:10.1006/jmla.2001.2827</p> <p>Joshi, R.M., Williams, K.A., & Wood, J.R. (1998). Predicting reading comprehension from listening comprehension: Is this the answer to the IQ debate? In C. Hulme & R.M. (Eds.), Reading and spelling: Development and disorder (pp. 319-327). Mahwah, NJ: Erlbaum.</p> <p>Keenan, J.M., Betjemann, R.S., & Olson, R.K. (2008). Reading comprehension tests vary in the skills they assess: Differential dependence on decoding and oral comprehension. Scientific Studies of Reading, 12(<reflink idref="bib3" id="ref26">3</reflink>), 281-300. doi:10.1080/10888430802132279</p> <p>Kintsch, W. (1988). The role of knowledge in discourse comprehension: A construction-integration model. Psychological Review, 95(<reflink idref="bib2" id="ref27">2</reflink>), 163-182. doi:10.1037/0033-295X.95.2.163</p> <p>Kintsch, W., & van Dijk, T.A. (1978). Toward a model of text comprehension and production. Psychological Review, 85(<reflink idref="bib5" id="ref28">5</reflink>), 363-394. doi:10.1037/0033-295X.85.5.363</p> <p>Kucera, H., & Francis, W.N. (1967). Computational analysis of present-day American English. Providence, RI: Brown University Press.</p> <p>Levy, R. (2008). Expectation-based syntactic comprehension. Cognition, 106(<reflink idref="bib3" id="ref29">3</reflink>), 1126-1177. doi:10.1016/j.cognition.2007.05.006</p> <p>MacCallum, R.C., Browne, M.W., & Sugawara, H.M. (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods, 1(<reflink idref="bib2" id="ref30">2</reflink>), 130-149.</p> <p>Mahony, D.L. (1994). Using sensitivity to word structure to explain variance in high school and college level reading ability. Reading and Writing, 6(<reflink idref="bib1" id="ref31">1</reflink>), 19-44.</p> <p>McCardle, P., Scarborough, H.S., & Catts, H.W. (2001). Predicting, explaining, and preventing children's reading difficulties. Learning Disabilities Research & Practice, 16(<reflink idref="bib4" id="ref32">4</reflink>), 230-239. doi:10.1111/0938-8982.00023</p> <p>Mellard, D.F., Fall, E., & Woods, K.L. (2010). A path analysis of reading comprehension for adults with low literacy. Journal of Learning Disabilities, 43(<reflink idref="bib2" id="ref33">2</reflink>), 154-165. doi:10.1177/0022219409359345</p> <p>Nagy, W., Berninger, V., & Abbott, R. (2006). Contributions of morphology beyond phonology to literacy outcomes of upper elementary and middle school students. Journal of Educational Psychology, 98(<reflink idref="bib1" id="ref34">1</reflink>), 134-147. doi:10.1037/0022-0663.98.1.134</p> <p>Nagy, W., Berninger, V., Abbott, R., Vaughan, K., & Vermeulen, K. (2003). Relationship of morphology and other language skills to literacy skills in at-risk second-grade readers and at-risk fourth-grade writers. Journal of Educational Psychology, 95(<reflink idref="bib4" id="ref35">4</reflink>), 730-742. doi:10.1037/0022-0663.95.4.730</p> <p>Nation, K. (2005). Reading comprehension difficulties. In M.J. Snowling & C. Hulme (Eds.), The science of reading (pp. 248-265). Oxford, UK: Blackwell. doi:10.1002/9780470757642.ch14</p> <p>Nation, K., & Snowling, M.J. (2004). Beyond phonological skills: Broader language skills contribute to the development of reading. Journal of Research in Reading, 27(<reflink idref="bib4" id="ref36">4</reflink>), 342-356. doi:10.1111/j.1467-9817.2004.00238.x</p> <p>Perfetti, C.A. (1987). Language, speech, and print: Some asymmetries in the acquisition of literacy. In R. Horowitz & S.J. Samuels (Eds.), Comprehending oral and written language (pp. 355-369). New York, NY: Academic.</p> <p>Perfetti, C.A., Landi, N., & Oakhill, J. (2005). The acquisition of reading comprehension skill. In M.J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 227-247). Oxford, UK: Blackwell. doi:10.1002/9780470757642.ch13</p> <p>Rubin, D.L., Hafer, T., & Arata, K. (2000). Reading and listening to oral-based versus literate-based discourse. Communication Education, 49(<reflink idref="bib2" id="ref37">2</reflink>), 121-133. doi:10.1080/03634520009379200</p> <p>Savage, R.S. (2006). Reading comprehension is not always the product of decoding and listening comprehension: Evidence from teenagers who are very poor readers. Scientific Studies of Reading, 10(<reflink idref="bib2" id="ref38">2</reflink>), 143-164. doi:10.1207/s1532799xssr1002_2</p> <p>Savage, R.S., & Wolforth, J. (2007). An additive simple view of reading describes the performance of good and poor readers in higher education. Exceptionality Education Canada, 17(<reflink idref="bib2" id="ref39">2</reflink>), 243-268.</p> <p>Steiger, J.H. (2007). Understanding the limitations of global fit assessment in structural equation modeling. Personality and Individual Differences, 42(<reflink idref="bib5" id="ref40">5</reflink>), 893-898. doi:10.1016/j.paid.2006.09.017</p> <p>Sticht, T.G., Beck, L.J., Hauke, R.N., Kleinman, G.M., & James, J.H. (1974). Auding and reading: A developmental model. Alexandria, VA: Human Resources Research Organization.</p> <p>Tomasello, M. (2003). Constructing a language: A usage-based theory of language acquisition. Cambridge, MA: Harvard University Press.</p> <p>Torgesen, J.K., Wagner, R.K., & Rashotte, C.A. (2012). Test of Word Reading Efficiency 2nd ed.. Austin, TX: PRO-ED.</p> <p>Townsend, D., Carrithers, C., & Bever, T. (1987). Listening and reading processes in college and middle school-age readers. In R. Horowitz & S.J. Samuels (Eds.), Comprehending oral and written language (pp. 217-242). New York, NY: Academic.</p> <p>Traxler, M. (2012). Introduction to psycholinguistics: Understanding language science. Sussex, UK: Wiley.</p> <p>Vellutino, F.R., Tunmer, W.E., Jaccard, J.J., & Chen, R. (2007). Components of reading ability: Multivariate evidence for a convergent skill model of reading development. Scientific Studies of Reading, 11(<reflink idref="bib1" id="ref41">1</reflink>), 3-32. doi:10.1080/10888430709336632</p> <p>Vitevitch, M.S., & Luce, P.A. (1998). When words compete: Levels of processing in spoken word perception. Psychological Science, 9(<reflink idref="bib4" id="ref42">4</reflink>), 325-329. doi:10.1111/1467-9280.00064</p> <p>Watson, R., & Olson, D.R. (1987). From meaning to definition: A literate bias on the structure of word meaning. In R. Horowitz & S.J. Samuels (Eds.), Comprehending oral and written language (pp. 329-354). New York, NY: Academic.</p> <p>Yuill, N., & Oakhill, J. (1991). Children's problems in text comprehension: An experimental investigation. New York, NY: Cambridge University Press .</p> <p>Submitted August 21, 2014</p> <p>Final revision received June 22, 2015</p> <p>Accepted July 6, 2015</p> <hd id="AN0112047212-25">Supporting Information</hd> <p>Additional supporting information may be found in the online version of this article on the publisher's website:</p> <p>* Appendix A: Assessment Battery: Details of Tests</p> <p>* Appendix B: Psychometric Properties of the Assessment Battery</p> <p>* Appendix C: SEMS Separate for Accuracy and Reaction Time</p> <aug> <p>By David Caplan, Boston University, Massachusetts General Hospital, and Harvard Medical School, Boston, Massachusetts, USA; Gloria Waters, Boston University, Massachusetts, USA; Julia Bertram, Boston University, Massachusetts, USA; Adam Ostrowski, Boston University, Massachusetts, USA and Jennifer Michaud, Massachusetts General Hospital, Boston, USA</p> <p></p> <p>DAVID CAPLAN (corresponding author) is an adjunct professor in the Department of Speech, Language and Hearing Sciences in the College of Health and Rehabilitation Sciences: Sargent College at Boston University, Massachusetts, USA. He is a professor of neurology in the Neuropsychology Laboratory in the Department of Neurology at the Massachusetts General Hospital, Boston, USA; e-mail: dcaplan@partners.org. He is also a professor of neurology in the Neuropsychology Laboratory at Harvard Medical School, Boston, Massachusetts, USA.</p> <p>GLORIA WATERS is a professor in the Department of Speech, Language and Hearing Sciences in the College of Health and Rehabilitation Sciences: Sargent College and the vice president and associate provost for research at Boston University, Massachusetts, USA; e-mail gwaters@bu.edu.</p> <p>JULIA BERTRAM is a research assistant in the Department of Speech, Language and Hearing Sciences in the College of Health and Rehabilitation Sciences: Sargent College at Boston University, Massachusetts, USA; e-mail jbertram@bu.edu.</p> <p>ADAM OSTROWSKI is a research assistant in the Department of Speech, Language and Hearing Sciences in the College of Health and Rehabilitation Sciences: Sargent College at Boston University, Massachusetts, USA; e-mail adamost@bu.edu.</p> <p>JENNIFER MICHAUD is the laboratory manager in the Neuropsychology Laboratory in the Department of Neurology at the Massachusetts General Hospital, Boston, USA; e-mail jmichaud1@partners.org.</p> </aug> <nolink nlid="nl1" bibid="bib790" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib2" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib6" firstref="ref4"></nolink> <nolink nlid="nl4" bibid="bib3" firstref="ref5"></nolink> <nolink nlid="nl5" bibid="bib1" firstref="ref6"></nolink> <nolink nlid="nl6" bibid="bib4" firstref="ref8"></nolink> <nolink nlid="nl7" bibid="bib5" firstref="ref14"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1087345
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Effects of Written and Auditory Language-Processing Skills on Written Passage Comprehension in Middle and High School Students
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Caplan%2C+David%22">Caplan, David</searchLink><br /><searchLink fieldCode="AR" term="%22Waters%2C+Gloria%22">Waters, Gloria</searchLink><br /><searchLink fieldCode="AR" term="%22Bertram%2C+Julia%22">Bertram, Julia</searchLink><br /><searchLink fieldCode="AR" term="%22Ostrowski%2C+Adam%22">Ostrowski, Adam</searchLink><br /><searchLink fieldCode="AR" term="%22Michaud%2C+Jennifer%22">Michaud, Jennifer</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Reading+Research+Quarterly%22"><i>Reading Research Quarterly</i></searchLink>. Jan-Mar 2016 51(1):67-92.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Wiley-Blackwell. 350 Main Street, Malden, MA 02148. Tel: 800-835-6770; Tel: 781-388-8598; Fax: 781-388-8232; e-mail: cs-journals@wiley.com; Web site: http://www.wiley.com/WileyCDA
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 26
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2016
– Name: SourceSuprt
  Label: Sponsoring Agency
  Group: SrcSuprt
  Data: Institute of Education Sciences (ED)
– Name: NumberContract
  Label: Contract Number
  Group: NumCntrct
  Data: R305A100261
– 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="%22Middle+Schools%22">Middle Schools</searchLink><br /><searchLink fieldCode="EL" term="%22Secondary+Education%22">Secondary Education</searchLink><br /><searchLink fieldCode="EL" term="%22Junior+High+Schools%22">Junior High Schools</searchLink><br /><searchLink fieldCode="EL" term="%22High+Schools%22">High Schools</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Middle+School+Students%22">Middle School Students</searchLink><br /><searchLink fieldCode="DE" term="%22High+School+Students%22">High School Students</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Skills%22">Language Skills</searchLink><br /><searchLink fieldCode="DE" term="%22Language+Processing%22">Language Processing</searchLink><br /><searchLink fieldCode="DE" term="%22Reading+Comprehension%22">Reading Comprehension</searchLink><br /><searchLink fieldCode="DE" term="%22Morphology+%28Languages%29%22">Morphology (Languages)</searchLink><br /><searchLink fieldCode="DE" term="%22Syntax%22">Syntax</searchLink><br /><searchLink fieldCode="DE" term="%22Written+Language%22">Written Language</searchLink><br /><searchLink fieldCode="DE" term="%22Oral+Language%22">Oral Language</searchLink><br /><searchLink fieldCode="DE" term="%22Auditory+Perception%22">Auditory Perception</searchLink><br /><searchLink fieldCode="DE" term="%22Factor+Analysis%22">Factor Analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Accuracy%22">Accuracy</searchLink><br /><searchLink fieldCode="DE" term="%22Reaction+Time%22">Reaction Time</searchLink><br /><searchLink fieldCode="DE" term="%22Difficulty+Level%22">Difficulty Level</searchLink><br /><searchLink fieldCode="DE" term="%22Sentences%22">Sentences</searchLink><br /><searchLink fieldCode="DE" term="%22Structural+Equation+Models%22">Structural Equation Models</searchLink><br /><searchLink fieldCode="DE" term="%22Evidence%22">Evidence</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1002/rrq.126/abstract
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0034-0553
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The authors assessed 4,865 middle and high school students for the ability to recognize and understand written and spoken morphologically simple words, morphologically complex words, and the syntactic structure of sentences and for the ability to answer questions about facts presented in a written passage and to make inferences based on those facts. Factor analysis of combined accuracy and reaction time results for the tests of simple words, complex words, and sentences resulted in three factors in both the auditory and written modalities, reflecting recognition of forms of simple and complex words, understanding of simple and complex words, and recognition of the structure and understanding the meaning of sentences. Structural equation models showed direct effects of these factors in both modalities on written passage comprehension. The results provide evidence that skills in processing language in the written modality are separate from, although related to, skills in auditory language processing and are important determinants of comprehension of written passages in middle and high school.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: Ref
  Label: Number of References
  Group: RefInfo
  Data: 66
– Name: CodeSource
  Label: IES Funded
  Group: SrcInfo
  Data: Yes
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2016
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1087345
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1087345
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1002/rrq.126/abstract
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 26
        StartPage: 67
    Subjects:
      – SubjectFull: Middle School Students
        Type: general
      – SubjectFull: High School Students
        Type: general
      – SubjectFull: Language Skills
        Type: general
      – SubjectFull: Language Processing
        Type: general
      – SubjectFull: Reading Comprehension
        Type: general
      – SubjectFull: Morphology (Languages)
        Type: general
      – SubjectFull: Syntax
        Type: general
      – SubjectFull: Written Language
        Type: general
      – SubjectFull: Oral Language
        Type: general
      – SubjectFull: Auditory Perception
        Type: general
      – SubjectFull: Factor Analysis
        Type: general
      – SubjectFull: Accuracy
        Type: general
      – SubjectFull: Reaction Time
        Type: general
      – SubjectFull: Difficulty Level
        Type: general
      – SubjectFull: Sentences
        Type: general
      – SubjectFull: Structural Equation Models
        Type: general
      – SubjectFull: Evidence
        Type: general
    Titles:
      – TitleFull: Effects of Written and Auditory Language-Processing Skills on Written Passage Comprehension in Middle and High School Students
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Caplan, David
      – PersonEntity:
          Name:
            NameFull: Waters, Gloria
      – PersonEntity:
          Name:
            NameFull: Bertram, Julia
      – PersonEntity:
          Name:
            NameFull: Ostrowski, Adam
      – PersonEntity:
          Name:
            NameFull: Michaud, Jennifer
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2016
          Identifiers:
            – Type: issn-print
              Value: 0034-0553
          Numbering:
            – Type: volume
              Value: 51
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: Reading Research Quarterly
              Type: main
ResultId 1