Citation
Effects of Causal Diagrams and Outlines on Fifth Grade Students Understanding of Scientific Texts

Material Information

Title:
Effects of Causal Diagrams and Outlines on Fifth Grade Students Understanding of Scientific Texts
Creator:
Gonzalez, Clairemarie
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (131 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Educational Psychology
Human Development and Organizational Studies in Education
Committee Chair:
FRANKS,BRIDGET ANN
Committee Co-Chair:
THERRIAULT,DAVID JAMES
Committee Members:
MILLER,DAVID
CRIPPEN,KENT J
Graduation Date:
5/3/2014

Subjects

Subjects / Keywords:
Causality ( jstor )
Diagrams ( jstor )
Genetic mapping ( jstor )
Learning ( jstor )
Memory ( jstor )
Oxidation ( jstor )
Oxygen ( jstor )
Reading comprehension ( jstor )
Statues ( jstor )
Trucks ( jstor )
Human Development and Organizational Studies in Education -- Dissertations, Academic -- UF
causal -- diagram -- fifth -- grade -- science -- sequence -- temporal -- text
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Educational Psychology thesis, Ph.D.

Notes

Abstract:
Many researchers and educators are interested in students' comprehension of expository text in science. The purpose of this study was to explore the role of a visual spatial display known as a causal diagram in helping readers to draw causal inferences and remember a causal sequence in a scientific text in which the causal sequence was presented in reverse chronological order (i.e., effect first, then the causes leading to the effect). The causal diagram was a visual display that explicitly represented the cause effect relationships presented in the text, but presented them in chronological order (i.e., causal sequence leading to effect). Participants were 99 fifth grade students who attended 4 different public schools in a Northeast Florida County. Students studied the text alone, the text accompanied by the causal diagram, and the text accompanied by an outline. Performance on three composite scores: memory for main idea, memory for the causal sequence, and understanding the causal sequence was analyzed via three separate one way Analyses of Covariance, with Florida Comprehensive Assessment Test (FCAT) Reading Equivalent Scale Scores (RESS) as the covariate. Results indicated that 5th grade students who studied the text accompanied with a causal diagram displayed a greater memory of the steps in the causal sequence compared with those in the Text Only condition and the Text with Outline condition. Students who studied the text accompanied with a causal diagram also displayed a greater understanding of the steps in the causal sequence compared with those in the Text Only condition and the Text with Outline condition. Studying a text with a chronologically organized causal diagram as a study aid facilitates both memory and understanding of causal sequences compared to studying text alone or text accompanied with an outline. The findings from this study provide groundwork for future experimental research and implications for educational practice. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: FRANKS,BRIDGET ANN.
Local:
Co-adviser: THERRIAULT,DAVID JAMES.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-05-31
Statement of Responsibility:
by Clairemarie Gonzalez.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
5/31/2016
Resource Identifier:
907379314 ( OCLC )
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 EFFECTS OF CAUSAL DIAGRAMS AND OUTLINES ON FIFTH UNDERSTANDING OF SCIENTIFIC TEXTS By CLAIREMARIE GONZLEZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 4

PAGE 2

2 2014 Clairemarie Gonzlez

PAGE 3

3 To my Mother

PAGE 4

4 ACKNOWLEDGMENTS I appreciate the contributions of my committee members. Dr. Bridget Franks has helped me extend myself beyond my expectations. She has taught me many things explicitly in the classroom and that a good mentor cares deeply and gives space for her apprentice to take responsibility for her educat ional goals. Each committee member has made a very intimidating and seemingly insurmountable process easier for me. Last, I thank Dr. Crippen for his approachability and willingness to serve on my committee. I appreciate each science teacher who gave me th eir valuable classroom time and each student who completed the study and provided the data without compensation or complaint. I thank my colleagues at the University of North Florida for their support. I thank each person who offered encouragement and cont inued to ask about my constantly offered her encouragement. I thank Dr. Paul Eggen for always giving me his time, knowledge and seeing the potential in me to pursue a Ph.D.. I am also grateful for the encouragement and support of my chair, Dr. Jeff Cornett and I am grateful that I work at a university where my personal and intellectual growth is valued.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 Statement of the Problem ................................ ................................ ....................... 11 Purpose of the Study ................................ ................................ .............................. 13 Theoreti cal Significance ................................ ................................ .......................... 16 Practical Significance ................................ ................................ .............................. 20 2 REVIEW OF THE LITERATURE ................................ ................................ ............ 21 Part I: Scientific Reasoning ................................ ................................ ..................... 21 Causal Inferences ................................ ................................ ........................... 21 The Role of Covariation in the Understanding of Causality .............................. 25 Part II: Comprehending Causality Through Reading ................................ .............. 30 Levels of Representation in Text Comprehension ................................ ............ 30 Causal Inferences with Narrative Texts ................................ ............................ 33 Causal Inferences with Expository Texts ................................ .......................... 39 Background Knowledge and Scientific Text ................................ ..................... 43 Part III: Visual Spatial Displays ................................ ................................ ............... 45 Partial vs. Complete Concept Maps ................................ ................................ 47 Knowledge Maps and Problem Solving ................................ ............................ 49 Technology vs. Paper and Pencil Generated Concept Maps ........................... 52 Collaborative vs. Individually Generated Concept Maps ................................ .. 54 Concept Mapping and Differences in Reading Skill and Text Difficulty ............ 56 Causal Diagrams ................................ ................................ .............................. 57 Outlines as Study Aids ................................ ................................ ..................... 58 Rationale for the Present Study ................................ ................................ ........ 61 3 METHODOLOGY ................................ ................................ ................................ ... 64 Research Questions ................................ ................................ ............................... 65 Research Question 1 ................................ ................................ ........................ 65 Research Question 2 ................................ ................................ ........................ 66 Research Question 3 ................................ ................................ ........................ 66 Particip ants ................................ ................................ ................................ ....... 67 Materials ................................ ................................ ................................ ........... 69 Construction of Test Items ................................ ................................ ................ 70

PAGE 6

6 Memory for Main Idea ................................ ................................ ...................... 71 Memory for Causal Sequence ................................ ................................ .......... 71 Understanding of Causal Sequence ................................ ................................ 71 Validity of Test Items ................................ ................................ ........................ 71 Procedures ................................ ................................ ................................ ....... 73 4 ANALYSIS OF DATA ................................ ................................ .............................. 76 Research Q uestion 1: Memory of Main Ideas ................................ ......................... 76 The FCAT RESS as Covariate with Memory of Main Ideas ............................. 77 Tests of Between Subjects Effects ................................ ................................ ... 77 Research Question 2: Memory of Causal Sequence ................................ ........ 78 The FCAT RESS as Covariate with Memory of Causal Sequence ................... 79 Tests of Between Subjects Effects ................................ ................................ ... 79 Pairwise Comparisons ................................ ................................ ...................... 80 Text only vs. text with outline ................................ ................................ ..... 80 Text only vs. text with causal diagram ................................ ....................... 80 Text with Causal Diagram vs. Text with Outline ................................ ............... 80 Research Question 3: Understanding of Causal Sequence ............................. 81 The FCAT RESS as Covariate with Understanding of Causal Sequence ........ 82 Tests of Between Subjects Effects ................................ ................................ ... 82 Pairwise Comparisons ................................ ................................ ...................... 83 Text only vs. text with outline ................................ ................................ ..... 83 Text only vs. text with causal diagram ................................ ....................... 83 Text with causal diagram vs. text with outline ................................ ............ 83 Summary ................................ ................................ ................................ .......... 83 5 DISCUSSION ................................ ................................ ................................ ......... 88 Memory of Main Idea ................................ ................................ .............................. 90 Memory of Causal Sequence ................................ ................................ .................. 91 Understanding Causal Sequence ................................ ................................ ........... 92 Limitations of the Study ................................ ................................ ........................... 95 Implications for Educators ................................ ................................ ....................... 97 Conclusion ................................ ................................ ................................ .............. 97 APPENDIX A SCIENCE TEXT IN CONSEQUENT ANTECEDENT FORM ................................ 100 B MULTIPLE CHOICE AND SHORT ESSAY ITEMS ................................ .............. 102 C RUBRIC ................................ ................................ ................................ ................ 105 D VALIDATION CHART ................................ ................................ ........................... 110 E CAUSAL DIAGRAM AND DIRECTIONS ................................ .............................. 114

PAGE 7

7 F CONSEQUENCE ANTECEDENT OUTLINE AND DIRECTIONS ........................ 115 REFERENCES ................................ ................................ ................................ ............ 117 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 131

PAGE 8

8 LIST OF TABLES Table page 4 1 Adjusted and Unadjusted Means and Standard Deviations for Composite score 1, Memory of Main Idea and FCAT RESS as a Covariate ........................ 86 4 2 Tests of Between Subjects Effects Composite score1: Memory of Main Idea .... 86 4 3 Adjusted and Unadjusted Means and Standard Deviations for Composite score 2, Memory of Causal Sequence and FCAT RESS as a Covariate ............ 86 4 4 Tests of Between Subjects Effects Composite score 2: Memory of Causal Sequence ................................ ................................ ................................ ........... 87 4 5 Adjusted and Unadjusted Means and Standard Deviations for Composite score 3, Understanding of Causal Sequence and FCAT RESS as a Covariate 87 4 6 Tests of Between Subjects Effects Composite score 3: Understanding of Causal Sequence ................................ ................................ ............................... 87

PAGE 9

9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECTS OF CAUSAL DIAGRAMS AND OUTLINES ON FIFTH UNDERSTANDING OF SCIENTIFIC TEXTS By Clairemarie Gonzlez May 201 4 Chair: Bridget Franks Major: Educational Psychology expos itory text in science. The purpose of this study was to explore the role of a visual spatial display known as a causal diagram in helping readers to draw causal inferences and remember a causal sequence in a scientific text in which the causal sequence was presented in reverse chronological order (i.e., effect first, then the causes leading to the effect). The causal diagram was a visual display that explicitly represented the cause effect relationships presented in the text, but presented them in chronolog ical order (i.e., causal sequence leading to effect). Participants were 99 fifth grade students who attended 4 different public schools in a Northeast Florida County. Students studied the text alone, the text accompanied by the causal diagram, and the text accompanied by an outline. Performance on three composite scores: memory for main idea, memory for the causal sequence, and understanding the causal sequence was analyzed via three separate one way Analyses of Covariance, with Florida Comprehensive Assess ment Test (FCAT) Reading Equivalent Scale Scores (RESS) as the covariate. Results

PAGE 10

10 indicated that 5th grade students who studied the text accompanied with a causal diagram displayed a greater memory of the steps in the causal sequence compared with those in the Text Only condition and the Text with Outline condition. Students who studied the text accompanied with a causal diagram also displayed a greater understanding of the steps in the causal sequence compared with those in the Text Only condition and the Text with Outline condition. Studying a text with a chronologically organized causal diagram as a study aid facilitates both memory and understanding of causal sequences compared to studying text alone or text accompanied with an outline. The findings from this study provide groundwork for future experimental research and implications for educational practice.

PAGE 11

11 CHAPTER 1 INTRODUCTION Statement of the Problem Most of the fundamental ideas of science are essentially simple, and may, as a rule, be expressed in a language comprehensible to everyone (Einstein & Infeld, 1938, p. 27). Written text is one of the most important sources of knowledge attainment. Much of the learning that takes place in and out of schools is based on successful c omprehension of texts. Text processing is entwined with virtually all cognitive functions and processes, including memory, perception, problem solving, and reasoning. Different types of texts place different cognitive demands on readers, however. Narrative text is easier to comprehend than expository text because the content is comparable to the familiar surroundings, experiences, cultural and social interactions that occur in everyday life. Narrative text topics are often human relationships or interperson al problem solving (Cote, Goldman, & Saul, 1998). They possess a causal temporal structure that is often more decipherable to readers than the logical structure of expository texts (van den Broek, Virtue, Everson, Tzeng, & Sung, 2002). During the comprehen sion of narrative texts, readers establish certain levels of coherence (mostly causal and referential), but comprehension of expository texts involves different standards of coherence (van den Broek, 2010). This is especially true when it comes to scientif ic text. Science text is taxing at several levels; its content is usually unfamiliar, and the concepts are often novel and expressed in a complex abstract logical sequence instead of the familiar narrative construction (Stein & Trabasso, 1981).

PAGE 12

12 Almost all science text is written to explain and describe new content to the way of organizing and explaining novel content that is based on the search for truth and stems from empirical evidence (Graesser, Len, & Otero, 2002). This logical propositional format leads mostly to causal network structures, in contrast to the goal structures Newton, 1995). Understanding science often amounts to grasping the meaning of some scienti students are led through a logical chain of evidence (Budiansky, 2001). Content should be organiz ed to show relationships among factual information, concepts, rules, and organizational structures at both the lesson and the curricular level (Harniss, Hollenbeck, Crawford, & Carnine, 1994). The recognition of an organizational pattern aids memory for te xt information because it enables the reader to form a mental representation of the information and to see the logical relationships (Ogle & Blachowicz, 2002). Generating such causal network structures is particularly taxing for readers because scientific explanations in texts often reverse the order of causality of the phenomena they are explaining, i.e., they start by describing effects (Consequences) and work their way back to the causes (Antecedents) (Newton, 1995). For example, the statue of liberty h as turned green (Consequence). It has turned green because it has

PAGE 13

13 undergone a chemical reaction known as oxidation (Antecedent). This structure is the opposite of what is typically presented in narratives with causal chains of events in chronological order where readers can follow events from cause to effect (i.e. a couple meets; they fall in love and marry) (van den Broek, 1997). This reversal may affect are attempting to c omprehend scientific explanations. Purpose of the Study The purpose of this study is to explore the role of a visual spatial display known as a causal diagram in helping readers to draw causal inferences and remember causal sequences in scientific texts in which the order of causality is reversed. Visual spatial displays represent objects, concepts, and their relations using symbols and spatial arrangements (Vekiri, 2002, p. 262). A causal diagram is a special type of visual spatial display that organizes t he causal relationships of a process or sequence of events with arrows indicating direction of causality. In a causal relationship, event A precedes and understanding of causal relationships within and across paragraphs, but their usefulness as study tools has not been explored extensively (McCrudden, Schraw, Lehman, & Poliquin, 2007). Studying a causal diagram may facilitate understanding of causal relationships because causal diagrams improve encoding by helping readers determine what is most pertinent to understanding causal sequences. McCrudden, Schraw, Lehman, and Poliquin (2007) examined the effect of studying a causal diagram on comprehension of causal relationships from an expository science text. Two groups of college students were presented with an expository text that described the effects of space travel on the

PAGE 14

14 human body. A causal diagram for the text showed five cause and effect sequences that occurred as a result o f a single cause (lack of gravity experienced during extended space travel). The control group (text only) and experimental group (text and causal diagram) performed similarly with respect to memory for main idea, but the readers who studied the causal dia gram while reading the text understood the five causal sequences better. A crucial factor in comprehending causal explanations in scientific texts, h Penalba (2002) investigated this factor, using a text that explained the death of a river from pollution by describing different polluting factors and listing in chronol ogical order the events that led to the final consequence, the death of the river. One version of the text had an Antecedent Consequent (AC) format, with the different polluting factors listed in chronological order. In the other version, a Consequent Ante cedent (CA) structure was used, in which the final consequence (death of the river) was offered first in the text, followed by its causes. Readers of both versions answered questions about the texts and also created causal diagrams based on what they had r ead. When the causal sequences produced by readers of the two different versions of the text were compared, participants had made more Antecedent Consequent associations than Consequent Antecedent associations in their responses, independently of the text s with two different chronological orders, regular science texts frequently present

PAGE 15

15 explanations in the Consequent Antecedent form, beginning with an observed effect and working backward to illustrate the causes of that effect (Stein & Trabasso, 1981; Brun This study explore d such texts ( in which a final consequence is offered first, followed by the causes) by providing them with a forward directional causal diagram in which the causes are given first, and lead in chronological order to the consequence. My hypothesis wa s that Antecedent structure text w ould be higher when the t ext wa s accompanied by an Antecedent Consequent causal diagram, because the causal diagram w ould reflect causal chains A different type of study aid, the outline, may also help readers comprehend and remember causal relationships in science texts An outline is an organized listing of the essential features or main aspects of concepts in a text. Outlines are useful as study aids because they include only the more important text information and convey hierarchical concept relations. The usefulness of outlines has been demonstrated by many researchers (e.g., Darch & Gersten, 1986; Glynn, Britton, & Muth, 1985). It is common for textbooks to inc lude chapter outlines, for instructors to write lecture outlines on the chalkboard, and for students to take outline notes. Outlines may also help readers comprehend and remember causal relationships in science texts, but only because they direct their att ention to essential elements of the text. Since outlines follow

PAGE 16

16 texts, an outline of a text with a Consequent Antecedent structure will also have a Consequent Antecedent structure. Although outlines are a popular method of organizing expository information there are likely problems that may result from studying them. Outlines, like text, have a linear format that discourages depicting relations among concepts. While outlines convey hierarchical, within concept relations, they obscure important coordinate, among concepts relations. Thus, the information that outlines contain is not the problem; it is their linear format that discourages students from integrating information across the concepts (Waller & Whalley, 1987). In this study, I will compare the effe cts of an outline in Consequent Antecedent form and a causal diagram in Antecedent Consequent form as study aids for a scientific text involving causal relations. Theoretical Significance This study explores how the order of presentation of causal sequenc es in scientific text affects comprehension of those texts. In science, the usual intention is to establish causal generalizations to explain a sample of observations (Hart & Honore, 1959; White, 1989). The processing of expository texts requires abstract categories, mechanisms, descriptions, and arguments, as well as novel vocabulary (Black, 1985; Graesser, Len, & Otero, 2002). These elements are often organized into abstract structures, such as linear chains and hierarchies, and into rhetorical networks, such as the comparison of two or more elements (Black, 1985). Science text is often encumbered with unfamiliar specialized technical terms that students are required to interpret and memorize ( In addition, expository text is often cognitively demanding, forcing the reader to employ strategic, time consuming analytical reasoning to generate causal consequence

PAGE 17

17 inferences. Scientific genres often assume readers are capable of generating the necessary inferences for text coherenc e. However, readers do not always make the inferences that are required in expository text for comprehension, and when they do, they are cognitively costly (Britton, van Dusen, Glyn, & Hemphill, 1990; Graesser, 1981). Because of the limitations of working memory, most of the information to be reinstated and used for inference generation is that from adjacent text segments; thus, readers retain in working memory the information that is the causal antecedent of the sequential text they read (Cote et al., 1998 ; Fletcher & Bloom, 1988). In circumstances where working memory is taxed, the global connections to the overall cause and effect chain are forfeited, reducing the number of inferences that would be used to assimilate the new information into the overall c ausal sequence. This cognitive overload limits the number of inferences generated and produces a surface level understanding of the text in which complex causal relationships are not always comprehended (Graesser & Bertus, 1998). For example, in their stud y of the causal information encoded by adult readers of scientific text, Graesser and Bertus (1998) observed that causal consequence inferences were more time consuming than causal antecedent inferences, and readers were more likely to construct inferences that referred to causal antecedents than to causal consequences of an event mentioned in the text. Visual displays promote learning because they reduce the amount of effort needed for understanding complex texts (Cheng, 2001). They provide a temporal and spatial organization that increases the salience of the relationships the text is expressing. The diagram communicates the implicit structure of the text (not necessarily

PAGE 18

18 the implicit causal relations, because sometimes those relations are actually stated in the text) and this affects learning of causal relationships. In other words, when we read science texts that are explaining what causes an effect, readers do not necessarily infer a cause that is not stated, as they might in a narrative. But they may f ind it difficult to generate for themselves the entire causal network conveyed by the text, especially when the text reverses the temporal order, as many science texts do. The diagram helps the reader to extract the causal structure out of the text. Extran eous cognitive load describes the working memory requirements of the format of the presentation of the information (Lee, Plass, & Homer, 2006). For visual displays, numerous effects that reduce extraneous load have been investigated. Among them is the spli t attention effect, which states that the integrated presentation of visual and verbal information is more effective than their split presentation (Chandler & Sweller, 1991, 1992; Sweller, Chandler, Tierney, & Cooper, 1990; Tarmizi & Sweller, 1988; Ward & Sweller, 1990) corresponding words and pictures are presented near rather than far from each other on While Science texts ar e cognitively demanding for a number of reasons, the reversed order in which causal sequences are often presented is an aspect of Science text processing that has not been studied extensively. Scientific explanations in texts typically reverse the order of Newton, 1995). This is the oppos ite of the order typically found in narratives with causal

PAGE 19

19 chains of events, where readers can follow events from cause to effect (van den Broek et al., 2002). understand causal relationships when they are attempting to comprehend scientific Perhaps just reducing the text to its essential points will improve comprehension. Will an outline fun ction just as well as a causal diagram? Outlines are useful as study aids because they include only the more important text information and convey hierarchical scientific texts may be affected simply by directing their attention to essential elements of text via an outline provided as a study aid (Robinson & Schraw, 1994; Robinson & Kiewra, 1995). Directing attention to essential elements may not be sufficient for improving compreh Linderholm, Everson, van den Broek, Mischinski, Crittenden, & Samuels, (2000). If temporal order is highly si especially in science, then a diagram that reflects their thinking (cause to effect) should having an outline o f that same text. Therefore, the theoretical implications of this study may be that understanding causality in science texts is not just difficult due to the essential elements of the texts in general; we also need to pay attention to the difficulty creat

PAGE 20

20 form of many science texts, because it is the opposite of what people have become accustomed to in their broader experience with narrative texts (van den Broek, 1997). Support for this idea would be demonstrated if an outline im proves performance over just reading the text, but a causal diagram in a forward direction improves performance more than an outline. Practical Significance From a practical perspective, it is important to establish whether directional causal diagrams can help learners understand complex causal relationships. My presumption is that by extracting the causal structure and making implicit causal sequences explicit, causal diagrams enable readers to better understand the direct causal relationships as well as t Newton, 1995); thus, science texts are often not organized in the Antecedent Consequent fashion. Providing learner s with a directional causal diagram during study is a relatively simple educational intervention. An instructor can help learners focus their attention on relevant text information by providing a causal diagram that displays direct effects and/or indirect effects of the integrated causal network. It would be far easier for teachers to make causal diagrams in Antecedent Consequent order for students to use to understand Consequent Antecedent texts than it would be for teachers to rewrite the texts into Antec edent Consequent form (AAAS, 2001).

PAGE 21

21 CHAPTER 2 REVIEW OF THE LITERATURE Part I: Scientific Reasoning Scientific thinking is defined as the use of the techniques or principles of scientific investigation to reason or solve problems. It involves the developm ent of skills that are essential in generating, testing, and revising theories and reflecting on the process of knowledge acquisition and change (Koslowski, 1996; Kuhn & Franklin, 2006; Wilkening & Sodian, 2005; des the skills involved in inquiry, experimentation, evidence evaluation, and inference that are done in the service Individuals employ some or all of the mechanisms of scientific inquiry when generating or revising theories about the phenomenon under examination (Zimmerman, 2007). An important strategic acquisition is the control of variables, a basic domain general strategy, which permits valid inferences as it constrains the sea rch of possible experiments (Klahr, 2000). Kuhn, Iordanou, Pease, and Wirkala (2008) identified the ability to explore multiple variables, not just the effect of a single variable, as essential for scientific reasoning (Kuhn et al., 2008). Causal I nferenc es Causal interpretation is the core and foundation of cognition about the objective world (Hong, Chijuna, Xuemeia, Shana, & Chongde, 2005). The generation of causal inferences about the physical world requires using a causal mechanism, such as the moveme nt of objects, as a basis for making a causal inference, e.g., she kicked the ball (cause), the ball rolled forward (effect). The direction of inferences in causal reasoning about objects in the physical world (i.e., from cause to effect versus effect to c ause)

PAGE 22

22 appears to be similar to the causal reasoning of text comprehension (i.e. forward inference, from cause to effect versus backward inference from effect to cause). The causal structure of a text is an essential element in how the reader constructs th e mental representation and how the text is understood by the reader (Linderholm, et al., 2000; O'Brien & Myers, 1987; Trabasso, Secco, & van den Broek, 1984; Trabasso, van den Broke, & Suh, 1989). Causal inferences explain why and how events, actions, and states occur in the context of reading. These inferences are an attempt to maintain coherence by activating episodic memory, retrieving memory of prior text, and/or accessing background knowledge and semantic memory. Texts that have numerous causal connec term memory; as a result, they are remembered and comprehended better than texts with fewer Sperry, 1985; Trabasso & van den Broek, 1985; van den Broek, 1988). Likewise, events that connect a series of events in a causal network are better remembered than events that are not connected causally (Black & Bower, 1980; Trabasso et al., 1984; Trabasso & van den Broek, 1985). The generation of causal inferences allows the reader to develop a mental continuously upgraded as new text events occur and further causal connections are revealed (Linderholm et al., 2000). Characteristically, the previous sentence remains active in working memory as reading continues, connecting it to the information contained in the current sentence. The average reader can maintain in working memory about one or t wo sentences at a time (Fletcher & Bloom, 1988) limiting the ability to

PAGE 23

23 process or sustain in memory all of the causal connections present in the text (Linderholm et al., 2000). If a reader cannot retrieve applicable information from working memory, a caus al connection cannot be made between contiguous sentences. Without a causal connection a causal inference cannot be made, thus incoherence transpires. At this point the reader makes a more comprehensive search of long term memory for previously encountere d text information to make a causal connection (Kintsch, 2004). The generation of a situation model involves the formation of a meaning based representation of the text for comprehension. The construction of an integrated and coherent mental model of a te xt relies upon the processes of integration and inference. The reader m ust integrate contiguous clauses to establish local coherence, and make inferences about diverse events, actions, and states to make the text come together as a whole. These processes r equire that the relevant information, either from the text or background knowledge, is available to the reader. Readers use this memory representation to recognize syntactic structures, recall facts, and apply this knowledge to the assessment of their comp rehension. Difficulty occurs when the reader does not have sufficient background knowledge about the topics in expository texts, so readers generate fewer inferences than they generate during the comprehension of narrative text (Britton & Glgz, 1991; Gra esser, 1981; Graesser, Singer, & Trabasso 1994). Scientific Understanding of Causality Often when trying to explain an event/outcome/result, there is a lack of information. Scientists generate inferences, utilizing their prior knowledge, to explain whether information is relevant to the explanation, that is, whether it is evidence. This reas oning from an existing situation to a hypothesized cause of that situation is termed

PAGE 24

24 to IBE, one explanation is chosen over its competitors because the chosen explanati on provides a better explanation than competing theories. This is achieved in part by attaining consistency with well established background information. As a result, the theory itself gains in credibility over its competitors because it can account for in formation that other explanations cannot (Koslowski, Marasia, Chelenza, & Dublin, 2008). Thus, beliefs about which things may cause an effect or the kind of effect they cause play critical roles in each phase of a scientific experiment (Masnick & Klahr, 20 03). Young children do not always understand the importance of inference to scientific work. For example, in a study of 62 sixth grade students' perceptions of science, Khishfe and Abd El Khalick (2002) asked participants how scientists use observation and inference to learn about dinosaurs. The initial results indicated that most students believed scientists used evidence such as bones and fossils to explain what dinosaurs looked like (shape of the dinosaurs). However, when asked how scientists knew what c olor the dinosaurs were, most children said the scientists just guessed the color or could look it up via the internet. In reality, children believed scientists had actually seen whole dinosaurs, and did not believe that scientists inferred what dinosaurs looked like based on fossil evidence (Khishfe & Abd El Khalick, 2002). difficulty with causal reasoning about objects in the physical world to the direction of inferences, forward vers us backward. In their study, every participant was given a cause effect inference task, in which children ages 3 to 5 were required to infer

PAGE 25

25 outcomes from causes. In addition, they were also given an effect cause inference task, in which children looked fo r causes referring to effects. This experiment revealed the cognitive limitations of 3 to 5 year old children; the majority could only infer causality with a simple if then rule from causes to effects. The Role of Covariation in the Understanding of Causality Hume (1978/1739) identified the covariation of perceptually significant events as one possible cue that two events are causally related (Hergenhahn, 2005). If event A happens, and is closely followed by event B, one infers that event A causes eve nt B. However, the problem with this sort of reasoning is that the association of A and B might be due to another event (C), which produces event B and is associated with event A. A does not cause B, but whenever C occurs, both A and B will occur. For exam ple, a man may notice that when he eats pizza in the evenings, he is most likely to have heartburn. It could be that the pizza is causing his heartburn; however, suppose he usually drinks beer when he eats pizza. The beer might be giving him heartburn, ind ependently of the effect of the pizza. The pizza might be associated with the heartburn, but it would not cause the heartburn. Eating pizza would be associated with heartburn; however, it would be incorrect to conclude that there was a causal relation betw een the two. The man could eat pizza (A) without drinking beer (C), or drink beer (C) without eating pizza (A) to see if the heartburn (B) was absent or present in either case. The causal relationship between A and B is removed by holding C constant. Thus, there is no direct causal relationship between A and B. Hans Reichenbach, a philosopher of science, termed this situation as: C screens off A from B (Reichenbach, 1956). This type of reasoning is used throughout science: It is the underlying principle be hind both methods of experimental design and statistical methods. Even children as

PAGE 26

26 young as 2 and 3 have a tendency to use the covariation of events (antecedent and outcome) as an indicator of causality. Gopnik, Sobel, Schulz, and Glymour (2001) gave child not others, were placed on it. Children used screening off to infer which objects had the causal power to make the machine go. Results indicated young children are able to use screening off to make accurate and genuinely causal inferences in the domain of sufficient cue for inferring a causal relationship, it is one of the bases for making ind When evaluating hypotheses, scientists often use evidence in the form of covariation patterns, e.g., the combinations of the presence and absence of antecedents (or potential causes) and outcomes. Koslow ski et al., (2008) investigated under what conditions information about an event is treated as evidence relevant to explaining the event. Participants were presented with events for which there were two possible and plausible explanations, as well as with two pieces of background information. Results indicated that such information is more likely to be viewed as evidentially significant to an event when there is an explanation that can place it into a broader causal framework. For example: the background in formation that two populations have languages with different grammatical structures is viewed as pertinent to explicating stature differences among the populations only if there is an explanation such as different gene pools that can place the information into a broader causal framework. Koslowski et al., (2008) observed that when an explanation for an occurrence contained background information, the explanation became more plausible

PAGE 27

27 e allows us to p. 482). Kuhn, Amsel, O'Loughlin, Schauble, Leadbeater, and Yotive (1988) examined how participants reconcile their theories about causal variables with covariation evidence presented to them. Adults, sixth and ninth graders were questioned about their beliefs concerning the types of foods that make a difference in whether a person caught a cold (35 foods in total). Four variables were selected: the causal theory, two factors that the participants believed make a difference in catching colds (e.g., types of fruit and cereal) and the noncausal theory, two factors that participants did not believe made a difference (e.g., type of potato and condiment). Participants were presented with evidence that confirmed one existing causal theory and one noncausal theory. In addition, non covariation evidence was offered that disconfirmed one prior causal theory and one noncausal theory. Participants were presented with a series of questions about what the evidence supported for each of the four variables. Results indicated that theory based responses were related to prior beliefs. Participants generally exhibited strategies to bring theory and evidence into alignm ent with one another. When participants were presented with cases in which evidence and theory were congruent, they provided a theory based response. For example, if given a pattern that indicated the type of cake covaried with catching a cold, participant s claimed the healthy children ate carrot cake and the sick children ate chocolate cake. Only after explicit probing for evidence did participants use evidence, such as children at table two have tissues and the children at table one do not have tissues, t o support

PAGE 28

28 their theories. Another strategy used was biased evaluation of the evidence to reduce its discrepancy with a theory, even when the strategy was difficult to maintain as discrepant evidence mounted. When participants were presented with evidence d epicting a pattern of non covariation between variables, they distorted, ignored, and/or selectively chose evidence consistent with their desired theory (Kuhn et al., 1988). In summary, adults and children reconcile their theories about causal variables wi th covariation evidence. They employ strategies to bring theory and evidence into alignment with one another even when discrepant evidence escalates. In marked contrast to the above results, researchers Ruffman, Perner, Olson, and Doherty (1993) observed the ability of 4 to 7 year olds to form hypotheses using covariation evidence with less complex tasks and fewer variables. When participants were given only one possible cause (type of food) that covaried with an outcome (tooth loss), children as young as 6 could form the hypothesis that the factor is causally responsible based on perfect or partial covariation evidence. Ruffman et al. (1993) argued that preschoolers have a basic understanding of the hypothesis evidence relation because it can be related t o the requirements of the false belief tasks in theory of mind research; such tasks are usually mastered by the age of four. Participants are evaluation abilities can based on the evidence presented. If participants have a basic understanding of the hypothesis essential for hypothes is formation, and that faked evidence will therefore lead people to entertain false hypotheses.

PAGE 29

29 Koerber, Sodian, Thoermer, and Nett (2005) used three evidence evaluation tasks (a Revision of Prior Belief task, a Faked Evidence task and a Partial Evidence evidence understanding, evidence evaluation skills, and causal beliefs. Each participant was presented with different patterns of covariation evidence (perfect covariation, imperfect covariation, non covariatio n). In the Revision of Prior Belief task, children were told that a character (Robby) believed green chewing gum causes teeth to fall out. The participant and Robby were then presented with ten pictures displaying children with red chewing gum and missing teeth and ten pictures displaying children with green chewing gum and healthy teeth (counterevidence in perfect covariation). The participants were then an absent character, by switching evidence (although red chewing gum makes teeth fall out, they arranged the evidence to show green chewing gum with children whose teeth were missing). Th e second character, Paul, appeared and saw the evidence. The (1993) experiment to test whether young children can predict causal beliefs on the basis cted access to covariation information (Partial Evidence task). In the which kind of handkerchief makes a red nose. After seeing evidence of imperfect

PAGE 30

30 covariation, the children were asked about their own interpretations. The findings were the hypothesis evidence relation can be obtained with the Faked Evidence technique. In addition 5 and 6 year olds can form beliefs based on evidence, and recognize that story characters may possibly change their beliefs based on covariation evidence. Moreover, even the 4 year olds demonstrated competence when tested with the Revision of Prior Belie f task and the Partial Evidence task (perfect covariation pattern). Thus, at age 4 most children can set aside their own beliefs and recognize that the limited evidence available to a story figure supports a different causal belief. The direction of infer ences in causal reasoning about objects in the physical world (i.e., from cause to effect versus effect to cause) appears to be similar to the causal reasoning of text comprehension (i.e. forward inference, from cause to effect versus backward inference ef fect to cause). Research studies have provided useful information about how people make causal inferences about the actual world and their experiences, but what happens when they are trying to understand what they read about the world rather than what they experience directly? To understand this, we need to know how text is generally represented by readers, and then how causality is understood and represented by readers in both narratives and expository texts. Part II: Comprehending Causality Through Readin g Levels of Representation in Text Comprehension Many factors determine the difficulty of reading: sentence length, familiarity of the words, the number of ideas expressed, coherence, and the text structure (Kintsch, 1994, 1998/2007). The representation of text divides into five major levels: surface code, textbase, situation model, problem model, and pragmatic communication level

PAGE 31

31 (Graesser, Millis, & Zwaan, 1997; Kintsch, 1994, 1998/2007; van Dijk & Kintsch, 1983). The first three of these levels are most pertinent to this discussion. The surface code ( the lines, shapes, angles, color, and size of the characters that form the words ) is the most superficial level (Graesser et al., 1997; Kintsch, 1994, 1998/2007; van Dijk & Kintsch, 1983). At this level, read ers construct a mental representation called the surface level memory. Surface level memory is an exact replica of words and phrases of the text but it is often short lived. The memory of exact words and phrases is often important for jokes, poems, or log ical arguments but for most instructional text, it is not necessary (Kintsch & Bates, 1977). For instructional text, the semantic or textbase level, the i dea expressed by the text is of more importance (Kintsch, 2004). The textbase is the memory for the meaning of words and their explicit associations (Thiede, Anderson, & Therriault, 2003) that preserves the meaning, but not the surface code. Propositions are one way to identify the elements that make up an idea in the text. A proposition consists of a pr edicate (main verb, adjective) that interconnects one or more arguments (noun referents, embedded propositions) (Graesser et al., 1997; Kintsch, 1994, 1998/2007). The global representation of what the text is about is the next level, called the situation model. The global representation involves the construction of a mental model the gist or a summary of the text (Kintsch, 2004). Situation models are not necessarily verbal. People often resort to imagery, constructing mental images of maps, diagrams, or pictures of what the text is describing. Text comprehension involves the construction of a meaning based depiction of the text (Gernsbacher, 1991; Johnson Laird, 1983;

PAGE 32

32 Kintsch, 1994, 1998/2007). It is rare that comprehension is evoked purely from the textbase. Readers amend the textbase with prior knowledge and experiences from their long term memory (Kintsch, 2004), often including information that was not in the text. However, the comprehension process is more than just filling in gaps; it allows the reader to organize the events, setting, people, and actions in the text ( Graesser et al., 1994; Johnson Laird, 1983, 1989; Kintsch, 1988, 2004; Lutz, & Radvansky, 1997; Tr abasso et al., 1989). This organized combination of the amended textbase comprises the mental model (Kintsch, 1994, 1998/2007). The amount of elaboration that transpires a vailable (Graesser et al., 1994; Kintsch, 1994, 1998/2007). These factors influence the An essential element in how the reader constructs the situation model is the causal structure of a text (Lin derholm et al., 2000; O'Brien & Myers, 1987; Trabasso, Secco, & van den Broek, 1984; Trabasso et al., 1989). Causal inferences explain why and how events, actions, and states occur in the context of reading. These inferences are an attempt to maintain cohe rence by activating episodic memory, retrieving memory of prior text, and/or accessing background knowledge, and utilizing semantic memory. Texts that have numerous causal connections are more likely to be represented in the term memory; as a result, they are remembered and comprehended Myers, 1987; Trabasso & Sperry, 1985; Trabasso & van den Broek, 1985; van den Broek, 1988). Likewise, events that connect a se ries of events in a causal network are

PAGE 33

33 better remembered than events that are not connected causally (Black & Bower, 1980; Trabasso et al., 1984; Trabasso & van den Broek, 1985). S urface code, textbase, and situation model are assumed to be present regardl ess of text type. (Graesser Singer, & Trabasso 1997; Kintsch, 1994, 1998/2007; van Dijk & Kintsch, 1983) Thus, causal inferences happen in both narrative and depending on whether they are comprehending the causes of events or goals in a story, or understanding causality in a scientific sense, in scientific texts. Causal Inferences with Narrative Texts Case (1999) attributed cognitive development to control structures, and to goals and strategies. According to Case, one type of control structure, the central narrative structure, emerges around 4 years of age. Children begin to use a theor y of mind schema that permits them to infer the thoughts or feelings of others, and to solve the classic false belief task. They also start to use an event sequence (theory of action) schema that permits them to verbalize social scripts and make causal sta tements and or predictions about what will happen next. Along the same line, Trabasso and Nickels (1992) argued that the more knowledge children have of goals, plans, actions, and outcomes, the more they demonstrate a coherent understanding of text. Trabas so and Nickels (1992) view the episodic structure as an efficient psychological unit, configured as a working memory chunk in the online processing of stories. The more knowledge children have of goals, plans, actions, and outcomes, the more they demonstra te coherent understanding of what characters do and the more likely they are to construct coherent understanding of new goals and actions. The

PAGE 34

34 causal inferences connect the components of a relating goal plan and form a straightforward episodic unit (settin g event internal response goal attempt outcome) These goal attempt outcomes categorize a sequence of events serving as building blocks for encoding multifaceted narratives when the episodes themselves are interrelated. The causal network model of Trabass o et al. (1989) represents and generates hierarchical goal plans. Each clause's content in the narration is classified into one of six categories: S (setting), E (event), IR (internal response), G (goal), A (attempt), and O (outcome). Outcomes are marked a s successful (+), unsuccessful ( ), or neutral (o) with respect to goal attainment. These categories together constitute an episodic unit. Trabasso and Nickles (1992) presented a picture story, Frog, Where Are You? by Mayer (1969), to four groups of childr en ages 3 4, 4 5, 5 6, and 9 10 years of age. The children were asked to tell a story based upon a booklet of pictures after paging once through the booklet. The causal network model of Trabasso et al. (1989) was used to represent and generate hierarchical goal plans. Each clause was coded by a letter that denotes its category in the causal network model. For example, the plan begins with a setting, coded (S), that is followed by an event coded (E) that happens to a protagonist. These, respectively, enable and psychologically cause an internal reaction, coded (IR). The reaction leads to a goal, coded (Gl). This goal motivates a sub goal, coded (G2), to obtain it. Trabasso and Nickels (1992) found that in encoding picture stories, the basic goal attempt outco me episodes emerge and increase numerically from age 5 to adulthood, becoming elaborately complex as the readers mature. Their results indicated that 9 year old children and adults described the pictorial story according to a

PAGE 35

35 hierarchical goal plan of acti on. As readers mature, relational and schematic processes combine in a single framework in which the story grammatical and causal roles of text elements can be represented simultaneously. The results also provide evidence of developmental differences in ch ildren ranging from 3 to 5 years of age. For example, the 3 year old children described outcomes that were unrelated to the central theme. In contrast, the 4 year olds encoded actions related to the central theme but omitted goals and purposes, whereas the 5 year olds added the goals and purposes. These developmental differences determined year olds failed to lay the groundwork of establishing who the main characters were and their relati onship to each other. The 4 year olds were better able to set the stage by using additional knowledge of the characters, supplied by the pictures and the addition of new the picture. The construction of causal coherence by the reader establishes when inferences will be generated. Backward inferences become increasingly necessary when one or more of the causal conditions is not met by clear cut statements in the story. The se backward inferences call for the reader to deduce critical information needed to fulfill the causal criteria. (1992) found that when encoding a picture story with children 5 years and younger, but this reverses as children mature. In contrast, forward inferences are optional and less likely to be involved in the construction of a causal pat hway, since they signify

PAGE 36

36 expectations or predictions about information presented later in the text. In summary, these episodic units ( setting event internal response goal attempt outcome) organize narrative events and serve as building blocks for encoding complex narratives when the episodes are interconnected. From age 5 to adulthood, when encoding picture stories, episodic units emerge and increase in number and complexity as the narrators mature. Casteel (1993) examined inferential processing during read ing of third, fifth, eighth graders, and adults. Readers were presented with one of three versions of text: 1) an explicit forward inference version, text containing a consequence inference that was strongly implied by the last sentence, but not necessary to understand the passage as a whole, or 2) an integrative text that contained an explicit backward inference, the same premise sentence now in the second sentence position, followed by a sentence consistent with the story that did not refer to the inferen ce, or 3) an explicit inference version with both a backward and forward inference. In the explicit inference versions (2 and 3), the premise sentence, in the second sentence position, was followed by a sentence that explicitly mentioned the occurrence of the consequent inference. In both explicit versions, the final sentence of each passage was the same inducing sentence. Results from a reading speed analysis indicated that the backward inference inducing sentences were read more slowly than the forward in ference inducing sentences by all four age groups. For third graders to adults the reading times of the stories in the superficial reading condition averaged only 535 ms, whereas reading times averaged 1,230 ms for stories in the integrative reading condi tion. Thus the backward inducing sentences were read more slowly than the explicit inducing sentences by all four age

PAGE 37

37 groups, implying that additional processing must have occurred for the backward versions of the text. van den Broek (1990 a 1990 b ; Trabass o & van den Broek, 1985) proposed that causal relationships must meet four criteria: the first two temporal priority and operativity state the cause must always precede the consequent and be active when the consequent occurs. The third criteria necessity states that the cause is necessary for the consequent to occur; the consequent cannot, in the circumstances of the story, occur in the absence of the cause. Lastly the sufficiency criterion basically argues that the presence of the cause is suf ficient for the presence of the consequent, again within sensitive to the causal constraints in text, but young children are as well. It appears that children as yo ung as third grade are responsive to a sufficiency criterion for establishing causal coherence (Casteel, 1993). The role inferences play in the construction of causal coherence by the reader determines when they will be generated. When one or more of the c ausal criteria is not met by the explicit statements in the text i nferences become increasingly necessary. As children mature, life experiences provide them with background knowledge. Readers routinely activate this background knowledge to assist them in generating causal and referential inferences. van den Broek (1997) investigated whether 8 11 ,14 and 18 ye ars of age children perceive an event as more central or salient if it has many causal relations than if it has few such relations. Two stories were constructed; the events in both stories were organized in three episodes, each of which revolved around a g oal. The episodes describe a boy wanting to have a bike, the boy trying to obtain

PAGE 38

38 money, and the boy wanting to have a paper route. The three episodes are connected to each other in the sense that the second episode (getting money) takes place in order to successfully complete the first one (buying a bike); likewise, the third episode (landing a job) takes place in order to complete the second (and first) one. Thus, the episodes are related in a hierarchical fashion. Children in each age group judged a goal to be more important the more causal consequences it had. Overall the 8 year old children judged goals to be less important than did the older children. This pattern dem onstrates that children as young as 8 years of age rely on the identification of caus al relations to comprehend the events that they experience. van den Broek (1997) concluded that the greater amount of causal relations that readers identify in a text, the more coherent they perceive the text to be, and the better they remember it. Childr en as young as 8 years of age are capable of judging statements as more significant if they contain numerous, rather than a small amount of intraepisodic causal connections. For the 8 year old children the average importance of the interepisodic anchor eve nts, goals, and successful/failed outcomes, was equal to that of the actions. In addition, the children in the three other age groups the average importance rating of the interepisodic anchor events was higher than that of the actions. The older children j udged the anchor events more important than did the 8 year olds. These results show that events that serve important interepisodic functions consistently are judged more important by children 11 years and older. Narrative representations described as causa l networks capture the causal properties of the events of the story, the chain of events, and the number of connections to other events. A variety of types of relations exist; however, in both narrative and expository text, causal relations have

PAGE 39

39 been found to play an essential role in the organization of mental representations ( Bartlett, 1932; Dewey, 1938; Piaget 1927 a 1927 b ) Readers of a wide range of ages and abilities construct these networks by inferring connections between causal antecedents and cons equences (van den Broek, 1997). Causal Inferences with Expository Texts During the comprehension of narrative texts, readers establish certain levels of coherence (mostly causal and referential), but comprehension of expository texts involves different st andards of coherence. This is especially true when it comes to scientific text, which has a unique way of organizing and explaining novel content that is based on the search for truth and stems from empirical evidence. Science text is taxing at several lev els because its content is usually unfamiliar, and the concepts are often novel and expressed in a complex abstract logical sequence instead of the familiar narrative construction. Almost all science text is written to explain and describe new content to t Another distinction between narrative and scien tific text is based on the extent of generalizations and the amount of observations that are needed to construct a causal explanation. In science, the usual intention is to establish causal generalizations to explain a sample of observations (Hart & H onore, 1959; White, 1989), in contrast to narratives that explain single events and cases. Expository texts are much less predictable in form than narratives. The processing of expository texts requires abstract categories, mechanisms, descriptions, and ar guments. These elements are often

PAGE 40

40 organized into abstract structures, such as linear chains and hierarchies, and into rhetorical networks, such as the comparison of two or more elements (Black, 1985). An additional factor that influences text comprehension is text structure. To organize information at a more global level, writers use conventional expository schemata (Baker, 1985; Graesser & Person, 1994). Meyer and Freedle (1984) distinguish five of these: collection, description, causation, problem solutio n, and compare contrast. Each of these text structures corresponds to a distinctive schema. In comprehending science text, readers make use of the conventional expository schemata typical for the particular scientific field to which they are being introduc ed. In addition, readers use other forms of linguistic and syntactic knowledge, including external cues about the text structure, such as enumeration markers, chronology, and logical connectors between sentences (Elshout Mohr, & van Daalen'Kapteijns, 2002) Chronology is an important aspect for organizing causality of events in narrative text. In the real world, causes precede effects; however, this is not always the format in scientific text. In scientific investigation, possible causes are evaluated in th e context of alternative accounts. Many scientific explanations reverse the order of causality, starting with the presentation of the problem and then attempt to answer the query of why the problem has transpired. Although chronological structures aid in t he construction of a causal model, they do not always ensure deep understanding of scientific text. When readers have only a general knowledge about the text topic, they have a tendency to use the chronological organization to construct causal mental struc

PAGE 41

41 Temporal Order and Causality in Expository Texts Causal interpretation is the cor e and foundation of cognition about the objective world (Hong, Chijuna, Xuemeia, Shana, & Chongde, 2005). The generation of causal inferences about the physical world requires using a causal mechanism, such as the movement of objects, as a basis for making a causal inference, e.g., she kicked the ball (cause), the ball rolled forward (effect). The direction of inferences in causal reasoning about objects in the physical world (i.e., from cause to effect versus effect to cause) appears to be similar to the c ausal reasoning of text comprehension (i.e., forward inference from cause to effect versus backward inference from effect to cause). 2002; Linderholm et al., 2000; Mackie, 1980; van den Broek, 1990). Readers are able to identify a causal relation without difficulty if the cause precedes its consequent in the text (Linderholm et al., 2000). Causal antecedent inferences are generated more quickly and consistently than predictive causal consequence inferences (Graesser & Bertus, 1998; Maury et al., 2002). For example, a causal antecedent inference precedes the consequent in the following: She touched the pot cooking on the stove. She was badly burned. In contrast, predictive causal conseq uence inferences are usually characterized by some indication to the reader of "what will happen next." So, in the example given by Potts, Keenan, and Golding (1988), No longer able to control his anger, the husband threw the delicate porcelain vase agains t the wall, a predictive causal consequence inference would most probably be the vase broke. When events are out of chronological order, that is, when the consequent event precedes its antecedent in a text, additional

PAGE 42

42 effort or attentional resources are ne eded, because readers must reorganize events mentally prior to determining if a causal relation is present (Maury et al., 2002). investigated whether causal structure, expressed in ch ronological order, is a vital factor in causal explanations in scientific texts. In this study, participants were presented with text that explained different polluting factors and listed in chronological order the events that led to the final consequence, the death of the river. One version had an antecedent consequent format, with the different polluting factors listed in chronological order. In the other version, a consequent antecedent structure was used, in which the final consequence (death of the riv er) was offered first in the text, followed by its causes. Participants answered a questionnaire to assess the levels of mental representations (surface code, textbase, and situation model) they reached as a result of the text's causal structure and their prior knowledge. Participants were also asked to make a causal diagram, and the researchers compared the causal sequences that participants produced to see if one of the causal structures was more common than the other one. A causal diagram is a type of vi sual spatial display (such as a map or matrix) that clearly signifies a cause effect relationship). There were significant differences in the type of causal links generated by readers. The participants made more antecedent consequent associations than con chronological criteria in the organization of causal chains, independent of the causal order in the text.

PAGE 43

43 Background Knowledge and Scientific Text A difficulty with scientific text is that even if students effectively comprehend the surface features of the text, they often lack the necessary activation of prior knowledge needed for comprehension. M oravcisk and Kintsch (1993) demonstrated that, without relevant domain knowledge, college students can retain text plausibly well, but still be unsuccessful at understanding the text at the situation model level. Most science texts do not provide enough pr ompts and cues for the generation of accurate mental representations of scientific concepts. The lack of prompts and cues is more detrimental for readers with little or inaccurate scientific knowledge. On the other hand, there are advantages for knowledgea ble readers to receiving texts with coherence gaps that must be filled in with inferences. McNamara, Kintsch, Songer, and Kintsch (1996) demonstrated that high knowledge readers learn substantially more from low coherence than high coherence texts. In thei r study, eighth grade readers either read a low or high coherence biology text twice, or read both the low or high coherence biology texts in one order and then in the opposite order. Bridging inference questions, problem solving questions, and a keyword abstract understanding of the text. Eighth graders who knew significantly less about the text domain benefited from a high coherence text on all comprehension measurements. In contrast, eighth graders with h igh levels of domain knowledge benefited significantly from the low coherence text. Both comprehension skill and domain knowledge can have differential influences on learning from text, depending on the contents of the text, as observed by Voss and Silfies (1996), who expanded history texts by making the causal relationships more explicit. When the text was expanded, learning was related to reading comprehension

PAGE 44

44 skill and not to prior knowledge, whereas learning from the unexpanded text, that did not contai n explicit causes, was a function of prior knowledge and not of reading comprehension skill. Obtaining information or meaning from text is a constructive process consisting of an interaction between the explicit contents of the text and the uniqueness of t he individual. Nevertheless, texts with coherence gaps place most readers at a disadvantage because of the limitations in their knowledge and processing strategies. Based on the results of the two studies above (McNamara et al., 1996; Voss & Silfies, 1996) it appears that before causal structure repairs can be effective, other reader characteristics, such as background knowledge, must be taken into account. When learners possess prior knowledge that is consistent with target information, learning of new inf ormation is more easily facilitated. The more the reader elaborates Inaccurate pr ior knowledge makes reading expository/scientific text and the ultimate goal of comprehension quite challenging, but with this kind of text, readers with inaccurate knowledge are the default case rather than the exception (Kendeou, Rapp, & van den Broek, 2 004; Perkins, & Simmons, 1988). Anderson (1977) and Spiro (1980) observed that existing background knowledge was not a single construct, but multidimensional. Therefore, it matters not only that individuals possess relevant prior knowledge, but also what k nowledge they possess. Misconceptions generate comprehension problems with all types of text, but they are knowledge impedes the development of accurate situation models of sc ientific

PAGE 45

45 principles. The situation models constructed on naive rather than scientifically founded beliefs form incorrect knowledge structures (Nussbaum & Novak, 1976; Sneider & Poulos, 1983; Vosniadou & Brewer, 1992, 1994). For example, Lipson (1982) found that students who did not have related background knowledge before reading a paragraph scored higher on comprehension measures than the students with related yet incorrect information about the paragraph. Given that students often lack accurate background knowledge in the sciences, what kinds of supports or activities help them to overcome their deficits? One possibility might be having (or producing) a visual representation of the complex scientific material. Part III: Visual Spatial Displays Visual spati examples of visual spatial displays include matrices, maps, and knowledge maps (Vekiri, 2002). Matrice s make use of columns and rows to display within column and across column comparisons across topics (Bera & Robinson, 2004; Robinson & Kiewra, 1995; Robinson, Robinson, & Katayama, 1999; Robinson & Schraw, 1994). Maps utilize a spatial arrangement of symbo ls to represent the features or locations of a particular territory (e.g., a map of a county or a weather map) (Schnotz & Bannert, 2003; Verdi & Kulhavy, 2002). Knowledge maps, also known as concept maps, show the structure of knowledge in propositional st atements that illustrate the relationships among the concepts in a map (Novak, 1981). A knowledge map illustrates the relationships with the use of nodes (terms or concepts) and linking lines with propositions (linking phrases) that describe the relationsh Hall, 2002; Wiegmann, Dansereau, McCagg, Rewey, & Pitre, 1992). The knowledge

PAGE 46

46 map arrows are labeled to represent a multitude of plausible types of relationships between nodes. Thus, a knowledge map a rrow can represent elaborative (e.g., Madame Curie is an example of a scientist), dynamic/causal (e.g., studying effectively leads to good grades), or structural (e.g., the heart is part of the circulatory system) K n owledge maps have been used extensively in science education t o promote meaningful learning (Heinze Fry & Novak, 1990; Jegede, Alaiyemola, & Okebukola, 1990; Okebukola, 1990; Horton, McConney, Gallo, Woods, Senn, & Hamelin, 1993; Kinchin, 2000; Kinchin, Ha y, & Adams, 2000; Martin, Mintzes, & Clavijo, 2000; Odom & Kelly, 2001; Sungur, Tekkaya, & Geban, 2001). A knowledge map is based on the principle that meaningful learning occurs when learners are able to construct their knowledge hierarchically and explor e the possible linkages between concepts (Novak & Gowin, 1984). When students generate knowledge maps, they present the hierarchical structure of i dea s with an emphasis on the relations between concepts. Research has shown that visual displays facilitate l earning of the kinds of relationships the displays are intended to communicate (Vekiri, 2002). For example, matrices tend to facilitate relational learning, while maps are likely to facilitate recall for facts (Vekiri, 2002). Researchers Robinson and Kiewr a (1995) compared the effects of studying matrices with text and outlines with text. There were no differences for factual learning as measured by a multiple choice test. However, as measured by an essay, the matrices, utilized as a study tool along with t ext, better facilitated the learning of relationships among concepts and the ability to communicate those relationships in an organized way than did the outlines with text.

PAGE 47

47 Partial vs. Complete Concept Maps Identifying causal relationships in science text s can be cognitively taxing because readers may have difficulty directing their attention to essential elements of text when constructing inferences. What are readers doing while they are reading science texts, as opposed to other texts, that reduces their cognitive resources? Readers need to connect information from adjacent text segments, as well as distant text segments with their prior knowledge (Graesser & Bertus, 1998). In addition readers often lack the necessary activation of prior scientific know ledge to connect information needed for comprehension. These challenges may hinder readers from constructing causal inferences and seeing the big causal picture in a text (McCrudden et al., 2007). Chang, Sung, and Chen (2001) investigated surface processin g and cognitive overload of 126 fifth graders in relationship to scientific text learning. The participants worked individually with a computer to read texts and construct concept maps. Chang, Sung, and Chen (2001) utilized 3 concept mapping approaches map correction, scaffold fading, and map generation and summarization abilities. The experimental groups received the same seven pieces of scientific writing, but each received a different map cons truction exercise. The map correction group was given an expert generated concept map that was altered to contain partly incorrect concepts and semantic links. The scaffold fading group progressed through five stages: (a) read an expert concept map, (b) fi lled in the blanks of the expert concept map, (c) completed a partial expert concept map, (d) constructed the concept map using the provided concepts and relation links, and (e) constructed the concept map on their own. The third group was asked to generat e a complete concept

PAGE 48

48 map. To conduct the text presentation and the concept mapping exercise the Concept Mapping Learning System was used (Chang, Chen, & Sung, 2001). The system provides diverse kinds of feedback according to different concept mapping pro cedures. The map correction group received the percentage of their accurate corrections as feedback. The map generation group received the percentage G score) derived fro Relationships: one point for each valid proposition, Hierarchy: five points for level of hierarchy, Cross Links: ten points for each valid cross link and General to Specific Example: one point for each valid example. The scaffold fading and generation group received the percentage of correctly filled blanks. The control group received instruction on how to rea d the texts presented in Microsoft Word. After a 1 week delay, t he participants were asked to read a text for 15 minutes and then summarize the contents. The students were allowed to refer to the text during their summarizing. In the text summarization tas k, readers were required to grasp the gist, delete unimportant messages, and compile and reorganize the main Idea. Participants then took a text comprehension test consisting of multiple choice items. Results indicated that the map correction method enhanc ed text comprehension and summarization abilities and that the scaffold fading method facilitated only the summarization ability of the map generation group. The text generation abilities of students using the map generation method were not significantly d ifferent from those of students who used no concept mapping strategy.

PAGE 49

49 Knowledge Maps and Problem Solving Several studies have been carried out to examine significant factors in problem solving performance and relevant approaches that support successful pro blem solving (Jonassen, 2000; Smith, 1991). Some researchers propose that the differences in success among problem solvers may be attributed more to the meaningful representation of knowledge than to the amount of their prior knowledge (Silver & Marshall, 1990). A knowledge map that represents the relationship between nodes and links (Jonassen, Beissner & Yacci, 1993) has been shown to be an effective method for solving problems. It has successfully enabled learners to understand the problem (Zhang, 1997), remember important information, and be aware of new relations among the concepts embedded in the problem (Hayes, 1989). Novak, Gowin and Johansen (1983) found that 7th and 8th grade science students who were engaged in a concept mapping strategy demonstrat ed better problem solving performance than those who were directly taught by another instructor. Okebukola (1990) examined students performance at solving genetics problems and found that the 63 students who generated concept maps significantly outperform ed the other 63 students who studied the material by themselves. Youngmin and Nelson (2005) investigated the effects of two types of maps (generative vs. completed) and the amount of prior knowledge (high vs. low) on well structured and ill structured prob lem solving performance. They examined forty four undergraduates from an introductory instructional technology course designed to introduce preservice teachers to basic concepts of instructional planning. A pretest was administered to measure the extent of designing instruction in the classroom. Participants were divided into four groups:

PAGE 50

50 completed map high knowledge; completed map low knowledge; generative map high knowledge; and generative map low knowledge. Problem solving performance was measured by two post tests. In the well structured problem test, the participants were required to solve the problems by applying principles and rules in relation to planning and designing the instruction. An example of a we ll structured problem in the study is structured problem solving test consisted of 20 multiple choice items, and the ill structured problems were compose d of four essay items. In the ill structured test, participants were required to solve problems by synthesizing and reasoning about the rules and principles. There were no absolute answers for the problems and it was more important to present their meaning ful thought reasonably and logically. An example of an ill the wind among educators. Over the last decade, research on how people learn has strengthened the case for teaching for understanding in a s tudent centered way. Constructivist approaches are particularly appropriate for that change. Describe a In the ill structured test, participants were required to solve problems by synthesizing and reasoning. Participants with high domain knowledge using generative concept maps significantly outperformed high domain knowledge participants who used the completed concept map on well structured problems, but not on ill str uctured problems. There was no significant difference between the low knowledge generative group and the low knowledge completed group on well or ill structured problem solving performance.

PAGE 51

51 What remains unanswered is why participants with high domain know ledge did not significantly outperform their counterparts on ill structured problem solving, even with the aid of a generated or a completed concept map. One explanation is that ill structured problem solving is more closely related to the structure of bac kward inference, and this is often the structure of scientific problems in the real world. Problem solving requires complex cognitive processes such as justification, evaluation, and hypothesis testing. When confronted with new or conflicting information, students not only activate their pre existing knowledge about the topic, but also their beliefs about knowledge itself (Mason & Gava, 2007). Lee (2010) examined the interactions between problem solving and conceptual change in a fifth grade elementary sci ence class where 50 students built a system dynamic model as a form of problem representation in Model It. Participants were presented with a real world problem pertaining to the water cycle where they had to identify and determine the parameters of the pr presented with 14 multiple choice items and asked to provide a short justification for their answers. Researchers then conducted interviews with the students. During the interviews, students were given paper and pencil to draw a water cycle concept map. the total number of nodes, total number of nodes associated with the central concepts (evaporation and condensation), and propositions. Students used a variety of strategies to study science in class. These included: asking someone (i.e., siblings, friends, parents, and school teacher), constructing mind maps, searching the internet for new information or to validate their understanding, and self questioning. Results indicated

PAGE 52

52 that the interplay of three emerging intervening conditions: epistemological beliefs, strate gies and the learning outcomes. Students with sophisticated epistemological beliefs performed better in the exploratory simulation problem representations. Both structural knowledge and domain knowledge were identified as additional intervening conditions. When students understood the organizations and relationships of the variables in the system, they were more likely to engage in self questioning, but limited domain knowledge deterred them from using the self questioning strategy Technology vs. Paper and Pencil Generated Concept Maps Results of some studies show that computer based concept mapping improves concept maps surpass page size, are easy to generate, and are fas ter to revise than their paper and pencil counterparts (Anderson Inman & Ditson, 1999; Royer & Royer, 2004). Anderson Inman and Zeitz (1993) found that computer based concept mapping using InspirationTM encouraged students to adjust their computerized conc ept maps more when compared to their maps drawn with paper and pencil. A number of researchers have found that when students are properly trained to use the computer, they can effortlessly and effectively construct, alter, or preserve their concept maps du ring conceptual learning (Anderson Inman & Zeitz, 1993; Jonassen, Reeves, Hong, Harvey, & Karen, 1997; Royer & Royer, 2004). Meaningful learning entails the integration of new concepts and propositions into existing cognitive structures (Novak, 2002). Mean ingful learners construct concept maps that contain more concepts, relationships and branching than rote learners

PAGE 53

53 investigated the effect of hypermedia hypertext and paper pen cil type concept maps with 30 undergraduate chemistry students. Concept maps prepared using hypertext hypermedia integrates all sorts of media accessible through the Internet (i.e. graphics, audio, moving images). Atom bonding was explained to the students in group 1 using concept maps prepared in a computer environment using hypertexts. The subject was explained to the students in group 2 using concept maps prepared with the paper pencil method. Afterwards the students were asked to generate their own conc ept maps regarding these concepts. The evaluation of concept maps followed the scoring method developed by Novak and Gowin (1984). Results indicated a significant difference in favor of the concept maps prepared using the hypermedia hypertext concept mappi ng technique. Using this technique, students produced a greater number of correctly established relationships between concepts compared to the students using paper pencil type concept maps. The difference between the pre test and post test results was not significant for the students who generated the paper pencil type concept maps ; however, a significant difference was observed between the pre test and posts test using the hypermedia hypertext concept mapping technique. Erdema et al. (2009) propose that th is is due to the fact that when the concept maps are prepared using the paper pencil technique, if students do not have adequate knowledge to indicate the link between any two concepts, they must read the entire chapter in order to find the required inform ation. In contrast, with the hypermedia hypertext technique, the picture, audio and visual information, and examples on the links placed on the concepts in the map, helped students achieve lasting and meaningful learning.

PAGE 54

54 In another use of computer based c oncept mapping, Kao, Lin, and Sun (2008) examined thirty two information management undergraduates utilizing an integrated concept map system (ICMSys) that accommodates group products while still preserving individual uniqueness. Each learner was required to construct his or her own concept map of computer systems and focus on individual conceptual awareness as stimulated poin t for inspection, thus allowing them to make comparisons among concept maps. Participants repeated this process as to view various combinations of ICMaps and discover what was lacking or incorrect in their own concept maps. The initial ICMaps and the revised versions were assessed to determine if differences between student and expert assessments decreased or increased during the study period; results indicated a statis tically significant improvement in examples and relationships but not in hierarchies or cross links. Collaborative vs. Individually Generated Concept Maps Concept maps offer learners a way to communicate and share representations of their cognitive struct ures with other learners (Novak & Gowin, 1984). Some researchers scientific knowledge construction more effectively than individual knowledge construction (Fischer, Bru hn, Grasel, & Mandl, 2002). Brown (2003) found that those students who collaboratively constructed concept maps outperformed students who individually constructed concept maps on a high school biology test, and, test scores were no higher for students who individually constructed concept maps than for students who did not construct concept maps at all. In contrast, Kwon and Cifuentes

PAGE 55

55 (2009) found that individually constructing concept maps on computers during study time positively influenced science concept learning above and beyond independent use of study time, but that collaboratively constructing concept maps on computers did not. Kwon and Cifuentes (2009) investigated the effects of individually constructed and collaboratively constructed computer based (InspirationTM software) concept mapping with one hundred and sixty one 7 th grade middle school students. They utilized achievement across groups. Student pairs consisted of one h igh achieving student and one low achieving student. Students were assigned to three groups: a self selected concept mapping group. Five short essays on weathering, the n ature of soil, soil erosion, erosion by gravity, and erosion by winds, were distributed to the students. Each group was told to study the essays in preparation for a test. Students in the control group individually followed their own learning strategies (i .e. highlighting, memorization, or taking notes) to prepare for their test. Students in the individual group created concept maps and studied independently using computers. Students in the collaborative pairs group created concept maps in pairs and studied together using computers. Concept maps were scored according to the four scoring components created by Novak and Gowin (1984). Students in each group were given the same 10 item multiple choice test (paper and pencil form) to measure their understanding o f the science concepts in the text that they studied. Results indicated the group that collaboratively constructed concepts maps did not significantly outscore the group that individually constructed

PAGE 56

56 concept maps in its performance, but the control group y ielded a significantly lower mean score than either of the two experimental groups. Concept Mapping and Differences in Reading Skill and Text Difficulty Concept mapping may help learners to organize individual knowledge meaningfully. The organization of co comprehension and memory of causal relationships (Tarmizi & Sweller 1988; Baddeley 1992; Sweller 2006). Conradty and Bogner (2010) examined 283 sixth maps as a means to promote meaningful learning of reading comprehension skills. the concept maps were predefined in order to reduce cognitive load; however, linking phrases were not defined. In addition to the two concept map assignments, each participant was given a 17 item multiple choice assessment to measure the individual learning achievement of all participants. Students generate d more complex concept maps in the context of the easier subject matter than with the difficult content, indicating that concept mapping was more sufficient with simple subject matters than with complex, difficult and multi faceted ones. Minor learning suc cess, with content xplore whether the presentation of an expert completed concept map for difficult versus easy subject matter would have an influence on performance.

PAGE 57

57 Causal Diagrams A causal diagram is a special type of visual spatial display that organizes the causal rela tionships of a process or sequence of events with arrows indicating direction of causality. While knowledge map arrows are labeled to embellish an array of potential types of relationships between nodes, causal diagram arrows depict causal relationships on ly. In a causal relationship, event A precedes and causes event B. For example, reduced stress on bones precedes and causes decreased production of bone building cells (McCrudden et al., 2007. Causal diagrams are distinctively designed to understanding of causal relationships within and across paragraphs. Causal diagrams can be used to evaluate aspects of text representations that cannot be captured entirely in other ways (McCrudden et al., 2007; Oestermeier & Hesse, 2000). Causal diagrams seem to be particularly suitable to assist learning when texts describe implicit or complex causal relationships. One possible reason that studying a causal diagram could facilitate understanding of causal relationships is that these diagrams improve enco ding by helping readers determine what is most pertinent to understanding causal sequences (McCrudden et al., 2007). A causal diagram adds value to a text that does not explicitly describe all causal sequences. Causal diagrams can facilitate at least three types of inferences that are implicit in a text. First, causal diagrams illustrate direct effects. A direct effect transpires when a change in one variable causes a direct change on a second variable. For example, a lack of gravity has a direct influence on the amount of stress on bones. As gravity decreases, the amount of stress on load bearing bones decreases. This relationship holds true regardless of other variables. Second, causal diagrams depict indirect effects. An indirect effect occurs when a chan ge in one variable causes a change in a second

PAGE 58

58 variable, which in turn, causes a change in a third variable. For example, lack of gravity does not have a direct influence on bone degeneration, but it does influence the amount of stress on bones, which in t urn influences bone degeneration. Third, a causal diagram can display multiple causal sequences that occur simultaneously. For example, it is possible to infer that each causal sequence begins with one main cause that leads to multiple sequences of causal events. The reader must identify each step and infer the relationships between and among the steps (McCrudden et al., 2007, p. 371). The causal diagram explicitly depicts the relationships that are implicit in the text. Thus, comprehension of a complex tex t should be enhanced when sophisticated relationships are depicted with a causal diagram. In essence, causal diagrams should make a very difficult task easier. A different type of study aid, the outline, may also help readers comprehend and remember causal relationships in science texts. Outlines as Study Aids An outline is an organized listing of the essential features or main aspects of concepts. Outlines are useful as study aids because they include only the more important text information and convey hie rarchical concept relations. The usefulness of outlines has been demonstrated by many researchers (e.g., Darch & Gersten, 1986; Glynn et al., 1985). It is common for textbooks to include chapter outlines, for instructors to write lecture outlines on the ch alkboard, and for students to take outline notes. Although outlines are a popular method of organizing expository information, there are likely problems that may result from studying them. Outlines, like text, have a linear format that discourages depictin g relations among concepts. While outlines convey hierarchical, within concept relations, they obscure important coordinate, among concepts relations. Thus, the information that outlines contain is not the problem; it is

PAGE 59

59 their linear format that discourage s students from integrating information across the concepts (Waller & Whalley, 1987). To enable students to learn important among concepts relations that are only implicit in text and outlines, researchers developed an adjunct display that uses a spatial format to represent key text ideas and graphic organizers. Winn and Holliday (1982) claim that the use of spatial displays presents concepts in a way that enables students to use the least amount of mental effort to understand the relations among those con cepts a phenomenon that has been referred to as visual argument. Visual argument transmits relations among ideas through a spatial arrangement of words, rather than ordinary written language. Thus, students are relieved of the cognitive burden of identifyi ng them from the text. According to this view, an outline uses visual argument to communicate hierarchical concept relations better than text but not coordinate concept relations. Thus, a graphic organizer uses visual argument to communicate both hierarchi cal relations and to communicate coordinate concept relations (interconcepts) better than outlines and text. One of the difficulties in learning from an expository text is understanding the mer, & Wolff, 1991; Winn, 1988). To understand the big picture, students must not only recognize important concepts but also comprehend the relations among those concepts. Robinson and Schraw (1994) investigated whether a matrix (a type of graphic organize r) communicates interconcept relations more effectively than an outline or text. In three experiments, college students judged the accuracy of interconcept relations after they read a text and studied a matrix, an outline, or the text again. Participants w ere

PAGE 60

60 presented with 40 pattern items and 32 comparison items which they identified as true or false statements. In Experiments 1 and 2, participants who studied only a matrix performed better than those who studied only text. In contrast in experiment 3, wh en testing was delayed, there was no advantage for studying a matrix. Results indicated that after a delay, performance decreased only for students studying a matrix and not for those who studied an outline or text. Perhaps these results are due to the com putational efficiency of the matrix. Students are not required to put forth sufficient effort in learning interconcept relations, thus resulting in low stability of memory traces and poorer performance when testing is delayed. If displays are constructed s o that they make relations in text explicit, then students may not be forced to disentangle implicit relations, which may be necessary to remember the information for a significant amount of time. Robinson and Schraw (1994) suggest that the type of display a student is offered may have a significant influence on the activities used to encode the information. Robinson and Kiewra (1995) investigated the effects of graphic organizers and ied a matrix, an outline, or the text again. Several tests were used to assess the different types of text information: A multiple choice test measured knowledge of text facts contained either in the adjunct displays (represented facts) or only in the text (nonrepresented facts). A cued recall test measured knowledge of hierarchical relations among concepts or what awareness of text structure is important in guiding the process of re ading comprehension; without it, students may treat the text as a list of facts. A matching test

PAGE 61

61 measured application of concepts in new situations. Lastly, an essay test measured both knowledge of coordinate relations among concepts and the ability to exp ress those relations in an organized manner (contrasting premises). Some researchers (Benton, Kiewra, Whitfill, & Dennison, 1993; Waller & Whalley, 1987) have indeed found evidence suggesting that studying a graphic organizer enables students to write more integrated essays. Results indicated students studying either graphic organizers or outlines learned more represented facts than those studying text alone. In contrast, students studying only text learned more nonrepresented facts than those studying grap hic organizers. In addition, students studying graphic organizers learned more hierarchical and coordinate relations, wrote more contrasting premises, and were more successful in applying new knowledge than those studying either outlines or text alone. No differences between the outline and text only groups on the aforementioned measures were observed. Rationale for the Present Study Different types of texts place different cognitive demands on readers. Narrative text possess a causal temporal structure tha t is often more decipherable to readers than the logical structure of expository texts (van den Broek et al., 2002). During the comprehension of narrative texts, readers establish certain levels of coherence (mostly causal and referential), but comprehensi on of expository texts involves different standards of coherence (van den Broek, 2010). Expository text is often unfamiliar, forcing the reader to employ strategic, time consuming, analytical reasoning to generate causal consequence inferences (Graesser & Bertus, 1998). Scientific genres often assume readers are capable of generating the necessary inferences for text coherence.

PAGE 62

62 However, identifying causal relationships in science texts can be cognitively taxing, limiting the construction of inferences (Grae sser & Bertus, 1998). o identify a causal relation without difficulty if the cause precedes its consequent in the text (Linderholm et al., 2000). However, scientific explanations in texts typically reverse the order of causality of the phenomena they are explaining, i.e., start by describing effects (or Consequences) and work their way backwards to the causes (or Antecedents). This is the reverse of what typically happens in narratives with causal chains of events, where readers can follow events from cause to effect. Causal dia grams seem to be particularly suitable to assist learning when texts describe implicit or complex causal relationships. of causal relationships within and across paragraphs. By c ommunicating the implicit structure of the text, and in some cases implicit causal relations as well, the diagram clarifies causal relationships and helps the reader to extract the causal structure out of the text. Thus, studying a causal diagram may facil itate understanding of causal relationships because causal diagrams improve encoding by helping readers determine what is most pertinent to understanding causal sequences (McCrudden et al., 2007). A different type of study aid, the outline, may also help r eaders comprehend and remember causal relationships in science texts. Outlines are useful as study aids because they include only the more important text information and convey hierarchical concept relations. While outlines convey hierarchical, within conc ept relations, they

PAGE 63

63 obscure important coordinate, among concepts relations. Thus, the information that outlines contain is not the problem; it is their linear format that discourages students from integrating information across the concepts (Waller & Whall ey, 1987). Outlines may help readers comprehend and remember causal relationships in science texts, but only because they direct their attention to essential elements of text or main points of a text.

PAGE 64

64 CHAPTER 3 METHODOLOGY Scientific explanations in texts typically reverse the order of causality of the phenomena they are explaining, i.e., they start by describing effects (hereinafter labeled Consequences) and work their way backwards to the causes (hereinafter Penalba, 2002; Newton, 1995). This is the o pposite of the order typically found in narratives with causal chains of events, where readers can follow events from cause to effect (van den Broek et al., 2002). remember causal sequences and/or understand causal relationships when they are attempting to comprehend scientific The main purpose of this study was to explore the e ffects of causal diagrams that do not sequences in scientific texts. There is ample research to show that many types of visual displays facilitate learning (Ainsworth & Loizou, 2 003; Bera & Robinson, 2004; Carney & Levin, 2002; DiCecco & Gleason, 2002; Jonassen, 2003; Lowe, 2003; Mayer & Moreno, 2002; & Hegarty, 1999; Vekiri, 2002; Verdi & Kulhavy 2002). In contrast, there is very little research on causal diagrams. The chronological order in which causal structures are expressed is potentially a crucial factor in comprehending causal explanations in scientific texts (Elshout Mohr & van Daalen'Kap teijns, 2002). This study explored the Consequence is offered first, followed by the Antecedents) by providing them with a

PAGE 65

65 forward directional causal diagram in whi ch the Antecedents, or causes, are given first, and lead in chronological order to the Consequence, or effect. This study also explored texts may be affected simply by directing their attention to essential elements of text via an outline provided as a study aid. Outlines may also help readers comprehend and remember causal relationships in science texts, because they present essential information or main points of a te xt. Unlike a causal diagram, however, an outline of a text must present the same chronological structure as the text does. Research Questions Based on the purposes of my study, I proposed the following research questions and hypotheses: Research Question 1 How is memory for main idea in scientific text affected by the following three different reading conditions? 1. reading text only in Consequent Antecedent form 2. reading text in Consequent Antecedent form accompanied by an outline in Consequent Antecedent form 3. reading text in Consequent Antecedent form accompanied by a causal diagram in Antecedent Consequent form My hypothesis was that performance on memory for main idea would be stronger with both causal diagrams and outlines tha n with text alone, but would not differ between the causal diagram and outline reading conditions, because these segments are mentioned explicitly in the text and can be understood in isolation.

PAGE 66

66 Research Question 2 How is memory for the causal sequence in scientific text affected by the following three different reading conditions? 1. reading text only in Consequent Antecedent form 2. reading text in Consequent Antecedent form accompanied by an outline in Consequent Antecedent form 3. reading text in Con sequent Antecedent form accompanied by a causal diagram in Antecedent Consequent form My hypothesis was that performance on memory for the causal sequence would be highest with the causal diagram, lower with the outline, and lowest with the text alone, bec chronological criteria in organizing their representations of causal chains. Research Question 3 How is understanding of the causal sequence in scientific text affected by the following th ree reading conditions? 1. reading text only in Consequent Antecedent form 2. reading text in Consequent Antecedent form accompanied by an outline in Consequent Antecedent form 3. reading text in Consequent Antecedent form accompanied by a causal diagram in Antecedent Consequent form My hypothesis was that performance on understanding the causal sequence would be highest with the causal diagram, lower with the outline, and lowest with the text alone, again because the causal diagram would reflect r chronological criteria in organizing their representations of causal chains.

PAGE 67

67 Participants A total of 99 fifth grade students, 60 females and 39 males, participated in the study. Eight students identified themselves as Asian, 42 as African American, 28 as Caucasian, 19 as Hispanic, and 2 as Mixed Ethnicity. Students attended 4 different Urban public schools in a Northeast Florida county. Regarding socioeconomic status of the students, 56 are automatically approved via their famil y Human Resource Service file for free lunches, an additional 31 applied for and were approved for free lunch, 9 receive reduced lunch and 3 are not eligible for free or reduced lunch. Although the students were taught by 10 different teachers, the teacher s covered the same state approved science curriculum content on similar schedules. Readers employ a range of strategies while processing text information and these strategies have di with the information they read (Chi, de Leeuw, Chiu, & LaVancher, 1994; Cot & Goldman, 1999; Cot et al, 1998). These skills include phonological awareness, decoding (e.g., Gough & Tunmer 1986; Perfetti, 1988; White, 2005), fluency (e.g., LaBerge & Samuels, 1974), and vocabulary knowledge (e.g., Beck, Perfetti, & McKeown, 1982; Stanovich, 1986). These skills are essential for successful reading (National Reading Panel, 2000; Oakhill, Cain & Bryant, 2003; Paris & Paris, 2003; Snow, 2002; van den Broek, Kendeou, et al., 2005). However, not all readers possess or utilize reading skills in the same way resulting in variations of comprehension levels. ividual reading skills influence on their comprehension of the text (scores for memory of main idea, memory of causal sequence, and understanding of causal sequence), a reading score (the FCAT Reading

PAGE 68

68 Equivalent Scaled Score, or RESS) was used as a covaria te. By using the FCAT RESS individual reading skills were taken into account, resulting in a more accurate measure of treatment differences. The FCAT 2.0 RESS measures s tudent reading achievement on the Next Generation Sunshine State Standards (NGSSS), compared across all grade levels. The NGSSS is the core content of the curricula taught in Florida. The NGSSS specifies the core content knowledge and skills that K 12 publ ic school students are expected to benchmarks identify what a student should know and be able to do at each grade level for each subject area. For grades 3 10, FCAT 2.0 rea ding consists of multiple choice all grade levels tested, FCAT 2.0 reading assesses what students know and are able to do (i.e. Vocabulary, Reading Application, Informati onal Text and Research Process) (FCAT 2.0 and Florida End of Course Assessments Achievement Levels, 2013). The difficulty of the concepts assessed on FCAT 2.0 reading progresses steadily from grade to grade, as does the complexity of the text presented to the student at each grade level, as described in the state curriculum framework. Developmental Scale Scores (DSS) range from 1 to 5, with level 5 being the highest and level 1 being the lowest level. For example, a level 3 indicates when students are pres ented with grade analyze information from a variety of text features (e.g., titles, subtitles, headings, subheadings, italicized text, sections, tables, charts, graphs, diagrams, illustrations,

PAGE 69

69 captions, maps, text boxes) to determine meaning and locate, interpret, and organize information for a variety of purposes (FCAT 2.0 Florida Comprehension Assessment Test, p. 8, 2013). To be considered on grade level, students must achieve a DSS level 3 or higher, a Reading Equivalent Scale Score (RESS) of 208 220 (FCAT 2.0 and Florida End of Course Assessments Achievement Levels, 2013). The mean FCAT om 161 DSS levels ranging from 1 5. It should be noted that 63.6% of the students fell at or below the 220 FCAT RESS, and were thus reading in the average to below average range. FCAT scores were missing for four students who had transferred from schools not required to take the FCAT. The mean FCAT RESS (M = 214) and the FCAT DSS reading achievement level mean (M = 3) were used as replacement scores for these students. Materials T he experimental text (Appendix A) is a 573 change) on the Statue of Liberty at an introductory level. The text begins with the effec t, or consequence (the Statue of Liberty is green) and moves backward to the causes, or antecedents. The text was adapted from several educational companion websites for the state wide curriculum: National Science Teachers Association (1999 2009), http://www.scilinks.org/ What is Oxidation? (Albers, 1999 2011) Demand Media, Inc. http://www.ehow.com/how does_4597366_oxidat ion affect heat transfer copper.html and What Made the Statue of Liberty Turn Green? ( http://www.ehow.com/how does_5439713_made statue liberty turn green.html (Pot 1999 2009). The causal

PAGE 70

70 diagram for the text (Appendix E) sequence is written in Antecedent Consequence order. The outline for the text (Appendix F) is written in Consequent Antecedent order, just as the text is. Consultation with residential faculty mem bers who work with the teachers and principals from the 4 county schools led to my choice of the concept of chemical change for the experimental text. Several Sunshine State Standard benchmarks of the 5th grade state wide science curriculum framework inclu de chemical change; therefore it is a concept that is assessed on the FCAT (Duval County Public Schools, 2009 2010). In addition, the specific example of oxidation (the statue of liberty turning green) was not an example commonly used in classroom textbook s, thereby making it a novel example of oxidation. Lastly, the science teachers agreed to have their classes participate in the research study, viewing the novel example as an enrichment activity for their future instruction about chemical change. Construc tion of Test Items Memory for main idea, memory for causal sequences, and understanding of causal sequences were assessed with a short answer test consisting of multiple choice and short essay questions. Multiple choice items can measure knowledge outcomes such as: knowledge of facts, terminology, principles, methods and procedures. Short essay questions are useful for measuring higher order thinking such as comprehension, application outcomes, interpretation of cause and effect relationships, and justifyin g methods and procedures (Miller, Linn, & Gronlund, 2009). The test consisted of 9 multiple choice questions to assess memory of main idea, 6 multiple choice items and 1 short essay question to assess memory of causal sequence, and 4 short essay questions to asses understanding of causal sequences

PAGE 71

71 and effects. Ite ms were arranged so they did not provide clues to other items on the same page or on succeeding pages. Examples of each type of test item appear below; Appendix B for the complete test. Memory for Main Idea Item 2. Which of the following materials are need ed in order for copper to oxidize? a. Rust, patina, oxygen b. Carbon dioxide, water, rust c. Water, oxygen, patina d. Oxygen, water, carbon dioxide Memory for Causal Sequence Item 17. Explain how the Statue of Liberty has turned green. List each of the steps beginning with the first step. Explain in as much detail as possible. Understanding of Causal Sequence Item 18. Does the process of oxidization continue on the surface of the Statue of Liberty after the patina has formed? Yes or No, explain your answer in as much detail as possible. Validity of Test Items Five experts (three science education professors and two cognitive psychology professors) offered feedback regarding the extent to which test items were consistent with the curriculum objective s, and the extent to which the test items actually measured what they purported to measure. Each expert was given a validation chart (Appendix D) and asked to evaluate the validity of each item. Examples of feedback for items deemed valid appear below: Ite

PAGE 72

72 concepts from the general oxidation paragraphs and apply them to the spe Items were removed if deemed irrelevant or invalid to the study; examples appear below: Item 6: Alexandre Eiffel designed the interior of the Statue of Liberty. Later he designed a. the Twin Towers b. the Empire State building c. the Eiffel Tower, d. none of the above Some items were appear green, but most of the pennies you see are brown, even though they are also wording of this item make knowledge of a steeple prerequisite to knowing the causal green, but most of the pennies you see are brown, even though they are also coated again reviewed by three of the teachers within the school district whose schools participated in the study. The teachers deemed text and all items as valid and consistent with the curriculum objectives.

PAGE 73

73 Procedures An application to perform research in the schools was submitted to a Northeast Florida county school district. Once the application was approved by the school district, principals were asked to participate in the research. With the permission of each principal of the participating schools, the aut hor addressed 5 th grade students in their Science classes regarding the nature and importance of the study. All students received informed consent forms, which were signed by a parent/guardian and returned to their respective Science teachers in order for them to participate in the study. Support with administering the experiment was provided from four undergraduate students who attended a local university (hereinafter referred to as research assistants). The experiment was conducted in a 5 th grade classroo m setting in each respective school. A research assistant with the project read students a child assent script that informed students of their rights as participants. Participants were then assigned randomly to one of three conditions: Group A) Consequent Antecedent ordered text only Group B) Consequent Antecedent ordered text with an accompanying outline, also in Consequent Antecedent order Group C) Consequent Antecedent ordered text with an accompanying causal diagram that presented the sequence in Antecedent Consequent order Group A stayed in the classroom and Groups B and C each moved to a different location (i.e. classroom, cafeteria, or study hall). This kept the students in the three different conditions apart from each other during the adminis tration of the task. A fourth group was assembled as needed for students who did not participate in the project, and an activity was provided by the teacher for those students.

PAGE 74

74 After random assignment to groups, participants were read an overview of tasks. All participants were informed that the text focused on what causes the green color of the Statue of Liberty and that they should read the text for understanding. They were also informed that they would be tested on their understanding of what they had re ad. Participants in the Text Only condition were given no further instructions. Participants in the Text and Outline condition were presented with the outline first and the text second in a single handout on 8.5 by 11 inch paper to reduce any potential eff ect for split attention (Chandler & Sweller, 1991, 1992; Sweller, Chandler, Tierney, & Cooper, 1990; Tarmizi & Sweller, 1988; Ward & Sweller, 1990). Students were given the following outline will help you understand information in the text about why the Statue of Liberty and Outline conditi on were prompted to examine the outline to ensure that they studied the outline while the text was available. Participants in the Text and Diagram condition were presented the causal diagram first and the text second in a single handout, again to reduce po tential split attention effects. the text about why the Statue of Liberty has turned green. The diagram flows from left to and diagram condition were prompted to examine the diagr am to ensure that they studied the diagram while the text was available.

PAGE 75

75 Next, participants read the passage and studied their handouts for 15 minutes. At the 10 minute mark, it was indicated that 10 minutes had elapsed. After 14 minutes, it was announced that there was one minute remaining. After 15 minutes, the students were asked to place the study materials in folders; the folders were collected prior to passing out the tests. The test items were also printed on an 8.5 by 11 inch paper for consistency. None of the participants had access to the text, causal diagram, or outline while being tested. After all participants completed the test, the researcher thanked them for participating in the study.

PAGE 76

76 CHAPTER 4 ANALYSIS OF DATA Three separate one way ANCOV As were conducted for performance on memory for main idea, performance on memory for the causal sequence, and performance on understanding the causal sequence under three conditions: text only in Consequent Antecedent form, text in Consequent Antecedent fo rm accompanied by an outline in Consequent Antecedent form and text in Consequent Antecedent form accompanied by a causal diagram in Antecedent Consequent form. individual reading skills on comprehension of the text, the FCAT RESS (Reading Equivalent Scaled Score) was used as the covariate. Research Question 1: Memory of Main Ideas How is memory for main ideas in scientific text affected by the following three different reading conditions? 1. reading text only in Consequent Antecedent form 2. reading text in Consequent Antecedent form accompanied by an outline in Consequent Antecedent form 3. reading text in Consequent Antecedent form accompanied by a causal diagram i n Antecedent Consequen t form My hypothesis was that performance on memory for main idea would be stronger with both causal diagrams and outlines than with text alone, but would not differ between the causal diagram and outline reading conditions. A one way Analysis of Covarianc e (ANCOVA) was performed using Text only, Text with Outline, and Text with Causal diagram as the independent variable, scores on memory for main idea questions as the dependent variable, and FCAT RESS as the covariate. Student responses to the memory for m ain idea items were scored with one

PAGE 77

77 point for each correct multiple choice answer. Items: 1, 2, 5, 6, 8, 9, 10, 11, 12 were aggregated to compute composite score 1 (Appendix C). Cronbach's alpha was used to measure the internal consistency of composite sco re 1, .432. The FCAT RESS as Covariate with Memory of Main Ideas The relationship between the FCAT RESS and composite score 1 met the assumption of a linear relationship between the covariate and the dependent variable scores. There was a positive relatio nship between the FCAT RESS and the dependent variable composite score 1, memory for main idea, with r = .267, n = 99, and p = .008. This relationship was significant at the 0.01 level (2 tailed). The assumption of homogeneity of regression slopes was met, as the interaction term was not statistically significant, F(2,93) = 1.502, p = .228. The assumption of normality was not met; standardized residuals for composite score 1, memory of main idea, were significant (p = .003). However, ANCOVA is fairly robus t to deviations from normality with sample sizes greater than 50; in this study, n = 99. The assumption of homogeneity of variance was met, as assessed by Levene's Test of Homogeneity of Variance (p = .530). There were no outliers in the data (no cases wi th standardized residuals greater than 3 standard deviations). Tests of Between Subjects Effects Table 4 1 illustrates the adjusted and unadjusted means and standard deviations for Composite score 1, memory of main idea, with FCAT RESS as a covariate. Aft er adjusting for FCAT RESS, no statistically significant differences were observed in Memory of Main Idea scores between the three conditions (Text Only, Text with Outline, Text with Causal Diagram), F (2,95) = 1.627, p = 202 2 = .033 ( t able 4 2)

PAGE 78

78 Research Question 2: Memory of Causal Sequence How is memory for the causal sequence in scientific text affected by the following three different reading conditions? 1. reading text only in Consequent Antecedent form 2. reading text in Consequent Antecedent form accompanied by an outline in Consequent Antecedent form 3. reading text in Consequent Antecedent form accompanied by a causal diagram in Antecedent Consequent form My hypothesis was that performance on memory for the causal sequ ence would be highest with the causal diagram, lower with the outline, and lowest with the text chronological criteria in organizing their representations of causal chains. A one way ANCOVA was performed using text only, text with outline, and text with a causal diagram as the independent variable, scores on memory of causal sequence questions as the dependent variable, and the FCAT RESS as the covariate. Student responses to the memory of the causal sequence multiple choice items (items 4, 7, 13, 14, 15, and 16) were scored with one point for each correct answer with a maximum of 6 points. In addition, the short essay item 17 required a listing of the steps in the cause and effect network for the main idea (oxidization) with a maximum of 9 points. Segments in every short essay answer protocol were evaluated to determine memory of the causal sequence was scored by tallying the number of steps they remembered in the process whereby the Statue of Liberty turns green. Steps were scored as present or absent and summed to create a memory of causal sequence score with a maximum of 15 points total. The experiment er and another rater, both blind to the

PAGE 79

79 experimental conditions, independently scored 20 percent of the responses for the memory of causal sequence item 17 using the scoring rubric (Appendix C). Cronbach's alpha was used to measure the agreement between tw o raters on the responses to item 17; .912. Multiple choice items 4, 7, 13, 14, 15, 16 and short answer item 17 were aggregated to compute composite score 2, memory of causal sequence (Appendix C). Cronbach's alpha was used to measure the internal consi stency of composite score 2 (consisting of items: 4, 7, 13, 14, 15, 16 and 17); .530. The FCAT RESS as Covariate with Memory of Causal Sequence The relationship between the FCAT RESS and composite score 2 met the assumption that there is a linear relationship between the covariate and the dependent variable scores. There was a positive relationship between the FCAT RESS and the dependent variable composite score 2, memory of causal sequence, r = .357, n = 99, p = 000. There was homogeneity of regression slopes as the interaction term was not statistically significant, F(2,93) = 2.501, p = .087. Standardized residuals for composite score 2, memory of causal sequence, p = .051 were normally distributed, as assessed b y Kolmogorov Smirnov Test (p > .05). There was homogeneity of variances, as assessed by Levene's Test of Homogeneity of Variance (p = .712). There were no outliers in the data (no cases with standardized residuals greater than 3 standard deviations). Tes ts of Between Subjects Effects Table 4 3 illustrates the adjusted and unadjusted means and standard deviations for Composite score 2, memory of causal sequence, with FCAT RESS as a covariate, in the three different reading conditions. After adjustment for FCAT RESS, there was a

PAGE 80

80 statistically significant difference in memory of causal sequence between the three conditions: (Text Only, Text with Outline, Text with Causal Diagram), F (2,95) = 15.690, p = 2 = .248 (Table 4 4). Pairwise Comparisons Text o nly vs. t ext with o utline The mean score on memory of causal sequence in the Text Only condition (M = 7.905a out of 15 possible points, SD = .413) was not significantly different (95% CI [ 1.179, 1.700], p = 1.000) from the mean score on memory of c ausal sequence in the Text with Outline condition (M = 7.645a out of 15 possible points, SD = .419). Bonferroni adjustment was chosen for multiple comparisons with the mean difference significant at the .05 level. Text o nly vs. t ext with c ausal d iagram The mean score on memory of causal sequence in the Text with Causal Diagram condition (M = 10.654a out of 15 possible points, SD = .424) was significantly higher (95% CI [1.307, 4.191], p = .000) than the mean score on memory of causal sequence in the Text On ly condition (M = 7.905a out of 15 possible points, SD = .413). Bonferroni adjustment was chosen for multiple comparisons with the mean difference significant at the .05 level. Text with Causal Diagram vs. Text with Outline The mean score on memory of caus al sequence in the Text with Causal Diagram condition (M = 10.654a out of 15 possible points, SD = .424), was significantly higher (95% CI [1.557, 4.462], p = .000) than the mean score on memory of causal sequence in the Text with Outline condition (M = 7. 645a out of 15 possible points, SD = .419).

PAGE 81

81 Bonferroni adjustment was chosen for multiple comparisons with the mean difference significant at the .05 level. Research Question 3: Understanding of Causal Sequence How is understanding of the causal sequence i n scientific text affected by the following three reading conditions? 1. reading text only in Consequent Antecedent form 2. reading text in Consequent Antecedent form accompanied by an outline in Consequent Antecedent form 3. reading text in Consequent Antecedent form accompanied by a causal diagram in Antecedent Consequent form My hypothesis was that performance on understanding the causal sequence would be highest with the causal diagram, lower with the outline, and lowest with the text alon chronological criteria in organizing their representations of causal chains. A one way ANCOVA was performed using text only, text with outline, and text with a causal diagram as the independent variable and scores on understanding of causal sequence questions as the dependent variable and the FCAT RESS as the covariate. Student res ponses to the short essay items for understanding of the causal sequence required evidence of comprehension of the cause and effect network for the main idea, oxidization. Responses for the short essay questions were evaluated on this task with a point for each correct connection and outcome, and one point for multiple choice item 3, with a maximum score of 26 points. Multiple choice item 3 and short essay items: 18, 19 and 20 were aggregated to compute composite score 3, understanding of causal sequence C ronbach's alpha was used to measure the internal consistency of c omposite score 3, (consisting of items: 3, 18, 19 and 20) ( .548).

PAGE 82

82 The experimenter and another rater, both blind to the experimental conditions, independently scored 20 percent of the respon ses for the understanding of causal sequence items using the scoring rubric (Appendix C). Cronbach's alpha was used to measure the agreement between two raters on the short answer items; for question 18, .975, for question 19, .912, and for question 20, .871 The FCAT RESS as Covariate with Understanding of Causal Sequence The relationship between the FCAT RESS and composite score 3 met the assumption that there is a linear relationship between the covariate and the dependent variable scores There was a positive relationship between the FCAT RESS and composite score 3, understanding of causal sequence, r = .293, n = 99, p = .003.There was homogeneity of regression slopes as the interaction term was not statistically significant, F(2,93) = .886, p = .416. Standardized residuals for composite score 3, understanding of causal sequence p = .092 were normally distributed, as assessed by Kolmogorov Smirnov Test (p > .05). There was homogeneity of variances, as assessed by Levene's Test of Homogeneity of Varianc e (p = .062). There were no outliers in the data (no cases with standardized residuals greater than 3 standard deviations). Tests of Between Subjects Effects Table 4 5 illustrates the adjusted and unadjusted means and standard deviations for Composite sco re 3, understanding of causal sequence, with FCAT RESS as a covariate, in the three different reading conditions. After adjustment for FCAT RESS, there was a statistically significant difference in score 3 between the conditions: Text Only, Text with Outli ne, and Text with Causal Diagram, F(2,95) = 15.694, p = .0005, 6).

PAGE 83

83 Pairwise Comparisons Text o nly vs. t ext with o utline The mean score on understanding of causal sequence in the Text Only condition (M = 4.336a out of a possible 2 6 points, SD = .543) was not significantly different (95% CI [ 1.878, 1.906], p = 1.000) from the mean score on understanding of causal sequence in the Text with Outline condition (M = 4.322a out of a possible 26 points, SD = .551). Bonferroni adjustment w as chosen for multiple comparisons with the mean difference significant at the .05 level. Text o nly vs. t ext with c ausal d iagram The mean score on understanding of causal sequence in the Text with Causal Diagram condition (M = 8.123a out of a possible 26 p oints, SD = .557) was significantly higher (95% CI [1.892, 5.682], p = .000) than the mean score on understanding of causal sequence in the Text Only condition (M = 4.336a out of a possible 26 points, SD = .543). Text with c ausal d iagram vs. t ext with o utline The mean score on Understanding of Causal Sequence in the Text wi th C ausal Diagram condition ( M = 8.123 a out of a possible 26 points SD = .557 ) was significantly higher (95% CI [ 1.892 5.710 ], p = .000) than the mean score on understanding of causa l sequence in the Text with Outline condition ( M = 4.322 a out of a possible 26 points SD = .551 ). Bonferroni adjustment was chosen for multiple comparisons with the mean difference significant at the .05 level. Summary The prediction (Research Question 1) that readers would perform better on memory of main ideas with both causal diagrams and outlines than with text alone, but

PAGE 84

84 would not differ in performance with the two different study aids was not supported. Readers performed equally well in all three con ditions, indicating they reached a similar level of memory for the main ideas in the text regardless of the presence of a study aid. As predicted (Research Question 2), scores on memory for causal sequence were significantly higher with a causal diagram as a study aid (M = 10.654) than with text alone (M = 7.905) or text with outline (M = 7.645). Not as predicted, students performed equally well on memory for causal sequence with an outline as a study aid or with text only; the outline did not enhance their memory of the causal sequence. As predicted (Research Question 3), scores on understanding the causal sequence were significantly higher with a causal diagram as a study aid (M = 8.123) than with text alone (M = 4.336) or text with outline (M = 4.322). Ag ain, not as predicted, students performed equally well on understanding the causal sequence with an outline as a study aid or with text only; the outline did not enhance their understanding of the causal sequence. Having an outline as a study aid did not i reading the text only, but having a diagram that reversed the chronological order presented in the text resulted in significantly better performance on both memory for the causal sequence and understanding of the causa l sequence. Internal consistency, as measured by Cronbach's alpha, was low for all three composite scores (Composite score 1, .432; Composite score 2, .530; Composite score 3, .548). Internal consistency of the combined test items from all three composi te scores was only moderate ( .686). While the agreement between raters on the short answer items was high ( for question 18, .975, for question 19,

PAGE 85

85 .912, and for question 20, .871 .) t hese results must be interpreted with some caution due to the low alpha coefficients.

PAGE 86

86 Table 4 1 Adjusted and Unadjusted Means and Standard Deviations for Composite score 1 Memory of Main Idea and FCAT RESS as a Covariate Unadjusted Adjusted Condition N M SD M SE Text only 34 6 4706 1 33110 6 405 a 256 Text with outline 33 6 8788 1 72767 6 947 a 260 Text with diagram 32 7 0000 1 5 4 50 2 6 999 a 262 a. Memory of main idea total of 9 possible points. Table 4 2 Tests of Between Subjects Effects Composite score1: Memory of Main Idea Source Type III Sum of Squares df MS F Sig. 2 Corrected Model 23.734 a 3 7.911 3.590 .017 .102 Intercept 4.511 1 4.511 2.047 .156 .021 SSRSS 18.609 1 18.609 8.443 .005 .082 Condition 7.171 2 3.586 1.627 .202 .033 Error 209.377 95 2.204 Total 4781.000 99 Corrected Total 233.111 98 a. R Squared = .102 (Adjusted R Squared = .073) Table 4 3 Adjusted and Unadjusted Means and Standard Deviations for Composite score 2, Memory of Causal Sequence and FCAT RESS as a Covariate Unadjusted Adjusted Condition N M SD M SE Text only 34 8.0588 2.49777 7.905 a .413 Text with outline 33 7.4848 2.84079 7.645 a .419 Text with diagram 32 10.6563 2.43111 10.654 a .424 a. Memory of causal sequence total 15 possible points.

PAGE 87

87 Table 4 4 Tests of Between Subjects Effects Composite score 2: Memory of Causal Sequence Source Type III Sum of Squares df MS F Sig. 2 Corrected Model 286.359 a 3 95.453 16.604 .000 .344 Intercept 3.065 1 3.065 .533 .467 .006 SSRSS 101.198 1 101.198 17.603 .000 .156 Condition 180.397 2 90.198 15.690 .000 .248 Error 546.146 95 5.749 Total 8338.000 99 Corrected Total 832.505 98 a. R Squared = .344 (Adjusted R Squared = .323) Table 4 5 Adjusted and Unadjusted Means and Standard Deviations for Composite score 3, U nderstanding of Causal Sequence and FCAT RESS as a Covariate Unadjusted Adjusted Condition N M SD M SE Text only 34 4.5000 3.08712 4.336 a .543 Text with outline 33 4.1515 2.99083 4.322 a .551 Text with diagram 32 8.1250 3.84162 8.123 a .557 a. Understanding of causal sequence 26 possible points. Table 4 6 Tests of Between Subjects Effects Composite score 3: Understanding of Causal Sequence Source Type III Sum of Squares df MS F Sig. 2 Corrected Model 429.145 a 3 143.048 14.406 .000 .313 Intercept 28.957 1 28.957 2.916 .091 .030 SSRSS 114.943 1 114.943 11.576 .001 .109 Condition 311.665 2 155.833 15.694 .000 .248 Error 943.300 95 9.929 Total 4428.000 99 Corrected Total 1372.444 98 a. R Squared = .313 (Adjusted R Squared = .291)

PAGE 88

88 CHAPTER 5 DISCUSSION The purpose of this study was to explore the possible effects of assisting by the causes, by providing them with a forward directional causal diagram in which the causes are given first and lead in chronological order to the consequence. I Consequent Antecedent structure t ba, 2002). Performance with the causal diagram was compared with two other reading conditions, one in which students read the text without a study aid, and one in which students were provided with an outline of the text that reflected the Consequent Antece dent form of the text. three reading conditions. In contrast, performance on memory for the causal sequence was significantly higher with the causal diagram than performance with the outline or with text alone. Performance on understanding of the causal sequence was also significantly higher with the causal diagram than performance with the outline or with text alone. Using an outline as a study aid did not facilitate either stude understanding of causal sequences in the text compared with reading the text alone, but having a diagram that reversed the chronological order presented in the text resulted in

PAGE 89

89 significantly better performance on both memory for the ca usal sequence and understanding of the causal sequence. In their study with adults, Len and Penalba (2002) asked readers to create their own causal diagrams of scientific texts, and found that readers pursued chronological criteria in organizing causal ch ains, regardless of the causal order actually presented in the text. Even when a text is not presented in chronological order, adult readers appear methodology diff propensity to make chronological sense out of causal information presented in scientific texts was highly relevant. It inspired me to think about the importance of chronology in reading and to look for research that used this idea to help people comprehend causal sequences in text. I found this in the work of McCrudden et al. (2007). McCrudden et al. (2007) gave adult readers a causal diagram of five cause and effect sequences, each on e in chronological order, and allowed them to study it along with an expository text that described the effects, but did not present them but did have a positive effect on their understanding of the causal sequences. Giving readers a chronologically organized diagram helps them to represent causal sequences accurately, even when texts do not present the sequences in cause effect order. McCrudden et al. (2007), however, di d not distinguish sufficiently between remembering a causal sequence and understanding it. All of their test questions were referred to as assessments of understanding, even though some questions only required students to remember the sequences.

PAGE 90

90 The presen t study expanded on the methods proposed by McCrudden et al. individual reading skills on the comprehension of text, and distinguishing between remembering a causal sequence and understanding it. In this chapter, I discuss the results of this study as they relate to prior research and discuss implications for theory and practice. In addition, I suggest directions for future research. Memory of Main Idea The findings f rom this study provide evidence that, as expected, use of a causal diagram had no effect on memory for main idea. The finding is consistent with prior research (McCrudden et al., 2007). Contrary to my expectations, however, students in all 3 conditions did not differ with respect to memory of main ideas; use of an outline had no effect on memory for main idea either. One would expect an outline to help memory for main ideas, since outlines express main ideas and convey hierarchical concept relations. The he lpfulness of outlines has been demonstrated by many researchers (e.g., Darch & Gersten, 1986; Glynn, Britton, & Muth, 1985). Not only was performance similar with the two study aids (as predicted) but readers reached a similar level of memory for the main idea of the text with or without either kind of study aid (not as predicted). A possible explanation for this finding is that answers to the questions in the Memory for Main Ideas section appeared explicitly within the text and did not require readers to r epresent causal chains; therefore, they were easy to remember without a study aid (Graesser & Bertus, 1998). These results are in marked contrast to the effects observed with both memory and understanding of causal sequences.

PAGE 91

91 Memory of Causal Sequence The causal structure of a text is an essential element in how the reader constructs the mental representation and how the text is understood by the reader (Linderholm, et al., 2000; O'Brien & Myers, 1987; Trabasso, Secco, & van den Broek, 1984; Trabasso, van d en Bro ek & Suh, 1989). Texts that have numerous causal term memory; as a result, they are remembered and comprehended better than texts with fewer causal connections (Mandler & Johnson, 19 1985; Trabasso & van den Broek, 1985; van den Broek, 1988). Likewise, events that connect a series of events in a causal network are better remembered than events that are not connected causally (Black & Bower, 1980; Trabasso et al., 1984; Trabasso & van den Broek, 1985). Although McCrudden et al. (2007) did not make a clear distinction between memory and understanding; they did find that providing adult readers with a diagram of the causal sequences in a text had a positive effect on their performance on test questions about the sequences. In this study, questions testing both memory and understanding were used, and the effect of a causal diagram on performance was similar (and positive) for both. The usefulnes s of causal diagrams as study aids was further underscored in this study by comparing them with another kind of study aid, the outline. The outline did not help readers to remember the causal sequences, but the causal diagram did. Readers who used the caus al diagram were the only participants who received a representation of the causal sequence that was actually in chronological order, since the sequence presented in the text was not. Similar to the findings of McCrudden et al. (2007), these

PAGE 92

92 results suggest that the students who studied the text accompanied with a causal diagram displayed a greater memory of the steps in the causal sequence compared to those who read the text alone and those who read the text with an outline (Table 4 3). an scores were not high, considering that 15 points were strategies while processing text play an important role in successful comprehension (Chi, de Leeuw, Chiu, & LaVa ncher, 1994; Cot & Goldman, 1999; Cot et al, 1998), and likely played a role in their performance on this task. As Table 4 3 illustrates, however, readers benefitted from an explicit representation of causal sequences in a visual diagram, even when their diverse reading levels were taken into account. A possible explanation for the effect of causal diagrams on memory for causal sequence is that the causal relation is easier to identify when the cause precedes its consequent (Linderholm et al., 2000) as wa readers have a tendency to pursue chronological criteria in the organization of causal chains, independent of the causal order in the text. One (2002), that is, constructing a causal diagram, to see if children make chronological associations independently of text structure, and if so, at what age. Understanding Causal Sequence As predicted, students who studied the text accompanied with a causal diagram not only remembered the causal sequence better, but also displayed a greater understanding of the steps in the causal sequence comp ared to those who read the text

PAGE 93

93 were not high (Table 4 5), considering that 26 points were possible for understanding of causal sequence. This likely reflects the avera ge to below average reading skills of the group, suggested by their FCAT RESS performance. Nonetheless, performance was highest with the causal diagram. One direction for future research might be to explore more broadly the possibility that providing causa l diagrams with science texts may improve performance for less skilled readers. This study extends our understanding of the usefulness of causal diagrams, because the effects are present with young readers, as well as with adult college students, and the e ffects are present even when variations in reading skill are taken into account. Access to an outline while reading the text resulted in no significant improvement on Understanding Causal Sequence compared with reading the text alone, but access to a caus al diagram in a forward direction resulted in significantly better understanding of the causal sequence than reading the text alone or reading the text with an outline. As noted earlier, outlines convey hierarchical, within concept relations, but they may obscure important coordinate, among concepts relations (Waller & Whalley, 1987). For example, in a study comparing the effects of studying matrices with text and outlines with text, researchers Robinson and Kiewra (1995) found no differences for factual le arning as measured by a multiple choice test. However, as measured by an essay, matrices better facilitated the learning of relationships among concepts and the ability to communicate those relationships in an organized way than did outlines. The results o bserved here indicate that studying a text with a causal diagram facilitates understanding of causal sequences more fully compared to studying text alone or text accompanied with an outline, regardless of the diverse reading levels of the children.

PAGE 94

94 Using a understanding of causal sequences in the text compared with reading the text alone. In contrast having a diagram that reversed the chronological order presented in the text result ed in significantly better performance on both memory for the causal sequence and understanding of the causal sequence. One possible reason that studying a causal diagram facilitated greater understanding of causal relationships is that these diagrams imp rove encoding by helping readers determine what is most pertinent to understanding causal sequences (McCrudden et al., 2007). A causal diagram adds value to a text that does not explicitly describe all causal sequences. A second explanation is that learner s demonstrate better understanding because causal diagrams provide information in a way that facilitates the development of an organized retrieval structure. This explanation is consistent with results obtained by several researchers that text with numerou s causal connections are term memory; as a result, they are remembered and comprehended better than texts with fewer causal connections 985; Trabasso & van den Broek, 1985; van den Broek, 1988). Readers are unlikely to understand what they cannot remember; the causal diagram facilitated both memory and understanding in this sample of young readers. As noted earlier, understanding causality in science texts is not just difficult due to the essential elements of the texts in general. The considered, especially for young readers who have less experience with ex pository text,

PAGE 95

95 because it is the opposite of what they experience with narrative texts (van den Broek, 1997). Limitations of the Study The results of this study provide information that could be useful in improving educational practice when students are pr esented with a text written in a consequent antecedent structure. It is important to note, however, that limitations of the study may have reduc ed the significance of the results Results of the Cronbach's alpha test for internal reliability were low for e ach composite score and even with the items combi ned internal reliability was at best moderate ( = 686) therefore the resul ts must be interpreted with caution. Participants in the study were fifth graders with average to below average reading skills, t he majority of whom qualified for free lunch at their urban schools. The facilitation of both memory and understanding provided by the causal diagram for these students may not generalize to students in different grades, students with stronger reading skil ls, or students from families with higher socioeconomic status. Visual displays have been widely used in science education to promote meaningful learning (Heinze Fry & Novak, 1990; Jegede, Alaiyemola, & Okebukola, 1990; Okebukola, 1990; Horton, McConney, G allo, Woods, Senn, & Hamelin, 1993; Kinchin, 2000; Kinchin, Hay, & Adams, 2000; Martin, Mintzes, & Clavijo, 2000; Odom & Kelly, 2001; Sungur, Tekkaya, & Geban, 2001). This study compared two study aids; one was a visual stimulus (causal diagram), and the o ther was verbal (an outline). A limitation of the design is that this difference in format between the two conditions is confounded with the difference in temporal order, with one study aid being in Antecedent Consequence order and the other one not. One g oal of future research

PAGE 96

96 might be to compare a visual display expressed in the same temporal order as the text (Consequent Antecedent) with the causal diagram, also a visual display, written in the Antecedent Consequence order. This would tease apart the eff ect of visual display from the effect of temporal sequence. Outlines are useful as study aids because they include only the more important text information. The outline used in this study was very comprehensive; it is possible that its length may have lim ited its usefulness as a study guide to these young readers. A second goal for future research would be to compare a short version of the outline to a very comprehensive outline, with both outlines in the same temporal order as the text. Many researchers have established the effectiveness of outlines (e.g., Darch & Gersten, 1986; Glynn et al., 1985). If carried out as described above, the comparisons would be between study aids that use the same formatting, whether verbal or visual. Most science texts do not deliver enough prompts, cues, or other supports for the generation of accurate mental representations of scientific concepts. The lack of prompts and cues is more harmful for readers with little scientific knowledge. In contrast, there are advantages for knowledgeable readers to receiving texts with coherence gaps that must be filled in with inferences. McNamara, Kintsch, Songer, and Kintsch (1996) demonstrated that high knowledge readers learn substantially more from low coherence than high coherence texts. Participants in the study were fifth graders with average to below average reading skills, the majority of whom qualified for free lunch at their urban schools. The facilitation of both memory and understanding provided by the causal diagram for these students may not generalize to students in different grades, students with stronger reading skills, or students from families with

PAGE 97

97 higher socioeconomic status. One direction for future research might be to explore more broadly the effect s of providing causal diagrams with science texts for low socioeconomic status vs. high socioeconomic status readers. I mplications for E ducators The present research has at least two implications for educators. First, providing learners with a directional causal diagram during study is a relatively simple educational intervention. An instructor can help learners focus their attention on relevant text information by providing a causal diagram that displays direct effects and/or indirect effects of the integ rated causal network. Second, studying a causal diagram can wield an influential effect on learning. Given the ease with which causal diagrams are constructed, it would be far easier for teachers to make AC diagrams for students to use to read CA texts tha n it would be for teachers to rewrite the texts themselves (AAAS, 2001). Teachers could create such materials for their students as handouts to be studied in conjunction with text. Conclusion Science text is taxing at several levels; its content is usually unfamiliar, and the concepts are often novel and expressed in a complex abstract logical sequence instead of the familiar narrative construction (Stein & Trabasso, 1981). Science text has a unique way of organizing and explaining novel content to the read 1995). When events are out of chronological order, that is, when the consequent event precedes its antecedent in a text, additional effort or attentional resources are needed (Maury et al.,

PAGE 98

98 nding of causal relationships, especially in science, as illustrated here by the finding that a causal t. While a causal diagram did not affect memory for main idea beyond reading text alone or text with an outline, the diagram did facilitate memory of causal sequence and understanding of causal sequences when studied with text. This is consistent with pre ceding research that has shown that visual displays facilitate understanding of particular types of information. For example, matrices tend to facilitate relational learning, whereas maps tend to facilitate recall for facts (Vekiri, 2002).The use of spatia l displays presents concepts in a way that enables students to use the least amount of mental effort to understand the relations among those concepts (Winn & Holliday, 1982). Causal diagrams seem to be particularly suitable to assist learning when texts de scribe implicit or complex causal relationships. Causal diagrams are distinctively paragraphs. By extracting the causal structure and making implicit causal sequences exp licit, causal diagrams enabled readers to better understand the direct causal relationships as well as the overall causal sequences. (National Reading Panel, 2000; Oakhill, Ca in, & Bryant, 2003; Paris & Paris, 2003; Snow, 2002; van den Broek, Kendeou, et al., 2005). Variations in the utilization of reading skills result in variations in comprehension. In this study, however, using the causal diagram facilitated understanding of causal relationships more fully compared to

PAGE 99

99 studying text alone or text accompanied with an outline, regardless of the diverse reading skill levels of the children.

PAGE 100

100 APPENDIX A SCIENCE TEXT IN CONSEQUENT ANTECEDENT FORM Why has the Statue of Liberty Turned Green? is green. The statue was a gift from the French to the United States. It was originally a reddish brown color, but now it is green. Did the French paint the statue green? No, t he reason for the color is a scientific one. The statue turned green because a chemical change took place. A chemical change occurs when one kind of matter changes into a completely different kind of matter with different properties. A property is a uniqu e characteristic of matter. For example, a property of garlic is its strong odor. So, an example of a chemical change is when you light a piece of paper, and it burns into ashes. The ashes are a different kind of matter than the original paper, and the as hes have different properties than the original piece of paper. A chemical change also has taken place when milk sours, because the sour milk has different properties than the original milk and it looks and tastes different. Chemical changes are often con trasted with physical change. For instance, ice melting is a physical change. The properties of the water and the properties of the ice are the same. The only difference is that the water is a liquid and the ice was a solid. One easy way to know a chemica l change has happened is to compare the color of the new material to the color of the original item. For example, a chemical change takes place when iron, such as a nail or the iron frame on your bike, rusts. We have all seen iron materials that have ruste d, and we know that rust is a reddish color. The red

PAGE 101

101 color is the result of the chemical change that has taken place. Rust looks different from iron, and the properties of rust are different from the properties of iron. Rust is the result of a particular type of chemical change called oxidation. Oxidation occurs when a material, such as iron, is exposed to oxygen, water, and carbon dioxide. Many materials will go through the process of oxidation, but only iron rusts. For example, oxidation occurs with copp er, but copper does not rust. What happens to copper? Instead of rust, a thin green film known as patina forms on the surface of the copper. This green film then prevents oxygen, water, and carbon dioxide from continuing to penetrate the copper, so the cop per is protected and oxidation slows. Patina is the result of the oxidation of copper. The patina forms when copper is exposed to oxygen, carbon dioxide, and water at the right temperature. A similar process occurs with pennies that are covered with copper If the pennies are left outside, where they are exposed to oxygen, water, and carbon dioxide, oxidation will also occur, and patina will also form on them. So, now we know why the Statue of Liberty is green. When it was first made it had the normal redd ish brown color of copper, like the color of a penny. This was because it is covered with a coat of copper. When the Statue of Liberty began to turn green, some people wanted to paint the outside of the statue and hide the patina. However, most other peopl

PAGE 102

102 APPENDIX B MULTIPLE CHOICE AND SHORT ESSAY ITEMS 1. Which of the following best identifies the main idea of the passage? a. The process of forming rust b. The process of oxidation c. The process of forming patina d. The process of ice melting 2. Which of the following materials are needed in order for copper to oxidize? a. Rust, patina, oxygen b. Carbon dioxide, water, rust c. Water, oxygen, patina d. Oxygen, water, carbon dioxide 3. Which of the following best describes how rust forming on a nail and patina forming on a copper penny are alike? a. They both occur when the nail and penny are exposed to oxygen. b. They both occur when the nail and penny are painted. c. They bo th occur when the nail and penny are exposed to oxygen, water, and carbon dioxide. d. They both occur when the nail or penny are put under water. 4. What does the patina on the Statue of Liberty protect? a. The paint on the Stature of Liberty b The copper coating of the Statue of Liberty c. The base on which the Statue of Liberty stands d. The lights that make the Statue of Liberty visible at night 5. One easy way to know a chemical change has happened is when: a. the material changes speed. b. the material changes direction. c. the material changes color. d. the material changes name. 6. The Statue of Liberty was a gift from what country? a. Italy b. France c. Spain d. Germany 7. What color was the Stature of Liberty when it was first created? a. Purplish blue b. Reddish brown c. Blackish grey d. Yellowish red

PAGE 103

103 8. When a chemical change takes place one kind of matter changes: a. into a completely different sub stance. b. into water and carbon dioxide. c. from a solid to a liquid. d. from rust to patina. a. gold. b. silver. c. copper. d. iron. 10. A metal that rusts is? a. Gold b. Silver c. Copper d. Iron 11. Which of the following is the best example of a chemical change? a. Water freezes to form ice. b. A log burns in a fireplace. c. Sugar dissolves in ice tea. d. A glass is broken into pieces. 12. Which of the following best illustrates the process of oxidation? a. Popcorn is popped. b. A tomato is cut into cubes. c. Ice cubes are ground up in a blender. d. An iron doorknob turns reddish in color. 13. Of the following, which is the first step in the process of forming patina on a copper surface? a. The copper turns green. b. The copper is exposed to oxygen, carbon dioxide, and water. c. The process of oxidation begins. d. The process of oxidation slows down. 14. Which of the follow ing is the second step that occurs when the Statue of Liberty turns green? a. The Statue of Liberty is exposed to oxygen, carbon dioxide, and water. b. The process of oxidation begins. c. The statue is painted. d. The Statue of Liberty is exposed to water as the result of rain.

PAGE 104

104 15. Which of the following is the last step that occurs when the Statue of Liberty turns green? a. The Statue of Liberty is exposed to oxygen, carbon dioxide, and water. b. The p rocess of oxidation begins. c. The statue is painted. d. The process of oxidation slows down. 16. How many steps does it take to turn the Statue of Liberty from reddish brown to green? a. 2 b. 3 c. 4 d. 5 17. Explain how the Statue of Liberty has turned green. List each of the steps beginning with the first step. Explain in as much detail as possible. 18. Does the process of oxidization continue on the surface of the Statue of Liberty after the patina has formed? Yes or No, explain your answer in as much detail as possible. 19. Does copper rust? Explain in as much detail as possible. 20. Why do you think copper roofs might appear green, but most of the pennies you see are brown, even though they are also coated with copper?

PAGE 105

105 APPENDIX C RUBRIC Memory of Main Idea Answer 1 point each 1. Which of the following best identifies the main idea of the text? a. The process of forming rust b. The process of oxidation c. The process of forming patina d. The process of ice melting B 2. Which of the following materials are needed in order for copper to oxidize? a. Rust, patina, oxygen b. Carbon dioxide, water, rust c. Water, oxygen, patina d. Oxygen, water, carbon dioxide D 5. One easy way to know a chemical change has happened is when: a. The material changes speed. b. The material changes direction. c. The material changes color. d. The material changes name. C 6. The Statue of Liberty was a gift from what country? a. Italy b. France c. Spain d. Germany B 8. When a chemical change takes place one kind of matter changes: a. into a completely different substance. b. into water and carbon dioxide. c. from a solid to a liquid. d. from rust to patina. A a. gold. b. silver. c. copper. d. iron. C

PAGE 106

106 Memory of Main Idea Answer 1 point each 10. A metal that rusts is? a. Gold b. Silver c. Copper d. Iron D 11. Which of the following is the best example of a chemical change? a. Water freezes to form ice. b. A log burns in a fireplace. c. Sugar dissolves in ice tea. d. A glass is broken into pieces. B 12. Which of the following best illustrates the process of oxidation? a. Popcorn is popped. b. A tomato is cut into cubes. c. Ice cubes are ground up in a blender. d. An iron doorknob turns reddish in color. D Memory of the Causal Sequence Answer 1 point each 4. What does the patina on the Statue of Liberty protect? a. The paint on the Stature of Liberty b. The copper coating of the Statue of Liberty c. The base on which the Statue of Liberty stands d. The lights that make the Statue of Liberty visible at night B 7. What color was the Stature of Liberty when it was first created? a. Purplish blue b. Reddish brown c. Blackish grey d. Yellowish red B 13. Of the following, which is the first step in the process of forming patina on a copper surface? a. The copper turns green. b. The copper is exposed to oxygen, carbon dioxide, and water. c. The process of oxidation begins. d. The process of oxidation slows down. Answer 1 point each B

PAGE 107

107 Memory of the Causal Sequence Answer 1 point each 14. Which of the following is the second step that occurs when the Statue of Liberty turns green? a. The Statue of Liberty is exposed to oxygen, carbon dioxide, and water. b. The process of oxidation begins. c. The statue is painted. d. The Statue of Liberty is exposed to water as the result of rain. B 15. Which of the following is the last step that occurs when the Statue of Liberty turns green? a. The Statue of Liberty is exposed to oxygen, carbon dioxide, and water. b. The process of oxidation begins c. The statue is painted. d. The proces s of oxidation slows down D 16. How many steps does it take to turn the Statue of Liberty from reddish brown to green? a. 1 b. 2 c. 3 d. 4 C 17. Explain how the Statue of Liberty has turned green. List each of the steps beginning with the first step. Explain in as much detail as possible. Answer 1 point each The copper is exposed to oxygen. The copper is exposed to water. The copper is exposed to carbon dioxide. Students are not required to list: oxygen, water, and carbon dioxide in any particular sequence. Oxidization or chem. change occurs when copper is exposed to all three components. The process of oxidation begins. Patina forms, turns copper green. The process of oxidation slows down. In correct sequence Total possible points 9

PAGE 108

108 Understanding of the Causal Sequence Answer 1 point each 3. Which of the following best describes how rust forming on a nail and patina forming on a copper penny are alike? a. They both occur when the nail and penny are exposed to ox oxygen. b. They both occur when the nail and penn y are painted. c. They both occur when the nail and penny are exposed to oxygen, water, and carbon dioxide. d. They both occur when the nail or penny are put under water. C 18. Does the process of oxidization continue on the surface of the Statue of Liberty after the patina has formed? Yes or No, explain your answer in as much detail as possible. Answer 1 point each No or Yes but it slows down. The proces s of oxidation slows down. Patina protects the copper. Patina slows down oxidation. Mentioning any of the components that the copper is not exposed to because the patina is there. Total possible points 19. Does copper rust? Explain in as much detail as possible. Answer 1 point each No It does not contain iron Iron rusts Stating oxidation causes rust on iron. Mentioning any of the components that the iron is exposed to as contributing to the formation of rust. Copper turns green because patina has formed. Mentioning any of the components that the copper is exposed to as contributing to the formation of patina. Total possible points 7 20. Why do you think copper roofs might appear green, but most of the pennies you see are brown, even though they are also coated with copper? Answer 1 point each The copper roofs are exposed to oxygen. The copper roofs are exposed to carbon dioxide. The copper roofs are exposed to water. The copper roofs oxidize, or chem. change happens. Patina forms on roofs. The pennies are not exposed to oxygen. The pennies are not exposed to carbon dioxide. The pennies are not exposed to water.

PAGE 109

109 The pennies do not oxidize. Patina does not form on pennies because they are covered, kept in Pennies are exposed to air, carbon dioxide but not in a way that allows patina to form (e.g. they are touched). If pennies are left out of the pocket, patina will form on the penny. Total possible points 13

PAGE 110

110 APPENDIX D VALIDATION CHART Validation Chart Memory of Main Idea Yes No Comments 1. Which of the following best identifies the main idea of the text? a. The process of forming rust b. The process of oxidation c. The process of forming patina d. The process of ice melting 2. Which of the following materials are needed in order for copper to oxidize? a. Rust, patina, oxygen b. Carbon dioxide, water, rust c. Water, oxygen, patina d. Oxygen, water, carbon dioxide 5. One easy way to know a chemical change has happened is when: a. The material changes speed. b. The material changes direction. c. The material changes color. d. The material changes name. 6. The Statue of Liberty was a gift from what country? a. Italy b. France c. Spain d. Germany 8. When a chemical change takes place one kind of matter changes: a. into a completely different substance. b. into water and carbon dioxide. c. from a solid to a liquid. d. from rust to patina.

PAGE 111

111 covered with a. gold. b. silver. c. copper. d. iron. Memory of Main Idea Yes No Comments 10. A metal that rusts is? a. Gold b. Silver c. Copper d. Iron 11. Which of the following is the best example of a chemical change? a. Water freezes to form ice. b. A log burns in a fireplace. c. Sugar dissolves in ice tea. d. A glass is broken into pieces. 12. Which of the following best illustrates the process of oxidation? a. Popcorn is popped. b. A tomato is cut into cubes. c. Ice cubes are ground up in a blender. d. An iron doorknob turns reddish in color. Memory of the Causal Sequence Yes No Comments 4. What does the patina on the Statue of Liberty protect? a. The paint on the Stature of Liberty b. The copper coating of the Statue of Liberty c. The base on which the Statue of Li berty stands d. The lights that make the Statue of Liberty visible at night

PAGE 112

112 Memory of the Causal Sequence Yes No Comments 7. What color was the Stature of Liberty when it was first created? a. Purplish blue b. Reddish brown c. Blackish grey d. Yellowish red 13. Of the following, which is the first step in the process of forming patina on a copper surface? a. The copper turns green. b. The copper is exposed to oxygen, carbon dioxide, and water. c. The process of oxidation begins. d. The process of oxidation slows down. 14. Which of the following is the second step that occurs when the Statue of Liberty turns green? a. The Statue of Liberty is exposed to oxygen, carbon dioxide, and water. b. The process of oxidation begins. c. The statue is painted. d. The Statue of Liberty is exposed to water as the result of rain. 15. Which of the following is the last step that occurs when the Statue of Liberty turns green? a. The Statue of Liberty is exposed to oxygen, carbon dioxide, and water. b. The process of oxidation begins c. The statue is painted. d. The process of oxidation slows dow n

PAGE 113

113 Memory of the Causal Sequence Yes No Comments 16. How many steps does it take to turn the Statue of Liberty from reddish brown to green? a. 1 b. 2 c. 3 d. 4 17. Explain how the Statue of Liberty has turned green. List each of the steps beginning with the first step. Explain in as much detail as possible. Yes No Comments Understanding of the Causal Sequence Yes No Comments 3. Which of the following best describes how rust forming on a nail and patina forming on a copper penny are alike? a. They both occur when the nail and penny are exposed to oxygen. b. They both occur when the nail and penny are painted. c. They both occur when the nail and penny are exposed to oxygen, water, and carbon dioxide. d. They both occur when the nail or penny are put under water. 18. Does the process of oxidization continue on the surface of the Statue of Liberty after the patina has formed? Yes or No, explain your answer in as much detail as possible. 19. Does copper rust? Explain in as much detail as possible. 20. Why do you think copper roofs might appear green, but most of the pennies you see are brown, even though they are also coated with copper?

PAGE 114

114 APPENDIX E CAUSAL DIAGRAM AND DIRECTIONS Effects on the Statue of Liberty Antecedent Consequence Causal Diagram Directions: will help you understand information in the text about the why the Statue of Liberty has Covered with a coat of copper tion to the order of the diagram.

PAGE 115

115 APPENDIX F CONSEQUENCE ANTECEDENT OUTLINE AND DIRECTIONS Consequence Antecedent Outline Directions: will help you understand information in the text about why the Statue of Liberty has Effects on the Statue of Liberty I. The Statue of Liberty was a gift from Franc e A. It was originally a reddish brown color, but now it is green. B. Did the French paint the statue green? II. A chemical change occurs when one kind of matter changes into a complet ely different kind of matter with different properties. A. A property is a unique characteristic of matter. B. A property of garlic is its strong odor. C. A chemical change happens when you light a piece of paper, and it burns into ashes. D. The ashes have different properties than the original piece of paper. E. Another example of chemical change is when milk turns sour. III. Chemical changes are often contrasted with physical change. A. Ice melting is a physical change. B. The properties of the water and the pr operties of the ice are the same. C. The only difference is that the water is a liquid and the ice was a solid.

PAGE 116

116 IV. One easy way to know a chemical change has happened is to compare the color of the new material to the color of the original item A. Iron materials that have rusted turn a reddish color. B. The red color is the result of the chemical change that has taken place. C. The properties of rust are different from the properties of iron. V. Rust is the result of a particular type of chemical change called oxidation A. Oxidation occurs when a material, such as iron, is exposed to oxygen, water, and carbon dioxide. B. Oxidation occurs with copper, but copper does not rust. D. When copper oxidizes a green film forms on its surface. E. The film is hydrated coppe r carbonate, or patina. VI. Patina is the end result of the oxidation of copper. A. Copper reacts with oxygen, water, and carbon dioxide, forming a green film. B. The film prevents more oxygen, water, and carbon dioxide from surface. C. Oxidization slows down. VII. The Statue of Liberty was covered with a coat of copper. A. When the Statue of Liberty was first made it had the standard reddish brown color of copper. B. When the Statue of Liberty began to turn green, some people w anted to paint the statue's surface. C.

PAGE 117

117 REFERENCES Ainsworth, S., & Loizou, A. T. (2003). The effects of self explaining when learning with text or diagrams. Cognitive Science 27, 669 681. Albers, J., How Does Oxidation Affect Heat Transfer in Copper? (1999 2011), Retrieved March 1, 2011, from http://www.ehow.com/how does_4597366_oxidation affect heat transfer copper.html American Association for the Advancement of Science. (2001). Project 2 061 AAAS science textbook conference [CD ROM]. Washington, DC: Author. Amsel, E., & Brock, S. (1996). The Development of Evidence Evaluation Skills. Cognitive Development, 11,523 550. Anderson, J. R. (1977). Memory for information about individuals. Memo ry & Cognition 5(4), 430 442. Anderson Inman, L., & Ditson, L. (1999). Computer based concept mapping: a tool for negotiating meaning. Learning and Leading with Technology, 26(8), 7 13. Anderson Inman, L., & Zeitz, L. (1993). Computer based concept mapping: Active studying for active learner. The Computer Teacher 6 8, 10 11. Baddeley, A. 1992. Working memory. Science 255: 556 9. Baker, L. (1985). Differences in the standards used by college students to evalu ate their comprehension of expository prose. Reading Research Quarterly 20(3), 297 313. Bartlett, E C. (1932). Remembering: A study in experimental and social psychology New York: MacMillan. Beck, I. L., Perfetti, C. A., & McKeown, M. G. (1982). Effects of long term vocabulary instruction on lexical access and reading comprehension. Journal of Educational Psychology, 74, 506 521. Benton, S. L., Kiewra, K. A., Whitfill, J. M., & Dennison, R. S. (1993). Encoding and external storage effects on writing proc esses. Journal of Educational Psychology 85, 267 280. Bera, S. J., & Robinson, D. H. (2004). Exploring the boundary conditions of the delay hypothesis with adjunct displays. Journal of Educational Psychology, 96(2), 381 388.

PAGE 118

118 Black, J. B. (1985). An exposi tion on understanding expository texts. In B. K. Britton & J. B. Black (Eds.), Understanding expository text: A theoretical and practical handbook for analyzing explanatory text (pp. 249 267). Hillsdale, NJ: Lawrence Erlbaum Associates. Black, J. B., & Bo wer, G. H. (1980). Story understanding as problem solving. Poetics, 9, 223 250. Britton, B. K., & Glgz, S. (1991). Using K intsch's computational model to improve instructional text: Effects of repairing inference calls on recall and cognitive structures. Journal of Educational Psychology 83(3), 329 345. Britton, B. K., van Dusen, L., Glynn, S. M., & Hemphill, D. (1990). The impact of inferences on instructional text. In A. C. Graesser & G. H. Bower (Eds.), The psychology of learning and motivation, 25, 53 70. San Diego, CA: Academic Press. Brown, D. S. (2003). High school biology: A group approach to concept mapping. The American Biology Teacher, 65(3), 192 197. Bruner, J. (1986). Actual minds, possible worlds Cambridge, MA: Harvard University Press. B udiansky, S. ( 2001 ) The Trouble With Textbooks. Prism February. from text. Educational Psychology Review 14(1), 5 26. Carnot, M., Dunn, B., Caas, A., Gram, P ., & Muldoon, J., (2007) Concept maps vs. web pages for information searching and browsing. Retrieved November 2010 from www.ihmc.us/users/acanas/Publications/CMaps. Carter, T., & Dean, E. (2006). Mathematics intervention for grades 5 11: Teaching mathemat ics, reading, or both? Reading Psychology 27(2), 127 146. Case, R. (1999). Developing Children's Understanding of the Rational Numbers: A New Model and an Experimental Curriculum. Journal for Research in Mathematics Education, 30(2) 122 147. Casteel, M. A. (1993). Effects of inference necessity and reading goal on children's inferential generation. Developmental Psychology 29(2), 346 357. Chandler, P., & Sweller, J. (1991). Cognitive load theory and the format of instruction. Cognition and Instruction 8, 293 332. Chandler, P., & Sweller, J. (1992). The split attention effect as a factor in the design of instruction. British Journal of Educational Psychology 62, 233 246.

PAGE 119

119 Chang, K. E., Chen, S F. & Sung, T.C. (2001) The Effect of Concept Mapping to Enhance Text Comprehension and Summarization Journal of Experimental Education, 71(1) 5 23. Cheng, P. C. H., Lowe, R. K., & Scaife, M. (2001). Cognitive science approaches to understanding diagrammatic re presentations. Artificial Intelligence Review 15(1 2), 79 94. Chi, M. T. H., de Leeuw, N., Chiu, M. H., & LaVancher, C. (1994). Eliciting self explanations improves understanding. Cognitive Science 18, 439 477. Conradty, C., & Bogner, F. (2010). Implemen tation of concept mapping to novices: reasons for errors, a matter of technique or content? Educational Studies 36(1), 47 58. Cot, N. C., & Goldman, S. R. (1999). Building representations of informational text: aloud protocol s In H. van Oostendorp & S.R. Goldman(Eds.), The construction of mental representations during reading (pp. 169 193). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Cot, N., Goldman, S. R., & Saul, E. U. (1998). Students making sense of informational text: Relations between processing and representation. Discourse Processes 25(1), 1 53. Darch, C, & Gersten, R. (1986). Direction setting activities in reading comprehension: A relation of two approaches. Learning Disabled Quarterly 9, 235 243. Dewey, J. (1938 ). Logic. The theory of inquiry New York: Holt. DiCecco, V. M., & Gleason, M. M. (2002). Using graphic organizers to attain relational knowledge from expository text. Journal of Learning Disabilities 35(4), 306 320. Duval County Public Schools Grade 5 Science (2009 2010) Curriculum 15 18. Einstein, A., & Infeld, L. (1938). The evolution of physics: The growth of Idea from early concepts to relativity and quanta NY: Simon and Schuster. Elshout Mohr, M., & van Daalen' Kapteijns, M. (2002). Situated regulation of scientific text processing. In A. C. Graesser, J. A. Len, & J. Otero (Eds.), The psychology of science text comprehension (pp. 223 252). Mahwah, NJ: L. Erlbaum Associates. (2009). The effect of concept mapping on meaningful learning of Atom and bonding. World Conference on Educational Sciences 2009 Procedia Social and Behavioral Sciences 1, 1586 1590.

PAGE 120

120 FCAT 2.0 and Florida EOC Assessments Achievement Levels (2013). Florida De partment of Education/Office of Assessment http://fcat.fldoe.org/fcat2/pdf/achlevel.pdf FCAT 2.0 Florida Comprehension Assessment Test (2013). Grade 4 FCAT 2.0 Reading Achievement Level Descripti ons Florida Department of Education/Office of Assessment http://fcat.fldoe.org/fcat2/pdf/readgrade4.pdf Fischer, F., Bruhn, J., Grasel, C., & Mandl, H. (2002). Fostering collaborative knowledge construction with visualization tools. Learning and Instruction, 12, 213 232. Fletcher, C. R., & Bloom, C. P. (1988). Causal reasoning in the comprehension of simple narrative texts. Journal of Memory and Language 27, 235 244. Florida Comprehensive Asses sment Test 2.0 (2013) Grade 4 FCAT 2.0 Reading Florida Comprehensive Assessment Test 2.0 (2013). Understanding FCAT 2.0 Reports Spring 2013. http://fcat.fldoe.org/mediapacket/2013/pdf/2 013UFR.pdf Gernsbacher, M. A. (1991). Language comprehension as structure building Hillsdale, N.J.: L. Erlbaum. Glynn, S. M., Britton, B. K., & Muth, K. D. (1985). Text comprehension strategies based on outlines: Immediate and long term effects. Journal of Experimental Psychology, 53, 129 135. Gopnik, A., Sobel, D. M., Schulz, L. E., & Glymour, C. (2001). Causal learning mechanisms in very young children: Two three and four year olds infer causal relations from patterns of variation and covariation. D evelopmental Psychology 37(5), 620 629. Graesser, A. C. (1981). Prose comprehension beyond the world NY: Springer Verlag. Graesser, A. C., & Bertus, E. L. (1998). The construction of causal inferences while reading expository texts on science and techno logy. Scientific Studies of Reading, 2(3), 247. Graesser, A. C., Len, J. A., & Otero, J. (2002). The psychology of science text comprehension Mahwah, N.J.: L. Erlbaum Associates. Graesser, A. C., Millis, K. K., & Zwaan, R. A. (1997). Discourse Comprehe nsion. Annual Review of Psychology 48, 163 189. Graesser, A. C., & Person, N. K. (1994). Question asking during tutoring. American Educational Research Journal, 31(1), 104 137. Graesser, A. C., Singer, M., & Trabasso, T. (1994). Constructing inferences d uring narrative text comprehension. Psychological Review 101, 371 395.

PAGE 121

121 Harniss, M., Hollenbeck, K., Crawford, D., & Carnine, D. (1994) Content Organization and Instructional Design Issues in the Development of History. Learning Disability Quarterly 17(3) 235 248. Hart, H., & Honore, T. (1959). Causation in the law Oxford, England: Claredon. Hayes, J. R. (1989). The complete problem solver (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum. Heinze Fry, J. & Novak, J. D. (1990) Concept mapping brings long term movement toward meaningful learning, Science Education 74, 461 472. Hergenhahn, B. R. (2005). An introduction to the history of psychology (5th ed.). Belmont, CA: Wadsworth. Hong, L., Chijun, Z., Xuemei, G., Sha n, G., & Chongde, L. (2005). The influence of Cognitive Development, 20, 87 101. Horton, P. B., McConney, A. A., Gallo, M., Woods, A. L., Senn, G. J. & Hamelin, D. (1993). An investigation of the effectiveness of concept mapping as an instructional tool. Science Education 77, 95 111. Hume, D. (1978). A treatise of human nature Oxford, England: Oxford University Press. (Original work published 1739) Jegede, O. J., Alaiyemola, F. F. & Okebu kola, P. A. O. (1990) The effect of concept Journal of Research in Science Teaching, 27, pp. 951 960. Johnson Laird, P.N. (1983). Mental models. Cambridge, MA: Harvard University Press. Johnson Laird, P. N., Legrenzi, P., & Legrenzi, M. (1972). Reasoning and a sense of reality. British Journal of Psychology 63, 395 400. Johnson Laird, P. N. (1989). Mental models. In M. I. Posner (Ed.), Foundations of cognitive science (pp. 469 500). Camb ridge, MA.:MIT Press. Jonassen, D. (2000). Computers as mind tools for schools (2nd ed.). Engaging critical thinking. Merrill. Jonassen, D. (2003). Using cognitive tools to represent problems. Journal of Research on Technology in Education 35(3), 362 381 Jonassen, H. D., Beissner, K., & Yacci, M. A. (1993). Structural knowledge: Techniques for representing, assessing, and acquiring structural knowledge. Hillsdale, NJ: Lawrence Erlbaum Associates.

PAGE 122

122 Jonassen, D. H., Reeves, T. C., Hong, N., Harvey, D., & Ka ren, P. (1997). Concept mapping as cognitive learning and assessment tools. Journal of Interactive Learning Research 8(3/4), 289 308. Kao a, G., Lin, S., & Sun, C., (2008). Breaking concept boundaries to enhance creative potential: Using integrated concep t maps for conceptual self awareness. Computers & Education 51, 1718 1728. Kendeou, P., Rapp, D. N., & van den Broek, P. (2004). The influence of reader's prior knowledge on text comprehension and learning from text. In R. Nata (Ed.), Progress in educatio n (Vol. 13, pp. 189 209). Hauppauge, NY: Nova Science. Kintsch, W. (1998). Comprehension : A paradigm for cognition Boulder, Colo.: Cambridge University Press. Kendeou, P., & v an d en Br oek P. (2005). The effects of readers' misconceptions on comprehensi on of scientific text. Journal of educational psychology 97(2), 235 245. Khishfe, V. L., & Abd El Khalick, F. (2002). Influence of Explicit and Reflective versus Implicit Inquiry Science. Journal of Research in Science Teaching 39(7), 551 578. Kinchin, I. M. (2000) Concept mapping in biology, Journal of Biological Education 34, pp. 61 69. Kinchin, I. M., Hay, D.B. & Adams, A. (2000) How a qualitative approach to cm analysis can be used to aid le arning by illustrating patterns of conceptual development, Educational Research 42, pp. 43 57. Kintsch, W. (1988). The use of knowledge in discourse processing: A construction integration model. Psychological Review 95, 163 182. Kintsch, W. (1994, 1998/2 007). Text comprehension, memory, and learning. American Psychologist 49(4), 294. Kintsch, W. (1998). Comprehension : A paradigm for cognition Boulder, Colo.: Cambridge University Press. Kintsch, W. (2004). The construction integration model of text co mprehension and its implications for instruction. In R. B. Ruddell, & N. Unrau (Eds.), Theoretical models and processes of reading (pp. 1270 1328). Newark, DE: International Reading Association. Kintsch, W, & Bates, E. (1977). Recognition memory for state ments from a classroom lecture. Journal of Experimental Psychology: Human Learning and Memory 3, 150 159.

PAGE 123

123 Klahr, D. (2000). Exploring science: The cognition and development of discovery processes Cambridge, MA, US: The MIT Press. Koerber, S., Sodian, B., Thoermer, C., & Nett, U. (2005). Scientific reasoning in young children: Preschoolers' ability to evaluate covariation evidence. Swiss Journal of Psychology Revue 64(3), 141 152. Konior, J. (1993). Research into the construction of mathematical texts. E ducational Studies in Mathematics 24(3), 251. Koslowski, B. (1996). Theory and evidence: The development of scientific reasoning. Cambridge, Mass.: MIT Press. Koslowski, B., Marasia, J., Chelenza, M., & Dublin, R. (2008). Information becomes evidence when an explanation can incorporate it into a causal framework. Cognitive Development 23(4), 472 487. Kuhn, D., Amsel, E., O'Loughlin, M., Schauble, L., Leadbeater, B., & Yotive, W. (1988). The development of scientific thinking skills San Diego, CA, US: Academic Press. Kuhn, D., & Franklin, S. (2006). The second decade: What develops (and how). Handbook of child psychology: Vol. 2, cognition, perception, and language (6th ed.). (pp. 953 993) Hoboken, NJ, US: John Wiley & Sons Inc. Kuhn, D., Iordanou, K., Pease, M., & Wirkala, C. (2008). Beyond control of variables: What needs to develop to achieve skilled scientific thinking? Cognitive Development, 23(4), 435 451. Kuhn, D., & Pease, M. (2008). What needs to develop in the development of inquiry skills? Cognition and Instruction 26(4), 512 559. Kwon, S., & Cifuentes, L., (2009). The comparative effect of individually constructed vs. collaboratively constructed computer based concept maps. Computers & Education, 52, 365 375. Lee, C., (2010). The interactions between problem solving and conceptual change: System dynamic modeling as a platform for learning. Computers & Education 55, 1145 1158. Lee, H., Plass, J., & Homer, B., ( 2006) Optimizing Cognitive Load for Learning From Computer Based Science Simulations. Journal of Educational Psychology 98(4), 902 913. scientific discourse. In J. Ote The psychology of science text comprehension (pp. 155 178). Mahwah, NJ: L. Erlbaum Associates.

PAGE 124

124 Linderholm, T., Everson, M. G., van den Broek, Mischinski, M., Crittenden, A., & Samuels, J. (2000). Effects of causal text revisions on more and less skilled readers' comprehension of easy and difficult texts. Cognition & Instruction 18(4), 525 556. Lipson, M. Y. (1982). Learning new information from text: The role of prior knowledge and reading ability. Journal of Re ading Behavior 14(3), 243 261. Loman, N. L., & Mayer, R. E. (1983). Signaling techniques that increase the understandability of expository prose Journal of Educational Psychology 75, 402 412. Lowe, R. K. (2003). Animation and learning: Selective proces sing of information in dynamic graphics Learning and Instruction 13(2), 157 176. Lutz, M. F., & Radvansky, G. A. (1997). The fate of completed goal information in narrative comprehension. Journal of Memory and Language 36(2), 293 310. Mackie, J. L. (19 80). The cement of the universe: A study of causality Oxford, England Clarendon. Mandler, J. M., & Johnson, N. S. (1977). Remembrance of things parsed: Story structure and recall. Cognitive Psychology 9,111 151. Martin, B. L., Mintzes, J. J. & Clavijo, I (2000) Restructuring knowledge in Biology: cognitive processes and metacognitive reflections, International Journal of Science Education, 22, pp. 303 323. Masnick, A. M., & Klahr, D. (2003). Error matters: An initial exploration of elementary school chil dren's understanding of experimental error. Journal of Cognition and Development, 4(1), 67 98. Mason, L., & Gava, M. (2007). Effects of epistemological beliefs and learning text structure on conceptual change. In S. Vosniadou, A. Baltas, & X. Vamvakoussi (Eds.), Reframing the conceptual change approach in learning and instruction (pp. 165 196). Amsterdam, Netherlands: Elsevier. technological contexts. In J. Otero, J. A. Le The psychology of science text comprehension (pp. 199 223). Mahwah, NJ: L. Erlbaum Associates. Mayer, M. (1969) Frog, where are you? Dial Press, New York. Mayer, R. E. (2001). Multimedia learning New York: Cambridge Univers ity Press. Mayer, R. E., & Moreno, R. (2002). Animation as an aid to multimedia learning. Educational Psychology Review 14(1), 87 99.

PAGE 125

125 McCrudden, M. T., Schraw, G., Lehman, S., & Poliquin, A. (2007). The effect of causal diagrams on text learning. Contempo rary Educational Psychology, 32(3), 367 388. Meyer, B. J., & Freedle, R. O. (1984). Effects of discourse type on recall. American Educational Research Journal, 21(1), 121 143. Miller, M. D., Linn, R. L. & Gronlund, N. E. (2009) Measurement and evaluation in teaching (10th. ed.). Upper Saddle River, NJ: Merrill. Moravcsik, J. E., & Kintsch, W. (1993). Writing quality, reading skills, and domain knowledge as factors in text comprehension. Canadian Journal of Experimental Psychology/Revue Canadienne De Psychologie Exprimentale 47(2), 360 374. National Reading Panel (US), National Institute for Literacy (US), National Institute of Child Health and Human Development (US), National Institutes of Health (US), National Reading Excellence Initiative (US), U nited States Dept. of Health and Human Services, et al. (2000). National reading panel: Teaching children to read: An evidence based assessment of the scientific research literature on reading and its implications for reading instruction : Reports of the s ubgroups. Washington, DC: National Institute of Child Health and Human Development, National Institutes of Health. Newton, D. P. (1995). Support for understanding: Discourse which aids the construction of a functional mental model of causal situations. Res earch in Science and Technological Education, 13, 109 122. Novak, J. D. (1981). Applying learning psychology and philosophy of science to biology teaching. The American Biology Teacher 43(1), 12 20. Novak, J.D. (2002) Meaningful learning: The essential f actor for conceptual change in limited or inappropriate prepositional hierarchies leading to empowerment of learners. Learning 548 571. Novak, J. D., & Gowin, D. B. (1984). Concept mapping for meaningful learning. Cambridge: Cambridge University Press. Nov ak, J. D., Gowin, D. B., & Johansen, G. T. (1983). The use of concept mapping and knowledge Vee mapping with junior high school science students. Science Education, 67(5), 625 645. Nussbaum & Novak, ( 1976 ) An assessment of children's concepts of the earth utilizing structured interviews. Science Education 60, 535 550. Oakhill, J. V., Cain, K., & Bryant, P. E. (2003). The dissociation of word reading and text comprehension: Evidence from component skills. Language and Cognitive Processes, 18, 443 468

PAGE 126

126 O'Br ien, E. J., & Myers, J. L. (1987). The role of causal connections in the retrieval of text. Memory and Cognition 15, 419 427. for cognitive processing. Educational Psych ology Review 14, 71 86. Odom, A.L. & Kelly, P.V. (2001). Integrating concept mapping and the learning cycle to teach diffusion and osmosis concepts to high school biology students, Science Education, 85 615 635. Oestermeier, U., & Hesse, F. W. (2000). Ve rbal and visual causal arguments. Cognition 75(1), 65 104. Ogle, D., & Blachowicz, C.L.Z. (2002). Beyond literature circles. In C.C. Block & M. Pressley (Eds.), Comprehension instruction: Research based best practices (pp.259 274). New York: Guilford. Ok ebukola, P. A. O. (1990) Attaining meaningful learning of concepts in genetics and ecology: an examination of the potency of the concept mapping technique, Journal of Research in Science Teaching 27, 493 504. Paris, A. H., & Paris, S. G. (2003). Assessing narrative comprehension in young children. Reading Research Quarterly 38, 36 76. Perkins, D. N., & Simmons, R. (1988). Patterns of misunderstanding: An integrative model for science, math, and programming. Review of Educati onal Research 58(3), 303 326. Piaget, J. (1927a ). La causaliu: physique chez l'enfant [The child's conception of physical causality]. Paris: Alcan. Piaget, J. (1927b). I'explication de l'ombre chez l'enfant [Children's explanations of shadows]. Journal d e Psychologie 24, 230 242. Pot, J. What Made the Statue of Liberty Turn Green? (1999 2009). Retrieved March 1, 2011, from (http://www.ehow.com/how does_5439713_made statue liberty turn green.html Potts, G. R., Keenan, J. M., & Golding, J. M. (1988). Asse ssing the occurrence of elaborative inferences: Lexical decision versus naming. Journal of Memory and Language, 27, 399 415. Reichenbach, H. (1956). The direction of time. Berkeley: University of California Press. Robinson, D. H., & Kiewra, K. A. (1995). Visual argument: Graphic organizers are superior to outlines in improving learning from text. Journal of Educational Psychology 87(3), 455 467.

PAGE 127

127 Robinson, D. H., & Kiewra, K. A. (1995). Visual argument: Graphic organizers are superior to outlines in improv ing learning from text. Journal of Educational Psychology, 87(3), 455 467. Robinson, D. H., Robinson, S. L., & Katayama, A. D. (1999). When words are represented in memory like pictures: Evidence for spatial encoding of study materials. Contemporary Educat ional Psychology 24(1), 38 54. Robinson, D. H., & Schraw, G. (1994). Computational efficiency through visual argument: Do graphic organizers communicate relations in text too effectively? Contemporary Educational Psychology 19(4), 399 415. Royer, R., & R oyer, J. (2004). Comparing hand drawn and computer generated concept mapping. Journal of Computers in Mathematics and Science Teaching 23(1), 67 81. Ruffman, T., Perner, J., Olson, D. R., & Doherty, M. (1993). Reflecting on scientific thinking: Children's understanding of the hypothesis evidence relation. Child Development, 64(6), 1617 1636. Schleppegrell, M. (2007). The linguistic challenges of mathematics teaching and learning: a research review Reading & Writing Quarterly 23, 139 159. Schnotz, W., & B annert, M. (2003). Construction and interference in learning from multiple representation. Learning and Instruction 13(2), 141 156. Shah, P., & Hoeffner, J. (2002). Review of graph comprehension research: Implications for instruction. Educational Psychology Review 14(1), 47 69. Shah, P., Mayer, R. E., & Hegarty, M. (1999). Graphs as aids to knowledge construction: Signaling techniques for guiding the process of graph comprehension. Journal of Educational Psychology 91(4), 690 702. Silver, E. A. & Marshall, S. P. (1990). Mathematical and scientific problem solving: findings, issues, and instructional implications. In B. F. Jones & L. Idol (Eds), Dimensions of thinking and cognitive instruction (pp. 265 290). Hillsdale, NJ: Lawrence Erlbaum. Smith, M. U. (1991). A view from biology. In M. U. Smith (Ed.), Toward a unified theory of problem solving (pp. 1 20). Hillsdale, NJ: Lawrence Erlbaum. Sneider, C., & Pulos, S. (1983). Children's cosmographies: Understanding the earth's shape and gravity. Science Education 67, pp. 205 221. Snow, C. (2002). Reading for understanding: Toward an R&D program in reading comprehension. Santa Monica, CA: RAND Corporation.

PAGE 128

128 Spiro, R. J. (1980). Accommodative reconstruction in prose recall. Journal of Verbal Learning & Ver bal Behavior, 19(1), 84 95. Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly 21, 360 364. n approach to comprehension and instruction. In R. Glaser (Ed.), Advances in instructional psychology 2, 213 267. Hillsdale, NJ: Erlbaum. Sungur, S., Tekkaya, C. & Geban, O. (2001). The contribution of conceptual change texts accompanied by concept mappin circulatory system, School Science and Mathematics 101, pp. 91 101. Sweller, J. ( 2006 ) How the human cognitive system deals with complexity. In Handling complexity in learning environments: Theory and research e d. J. Elen and R. Clark, 13 25. Amsterdam, The Netherlands: Elsevier. Sweller, J., & Chandler, P. (1991). Evidence for cognitive load theory. Cognition and Instruction, 8(4), 351 362. Sweller, J., Chandler, P., Tierney, P., & Cooper, M. (1990). Cognitive load as a factor in the structuring of technical material. Journal of Experimental Psychology 119, 176 192. Tarmizi, R.A., & Sweller, J. ( 1988 ) Guidance during mathematical problem solving. Journal of Educational Psychology 80(4), 424 36. Thiede, K. W., Anderson, M. C. M., & Therriault, D. (2003). Accuracy of metacognitive monitoring affects learning of texts. Journal of Educational Psychology 95(1), 66 73. Trabasso, T., & Nickels, M. (1992). The development of goal plans of action in the narration of a picture story. Discourse Processes 15(3), 249. Trabasso, T., Secco, T., & van den Broek, P.W. (1984). Causal cohesion and story coherence. In H. Mandl, N. L. Stein, & T. Trabasso (Eds.), Learning and comprehension of text (pp. 83 11). Hillsdale, NJ: Erlbaums. Trabasso, T., & Sperry, L. L. (1985). Causal relatedness and importance of story events. Journal of Memory and Language 24(5), 595 611. Trabasso, T., & van den Broek, P. (1985). Causal thinking and the representatio n of narrative events. Journal of Memory and Language 24, 612 630. Trabasso, T., van den Broke, P., & Suh, S. (1989). Logical necessity and transitivity of causal relations in stories. Discourse Processes 12(1), 1.

PAGE 129

129 van den Broek, P. (1988). The effects of causal relations and hierarchical position on the importance of story statements, Journal of Memory and Language 27(1), 1 22. van den Broek, P. (1990a). The causal inference maker: Towards a process model of inference generation in text comprehension. In D. A. Balota, G. B. Flores Comprehension processes in reading (pp. 423 445). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. van den Broek, E (1990b). Causal inferences and the comprehension of narrative texts. In A. C. G raesser & G. H. Bower (Eds.), Psychology of learning and motivation: Inferences and text comprehension 25, 175 196. San Diego: Academic Press. van den Broek, P. (1997). Discovering the cement of the universe: The development of event comprehension from ch ildhood to adulthood. In P. van den Broek, P. J. Bauer, & T. Bourg (Eds.), Developmental spans in event comprehension and representation (pp. 321 342). Mahwah, NJ: Lawrence Erlbaum Associates. van den Broek, P. (2010). Knowledge Representation Using Texts in Science Education: Cognitive Processes and Science 328, 453. van den Broek, P., Kendeou, P., Kremer, K., Lynch, J. S., Butler, J., White, M. J., et al. (2005). Assessment of comprehension abilities in young children. In S. Stahl & S. Paris (Eds.), Chil (pp. 107 130). Mahwah, NJ: Lawrence Erlbaum Associates. van den Broek, P., Virtue, S., Everson, M., Tzeng, Y., & Sung, Y. (2002). Comprehension and Memory of Science Texts: Inferential Processes and the Construct ion of a Mental Representation. In A. C. Graesser, J. A. Len, & J. Otero (Eds.), The psychology of science text comprehension (pp. 131 154). Mahwah, NJ: L. Erlbaum Associates. van Dijk, T. A., & Kintsch, W. (1983 ). Strategies of discourse comprehension N ew York: Academic. Vekiri, I. (2002). What is the value of graphical displays in learning? Educational Psychology Review, 14(3), 261 312. Verdi, M. P., & Kulhavy, R. (2002). Learning with maps and texts: An overview. Educational Psychology Review 14(1), 27 46. Vosniadou, S., & Brewer, W. F. (1992). Mental models of the earth: A study of conceptual change in childhood. Cognitive Psychology 24(4), 535 585. Vosniadou, S., & Brewer, W. F. (1994). Mental models of the day/night cycle. Cognitive Science: A M ultidisciplinary Journal, 18(1), 123 183.

PAGE 130

130 Voss, J. F., & Silfies, L. N. (1996). Learning from history text: The interaction of knowledge and comprehension skill with text structure. Cognition & Instruction 14(1), 45. Waller, R., & Whalley, P. (1987). Gr aphically organized prose. In E. de Corte (Ed.), Learning and instruction: European research in an international context, 369 381. Leuven, Belgium: Leuven University Press. Ward, M., & Sweller, J. (1990). Structuring effective worked examples. Cognition & Instruction 7, 1 39. West, C. K., Farmer, A., & Wolff, P. M (1991). Instructional Design: Implications from Cognitive Science Englewood Cliffs, N I : Prentice Hall White, E A. (1989). A theory of causal processing. British Journal of Psychology 80, 431 45 4. Wiegmann, D. A., Dansereau, D. F., McCagg, E. C., Rewey, K. L., & Pitre, U. (1992). Effects of knowledge map characteristics on information processing. Contemporary Educational Psychology 17(2), 136 155. Wilkening, F., & Sodian, B. (2005). Editorial: S cientific reasoning in young children: Introduction. Swiss Journal of Psychology 64(3), 137 139. Winn, W. (1988). Recall of the pattern, sequence, and names of concepts presented tn instructional diagrams. Journal of Research in Science Teaching 25, 375 386. Winn, W., & Holliday, W. G. (1982). Design principles for diagrams and charts. In D. Jonassen (Ed.), The technology of text 1 277 299). Englewood Cliffs, NJ: Educational Technology Publications. Youngmin, L., & Nelson, D., (2005). Best Concept Map S trategy on Problem Solving Performance British Journal of Educational Technology Viewing or visualizing which concept map strategy works best on problem solving performance? 136(2) 193 203. Zhang, J. (1997). The nature of external representations in proble m solving. Cognitive Science 21 179 217. Zimmerman, C. (2007). The development of scientific thinking skills in elementary and middle school. Developmental Review 27(2), 172 223.

PAGE 131

131 BIOGRAPHICAL SKETCH Clairemarie Gonzlez was born in Philadelphia, Pen nsylvania and raised in education from the University of North Florida. Continuing her education at the sional education education at the University of North Florida, she was encouraged to pursue a Ph.D. in educational psychology at the University of Florida. Claire current ly holds a position as an instructor of educational psychology at the University of North Florida in Jacksonville.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EEEKOLTB0_J7740D INGEST_TIME 2014-10-03T21:34:46Z PACKAGE UFE0046482_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES