1' ....— .——.——- (I‘ _-.r_ '3" '0I . Ib1.. 11.9111. 11"; .,‘..‘-V?7:“'.- ° 7-... ~ ‘_~. ‘ _ -. .. 932°33’5- {-CTN' S—J'Tffi‘fi‘” ' - « . - 2&5”: ' 1| 1. ' o . - - 0.... -' . -.I Q .errt-z‘kJ-tu .. x-I ... 3 .a .k' -:.;', i—n'j; ‘ U . _ 1 . ‘ ._. - ' agafi‘- ' A ' v- ‘> no. u . 3 - - _‘ ‘ f;‘ 3}- ‘. I. 11.11," 1111111 ”111%“ 1 1 1‘II' 11.1? ‘0'» ’ . .. :.~ 2 51%. 1 .1 . ' . . r..‘~;- '1 i' 3:" . Wu) - . '.r . _ :ii 19:41:32 . ' ' .‘ ' 3 . 31.311111 3:! f'v'IF‘F ." 1' 1 ‘1' I 1' 1 1'11II""'({1'WI'\§ " 3' - 1'1 ' 1‘ | '1' .. 1:1 ”11. ' ’1' -' 1'11 ".4 :: : ‘ 3:1 I l’ ,1" ‘ ' . 1, u: 2 ' I": :1 . ' I .-.:11”L:: 1' \I o- 1"}: 1'1: 1.12:: .7 ‘ ‘. .- .' . '. - .711’ 111111 ' 'z. - ' "1‘32 1.?" 1..) ""1"'l" I1 41:1 [111111. '1 1 1' ’Ill1l11'l' 1}"vr. '1'131'1 1111.1“ .1111; '111-1'. ‘21,} IF111111111111HM111111111115H11111N11H11hhgyfiw11fiug :3; 1111111111'11. :11 111 I 1111'- 111,... “1'1 1' 1:1 1:11.911: '11-..I1 1' .21.. 11 1", .11 » 1 w. IW111‘3111V1w11%"M%MU%51 -1 . 1:111” '11::“11111. 1111:1111'11111'1'11111'1'11'1'1131"" "1111'; “:1 '1 11:131. .1 '1‘”:‘111"131':1.‘1'1':f":1: 11111111111111.13 "114541111111 :1: 1;: i ,1; 3 I; v' 1 -wr..m1 ‘1 'IuNW“”‘ ' qqfiawfi ..u M! . 1111.11111.11.11.111'1'11'1'1111211'11 ',1:111".1;:::1 1:1." 11 (1:111 :1 11! 1"1': : 111515121131;1:1;11'11 '11 113111;:111 :43? 1 " .'. ‘I' :_ " ' '1"I»r~‘,-’~'tf If," I I '1‘ ' 3 1' '21.,‘1’,1 11'11'1"111':i.'1.':1‘:-’|':.1’L‘."’I'Jfa"'"‘ 2.13111 . . : af' 1' ‘ 1:11:11“? . l 1 : . , (1‘ 3.11 '1fl1'1'1'11’1:1'11111( 111'. 1 . .3 o '11": ‘15; II 111'? '1' 111' '11_ '1-"1'V1 ':'1l" I11. ' ‘11‘1! : ‘.I ' 11"1111 ’111 a _. -. _ . w < ‘l w 3: 1 . -. ‘_ - .u ipg‘ ~- 1?? a .1 L‘ . .l , ‘ . 3 ' , _ Q .0. 1e . 134111411711: 1:41:11 : I ‘I: .2111 ' 1HWN'HK11MI ‘11, :'éw« 1 1.111921}! I1 11111.11 ..... 14:11-11: I1: 1, 1 I 1111113114.. . .1 -. 1 ~. ' 1.111111 13.5.1.4} ‘ .3'u3:.~:1:~;. . I ; :'l.1.11"1:;j.f1i';11‘. 1H'l1':"'1"111'1~'1'1"1 .1 .131 1.1 '1‘ t" 1” 11511.11 ‘~—-'L_ *cvfi «7. :1'1 1?" ‘5'; . ‘ .1 1',” 1'1" ' .1171 1131:1131: M ' (1":1'i3‘l'1'z11' 1"":'1"":I‘1.;', \— .ng': v -~;. ..”I 111'1'11’."" ”1'11" :11} «1“! y. b 1"...1 11.1 1 11 1111' 1I I 1": ”Vi-1111'- 11131111111 11111111 ID. 1111 .l . . 1"11 '111 41111:. 11:11"1'111111'111111'1'1'1111‘u'1'11 1:1 ,, 11.1111. 111 ...... 11-: m .01 “-4.4“. (111111 I'1II,::'~\1,II1.”I 1 11111111111111 111.1111; 1I11I 11111.1: 1 1. .I1 1 11111111. 1113111 Iuwwi 11,11f1 1:1:1 '11- I1" 13W 111111111111 11141111“ . ‘1 51.1,] 911.4“ I- ”111:1 l‘ .1 1 111 11:11:11 1 1' 1311'.» '1', 111111111'1 ' 1'1 1111111111 I 11 ' '11111 1151' .11'1 1 I 1 I‘ ‘1'" 1 11 I1 1' "1111' .111.'I'..11 1.“ I ru111111'111r1‘ 1 '"11 '1'1'lsIII‘1II .1 11.111111511111411: ' ' 1.11.1111 111.: 711.111'1'11111' 1'" 11' ., ' 1.1‘1'111' “.1114 ."II :1 "1‘.“ .1111 1.141 1 1 11";11'1111'I111'1'3xl1:1“17111'1111'1'1:111111'1111‘1911'1'1111“11111"'[1|1:1':;I.:1 11' ”1111111; .1111 )1 1' :11'11'1;§1'fi+'1111’111'1‘11‘1111‘1'1'{111*}1'1'1"1'“qu 1 ..: 1:,” 1:1. 11.1.I111111'1I11 1 I": 11'1”] .1 [:1 114:1". 1 1 In” 11:1 I! 1 ' 1 .11111' 11111111 1"1'11I ' '1' '11'1‘ 11' 1111'} 1.111.111 1‘1 11111 "'1 1114, 11.11111 I 11111 wig-1'1"" ‘ 1'1" «1......11 1 .11 .. 11" " ' 11.11111 11111 11111 '111 I W ‘ ' II 1 11.1.1: 1'11 II.1I'JII I1I11IIIII'I'1111I 11111111 1'11“.""011'1'1'1'1 . .. 111% ”Mk III III] 230 ..-..l IIIIIII IIIIIIIIIIII 070 4399 I ¥‘-‘, _f*” ’4; a‘ .. .- . . a: D b v fa,1' ~___a~-- 1:45;: bum—v- — I O .1 t at q-A-v-f '3' J?Nfl"-§~i-i mesh This is to certify that the dissertation entitled Effects of Self-Questioning Guided by Bloom's Taxonomy on the Readino Comprehension of Science Materials presented by Nora Josefine Sabater Valentin has been accepted towards fulfillment of the requirements for Ph . D . degree in Educatign 2 [.22. Major professor November 10, 1986 Date MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771 MSU LIBRARIES m. ht RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. EFFECTS OF SELF-QUESTIONING GUIDED BY BLOOM'S TAXONOMY ON THE READING COMPREHENSION OF SCIENCE MATERIALS BY Nora Josefina Sabater Valentin A DISSERTATION Submitted to Michigan State University in partial fullfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Educational Psychology and Special Education 1986 Copyright by NORA JOSEFINA SABATER VALENTIN 1986 ABSTRACT EFFECTS OF SELF-QUESTIONING GUIDED BY BLOOM'S TAXONOMY on THE READING COMPREHENSION OF SCIENCE MATERIALS by Nora Josefina Sabater Valentin The study was designed to investigate the effectiveness of the acquisition and maintenance of self-questioning technique, structured from using Bloom's Taxonomy in the reading comprehension of science materials by poor comprehenders. Poor comprehenders were selected from the seventh grade of an urban middle school and assigned to three groups. Four hundred to five hundred word passages of science material different from the one used in class by these students were selected as assessment measures for this study after being screened for apprOpriate readability levels. A series of 10 questions were developed for each passage according to Bloom's Taxonomy of Intellectual Skills/Cognitive Domain. A multiple baseline design across subjects was used which consisted of four phases: Baseline, Intervention #1, Intervention #2, and Maintenance. During the four phases students were asked to read passages and answer questions, and to generate questions. They were instructed in the use of questions through a combination of modeling and direct instruction taking one question type at a time to distinguish it from other question types and practice making similar questions to be able to read a passage and answer questions following instruction. During phase four, Maintenance, students read and aswered questions independently. Analysis of results showed that the use of self-generated questions alone does not increase the level of reading comprehension but the training in question generation does, especially when trained using Bloom's Taxonomy of Intellectual Skills/Cognitive Domain. Also, the results of the measures taken during maintenance one week after the end of Intervention #2 showed that an increase in the level of comprehension was maintained over the short period of time. Further research is indicated in the use of self-generated questions structured according to Bloom's Taxonomy across settings in the regular science class, observed over a long period of time. Its use is suggested in the training of science teachers to increase overall learning and comprehension of science materials. This work is dedicated to my family. They always encouraged and supported my pursuit in learning. ACKNOWLEDGEMENTS I would like to acknowledge and thank my dissertation advisor and chairman of my committee, Dr. Annemarie Palincsar. I would like to express my gratitude to my committee members Dr. Taphy Raphael for her encouragement and support, Dr. James Gallagher for his guidance, encouragement and support in the area of science education, and Dr. Stephen Yelon for his continuous advice and support throughout my doctoral program. I would like to thank the principal, teacher, parents, and students who gave me full cOOperation in the study. I would like to thank all my teachers and professors who have led toward this goal through their confidence in my ability. I would like to thank all my friends, especially Roberto, Carol, Nayda, Carlos, Lucy, Marlio, Becky, Eduardo, Natacha, Daniel, Debbie, Ken, Shoba, Armando, Consuelo, Carmen, Marisol, Martin, and Francesita whose support, advice and faith in me has been a continued encouragement. I would like to thank Carol, Carlos and Martin for their technical assistance in the design and elaboration of the graphs. iv I would like to express my gratitude and patience to my family, all of them, my parents Angela Valentin de Sabater and Alfredo Sabater D. who always gave me support in my pursuit in learning. To my grandmother Maria de Valentin whose faith in me led me to keep confidence in my ability. To my brothers and sisters, Alfredo, Leo, Maria, Juan Carlos, Isabella, Laura, and their spouses, especially Bernardo, who assisted me to accomplish this goal. To my Aunt Lola de Gomez for her support in her letters and my godmother Maria de Hernandez. At last, very deeply to my Aunt and Uncle Amalia de Veluntini and Andres Veluntini together with my cousins for their constant support, advice, confidence and trust in me, who made it possible to accomplish this goal. II. III. Chapter TABLE OF CONTENTS One: Introduction. . . . . . . . The Problem . . . . . . . Statement of the Problem. Research Objectives . . . Potential Contributions . Chapter Chapter Summary . . . . . Two: Review of the Literature. . Questioning Technique . . Section Bloom's Section Science Section Chapter Chapter Summary . . . . . Taxonomy and High Summary . . . . . Learning and High Summary . . . . . Summary . . . . . Level Questions Level Questions Three: Methods and Procedures. . Introduction. . . . . . . Papulation and Sample Population. . . . . . . . Sample and Selection Procedures . . . . . Instruments . . . . . . . Assessment and Training Validity of Instruments Design of Study . . . . . Procedures. . . . Baseline. . . . . . . . . Intervention #1 . . . . . vi Material. . . . . 15 16 28 3O 34 34 36 36 36 36 36 38 38 42 42 44 44 45 IV. Intervention #2 . . . . . . . . . . . . . . . . . . 45 Maintenance . . . . . . . . . . . . . . . . . . . . 48 Chapter 1!: Results. . . . . . . . . . . . . . . . 49 Will self-generated questions increase the level of reading comprehension of poor readers? Baseline. . . . . . . . . . . . . . . . . . . . . . 51 Group #1. . . . . . . . . . . . . . . . . . . . . . 54 Group #2. . . . . . . . . . . . . . . . . . . . . . 54 Group #3. . . . . . . . . . . . . . . . . . . . . . 56 Intervention #1 . . . . . . . . . . . . . . . . . . 57 Group #1. . . . . . . . . . . . . . . . . . . . . . 57 Group #2. . . . . . . . . . . . . . . . . . . . 61 Group #3. . . . . . . . . . . . . . . . . . . . . . 64 Will instruction of self-generated questions using Bloom's Taxonomy increase the level of reading compre- hension of poor readers and the ability to answer high level questions? Intervention #2 . . . . . . . . . . . . . . . . . . 68 Group #1. . . . . . . . . . . . . . . . . . . . . . 68 Group #2. . . . . . . . . . . . . . . . . . . . . . 71 Group #3. . . . . . . . . . . . . . . . . . . . . . 71 Individual's performance per question type, per day of training. . . . . . . . . . . . . . . . . . . . . . 73 Will the change observed in comprehension maintain over time? Maintenance . . . . . . . . . . . . . . . . . . . . 82 Group #1. O O O O O O O O O I O O O O O O O O O O O 82 VI. Group #2. . . . . . . . . . . . . . . . . . . . . . Group #3. . . . . . . . . . . . . . . . . . . . . . Student's Performance Throughout the Study. . . . . Chapter Summary . . . . . . . . . . . . . . . . . Chapter 1: Discussion. . . . . . . . . . . . . . Self Generation of Questions and Students' Level of Reading Comprehension . . . . . . . . . . . . . . Effect of Instruction of Self Generation Questions Structured Using Bloom's Taxonomy on the Reading Comprehension of Poor Readers . . . . . . . . . . . Change of Comprehension Maintained Over Time. . . . Chapter Summary . . . . . . . . . . . . . . . . . . Chapter 11: Conclusions. . . . . . . . . . . . . . Limitations of Study. . . . . . . . . . . . . . . . Implications and Future Directions for Research Implications for Classroom Teachers . .I. . . . Chapter summary 0 O O O O O O O O O O O O O O O O 0 Appendix A Letter to Parents . . . . . . . . . . . Appendix B Results of Screening. . . . . . . . . . Appendix C Validation of Questions . . . . . . . . Appendix D Sample of Lesson Plans. . . . . . . . Appendix E Question Sample List. . . . . . . . Appendix F Sample of Passages Used and Sample questions 0 C O O I O O O O O O O I Appendix G Sample of Students' Responses Through- out the Experiment. . . . . . . . . . Appendix H Sample Test Given in School . . . . . Bibliography. . . . . . . . . . . . . . . . . . . viii 85 85 87 91 92 93 96 105 108 109 111 113 115 116 117 118 119 139 148 150 168 177 180 TABLE 10. 11. LIST OF TABLES Total number and type of question given and answered correctly per phase, per group . . . . . Report of the mean, scores and trends per subject in Baseline . . . . . . . . . . . . . Number and type of question answered correctly in Baseline . . . . . . . . . . . . . . . . . . . Report of the mean, scores and trends per subject in Intervention #1. . . . . . . . . . Number and type of questions generated by each student during Intervention #1. . . . . . . . . Number of questions generated, means and standard deviation per group in Intervention #1. . . . . . Number and type of question answered correctly in Intervention #1. . . . . . . . . . . . . . . . Ratio of number and type of question answered correctly during Intervention #1. . . . . . . . Report of the mean, scores and trends per subject in Intervention #2. . . . . . . . Number and type of question answered correctly in Intervention #2. . . . . . . . . . . . . . Ratio of number and type of question answered correctly during Intervention #2. . . . . . . ix PAGE 52 53 55 58 59 6O 62 67 69 74 75 12. 13. Report of the mean, scores and trends per subject in Maintenance. . . . . . . . Number and type of question answered correctly in Maintenance. . . . . . . . . . . 83 84 FIGURE 1. LIST OF FIGURES PAGE A comparison of Bloom's Taxonomy with other frequently used Taxonomies. . . . . . . . . . . . 24 Textbook used, units, chapters, pages selected to design passages used in the study. . . . . . . 40 Design strategy training study - Timeline . . . . 46 Percentage of comprehension, per subject, per group, per phase. . . . . . . . . . . . . . . . . SO Diagram of students' responses to understanding question type throughout Intervention #2. . . . . 76 Diagram of students' responses to application question type throughout Intervention #2. . . . . 78 Diagram of students' responses to analysis question type throughout Intervention #2. . . . . 79 Diagram of of students' responses to synthesis question type throughout Intervention #2. . . . . 81 xi CHAPTER I INTRODUCTION The acquisition of reading is a continuous challenge for those involved in the study of learning. From the key moment when a child or an adult discovers the connection between isolated printed symbols and the language he/she uses to communicate, a series of processes take place. Most of them are unobservable and are usually inferred or hypothesized based upon observable performance. The eye movement displayed over the page, the ability to interpret the graphemes seen, all in a matter of seconds to announce the names of an object or a person, and the sudden understanding of what commercial signs or a passage mean are all pieces of the complex Operation called reading. There is a diversity of Opinion as to what reading encompasses. There are those who believe that reading, like speaking and writing, is an active language process (Goodman and Goodman, 1981). Other refer to reading as a complex cognitive process that if ever understood would enable us to discover the mysteries of the human mind (Pearson and Johnson, 1978). Recent work with computer simulation of mental processes has led the cognitive psychologists to view reading as possessing the same constraints as those of human memory and problem solving. As readers in a problem solving situation we try to discover what the author means while simultaneously building meaning for ourselves. However, 1 2 these interpretations are limited by what we know (Pearson and Johnson, 1978). Regardless of how reading takes place, there are certain characteristics that seem to influence the quality of performance of those who read. Some are internal in nature such as motivation, interest, competency and reading ability which vary from individual to individual. Others are external such as the written message and the environment that surrounds the individual (Pearson and Johnson, 1978). When a difficulty arises in any one of these areas the probability of grasping the meaning of what is read is reduced. The efficient reader, when confronted with difficulty in understanding the material read, is capable in one way or another of monitoring and regulating his comprehension. In some cases, certain strategies have been introduced, suggested or observed that seem to account for the student's success in reading. However, there is insufficient information to explain what goes on inside the head of the reader when he reads, what mechanism he activates or uses to regulate and monitor what he reads, which enables him to understand and assimilate the information in a productive manner. The Problem In research, when we divert our attention from the efficient reader, the issue of reading comprehension becomes far more complicated to assess. Recent research in the field 3 of cognitive psychology has focused its attention tO the study Of reading comprehension monitoring or metacognition (Brown, 1975; Brown and DeLoache, 1978; Flavell and Wellman, 197i; Boss and Phillip, 1982; Wagoner, 1983; Brown and Palincsar, 1982). Metacognition has been defined as: "That process which is affected by person strategy and task variables. It is an executive function for competent reading which directs reader's cognitive processes as he or she strives to make sense Of incoming textual information." (Brown and Palincsar, 1982, pg. 1) "Conscious access tO one's cognitive Operation and reflexion about those of others. It's a declarative knowledge about the domain Of thinking." (Brown and Palincsar, 1982, pg. 1) "Metacognition refers to that secondary level of understanding in which a person addresses his/her own thinking or whose knowledge concerning one's own cognitive processes and products." (Flavell, 1976, pg. 232) These metacognitive processes involve two parts: 1) monitoring and 2) regulation. Monitoring consists Of judging the degree Of understanding when reading. Regulation is related to the activation Of appropriate strategies when comprehension fails to meet the desired levels (Baker, 1979). Research indicates that poor readers with comprehension difficulties when compared with normal pOpulations appear to have a lack Of awareness Of task demands (Loper, 1980). Also, they seem to be different in three general classes Of cognitive skills: strategic, metacognitive and processing efficiency (Brown and Palincsar, 1982). Some Of these deficits include spontaneous use Of various types Of attentional and mnemonic strategies (Hallahan, Kauffman and Ball, 1973; Tarver, Hallahan, Kauffman and Bale, 1976; Torgesen 197]; Torgesen and Goldman, 1977; Loper, 1980; and Hallahan and Kneedler, 1981). Also, they are deficient in various metacognitive skills such as planning, monitoring and checking (Torgesen, 1977), application Of appropriate memory strategies (Loper, 1980) or simply failure to remember academic material presented under normal conditions (Torgesen, 1977). To summarize, it seems that the lack Of spontaneous use of various attentional and mnemonic strategies affects the comprehension Of the material the poor comprehender is exposed to. A frequent characteristic is the failure to remember the material read as suggested by Torgesen. Nevertheless, it is possible to assume that the weakness in remembering is a consequence of the failure to understand, perhaps due tO the inability to focus spontaneously on the important concepts, principles or elements in a given story. Among the many strategies suggested to assist students to compensate for their comprehension difficulties is the use of questions. Teacher generated questions are commonly used in the classroom to clarify concepts and guide students in the important areas of reading. Student generated questions are less common but are also very important since they tell us what the student is focusing his attention on, whether he understood the passage or not. 5 Initial investigation supports the use of self- generated questions when reading stories (Andre and Anderson, 1978—1979). Question generating has been shown tO be particularly beneficial for poor comprehenders. It appears tO assist them monitor their comprehension of the passages read. Statement pf the Problem The ability tO adequately monitor comprehension processes while reading influences the degree of understanding of what is read. Poor comprehenders seem to be unaware Of their comprehension monitoring processes (Gardner and Reis, 1981) and therefore fail to adequately apply apprOpriate regulatory strategies (Torgesen, 1977; LOper, 1980; Hallahan, Kauffman and Ball, 1973). Self-questioning seems to be a useful strategy to assist poor readers to monitor their Own comprehension processes. Furthermore, questioning in general is a strategy that is not content specific. Thus, it may be used effectively across several content areas like science, where a certain amount of reading comprehension is required to be able to perform the experiments correctly and internalize underlying scientific principles (Porterfield, 1974; Russell, 1981). Therefore, the purpose Of this study is to investigate the effectiveness Of the instruction Of self- generated questions, structured using Bloom's Taxonomy, on the acquisition and maintenance Of comprehension Of science 6 material by poor comprehenders. Research Objectives This study has one major goal but several Objectives. The first is to determine if the instruction Of self- generated questions structured using Bloom's Taxonomy would produce differences on the acquisition and maintenance Of comprehension skill Of science material by poor comprehenders. Specifically, if differences do exist, there are several questions that need tO be answered in relation to the use Of self-generated questions: 1. Will self-generated questions increase the level of reading comprehension Of poor comprehenders? 2. Will instruction in question-generating using Bloom's Taxonomy increase the level Of reading comprehension of poor readers and the ability to answer high level questions? 3. Will the change Observed in comprehension be maintained over time? Potential Contributions More research is needed to explore the possibility Of whether training in the spontaneous use Of self-generated questions can be of assistance tO poor comprehenders in other content areas like science. Perhaps through the use of self-generated questions, the poor comprehenders may be able to acquire the necessary skills required in the comprehension of scientific materials. In general, researchers have showed strong support for the use of questions in the classroom (Cohen, 1983; Sadker and COOper, 1974), but many have observed that there is an excessive use of factual-recall type of questions (Gutzak, 1967; Durkin, 1978) not necessarily increasing reading comprehension skills of poor comprehenders. If the student is able to generate high level questions (especially analysis, synthesis, evaluation) such as those that could be provided in Bloom's Taxonomy one might speculate that students can assume control of their own comprehension, assisting them to focus on the important aspects of the material at hand not only in science materials, but also in other content areas such as history, geography, etc. Chapter Summary In this chapter, an introduction to reading comprehension was given. The purpose of this study was stated. The statement of the problem was presented within the context of a brief overview of the relevant literature. The research objectives were explained and the chapter concluded with the description of potential contributions and compensatory strategies to increase the comprehension skills of poor comprehenders when reading science materials and possibly other content areas too. The next chapter is a more extensive review of the relevant literature. CHAPTER II REVIEW OF THE LITERATURE This chapter is divided into three sections. The first section focuses on the questioning technique. The second section describes Bloom's Taxonomy and its use in the generation Of different types of questions. The third section focuses on science learning and its effect on the use of high level questions. I. Questioning Technique One of the most common and widely used strategies among teachers and students in the classroom to enhance reading comprehension is questioning. Usually the questions, whether teacher generated, textbook generated or student generated, help to gather new information, clarify concepts, analyze and evaluate situations or organize steps to follow to accomplish a given task. Also, often the student seeks reassurance that he is headed in the right direction when generating questions. Research studies have indicated that very little instruction in reading comprehension occurs in classrooms. In one study, out of 2174 minutes of reading observed, only 13.04 minutes (0.60%) concerned comprehension instruction (Durkin, 1978). In her study Durkin observed that teachers spent most of their time in comprehension assessment, that is, asking questions to students (Durkin, 1978). 10 Other studies revealed that when questions are used with students, these emphasize the recall of facts and events, but very few stress other aspects like application of a principle, inferencing information to enhance compre- hension, etc. (Durkin, 1978; Gutzak, 1967; Elias and Legenza, 1978). Also, some of the studies have used younger readers (Crowell and Hupei, 1981) where it has been observed that the ability to make inferences, draw conclusions and evaluate a situation is usually evidenced at a later age. There is no explanation as to why this is so. Therefore, teachers feel that the ability to retell the events of a given story is sufficient evidence indicating comprehension. Historically, question formulation and its use to interpret and solve situations can be traced to the great philosophers like Aristotle and Socrates, who used questions to understand the world. Writers Of all times, such as Shakespeare, Poe and Verne, formulated questions through their characters to be able to solve, judge, or raise their uncertainties about life, men, society, moral or religious issues, scientific principles, etc. Today, there are several reasons according to Marksberry, (1979) for the importance of questions by students in the classroom. These are: 1. Curriculum development is Often based on inquiry and discovery techniques and stresses the importance of children asking questions, especially high order questions. 11 2. Students' questions aid in clarifying and defining dimensions of the problems of being attacked and hypotheses being formulated, for gathering information and for assisting in the control of premature and faulty inference. 3. Having children ask questions and seek answers is believed to arouse their interest and participation in self-directed learning. 4. Through question-asking, children increase their perception and their ability to think and express ideas on both discursive and non-discursive levels (Marksberry, 1979, pg. 190). 5. Once you have learned how to ask relevant, appropriate, and substantial questions, you have learned how to learn and no one can keep you from learning whatever you want or need to know (Postman and Weingartner, 1969, pg. 23). In addition, the use Of questions, properly elaborated to raise high thinking levels, facilitates the understanding of the content material especially by poor readers who seem to have great difficulty comprehending what they read. Furthermore, questions increase the inspection time and the cognitive effort that a reader gives to what is considered relevant to his or her purposes; in this case, answering the questions (Durkin, 1981). However, the questioning technique is by no means the only strategy used as a study skill in the process of reading comprehension. There are other 12 strategies that are used as an attempt to help poor readers with their comprehension difficulties. Some of these are: l. summarizing (Taylor, 1982; Brown and Palincsar, 1982) 2. outlining 3. advance organizers 4. idea mapping 5. re-telling 6. connecting the old with the new (Wilson, 1983) 7. look-backs (Gardner and Reis, 1981) 8. problem-solving (Ross and Haynes, 1983) 9. inquiry - discovery (Maria, 1981; Porterfield, 1974) All these strategies and many others have been used in isolation or in combination to assist students in the learning process. Some of these strategies are content specific, that is, they are applicable with best results in certain types of areas like science, literature or social studies (i.e., problem solving in science, etc.). Others, like questioning are not content specific and can be used across several areas. Any social studies or science textbook will carry a section of questions in each chapter or unit of instruction. These questions are frequently used to guide students in their learning. In addition, teachers often ask student questions in relation to the material the students read for a variety of purposes, such as to evaluate the attainment of a specific concept or principle and to recall facts. On occasion the students ask questions to 13 clarify ideas or to compare their preconceived notions of a given event with the printed material to which they are being exposed. In spite of the controversy that surrounds the area of reading comprehension and the processes that may influence it, reading comprehension can be safely defined as the "interaction between the resources of the reader, and the characteristics of the text" (Adams and Collins, 1977; Rumelhart, 197]). Thus, it is important to examine the teachers' and the students' use of questions in the classroom and their relationship to reading comprehension. The first thing the teacher examined was the type of questions the students could generate that would provide the best results for enhancing comprehension of content material. It has been mentioned previously that most Of the questions usually observed in classrooms are those that promote lower levels of thinking activity, like the recall of facts or sequence Of events (Gutzak, 1967). In order to be able to fully understand the content material, it is necessary that higher levels of questions are generated. Some examples are those that promote analysis, comparison, evaluation and interpretation, and those that facilitate association of ideas and cause-effect relationships. Second, content area material like science requires that the student be able to problem-solve situations presented to him. When questions are generated to promote higher levels of thinking they also enable the student to solve problems related to his content area. Research 14 indicates that a significant increase in comprehension can be noticed when high level questions are encouraged (Sadker and COOper, 1974; Redfeld and Rousseau, 1981; Newton, 1978; Rhoades, 1980) in the classroom. Sadker and COOper (1974) trained eight elementary school students of whom four received micro-teaching in higher order questioning. The authors considered the following categories as representative of higher order questioning: evaluation, comparison, problem-solving, and cause-effect divergent questions. A reinforcement procedure was later introduced to compare the effect of reinforcement with use of the questioning techniques. Results indicated that trained students responded to both higher order questioning alone and high order questioning with reinforcement. However, it was questioned whether training alone or with reinforcement caused increase in question asking. Redfeld and Rousseau (1981) investigated the relationship between the level of teacher questioning and student achievement. They reviewed 20 studies on teachers' use of higher cognitive questioning. Fourteen studies out of the 20 were selected for the study since many did not fulfill requirements Of internal validity. Authors found out that regardless of the type of study the predominant use of higher order questions has a positive effect on students' achievement (Redfeld and Rousseau, 1981). As far as guidelines for question elaboration, Newton (1978) suggested Bloom's Taxonomy as one alternative way to promote high level 15 questions while enhancing comprehension in the classroom. Rhoades (1980) provided a series of guidelines to instruct teachers in questioning strategies to promote generation of higher level questions. Third, the possibility of students formulating or generating their own questions is an issue of recent concern by researchers in the area to determine potential beneficial effects on reading comprehension. There is initial support for the use of self-generated questions and investigation seems to favor its use (Andre and Anderson, 1978-1979; Frase and Schwartz, 1975). Andre and Anderson (1978-1979) and Frase and Schwartz (1975) investigated to determine whether or not generating good comprehension questions while studying prose material was an effective study technique. Both found similar positive results with both high school students and college freshmen. However, due to the type of material used in the experiment most of the questions generated were related to knowledge of facts and not higher ordered knowledge. Nevertheless, further research is needed to determine the usefulness of self-generated questions. Section Summary To summarize, the use of questions has been discussed. Several types of questions have been described according to their source. Similarly, questions have been approached according to 16 the type Of information Obtained from them (i.e., factual, recall, etc.) and issues have been raised in terms of student's ability to generate good questions, guidelines to elaborate good questions, etc. This study will focus on the student generated question as a study skill strategy to assist the poor reader to enhance his reading comprehension. Bloom's Taxonomy and High Level Questions It was conceived by Benjamin Bloom (Bloom, Engelhart, Furst, Hill and Krathwolh, 1956; Gronlund, 1978; Bloom, et. al., 1956). This is a taxonomy of educational objectives created in an attempt to provide a classification system of goals for the American educational program. As such, it has six levels including: knowledge, understanding, application, analysis, synthesis and evaluation. It was Bloom's purpose to classify intended behavior of students, specifically the ways in which students are to act, think or feel as the result of participation in some unit of instruction (Bloom, et. al., 1956). Bloom's Taxonomy was develOped after the idea that taxonomies in general are commonly used to facilitate communication among scientists and to insure accuracy in understanding the organization and interpretation of the various parts of the animal and plant world (Bloom, et. al., 1956). Therefore, the possibility of creating a taxonomy in 17 education to include the goals and objectives of American curriculum was conceived. The author foresaw several advantages to it. These are: (Gronlund, 1978; Bloom, et. al., 1956) 1. It helps to identify and define instructional objectives. 2. It provides a classification Of educational objectives analogous to the classification used for plants and animals. 3. It consists of a set of general and specific categories that encompasses all possible learning outcomes that might be expected from instruction. 4. It was developed by individuals related to the field Of education such as psychologists, teachers, and test experts to be used in curriculum develOpment, teaching and testing. 5. It's a tool to help design instruction. However, a word of caution is included here. Authors recommend that the individual should not become a slave of the taxonomy. It should serve as a guide not as a master. 6. Taxonomy should facilitate the communication exchange of information about curricular develOpment and evaluation devices among teachers, researchers, test developers, etc. 7. It's a tool in curriculum development since teachers can find a range of possible educational 18 goals or outcomes in the cognitive area like: remembering and recall, thinking, problem solving, and creating. Taxonomy can also help one to gain a perspective on the emphasis given to certain behavior by particular set of educational plans. Nevertheless, authors considered several assumptions in the elaboration of the taxonomy of educational objectives in the area (Gronlund, l. of cognitive domain. These assumptions are: 1978; Bloom, et. al., 1956). Neutrality The taxonomy is intended to maintain neutrality in relation to educational principles and philosOphies. Hierarchy The taxonomy is based on the principle that it‘ should be an educational logical psychological classification system. There are six levels - knowledge, understanding, application, analysis, synthesis, and evaluation. These categories are ranked in hierarchical order from lowest, simplest behavior to the highest, most complex behavior. In addition, each category is assumed to include the behavior of previous lower levels. Psychological Classification System The system is based on the assumption that learning outcomes can be best described in terms The respect 19 Of changes of students' behavior. Actually, the author's purpose was to classify intended behavior of students, the ways in which students are to think, act or feel as a result of participating in some unit of instruction. authors Of the taxonomy are very specific with to the behaviors students should exhibit at each level. For example: (Bloom, 1956) (Gronlund, 1978, pg. 26) Level Assumed Behaviors knowledge To remember by recall of facts or reorganization of information in a form very close to that in which it was originally encountered. Sometimes filed or stored in memory. Role Learning. comprehension Behavior that represents an understanding of literal message contained in communi- cation. Three types of comprehension are con- sidered: -Translation -- Individual can put com- munication into another's language. -Interpretation -- Communication involves reordering of ideas into a new configura- tion in the mind of the individual, importance of idea, and establishment of application analysis synthesis evaluation 20 interrelationships. -Extrapolation -- Making prediction on estimates. Being able to demonstrate an abstraction as asked to do by the teacher. TO apply something requires comprehen- sion. Being able to use an abstraction without having to be shown how to use it in that situation provided. Emphasizes breakdown of material into their constituent parts. Detection of relationships of the parts and Of the way they are organized. Techniques and devices used to convey meaning or to establish the conclusion of communication. Putting together parts, pieces into a whole. Making judgements about the value for some purpose Of ideas, works, solutions, methods materials, etc. It involves the appraisal in which particulars are accurate, effective, economical or satisfying judgements are qualitative or quantitative. Placed at the end since it requires the combination of the previous behaviors. 21 However although placed at the end, this is not the last of stages. Some other behaviors may involve evaluation. Some of the categories, especially the latter ones, increase in difficulty when the authors try to explain what behaviors to expect, how the student is supposed to act, think or feel. This may affect the precision needed to evaluate students in a given category of the taxonomy. Thus, recent research has speculated on the validity of the assumptions made by Bloom, et. al. on the taxonomy. Furst (1981), Hill and McGaw (1981), Andre (1979), Miller, et. a1. (1979) and Furst (1981) reviewed the philosOphical assumptions of the taxonomy specifically the issues of neutrality, comprehensiveness and cumulative hierarchical structure. He indicated that when neutrality was defined it meant "impartiality in relation to the source, such as educational philosophy of education and with respect to the relative worth of the goals." He also suggested that Bloom, et al. ruled out those goals that did not reflect student intended behaviors or could not be described as changes of that type. Therefore, he concluded that the issue of neutrality was violated since ”classifications tend to throw emphasis on certain qualities and, in turn, to diminish the apparent significance of other qualities" (Furst, 1981, pg. 442). The assumption of comprehensiveness was discussed by Furst (1981). He claimed that since an educational 22 objective was described as changes in behavior and ways of acting, thinking, and feeling, it is difficult to include and describe all educational Objectives in terms of change of behaviors. He suggests that "many processes and states" are unobservable. Furthermore, goals that cannot be specified in behavioral terms are unfortunately neglected (Furst, 1981). Also, categories like "understanding” do not seem to adequately satisfy some researchers. Ormell (1974, 1979) argues that the behaviors to be observed here are not sufficient to define all the characteristics and information related to comprehension. The principle of hierarchy has been investigated by several researchers (Miller, et. al., 1979; Hill and McGaw, 1981; Furst, 1981). This has been attacked strongly by investigators who appear to disagree with what seems to be the central core to this taxonomy. Furthermore, some have submitted the hierarchical structure of the taxonomy to experimentation and complex statistical analysis to investigate the validity of this assumption (Furst, 1981; Miller, et. al., 1979). From the philosOphical standpoint, when Ormell (1974) tried to apply the taxonomy, he found contradictions in the frequent inversion of various Objectives and tasks. He fOund that certain demands on knowledge are more complex than certain demands for analysis or evaluation. He solved them by setting levels within each category. Others elaborated arguments about the definitions and skills 23 assumed for each category (Furst, 1981; Miller, et. al., 1979). They felt that definitions are restrictive and sometimes one category early in the hierarchy may include others which are classified at the end of it (Miller, et al., 1979). For instance, Orlandi (1971) implied that meaningful comprehension in social sciences must be accompanied or preceeded by analysis. However, Bloom, et a1. (1956) acknowledge that it was not possible to make distinctions as clear-cut as one would like among the categories (1956, pg. 15, 144-145). Another group has investigated the construct validity of the hierarchy from the statistical point of view (Miller, et al., 1971; Hill and McGaw, 1981), applying a series of analytic methods. Hill and McGaw (1981) found that when the knowledge category was deleted the hierarchy assumption of Bloom's Taxonomy is supported. There are many other classifications or categories upon which questions could be elaborated like Sanders (1966), Pearson and Johnson (1978), Barrett (1967). However, most of these are derivations, adaptations or combinations of the same original source -- Bloom's Taxonomy (Figure 1). Therefore, the cognitive domain of Bloom's Taxonomy was selected for this study for the following reasons: 1. Convenience To teach poor comprehenders to generate good questions while reading science materials, it is necessary to provide them with an adequate sample that can meet the instructional Figure 1 A comparison of Bloom's Taxonomy With Other Frequently Used Taxonomies Bloom Barrett Sander Pearson é Johnson Knowledge Literal Meaning Memory Textually Explicit -recognition -recognition -recall -recall Understanding Inferences Translation -translation -cause and effect -interpretation -interpretation -like and differences -extrapolation Application Application Textually Implicit -abstraction no solution Analysis Analysis -break into pieces Synthesis Synthesis ~put together EvaluatiOn Evaluation Evaluation Appreciation Scriptically Implicit 24 25 goals of science classes and aid the students to increase their understanding of the scientific processes, as they read about them. Also, if successful, as a consequence of learning to generate their own questions and being able to comprehend science better, the students would be able to apply in given situations the scientific concepts just learned. In addition, the taxonomy specifies clearly what behaviors are expected from the students as a consequence of a unit of instruction. Therefore, the student's performance in relation to science materials can be measured in terms Of the behaviors established for each of the six levels. Moreover, instructions regarding the most effective procedures in the instruction of science include: discovery learning, mastery learning, problem solving and deductive learning (Burton, 1983). Thus, Bloom's Taxonomy would allow this investigator to expose and train the poor comprehenders in how to think analytically, to interpret information, to apply concepts learned, summarize information, or give an evaluative judgement. Also, students would be able to generalize scientific principles through the elaboration of questions according to the behavioral guidelines set for each level of the taxonomy. In addition, the use of questions derived from Bloom's Taxonomy may provide both the researcher and the student With a means to meet the needs of the poor comprehender. 26 2. Familiarity The researcher also selected Bloom's Taxonomy for reasons of familiarity with the design of regular classroom activities. It is a clear, precise, organized way of planning, and evaluating the content of instruction in a given course or subject. Additionally, another assumption of this taxonomy claimed by the author is that of generality; it can be used across subject matters, curriculum areas (Bloom, et. al., 1956). Due to this acquaintance with the taxonomy, this author thought of the possibility of using it to guide the model questions to train the poor comprehenders. 3. Variety Science reading materials in general require the ability to master several skills that are content specific such as the ability to: interpret a given scientific conceptual issue, apply it in an experimental situation, analyze a situation, summarize a fact, generalize and evaluate scientific principles, demonstrate comprehension of a given concept by facilitating an example, etc. The levels of the taxonomy allow for the production of a number Of different questions that appear to satisfy those specific study skills required in science. In particular, the last four levels, application, analysis, synthesis and evaluation seem to suggest that questions elaborated this way would help students elicit the behaviors necessary to develop those skills indispensable to succeed in science. 27 4. Completeness Another feature of the taxonomy relies on its completeness when compared with other classifications. Bloom's Taxonomy contains more areas and various modes, example of behaviors and/or thinking skills. Each area is very detailed in relation as to what behaviors are expected, how to elicit them, how to evaluate them, etc. Very little literature is available in regarding the use of Bloom's Taxonomy to develop questions to assist students in the classroom (Elijah and Legenza, 1978; Masland, 1978-1979; Newton, 1978). There is even less information about the use of Bloom's Taxonomy to train students in self—questioning to assist them in the comprehension of science material. Due to the limited amount of research done in the area, the use of the taxonomy to assist students in the self- generation of questions with science materials was considered a unique Opportunity to develop the critical thinking skills required to understand science materials. In addition, the taxonomy provides a technique for use beyond the remedial reading classes and into regular classroom activities like science. 5. Originality Bloom's Taxonomy was described point in detail in order for the reader to understand its unique applicability in the design of questions, when compared to other taxonomies so that evaluation of this taxonomy can be made and usefulness 28 with science reading materials can be determined. Section Summary To summarize, there are several taxonomies used to assist teachers, test makers, and supervisors design instruction. But analysis of the taxonomies suggests that most Of them are variations of Bloom's Taxonomy of Educational Objectives Cognitive Domain. Therefore, this taxonomy was selected to design the questions to aid the poor comprehenders in the understanding of science material. The taxonomy offers completeness, familiarity, convenience, variety and originality. Research already indicates that the use of questions in the classroom has positive effects on achievement (Cohen, 1983; Sadker, and Cooper, 1987). Bloom's Taxonomy Offers us an organized efficient way of elaborating questions to be able to assist the poor comprehenders to grasp the content of the science material they read. Still, very little information is available in relation to the use and effectiveness of questions made with Bloom's or any other's taxonomy (Elijah and Legenza, 1978; Masland, 1978-1979; Newton, 1978; Andre, 1979) to assist students in the classroom. Newton (1978) suggested that the role of school is to provide Opportunities to develop critical reading skills in students as intelligent beings. Newton suggested the use of Bloom's Taxonomy to elaborate questions to help students raise their cognitive thinking abilities, 29 whenever emphasis is made on questions that tap the last three categories of the taxonomy: analysis, synthesis and evaluation. Newton considered that these processes involve higher cognitive thinking. She indicated that several advantages can be observed when teachers ask students questions which are included in those three categories. Higher cognitive questioning can assist students develOp critical reading skills since the student “not only literally comprehends the printed page but interacts critically as he reads its contents" (Newton, 1978, pg. 27). Also, the student becomes an "active participant in the reading situation" promoting discussions and enabling the student to use questions as an instrument to formulate evaluative judgements (Newton, 1978, pt. 27). Andre (1979) reviews a series of studies in which it was shown that high level questions defined as those concerned with analysis, evaluation in Bloom's Taxonomy have a positive effect on learning. However, the conditions under which they occur is not very well understood yet by researchers. Elijah and Legenza (1978) and Masland (1979) discussed the existence of several taxonomies, including Bloom's, in their effort to help teachers ask students questions over the material they read. Both prOposed and develOped their own taxonomies and suggest their use to enhance reading comprehension. Elijah and Legenza (1978) specifically outline theirs in detail and provide suggestions for teachers to use in the preparation of their questions. Similarly Duffelmeyer (1980) advises the 30 use of Barrett's taxonomy to promote reading comprehension beyond literal level. Yet none of them has examined scientifically the effect of a given taxonomy used to develop questions on student's learning. Furthermore, none of them has studied the effects of self questioning using Bloom's Taxonomy in the comprehension of science material of poor comprehenders. It would seem that self-questioning techniques using Bloom's taxonomy promote the development of questions which are compatible with the problem solving discovery methods frequently used in science if we emphasize the production of high level questions, that is analysis, synthesis and evaluation. In addition, it would provide poor comprehenders with an efficient tool which they can use to control and monitor their comprehension as active readers which can be self- rewarding and perhaps lead to a positive attitude towards reading. Encouraging student generated questions using Bloom's Taxonomy may allow poor comprehenders and their science teacher the opportunity to confront scientific principles in a manner that poor comprehenders can understand; in a way that permits them to identify the relationship of ideas and evaluative judgements necessary to engage efficiently in laboratory experiences. Science Learning and High Level Questions Science instruction tries to determine those teaching practices that lead to effective learning of the underlying 31 principle within scientific knowledge. It's a field constantly evolving due to the advancement of science therefore science educators and publishers alike try to adjust, modify their curriculum to incorporate new updated material and scientific discoveries. Several methods are used in science to assist the students in the understanding and application of a variety of scientific principles such as mastery learning, meaningful learning which includes the use of problem solving strategies, discovery, learning, links to prior knowledge, etc. (Burton, 1983). Similarly, there are several techniques used to increase the meaningfulness of scientific knowledge like the use of concrete analogies, notetaking, and summarizing advance organizers, conceptual cues, and improving the content of science textbooks. (Burton, 1983). It would seem that the existence of such a variety of methods and techniques would guarantee the success of science instruction. However, this does not appear to be the case. A crisis seems to be a more likely description of the state of the art in science instruction (Burton, 1983, pg. 293). While it is not the intention of the author to discuss the state of the art in science instruction and the factors that may affect science curriculum, instead it is the purpose of this investigation to provide the poor comprehender with a study skill, as a tool that will assist 32 him in the understanding of science material and serve as a bridge between the general process of comprehension and the thinking skills demands of the science curriculum. Recent research indicates initial support for the use of questions to enhance reading comprehension in science. Ackerman (1981) studied the use of several forms of adjunct questions in science, to establish encoding by visual imagery and and encoding by semantic stimulation to test whether either coding enhanced performance. Results indicated that there were no differences or main effects but that students who scored higher on visual imagery scored significantly better on criterion tests here adjunct questions had partial diagrams, than did those students whose questions had verbal coding only. Andre and Anderson (1978-1979) investigated the effects of student generated questions on how it facilitates learning from text. They designed two experiments. Experiment #1: 29 student seniors were exposed to two conditions, questioning training and no questioning. Results indicated that the question generation strategy seems to work better for low ability students. Experiment #2: 81 juniors and seniors were placed into three groups, questioning and training, questioning and not training, and no question with three different types of passage requirements. Results showed that those students involved with questioning and training and questioning without training performed better than those with no training in 33 questioning. Although the information is limited it seems to encourage positive results towards the use of self generation Of questions to enhance comprehension of science materials. Andre and Anderson (1978-1979) point out several explanations for the beneficial effects of self questioning. 1. Accordingly input is analyzed in a hierarchy of processing stages where increasing depth implies greater degree of semantic or cognitive analysis, hence greater retention. 2. Improved retention of textual material by questioned groups is simply a function of extended study time. 3. The combination of metacognitive and cognitive characteristics frequent in reading (promoting comprehension, etc.) (Andre and Anderson, 1978- 1979, pg. 620). Lastly there is the study by Porterfield (1974) where he investigated the effects of teacher questioning students in science with the use of high level questions (comprehension, analysis, evaluation). He found that those trained in (SCIS) method of inquiry discovery science instructional approach using questions asked more questions than those who were not trained in it. Yet, little information was available in terms of the student's level of achievement, to account for a possible relationship between questioning technique, science materials and student 34 achievement. Section Summary In summary, the use of questions in the science content area seems to have initial positive results, especially with low ability students (Andre and Anderson, 1978-1979; Porterfield, 1974). However, little information is available and there is need for further research to determine the usefulness Of techniques with students in science content area. Yet, some initial advantages cited by Andre and Anderson (1978-1979) indicate that the self-questioning technique encourages the readers to: a. Set purposes in reading. b. Identify important segments of the material. c. Generate questions which require comprehension of the text to be correctly answered. d. Think of possible answers to the questions. e. Actively monitor the learning activity to achieve efficiency (Andre and Anderson, 1978-1979, pg. 620). Chapter Summary This chapter was a review of literature related to the present study. The first part of research focused on the study of the importance of questioning as a technique to facilitate comprehension. The second part focused on 35 Bloom's Taxonomy and its use in the generation of different types of questions. The last section focused on science learning and the effect of high level questions on learning. CHAPTER III METHODS AND PROCEDURES Introduction This chapter on methods and procedures consists of five sections. The first section identifies the pOpulation and the sample and describes how they were selected for the study. The second section presents the design employed and addresses the issues of internal and external validity. The third section includes a presentation of the instruments used to measure the students' reading comprehension and the instructional strategy used to assist the students in reading, as well as a discussion of the validity of the instruments selected. The fourth section includes the procedures used in data collection. The last section describes the analysis of procedures. Papulation and Sample Population The pOpulation for this study was poor readers in seventh and eighth grade of a public middle school in the Lansing School District. Sample and Selection Procedures The sample consisted Of 11 students nominated by the remedial reading teacher from seventh and eighth grades. The students were chosen from a develOpmental reading class from 36 37 a selected middle school in the Lansing School District area. All subjects volunteered to participate in the study. The investigator supplied consent forms to the students' parents (Appendix A). The principal from the middle school in Lansing was contacted to request the use of their students and teachers for participation in the study. After the study was approved by the Lansing School District Office of Evaluation and the Human Subjects Committee at Michigan State University, the teachers were contacted again to explain the extent of the students' involvement in the study. The 11 students selected for this study were adequate decoders but poor comprehenders. The students were screened to determine that they were indeed adequate decoders and to assess their reading comprehension abilities. To determine students' ability to decode and comprehend all nominees were asked to read silently a 400-word passage assessed to be written at a seventh grade level according to Fry Readability Formula (Fry, 1977). The students were then asked to answer in writing 10 comprehension questions. Following the comprehension questions, the students were asked to read the passage aloud to ascertain correct and incorrect reading rates. Students who achieved a minimum of 80 w.p.m. correct and a maximum of 2 w.p.m. incorrect (Lovitt and Hansen, 1976) and a score of 50% or below on the comprehension assessment were eligible to participate in the study. The 11 subjects were randomly selected from a pool 38 of eligible students (Appendix B). Following this screening, the 11 students who met the criteria were selected from a pool of eligible students. There were seven boys and four girls assigned to three groups. The students were assigned to three different groups on the basis of scheduling needs. The schedule needed to allow the researcher to meet at least four students at one time two to three times per week. Therefore, Group #1 had three boys and one girl and Group #2 had two boys and two girls. Group #3 had two boys and one girl. The cultural and ethnic background of these children were the same except for the fact that all but one had Spanish surnames. One had a native American Indian surname. Nine students were hispanic, one student was black, and one student was native American Indian. In Group #1 were Beatriz (S ), Humberto (S ), Vicente _ l 2 (S ), and Federico (S ). In Group #2 were Carol (8 ), 3 4 5 Antonio (8 ), Jeronimo (S ), and Laura (S ). In Group #3 6 7 8 were Alfredo (S ), Roberto (S ) and Adelaide (S ). 9 10 11 Instruments Assessment and Trainipg Material Passages used for training and assessment were taken from material different than the students used in their science classes. Six chapters were taken from Life Science (Webster, et. al., 1980) to develop the assessment materials for the study. The topics were cells, vascular plants with 39 seeds, vertebrates, invertebrates, new directions in life science and seed plants. The passages were screened for appropriate readability level according to (Fry 1977) readability formula (Figure 2). Also, a series of 10 questions were developed for each passage according to Bloom's Taxonomy of Intellectual Skills/Cognitive Domain. The questions developed covered most levels of the hierarchy with the exception of the level of evaluation. The questions were elaborated with strong emphasis on application, analysis, synthesis to be used in all the conditions baseline and experimental as well as in the classroom generalization probes. A total of 186 passages were used in this study divided into 2 for Screening, 3, 5, 7 for Baseline, 3 for Intervention #1, 8 for Intervention #2, 3 for Maintenance per subject, per group. However, there were several problems encountered by the researcher in the develOpment of the assessment/training materials used in the study. First, there was the selection of the science topics to be used. The selections were influenced by familiarity with subject matter by the researcher, since the researcher was not a Science Education major. Therefore, those tOpics that were well known to the researcher were selected. Second, although the length of the passages was established to be between 400—500 words, on many occasions it was necessary to adjust the length to insure that the presentation of a concept was not negatively affected. 40 Life Science - Webster et a1 - Prentice Hall, Inc., 1980 Unit Unit Unit Unit Unit Unit Unit Unit Chapter I Chapter I Chapter I Chapter I Chapter III Chapter III Chapter VII Chapter Units, chapters Description 1 Vascular Plants with Seeds (pg. 52-61) 1 Some Flowering Plant Families 2 Invertebrate Animals (pg. 67-94) 3 Vertebrate Animals (pg. 97-124) 7 Cells (P8. 207-222) 8 Seed Plants (pg. 225-248) 18 New Directions in Life Science (pg. 469-490) Textbook Used and tOpics selected to design the passages used in the study. (a) Figure 2 a) except for reading concerning laboratory experiments 41 Therefore, in some instances, passages were extended to 550 words, with no more than 2-3 concepts introduced per passage. Third, the writing of the questions according to Bloom's Taxonomy area of Cognitive Domain posed problems due to the nature of the content in some passages used and the style in which they were written (EX: descriptive vs. inquiry or discussion). In many cases it became very difficult to produce some of the question types. For most cases it was very easy to make factual-recall questions, knowledge and understanding questions. On the other hand, it was extremely difficult to make application, analysis, and synthesis questions. However, there were always 10 questions always containing at least one application, analysis, and synthesis question with the remainder being knowledge and understanding questions. Therefore, in each passage there were approximately four knowledge, three understanding question types and at least one application, one analysis and one synthesis question. Fourth, on other occasions, due to the nature of the tOpic (e.g., cells), it was necessary to add pictures to the reading selection. Otherwise it was very difficult for the reader to understand the science tOpic without the additional laboratory experience or discussion. In addition, pictures were necessary in many cases to formulate question types like application and analysis. Passages were adapted and questions constructed. Both 42 were typed separately on letter size paper double spaced so that no relationship or familiarity with the textbook was made by the student. Passages were administered randomly so that no two students had the same passage at one time. Validity pf Instruments The validation of the instruments used in this study was achieved by having one independent individual familiar with Bloom's Taxonomy rate the questions to confirm that they indeed represented the different levels of the Taxonomy (see Appendix C). In addition, a reliability check was used where a pool Of questions of different types was elaborated according to the taxonomy chosen. Two individuals rated them again according to their type. There was 90% reliability agreement. Then those questions disagreed on were rewritten to conform to the types. Design pf the Study The study was designed to investigate the effectiveness of the instruction Of self-generated questions structured using Bloom's Taxonomy on the acquisition and maintenance of comprehension skills of science material by poor comprehenders. The design selected for this study to investigate the effectiveness of this strategy was a multiple baseline design across subjects (Gay, 1981; Kratochwill, 1978). There were four phases: Baseline, Intervention #1, 43 Intervention #2 and Maintenance. Baseline lasted for 3, 5, or 7 days for students in Group #1, Group #2 or Group #3, respectively. Intervention #1 lasted for 3 days for all individuals in all three groups. Intervention #2 lasted for 8 days for all individuals in all three groups, and Maintenance lasted 3 days for all students in the groups. The design and timeline for the study are presented in Figure 3. The research questions this study was designed to address include the following: 1. Will self-generated questions increase the level of reading comprehension of poor comprehenders? 2. Will instruction in question-generating using Bloom's Taxonomy increase the level Of reading comprehension of poor readers and the ability to answer high level questions? 3. Will the change observed in comprehension maintain over time? There are several dependent variables to these research questions: 1. The number of comprehension questions answered correctly in the passages read independently in experimental setting. 2. The number of high level questions generated by students. 3. Maintenance measures; that is, comprehension questions answered correctly some time later after the end of the study. 44 Procedures The study was carried out by the investigator with the cOOperation of the developmental reading and science teachers. The students selected were informed that they were going to do some reading activities that would help them understand and learn science material better and faster. They were told that these activities were varied. At specific times (i.e., Baseline, Intervention #1, Intervention #2, and Maintenance) the students were given an explanation about the activity they were to do. In Baseline they were told that in order for the researcher to assist them in learning science material better, she needed to observe if they knew how to read well and that it would be reflected on how accurately they answered the questions given later. In Intervention #1 the students were told that it was important to know if they knew how to ask good questions about their readings. In Intervention #2 the students were told that the researcher would instruct them in the different question types so that they would assist them to understand science material and learn it better and faster. In Maintenance students were told that the researcher needed to Observe how they would apply instruction on the question types independently. Baseline Baseline measures were taken in the experimental setting immediately following subject selection. The 45 student was told to ask for assistance with any word that he/she did not understand. Upon completion of the passage, the student was given 10 comprehension questions based on Bloom's Taxonomy. The student was praised for correct responses. The daily measures taken throughout the study and referred to as assessment passages were administered in the same manner. As was mentioned before in the subject selection process there were 11 students selected from a pool of eligible students who were divided into three groups (Figure 3). Intervention #1 During Intervention #1 the student received a passage to read silently. The student was asked to generate questions about the passage and write them down on a piece of paper as he/she read. Afterwards the student received 10 comprehension questions to answer about the passage read (Figure 3). Intervention #2 The student was instructed in the use of questions through a combination of both modeling and direct instruction taking one question type at a time. First, the investigation discussed the use Of self-questioning in reading and its importance to enhance comprehension of science materials. Consequently, the student was informed 46 DESIGN: STRATEGY TRAINING STUDY - TIMELINE BASELINE QUESTION INSTRUCT. QUEST. MAINTENANCE GENERATION GENERATION H Group # 3 days 3 days 8 days 3 days I CORRECT OF COMPREHENSION N Group # .--q --T -0- 3 days.‘ 3 days 8 days 3 days 1 CORRECT OF COMPREHENSION O '1 O C '0 Q h.) 7 days 3 days 8 days 3 days I CORRECT OF COMPREHENSION SCHOOL DAYS Figure 3 47 of the different types of questions that could be made according to Bloom's Taxonomy and the different types of information they provided the student with. The researcher selected one question type (i.e., knowledge or factual recall) and explained to the student its importance and the kind of information he/she was able to obtain by formulating this type of question. Next, the student distinguished the questions that emphasized the category of knowledge in Bloom's Taxonomy (i.e., recall of facts) from other types of questions in the taxonomy from a list given to him, with the help of the researcher. Then, the researcher modeled how to make a knowledge type questions. Then, the student practiced making questions. Finally, the student was given a passage to read silently. Afterwards, the student received 10 comprehension questions to answer in writing (Appendix D). Emphasis on question generation was made on the level of application, analysis, synthesis types since these were the ones believed by the investigator to be most useful in assisting the student to understand the issue involved in the reading of science materials. There was one passage given per student, per day. Each was used for assessment purposes. The same procedure was used for the rest of the categories in the Bloom's Taxonomy. :0 v 48 There was no criteria established to achieve mastery of each question type, rather the time allowed each question type was controlled; one day for knowledge, one day for understanding, and two days for each of the rest of the categories of questions application, analysis, synthesis. When the student was exposed to all question types possible according to Bloom's Taxonomy area of cognitive domain, the student entered the maintenance phase (Figure 3). Maintenance The maintenance phase was similar to Baseline. The student read a 4004500 word passage silently. Upon completion of the passage, the student was given 10 comprehension questions based on Bloom's Taxonomy to answer on a separate piece of paper (Figure 3). Throughout all of these phases students were given global feedback on their performance encouraging them to do better. Chapter Summary This chapter was a description of the methodology and procedures used in this study. A description Of the population and subjects was provided. A discussion of the materials was introduced. At last, a description of the design selected and procedures followed in the study. CHAPTER IV RESULTS This chapter focuses on the results of the teaching strategy of self questioning. The analysis of the four phases of the study, Baseline, Intervention #1, Intervention #2, and Maintenance, is reported. Within each phase the means and general trends are described as they relate to the three research questions previously stated. The means and general trends within and between phases are described, followed by a discussion of group differences in the percent of accuracy with the questions comprehension, ability to generate questions, and type of questions generated and answered. The number of students selected for the study was 11. All but one completed the study. Student NO. 11 was dropped out of the study, due to lack of motivation and cooperation, and overall indifference towards the research (Figure 4). There were three research questions: 1) Will self-generated questions increase the level of reading comprehension of poor readers? 2) Will instruction of questioning using Bloom's Taxonomy, increase the level of reading comprehension of poor readers and the ability to answer high level questions? 3) Will the change observed in comprehension maintain over time? 49 50 j- Illl In I'llfdlli I'H'l m><= fizz—cm q 1 o \0 Kb. .IndcmmI/wm.\. ...GW.IQ.I.\\G 3 i...u.t........+. «ml :0“. ~ . I.-. ...... I . .. I I J u q u u u 1 c 1 q 1 q a u a q u ....l. .a1 a. \h/ \x... r- . .\ : . . a. .\ x... A.) .Vo. \ / \NH/O M... .c u . / \ V\. I . ./. . .. .... D D .9 . a \ HYI. ..\ PHI: ...h\ s... . .. ,.... I. a s ..\\\VQ. “(Tirm \\....woof ..... - m? .C o .I sat. 5d. .o. . ”—6.22...— h... —<—. 2 2:53.252: aha-:9...— .one== so; .szscw so; .uuu.::o he; :c.z:o:aa:5=u _— ==—._.=:>=x._.=_ .O eucd=oucsa ...I.....II..I: II .... c x x 1w. /.R \.r« .wll m:..-wmuuuaN No._ No.N Nn.m NIN Nn.a NN.on um NS.N NG.N~ Nos NoN noN non use a a a. an aN Ia NN SN an on on an em oNI con so oNI oNN N. coau=u>uouau no._ No._ N_.N N0.m NN.N NN.nI NN._ NN.G NN.G NoI uoI noI non Nos n m c_ SN A— Is a oN NN as as as was «NI cos osI ocN oNN uaNNouum n N N m N I m z< m< a x N 3ao mzoNNmmaa mm<=m >AHumx¢ou Gmxmzmz< amuse zoNamuac axe—amuse so .02 zw>No amass chNmmao no .02 aaoun Non manna use Aunoouuou venomous can co>uu cecauaoao «0 scan was non-a: usuOH ~ UHAOH Table 2 Report of the mean, scores and trends per subject in Baseline. Student x Mean Scores Trend Group #1 Subject #1 53.3 35, 60, 65 high, i t 22.9 accelerating Subject #2 30.0 40, 10, 40 low, unstable Subject #3 8.3 10, 10, 5 stable, low Subject #4 O O, 0, O stable, low Group #2 Subject #5 28 10, 30, 40, stable i - 19.97 15, 45 Subject #6 13 10, 30, 0, variable 15, 10 Subject #7 10 0, 40, O, stable, low 0, 10 Subject #8 6 0, 20, 0, stable, low 0, 10 Group #3 Subject #9 22.1 30, O, 0, variable i - 44.95 10, 15, 70, 30 Subject #10 32.8 45, 10, 0, variable 0, 50, 65, 54 Afterwards, the subject had to answer from memory 10 comprehension questions given to him/her about the passage read. This phase lasted 3, 5, or 7 days for each subject depending whether in Group #1, Group #2 or Group #3. 91222.}. The mean percent correct on the comprehension measures in Group #1 was R a 22.9 and scores ranged from 0% to 65% (Table 2). The trend for S and S was variable, with a high positive trend for S . 1The trfind for S and S was low and stable (Figure 4). 1 During Baseline, Group #I received a total of 120 questions, 10 per passage, per student, which included at least one of each type according to Bloom's Taxonomy. A total of 32 (26.6%) questions were answered correctly (Table 3). There were 7 (21.8%) questions knowledge type and 15 (46.8%) questions understanding type, I (21.8%) questions application type, 2 (6.2%) questions analysis type and 1 (3.1%) question synthesis type. That is out of 32 questions answered correctly, 22 (68.6%) of them were knowledge and understanding type. 93.2223. The mean percent correct on the measures for this group was i = 14.25 and scores ranged from 0% to 40% (Table 2). SS, 87, S8 had a low and consistent trend. The trend for 86 was low but variable (Figure 4). Group #2 received a total of 200 questions. A total of 20 (10%) questions were 55 Table 3 BASELINE Number and type of questions answered correctly in Baseline NO. OF UESTIONS CORRECT GROUP STUDENTS R U AP AN 3 TOTALS #1 Subject #1 4 8 2 2 1 17 (56.6%) Total NO. Subject #2 3 6 2 0 0 11 (34.3%) of Questions Subject #3 0 1 2 0 0 3 (9.3%) 120 32.26.6% Subject #4 O 0 1 0 0 1 (3%) Suthtal 7 15 7 2 1 32 % 21.8 46.8 21.& 6.3 3.1 #2 Subject #5 4 5 0 1 O 10 (5%) Total NO. Subject #6 0 1 2 1 1 5 (25%) of Questions Subject #7 2 1 O 0 0 3 (15%) 200 20310% Subject #8 0 2 0 0 0 2 (10%) Subtotal 6 9 2 2 1 20 % 30 45 10 10 5 #3 Subject #9 4 3 3 1 2 13 (35%) Total No. of Subject #10 4 11 3 3 3 24 (64.86%)Questionm 140 Subtotal 8 14 6 4 5 37 37-26.42% % 21.6 37.8 16.2 10.8 13.5 56 answered correctly (Table 3). There were 6 (30%) questions knowledge type, 9 (45%) questions understanding type, 2 (10%) questions application type, 2 (10%) questions analysis type, 1 (5%) question synthesis type. That is out Of 20 questions answered correctly 15 (75%) of them were knowledge and understanding types. §£°_“RL The mean percent of accuracy on comprehension measures for this group was E - 27.4 and scores ranged from 0% to 70% (Table 2). The trend for S and S was variable but still low (Figure 4). Group #3 rezeived iototal of 140 questions. A total of 37 questions were answered correctly (Table 3). There were 8 (21.6%) questions knowledge type and 14 (37.8%) questions understanding type, 6 (16.2%) questions application type, 4 (10.8%) questions analysis type, 5 (13.5%) questions synthesis type. That is that out of 37 questions answered correctly 22 (59.4%) of them were knowledge and understanding types. The difference in time allocated for Baseline for all individuals in all groups does not seem to cause any noticeable difference in performance. Regardless of amount of days spent in Baseline the overall level of comprehension remained low as expected except for subject 9 and 10. S exhibited one day where he demonstrated his level 0: comprehension to be unusually high (70%) and S 0 showed two 1 days where he demonstrated his level of comprehension to be slightly high (65% and 60%). The degree of difficulty of the 57 content of previous passages as Opposed to the ones given on these occasions may account for the student's difference in performance. Intervention #1 During this phase the subjects received 3 passages each, one per student per day, to be read silently. The subjects were asked to generate questions about each passage read and write them down on paper as he/she read. Afterwards the subject had to answer from memory 10 comprehension questions given to him/her about the passage read. This phase lasted three days for all subjects in all three groups. The results of this phase per student per group were as follows: In Group #1 The mean percent of accuracy on comprehension measures for the subjects in this group was i = 29.7 and scores ranged from 10.6% to 56.6% (Table 4). The trend for S and S was variable and the trend for S was low and coisistenz; S was high and consistent abov: 50% level. In Group 11 szbjects generated a total of 42 questions, that is an average 10.5 questions per subject (Table 5 and Table 6). These ranged from 7 to 16 questions. Out of a total of 42 questions, there were 38 (90.4) knowledge type, 1 (2.3%) comprehension type, 2 (4.6%) application type, 1 (2.3%) analysis type. The relationship between the number of questions generated and percent of comprehension for subjects in Group Table 4 Report of the mean, scores and trends per subject in Intervention #1 Student 3 Mean Scores Trend Group #1 Subject #1 15 0, 30, 15 variable i - 29.7 Subject #2 56.6 50, 65, 55 stable, high Subject #3 10.6 0, 10, 22 stable, positive accelerated Subject #4 36.6 0, 75, 40 variable Group #2 Subject #5 8.3 5, S, 5 stable i = 19.97 Subject #6 30.0 0, 50, 40 stable Subject #7 16.6 45, 5, O decreasing, stable Subject #8 25.0 10, 45, 20 stable Group #3 Subject #9 56.6 30, 40, 100 accelerating i - 44.95 Subject #10 33.3 30, 10, 60 accelerating 59 S.S . I u N .NSS .SaoscN NNSS S u S S .aso .Smsuos NNSN.SS S . z< SN.NN u u o NN.a .N .azo .NN .uou> NNSS.NS S u N< SN .aSo .NNS auz NN.SN NS.SS NS.SS NNNN.SS S n a S.oN u I S S .aao .NN SSSS SNN SN SS NS NNSN.NSS SSN n u SN SSSuoSoN N .aso SNNSO S N N NSNSN S S S SNN IN NSNSS S .N SS SS SS N o S S oNNSSsss NNNSS S .N SN SS S N N S S S S SNNNooS NNNSS .N .S S Nxmth oz~¢=n hzunahm zo€ent too much time and attention to small pieces of reading to make questions rather than read the material to answer t”Tie. questions that followed it. Group #1 received a total of 120 questions, 4 students, 3 Passages each, 10 questions per passage, per student, and 98011 had at least one of each type according to Bloom's Takenomy. A total of 40 (32.5%) questions were answered correctly, 13 (32.5%) were knowledge type questions, 16 (402) understanding type questions, 6 (15%) application t137133 quesrions, 3 (7.5%) analysis type questions, 2 (51) 8yuthesis type questions. That is out of 40 questions ansWered correctly 29 (72.5%) of them were knowledge and undarstanding question types (Table 7). 1:1 Group _#_2_ the average percent of accuracy on 62 Table 7 INTERVENTION #1 Number and type of questions answered correctly in Intervention #1 NO. OF QUESTIONS CORRECT ROUP STUDENTS K U AP AN TOTALS if 1L Subject #1 1 4 1 0 5 (16.6%) Total No. Subject #2 6 9 3 3 22 (55%) of Questions Subject #3 3 1 1 0 5 (12.5%) 120 40-33.3% Subject #4 3 3 1 0 8 (26.6%) Subtotal 13 16. . 6 3 40 % 32.5 40 15 7.5 #2 Subject #5 2 O 1 O 3 (13%) Total No. Subject #6 1 2 2 1 6 (26%) of Questions Subject #7 1 3 1 0 5 (21.7%) 120 ’ 23-19.1% Subject #8 1 2 4 1 9 (39.1%) Subtotal 5 6 8 2 23 % 21.7 26 34.7 8.6 8. #73 Subject #9 6 7 1 2 17 (62.9%)Total No. of Subject #10 4 14 1 1 10 (37%) Question 60 Subtotal 10 11 2 3 27 37a45% I % 37 40. 7.4 11.1 5. an LI 511 Pt 5t 0f 95c 63 comprehension measures for the subjects in this group was SE = 19.97 and scores ranged from 8.3% to 30.0% (Table 4). The trend for S and S was low and consistent, the trend for S was variaisale and the trend for S was average an: c o n sistent. 6 In Group #1 the students generated a total of 55 que stions (Table 5). There were 47 (85.4%) questions knowledge type, 4 (7.2%) comprehension type and 4 (7.2%) analysis type. That was an average of 12.75 questions per subject (Table 6). These ranged from 3 to 25 questions. The relationship between the number of questions generated and the percent of comprehension for subjects in GrOup #2 was as follows: S , S and S generated 24, 3, and 5 questions respectively (35‘: = I2.75, 7SD - 12.31) (Table 5 and 6). The percent of comprehension was low but consistent for all of them and S generated 25 questions and percent of Comprehension was viriable. Similarly in Group #2 the higher the number of questions generated the lower the Pel’cent of comprehension for S and S . Also, the less amOunt of questions generate: the higher the percent of c0"1p1'ehension for S and S . Again, it was observed that 8t“dents spent tooémuch time and attention to small pieces of information in the passage to make questions rather than reading to be able to answer the questions that followed it. Group #1 received a total of 120 questions, 4 students, 3 Pilssages each, 10 questions per passage, per student and each had at least one of each type according to Bloom's 64 Taxonomy. Twenty-three questions were answered correctly (19.1%) (Table 7). There were 5 (21.7%) knowledge type questions, 6 (26%) understanding type questions, 8 (34.7%) application type questions, 2 (8.6%) analysis type que stions, 2 (8.6%) synthesis type questions. That is out of 23 questions answered correctly 11 (47.7%) were knowledge and understanding type. In Group #1 the average percent of accuracy on c omprehension measures was if - 44.95 (Table 4). The trend for S and S was variable. In Group #1 the students 9 10 generated a total of 19 questions (Table 5). There were 18 (94.72) questions knowledge type and 1 (5.22.) comprehension question. That was an average of 9.5 questions per subject (Table 6). These ranged from 5 to 14 q Ue stions. The relationship between the number of questions generated and the percent of comprehension for subjects in GI‘Oup #3 was as follows: S and S generated 5 and 15 9 10 questions respectively (5: = 9.5, SD - 4.5) (Table 5). The Percent of comprehension was variable for both S and S . 9 10 However, the percent of comprehension for S was above 50% 9 comprehension (if = 56.6) (Table 4). Similarly, in Group #3 the higher the number of questions generated the lower Percent of comprehension for S and the less the amount of 10 clueStions generated the higher the percent comprehension for S . 9 Group #3 received a total of 60 questions, 2 students, 65 3 passages each, 10 questions per passage, per student and each had at least one of each type according to Bloom's Taxonomy. Twenty-seven questions were answered correctly (45%) (Table 10). There were 10 (37%) knowledge type que stions, 11 (40.7%) understanding type questions, 2 (7.4%) application type questions, 3 (11.1%) analysis type que stions, 1 (5.8%) synthesis type questions. That is that out: of 27 questions answered correctly 21 (77.7%) of them we re knowledge and understanding type. The trends observed in Intervention #1 indicated that there is not a positive relationship between the number of questions generated and the percent of comprehension exhibited at this point; on the contrary a general low Performance below 50% comprehension was observed for all s‘-1b;_‘|ects that ranged from i= 19.97 to 44.95 and was Consistent with Baseline. The higher the number of questions generated the lower the comprehension exhibited exCept for S , S and S where comprehension increased but 2 6 9 the number of questions generated was low. It was noticed on all three groups that students spent too much time and attention to small pieces of reading to write questions and did not read to answer the questions which may account for the low comprehension exhibited. Thus, the self-generation of questions alone does not improve poor readers' accuracy ansWering comprehension of questions except when results are °°mpared with those of Baseline then, only a small moderate 1mPl‘ovement was observed. Also, it seems that the students 66 (1C! not appear to have a specific strategy-study skill to assist them in understanding what is read. In addition it was found that students can generate questions about the pa 8 sage but out of a total of 118 questions generated, 103 (’£3 7' .28%) were knowledge type, followed by 8 (6.77%) comprehension type, 3 (2.54%) application type and 5 (4.23%) aLt).£1.lysis type. There were no synthesis questions generated (,7I?za.ble 6). That is 94.05% of all questions generated were e 1 ther knowledge or comprehension type. Very few students ‘vrea:tre able to generate questions beyond this level. Also, the ratio of the number and type of question answered correctly, F’Gel? student each day of Intervention #1 indicated that most s‘ijects knew how to answer knowledge and understanding question types (Table 8). Similarly, out of a total of 300 questions given to G 9 1 G2 9 G 91 of them were answered correctly (30.3%) (Table 3 1) - There were 28 (30.7%) knowledge type, followed by 33 (36 .2%) comprehension type, 16 (17.5%) application type, 8 (8 - 7%) analysis type, 5 (5.4%) synthesis type. Again out of 91 questions answered correctly 61 of them were knowledge and comprehension type (67%). These factual recall question tYpes are the ones students are most accustomed to answering (Durkin, 1978). 2' Will instruction of self~generated questions using Bloom's Taxonomy increase the level of reading comprehension of poor readers and the ability to answer high level questions? 67 Table 8 Ratio of number and type of question answered correctly during Intervention #1 S 1 Knowledge Ijiaa.y 1 0/4 Ijlza.y 2 0/4 ‘I)u£1y 3 1/4 7 Understanding T[)aay 1 0/3 ‘1)1ay 2 2/3 13:3y 3 1/3 6 Anglication 13£1y 1 0/1 13£1y 2 1/1 I3£iy 3 0/1 3 _Analysis 13£1y 1 0/1 Day 2 0/2 Day 3 0/1 1 .13 1”lthesis Day 1 0/1 Day 2 0/1 Day 3 0/1 1 S S 2 3 2/3 2/4 2/3 0/3 2/4 1/4 out of 10 2/4 1/3 4/4 0/4 3/3 0/3 out of 10 1/1 1/1 1/1 0/1 1/1 0/1 out of 10 1/1 0/1 1/1 0/1 1/1 0/1 out of 10 0/1 0/1 0/1 0/1 0/1 0/1 out of 10 S S S S S 4 5 6 7 8 0/4 1/3 0/4 1/4 2/3 3/4 1/4 2/4 0/2 1/3 1/2 1/3 1/4 0/2 0/4 answered correctly (70%) 0/3 0/2 0/3 0/3 0/3 3/3 0/3 1/3 0/4 1/4 1/4 0/3 1/3 0/3 1/3 answered correctly (60%) 0/1 0/2 0/1 0/1 1/2 1/1 1/1 1/1 0/2 1/1 0/2 0/1 1/1 0/2 1/1 answered correctly (30%) 0/1 0/1 0/1 0/1 0/1 0/1 0/1 1/1 0/1 1/1 0/1 0/2 0/1 0/1 0/1 answered correctly (10%) 0/1 0/2 0/1 0/1 0/1 1/1 0/1 0/1 0/1 1/1 1/1 0/1 0/1 0/2 0/1 answered correctly (10%) 1/4 2/4 3/3 2/3 2/3 3/3 0/1 0/1 1/1 0/1 0/1 1/2 0/1 0/1 1/1 10 1/3 1/4 2/4 2/3 0/1 2/3 0/1 0/1 1/1 0/2 0/2 1/1 0/1 0/2 0/1 68 The results of the data collected during Intervention IFZ were as follows: I n t: ervention fl During this phase the students were instructed in the di f ferent kinds of question types that they could learn to geherate when they read to assist them to comprehend better the passages they read. The training period lasted eight days for all three groups. All students were instructed in each question type f or two days, with the exception of knowledge and comprehension types for which there was one training day. The results of both Baseline and Intervention #1 indicated tfilial: the greater amount of questions answered correctly were kflowledge and comprehension types. In addition, most of the questions generated in Intervention 771 were knowledge and cOInprehension type. Therefore, it was not necessary to tra in the students in these two question types for more than a day, since they were already familiar with them in their re8111ar classroom work. In Group 11, the mean percent accuracy of comprehension mefigures was if = 52.9% and ranged from 0% to 100% (Table 9). The trend for S was positive, accelerated and was highly vat‘ileable and theltrend for S was positive and stable. Group £1 received a totgl of 320 questions 4 students, 8 Passages each, 10 questions each per passage, per student to answer, which included at least one of each question type trcc>up #1 =- 52.9 lrc>up #2 " 69.0 ErOUp #3 x “ 85.6 69 Table 9 Report of the mean, per subject in Intervention #2 Student Subject Subject Subject Subject Subject Subject Subject Subject Subject Subject #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 scores and trends 3 Mean 38.1 70.0 37.5 66.25 74.3 66.2 63.7 71.8 88.1 83.1 Scores 0. 20, 85, 7o, 40, 9o, 30, 25, 40, 20, 100, 90, 20, 9o, 85, 20, 100, 75, 10, 100, 85, 10, 100, 100, 100, 100, 60, 55, 65, 100, 10, 40, so 60, 100 65, 9o, 80 1o, 75, 55 3o, 75, 9o, 30, 25, 100 50, 75, 9o 60, 9o, 60 0. 7o, 55, 100, 35, 1o, 55, 100, 70, 100, 65 90 100, 20, so, 65, 90, 100, 55, 100, 55, 90, 100 90, 100 100, 100, 33:14 accelerating variable, delayed highly, variable upward, stable upward, variable variable upward variable upward positive upward stable upward stable positive upward 70 Table 10 Intervention #2 Number and type of questions answered correctly in Intervention #2 NO. OF QUESTIONS CORRECT GROU STUDENTS K U AP AN 8 TOTALS #1 ubject #1 9 6 4 2 3 24 (30%) Total No. ubject #2 25 13 8 6 4 56 (34%) of Questions Subject #3 15 16 4 3 4 42 (25.6%) 320 164=51.2% Subject #4 15 14 4 4 5 42 (52.%) Subtotal 64 49 20 15 16 164 % 39 29.7 12.1 9.1 9.7 #2 Subject #5 10 15 4 3 3 35 (18.6%) Total No. Subject #6 19 9 5 6 3 42 (22.3%) of Questions Subject #7 14 16 4 9 4 47 (25%) 320 188-58.7% Subject #8 25 14 12 7 6 64 (34%) Subtotal 68 54 43 25 16 188 % 36 28.7 22.8 13.2 8.5 #3 Subject #9 24 25 7 7 8 71(48.6%) Total No. of Subject #10 22 25 8 11 9 75(51.36%)Questionfi 160 Subtotal 46 50 15 18 17 146 146=91.2% % 31. 34.2 10.2 12.3 11.6 71 according to Bloom's Taxonomy. A total of 164 (51.2%) questions were answered correctly (Table 10). There were 64 (39%) knowledge type questions, 49 (29.8%) comprehension type questions, 20 (12.1%) application type questions, 15 (9.1%) analysis type questions and 16 (9.7%) synthesis type questions (Table 10). In Eggpp #2 the mean percent accuracy of comprehension measures was i = 69.0% and scores ranged from 10% to 100%. The trend for S was upward, variable but consistent. The trend for S wa: variable, upward. The trend for S and S was high, agcelerated, stable and consistent (Table 9). 8 Epppp 12 received 320 questions to answer. A total of 188 (58.7%) questions were answered correctly. There were 68 (36%) knowledge type questions, 54 (28.7%) comprehension type questions, 43 (22.8%) application type questions, 25 (13.2%) analysis type questions, and 16 (8.5%) synthesis type questions (Table 10). Eggpp #3 had a mean percent accuracy of comprehension measures of i = 85.6 and scores ranged from 55% to 100%. The trend for S and S was positive, accelerating, upward and consistent Zr stablg (Table 9). Gpppp 1; received 160 questions to answer, 2 students 8 passages each, 10 questions each per passage, per student which included at least one of each question type according to Bloom's Taxonomy. A total of 146 (91.2%) questions were answered correctly. There were 46 (31.5%) knowledge type questions, 50 (34.2%) comprehension type questions, 15 72 (10.2%) application type questions, 19 (12.3%) analysis type questions, and 17 (11.6%) synthesis type questions (Table 10). Overall, mean percent correct during Intervention #2 was above 50% comprehension and shows a marked improvement when compared to the mean percent of comprehension in the previous phase (Table 6, Table 9). Intervention £1 Intervention #2 G i - 29.7 G x B 52.9 1 1 1 2 G i - 19.97 G i - 69.0 2 1 2 2 G x = 44.95 G x = 85.6 3 1 3 2 Also, the number of questions answered correctly markedly increased when compared to Intervention #1 (Table 1). Intervention #1 Intervention £2 Total No. of Questions Given Total No. of Questions Given 300 800 No. of Questions Correct No. of Questions Correct 90 498 30.0% 62.25% In addition, the number and percent of question types answered correctly moderately increased during Intervention #2 when compared with Intervention #1 (Table 1). Question Type Intervention #1 Intervention #2 Knowledge 28 (9.3%) 178 (22.25%) Understanding 33 (11%) 153 (19.1%) Application 16 (5.3%) 60 (7.5%) Analysis 8 (2.6%) 58 (7.2%) Synthesis 5 (1.6%) 49 (6.1%) TOTAL 90 (30.0%) 498 (62.25%) 73 Also, the performance of each subject per question type per day of training was as follows (Table 11). Knowledge Z Day 1 In Group #1 S , S , S answered 3/4, 1/4, 1/3 knowledge 2 3 4 type questions correctly. S - 0/4. 1 In Group #2 S , S , S , S answered 1/4, 2/4, 1/2, 2/4 5 6 7 8 knowledge type questions correctly. In Group #3 S answered 3/3 knowledge type questions 9 correctly. S = 0/4. 10 In Group #1, Group #2 and Group #3 S , S , S , S , S , 2 6 7 8 9 that is 5 out of 10 subjects were able to answer knowledge type questions correctly consistently. In addition, 8 , S and S were able to answer at least one knowledge queztio: type cirrectly. Thus, as anticipated most of these subjects knew how to answer factual recall questions correctly (Table 11). Understanding : Day 2 In Group #1 S and S answered 1/2, 3/4 understanding type questions coriectly.3 S = 0/2 and S - 0/2. In Group #2 S , S t S answergd 2/3, 1/3, 3/3 understanding type qqution: corZectly. S - 0/2. In Group #3 89, $10 answered 4/4, 2/3 understanding type questions correctly. In Group #1, Group #2 and Group #3 S , S , S , S , S , 2 3 5 7 9 S that is 6 out of 10 subjects were able to answer 10 understanding type questions correctly consistently. In 74 ~\~ «\N _\— n\n nxn ~\~ ~\— ~\_ ~\~,m\e ~\~ ~\— —\~ nxu n\~ _\_ ~\~ ~\¢ n\n n\o m ~\— m\o “\o ~\N n\n ~\~ ~\~.~\~ «\q n\n ~\~ ~\— ~\~ «\c n\n _\~ ~\~ —\~ «\q nxn am ~\~ ~\_ ~\~ n\m axe _\— ~\~ ~\~ Nx— m\n ~\c ~\o ~\~ «\c axe .\c ~\c ~\— nxm «\N am dxu n\n wa ~\~ «\N ~\c ~\c uxo «\c exu ~\~ Nxo ~\~ mxm M\c Nxc «\c «\c «\c ~\_ hm dx— d\— _\~ m\~ «\c —\~ ~\~ ~\~ nxu exe ~\o ~\c ~\~ n\~ «\d axe uxc ~\o n\c «\w om _\~ ~\~ ~\~ mxn ~\o ~\c ~\o ~\~ n\n ~\~ ~\o uxo ~\o nxm «\~ ~\_ ~\~ ~\c ~\c «\g mm _\o ~\~ _\~ n\~ n\n a\e a\c «\N nxn n\n «\c «\o ~\c «\o ~\_ _\o _\c ~\o n\~ n\_ em _\o Nxo _\c nx~ nx~ ~\c «\o ~\c nxn nx— ~\~ —\a ~\~ «\n ~\~ ~\~ ~\_ ~\— n\c ex. mm a\c «\c _\~.~\o e\m a\o a\a ~\a n\~ e\n ~\o [\o ~\a «\a e\n _\c _\o ~\a n\n e\n Nm ~\c ~\c _\c «\o m\c ~\o ~\o _\_ axe exa _\: uxe ~\_ n\= N‘s ex: _\= _\_ me: ex: ea m z< ac = x m \z< e< = a m z< a< = x m- z< a< = x wasp a< a< = x coaueoao q n N _ axes Na zouhzm>¢mhz~ UZH¢=Q NAHummzoo amxmzmz< mzcuhmmac no mmxh nz< amazaz mo omh