<%BANNER%>

An Evaluation of Familial Involvements' Influence on Student Achievement in K-12 Virtual Schooling

Permanent Link: http://ufdc.ufl.edu/UFE0024208/00001

Material Information

Title: An Evaluation of Familial Involvements' Influence on Student Achievement in K-12 Virtual Schooling
Physical Description: 1 online resource (166 p.)
Language: english
Creator: Black, Erik
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: achievement, distance, education, family, k12, online, survey
Teaching and Learning -- Dissertations, Academic -- UF
Genre: Curriculum and Instruction (ISC) thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Virtual schooling is fast becoming a mainstream option for today?s generation of learners. Greater numbers of students and parents are demanding access to distance education opportunities, increasing the amount of public funding allocated to. While virtual schooling at the K-12 level has grown in popularity, research-based investigations into successful teaching, learning and student support processes are limited. There are guidelines and standards for instructional practices in online settings, mainly produced by leading organizations in teaching and learning, including The American Federation of Teachers, Sloan-C, American Distance Education Council, Southern Regional Educational Board , National Education Association, and the North American Council for Online Learning. However, most guidelines remain non-empirical adaptations of face-to-face practices, whose primary focus is on guidelines for course development and pedagogical practice. There is reason to believe that the role of familial involvement in virtual schooling could be as important, if not more important than its role in traditional schooling. Unfortunately, contemporary studies on parental involvement only address face-to-face student populations. Therefore, the study of parental involvement and its relationship to virtual school student achievement could assist the development of new communication strategies between virtual schools, teachers and parents that will lead to improvement in student achievement. The purpose of this study is to investigate the role of familial participation in student's achievement in K-12 virtual schools. To address this question, this study employed an online survey adapted from research by Hoover-Dempsey and Sandler, sampling parents from a virtual school in the Southeastern U.S. Quantitative statistical procedures were utilized to analyze the resulting data. Outcomes indicate that parent-student interactions do have predictive effect related to student achievement, parents and students differ in their perceptions of each other?s involvement in the academic process, and demographic variables such as gender and socioeconomic status affect the level of parental involvement perceived by students. Implications related to these findings can be used to increase the effectiveness of homeschool communications, develop comprehensive communications policies for virtual school employees, develop and deliver instructional materials for parents in order to promote efficacious interaction with their children regarding academic work and develop a broader understanding of the ancillary environment associated with virtual school students.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Erik Black.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Ferdig, Richard E.
Local: Co-adviser: Dawson, Kara M.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024208:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024208/00001

Material Information

Title: An Evaluation of Familial Involvements' Influence on Student Achievement in K-12 Virtual Schooling
Physical Description: 1 online resource (166 p.)
Language: english
Creator: Black, Erik
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: achievement, distance, education, family, k12, online, survey
Teaching and Learning -- Dissertations, Academic -- UF
Genre: Curriculum and Instruction (ISC) thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Virtual schooling is fast becoming a mainstream option for today?s generation of learners. Greater numbers of students and parents are demanding access to distance education opportunities, increasing the amount of public funding allocated to. While virtual schooling at the K-12 level has grown in popularity, research-based investigations into successful teaching, learning and student support processes are limited. There are guidelines and standards for instructional practices in online settings, mainly produced by leading organizations in teaching and learning, including The American Federation of Teachers, Sloan-C, American Distance Education Council, Southern Regional Educational Board , National Education Association, and the North American Council for Online Learning. However, most guidelines remain non-empirical adaptations of face-to-face practices, whose primary focus is on guidelines for course development and pedagogical practice. There is reason to believe that the role of familial involvement in virtual schooling could be as important, if not more important than its role in traditional schooling. Unfortunately, contemporary studies on parental involvement only address face-to-face student populations. Therefore, the study of parental involvement and its relationship to virtual school student achievement could assist the development of new communication strategies between virtual schools, teachers and parents that will lead to improvement in student achievement. The purpose of this study is to investigate the role of familial participation in student's achievement in K-12 virtual schools. To address this question, this study employed an online survey adapted from research by Hoover-Dempsey and Sandler, sampling parents from a virtual school in the Southeastern U.S. Quantitative statistical procedures were utilized to analyze the resulting data. Outcomes indicate that parent-student interactions do have predictive effect related to student achievement, parents and students differ in their perceptions of each other?s involvement in the academic process, and demographic variables such as gender and socioeconomic status affect the level of parental involvement perceived by students. Implications related to these findings can be used to increase the effectiveness of homeschool communications, develop comprehensive communications policies for virtual school employees, develop and deliver instructional materials for parents in order to promote efficacious interaction with their children regarding academic work and develop a broader understanding of the ancillary environment associated with virtual school students.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Erik Black.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Ferdig, Richard E.
Local: Co-adviser: Dawson, Kara M.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024208:00001


This item has the following downloads:


Full Text

PAGE 1

AN EVALUATION OF FAMILIAL INVOL VEMENTS INFLUENCE ON STUDENT ACHIEVEMENT IN K-12 VIRTUAL SCHOOLING By ERIK W. BLACK A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Erik W. Black 2

PAGE 3

To my wife Nicole 3

PAGE 4

ACKNOWLEDGMENTS The completion of this dissertation signifies the end of a very l ong and unlikely journey begun on September 10, 2001. Throughout my travels (physical, emotional and cognitive) I have been assisted by family, friends and of course by my academic faculty. Ultimately, this experience would not have been possible had it not been for the love and support of my wife, Nicole. She has been more than just a partner, but a mentor, role model and one who recognized and encouraged my true potential from our fi rst meeting. Throughout our relationship she has served as an inspiration, show ing me that one could truly find happiness and fulfillment in a career. During our six years of marriage it has been her who has sacrificed to enable my aspirations, growth and development. In answer to a question constantly asked, not just by her, but by many others, I adapt a quote from Steve Prefontaine, To give anything less than [my] best would be to sa crifice [her] gift. Dr. Rick Ferdig has served as my primary mentor during the la st four years. His unautocratic management style, in terdisciplinary focus, entrepreneurial spirit and dedication to research has both promoted and accelerated my ach ievement and served as an inspiration. I find comfort in the knowledge that a fe llow academic late bloomer can not just achieve, but become a leader at a national university. His philosophies related to doctor al study and managing a hectic and diverse workload have proven to be invalu able. Few advisors would allow their doctoral students the freedom that I have been gran ted, my educational experience has surpassed expectations because of his trust. Dr. Kara Dawson has served as a second, but no less important, mentor during my tenure at the University of Florida. She provided an introduction to constructivism, served as a benchmark for teaching excellence at the university level, and a willing mentor in the writing process. Dr. Dawson has always made time to lend advice, support, and her professional opinion. 4

PAGE 5

She provides a exemplary model of someone who succe ssfully balances a role as a leader in our field and a family. Dr. Cathy Cavanaugh, Dr. David Miller and Dr. Diane Yendol-Hoppey have each provided exceptional support and guidance as in structors and members of my dissertation committee. Dr. Cavanaughs status as one of, if not the, foremost researcher in K-12 virtual schooling has inspired both confidence and vigilan ce, considering she has e ither written or edited a significant number of the studies and papers re ferenced in this dissertation. Since joining my committee, Dr. Cavanaugh has served as an e fficient and meticulous advisor and mentor. As a member of a program with many gradua te students and too few faculty, the words thank you are not heard often enough by Drs. Ferdig, Dawson and Cavanaugh. They have each left such an indelible and profound imprint upon who I am and who I will become, that those same words, thank you, do not do them justice. In addition to my committee, I have been fortunate to have an amazing cohort of doctoral students who have supported my growth. Meredith DiPietro deserves special thanks for her guidance, encouragement and organizational prowess. Finally, this work would not have been po ssible without the assi stance of the AT&T Foundation, who has encouraged the development and growth of state led vi rtual schools in the U.S. Kim Mulkey of Integrity Consulting, who has act ed as an invaluable external liaison for the AT&T Virtual School Clearinghouse. And the Un iversity of Florida Graduate Alumni Fellowship Program, which has provided four y ears of generous funding, enabling my studies. I wish to thank the alumni and friends of the Un iversity of Florida who have graciously provided graduate students with the opportunity for advancement. 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4LIST OF TABLES................................................................................................................. ..........9LIST OF FIGURES.......................................................................................................................11ABSTRACT...................................................................................................................................12 CHAPTER 1 INTRODUCTION................................................................................................................. .14Background.............................................................................................................................14Defining Family Involvement.................................................................................................16The Role of Familial Involvement in Traditional Schooling..........................................17The Role of Familial Involvement in Virtual Schooling.................................................18Models of Parental Involvement.....................................................................................20Problem Statement.............................................................................................................. ....22Significance................................................................................................................... .........22Purpose...................................................................................................................................23Research Questions............................................................................................................. ....232 REVIEW OF LITERATURE.................................................................................................24Introduction................................................................................................................... ..........24An Overview of Virtual Schooling in the US.........................................................................25K-12 Virtual Schools and Student Achievement....................................................................29Family Involvement................................................................................................................30Family Involvement in Virtual Schools..................................................................................32The Hoover-Dempsey and Sandler Model of Family Involvement.......................................34Conclusion..............................................................................................................................383 METHODOLOGY.................................................................................................................4 5Introduction................................................................................................................... ..........45Research Overview.............................................................................................................. ...45Research Design..............................................................................................................45Population..................................................................................................................... ...46Privacy and Confidentiality.............................................................................................47Data Integrity and Security..............................................................................................49Power Calculation and Sample Size................................................................................50Sampling Procedure.........................................................................................................51 6

PAGE 7

Instrumentation................................................................................................................ .......51Achievemen t Data...........................................................................................................52Revision and Proposed Path Model.................................................................................52Data Collection and Analysis.................................................................................................53Data Analysis Techniques Employed.....................................................................................54Limitations.................................................................................................................... ..........564 RESULTS...................................................................................................................... .........60Introduction................................................................................................................... ..........60Sample....................................................................................................................................60Parent Group A: All Parent Respondents........................................................................61Parent Group B: Parents W hose Child Did Not Respond...............................................62Parent Group C: Parents Whose Child Did Respond......................................................62Student Group..................................................................................................................64A Summary of Survey Factors and Survey Reliability...................................................65Research Question 1............................................................................................................ ...67Research Question 2............................................................................................................ ...71Research Question 3............................................................................................................ ...72Interaction 1.....................................................................................................................74Interaction 2.....................................................................................................................75Interaction 3.....................................................................................................................75Interaction 4.....................................................................................................................75Main Effect......................................................................................................................75Summary of Findings............................................................................................................ .765 DISCUSSION AND IMPLICATIONS..................................................................................98Introduction................................................................................................................... ..........98Research Question 1............................................................................................................ ...99Parental Instructions Negative Rela tionship With Student Achievement....................102Parental Reinforcements Positive Rela tionship With Student Achievement...............104Implications Related to Research Question 1................................................................104Research Question 2............................................................................................................ .107Research Question 3............................................................................................................ .112Derivative Outcomes............................................................................................................ 116Outcomes Related to the Survey...................................................................................116Outcomes Related to the Study Population...................................................................117Implications Related to Derivative Outcomes...............................................................120Broad Outcomes Associat ed With This Study.....................................................................123Additional Framework for Research: Understa nding the Variability Associated with Virtual Schools in the United States.................................................................................127Conclusion............................................................................................................................128 APPENDIX A IRB AND INFORMED CONSENT.....................................................................................133 7

PAGE 8

B VISUALIZING THE PARENTAL INTERACTION AND OUTLIER EFFECT ACROSS ALL FOURD STUDY VARIABLES..................................................................134Interaction 1..........................................................................................................................134Interaction 2..........................................................................................................................135Interaction 3..........................................................................................................................135Interaction 4..........................................................................................................................136C VISUALIZING THE STUDENT INTERAC TION AND OUTLIER EFFECT ACROSS ALL FOUR STUDY VARIABLES.....................................................................................138Interaction 1..........................................................................................................................138Interaction 2..........................................................................................................................139Interaction 3..........................................................................................................................139Interaction 4..........................................................................................................................140D PARENT SURVEY..............................................................................................................14 2Demography.........................................................................................................................142School Valence.....................................................................................................................144Variable A Parental Report of Encouragement..................................................................145Variable B Parental Report of Modeling............................................................................146Variable C Parental Report of Reinforcement...................................................................147Variable D Parental Report of Instruction..........................................................................147E STUDENT SURVEY...........................................................................................................149Variable A Student Report of Pa rents Use of Encouragement.........................................149Variable B Student Report of Parents Use of Modeling...................................................150Variable C Student Report of Pa rents Use of Reinforcement...........................................151Variable D Student Report of Parent's Use of Instruction..................................................152LIST OF REFERENCES.............................................................................................................153BIOGRAPHICAL SKETCH.......................................................................................................166 8

PAGE 9

LIST OF TABLES Table page 2-1 Success factors in K-12 virtual schooling..........................................................................40 2-2 Correlations among parents reports of i nvolvement mechanisms, student perceptions of parents involvement, and a summary measure of student achievement......................43 3-1 Correlations among parents reports of i nvolvement mechanisms, student perceptions of parents involvement, and a summary measure of student achievement......................58 4-1 Parent group A demographics............................................................................................79 4-2 Parent group B demographics............................................................................................80 4-3 Parent group C demographics............................................................................................81 4-4 Student group demographics..............................................................................................82 4-5 Reliability coefficients for survey variables......................................................................83 4-6 Results of regression one...................................................................................................84 4-7 Results of regression two.................................................................................................. .85 4-8 Results of regression three................................................................................................ .86 4-9 Mean reinforcement and instruction for parent groups B & C..........................................88 4-10 Racial/ethnic data fo r parent groups B & C.......................................................................88 4-11 Wilcoxon W for parent groups B & C...............................................................................89 4-12 A summary of regression re sults, research question 1.......................................................89 4-13 Descriptive statistics, st udent group and parent group C...................................................89 4-14 Test of within subjects contra sts, student group and parent group C................................90 4-15 Correlations between parent a nd corresponding student study factors..............................92 4-16 Between subjects ANOVA: Student data with outlier.......................................................92 4-17 Between subjects ANOVA: Parent data with outlier.........................................................92 4-18 Gender by income interaction (student factors).................................................................92 4-19 Income main effect (Student instruction only)..................................................................96 9

PAGE 10

5-1 Parent respondent race/ethnicity......................................................................................131 5-2 Student respondent race/ethnicity....................................................................................131 5-3 AVS enrollment by ethnicity, Fall 05 Summer 07.......................................................131 10

PAGE 11

LIST OF FIGURES Figure page 2-1 The Hoover-Dempsey and Sandler model of parental involvement..................................412-3 Path model for the Hoover-Dempsey and Sandler model of pare ntal involvement..........443-2 Proposed path model for Hoover-Dempsey & Sandler model of parental involvement...594-1 A representation of the c onflicting results associated with research question 1................874-4 The interaction between student inst ruction and student gender and income...................944-5 The interaction between student enco uragement and student gender and income............954-6 The interaction between student rein forcement and student gender and income..............964-7 Income main effect for student instruction........................................................................97B-1 The interaction between parental modeling and parent gender and parent race/ethnicity....................................................................................................................134B-2 The interaction between parental instruction and parent gender and parent race/ethnicity....................................................................................................................135B-4 The interaction between parental rein forcement and parent gender and parent race/ethnicity....................................................................................................................137C-2 The interaction between student modeling and parent education and parent gender......139C-3 The interaction between student encour agement and parent education and parent gender...............................................................................................................................140C-4 The interaction between student reinfo rcement and parent education and parent gender...............................................................................................................................141 11

PAGE 12

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy AN EVALUATION OF FAMILIAL INVOL VEMENTS INFLUENCE ON STUDENT ACHIEVEMENT IN K-12 VIRTUAL SCHOOLING By Erik W. Black May 2009 Chair: Richard E. Ferdig Cochair: Kara Dawson Major: Curriculum and Instruction Virtual schooling is fast becoming a main stream option for to days generation of learners. Greater numbers of students and pa rents are demanding access to distance education opportunities, increasing the amount of public funding allocated to. While virtual schooling at the K-12 level has grown in popularity, research-based investigations into successful teaching, learning and student support pr ocesses are limited. There are guidelines and standards for instructional practices in online settings, main ly produced by leading organizations in teaching and learning, including The American Federatio n of Teachers, Sloan-C, American Distance Education Council, Southern Regional Educatio nal Board National Education Association, and the North American Council for Online Learni ng. However, most guidelines remain nonempirical adaptations of face-to-face practices, w hose primary focus is on guidelines for course development and pedagogical practice. There is reason to believe that the role of familial involvement in virtual schooling could be as important, if not more important than its role in traditional schooling. Unfortunately, contemporary studies on parental involvemen t only address face-to-face student populations. Therefore, the study of parental involvement and its relationship to virtual school student 12

PAGE 13

13 achievement could assist the development of new communication strate gies between virtual schools, teachers and parents that will l ead to improvement in student achievement. The purpose of this study is to investigate the role of familial participation in student's achievement in K-12 virtual schools. To addre ss this question, this study employed an online survey adapted from research by Hoover-Dempsey and Sandler, sampling parents from a virtual school in the Southeastern U.S. Quantitative statistical procedures were utilized to analyze the resulting data. Outcomes indicate that parent-student interac tions do have predictive effect related to student achievement, parents and st udents differ in their perceptions of each others involvement in the academic process, and demographic variables such as gender and socioeconomic status affect the level of parental i nvolvement perceived by students. Implications related to these findings can be used to incr ease the effectivenes s of homeschool communications, develop comprehensive communications policies for vi rtual school employees, develop and deliver instructional materials for parents in order to pr omote efficacious interaction with their children regarding academic work and develop a broade r understanding of the ancillary environment associated with virtual school students.

PAGE 14

CHAPTER 1 INTRODUCTION Background The United States (US) has experienced an unprecedented growth in both the availability and participation in online education programs. At present, there are over 3.2 million online students at the college and uni versity level (Allen & Seaman, 2007) and over 96 percent of the very largest higher educational institutions have online course offerings (Allen & Seaman, 2006). At the K-12 level, over 1,000,000 students part icipate in online learning (Watson, Gemin & Ryan, 2008); by 2010, this number is expected to double (Tucker, 2007). At the end of 2002, only 12 states had significant online learning pr ograms (Kozma & Zucker, 2003). Now 44 states have taken the initiative to crea te online offerings. This number reflects the extraordinary need for online learning (Watson, Gemin & Ryan, 2008). An active body of researchers is engaged in studies to identify best practices and insight into how to best support and promote online students success. Research by Watson and Ryan (2007) re veals that virtual schools are focused on the development of successful practices in a myriad of areas, including: teacher professional development, student teacher communication, student services, data management and course development. One area of research that has not received adequate attention is the effect that parental involvement has on student achievement in virtual schools. Several virtual schools collect data from parents, utilizing th e information to ascertain an understanding of parental attit udes toward virtual schooling and curriculum. Unfortunately, there are no published studies that empi rically investigate this impact (Russell, 2004; Black, Ferdig & DiPietro, 2008). A better understanding of factors related to online student achievement is of critical importance to K-12 online education (NCREL, 2002) Parental involvement is seen as a key 14

PAGE 15

component of other non-traditi onal forms of education, including charter schools and home schooling (Green & Hoover-Dempsey, 2007; Bulk ley & Fisler, 2003). Online learning and traditional schooling have both shared and unique factors and components that affect student achievement (DiPietro, 2008; Ferdig, DiPi etro, Papanastasiou, 2005; Cavanaugh, Gillan, Kromrey, Hess, & Blomeyer, 2004; Mills, 2003; McConnell, 2000). Research by Roblyer and Marshall (2003) has identified several psychologi cal factors necessary for academic success in online environments; these factors include: technological skills, self-discipline, timemanagement skills, internal locus of contro l, self-esteem, responsibility and achievement motivation. The physical presence afforded by th e teachers and the classroom has a critical impact on the development and shaping of thes e factors (Harter, 1996) Given the lack of physical presence of the teacher inherent to onlin e learning, it remains to be determined how to best provide the support to keep online learners focused on assigned tasks. In light of the proximity issue associated with online learning, and the relative uncerta inty regarding face-toface educational practices that transfer into an online learning environment, the assumption that the effects of parental involvement on student ac hievement in online learning will be similar to those found in tradition schooling cannot be made. Family involvement has been recognized as an intangible ideal that can be connected to a large number of activities that focus on a rela tionship between the home and the school (Sy, Rowley & Schulenberg, 2007). Fami lial participation in student learning in traditional schooling environments has a positive relationship with student achievement, attendance and pro-social behaviors (Henderson, 1981, 1987; Anderson, Hieb ert, Scott & Wilkinson, 1985; Cotton & Reed-Wikelund, 1989; Edwards, 2004). This relationshi p has held considerable appeal to school administrators, politicians, parents and students. Thus a considerable body of research has 15

PAGE 16

developed to explore the role of the family a nd its effects on students academic achievement (Christenson, Rounds & Gorney, 1992; Epstei n, 1991; Keith, 1991; National Center for Education Statistics [NCES], 1997). It is essential that a better understanding of the role of the parent with regards to virtual school achievement be developed. Understanding th e impact that family involvement contributes to student learning is a critical element to di scerning factors that prom ote student success in online learning environments and a primary concer n of virtual school educ ators and leaders. The purpose of this dissertation is to investigate the role of familial participation in students achievement in K-12 virtual schools. Defining Family Involvement Family involvement is an abstract ideal associated with a multitude of behaviors and interactions that focus on a relationship between the home and the school (Sy, Rowley & Schulenberg, 2007). The homeschool relationship t ypically involves several different types of involvement: family involvement, parent invol vement, parent education and family-school partnerships (Edwards, 2004). Differing ideas abou t the role of the involved family reflect the many social strata and structures in the U.S. (Morrow, 1989; Brito & Waller, 1994; Seely, 1981; Grolnick & Slowiaczek, 1994). It is further purporte d that cultural identity affects the role and type of parental involvement (Sy, Rowley & Schulenberg, 2007; Kerbow & Bernhardt, 1993; Mau, 1997; Yao, 1985). Rather than differentiati ng between the multiple typologies, Edwards (2004) uses the term family involvement as an omnibus expression. Edwards describes the term as broad enough to encompass the radical change undergone in what constitutes a family and the roles and responsibilities in the lives of the adults who nurture the children in todays schools (p.11). For the purposes of this study, the term family involvement or parental involvement will 16

PAGE 17

be limited to the involvement of biological parents or those placed in a guardianship role as supervisors of a student enro lled in a virtual school. Successful family involvement is indicated by the process of helping parents and family members use their abilities to simultaneously benefit themselves as well as their children (Edwards, 2004). By adopting an Edwardian und erstanding of family involvement it can be asserted that parental involvement could involve a multitude of different activities, some of which may not be associated with physical pres ence within a school (Sy, Rowley & Schulenberg, 2007). The Role of Familial Involvement in Traditional Schooling Research demonstrates that familial partic ipation in students learning has a positive relationship with student achievement (Anders on, Hiebert, Scott & Wilkinson, 1985; Cotton & Reed-Wikelund, 1989; Edwards, 2004). Family invol vement in childrens learning has been linked to higher achievement, enhanced conduct a nd attitude, and better attendance for students (Henderson, 1981, 1987; Kellaghan, Sloane, Alva rez, & Bloom, 1993; Henderson & Berla, 1994; Henderson & Mapp, 2002). Research also indicates that early involv ement by parents has a profound effect on the magnitude of educational outcomes. Early ch ildhood education programs with strong parent involvement components have amply demonstrated the effectiveness of th is approach (Ziegler, 1987; Schweinhart and Weikart, 1992). Despite the significant body of research that exists addressing the role of the family in traditional schooling, much remains to be researched. For example, the relatively recent move from synchronous (telephone and parent-teacher conferences) to asynchronous communication methods (email, voicemail) between home and school has specific implications for how teachers and parents communicate (Bouffard, 2006); and the effect of familial involvement on specific 17

PAGE 18

academic content areas remains an area with limited research (Fan & Chen, 2001). Additionally, the role of social and economic variables remain s a source of considerable controversy (Lareau, 2000, 1989), as is the abject visibility of social class stratification in the nature of familial participation (DeCarvalho, 2000). The Role of Familial Involvement in Virtual Schooling Virtual schools provide students access to onl ine educational opportuni ties that fulfill a need in the contemporary educational system by giving students access to learning (Joy & Garcia, 2000). Virtual schooling fills a distinc tive role in the nations education agenda by serving as an additional option for enacting school choice legislation (Cavanaugh, Gillan, Kromrey, Hess, & Blomeyer, 2004; NCES, 2005; Watson, Winograd, & Kalmon, 2004). Several virtual schools collect data from families, including the Ohio Virtual Academy, Florida Virtual School and Kiel eSchool, but none focus their e fforts on the specific roles that parents play with regards to virtual curriculum and student ac hievement. According to Clark (2001) and Rose and Gallup (2001), family attitudes regarding the va lidity of distance education will play an important role in the continued gr owth of the virtual school movement. Research by Cavanaugh et al. (2004) identified factors that influence the success of a distance education program, these factors include Abilities and disabilities of the student Quality of the teacher Demands of the content Design of the distance learning environment Del Litke (1998), Russell (2004) and Black, Ferd ig and DiPietro (2008) encourage future researchers to consider these factors and bu ild upon them, specifically, to identify factors external to the student that influence student achievement, in cluding the role of the family. Furthermore, Russell (2004) disc usses the value of the parent from a non-empirical standpoint, 18

PAGE 19

asserting that the family is a critical com ponent of the virtual school dynamic and recognizing that there will be significant variance in the qua lity of familial support that students receive at home. Del Litkes (1998) analysis of a Canadian vi rtual school program identified a general lack of communication between families and teachers, and identified suppor tive parents as a key component to virtual school success. Del Litke argue s that the familial role in virtual schooling is more complex than the role the family plays in traditional schooling. Re search by Russell (2004) asserts that a portion of the responsibility for a stude nts education is returned to the family in the virtual schooling model. Thus it is important th at parents provide approp riate support to enable the timely completion of schoolwork, and appropriate use of an online co mputer at home or elsewhere. A virtual schools lack of physi cal facilities may in fact pr ovide an opportunity to promote more equitable involvement by families from a spectrum of social groups and strata. Parents differ by social class, race, and et hnicity in their access to school s and in their effectiveness in dealing with educators (Lareau, 1989; Lareau & Shumar, 1996; Wells & Serna, 1996). Parents from higher socioeconomic classes have more opportunity, as compared to those in lower classes, for school involvement, particularly when physical presence is required (McGrath & Kuriloff, 1999). Many virtual schooling models rem ove the time constraints typically associated with the traditional school day and the distance between parent and child, potentially making it easier for busy parents and other family members to engage in educational interactions with children (Kozma & Shank, 1998). Involved parents in traditional schools have an unintentional tendency to focus their efforts toward promo ting advantages for their own children. This situation encourages school administrators to focus on the needs of those parents and families 19

PAGE 20

who have high access to schooling, typically th ose from a higher socioeconomic stratum (McGrath & Kuriloff, 1999). Because of their lack of physical structure virtual schools may, in fact, contribute to a leveling of the playing field between social classes by effectively handicapping those with greater time availability and eliminating involvement opportunities that involve physical presence in the school. Models of Parental Involvement Several contemporary models of parental involvement are ac tively used by the research community writ large. Specific models focus on asp ects of parental involvement associated with family school partnerships (Edwards, 2004); th e quantification of a continuum for parent involvement (Cervone & OLeary, 1982); the asserti on that parental involvement is a form of social capital (Grolnick, Benjet, Kuriwski & Apostoleris, 1997); and the categorization of relationships between the home and the school (Epstein, 1995). Each of these models can be applied in a virtual school setting, though they do not focus on a measurable underlying psychology of the involvement activity (Franke, 2005) Rather, their focus is on actions. To gain a comprehensive understanding of the parental involvement construct and the psychological underpinnings of this construct within the context of virtual schooling, the Hoover-Dempsey and Sandler model (1995, 2005) of parental involvement was select ed for this study. The models focus on the psychology of involvement provides for an ideal opportunity to explore parental involvement in a new domain, that of th e K-12 virtual school (Fan & Chen, 2001). Hoover-Dempsey and Sandler (1995 ) built upon the work of Eccl es and Harold (1993) to discern motivations for parental involvement. They suggested that parental involvement is predicated upon personal constr uction of the parental role, personal sense of efficacy for helping children succeed in school, and parental reaction to the opportunities and demand characteristics presented by both their children and their childrens schools (p. 311). Armed 20

PAGE 21

with the theoretical antecedents of involvement, Hoover-D empsey and Sandler (1995) hypothesized that they could then discern the influence that i nvolvement asserted upon student achievement. Hoover-Dempsey and Sandler (1995) downplay, though do not marginalize, the prominence ascribed by researchers (eg: La reau, 1989; Heymann, & Earle, 2000; Aughinbaugh, Pierret & Rothstein, 2005) to fact ors such as parents' education, income, marital status, and other indicators of family status in efforts to understand parents' involvement decisions. It is their assertion that these factors did not effectively explain parents' decisions to become involved, their choice of involvement forms, or the e ffects of their involvem ent on student outcomes. According to Hoover-Dempsey and Sandler (1995) parents become involved in children's education based on three factors: The successful development of a personal construction of the parental role that includes participation in their children's education. The successful development of a positive sense of efficacy for helping their children succeed in school. The perception of opportunities for invol vement from children and the school. Parents then choose specific forms of involve ment in response to the explicit domains of skill and knowledge they possess, the total de mands on their time and energy, and specific requests for involvement from children and th e school. The Hoover-Dempsey and Sandler model asserts that parental involvement then influe nces children's developmental and educational outcomes through such mechanisms as modeli ng, reinforcement, and instruction (HooverDempsey, Battiato, Walker, Reed, DeJong, & Jones, 2001; Hoover-Dempsey & Sandler, 2005; Green & Hoover-Dempsey, 2007). The ultimate result of the Hoover-Dempsey/Sandler assertions is the framework re presented in Figure 3, developed through empirical analysis by The Family School Partnership Lab at Vanderbilt University (2007). By draw ing upon this theoretical 21

PAGE 22

framework, it is possible to quantify beliefs that affect parental involve ment and discern their relationships to student achi evement in virtual schools. Problem Statement The role of familial involvement in virtual schooling could be as important, if not more important than its role in traditional schooling (R ussell, 2004). Current research does not address the role of the family in online learning (Russell, 2004). This dissertation w ill investigate the role that familial involvement plays in achie vement for virtual school students. Significance Virtual schooling is fast becoming a mainstr eam option for todays ge neration of learners (Watson, Gemin & Ryan, 2008). Virtual school administrations have s een fit to develop curriculum that addresses the br oad range of students from K-12 (Clark, 2001). Greater numbers of students and parents are demanding access to distance education opportunities, increasing the amount of public funding alloca ted to providers (Black, Ferdig & DiPietro, 2008). While virtual schooling at the K-12 level has grown in popularity, research-b ased investigations into successful teaching, learning and student support processes are limited (Cavanaugh et al., 2004). Instructional practices for onlin e settings and the guidelines a nd standards produced by leading organizations in teaching and learning, includ ing The American Federa tion of Teachers (AFT, 2001), Sloan-C (Sloan-C, 2002), American Distan ce Education Council (ADE C, 2003), Southern Regional Educational Board (SREB, 2006), Nationa l Education Association (NEA, 2006), North American Council for Online Learning (NACOL, 2007-2008) remain non-empirical adaptations of face-to-face practices whose primary focus is on guideli nes for course development and pedagogical practice (DiPietro, Ferd ig, Black & Preston, 2008). There is reason to believe that the role of familial involvement in virtual schooli ng could be as important, if not more important than its role in traditional schooling (Russell, 2004). Unfort unately, contemporary studies on 22

PAGE 23

parental involvement only address face-to-face student populations. Therefore, the study of parental involvement and its relationship to virt ual school student achievement could assist the development of new communication strategies betw een virtual schools, teachers and parents that will lead to improvement in student achievement. Purpose The purpose of this study is to investigate the role of familial participation in student's achievement in K-12 virtual schools. Research Questions The following questions frame the research: What quantifiable impact, if any, does familia l involvement have on student achievement in K-12 virtual schooling? Do students and parents differ in their percep tions of familial involvement in K-12 virtual schooling? Do factors including, socioeconomic status, race and gender, eff ect involvement of families in virtual schooling? 23

PAGE 24

CHAPTER 2 REVIEW OF LITERATURE Introduction Parental involvement has been identified as a critical component of student achievement in traditional schooling environment. Several research ers have suggested that parental involvement is of importance to K-12 virtual schooling (Clark, 2001; Johnston, 2004). It is tempting to assume the commonality between factors infl uencing achievement in traditional schooling environments and factors that influence virtual school achie vement (Reeves & Nass, 1996). Unfortunately, there are no published studies that empirically investig ate the impact of this type of involvement on student achievement (Russe ll, 2004; Black, Ferdig & DiPietro, 2008). An experimentally based understandi ng of the role of the parent with regards to virtual school achievement is an important research component that will shape the future of K-12 virtual schooling in the U.S. Understandi ng the impact that family involvement contributes to student learning is a critical element for discerning factors extern al to the school environment that promote student success in virtual schooling. The goal of this chapter is to explore the current state of virtual school research by establishing it within the context of existing research that e xplores family involvement and its impact on students academic achievement. This chap ter is organized into five sections, the first of which will introduce K-12 virtua l schooling and research relate d to achievement. The second section discusses research underlying the f oundation of contemporary notions of family involvement in U.S. public education. The third section discusses resear ch related to family involvement in virtual schooling. The fourth section addresses the theoretical model developed by Hoover-Dempsey and Sandler (2005) and the fina l section provides a summary of the chapter. 24

PAGE 25

This review of literature will function to establ ish a context for virtual schooling and provide a basis for discussing the role of the pare nt in virtual student achievement. An Overview of Virtual Schooling in the US Moore and Kearsley (1996) defi ne distance education as: p lanned learning that normally occurs in a different place from teaching and as a result requires special techniques of course design, special instruct ional techniques, special methods of communication by electronic and other technology, as well as sp ecial organizational and admini strative arrangements (p.2). Distance education has a rich history in the United States and internationally as a form of instruction for both children and adults. Virtua l high schools constitute a relatively recent addition to the broader concept of distance education. These school s are an evolutionary offshoot of correspondence programs, whic h have facilitated distance edu cation practices since the 1870s (Yates, 2003). Distance learning programs provi de a complementary opportunity for those not able to attend traditional learning environments and have played a role in the education of disparate populations, including displaced indivi duals, the active-duty military and those serving sentences in state and federal prison system s (Keegan, 1996). These programs, dependant upon the prevalent communications technology availa ble to the public, have relied on multiple communication mediums, including postal, radi o and uni-directional video to reach their audience (Keegan, 1996). With the widespread adop tion and mainstreaming of the Internet in the 1990s, many distance learning programs, lead first by universities, looked to the Internet as an efficient medium for instruction (Roblyer, 1999). T oday, the Internet is a principal component of distance education, resul ting in a morphing of terminology in which distance education has now come to be termed online or virtual education. The first online K-12 public school began in 1995 with the Cyber School Project in Eugene, Oregon. This program was started by ni ne teachers to offer supplemental high school 25

PAGE 26

classes (Greenway & Vanourek, 2006). By 1996, there were three online schools: the WebSchool in Orange County, Fl orida (later renamed as Flor ida Virtual School); The CyberSchool Academy, started by the Federal Way School District in the state of Washington; and the Concord Virtual High School (later renamed as Virtual High School) based in Maynard, Massachusetts (Greenway & Vanourek, 2006). From 1996 to 2002, just twelve different state de partments of education started virtual high schools in their respective states (Zucker & Kozma, 2003). However, by October, 2008, 44 states had established state-led online learning pr ograms (Watson, Gemin & Ryan, 2008). Based on a comprehensive survey of school administrators, the Sloan Consortium estimates that more than 1,000,000 K students participated in online learning during the 2007-2008 school year. Greater than half of the administrators respondi ng to the Sloan survey (57.9 percent) reported members of their student body participating in online courses during 2005 (Picciano & Seaman, 2007). Further, 25% of those without virtual school enrollees anticipated a need to enroll students in online courses within the ne xt three years (Picciano & Seaman, 2007). The Sloan report supports Maeroffs (2003) prediction that distance education student populations in the K-12 setting, having already experienced dou ble digit growth since 1996, will continue growing at double digit rates for the foreseeable future. While many students utilize virtual schools to au gment courses taken as part of a regular instructional program within a public school, a gr owing body of students is engaging in entirely online instruction and forgoing tr aditional classrooms all togeth er (Clark, 2001). Unfortunately, due to data collection limitations inherent in the U.S. virtual sc hools, no accurate count of dual enrollment versus full-time enrolled students can be made (Black, Ferd ig & DiPietro, 2008). 26

PAGE 27

Motivations for establishing a K-12 online learning programs typically center on expanding course offerings to those unable to ac cess due to geography, disa bility or lack of course offering (Watson & Ryan, 2007; Mills & Roblyer, 2004). A prime example of the appeal of virtual schooling is represented by students and parents requesting increased access to foreign language programs, such as Mandarin Chinese a nd Arabic. Given the sh ortage of qualified foreign language instructors with specific expe rience in these languages, online learning offers the opportunity for students to gain access to high quality instructors regardless of their geographic location and the limita tions of the course offerings in their brick and mortar high school (Watson & Ryan, 2007; USDOE, 2007). K-12 virtual schools are seen as vehicles of school change and have become increasingly popular from a political standpoint at both the state and local level (Hassell & Goda rd-Terrell, 2004; Ferdig & Cavanaugh (eds), in press). Virtual schools, similar to their brick and mo rtar counterparts, have been developed and implemented in many different formats. Several di fferent administrative structures are presently practiced throughout the U.S., including, statewide supplemental programs, district-level supplemental programs, single-district programs, multidistrict cyberschoo ls, and online charter schools (Rice, 2006; Watson, Winograd & Kalo man, 2004). Two character istics define the aforementioned programs: whether the online pr ogram enrolls students and grants credit and diplomas directly and the jurisdic tional level of the program (such as state or district) (Watson, et al., 2004). States, including Florida, Mich igan, Louisiana and at least 38 others have commissioned state or district/county sponsored virtual school s (Watson, & Ryan, 2007). Increasingly colleges and universities are offering advanced high sc hool students the opportunity for dual enrollment utilizing the Internet to take college level courses while completing their high school degrees. 27

PAGE 28

States and county regions have utilized economies of scale by build ing consortia to offer virtual schooling opportunities to students within their geographic locali ties (Hassel & Te rrell-Godard, 2004). In addition, many public school systems ha ve taken it upon themselves to offer virtual schooling opportunities to supplem ent educational needs and to reach out to homeschooled students (Hassel & Terre ll-Godard, 2004). State-chartered schools and for-profit entities make up the final subset of the virtual schooling comm unity (Clark, 2001). As increasing numbers of students and parents are demanding access to di stance education opportunities, the amount of funding allocated to virtual schools has incr eased along with access. Given their numerous models and bureaucratic formats, virtual high schools cannot be eas ily nominalized as either an educational or technological phenomenon, though all rely on the In ternet for student-instruction interaction (Roblyer, 1999). Much of the research regarding the advantages of online learning has focused on collegiate and other adult populations. Some, but not all, of this research can be applied to K-12 populations. By utilizing a body of research from post-secondary populations accepted by the K12 online research community writ large, in addition to research that focuses exclusively on K12 online student, specific affordances of the medium can be discerned (Kiekel, 2007). Though virtual high schools, traditional schools can supple ment the depth and breadth of their course offering without incurring the costs associated with infrastructure or increased staffing (Roblyer & Elbaum, 2000). Research indicates that online le arning tends to facilitate greater individual student and teacher interactions (Meyer, 2003). Meyer (2003) and Weisgerber and Butler (2005) report that asynchronous mediums of instruction, most oft employed in present day K-12 virtual schooling (Watson & Ryan, 2007), prom ote a myriad of skills includ ing, self discipline, sense of community, communication skills, and reflect ive capacity (Romiszowski & Ravitz, 1997; 28

PAGE 29

Chickering & Ehrmann, 1996 ; Debard & Guidera, 1999-2000 ; Dutt-Doner & Powers, 2000). Given the lack of physical presence associated with online learning, students who may not normally participate in traditional classroom setti ngs often find themselves able to contribute in an online medium (Chong, 1998; Dutt-Doner & Powers, 2000). Asynchronous discussions allow students and teachers the opportunity for consider ation and reflection upon postings resulting in the opportunity for rich discourse (Ferdig & Roehler, 2003). Twigg (2003) and Palloff and Pratt (1999) note that this style of di scussion changes the fundamental ro le of the teacher, encouraging a more constructivist approach with students. The concept of anytime, anywhere learning is an affordance of online instruction, in many instances, particularly those involving asynchr onous online mediums, students have access the course 24 hours a day, seven days a week. K-12 Virtual Schools and Student Achievement Since their emergence and rapid growth, questi ons related to the efficacy of K-12 virtual schoolings have been raised by school admini strators, policy makers, parents and students (Cavanaugh et al., 2004; Dickson, 2005). As such, researchers have spent considerable time and effort comparing achievement rates between virtual schools and their traditional counterparts. One of the earliest analyses conducted by Russell reviewed 355 studies, ultimately reporting no significant difference in achievement between online and traditional courses, but a wide variability in effect size (Ru ssell, 1999). A second meta-analy sis conducted by Bernard, Abrami, Lou, Borokhovski, Wade, Wozney, Wallet, et al (2004) also confirm a net no significant difference. A more granular inspec tion of Bernard et als. reveals a large variance in effect sizes, from -1.31 to +1.41. These effect sizes indicate th at while the average effect is indeed zero, groups of students excel in online learning enviro nments and groups of students are clearly not succeeding in an online learning environment (Dickson, 2005). 29

PAGE 30

The Cavanaugh et al. 2004 meta-analysis support s the Bernard et al. findings, where again, an average effect size of zero was found but a wide variance in e ffect sizes occurs (between -1.2 and .6). The Cavanaugh et al. resu lts were based on 116 different e ffect sizes gathered from 14 different online K-12 programs. These three meta-a nalytic findings indicate that while there is indeed no significant difference between student achievements in K-12 virtual schooling, there is a body of students whose achievement can be improved. Family Involvement Family involvement in schools is commonly recognized as an activity that has substantial benefit to those involved. Schools, students and parents all derive sp ecific benefits from parental involvement (Comer & Haynes, 1991; Epst ein, 2001; Fan, 2001; Henderson & Mapp, 2002; Jeynes, 2005). Research shows th at children receive greater benefit the earlier the parent involvement process begins, t hough benefit can be derived for students at all grade levels regardless of gender or racial status of the children (Je ynes, 2005; Cotton & Reed-Wikelund, 1989). Unfortunately, there is considerable deba te when attempting to define parental involvement (Jaynes, 2005; Fan & Chen, 2001). Due to its multidimensional nature, defining family involvement proves to be a difficult task as the concept bridges many different activities and endeavors. This complexity makes it difficult to label and measure (Fan & Chen, 2001; McLaughlin, 2006). The No Child Left Behind Act of 2001 (White House, 2001) defines parent involvement in Sec. 9101(32) as the participation of parents in regular, two-way and meaningful communication involving student academic learning and othe r school activities including, ensuring: That parents play an integral role in assisting in their childs learning; That parents are encouraged to be actively involved in their childs education at school; 30

PAGE 31

That parents are full partners in their childs education and ar e included, as appropriate, in decision making and on advisory committees to assist in the education of their child; The carrying out of other activities, such as those described in Sec. 1118. Pat Edwards (2004) uses the term family involvement as an omnibus expression recognizing the wide variety of parental behavi oral patterns and practices (p.11) congruent to ideals noted by Balli (1996) and Brown (1994). Edwardss description is broad enough to encompass the radical change undergone in what constitutes a fa mily and the roles and responsibilities in the lives of the adults who nurture the children in todays schools ( p.11). For the purposes of this study, the term family involvement or parental in volvement will be limited to the involvement of biological parents, or those placed in a guardians hip role, as supervisors or tenants of a student enrolled in a virtual school. After surveying the myriad of definitions associated with family involvement, McLaughlin (2006) recognized that the most descriptive and useful definitions of the term focus on the multiple dimensions of family involvement by conceptualizing and considering the behaviors from which family involvement is instigated. Drawing on several descriptive definitions, Epstein (1995) desc ribes commonly accepted categories of family involvement to assist educator s in the development of compre hensive school-family-community partnerships. Epsteins theoretical model include s six types of homeschool relationships, they are: (1) Parenting: Promoting home environments that are supportive of ch ildren as students; (2) Communicating: Establishing eff ective school to home communications about student programs and child progress; (3) Volunteerin g: Parental help and support in classroom or school activities; (4) Learning at Home: Involving families with their children in learni ng activities at home, including homework and other cu rriculum-linked activit ies and decisions; (5) Decision Making: Inclusion of parents in school decision, developing parent le aders and representatives; (6) 31

PAGE 32

Collaborating with Community: Integrating reso urces and services from the community to strengthen school programs. Hoover, Dempsey and Sandler (1985, 1995, 2005) define parental involvement through mechanisms of influence: modeling of behavi ors, reinforcing school values and home-based instruction. Involved parents encourage pro-scho ol ideals through reinfo rcement of behaviors likely to increase learning (eg: study, regular attendance). Fan and Chen (2001) assert that the Hoover-Dempsey and Sandler model holds consider able appeal as a co mprehensive model of family involvement; it attempts to both explain parents rationale for involvement and describe mechanisms for involvement. Desforges and Abouchaar (2003) describe the most efficacious forms of parent involvement are those which encourage parents to interface directly with their children on learning activities in the home. Cotton and Reed-Wikelund (1989) found that active forms of parent involvement, such as participating in phone calls with teachers, reading and signing homeschool communications and participating in parent teacher conferences, produce greater achievement benefits than the more passive ones. Considerably greater achievement benefits are noted when parent involvement is active and includes participation in school and classroom activities such as direct classroom participation or field trips. Family Involvement in Virtual Schools Scant literature regarding the role of parents in K-12 virtua l schooling exists. A comprehensive literature search utilizing the ProQuest Database, EBSCOhost and Google Scholar with the following search terms: Virt ual School, Online Learning, Parents, Family, Involvement, Achievement, and Student reveal ed relatively few mentions of parents in contemporary literature, limited to: Del Litke (1998); Clark (2001); Ru ssell (2004); and Black, Ferdig and DiPietro (2008). No quantitativ e studies with a primary focus on parental 32

PAGE 33

involvement were found. Further investigati on of Del Litkes 1998 qualitative study of a Canadian Online Middle School offers several important insights: Teachers identified supportive parents as a significant factor in student success. Teachers discovered that attemp ts to facilitate online rela tionships with parents proved difficult. Reliance on parents for authoritative pur poses provided inconsistent outcomes. Del Litke characterized parent involvement based on the students perception, dividing parents into three separate groups: absentee, supp orters and participativ e parents. Through this process, Del Litke acknowledges the critical role that parental pe rceptions play in online student achievement in congruence with Hoover-Dempsey and Sandlers (2005) model of parental involvement. Russell (2004) supports Del Litk es commentary on virt ual school parents, asserting that a virtual school in structors lack of physical pres ence results in a portion of the responsibility for a students educ ation returned to the family in the virtual schooling model. Thus, it is important that parents provide approp riate support to enable the timely completion of schoolwork, and appropriate use of an on line computer at home or elsewhere. A significant amount of research and effort has focused on identifyi ng traits of online learners that correlate value to success. Attributes such as self-motivation, technological experience, interest in subj ect matter and self confidence have all been recognized as contributory to a successful online learning ex perience (Gibson & Graff, 1992; Coussement, 1995; Richards & Ridley, 1997; Hardy & Boaz, 1997). In addition, a body of research concerning social constructs has assisted in providing greater unde rstanding of social variables that contribute to a student satisfaction, learning and reten tion (Cereijo, Young & Wilhelm, 2001; Curry, 2000; Rovai & Whiting, 2005; Haythornt hwaite, Kazmer, Robins & Shoemaker, 2000; Eastmond, 1995; Rovai, 2001a; Rovai, 2001b). The Cavanaugh et al. 2004 meta-analytic 33

PAGE 34

assessment of factors contributory to success in K-12 virtual schooling led to the development of a systematic framework useful for outcomes ba sed research. The four factors identified by Cavanaugh et al. were built on by Black, Ferdig and DiPietro (2008) drawing upon Del Litke (1998) to include additiona l categories, including a category that addresses the role of the parent and/or family in virtual student success, as shown in Table 2-1. Because of the limited scope of research inves tigating the role of family involvement in virtual schooling a significant number of questions remain unanswered. Therefore, it is necessary to engage in a comprehensive study of parental involvement to determine its importance in K-12 virtual schooling. The Hoover-Dempsey and Sandler (1995; 2005) model of parental involvement provides a means by which the constr uct can be understood a nd interpreted, and its effect on a student populati on can be ascertained. The Hoover-Dempsey and Sandler Model of Family Involvement In spite of the overwhelming evidence supporting family involvement in students education (Epstein, 1996; Fan & Chen, 2001; Henderson & Mapp, 2002), significant barriers exist, consciously or unconsciously hindering implementation. These barriers are enacted by teachers, administrators and even parents th emselves (Hoover-Dempsey, Walker, Jones & Reed, 2002; Hoover-Dempsey, Walker & Sandler, 2005; Edwards, 2004). To increase familial involvement, researchers have pr oposed several contemporary models of parental involvement. These models focus on aspects of parental involvement associated with family school partnerships (Edwards, 2004); th e quantification of an underlying continuum for the measure of parental involvement (Cervone & OLeary, 1982); the hypothesis that parent al involvement is a form of social capital (Grolnick, Benjet, Kuriwski & Apostoleris 1997); a nd the categorization of relationships between the home and the school (Epstein, 1995). Each of these models can be applied in a virtual school se tting, though, the previously men tioned models do not focus on a 34

PAGE 35

measurable underlying psychology of the involvement activity (Franke, 2005). Rather, their focus is on actions. Because cognition precedes action, models that focus on actions do not provide for an adequate understanding of the an tecedents that promote action (Leahy & Harris, 2001). To gain a comprehensive understanding of the latent psychological construct of parental involvement construct and observabl e factors associated with this construct within the context of virtual schooling, the Hoover-Dempsey and Sandler model (1995, 2005) of parental involvement was selected for this study. The models focus on the psychology of involvement rather than involvement activities and actions provide for an ideal opportunity to e xplore the psychological construct of parental involvement in a new domain, K-12 virtual schools (Fan & Chen, 2001). Hoover-Dempsey and Sandler (1995) built on th e work of Eccles and Harold (1994, 1996) to discern motivations for parent al involvement. They suggested that parental involvement is predicated on personal constructio ns of the parental role, personal senses of efficacy for helping children succeed in school and parental reacti on to the opportunities and demand characteristics presented by both their children and their childrens schools (p. 311). Armed with the theoretical antecedents of involvement, Hoove r-Dempsey and Sandler (1995) hypothesized that they could then discern the influence that involvement had on student achievement. Hoover-Dempsey and Sandler (1995) downplay, though do not marginalize, the prominence ascribed by researchers (eg: La reau, 1989; Heymann, & Earle, 2000; Aughinbaugh, Pierret & Rothstein, 2005) to fact ors such as parents' education, income, marital status, and other indicators of family status in efforts to understand parents' involvement decisions. It is their assertion that these factors did not effectively explain parents' decisions to become involved, their choice of involvement forms, or the e ffects of their involvem ent on student outcomes. 35

PAGE 36

Research by Hoover-Dempsey and Sandler (1995, 1996) focuses on understanding the psychological variables associated with parental involvement rather than solely focusing on involvement activities. By doing so, Hoover-Dempsey and Sandler subvert potential barriers or conceptualizations that may exist. As such, they have proposed a multi-level model describing the parental involvement proce ss (Figure 2-1). Further, the Hoover-Dempsey and Sandler model (1995, 2005) complies with Fan and Chens (2001) recommendation for measurable multidimensionality. The Hoover-Dempsey and Sandler (2005) model of parent involvement is a five tiered construct; the first tier su ggests that parents become involved in their childrens education for three major reasons: Personal motive (Do parents belie ve they should be involved?) Life context (Do parents have the knowledge/skills and time necessary to help their child?) Invitations (Do parents believe that the school wants their involvement?) The second tier of the model describes th e parents mechanisms of involvement: Encouragement (What are the methods of academic encouragement?) Modeling of appropriate school-related skills (Are parents modeling academic skills? eg: reading, writing, mathematics) Reinforcement of learning and attributes re lated to learning (By what means do parents reinforce leaning behaviors?) Instruction (What instructional methods are used by pare nts to assist children?) The third tier of the model focuses on the ch ilds perception of pa rental involvement: Encouragement (What are student perceptions of their parents me thods of academic encouragement) Modeling of appropriate school-related skills (What are student pe rceptions of their parents modeling of academic skills?) 36

PAGE 37

Reinforcement of learning and attributes rela ted to learning (What are student perceptions of parental reinforcement beha viors related to academics?) Instruction (What are student perceptions of the instructional methods used by parents?) The fourth tier of the model focuses on attr ibutes associated with student learning: Academic self-efficacy Intrinsic motivation to learn Self-Regulatory Strategy Use Social Self-Efficacy The fifth tier of the model represents an outcome variable influenced by parental involvement. According to Hoover-Dempsey and Sandler (1995), these outcome variables could include measures of achievement, measures of knowledge or a measure of school-based efficacy. Investigations by Hoover-Dem psey and Sandler (2005) and Green and Hoover-Dempsey (2007) provide validation for this model within the context of elementary school-age students and their parents. This research indicates that parents motiva tional beliefs, perceptions of invitation for involvement from others and perceived life context cont ribute to parental involvement behaviors. These i nvolvement behaviors are mediated by the childs perception of parental involvement, which in tu rn effect attributes within th e child, such as self-efficacy and achievement motivation, which have e ffect on achievement (Figure 2-2). Regression results suggest that parent reports of involvement were significant in predicting student academic outcomes (Adj. R2 = .039, F = 17.890, p < .000; t = 4.230, p < .000) and student reports of parental invo lvement were significant in predicting student outcomes (Adj. R2 = .357, F = 234.393, p < .000; t = 15.310, p < .000). When combined, the direct path between parent reports and student outcomes became insigni ficant, while the path between student reports 37

PAGE 38

and student outcomes remained significant (Adj. R2 = .361, F = 119.431, p < .000). These results provide evidence to support the utilization of this instrument with the study population. The assessment was validated and refined on a divers e population of parents and students in four separate studies from 2002-2003 (Hoover-Dempsey & Sandler, 2005). Correlations related to individual attributes ca n be found in Table 2-2. Figure 2-3 is a path diagram of the Hoover-D empsey and Sandler model (2007). This path diagram shows the tiered contributory fashion by wh ich parent motives and beliefs effect student achievement. The Hoover-Dempsey and Sandler model presents just one of many co nceptualizations of the family involvement construct utilized in educational research. The specific affordances of the model: multi-dimensionality, quantitative nature and focus on psychological factors rather than socioeconomic factors and involvement activities make it an ideal tool for application with a sample of virtual school students. Conclusion Recent research has provided evidence of the effectiveness of virtual high schools (Cavanaugh, Gillan, Kromrey, Hess, and Blomeyer 2004), leading some educational leaders to propose that online learning is one of the most important new approaches for K-12 schools (Blomeyer, 2002). Unfortunately, there is still not enough research to inform new policy and practice in this area, particularly concerning the role of the parent in st udent achievement (Black, Ferdig & DiPietro, 2008). In this chapter, a review of literature esta blished the foundation for an understanding of K12 virtual schooling in the U.S ., current research related to achievement in K-12 virtual schooling, and an overview of family involvement and literature rela ted to family involvement in K-12 virtual schools. Finally, a model of family involvement and its relationship to student 38

PAGE 39

achievement was described. This literature review sets the stage for the argument that an understanding of the role of the family in vi rtual schooling may provide an understanding of a factor that is purported to affect student achievement. 39

PAGE 40

Table 2-1. Success factors in K-12 virtual schooling Abilities and disabilities of the student Cavanaugh et al. (2004)s success factor Quality of the teacher Cavana ugh et al. (2004)s success factors Demands of the content Cavanaugh et al. (2004)s success factors Design of the distance learning system Cavanaugh et al. (2004)s success factors Course Instance Added by Black, Ferdig & DiPietro (2008) Other (including: parent s, guardians, mentors) Added by Black, Ferdig & DiPietro (2008) 40

PAGE 41

Level 5 Student Achievement Level 4 Student Attributes Conducive to Achievement Academic SelfEfficacy Intrinsic Motivation to Learn Self-Regulatory Strategy Use Social Self-Efficacy Teachers Level 3 Mediated by Child Perception of Parent Mechanisms Encouragement Modeling Reinforcement Instruction Level 2 Parent Mechanisms of Involvement Encouragement Modeling Reinforcement Instruction Parent Involvement Forms Values, goals, etc. Home Involvement School Communication School Involvement Level 1 Personal Motivation Invitations Life Context Parental Role Construction Parental Efficacy General School Invitations Specific School Invitations Specific Child Invitations Knowledge and Skills Time and Energy Family Culture Figure 2-1. The Hoover-Dempsey and Sandler model of parental involvement, adapted from Hoover-Dempsey & Sandler, 1995; 2005 41

PAGE 42

42 Figure 2-2. Diagram describing the results of Hoover-Dempsey and Sandlers 2005 study

PAGE 43

Table 2-2. Correlations among parents reports of involvement mechanisms, student perceptions of parents involvement, and a summary measure of student achieveme nt (Hoover-Dempsey & Sandler, 2005) Parent report of involvement mechan isms Parent report encouragement Parent report modeling Parent report reinforcement Parent report instruction Student report parental encouragement Student report of parental modeling Student report of parental reinforcement Student report of parental instruction Achievement TCAP Encouragement -Modeling .67** -Reinforcement .70** .75** -Instruction .61** .72** .70** -Student report of parental involvement Encouragement ns .17** .14** ns -Modeling .11* .22** ns ..20* .59** -Reinforcement .10* .22** .16** .22** .82** .61** -Instruction ns .16** ns .17** .76** .71** .74** -Student achievement .18** .15** .25** ns ns ns ns -.12* -43 p = .05 ** p = .01 TCAP stands for Tennessee Comprehensive Achievement Test, stude nts TCAP scores were used as achievement variables by Hoover-Dempsey.

PAGE 44

Figure 2-3. Path model for the Hoover-Dempsey and Sandler model of parental involvement (Hoover-Dempsey, 2007) 44

PAGE 45

CHAPTER 3 METHODOLOGY Introduction The research study surveyed both virtual sc hool students and their parents regarding perceptions of their involvement in the stude nts schoolwork. Utilizing existing partnerships established through work with the AT&T Foundation, it was determined that state virtual schools are actively voicing questions rega rding the effect of parental involvement on virtual school student achievement. A partnershi p with a state-led virtual school in the Southeastern United States was established to facilita te the collection of da ta; respondents consisted of students and a parent enrolled within this institution during the summer 2007, fall 2007 and spring 2008 semesters. Research Overview This study employed an online survey adap ted from research by Hoover-Dempsey and Sandler (2005). Data were collected from a population of virtual sc hool students and their parents in a two-tiered fashion: Tier one targeted parents of virtual school students. Tier two targeted the child enrolled in the virtual school related to a responding parent. Responses were analyzed and achievement data, in the form of a semester grade for the course (or average grade for the courses) in which the child was enrolled during the summer 2007, fall 2007 and spring 2008 semesters, were collected and evaluated in relation to the surveys. Quantitative statistical procedures were utilized to analyze the data. Research Design Educational leaders and policy makers have articulated unease with the lack of scientifically based data upon which funding decisi ons for educational policy issues can be based (Schneider, Carnoy, Kilpatrick, Schmidt & Shav elson, 2007). Schneider et al. call for an 45

PAGE 46

increased emphasis on methods to discern causali ty; one such suggested method is quantitative research. To answer the call for empirical data, a quantitative research design was used to assess the role of parental participat ion in the educational process in their childrens virtual school academic achievement. This outcomes-focused, pos itivistic method was selected to provide for a scalable, replicable, objective, empirical investigation. Through the adoption of a positivistic point of view it is possible to test theory through analysis of data collected from research subjects, and to develop optimal estimations and an indication of the precision of ones outcome (Dooley, 2001; Duncan, 1975). Further, through th e use of quantitative data collection it is possible to sample a large number of individu als and project these results to a population. Quantitative data generated by this study were analyzed using statistical procedures. Population The state-led virtual school utilized for this study has established data infrastructures that allow access to student and parent email, a ddress and phone number information and student achievement data. Thus, responses could be co llected using a multi-phase d approach. Given the ease and speed by which Internet based surveys ca n be conducted, and the validity of the data collection medium when compared to traditional means, utilizing the Internet to obtain student and parent responses is optim al (Chang & Krosnick, 2002). The virtual school surveyed in this study is a state-level institution, meaning its central administering agency is a state department of education and its primary means of funding is through state level funds. The school describes itself as a supplementary online educational program created to serve public, private and hom e school study student populations from across the state. Online courses are provided within the traditional agrarian school calendar. These courses consist of 78 core curriculum offerings, AP and elective courses. Virtual school students are primarily high school level st udents; though there is a limite d set of courses available to 46

PAGE 47

middle school level students. The virtual school o ffers both state funded and tuition funded seats to the states students. Local sc hool districts are responsible for establishing policy regarding the number of state funded course enrollments in wh ich a student may participate at a given time. Individuals were recruited from the grade 9-12 courses. Research indicates that the virtual school population consists of a racially dive rse student body, that while not completely representative of the states face-to-face school population, compar es favorably with other state virtual schools in terms of the ratio of virtua l school to statewide enrollment percentages for minority students (Clark & Blomeyer, 2007). Course content was not be taken into account when recruiting students as the composition of the student population limits th e ability to build a substant ive sample population within a specific content area. Evidence for this is prov ided through analysis of the virtual schools enrollment data for spring 2007. These data indi cate a mean enrollment of 15 students (SD = 9.7) in 181 courses in 8 course content areas. This diversity would hinder the composition of a sample of substantive size and statistical relevance. Additionally, re search indicates that intrastate virtual school populat ions are relatively homogeneous (Zucker & Kozma, 2003), which may serve as a potential confound, though it represen ts an opportunity for future analysis. Group-wise deletion was u tilized to assure that only students and their corresponding parents are included in the data analysis. Privacy and Confidentiality This research study falls under the scrutiny of both the University of Florida Institutional Review Board (IRB) and U.S. Federal Educationa l Rights and Privacy Act (FERPA). Behavioral and Non-Medical IRB, designated IRB02 within the University of Florida system, is responsible for reviewing and monitoring research with hum an subjects conducted at the University of Florida. The board reviews resear ch studies that involve behavior al observations and recordings, 47

PAGE 48

non-invasive physiological recordings analysis of documents that were previously gathered for nonresearch purposes, evaluation of behavioral/social interventions or manipulations, educational assessments, interviews, surveys, cognitive tests, and taste/food evaluations of wholesome food within FDA regulations. The University of Florida IRB ( 2007) accepts as basic principles those expressed in the Nuremberg Co de (1947), the Declaratio n of Helsinki (revised 1975), and the Belmont Report (1979). Before enga ging study participants, a research proposal was prepared and presented and approved by th e University of Florida IRB (Appendix A). Additionally, because the research project concerns the collecti on of data from the records of minors, it was incumbent upon bot h the researcher and the virtua l school to comply with the tenets of FERPA. FERPAs la test revision, 2001, reforms its purpose statement to read: preventing an education agency or institution from sharing student records, or personally identifiable information in these records, without the written consent of a parent. (NCES, 2006). FERPA specifications, which outline the conditions for exceptions to the requirement of parental consent based on the organization or circumstance on which the re quest is based, are vague. To provide some clarification and proactively addres s the potential obstacles presented by FERPA, it is important to outline the impact of th e act on the realm of educational research. The requirements outlined by FERPA do not stipulate the terms of compliance; individual states are responsible for completing this task. Though many states have as yet to formally publish procedures for compliance, th ere is considerable apprehension regarding sharing information with enti ties outside a specific stat e educational organization. FERPA allows for several exceptions to its privacy restrictions, enabling state and local agencies to provide external en tities access to student information. The first exception is termed directory information, which includes student information that is not generally considered 48

PAGE 49

harmful or an invasion of privacy such as: name address, telephone number, date and place of birth, honors and awards, and dates of attend ance. Under FERPA, school systems have flexibility in deciding what studen t data is directory information; the only requirement being that parents have the opportunity to request the exclusion of their childs data. In addition to directory information, FERPA allows disclosure, without consent, to the following parties or under the following conditions (except as note d, conditions are listed in 34 CFR 99.31): School officials are permitted to access student information to perform their professional responsibility. Representatives of federal state or local education agencies are permitted to access student information as part of an audit, evaluati on, or to enforce educational programs. Organizations are permitted to access stude nt information for conducting a study on the schools behalf to develop, validate, or administ er predictive tests, administer student aid programs, and improve instruction. Upon agreeing to participate in an evaluation by external re searchers or evaluators, an educational entity may provide access student information and da ta. The responsibility falls on the educational entity to comply in their repor ting with the stipulation that students not be individually identified, or even recognizable through outcomes. Data Integrity and Security The advent and evolution of the World Wide Web and other electronic methods of mass communications has made data privacy and protection issues of prime importance for government entities, educational inst itutions, private industry and indivi duals. It is the priority of the researcher to ensure the privacy and protecti on of the data collected from the virtual school. Therefore, a structured policy has been adopted for data storage, transformation and reporting. The key principles instituted to ensure data protection and integrity are: Notice: The virtual school has been notified about the purposes for which the data was collected, the data will not be used for any other purpose. The virtua l school can contact the University of Florida researcher at anytime with any inquiries about data. 49

PAGE 50

Individual Privacy: The researcher works to li mit the collection of personally identifiable information related students or teachers including, their name, address and phone numbers. Sharing: It is the standard pract ice of the researcher to provide interstate data in aggregate to ensure that only non-identifiable data are pr esented through aggregat e reports in the data portal. Access: The virtual school has full access to it s data and the ability to correct or amend information where it is inaccurate. Security: The researcher has taken precautions to protect individual virtual school information from loss, misuse and unaut horized access, disclo sure, alteration and destruction. The researcher strictly follows th e industrys best practi ces to protect servers and data infrastructure from unauthorized ac cess. All web applications are tested and scanned for security vulnerabilities before deployment. Database servers are updated and backed up regularly and connecti ons are audited to detect secu rity related issues. Standard techniques to secure system f iles, such as: encryption, utiliz ation of firewall sand antivirus solutions are used to make servers secure. Data integrity: The researcher strives to assure that all the data is relevant for the purposes for which it is to be used and is reliable for its intended use, accurate, complete, and current. Power Calculation and Sample Size The power of a statistical test is the probability that the test will reject a false null hypothesis (that it will not make a Type II error). As power incr eases, the chances of a Type II error decrease. The probability of a Type II error is referred to as the false negative rate ( ). Therefore, power is equal to 1 (Cohen, 1988). Cohen (1992) desc ribes power of .80 as an acceptable convention that effectively minimizes th e risk of Type II error while not increasing the researchers material costs for collection of data. Utilizing GPower, a power analysis tool, a one-tailed, post hoc calculation of power was made: d = .385 = .05, sample size = 168 pairs, estimates power of .80. According to Gatti and Harw ell (1998) the utilization of software to calculate power provides estimations that are preci se and easier to extrapolate than traditional power charts. 50

PAGE 51

Sampling Procedure To obtain a proportional sample of students from the institu tion, the entire body of students enrolled in the virtual school during su mmer 2007, fall 2007 and spring 2008 semesters was contacted. This procedure allowe d for an equal possibility of selection by all members of a population for which AVS mainta ins data (Dillman, 2007). Instrumentation A survey instrument designed by Hoover-Dempsey and Sandler (1995) was adapted for this research study. The instrument was develo ped as a part of a comprehensive research program intended to a) develop and refine scales necessary to test the Hoover-Dempsey and Sandler (1995, 1997) model of pare ntal involvement and b) examine elements of the parental involvement process described by the model. The survey successf ully described the relationship between parent perceptions of school involveme nt, student perceptions of school involvement and student academic achievement with suffici ent reliability and validity (Hoover-Dempsey & Sandler, 2005; Green, Walker, Hoover-Dempsey & Sandler, 2007). Results of Hoover-Dempsey and Sandlers 2005 study can be summarized in Figure 3-1. Regression results suggest that both parent and student re ports of involvement were significant in predicting stude nt academic outcomes (Adj. R2 = .039, F = 17.890, p < .000; t = 4.230, p < .000) and student reports of parental involvement were significant in predicting student outcomes (Adj. R2 = .357, F = 234.393, p < .000; t = 15.310, p < .000). When combined, the direct path between parent reports and student outcomes became insignificant, while the path between student reports and student outcomes remained significant (Adj. R2 = .361, F = 119.431, p < .000). These results provide evid ence to support the utilization of this instrument with the study population. The assessment was validated a nd refined upon a diverse population of parents and students in four separate studies from 2002-2003 (Hoover-Dempsey & Sandler, 2005). Table 51

PAGE 52

3-1 provides correlations between factors fr om Hoover-Dempsey and Sandler 2005 study. Both the parent and the student surveys are f eatured in Appendix D and E respectively. Achievement Data Under state law applicable to the virtual sc hool studied (160-4-2-.13 Statewide Passing Score), the lowest possible score that a student can earn and still meet the requirements for completion of a subject or grade is 70 percent based on a 100 percent sc ale. For this study, achievement data will be collected in the form of the student s numeric course grade, based on a scale from 0 to 100, for the summer 2007, fall 2007 and spring 2008 semesters. Jeynes (2005) notes in his 2005 meta-analysis that numeric co urse grade data has been found as a statistically significant dependent variable in family involvement studies Jeynes encourages the use of course grades rather than standardized tests as a measure of achievement as parents are more likely to focus their involvement on classroom assignments rather than preparing their ch ild for a standardized test. The virtual school, in conjunction with its state department of education provided detailed course grade information on 453 of th e students whose parents completed the survey. This number represents 48% of the sample. This grade data had a range of 90, with a minimum of 10 and a maximum of 100, a mean of 82.5 with a standard deviation of 16.3 and a variance of 265.5. The virtual school reports grades in a perc entage format as schools and school districts may have different grading scales. Revision and Proposed Path Model Hoover-Dempsey and Sandlers survey was revised through the excl usion of questions from the parent survey pertaining to face-to-f ace school involvement. Two factors influenced the reduction of the Hoover-Dempsey and Sandler Model: geography and exis ting research. Given that virtual schools do not have a physical location, there are specific limits to a parents ability 52

PAGE 53

to self-involve specific to the virtual school. Further, an existing body of research has been developed that focuses on the proximal factors associated with student success including, Roblyer and Marshall (2002), Cavanaugh (2002), Dorman (2003), Swan (2003) and Ferdig, Papanastasiou and DiPietro (2005 ). There are few empirical st udies in K-12 virtual schooling that incorporate achievement based outcomes as a dependent variable, thus a simplified model is elicited from the Hoover-Dempsey and Sandler model of parental involvement (Fi gure 3-2). This model recognizes that student f actors exist but chooses to instead focus on the outcomes of these factors: student achievement data. With the assistance of two focus groups consis ting of a total of six high school students, slight wording changes were made to the student survey to make the survey more appropriate for the sample. The refined student survey was then subjected to a content review by three subject matter experts. Through this process, the survey was made more approp riate for use with an online learning population. The revised assessment c onsists of two forms, a 78 question parent form and a 50 question student form. Data Collection and Analysis Data was collected utilizing the Zoomera ng (http://www.zoomerang.com) a secure, webbased survey interface. Students and parents were contacted via email with an email explaining the survey and inviting participa tion. This email briefly explaine d the study and offered students and parents the opportunity to remove themselves through the use of an unsubscribe link, from the sample and to contact the research team for more information about the study. Due to the group-wise deletion procedure empl oyed by this study, if a parent vo luntarily withdrew from the sample population, his/her student would not be sent survey materials. 53

PAGE 54

Approximately three days af ter the initial survey invita tion, a subsequent reminder was sent to non-respondents. Two additional reminders we re sent to parents before closing the parent survey. Data were organized and sort ed utilizing Microsoft Excel and SPSS 13.0 to efficiently code and transform data for analysis. Data an alysis was conducted us ing SPSS 13.0 and SAS 9.1. SAS 9.1 was employed to account for missing data through multiple imputation procedures. Multiple imputation is a statistical calculati on of missing values based upon probability. The specific method utilized for analysis, Markov Ch ain Monte Carlo (MCMC), specifies that all data are missing completely at random (Rubin, 1976; Little & Rubin, 1987). In the parent survey, excluding demographic information, missing data acc ounted for as little as .003% of responses to a question and as many as 13% resp onses to a question. Missing data accounted for as little a 0% of responses to as many as 7% responses to a question in the student survey. This analysis procedure decreased the amount of data that would have been lost utilizing a procedure such as group-wise deletion. Data analysis procedures allowed for the gr ouping and comparison of between and within group responses of student and parent data. In addition, regression techniques were applied utilizing both achievement data and student academic self report variables as dependent variables to both explain and predict potential implications for policy and practice. Data Analysis Techniques Employed Research question 1: What quantifiable impact, if any, does familial involvement have on student achievement in K-12 virtual schooling? 0H : Familial involvement does not effect virtual school achievement. 54

PAGE 55

Regression procedures were empl oyed utilizing achievement data as a dependant variable and student perception and parent pe rception as independent variab les. Correlations and partial correlations were calculated to assist in the analysis of this question. Research question 2: Do students and parents differ in their perceptions of familial involvement in K-12 virtual schooling? 0H : Parental perceptions of familiar involvement do not differ from student perceptions of familial involvement. Responses between student and their parent we re compared across gr ouping variables that included: student gender, parent ge nder, race and socioeconomic status. Research question 3: Is there an interaction amongst factors, including, socioeconomic status, race and gender, when measuring the effect of parental involvement of families in virtual schooling? 0H : Race does not have a significant interaction eff ect with other factors when measuring f involvement in virtual schooling. amily 0H : Socioeconomic status does not ha ve a significant interaction effect with other factors when measuring family involvement in virtual schooling. 0H : Student gender does not have a significant interaction eff ect with other factors when measuring family involvement in virtual schooling. Two subjects multiple analysis of variance (MANOVA) procedures were employed, the first utilized student responses as dependant variables, fixed fact ors included student gender in addition to socioeconomic status, race, and pare ntal education. The seco nd procedure utilized 55

PAGE 56

parental response data as dependant variables and gender, ra ce, socioeconomic status and education as fixed factors. Limitations Limitations of this study include: The single state virtual school a ssessed, potentially preventing the generalizability of this study to other states. The survey was only provided in an English language format, thus non or limited English speaking individuals may have been unde rrepresented in the sample population. The possibility for socially desirable responses by respondents, particularly parents, exists as the nature of topical matte r covered by the survey may have lead respondents to skew responses toward a higher degree of involvement. Given that the student survey instrument adapted for this study was designed and validated upon an elementary population, there existed the possibility of a signi ficant threat to validity. The student survey was subjected to redesign and provided with a limited external review; therefore, there exis ts the possibility of measuremen t error due to poor wording or presentation of questions. 56

PAGE 57

57 Figure 3-1. A detail of the inte raction between factors from Hoover-Dempsey and Sandlers 2005 study

PAGE 58

Table 3-1. Correlations among parents reports of involvement mechanisms, student perceptions of parents involvement, and a summary measure of student ach ievement (Hoover-Dempsey, 2005) Parent report of involvement mechanisms Parent report encourag ement Parent report modelin g Parent report reinforce ment Parent report instructio n Student report parental encourage ment Student report parental model-ing Student report parental reinforcement Student report parental instruction Achie vemen t TCAP Encouragement -Modeling .67** -Reinforcement .70** .75** -Instruction .61** .72** .70** -Student report of parental involvement Encouragement ns .17** .14** ns -Modeling .11* .22** ns ..20* .59** -Reinforcement .10* .22** .16** .22** .82** .61** -Instruction ns .16** ns .17** .76** .71** .74** -Student achievement .18** .15** .25** ns ns ns ns -.12* -58 p = .05 ** p = .01

PAGE 59

Figure 3-2. Proposed path model for Hoover-Dempsey & Sandler model of parental involvement 59

PAGE 60

CHAPTER 4 RESULTS Introduction This chapter will report results obtained thr ough the investigation of the role of familial participation in student's achievement in K-12 vi rtual schools. The chapter will: (a) describe the sample and report findings associated with the res earch questions described in Chapter 1; and (b) examine the impact of familial involvement on student achievement in K-12 virtual schooling, discerning differences in student and parental pe rception of familial involvement in K-12 virtual schooling, and describing demograp hic factors that effect pare ntal involvement in virtual schooling. Sample Utilizing a list provided by the virtual school, 10,169 parents of students enrolled in 20072008 were contacted via email with a request to ta ke part in the study. In order to expedite the contact and data collection, an online survey application, Zoomerang, was utilized. The respondent population can be divided into four distinct gr oups: Parent Groups A, B, C and Students. The first group, Parent Group A, consis ts of all the parent re spondents (n = 940, a 9% survey response rate). These individuals repres ent the entire sample of parents whose child enrolled in the virtual school and responded to the parent survey. The second group, Parent Group B, (n = 776) consisted of parents who re sponded to the survey and whose child did not respond to the student survey. The third group, Pa rent Group C, (n = 164) represent parents respondents whose child also res ponded to the student survey. 665 individuals (71%) in Parent Group A elected to allow their st udent to participate in the su rvey, the students who responded comprised the fourth and final group, Student Gr oup, (n = 164). This group consisted of students who were enrolled in the virtua l school and responded to the survey. In this group, demographic 60

PAGE 61

data pertaining to the student wa s obtained through his/her parents responses. It should be noted that not all of the 665 students whose parents had allowed contact could be reached via email. The virtual school could not produce email addr esses for 180 of the 665 students, limiting the total pool of potential student respondents to 485. Thus, the student re sponse rate based on a possible 485 responses was 36%. Parent Group A: All Parent Respondents The parent group A sample consisted of 940 individuals. The respondents were asked to provide demographic information including inform ation related to gender, employment, income, average time spent at work and ed ucation. This information is su mmarized in table 4-11. Parent Group A can be described as primarily white, female, upper-middle class, well educated individuals, who are employed in full-time professional occupations The majority of respondents were female (72%). Of those providi ng employment information the largest segment (27%) of the sample described the nature of th eir employment as prof essional or executive, followed by teacher (18%) and unemployed, retired, student or disabled (11%). Only 6% of the sample reported annual household incomes of less than $30,000, while 51% of respondents disclosed household income greater than $60,000 annually. A majority of respondents reported working mo re than 20 hours during the average week, with 41% of respondents indicating that they worked more than 41 hours during the average week. With the exception of 3 respondents, the sample consisted of high school graduates. 46% of respondents had obtained a bachelors degree, and 27% of respondents had post-baccalaureate educational experience at the graduate level. Ac hievement scores were available for 452 children of responding parents; this number represen ts 48% of the sample population. The average achievement of students corresponding to pa rents in the population was 82.5 (SD = 16.40, n = 452). 61

PAGE 62

Parent Group B: Parents Whose Child Did Not Respond The parent group B sample consisted of 776 i ndividuals. Information related to this group can be found in table 4-2. This parent gr oup consists of predominately white, female, professionals employed in full-time occupations. The majority of respondents within this group are considered upper-middle class and well educated. Of those providing employment information the largest segment (28%) of the sample described the nature of their employment as professional or executive, followed by teacher (18%) and unemployed, retired, student or disabled (10%). A majority of the respondents were female (71%). 6% of the sample reported a nnual household incomes of less than $30,000, while 51% of respondents disclosed household inco me greater than $60,000 annually. The sample consisted of a racially diverse, though predom inantly white population (58%). A majority of respondents reported working more than 20 hours during the average week; 40% of respondents indicated that they worked more than 41 hours du ring the average week. W ith the exception of 2 respondents, parents who responded to the survey were high school graduates. 47% of respondents had obtained a bachelors degree, 28% of respondents ha d post-baccalaureate educational experience at the graduate level. Ac hievement scores were available for 350 children of responding parents; this number represen ts 45% of the sample population. The average achievement of students corresponding to pare nts in the population was 81.65 (SD = 17.16). This achievement data was not significantly different (p = .12) from the achievement data associated with the two parent groups (Parent Group A and Parent Group B). Parent Group C: Parents Whose Child Did Respond Parent Group C consisted of 164 individuals. De tailed information related to this parent group can be found in table 4.3. The 164 individual s comprising parent group C were primarily white females, employed full-time in professional occupations, the majority of these individuals 62

PAGE 63

had household incomes placing them in the upper middle class and possessed more than a high school education. 72% of the individuals within this group were female and 64% were white. Members of Parent Group C (Table 4-3) were employed in a wide range of categories, including professional or executive (23%), teaching (18%) and social and public services ( 12%). Only 5% of this group came from households with annual incomes less than $30,000, 55% of this group came from households with reported annual in comes of $60,000 or greater. Pa rent Group C is slightly less educated than the overall respondent populati on, 43% of these parents reported having a bachelors degree or greater. Achievement data was available on 102 students related to parents in this group (62%). The average achievement of students corresponding to parent group C was 85.6 (SD = 12.8). While this number is slightly hi gher than the average ac hievement for students associated with parent group A (82.5) and pa rent group B (81.65), ther e is no statistically significant difference in average achieveme nt (p = .12) between the three groups. 63

PAGE 64

Student Group The Student Group consisted of 164 individuals Detailed information related to these individuals can be found in Tabl e 4-4. Students participating in this study were predominately white females who come from upper-middle class homes and whose parents have both completed high school and have some formal co llegiate experience. Students parents are employed, most in full-time jobs. The preponderan ce of students report that their mothers provide the majority of s upport for their school work. Student respondents to the survey were pre dominantly female (63%) and primarily white (63%). Student respondents pare nts were employed in a wide range of categories, including professional or executive (23%), teaching (18%) and social and public services (12%). Only 5% of student respondents came from househol d with annual incomes less than $30,000, 55% of student respondents came from households with reported annual incomes of $60,000 or greater. Student respondents came from households that were slightly less educated than the overall respondent population, 43% of stude nt respondents parents reported having a bachelors degree or greater. 60% of student respondents disclose d that their mothers provided a majority of support for their school work. 13% of respondents received a majority of support from their fathers. Achievement scores were available for 102 students; this number represents 62% of the sample population. The average achievement of student respondents was 85.6 (SD = 12.8). 64

PAGE 65

A Summary of Survey Factors and Survey Reliability The purpose of this study was to investigate the role of familial participation in student's achievement in a virtual high school Three specific research ques tions guided the collection and analysis of data, question one explored the impact of familial involvement on student achievement in K-12 virtual schooling; question tw o sought to discern diffe rences in student and parental perception of familial involvement in K-12 virtual schooling; question three sought to describe demographic factors that effect parental involve ment in virtual schooling. Detailed information regarding the survey, factors and methods utilized for data collection can be found in Chapter 3. As a brief summary, the study employed an online survey adapted from research by Hoover-Dempsey and Sandler ( 2005). Data were collected from a population of virtual school students and their parents in a two-tiere d fashion: Tier one targeted parents of virtual school students. Tier tw o targeted the child enrolled in the virtual school related to a responding parent. Responses were analyzed and ach ievement data, in the form of a semester grade for the course (or average grade for the cour ses) in which the child was enrolled during the summer 2007, fall 2007 and spring 2008 semesters, were collected and evalua ted in relation to the surveys. Quantitative statistical proce dures were utilized to analyze the data. The parent survey included four separate measurement variables: Parent Report of Modeling (Modeling): The 10 items comprising this factor sought to measure the number of reciprocal interactions between parent and ch ild related to school activities (Hoover-Dempsey and Sandler, 2005). Parent Report of Instruction (Instruction): Th e 15 questions associat ed with this factor draw gauge both indirect and direct instructional activitie s undertaken by both child and parent associated with learning and a cademics (Hoover-Dempsey and Sandler, 2005). Parent Report of Encouragement (Encourag ement): The 13 questions encompassing this factor elicited information related to parent s explicit affective support for the students interest in school and learning (Hoover-Dempsey and Sandler, 2005). 65

PAGE 66

Parent Report of Reinforcement (Reinforcem ent): The 13 items compromising this factor measure parents reinforcing behaviors used to shape student characteristics associated with positive learning outcomes (Hoover-Dempsey and Sandler, 2005). The student survey also included four separate measurement variables: Student Perception of Parental Modeling of appropriate school-related skills (Student Modeling): These 10 items measured the studen ts perception of reci procal interactions between parent and child related to school activities (Hoover-Demp sey and Sandler, 2005). Student Perception of Parental Instruction (Student Instructio n): This factor was comprised of 15 items which measured the students perception of both i ndirect and direct instructional activities undert aken by both child and parent associated with learning and academics (Hoover-Dempsey and Sandler, 2005). Student Perception of Parental Encouragemen t (Student Encouragement): The 12 questions encompassing this factor elicited information related to the students perception of parents explicit affective support for the students interest in school and learning (HooverDempsey and Sandler, 2005). Student Perception of Parental Reinforcement of learning and attributes related to learning (Student Reinforcement): The 13 questions co nstituting this factor measure students perceptions of parents rein forcing behaviors used to shape student characteristics associated with positive learning outc omes (Hoover-Dempsey and Sandler, 2005). Cronbach reliability coefficients were calcula ted for each of the four parent and student measurement variables. Alpha reli abilities (Table 4-5) for each of the factors is greater than .7, indicating a satisfactory level of internal consistency across a ll variables (Nunnally & Bernstein, 1994). 66

PAGE 67

Research Question 1 What quantifiable impact does familial involvement have on student achievement in K-12 virtual schooling? : In the U.S., several virtual school s collect data from parents (W. Scott, personal communication, April 2, 2007), utilizing the information to ascertain an understanding of parental attit udes toward virtual schooling and curriculum. Unfortunately, there are no published studies that empi rically investigate this impact (Russell, 2004; Black, Ferdig & DiPietro, 2008). A better understand ing of factors related to online student achievement is of critical importance to K-12 onlin e education (NCREL, 2002), as pare ntal involvement is seen as a key component of other non-traditional forms of education, including ch arter schools and home schooling (Green & Hoover-Dempsey, 2007; Bulkle y & Fisler, 2003). In or der to effectively answer question one, three separa te regression analyses were performed (re gression analyses 13). Regression analysis one was conducted utilizi ng student achievement (in the form of student virtual school grade(s)) as a dependant vari able and the four involve ment factors reported by parents: Encouragement, Modeling, Reinforcemen t and Instruction as independent variables. The analysis detailed in Table 4-6, was c onducted on student achievement data from 452 students and their corresponding pare nt data. It reveals a non-si gnificant relationship (p = .46) between the variables and student achievement, i ndicating that the four va riables did not predict student achievement within this sample (Adjusted R = -.001, F(5, 447) = .89, p = .46). 2The second regression analysis (regressi on two) was conducted utilizing student involvement factors measured by the survey. In this analysis, student achievement (in the form of student virtual school grade(s)) served as a depe ndant variable and there were four independent variables: Student Encouragement, Student Modeling, Student Reinforcement and Student 67

PAGE 68

Instruction. The analysis detailed in table 4-7, was conduc ted upon 101 students. This reduced sample size was related to grade data which could not be obtained from the Student response data reveals a non -significant relationship (p = .21) indicating that within this sample the four variables did not pr edict student achievement (Adjusted R = .02, F(5, 96) = 1.47, p = .21). 2A third and final regression an alysis (regression three) was conducted utilizing both parent and student involvement factors elicited by the su rvey (Table 4-8). In this analysis student achievement (in the form of student virtual school grade(s)) served as the dependant variable and there were eight independent variables: Encour agement, Modeling, Reinforcement, Instruction, Student Encouragement, Student Modeling, Studen t Reinforcement and Student Instruction. The analysis conducted on 101 stude nts and their corresponding parent data reveals a significant relationship (p = .03) between the variables and student achievement, within this sample, the regressed variables explained 9.4% of the variance in student achievement (Adj R2 = .09, F(9,83) = 2.31, p = .02). Specifically, two of the ei ght independent variab les held significant value in the prediction of student achievement. Both of these vari ables were parent variables: reinforcement (B = .58) and instruction (B = .61). The analysis indi cates that a one point increase in parent reinforcement as scored by the survey will translate to a .583 point increase in student achievement. This result i ndicates that there is a positive relationship between a parents perception that they praise their child for scholastic activities and a childs academic achievement in a virtual school course. Conve rsely, for each one point increase in parent instruction as score by the survey will translate to a .61 decrease in student achievement. This result indicates that there is a negative relatio nship between a parents perception that they engage in instructional activities, such as teachi ng their child to engage in challenging problem 68

PAGE 69

solving activities and to persist when a speci fic task is challenging, and a childs academic achievement in a virtual school course. The results of regression one and regression three provide context fo r the assertion that there is a conflicting result associated with res earch question one. The results of regression one, conducted upon the 452 members of Parent Group A for which achievement data existed for their corresponding student, prov ide evidence that there is no relationship between parental involvement and student achievement. Resu lts of regression three, conducted upon a subpopulation of the sample (n = 164) utilized in regression one (Par ent Group C), directly contradicts this outcome, indicating that within this sample, there is indeed a relationship between parental involvement a nd student achievement. These resu lts are detailed in Figure 4-1. Due to this discrepancy, a mo re granular investigation of the analysis was conducted. This investigation focused on differences in the independent variables associated with the samples. As the data were drawn from the same population, because of the difference in prediction, it was plausible to assume that a difference exists betw een the parents in the two samples (parents of students who did not respond to the survey (Par ent Group B) and parent s of students who did respond to the survey (Parent Group C)). The stud ents could be eliminated from scrutiny as a differentiating factor, as student results were fou nd to be non-significant. Several issues could contribu te to the outcome discrepanc y described above, including parent and student demographics, differences in course content, instructors and schools attended by corresponding students in each group. Given the limitations of data collection, data were collected upon some, but not all of these potential variables. Therefore, survey factors and available demographics were analyzed to id entify differences between members of Parent Groups B and C. These factors and demogra phics included: average reinforcement and 69

PAGE 70

instruction scores for the two parent groups, ra ce and ethnic data describing the members of the groups, and gender composition and income st ratification information for the groups. Average scores for reinforcement and instruction for parents in these groups are featured in table 4-9. An analysis of variance upon these factors reveals nonsignificant differences between the two groups of respondents for both the reinfo rcement and instruction factors, F(1,940) = .49, p = .48 and F(1,940) = .17, p = .68 respectively. Analysis of income data between the groups reveals no statistically significant difference between the income distributes of parents of non-responding students and parents of responding students (p = 1.00 for the low, middle and high income classifications). An alysis of differences in the gender composition of the groups also reveals no statistically significant differences, nor were statistical differences found in the edu cational levels between parents of non-responding and responding students. An analysis of demographic ch aracteristics reveals significant differen ces in the racial make up of the groups. Group means are detailed in table 4-10, statistical results of a between groups analysis are featured in table 4-11; results indicate the following st atistical differences: p = .01 for Asian/Asian Americans, p = < .01 for Black/African Americans, p = .05 for Hispanic/Hispanic Americans, p = < .01 for White/Caucasian Americans and p = <.01 for individuals describing themselves as Other. Thus, available information indicates that ther e are statistically significant differences in racial and ethnic composition between the members of parent groups B and C. Specifically; there are more individuals in Parent Group B who identify themselves as Asian/Asian-American and Black/African-American. In Parent Group C there ar e more individuals who describe themselves 70

PAGE 71

as Hispanic/Hispanic-American, White/Caucasian and Other. Each of these differences is statistically significant. The results related to research question 1 reveal conflicting info rmation regarding the impact of parental involvement on student achievement in the research sample. Two different outcomes can be concluded based upon the anal yses, these outcomes point to parental involvement having both a significant and a non-significant relations hip to student achievement, depending upon the sample analyzed. A summary of the results is presented in table 4-12. Research Question 2 Do students and parents differ in their p erceptions of familial involvement in K-12 virtual schooling? : The theoretical models built upon durin g this study were developed and utilized with elementary and mi ddle school populations, thus, it is important to assess the ability of the survey instrument utilized to effectivel y measure student and parent perceptions. Further, by understanding the base-line nature of both parent and student perceptions of involvement it is then possible to plan for further research and begi n to construct and test interventions that work to mediate perceptions between parents and students. In order to effectively ad dress research question two, a within-subjects analysis of variance was conducted, utilizi ng two groups, student group and pa rent group C, with the four measurement variables nested in each group. The an alysis indicates that the data do not violate the sphericity assumption ( = 1.00). Sphericity is a fundament al property necessary for withinsubject ANOVA procedures, sphericity assumes that there is a homogeneity of variance between the levels of the factors utilized in this statistical analysis (Max & Onghena, 1999). The results of the within-subjects ANOVA reveal a statistically significant indication, F(1,163) = 1510.37, p < .00, R2= .90, E = .90, that students and parents do indeed have differing perceptions regarding 2 71

PAGE 72

the amount familial involvement present in thei r K-12 school work. Examination of group means across the four factors (Table 413 and Figure 4-2) reveals that st udents tend to rate a parents involvement lower than parents ra te their own involvement. This conclusion is supported (Table 4-14) by a statistically signifi cant linear trend to the fact or data: F(1,163) = 1242.22, p <.00, R2= .88, E = .88. Finally, correlation data provides evidence of a statistic ally significant relationship between three of the four variables (Table 4-15). 2The analysis associated with research ques tion 2 revealed that parents and K-12 virtual school students have statistical differences in their perceptions of the amount of involvement parents have in their academic work. Outcomes i ndicate that parents data is compared to student data, the parents perceive themselves to be more involved in their childrens academic. Research Question 3 Do factors including, socioeconomic status, race and gender, effect involvement of families in virtual schooling? : Hoover-Dempsey and Sandlers (1995) research, which this study draws heavily upon, minimizes factors such as parents' demographics (eg: education, income, marital status, and other indicators of family status) in efforts to understand parents' involvement decisions. Nevertheless the role of social and economic variables remains a source of considerable controversy in research concer ning parental involvement in traditional schools (Lareau, 2000; 1989). Therefore demographics fact ors influence on familial involvement in virtual schooling was an important aspect of this study. An analysis of demographic factors reveals several interact ions between demographic variables; an interaction exists when the association between two variables changes as the va lue of a third variable increases or decreases (Agresti & Finley, 1999). Prior to an investigation of the eff ect of individual factors, the interaction between factors must be explored. 72

PAGE 73

In the investigation associated with research question three, two separate between subjects analyses were conducted. The first analysis fo cused on student respondents and the demographic variables associated with thes e individuals (gender, race/et hnicity, education and household income). Results of the analysis (Tables 4-16 and 4-17) of students reveal a statistically significant interaction between gender and education across all four vari ables (student modeling, student instruction, student encour agement and student reinforcement) Results of the analysis of parent data reveal a statistically significant inte raction between gender and race/ethnicity for all four variables (modeling, instru ction, encouragement and reinforcement). When the data for these interactions were subjected to closer scrutiny it was apparent that outliers were influencing the results of both ANOVAs. An outlier is an observ ation that lies outside the overall pattern of a distribution (Moore & McCabe, 1999) There is no strict mathematical definition of what constitutes an outlier; the process of determini ng whether or not data should be considered an outlier is a subjective exercise (Hill & Lewicki, 2007). The firs t outlier, was an Asian/AsianAmerican male parental respondent (n=1) whose responses to the parent survey score far below the rest of the sample in each of the four pa rent factors (modeling, instruction, encouragement, and reinforcement). The second outlier, a student (n=1) whose parent is female with a doctoral degree, generated student survey scores far belo w the rest of the sample in each of the four student factors (student mode ling, student instruction, student encouragement, and student reinforcement). A detailed analys is of the data with the outl iers included can be found in Appendix B. The student and parent outliers were eliminated from both analyses in order to remove their influence upon the sample. The data was re analyzed utilizing be tween subjects ANOVAs, the first analysis focused on the parental va riables. This ANOVA provided no evidence of a 73

PAGE 74

statistically significant relationship betw een the study factors (modeling, instruction, encouragement and reinforcement) and the demogr aphic data (gender, ra ce/ethnicity, education and household income). This result indicates that there is no evidence, within the sample, of differences in parent gender, race/ethnicity, parent education or household income having an effect on a parents perception of involvement. The second ANOVA focused on the student respondents and th e demographic variables associated with these individuals (gender, ra ce/ethnicity, education a nd household income). Results (table 4.19) indicate a gender by income interaction, detailed in table 4.18, for the four dependent variables (student encouragement, st udent modeling, student in struction and student reinforcement). In addition, results indicate a sign ificant main effect between student instruction and income, which is described in table 4-18. These interactions reveal two distinct patterns, (a) male st udent respondents in the upper (>$60,000 annually) and lower income (< $30,000 annually) categories perceived a higher degree of encouragement, modeling, instruction and reinforcement from parents than male student respondents in the middle income category (>$30,000 < $60,000 annually) and (b) female student respondents in the middle inco me category perceived higher levels of encouragement, modeling, instruction and reinforcement from parents than female student respondents in the lower and upper income categories. Charts 4.4 4.8 describe the interaction effect between gender and income across all four variables. Interaction 1 The first interaction, interaction 1, details the relationship between student modeling and household income, which is influenced by student gender (Figure 4-3). ANOVA results indicate a statistically significant inte raction between average studen t encouragement and household income and gender, F(2, 164) = 3.87, P = .05. 74

PAGE 75

Interaction 2 The second interaction, interaction 2, details the relationship between student instruction and household income, which is influenced by student gender (Figure 4-4). ANOVA results indicate a statistically significant interaction between aver age student instruction and household income and gender, F(2, 164) = 5.36, P = .02. Interaction 3 The third interaction, interaction 3, details th e relationship between student encouragement and household income is influenced by student gender (Figure 4-5). ANOVA results indicate a statistically significant intera ction between average student encouragement and household income and gender, F(2, 164) = 5.51, P = .02. Interaction 4 The forth and final interacti on, interaction 4, deta ils the relationship between student reinforcement and household income, which is influenced by student gender (Figure 4-6). ANOVA results indicate a statistically signi ficant interaction between average student encouragement and household income and gender, F(2, 164) = 6.61, P = .01. Main Effect The main effect, describes the effect of an i ndependent variable, inco me in this instance, upon a dependant variable, student interaction. The ANOVA results indicate that there is a difference in student instruction based upon household income level. Th is effect, detailed in table 4-19 and figure 4-7 provides evidences that w ithin the sample, students in lower and upper income levels perceived parents offering a similar level of instructional activities, while students from middle income households reported lower levels of parental instructi on. Parental instruction was measured by obtaining information related to a students perception of parental guidance related to activities such as homework and studying strategies. 75

PAGE 76

The results of question 3 reveal several distinct phenomena, (a) male student respondents in the upper (>$60,000 annually) and lower income (< $30,000 annually) categories perceived a higher degree of encouragement, modeling, inst ruction and reinforcement from parents than male student respondents in the middle income category (>$30,000 < $60,000 annually) and (b) female student respondents in the middle inco me category perceived higher levels of encouragement, modeling, instruction and reinforcement from parents than female student respondents in the lower and upper income categories. Finally, (c ) a significant effect was found related to household income, this income effect closely mirrors the interaction effects described previously, but it is a main effect, only conc erning a single variable (household income). The main effect indicates that students in lo wer (< $30,000 annually) and upper (>$60,000 annually) income categories perceive greater levels of instruction from their parents than individuals in middle income categories (>$30,000 < $60,000 annually). Summary of Findings The three research questions sought to (a) ex plore the impact of familial involvement on student achievement in K-12 virt ual schooling; (b) discern differe nces in student and parental perception of familial involvement in K-12 virtual schooling and; (c) describe demographic factors that effect parental i nvolvement in virtual schooling. The results of question one, the goal of which was to explore the impact of familial involvement on student achievement, provided a conflicting picture. Within the larger sample of parents and both responding and no n-responding students, no signi ficant statistical relationship (p = .46) was found to exist between the familial involvement factors and student achievement in K-12 virtual education. Within a smaller sample consisting of parents and students who did respond to the student survey, two factors were found to be predictive of student achievement, parent reinforcement and parent instruction (p = .026). Based upon the outcomes of the analyses, 76

PAGE 77

demographic differences between the parents in both samples were explored. This exploration provided evidence that there were significant diff erences (p < .01 p = .05) in the racial/ethnic composition of the parent groups in the samples. Results of research question two, which investigated differences in parent and student perception of familial involvement reveals that parents and K-12 virtual school students do indeed have statistically signifi cant differences (p < .00) in thei r perceptions of the amount of involvement parents have in their academic wo rk. Within the sample, on average, students described a much lower level of invo lvement than their corresponding parent. The third and final research questions goal was to illustrate demographic factors effect on parental involvement in virtual schooling. This analysis provided ev idence that household income plays a direct role (p = .05) in the amoun t of parental instructi on student perceive they are receiving. The pattern of this direct effect mirrors significant interaction effects occurring with student perceptions of pare ntal involvement. These interaction effects reveal disparities in the way male and female students perceive the amount of encouragement, modeling, instruction and reinforcement they are receiving from their parents based upon household income. The pattern that emerges within the research indicates that male students perceive higher levels of involvement in lower and upper income categories, than male individuals in middle income categories. Conversely, female students respondents in the middle income category perceived higher levels of encouragement, modeling, inst ruction and reinforcement from parents than female student respondents in the lower and upper income categories. The outcomes presented within th is chapter provide evidence that family involvement and family demographics are indeed el ements to that need to be c onsidered when discussing factors that promote student success in K-12 virtual school s. In addition, the results of this study reveal 77

PAGE 78

78 that the Hoover-Dempsey and Sadler (2005) asse ssment, adapted and utilized for this study, elicited internally reliable results. These results we re applied to three research questions that seek to provide a basis for investigating the role of familial participation in students achievement in K-12 virtual schools.

PAGE 79

Table 4-1. Parent group A demographics Male 135 14% Female 674 72% Gender Missing 131 14% Asian/Asian-American 21 2% Black/African-American 172 18% Hispanic/Hispanic-American 20 2% White/Caucasian 561 60% Other 21 2% Ethnicity Missing 145 15% Unemployed, retired, student, disabled 103 11% Labor, custodial, maintenance 7 0% Factory worker, construction 9 1% Driver (taxi, delivery, bus, truck) 8 1% Food service, restaurant 4 0% Skilled craftsman (plumber, etc) 13 1% Retail sales, customer service 32 3% Service technician (cars, appliances, etc) 4 0% Accounting, bookkeeping 64 7% Creative arts (writer, musician) 13 1% Sales (real estate, co mmodity goods, etc) 27 3% Social services, public services 78 8% Teacher 171 18% Professional, executive 253 27% Employment Missing 154 16% Lower (< $30,000) 53 6% Middle ($30,000-$60,000) 167 18% Upper (> $60,000) 469 51% Household Income Missing 225 25% 0-5 92 10% 6-20 72 8% 21-40 252 27% 41-50 292 31% 50 or more 94 10% Average Hours Worked During the Week Missing 138 15% Less than high school 3 0% High school or GED 82 9% Some college, 2-year college/vocational 292 31% Bachelor's degree 179 19% Some graduate work 54 6% Master's degree 163 17% Doctoral degree 40 4% Parent's Educational Attainment Missing 127 14% 79

PAGE 80

Table 4-2. Parent group B demographics Male 112 14% Female 553 71% Gender Missing 111 14% Asian/Asian-American 18 2% Black/African-American 143 18% Hispanic/Hispanic-American 16 2% White/Caucasian 447 58% Other 15 2% Ethnicity Missing 126 18% Accounting, bookkeeping 48 6% Creative arts (writer, musician) 10 1% Driver (taxi, delivery, bus, truck) 5 1% Factory worker, construction 6 1% Food service, restaurant 4 1% Labor, custodial, maintenance 6 1% Missing 144 19% Professional, executive 213 28% Retail sales, customer service 24 3% Sales (real estate, co mmodity goods, etc) 23 3% Service technician (cars, appliances, etc) 4 1% Skilled craftsman (plumber, etc) 11 1% Social services, public services 59 8% Teacher 138 18% Employment Unemployed, retired, student, disabled 81 10% Lower (< $30,000) 48 6% Middle ($30,000-$60,000) 137 18% Upper (> $60,000) 385 51% Household Income Missing 206 24% 0-5 68 9% 6-20 55 7% 21-40 203 26% 41-50 243 31% 50 or more 73 9% Average Hours Worked During the Week Missing 133 17% Less than high school 2 0% High school or GED 62 8% Some college, 2-year college/vocational 231 30% Bachelor's degree 144 19% Some graduate work 43 6% Master's degree 133 17% Doctoral degree 37 5% Parent's Educational Attainment Missing 123 16% 80

PAGE 81

Table 4-3. Parent group C demographics Male 24 15% Female 118 72% Gender Missing 22 13% Asian/Asian-American 2 1% Black/African-American 27 16% Hispanic/Hispanic-American 5 3% White/Caucasian 103 63% Other 5 3% Ethnicity Missing 22 13% Accounting, bookkeeping 15 11% Creative arts (writer, musician) 3 2% Driver (taxi, delivery, bus, truck) 3 2% Factory worker, construction 3 2% Food service, restaurant 0 0% Labor, custodial, maintenance 1 1% Missing 24 15% Professional, executive 37 23% Retail sales, customer service 8 6% Sales (real estate, co mmodity goods, etc) 3 2% Service technician (cars, appliances, etc) 0 0% Skilled craftsman (plumber, etc) 2 1% Social services, public services 17 12% Teacher 29 18% Parental Employment Unemployed, retired, student, disabled 19 10% Household Income Lower (< $30,000) 8 5% Middle ($30,000-$60,000) 32 20% Upper (> $60,000) 92 55% Missing 32 20% 0-5 22 13% 6-20 14 9% 21-40 43 26% 41-50 47 29% 50 or more 19 12% Average Hours Worked During the Week by Parent Missing 19 12% High school or GED 18 11% Some college, 2-year college/vocational 56 34% Bachelor's degree 34 21% Some graduate work 10 6% Master's degree 23 14% Doctoral degree 3 2% Missing 29 11% 81

PAGE 82

Table 4-4. Student group demographics Male 60 37% Female 104 63% Gender Asian/Asian-American 2 1% Black/African-American 27 16% Hispanic/Hispanic-American 5 3% White/Caucasian 103 63% Other 5 3% Ethnicity Missing 22 13% Accounting, bookkeeping 15 11% Creative arts (writer, musician) 3 2% Driver (taxi, delivery, bus, truck) 3 2% Factory worker, construction 3 2% Food service, restaurant 0 0% Labor, custodial, maintenance 1 1% Missing 24 15% Professional, executive 37 23% Retail sales, customer service 8 6% Sales (real estate, co mmodity goods, etc) 3 2% Service technician (cars, appliances, etc) 0 0% Skilled craftsman (plumber, etc) 2 1% Social services, public services 17 12% Teacher 29 18% Parental Employment Unemployed, retired, student, disabled 19 10% Household Income Lower (< $30,000) 8 5% Middle ($30,000-$60,000) 32 20% Upper (> $60,000) 92 55% Missing 32 20% 0-5 22 13% 6-20 14 9% 21-40 43 26% 41-50 47 29% 50 or more 19 12% Average Hours Worked During the Week by Parent Missing 19 12% Less than high school 1 1% High school or GED 18 11% Some college, 2-year college/vocational 56 34% Bachelor's degree 34 21% Some graduate work 10 6% Master's degree 23 14% Doctoral degree 3 2% Parent's Educational Attainment Missing 29 11% 82

PAGE 83

83 Table 4-5. Reliability coefficients for survey variables Variable Reliability Parent Report of Modeling .91 Parent Report of Instruction .88 Parent Report of Encouragement .90 Parent Report of Reinforcement .93 Student Perception of Modeling .92 Student Perception of Instruction .94 Student Perception of Encouragement .91 Student Perception of Reinforcement .92

PAGE 84

Table 4-6. Results of regression one, expl oring the relationship between achievement and encouragement, modeling, reinforcement and instruction. R R Square Adjusted R Square Std. Error of the Estimate 0.09 0.008 -0.001 16.43 Sum of Squares df Mean Square F Sig. Regression 969.91 4 242.47 0.89 0.46 Residual 120,740.27 447 270.11 Total 121,710.18 451 Unstandardized Coefficients Standardized Coefficients Correlations B Std. Error Beta t Sig. Zero-order Partial Part (Constant) 71.68 7.31 9.8 0 Modeling 0.24 0.19 0.09 1.27 0.2 0.08 0.06 0.06 Instruction -0.08 0.11 -0.05 -0.76 0.44 0.02 -0.03 -0.03 Encourage ment -0.01 0.13 0 -0.1 0.91 0.04 -0.005 -0.005 Reinforce ment 0.06 0.14 0.03 0.43 0.66 0.06 0.02 0.02 84

PAGE 85

Table 4-7. Results of regression two, exploring th e relationship between student achievement and student encouragement, student modeli ng, student reinforcement and student instruction. R R Square Adjusted R Square Std. Error of the Estimate .24 .06 .02 13.18 Sum of Squares df Mean Square F Sig. Regression 1023.68 97 255.92 1.47 .21 Residual 16871.19 101 173.93 Total 17894.87 Unstandardized Coefficients Standardized Coefficients Correlations B Std. Error Beta t Sig. Zeroorder Partial Part (Constant) 68.26 8.09 8.43 0 Modeling 0.002 0.412 0.001 0.006 0.99 0.18 0.001 0.001 Instruction 0.14 0.36 0.09 0.37 0.7 0.21 0.038 0.037 Encouragement -0.18 0.47 -0.103 -0.38 0.7 0.2 -0.039 -0.038 Reinforcement 0.39 0.37 0.25 1.06 0.29 0.23 0.1 0.1 85

PAGE 86

Table 4-8. Results of regression exploring the relationship between encouragement, modeling, reinforcement, instruction, stude nt achievement and student en couragement, student modeling, student reinforcement and student instruction. R R Square Adjusted R Square Std. Error of the Estimate 0.40 0.16 0.09 12.67 Sum of Squares df Mean Square F Sig. Regression 2,970.74 8 371.34 2.31 0.03 Residual 14,924.13 93 160.47 Total 17,894.87 101 Unstandardized Coefficients Standardized Coefficients Correlations B Std. Error Beta t Sig. Zeroorder Zeroorder Std. Error (Constant) 71.17 14.37 4.95 0.000 Modeling -0.29 0.46 -0.11 0.65 0.518 0.008 0.067 -0.06 Instruction -0.61 0.22 -0.44 2.82 0.006 -0.13 -0.28 -0.27 Encouragement 0.26 0.25 0.16 1.07 0.289 0.06 0.11 0.10 Reinforcement 0.58 0.27 0.36 2.14 0.035 0.11 0.22 0.20 Student Modeling -0.10 0.43 -0.04 0.25 0.804 0.19 -0.03 -0.02 Student Instruction 0.30 0.35 0.20 0.86 0.391 0.21 0.09 0.08 Student Encouragement -0.35 0.47 -0.19 0.74 0.464 0.20 -0.07 -0.07 Student Reinforcement 0.42 0.36 0.26 1.15 0.255 0.24 0.12 0.11 86

PAGE 87

Figure 4-1. A representation of the conflicting results associat ed with research question 1. 87

PAGE 88

Table 4-9. Mean reinforcement and in struction for parent groups B & C Group Factor Mean Standard Deviation Parent Group B Reinforcement 71.69 8.58 Parent Group B Instruction 75.50 10.37 Parent Group C Reinforcement 72.20 8.34 Parent Group C Instruction 75.88 9.66 Table 4-10. Racial/ethnic data for parent groups B & C Parent Group B Parent Group C Race/Ethnicity n % n % Asian/Asian-American 7 1.50% 1 1% Black/African-American 87 19.20% 14 13.70% Hispanic/Hispanic-American 5 1.10% 2 2% White/Caucasian 275 60.80% 67 65.70% Other 14 3.10% 5 4.90% 88

PAGE 89

Table 4-11. Wilcoxon W for parent groups B & C Asian/Asian American Black/African American Hispanic/Hispanic American White/ Caucasian Other Wilcoxon W 21.00 2701.00 6.00 21736.00 45.00 Z -2.45 -9.27 -2.00 -16.58 -3.61 Significance 0.01 0.00 0.05 0.00 0.00 Table 4-12. A summary of regression results, research question 1 Regression Independent Variables Results 1 Modeling, Instruction, Encouragement, Reinforcement Non-significant relationship with student achievement 2 Student Modeling, Student Instruction, Student Encouragement, Student Instruction Non-significant relationship with student achievement 3 Modeling, Instruction, Encouragement, Reinforcement, Student Modeling, Student Instruction, Student Encouragement, Student Instruction Significant relationship with student achievement Table 4-13. Descriptive statistics student group and parent group C Group Factor Dependent Variab le Mean Std. Deviation N Student Group 1 Student Modeling 33.03 5.56 164 2 Student Instruction 49.07 8.70 164 3 Student Encouragement 39.87 7.46 164 4 Student Reinforcement 44.68 8.20 164 Parent Group C 1 Modeling 54.85 5.57 164 2 Instruction 76.68 9.66 164 3 Encouragement 70.05 8.32 164 4 Reinforcement 71.91 8.43 164 89

PAGE 90

Table 4-14. Test of within subjects c ontrasts, student group and parent group C Source Student Parent Factor df Mean Square F Sig. Partial Eta Squared Group (Student vs. Parent C) Linear 1 233,985.10 1,510.37 0.00 0.90 Error(Student Parent C) Linear 163 154.92 Factors Linear 1 20,278.44 1,242.22 0.00 0.88 Quadratic 1 19,935.62 821.36 0.00 0.83 Cubic 1 23,817.19 1,189.33 0.00 0.88 Error(Factors) Linear 163 16.32 Quadratic 163 24.27 Cubic 163 20.03 Student Parent C Factor Linear Linear 1 1,451.47 110.51 0.00 0.40 Quadratic 1 1,567.89 107.84 0.00 0.39 Cubic 1 21.59 1.13 0.29 0.007 Error(Student Parent C Factor) Linear Linear 163 13.13 Quadratic 163 14.54 Cubic 163 19.16 90

PAGE 91

2 1 1 = Students 2 = Parents 80 70 60 50 40 30 Estimated Marginal Means 72 45 70 40 76.7 49.2 54.9 33.2 Reinforcement Encouragement Instruction Modelingfactor Estimated Marginal Means of With Subjects Anova Figure 4-2. Visualizing the perceived differences in involvement between students and parents 91

PAGE 92

Table 4-15. Correlations between parent and corresponding student study factors Variable Parent Modeling Parent Instruction Parent Encouragement Parent Reinforcement Student Modeling .21* Student Instruction .20* Student Encouragement .19* Student Reinforcement .15 p .05 Table 4-16. Between subjects ANOVA: Student data with outlier Source Dependent Variable Type III Sum of Squares df Mean Square FSig. Student Modeling 501.81 5 100.36 3.71 0.004 Student Instruction 1,134.44 5 226.88 3.44 0.007 Student Encouragement 800.00 5 160.00 3.05 0.014 Gender Education Student Reinforcement 1,083.70 5 216.74 3.38 0.008 Table 4-17. Between subjects ANOVA: Pare nt data with outlier Source Dependent Variable Type III Sum of Squares df Mean Square FSig. Modeling 634.15 4 158.53 3.85 0.004 Instruction 1,790.13 4 447.53 4.00 0.003 Encouragement 827.79 4 206.95 2.50 0.041 Gender Race/Ethnicity Reinforcement 1,154.48 4 288.62 3.36 0.010 Table 4-18. Gender by income inte raction (student factors) Source Dependent Variable Type III Sum of Squares df Mean Square FSig. Student Modeling 108.02 2 108.02 3.87 0.05 Student Instruction 347.43 2 347.43 5.36 0.02 Student Encouragement 276.86 2 276.86 5.51 0.02 Gender Income Student Reinforcement 440.23 2 440.23 6.61 0.01 92

PAGE 93

Upper Middle Lower income 34 32 30 28 26 24 22 Student Modeling Female MaleGender Figure 4-3. The interaction between m odeling and student gender and income 93

PAGE 94

Upper Middle Lower income 50 45 40 35 Student Instruction Female MaleGender Figure 4-4. The interaction be tween student instruction a nd student gender and income 94

PAGE 95

Upper Middle Lower income 44 42 40 38 36 34 32 Student Encouragement Female MaleGender Figure 4-5. The interaction betw een student encouragement a nd student gender and income 95

PAGE 96

Upper Middle Lower income 50 45 40 35 Student Reinforcement Female MaleGender Figure 4-6. The interaction be tween student reinforcement and student gender and income Table 4-19. Income main effect (Student instruction only) Source Dependent Variable Type III Sum of Squares df Mean Square FSig. Income Student Instruction 400.84 2 200.42 3.09 0.05 96

PAGE 97

46.5 47 47.5 48 48.5 49 49.5 50 Lower Middle Upper IncomeAverage Student Instruction Figure 4-7. Income main effect for student instruction 97

PAGE 98

CHAPTER 5 DISCUSSION AND IMPLICATIONS Introduction In an effort to understand the impact of familial involvement upon virtual school achievement, this study employed an online surv ey adapted from research by Hoover-Dempsey and Sandler (2005), sampling parents from the a virtual school in th e Southeastern U.S. Presently, several virtual schools co llect data from parents, utiliz ing the information to ascertain an understanding of parental attitudes toward virtual schooling and curr iculum. Unfortunately, there are no published studies that empirically inves tigate this impact (Russell, 2004; Black, Ferdig & DiPietro, 2008). A better understanding of factors related to online student achievement is of critical importance to K-12 online education (NCREL, 2002) Parental involvement is seen as a key component of other non-traditi onal forms of education, including charter schools and home schooling (Green & Hoover-Dempsey, 2007; Bulkley & Fisler, 2003). There is reason to suspect that family involvement also plays an equal, if not more important role in student achievement within U.S. virtual schools (Ru ssell, 2004). In order to explore th is concept, data was collected from a sample of virtual school students and th eir parents in a two-tier ed fashion: Tier one targeted parents of virtual school students, tier two targeted the child enrolled in the virtual school related to a responding parent. Responses were analyzed and achievement data, in the form of a semester grade for the course (or aver age grade for the courses) in which the child was enrolled, was collected and eval uated in relation to the surveys. Quantitative statistical procedures were utilized to an alyze the data. This chapter will discuss the outcomes and review the implications associated with the three re search questions providi ng foundation to this study, these questions sought to 98

PAGE 99

. The chapter will also discuss and review the implications associ ated with findings elicited by this study not associated with the three research questions; and fina lly, discuss and review the broader outcomes and implications associated with the results of this study. Research Question 1 What quantifiable impact does familial involvement have on student achievement in K-12 virtual schooling? : Several researchers, in cluding Del Litke (1998), Clark (2001), Russell (2004), Cavanaugh et al. (2004) and Rose and Gallup (2001) have postu lated that parents play an important role in virtual stude nt achievement. According to Ru ssell, parents constitute the physical presence normally occupied by the teacher in traditional classroom environments. This study represents the first comprehensive empirica l investigation of pare nts effect on virtual school achievement. The results re lated to research question one, which seeks to quantify the effect of familial involvement, reveal conflic ting outcomes regarding the impact of parental involvement on student achievement in the re search sample. Data reveals that familial involvement was predictive of student achieveme nt in a subset of parents (Parent Group C, Parents Whose Child Did Respond to the Survey, n=164). But, in the larger parent group (Parent Group A, All Parents, n=940) there was no statisti cally significant relations hip between parental involvement and student achieveme nt. Within Parent Group C, two variables, parent instruction and parent encouragement held statistical si gnificance. Further, no link between students perceptions of parental involve ment could be made to student achievement. The conflicting results support Del Litkes (1998) assertion that the familys role and impact on virtual schooling 99

PAGE 100

is considerably complex and confirm the myriad of ideas and strategi es found in traditional schooling research regarding th e role of the parent (Ja ynes, 2005; Fan & Chen, 2001; McLaughlin, 2006). In light of the results of this research study, it can be claime d that Russell (2004) may have overstated the role of the parent in virtual schooling. Though, it is worth noting that Russell was referring to students who were en rolled in virtual school for all of their academic curricula, the students involved in this study have the ability to take, at most, two on line courses in a given semester. Thus, the majority, it not all, of students studied were receiving instruction in traditional schooling environments. The lack of predictive effect between parental involvement and its four antecedent variables (parent reports of modeling, instruction, encouragem ent and reinforcement) in Parent Group C would, on the surface, seem to indicate that virtual schools could ignore parents and their contributions to achievement online. This could not be further from the truth, as will be discussed; there are several reason s why this result should not be he ld as doctrinaire. First, metaanalytic research by Fan and Chen (2001) and Ja ynes (2007), while contradicting the previously disclosed non-predictive outcome associated with Parent Group A, provide evidence of the inconsistencies associated with quantitative research focused on parental involvement. Fan and Chen (2001) in particular discuss discretion that should be exercised when individual measures of achievement (eg: a singular course grade as wa s used in this study) are utilized as dependant variables due to their instability. Second, a robust analysis of research by Desforges and Abouchaar (2003) and Cotton and Reed-Wikelund (1989) provide indication that the most efficacious forms of parent involvement are active forms of involvement, those which encourage parents to interf ace directly with their 100

PAGE 101

children on learning activities in the home. Active forms of i nvolvement, which include phone calls with teachers, parent-t eacher conferences and homeschool communication logs produce greater achievement benefits than passive fo rms of involvement. Unfortunately, these active involvement activities place a considerable burde n upon teachers. While data was not collected regarding AVS course enrollments or student-to-t eacher ratios, student-teach er ratios associated with other K-12 Virtual Schools can approach 185:1 (Florida TaxWatch, 2007). With such a large number of students it is nearly impossible for instructors to facilita te active, meaningful involvement and communication with parents. Virt ual schools purport to facilitate involvement by providing parents entre to their childs content, grades and other course information through access to the schools learning management syst em, however, there is no empirical data evaluating parental use of this information or its impact on student achievement. Further, there is evidence that parental involvement declines as students approach adulthood (Eccles & Harold, 1996), which is partic ularly interesting as this study focused on secondary school students. This issue may account for some of the conflict ev ident in the results. Additionally, as reported in Chapter 4, Parent Group A contained a larger number of minorities as compared to Parent Group C. Research by Ep stein (1990), Griffith ( 1998) and Lareau (1987) associate minority group membership with lower levels of parental involvement, possibly contributing to the non-significan t nature of the results. Other variables unaccounted for in the study could also have contributed to the discrepancy. As discusse d in Chapter 4, data pertaining to teachers, content and community was not co llected, this information has shown to have considerable influence on parental involvement in traditional schooling (Eccles & Harold, 1996) and could have contributed to th e non-significant relationship. 101

PAGE 102

Finally, anecdotal evidence drawn from in formation published by the Florida Virtual School (Optimal Performance, Inc., 2007), provides indication that parental outreach programs that include nothing more than a monthly phone call to parent s of virtual school students contribute to high levels of pare ntal satisfaction. These satisfie d parents can aid in a virtual schools efforts to market its services, turning pa rents into an effective public relations machine (Bulkley & Fisler, 2003 ; Lubienski, 2005). When considering the other half of the conflicting result, the sample consisting of parent respondents whose children participated in the survey (Par ent Group C), a relationship was found between parental involvement a nd student achievement. Within Pa rent Group C, two variables, parental instruction and parent al encouragement, account for 9.4% of the variance found in a corresponding students achievement, a number representing the approximate equivalent of one letter grade. Russell (2004) theorized that due to a virtual school instructors lack of physical presence, a portion of the responsib ility for a students education is placed upon the family in a virtual schooling model. The outcomes evident with Parent Group C support Russells thoughts. Further, these results mirror Antoscas (1996) re search and partial results of Hoover-Dempsey and Sandlers (2005) studies. In these two examples (Antosca and Hoover-Dempsey and Sandler) parental perception of involvement held a significant relationship to student achievement. In order to further explore the outcomes associated w ith Parent Group C in research question one, the significant va riables will be discussed individually. Parental Instructions Negative Relationship With Student Achievement As previously disclosed, two variables co mprising parental involvement were found to have significance with in Parent Group C. The first variable, parental instruction, held a negative relationship with student achievement. That is, when parents reported engaging in more instructional activities with their child, the ch ilds achievement scores decreased. Pomerantz, 102

PAGE 103

Moorman and Litwak (2007) describe the limited nu mber of studies that investigate the quality of home-based parental instru ctional strategies. When delvi ng into this body of research concerning home-based parental instructional strategies, Fan and Chen (2001), Shumow and Miller (2001) and Ma (1999) report weak positive or ev en negative correlations with student achievement, supporting the research findings associated with th is study. Presented with this evidence, quality of parental instruction could be the basis for the negative relationship between parent instruction and student ach ievement. Thus, it is plausible to conclude, based upon the data from the subset of parents in this study and a review of literature that parent instructional strategies were of limited and even negative value. This a ssertion is further supported by Hoover-Dempsey and Sandler (2005) who believe that parents of older children experience concerns about the limits of their personal knowledge and skill when assisting children with school work. These reservations may translate to substandard instructional practices. An additional explanation for the negative co rrelation between parental instruction and student achievement may be that increased parental instruction occurs only after students are performing poorly in school. That is, parents begin to spend more time focusing on their students achievement only after the student brings home sub-st andard grades (Fan & Chen, 2001), too late to repair damage that has al ready been brought upon a students grade in a specific course. As noted by Cotton and Reed -Wikelund (2001) parent al involvement in academic activities decrease as students approach adulthood, with some parents reporting that participation in a childs homework is a tedious and undesirable activity (Griswold, Cotton & Hansen, 1986). Further, given the age of the population evaluated, a parents inser tion into the academic process may cause tension between a pare nt, who might be reluctantly participating in 103

PAGE 104

the academic endeavor, and a child who is bot h exploring autonomy and may be struggling academically (Solomon, Warin & Lewis, 2002). Parental Reinforcements Positive Rela tionship With Student Achievement A second outcome of this study indicates that parental reinforcement held a significant positive relationship with student achievement within Parent Group C. Indicating that as parents employed more positive reinforcement strategi es associated with school work, academic achievement increased. Reinforcement, as co nceptualized by Hoover-Dempsey and Sandler (2005), is defined by the behavior istic notion that actions occur and are maintained because of their consequences (Skinner, 1989). In the cont ext of student learning, reinforcement theory predicts that children will repeat behaviors when they are consiste ntly associated with receiving positive reinforcement (Hoover-Dempsey & Sandler, 2005). Support for parental reinforcements positive relationship with student achievement, as seen in this study, can be found in contemporary literature associated with parental involvement (eg: Power, 2004; Martinez-Pons, 2001; Steinberg, 2001). Accord ing to Hoover-Dempsey and Sandler (1995) when parents engage in praise and positive rein forcement of behaviors such as listening and contributing in class, earning good grades, or th e timely completion of homework on a consistent basis, the results are likely to increase the child's acquisition of skills and behaviors beneficial to school success. Others, including Yap and Enoki (1995) and Walberg (1984) similarly suggest that increased student achievement can best be achieved through parent al supplementation of a students school work at home, thro ugh monitoring and encouragement. Implications Related to Research Question 1 Several implication for research, policy a nd practice can be drawn from the outcomes associated with research question one. Thes e implications provide a roadmap for future investigation into the role and effect of parents on virtual sc hool achievement. While family 104

PAGE 105

involvement is uniformly recognized as an e ssential component of co mprehensive academic programs, existing policies, including No Child Left Behind, are not formul ated to reflect this (Harvard Family Research Projec t, 2008). Many of the problems associated with the translation of issues in practice concerning families and academics to policy mandates concerning family and academics center on the complexities associat ed with evaluation of familial involvement. Virtual schools have the opportunity to lead the entire educational field in the incorporation and adoption of specific, methodologica lly sound standards for the evaluation of family involvement. A comprehensive understanding of the impact of the family, communicated effectively to policy makers could have a tremendous impact on future virtual school funding. In order to facilitate this understanding several steps can be taken by virtual schools. Foremost, a longitudinal assessment of parental involvements impact on students who participate in virtual schooling should be initiated. This assessment should not be limited to just survey data, the incorporation of qualitative data is necessary to build a comprehensive visualization of the role of the parent. Further, this qualitative data wi ll assist in the triangulation of survey outcomes. Fan and Chen (2001) encourage the use of measures of achievement over time as they provide a more reliable and valid construct for measure. Presently, ther e are several obstacles that impede a virtual schools ability to analyze student data longitudinally. One obstacle of particular importance to virtual school administrators and researchers is access to student data. Because many virtual schools are not diploma granting institutions, they do not have access to state and district level comprehensive student information. This may include data pertaining to standardized test scores, student disabilities, individualized education plans and disciplinary issues, information that can have critical impact on the manner in which a te acher delivers course content. When virtual schools are given access to state and district level database s, virtual school teachers will have the 105

PAGE 106

opportunity to fully understand their students and virtual school administrators will have the ability to monitor student growth both within th e context of the virtual school and traditional school environment. Second, virtual schools should develop outre ach strategies and formal guidelines for encouraging parental participation in the instruct ional process that are beneficial for students. Walker, Hoover-Dempsey, Whetsel and Green (2004) recognize two specific domains for parents to engage in teaching st rategies, direct instances, in wh ich a child may be engaged in homework and indirect instances, when a child is not engaged in academic activities, but a teachable moment arises (eg: measuring ingr edients while cooking, calculating the amount of paint needed to cover a wall). Walker, et al acknowledge that parents may need support to develop strategies that will have positive impact on an individual student. For example, strategies that focus on direct instances may include creating and practicing with flash cards in an effort to memorize multiplication tables. Strategies that fo cus on indirect instances include encouraging a child to journal or write crea tively or making parents aware of cultural or other learning opportunities available in their geographic area. Additionally, in formation on developmentally appropriate, content specific teachi ng strategies can help to build successful home-based parent instructional skills (Cancio, West & Young, 2004) Indeed, Shaver and Wallis (1998) linked workshops for parents on helpi ng their children at home to higher reading and math scores. In light of the geographic constraints associat ed with virtual schooli ng, and virtual schools pre-existing multi and hyper-media infrastructure, institutions should utilize the Internet for delivery of parental instructi onal resources. The util ization of Internet-based instructional resources for parents has been successfully demonstrated with ma ny different student populations, including those with disabilities (Ferdig, Amberg, Elder, Donaldson, Valcante & 106

PAGE 107

Bendixen, in press; Ferdig, Amberg, Elder, Valcante, Donaldson & Bendixen, 2008). Similarly, Internet-based instructio nal resources can also be used to in struct parents on appropriate methods of parental encouragement relating to school work and achievement. Future researcher should investigate the eff ect that providing parent al access to student learning management systems and course material s has upon student achievement. At present, there is a general if you build it, they will com e assumption in virtual schooling. As previously mentioned, parent LMS access constitutes a passiv e form of involvement; theoretically, it should not translate to significant ach ievement gains. Researchers should quantify the frequency in which parents access course materials, what they access and whether access to course information changes the manner in which parent s interact with thei r children regarding academics. Finally, future research should investigate parental knowledge and pe rceptions related to the processes of learning in online environments. This inform ation could be used to better educate and diffuse parental misconceptions related to online pedagogy and learning processes experienced by students. Armed with an compre hensive understanding of the cognitive processes associated with proactive learning experience in vi rtual school courses, pa rents could potentially become better participants in the educational process. Research Question 2 Do students and parents differ in their p erceptions of familial involvement in K-12 virtual schooling? : In previous chapters, it was posited th at the study of parental involvement and its relationship to virtual school student achievement could assist the development of new communication strategies between both homes and virtual schools and parents and virtual schools. These communications tactics would be used to facilitate inter actions between virtual schools, teachers and parents. These interactions would, in turn, lead to improvements in student 107

PAGE 108

achievement. To do so, it is necessary to quantif y and establish a baseline for the perception of intra-family communication regarding academics pr ior to the development of these strategies. Research question two sought to do this by dete rmining whether students and parents differ in the perception of familial involvement. The analysis associated with this research question revealed that parents and K-12 virtual school students have statistical differences in their perceptions of the amount of involvement parents have in the students academic work. Th e data provides evidence that students perceive a lower level of involvement in academic activitie s than parents. These results are analogous to data reported by DePlanty, Coulter-Kern and Duchane (2007) and Hoover-Dempsey & Sandler (2005). DePlanty et als study indi cates that parents ha ve a tendency to perceive themselves as more involved in academics than both student s and teachers. Hoover-Dempsey and Sandler (2005) describe weak correlative results between student percepti ons and parent perceptions of parent reinforcement, instruction, modeli ng and encouragement, at .16, .16, .14, and .16 respectively. These results are quite similar to th e correlative results elic ited by this study and discussed in chapter 4. The results associated wi th question two will be discussed in terms of psychometric validity and social desirability, and within the context of the outcomes associated with research question one. It is possible that the dispar ity in responses between parents and students exist due to validity issues inherent to the su rvey. It is useful to consider that the survey adapted for this study was designed for elementary aged students, and it should be acknowledged that some instruments designed to assess aspects of the parent-child milieu have been found to have questionable validity outside of specific age ra nges (Trost, Biesecker, Stattin & Kerr, 2007; Locke & Prinz, 2002). After the su rvey was adapted it was subject ed to a level of validation, 108

PAGE 109

detailed in chapter 3, but this process did not include an empirical component. Therefore, a structured empirical validation process that includes a confir matory factor analysis would provide insight as to the reliab ility and validity of the measurements associated with both students and parents. As rese archers and practitioners adopt and transform conventional assessments for use in virtual schools, validity needs to be a constant concern. Within the confines of this study, it is possible that the survey, which made fundamental assumptions about the nature of respondent views about the definition of parents and students, provide context appropriate for elementary aged students but in congruent with more ma ture adolescents. HooverDempsey and Sandlers survey wa s designed with a specific bias: that parents and children are talking about each other in exclusion of others. This bias would seem to directly contradict the very notions of family involvement described in chapter 1 by Pat Edwards (2004) as an omnibus expression broad enough to incorporate the myriad of family structures in practice in the U.S. Additionally, it is plausible that some of the di screpancy associated with parent and student perceptions of involvement could be attributed to social desira ble parent responses. Locke and Prinz (2002) discuss social desirab ility and its role in surveys that seek to detail aspects of the parent-child relationship, a concept harmonious w ith the notion that pare ntal involvement is a form of social capital (Grolnick, Benjet, Kuriwski & Apostoleris, 1997; McNeal, 1999). It is Locke and Prinzs assertion that additional re search correlating survey responses with observational data would advance understanding related to the self-report of parent-child information. Finally, when the results of this research question are placed within the context of the results of research question 1 (the relations hip between parental i nvolvement and student achievement) and Balli and Deplanty, et als research it can be a sserted that parents may indeed 109

PAGE 110

be engaging in activities meant to support stud ent achievement. Unfortunately, these activities may not be effective or enjoyable for students, as noted by the statistically significant parent instructional strategies variable, which was found to have a negative effect (B = -.61) on achievement. Given the wide disparity between pe rceptions of involvement held by parents and students, it is plausible that th e efficacy of parent involvement may be hindered by a students inability to comprehend and see the value of parent involvement activities. Additionally, contemporary theoretical and empiri cal literature suggests that a ch ilds acuity in perceiving and comprehending a parents academic involvement dire ctly influence the eff ect of that parents involvement on both student achievement and student psychological well-being (HooverDempsey & Sandler, 2005; Flouri & Buchana n, 2003) Thus, an accurate perception of involvement is a contributing component to effective parent interventions. Implications related to research question 2 : Results indicate that administrators, researchers and policy-makers w ho inquire about the role of th e parent may receive different results based on who is asked. Thus it will be im portant for practitioners and future researchers to consider their target audien ce and triangulate results with qual itative data elicited from the audience. Further, research exploring the per ceptional mismatch identified by this study is necessary; this research should s eek to discern the antecedents of the discrepancy and whether these issues are endemic in virtual schooling. Additional research should explore whether th e inequities in parent and student perception of involvement may be based upon social desirable response patte rns by parents. This social desirability issue, explained within the context of the social cap ital associated with education in contemporary American society by McNeal (1999) and Benjet, Ku riwski and Apostoleris (1997), may account for the discrepancy in perceived re sponses between students and parents. Future 110

PAGE 111

research should explore the perceptional disparity within the context of social desirability and social capital. The survey utilized in this study should be s ubjected to a rigorous validation process that includes structural equati on modeling in order to estimate the fit of the data. While the parent sample is comprised of a suitable number of participants, because of its limited sample size, accurately estimating fit with th e student population will be difficult. An appropriate student sample should include at least 280 individuals (MacCallum, Browne, & Sugawara, 1996). Further, parent and student survey responses need to be triangulated with qualitative data in order to ascertain whether parents are indeed responding with a level of accuracy. Virtual school administrators need to questi on whether parents are receiving timely reports and updates related to their stude nts progress and achievement in a course. As previously mentioned, often parents become involved in th eir childs academics onl y after the child is failing. Interactions between parent and child, pote ntially strained due to developmental concerns can be further exacerbated under an academical ly stressful environment. Researchers and practitioners need to develop consistent, reliable and relevant means for communicating student performance to parents. The incorporation of pred ictive analytical methods for student evaluation should be an area of increased in terest and research within the virtual sc hool community. Black, Dawson and Priem (2008) provide evidence that simple data l ogging applications can predict affective states related to student achievement in a sample with insufficient grade variability for direct achievement measure analysis. Similar research holds specific promise for direct translation to student achievement in virtual sc hools due to the high leve l of grade variability found in a majority of K-12 virtual institutions. Predictive analytics may assist in providing 111

PAGE 112

parents, instructors, administ rators and students with an ea rly warning system related to academic achievement. Finally, practitioners should consider build ing parent-student in terventions based upon activities that students readily comprehend as va luable contributors to their academic success. These interventions could include targeted out research to father s. Fathers were overwhelmingly absent in our respondent populat ion, providing an indication that they may represent an underutilized resource for their ch ildren. Given that virtual course s do not have to adhere to the traditional notions of time and space that are asso ciated with face-to-face schooling, fathers and other working parents have increased opportunity for participation. Research Question 3 Do factors including, socioeconomic status, race and gender, effect involvement of families in virtual schooling? : The results of question 3 reveal several distinct phenomena, (a) male student respondents in the upper (> $60,000 annually) and lower income (< $30,000 annually) categories per ceived a higher degree of encouragement, modeling, instruction and reinforcement from parents than male stude nt respondents in the middle income category (>$30,000 < $60,000 annually) and (b) female student respondents in the middle income category perceived higher levels of encouragem ent, modeling, instruction and reinforcement from parents than female student respondents in the lower and upper income categories. Finally, (c) a significant effect is found related to househ old income, where this income effect closely mirrors the interaction effects described previ ously, but it is a main effect, only concerning a single variable (household income). The main ef fect indicates that students in lower (< $30,000 annually) and upper (>$60,000 a nnually) income categories perceive greater levels of instruction from their parents than individuals in mi ddle income categories (>$30,000 < $60,000 annually). 112

PAGE 113

The research results reveal differences in st udent perception of parent involvement based on both the students gender and household income. These differe nces are patterned by male students of lower and higher household income s perceiving a greater level of parental involvement than male students of middle househol d incomes. Conversely, female students of lower and higher income households perceived lowe r parental involvement than female students of middle household incomes. In addition, when all other variables were held constant, household income was found to have an effect on the amount of parent instruction perceived by students. It has been well documented th at demography, particularly socioeconomic status, plays a role in the amount of involvement perceived by students and parents (B alli, 1996; Bracey, 1996; Brody, 1995). Hendersons (2006) resear ch indicates that some parent s, particularly those from lower socioeconomic strata may feel uncomfortable questioning the nature of an assignment or discussing general concerns regarding their childs academic performance. Research by Ritblatt, Beatty, Cronan and Ochoa (2002) discusses the importance of school personnels sensitivity to parents situations, including their cultural and physical environment. Further, Ritblatt et al note that socioeconomic status plays a significant role in parents school involvement activities. Their findings indicate that parents in lower income groups spent less time involved in school-based extracurricular activit ies, perhaps due to work obligations. Interestingly, these lower-income families spent more time working directly with their child on academic concerns than those families in middle or high-income groups. While this pattern is not specifically seen within the data from this research study, it is possible that the limited sample size and lack of specific acuity present in the in come independent variable may have limited the researchers ability to discern a similar model. 113

PAGE 114

Given the outcomes associated with resear ch question three and the conflicting results associated with gender, it would seem that a vi rtual schools lack of phys ical facilities do not necessarily provide a specific parental involvement advantage to individuals based upon economic status. This outcome runs contrary to McGrath and Kuriloff (1999). But, when gender is held constant, we find that lower and upper in come parents engaged in similar patterns of instructional activities, an outcome supporte d by Kozma and Shank (1998). Unfortunately, as evidenced from research question one, these inst ructional activiti es were not found to have a positive influence on student achievement. Implications related to research question 3 : The results of question three seem to provide the opportunity to addres s whether virtual schools are truly bridging the digital divide. Hoover-Dempsey and Sandler (1995 ) downplay indicators of family status in efforts to understand parents' involvement decisions. The c onscious choice to disregard demographics may have been an appropriate choice with elementary age students. Future research should discern whether demography plays a more expansive role, as evidenced by the resu lts of question three, in the exploration of familial involvement in tr aditional secondary students and virtual students. This future research should include qualitati ve follow-up with respondents to gain a deeper understanding of the motivations behind participation in virtual courses. In addition, future research should call into question the notio n that a virtual schools lack of physical facilities inherently promotes pare ntal involvement (Kozma & Shank, 1998). This is particularly salient in situations similar to th e Virtual School, where stude nts are not afforded the opportunity to be 100% online. A robust analysis of the role of demography, including race, gender and socioeconomic status and its effect on virtual school student enrollment, achieve ment and parental involvement 114

PAGE 115

should be initiated. Future re search should employ a more granular stratification of socioeconomic status than utilized in this study. For example, additional income strata may promote more variability within the data. In addition, researchers should seek to focus on both quantitative and qualitative methodologies, to tria ngulate and confirm the va lidity of statements from both students and parent s (Locke & Prinz, 2002). Respondents to the survey represented a racial ly diverse, but predominantly white group. The majority of respondents were educated, pr ofessional females who came from households that earn $60,000 or more annually. We were gi ven empirical evidence th at the majority of households participating in the study earned more than the median household income in the state, reported by the US Census Bureau as $42,679 (2008) This data does not discourage the notion that in the Virtual School, as with other K-12 Virtual Schools in the U. S., equal access is an issue. A consistent criticism of K-12 online schools is that they are atte nded by individuals with higher household incomes (Watson, 2007). The demographics of parent and student res pondents, detailed in Tables 5.1 and 5.2, mirror the demographics described by Clark and Blomey er (2007) and featured in Table 5.3, in their external evaluation of the Virtual School. The AVS constituency is diverse, but it does not yet emulate the diversity found in the traditional sch ool population in the state, which is represented by a 47% non-white population (D epartment of Education, 2008). In an effort to determine geographic dist ribution of study respondents, respondent zip codes were collected and mapped utilizing ESRI ArcGIS. ArcGIS is a geographic information system that allows for the multidimensional repr esentation of data with geographic components. The map, Map 5.1, reveals that many of the partic ipants in this study ha iled from urban areas, this geographic phenomenon may potentially bias the data. Geography needs to be considered in 115

PAGE 116

both the present and future discussions of virt ual school student demogr aphy. Without the ability to visualize the respondents to th is survey, it could be incorrec tly concluded that the respondents were representative of students in the state. GIS provides the opportuni ty to view respondents from a geospatial perspective, allowing for the co nsideration that the ma jority of respondents were clustered in higher density urban locales. This geographic visualization also assists in calling into question another common ly held belief: that virtual sc hools are impacting students in rural areas (Watson & Ryan, 2007; Mills & Roblyer, 2004; USDOE, 2007). In order to address issues associated with geographic, economic and minority representation, AVS and other inst itutions should reconsider its admissions process. AVS and several other institutions currently admit students on a first-come-fir st-serve basis. This process may inadvertently keep minority, rural and stude nts from lower socioeconomic strata from enrolling in AVS courses. In order to counter the possibility of unequal access, AVS and other institutions choosing to enroll on a first-come-first-serve basis may want to institute a lottery program for over-subscribed courses. Lottery pr ograms have been successfully utilized in Milwaukee, Wisconsin in order to manage enrollment in a popular school choice voucher program (Greene, Peterson & Du, 1999). Derivative Outcomes Ancillary findings, not related to the three research questions became apparent through the investigation and analysis of data. These findings provide specific opportunity to impact virtual school practice, research and policy and fall into two categ ories, outcomes related to the survey and outcomes related to the study population. Outcomes Related to the Survey Black, Ferdig & Dipietro (2008) call for the exploration and use of appropriate surveys from traditional K-12 research in virtual schools. Black et al. di scuss the caution that should be 116

PAGE 117

exercised with this process, as there are si gnificant differences between K-12 face-to-face and virtual schools. The incorporation and augmen tation of Hoover-Dempsey and Sandlers (2005) survey gave opportunity to exerci se Black et als call, as the survey, designed for and used in conjunction with investigations of elementary and middle school family involvement research, had not been used in conjunction with a population of secondary school students. Further, it had not been used in an online format or with st udents in non-traditional schooling environments like virtual schools. The data described in Chapter 4 reveals that the survey proved to be a reliable instrument, with Cronbachs Coefficient ranging from .89 .93 across th e four variables constituting parents perception of involvement and .92 .95 acr oss the four variables constituting students perception of involvement. These reliability figu res provide empirical evidence that there is a minimal amount of random error associated w ith the results genera ted by this study. The reliability figures also closely emulate the figures reported by Hoove r-Dempsey and Sandler (2005). It is important to note that Hoover-Demps ey and Sandlers survey was revised with the exclusion of questions from the parent survey pertaining to face-to-face school involvement. Two factors influenced the reduction of the Hoover-Dempsey and Sandler Model: geography and existing research. Given that vi rtual schools do not have a physica l location, there are specific limits to a parents ability to self-involve spec ific to the virtual school. Based upon the results of the survey it can be asserted that the alteration ma de to the original format of the survey did not have an effect on internal consistency. Outcomes Related to the Study Population Utilizing a list compiled by the Virtual Sc hool, 10,169 parents were contacted via email with a request to take part in this study, 940 parents respond ed representing a 9% survey response rate. 665 of the 940 responding parents a pproved their childs part icipation in the study. 117

PAGE 118

Of these eligible students, email addresses fo r 485 were obtained thr ough cooperation with the Virtual School and Department of Education. Out of the 485 stude nts contacted via email with a request to take part in the s econd phase of the study, 164 students participated in the student survey. This represents a 36% response rate associated with the students. Further analysis of the email solicitation ma de to parents reveals that of the 10,169 email invitations, 2270 solicitations, representing 22% of the study population, were sent to email addresses no longer util ized. 212 parents, 2% of the study popu lation, requested that their names be removed from the list of survey participants by utilizing the opt-out feature embedded in the email solicitation; these individuals and thei r corresponding students were removed from the studys email roster. When individuals associ ated with unreachable email addresses are eliminated from the overall invitation figures, the parental response rate rises to 11.8%. Though there is no literature that has established a benchmark for webbased survey response, response rates associated with parents for this survey we re low as compared to rates discussed by Cook, Heath and Thompson (2000), but fall in line with those reported in Manfreda, Bosnjak, Berzelak, Haas and Vehovars 2008 meta-analy sis. Manfreda et als study reviewed 45 different content non-specific cases in which web-based surveys we re utilized, their findi ngs detail a range of response rates, from 11% to 82%. Often, unsolicited mass emailing, such as those utilized by this research study, are associated with SPAM a nd phishing, reducing legitimacy and leading recipients to be wary of participating (Tut en, 1997; Porter & Whitcomb, 2003). According to Dillman (2007) and research by Kent and Turner (2002) there is statistically significant evidence that web-based surveys that include a pre-notification email receive more responses than those that do not utilize prenotification. This pre-notification em ail provides a poten tial participant with a sense of legitimacy. Unfortunately, the Zoomerang application utilized for this study does 118

PAGE 119

not feature the ability to send a pre-notification to potential pa rticipants. This lack of prenotification may have led to decreased participation. Given the relatively low level of particip ation by parents compri sing the study population, it is necessary for AVS and other virtual schools to investigate email practices that are currently employed by teachers and administrators. Evidence indicates that many individuals experience email overload, characterized by overflowing em ail in-boxes and the stress of attempting to respond to this deluge of communications (Whittaker & Sidner, 1996; Jackson, Burgess & Edwards, 2006). Additionally, research indica tes that email users develop patterns of responsiveness based upon reciprocity and perc eived communicative value. Over time, the perception that an email senders communications have little value or are non-reciprocative will hinder participation (Tyler & Ta ng, 2003). Non-response to email can also be associated with poor email communications practices, these communications practices compound over time, leading individuals to devel op a negative opinion of all communications emanating from the specific source. Because email is a critical co mmunications medium for K-12 Virtual Schools, administrators should consider implementing ema il policies and trainings to maximize the utility of email communications sent to recipients, including, co-workers, students and parents. Jackson, Burgess and Edwards (2006) recommend developing tr ainings that target the following areas for more effective email communications: (a) Is an email message necessary? (b) Effective targeting of your email (c) Effective s ubject lines (d) Getting your message across and (e) Sending attachments. An additional area of concern involves the number of email addresses for students that could not be located. As noted in Chapter 4, 180 of the 665 email addresses needed to contact student participants could not be located. Thes e 180 individuals represen t 27% of the potential 119

PAGE 120

student respondent population. Reas ons for the data loss are uncle ar, though interactions with AVS have provided evidence that a comprehensiv e student management system either does not exist or is inadequate. Such a system would pr ovide tracking and maintenance of students from the time of their application to the end of thei r association with the AVS. AVS and other virtual schools may benefit from the experience of th e Florida Virtual School (FLVS), who, after experiencing tremendous growth in its populatio n developed a student management application called Virtual School Administrator (VSA). FLVS describes VSA as a performance management system which is designed specifically to meet the unique needs of organizations providing online learning opportunities to students, VS A is a dynamic, performance-based system developed to provide program admi nistrators with the tools they need to effectively manage the successful operations of an online learning program. In creating VS A, FLVS has built a comprehensive solution for tracking overall perf ormance through monitori ng activities in four areas: student performance a nd data management; student registration and enrollment management; role-based reporting; and communi cations (Florida Virtual School, 2006, 1). Implications Related to Derivative Outcomes The high reliability elicited in this study provi des evidence that the su rvey utilized is a reliable instrument that should be considered in future research investigating the role of parents in Virtual School achievement. It is important to note that the validity of this instrument has not been investigated, though the results reported in ch apter 4 and discussed in this chapter indicate that a level of generalizabil ity exists with outcomes descri bed by Hoover-Dempsey and Sandler (2005). This generalizability is one aspect of the evidence that will be necessary for validity to be concluded. Thus, future research ers should work to build a body of evidence that will provide affirmation of the reliability statistics repor ted in this study and substantiation of the 120

PAGE 121

assessments validity, both through use of genera lizability and using crit erion related validity. This is, correlating outcomes with other measures of parental involvement. It goes without saying that the logistics involved in a co mprehensive survey are heady. Even with the assistance of databases and onlin e survey applications, th e process of cataloging and matching student and parent responses is a complex undert aking that should be avoided unless absolutely necessary. Outcomes from this study, if accepted unconditionally, indicate that future inquiry into the role of the family could derive efficiency through the exclusion of student surveys. This study found that student perceptio ns had little impact on student achievement. Unfortunately, students were much more receptive to responding to the survey as compared to parents. Further, the outcomes provide evidence that a quantitative investigation may not be the most effect manner by which to solicit valid parent information. In light of these issues and the aforementioned complexity associated with pare ntal involvement resear ch, future researchers should consider looking towards focus groups bui lt upon stratified parent samples and individual interviews to discern the prevalence and proce ss by which parents are pa rticipating in their childs virtual education. Virtual schools must employ a modular data infr astructure that has the ability to track and update changes in student contact data. This da ta infrastructure shoul d include a dedicated student information system (SIS) such as VSA or BocaVoxs Maestro. SISs streamline and automate the process of updating student data. Fu rther, student contact information should be verified on a regular basis to prevent instances in which a student or pa rent is unable to be contacted. Emergencies can occur, even in onlin e environments, a prime example is provided by DiPietro (2008) who describes an incident in which a virtua l school student posted suicidal ideation in an online course forum. Due to the prompt actions taken by the virtual school 121

PAGE 122

instructors and staff and accurate contact informa tion related to the student, appropriate help was provided in a timely manner. At the present time the federal government and the majority of state governments have not initiated oversight or reporti ng guidelines for virt ual schools. Instead, government is relying upon traditional school data infrastructures to maintain records and student accountability information. Recent developments in several states, including Colorado, Wisconsin and Pennsylvania provide examples where virtual sch ools, in environments of lax oversight, could not justify or provide specific evidence of the benefit they provide to their constituency. Specifically, the Trujillo commission, which was conve ned to address issues associated with the state department of education in Colorado s administration of vi rtual schools, found individual student data were insufficient to facilitate a meaningful evaluation of online student assessment scores or attrition rates (Donnell-Kay Foundatio n, 2007, p. 4). The Trujillo commissions report and subsequent recommendations for virtual school s should be used as a template and guideline for state departments of education to effec tively manage and appropriately account for the growth and oversight of virtual schools until federal guidelines for oversight, data collection and data reporting can be developed and initiated. Finally, further research needs to be conducte d seeking to explore th e relationship between virtual school achievement and socioeconomic stat us, race, gender and the role of the parent. An additional variable that should be considered is geography. Evidence exists that the geographic makeup of the sample in this study was highly sk ewed towards urban and suburban students (see chart 5.4). While GVS and most K-12 virtua l schools do indeed serve diverse student populations, they could be doing more to encour age additional minority and rural enrollments. AVS in particular, has a unique history of engaging schools in a missionary style, that is, 122

PAGE 123

conducting informational education and outreach in school districts to inform interested parties about their services. These outre ach activities have in cluded targeting sc hool administrators, counselors, teachers, parents and students (Fer dig & Cavanaugh, in press). Additional resources should be dedicated to these missi onary-style activities to increase rural an d minority enrollment. Other virtual schools should consid er the missionary approach that has enabled AVS to establish diversity. Broad Outcomes Associated With This Study In the process of conducting and analyzing the data produced in this study, several broadbased outcomes that could be applied in research, policy and practice became apparent. These five outcomes touch upon notions of contem porary theory, homeschool and intra-school communications and technological development. Results associated with research questions 1 and 2 provide evid ence that, within the context of virtual schools, it is may not be appr opriate to rely upon the notions of technological and psychological transference po sited by Reeves and Nass (1996). Reeves and Nass postulate that computers are treated as social actors by users. Thus rules which people apply to social interactions with people should apply equally well to their inte ractions with computers. As evidenced by this study and by research by Black, Greaser and Dawson (2008) the subtle differences found between outcomes and interactions amongst individuals, technology and corresponding others when comparing traditional and online venues reveal a complexity easily ignored by the blanket applicati on of Reeves and Nass theory. Overlooking this complexity, a process that is not difficult give n the relative dearth of empiri cally based research in virtual schooling, has the potential to rele gate online teaching and learning to an antiquarian state, one that ignores much of the progress made duri ng the last 20 years by assuming that all online learners are the same. 123

PAGE 124

It may be appropriate to further situate futu re conversations about virtual schooling in a broader theoretical context. This context s hould acknowledge the multiple complex interacting layers associated with todays virtual schools. An ecological approach provides one framework for exploring this type of environment (Br onfenbrenner, 1979, 1955). This theory of human development and interactions cons iders the way relationships within the family and between the family and social environment influence i ndividual development and family functioning (Bronfenbrenner, 1979). From an ecological perspective, the most l ogical model of a family is as a system. Accordingly, a majority of researchers now view the family from what could be termed a "systems perspective" (Kreppner & Lerner 1989). Bronfenbrenners (1979) ecological perspective views human developmen t and interactions from a systemic perspective. His theory combined sociology and developmental psycholo gy, asserting that the relationships between individuals and their environmen ts are viewed as "mutually sh aping." Brofenbrenner asserted that the individual's experience was likened to a set of nested structures, each inside the next, like a set of Russian dolls" (B ronfenbrenner 1979, 22). Accord ing to Bronfenbrenner when studying human interactions, one has to see within, beyond, and "acro ss" how the several systems interact (family, workplace, and econom y). By viewing virtual schools from this systemic or ecological perspective it is possible to postulate that the category of the other as described by Black, Ferdig and Di pietro (2008) may play a signific antly large role in the success or failure of a virtual school st udent. Given the dominant role ascr ibed to the other category in an ecological model, there is a critical need for additional evaluative instruments to enable outcomes-based research in virtual schooling. Th ese instruments should be reflective of the diversities inherent in K-12 vi rtual schooling and should seek to describe the manner in which factors external to the student effect student achievement (B lack, Ferdig & DiPietro, 2008). 124

PAGE 125

In traditional schooling environments, teacher outreach that includes active involvement practices such as: parent teacher meetings, regular progress updates and a consistent exchange of learning materials between the home and school re sult in improved student performance (Westat & Policy Studies Associates, 2001). Effective outre ach practices that include prosocial academic modeling and instructio nal activities should be incorporat ed into a comprehensive support package for virtual school parents. This pack age should include a clea r and direct line of communication to the students virtual school teache r, a point of contact for technical issues and concerns and the opportunity to in teract with other parents of virtual school students. This interaction could be faci litated thorough an online forum for virtual school parents. Research by Romiszowski and Ravitz (1997), Ti ene (2000) and Ferdig and Roeh ler (2003) has promoted the importance of the flexibility and convenience that discussion boards offer, as they are accessible to any individual with a computer and Internet access. In addition, discussion boards allow those who are less outgoing to have the space and time necessary to speak their mind (Chong, 1998; Dutt-Doner & Powers, 2000). The de velopment of a parent forum could lead to more complex forms of parental engagement, for example, th e formation of parental advisory committee to serve as a liaison between parents, teachers, students and virtual school administration. Hoover Dempsey and Sandler (1995); Hoover-Dem psey, Walker, Sandler, Whetsel, Green, Wilkins and Closson (2005) and Baumrind (1971; 1991) suggest that both explicit and implicit student solicitations for parental involvement in academic work have the potential to prompt parental action. Student requests for assistance may be either explicit, such as a direct request for parental assistance, or implicit, such as a pare nt noticing signs of a childs frustration or receiving notification from a teacher. A parent w ho understands that his/her child is struggling in a specific course or content is often more likely to closely monitor homework (Hoover-Dempsey, 125

PAGE 126

et al, 2005). The problem is co mmunicating this information to parents who lead busy lives and have obligations related to work, home and other children. Many virtual schools seem to feel that by allowing parents the ability to access course materials and monitor progress they are fulfilling their obligation towards promo ting communication between school and parent. Unfortunately, as previously stated, there is no empirical data that has been publis hed related to parent access and whether this access translates to gains in achievement. Research by Abouchaar (2003) and Cotton and Reed-Wikelund (1989) allows us to generalize that course acce ss, as a passive form of involvement, would not have as dramatic eff ect as more active forms of involvement. This does not mean that parent access should not be a part of a communications strategy, it should, but the cornerstone of a parent -school communications strategy, whose purpose is to increase achievement, should involve active forms of invol vement (eg: parent-teacher conferences, phone calls to update parents rega rding student progress). In order to establish regular communications and feedback virtual sc hools should make use of the inherent automated affordances of learni ng management systems. LMSs can be used to track student activity automatically. While no LMS on the market currently automates communication of course related data, it would not be particularly difficult to design a system that would send out periodic updates to parents and other concerned individuals on a regular basis via email or SMS (Black, Dawson & Prie m, 2008). In this manner, parents could be informed of their students progr ess in a course, or of course activity and inactivity over a specified period of time. This process of regularly updating parents could eliminate the last minute, hectic scramble to fix a sub-par course grade. The utilization of online tec hnologies, including learning management systems and the dissemination of training materials to parents should not be c onstrained to virtual schools. 126

PAGE 127

Traditional schools can benefit from the structure and utility that a learning management system can offer to students. With access to course materials and classmates via a learning management system, students would be provided with a contro lled, safe environment for group activities, outof-class discussion and assignments. This same learning management system could be used to disseminate training and information to pare nts, circumventing the cost and time delay associated with mailings or sending im portant materials home through students. Additional Framework for Research: Unders tanding the Variability Associated with Virtual Schools in the United States It is important for readers to consider the pres ent state of K-12 virtual schooling in the U.S. and its inherent variability. 42 states currently fund state led virtual schools programs, no two states has a similar management or bureau cratic structure (Watson & Ryan, 2007). The inconsistencies associated with these instituti ons provide unique challe nges for individuals who seek to understand this contemporary medium for learning. Prior to gene ralizing the results of this study there are several critic al points that must be addr essed. First, readers should be reminded that the virtual school studied in this dissertation does not allow full-time student enrollment. Students are limited to enrollment in two courses during a given semester. Second, the virtual school studied is not a diploma granting institution, thus it must report course grades for student to the traditional school in which th e student is enrolled. The traditional school has the option of adjusting virtual sc hool grades in any manner it wish es before publicizing them to the student. It is unknown whethe r the traditional schools made ad justments to students final grades. Additionally, a broad definition of the pare nt was utilized for this study; this definition derived from Edwards (2004), incorporates the multitude of different family structures in existence in the U.S. today. Finally, this resear ch was conducted through contact with the virtual schools parent or legal guardian of record. It is possible that the individual who responded to the 127

PAGE 128

survey was not the person or family member wh o provided the majority of academic assistance to the student. Future research should invest igate the new family structure that Edwards describes and whether these non-traditional family units have unique means of interacting with students and schools. Conclusion This dissertation explored the impact of familial involvement upon virtual school achievement utilizing a sample of parents and st udents from a state led virtual school in the Southeastern United States. Outcomes of this stu dy have specific implications for researchers, policy-makers and practitioners. Results indicate that familial impact is a co mplex construct which may not be effectively measured utilizing quantitative methodology independe ntly. Therefore, future research seeking to explore familial impact should incorporate mixed methodology (quantitative and qualitative methods) and a longitudinal asse ssment strategy. In order for engagement in an effective longitudinal study, virtual schools must provide researchers wi th the ability, amongst other things, to track students longit udinally. Data infrastructures must be improved to provide seamless access to comprehensive student informa tion. Educational policymakers, either at the state or federal level, may ultimately be th e impetus for the integration of the multiple educational data systems that exist in some states in the U.S. The establishment of a comprehensive student data infrastructure will take time and a considerable amount of money. Until this can be accomplished virtual school practiti oners must also take a more progressive role establishing and maintaining virt ual school student and parent in formation. A starting point for this process is the adoption and utilization of a student information system. The SIS will serve as a foundation for maintaining accurate informa tion about current and former students. 128

PAGE 129

Results associated with a subset of the parent sample indicate that pa rental instruction has a negative effect on student achievement and pa rental encouragement has a positive effect on student achievement. Practitioners seeking to increase th e role of the parent in virtual schools should consider building media applications to pr ovide instruction to parents on age appropriate educational interventions aimed at improving th e effectiveness of pa rent instruction and encouragement activities. Outcomes indicate that socioeconomic status pl ays a key role in the level of involvement parental involvement perceived by virtual school st udents. Further, gender and a parents level of education also effect the amount of parental support perceived by students. It is important to consider that these two outcomes must also be paired with finding that indicate parents and corresponding students perceive very different amounts of parental involvement. Future research should investigate the validity of the survey utilized in this study. In addition, practitioners seeking to encourage and improve parental involv ement in their childs virtual school education should look to direct involvement activities, such as parent-teacher conferences, the creation of a parent advisory committee or establishing a foru m for virtual school parents to discuss topics pertaining to their children. Furthe r, virtual school instructors with a large student to teacher ratio are inhibited from engaging parents in a mean ingful and thoughtful manner. The process of making 180 phone calls every month to virtual school parents may in-effect turn a teacher into the equivalent of a telemarketer. Finally, practitioners and resear chers need to utilize learning management systems to their full capacity. The investigation a nd construction of LMS applications that provide analytical capabilities in the learning envir onment will allow the instructor to spend more of his or her time 129

PAGE 130

engaging students in rich and m eaningful discourse rather than policing message boards or trying to ascertain the level of student engagement in a non-empirical fashion. K-12 virtual schools are still a relatively novel concept for the vast majority of American households. A growing cadre of researchers is working to understand th e nuances associated with educating children and adoles cents in an online environment. This dissertation should serve as the starting point for a more detailed and co mprehensive conversation concerning the role of parents and their impact on student achievement in virtual schooling. 130

PAGE 131

Table 5-1. Parent respondent race/ethnicity Asian/Asian-American 21 2% Black/African-American 172 18% Hispanic/Hispanic-American 20 2% White/Caucasian 561 60% Other 21 2% Missing 145 15% Table 5-2. Student resp ondent race/ethnicity Asian/Asian-American 2 1% Black/African-American 27 16% Hispanic/Hispanic-American 5 3% White/Caucasian 103 63% Other 5 3% Missing 22 13% Table 5-3. AVS enrollment by ethnicity, Fall 05 Summer 07 24% 62% 4% 3% 4% 3% 0% American Indian Asian Black Hispanic Multi-Racial White Not Given 131

PAGE 132

Figure 5-1. This map describes the geogra phic distribution of survey respondents 132

PAGE 133

APPENDIX A IRB AND INFORMED CONSENT 133

PAGE 134

APPENDIX B VISUALIZING THE PARENTAL INTERACTI ON AND OUTLIER EFFECT ACROSS ALL FOUR STUDY VARIABLES Graphing the parental interaction across the f our factors reveals an Asian/AsianAmerican male parental respondent (n=1) whose responses to the parent survey score far below the rest of the sample in each of the four pa rent factors (modeling, instruction, encouragement, and reinforcement). The following charts (2-5), pr esent a visual description of the interactions. The outlier response is circle d in red in charts 2-5. Interaction 1 The first interaction, interaction 1 details a relationship between parental modeling and parent gender which is influenced by parent race/ethnicity (C hart 2). ANOVA results indicate a statistically significant interaction between average modeling and parent gender and race/ethnicity, F(4, 688) = 3.858, p = .004. Asian/Asian-American Female Black/African-American Male Asian/Asian-American Male Hispanic/Hispanic-American Male White/Caucasian Male Other Male Black/African-American Female Hispanic/Hispanic-American Female White/Caucasian Female Other Female35.00 40.00 45.00 50.00 55.00 60.00 Race/EthnicityAverage Modeling Figure B-1. The interaction between parental modeling and parent gender and parent race/ethnicity 134

PAGE 135

Interaction 2 The second interaction, interaction 2, detail s a relationship between parental instruction and parent gender which is influenced by pare nt race/ethnicity (Cha rt 3). ANOVA results indicate a significant interacti on between average parental inst ruction and parent gender and parent race/ethnicit y, F(4, 688) = 4.009, P<.01. Asian/Asian-American Female Other Female White/Caucasian Female Hispanic/Hispanic-American Female Black/African-American Female Other Male White/Caucasian Male Hispanic/Hispanic-American Male Black/African-American Male Asian/Asian-American Male40.00 45.00 50.00 55.00 60.00 65.00 70.00 75.00 80.00 85.00 Race/EthnicityAverage Instruction Figure B-2. The interaction between parental instruction and parent gender and parent race/ethnicity Interaction 3 The third interaction, interaction 3, de tails the relationship between parental encouragement and parent gender which is infl uenced by parent race/ethnicity (Chart 4). ANOVA results indicate a statistically significant interaction between average parental encouragement and gender and race/ethnicity, F(4, 688) = 2.507, P = .041. 135

PAGE 136

Asian/Asian-American Male Black/African-American Male Hispanic/Hispanic-American Male White/Caucasian Male Other Male Asian/Asian-American Female Black/African-American Female Hispanic/Hispanic-American Female White/Caucasian Female Other Female40.00 45.00 50.00 55.00 60.00 65.00 70.00 75.00 Race/EthnicityAverage Encouragement Figure B-3. The interaction between parental encouragement and parent gender and parent race/ethnicity Interaction 4 The forth interaction, interaction 4, de tails the relationship between parental reinforcement and parent gender is influenced by parent race/ethnicity (C hart 5). ANOVA results indicate a statistically significant interaction be tween average parental reinforcement and parent gender and parent race/ethni city, F(4, 688) = 3.361, P = .01. 136

PAGE 137

Asian/Asian-American Male Black/African-American Male Hispanic/Hispanic-American Male White/Caucasian Male Other Male Asian/Asian-American Female Black/African-American Female Hispanic/Hispanic-American Female White/Caucasian Female Other Female40.00 45.00 50.00 55.00 60.00 65.00 70.00 75.00 80.00 Race/EthnicityAverage Reinforcement Figure B-4. The interaction between parental reinforcement and parent gender and parent race/ethnicity 137

PAGE 138

APPENDIX C VISUALIZING THE STUDENT INTERACTION AND OUTLIER EFFECT ACROSS ALL FOUR STUDY VARIABLES Graphing student interaction ac ross the four studen t factors reveals a student (n=1) whose parent is a female with a docto ral degree whose survey responses to the student survey score far below the rest of the sample in across all four student factors (see charts 6-9). This individuals response is considered an outli er within the sample (circled in red in charts 6-9). Interaction 1 The relationship between student modeling and parent education is influenced by parent gender (Chart 6). Analysis of th e significant interaction between average parental reinforcement by gender and race/ethnicit y, F(5, 129) = 3.717, P = <.01. Male high school or GED Male bachelor's degree Male some graduate work Male master's degree Male doctoral degree Female less than high school Female high school or GED Female some college, 2-year college/vocational Female bachelor's degree Female some graduate work Female master's degree Female doctoral degree Male some college, 2-year college/vocational15 20 25 30 35 40 Parent EducationAverage Student Modeling Figure C-1. The interaction betw een student modeling and parent education and parent gender 138

PAGE 139

Interaction 2 The relationship between student instruction a nd parent education is influenced by parent gender (Chart 7). Analysis of th e significant interaction between average parental reinforcement by gender and race/ethnicit y, F(5, 129) = 3.447, P = <.01. Male high school or GED Male some college, 2-year college/vocational Male bachelor's degree Male some graduate work Male master's degree Male doctoral degree Female less than high school Female high school or GED Female some college, 2-year college/vocational Female bachelor's degree Female some graduate work Female master's degree Female doctoral degree20 25 30 35 40 45 50 55 60 65Parent EducationAverage Student Instruction Figure C-2. The interaction betw een student modeling and parent education and parent gender Interaction 3 The relationship between student encouragemen t and parent education is influenced by parent gender (Chart 8). Analysis of the si gnificant interaction between average parental reinforcement by gender and race/e thnicity, F(5, 129) = 3.058, P = .01. 139

PAGE 140

140 Male high school or GED Male some college, 2-year college/vocational Male bachelor's degree Male some graduate work Male master's degree Male doctoral degree Female less than high school Female high school or GED Female some college, 2-year college/vocational Female bachelor's degree Female some graduate work Female master's degree Femal gree 10 20 30 40 50Parent EducationAverage Student Encouragemente doctoral de Figure C-3. The interaction betw een student encouragement and parent education and parent gender Interaction 4 The relationship between student reinforcemen t and parent education is influenced by parent gender (Chart 9). Analysis of the si gnificant interaction between average parental reinforcement by gender and race/e thnicity, F(5, 129) = 3.380, P = <.01.

PAGE 141

Male high school or GED Male some college, 2-year college/vocational Male bachelor's degree Male some graduate work Male master's degree Male doctoral degree Female less than high school Female high school or GED Female some college, 2-year college/vocational Female bachelor's degree Female some graduate work Female master's degree Femal ree10 15 20 25 30 35 40 45 50 55 Parent EducationAverage Student Reinforcemente doctoral deg Figure C-4. The interaction betw een student reinforcement and parent education and parent gender 141

PAGE 142

APPENDIX D PARENT SURVEY The parent survey consists of a demogra phy section, a school vale nce section and four separate measurement variables: parental report of encouragement, pare ntal report of modeling, parental report of reinforcement and parental report of instruction. Both the demography and valence section have unique formats and in the ca se of the valence section a unique Likert-style scale. For each of the four m easurement variables, parents were instructed to respond to the following prompt: Parents and families do many different things when they help their children with schoolwork. We would like to know how true the following thi ngs are for you and your family when you help your child with schoolwork. Please think about the current school year as you read and respond to each item. Demography 1 Your Gender: ____ Female ____ Male 2 Please choose the job th at best describes yours (please choose only one ): ___ Unemployed, retired, student, disabled ___ Labor, custodial, maintenance ___ Warehouse, factory worker, construction ___ Driver (taxi, truck, bus, delivery) ___ Food services, restaurant ___ Retail sales, clerical, customer service ___ Service technician (app liances, computers, cars) ___ Bookkeeping, accounting, related administrative ___ Singer/musician/writer/artist ___ Real Estate/Insurance Sales ___ Teacher, nurse ___ Professional, executive ___ Other: _______________________ 3 On average, how many hours per week do you work? ___ 0-5 ___ 6-20 ___ 21-40 142

PAGE 143

___ 41 or more 4 Your level of education (please check highest level of education) ___ less than high school ___ high school or GED ___ some college, 2-year college or vocational ___ bachelors degree ___ some graduate work ___ masters degree ___ doctoral degree 5 Please choose the job that best describes your spouse or partners: ___ No spouse or partner ___ Unemployed, retired, student, disabled ___ Labor, custodial, maintenance ___ Warehouse, factory worker, construction ___ Driver (taxi, truck, bus, delivery) ___ Food services, restaurant ___ Skilled craftsman (plumber, electrician, etc) ___ Retail sales, clerical, customer service ___ Service technician (app liances, computers, cars) ___ Bookkeeping, accounting, related administrative ___ Singer/musician/writer/artist ___ Real Estate/Insurance Sales ___ Teacher, nurse ___ Professional, executive ___ Other: _______________________ 6 Your spouse or partners level of education (please check highest level of education) ___ less than high school ___ high school or GED ___ some college, 2-year college or vocational ___ bachelors degree ___ some graduate work ___ masters degree ___ doctoral degree 7 On average, how many hours per week does your spouse or partner work? ___ 0-5 ___ 6-20 ___ 21-40 143

PAGE 144

___ 41 or more 8 Family income per year (check one): ___ less than $5,000 ___ $5,001-$10,000 ___ $10,001-$20,000 ___ $20,001-$30,000 ___ $30,001-$40,000 ___ $40,001-$50,000 ___ $50,001-$60,000 ___ $60,001-$70,000 ___ $70,001-$80,000 ___ $80,001-$90,000 ___ $90,001-$100,000 ___ $100,001-$150,000 ___ over $150,000 9 How many children (under the ag e of 19) live in your home? ___ 1 ___ 2 ___ 3 ___ 4 ___ 5 ___ 6 or more 10 Your race/ethnicity: ___ Asian/Asian-American ___ Black/African-American ___ Hispanic/Hispanic-American ___ White/Caucasian ___ Other School Valence This scale assesses the parents attraction to or general disposition toward schools, based on his or her prior personal experience with schools. This scale wa s adapted from Walker, Wilkins, Dallaire, Sandler, and Hoover-Demps ey (2005) and Hoover-Dempsey and Sandler (2005); the scale was administered to a sample of 358 parents of students in public school grades 4-6 with an alpha reliability of .84 (Walker et al., 2005; Hoover-Dempsey & Sandler, 2005). 144

PAGE 145

The valence questions employ a 6point Likert-style response form at in which parents are asked to rate their experience regarding selected elem ents of schooling. Each of the elements is on a continuum; one end is anchored by negative ex perience, the other by positive experience (e.g., My school: 1 = disliked, 6 = liked). Participants were asked to respond to the following prompt: People have different feelings about school. Please mark the number on each line below that best describes your fee ling about your school experiences when you were a student. The following items constitute the valence section: 1 My school: disliked 1 2 3 4 5 6 liked 2 My teachers: were mean 1 2 3 4 5 6 were nice 3 My teachers: ignored me 1 2 3 4 5 6 cared about me 4 My school experience: bad 1 2 3 4 5 6 good 5 I felt like: an outsider 1 2 3 4 5 6 I belonged 6 My overall experience: failure 123456 success Variable A Parental Report of Encouragement This scale assesses parents self-reports of behaviors during involvement focused on encouraging the child in his or her schoolwork and learning. Adapted from Martinez-Pons (1996) by Hoover-Dempsey and Sandler (2005); the scale wa s administered to a sample of 358 parents of public school students in grades 4-6 with an alpha reliability of .92 (Hoover-Dempsey & Sandler, 2005). 145

PAGE 146

We encourage this child 1. when he or she doesnt feel like doing schoolwork. 2. ... when he or she has trouble organizing schoolwork. 3. ... to try new ways to do schoolwork wh en he or she is having a hard time. 4. ... to be aware of how he or she is doing with schoolwork. 5. ... when he or she has trouble doing schoolwork. 6. ... to look for more information about school subjects. 7. ... to develop an interest in schoolwork. 8. ... to believe that he/she can do well in school. 9. ... to stick with problems until he/she solves it. 10. ... to believe that he/she can learn new things. 11. ... to ask other people for help when a problem is hard to solve. 12. ... to explain what he/she thinks to the teacher. 13. ... to follow the teachers directions. Variable B Parental Report of Modeling This scale assesses parents self-reports of modeling strategies for so lving problems, selfregulating, and learning. The scale was adapted from Martinez-Pons (1996) by Hoover-Dempsey and Sandler (2005). The scale wa s administered to a sample of 358 parents of public school students in grades 4-6 with an alpha relia bility of .94 (Hoover-Dempsey & Sandler, 2005). We show this child that we 1. like to learn new things. 2. ... know how to solve problems. 3. ... enjoy figuring things out. 4. ... do not give up when things get hard. 5. ... ask others for help when a problem is hard to solve. 6. ... can explain what we think to others. 7. ... can learn new things. 8. ... want to learn as much as possible. 9. ... like to solve problems. 10. ... try different ways to solve a problem when things get hard. 146

PAGE 147

Variable C Parental Report of Reinforcement This scale assesses parent self -reports of parental reinfor cement behaviors with the child. The scale was adapted from Martinez-Pons ( 1996) by Hoover-Dempsey and Sandler (2005) and was administered to a sample of 358 parents of public school students in grades 4-6. The scale has an alpha reliability of .96, as repor ted in Hoover-Dempsey and Sandler (2005). We show this child we like it when he or she 1. wants to learn new things. 2. ... tries to learn as much as possible. 3. ... has a good attitude about doing his or her homework. 4. ... keeps working on homework even when he or she doesnt feel like it. 5. ... asks the teacher for help. 6. ... explains what he or she thinks to the teacher. 7. ... explains to us what he or she thinks about school. 8. ... works hard on homework. 9. ... understands how to solve problems. 10. ... sticks with a problem until he or she solves it. 11. ... organizes his or her schoolwork. 12. ... checks his or her work. 13. ... finds new ways to do schoolwork when he or she gets stuck. Variable D Parental Report of Instruction This scale assesses parent self -reports of instruc tional behaviors with children during the course of involvement activiti es. The scale was adapted from Martinez-Pons (1996) by HooverDempsey and Sandler (2005) and was used with a sample of 358 parents of public school students in grades 4-6. The scale has an alpha reliability of .92, as reported in Hoover-Dempsey and Sandler (2005). We teach this child 1. to go at his or her own pace while doing schoolwork. 2. ... to take a break from his or her wo rk when he or she gets frustrated. 3. ... how to check homework as he or she goes along. 4. ... how to get along with othe rs in his or her class. 147

PAGE 148

5. ... to follow the teachers directions. 6. ... how to make his or her homework fun. 7. ... how to find out more about the things that interest him or her. 8. ... to try the problems that he lp him or her learn the most. 9. ... to have a good attitude about his or her homework. 10. ... to keep trying when he or she gets stuck. 11. ... to stick with his or her homework until he or she finishes it. 12. ... to work hard. 13. ... to communicate with the teacher when he or she has questions. 14. ... to ask questions when he or she doesnt understand something. 15. ... to make sure he or she understand s one part before going onto the next. 148

PAGE 149

APPENDIX E STUDENT SURVEY This survey consists of four separate meas urement variables: stude nt report of parents use of encouragement, student re port of parents use of modeli ng, student report of parents use of reinforcement, and student report of parent's use of instruction. For each of the measurement variables, students were asked to respond to the following prompt: Dear Student, Families do many d ifferent things when they help children with school. Please think about how your family helps y ou with school and fill in the circle that ma tches what is most true for them. Thank you! The assessment employs a four-point Likert-type s cale: 1 = not true, 2 = a little true, 3 = pretty true, 4 = very true. Variable A Student Report of Parents Use of Encouragement This scale assesses the extent to which a student perceives that his or her parent (or other family member identified by the student) encourag es student behaviors, interests, and beliefs conducive to achievement during a representative parental invol vement activity, monitoring or helping the student with homework. The scal e was adapted from Martinez-Pons (1996) by Hoover-Dempsey and Sandler (2005). The scale was administered to a sample of 358 public school students in grades 4-6 with an alpha reliability of .87 (Hoover-Dempsey & Sandler, 2005). The person in my family who usually help s me with my homework encourages me 1. when I don't feel like doing my schoolwork. 2. when I have trouble organizing my schoolwork. 3. to be aware of how I'm doing with my schoolwork. 4. to try new ways to do schoolwork when I'm having a hard time. 5. when I have trouble doing my schoolwork. 149

PAGE 150

6. to look for more information about school subjects. 7. to develop an interest in schoolwork. 8. to believe that I can do well in school. 9. to believe that I can learn new things. 10. to ask the teacher for help when a problem is hard to solve. to follow the teachers directions. 12. to explain what I think to the teacher Variable B Student Report of Parents Use of Modeling This scale assesses the extent to which a student perceives that his or her parent (or other family member identified by the student) encourag es student behaviors, interests, and beliefs conducive to achievement during a representative parental invol vement activity, monitoring or helping the student with home work. The scale was adapted from Martinez-Pons (1996) and reported in Hoover-Dempsey and Sa ndler (2005). It includes items in the Parental Report of Modeling Scale, altered as appropriate for student perspective and response. The scale achieved an alpha reliability of .75 as administered to a sample of 358 public school students in grades 4-6 (Hoover-Dempsey & Sandler, 2005). The person in my family who usually helps me with my homework 1. likes to learn new things. 2. wants to learn as much as possible. 3. likes to solve problems. 4. enjoys figuring things out. 5. knows how to solve problems. 6. tries a different way if he or she has trouble solving a problem. 7. doesnt give up when things get hard. 8. can learn new things. 9. asks other people for help when a problem is hard to solve. 10. can explain what he or she thinks to other people. 150

PAGE 151

Variable C Student Report of Parents Use of Reinforcement This scale assesses the extent to which a student perceives that his or her parent (or other family member identified by the student) reinfor ces student behaviors, interests, and beliefs conducive to achievement during a representative parental invol vement activity, monitoring or helping the student with home work. The scale was adapted from Martinez-Pons (1996) and reported in Hoover-Dempsey and Sa ndler (2005). It includes items in the Parental Report of Reinforcement Scale, altered as appropriate for student perspe ctive and response. The scale achieved an alpha reliability of .87 when admini stered to a sample of 358 public school students in grades 4-6 (Hoover-Dempsey & Sandler, 2005). The person in my family who usually helps me with my homework shows me that he or she likes it when I 1. try to learn as much as possible. 2. have a good attitude about doing my homework. 3. want to learn new things. 4. check my work. 5. understand how to solve problems. 6. organize my schoolwork. 7. find new ways to do my work when I get stuck. 8. stick with a problem until it gets solved. 9. work hard on my homework. 10. keep working on my homework even when I don't feel like it. 11. ask the teacher for help. 12. explain what I think to the teacher. 13. explain what I think about school to him or her. 151

PAGE 152

Variable D Student Report of Parent's Use of Instruction This scale assesses the extent to which a student perceives that his or her parent (or other family member identified by the st udent) instructs or teaches the student during a representative parental involvement activity, monitoring or help ing the student with homework. The scale was adapted from Martinez-Pons (1996) and reporte d in Hoover-Dempsey and Sandler (2005). It includes items in the Parental Report of Instruct ion Scale, altered as appropriate for student perspective and response. The s cale achieved an alpha reliability of .86 when administered to a sample of 358 public school students in grades 4-6 (Hoover-Dempsey & Sandler, 2005). The person in my family who usually helps me with my homework teaches me 1. ways to make my homework fun. 2. how to find out more about things that interest me. 3. to try the problems that help me learn the most. 4. to have a good attitude about my homework. 5. to make sure I understand one part before I go on to the next. 6. to take a break from my work when I get frustrated. 7. how to check my homework as I go along. 8. to go at my own pace while doing my homework. 9. to keep trying when I get stuck. 10. to stick with my home work until I get it all done. 11. to work hard. 12. to ask questions when I don't understand something. 13. how to get along with others in my class. 14. to follow the teacher's directions. 15. to communicate with the te acher when I have questions. 152

PAGE 153

LIST OF REFERENCES Allen, I.E. & Seaman, J. (2006). Making the Gr ade: Online Education in the United States. Needham, MA: Sloan Consortium. Allen, I.E., Seaman, J. (2007). Online Nation: Five Years of Growth in Online Learning. Needham, MA: The Sloan Consortium. Anderson, R.C., Hiebert, E., Scott, J.A. & Wilkinson, I.A.G. (1985). Becoming a nation of readers: The report of the commission on reading. Washington, DC: The National Institute of Education. Antosca, F.E. (1996). The effect of student perception of parental involvement on student achievement. University of Massachuse tts, Amherst Dissertation. DAI-A 57/10, p. 4315, Apr 1997. Aughinbaugh, A., Pierret, C.R., & Rothstein, D.S. (2005). The impact of family structure transitions on youth achievement: Evidence from the children of the NLSY79. Demography, (42)3, 447-468. Balli, S. J. (1996). Family divers ity and the nature of parental involvement. Educational Forum 60: 149. Baumrind, D. (1971). Current patterns of pa rental authority. Developmental Psychology Monographs, 4(1, 2). Baumrind, D. (1991). Parenting styles and adolescent development. In R. M. Lerner, A. C. Petersen, & J. Brooks-Gunn (Eds.), En cyclopedia of adolescence (Vol. 2 746.) New York: Garland. Bernard, R. M., Abrami, P. C., Lou, Y., Borokhovs ki, E., Wade, A., Wozney, L., Wallet, P. A., et al. (2004). How does distance education compare with classroom instruction? A metaanalysis of the empirical literature. Revi ew of Educational Research, 74(3), 379-439. Black, E.W., Dawson, K. & Priem, J. (2008). Data for free: Using LMS activity logs to measure community in an online course. The In ternet and Higher Education, 11, 65-70. Black, E.W., Ferdig, R.E. & DiPietro, M. (2008). An overview of evaluative instrumentation for virtual high schools. The American Jour nal of Distance Education, 22(1), 24-45. Black, E.W., Greaser, J. & Dawson, K. (In press) Academic dishonesty in traditional and online classrooms: Does the Media Equation hold true? Paper submitted to the Journal of Asynchronous Learning Networks. Blomeyer, R. (2002). Online learning for k-12 st udents: What do we know now? Naperville, IL: North Central Regional Educational Labor atory. Retrieved September 22, 2007 from http://www.ncrel.org/tech/elearn/synthesis.pdf 153

PAGE 154

Bracey, G.W. (1996). SES and involvement. Phi Delta Kappan 78: 169. Brito, J. & Waller, H. (1994). Part nership at a price? In R. Merttens, D. Maers, A. Brown, & J. Vass (Eds.), Ruling the margins: Problematizin g parental involvement. London: University of London, Institute of Education, pp. 157-166. Brody,G.H. (1995). Linking family processes and academic competence among rural African American youths. J. Marr. Family 57: 567. Bulkley, K., & Fisler, J. (2003). A decade of charter schools: From theory to practice. Educational Policy 17 (3): 317 42. Cancio, E. J., West, R. P., & Young, K. R. (2004). Improving mathematics homework completion and accuracy of students with EBD through self-management and parent participation. Journal of Emotional and Behavioral Disorders, 12(1), 9. Cavanaugh, C. (2002). Distance education quality: Success factors for resources, practices and results. In R. Discenza, C.Howard, K& K. Schenk (Eds.) The Design and Management of Effective Distance Learning Programs (171189). Hershey, PA: Idea Group Publishing. Cavanaugh, C., Gillan, K. J., Kromrey, J., Hess, M., & Blomeyer, R. (2004). The effects of distance education on k-12 student outcomes: A meta-analysis. Naperville, IL: Learning Point Associates. Cereijo, M. V. P., Young, J., & Wilhelm, R. W. (2001). Factors Facilitati ng Student Participation in Asynchronous Web-Based Courses. Journal of Computing in Teacher Education, 18. Cervone, B. T., and O'Leary, K. (1982). A conc eptual framework for parent involvement. Educational Leadership 40, 48-49. Chong, S.M. (1998). Models of asynchronous com puter conferencing for collaborative learning in large classrooms. In C.J. Bonk & K.S. Kind (Eds.), Electronic collaborators: Learnercentered technologies for lite racy, apprenticeship and di scourse (pp. 157-182). Mahwah, NJ: Lawrence Erlbaum Associates. Chickering, A., & Ehrmann, S.C. (1996). Implem enting the seven principles: Technology as level. AAHE Bulletin, October, 3-6. Christenson, S.L., Rounds, T., and Gorney, D. (1992). Family factors an d student achievement: An avenue to increase students succ ess. School Psychology Quarterly, 7, 178-206. Clark, T. (2001). Virtual schools: Trends and issues. Distance Learning Resource Network. Retrieved November 14, 2004, from http://www.wested.org/onlin e_pubs/virtualschools.pdf. Clark, T. & Blomeyer, B. (2007). Georgia virtua l school evaluation: Final report. Submitted to K. Clements, Georgia Department of Education. 72 manuscript pages. 154

PAGE 155

Cohen, J. (1988). Statistical power analysis for the behavioral sciences 2nd ed. Hillsdale, NJ: Earlbaum. Cohen, J. (1992). A power primer. Ps ychological Bulletin, 112(1), 155-159. Cook, C., Heath, F., & Thompson, R.L. (2000). A meta-analysis of response rates in webor internetbased surveys. Educational and Psychological Measurement, 60, 821-836. Cotton, K., & Reed-Wikelund, K. (1989). Pare nt involvement in education. NWREL, http://www.nwrel.org/scpd/sirs/3/cu6.html. Coussement, S. (1995). Educational telecommu nication: Does it work? An Attitude study. (ERIC No. ED 391 465). Curry, D. B. (2000). Collaborative, connected and experiential learning: Re flections of an online learner. Retrieved 7/15/2007, from http:/ /www.mtsu.edu/~itconf/proceed01/2.html. DeBard, R., & Guidera, S. (1999-2000). Adap ting asynchronous communication to meet the seven principles of effective teaching. E ducational Technology Systems, 28(3), 219-230. Del Litke, C. (1998). Virtual schooling at the middle grades: A case study Journal of Distance Education, 13(2). Department of Education (2008). 2006-2007 report card. Georgia Depart ment of Education. Retrieved 9/4/2008, from http://public.doe.k12.ga.us/ReportingFW.aspx?PageReq=102&StateId=ALL&T=1 Desforges, C. & Abouchaar, A. (2003). The imp act of parental involve ment, parental support and family education on pupil achievement and adjustment: a literature review. Research Report no. 433 (London, Department for Education and Skills). Donnell-Kay Foundation (2007). Trujillo commission on online education: Final findings and recommendations. Denver, CO: Donnell-Kay Foundation. Dickson, P. (2005). Toward a deeper understanding of student performance in virtual high school courses: Using quantitative analysis and da ta visualization. Nape rville, IL: NCREL. DiPietro, M.K. (2008). Understanding virtual sc hool pedagogy: An an alysis exploring the beliefs, goals and practices of virtual school teachers. Doctoral dissertation. University of Florida, Gainesville, FL. Dillman, D. A. (2007). Mail and internet surveys: The tailored design method, 2nd ed. Hoboken, NJ: John Wiley & Sons. Duncan, O.D. (1975). Introduction to structural equation models New York: Academic Press. 155

PAGE 156

Dutt-Doner, K.M., & Powers, S.M. (2000). The use of electronic communication to develop alternative avenues for classroom disc ussion. Journal of Technology and Teacher Education, 8(2), 153-172. Eastmond, D.V. (1995). Alone but together: Adult distance study through computer conferencing. Cresskill, NJ: Hampton Press. Eccles, J.S. & Harold, R.D. (1993). Parent-schoo l involvement during the early adolescent years, Teachers College Record 94, 568. Eccles, J. S., & Harold, R. D. (1996). Family involvement in childre ns and adolescents schooling. In A. Booth & J. F. Dunn (Eds.) Family-school links: How do they affect educational outcomes? (pp. 3). Mahwah, NJ: Erlbaum. Edwards, P.A. (2004). Childrens literacy development: Making it happen through school, family, and community involvement. Boston: Pearson. Epstein, J. L. (1990). Single parents and the schools: The effect of marital status on parent and teacher evaluations. In M. Hallinhan (Ed.), Change in societal institutions (pp. 911). New York: Plenum. Espstein, J.L. (1991). Paths to partnership. Phi Delta Kappan, 72(5), 344-350. Epstein, J.L. (1995). School/f amily/community partnerships. Phi Delta Kappan, 76(9), 701-713. Epstein, J.L. (1996). New connections for soci ology and education: Contributing to school reform. Sociology of Education, 69(2), 6-23. Epstein, J.L. (2001). Building bridges of ho me, school, and community: The importance of design. Journal of Education for Studen ts Placed at Risk, 6(1/2), 161-168. The Family School Partnership Lab at Vanderbil t University. (2007). Scale descriptions. The Family School Partnership Lab at Vanderb ilt University. Retrieved 11/20/2007, from http://www.vanderbilt.edu/Peabody/fam ily-school/scaledescriptions.html. Ferdig, R.E. (2006). End of course survey parents. Retrieved June 3, 2006, from http://ferdig.coe.ufl.edu/vhs/surveys/parent.htm Ferdig, R.E., Amberg, H.G., Elder, J.H., Donaldson, S.A., Valcante, G., Bendixen, R (submitted). Autism and family interventions: Th e use of the Internet to train fathers of autistic children. Paper submitted to the Inte rnational Journal of Web-based Learning and Teaching Technologies. Ferdig, R.E., Amberg, H.G., Elder, J.H, Valcante, G., Donaldson, S., & Bendixen, R. (2008). The development of a web-based platform for fathers of children with autism. Learning Technology [Online serial], 10 (2), 11-15. Available from: http://lttf.ieee.org/learn_tech/issues/april2008/lt_april2008.pdf 156

PAGE 157

Ferdig, R.E & Cavanaugh, C. Ed. (in press). Effective practice in virt ual schools. To be published by the North American Council fo r Online Learning (NACOL); Vienna, VA. Ferdig, R.E., Papanastasiou, E., & DiPietro, M. (2005). Teaching and learning in collaborative virtual high schools. Report submitted to the North Central Regional Educational Laboratory as part of the K12 Online Lear ning Initiative. 65 manuscript pages. Ferdig, R.E. & Roehler, L.R. (2003). Student engagement in electronic discussions: Examining online discourse in literacy pr e-service classrooms. Journal of Research on Technology in Education, 36(2), 119-136. Florida TaxWatch (2007). Final report: A compre hensive assessment of fl orida virtual school. Florida TaxWatch Center for Educational Performance and Accountability. Retrieved 10/1/2008 from http://www.nacol.org/docs/FL VS_Final_Final_Report(10-15-07).pdf Florida Virtual School (2006). Q& A for flvs staff on vsaA performance management system for virtual schools. Re trieved 9/5/2008 from www.flvs.net/general/documnets/Press%20Release%20VSA.F.doc Flouri, E. & Buchanan, A. (2003). The role of father involvement and mother involvement in adolescents psychological well-being. British Journal of Social Work, 33, 299-406. Frank, C. (2005). Measurement of social capital: Reference docume nt for public policy research, development, and evaluation. Otta wa: Policy Research Initiative. Gatti, G. & Harwell, M. (1998). Advantages of computer programs over power charts for the estimation of power. Journal of Statistics Education, 6(3). Green, C.L. & Hoover-Dempsey, K. (2007). W hy do parents homeschool? A systematic examination of parental involvement. E ducation and Urban Society, 39(2), 264-285. Green, C., Walker, J., Hoover-Dempsey, K. & Sandler, H. (2007). Parents' motivations for involvement in children's education: An empiri cal test of a theoretical model of parental involvement. Journal of Edu cational Psychology, 99(3), 532-533. Greene, J.P., Peterson, P.E., & Du, J. (1999). E ffectiveness of school choice: The Milwaukee experiment. Education and Urban Society, 31, 190-213. Griffith, J. (1998). The relation of school structure and social envi ronment to parent involvement in elementary schools. Elementary School Journal, 99, 53. Griswold, P.A., Cotton, K.J., & Hansen, J.B. (1986). Effective compensatory education sourcebook. Washington, D.C.: U.S. Department of Education. Gibson, C. & Graf, A. (1992). Impact of adults preferred learning styl es and perception of barriers on completion of external baccalaure ate degree programs. Journal of Distance Education, 7(1), 39. 157

PAGE 158

Grolnick, W.S. & Slowiaczek, M.L. (1994). Pare nts involvement in childrens schooling: A multidimensional conceptualization and motiv ational model. Child Development, 65, 237252. Grolnick, WS, Benjet, C, Kuri wski, CO, & Apostoleris, NH. (1997). Predictors of parent involvement in children's schooling. J ournal of Educational Psychology, 89, 538-548 Hardy, D.W. & Boaz, M.H. (1997). Learner de velopment: Beyond the technology. In T.E. Cyrs. (Ed.) Teaching and learning at a distance: What it takes to effectively design, deliver, and evaluate programs (pp. 41 48) San Francisco: Jossey-Bass. Harter, S. (1996). Teacher and classmate influences on scholastic motivation, self-esteem, and level of voice in adolescents. In Social Motivation: Understanding Childrens School Adjustment, ed. J Juvonen, K Wentzel, pp. 11. New York: Cambridge Univ. Press Harvard Family Research Project (2008). St rengthening parent involvement policy. The Evaluation Exchange, 14(1&2). Hassel, B. & Terrel-Godard, M. (2004). How can virtual schools be a vibrant part of meeting the choice provisions of the no child left behind act? Virtual School Report. http://www.doe.virginia.gov/VDOE/Instruction/title1/Hassel-TerrellVirtualSchools.pdf Haythornthwaite, C., Kazmer, M., Robins, J., and Shoemaker, S. (2000). Making Connections: Community among computer-supported distan ce learners. Paper presented at the Association for Library and Information Science Education 2000 Conference. San Antonio, Texas. Retrieved July 15, 2007 from: http://www.alise.org/conferences/conf00_Haythornthwaite_Making.htm Henderson, A. (Ed.). (1981). Parent participation-student achie vement: The evidence grows. [occasional paper]. Columbia, MD: National Committee for Citizens in Education. (ERIC Document Reproduction Service No. ED206963). Henderson, A. (1987). The evidence continues to grow: Parental invol vement improves school achievement. Columbia, MD: National Committee for Citizens in Education. Henderson, J.M. (2006). Power relations w ithin the homework process. Unpublished dissertation. University of Sterling. Henderson, A. T., & Berla, (Eds.). (1994a). A new ge neration of evidence: The family is critical to student achievement. National Committee for Citizens in Education, Washington, DC. ERIC (ED 375968). Henderson, A. T., & Berla, N. (1994b). The fami ly is critical to student achievement. Washington, DC: Center for Law and Education. Henderson, A. T.& Berla, N. (1995). A new generation of evidence: The family is critical to student achievement. Washington DC: Center for Law and Education. 158

PAGE 159

Henderson, A., & Mapp, K. L. (2002). A new wave of evidence: The impact of school, family, and community connections on student achieve ment. Southwest Educational Development Laboratory, Austin, TX. (ERIC Docume nt Reproduction Service No. ED474521) Heymann, S.J., & Earle, A. (2000). Low-income pa rents: How do working conditions affect their opportunity to help school-age children at risk? American Educational Research Journal, 37(4), 833-848. Hoover-Dempsey, K.V., & Sandler, H.M. (1995). Pare ntal involvement in children's education: Why does it make a difference? Te achers College Record, 97, 310-331. Hoover-Dempsey, K.V., Bassler, O.C., & Brissie, J.S. (1992). Explorations in parent-school relations. Journal of Edu cational Research, 85, 287-294. Hoover-Dempsey, K.V., Battiato, A.C., Walker, J.M. T., Reed, R.P., DeJong, J.M., Jones, K. P. (2001). Parental involvement in homework. Educationa l Psychologist, (36)3, 195-209. Hoover-Dempsey, K. V., Walker, J. M. T., Sandler H. M., Whetsel, D., Green, C. L., Wilkins, A. S., & Closson, K. E. (2005). Why do parent s become involved? Research findings and implications. Elementary School Journal, 106(2); 105-130. Hoover-Dempsey, K.V., & Sandler, H.M. ( 2005). Final performance report for oeri grant # r305t010673: The social cont ext of parental involvement: A path to enhanced achievement. Presented to Project Monitor, Institute of Education Sciences, U.S. Department of Education, March 22, 2005. Jackson, T.W., Burgess, A., & Edwards, J. (200 6). Technical opinion: A simple approach to improvising email communication. Communi cations of the ACM, 49(6), 107-109. Jeynes, W.H. (2007). The relationship between pa rental involvement and urban secondary school student academic achievement: a meta-a nalysis. Urban Education, 42, 82-110. Joy, E. H., & Garcia, F. E. (2000). Measuri ng learning effectiveness: A new look at nosignificant-difference findings. Journal of Asynchronous Learning Networks, 4(1), 33-39. Keegan, D. (1996). Foundations of distance education, 3rd Ed. New Jersey: Routledge. Keith, T.Z. (1991). Parent involvement and achie vement in high schools. In Silvern, S. (Ed.), Advances in Reading/Language Research: Literacy Through Family, Community, and School Interaction (Vol. 5), Greenwich, CT: JAI Press. Kellaghan, T., Sloane, K., Alvar ez, B., & Bloom, B. S. (1993). The home environment and school learning: Promoting parental involve ment in the education of children. San Francisco: Jossey-Bass, Inc. Kerbow, D. & Bernhardt, A. (1993) Parent inte rvention in the school: Th e context for minority involvement. In B. Schneider and K. Coleman (e ds.), Parents, their children, and schools. San Francisco: Westview Press. 159

PAGE 160

Kiekel, J. (2007). Characteristics of high school online educational programs: A multiple case study. King, F. B., Young, M. F., Drivere-Richmond, K., & Schrader, P. G. (2001). Defining distance learning and distance education. Educationa l Technology Review, 9 (1). Available at: http://www.aace.org/pubs/etr/king2.cfm. Kozma, R. and Shank, P. (1998). Connecting with the 21st century: Tech nology in support of educational reform, in ASCD Yearbook: Learning With Technology, C. Dede, Editor. Association for Supervision and Curr iculum Development: Alexandria: VA. Kreppner, K. and Lerner, R. (1989). Family systems and the life span development : Issues and perspectives. In Kreppner, K. and Lerner, R, (Eds.), Family Systems and Life-Span Development. Lawrence Erlbaum Asso ciates: Hillsdale, New Jersey. Lareau, A. (1987). Social class differences in family-school relationshi ps: The importance of cultural capital. Sociology of Education, 60, 70. Lareau, A. (1989). Home advantag e: Social class and parental intervention in elementary education. New York: Falmer Press. Lareau, A. (2000). Home advantag e: Social class and parental intervention in elementary education 2nd Ed. Lanham, MD: Rowman & Littlefield Publishers, Inc. Lareau, A., & Shumar,W. (1996). The problem of individualism in family-school policies. Sociology of Education, 69, 24-39. Leahy, T.H., & Harris, R.J. (2001). Learning and cognition, 5th Ed. Upper Saddle River, NJ: Prentice Hall. Little, R.J.A. & Rubin, D.B. (1987) Statistical an alysis with missing data. J. Wiley & Sons, New York. Lubienski, C. (2005). Public schools in mark etized environments: Shifting incentives and unintended consequences of competition-based educational reforms. American Journal of Education, 111(4), 464-486. Ma, X. (1999). Dropping out of advanced mathematics: The effects of parental involvement. Teachers College Records, 101(1). 60-81. Maeroff, G. (2003). A classroom of one. New York: Palgrave MacMillan. Manfreda, K.L., Bosnjak, M., Berzelak, J., Hass, I., & Vehovar, V. (2008). Web-surveys versus other survey modes: A meta-a nalysis comparing response rate s. International Journal of Market Research, 50(1), 79-104. Martinez-Pons, M. (1996). Test of a model of parental inducem ent of academic self-regulation. Journal of Experimental Education, 64, 213-227. 160

PAGE 161

Martinez-Pons, M. (2001). The psychology of teaching and learning: A three step approach. New York, Continuum. Mau, W. (1997). Parental influences on the high school students academic achievement: A comparison of asian immigrants, asian americans, and white americans. Psychology in the Schools, 34, 267-277. Max, L. & Onghena, P. (1999). Some issues in the statistical analysis of completely randomized and repeated measures designs for speech, la nguage, and hearing research. Journal of Speech, Language, and Hear ing Research, 42, 261. MacCallum, R.C., Browne, M.W., & Sugawara, H.M. (1996). Power analysis and determination of sample size for covariance structure modeling, Psychological Methods 1(2): 130-149. McConnell, D. (2000). Implementing computer supported cooperative learning. London: Kogan Page Limited. McGrath, D.J. and Kuriloff, P.J. (1999) "Theyre going to tear the doors off this place": Uppermiddle class parent involvement and the e ducational opportunities of other peoples children, Educational Policy, 13, 603-629. McNeal, R.B. (1999). Parental involvement as social capital: Differential effectiveness on science achievement, truancy, and dropping out. Social Forces, 78(1), 117-145. Mills, S.C. (2003). Implementing online secondary education: An evaluation of a virtual high school. Proceedings of the 14th International Conference of the Society for Information Technology and Teacher Education. Albuquerque NM: Association for the Advancement of Computers in Education. Mills, S.C. & Roblyer, M.D. (2004). Implementing online high school programs: An evaluation study. In Simonson, M.R. (Ed.), Proceedings of selected research and development presentations at the 2003 inte rnational convention of the association for educational communications and technology. Ames, IA: Research and Theory Division, AECT. Moore, M. G., & Kearsley, G. (1996). Distan ce education: A systems view. Belmont, CA: Wadsworth. National Center for Education St atistics (NCES). (1997). Fathers involvement in their childrens schools. Washington, DC: U.S. Department of Education National Center for Education Statistics (NCES). (2005). The condition of education. Jessup, MD: U. S. Department of Education, Institute of Educational Sciences. North Central Regional Educa tional Laboratory (NCREL). (2002) Policy issues 11: Virtual schools and e-learning in k-12 environments: emerging policy and practice. Naperville, IL.: Learning Point Associates. 161

PAGE 162

Nunnally, J.C. & Bernstein, I.H. (1994). Psychometric theory (3rd edition). McGraw-Hill Series in Psychology, McGraw-Hill, Inc., New York: NY, 264-265. Optimal Performance, Inc. (2007). Parent su rvey report, 2006-2007: Florida virtual school. Retrieved 10/10/2007 from http://www.flvs .net/educators/documents/pdf/currentevals/Parent%20Survey%20Report2007xx.pdf Picciano, A.G. & Seaman, J. (2007). K online learning: A survey of U.S. school district administrators. Needham, MA: The Sloan Consortium. Pomerantz, E.M., Moorman, E.A, Litwack, S.D. (2007). The how, whom, and why of parents involvement in childrens academic lives: More is not always better. Review of Educational Research, 77, 373-410. Porter, S. R., & Whitcomb, M. E. (2003). The imp act of contact type on web survey response rates. Public Opinion Quarterly, 67(4), 579-588. Power, T.G. (2004). Stress and coping in chil dhood: The parents role Parenting, 4(4), 271-217. Reeves, B. & Nass, C. (1996). The media equati on: How people treat com puters, television, and new media like real people and places. New York: Cambridge University Press. Rice, K. L. (2006). A comprehensive look at distan ce education in the k-12 context. Journal of Research on Technology in Education, 38(4), 425-448. Richards, C., & Ridley, D. (1997) Factors affecting college students persistence in online computer-managed instruction. Co llege Student Journal, 31, 490-495. Ritblatt, S.N., Beatty, J.R., Cronan, T.A., & Ochoa, A.M. (2002). Relationships among perceptions of parent involvement, time allocation and demogra phic characteristics: Implications for policy formation. Journal of Community Psychology, 30(5), 519-549. Roblyer, M.D. (1999). Is choice important in distance learning? A study of student motives for taking Internet-based courses at the high school and community college levels. Journal of Research on Computing in Education, 32(1). Roblyer, M.D. & Elbaum, B. (2000). Virtual learning? Research on virtual high schools. Learning & Leading with Technology, 27(4), 58-61. Romiszowski, A.J. & Ravitz, J. (1997). Comput er-mediated communication. In C.R. Dilles & A. Romiszowski (Eds.). Instruc tional developmental paradi gms. (pp. 745-768). Englewood Cliffs, NJ: Educational Technology Publications. Rose, L. C., & Gallup, A. M. (2000). The 32nd annual phi delta kappa/gallup poll of the public's attitudes toward the public schools. Bloomington, IN: Phi Delta Kappa International. Available online: http://www.pdki ntl.org/kappan/kimages/kpollv82.pdf 162

PAGE 163

Rovai, A. (2002a). Development of an instrument to measure classroom community. Internet and Higher Education, 5. Rovai, A. P. (2002b). Sense of community, perc eived cognitive learning, and persistence in asynchronous learning networks. Internet and Higher Education, 5 (4), 319-332. Rovai, A.P. & Wighting M. J. (2005) Feelings of alienation and community among higher education students in a virtua l classroom. The Internet and higher education, 8(2), 97-110 Rubin, D.B. (1976) Inference and missing data. Biometrika, 63, 581-592. Russell, G. (2004). Virtual Schools: A Critical View. In C. Cavanaugh (Ed.), Development and Management of Virtual Schools: Issues and Trends, (pp. 1-26) Schneider, B., Carnoy, M., Kilpatrick, J., Schmid t, W.H. & Shavelson, R.J. (2007). Estimating causal effects using experimental and observational designs: A think tank white paper. Washington, D.C.: American Educational Research Association Schweinhart, L. J., & Weikart D. P. (1992). The high/scope perry Preschool program study, similar studies and their implications for Public Policy in the U.S. In D. Stegelin, (Ed.), Early Childhood Education: Policy Issues for the 1990s. Norwood, NJ: Ablex Publishing Corporation. Shaver, A. V, & Walls, R. T. (1998) Effect of title I parent involvement on student reading and mathematics achievement. Journal of Resear ch and Development in Education, 31(2), 9097. Shumow, L. & Miller, J.D. (2001). Parents at -home and at-school academic involvement with young adolescents. Journal of Early Adolescence, 21, 69-91. Skinner, B.F. (1989). Skinner, B. F. (1989).The origins of cognitive thought. American Psychologist, 44, 13-18. Smith, M., Clark, T. & Blomeyer, B. (2005). A synthesis of new research on k-12 online learning. North Central Regional Educatio nal Laboratory (NCREL). Naperville, IL.: NCREL. Solomon, Y., Warin, J. & Lewis, C. (2002). Helping with homework? Homework as a site of tension for parents and teenagers. British E ducational Research Journal, 28(4), 603-622. Steinberg, L. (2001). We know some things: Parentadoles cent relationships in retrospect and prospect. Journal of Research on Adolescence, 11(1), 1-19. Swan, K. (2003). Learning effectiveness: what the research tells us In J. Bourne & J. C. Moore (Eds) Elements of Quality Online Education, Practice and Direction. Needham, MA: Sloan Center for Online Education, 13-45. 163

PAGE 164

Sy, S.R., Rowley, S.J. & Schulenberg, J.E. ( 2007). Predictors of parent involvement across contexts in asian american and european american families. Journal of Comparative Family Studies, 38(1), pp.1-29. Tiene, C.D. (2000) Online discussions: a survey of advantages and disa dvantages compared to face-to-face discussions. Journal of Educa tional Multimedia and Hypermedia, 9(4), 371384. Trost, K., Biesecker, G., Stattin, H., & Kerr, M. (2007). Not wanting pare nts involvement: Sign of autonomy or sign of problems. European Journal of Developmental Psychology, 4(3), 314-331. Tucker, B. (2007). Laboratories of reform: Virtual high schools and innovation in public education. Education Sector Repor ts. Retrieved 2/15/2008, available: http://www.educationsector.or g/usr_doc/Virtual_Schools.pdf Tuten, T. L. (1997). Getting a foot in the el ectronic door: Understanding why people read or delete electronic mail (ZUMA Working Paper No. 97/8). Mannheim, Germany: ZUMA. Twigg, C. A. (2003). Improving quality and re ducing cost: Designs for effective learning. Change, 35(4), 22-29. Tyler, J.R. & Tang, J.C. (2003). When can I exp ect an email response? A study of rhythms in email usage. Proceedings of the Eighth European Conference on Computer-Supported Cooperative Work: Helsinki, Finland. University of Florida Institutional Review Bo ard (IRB) (2007). Policy and procedures manual: University of florida irb02. Retrieved 1/16/ 2008, available: http://irb.ufl.edu/irb02/PPM.doc United States Department of Education (USDOE ) (2007). Innovations in education: Connecting students to advanced courses online. Je ssup, MD: Education Publications Center. US Census Bureau (2008). Geor gia quick-facts from the us census bureau. US Census. Retrieved 9/4/2008 from http://quickfacts. census.gov/qfd/states/13000.html. Vanderbilt Family-School Partnership Lab. ( 2007). The hoover-dempsey and sandler model of parental involvement. Retrieved December 20, 2007, from Vanderbilt Family School Partnership Lab Web site: http://www.vanderb ilt.edu/Peabody/family-school/model.html Walberg, H.J. (1984). Families as partners in e ducational productivity. Phi Delta Kappan, 65(6). 397-400. Watson, J. (2007). A national primer on k-12 online learning. Wash ington, D.C.: North American Counsel for Online Learning. Watson, J., & Ryan, J. (2007). Keeping pace with k-12 online learning: A review of state level policy and practice. Naperville, IL: North Central Regional Educational Laboratory. 164

PAGE 165

Watson, Winograd, & Kalmon (2004). Keeping pace w ith k2 online learning: A snapshot of state-level policy and practice. Naperville IL: North Central Regional Educational Laboratory. Weisgerber, C. & Butler, S.H. (2005). Towards a theoretical framework for the design of interactive online distance learning systems. The Texas Journal of Distance Learning, 2(2), 44-57. Wells, A., & Serna, I. (1996). The politics of cult ure: Understanding local political resistance to detracking in racially mixed schools. Harvard Educational Review, 66, 93-118. Westat and Policy Studies Associates (2001). The longitudinal evaluation of school change and performance in title I schools. Washington, DC: U.S. Department of Education, Office of the Deputy Secretary, Planni ng and Evaluation Service. Walker, J.M.T., Hoover-Dempsey, K.V., Whetse l, D.R., & Green, C.L. (2004). Parental involvement in homework: A review of current research and its impli cations for teachers, after school program staff, and parent leaders. Harvard Family Research Project. Accessed 9/21/2008. Available: http://www.hfrp.or g/publications-resources/browse-ourpublications/parental-involvement-in-homew ork-a-review-of-current-research-and-itsimplications-for-teacher s-after-school-program-staff-and-parent-leaders Walker, J. M., Wilkins, A. S., Dallaire, J., Sandler, H. M., & Hoover-Dempsey, K. V. (2005). Parental involvement: Model revision thr ough scale development. Elementary School Journal, 106(2); 85-104. The White House. (2001). No child left behind. Washington, DC: The White House. Yao, E.L. (1985). A comparison of family character istics of asian American and anglo-american high achievers. International Journa l of Comparative Sociology, 16, 198-208. Yap, K.O. & Enoki, D.Y. (1995). In search of the elusive magic bullet: Parental involvement and student outcomes. School Community Journal, 5(2), 97-106. Yates, J. (2003). Interactive distance learning in pre-k-12 settings: A handbook of possibilities. Westport, CT: Libraries Unlimited. Zucker, A.A., & Kozma, R. (2003). The virtua l high school: Teaching generation v. New York: Teachers College Press. 165

PAGE 166

166 BIOGRAPHICAL SKETCH Erik Wade Black was born in 1975 in Ka nsas City, Missouri. The younger of two children, he grew up in Northern Virginia, attending W.T. Woods on High School in Fairfax City, VA. Erik graduated from Virginia Techs Pa mplin School of Business in 1997 with a BS in Marketing Management. He subsequently spent several years in sales and sales leadership positions with different technology and telecommuni cations companies in both Dallas, Texas and the Washington, D.C. area. Erik earned an MA from The College of Ne w Jersey in May 2005 where he specialized in career counseling and collegiate student services. In August of 2005, Erik enrolled at the University of Floridas School of Teaching and L earning as a doctoral fell ow in the educational technology program. During his tenure as a doctor al fellow, Erik has focused on research in the learning sciences. His investigative initiatives have in cluded studies in online learning, quantitative methods and computational social science applic ations and identity in immersive online and social networking environments. Erik is married to Dr. Nicole M. Paradise Bl ack, an assistant professo r of pediatrics and a pediatric hospitalist currently em ployed by the University of Fl oridas College of Medicine. They are the proud parents of two children: a s on, Brennan, age 2; and a daughter, Ryan, age 1.