<%BANNER%>

The Cost Effectiveness of a Bonus Pay Plan for National Board Certified Teachers in High Poverty Elementary Schools in a...

Permanent Link: http://ufdc.ufl.edu/UFE0044416/00001

Material Information

Title: The Cost Effectiveness of a Bonus Pay Plan for National Board Certified Teachers in High Poverty Elementary Schools in an Urban School District in Florida
Physical Description: 1 online resource (97 p.)
Language: english
Creator: Brown, Anna L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: achievement -- certification -- fcat -- national -- nbct -- nbpts -- poverty -- schools -- student -- teacher -- value-added
Human Development and Organizational Studies in Education -- Dissertations, Academic -- UF
Genre: Educational Leadership thesis, Ed.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Improving standards for teacher certification along with improving student achievement in our nation’s public schools has been an emphasis of public school educators for decades.  The National Board for Professional Teaching Standards developed a national teacher certification process that is recognized by many states as meeting the criteria for certification to teach in the state.  Also recognized as one criteria for meeting “Highly Qualified Teacher” requirements under NCLB, teachers who hold a National Board Certification Credential are often recruited to an paid bonuses from school districts and states nationwide.  The state of Florida ceased its bonus payment to National Board Certified teachers in 2010.  The school district in this study continues to pay bonuses to National Board Certified teachers who teach in high poverty schools. The purpose of this study was to examine the federal and state of Florida legislation related to teacher effectiveness and student achievement and the policies in a central Florida urban school district related to compensation to teachers for holding a NBPTS certification credential.  The study sought to answer the research question: To what extent did Nationally Board Certified teachers in high poverty elementary schools in a central Florida urban school district demonstrate positive student achievement gains in mathematics in fourth and fifth grade as measured by FCAT that was greater than the gains demonstrated by non-nationally board certified teachers in the same setting? The study examined a body of research to gain insight into the polices related to National Board Certified Teachers both in the state of Florida and the school district included in this study.  Student scores from FCAT Mathematics in grades four and five were examined and divided into two groups:  Students taught by teachers who were National Board Certified and those who were not.  The means of the two group were compared using a t-test.  Results indicate that there is no clear relationship between National Board Certification and student achievement in Mathematics in the grade levels included in this study.  Further, this study indicates that the district policy for paying teachers who hold the National Board Certification Credential is not cost effective.
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 Anna L Brown.
Thesis: Thesis (Ed.D.)--University of Florida, 2012.
Local: Adviser: Wood, R. Craig.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044416:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044416/00001

Material Information

Title: The Cost Effectiveness of a Bonus Pay Plan for National Board Certified Teachers in High Poverty Elementary Schools in an Urban School District in Florida
Physical Description: 1 online resource (97 p.)
Language: english
Creator: Brown, Anna L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: achievement -- certification -- fcat -- national -- nbct -- nbpts -- poverty -- schools -- student -- teacher -- value-added
Human Development and Organizational Studies in Education -- Dissertations, Academic -- UF
Genre: Educational Leadership thesis, Ed.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Improving standards for teacher certification along with improving student achievement in our nation’s public schools has been an emphasis of public school educators for decades.  The National Board for Professional Teaching Standards developed a national teacher certification process that is recognized by many states as meeting the criteria for certification to teach in the state.  Also recognized as one criteria for meeting “Highly Qualified Teacher” requirements under NCLB, teachers who hold a National Board Certification Credential are often recruited to an paid bonuses from school districts and states nationwide.  The state of Florida ceased its bonus payment to National Board Certified teachers in 2010.  The school district in this study continues to pay bonuses to National Board Certified teachers who teach in high poverty schools. The purpose of this study was to examine the federal and state of Florida legislation related to teacher effectiveness and student achievement and the policies in a central Florida urban school district related to compensation to teachers for holding a NBPTS certification credential.  The study sought to answer the research question: To what extent did Nationally Board Certified teachers in high poverty elementary schools in a central Florida urban school district demonstrate positive student achievement gains in mathematics in fourth and fifth grade as measured by FCAT that was greater than the gains demonstrated by non-nationally board certified teachers in the same setting? The study examined a body of research to gain insight into the polices related to National Board Certified Teachers both in the state of Florida and the school district included in this study.  Student scores from FCAT Mathematics in grades four and five were examined and divided into two groups:  Students taught by teachers who were National Board Certified and those who were not.  The means of the two group were compared using a t-test.  Results indicate that there is no clear relationship between National Board Certification and student achievement in Mathematics in the grade levels included in this study.  Further, this study indicates that the district policy for paying teachers who hold the National Board Certification Credential is not cost effective.
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 Anna L Brown.
Thesis: Thesis (Ed.D.)--University of Florida, 2012.
Local: Adviser: Wood, R. Craig.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044416:00001


This item has the following downloads:


Full Text

PAGE 1

1 THE COST EFFECTIVENESS OF A BONUS PAY PLAN FOR NATIONAL BOARD CERTIFIED TEACHERS IN HIGH POVERTY ELEMENTARY SCHOOLS IN AN URBAN SCHOOL DISTRICT IN FLORIDA By ANNA L. BROWN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Anna L. Brown

PAGE 3

3 ACKNOWLEDGMENTS I would like to thank all of the individuals who have supported me t hroughout the process of completing my dissertation. First, I would like to thank my advisor, Dr. R. Craig Wood whose patience and encouragement have been immeasurable. His guidance and support have helped me grow personally and professionally. Next, I would like to thank my committee members Dr. Linda Eldridge, Dr. Bruce Mousa, Dr. Cynthia Griffin, and Dr. Bernard Oliver for their support with my dissertation. Also, I would like to thank Angela Rowe for her assistance throughout coursework and dissert ation work. She has been the glue that kept all things moving forward for me in this program. This dissertation marks the culmination of a doctoral program marked with unique attributes. The Collier Cohort of the LEAD program required a great deal from e ach of the professors who agreed to teach courses and I thank them for committing their time to what was an important opportunity for students. I want to thank my colleagues from Collier County who welcomed me into their cohort. I value the professional and personal relationships developed over the time we spent together. I S heryl Rogers who supported me from day one to the end. A special thank you goes to Allison Stewart Fo ster from Sarasota County who shared the long commutes to Collier County for classes and provided much support both during coursework and the dissertation phase. Most importantly, I would like to thank my family. First, I thank Sheila and our son Eli fo r their continued support and patience with my studies. Sheila is the reason I enrolled in and remained committed to the doctoral program as she has been my

PAGE 4

4 biggest cheerleader. With the birth of our son right at the beginning of the dissertation phase, I know I would not be at completion without her unrelenting support. Next, I would like to thank my parents. Each of them has encouraged me and supported me along the way. They raised me to believe in myself, dream big, and make things happen. Their su pport has helped me to this completion of my degree.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION TO THE STUDY ................................ ................................ ......... 13 Population ................................ ................................ ................................ ............... 18 Statement of Purpose ................................ ................................ ............................. 19 Research Questions ................................ ................................ ............................... 20 Hypothesis ................................ ................................ ................................ .............. 20 Assumptions ................................ ................................ ................................ ........... 20 Limitations ................................ ................................ ................................ ............... 21 Summary ................................ ................................ ................................ ................ 21 2 LITERATURE REVIEW ................................ ................................ .......................... 22 Teacher Quality ................................ ................................ ................................ ...... 22 N BPTS Certification ................................ ................................ ................................ 25 Teachers A re C ommitted to S tudents a nd T heir L earning ................................ 26 Teachers K now t he S ubjects T hey T each and H ow to T each T hose S ubjects to S tudents ................................ ................................ ..................... 27 Teachers A re R esponsible for M anaging and M onitoring S tudent L earning .... 27 Teachers T hink S yst ematically A bout T heir P ractice and L earn F rom E xperience ................................ ................................ ................................ .... 28 Teachers A re M embers of L earning C ommunities ................................ ........... 29 NBPTS and Student A chievement ................................ ................................ .......... 30 Stu dies W ith R esults R elating NBC T eachers and S tudent A chievement ........ 31 Use of V alue A dded M odeling ................................ ................................ .......... 45 Recent Dissertations Linking Student Achievement Results to NBC Teachers ...... 46 Summary ................................ ................................ ................................ ................ 49 3 RESEARCH DESIGN ................................ ................................ ............................. 51 Research Questions ................................ ................................ ............................... 51 Research Design ................................ ................................ ................................ .... 51 Research Population ................................ ................................ ............................... 52 Data Source ................................ ................................ ................................ ............ 55

PAGE 6

6 Data Analysis ................................ ................................ ................................ .......... 56 Summary ................................ ................................ ................................ ................ 56 4 OBSERVATIONS ................................ ................................ ................................ ... 57 Introduction ................................ ................................ ................................ ............. 57 Descriptive Statist ics ................................ ................................ ............................... 59 2008 2009 R esults ................................ ................................ ................................ .. 59 Overall G ain ................................ ................................ ................................ ............ 60 Fifth G rade ................................ ................................ ................................ ....... 60 Fourth G rade ................................ ................................ ................................ .... 60 Combined ................................ ................................ ................................ ......... 60 Improving a L evel ................................ ................................ ................................ ... 61 Fifth G rade ................................ ................................ ................................ ....... 61 Fourth G rade ................................ ................................ ................................ .... 61 Combined ................................ ................................ ................................ ......... 62 Maintaining L evel 3, 4, or 5 ................................ ................................ ..................... 62 Fifth G rade ................................ ................................ ................................ ....... 62 Fourth G rade ................................ ................................ ................................ .... 62 Combined ................................ ................................ ................................ ......... 63 Improving Developmental Scale Score ................................ ................................ ... 63 Fifth G rade ................................ ................................ ................................ ....... 63 Fourth G rade ................................ ................................ ................................ .... 64 Combined ................................ ................................ ................................ ......... 64 Value Added Score ................................ ................................ ................................ 65 Fifth G rade ................................ ................................ ................................ ....... 65 Fourth G rade ................................ ................................ ................................ .... 65 Combined ................................ ................................ ................................ ......... 65 2009 2010 R e sults ................................ ................................ ................................ .. 66 Overall G ain ................................ ................................ ................................ ............ 67 Fifth G rade ................................ ................................ ................................ ....... 67 Fourth G rade ................................ ................................ ................................ .... 67 Combined ................................ ................................ ................................ ......... 67 Improving a L evel ................................ ................................ ................................ ... 68 Fifth G rade ................................ ................................ ................................ ....... 68 Fourth G rade ................................ ................................ ................................ .... 68 Combined ................................ ................................ ................................ ......... 69 Maintaining L evel 3, 4, or 5 ................................ ................................ ..................... 69 Fifth G rade ................................ ................................ ................................ ....... 69 Fourth G rade ................................ ................................ ................................ .... 69 Combined ................................ ................................ ................................ ......... 70 Improving Developmental Scale Score ................................ ................................ ... 70 Fifth G rade ................................ ................................ ................................ ....... 70 Fourth G rade ................................ ................................ ................................ .... 71 Combined ................................ ................................ ................................ ......... 71

PAGE 7

7 Value Added Score ................................ ................................ ................................ 72 Fifth G rade ................................ ................................ ................................ ....... 72 Fourth G rade ................................ ................................ ................................ .... 72 Combined ................................ ................................ ................................ ......... 72 2010 2011 R esults ................................ ................................ ................................ .. 73 Overall G ain ................................ ................................ ................................ ............ 74 Fifth G rade ................................ ................................ ................................ ....... 74 Fourth G rade ................................ ................................ ................................ .... 74 Combined ................................ ................................ ................................ ......... 74 Improving a L evel ................................ ................................ ................................ ... 75 Fifth G rade ................................ ................................ ................................ ....... 75 Fou rth G rade ................................ ................................ ................................ .... 75 Combined ................................ ................................ ................................ ......... 76 Maintaining L evel 3, 4, or 5 ................................ ................................ ..................... 76 Fifth G rade ................................ ................................ ................................ ....... 76 Fourth G rade ................................ ................................ ................................ .... 76 Combined ................................ ................................ ................................ ......... 77 Improving Developmental Scale Score ................................ ................................ ... 77 Fifth G rade ................................ ................................ ................................ ....... 77 Fourth G rade ................................ ................................ ................................ .... 78 Combined ................................ ................................ ................................ ......... 78 Value Added Score ................................ ................................ ................................ 79 Fifth G rade ................................ ................................ ................................ ....... 79 Fourth G rade ................................ ................................ ................................ .... 79 Combined ................................ ................................ ................................ ......... 80 Summary of Results ................................ ................................ ................................ 80 Research Q uestion 1 ................................ ................................ ........................ 80 Gra d e 4 ................................ ................................ ................................ ............ 80 Grade 5 ................................ ................................ ................................ ............ 81 Combined ................................ ................................ ................................ ......... 82 Summary ................................ ................................ ................................ ................ 83 5 CONCLUSIONS ................................ ................................ ................................ ..... 84 Summary of Findings ................................ ................................ .............................. 84 Recommendations for Future P olicy ................................ ................................ ....... 89 Recommendations for Future R esearch ................................ ................................ 89 APPENDIX : CITED CONSTITUTIONS, STATUTES, AND DISTRICT CONTRACT ..... 90 WORKS CITED ................................ ................................ ................................ ............. 91 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 97

PAGE 8

8 LIST OF TABLES Table P age 4 1 Number of National Board Certified Teachers 2008 2009 ................................ .. 59 4 2 Count of students assigned to National Board Certified Teachers 2008 2009 ... 59 4 3 Percent of students making gain 2008 2009 ................................ ...................... 60 4 4 Percent of students improving a level 2008 2009 ................................ ............... 61 4 5 Percent of students maintaining level 3, 4, or 5 ................................ .................. 62 4 6 Percent of students improving developmental scale s core ................................ 63 4 7 Average teacher value added s core ................................ ................................ ... 65 4 8 Number of National Board Certified Teachers 2009 2010 ................................ .. 66 4 9 Count of students a ssigned to National Board Certified Teachers 2009 2010 ... 66 4 10 Percent of students making gain 2009 2010 ................................ ...................... 67 4 11 Percent of st udents improving a level 2009 2010 ................................ ............... 68 4 12 Percent of students maintaining level 3, 4, or 5 ................................ .................. 69 4 13 Percent of students imp rov ing developmental scale s core ................................ 70 4 14 Average teacher value added s core ................................ ................................ ... 72 4 15 Number of National Board Certified Teachers 20 10 2011 ................................ .. 73 4 16 Count of students assigned to National Board Certified Teachers 2010 2011 ... 73 4 17 Percent of students making g ain 2010 2011 ................................ ...................... 74 4 18 Percent of students improving a level 2010 2011 ................................ ............... 75 4 19 Percent of students maintaining level 3, 4, or 5 ................................ .................. 76 4 20 Percent of students improving developmental scale s core ................................ 77 4 21 Average teacher value added s core ................................ ................................ ... 79 4 22 Average v alue added score for grade 4 and 5 teachers combined ..................... 82

PAGE 9

9 LIST OF ABBREVIATION S E LEMENTARY AND S ECONDARY E DUCATION A CT OF 1965 (ESEA) The first comprehensive fe deral education law that provided funding for kindergarten through twelfth grade education. It has been reauthorized by the No Child Left Behind Act of 2001 (NCLB) Blueprint for Reform currently under consideration by US legislators. 1 F LORI DA C OMPREHENSIVE A SSESSMENT T EST (FCAT) A State of Florida specific test measuring elements in the Sunshine State Standards for reading, Math, Science, and Writing administered in grades 3 10. 2 F LORIDA A+ P ROGRAM on proficiency and gains in student achievement using the FCAT assessment program. 3 F REE AND R EDUCED L UNCH P ROGRAM The national school lunch and breakfast program providing meals at a free or reduced cost based on eligibility guidelines set by the Unite d States Department of Agriculture (USDA). Eligibility factors include household size and household income. 4 H IGH P OVERTY S CHOOL For the purpose of this study, schools where the number of students eligible for free and reduced price lunch is at or abov e 90 % of the total student population at that school. 5 N O C HILD L EFT B EHIND A CT OF 2001(NCLB) R eauthorized the Elementary and Secondary Education Act and outlined specific requirements for highly qualified teachers. Also included requirements to measure school performance using student achievement gains. 6 O BAMA S B LUEPRINT FOR R EFORM : T HE R EAUTHORIZATION OF TH E R eleased in 2010 to reauthorize the Elementary and Secondary Education Act. Preserved language and requir ements from NCLB regarding highly qualified 1 20 U.S.C. §§ 6301 et seq. (1965); 20 U.S.C. §§ 6301 et seq. (2002); 20 U.S.C. §§ 6301 et seq. (2011) 2 Fl. Stat. § 1008.22 3 Fl. Stat. § 1008.34 4 [online] Retrieved may 5, 2010 from http://www.fns.usda.gov/cnd/Governance/notices/iegs/EligibilityManual.pdf 5 line] Retrieved May 5, 2010 from http://www.sdhc.k12.fl.us/SSFPDiv/Salary % 20Differential % 202010.11A.pdf 6 20 U.S.C. §§ 6301 et seq. (2002)

PAGE 10

10 E LEMENTARY AND S ECONDARY E DUCATION A CT teachers. Required teacher effectiveness to be measured using students achievement gains. 7 N ATIONAL B OARD FOR P ROFESSIONAL T EACHING S TANDARDS (NBPTS) A n independent, non governmental board established to develo p national criteria for teacher certification. Certified teachers using a rigorous protocol. The certification is recognized as meeting the criteria for highly qualified under NCLB. 8 H IGHLY Q UALIFIED T EACHER state licensure or certification, demonstrates knowledge of the subject they teach. 9 N ATIONAL B OARD C ERTIFIED T EACHER (NBCT) A teacher who holds the credential offered through the NBPTS. 10 S TUDENT GAIN A specified increase in the developmental scale sc ore (DSS) on the FCAT mathematics assessment. This level is defined by the Florida Department of Education Technical Assistance Papers for school grades. Gain is defined as increasing an entire level on the assessment (Levels 1 5), by maintaining a prof icient level (Levels 3 5), or by increasing the specified number of DSS points DSS level in the same subject. 11 7 20 U.S.C. §§ 6301 et seq. (20 11) Note: refers also to document titled: U.S. Department of education. http://www2.ed.gov/policy/elsec/leg/blueprint/blueprint.pdf 8 2010 http://www.nbpts.org/about_us/mission_and_history/history 9 h ttp://www2.ed.gov/programs/teacherqual/hqtplans/fl.doc 10 from http://www.nbpts.org/a bout_us/mission_and_history/history 11 Ibid.

PAGE 11

11 Abstract of Dissertation Presented to the Graduate School of the University of Florid a in Partial Fulfillment of the Requirements for the Degree of Doctor of Education THE COST EFFECTIVENESS OF A BONUS PAY PLAN FOR NATIONAL BOARD CERTIFIED TEACHERS IN HIGH POVERTY ELEMENTARY SCHOOLS IN AN URBAN SCHOOL DISTRICT IN FLORIDA B y Anna L. Brown August 2012 Chair: R. Craig Wood Major: Educational Leadership Improving standards for teacher certification along with improving student educators for deca des. The National Board for Professional Teaching Standards (NBPTS) developed a national teacher certification process that is recognized by many states as meeting the criteria for certification to teach in the state. Also recognized as one criterion for who hold a National Board Certification Credential are often recruited to an d paid bonuses from school districts and states nationwide. The state of Florida ceased its bonus payment to National Board Certified teachers in 2010. The school district in this study continues to pay bonuses to National Board Certified teachers who teach in high poverty schools. The purpose of this study was to examine the federal and state of Florida legisla tion related to teacher effectiveness and student achievement and the policies in a central Florida urban school district related to compensation to teachers for holding a NBPTS certification credential. The study sought to answer the research question: T o

PAGE 12

12 what extent did Nationally Board Certified teachers in high poverty elementary schools in a central Florida urban school district demonstrate positive student achievement gains in mathematics in fourth and fifth grade as measured by FCAT that was greater than the gains demonstrated by teachers without this certification in the same setting? The study examined a body of research to gain insight s into the polices related to National Board Certified Teachers both in the state of Florida and the school distri ct included in this study. Student scores from FCAT Mathematics in grades four and five were examined and divided into two groups: Students taught by teachers who were National Board Certified and those who were not. The means of the two group s were com pared using a t test. Results indicate d that there wa s no clear relationship between National Board Certification and student achievement in Mathematics in the grade levels included in this study. Further, this study indicates that the district policy fo r paying teachers who hold the National Board Certification Credential is not cost effective.

PAGE 13

13 CHAPTER 1 INTRODUCTION TO THE STUDY The quality of the American system of Public education has been a priority among Americans for the past several decades. T he Elementary and Secondary Education Act (ESEA) of 1965 1 began the age of reform and the desire to achieve continuity in educational opportunity throughout the United States. The ESEA act was amended and reauthoriz ed by the No Child Left Behind A ct (NCLB ) of 2001 2 which was signed into law by President Bush on January 8, 2002. This act outlined for the first time the requirement for highly qualified teachers to be hired in order to teach students in high needs schools. In Title II of the NCLB legislatio n, 3 language focuses on providing every child with a highly qualified teacher. Following a similar time frame, the National Board for Professional Teaching Standards (NBPTS) originated in 1987. 4 This board set about to determine the necessary qualificati ons for teachers at a national level. The certification process for National Board Certified Tea chers began in 1993 and by 2010, the number of teachers who had achieved NBPTS certification was 91,013. 5 Because the rigorous certification process incorpora ted standards that exceeded most state certification requirements, this specific certification was deemed one of the ways that a teacher could document highly qualified status as outlined in 1 20 U.S.C. §§ 6301 et seq. (1965) 2 20 U.S.C. §§ 6301 et seq. (2002) 3 Ibid. 4 http://www.nbpts.org/about_us/mission_and_history/history 5 National Board for Professional Teaching Standa Retrieved May 5, 2010 from http://www.nbpts.org/userfiles/File/Profiles/FL_State_Profile_2010_certday.pdf

PAGE 14

14 NCLB. 6 on of the Elementary and Secondary Education Act 7 was released in March 2010 and preserves the language and requirements for highly qualified teachers. Again, the focus of this reform was to ensure that there was a n effective teacher in every classroom. President entities to recruit, develop, retain and reward outstanding teachers. In the state of Florida, legislation was passed in 1998 8 that recognized the essential role that teachers renamed the Dale Hickman Excellent Teaching Program Act in 2002. 9 This program outlined the payment of bonuses for teachers who held the NBPTS Natio nal Board Certification (NBC) credential. This annual bonus was required to be equal to 10 % of years for initial certification. 10 An additional 10 % bonus was availabl e for National Board Certified (NBC) teachers who provide twelve days of mentoring to public school teachers who do not hold the NBPTS certification. 11 In 2010, the number of teachers holding the NBPTS credential in the State of Florida was 13,532. 12 The number of 6 20 U.S.C. §§ 6301 et seq. (2002) 7 n of the http://www2.ed.gov/policy/elsec/leg/blueprint/blueprint.pdf find the legal 8 Fla. Stat. § 1012 .72 (2002) 9 Id 10 Id 11 Fla. Stat. § 1012.72 (2002) 12 from ht tp://www.nbpts.org/userfiles/File/Profiles/FL_State_Profile_2010_certday.pdf

PAGE 15

15 NBPTS credentialed teachers in the selected school district in this study was 880 teachers. 13 The school district in this study had focused on maintaining an equal distribution of highly qualified teachers in all schools. In an effort to recruit and retain national Board Certified Teachers to high poverty schools, the district has an additional bonus payment for NBPTS credentialed teachers who taught in schools with greater than 90 % of students receiving free and reduced lunch. For these teacher s, the district paid an additional bonus equal to the state bonus for holding the credential. 14 Another component of the NCLB legislation was the use of student growth to 15 Adequate Yearly Progress targets were required by th e legislation 16 and each state developed its own definition of these targets. In the state of Florida, these components were outlined in 2002. 17 In order to measure these targets, the state developed the Florida Comprehensive Achievement Test or FCAT 18 as a yearly measure of student achievement in grades 3 10. In addition, the state developed the A+ plan 19 which rated schools with grades of A F based on several student growth criteria. Student proficiency and gain are measured using the developmental scale scores from these yearly assessments in reading, mathematics, 13 Ibid. 14 Differentiated Pay: Contract Section 21.4.8. Hillsborough Classroom Teachers Association, Inc. Teacher Contract 2010 2013. Hillsborough County Public Schools, Hillsborou gh, FL. 2010. [online] retrieved May 5, 2010 from http://www.hillscta.org/Welcome_files/teachercontract.pdf 15 20 U.S.C. §§ 6301 et seq. (2002) 16 Id 17 Fla. Stat. § 1008.34 (2002) 18 F la. Stat. §1008.22 (2002) 19 Fla. Stat. § 1008.34 (2002)

PAGE 16

16 and science. 20 FCAT Writing is also utilized, but a different scoring mechanism was used. Each year, a simple growth model was used to calculate the gain in student achievement by comparing the developmental scale score from the prior year to the current year score. Calculations were used to rate performance at the school level but not at the teacher level. 21 The school distric t in this study defined teacher effectiveness by student achievement gain s when it began bonus payments in the Merit Award Program (MAP) in 20 07 22 These bonuses were paid to all eligible teachers based on 23 Even though the MAP program requi red demonstrated student achievement gains before a bonus payment was made, the bonuses paid for NBPTS credential were different and in addition to the MAP bonuses. The payment of additional bonuses for National Board Certified teachers, in this school di strict, who teach in high poverty schools, was not linked to any student achievement measure as the bonus was paid based on holding the credential alone. 24 Therefore, in this district, a teacher holding the NBPTS credential had the ability to earn several bonuses above the base salary. These options included: 1) State bonus for NBPTS credential (10 % of the average state teacher salary); 25 2) State bonus for mentoring a non NBC teacher (10 % of the average 20 Id 21 Id 22 Fla. Stat. § 1 0 12.225 et. seq. (2007) 23 Hillsborough County Public Schools (2010). 24 Differen tiated Pay: Contract Section 21.4.8. Hillsborough Classroom Teachers Association, Inc. Teacher Contract 2010 2013. Hillsborough County Public Schools, Hillsborough, FL. 2010. [online] retrieved May 5, 2010 from http://www.hillscta.org/Welcome_files/teachercontract.pdf 25 Fla. Stat. § 1012.72 (2002)

PAGE 17

17 state teacher salary); 26 3) District bonus for NBPTS c redential and teaching in a high poverty school ($4500.00); 27 4) District bonus for teaching in a high poverty school (automatic 5 % solely for teaching in the high poverty school); 28 5) District bonus for demonstrated school gains f or teachers teaching a high poverty school (5 % 29 6) District bonus for student gains under the Merit Award Program (5 % of the average district teacher salary); 30 7) Federal TIF bonus for demonstrated student gain in an eligible scho ol (5 % of the average district salary); 31 and 8) District bonus if recommended by supervising principal for performance pay (5 % of average district teacher salary). 32 If all bonuses were earned, the average sum of these bonuses for a teacher in this district was $23,000.00 in 2010. In this list of bonuses, four require demonstrated student achievement gains while four are credential or placement based. With approximately $14,500.00 (from credential based bonuses) per teacher at stake, the question of cost e ffectiveness was asked. Was there a return on investment for these credential and placement based bonuses? The NBPTS commissioned studies of the effectiveness of teacher s who hold the NBPTS 26 Fla. Stat. § 1012.72 (2002) 27 Differentiated Pay: Contract Section 21.4.8. Hillsborough Classroom Teachers Association, Inc. Teacher Con tract 2010 2013. Hillsborough County Public Schools, Hillsborough, FL. 2010. [online] retrieved May 5, 2010 from http://www.hillscta.org/Welcome_files/teachercontract.pdf 28 ibid 29 i bid 30 Differentiated Pay: Contract Section 21.4.9. Hillsborough Classroom Teachers Association, Inc. Teacher Contract 2010 2013. Hillsborough County Public Schools, Hillsborough, FL. 2010. [online] retrieved May 5, 2010 from http://www.hillscta.org/Welcome_files/teachercontract.pdf 31 ibid 32 2013. Hillsborough County Public Schools, Hil lsborough, FL. 2010. [online] retrieved May 5, 2010 from http://www.hillscta.org/Welcome_files/teachercontract.pdf

PAGE 18

18 credential. 33 These studies and others conducted independently a re split on the effectiveness of National Board Certified teachers as measured by gains in student achievement. Some studies show that student achievement is higher among teachers who hold the NBPTS credential. 34 Others maintain that there is no significa nt difference in student achievement between teachers who hold the credential and those that do not hold the credential 35 In 2011, the NBPTS released a report f rom the NBPTS Student L earning, Student Achievement Task Force which suggests that measuring te acher effectiveness requires the incorporation of both observed professional practice and gains in student achievement. 36 Population The population examined for this study is a group of teachers of fourth and fifth grade students in high p overty elementary schools in a central Florida urban school district. These specific grade levels were chosen because scores for these two grade levels are readily available in the public database. Fourth and fifth grade FCAT scores are the only elementary grade levels fo r which a gain in the developmental scale score is determined and reported. No other student achievement data was available for any other grade level that was reported on a state scale and with gain calculated. The subject of mathematics was chosen becau se in this school district the curriculum focus in mathematics in grades four and five had been stable over a period of six years. 33 Retrieved May 5, 2010 from http://www.nbpts.org/resources/research/impact_of_certification#bottom 34 ibid 35 ibid 36 National Board f http://www.nbpts.org/abo ut_us/events_and_photos/2011_student_learning_st

PAGE 19

19 Reading scores were not considered for use because the school district experienced a change in reading curriculum during 201 0 2011 when a new textbook series was adopted. This district had 194,737 enrolled students. Student demographics for the enrolled student population were: 3.37 % Asian, 21.65 % Black, 29.45 % Hispanic, .29 % Indian, 4.98 % Multiracial, and 40.26 % White. 37 Sp ecifically included are students and teachers assigned to schools that meet the definition of a high poverty school (at or above 90 % of the student population receiving free and reduced lunch). In this district the number of schools that met this criterio n was thirty nine. Of these, twenty eight were e lementary schools 38 Included in the study were forty National Board Certified fourth grade teachers and 702 four th grade teachers who were not National Board C ertified. Also included were forty National Bo ard Certified fifth grade teachers and 654 fif th grade teachers who were not National Board C ertified. Statement of Purpose The purpose of this study was to examine: The federal and state of Florida legislation related to teacher effectiveness and student achievement. The policies in the state of Florida related to teacher effectiveness and student achievement. The policies in a central Florida urban school district related to teacher effectiveness and student achievement. The policies in the state of Flor ida related to compensation to teachers for holding a NBPTS certification credential. 37 http://publicaffairs.mysdhc.org/files2010 11/FACT SBrochureFINAL.pdf 38 http://www.sdhc.k12.fl.us/SSFPDiv/Salary % 20Differenti al % 202010.11A.pdf

PAGE 20

20 The policies in a central Florida urban school district related to compensation to teachers for holding a NBPTS certification credential. The extent to which students of National Board Certified teachers in fourth and fifth grade differ in achievement on the math portion of the FCAT, from students of non National Board Certified teachers. Research Questions This study will answer the following research questions: To what extent did National Board Certified teachers in high poverty elementary schools in a central Florida urban school district demonstrate positive student achievement gains in mathematics in fourth and fifth grade as measured by FCAT that was greater than the gain s demonstrated by non National Board C ertified teachers in the same setting? To what extent were the state and district level polices relat ed to compensation of National Board Certified Teachers cost effective? Hypothesis National Board Certification had a positive impact on the achievement results of fourth and fifth grade students on the mathematics portion of the FCAT in high poverty elementary schools in this particular school district. Assumptions The following assumptions are made regarding this study: Student level data were collected and measured without error Student teacher linkages relating student achievement information to teacher assignment followed the state A+ plan definitions for eligible students. Students were assigned to their classr ooms in a normal manner with no pre determined conditions This study examined intact groups of students that wer e normally assigned to teachers by school principals. Student characteristics in similar settings were relatively similar to each other.

PAGE 21

21 Limitat ions Results from this analysis may lack generalizability outside of the state of Florida or the selected school district because the test data used was limited to the Florida Comprehensive Achievement Test, mathematics portion. Results from this analysis may lack generalizability beyond the population of teachers observed (e lementary mathematics teachers in fourth and fifth grade teaching students in high poverty elementary schools). Results of this analysis may be limited because they rely on the use of a simple growth model for measuring student achievement gains. Individual student variables may impact results beyond the scope of what can be controlled for in this analysis. The researcher for this study holds a National Board Certification credential in Middle Grades. Though efforts for objectivity are in place, there was potential for bias on the part of the researcher. Summary This study examined national, state, and district policies regarding compensation for National Board Certified Teachers. Rese arch regarding the student achievement gains for National Board Certified teachers was reviewed and summarized. Student achievement gains in mathematics were analyzed for fourth and fifth grade teachers in high poverty elementary schools in one central Fl orida, urban school district. Teachers were grouped into two categories: those holding the NBPTS credential and those without the NBPTS credential. The gains of these groups were compared and the results reported.

PAGE 22

22 CHAPTER 2 LITERATURE REVIEW Teacher Quality The quality of teachers in American schools is an increasingly prevalent topic and focus of educational reform. Research studies have focused on the effect of having a quality teacher. In 1992, Hanushek found that with all else held equal, studen ts in a those in a classroom with a low quality teacher had an average learning gain of 0.5 years. 1 The National School Boards Association published a report in 2006 titled Why Teachers Matter. 2 This report stated that teacher quality had more of an influence on student performance than any other factor including race, class, or school and that disadvantaged students benefit more from good teachers than advantaged students. 3 Several studies reported similar findings and clearly stated that the quality of the teacher was the strongest predictor of student achievement outranking all other variables in the classroom. 4 Still other studies indicate that the single consistently 1 Journal of Political Economy 1 (100). February 1992. [online] Retrieved May 5, 2010 from http://hanushek.stanford.edu/sites/default/files/publications/Hanushek % 201992 % 20JPE % 20100 % 281 % 2 9.pdf 2 American School Boards Journal, 193 (3). 2006. 3 Ibid 4 See these studies: Darling Educational Researcher, 31(9). 2002 [online] Retrieved May 5, 2010 from http://edr.sagepub.com.lp.hscl.ufl.edu/content/31/9/13.full.pdf+html Equity and Opportunity 60(4) 2003. [online] Retrieved M ay 5, 2010 from http://www.edpal.com/courses/upload/ELL_002/Toward % 20A % 20Fair.pdf Rivkin, S., Hanushek, E., & K Achievement Econometrica 73(2). 2005. [online ] Retrieved May 5, 2010 from http://www.economics.harvard.edu/faculty/staiger/files/HanushekRivkinKain % 2BEcta % 2B2005.pdf

PAGE 23

23 influential factor that affected student outcomes was the classroom teacher. 5 Each study reported that the teacher made the difference. The differentiating factor, however, was the definition and measure of quality Defining quality in research was probl ematic as each researcher measured interrelated variables and measured student achievement in statistically different ways. Defining quality at a state and national level was problematic as there was great variation among states regarding qualifications f or earning the certification to teach in public schools, as well as the standards used to measure performance of students and student learning. Coupled with the financial requirements for funding public schools, reform has focused on measuring the effecti veness of each teacher relative to some common standard. In 1983, the U.S. Department of Education released a report titled, A Nation at Risk 6 This report sparked the transition of teacher quality from an isolated educational issue into a public policy issue. The report raised public awareness that the country. 7 The Carnegie Foundation became involved, creating its own report titled, A 5 See these studies: Clotf Economics of Education Review 26, (2007): 673 682. [online] Retrieved May 5, 2010 from http://www.nber.org/papers/w13617.pdf?new_window=1 Stronge, J., Ward, T., Tucker, P., Hindman, J., McColsk e Teachers and non National Board Certified Teachers: Is there a d ifference in teacher effectiveness and student achievement Personnel Evaluation in Education 20, 2007: 185 210. [online] Retrieved May 5, 2010 from http://strongeandassociates.com/files/Stronge % 20et % 20al % 202007_National % 20Board % 20Certified % 20 Teachers.pdf 6 National Commission on Excellence in Education. A Nation at Risk: The Imperatives for Education Reform. Washington D.C. : U.S. Government Printing Office, 1983. [online] Retrieved May 5, 2010 from http://www2.ed.gov/pubs/NatAtRisk/index.html 7 Ibid.

PAGE 24

24 N ation Prepared: T eachers for th e 21 st Century 8 This report called for creating a set of national standards for teachers which became known as the National Board for Professional Teaching Standards (NBPTS). In 1987, the NBPTS joined forces with both the American Federation of Teachers and the National Education Association to develop a national teacher certification program. 9 In 1989, the NBPTS published a policy statement titled, What Teachers Should Know and Be Able to Do. 10 This publication became the foundation for the process and s tandards for a performance based assessment of the quality of teaching. The NBPTS model was designed to provide a rigorous assessment of teaching teaching. 11 Outlined in t his document are the five core propositions of the NBPTS. These propositions include: 1) teachers are committed to students and their learning; 2) teachers know the subjects they teach and how to teach those subjects to students; 3) teachers are responsib le for managing and monitoring student learning; 4) teachers think systematically about their practice and learn from experience; and 5) teachers are 8 Carnegie T ask Force on Teaching as a Profession. New York: Carnegie Forum on Education and the Economy, 1986. 9 from http://www.nbpts.org/about_us/mission_and_history/history 10 http://www.nbpts.org/UserFiles/File/what_teachers.pdf 11 Recruiting and retaining National Board Certified Teachers for hard to staff low performing schools Southwest Center for T eaching Quality. Chapel Hill, North Carolina.2005. [online] Retrieved May 5, 2010 from http://www.teachingquality.org/legacy/RecruitRetainHTSS.pdf Sato, M., Chung, W., & Darling American Educational Research Journal vol. 45(3), 2008: 669 700. [online] Retrieved M ay 5, 2010 from http://www.stanford.edu/~rchung/AERJ316955 % 20NBPTS % 20Study % 204.15.08.pdf

PAGE 25

25 members of learning communities. 12 These propositions were the foundation for the NBPTS certification pro credential. Since 1993 the number of national board certified teachers had grown from the initial 93 to over 91,000 teachers earning the credential by 2010. 13 During this time period, the reautho rization of ESEA resulted in the new legislation known as No Child Left Behind (NCLB) 14 NCLB incorporated requirements for measuring student learning islation was supported by a large body of research linking the effectiveness of the teacher to student outcomes. 15 National Board Certified Teachers are recognized by the U.S. Department of Education as meeting the requirements for 16 N BPTS Certification The National Board for Professional Teaching Standards (NBPTS) set as its underlying premise five core propositions, or standards, that outline the core of a quality 12 Retrieved May 5, 2010 http://www.nbpts.org/the_standards/the_five_core_propositio 13 Retrieved May 5, 2010 from http://www.nbpts.org/resources/nbct_directory 14 20 U.S.C. §§ 6301 et seq. (2002) 15 See these studies: Darling Does Scientifically Based Educational Researcher, 31(9). 2002 [online] Retrieved May 5, 2010 from http://edr.sagepub.com.lp.hscl.ufl.edu/content/31/9/13.full.pdf+html Goldhaber, Dan Urban Institute (2004). [online] Retrieved May 5, 2010 from: http://www.urban.org/publications/410958.html rade Journal of Political Economy 1 (100). February 1992. [online] Retrieved May 5, 2010 from http ://hanushek.stanford.edu/sites/default/files/publications/Hanushek % 201992 % 20JPE % 20100 % 281 % 2 9.pdf 16 20 U.S.C. §§ 6301 et seq. (2002)

PAGE 26

26 of teacher quality. The Certification process is designed to measure instructional practice as the NBPTS sought to identify teachers who demonstrated a high level of knowledge, skills, and commitment to the five core propositions. A discussion of each o f the five core propositions follows. Teachers A re C ommitted to S tudents and T heir L earning 17 This proposition reflects the core belief that all students can learn. The NBPTS believes that accomplished teachers are those that reflect this belief throughout their classroom practices. They are skilled at making knowledge accessible to all students and they treat students equitably. Components to this standard include the belief that accomplished teachers: recognize individual differences in their students an d adjust their practice accordingly; have an understanding of how students develop and learn; and have a mission that extends beyond developing the cognitive capacity of their students. 18 The NBPTS also asserts that accomplished teachers are concerned with are concerned with the development of character and civic responsibility. 19 17 5, 2010 http://www.nbpts.org/the_standards/the_five_core_propositio 18 National Board for Professional Teachi n 10 from http://www.nbpts.org/UserFiles/File/what_teachers.pdf 19 5, 2010 http://www.nbpts.org/the_standards/the_five_core_propositio

PAGE 27

27 Teachers K now the S ubjects T hey T each and H ow to T each T hose S ubjects to S tudents 20 The NBPT S outlines in proposition two that accomplished teachers have mastery of the subjects they teach and understand applications of the subject. 21 They appreciate how knowledge is created and linked to other disciplines. Accomplished teachers are able to take command of their subject both in the substance of the content knowledge of these underlying structures is critical to the teacher s ability to teach students to think analy tically. Accomplished teachers have a specialized knowledge of how to convey a subject to students. Knowledge of how to reveal content to students is a critical skill that requires an instructional toolkit filled with instructional techniques, curricular resources, and highly refined decision making skills in order to account for the variety of students and contexts that the teacher faces. Accomplished teachers generate multiple paths to knowledge by demonstrating the value of both structured and inductive learning. 22 Teachers A re R esponsible for M anaging and M onitoring S tudent L earning 23 In this proposition, the NBPTS presents its belief that accomplished teachers hold high expectations for all students and strive to facilitate learning. They make the mos t effective use of time and find multiple ways to capture and sustain the interest of their students. They call on multiple methods to meet their goals by using a variety of 20 Ibid. 21 Ibid. 22 rieved May 5, 2010 from http://www.nbpts.org/UserFiles/File/what_teachers.pdf 23 5, 2010 http://www.nbpts.org/the_standards/the_five_core_propositio

PAGE 28

28 instructional skills to create a varied learning environment engaging each type of student. This engagement is a priority for accomplished teachers as they realize it is their responsibility to motivate and build on students attention in order to generate new enthusiasm for learning. Accomplished teachers also regularly assess studen t progress. They monitor successes and failures in order to understand both the impact of their strategies and the development of understanding within the learner. Accomplished teachers can track what students are learning and modify practice based upon their judgments about their own teaching practices. This reflection is paramount to the success of an accomplished teacher. It is this professional learning that refines the instructional practices of the accomplished teacher. 24 Teachers T hink S ystemati cally A bout T heir P ractice and L earn F rom E xperience 25 Accomplished teachers have a professional obligation to be lifelong students of their own craft. They are inventive in their teaching and continually make choices to test their judgment. 26 They are wil ling to try new things and stay abreast of current issues in education. Accomplished teachers make decisions about the breadth and scope of the content they present. They must reconcile the demand of covering a topic versus the demands for in depth under standing and they understand that this decision differs with the developmental level of the students they teach. They critically examine their practice on a regular basis and incorporate new findings into their practice. 27 24 Ibid. 25 ine] Retrieved May 5, 2010 from http://www.nbpts.org/UserFiles/File/what_teachers.pdf 26 Do http://www.nbpts.org/UserFiles/File/what_teachers.pdf 27 trieved May 5, 2010 http://www.nbpts.org/the_standards/the_five_core_propositio

PAGE 29

29 Accomplished teachers value the t ime they spend reflecting on their practice. They understand that they must push their beliefs about a particular practice and its relative impact on student learning. They incorporate their own observations and classroom data as well as they findings of educational research for this purpose. Teachers A re M embers of L earning C ommunities 28 The NBPTS asserts that accomplished teachers are those that understand that their work extends beyond the walls of their classroom. They seek to develop professional r elationships with other educators in order to expand their knowledge by capitalizing on the expertise that resides within their school. They collaborate with other professionals in order to learn, grow, and develop as a teacher as well as to influence pol icy decisions within their school. They share the responsibility for developing what is quality learning for students and are proactive in leading discussions about policy implementation rather than blindly accepting curricular conventions. 29 Accomplished teachers also work to develop relationships with parents and the community. They work collaboratively with parents in order to engage them in the work of the school. 30 They work to build relationships that empower parents to be mutual partners in their s education. They build community partnerships and take advantage of community resources as a valuable resource for learning. Accomplished teachers have a great 28 Ibid. 29 Nation le to ] Retrieved May 5, 2010 from http://www.nbpts.org/UserFiles/File/what_teachers.pdf 30 line]. Retrieved May 5, 2010 http://www.nbpts.org/the_standards/the_five_core_propositio

PAGE 30

30 understanding of the culture of their community including an appreciation of the ethnic and linguistic differences and the effects of poverty and affluence. 31 The National Board Certification process places the certification of teachers on a national playing field and outlines a rigorous approach to demonstrating the ability to interpret an d reflect on student learning in the classroom. Measuring instructional practices and requiring teachers to demonstrate in their practice their internalization of the five core propositions leads to a credential with much merit. Though student learning i s a component of the certification process, much of the assessment focuses on the ability of the teacher to interpret student work samples for mastery of the intended objectives. The link, however, between the ability to obtain the credential and the demo nstration of student achievement gains is not clearly documented by educational research. NBPTS and Student Achievement There is consistent and ongoing debate over the statistics linking National Board Certified teachers to student achievement gains. Much of this debate is centered on studies where relatively small yet statistically significant achievement gains are reported for the group of teachers with National Board Certification. 32 Results of each study need to be carefully analyzed for sample size, ef fect size, statistical significance, and practical 31 ble to http://www.nbpts.org/UserFiles/File/what_teachers.pdf 32 Note: The authors make this claim in their review of research that compared severa l studies relating Value Added Assessment of Teacher Quality as an Alternative to the National Board for Professional Teaching standards: What Recent Studies Theory and Applications. Maple Grove, MN: Jam Press, 2005. [online] Retrieved May 5, 2010 from http://www.education cons umers.com/Cunningham Stone.pdf

PAGE 31

31 significance. There is also new interest in analyzing th e results of value added analyse s and the relative position of overall teacher effect correlated with the NBCT credential. 33 Studies W ith R esults R elating NBC T eachers and S tudent A chievement Stone conducted a study in 2002 that compared student achievement results from Tennessee. Stone included the sixteen NBC teachers teaching grades three through eight at the time of his study. He utilized the T ennessee Value Added Assessment System (TVAAS) results as calculated for the state of Tennessee by Sanders. The TVAAS calculation uses a mixed model that measures student gain s over time. This model does not include any covariates for unique student chara cteristics and assumes that teachers have control over student learning gains. Teacher effect scores were reported on a scale of 0 50. Stone designed his study to investigate whether Tennessee ievement 34 Exceptional was defined as improving student achievement equal to 115 % of 35 This definition was borrowed f ro m the state of Tennessee as it defined exemplary as equal to or exceeding 115 % In Chattanooga, Ten nessee, teachers making this gain in three core subjects were rewarded with a state bonus of $5000.00. 36 identified would qualify for the bonus mentioned above as n one of the NBC teachers met the required gain threshold in three core subjects. A nalyzing the effect scores for 33 Ibid. 34 Added Achievement Gains of NBPTS Certified Teachers in Tennessee: A Brief [online] Retrieved M ay 5, 2010 from http://www.educationconsumers.com/oldsite/briefs/stoneNBPTS.shtm 35 Ibid. 36 Ibid.

PAGE 32

32 the sixteen teachers and incorporating the individual effect score s for each subject for each of the thr ee years of data analyzed yielded 123 records for analysis. Of the se, only eighteen (15 % ) reach ed the definition of exemplary. Of note, thirteen (11 % ) fe ll 37 Stone concluded that the achievement gains of students with NBC teachers we re no greater than those made by stu dents who had other teachers without this certification 38 In 2004, Cavaluzzo conducted a study of student achievement in Miami Dade Public Schools, Miami, Florida. She examined 108,000 student records for ninth and tenth grade student achievement in mathe matics. 39 Student gains were analyzed across a three year period using scores from the Florida Comprehensive Achievement Test (FCAT) in Mathematics. The study incorporated several model designs which included controls for varying student and teacher charac teristics. By controlling for each variable separately in models, the study shows that after controlling for all other variables, students assigned to a NBC teacher are likely to gain 12 % of a standard deviation more than others. 40 After taking into accou nt the school effect, students assigned to a NBC teacher are likely to gain 7 % to 8 % of a standard deviation m ore than others. 41 When examining student characteristics, the study results show ed that NBC teachers had greater eff ect on the achievement of bla ck and H ispanic students than other students 37 Ibid. 38 Ibid. 39 Corporation (2004). [online]. Retrieved Ma y 5, 2010 from: http://www.cna.org/documents/CavaluzzoStudy.pdf 40 Ibid. 41 Ibid.

PAGE 33

33 with effect sizes of 0.14 and 0.15 respectively. Similarly, retained students were at an advantage when assigned to NBC teachers with an effect size of 0.16. 42 Interestingly, however, the effect on students who were eligible for free or reduced lunch was small and statistically insignificant. 43 Because correlations between race and free and reduced price lunch eligibility were very strong, the researcher ran a combined model in order to see if the effect on free and reduced lunch would surface. The NBC teachers were not found to add any benefit to students who were eligible for free and reduced price lunch. 44 made larger gains in mathematics, on average, than did their counterparts without such 45 Also in 2004, Vandervoot, Amrein Beardsley, and Berliner conducted a study of National Board Certified Teachers and the relationship to student achievement. This study included 208,650 stu dents in grades three through six in Arizona. The researchers collected data from the entire state fi le and included only the twenty four schools in fourteen districts with NBC teachers who agreed to participate in the study and had qualifying da ta. A to tal of thirty five NBC teachers were included in the analysis. The study included results from the Stanford Achievement Test (SAT 9) in reading, mathematics and language across four years from 1999 to 2003. The 42 Ibid. 43 Ibid. 44 Ibid. 45 Corporation (2004). [online]. Retrieved May 5, 2010 from: http://www.cna.org/documents/CavaluzzoStudy.pdf

PAGE 34

34 46 Results from this study were reported for each year and each subject along w ith results across the four years and across the three subjects. The overall average effect size over the four year period for having a NBC teacher was .122. According to the authors, this was equivalent to slightly more than one month advantage for the students assigned to NBC teachers. 47 The scaled score differences for each subject were also reported. In Reading, students assigned to NBC teachers had an average increase of 2.39 scale score points. In Mathematics, the increase was 3.11 and in language the increase was 1.86 scale score points. 48 Vandervoot and her fellow researchers assert that their study showed NBC teachers to be superior to non NBC teachers and that NBC teachers gained more in student achievement over the course of a year than did n on NBC teachers. 49 Goldhaber and Anthony published their study in 2004. 50 This study focused on student achievement gain in mathematics and reading for elementary students in grades three, four and five in the state of North Carolina. The researchers calcu lated value added models with controls for free and reduced lunch, race, gender, ethnicity, English proficiency status, and learning disability status of students in the study. They 46 Vandevoort, L., Amrein d Teachers and Their Education Policy Analysis Archives, 12 (46). (2004) [online] Retrieved May 5, 2010, from http://epaa.asu.edu/epaa/v12n46/v12n46.pdf 47 Ibid. 48 Vandevoort, L ., Amrein Education Policy Analysis Archives, 12 (46).(2004) [online] Retrieved May 5, 2010, from http: //epaa.asu.edu/epaa/v12n46/v12n46.pdf 49 Ibid. 50 Urban Institute (2004). [online] Retrieved May 5, 2010 from: http:/ /www.urban.org/publications/410958.html

PAGE 35

35 studied three years of student achievement data; 1996 1999. 51 Student gai n scores were calculated by subtracting the previous end of grade test from the subsequent year score. 52 The researchers chose to focus on elementary grade students because this group of students tended to have only one teacher per grade level and the link between the student gain and the teacher was more direct. The study included approximately 390,450 students and 6,000 teachers. A total of 303 NBC teachers were included in the sample for the three year period. Results show that NBC teachers produced st udent gains greater than non NBC teachers by 4% of a standard deviation in reading and 5% of a standard deviation in math. 53 The differences in student scores showed that students with NBC teachers scored 0.49 of a point higher in reading and 0.46 of a poi nt higher in math. 54 All other coefficients were reported in groups of teachers who earned their NBCT credential rather than directly comparing to non NBC teachers. obtain ing their licensure and that obtaining the certification does not seem to add human capital value to the professional. 55 In 2005, Sanders, Ashton, and Wright completed an analysis of the effect of National Board Certified teachers on student achievement. This study was funded by a 51 Ibid. 52 Ibid. 53 Urban Institute (2004). [online] Retrieved May 5, 2010 from: http://www.urban.org/publications/410958.html 54 Ibid. 55 Ibid.

PAGE 36

36 grant from the USDOE and partially funded by the NBPTS. 56 This study compared the results from end of grade mathematics and reading tests from North Carolina students in grades four through eight. The study focused on two larg e school districts in North Carolina: Charlotte Mecklenberg and Wake County incorporating results from 2000 2003. The study sample included approximately 260,000 student records linked to over 4,600 teachers. Of this group, 281 teachers were NBC teachers of mathematics and 306 were NBC teachers of reading. 57 Sanders calculated student gain scores and prior gain scores as a covariate. He included four models. Two of these models used simple gain and fixed teachers effects. The purpose for including thes e models was to compare results to previous research. Sanders Model 1 is similar to the model used by Cavaluzzo in 2004. Sanders Model 3 is similar to the model used by Goldhaber and Anthony in 2004. 58 Sanders Model 2 and 4 used prior scores as a covaria te and included random teacher ef fects. Results from his analyse s showed that NBC teachers did not achieve significantly better academic gain s in their students than other teachers without this certification 59 In M odels 1 and 3, NBC teachers tended to ha ve better average performance both by large r end of year scores (M odel 1 ) and larger gain 56 Comparison of the Effects of NBPTS Certified Teachers with Other Teachers on the Rate of Student Academic Progress National Board for Prof essional Teaching Standards Arlington, VA. 2005. [online] Retrieved May 5, 2010 from http://www.nbpts.org/UserFiles/File/SAS_final_NBPTS_report_D_ _Sanders.pdf 57 I bid. 58 Comparison of the Effects of NBPTS Certified Teachers with Other Teachers on the Rate of Student Academic Progress National Board for Professional Teaching Standards Arlington, VA. 2005. [online] Ret rieved May 5, 2010 from http://www.nbpts.org/UserFiles/File/SAS_final_NBPTS_report_D_ _Sanders.pdf 59 Ibid.

PAGE 37

37 scores (M odel 3) with greater results in mathematics than in reading. 60 The effect sizes reported were 0.09 for math and 0.04 for reading and were similar to the re sults published by Goldhaber and Anthony, and Cavaluzzo. 61 When Sanders calculated the models using random teacher effects, he found fewer significant effects in both math and reading. Model 2 results yielded effect sizes of 0.07 for reading and 0.04 for math. Model 4 results yielded effect sizes of 0.05 for reading and 0.01 for math. Sanders explained that this difference may have been caused by using a hierarchical model with random teacher effects in order to avoid clutter in the data caused by nestin g of students with teachers. He claimed that the studies conducted by Cavaluzzo, Goldhaber and Anthony, and Vandervoot, Amrein Beardsley, and Berliner fail ed to account for this nesting of data as they each used a linear regression model. 62 Sanders assert ed that in order to accurately assert the effect size of the teacher on student achievement, a model incorporating random teacher effects must be used. 63 Sanders discussed the fact that the small effect sizes from the previous studies along with the linear regression used, were reasons to use caution when interpreting the results relating to policy and the effectiveness of NBC teachers. He summarized his findings 60 Ibid. 61 Ibid. 62 Sanders, W. L., Ashton, J. J. & Wright, Comparison of the Effects of NBPTS Certified Teachers with Other Teachers on the Rate of Student Academic Progress National Board for Professional Teaching Standards Arlington, VA. 2005. [online] Retrieved May 5, 2010 from http://www.nbpts.org/UserFiles/File/SAS_final_NBPTS_report_D_ _Sanders.pdf 63 Ibid.

PAGE 38

38 distinguishing effective from ineffective teachers with respect to student achievement results. 64 In 2005, McColskey and Stronge led a team of researchers on a study comparing teacher effectiveness and student achievement of NBC teachers and non NBC teachers. The resear ch team answered a request for proposal from NBPTS and was funded to complete the study. This study examined NBC teachers in three North Carolina districts. Teachers were selected that had longitudinal student achievement data linked to teacher assignmen t available for the research team. A second phase of the study examined the teaching practices of these teachers. For this literature review, only phase one is discussed. Student data examined were the results from the North Carolina End of Grade tests in reading and math for fourth and fifth grade students. The researchers used statistical modeling to make predictions for student achievement and control for some student characteristics. Controls were used for gender, ethnicity, free and reduced lunch status, attendance, school size, and the percent age of students receiving English as a second language services and were applied at the individual, classroom, and school level. 65 Two years of student data were included from approximately 307 teachers. App roximately 4,600 students were included in the analysis with approximately 4,200 taught by non NBC teachers and 417 taught by NBC teachers. 66 A two stage, two level hierarchical linear model was used. Student 64 Ibid. 65 McColsk e y, W., Stronge, J. H., Ward, T. J., Tucker, P. D., Howard, B. & Lewis, K. ef fectiveness, student achievement, and National Board Certified teachers National Board for Professional Teaching Standards Arlington, VA. 2005 [online]. Available from http://www.nbpts.org/UserFiles/File/Teacher_Effectiveness_Student_Achievement_and_National_Board_ Certified_Teachers_D_ _McColskey.pdf 66 Ibid.

PAGE 39

39 residuals were averaged to roll up to the teac her level. 67 Student residuals from both NBC teachers and Non NBC teachers were compared. Results showed no statistically significant difference between the two groups. 68 The mean student residual for students with NBC teachers was higher, but the differe nce between them was so small that there is no statistical difference. The researchers also placed the teacher level averages into quartiles and compared the results across both subjects related to being board certified or non board certified. This analy sis did not consistently place NBC teachers in the top quartiles of all the teachers. 69 The researchers clearly caution against using the results and generalizing to larger populations. 70 They express concern over the small sample size and the limit of app lying results beyond the grade level analyzed (fifth grade). 71 In 2007 Clotfelder, Ladd, and Vigdor published a study of student achievement results with NBC teachers. This was a longitudinal study covering ten years. The study included student data from all students in grades three, four, and five from the North Carolina End of Grade tests in years 1995 2004. 72 The researchers were able to match 75 % of the students with their teacher of math or reading over the ten year period. Grades three, four, and f ive were included because they had the closest student 67 Ibid. 68 Ibid. 69 Ibid. 70 McColsk e y, W., Stronge J. H., Ward, T. J., Tucker, P. D., Howard, B. & Lewis, K. effectiveness, student achievement, and National Board Certified teachers National Board for Professional Teaching Standards Arlington, VA. 2005 [online]. Available from http://www.nbpts.org/UserFiles/File/Teacher_Effectiveness_Student_Achievement_and_National_Board_ Certified_Teachers_D_ McColskey.pdf 71 Ibid. 72 Economics of Education Review 26, (2007): 673 682. [online] Retrieved May 5, 2010 from http://www.nber.org/papers/w13617.pdf?new_window=1

PAGE 40

40 teacher link. 73 Several levels of teacher characteristic were analyzed in this study including years of experience, level of degree, and National Board Certification. The matched sample of math and re ading teachers included 900 NBC teachers with approximately 300 in each of the grades levels included. 74 Positive coefficients were reported for NBC teachers both in the year that they became certified and the years prior to their certification. This find ing led the researchers to conclude that the NBPTS certification process accurately identified effective teachers. The coefficients were reported over a four year period showing two years prior to certification, one year prior to certification, the certif ication year, and one year post certification. Interestingly, the strength of the coefficient declines over time. These findings suggest that the process of becoming certified does not add to the effectiveness of the teacher. Coefficients for math are stronger than coefficients for reading. Clotfelder, Ladd, and Vigdor claimed that this study confirmed the results from the earlier study by Goldhaber and Anthony. 75 In January 2007, Harris and Sass published the results of their study identifying the ef fects of NBPTS certified teachers in student achievement in Florida. 76 The study focused on math and reading results from students in grades three through ten over a four year period from school year 2000 2001 to school year 2003 2004. 77 73 Ibid. 74 Ibid. 75 Longitudinal Analysis with F Economics of Education Review 26, (2007): 673 682. [online] Retrieved May 5, 2010 from http://www.nber.org/papers/w13617.pdf?new_window=1 76 Ibid. 77 Harris, D. N. & S National Center for Analysis of Longitudinal Data in Educational Research (January 13, 2007). [online]. Available from http://www.nbpts.org/UserFiles/File/Harris_Sass_Final_2007.pdf

PAGE 41

41 The study was uni que in that it was able to compare results across elementary, middle, and high school teachers. Often statistical modeling has limits in interpretation due to the school level being analyzed. Elementary school student data is easier to link to specific t eachers, but has the limitation of masking other factors effecting student achievement like peer effects. In high school, however, the ability to link the student to multiple teachers and student groups allows for the peer effect to be indentified and rem oved from the analysis thus more clearly identifying the teacher effect. 78 The researchers also eliminated from the analysis, any student where more than one primary content teacher was linked in the data set. In addition, only courses with ten to fifty e 79 The resulting sample used in the study was extremely large including achievement measures for over one million students. Over 30,000 math teachers were linked to student data and of these, 1,250 were NBPTS certified. Over 33,000 reading teachers were linked to student data and of these, 1,500 were NBPTS certified. This was significantly larger than the samples used in the Cavaluzzo and Goldhaber and Anthony studies from 2004. 80 This large sample revealed some important factors related to the distribution of NBC teachers in the state of Florida. In this data set, NBC teachers are less likely to be assigned to black students and students receiving free and reduced lunch. Only 16 % of students assigned to NBC teachers are black. Only 31 % of students receiving free and reduced lunch are assigned to NBC teachers. Non NBC 78 Ibid. 79 National Center for Analysis of Longitudinal Data in Educational R esearch (January 13, 2007). [online]. Available from http://www.nbpts.org/UserFiles/File/Harris_Sass_Final_2007.pdf 80 Ibid.

PAGE 42

42 teachers have 23 % black students and 43 % students receiving free and reduced lunch. 81 Analysis of studen t achievement data was limited to the use of FCAT SSS (Sunshine State Standards) scores in reading and math. 82 The FCAT SSS is aligned to curriculum standards as outlined by the state, but it is not a vertically aligned test. Results from this study showe d that teachers who obtained NBPTS certification were more productive in reading. No significant difference was found in mathematics between NBC teachers and non NBC teachers. The researchers also examined the periods before and after a teacher becomes N BPTS certified. Similar to the findings from the Clotfelder, Ladd and Vigdor study, Harris and Sass found that teachers who obtained NBPTS certification were more productive prior to the year they obtained the credential and were less productive after obt aining the credential. Again, they note that differences in productivity are only noted in reading and no differences are noted for math. 83 This study also supports earlier research by Goldhaber and Anthony that suggests that effectiveness of NBC teachers can vary by grade level. In this study, NBC teachers in Florida were more productive in mathematics at the elementary level after receiving the credential. In middle school, the opposite trend was revealed. The effectiveness of NBC teachers on student subgroups was also analyzed. NBC teachers were more productive with black students in mathematics ( twenty scale score points) 81 Ibid. 82 Note: The FCAT test is defined to measure the Sunshine State Standards (FL. Stat. § 1008.22) When FCAT scores are presented, the Sunshine State Standards Scale Score is produced. At the time of this test administration, the state of Florida was also administering a Norm Referenced Reading and Ma th test (SAT 10) to all students in grades 3 10. The use of the term FCAT SSS is merely to differentiate between the types of tests and scores. 83 National Cente r for Analysis of Longitudinal Data in Educational Research (January 13, 2007). [online]. Available from http://www.nbpts.org/UserFiles/File/Harris_Sass_Final_2007.pdf

PAGE 43

43 before they applied for certification. After certification, there is no significant difference of the effectiveness of NBC teach er and non NBC teachers. In reading, NBC teachers are more productive with students receiving free and reduced lunch ( twenty five scale score points) before they apply for certification and their effectiveness is only significant after certification with students whose prior achievement level was in the highest range on the test. This indicates that NBC teachers may be most effective in raising reading achievement in only high achieving students. 84 Harris and Sass concluded that the small positive effects for some sub groups of students in the data set may have been related to the type of test and the quality of the test as well as the alignment of the test to standards at that grade level as much as they are to the effectiveness being related to the crede distinctive mixture of certification, preparation, and merit pay, but that does not necessarily make it a more co s t 85 Also in 2007, Ladd, Sa ss, and Harris published an analysis of the impact of national board Teachers on student achievement in Florida and North Carolina. 86 This study built upon the previous study using Florida data by adding the results from North Carolina in order to provide a cross state comparison. In North Carolina, the link between students and teachers in only accurate at the elementary level, so the study only used elementary grades four and five for the cross state comparison. Results from the study reveal that the co efficients for non NBC teachers in both states are extremely 84 Harri National Center for Analysis of Longitudinal Data in Educational Research (January 13, 2007). [online]. Available from http://www.nbpts.org/UserFiles/File/Harris_Sass_Final_2007.pdf 85 Ibid. 86 Achievement The Nat ional Academies Committee of the Evaluation of the Impact of Teacher Certification by NBPTS. (February 28, 2007) [online] Retrieved May 5, 2010 from http://www7.nationalacadem ies.org/bota/NBPTS MTG4 Sass paper.pdf

PAGE 44

44 similar. In North Carolina, NBC teachers are more effective than non NBC teachers by 4% of a standard deviation in reading and 7% of a standard deviation in math. In Florida, the results are on ly positively correlated in reading for NBC teachers but only at about 2% of a standard deviation. The results from both, though positive, are relatively low in magnitude. 87 The data from both states revealed a similar trend in the decline of effectiveness post NBPTS certification. In 2010, Marietta embarked on a multi year study of the implementation of a new teacher evaluation system in Hillsborough County Public Schools in Florida. 88 Concurrently, he released a series of case studies outlining the deta ils and interim data reported yearly during the implementation. Data from the first year of implementation (2010 2011) reveal a positive effect from NBC teachers. Analysis of the teacher evaluation scores from both sides of the new teacher evaluation was reported. The Teaching 89 were reported at the district level for this report. The student achievement results were also reported at the district level. These achievement re sults used a value Center (VARC) led by Rob Meyer. 90 At the district level, results for 11,119 non NBC teachers and 577 NBC teachers were reported. Evaluation scores for thes e groups differed greatly. NBC teachers had a mean overall evaluation score of 68.70 where non NBC teachers had a mean overall evaluation score of 60.59. Overall evaluation 87 Achievement The National Academies Committee of the Evaluation of the Impact of Teac her Certification by NBPTS. (February 28, 2007) [online] Retrieved May 5, 2010 from http://www7.nationalacademies.org/bota/NBPTS MTG4 Sass paper.pdf 88 htfully Implementing Value Added Measures in Hillsborough County Public Harvard University 2012. 89 Ibid. 90 Ibid.

PAGE 45

45 scores have a possible range of 0 100. In both the observed practices portion an d the student achievement portion, the scores of NBC teachers were greater than one standard deviation higher than non NBC teachers. The mean Value Added score for NBC teachers was 25.90 and the mean Value Added score for non NBC teachers was 24.23. It is important to note that in the Hillsborough County Public School District, the new teacher evaluation has not yet evaluated all teaching groups. Several NBC teachers in this district serve in school based resource teacher roles. Scores for these teache rs are not included in these figures. The Merit Pay program in this school district provides awards based on the student achievement results calculated. In the 2010 2011 school year, 47.6 % of NBC teachers received a merit award where only 18.8 % of non NB C teachers received a merit award. 91 Use of V alue A dded M odeling The use of covariate adjustment models to calculate teacher effects is not a new statistical procedure, but has implications in the current state of Florida law linking teacher pay levels to s tudent achievement results. 92 Due to the use of a variety of individual calculation models, a wide array of included covariates, and instability of scores over time, many researchers continue to caution against the use of VAM scores as a solitary determina nt for evaluating teacher performance. 93 While many will argue 91 Added Measures in Hillsborough County Public npublished Case Study. Harvard University 2012. 92 Fl. Stat. § 1012.34 (2011) 93 Educational Testing Service. (2005). [online] Retrieved May 17, 2012 from http://www.ets.org/Media/Research/pdf/PICVAM.pdf Center for Reinventing Public Education (20 08). [online] Retrieved May 17, 2012, from http://www.crpe.org/cs/crpe/download/csr_files/wp_crpe5_badclass_nov08.pdf

PAGE 46

46 that the value added calculation offers better insight than other simple growth models, all caution that without the proper use of standard errors and confidence intervals for appropriately rep orting results, teachers can be misidentified as effective or ineffective. 94 Results reported can be shown to have variance from year to year and can occur due to inadequate controls in the model or due to the failure of reporting agencies to appropriate ly report confidence intervals in the results. 95 Several cautions exist due to non random assignment of students, multiple influences on student learning, instability due to lack of validity in statistical controls and inadequate tests from which useful and r eliable measures of student mastery of knowledge are gathered 96 In short, c urrent disagreement in the field has led to lack of confidence in the value added approach to measuring student gain over time. Recent Dissertations Linking Student Achievement R esults to NBC T eachers In 2003, Angela Stephens completed a dissertation at the University of South Carolina. 97 She examined the relationship between NBC teachers and student CALDER. (2009). [online] Retrieved May 17, 2012, from http://www.urban.org/UploadedPDF/1001431 what makes for a good teacher.pdf McCaffrey Value Added Models: Analytic Issues the National Academy of Education, Board on Testing and Accountability Workshop on Value Added Modeling, Washington, DC. (2008). [online] Retrieved May 17, 2012, fr om http://www7.nationalacademies.org/bota/VAM_McCaffrey_Lockwood_Paper.pdf Education Finance and Policy. Fall 2009, Vol. 4, No. 4, Pages 572 606. [online] Retrieved May 17, 2012, from https://www.utla.net/system/files/mccaffrey_study.pdf 94 Ibid. 95 Ibid. 96 Ibid. 97 Achievement. Doctoral Dissertation. University of South Carolina, 2003

PAGE 47

47 achievement in two school districts in South Carolina. Her study included studen t achievement results for fourth and fifth grade students in mathematics from the Palmetto Achievement Challenge Test. Data for this study was obtained from the school year 2002 and matched to the prior year score from 2001. Four clas ses of students in g rade four, and four classes from grade five were included in the study. The researcher attempted to match the classes closely and required that the poverty index be within a similar range and that the teacher have at least three years of teaching experien ce in the district. The study included approximately 160 students and sixteen teachers. Eight teachers were NBPTS certified and eight were non NBC teachers. Findings showed no significant difference in the mathematics achievement of students assigned t o NBC teachers and those assigned to non NBC teachers. 98 In 2008, Tom Vitale completed a dissertation for the University of Central Florida that examined the relationship between NBPTS certification and student achievement of third grade students in a cen tral Florida school district. 99 He included student achievement results from the third grade reading and mathematics state exams (FCAT). Included in this study were four NBC teachers and four non NBC teachers. Approximately 160 student scores were exami ned from the school year 2004 2005. In this study, hierarchical linear modeling was used in order to control for student experience. It is important to note that this study exp erienced a shift in focus due to the limited number of NBC teachers that taught full time in this district. Findings show that 98 Ibid. 99 and the Achievement Results Central Florida, 2008.

PAGE 48

48 scores. 100 In 2010, Katrina McDaniel c ompleted a dissertation at the University of Oklahoma examining the link between student achievement and NBC teachers in Title I schools. 101 She included student achievement results from benchmark tests in reading and math from 610 third, fourth, and fifth g rade students across sixteen schools in an urban district in Oklahoma for the school year 2008 2009. Nine NBC teachers and twenty one non NBC teachers were matched to student data for analysis. The researcher used hierarchical linear modeling in order to c ontrol for student characteristics such as gender, ethnicity, SES, educational attainment of the teacher and number of years teaching experience. For reading achievement, there was no significant difference among the two groups of teachers. An important finding was that the students assigned to NBC teachers had a higher mean prior achievement score in reading which suggests that there was selective assignment to these classes. In Mathematics, there was no significant difference between the two groups of teachers. In fact, the NBC teachers had a slightly negative coefficient related to student achievement in mathematics. 102 In 2010, David Gardner completed a dissertation at the University of Central Florida that examined the relationship between student a chievement and National 100 Ibid. 101 Dissertation, University of Oklahoma, 2010. 102 Dissertation, University of Oklahoma, 2010.

PAGE 49

49 Board Certification. 103 This study included student reading scores of third, fourth, and fifth grade students on a Student Reading Inventory (SRI) that was administered at the beginning and end of the school year. Lexile scores were gathered and compared using a pre post analysis. Student data was obtained from two Florida school districts during the school year 2008 2009. Approximately 16,000 student scores were examined. Approximately 15,000 teachers were included with 2,718 bein g NBC teachers. No significant difference was found between NBC teachers and non NBC teachers. In each grade level, it was noted, however, that NBC teachers were consistently assigned higher achievement scores upon entering their class. This calls into q uestion the random assignment of students. Given this finding, the researcher still found no significant difference between NBC teachers and non NBC teachers in the amount of gain in student achievement in reading from beginning of the school year to end of the school year with respect to the Lexile levels for students. Summary National Board Certification was developed by the National Board for Professional Teaching Standards as an effort to raise the bar for standards for teacher certification nationwi de. Studies commissioned by NBPTS with the purpose of determining if there is a relationship between National Board Certification and student achievement have mixed results. 104 Researchers disagree about the relationship between National Board Certificatio n and student achievement. Studies that 103 raduate Degreed, and National Board Certified Florida, 2010. 104 Note that studies included in this review of literature are all published by commissions, po licy groups, etc. Results from these studies, though heavily cited, have not been peer reviewed.

PAGE 50

50 demonstrate a positive impact on student achievement 105 are criticized upon analysis of the study due to limited effect sizes or sample sizes. Several studies have results that show no relationship between National B oard Certification and student achievement. 106 Debate exists over exactly how to measure gains in student achievement. The use of linear regression models and value added modeling has increased due to the perceived ability to isolate the effect of the teac her by controlling for other known variables such as ethnicity, free and reduced lunch status, exceptionality, E nglish language learner status, etc. Debate over the potential instability of value added residual scores over time has kept the issue of linki 107 105 In these studies, the researchers found that NBC teachers had a greater impact on student achievement in mathematics than non NBC teachers: Vandervoot, Amre in Beardsley and Berliner (2004); Goldhaber and Anthony (2004); and Clotfelder, Ladd, and Vigdor (2007) 106 In these studies, the researchers found no significant difference between NBC and non NBC teachers with respect to their impact on student achievem ent results in mathematics: Sanders, Ashton, and Wright (2005); McColskey, Stronge, Ward, and Tucker (2005); Harris and Sass (2007), and the dissertations studies of Stephens (2003), Vitale (2008), McDaniel (2010), and Gardner (2010). 107 See comments in p ublications from: Braun, H. (2005); Goldhaber (2010);Harris and Sass (2009); McCaffrey and Lockwood (2008); Sass (2008); Raudenbush (2004); McCaffrey, Sass, Lockwood, and Mihaly (2009).

PAGE 51

51 CHAPTER 3 RESEARCH DESIGN The purpose of this study was to examine the relationship between National Board Certification and student achievement in mathematics on the Florida Comprehe nsive Achievement Test (FCAT) in high poverty elementary schools in a large urban district. Further, the study results were intended to inform a discussion on the cost effectiveness of the bonus program for National Board Certified teachers in this school district. Questions addressed in this research were: Research Questions 1. To what extent do National Board Certified teachers in a central Florida urban school district demonstrate positive student achievement gains in mathematics in fourth and fifth grade as measured by FCAT that is greater than the gains demonstrated by no n National Board C ertified teachers in the same setting? 2. To what extent are the state and district level polices relat ed to compensation of National Board Certified Teachers cost effecti ve? Research Design This study used an ex post facto design to examine differences in the performance of students assigned to National Board Certified teachers and students assigned to non NBC teachers in high poverty elementary schools. This design was a ppropriate because data analyzed were historical achievement data. Ex post facto research is utilized as a research design when the independent variables have already occurred and when the researcher starts with the observation of the dependent variable o r variables. In this study, the independent variable, stud ent achievement, was analyzed at a five levels. Four of these levels used raw scores from the administration of FCAT mathematics and were broken into analysis levels based on the state definitions of making gain. These definitions included: increasing a level from the prior year,

PAGE 52

52 maintaining a level 3 or higher, or increasing a pre defined number of developmental scale score points. 1 The fifth level of analysis was the state reported value added calculation for each teacher. This level was mentioned separately as it is the only level that included student demographic information in the calculation and was based on a covariate adjustment model using hierarchical linear modeling. These calculation s were not completed by the researcher, but were obtained from the historical database provided by the state of Florida Department of Education to the school district in this study. 2 The dependent variable, National Board Certification was analyzed on a d ichotomous scale with two levels: National Board Certified and non National Board Certified. Research Population This school district studied is a large urban district in central Florida that has consistently been ranked in the top ten largest urban dis tricts in the nation. This district is comprised of greater than 197,000 students in 243 school sites and has 58 % of students living in poverty as defined by national free lunch standards. Thirty eight school sites were grouped in this district under th criterion for becoming a Renaissance School was to have a student enrollment with 1 Grades: Technical Assistance [online] Retrieved May 5, 2010 from http://schoolgrades.fldoe.org/pdf/0809/2009SIRTAP.pdf ng Schools Grades: Technical Assistance [online] Retrieved May 5, 2010 from http://schoolgrades.fldoe.org/pdf/0910/2010SchoolGradesTAP.pdf Florida Department of Educati [online] Retrieved April 9, 2011 from http://schoolgrades.fldoe.org/pdf/1011/SchoolGradesTAP2011.pdf 2 Office of Assessment and Accountability, Hillsborough County Public Schools. Electronic Spreadsheet Files received February, 2012.

PAGE 53

53 90 % or greater of the students qualifying for free or reduced price lunch. 3 For this study, data analyzed was reduced to a subset of the larg er district. Elementary Renaissance school sites were selected for analysis. These schools were chosen for a variety of reasons. First, National Board Certified teachers receive a monetary bonus for teaching in these sites. 4 Second, though these schoo l sites have some variance in student population, the number of students receiving free or reduced price lunch is reasonably constant at 90 % or greater in each site. Third, each of these school sites has a similar set of school supports such as additional reading teachers, math teachers, and student services personnel. All school sites use common district developed curriculum with common content and pacing guides. Student data from twenty eight school sites were included in this study. Student data from fourth and fifth grade students were analyzed over a three year period (2008 2011) which resulted in the examination of 15,698 student records and the student teacher assignments for 1,436 teachers. Each year was analyzed separately. For the 2008 2009 d ata set, 5,559 student records were analyzed; 2,780 fourth grade and 2,779 fifth grade. These students were assigned to 531 teachers. For the fourth grade data set, 224 students were assigned to seventeen National Board Certified Teachers and 2,556 stude nts were assigned to 265 non National Board Certified Teachers. For the fifth grade data set, 214 students were assigned to nineteen National Board Certified Teachers and 2,565 students were assigned to 230 non National Board Certified Teachers. For the 2009 2010 data set, 3 http://www.sdhc.k12.fl.us/SSFPDiv/Salary % 20Differential % 202010.11A.pdf 4 Differentiated Pay: Contract Section 21.4.10. Hillsborough Classroom Teachers Association, Inc. Teacher Contract 2010 2013 Hillsborough County Public Schools, Hillsborough, FL. 2010. [online] retrieved May 5, 2010 from http://www.hillscta.org/Welcome_files/teachercontract.pdf

PAGE 54

54 4,780 student records were analyzed; 2,440 fourth grade and 2,340 fifth grade. These students were assigned to 450 teachers. For the fourth grade data set, seventy students were assigned to nine National Board Certified Teachers and 2,370 students were assigned to 220 non National Board Certified Teachers. For the fifth grade data set, 121 students were assigned to nine National Board Certified Teachers and 2,219 students were assigned to 212 non National Board Certified Teachers. F or the 2010 2011 data set, 5,359 student records were analyzed; 2,650 fourth grade and 2,709 fifth grade. These students were assigned to 455 teachers. For the fourth grade data set, 195 students were assigned to fourteen National Board Certified Teacher s and 2,455 students were assigned to 217 non National Board Certified Teachers. For the fifth grade data set, 209 students were assigned to twelve National Board Certified Teachers and 2,500 students were assigned to 212 non National Board Certified Teac hers. Only students who counted in the calculation of school grade (those with a FTE Survey 2/3 match) were included in the data files for this study. 5 This means that nd were assigned to the teacher for the entire time. 6 Fourth and f ifth grade scores were chosen because these are the only two grades at the elementary level for which gain from the prior year could be calculated as they were the only elementary grade leve ls with prior year FCAT mathematics scores from which to calculate the gain in developmental scale score. 5 Florida Department of E [online] Retrieved April 9, 2011 from http://schoolgrades.fldoe.org/pdf/1011/SchoolGradesTAP201 1.pdf 6 Ibid.

PAGE 55

55 Data Source Data records were obtained from the school district assessment office following a specific study approval and data request. The file rece ived included student scores from the FCAT SSS Mathematics tests for grades four and five for each year for the school years 2008 2009, 2009 2010, and 2010 2011. Students were matched using encoded student identification numbers and were linked to encoded teacher identification numbers. Records coded each teacher as being National Board Certified or non National Board Certified. Student data were also provided to indicate whether each student did or did not make gain for the school year as well as an ind ication of the way(s) that the student demonstrated that gain. The state of Florida defines making gain as: maintaining a level 3 or higher, improving a performance level form the prior year, or increasing a specified number of developmental scale score p oints from the prior year. 7 Finally, the file received from the district contained the value added score calculated by the American Institute for Research (AIR), a vendor calculating value added scores on behalf of the state of Florida for each teacher fo r each of the school years in the study. The teacher level residual was provided for each teacher. These residuals were calculated by AIR using a covariate adjustment model that included student demographic variables. 8 7 [online] Retrieved April 9, 2011 from http://schoolgrad es.fldoe.org/pdf/1011/SchoolGradesTAP2011.pdf 8 [online] Retrieved June 4, 2012 from http://www.fldoe.org/committees/pdf/SummaryFinalRecommendation.pdf

PAGE 56

56 Data Analysis I ndependent sample s t test s w ere conducted to compare the gains made by students assigned to National Board Certified teachers and the gains made by students assigned to non National Board Certified teachers. The t test s were appropriate for analysis because the means of t he two groups were compared. The t test was used to determine if the differences in the means were statistically significant. Individual t tests were performed for each category of scores in each school year. Resulting p values were used to determine si gnificance at the .05 level with 95% confidence intervals. Summary Chapter three contains the general methodology, research setting, population, and data analysis procedures utilized. The design of this study was intended to address the relationship bet ween student achievement and National Board Certification by comparing the student achievement results of National Board Certified Teachers as compared to their non National Board Certified peers. After it was determined whether a relationship between stu dent achievement and National Board Certification existed, plan could be made.

PAGE 57

57 CHAPTER 4 OBSERVATIONS Introduction This study examined the relationship between National Board Certification and student performance in high poverty schools in a large urban school district in Central Florida. Student scores from FCAT mathematics administrations in 2008 2009, 2009 2010, and 2010 2011 were examined. Scores were limit ed to fourth and fifth grade students in twenty eight high poverty elementary schools in the same school district. The primary question of this study was to determine if there was a difference in student status and therefore, if the school The study utilized a t test to determine statistical significance of the means of the groups examined. When calculating school grades in the state of Florida, the Department of Education calculates the percent age of students who make gain in reading and mathematics. 1 Following all of the state defined rules for inclusion in school grade, 2 the overall percentage of students making gain was calculate d for each type of teacher. A t test was completed on the overall means in order to determine statistical significance of the differences in the means of the two groups. 1 [onl ine] Retrieved April 9, 2011 from http://schoolgrades.fldoe.org/pdf/1011/SchoolGradesTAP2011.pdf 2 Ibid

PAGE 58

58 Using Florida State Department of Education grading criteria, 3 there are three diff erent ways that a student could be considered to make gain from the prior year. The three ways that a student could meet the criteria for making gain were: 1) improving a level from the prior year, 2) maintaining a level three, four, or five from the prio r year, and 3) improving a pre determined number of developmental scale points within level one or two. In order to determine if there was a relationship between the way in which a student made gain and the National Board Certif ication status of the teach er, t test s were performed for each category. Finally, the state of Florida has recently contracted with American Institute for Research (AIR) to calculate value added score for teachers using FCAT reading and math results. Scores were provided for each of the years included in this study. 4 The average teacher score was calculated for each group as well as the overall standard deviation of the variance among these scores. A t test was completed to determine if there was a relationship between the value added scores of National Board Certified Teachers and the student achievement results in high poverty elementary schools. 3 early Progress (AYP) http://schoolgrades.fldoe.org/pdf/0910/AYPTAP2010.pdf to Calculating Schools Grades: Technical Assistance [online] Retrieved May 5, 2010 from http://schoolgrades.fldoe.org/pdf/0809/2009SIRTAP.pdf 2010 Guide to Calculating Schools Grades: Technical Assistance [online] Retrieved May 5, 2010 from http://schoolgrades.fldoe.org/pdf/0910/2010SchoolGradesTAP.pdf Florid [online] Retrieved April 9, 2011 from http://schoolgrades.fldoe.org/pdf/1011/S choolGradesTAP2011.pdf 4 [online] Retrieved June 4, 2012 from http://www.fldoe.org/committees/pdf/SummaryFinalRecommendation.pdf

PAGE 59

59 Descriptive Statistics Student data for this study were examined across three school years and the t tests were run for each year in dependently. The total sample included 15,698 student records for students assigned to 1,436 teachers in twenty eight elementary schools. This included eighty National Board Certified teachers and 1,356 non National Board Certified teachers. 2008 2009 R esults The 2008 2009 data sample included 5,559 students. Student scores for 438 students assigned to thirty six National Board Certified Teachers and 5,121 students assigned to 495 non National Board Certified teachers were i ncluded and examined. Table s 4 1 and 4 2 include the count of teachers and students included in the analysis Table 4 1 Number of National Board Certified Teachers 2008 2009 Number of teachers 20 08 2009 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 19 17 36 Not Certified 230 265 495 Total 249 282 531 Table 4 2 Count of students assigned to National Board Certified Teachers 2008 2009 Number of students 20 08 2009 FCAT Math FIFTH GRADE FOURTH GRADE Total Teacher is National Board Certified 214 2 24 438 Teacher is not National Board Certified 2565 2556 5121 Total 2779 2780 5559

PAGE 60

60 Overall G ain Table 4 3 Percent of students making gain 2008 2009 Percent of Students making gain 20 08 2009 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 73.364486 62.9464286 68.0365297 Not Certified 63.5477583 65.258216 64.4014841 p value from t test .0040* .4887 .1351 (*p<.05) Fifth G rade The percent of students making gain from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2008 2009 was 73.36 % The percent of students making gain from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2008 2009 was 63.54 % The t test determined the difference of these means to be statistically significant with a p value of .0040. Fourth G rade The percent of students making gain from the prior school year on FCAT Mathematics for fourth grade students assigned to National Bo ard Certified teachers in 2008 2009 was 62.94 % The percent of students making gain from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2008 2009 was 65.25 % The t test determined the dif ference of these means not to be statistically significant with a p value of .4887. Combined When the grade levels were combined, 68.03 % of students assigned to National Board Certified teachers in 2008 2009 made gain from the prior year. 64.40 % of the

PAGE 61

61 st udents assigned to non National Board Certified teachers in 2008 2009 made gain from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .1351. Improving a L evel Table 4 4 Percent of st udents improving a level 2008 2009 Percent of students improving a level 20 08 2009 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 24.7663551 16.9642857 20.7762557 Not Certified 18.4015595 19.8748044 19.1368873 p value from t test .0235* .2948 .4418 (*p<.05 ) Fifth G rade The percent of students improving a level from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2008 2009 was 24.76 % The percent of studen ts improving a level from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2008 2009 was 18.4 % The t test determined the difference of these means to be statistically significant with a p va lue of .0235. Fourth G rade The percent of students improving a level from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2008 2009 was 16.96 % The percent of students improving a level from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2008 2009 was 19.87 % The t test determined the difference of these means not to be statistically significant with a p value of .2948.

PAGE 62

62 C ombined When the grade levels were combined, 20.77 % of students assigned to National Board Certified teachers in 2008 2009 improved a level from the prior year. 19.13 % of the students assigned to non National Board Certified teachers in 2008 2009 improved a level from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .4418. Maintaining L evel 3, 4, or 5 Table 4 5 Percent of students maintaining level 3, 4, or 5 Percent of students Main taining 3+ 20 08 2009 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 26.1682243 31.25 28.7671233 Not Certified 22.3001949 31.3771518 26.8306971 p value from t test .2014 .9753 .3443 (*p<.05) Fifth G rade The percent of stud ents maintaining a level three, four, or five from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2008 2009 was 26.16 % The percent of students maintaining a level three, four, or five f rom the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2008 2009 was 22.30 % The t test determined the difference of these means not to be statistically significant with a p value of .2014. Fou rth G rade The percent of students maintaining a level three, four, or five from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board

PAGE 63

63 Certified teachers in 2008 2009 was 31.25 % The percent of students maintaining a level three, four, or five from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2008 2009 was 31.37 % The t test determined the difference of these means not to be statistically significa nt with a p value of .9753. Combined When the grade levels were combined, 28.76 % of students assigned to National Board Certified teachers in 2008 2009 maintained a level three, four, or five from the prior year. 26.83 % of the students assigned to non Nat ional Board Certified teachers in 2008 2009 maintained a level three, four, or five from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .3443. Improving Developmental Scale Score T able 4 6 Percent of students improving developmental scale s core Percent of students improving DSS 2008 2009 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 22.4299065 14.7321429 18.4931507 Not Certified 22.8460039 14.0062598 18.4 338996 p value from t test .8933 .7726 1.000 (*p<.05) Fifth G rade The percent of students meeting criteria for making gain by improving the developmental scale score from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2008 2009 was 22.42 % The percent of students meeting criteria for making gain by improving the

PAGE 64

64 developmental scale score from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2008 2009 was 22.84 % The t test determined the difference of these means not to be statistically significant with a p value of .8933. Fourth G rade The percent of students meeting criteria for making gain by improving the developmen tal scale score from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2008 2009 was 14.73 % The percent of students meeting criteria for making gain by improving the developmental scale s core from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2008 2009 was 14.00 % The t test determined the difference of these means not to be statistically significant with a p value of .77 26. Combined When the grade levels were combined, 18.49 % of students assigned to National Board Certified teachers in 2008 2009 met criteria for making gain by improving the developmental scale score from the prior year. 18.43 % of the students assigned to non National Board Certified teachers in 2008 2009 met criteria for making gain by improving the developmental scale score from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of 1.00 0.

PAGE 65

65 Value Added Score Table 4 7 Average teacher value added s core Average of teacher VAM e stimate 2008 2009 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 12.991245094 5.853849702 9.095043809 Not Certified 5.975947031 5.8976025 54 5.936556765 p value from t test .0002* .9816 .0125* (*p<.05 ) Fifth G rade The average value added score using student scores on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2008 2009 was 12.99. The av erage value added score using student scores on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2008 2009 was 5.97. The t test determined the difference of these means to be statistically significant with a p value of .0002. Fourth G rade The average value added score using student scores on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2008 2009 was 5.85. The average value added score using student scores on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2008 2009 was 5.89. The t test determined the difference of these means not to be statistically significant with a p value of .9816. Combined When the grade levels were combined, the average value added score using student scores on FCAT Mathematics for students assigned to National Board Certified

PAGE 66

66 teachers in 2008 2009 was 9.09. The average value added score using student scores on FCAT Mathematics for students ass igned to non National Board Certified teachers in 2008 2009 was 5.93. The t test determined the difference of these means to be statistically significant with a p value of .0125. 2009 2010 R esults The 2009 2010 data sample included 4,780 students. Studen t scores for 191 students assigned to eighteen National Board Certified Teachers and 4,589 students assigned to 432 non National Board Certified teachers were i ncluded and examined. Tables 4 8 and 4 9 include the count of teachers and students included in this analysis. Table 4 8 Number of National Board Certified Teachers 2009 2010 Number of Teachers 2009 2010 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 9 9 18 Not Certified 212 220 432 Total 221 229 450 Table 4 9 Count o f students assigned to National Board Certified Teachers 2009 2010 Number of Students 2009 2010 FCAT Math FIFTH GRADE FOURTH GRADE Total Teacher is National Board Certified 121 70 191 Teacher is Not national Board Certified 2219 2370 4589 Total 23 40 2440 4780

PAGE 67

67 Overall G ain Table 4 10 Percent of students making gain 2009 2010 Percent of Students making gain 2009 2010 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 65.2892562 65.7142857 65.4450262 Not Certified 56.42181 16 61.7299578 59.1632164 p value from t test .0566 .4972 .0820 (*p<.05) Fifth G rade The percent of students making gain from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2009 2 010 was 65.28 % The percent of students making gain from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2009 2010 was 56.42 % The t test determined the difference of these means not to be statistically significant with a p value of .0566. Fourth G rade The percent of students making gain from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2009 2010 was 65.71 % The percent of students making gain from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2009 2010 was 61.72 % The t test determined the difference of these means not to be statistically significant w ith a p value of .4972. Combined When the grade levels were combined, 65.44 % of students assigned to National Board Certified teachers in 2009 2010 made gain from the prior year. 59.16 % of the

PAGE 68

68 students assigned to non National Board Certified teachers in 2009 2010 made gain from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .0820. Improving a L evel Table 4 11 Percent of students improving a level 2009 2010 Percent of students impr oving a level 2009 2010 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 16.5289256 11.4285714 14.6596859 Not Certified 11.5817936 17.6793249 14.7308782 p value from t test .0973 .1780 .9695 (*p<.05) Fifth G rade The percent of students improving a level from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2009 2010 was 16.52 % The percent of students improving a level from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2009 2010 was 11.58 % The t test determined the difference of these means not to be statistically significant with a p value of .0973. Fourth G rade The percent of students impro ving a level from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2009 2010 was 11.42 % The percent of students improving a level from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2009 2010 was 17.67 % The t test determined the difference of these means not to be statistically significant with a p value of .1780.

PAGE 69

69 Combined When the grade levels were combined, 14.65 % of s tudents assigned to National Board Certified teachers in 2009 2010 improved a level from the prior year. 14.73 % of the students assigned to non National Board Certified teachers in 2009 2010 improved a level from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .9695. Maintaining level 3, 4, or 5 Table 4 12 Percent of students maintaining level 3, 4, or 5 Percent of students Maintaining 3+ 2009 2010 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 31.4049587 32.8571429 31.9371728 Not Certified 25.8675079 30.1687764 28.0889083 p value from t test .1724 .6280 .2403 (*p<.05) Fifth G rade The percent of students maintaining a level three, four, or five from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2009 2010 was 31.40 % The percent of students maintaining a level three, four, or five from the prior year on FCAT Mathematics for fifth gr ade students assigned to non National Board Certified teachers in 2009 2010 was 25.86 % The t test determined the difference of these means not to be statistically significant with a p value of .1724. Fourth G rade The percent of students maintaining a lev el three, four, or five from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board

PAGE 70

70 Certified teachers in 2009 2010 was 32.85 % The percent of students maintaining a level three, four, or five from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2009 2010 was 30.16 % The t test determined the difference of these means not to be statistically significant with a p value of .6280. Combined When the grade levels were combined, 31.93 % of students assigned to National Board Certified teachers in 2009 2010 maintained a level three, four, or five from the prior year. 28.08 % of the students assigned to non National Board Certified teachers in 2009 2010 maintai ned a level three, four, or five from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .2403. Improving Developmental Scale Score Table 4 13 Percent of students improving development al scale s core Percent of students improving DSS 2009 2010 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 17.3553719 21.4285714 18.8481675 Not Certified 18.9725101 13.8818565 16.3434299 p value from t test .6615 .0711 .3602 (*p<.05) Fifth G rade The percent of students meeting criteria for making gain by improving the developmental scale score from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2009 2010 wa s 17.35 % The percent of students meeting criteria for making gain by improving the

PAGE 71

71 developmental scale score from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2009 2010 was 18.97 % The t test determined the difference of these means not to be statistically significant with a p value of .6615. Fourth G rade The percent of students meeting criteria for making gain by improving the developmental scale score from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2009 2010 was 21.42 % The percent of students meeting criteria for making gain by improving the developmental scale score from the prior year on FCAT Mathematics for f ourth grade students assigned to non National Board Certified teachers in 2009 2010 was 13.88 % The t test determined the difference of these means not to be statistically significant with a p value of .0711. Combined When the grade levels were combined, 18.84 % of students assigned to National Board Certified teachers in 2009 2010 met criteria for making gain by improving the developmental scale score from the prior year. 16.34 % of the students assigned to non National Board Certified teachers in 2009 201 0 met criteria for making gain by improving the developmental scale score from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .3602.

PAGE 72

72 Value Added Score T able 4 14 Average teacher value added s core Average of Teacher VAM Estimate 2009 2010 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 9.297646523 12.79599025 10.56809767 Not Certified 0.546740504 8.437261298 4.098250979 p value from t test .0001* .2 639 .0015* (*p<.05) Fifth G rade The average value added score using student scores on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2009 2010 was 9.29. The average value added score using student scores on FCA T Mathematics for fifth grade students assigned to non National Board Certified teachers in 2009 2010 was 0.54. The t test determined the difference of these means to be statistically significant with a p value of .0001. Fourth G rade The average value ad ded score using student scores on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2009 2010 was 12.79. The average value added score using student scores on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2009 2010 was 8.43. The t test determined the difference of these means not to be statistically significant with a p value of .2639. Combined When the grade levels were combined, the average value added score usin g student scores on FCAT Mathematics for students assigned to National Board Certified

PAGE 73

73 teachers in 2009 2010 was 10.56. The average value added score using student scores on FCAT Mathematics for students assigned to non National Board Certified teachers i n 2009 2010 was 4.09. The t test determined the difference of these means to be statistically significant with a p value of .0015. 2010 2011 R esults The 2010 2011 data sample included 5,359 students. Student scores for 404 students assigned to twenty si x National Board Certified Teachers and 4,955 students assigned to 429 non National Board Certified teachers were i ncluded and examined. Tables 4 15 and 4 16 include the count of teachers and students included in this analysis. Table 4 15 Number of Natio nal Board Certified Teachers 2010 2011 Number of teachers 2010 2011 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 12 14 26 Not Certified 212 217 429 Total 224 231 455 Table 4 16 Count of students assigned to National Board C ertified Teachers 2010 2011 Number of Students 2010 2011 FCAT Math FIFTH GRADE FOURTH GRADE Total Teacher is National Board Certified 209 195 404 Teacher is not National Board Certified 2500 2455 4955 Total 2709 2650 5359

PAGE 74

74 Overall G ain Table 4 17 Percent of students making gain 2010 2011 Percent of Students making gain 2010 2011 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 40.6698565 44.6153846 42.5742574 Not Certified 39.88 45.5804481 42.7043391 p value from t test .8204 .8226 .9402 (*p<.05) Fifth G rade The percent of students making gain from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2010 2011 was 40.66 % The percent of students maki ng gain from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2010 2011 was 39.98 % The t test determined the difference of these means not to be statistically significant with a p value of 8204. Fourth G rade The percent of students making gain from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2010 2011 was 44.61 % The percent of students making gain from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2010 2011 was 45.58 % The t test determined the difference of these means not to be statistically significant with a p value of .8226. Combined When the gra de levels were combined, 42.57 % of students assigned to National Board Certified teachers in 2010 2011 made gain from the prior year. 42.70 of the

PAGE 75

75 students assigned to non National Board Certified teachers in 2010 2011 made gain from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .9402. Improving a L evel Table 4 18 Percent of students improving a level 2010 2011 Percent of students improving a level 2010 2011 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 34.9282297 28.7179487 31.9306931 Not Certified 35.08 31.1201629 33.1180626 p value from t test .9768 .4854 .6215 (*p<.05) Fifth G rade The percent of students improving a level from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2010 2011 was 34.92 % The percent of students improving a level from the prior year on FCAT Mathematics for fifth grade students assigned to non National B oard Certified teachers in 2010 2011 was 35.08 % The t test determined the difference of these means not to be statistically significant with a p value of .9768. Fourth G rade The percent of students improving a level from the prior school year on FCAT Mat hematics for fourth grade students assigned to National Board Certified teachers in 2010 2011 was 28.71 % The percent of students improving a level from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2010 2011 was 31.12 % The t test determined the difference of these means not to be statistically significant with a p value of .4854.

PAGE 76

76 Combined When the grade levels were combined, 31.93 % of students assigned to National Board Certified teache rs in 2010 2011 improved a level from the prior year. 33.11 % of the students assigned to non National Board Certified teachers in 2010 2011 improved a level from the prior year. The t test determined the difference of these means not to be statistically significant with a p value of .6215. Maintaining level 3, 4, or 5 Table 4 19 Percent of students maintaining level 3, 4, or 5 Percent of students maintaining 3+ 2010 2011 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 1.4354067 15 .8974359 8.4158416 Not Certified 1.76 14.4602851 8.0524723 p value from t test .7488 .7027 .7766 (*p<.05) Fifth G rade The percent of students maintaining a level three, four, or five from the prior school year on FCAT Mathematics for fifth grade st udents assigned to National Board Certified teachers in 2010 2011 was 1.43 % The percent of students maintaining a level three, four, or five from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teacher s in 2010 2011 was 1.76 % The t test determined the difference of these means not to be statistically significant with a p value of .7488. Fourth G rade The percent of students maintaining a level three, four, or five from the prior school year on FCAT Mat hematics for fourth grade students assigned to National Board

PAGE 77

77 Certified teachers in 2010 2011 was 15.89 % The percent of students maintaining a level three, four, or five from the prior year on FCAT Mathematics for fourth grade students assigned to non Na tional Board Certified teachers in 2010 2011 was 14.46 % The t test determined the difference of these means not to be statistically significant with a p value of .7027. Combined When the grade levels were combined, 8.41 % of students assigned to National Board Certified teachers in 2010 2011 maintained a level three, four, or five from the prior year. 8.05 % of the students assigned to non National Board Certified teachers in 2010 2011 maintained a level three, four, or five from the prior year. The t tes t determined the difference of these means not to be statistically significant with a p value of .7766. Improving Developmental Scale Score Table 4 20 Percent of students improving developmental scale s core Percent of students improving DSS 2010 2011 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 4.3062201 0 2.2277228 Not Certified 3.04 0 1.5338042 p value from t test .2986 0 .2755 (*p<.05) Fifth G rade The percent of students meeting criteria for making gain by improving th e developmental scale score from the prior school year on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2010 2011 was 4.30 % The percent of students meeting criteria for making gain by improving the

PAGE 78

78 development al scale score from the prior year on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2010 2011 was 3.04 % The t test determined the difference of these means not to be statistically significant with a p valu e of .2986. Fourth G rade The percent of students meeting criteria for making gain by improving the developmental scale score from the prior school year on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2010 2011 was 0 % The percent of students meeting criteria for making gain by improving the developmental scale score from the prior year on FCAT Mathematics for fourth grade students assigned to non National Board Certified teachers in 2010 2011 was 0 % The t te st could not be completed because there was no variance in these means. Combined When the grade levels were combined, 2.22 % of students assigned to National Board Certified teachers in 2010 2011 met criteria for making gain by improving the developmental scale score from the prior year. 1.53 % of the students assigned to non National Board Certified teachers in 2010 2011 met criteria for making gain by improving the developmental scale score from the prior year. The t test determined the difference of the se means not to be statistically significant with a p value of .2755.

PAGE 79

79 Value Added Score Table 4 21 Average teacher value added s core Average of Teacher VAM Estimate 2010 2011 FCAT Math FIFTH GRADE FOURTH GRADE Total National Board Certified 6.49 1595017 34.60469906 10.87783314 Not Certified 1.411677836 5.069407262 1.80394755 p value from t test .0415* .0001* .0001* (*p<.05) Fifth G rade The average value added score using student scores on FCAT Mathematics for fifth grade students assigned to National Board Certified teachers in 2010 2011 was 6.49. The average value added score using student scores on FCAT Mathematics for fifth grade students assigned to non National Board Certified teachers in 2010 2011 was 1.41. The t test determined the difference of these means to be statistically significant with a p value of .0415. It is important to note in this case that the average value added score for non National Board Certified teachers is higher than the average value added score for fifth grade National Board Certified teachers. Therefore the statistically significant results indicate that the fifth grade non National Board Certified teachers had greater impact on student achievement than the fifth grade teachers who were National Board Cer tified. Fourth G rade The average value added score using student scores on FCAT Mathematics for fourth grade students assigned to National Board Certified teachers in 2010 2011 was 34.60. The average value added score using student scores on FCAT Mathemat ics for fourth grade students assigned to non National Board Certified teachers in 2010 2011

PAGE 80

80 was 5.06. The t test determined the difference of these means to be statistically significant with a p value of .0001. Combined When the grade levels were combine d, the average value added score using student scores on FCAT Mathematics for students assigned to National Board Certified teachers in 2010 2011 was 10.87. The average value added score using student scores on FCAT Mathematics for students assigned to no n National Board Certified teachers in 2010 2011 was 1.80. The t test determined the difference of these means to be statistically significant with a p value of .0001. Summary of Results The research questions guiding this study are divided into two group s. Research question one required data analysis and is therefore addressed in this chapter. Research question two required application of public policy and policy analysis and are addressed in chapter five. Research Q uestion 1 : To what extent do National ly Board Certified teachers in a central Florida urban school district demonstrate positive student achievement gains in mathematics in fourth and fifth grade as measured by FCAT that is greater than the gains demonstrated by non nationally board certified teachers in the same setting? Grade 4: The results from this study provide only one significant finding that the student achievement gains of students assigned to National Board Certified teachers in fourth grade is greater than those of non National Boa rd Certified teachers in the same setting. In the 2010 2011 data set, the difference in value added scores for fourth grade

PAGE 81

81 teachers in these two groups was significant with a p value of .0001. National Board Certified Teachers had a higher value added s core which means that they had more students meeting the expected growth for the 2010 2011 school year. Across all other data sets there w ere no significant difference s in fourth grade scores. Grade 5: The results from this study provided several signific ant findings that the student achievement gains of students assigned to National Board Certified teachers in fifth grade was greater than those of non National Board Certified teachers in the same setting. 1. In the 2008 2009 data set, the difference in the percent of overall students making gain from the prior year for fifth grade teachers in these two groups was significant with a p value of .0040. National Board Certified teachers had a greater percent age of students making gain overall for the 2008 2009 school year. 2. In the 2008 2009 data set, the difference in the percent of students improving a level from the prior year for fifth grade teachers in these two groups was significant with a p value of .0235. National Board Certified teachers had a greate r percent age of students improving a level over the prior year for the 2008 2009 school year. 3. In the 2008 2009 data set, the difference in the average teacher value added score for fifth grade te achers in these two groups was significant with a p value of .0002. National Board Certified teachers had a higher average value added score for the 2008 2009 school year. 4. In the 2009 2010 data set, the difference in the average teacher value added score for fifth grade te achers in these two groups was significant with a p value of .0001. National Board Certified teachers had a higher average value added score for the 2009 2010 school year. The results of this study also provided one significant finding that the student achievement gains of students assigned to non National Board Certified teachers in fifth grade is greater than those of National Board Certified teachers in the same setting.

PAGE 82

82 In the 2010 2011 data set, the difference in the average teacher value added score for fifth grade teachers in these two group s was significant with a p value of .0415. Non National Board Certified teachers had a higher average value added score for the 2010 2011 school year. Combined: The results from this study provide one significant finding that the student achievement gains of students assigned to National Board Certified teachers in fourth and fifth grade combined is greater than those of non National Board Certified teachers in the same setting. In the 2008 2009, 2009 2010, and the 2010 2011 data set s the difference in v alue added scores for fourth and fifth grade teachers combined in these two groups was significan t with a p value <.01. Table 4 22 displays this finding. Table 4 22 Average v alue added score for grade 4 and 5 teachers combined Average of Teacher VAM E stimate 0809 FCAT Math 0910 FCAT Math 1011 FCAT Math Total Total Total National Board Certified 9.095043809 10.56809767 10.87783314 Not Certified 5.936556765 4.098250979 1.80394755 p value from t test .0125 .0015 .0001 (*p<.01) National Board Certified Teachers in grade four and five combined had a higher value added score each year which means that they had more students meeting the expected growth for all three years in the study. Across all other data sets there was no significant differenc e in fourth and fifth grade combined scores.

PAGE 83

83 Summary Results obtained from this study do not clearly support or disprove the hypothesis that National Board Certification has a positive impact on the achievement results of fourth and fifth grade students on the mathematics portion of the FCAT in high poverty elementary schools in this particular school district. Statistically significant findings emerged between NBC teachers and non NBC teachers when using a value added measure. That is, National Board Cer tified Teachers had a positive impact on the achievement results of fourth and fifth grade students in high poverty elementary schools in this school district on the mathematics portions of the FCAT. When using the attainment measures as defined for deter mining Florida school grades, the results only show ed significant findings that National Board Certification had a positive impact in two categories for fifth grade teachers for the 2008 2009 school year.

PAGE 84

84 CHAPTER 5 CONCLUSIONS This study examined the st udent achievement results of fourth and fifth grade students enrolled in high poverty elementary schools in an urban school district in Central Florida. Student results on FCAT Mathematics were examined. Means and standard deviations were calculated for two groups: students assigned to National Board Certified teachers and students assigned to non National Board Certified teachers. The study included student data from three school years from 2008 2011 which included 15,698 student records for students a ssigned to 1,436 teachers. The study included a total of eighty National Board Certified teachers and 1,356 non National Board Certified teachers in twenty eight high poverty elementary schools. Summary of Findings: Results from this study were varied de pending on the measures included. 1. When analyzing standard test score results using the definitions for making gain as utilized by the state of Florida in calculating school grades, the results show no consistently significant finding supporting the hypot hesis that National Board Certification has a positive impact on student achievement in these high poverty schools. From the literature review conducted for this study, this finding supports the studies by Sanders, McColskey and Stronge, Harris and Sass, and the dissertations studies of Stephens, Vitale, McDaniel, and Gardner. In each of these studies, the researchers found no significant difference between NBC and Non NBC teachers with respect to their impact on student achievement results in mathematics 2. When using the teacher value added score, which measures gain overall without limiting the definition of gain to specific cut levels, the results show highly significant findings supporting the hypothesis that National Board Certification has a positive impact on student achievement in these high poverty schools.

PAGE 85

85 From the literature conducted for this study, this finding supports the findings from the studies by Vandervoot, Goldhaber and Anthony, and Clotfelder. In these studies, the researchers found that NBC teachers had a greater impact on student achievement in mathematics than non NBC teachers. This finding does not support the prior studies by Sanders, McColskey and Stronge, Harris and Sass, and the dissertations studies of Stephens, Vitale, McDa niel, and Gardner. Research q uestion 2 : To what extent are the state and district level polices related to compensation of Nationally Board Certified Teachers cost effective? Results from this study indicate that determining cost effectiveness will depend on the type of student achievement gain the entity is interested in measuring. If the goal was to reward teachers who were able to impact student achievement in ways that will have an immediate and positive impact on school grade, then the policies in pla ce are not cost effective as NBC teachers in this study did not demonstrate positive impact in this manner. If, however, the goal was to reward teachers who were able to impact student ere the student started at the beginning of the school year, then the results from the study indicate that the policies in place are cost effective as the value added scores for NBC teachers are greater than those of non NBC teachers. The school district in this study paid bonuses above and beyond those provided by the state bonus policy. In 2008 2009, the school district paid a bonus amount of $4,500.00 to each of the thirty six elementary NBC teachers in this study. This was a total cost expenditure fo r the district of $162,000.00. In 2009 2010, there were eighteen teachers paid this bonus by the district which was a total cost expenditure of $81,000.00 for the district. In 2010 2011,

PAGE 86

86 the $4,500.00 bonus was paid to twenty six elementary teachers in t his study which was a total cost expenditure of $117,000.00 for the school district. Over the three year period, the total cost to the district was $360,000.00. The findings using the value added scores for these teachers were significant and indicate th at the positive impact on student results was not due to chance and could be attributed to the effectiveness of the NBC teachers. The value added scores were not directly related to the cut levels required in the school grading criteria. The findings usi ng only the school grading criteria would lead to the conclusion that the expenditure by the district was not a good use of funds. Further, it is important to note that this study was limited to impact on student achievement in mathematics in fourth and fifth grade only. Inclusion of additional grade levels and subjects may yield different results. The results and conclusions drawn from this study are also complicated by the fact that in this school district, these high poverty elementary schools benefi ted from additional staffing in the form of school wide subject specialists who provided additional instructional support to students in small groups as well as coached teachers in instructional best practices. Student achievement results could have been impacted by the effort of these individuals and there may exist spillover from their combined effort in the results attributed to the teachers in this study. An analysis of the teaching assignments of NBC teachers in these schools reveals that there were an additional twenty nine NBC teachers with school wide impact from the 2008 2009 data set, an additional twenty five NBC teachers with school wide impact from the 2009 2010 data set, and an additional twenty seven NBC teachers with school wide impact from the 2010 2011 data set. These teachers were not clearly assigned to students, so it is not

PAGE 87

87 possible to determine how their effectiveness spills over into the overall results. Results are consistent for comparison, however, as there was a similar distrib ution of the school wide resources in each of the schools included in the study. It should also be assumed that these school wide resources impacted the National Board Certified teacher and non National Board Certified teacher in similar fashion. It is i mportant to note this significant number of NBC teachers in these school sites prior to making conclusions about the cost effectiveness of this bonus program. These school wide specialists were paid for holding the NBC credential and add a cost of $130,50 0.00 for 2008 2009, $112,500.00 for 2009 2010, and $121,500.00 for 2010 2011. In this school district, there were 144 Elementary schools. Schools that had a student enrollment where 90 % or greater qualify for free and reduced lunch were considered rena issance schools. 19.44 % (twenty eight) of the elementary schools in this district were designated as renaissance schools. National Board Certified teachers at each of these schools received a $4,500.00 bonus each year for holding the NBC credential. The purpose for providing bonus monies was to recruit effective teachers to teach in high poverty, high need schools. In 2008 2009, there were a total of 752 National Board Certified teachers teaching in the school district. 444 of these taught in Elementar y school settings. 21.97 % (eighty) of the Elementary teachers with NBC credentials taught in renaissance schools. In 2009 2010, there were a total of 741 National Board Certified teachers teaching in the school district. 439 of these taught in Elementar y school settings. 19.95 % (seventy three) of the Elementary teachers with NBC credentials taught in renaissance schools. In 2010 2011, there were a total of 731 National Board Certified teachers teaching in the school district. 452 of these taught in

PAGE 88

88 El ementary school settings. 19.58 % (seventy four) of the Elementary teachers with NBC credentials taught in renaissance schools. This data indicates that the district is able to recruit a proportion of NBC teachers to renaissance schools equal to the propo rtion of elementary schools that are renaissance schools themselves. Though this proportion is similar, the inverse fact remains: an average of 80% of the Elementary NBC teachers in the district do not teach in elementary renaissance schools. The questio n that remains is whether this policy for recruiting national Board Certified teachers to these high needs schools is cost effective when measured by student achievement results. The percent age of Elementary NBC teachers was similar across all three years and the proportion of NBC teachers that teach in renaissance schools was also similar across the three years. This could indicate that the teachers remained in the school setting and that the bonus money was effectively retaining them to teach in these s ettings. It is difficult to draw this conclusion with power as it is impossible to tell whether teachers remained in the same school or position over the three year period. There was a complicating factor in this analysis. Beginning in the school year 2 009 2010, this school district implemented a new teacher evaluation system that required hiring peer evaluators. A large number of NBC teachers were hired for these new positions, and many were NBC Elementary teachers who previously taught in renaissance schools. These teachers were on leave from the school they were teaching in and will return after the temporary assignment ends. With a competing incentive for teachers, it is impossible to draw conclusions about the effectiveness of the bonus payment fo r NBC teachers to teach in Elementary renaissance schools. More

PAGE 89

89 detailed analysis is required to determine the cost effectiveness of this bonus plan expenditure. Recommendations for F uture P olicy 1. Payment of bonus monies to teachers who hold National Boar d Certification credentials should be made with intent to link the receipt of the monies with specific demonstrated practice and student achievement requirements in order to determine the cost effectiveness of the expenditure. 2. School districts need to defi ne how they intend to measure student learning and what measures of student achievement will be used. It is important to understand the limits of each measure as well as the ability to measure intended outcomes. 3. Due to individual variance in performance, bonus monies should not be provided for holding a credential alone. 4. School districts should attend to the contribution of NBC teachers with school wide impact when determining the overall effectiveness of a bonus payment for NBC teachers. The spillover from the school wide NBC teacher can impact the achievement of students in all classes in the school. 5. School districts should attend to the lack of random assignment of students to NBC teachers prior to comparing student achievement results for the purpose of decision making. 6. School districts should pay close attention to the sample size and respective impact on means and distributions of scores prior to decision making. 7. School districts will need to examine data related to school culture and strength of le adership along with retention and results data in order to make accurate decisions about the effectiveness of the policy put in place. Recommendations for F uture R esearch 1. Future research incorporating the results of multiple content areas would strengthen the conclusions related to the effectiveness of NBC teachers. 2. Continued research incorporating larger data sets so that the distribution of teachers and the random assignment of students can be examined to inform this body of research. 3. Continued research c omparing student achievement results as measured by attainment models versus value added models.

PAGE 90

90 APPENDIX CITED CONSTITUTIONS, STATUTES, AND DISTRI CT CONTRACT United States Codes 20 U.S.C. §§ 6301 et seq. ( 1965 ) 20 U.S.C. §§ 6301 et seq (2002) Florida State Statutes Fl. Stat. § 1008.22 Fl. Stat. § 1008.34 Fl. Stat. § 1012.34 Fl. Stat. § 1012.72 (2002) Fl. Stat. § 1012.225 et. seq. (2007) HCPS District Policy HCPS. Differentiated Pay: Contract Section 21.4.10. Hillsborough Classroom Teachers Association Inc Teacher Contract 2010 2013. Hillsborough C ounty Public Schools. Hillsborough, FL. 2010.

PAGE 91

91 WORKS CITED Carnegie Task Force on Teaching as a Profession. New York: Carnegie Forum on Education and the Economy, 1986. Recruiting and Retaining National Board Certified Teachers for Hard to Staff Low Performing schools Southwest Center for Teaching Quality. Chapel Hill, North Carolina (2005). [online] Retrieved May 5, 2010 from http://www.teachingquality.org/legacy/RecruitRetainHTSS.pdf Added Educational Testing Service. (2005) [online] Retrieved May 17, 2012 from http://www.ets.org/Media/Research/pdf/PICVAM.pdf A Corpora tion (2004). [online] Retrieved May 5, 2010 from: http://www.cna.org/documents/CavaluzzoStudy.pdf Achievement: Lon Economics of Education Review 26, (2007): 673 682. [online] Retrieved May 5, 2010 from http://www.nber.org/papers/w13617.pdf?new_window=1 Cu nn Value Added Assessment of Teacher Quality as an Alternative to the National Board for Professional Teaching Standards: What Value Added Models in Education: Theory and Applicati ons. Maple Grove, MN: Jam P ress, 2005. [online] Retrieved May 5, 2010 from http://www.education consumers.com/Cunningham Stone.pdf Darling Educational Researcher, 31(9). 2002 [online] Retrieved May 5, 2010 from http://edr.sagepub.com.lp.hscl. ufl.edu/content/31/9/13.full.pdf+html [online] Retrieved May 5, 2010 from http://schoolgrades.fldoe.org/pdf/0910/AYPTAP2010.pdf [online] Retrieved May 5, 2010 from http://schoolgrades.fldoe.org/pdf/0809/2009SIRTAP.pdf [online] Retrieved May 5, 2010 from http://schoolgrades.fldoe.org/pdf/0910/2010SchoolGradesTAP.pdf

PAGE 92

92 [online] Retrieved April 9, 2011 from http://schoolgrades.fldoe.org/pdf/1011/SchoolGradesTAP2011.pdf May 5, 2010 from http://www2.ed.gov/programs/teacherqual/hqtplans/fl.doc Retrieved May 5, 2010 http://www.fldoe.org/etp/ http://www.fldoe.org/etp/bonuses.asp Florida Department of Commissioner of Education From the Student Growth Implementation Committee [online] Retrieved June 4, 2012 from http://www.fldoe.org/committees/pdf/SummaryFinalRecommendation.pdf May 5, 2010 from http://www.leg.state.fl.us/statutes/index.cfm?mode=View % 20Statutes&SubMenu =1&App_mode=Display_Statute&Search_String=school+accountability&URL=C H1008/Sec22.HTM Florida Legislatur [online] Retrieved May 5, 2010 from http://www.leg .state.fl.us/statutes/index.cfm?mode=View % 20Statutes&SubMenu =1&App_mode=Display_Statute&Search_String=merit+award+program&URL=C H1012/Sec225.HTM Board Certified Teachers as D Doctoral Dissertation, University of Central Florida, 2010. Urban Institute (2004). [online] Retrieved May 5, 2010 from: http://www.urban.org/publications/410958.html Center for Reinventing Public Education (2008). [online] Retrieved May 17, 2012, from http://www.crpe.org/cs/crpe/download/csr_files/wp_crpe5_badclass_nov08.pdf off Between Child Quantit Journal of Political Economy 1 (100). February 1992. [online] Retrieved May 5, 2010 from http://hanushek.stanford.edu/sites /default/files/publications/Hanushek % 201992 % 20JPE % 20100 % 281 % 29.pdf Certified Teachers on Student National Center for Analysis of Longitudinal Data in Educational

PAGE 93

93 Research (January 13, 2007) [online]. Available from http://www.nbpts.org/UserFiles/File/Harris_Sass_Final_2007.pdf CALDER. (200 9). [online] Retrieved May 17, 2012, from http://www.urban.org/UploadedPDF/1001431 what makes for a good teacher.pdf Equity and Opportunity 60(4) 2003. [online] Retrieved M ay 5, 2010 from http://www.edpal.com/courses/upload/ELL_002/Toward % 20A % 20Fair.pdf HCPS. Differentiated Pa y: Contract Section 21.4.10. Hillsborough Classroom Teachers Association, Inc. Teacher Contract 2010 2013. Hillsborough County Public Schools, Hillsborough, FL. 2010. [online] retrieved May 5, 2010 from http://www.hillscta.org/Welcome_files/teachercontract.pdf HCPS http://publicaffairs.m ysdhc.org/files2010 11/FACTSBrochureFINAL.pdf 2012 Merit Award Program (MAP): Teacher Performance Assessment e] Retrieved May 5, 2010 from http://www.sdhc.k12.fl.us/SSFPDiv/Salary % 20Differential % 202010.11A.pdf Board Certified Education Policy Analysis Archives 13(18) 2005: 1 50. [online] Retrieved May 5, 2010 from http://epaa.asu.edu/ojs/article/view/123/ 249 Education Policy Analysis Archives 15(7) 2007: 1 30. [online] Retrieved May 5, 2010 from http://epaa.asu.edu/ojs/article/view/55/181 on Student Achievement The National Academies Committee of the Evaluation of the Impact of Teacher Certification by NBPTS. (February 28, 2007) [online] Retrieved May 5, 2010 from http://www7.nationalacademies.org/bota/NBPTS MTG4 Sass paper.pdf Added Measures in Hillsborough County Harvard University 2012. Value Added Models: Analytic Issues Research Coun cil and the National Academy of Education, Board on Testing and Accountability Workshop on Value Added Modeling, Washington, DC. (2008).

PAGE 94

94 [online] Retrieved May 17, 2012, from http://www7.nationalacademies.org/bota/VAM_McCaffrey_Lockwood_Paper.pdf Education Finance and Policy. Fall 2009, Vol. 4, No. 4, Pages 572 606. [onli ne] Retrieved May 17, 2012, from https://www.utla.net/system/files/mccaffrey_study.pdf McColsk e y, W., Stronge, J. H., Ward, T. J., Tucker, P. D., Howard, B. & Lewis, K. eness, Student Achievement, and National Board Certified Teachers National Board for Professional Teaching Standards Arlington, VA. 2005 [online]. Available from http://www.nbpts.org/UserFiles/File/Teacher_Effectiveness_Student_Achieveme nt_and_National_Board_Certified_Teachers_D_ _McColskey.pdf Doctoral Dissertation, University of Oklahoma, 2010. Retrieved May 5, 2010 from http:/ /www.nbpts.org/resources/nbct_directory May 5, 2010 from http://www.nbpts.org/about_us/mission_an d_history/history [online]. Retrieved May 5, 2010 from http://www.nbpts.org/the_standards/the_five_c ore_propositio nline]. Retrieved May 5, 2010 from http://www.nbpts.org/userfiles/File/Profil es/FL_State_Profile_2010_certday.pdf [online] Retrieved May 5, 2010 from h ttp://www.nbpts.org/userfiles/File/Profiles/FL_State_Profile_2010_certday.pdf Retrieved May 5, 2010 from http://www.nbpts.org/about_us/events_and_photos/2011_student_learning_st http://www.nbpts.org/UserFiles/File/what_teachers.pdf http://www.nbpts.org/resources/research/impact_of_certification#bottom

PAGE 95

95 Achievement: How Do Teachers Measure Up? [ online]. Retrieved May 5, 2010 from http://www.nbpts.org/about_us/events_and_photos/2011_student_learning_st National Commission on Excellence in Education. A Nation at Risk: The Imperatives for Education Reform. Washington D.C.: U.S. Government Printing Office, 1983. [online] Retrieved May 5, 2010 from http://www2.ed.gov/pubs/NatAtRisk/index.html National S American School Boards Journal, 193 (3). 2006. Rivkin, S., Hanushek, E., & K Achievement Econometrica 73(2). 2005. [online] Retrieved May 5, 2010 from http://www.economics.harvard.edu/faculty/staiger/files/HanushekRivkinKain % 2B Ecta % 2B2005.pdf Comparison of the Ef fects of NBPTS Certified Teachers with Other Teachers on the Rate of Student Academic Progress National Board for Professional Teaching Standards Arlington, VA. 2005. [online] Retrieved May 5, 2010 from http://www.nbpts.org/UserFiles/File/SAS_final_NBPTS_report_D_ _Sanders.pdf Sato, M., Chung, W., & Darling Practices Through Professional Development: The case of Nationa l Board American Educational Research Journal vol. 4 5(3), 2008: 669 700. Retrieved M ay 5, 2010 from http://www.stanford.edu/~rchung/AERJ316955 % 20NBPTS % 20Study % 204.15.08 .pdf Student Achievement. Doctoral Dissertation. University of South Carolina, 2003. Added Achievement Gains of NBPTS Certified T eachers in Tenness ay 5, 2010 from http://www.education consumers.com/oldsite/briefs/stoneNBPTS.shtm Stronge, J., Ward, T., Tucker, P., Hindma n, J., McColsk e Board Certified Teachers and Non National Board Certified Teachers: Is There a Difference in Teacher Effectiveness and Student Achievement Personnel Evaluation in Education 20, 2007: 185 210. Retrie ved May 5, 2010 from http://strongeandassociates.com/files/Stronge % 20et % 20al % 202007_National % 2 0Board % 20Certified % 20Teachers.pdf

PAGE 96

96 U.S. [o nline]. Retrieved M ay 5, 2010 from http://www.fns.usda.gov/cnd/Governance/notices/iegs/Elig ibilityManual.pdf [o nline]. Retrieved May 5, 2010 from http://www2.ed.gov/policy/elsec/leg/esea02/107 110.pdf U.S. De Retrieved May 5, 2010 from http://www2.e d.gov/policy/elsec/leg/blueprint/blueprint.pdf Retrieved May 5, 2010 from http://www2.ed.gov/policy/el sec/leg/esea02/107 110.pdf Vandevoort, L., Amrein Education Policy Analysis Archives, 12 (46). (2004) [online]. Retrieved May 5, 2010, from http://epaa.asu.edu/epaa/v12n46/v12n46.pdf Achievement Results of Third Grade Students in a Local Central Florida School toral Dissertation, University of Central Florida, 2008.

PAGE 97

97 BIOGRAPHICAL SKETCH Anna L. Brown resides in Tam pa, Florida. She received her b e lementary e ducation from the University of Florida in 1989, her e lementary e d ucation from the University of Florida in 1990, and her certification in e duca tional l eadership from the University of South Florida in 2000. She began her career in e ducation in 1990 and has taught students in a variety of settings encompassing each grad e level in grades K 8. In 2001, she became an Assistant Principal for Elementary Instruction. In 2004, she became an Elementary School Principal. She is currently the Director for Assessment and Performance Management for the Hillsborough County Public School System.