EXAMINING RELATIONSHIPS AMONG PRE -SERVICE PROFESSIONALSÕ COURSEWORK, KNOWLEDGE, BELIEFS, AND PRACTICES RELATED TO CHILDRENÕS SOCIAL -EMOTIONAL DEVELOPMENT By Tamesha N. Harewood A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Human Development and Family Studies - Doctor of Philosophy 2016 ABSTRACT EXAMINING RELATIONSHIPS AMONG PRE -SERVICE PROFESSIONALSÕ COURSEWORK, KNOWLEDGE, BELIEFS, AND PRACTICE S RELATED TO CHILDRENÕS SOCIAL -EMOTIONAL DEVELOPMENT By Tamesha N. Harewood ChildrenÕs early social -emotional development is important to their lifelong learning, academic success, and school readiness. With greater focus on high -quality earl y care experiences and increasing numbers of children enrolled in early education and care settings, the role of early childhood teachers has become more important to childrenÕs development. However, many children enter kindergarten lacking social -emotiona l skills and many early childhood teachers report being unprepared to support childrenÕs social -emotional development. Institutions of higher education provide opportunities for teachers to improve their knowledge and skills, and meet standards in the earl y childhood education field , which expect teachers to have knowledge of childrenÕs development and strategies to support childrenÕs learning and development across domains. Yet , research has only examined basic links between education and classroom quality, limiting our understanding of what educational experiences contribute to teachersÕ development of knowledge and skills to work with young children. One current impediment to t his line of research is the lack of existing tools to assess teacher knowledge. In the current research , a tool was created to assess pre -service teachersÕ knowledge of social -emotional development and its psychometric properties tested in Study 1 , this to ol was then used to answer more substantive questions related to teachersÕ coursework, knowledge, beliefs, and practices in Study 2 . These were cross -sectional studies using a convenience sample of undergraduate students at a Midwestern university in the U nited States. Participant s included 160 students enrolled in courses through a Human Development and Family Studies department . In Study 1 , the Knowledge of Social -Emotional Milestones and Support Strategies (K -SEMS) tool was determined as hav ing two indices (the Knowledge of Social -Emotional Milestones index and the Knowledge of Social -Emotional Support Strategies Index), which were valid and moderately reliable. Study 2 found pre -service teachers who took more domain -specific coursework had more accurate knowledge of preschoolersÕ social -emotion al milestones and strategies, as well as higher endorsements of beliefs about expressing and support emotions in the classroom. Observing a subsample o f the larger 160 participants, S tudy 2 also found p re-service teachers (n = 33) with more accurate knowledge and higher endorsements of instruction/modeling beliefs about emotions more frequently used developmentally supportive social -emotional practices in the classroom. These findings have implications f or research, policy, and practice related to the education and development of early childhood teachers. iv !ACKNOWLEDGMENTS If God brought you to it, He will bring you through it! This has been my motto and my belief throughout my graduate career, so I would first like to acknowledge God for bringing me through th is process and placing the people I needed along the way in my life. What a process and experience graduate school has been. I have grown in so many ways and have learned many things from many wonderful people. I would like to thank you Claire, my mentor, and cheerleader in so many ways, for your relentless support and for caring so much about my success, and my professional development. You truly are a rare gem, with a beautifu l mind. I will forever cherish my time working with you as a student and look forward to working with you as a colleague. Your example of mentorship should be published everywhere, but until it is, I will take the experience with me in my future career and prayer I can be to my students what you were for me, a place of hope and encouragement. Thank you! To my other committee members Ð Dr. McNall, Dr. Fi tzgerald, and Dr. Brophy -Herb Ð thank you for your guidance, for your strong and valued feedback, and for not letting me try to examine the whole world in this dissertation. I would also like to thank Dr. Johnson , who has cheered me on and supported me in so many ways. You truly are an awesome person. To my loving family and friends, I cannot even begin to sa y how eternally grateful I am to all of you. You have sacrificed so much for me to accomplish this goal. To my husband Karl, thank you for being my rock and cheerleader. You picked up the slack with our children when I was too tired to be mommy and you mad e sure we did not lose all of our family time to school work. Thank you for keeping me focused and balanced and not letting me work myself to death. I love you forever. v !To my wonderful children, Jordan and Jesse, thank you for coming into our lives when you did. You made me prioritize and manage my time more effectively. I love you both and look forward to spending a lot more time with you my darlings. To my mom, Margaret Watson, my dad, sisters and brother, my nieces and nephews, I love you and thank you for your continuous prayers, checking in on me, and support from far away. You didnÕt understand this journey but you always believed I could do it. Telisa, I hope I was a good example for you; I am looking forward to your graduation next. To Walton and K arlene Phillips, I do not know how I would have done this without your sacrifice. I am forever indebted to you and cannot tell you enough how much I appreciate a nd love you both. Thank you for your prayers, encouragement, and love. Glenn, thank you for making me even think about pursuing a Ph.D. and for always being there and making me laugh. Tina my dear, you are next. I couldnÕt have asked for a sweeter more loving sister -in-law. See you on the other side , honey!! Last but certainly not least; I would li ke to thank my APOC church family, too many names to mention. I am so blessed to have each of you in my life. Thank you all for your unending love. Thank you to tho se who sent me care packages, message s of encouragement along the way, hugged me until I cou ldnÕt breath e, and definitely for your many thoughtful prayers. I love you all! To all my other family in Barbados , New York, Maryland, and everywhere, thank you, thank you, thank you! vi !TABLE OF CONTENTS LIST OF TABLES ix LIST OF FIG URES x CHAPTER 1 1 Justification for Current Research 1 REFERENCES 7 CHAPTER 2: STUDY 1 Development of the Knowledge of Social -Emotional Milestones and Support Strategies (K -SEMS, Preschool Version) Tool 12 Early Childhood Education and Practices 13 Knowledge of Social -Emotional Development 13 Current Measures Related to Knowledge of Social -Emotional Development 15 Developing the Knowledge of Social -Emotional Milestones and Support Strategies Survey (K -SEMS) 16 Method 17 Procedures 17 Overview 17 Item d evelopment 18 Knowledge of m ilestones 18 Knowledge of s trategies 18 Validation procedures 19 Data collection p rocedures 20 Retest 21 Participants 21 Observation subsample 21 Test-retest s ubsample 23 Measures 23 Coursework with social -emotional c ontent 23 Classroom practices that support social -emotional d evelopment 24 Analyses and Results 25 Exploratory Factory Analyses 25 Fit c riteria 25 Model s election 26 Knowledge of preschoolersÕ social -emotional m ilestones 26 Knowledge of social -emotional support s trategies 29 Validity 29 Content v alidity 29 Concurrent v alidity 30 Predictive validity 31 Reliability 32 Internal consistency 32 vii ! TestÐretest 32 Discussion 33 Knowledge of Social -Emotional Milestones Scale 33 Improving the reliability and validity of the K nowledge of Social -Emotional M ilestones s cale 35 Knowledge of Support Strategies Scale 36 Limitations and Recommendations 37 Conclusion 41 APPENDIX 43 REFERENCES 48 CHAPTER 3: STUDY 2 A Heuristic Model of Pre -service Early C hildhood Teacher Preparation to Support ChildrenÕs Social -Emotional Development 54 Links Between Teacher Education and Classroom Practices 57 Coursework, Knowledge, and Beliefs 59 Knowledge of Child Development 60 Beliefs About Child Development 63 Theoretical Framework 64 Methods 66 Overview 66 Procedures 66 Participants 67 Survey p articipa nts 67 Observation p articipants 67 Measures 68 Knowledge of social -emotional milestones and support s trategies 68 Teacher beliefs about e motions in the c lassroom 69 Teacher p ractices to support preschoolersÕ social -emotional development 70 Observer t raining and reliability 71 Subscales of observed s kills 71 Courses with social -emotional content 72 Preliminary Analyses 73 Results 77 Relationships Between Coursework and Knowledge and Beliefs 77 Relationships between c oursework a nd knowledge 77 Comparisons of average knowledge scores b ased on number of prior courses taken 77 Relationships between coursework and b eliefs 78 Moderation Effects of Beliefs on the Rela tionships Between Knowledge and Practices 78 Positive an d proac tive behavior m anagement 79 Knowledge of m ilestones 79 Encouraging the use of social -emotional s kills 80 Knowledge of support s trategies 80 Social -emotional g uidance 81 viii !Discussion 81 Relationships Among Coursework, Knowledge, and Beliefs 82 Relationships Among Knowledge, Beliefs , and Practices 83 Main effec ts of k nowledge on practices 83 Beliefs moderated relationships b etween knowledge and p ractices 84 Limitations 86 Conclusions 87 APPENDIX 89 REFERENCE S 92 CHAPTER 4: INTEGRATIVE CO NCLUSION 99 Implications for Practice, Research , and Policy, and Future Considerations 101 REFERENCES 103 ix !LIST OF TABLES Table 2.1 Characteristics of the Study 1 Participants ÑPercentages 22 Table 2.2 Factor Loadings for Knowledge of Social -Emotional Milestone Items (15 Items) 28 Table 2.3 Factor Loadings for Knowledge of Support Strategy Items (9 Items) 30 Table 2.4 Correlations Between K -SEMS and TSRS Subscales 31 Table 3.1 Characteristics of the Study 2 Participants ÑPercentages 68 Table 3.2 Means and Standard Deviations of Study Variables 74 Table 3.3 Correlations Among Study Variables for the Entire Sample (n = 160) 75 Table 3.4 Correlations Among Study Variables for Subsample (n = 33) 76 x !LIST OF FIGUR ES Figur e 3.1. Conceptual model 65 Figure 3.2. Instruction/model ing beliefs moderated the relationship between knowledge of milestones and behavior management practices 80 Figure 3.3. Instruction/modeling beliefs moderated the relationship between knowledge of support strategies and practices encouraging the use of social -emotional skills 81 1 !CHAPTER 1 Justification for Current Research Social -emotional (SE) development in early childhood predicts subsequent school and life success (Boyd, Barnett, Bodrova, Leong, & Gomby, 2005; Cohen, Onunaku, Clothier, & Poppe, 2005; Roman o, Babchishin, Pagani, & Koh en, 2010; Shonkoff & Phillips, 2000). Further, children who lack social and emotional skills at school entry may be subjected to peer rejection and experience conflictual or distant relationships with teachers (Berry , 2012; Miller et al. , 2005), which can further affect their academic performa nce and social adjustment (Engle, McElwain , & Lansky, 2011; Howes, 2000; Rhoades, Warren, Domitrovich, & Greenberg, 2011; Trentacosta & Izard, 2007; Vitaro, Boivin, Brendgen, Gi rard, & Dionne, 2012). A growing body of child development research and numero us reports demonstrate that the quality of early care and education (ECE) experiences is important for young childrenÕs school readiness and later outcomes (Burger, 2010; Mashburn et al., 2008). Early care and education settings provide opportunities for c hildren to receive much -needed support for their social -emotional skills, with both immediate and long -term impacts for their development, well -being, and academic success. More specifically, the most important element in ECE settings is the teacher/care p rovider and the quality of his or her interactions with children (Baker, 2006; Baker, Clark, Maier, & Viger, 2008; Burchinal, Roberts, Zeisel, Hennon , & Hooper , 2006; Hamre & Pianta, 2001; Howes, 2000; Mitchell -Copeland, Denham, & DeMulder, 1997; Pianta & Stuhlman, 2004; Rimm -Kaufman & Hamre, 2010). When the teacher -child relationship is negative it affects childrenÕs social behaviors, with children more likely to be disruptive in class, cooperate less, dislike school, and display aggressive behaviors towar ds others (Elicker & Fortner -Wood, 1995; Hamre & Pianta, 2001; Howes, 2000; Ladd, Birch, & Buhs, 1999; Ladd & 2 !Burgess, 2001; Pianta & Stuhlman, 2004). But when teachers interact with children in positive ways, these interactions foster childrenÕs positive emotional development and build a sense of security that encourages childrenÕs active engagement in the social and academic opportunities provided by the classroom (Pianta, La Paro, Payne, Cox, & Bradley, 2002). Further, research indicates that the develo pmental needs of children may best be met by teachers with specialized education and training from teacher preparation programs in child development and early childhood education (Burchinal, Cryer, Clifford & Howes, 2002; Howes, 1997). Hence, teachersÕ abi lities to support the social -emotional development of all children, even those with challenging behaviors, are crucial, but may depend in large part on the preparation they receive to provide such support. Essentially, what teachers know and do are fundame ntal factors in the quality of educational and developmental experiences they provide to young children, and can have long -lasting effects on childrenÕs subsequent development. However, the teacher education literature related to development of teachersÕ k nowledge and skills to work with young children is quite sparse, and there is very little understanding about how pre -service teachersÕ educational experiences contribute to their effective teaching and support in the classroom. Thus, educators of pre -serv ice early childhood teachers in institutions of higher education are left to assume that the education these teachers receive improves their knowledge and skills and adequately prepares them for their futur e work with young children. However, reports of ea rly care and education professionals feeling unprepared t o meet the social -emotional needs of children, especially those with difficult behavior, coupled with the high rates of preschool expulsions (Gilliam, 2005; Hemmeter , Corso, & Cheatham, 2006) suggest a need to more closely examine early childhood teachersÕ preparation, such as associations between their 3 !educational experiences and educational outcomes, to better understand which educational factors produce effective teachers of young children. Institu tions of higher education provide a context for early childhood teachers not only to gain knowledge, but to learn and practice skills for their future work with children. It is expected that early childhood programs will improve teachersÕ knowledge of chil drenÕs development in all domains and equip them with effective teaching strategies to support childrenÕs development. In the US, however, there is no single level of education required for early care and education teachers/caregivers. Different levels of formal education, from high school diplomas through masters degrees, as well as non-credit -bearing early childhood certifications (e.g., the CDA), can all be considered acceptable qualifications for individuals working directly with children in early child hood settings, depending upon the specifications of the individual programs and their funding sources. Given this wide range of acceptable levels of training, it is not surprising that debates about the relevance of education to teachersÕ practices continu e to arise. Research in this area has produced mixed findings, with recent meta analyses refuting the effects of education on teacher outcomes (Early et al., 2006; Early et al., 2007), which does not contribute clarity to this debate. Nonetheless, research in this area generally suggests that more formal education -- used in both research and hiring practices as a proxy for knowledge and skills -- is associated with overall classroom quality and linked to better cognitive and social outcomes for children (A rnett , 1989; Bowman , Donovan, Burns, & the Committee on Early Childhood Pedagogy !, 2000; Burchinal et al. , 2002; Cassidy, Buell, Pugh -Hoese, & Russell, 1995; Howes, 1997 ; Whitebook, 2003; Whitebook et al., 1989). But what pre -service teachers know about child development, how they obtain this knowledge, and how this knowledge influences their classroom practices remain substantial gaps in the research on early care and education, despite the generally 4 !accepted idea that teachersÕ knowledge of child development is important ( National Association for the Education of Young Children [ NAEYC], 2009). While the examination of the relationship between knowledge and practices is important, it is incomplete without consider ing the influence of beliefs. Beliefs are defined as concepts individuals hold to be true that are based on personal judgment and evaluation regardless of an evidence -base (Pajares, 1992). Research indicates that teachersÕ beliefs predict their classroom behaviors and affect their planning and execution of classroom activities and curriculum, and interactions with children (Faulkner -Schneider, 2005; Hur, Buettner, & Jeon, 2013; Stipek & Byler, 1997). Further, earlier researchers suggest that teachersÕ instr uctional judgments and classroom decisions are based less on their knowledge and more on their beliefs about a topic or teaching area (Nisbet t & Ross, 1980; Shavelson, 1983). However, only a few studies have incorporated both knowledge and beliefs as predi ctors of teachersÕ practices. Even fewer have addressed the role of domain -specific beliefs, such as those about early math, literacy, or social -emotional development, related to knowledge and teacher practices. Findings indicate domain -specific beliefs ar e important because they may influence teacher expectations of and behaviors toward children, thereby determining the teaching strategies used by teachers related to the subject (Schirmer, Casbon, & Twiss, 1997); these findings further support the inclusio n of beliefs as an important factors to consider when examining associations between teachersÕ knowledge and practices. Thus, a goal of the current study was to determine whether an aspect of teachersÕ educational experiences Ñcollege coursework Ñinfluences teachersÕ domain -specific knowledge and beliefs (about the social -emotional domain), whether teachersÕ domain -specific knowledge is associated with their domain -specific classroom practices, and what role, if any, beliefs play in the relationship between k nowledge and practices. 5 !One current barrier in this line of research is the lack of adequate measurement tools to assess teachersÕ domain -specific knowledge related to social -emotional development. Instruments measuring pre -service teachersÕ knowledge rele vant for their future work performance are limited in the field of early childhood. Thus far, there have been efforts to develop instruments about teachersÕ knowledge of early literacy, science, technology, and math (Blomeke, Buchholtz, Suhl, & Kaiser, 201 4), but no known work to develop a measure appropriate for measuring teachersÕ knowledge of childrenÕs social -emotional development or the teaching practices which support it. There is one measure of adultsÕ knowledge of child development used in research and linked to adultsÕ education, the Knowledge of Infant Development Inventory (KIDI; M acPhee, 1981). However, the relevance of this measures is limited for preschool teachers because although it covers a broad range of knowledge across numerous domains in addition to social -emotional d evelopment, it focuses on parenting and the home context, and on children between birth and 2 years old; it does not address the classroom setting or teaching practices, and does not have questions relevant to preschool aged children. Thus, the field of early child education, and specifically teacher preparation, is lacking an adequate measure of pre -service professionalsÕ knowledge of young childrenÕs social -emotional development and related classroom practices. Another goal of the current study was to develop an instrument to measure pre -service early childhood teachersÕ knowledge of preschoolersÕ social -emotional development and strategies to support SE development. This measure will be used to gather data to address substan tive questions about the relations among pre -service teachersÕ courses, knowledge, beliefs, and skills. The present study aimed to resolve several limitations in the current scientific literature on the professional development of early childhood educators by (1) developing an instrument to 6 !measure knowledge of childrenÕs social -emotional development and strategies to support this development, and (2) testing a theoretical model of relationships among coursework, knowledge, beliefs, and practices. 7 ! REFERENCES 8 !REFERENCES Arnett, J. (1989). Caregivers in day -care centers: Does training matter ? Journal of Applied Developmental Psychology, 10 , 541Ð552. Baker, J. A. (2006). Contributions of teacher Ðchild relationships to positive school adjustment during elementary school. Journal of School Psychology, 44 , 211Ð229. Baker, J. A., Clark, T. P., Maier, K. S., & Viger, S. (2008). The differential influence of instructional context on the academic engagement of students with behavior problems. Teaching and Teacher Education, 24 ,1876Ð1883. Blomeke, S., Buchholtz, N., Suhl, U., & Kaiser, G. (2014). Resolving the chi cken -or-egg causality dilemma: The longitudinal interplay of teacher knowledge and teacher beliefs. Teaching and Teacher Education, 37 , 130Ð139. Boyd, J., Barnett, S., Bodrova, E., Leong, D., & Gomby, D. (2005). Policy report : Promoting childrenÕs social emotional development through preschool education. National Institute for Early Education Research (NIEER) , New Brunswick: NJ . Burchinal, M. R., Cryer, D., Clifford, R. M., & Howes, C. (2002). Caregiver training and classroom quality in child care centers. Applied Developmental Science, 6 , 2Ð11. Burchinal, M., Roberts, J. E., Zeisel, S. A., Hennon , E. A., & Hooper , S. (2006).!Social risk and protective child, parenting, and child care factors in early elementary school years . Parenting: Science and Practice, 6 , 79Ð113. Burger, K. (2010). How does early childhood care and education affect cognitive development? An international review of the effects of early interventions for children from different social backgrounds. Early Childhood Research Quarter ly, 25 , 140Ð165. doi: 10.1016/j.ecresq.2009.11.001 Cassidy, D., Buell, M., Pugh -Hoese, S., & Russell, S. (1995). The effects of education on child care teachersÕ beliefs and classroom quality: Year one evaluation of the TEACH early childhood associate degre e scholarship program. Early Childhood Research Quarterly, 10, 171Ð183. Cohen, J., Onunaku, N., Clothier, S., & Poppe, J. (2005 , September ). Helping young children succeed: Strategies to promote early childhood social and emotional development. Washington , DC: Zero to Three and the National Conference of State Legislatures. Early, D., Bryant, D., Pianta, R., Clifford, R., Burchinal , M., Ritchie, S., É Barba rin, O. (2006) . Are teacher education, major, and credentials related to classroom quality and childre nÕs academic gain in pre -kindergarten? Early Childhood Research Quarterly, 21 , 174Ð195. 9 !Early, D. M., Maxwell, K. L., Burchinal, M., Alva, S., Bender, R. H., Bryant, D., É Zill, N. (2007). TeachersÕ education, classroom quality, and young childrenÕs academ ic skills: Results from seven stud ies of preschool programs. Child Development, 78 , 558Ð580. Elicker , J. & Fortner -Wood, C. (1995). Research in review: Adult Ðchild relationships in early childhood settings. Young Children, 51 , 69Ð78. Engle, J. M., McElwain, N. L., & Lasky, N., (2011). Presence and quality of kindergarten childrenÕs friendships: Concurrent and longitudinal associations with child adjustment in the early school years. Infant and Child Development, 20 , 365Ð386. Faulkner -Schneider, L. (2005). Child care teachersÕ attitudes, beliefs, and knowledge regarding science and the impact of early childhood learning opportunities . (Unpublished thesis ). Oklahoma State University , Norman: Oklahoma. Gilliam, W. (2005). Prekindergartene rs left behind: Expulsion rates in state prekindergarten systems. New York , NY : The Foundation for Child Development. Hamre, B. K., & Pianta, R. C. (2001). Early teacher Ðchild relationships and the trajectory of childrenÕs school outcomes through eighth gr ade. Child Development, 72, 625Ð638. Hemmeter, M. L., Corso, R., & Cheatham, G. (2006). Issues in addressing challenging behaviors in young children: A national survey of early childhood educators . Paper presented at the Conference on Research Innovat ions in Early Intervention, San Diego, CA. Howes, C. (1997). ChildrenÕs experiences in center -based child care as a function of teacher background and adult Ðchild ratio. Merrill -Palmer Quarterly, 43 , 404Ð425. Howes, C. (2000). Social -emotional classroom cl imate in child care, child Ðteacher relationships and childrenÕs second grade peer relations. Social Development, 9 , 191Ð204. Ladd, G. W., Birch, S. H., & Buhs, E. S. (1999). ChildrenÕs social and scholastic lives in kindergarten: Related spheres of influence? Child Development, 70 , 1373Ð1400. Ladd, G. W., & Burgess, K. B. (2001). Do relational risks and protective factors moderate the linkages between childhood aggression and early psychological and school adjustment? Child Development, 72 , 1579Ð1601. Mashburn, A., Pianta, R., Hamre, B., Downer, J., Barbarin , O., Bryant, D., É Howes, C. (2008). Measures of classroom quality in prekindergarten and childrenÕs development of academic skills. Child Develoment, 79 , 732-749. MacPhee, D. (1981). Knowledge of Infant De velopment Inventory . Princeton, NJ: Educational Testing Service. Miller, A. L., Gouley, K. K., Seifer, R., Zakriski, A., Eguia, M., & Vergnani, M., (2005). Emotion knowledge skills in low -income elementary school children: Associations with social status and peer experiences. Social Development, 14 , 637Ð651. 10 !Mitchell -Copeland, J., Denham, S., & DeMulder, E. (1997). Q -Sort assessment of child Ðteacher attachment relationships and social competence in the preschool. Early Education and Development, 8 , 27Ð39. National Association for the Education of Young Children [NAEYC]. (2009). Position statement: Developmentally appropriate practice (3rd ed.). Washington, DC: Author. Nisbett, R., & Ross, L. (1980). Human inferences: Strategies and shortcomings of social judgment. Englewood Cliffs, NJ: Prentice -Hall. Pajares, M. F. (1992). Teacher beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62 , 307Ð332. Pianta, R. C., La Paro, K. M., Payne, C., Cox, M. J., & Bradley, R. (2002). The relation of kindergarten classroom environment to teacher, family, and school characteristics and child outcomes. Elementary School Journal, 102 , 225Ð238. Pianta, R. C., & Stuhlman, M. W. (2004). Conceptualizing risk in relational terms: Associations among the quality of child -adult relationships prior to school entry and childrenÕs developmental outcomes in first grade. Educational and Child Psychology, 21 , 32Ð45. Rhoades, B. L., Warren, H. K., Domitrovich, C. E., Greenberg, M. T. (2011). Examining the link between preschool social -emotional competence and first grade academic achievement: The role of attention skills. Early Childhood Research Quarterly, 26 , 182Ð191. Rimm -Kaufman, S. E., & Hamre, B. K. (2010). The role of psychological and developmental science in efforts to improve teacher quality. Teachers College Record, 112, 2988Ð3023. Roman o, E., Babchishin, L., Pagani, L., & Koh en, D. (2010). School readiness a nd later achievement: Replication and extension using a nationwide Canadian survey. Developmental Psychology, 46 , 995Ð1007. doi:10.1037/a0018880 Schirmer, B., Casbon, J., & Twiss, L. (1997). Diverse lear ners in the classroom: Teacher b eliefs about learning: What happens when the child doesnÕt fit the schema? The Reading Teacher, 50 , 690Ð692. Shavelson, R. (1983). Review of research on teachersÕ pedagogical thoughts, judgments, plans, and decisions. Elementary School Journal, 83, 392-413. Shonkoff, J. P., & Phillips, D. A. (Eds.). (2000). From ne urons to neighborhoods: The science of early childhood development . Washington, DC: National Academy Press. Stipek, D., & Byler, P. (1997). Early childhood education teachers: Do they practice wh at they preach? Early Childhood Research Quarterly, 12 , 305Ð325. Trentacosta, C., & Izard, C. (2007). Kindergarten childr enÕs emotion competence as a predictor of their academic competence in first grade. Emotion, 7 , 77Ð88. 11 !Vitaro, F., Boivin, M., Brendgen, M., Girard, A., & Dionne, G. (2012). Social experiences in kindergarten and academic achievement in grade 1: A monozygotic twin difference study. Journal of Educational Psychology, 104, 366Ð380. Whitebook, M. (2003). Early education quality: Higher teacher qualifications for better learning environments ÑA review of literature. Berkeley, CA: Institute of Industrial Relations, University of California. Whitebook, M., Gomby, D., Bellm, D., Sakai, L., & Kipnis, F. (2009). Prepa ring teachers of young children: The current state of knowledge, and a blueprint for the future. Berkeley, CA: Center for the Study of Child Care Employment. 12 !CHAPTER 2 : STUDY 1 Development of the Knowledge of Social -Emotional Milestones and Support Strategies (K -SEMS, Preschool Version) Tool The last two decades of research have made it clear that childrenÕs social -emotional development is important to their school readiness and success ( Denham et al. , 2013; Eisenberg, Valiente, & Eggum, 2010; McClelland et al., 2007; Raver, 2002). Early care and education settings are perceived as preparation for Kindergarten school -readiness (Hatcher, Nuner, & Paulsel, 2012). These settings provide opportunities for children to receiv e much -needed support for their social -emotional skills, with both immediate and long -term impacts for their development, well -being, and academic success. Further, the key elements in these settings are the characteristics of the teacher/care provider and the quality of his or her interactions with children (Baker, 2006; Baker, Clark, Maier, & Viger, 2008; Burchinal, Roberts, Zeisel, Hennon , & Hooper , 2006; Hamre & Pianta, 2001; Howes, 2000; Mitchell -Copeland, Denham, & DeMulder, 1997; Pianta & Stuhlman, 2004; Rimm -Kaufman & Hamre, 2010). Studies have shown that negative teacher -child relationships and interactions predict childrenÕs poor social behavior such as disruptions in class, lack of cooperation, disliking school, and aggressive behavior towards ot hers (Elicker & Fortner -Wood, 1995; Hamre & Pianta, 2001; Howes, 2000; Ladd, Birch, & Buhs, 1999; Ladd & Burgess, 2001; Pianta & Stuhlman, 2004 : Garner, Mahatmya, Moses, & Bolt, 2014 ). Conversely, teachers who interact with children in positive ways foster constructive social -emotional experiences for children and build a sense of security that encourages childrenÕs active engagement in the social and academic opportunities provided within the classroom (NICHD ECCRN, 1999; Pianta, La Paro, Payne, Cox, & Bra dley, 2002). But many preschool teachers report feeling unprepared to address childrenÕs challenging 13 !behaviors and effectively support their social -emotional development (Hemmeter, Corso, & Cheatham, 2006). Additionally, reports of high rates of preschool expulsions (Gilliam, 2005 ; Perry , Holland, Darling -Kuira, & Nadiv, 2011 ) seem to suggest that preschool teachers may be lacking the knowledge, skills, and/or dispositions necessary to support childrenÕs social -emotional development, particularly when child ren exhibit challenging behaviors. Early Childhood Education and Practices Federal policy initiatives and national programs requiring pre -K teachers to have bachelorÕs degrees suggest that one method to improve knowledge and skills is through education. Institutions of higher education provide a context for pre -service teachers to learn and practice such skills . But studies examining the relationships between teacher education and teacher Õs practices or overall classroom quality have revealed mixed findin gs, which have contributed to debates about the relevance of formal education to early childhood teachersÕ effectiveness in the classroom (Burchinal, Cryer, Clifford, & Howes, 2002; Early et al., 2006; Early et al., 2007) . In general these studies measure basic association s between education level and classroom quality but do not examine more distinct features of teachersÕ educational experiences and teachersÕ educational outcomes (e.g., knowledge or skills ). These inconsistent findings should not lead to t he assumption that education is not important, but rather, encourage a closer examination of early education programs in institutions of higher education to better understand the variations within studentsÕ educational experiences which may explain inconsi stencies in the links between education level and quality. Knowledge of Social -Emotional Development Professional standards set by influential agencies within the field of Early Care and Education (ECE), such as the National Association for the Education of Young Children 14 !(NAEYC), Administration for Children and Families (ACF), and Division of Early Child hood Council for Exceptional Child (DEC/CEC), indicate that in order to guide childrenÕs development and promote learning, early childhood teachers should be knowledgeable about childrenÕs typical early development in all domains and have skills to support childrenÕs development. This includes knowing about the timing of crucial developmental milestones , as well as strategies , which support the development of these milestones . Social -emotional milestones may be defined as childrenÕs age -varying abilities to express and regulate emotions and respond to social interactions with others (Giles, 2015). As children grow older , their abilities to express and regulate emotions and respond to social interactions continue to expand and become more refined. Knowing abo ut social -emotional milestones can inform teachersÕ expectations of children Õs behavior, and help teachers decide how to best support childrenÕs social -emotional behavior and development. Social -emotional support strategies, defined for this study as appro aches taken by adults to support the development of social -emotional skills, may include encouraging childrenÕs expression of emotions, managing and redirecting behavior, modeling caring and positive regards for others, labeling oneÕs own and childrenÕs em otions, and building positive relationships with children and their families (Kemple & Hartle, 1997 ). Having such knowledge may better prepare teachers to support childrenÕs social -emotional development and manage childrenÕs difficult behaviors. It is thus expected that early childhood teachers who lack knowledge of childrenÕs typical social -emotional development and the strategies to support this development are less likely to respond appropriately to childrenÕs social -emotional needs or use effective prac tices to promote their development. With standards in the field communicating the expectation that teachers of young children should have knowledge of child development and strategies to support development in 15 !order to promote childrenÕs development and l earning across domains (NAEYC, 2009), it is surprising that little research has been conducted examining teachersÕ child development knowledge and relations between knowledge and practices. Given that early childhood teachersÕ child development knowledge, let alone knowledge of the social -emotional domain , has not been measured , it is difficult to adequately assess how pre -service early childhood teachers develop skills needed to support childrenÕs social -emotional learning and development, or to determine the training and professional development needs these teachers may have. One current impediment to this line of research is the lack of adequate measurement tools to assess teachersÕ domain -specific knowledge of development and strategies related to supporting childrenÕs social -emotional development. Thus far, there have been e fforts to develop instruments about teachersÕ knowledge of early literacy, science, technology, and math (Blomeke, Buchholtz, Suhl, & Kaiser, 2014), but no published work on measures appropriate for assessing teachersÕ knowledge of childrenÕs social -emotio nal development or strategies to support it . Current Measures Related to Knowledge of Social -Emotional Development Measures of adultsÕ practices of caring for or interacting with young children , which are linked to childrenÕs social -emotional development have been created for parents, and are typically specific to the context of parenting. That is, they focus on the context of home rather than school, on responses to one child rather than children in groups, and on the many roles of parents, but not the ro les of teachers. Further, these typically cover a whole range of parenting practices and care routines which support childrenÕs growth, health, and development across domains, and do not focus largely on the social -emotional domain. Examples of such measur es include Caregiver Knowledge of Child Development Inventory (E rtem et al., 2007), Epstein High/Scope Knowledge Scale (1980), and the Knowledge of Infant Development Inventory 16 !(KIDI; M acPhee, 1981). Other measures such as the Coping with ToddlersÕ Negativ e Emotion Scale (CTNES) and the Maternal Emotional Styles Questionnaire (MESQ) are more focused on childrenÕs social -emotional development and much closer to measuring parental socialization practices or responses to childrenÕs emotions, but still do not g et at knowledge of social -emotional development. Meanwhile, measures such as the Parental Modernity Scale (Schaefer & Edgerton, 1985), the Parent Opinion Survey (Luster, Rhoades, & Haas, 1989), and the Adolescent Adult Parenting Inventory (Bavolek, 1984), while not specifically designed for teachers, include some practices, and might capture important attitudes and beliefs related to childrenÕs development, but do not measure knowledge of development or strategies for supporting development. A measure that is more focused on assessing early childhood teachersÕ knowledge related to the social -emotional domain is therefore needed. Developing the Knowledge of Social -Emotional Milestones and Support Strategies Survey (K-SEMS) As knowledge of child development and support strategies have been indicated as important factor s in teachersÕ classroom decision -making and practices, developing an instrument to measure teachersÕ knowledge of social -emotional development is central to assessing how early childhood teach ers develop skills to effectively support childrenÕs social -emotional development and learning . Such a tool could move the early childhood field forward by providing a measure of domain -specific knowledge which could gather data to inform teaching and inst ruction in early childhood programs at institutions of higher education , identify the achievements and learning needs of students in these programs , inform course development, and be a tool for research focused on the relationships between studentsÕ development of knowledge and application of skills. Thus, the aim of the current study is to evaluate the 17 !psychometric properties of a self -report tool, the Knowledge of Social -Emotional Milestones and Support Strateg ies (K -SEMS) designed to assess pre -service teachersÕ knowledge of general social -emotional milestones of children 3 Ð5 years old and knowledge of strategies to support childrenÕs development in this domain. Items for the K -SEMS were designed to assess pre -service ear ly childhood teachersÕ knowledge of childrenÕs social -emotional milestones and the strategies that can be used to support this development in children 3 Ð5 years old . The development of items was informed by the early childhood scientific literature on typical development from 3 to 5 years, and evidence based practices, as well as milestone information collected by federal institutions (e.g., the Centers for Disease Control [CDC] ), and the recommendations of national associations of early childhood recommendations ( CDC ,gov; E rtem et al., 2007 ; Fox & Lentini, 2006; Joseph, Strain, Yates, & Hemmeter, 2010; Kemple & Hartle, 1997; M acPhee, 1981; Squires et al. , 2002). Method Procedures Overview. This was a cross sectional study using a convenience sample of undergraduate students at one large university in the Midwest ern United States to develop and validate a tool to assess pre -service early childhood teachersÕ knowledge of preschoolersÕ social -emotional milestones and support strategie s ( the K-SEMS). The procedures used to develop and validate the K -SEMS included an item development phase, a validation phase, and a data collection phase, which are described below. These are followed by the analytic procedures used to provide evidence of the reliability (internal and test Ðretest) and validity (content , concurrent and predictive) of the instrument. 18 !Item d evelopment . Knowledge of milestones. Four to six items were created to capture knowledge of milestones in four broad area s of social -emotional development for children ages 3 to 5 years: self-awareness, self -regulation, relationship skills and social interaction, and social emotional understanding . A total of 25 original items were crafted based on the social -emotional milestones identified for children 2Ð6 years old (CDC .gov; Squire s et al. , 2002) and using examples of skills form the California Preschool Learning Foundations publication . Current measures of caregiversÕ knowledge of child development (e.g., KIDI [MacPhee, 1981 ], Caregiver Knowledge of Child Development Inventory [Ertem et al., 2007 ]) were considered to help structure the response options for participants, and inform the structure of questions. Following the answer structure of the KIDI, which assesses knowledge of infant and toddler development across domains, respondents were asked to respond to each item by indicating whether they ÒagreeÓ (1), disagree, and thus selecting either ÒyoungerÓ (2 Ð indicating a child develops this skill at a younger age) or ÒolderÓ (3 Ð indicating a child develops this skill at an older age) option, or are Ònot sureÓ (4), to determine accuracy of knowledge. These were later recoded to indicate the participantsÕ response as 1 (correct ) or 0 (incorrect ). Selection of the Ònot sureÓ optio n was coded as incorrect because it suggested that participants attempted the questions but did not have the knowledge to select the correct answer. Higher scores on the Knowledge of Milestones scale indicated pre-service teachers had more accurate knowled ge of childrenÕs social -emotional milestones. Knowledge of strategies . Items for the strategies section of the K -SEMS were gathered primarily from research providing evidence -based strategies used to support childrenÕs social -emotional development (Joseph et al. , 2010; Kemple & Hartle, 1997) ; websites and projects presenting strategies were also explored ( CDC.gov, California Department of Education, and 19 !VirginiaÕs Early Childhood Development Alignment Project ). The research emphasized a number of important strategies, but for this study, focus was placed on strategies that could be used while pre-service teachers are interacting with children during free -play and clean -up activities, such as modeling behaviors, label ing emotions, encouraging empathic thinking, and redirecting childrenÕs behavior in positive ways. Since no existing instruments specifically measure knowledge of strategies to support social -emotional development, statements were created based on these re search -based strategies shown to promote childrenÕs social -emotional learning. Ten original statements were created to measure participantsÕ knowledge of support strategies . Respondents were required to indicate whether the statements were true or false; answers were later recoded as correct (1) incorrect (2). Higher scores on the knowled ge of strategies scale indicate pre-service teachers have greater knowledge of effective strategies to support childrenÕs social -emotional milestones. Validation procedures . After the initial items and response options were created, nine individuals with expertise in the science of child development and practice of early child education reviewed them for face and content validity . The group of experts included six doctoral s tudents in child development and three professors in the same discipline. The doctoral students all had masterÕs degrees and had direct experience with young children; the professors each had over 15 years of experience in the early childhood field. The re viewers were asked to provide feedback on the clarity, accuracy, and appropriateness of each item for measuring knowledge of social -emotional milestones and strategies to support social -emotional development for 3 - to 5 -year -old children. Based on the preliminary feedback, three initial items were deleted, leaving 32 items. Ten of those items were reworded for clarity, and another five items received minor edits to improve relevance of examples for the 3 -to-5 age group. Two of 20 !the original experts plus six new experts complete d the revised 32 -item K -SEMS and offer ed additional feedback. Two additional items were discarded after this process because they were considered controversial Ñanswers were inconsistent Ñbased on expertsÕ answers and feedback. There were therefore a total of 30 items in the final version of the K -SEMS administered in this study . Data collection procedures. A convenience sample of undergraduate students enrolled in 10 different courses offered through a department of human development and family studies (HDFS) were recruited for this study. Approval to conduct the study was granted by the universityÕs Institution Review Board prior to the distribution of the survey. Courses ranged from lower level courses to upper level courses and incl uded early childhood courses (e.g., Child Growth and Development ) and courses not in the early childhood major (e.g., Lifespan Human Development in the Family). An online survey including (a) the 30 -item K-SEMS tool which included the knowledge of social -emotional milestones (20 items) and support strategies (10 items) ( Appendix A ), and (b) background questions such as demographics and experiences with children, were used to gather the data for this study. Surveys were distributed solely online during the f all semester (2014) via course instructors, but due to a low participation rate, a second round of surveys were distributed both in -class and online in the spring semester (2015). Students consented to participate in the study by selecting the ÒyesÓ option after reading the online consent form. For the paper -and -pencil option students selected Òyes,Ó signed, and dated a paper copy of the consent form. Students also had the option to decline consent by selecting ÒnoÓ online or on the paper copies, and could cho ose to withdraw their participation at any time during the study. Students received weekly e -mail reminders with the survey link and the survey closing date via their course instructors. Survey respondents were offered a chance to win 1 of 10 21 !$25 Amazon gift cards as an incentive to participate. Of the students participating in this study, a subsample of students taking three courses in the early childhood major with required field placements were observed for their social -emotional practices during inte ractions with children. This subsample of participants were observed one time for 1 hour during free play, clean -up, and transition to the next activity by undergraduate research assistants trained to used the observation tool for this study. Retest. A new survey link was directly e -mailed to participants 6 weeks after initial surveys were completed asking participants to complete the K -SEMS a second time. All participants were sent the e -mail in an effort to gain the needed responses to conduct test Ðretest analysis. An additional incentive of a chance to win 1 of 5 $15 Amazon gift cards was offered to participants to complete the survey the second time. Participants A total of 211 students filled out the initial survey. Fifty -one participant surveys were omitted from analys is because of missing data on all of the K -SEMS items. Thus, d ata from 160 participants were used to conduct analyses. The majority of participants were White (82%), female (96%), and ranged in ages from 18 to 20 (54%), 21 to 24 (39%), and 25 years and older (7.4%). The median household income while participants were in high school was $100,000 Ð$150,000 per year (see Table 2.1). Students were at various stages in their programs of study, and reported a variety of majors (see Table 2.1). Observation subsample. Of the 160 participants from the larger sample, 33 were observed using an adapted version of the Teaching Styles Rating Scale (TSRS) described in the measures section . The majority of observation participants were juniors and seniors in college 22 !(81%), white (91%) non -Hispanic (94%), and between the ages of 18 and 24 (94%). All observation Table 2.1 Characteristics of the Study 1 Participants ÑPercentages Characteristics Survey participants (n = 160) % Observation participants (n = 33) % Gender Female 95 100 Ethnicity Not Hispanic 95 94 Race White 82 91 African American/B lack 10 3 Asian/Pacific Islander 4 3 Other 4 3 Age 18Ð20 54 38 21Ð24 39 56 25+ 7 6 Major Early childhood 55 100 Communication 8 Kinesiology 8 Nursing/Pre -Nursing 8 Other 21 Year in College Freshman 13 0 Sophomore 21 9 Junior 34 38 Senior 24 44 Other 7 9 Family Income Under 10,000 1 10,000Ð49,000 21 16 50,000Ð99,000 25 26 100,000+ 31 23 Would rather not say 22 35 23 !participants were female and were early childhood majors (child development, or elementary education: early childhood track) . Test Ðretest subsample. Forty -eight participants took the K -SEMS a second time. Twelve of the participants were missing background data. The demographic data from the remaining 36 participants showed these participants were representative of the whole sampl e based on gender, race, ethnicity, family income, and major. Participants differed significantly on their year in college; a greater percentage of sophomores (36% vs. 21%) and fewer juniors (25% vs. 34%) and seniors (16% vs. 24%) retook the survey. Measures Coursework with social -emotional content. Copies of course syllabi were obtained from instructors to determine which courses included social -emotional content; three courses had social -emotional content embedded in them ÑChild Growth and Development, Interaction Processes with Children in groups, and Curriculum for Early Childhood Programs. In the background section of the survey participants were asked to indicate whether they (1) had not yet taken a course, (2) were taking it currently, o r (3) took it in the past for each of the 10 courses in which students were recruited. For each of the courses with social -emotional content, a binary variable was created to indicate whether the student had already taken the course; courses in which stude nts were currently enrolled were coded as 0 because the survey was collected early in the semester, thus students were not expected to have gotten much of the content. A variable was then created to summarize the number of courses students took in the past with social -emotional content. Codes for the coursework variable were as follows: 0 (took no prior courses with social -emotional content ); 1 (took 1 course with social -emotional content ); 2 ( took two courses with social -emotional content ); 3 (took all 3 courses with social -emotional content ). 24 ! Classroom practices that support social -emotional development . Observation data were collected from a subsample of participants using an adapted version of the Teach Styles Rating Scale (TSRS ; Domitrovich, Cortes, & Greenberg, 2000). This tool was designed to assess teachersÕ teaching styles across three domains: Positive Discipline, Classroom Management, and Positive Emotional Climate (Domitrovich et al. , 2000). It was chosen because it captur es teachersÕ interaction practices with children in both the social and emotional child development areas. This meant it could be used to provide predictive validity of the K -SEMS measure, which is designed to capture pre-service teachersÕ knowledge coveri ng both the social and emotional child development areas. The Classroom Management domain (3 items) from the original tool was removed because it was no t appropriate for observing the sample for this study, i.e., pre -service teachers in a classroom being s upervised by a lead teacher who was responsible for managing the classroom. One item under the Positive Discipline domain (negative behavior management) was also removed because it was believed the positive behavior management item in this domain would cap ture this information. Three items were then added to the tool to create a dimension to measure p ositive social climate and one item was added to the Positive Emotion Climate domain. Participants were rated by observers on their use of practices on a scale of 1 (almost never) to 5 ( almost always). Exploratory factor analyses conducted in Mplus on the adapted version of the TSRS for this study produced three subscales: Positive and Proactive Behavior Management includes three items ( ! = .74), Social -Emotiona l Guidance includes 3 items ( ! = .69), and Encouraging the Use of Social -Emotional Skills includes three items ( ! = .83). Average scores were calculated for each subscale. When there were no opportunities to observe participants using a practice this was c oded not observed and identified as missing data. Higher scores on the adapted TSRS indicated more frequent use of social -emotional practices. 25 !Analyses and Results In order to examine the psychometric properties of the K -SEMS, Exploratory Factor Analyses ( EFA) w ere conducted to first determine what conceptually meaningful latent factors might be extracted, and evaluate construct validity (DeVon et al., 2007 ). Internal and ext ernal consistencies of the extracted factors were measured using Cronbach Õs alpha analyses and PearsonÕs product moment correlations for test Ðretest. CronbachÕs alphas estimated at a .70 or higher indicate strong internal reliability and construct validity. Other measures of validity Ñconcurrent and predictive validity Ñwere calcul ated using the personÕs correlation analyses. Exploratory Factory Analyses EFAs of the knowledge of social -emotional milestones items (20) and support strategies items (10) were conducted separately. Version 6 of Mplus was used to conduct analyses of the 160 cases since the data for the K -SEMS were categorically coded as correct (1) or incorrect (0) and Mplus computes a polychoric correlation matrix for categorical data (Muth ”n & Muth ”n, 2010). Seven single responses were missing on different items from di fferent participants, and were therefore coded as missing; Mplus uses all cases that have at least partial data. If, however a case is missing values on all the variables in the analyses, Mplus excludes it from the model. The EFAs were estimated using the weighted least square estimator (WLSMV), which is considered a strong estimation method for factor analysis with categorical data ( Brown, 2006 ). An oblique rotation was chosen because it was assumed that the factors would be correlated (Costello & Osborne, 2005). Fit criteria. To evaluate model fit , the chi -square test of goodness -of-fit was used . For this test, the null hypothesis is that the model is a good fit for the data; thus, a p value greater than .05 is considered an indicator of good mod el fit ( Raykov & Marcoulides , 2011). The root Ð 26 !mean square error of approximation (RMSEA) was also used to evaluate fit. Values below .05 on the RMSEA usually indicate close fit (Browne & Cudeck, 1992), but a .0 6 cutoff is also suggested (Yu & Muth ”n, 2002). Other fit indices including the Comparative Fit Index (CFI) and the Tucker Lewis Index (TLI) can also be used to determine fit Ñcloser to 1 or 1 indicates the considered model is a better fit for the data than other models ( Raykov & Marcoulides , 2011). Model selection. The Kaiser -Guttmann criterion (eigenvalues > 1.0) was initially used to gain an idea of the number of possible factors in each of the K -SEMS scales, however, using this criterion is often shown to overestimate the appropriate number of fac tors (Gorsuch , 1983). Eigenvalues were also used to calculate the percentage of variance accounted for, once the number of factors that best fit the model were determined. CattellÕs (1966) scree test, which is a graphical representation of the eig envalues, was then used as a more objective method for determining the number of factors in the K -SEMS and the number of factors to rotate. Following this, each model, starting from the one -factor solution, was examined, and the first model that fit the data accord ing to the fit criteria was selected. The models were then evaluated for interpretability of loadings. Selection of the final factor structure for the two main subscales of the K -SEMSÑknowledge of milestones and knowledge of strategies Ñwas based on the abi lity of the models to produce subscales that (a) had salient factor loadings > .30, (b) indicated three or more items for retention on a subscale, and (c) exhibited conceptual clarity. Knowledge of preschoolersÕ social -emotional milestones. The Kaiser -Guttmann criterion revealed eight factors with eigenvalues greater than 1.0, using the 20 items for this section of the K -SEMS, but these were not interpretable. Examining the scree test showed the 4 -factor solution as the most promising factor solution for the initial EFA. Exploring the fit indices in each model from the 1 -factor solution to the 4 -factor solution indicated the 4 -factor solution 27 !was the best fit for the data. Two items (2 and 12) in the 4 -factor solution did not load on to any of the four factors and while this solution was indicated as best explaining the data, the 3 -factor solution made more conceptual sense related to age -based social -emotional milestones. Deleting items 2 and 12, the remaining18 items were therefore analyzed using a forced 3 -factor solution. The 3 -factor solution revealed three items (items 1, 6, and 18) with loadings < .30 on all three factors. Removing those items produced a 3 -factor solution with good model fit ( !2 = 72.55, p = .19, df = 63, RMSEA = .03; CFI/TLI = .94/.9 1) and factors that were interpretable, and was therefore chosen as the final model. The three eigenvalues account ed for approximately half of the original variance. Because there were 15 observed variables each with a variance of 1, there were 15 units of variance, hence the share of the three eigenvalues is (3.259 + 2.220 + 1.870)/15 = 49% (Raykov & Marcoulides , 2011). Conceptually, the three factors extracted measured (1) knowledge of 3 -year -olds social -emotional milestones, (2) knowledge of 4 -year -olds social -emotional milestones, and (3) knowledge of 5 -year -old social -emotional milestones (see Table 2.2 for final factor loadings), and were thus labeled in this manner . Although items were crafted around age -based milestones, there was an expectation that the items would load based on the four social -emotional development areas focused on for this study (i.e., self-awareness, self -regulation, relationship skills and social interaction, and social emotional understanding ). Nonetheless, items did not load in this manner , and even though the three factors were conceptually identified as knowledge of 3 -, 4- and 5 -year -olds Õ milestone s, the loadings were not clean and distinct. Some probing of loadings seemed to suggest that loadings were not a r esult of underlying latent variables, but could be reflecting participantsÕ choices of associated response options of agree, younger, and older. 28 !!Note: N = 160 Table 2. 2 Factor Loadings for Knowledge of Social -Emotional Milestone Items (15 Items) Component Knowledge of 3 -year - olds (1) Knowledge of 4 -year -olds (2) Knowledge of 5 -year -olds (3) 4. At around the age of 4 yea rs old , children typically begin to assert their independence, test limits and say ÒNoÓ a lot. (Younger) .84 .20 .00 8. The average 4-year -old is able to manage transitions in the classroom routine, such as moving from playtime to cleanup without resist ing or throwing a tantrum. ( Younger) .37 .08 .03 10. At around 3 years old , children typically begin to notice and respond to other peopleÕs moods and feelings. (Younger) .66 .04 .06 11. At around 4 years old, typical children may cling to parents/teachers in new situations. (Younger) .61 .00 .54 13. At around 6 years old , ch ildren typically begin to show a strong sense of self through assertiveness and directing others. (Younger) .79 Ð.04 Ð.22 20. Most 5 -year -old children typically play Ònext to,Ó rather than Òwith,Ó others. (Younger) .66 .42 .03 7. The average 3-year -old understands that another child might be mad because he/she couldnÕt finish a hard task he/she was trying to do. (Older) .19 .44 .17 14. On average, children begin to compare their abil ities to those of others aro und the age of three (3). (Old er) Ð.04 .49 .03 15. A typical 3 -year -old may explain the reason for a behavioral rule (e.g., ÒWe walk inside so we donÕt bump into other peopleÓ). (Older) Ð.05 .52 Ð.00 16. A typical 3 -year -old child may suggest taking turns riding the tricycle on the playground. (Older) .00 .51 .26 17. Typical 4 -year -old children may coordinate simple sequences of pretend play by communicating to another child ÒIÕll be the mommy, you can be the baby.Ó (Agr ee) Ð.32 .55 Ð.00 21. On average, children around 3 years old are motivated to please their friends. (Older) Ð.14 .42 .02 3. Children around 5 years old could generate an explanation for another childÕs emotions . (Agree) Ð.09 Ð.11 .32 5. At ar ound 3 years old children are able to regulate their impulses with little adult guidance. (Older) Ð.01 .13 .94 9. Children around 5 years old may come to the defense of a friend who is teased by a peer. (Agree) .01 Ð.12 .39 29 !Knowledge of social -emotional support strategies. A separate EFA was conducted for the knowledge of strategies section of the K -SEMS and revealed a one -factor solution, which was confirmed by the scree test. This one factor accounted for 29% of the variance and model fit indices indicated tha t one factor best explained the data ( !2 = 29.42 p > .05, df = 27, RMSEA =. 02, CFI/TLI = .96/.95). The 29% of variance accounted for is low and could mean that a one -factor solution was not the best fit for the data, however these were the only output pro duced. One item (item 1) loaded < .10 and was eliminated from the final model (see Table 2.3 for factor loadings). Items two and five were retained in the model because they were conceptually important to the factor. This one-factor solution conceptually measured support strategies for preschoolersÕ social -emotional development and was thus named. Validity. Content validity and two types of criterion validity were tested to establish the validity of the K -SEMS, including concu rrent and predictive validity. Content validity. Content validity was established by conducting independent samples t -test to examine differences in knowledge scores between participants in an early childhood major and those in other majors. Results showe d significant mean differences in the scores on the Knowledge of 3 -year -olds Õ Milestones subscale of participants in an early childhood major ( M = .49, SD = .25) compared to those in other majors ( M =.33, SD = .25). Significant differences were also found in the mean scores on the Knowledge of Support Strategies scale of participants in an early childhood major (M = .85, SD = .16) compared to those in other majors (M = .69, SD = .16). Results indicate this knowledge measure has appropriate content relative to the population for which it is intended. 30 !!Table 2.3 Factor Loadings for Knowledge of Support Strateg y Items (9 Items) Component Support Strategies for PreschoolersÕ Social -Emotional Development 2. Drawing childrenÕs attention to othersÕ feelings (e.g. , ÒLook at her face. Can you tell how she feels?Ó) can help them develop empathy. .28 3. Children learn most from other children, so i tÕs important to point out what other children are doing right and wrong in the classroom. .45 4. It is best to ignore children who are acting out in the classroom; most children just act out to get attention, and itÕs best not to reward that behavior by giv ing them attention. .39 5. Establishing warm, close relationships with children who exhibit challenging or negative beh aviors will help them develop self -control. .17 6. Allowing children to play together freely may allow conflicts to occur between children, b ut that is a way for them to learn problem -solving and negotiation skills .43 7. Giving children specific words to say to each other during social conflicts will hinder their abilities to develop their own skills for negotiation and managing conflicts. .43 8. Labeling childrenÕs emotions helps children to identify and understand emotions .36 9. Labeling your own negative emotions for children could scare children. .72 10. Teachers should not express negative emotions in the classroom because it creates a negative at mosphere that harms childrenÕs abilities to learn. .76 Note: N = 160 Concurrent validity . Concurrent validity was established by correlating factors on the K -SEMS with courses including social -emotional content using the PearsonÕs correlation method ParticipantsÕ knowledge of 3 -year -olds Õ social -emotional milestones was significantly and positively correlated with the number of courses participants took which included social -emotional content ( r = .36, p < .01), suggesting a positive association between knowledge of 3 -year -olds Õ milestones and courses with social -emotional content. ParticipantsÕ knowledge of 4 -year -olds Õ social -emotional milestones significantly but negatively correlated with relevant coursework ( r = Ð.28, p < . 01), suggesting associations between more courses with social - 31 !emotional content and less accurate knowledge of 4 -year -oldsÕ social -emotiona l milestones. Knowledge of 5 -year -olds Õ social -emotional milestones was not significantly correlated with coursework ( r = .14, ns). Knowledge of support strategies was positively and significantly correlated with coursework ( r = .28, p < .01), suggesting a positive association between participantsÕ knowledge of support strategies and courses taken with social -emotional content. Predictive validity. Predictive validity was established by correlating participantsÕ scores on both knowledge scales with their scores on observed classroom practices for supporting social -emotional development (see Table 2.4) , which included three sub -scales: (1) positive and proactive behavior management, (2) Encouraging the Use of Social -Emotional Skills, and (3) Social -Emotional Guidance. Knowledge of 3 -year -olds Õ Social -Emotional Milestones correlated only with the Positive and Proactive Behavior Management subscale on the adapted TSRS ( r = .38, p < .05), but not the other two subscales. The Knowledge of Support Strategies scale was positively correlated with the Positive and Proactive Behavior Management subscale ( r = .36, p < .05) and encouraging the use of social -emotional skills subscale on the TPRS ( r = .38, p < .05); the association between knowledge of support strategies and the social -emotional guidance subscale on the TSRS was not significant. Table 2.4 Correlations Between K -SEMS and TSRS Subscales TSRS Subscales K-SEMS Subscales Knowledge of 3 -Year-Olds Knowledge of 4 -Year-Olds Knowledge of 5 -Year-Olds Knowledge of Support Strategies Positive and Proactive Behavior management .38* Ð.30 .07 .36* Social -emotional Gui dance .15 Ð.29 Ð.35 .31 Encouraging the Use of Social -emotional Skills .08 Ð.18 Ð.13 .38* N 29 29 29 30 *p < .05 32 !Reliability. Two estimators of reliability were used to examine the reliability of the K -SEMS. First, internal consistency was used to determine how well the items in the revealed K -SEMS factors fit together and second, test Ðretest was used to determine the stability o f the K -SEMS factors (DeVon et al., 2007). Internal consistency. Internal consistency for each of the K -SEMS factors was calculated using CronbachÕs alpha in SPSS 22.0 . CronbachÕs alphas for the factors were Factor 1 (Knowledge of 3 -year old social -emotio nal development, 6 items) .61; Factor 2 (Knowledge of 4-year old social -emotional milestones, 6 items) .50; and Factor 3 (Knowledge of 5 -year old social -emotional milestones, 3 items) .39. For the Knowledge of Social -Emotional Support Strategies , the C ronbachÕs alpha (9 items) was .51. Based on internal consistency, Factor 1 on the Knowledge of Milestones scale (Knowledge of 3 -year -olds) was the most robust of all factors, but still did not meet the .70 criteria for acceptable reliability recommended for ne w scales (DeVellis, 2003). This indicates that these factors should be interpreted with caution and that the items in these factors may not consistently measure pre -service professionalsÕ knowledge of social -emotional milestones and support strategies. TestÐretest. This form of reliability measures the stability in a test given at two time points and was established by correlating participantsÕ responses on the knowledge of milestones and strategies instruments at time -1 with their time -2 responses. Usin g the Pearson correlation the test Ðretest reliability for knowledge of 3 -year -old s factor was significant at the .10 level ( r = .26, p = .07), the knowledge of 4 -year -olds factor was significant ( r = .58, p < .01) and the knowledge of 5 -year -olds factor wa s not significant ( r = .12, p = .41). The t est Ðretest correlation coefficient for Knowledge of Support Strategies between Time 1 and Time 2 was strong and statistically significant ( r = .72, p < .01). 33 !Discussion The aim of this study was to evaluate the psychometric properties of a tool developed to measure pre -service early childhood teachersÕ knowledge of childrenÕs social -emotional milestones and strategies to support children in this domain using a sample of undergraduate students enrolled in courses through an HDFS department at one university. Knowledge of Social -Emotional Milestones Scale The three factors identified on the Knowledge of Social-Emotional Milestones scale were indicative of a general knowledge of preschoolers Õ age -based social -emotional milestones. However, i t was expected that items on the Knowledge of Social -Emotional Milestones scale would have loaded based on the four social -emotional areas initially used to developed items Ñself-awareness, self -regulation, r elationship skills and social interaction, and social e motional understanding Ñto create a multidimensional construct of knowledge of social -emotional milestones. Based on the extracted factors, however, it appeared as though items with correct responses id entified as Òagree ,Ó Òyounger , or ÒolderÓ were loading onto each other to create a multi -dimensional construct based on age and thus identified as K nowledge of 3 -Year-Olds, Knowledge of 4 -Year-Olds, and Knowledge of 5 -Year-Olds . Because of the response options included in the Knowledge of Social Emotional Milestones section, the item loadings were untidy, and some items could have loaded onto multiple factors and still be conceptualized as fitting for that factor . For example, Ò A typical 3 -year -old may explain the reason for a behavioral rule ÓÑanswer, Older Ñcould have loaded onto knowledge of 4 - or 5-year-olds. It could be that the factors extracted are not producing age -based constructs but rather are artifact s of the responses even though the underlying construct is unidimensional (Spector, Katwyk, Brannick, & Chen, 1997) . Treating the K -SEMS as a scale rather than an index may have been 34 !inappropriate for this tool. While a scale and an index are both composite measures of variables , they have differen t properties and account for composite score s in different ways . A scale is a measure based on intensity of a concept such as attitudes or emotions and assign s numerical values to a concept based on a range of response -related intensities to provide an interpretation of the relative intensity of items on a scale (Babbie, 200 1). It is comprised of several items having an empirical or a logical structure amo ng them. The most commonly known and used scale is the L ikert s cale which may include responses such as strongly agree, agree, disagree, strongly disagree . On the other hand , an index is a measure that accumulates values from a variety of items and represents more of a general dimension (Babbie, 2001). Studies should be cognizant of these differences in order to appropriate ly design and analyze new instruments. Based on these differences, the Knowledge of Social -Emotional Milestones and Strategies may be considered an index rather than a scale. Results u sing all 20 items initially included on the Knowledge of Milestones sub -scale to calculate an index score and then estimate internal consistency show the Knowledge of Milestones index is not a good index or measure of pre-service teacher sÕ knowledge of social -emotional milestones (! = .44). Further, there were no significant associations between overall Knowledge of Milestones index and coursework, or observed practices. However, c orrelation s of all items showed the items which had loaded as the Knowledge of 3 -Year-Olds factor in the earlier factor analysis were the only items significantly correlated with each other, suggesting these six items (4, 8, 10, 11, 13, and 20) as a more concise and adequate index of Knowledge of Social -Emotional Milestones . Based on reliability and validity analyses of these six items previously identified as the Knowledge of 3 -Year-Olds Social -Emotional Milestones, this more concise index of Knowledge of Milestones was 35 !determined as the most valid and reliable for measuring teacherÕs knowledge of social -emotional milestones. Improving the r eliability and validity of the Knowledge of Social -Emotional Milestones s cale . The subscales of the Knowledge of Social-Emotional Milestones scale had poor to moderate internal reliability ranging from . 39 to .61 and only the Knowledge of 4 -year -old subscale was stable from Time1 to Time2 ( r =.58, p < .01), bringing into question the ability of this scale to consistently measure pre -service ECEPsÕ knowledge of milestones. The moderate relia bility on all factors may be due to a number of reasons. First , the value of alpha is significantly reduced when items are not correlated, and many of the items in Factor 2 (Knowledge of 4 -year -olds) and 3 (Knowledge of 5 -Year-olds) were not correlated, th us producing lower alphas than Factor 1 (Knowledge of 3 -Year-olds) (De Von et al., 2007). Second, the number of items included in each subscale may not have been adequate to reach the desired alpha level. It was not surprising that the Cronbach Õs alpha for Factor 3 (knowled ge of 5 -year -olds social -emotional milestones ), which included three items , would have lower reliability than the other factors , which included six items , since calculations of internal consistency are sensitive to the number of items included in a factor (DeVe llis, 2003) . Thus, adding items may help to improve the reliability of the subscales. Although not used for this study, the Spearman -Brown Prophesy F ormula can be used to determine the number of items that are needed to improve reliability to a desired level ( DeVellis, 2003 ) and should be considered in future studies. Along with increasing the number of items included in each of the subscales, the next phase in developing this instrument should consider other knowledge of social -emotional development, such as sequences of development and learning processes. For this study, o nly milestones of development were included as knowledge of social -emotional development. 36 !However, items should be crafted to assess teachersÕ knowledge of childrenÕs social -emotional sequences of developmental and different learning processes. Combined, these different components of knowledge of development could produce a better quality tool that provides a comprehensive assessment of pre-service early childhood teachersÕ knowledge, and highlight areas where teachers may need additional professional development. Poor test Ðretest stability could have occurred due to four reasons id entified by Kelly and McGrath (1988). First, it is possible participants knowledge could have increased during the 6 -week lag between tests, thus, affecting differences in responses from Time 1 to Time 2. Second, it is possible participant responses could have been influenced by the time of day or test administration (paper -pencil at Time 1 and online at Time 2). Participants may have felt more pressured and rushed taking the test in class as opposed to more relaxed taking the survey at their leisure online. Thirdly, fatigue could have played a role in how participants read items and therefore responded. Lastly, it is possible that the instability of the test may have been due to inherent unreliability of the measurement . Future studies therefore need to consider many factors when conducting this form of reliability and careful acknowledgment should be given to time1 and time2 confounding factors to help explain differences. Finally , the quality of the ite ms may need to be improved to better measure knowledge. Items could be improved by using broader constructs of social -emotional skills instead of specific examples, which despite the best efforts to craft, can lead to misinterpretation by participants and affect reliability. Knowledge of Support Strategies Scale The items on this scale produced a unidimensional factor , and two items with factor loading s < .30 were retained, which may have affect ed the reliability of this factor . Removing 37 !items with low loading s are suggested because these items may not be important to the factor (Rahn, 2015 ). However, examination of the CronbachÕs alpha results showed that removing these items did not change the alpha level. Even thought this scale had only moderate inte rnal consistency estimated at ! = .51, the stability of the test from Time 1 to Time 2 was r = .72, p < .01, indicating strong stability . However, the true/false answer format for the knowledge of support strategies statements could have made it easier for participantsÕ to select consistent responses at Time 1 and Time 2. This format gave respondents a 50/50 chance to select the correct answer , which limited the amount of variance in answers , and may have resulted in participants Õ answering some items correctly by chance. But since response s were highly consistent from Time 1 to Time 2 and there was a 6-week lag between tests , it is possible that those who responded correctly did have more accurate knowledge of strategies. A better question and answer format may eliminate these ambiguities and provide a better way to capture teachersÕ knowledge of strategies to support social -emotional development. Limitations and Recommendations The sample of participants used for this study were al l from the same university, were homogenous in terms of gender and race , and came from middle -income families, thus limiting the generalizability of the findings. Many other unmeasured factors could have played a role in the associations between the scores on the K -SEMS and the scores used to establish validity. That is, the positive associations between participantsÕ scores on the K -SEMS, coursework taken, and observed practices could be explain ed by associations with var iables not collected for this study. For example, factors such as IQ and GPA may have been important to consider since there is a possibility that participants with higher IQs or GPAs may have answered question s more accurately due to intelligence , and 38 !hav ing performed better in courses related to the content of the study . Also, work or personal experience with children in the preschool years may also be factors related to participantsÕ knowledge of social -emotional development, the courses they do or do no t take, and the ways they interact with children to support thei r development . It is possible that some participants responded to questions on the K -SEMS based on their own implicit theories of social -emotional development rather than on research -based knowledge they receive through courses (Gebel & Schrier, 2002; Kuzborska, 2011) . It is further possible these beliefs could have affected the ratings observed participants received during th eir interactions with children , and thus the relations between knowle dge and practices. It is also possible that participants may believe they already have knowledge of childrenÕs social -emotional development and may forego taking courses with social -emotional content, or may attend less to that particular content, which may affect the accuracy of their answers on the K -SEMS ( Cunningham, Zibulsky, & Callahan, 2009) . Future studies should consider each of these factors and their potential relationships to coursework, knowledge, and practices in order to assess psychometric properties of a knowledge tool. Many of the items on the K -SEMS referred to a gender as part of the example (e.g., ÒLook at her faceÉ Ó) and could have unconsciously influenced responses on items, especially if participants implicitly were accounting for differences that may exist between girls and boys social -emotional development. Revisions to items and the addition of items should ensure items are balanced by making references to both genders and making sure items account for differences related to gende r. Only 3 5 items were developed for the K -SEMS. A large pool of potential items, which can be reduced based on expert feedback as well as participant responses , should be developed in 39 !the early stages of instrument development (Netemeyer et al., 2003). Having a larger pool of items may help to better capture the intended construct and may permit similar questions to be asked in multiple ways and formats to allow a more thorough evaluation from experts and participant. The K -SEMS was not pilot -tested with student respondents, only with experts, prior to distribution of the survey. This is a critical stage of instrument development and may have contributed to the poor to moderate reliability and validity o f the K -SEMS. Future development of this tool should include a more thorough pilot testing stage. The response optio ns to questions may not have been the most appropriate to assess knowledge on either of the K -SEMS scales. If a participant selected ÒyoungerÓ or ÒolderÓ on an item it did not suggest the par ticipant knew exactly at what age a social -emotional skill appears , just that they have enough knowledge to know it does not appear at the age associated with the question , leaving the researcher to assume these participants have accurate knowledge of soci al-emotional milestones . Future studies could consider using vignette -type questions and responses reflecting a correct action, a partially correct action, and a n in appropriate action to better assess pre-service teachersÕ content and practical knowledge of social -emotional development . This would alleviate some of the problems incurred from true or false responses and from response options including a younger -older option, which led t he researcher to make many assumptions about participantsÕ actual knowledge. F uture studies might consider the use of open -ended question s at the item -development phase to design appropriate items for a final measure. Allowing for open -ended responses on the strategies scale could offer valuable insight into what strategies pre -service teachers are being taught through coursework or are learning while working with children. 40 !Only two component s of child development knowledge were measured Ñknowledge of milestones and support strategies Ñlimiting the thoroughness of the measure to assess knowledge of development. Other components such as sequences of skill development and learning processes should be included in future development of this instrument. Furth er, a number of other strategies were not represented in this scale but are recommended as best practices to promote childrenÕ social -emotional development. These include modeling positive social behavior, role playing, encouraging children in their social -emotional efforts, setting up the environment to support social -emotional development, providing materials to promote social -emotional skills, and developing activities to promote, support and encourage childrenÕs social -emotional development (Fox & Lenti ni, 2006; Joseph et al. , 2010; Kemple & Hartl e, 1997). The reason for this omission of strategies was that many of the students in field placements are not the ones making decisions about what is done in the classroom (materials, design of space and enviro nment) as they are under the direction of the lead teachers , and it was thought that the strategies developed should be aligned with what students were allowed to do in the classroom. Thus, it was assumed that there would be a lot of missing data on the kn owledge of strategies subscale, or that they would not be valid for this population, if the strategies did not take into account the role of the participants in the classroom. However, including these other variables may have provided a more thorough test of knowledge related to support strategies , particularly if this measure were to be used with a broader population of pre -service and in -service early childhood professionals . These other strategies should therefore be included in the next phase of measure development especially if this instrument will be used to assess the knowledge of in -service teachers. 41 !Conclusion The K -SEMS was designed to capture pre -service teachersÕ knowledge of preschoolersÕ social -emotional milestones and support strategies. After careful review of factors and further research related to scales and indices , it was determined that the K -SEMS should be treated as an index rather than a scale. Further, the most robust factor on the Knowledge of Social -Emotional Milestones sub -scale (Knowledge of 3 -Year-Old Milestones) and the K nowledge of Support Strategies sub -scale should be used as an index and compose the final measure of pre-service teachersÕ knowledge prior to any further modifications and testing of the measure . Caution should be taken in interpreting this index as a thorough measure of preschoolersÕ social -emotional milestones and support strategies. Additional items should be crafted and tested for both sections of the index a more diverse sample should be surveyed and o bserved to make this tool more adequate for the early childhood population. Despite the need to revise and enhance this instrument, these results indicate the importance of domain -specific measures of knowledge to understand the professional development n eeds of pre -service and in -service teachers of young children. Only one factor of this instrument was sufficiently reliable and valid to be considered for further development, but the process taken and the limitations outlined provide useful information to guide the development of a better quality tool. This study provided an instrument designed to capture preschool teachersÕ actual knowledge of social -emotional milestones and strategies to support social -emotional development. This instrument can serve as an important support for early childhood programsÕ curricula development , course enhancements, and professional development for pre -service teachers . Opportunities to take courses with relevant content , plus more and better learning opportunities , can help teachers with their competence in supporting 42 !childrenÕs development in all domains and produce effective teachers of young children. Enhancing teachersÕ knowledge of social -emotional skills and support strategies could significantly affect childrenÕs soci al-emotional development and thus their school readiness. 43 ! APPENDIX 44 !Appendix Instructions (Please read carefully before starting): This questionnaire asks you about childrenÕs social and emotional development. Each of the following statements asks you about the average age at which typically developing children can do something. If you think the age is correct, check ÒAgree.Ó If you d onÕt agree, then decide whether a Younger or Older child could do it. If you arenÕt sure of the age, check ÒNot Sure.Ó 1. A typical 5 -year -old may have the self -awareness to recognize he/she is currently able to do something he/she wasnÕt able to do when he/ she was younger, such as riding a bike. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 2. By age 4, children are able to verbally label their own feelings of anger and frustration by saying ÒIÕm madÓ or ÒIÕm upset.Ó (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 3. Children around 5 years old could generate an explanation for another childÕs emotions (e.g., Jessica is sad because she wanted to play with the doll and someone else took it). (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 4. At around the age of 4 years old , children begin to assert their independence, test limits, and say ÒNoÓ a lot. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 5. At around 3 years old , children are ab le to regulate their impulses with little adult guidance. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 6. At a round 4 years old, preschoolers consistently play with one or two special buddies, especially those of the same sex. (A) Agre e _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 7. Most 3 -year -olds understand that another child might be mad because he/she couldnÕt finish a hard task he/she was trying to do. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 8. Most 4 -year -olds are able to manage transitions in the classroom routine, such as moving from playtime to cleanup without resisting or throwing a tantrum . (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ Knowledge of ChildrenÕs Social -Emotional Milestones and Support Strategies (K -SEMS ) ! 45 !9. Children at around 5 years old may come to the defense of a friend who is teased by a peer. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 10. Around 3 years old children begin to notice and respond to other peopleÕs moods and feelings. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 11. Around 4 years old, children may cling to parents/teachers in new situations. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 12. Children around 4 years old understand basic moral reasoning about good and bad. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 13. Around 6 years old children begin to show a strong sense of self through assertiveness and directing others. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 14. Children begin to compare their abilities to those of others around the age of 3. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 15. A typical 3 -year -old may explain the reason for a behavioral rule (e.g., ÒWe walk inside so we donÕt bump into other peopleÓ). (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 16. A 3 -year -old child may suggest taking turns riding the tricycle on the playground. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 17. Typical 4 -year-old children may coordinate simple sequences of pretend play by communicating to another child ÒIÕll be the mommy, you can be the baby.Ó (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 18. Children around 5 years old may negotiate with each other, seeking adult assistance when needed. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ 19. Most 5 -year -old children typically play next to , rather than with , others. (A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure _ _____ 20. Children around 3 years old are motivated to please their friends. 46 !(A) Agree _____ (B) Younger____ (C) Older____ (D) Not Sure ______ The next section is about the things that support preschool -aged childrenÕs social skills. In this section, please read the following statements and respond by selecting true or false. Research suggests thatÉ 1. When young children are upset and acting out , you should use that moment as an opportunity to explain the best way to handle their emotions. a. True b. False 2. Drawing childrenÕs attention to othersÕ feelings (e.g. , ÒLook at her face. Can you tell how she feels?Ó) can help them develop empathy. a. True b. False 3. Children learn most from other children, so itÕs important to point out what other children are doing right and wrong in the classroom. a. True b. False 4. It is best to ignore children who are acting out in the classroom; most children just act out to get attentio n, and itÕs best not to reward that behavior by giving them attention. a. True b. False 5. Establishing warm, close relationships with children who exhibit challenging or negative behaviors will help them develop self -control. a. True b. False 6. Allowing children to play together freely may allow conflicts to occur between children, but that is a way for them to learn problem -solving and negotiation skills. a. True b. False 7. Giving children specific words to say to each other during social conflicts will hinder their abilities t o develop their own skills for negotiation and managing conflicts. a. True b. False 8. Labeling childrenÕs emotions helps children to identify and understand emotions. a. True b. False 9. Labeling your own negative emotions for children could scare children. a. True b. False 47 !10. Teachers should not express negative emotions in the classroom because it creates a negative atmosphere that harms childrenÕs abilities to learn. a. True b. False 48 ! REFERENCES 49 !REFERENCES Babbie, E. (2001). The practice of social research (9th ed.). Belmont, CA: Wadsworth Thomson. Baker, J. A. (2006). Contributions of teacher Ðchild relationships to positive school adjustment during elementary school. Journal of School Psychology, 44 , 211Ð229. Baker, J. A., Clark, T. P., Maier, K. S., & Viger, S. (2008). The differential influence of instructional context on the academic engagement of students with behavior problems. Teaching and Teacher Education, 24 , 1876Ð1883. Bavolek, S. J. (1984). Adult -adolescent parenting inventory (AAPI). Park City, UT : Family Development Resources, Inc. Blomeke, S., Buchholtz, N., Suhl, U., & Kaiser, G. (2014). Resolving the chicken -or-egg causality dilemma: The longitudinal interplay of teacher knowledge and te acher beliefs. Teaching and Teacher Education, 37 , 130Ð139. Brown, T. A. (2006). Confirmatory Factor Analysis for Applied Research . New York , NY : Guilford . Browne, M. W., & Cudeck, R. (1992). Alternative ways of assessing model fit. Sociological Methods and Research, 21, 230Ð258. Burchinal, M. R., Cryer, D., Clifford, R. M., & Howes, C. (2002). Caregiver training and classroom quality in child care centers. Applied Developmental Science, 6 , 2Ð11. Burchinal, M., Roberts, J. E., Zeisel, S. A., Hennon , E. A. , & Hooper , S. (2006).!Social risk and protective child, parenting, and child care factors in early elementary school years . Parenting: Science and Practice, 6 , 79Ð113. Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral Research, 1, 245Ð276. Costello, A. B., & Osborne, J. W. (2005). Exploratory factor a nalysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research, and Evaluation, 10 , 1Ð9. Cunningham, A. E., Zibulsky, J., & Callahan, M. D. ( 2009). Starting small: Building preschool teacher knowledge that supports early literacy development . Reading & Writing, 22 , 487Ð510. Denham, S. A., Kalb, S., Way, E., Warren -Khot, H., Rhoades, B. L., & Bassett, H. H. (2013). Social and emotional information processing in preschoolers: indicator of early school success? Early Child Development and Care, 183 , 667Ð688. doi:10.1080/03004430.2012.682728 50 !DeVellis, R. F. (2003). Scale development: Theory and a pplication (2nd ed.) . Thousand Oaks, CA: Sage. DeVon, H. A., Block, M. E., Moyle -Wright, P., Ernst, D. M., Hayden, S. J., É Kostas -Polston, E. (2007). A psychometric toolbox for testing validity and reliability. Journal of Nursing scholarship, 39 , 155Ð164. Domitrovich, C., Cortes, R. C., & Greenberg, M. T. (200 0). Head Start Competence Scale technical report . Unpublished manuscript, Pennsylvania State University. Early, D., Bryant, D., Pianta, R., Clifford, R., Burchinal , M., Ritchie, S., É Barba rin, O. (2006). Are teacher education, major, and credentials related to classroom quality and childrenÕs academic gain in pre -kindergarten? Early Childhood Research Quarterly , 21, 174Ð195. Early, D. M., Maxwell, K. L., Burchinal, M., Alva, S., Bender, R. H., Brya nt, D., É Zill, N. (2007). TeachersÕ education, classroom quality, and young childrenÕs academic skills: Results from seven stud ies of preschool programs. Child Development, 78 , 558Ð580. Eisenberg, N., Valiente, C., & Eggum, N. D. (2010). Self -regulation and school readiness. Early Education and Development, 21 , 681Ð698. doi:10.1080/10409289.2010.497451 Elicker, J. & Fortner -Wood, C. (1995). Research in review: Adult Ðchild relationships in early childhood settings. Young Children, 51 , 69Ð78. Ertem, I.O., Atay, G., Dogan, D.G., Bayhan, A., Bingoler, B.E., Goc, C.G., Ozbas, S., Haznedaroglu, D. & Isikli, S. (2007). MotherÕs knowledge of young child development in a developing country. Child: Care Health and Development, 33, 728-737. Fox, L., & L entini, R. (2006). ÒYouÕve got it !Ó Teaching social and emotional skills. Beyond the Journal: Young Children on the Web , November , 1Ð7. Garner, P. W., Mahatmya, D., Moses, L. K., & Bolt , E. N. (2014) Associations of preschool type and teacher Ðchild relational quality with young children Õs social-emotional competence. Early Education and Developmen t, 25 , 399Ð420. doi: 10.1080/10409289.2013.801706 Gebel, A., & Schrier, L. L. (2002). Spanish language teachersÕ beliefs and practices about reading in a sec ond language. In J. H. Sullivan (Ed.), Literacy and the second language learner (pp. 85Ð109). Charlotte, NC: Information Age Publishing. Giles, G. (2015). What are developmental milestones in children? Table, definitions and examples . Retrieved from http://study.com/academy/lesson/what -are -developmental -milestones -in-children -table -definition -examples.html#lesson Gilliam, W. (2005). Prekindergarteners left behind: Expulsion rates in state prekindergarten systems. New York , NY : The Foundation for Child Development. Gorsuch, R. L. (1983). Factor analysis . Hillsdale , NJ: Erlbaum. 51 !Hamre, B. K., & Pianta, R. C. (2001). Early teacher Ðchild relationships and the trajectory of childrenÕs school outcomes through eighth grade . Child Development, 72, 625Ð638. Hatcher, B., Nuner, J., & Paulsel, J. (2012) . Kindergarten readiness and preschools: TeachersÕ and parentsÕ beliefs within an d across programs. Early Childhoo d Research and Practice, 14. Hemmeter, M. L., Corso, R., & Cheatham, G. (2006). Issues in addressing challenging behaviors in young children: A national survey of early childhood educators . Paper presented at the Conference on Research Innovat ions in Early Intervention, San Diego, CA. Howes, C. (2000). Social -emotional classroom climate in child care, child Ðteacher relationships and childrenÕs second grade peer relations. Social Development, 9 , 191Ð204. Joseph, G., Strain, P., Yates, T., & Hemmeter, M. L. (2010). Social emotional teaching strategies: PresenterÕs script, module 2 . Retrieved from http://csefel.vanderbilt.edu/modules/modul e2/script.pdf Kelly, J. R. , & McGrath, J. E. (1988). On time and method . Newbury Park, CA : Sage .!Kemple, K., & Hartl e, L. (1997). Ge tting along: How teachers can support childrenÕs peer relationships. Early Childhood Education Journal, 24, 139Ð146. Kuzborska, I. (2011). Links between teachersÕ beliefs and practices and research on reading. Reading in a Foreign Language, 23 , 102Ð128. Ladd, G. W., Birch, S. H., & Buhs, E. S. (1999). ChildrenÕs social and scholastic lives in kindergarten: Related sphere s of influence? Child Development, 70 , 1373Ð1400. Luster, T., Rhoades, K., & Haas, B. (1989). The relation between parental values and parenting behavior: A test of the Kohn hypothesis. Journal of Marriage and the Family, 51, 139-147. MacPhee, D. (1981). Knowledge of Infant Development Inventory . Princeton, NJ: Educational Testing Service. McClelland, M. M., Cameron, C. E., Connor, C. M., Faris, C. L., Jewkes, A. M., & Morrison, F. J. (2007). Links between behavioral regulation and pr eschoolersÕ literacy, vocabulary, and math skills. Developmental Psychology, 43, 947Ð959. doi:10.1037 /0012Ð1649.43.4.947 Mitchell -Copeland, J., Denham, S., & DeMulder, E. (1997). Q -Sort assessment of child Ðteacher attachment relationships and social compet ence in the preschool. Early Education and Development, 8 , 27Ð39. Muth”n, B. O., & Muth”n, L. K. (2010). Mplus userÕs guide: Statistical analysis with latent variables. Los Angeles, CA: Muth”n & Muth”n. 52 !National Association for the Education of Young Children [NAEYC]. (2009). Position statement: Developmentally appropriate practice (3rd ed.). Washington, DC: Author. NICHD Early Child Care Research Network [NICHD ECCRN]. (1999). Child outcomes when child c are center classes meet recommended standards for quality. American Journal of Public Health, 89, 1072-1077. Perry, D. , Holland, C ., Darling -Kuria, N ., & Nadiv, S . (2011). Challenging behavior and expulsion from child care: The role of mental health consultation. Zero to Three, 32, 4Ð11. Pianta, R. C., La Paro, K. M., Payne, C., Cox, M. J., & Bradley, R. (2002). The relation of kindergarten classroom environment to teacher, family, and school characteristics and child outcomes. Elementary School Journal, 102 , 225Ð238. Pianta, R. C., & Stuhlman, M. W. (2004). Conceptualizing risk in relational terms: Associations among the quality of child Ðadult relationships prior to school entry and chi ldrenÕs developmental outcomes in first grade. Educational and Child Psychology, 21 , 32Ð45. Rahn, M. (2015). Factor analysis: A short introduction . Retrieved from http://www.theanalysisfactor.com/factor -analysis -1-introduction/ Raver, C. C., & Knitzer, J. (2002). Ready to enter: What research tells policymakers about strategies to promote social and emotional school readiness among three - and four -year -old children (Promoting the Emotional Well -being of Children and Families Policy Paper No.3 ). New York, NY: National Center for Children in Poverty. Raykov, T., & Marcoulides, G. A. (2011). Introduction to psychometric theory . New York , NY : Routledge . Rimm -Kaufman, S. E., & Hamre, B. K. (2010). The role of psychological and developmental science in efforts to improve teacher quality. Teachers College Record , 112, 2988Ð3023. Schaefer, E. S., & Edgerton, M. (1985). Parent and child correlates of parental modernity. In I. E. Sigel (Ed.), Parental belief systems: The psychological consequences for children (pp. 287-318). Hillsdale, NJ: Lawrence Erlbaum Associates. Spector, P., Katwyk, P., Brannick, M, & Chen, P. (1997). When two factors donÕt reflect two constructs: How items characteristics can produce artifactual factors. Journal of Management, 32, 659Ð677. Squires, J., Bricker, D., & Twombly, E., Yockelson, S., Davis, M.S., & Kim, Y. (2002). Ages & Stages Questionnaires¨: Social -Emotional (ASQ: SE): A parent -completed, child -monitoring system for social -emotional behaviors. Baltimore , MD : Paul H. Brookes Publishing Co. 53 !Yu, C. Y., & Muth”n, B. (2002). Evaluation of model fit indices for latent variable models with categorical and continuous outcomes (Tec hnical report) . Los Angeles, CA: University of California at Los Angeles, Graduate School of Education & Information S tudies . 54 !CHAPTER 3 : STUDY 2 A Heuristic Model of Pre -service Early Childhood Teacher Preparation to Support ChildrenÕs Social -Emotional De velopment Public awareness of the value of high quality early childhood education has grown stronger in the last decade, especially as greater attention is being placed on childrenÕs school readiness. Research is quite clear that there is a link between s ocial -emotional development and school readiness, specifically, early social -emotional development predicts later school and life success (Boyd , Barnett, Bodrova, Leong, & Gomby, 2005; Cohen, Onunaku, Clothier, & Poppe, 2005; Roman o, Babchishin, Pagani, & Koh en, 2010; Shonkoff & Phillips, 2000). Social -emotional development inclu des the childÕs abilities to identify and understand his or her own feelings and the feeling of others, manage emotions, regulate behavior, and establish and sustain relationships ( National Scientific Council on the Developing Child, 2005). Failu re to promote young childrenÕs social -emotional development can lead to significant deficits in social skills, school readiness, academic success, as well as internalizing and externalizing p roblems during adolescence, juvenile delinquency, and increased risk of drug abuse and imprisonment (Bornstein, Hahn, & Haynes, 2010; Dish ion, French, & Patterson, 1995). With increasing numbers of children 0 Ð5 years old being enrolled in early childhood programs for longer hours, the role of teachers in the development of young children is becoming even more important (U.S. Department of Education, National Cente r for Education Statistics, 2015). TeachersÕ interactions with and the supports they provide t o children are indicated as vital to childrenÕs social -emotional development, and teachers are expected to share with parents the responsibility of supporting young childrenÕs social -emotional development. Further, early childhood teachers are expected to be familiar with knowledge of child development and 55 !practices to promote childrenÕs development as part of their work with children (NAEYC, 2009; Allen & Kelly, 2015). However, high rates of preschool expulsions seem to suggest teachers may be lacking the knowledge and skills needed to promote social -emotional developmen t (Gilliam, 2005; Perry , Holland, Darling -Kuira, & Nadiv, 2011). Systems of learning such as degrees obtained in institutions of higher education provide a context for early childhood teache rs to learn about child development and best practices for workin g with children, and help them translate their knowledge into practice (Allen & Kelly, 2015). Early childhood teachersÕ abilities to support the social -emotional development of all children, including those with challenging behaviors, may depend in large part on the educational preparation they receive. But, research indicating social -emotional development as one of the least focused on domains in higher education early childhood programs sugg ests pre -service teachers may not receive adequate professional development in this area of child development (Allen & Kelly, 2015; Andrew s, Buettner, Hur, & Jeon, 2015). Further, many teachers of young children and graduates of early childhood programs re port feeling unprepared to address childrenÕs challenging behaviors and effective ly support their social -emotional development (Hemmeter, Corso, & Cheatham, 2006; Hemmeter, Santos, & Ostrosky , 2008; National Survey of Early Care and Education Project Team, 2013). Findings regarding the impacts of variation in early childhood teacher education have been mixed. Although, in general, studies show that teachers with higher levels of education have more child development knowledge, better interactions with child ren, better classroom quality, and their students have better academic achievements and social outcomes than those of teachers with less or general education (Arnett, 1989; Bowman , Donovan, Burns, & the Committee on Early Childhood Pedagogy , 2000 Burchinal, Cryer, Clifford, & Howes, 2002; Goble, Horm, Atanasov, Williamson, & Young Choi, 2015; Howes, 1997; Pianta et al., 2005; 56 !Whitebook, 2003), research has not examined links between knowledge and practices despite standards in the field suggesting teachers of young children should have knowledge of child development and the strategies which support it in order to guide childrenÕs learning and development (NAEYC, 2009). Further, very little research has gone beyond examining basic associations betwe en levels of formal education (e.g. associate Õs degree, bachelorÕs degree) and teacher or classroom outcomes, limiting our understanding of teachersÕ educational experiences that contribute to their development of relevant knowledge and skills to support c hildrenÕs development. Further, because education level is often used as a proxy for knowledge, there is an assumption that the more content knowledge teachers gain the more prepared they are to meet the needs of children in their care. However, this has n ot been explicitly tested and it is not clear from the existing literature what domain -specific knowledge early childhood teachers have and how this knowledge is related to their practices. While early childhood teachersÕ knowledge of child development ma y be an important factor accounting for teachersÕ practices, it is not the only factor that may explain teacher practices related to childrenÕs social -emotional development. TeachersÕ beliefs play a vital role in what teachers do and how they interact with children in the classroom. The influence of teachersÕ beliefs on practices has been well documented, and studies suggest that teachersÕ classroom decisions are influenced by their beliefs about students and learning (Faulkner -Schne ider, 2005; Stipek & Byl er, 1997). Hence, domain -specific beliefs may be important to examine because they may determine the teaching strategies used by teachers related to the subject (Schirmer, Casbon, & Twiss, 1997). For this study, what teachers believe about emotions and the ir place in the classroom may be another significant factor related to the practices they use to support childrenÕs SE development. 57 !To go beyond the basic link between teachersÕ education levels and classroom practices, the current study focuses on (1) th e relationship between coursework and early childhood teachersÕ knowledge and beliefs related to social -emotional development, (2) relationships between teachersÕ knowledge of childrenÕs social -emotional age -based milestones and support strategies, and the ir classroom practices, and (3) the role of teachersÕ beliefs in the relationship between knowledge and practice. Below is a review of the literature examining basic associations between education and teacher practices showing mixed findings for the effec ts of education on early childhood teacher practices, and calling for a closer look at teachersÕ education to understand teachersÕ development of knowledge and practices. This is followed by a review of literature on teacher education studies in the elemen tary through high school field showing mediating and moderating factors, which may play a role in teachersÕ classroom practices, and a proposed theoretical model that may explain how these factors are related. Links Between Teacher Education and Classroom Practices In the U nites States there is no single level of education required for early care and education teachers/caregivers. Different levels of formal education such as masterÕs degrees, bachelorÕs degrees, associa te degrees, high school diplomas, and early childhood certifications are all acceptable qualifications for individuals working directly with children in early childhood settings (Morgan, 2003). Hence, res earch has primarily focused on comparing the influences of these different education leve ls on teacher outcomes. Findings from such research have been mixed, but the majority of the research shows that preschool teachersÕ formal education and specialized training predict their knowledge, classroom quality and influence childrenÕs learning and development (Arnett, 1989; Burchinal et al. , 2002; Cassidy, Buell, Pugh -Hoese, & Russell, 58 !1995; Clarke -Stewart & Gruber, 1984; Goble et al. , 2015; Howes, Whitebook, & Phillips, 1992; Pianta et al ., 2005; Snider & Fu, 1990). For example, Pianta and colleagu es found evidence indicating teachers with bachelorÕs degree, specifically in early childhood education or child development, had classrooms with more positive emotional climates and better provision for learning than teachers holding other BA degrees. Sim ilarly, Burchinal and colleagues found that teachers with BA/BS degrees in early childhood were rated substantially higher on global measures of classroom quality (including the ECERS and ITERS) and teacher sensitivity (Arnett Caregiver Interaction Scale [ CIS]) than teachers with either a CDA/Associates degree, vocational courses only, or no training. Further more , children with higher receptive language scores had teachers with more formal (specialized) education (e.g., BA/BS in early childhood) compared to children of teachers in the other three education categories (Burchinal et al. , 2002). Other recent research, however, did not find completely consistent results to those previously presented. For instance, Early and colleagues (2007) reanalyzed data fro m seven large studies of early care and education examining the relationship between teachersÕ level of education and classroom quality. Results showed null findings from four of the seven studies reviewed. The remaining three studies showed conflicting re sults, where two of the studies found that teachers with BAs or higher had higher classroom quality, while the other study indicated lower classroom quality for teachers with BAs or higher compared to those with less training. Similarly, when examining tea chersÕ years of education, level of education, degree status, and degree content (e.g., majoring in early childhood versus majoring in other topics), no significant associations were found related to measures of quality (Early et al., 2006). In another stu dy, the associations between teacher education and classroom quality disappeared once adult/child ratio and teachersÕ wages were taken into account (Phillipsen, Burchinal, Howes, & Cryer, 1997) 59 !indicating that these structural features which are associated with teacher education were the actual predictors of classroom quality, rather than teacher education itself. The moderate literature on early childhood teacher education provides an elementary understanding of the link between formal education and teach er practices. However, mixed findings suggest the need to go beyond this basic relationship to better understand specific aspects of teachersÕ preparation (e.g. , coursework) that contribute to the development of knowledge, and thus influence their practice s. Coursework, Knowledge, and Beliefs There is little empirical evidence in the field of early care and education linking pre -service teachersÕ coursework to their knowledge, despite broad associations between degrees in early childhood and teacher practices. A closer look at the relations between coursework Ña fundamental aspect of teachersÕ educational experiences Ñand knowledge and beliefs, which are expected to have an effect on teachersÕ practices is therefore needed to understand how teachers dev elop knowledge and skills to teach young children. The limited research on coursework and knowledge converges on the idea the coursework does not always improve teachersÕ knowledge (Floden & Meniketti, 2005; Koh & Neuman, 2009; Neuman & Cunningham, 2009). Research by Floden and Meniketti suggest ed that coursework in specific subject matter is significant for teachersÕ knowledge but is not always effective in developing all pre -service teachersÕ knowledge in the subject matter. Further more , although coursewo rk provides teachers with breadth of knowledge, they may lack depth in understanding concepts related to teaching the subject matter (Shulman, 1987; Whitebook, Gomby, Bellm, Sakai, & Kipnis, 2009). This suggests that coursework enhances teachersÕ knowledge about a subject (e.g., childrenÕs social and emotional development ), but does not necessarily guide teachers in how to apply the 60 !knowledge to practice in their classrooms in order to teach it to children. On the other hand, studies of pre -service and in -service populations show that teachersÕ coursework alone in early literacy development does not improve their knowledge of early literacy (Koh & Neuman, 2009; Neuman & Cunningham, 2009). However, coursework plus coaching leads to improved knowledge of early literacy, suggesting coaching is a vital component for improving knowledge (Koh & Neuman, 2009; Neuman & Cunningham, 2009). Coaching provides one -on-one support to teachers and often enhances teachersÕ knowledge of strategies to support childrenÕs learnin g in a focused area (Joyce & Showers, 2002). Research shows that teachers who receive coaching make sustained changes to their teaching practices compared to teachers who receive other types of professional development, for example, workshop -style training or formal coursework (Joyce & Showers, 2002; Neuman & Wright, 2010). It is therefore likely that coursework which includes field placements with coaching from lead teachers may have a greater effect on teachersÕ knowledge of strategies to support children Õs development than courses without field placements that include coaching. Because education is a developmental process we need to not only understand how the different aspects of education influence teacher development, i.e., their knowledge of developme ntal milestones and support strategies, their beliefs, and ultimately their practices, but also how these developmental components influence each other and interact to influence teacher, classroom, and child outcomes. Knowledge of Child Development Early c hildhood education has been viewed as a Òpractical application of the scientific field of child developmentÓ (Spodek, 1987, p. 206). Leaders in the science and practice of early child development and education indicate that teachersÕ knowledge of child dev elopment is, or should be, central to their practice and pedagogy in the classroom. For example, NAEYCÕs 61 !(2009) position statement on Developmentally Appropriate Practice in Early Childhood Programs Serving Children from Birth Through Age 8 states , ÒTeache rs who are knowledgeable about child development and learning are able to make broad predictions about what children of a particular age group typically will be like, what they typically will and will not be capable of, and what strategies and approaches w ill most likely promote their optimal learning and developmentÓ (p. 9). These standards in the field set the expectation that future professionals will know about and understand childrenÕs development across domains in order to promote childrenÕs developme nt and learning, and serve as guidelines for many early care and education programs and courses in institutes of higher education. These views and standards are redolent of the idea that knowledge of child development Ñincluding milestones of development Ñpredicts teachersÕ classroom practices; further, that a lack of such knowledge may lead to inappropriate expectations and the use of ineffective strategies to support childrenÕs development (Qi & Kaiser, 2003). Yet very few studies have examined associations between teachersÕ knowledge about childrenÕs development and their practices, and even fewer about knowledge of childrenÕs social -emotional development. As reviewed above, extant literature does suggests that more formal education Ñoften used as a proxy fo r knowledge Ñis associated with overall classroom quality and linked to better cognitive and social outcomes for children (Arnett , 1989; Bowman et al., 200 0 Cassidy et al. , 1995; Early et al ., 2006; Whitebook, 2003 ; Whitebook et al., 1989). But what early childhood teachers know about child development and whether their knowledge of child development influences their classroom practices remains a gap in the research on early care and education, despite the idea that knowledge of child development is impor tant for the field of practice. Research on teacher education for the elementary and high school (K Ð12) age range shows that 62 !teachers who are more knowledgeable about a subject matter are better able to engage students with the subject matter, and better abl e to evaluate and use instructional materials related to that subject matter (Cai, 2005; Cunningham, 1998; Fern⁄ndez , 1997; Gess -Newsome & Lederman, 1995; Ma, 1999; Sanders, Borko , & Lockard, 1993; Sowder, Phillip, Armstrong, & Schappelle, 1998; Wilkins, 2 002). Further, drawing from studies of parentsÕ child development knowledge, there is evidence supporting the expectation that greater knowledge of childrenÕs normative development predicts parenting behaviors (Haung, Caughy, Genevro, & Miller, 2005; Steve ns, 1984). Findings from the K Ð12 and parenting research linking knowledge to practices suggest that understanding childrenÕs normative social -emotional development is likely an important component of the knowledge early childhood teachers need to guide th e practices they use to support childrenÕs development in this domain. While having more formal education (e.g., a BA degree) does imply and has been shown to predict greater knowledge of child development (Goble et al. , 2015), it does not provide evidence for the depth of, or domain -specificity of, knowledge a teacher may gain while obtaining their degree. Further, the focus given to different domains in many higher education programs is generally unequal, with social -emotional development being one of the areas receiving less attention, and often taught as part of much broader courses on development across domains, rather than a domain -specific course (Andr ews, Buettner, Hur, & Jeon, 2015 ). This lack of focus on the social -emotional domain, despite having and early childhood degree, may be the reason teachers report feeling unprepared to support childrenÕs social -emotional needs. Given the general idea that knowledge influences practices, we hypothesize that early childhood teachers with more accurate knowl edge of childrenÕs SE milestones and strategies which support social -emotional development will more often use 63 !practices to support childrenÕs social -emotional development than those with less accurate knowledge. Beliefs About Child Development While knowl edge may be an important factor guiding pre -service teachersÕ practice, it is not the only factor that accounts for teachersÕ practices. Beliefs Ñdescribed as Òa proposition which may be consciously or unconsciously held, is evaluative in that it is accepte d as true by the individual, and is therefore imbued with emotive commitment; further, it serves as a guide to thought and behavior Ó (Borg, 2001, p. 186)Ñmay also explain teacher practices. It has been suggested that teachers approach teaching on impulse a nd intuition, relying very little on their professional education, and that teaching is really about common sense and practice, and less about professional knowledge (OÕBrien & Stewart, 1990) such as knowledge of childrenÕs development. It is expected that teachers will bring their beliefs to their learning and practice, which in some cases may contradict the research -based knowledge they receive from institutions of higher education, but nonetheless shape their classroom practices (Davis & Wilson, 1999; Gebel & Schrier, 2002; Johnson, 1992; Kuzborska, 2011; Richardson, Anders, Tidwell, & Lloyd, 1991). Research shows that teachersÕ decisions to use specific teaching practices in the classroom may reflect their beliefs about what is important for children to learn (Faulkner -Schne ider, 2005; La Paro, Siepak, & Scott -Little, 2009; Stipek & Byler, 1997). For example, Faulkner -Schneider (2005) found that teachers with more positive beliefs and attitudes about science and science teaching more frequently provided d ifferent types of science activities. Also, early childhood teachers who believed that early child education should be about teaching children basic skills tended to place emphasis on teaching numbers, letters, and days of the week ; teachers who believed children learn best through child -centered practices tended to place more 64 !emphasis on following childrenÕs lead by being more responsive, respectful, accepting, and nurturing (Stipek & Byler, 1997). Some studies, however, failed to find relat ionships between beliefs and practices or found incongruences between beliefs and practices (e.g., McClintic & Petty, 2015; Wen, Elicker, & McMullen, 2011; Wilcox -Herzog, 2002). These null findings and incongruences may be a result of teachersÕ self -report ing beliefs that are based on their educational knowledge rather than deep rooted beliefs (Green, 1971). Thus, when observed in the classroom teachers may demonstrate deep -rooted beliefs about teaching which may contradict their self -reported beliefs. If t eachersÕ beliefs are not addressed during their preparation this may have strong implications for what teachers do in the classroom, especially if the beliefs are contrary to recommended practices. Because beliefs can have serious implications for teachers Õ decision -making and behaviors toward children, understanding how knowledge and beliefs interact to influence practices can benefit early childhood programs seeking to better prepare their students for their future work with young children (Monroe, 2005; Silverman, 2007). This information may provide support for the need for pre -service teachers to examine their own deep -rooted beliefs, which may be misaligned with current evidence -based practices (Thomas, 2014). Theoretical Framework The purpose of the cu rrent research was to investigate relationships among coursework, knowledge, beliefs, and practices. Little is known about the factors beyond the broad construct of education Ñtypically measured as degree level Ñthat influence teachersÕ development of knowle dge and practices (Whitebook et al., 2009). Research, however shows that education level influences knowledge (Goble et al. , 2015) and standards in the field expect that knowledge of child development influences teacher practices (NAEYC, 2009). Also, there are numerous 65 !studies showing that beliefs are important to, and often predict, teacher practices, but have not examined interactions between knowledge and beliefs, nor related this relationship to practices (Pajares, 1992; Stipek & Byler, 1997; Wilkins, 2 008). Thus, we hypothesized that content -specific coursework Ñan aspect of education Ñwould influence early childhood teachersÕ knowledge of child development and support strategies as well as their beliefs. Also, that early childhood teachersÕ knowledge of child development and support strategies would directly and positively influence the use of practices (NAEYC, 2009; Wilkins, 2008), but this might depend on their beliefs. Figure 3.1 represents the conceptual model that guided the development of this study and led to the following research questions: Figure 3.1. Conceptual model . 1. What is the relationship between early childhood teachersÕ prior coursework containing social -emotional content and (a) knowledge of childrenÕs SE milestones, (b) knowledge of strategies to support SE development, and (c) beliefs about supporting and expre ssing emotions? Coursework Knowledge: Social -Emotional Milestones & Support Strategies Beliefs: Expressing & Supporting Emotions Social -Emotional Teaching Practices 66 !2. Do early childhood teachersÕ beliefs about expressing and supporting emotions in the classroom moderate the relationships between early childhood teachersÕ knowledge of childrenÕs social -emotional milestones and strategies to support social -emotional development and their use of social -emotional practices in the classroom? Methods Overview This is a cross -sectional study employing both self -report and observation methods, using a convenience sample of college students at one large public uni versity, including those majoring in Early Childhood Education (ECE) and other majors. Procedures The universityÕs IRB and the research board of the universitiesÕ Child Development Laboratories (CDLs) approved this study. Students were recruited from 10 c ourses offered through the Human Development and Family Studies Department (HDFS). These courses ranged from lower to upper level courses and included courses both in an out of the early childhood major (e.g., Child Growth and Development and Lifespan Huma n Development in Family). The researcher visited each of the 10 college courses from which students were recruited prior to distribution of the survey. Data were collected over the two semesters of the 2014/2015 academic year. A survey was distributed both online and in paper -and -pencil formats to collect data on studentsÕ educational and demographic backgrounds, knowledge of social -emotional developmental milestones and strategies to support social -emotional development, and beliefs about expressing and su pporting emotions in the classroom context. Survey items used to gather background information on participating students included current enrollment in courses, prior 67 !courses taken, prior and current work and field experiences with children, year in colleg e, major, and basic demographic data ( e.g., gender, age, family income, race, and ethnicity). Observations were also conducted with a subsample of the participants who were taking courses with field placements at the universityÕs child development laborat ories. The three courses, which each included supervised field work, in which participants were observed included Interactions with Children, Curriculum in Early Childhood Education, and Student Teaching in Early Childhood Education. Observations allowed t he researcher to gather data on pre-service teachersÕ practices without participant response bias, which may occur through self -report measures, and may have inflated associations. All participants were entered into a drawing to win 1 of 10 $25 gift cards from Amazon. Participants Survey participants. The participants in this study were undergraduate students enrolled in one or more of 10 courses through a Human Development and Family Studies department in one large public university in the Midwest. Surve y participants (n = 160) ranged in age, year in college, majors, and family income. The majority of survey respondents were female (95%), White (82%), and non -Hispanic (95%). Approximately half (55%) of students were majoring in either early child education or child developme nt; other common majors included kinesiology, nursing, and communications and participants were at various stages in the education (see Table 3.1).!Observation participants. The participants observed on classroom practices (n = 33) were a subsample of the survey participants for this study who were taking courses with supervised field placements in preschool classrooms. Observation participants were mostly juniors and seniors in college (81%), white (91%), non -Hispanic (94%), and between the ages of 68 !18 and 24 (94%). All observation participants were female and were early childhood majors (child development or early child education). !Table 3.1 Characteristics of the Study 2 Participants ÑPercentages Characteristics Survey participants (n = 160) % Observation participants (n = 33) % Race White 82 91 African American/Black 10 3 Asian/Pacific Islander 4 3 Other 4 3 Age 18Ð20 54 38 21Ð24 39 56 25+ 7 6 Major Early childhood 55 100 Communication s 8 Kinesiolog y 8 Nursing/Pre -Nursing 8 Other 21 Year in College Freshman 13 0 Sophomore 21 9 Junior 34 38 Senior 24 44 Other 7 9 Family Income Under 10,000 1 10,000Ð49,000 21 16 50,000Ð99,000 25 26 100,000+ 31 23 Would rather not say 22 35 Note: Extracted from Harewood, 2015 Measures Knowledge of social -emotional milestones and support strategies. Knowledge of social -emotional milestones and support strategies w as measured using the Knowledge of Social - 69 !Emotional Milestone Index and the Knowledge of Support Strategies Index (K -SEMS, preK version ; Harewood, 2015). The Knowledge of Social -Emotional Milestones index consists of six items and the Knowledge of Social -Emotional Support Strategies consists of nine items. These indices were determined in a previous study as valid (e.g., face and content validity were established via experts and predictive validity established via associations with obse rved social -emotional teacher practices) and moderately reliable (e.g., CronbachÕs alphas for the two indices were .61 and .51 respectively and test Ðretest and correlations of items were moderate to good) Responses on the indices were coded as 1 (correct) or 0 (incorrect) and scores for knowledge on both indices were calculated by averaging the scores across items, resulting in scores on a 0 Ð1 scale, with scores closer to 1 indicating more accurate knowledge. An example item on the Knowledge of Social -Emot ional Milestones Index is, ÒC hildren begin to compare their abilities to those of others around the age of 3 .Ó An example item on the Knowledge of Support Strategies index is Ò Giving children specific words to say to each other during social conflicts will hinder their abilities to develop their own skills for negotiation and managing conflicts.Ó Teacher beliefs about emotions in the classroom. The Teacher Beliefs About Emotions (TBAE; Hyson & Lee, 1996; Appendix B) survey was used to measure participantsÕ beliefs about expressing and supporting emotions in the classroom. The TBAE consists of 23 Likert -scale items in 6 subscales ranging from 1 ( strongly disagree ) to 5 ( strongly agree ). The six subscales include bonds (beliefs concerning adult -child connecti ons [e.g., People are better teachers if they arenÕt emotionally involved with the children ]), expressiveness (beliefs in adultsÕ expression of emotions around children [e.g., ItÕs good for a teacher to let children know when she is feeling angry ]), instru ction/ modeling (beliefs in using direct instruction and demonstration to help illustrate to children appropriate emotion expression [e.g., I think itÕs better for children 70 !to figure out how to express their feelings on their own, instead of having the teac her show them how]), Talk/Label (beliefs in helping children identify and discuss their current emotion states [e.g., When children are upset or angry about something, itÕs not the best time to talk about their feelings ]), protect (beliefs in shielding children from upsetting emotions [e.g., If a class pet died, I would not tell the children because they might become too upset ]), and display/ control (beliefs in childrenÕs ability to regulate and exhibit emotions in a socially accepta ble manner [e.g., ItÕs important for teachers to teach children socially acceptable ways of expressing their feelings ]). Item responses in each subscale are averaged with higher scores reflecting higher endorsement of the statements in a specific belief ar ea. Higher scores on all subscales except the Protect subscale were considered better. For this study, a total average score was calculated by averaging the average scores on each subscale. In the original study in which this instrument was developed, alph as for subscales ranged from .41 to .62. CronbachÕs alphas for this study were similar to those of the original study ranging from .42 to .67. Teacher practices to support preschoolersÕ social -emotional development. Observations of pre -service early childh ood teachers in the classroom context were conducted using an adapted version of the Teacher Styles Rating Scale (TSRS; Domitrovich, Cortes, & Greenberg, 2001). The original instrument was developed as a complementary tool to the CLASS and included nine to tal items, three items in each subscale (Positive Discipline, Classroom Management, Positive Emotional Climate). For this study, five items from the original scale were retained and four items were added. Two of the four added items were taken from an adap ted version of the TSRS and the remaining two items were taken from the CSEFEL Teaching Practices Inventory (2006). These added items aligned with the teaching skills taught in the HDFS coursework, which are modeled and reinforced by master teachers in par ticipantsÕ 71 !field placements; thus, the modifications were made to provide more curricular validity to the observational measure. Three of these added items were expected to create a subscale for social climate and one item was added to the positive emotion s subscale. Observer training and reliability . Three undergraduate research assistants (URAs) were trained by the researcher to use the observation instrument through live observations at the CDLs. Inter -observer agreement was assessed using percentage of agreement within one scale point. During the training period, randomly selected early childhood teachers in preschool -aged classrooms were rated independently and simultaneously by all three URAs and assessed for inter -observer agreement. Training cont inued until all URAs reached 90% agreement within one scale point on a minimum of 5 teachers. RAs observed each participant one time during their field placement and rated pre -service teachers for SE teaching practices during free play, clean -up, and trans ition. Observations lasted for 1 hour, beginning when children were dismissed to free play and ending when all children finished cleaning up. These three contexts were selected for observation because these are often child -directed activities that allow pr e-service teachers to interact freely with children, potentially eliciting a broader range of teacher behavior relevant to the observation instrument than would a more teacher directed activity. At the end of the hour, the URA rated how often pre-service t eachers used the SE practices on a scale of 1 (never) to 5 (always) using notes taken during observation to help support their ratings. Subscales of observed skills. Since the original version of the instrument was adapted, we tested the items using confirmatory factor analysis (CFA) to determine if the theoretical structure of the adapted instrument would hold. Results from the CFA revealed that the nine items in the adapted TSRS for this study did not load well together in the proposed three factor 72 !model ( !2 = p < .001; RMSEA = .39; CFI/TLI = .70/.57). Thus, an exploratory factor analysis (EFA) was conducted to identify a factor solution that best fit the data and ite m loadings. Using a varimax rotation, we examined the eigenvalues and the scree test, which suggested a 2 -factor model. When the 1 - and 2 -factor solutions were examined, test statistics did not show good fit for either, thus, the 3 -factor solution was exam ined. Chi -square test statistic, and RMSEA indicated the 3 -factor solution was the best fit for the data ( !2 = .92, RMSEA < .001) and accounted for 89% of the variance (6.049 + 1.106 + .885) / 9 = .89). The EFA revealed slight differences in the items that loaded together in thi s 3 -factor solution compared with the proposed 3-factor model analyzed for the CFA. Thus, factors were renamed to reflect the items in each factor. Factor 1, Positive and Proactive Behavior Management , included three items ( ! = .74) positive behavior management, classroom awareness, and social awareness. Factor 2 , Social -Emotional Guidance , included three items ( ! = .69) including identifies and labels emotions, identifies and corrects misbehavior, models emotion expression. Factor 3, Encouraging the Use of Social -Emotional Skills ( ! = .83), included three item s: emotion regulation, emotion expression, and social problem -solving. Average scores across items in each factor were used in analyses. Higher scores indicated pre -service teach ers used practices more frequently. Courses with social -emotional content . Background questions on the survey required participants to indicate whether they (1) had not yet taken a course, (2) were taking it currently, or (3) took it in the past for each of the 10 courses from which students were recruited. Instructors provided copies of course syllabi, which were used to d etermine the courses with social -emotional content. Based on the weekly topics and assignments listed in course syllabi, three courses had social -emotional content embedded in them Ð Child Growth and Development, Interaction Processes with Children in Grou ps, Curriculum for Early Childhood Programs. For 73 !each of these three courses, binary variables were created to indicate participants had either already taken the course or they had not; courses in which students were currently enrolled were coded as 0 beca use surveys were collected early in each semester, thus students were not expected to have gained much of the content. A variable was then created indicating the number of prior courses students took with social -emotional content. Codes for the coursework variable were: 0 ( took no prior courses with social -emotional content ); 1 ( took 1 course with social -emotional content ); 2 ( took two courses with social -emotional content ); 3 ( took all 3 courses with social -emotional content ). Preliminary Analyses Means and standard deviations of study variables are reported in Table 3.2 for the 160 survey participants as well as for the 33 observation participants. Correlations were conducted to determine associations among study variables. Age and year in college significantly correlated with predictor or outcome variables for the larger sample and were included in regression models using the larger sample (see Table 3.3). In the observation subsample, neither age nor year in college correlated with the predictor o r outcome variables (see Table 3 .4), hence, these variables were not included in the regression analyses for the subsample. 74 ! Table 3.2 Means and Standard Deviation s of Study Variables Variables Survey Sample Observation Subsample N M (SD) Min ÐMax N M (SD) Min ÐMax TBAE: Bonds 156 3.87 (.65) 2.25 Ð5.00 33 4.19 (.57) 2.75 Ð5.00 TBAE: Expressiveness 154 3.14 (.63) 1.50 Ð4.50 33 3.39 (.62) 22.00 Ð4.50 TBAE: Instruction and Modeling 156 3.72 (.78) 1.67 Ð5.00 33 4.08 (.62) 3.00 Ð5.00 TBAE: Talk/Label 156 3.83 (.57) 2.67 Ð5.00 33 4.18 (.49) 3.33 Ð5.00 TBAE: Protect 158 2.25 (.62) 1.00 Ð4.00 33 1.79 (.49) 1.00 Ð2.67 TBAE: Display/Control 158 3.92 (.64) 2.33 Ð5.00 33 4.30 (.50) 3.33 Ð5.00 TBAE: Total 151 3.46 (.36) 2.36 Ð4.38 33 3.66 (.29) 3.04 Ð4.28 Positive and Proactive Behavior Management 32 3.79 (.79) 2.00 Ð5.00 SE Guidance 30 3.58 (.77) 1.67 Ð4.67 Encouraging Use of SE Skills 31 2.97 (.95) 1.00 Ð4.67 Knowledge of SE Milestones 160 .41 (.26) 0.00 Ð1.00 30 .52 (.23) .00 Ð.83 Knowledge of Support Strategies 161 .78 (.18) .22 Ð1.00 31 .87 (.13) .56 Ð1.00 Age Range 129 1.64 (.95) 1Ð5 32 1.75 (.80) 1Ð5 Year in College 128 2.87 (1.14) 0Ð5 32 3.53 (.80) 2Ð5 No. of prior courses with social -emotional content 140 1.08 (1.19) 0Ð3 33 2.15 (.94) 0Ð3 N (listwise) 106 25 75 !!Notes: TBAE = TeachersÕ Beliefs About Emotions; SE = Social -Emotional *p < .05; ** p < .01. Table 3.3 Correlations Among Study Variables for the Entire Sample (n = 160) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 1. Bonds 1 2. Expressiveness .44** 1 3. Instruc tion/ modeling .42** .21* 1 4. Talk/label .57** .36** .59** 1 5. Protect Ð.41** Ð.23** Ð.36** Ð.48** 1 6. Display/ control .59** .25** .47** .65** Ð.57** 1 7. TBAE: Total Ave .78** .60** .74** .81** Ð.32** .71** 1 8. Age .24** .12 .07 .12 .00 .17 .21** 1 9. Year in college .30** .17 .27** .29** Ð.26** .34** .32** .53** 1 10. Knowledge of SE Milestones .35** .20* .31** .41** Ð.32** .41** .42** .31** .20** 1 11. Knowledge of SE strategies .40** .24** .37** .45** Ð.27** .37** .47** .16 .27** .47** 1 12. Courses with SE content .49** .40** .46** .51** Ð.37** .43** .60** .37** .58** .42** .48** 76 !Table 3.4 Correlations Among Study Variable s for Subsample (n = 33) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 1. Bonds 1 2. Expressiveness .49** 1 3. Instruction/ Modeling .27 .04 1 4. Talk/Label .56** .13 .43* 1 5. Protect Ð.42* Ð.19 Ð.29 Ð.44* 1 6. Di splay/Control .54** .25 .43* .73** Ð.58* 1 7. Total: TBAE Ave .78** .58** .62** .74** Ð.31 .74** 1 8. Age Range .19 Ð.10 .00 Ð.09 Ð.01 .03 .01 1 9. Year in College .16 Ð.11 .12 .03 Ð.16 .25 .09 .76** 1 10. PPBM .01 .13 .18 .07 Ð.01 .13 .16 .11 .32 1 11. SEG .17 Ð.04 Ð.07 .00 .02 .11 .06 .26 .43* .42* 1 12. EUSES .05 Ð.07 .25 .18 Ð.06 .10 .14 .13 .19 .61** .53** 1 13. K-SEM .00 .14 .43* .33 Ð.18 .39* .34 .08 .19 .38* Ð.13 .11 1 14. K-SES .02 Ð.06 .07 .10 .26 Ð.21 .05 Ð.01 .04 .36* .19 .38* .18 Notes: PPBM = Positive and Proactive Behavior Management; SEG = Social -Emotional Guidance ; EUSES = Encouraging the Use of Social -Emotional Skills; K-SEM = Index of Knowledge of Social -Emotional Milestones; K -SES = Index of Knowledge of S ocial -Emotional Support Strategies *p < .05; **p < .01. 77 !Results Relationships Between Coursework and Knowledge and Beliefs What is the relationship between courses with SE content and pre -service teachersÕ (a) knowledge of SE milestones, (b) knowledge of SE support strategies (c) beliefs about expressing and supporting emotions? Multiple regression analyses were conducted to examine the influence of coursework on knowledge and beliefs using the larger sample ( n = 160). Results showed that domain -specific coursework predicted both domain -specific knowledge and beliefs. Relationships between coursework and knowledge. Multiple regressions showed participantsÕ prior coursework with SE content predicted knowledge of social -emotional milestones ( ! = .38, p < .001; effect size = .34) as well as their knowledge of support strategies (! = .41, p < .001; effect size = .35) even when ac counting for year in college and age. Results indicate that participants who took more courses with social -emotional content had more accurate knowledge about childrenÕs social -emotional milestones and more accurate knowledge of strategies to support child renÕs social -emotional development. Coursework also explained a significant portion of the variance in the Knowledge of Milestones scores (R 2 = .21, F(3, 106) = 9.46, p < .001) and the Knowledge of Support Strategies scores (R 2 = .23, F(3, 112) = 11.29, p < .001). Students who took all courses with SE content were all in an early childhood major. Comparisons of average knowledge scores based on number of prior courses taken. Post hoc analyses showed significant differences between the knowledge of milestone s scores for students who took all three courses with social -emotional content and those who did not take any of these courses. There were no differences between scores for those who had taken only one or two courses and those who had not taken any courses . The mean scores on the knowledge of 78 !milestones scale for students who took courses with SE content ranged between .32 for those who took one course to .65 for those who took three courses and compared to .37 for students who took no courses with SE conte nt. Mean scores on the knowledge of support strategies index were generally higher than scores on the social -emotional milestones index. There were significant differences between the knowledge of support strategies scores of participants who took two or more social -emotional content -courses scoring between .88 and .90 compared to scores of .72 for those who did not take courses with social -emotional content. Relationships between coursework and beliefs. Participants who took more courses with social -emoti onal content reported higher endorsements of beliefs about expressing and supporting emotions. Results related to each of the beliefs subscales on the TBAE scale showed that coursework had a significant and positive effect on all belief areas: Bonds (! = .52, p < .001; effect size = .37); Expressiveness (! = .40; p < .01; effect size = .34); Instruction/Modeling (! = .38, p < .001; effect size = .32); Talk/Label (! = .37, p < .001; effect size = .31); Display/Control (! = .30, p < .01; effect size = .25) ; and Total Beliefs about Expressing and Supporting Emotions ( ! = .59, p < .001; effect size = .47). The only exception was in the Protect belief area, where coursework was negatively related to this subscale: Protect ( ! = Ð.29 p < .001; effect size = .24). These results suggest that participants with more social -emotional coursework believed emotions should be expressed and supported in the classroom and children should not be protected from experiencing or expressing emotions. Mod eration Effects of Beliefs on the Relationships Between Knowledge and Practices Do early childhood teachersÕ beliefs about expressing and supporting emotions in the classroom moderate the relationships between early childhood teachersÕ knowledge of childr enÕs 79 !SE milestones and strategies to support SE development and their use of SE practices in the classroom? Basic regression s were conducted in SPSS using the subsample of participants (n = 33) to test the effects of pre -service teachersÕ knowledge and bel iefs on each of the three outcomes (Positive and Proactive Behavior Management, Social -Emotional Guidance, and Encouraging the use of Social -Emotional Skills). Independent variables (knowledge of milestones, knowledge of support strategies, and beliefs) we re centered and the knowledge by beliefs interaction terms were computed for knowledge of milestones and total beliefs, as well as knowledge of milestones and each belief subscale (Aiken & West, 1991). This process was repeated for knowledge of support str ategies. The two independent variables and the interactions for each model were entered into a regression model sequentially. Positive and proactive behavior management. Knowledge of milestones . There were no main effects for knowledge of milestones, or any of the beliefs subscales on positive and proactive behavior management practices. There was a significant interaction between teachersÕ knowledge of milestones and instruction/modeling beliefs related to positive and proactive behavior management practic es ( ! = .35, p = .05; effect size = .50). the interaction between knowledge of milestones and instruction /modeling beliefs accounted for a significant portion of the variance in positive and proactive behavior management practices ( R2 = .26, F(2, 26) = 2.94, p = .05). As seen in Figure 3.2, results show participants with higher endorsements of instruction/modeling beliefs and more accurate social -emotional knowledge of milestones used developmentally appropriate positive and proactive behavior management practices more frequently. Converse ly, participants with higher endorsements of instruction/modeling beliefs and less accurate social -emotional knowledge of 80 !milestones used developmentally appropriate positive and proactive behavior management practices less frequently. Other models includi ng two -way interactions between knowledge of support strategies and beliefs were not significant related to positive and proactive behavior management practices. Figure 3.2. Instruction/modeling beliefs moderated the relationship between knowledge of milestones and behavior management practices . Encouraging the use of social -emotional skills. Knowledge of support strategies. There was a main effect for knowledge of support strategies on practices encouraging the use of social -emotional skills ( ! = .38, p = .05; effect size = .35), but not for any of the beliefs subscales. There was a significant interaction between teachersÕ knowledge of support strategies and instruction/modeling beliefs related to practices encouraging the use of social -emotional skills (! = .38, p = .05; effect size = .37). the interaction between knowledge of milestones and instruction /modeling beliefs accounted for a significant portion of the variance in practices encouraging the use of social -emotional skills ( R2 = .31, F(2, 26) = 3.69, p < .05). Results in Figure 3.3 show that participants with higher endorsements of 1 2 3 4 5 Less Knowledge More Knowledge Behavior Management Practices Low I&M Beliefs High I&M Beliefs 81 !instruction/modeling beliefs and more accurate knowledge of strategies to support social -emotional development used practices encouraging the use of social -emotional s kills more frequently than those with less accurate knowledge and higher endorsements of instruction/modeling beliefs. Other models including two -way interactions between knowledge of milestones and beliefs were not significant related to practices encour aging the use of social -emotional skills. Figure 3. 3. Instruction/ modeling beliefs moderated the relationship between knowledge of support strategies and practices encouraging the use of social -emotional skills . Social -emotional guidance. None of the tested models related to social -emotional guidance practices were significant. Discussion The current study used a sample of undergraduate students at one university to identify relationships between coursework, knowledge, beliefs, and practices. Results revealed that taking courses with domain -specific content is an important factor related to pre -service teachersÕ development of domain -specific knowledge and beliefs. Using a smaller sample (n = 33), results 1 2 3 4 5 Less Knowledge More Knowledge Encouraging Use of SE skills Low I&M Beliefs High I&M Beliefs 82 !showed that domain specific knowledge is an imp ortant factor in pre -service teachers use of domain specific practices, Further, together domain -specific knowledge and beliefs are significant factors related to pre -service teachers use of domain -specific practices. Relationships Among Coursework, Knowle dge, and B eliefs The findings from this study are consistent with studies broadly defining education as degree level which imply that educational experiences are important to the development of teachersÕ knowledge and skills (Goble et al. , 2015). Also, taking courses with domain specific content contributes to domain specific knowledge (Floden & Meniketti, 2005) as well as beliefs. This study showed that participantsÕ who took HDFS courses with social -emotional content had more accurate knowledge and sig nificantly higher endorsements of beliefs about expressing and supporting emotions than participants who did not take these courses. These findings align with research indicating that education predicts greater child development knowledge (Goble et al. , 2015). Also, consistent with work by Floden and Meniketti, these findings suggest that taking courses with domain specific content contributes to pre -service teachersÕ to domain -specific child development knowledge and beliefs. Post hoc analyses for differe nces in knowledge for those who took domain -related coursework showed that although participants who took courses with SE content had higher scores on the knowledge of milestones scale compared to those who did not take these courses, these scores only dif fered significantly when students took cumulative courses with embedded social -emotional content and these students only had a 65% accuracy on the knowledge of milestones scale. When students took only one course with embedded social -emotional content thei r scores did not differ significantly from students who did not take any courses with social -emotional content. It could be that since SE content was embedded in courses, that focus on SE 83 !development and/or age -based milestones were underemphasized (Allen & Kelly, 2015; Andrew s, Buettner, Hur, & Jeon , 2015) leaving participants to make educated guesses on some of the knowledge items. Further, when social -emotional content is embedded in other courses students would have to take multiple courses in order to gain a depth of knowledge on social -emotional development. Embedding social -emotional content in other courses could therefore leave students feeling unprepared to meet the social -emotional needs of children (Hem meter et al., 2006; National Survey of Earl y Care and Education Project Team , 2013. A stand -alone course in SE development may provide pre -service teachers with more thorough and concentrated knowledge of SE development. Future studies should test the effects of a stand -alone SE course on pre -servi ce teachersÕ SE knowledge. Also, studies should compare the effects of a stand -alone course and courses with SE content embedded in them on studentÕs knowledge of childrenÕs social -emotional development. Alternatively, not all courses with social -emotional content had a field placement which allowed students to receive coaching from highly qualified lead teachers. Thus, it is possible that when students took courses having both social -emotional content and field placements their knowledge was more accurate (Joyce & Showers, 200 2; Koh & Neuman, 2009; Ne uman & Cunningham, 2009; Ne uman & Wright, 2010). Relationships Among Knowledge, Beliefs, and Practices Main effects of knowledge on practices. When pre -service teachers had more accurate knowledge of social -emotional support strategies they more frequently used social -emotional practices in the classroom. This finding validates standards in the field indicating teachersÕ should have knowledge of strategies, which g uide their developmentally appropriate practices to support childrenÕs learning and development (NAEYC, 2009). Specifically this study found that pre-service teachersÕ knowledge of strategies to support social -emotional development affe cted 84 !their application of practices encouraging the use of social -emotional skills. It could be that pre -service teachers with more accurate knowledge of strategies are more confident in how to encourage children to use their skills and thus more often res ponded to children in situations requiring emotional regulation or problem solving, and that those who had less accurate knowledge may have avoided such situations. Beliefs moderated relationships between knowledge and practices. When pre -service teachers had more accurate knowledge of child development and support strategies, but less endorsements of beliefs about expressing and supporting emotions, they used SE practices less frequently than when they had more accurate knowledge and high endorsements of beliefs. These findings support research indicating beliefs are important to teachersÕ decision -making and help to explain teachersÕ practices (Faulkner -Schneider, 2005; Hur, Buettner, & Jeon, 2013; La Paro , Siepak, & Scott -Little, 2009; Pajares, 1992; Sti pek & Byler, 1997). Figures 3.2 and 3.3 show that the accuracy of pre -service teachersÕ knowledge of child development and support strategies interacted with their beliefs about expressing and supporting emotions to predict their use of emotionally support ive practices. While not specifically testing change, the results in this study imply that solely addressing knowledge is not a complete path to educating pre -service teachers of young children. While knowledge is necessary, it is not sufficient. If pre -service teachers come into programs of early childhood with beliefs that are not developmentally supportive for children, and only their knowledge, not beliefs, is influenced by their educational experiences, we may not be helping them to change their practi ce. Tatto and Coupland (2003) assert that while college courses are often used as the primary context for changes in beliefs, it is premature to think that simply teaching students content knowledge (e.g., theories and empirical research ) on a specific top ic will effectively influence their beliefs (Tatto & Coupland, 2003). 85 !Instead, college studentsÕ beliefs about a specific subject matter are more likely to align with the content knowledge provided in the preparation program if there are opportunities for self- or peer -mediated reflection on existing beliefs (Thomas, 2014). Self -reflective assignments could help teachers identify biases or misconceptions gained through their upbringing and societal experiences and if not change their beliefs at least help t hem to be aware that they have biases that are not aligned with recommended practices for working with children (Thomas, 2014). Similar to other research (McClintic & Petty, 2015; Wen et al. , 2011; Wilcox -Herzog, 2002), beliefs themselves did not predict practices in this study when knowledge was accounted for in the model. These findings may be linked to the idea that teachersÕ self -reported beliefs are based on their knowledge of a subject (Green, 1971), thus, making beliefs appear insignificant once the relationship between knowledge and practices was accounted for. Pre - and post - term studies could be used to examine changes in pre -service teachersÕ domain -specific knowledge and beliefs and consider the effects of other educational factors such as the p rogramsÕ goals, and instructorsÕ beliefs to see how these might be related to such changes. Also, future studies should conduct path analyses from coursework to knowledge and beliefs and then to practices to better understand these relationships. While thi s study provides initial associations between domain -specific coursework and one component of child development knowledge, future studies should include other aspects of pre-service teachersÕ education such as field experiences. This study calls attention to the need for researchers to conduct more domain -specific studies to further understand what aspects of pre-service teachersÕ educational experiences produce effective teachers of young children and inform policy and practice. Although not included in th is study, other components of child 86 !development knowledge should be included in future research to provide a more complete understanding of teachersÕ development of knowledge and skills. Limitations This study was conducted using a small sample of pre -service early childhood teachers at one university in the Midwest. Participants were mostly white, middleclass females, thus, generalization of results are limited in terms of educational context, race, gender, and socioeconomic status. Surveys and observatio ns were completed on average 6 weeks apart; studentsÕ interim exposure to coursework and field experiences could have influenced changes in knowledge of milestones and support strategies, as well as their interactions with children in the classroom. Future studies should lessen the time between survey completion and observations to minimize the effects of coursework exposure. This study was based on the assumption that students would be able to recall information and respond to questions based on their kno wledge of the specific topic gained through their coursework. Thus, results may not reflect an accurate association between knowledge and practices or beliefs and practices; instead, associations between concurrent knowledge and practices might be stronger than reflected in the current results. The instrument used in this study to measure knowledge of SE milestones included only six items that minimally covered preschoolersÕ social -emotional milestones, thus caution should be taken in generalizing this as a thorough measure of social -emotional milestones. Further, the reliabilities for the two Knowledge of Social -Emotional indices were not strong, which suggests these indices may not consistently measure pre -service teachersÕ knowledge of social -emotional development and strategies, making it difficult to replicate this study. 87 !Some potential relationships were not explored in this study but could be of interest in future studies. ParticipantsÕ characteristics which may be related to associations between knowledge, beliefs, and practices Ñsuch as IQ, GPA, and previous experiences working with preschool -aged children Ñwere not included as potential influences of participantsÕ knowledge and beliefs, but may have influenced their responses on the K -SEMS and the bel iefs measures. That is, participants could have answered more accurately because they were more intelligent, they performed better in college, they have experiences working with children of this age group from which they are basing their knowledge, beliefs , and practices. Also, participantsÕ field experiences and support from lead teachers may be important factors associated with relationships among knowledge, beliefs, and practices. Including these variables in future studies could provide a better underst anding of the relationships among knowledge, beliefs, and practices. Conclusions This study contributes to the early childhood education literature by examining questions about the links between pre -services teachersÕ coursework, knowledge, and practices, links that so far has only been implied or expected, but not explicitly tested. The findings from this study indicate moderate effects of prior coursework on pre -service teachersÕ knowledge and beliefs, and moderate effects of prior knowledge and beliefs o n observed practices. This current study contributes to the existing literature by providing evidence for the importance of professional development as a way to enhance teachersÕ knowledge about child development and strategies to support development. Thi s study provides evidence that pre -service teachersÕ knowledge and beliefs affect what they do in the classroom to support childrenÕs development, and that coursework is an important 88 !factor affecting teachersÕ domain -specific beliefs, not just their knowle dge of development. Thus, it may benefit higher education programs to include reflective assignments in courses as a way to help teachers become aware of their own beliefs, including any biases they may hold about child development, therefore becoming more able to address them explicitly. Results from this study indicate the need for further research examining relationships between educational experiences and educational outcomes in order to inform practices in early childhood teacher preparation programs, and to inform policy related to the baseline educational requirements for teachers of young children. Knowledge of developmental milestones, in addition to other components of knowledge, should be associated with child outcomes to determine what knowledge is most essential to teachersÕ effectiveness in the classroom. 89 ! APPENDIX 90 ! Appendix For each sentence, circle the number that describes HOW TRUE it is for you. Strongly Disagree Disagree Neither agree or disagree Agree Strongly Agree 1. People are better teachers if they arenÕt emotionally involved with the children. 1 2 3 4 5 2. ItÕs good to hug and touch children affectionately throughout the day. 1 2 3 4 5 3. Teachers avoid being physically affectionate or ÒhuggyÓ with the children 1 2 3 4 5 4. Children need to feel emotionally close to their teachers. 1 2 3 4 5 5. ItÕs good for a teacher to let children know when she is feeling angry. 1 2 3 4 5 6. Teachers should Òlet their feelings outÓ in the classroom. 1 2 3 4 5 7. Teachers should try hard no to show when they are upset with childrenÕs behavior. 1 2 3 4 5 8. Teachers should constantly show the children how much they love them 1 2 3 4 5 9. When a child is angry because another child wonÕt share a toy, the teacher should tell the child exactly what words she/he could use to express her/his feelings. 1 2 3 4 5 10. Teachers should avoid showing children how to express their feelings. 1 2 3 4 5 11. I think itÕs better for children to figure out how to express their feelings on their own, instead of having the teacher show them how. 1 2 3 4 5 12. When a child is upset about something, teachers should try to put into words how he or she is feeling. 1 2 3 4 5 13. Teachers should often label the childrenÕs feelings for them, ( e.g., ÒYou seem worried about our trip to 1 2 3 4 5 Teacher Beliefs About Emotions 91 !the swimming poolÓ). 14. When children are upset or angry about something, itÕs not the best time to talk about their feelings. 1 2 3 4 5 15. I believe that some teachers spend too much time talking to childre n about their feelings. 1 2 3 4 5 16. Teachers should spend a lot of time talking to children about why they feel the way they do. 1 2 3 4 5 17. Children ages 3 Ð5 years are too young for me to discuss the causes of their feelings with them. 1 2 3 4 5 18. Teacher s should not read children stories that might make them sad or worried. 1 2 3 4 5 19. Children should be taken to funerals and other family events, even if they might feel sad or upset as a result. 1 2 3 4 5 20. If a class pet died, I would not tell the childr en because they might become too upset. 1 2 3 4 5 21. Children ages 3 Ð5 years are really not ready to control the way they express their feelings. 1 2 3 4 5 22. Children ages 3 Ð5 years are really too young to display their feelings in Òsocially acceptableÓ way s. 1 2 3 4 5 23. ItÕs important for teachers to teach children socially acceptable ways of expressing their feelings. 1 2 3 4 5 THANK YOU SO MUCH FOR COMPLETING THIS QUESTIONNAIRE. 92 ! REFERENCES 93 !REFERENCES Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park: Sage. Allen, L. R., & Kelly, B. B . (2015). Transforming the workforce for children birth through age 8: A unifying foundation. Washington , DC: National Academies Press. Andrews, D., Buettne r, C., Hur, E., & Jeon, L. (2015 ). What are we teaching the teachers? Child development curricula in US higher education . Paper symposium presented at the Society for Research in Child Development Biennial Conference, Philadelphia, PA. Arnett, J . (1989). Caregivers in day -care centers: Does training matter ? Journal of Applied Developmental Psychology, 10 , 541Ð552. Bornstein, M., Hahn, C., & Haynes, O. M. (2010). Social competence, externalizing, and internalizing behavioral adjustment from early childhood through early adolescence: Developmental cascades. Development and Psychopathology, 22 , 717Ð735. Bowman, B., Donovan, M., Burns, S., & the Committee on Early Childhood Pedagogy of the National R esearch Council . (2000). Eager to learn: Educating o ur preschoolers . Washington, DC: National Academy Press. Borg, M. (2001). Key con cepts in ELT. ELT Journal, 55 , 186Ð188. Boyd, J., Barnett, S., Bodrova, E., Leong, D., & Gomby, D. (2005). Policy report: Promoting childrenÕs social emotional development through preschool education. National Institute for Early Education Research (NIEER) New Brunswick: NJ . Burchinal, M. R., Cryer, D., Clifford, R. M., & Howes, C. (2002). Caregiver training and classroom quality in child care centers. Applied Developmental Science, 6 , 2Ð11. Cai, J. (2005 ). U.S. and Chinese teachersÕ constructing, knowing, and evaluating representations to teach mathematics. Mathematical Thinking and Learning: An International Journal, 7, 135Ð169. Cassidy, D., Buell, M., Pugh -Hoese, S., & Russell, S. (1995). The effects of education on child care teachersÕ beliefs and classroom quality: Year one evaluation of the TEACH early childhood associate degree scholarship program. Early Childhood Research Quarterly, 10, 171Ð183. Clarke -Stewart, K. A., & Gruber, C. (1984). Day care forms and features. In R. C. Ainslie (Ed.), The child and the day care setting (pp. 35Ð62). New York , NY : Praeger Special Studies. 94 !Cohen, J., Onunaku, N., Clothier, S., & Poppe, J. (2005 , September). Helping young children succeed: Strategies to promote early childhood social and emotional development. Washington , DC: Zero to Three and the National Conference of State Legislatures. CSEFEL Inventory of practices for promoting childrenÕs social and emotional competence . (2006). Available from: www.vanderbilt.edu/csefel. Center on the Social and Emotional Foundations for Early Learning (CSEFEL). Cunningham, C. M. (1998). The effect of teachersÕ sociological understanding of science (SUS) on curricular innovation. Research in Science Education, 28, 243Ð257. Davis, M. M., & Wilson, E. K. (1999). A Title 1 teacherÕs beliefs, decision -making, and instruction at the third and seventh grade levels. Reading Research and Instruction , 38, 290Ð299. Dishion, T. J., French, D. C., & Patterson, G. R. (1995). The development and ecology of antisocial behavior. In D. Cicchetti & D. J. Cohen (Eds.) , Developmental psychopathology (Vol. 2, pp. 421Ð471). New York , NY : John Wiley & Sons. Domitrovich, C., Cortes, R. C., & Greenberg, M. T. (2 001). Head Start Competence Scale technical report . Unpublished manuscript, Pennsylvania State University. Early, D., Bryant, D., Pianta, R., Clifford, R., Burchinal , M., Ritchie, S., É Barba rin, O. (2006) . Are teacher education, major, and credentials related to classroom quality and childrenÕs academic gain in pre -kindergarten? Early Childhood Research Quarterly 21, 174Ð195. Early, D. M., Maxwell, K. L., Burchinal, M., Alva, S., Bender, R. H., Bryant, D., É Zill, N. (2007). TeachersÕ education, classroom quality, and young childrenÕs academic skills: Results from seven stud ies of preschool programs. Child Development, 78 , 558Ð580. Faulkner -Schneider, L. (2005). Child care teachersÕ attitudes, beliefs, and knowledge regarding science and the impact of early childhood learning opportunities . (Unpublished thesis ). Oklahoma State University , Norman Oklahoma. Fern⁄ndez, E. (1997). The Òs tandards -likeÓ role of teachersÕ mathematical knowledge in responding to unanticipated student observations. Paper presented at the annual meetings of the American Educational Research Association, Chicago, IL. Floden, R., & Meniketti, M. (2005). Research on the effects of coursework in the arts and sciences and in the foundati ons of education. In M. Cochran -Smith & K. M. Zeichner (Eds.), Studying t eacher education (pp. 261Ð308). Mahwah, NJ: Erlbaum . Gebel, A., & Schrier, L. L. (2002). Spanish language teachersÕ beliefs and practices about reading in a second language. In J. H. Sullivan (Ed.), Literacy and the second language learner (pp. 85Ð109). Charlotte, NC: Information Age Publishing. 95 !Gess-Newsome, J. & Lederman, N. G. (1995). Biology teachersÕ perceptions of subject matter structure and its relationship to classroom practice. Journal of Research in Science Teaching, 32 , 301Ð325. Gilliam, W. (2005). Prekindergarteners left behind: Expulsion rates in state prekindergarten systems. New York , NY : The Foundation for Child Development. Goble, C. B., Horm, D. M., Atanasov, A. M., Williamson, A. C., & Young Choi, J. (2015). Knowledge and beliefs of early childhood education students at different levels of professional preparation. Journal of Early Childhood Teacher Education, 36, 211Ð231. doi: 10.1080/10901027.2015.1062831 Green, T. E. (1971). The activities of teaching. New York , NY : McGraw -Hill. Haung, K., Caughy, M., Genevro, J., & Miller, T. (2005). Maternal knowledge of child development and quality of parenting among Whit e, African -American and Hispanic mothers. Applied Developmental Psychology, 26 , 149Ð170. Hemmeter, M. L., Corso, R., & Cheatham, G. (2006). Issues in addressing challenging behaviors in young children: A national survey of early childhood educators . Paper presented at the Conference on Research Innovations in Early Intervention, San Diego, CA. Hemmeter, M., Santos, R., & Ostrosky, M. (2008). Preparing early childhood educators to address young childrenÕs social -emotional development and challenging behavior : A survey of hig her education programs in nine s tates. Journal of E arly Intervention, 30 , 321Ð340. Howes, C. (1997). ChildrenÕs experiences in center -based child care as a function of teacher background and adult Ðchild ratio. Merrill -Palmer Quarterly, 43 , 404Ð425. Howes, C., Whitebook, M ., & Phillips, D. A . (1992), Thresholds of Quality: Implications for the Social Development of Children in Center -based Child Care. Child Development, 63, 449Ð460. Johnson, K. E. (1992). The relationship between teachersÕ beliefs and practices during literacy instruction for non -native speakers of English. Journal of Reading Behavior , 24, 83Ð108. Joyce, B., & Showers, B. (2002). Student achievement through staff deve lopment (3rd ed.). Alexandria, VA: Association for Supervision and Curriculum Development. Koh, S., & Neuman, S. B. (2009). The impact of professional development on family child care: A practice -based approach. Early Education and Development, 20 , 537Ð562. Kuzborska, I. (2011). Links between teachersÕ beliefs and practices and research on reading. Reading in a Foreign Language, 23 , 102Ð128. 96 !La Paro, K., Siepak, K., & Scott -Little, C. (2009). Assessing beliefs of preservice early childhood education teacher s using Q -sort methodology. Journal of Early Childhood Teacher Education, 30, 22-36. doi: 10.1080/10901020802667805 Ma, L. (1999). Knowing and teaching elementary mathematics: T eacher sÕ understanding of fundamental mathematics in China and the United States. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. McClintic, S., & Petty, K. (2015). Exploring early childhood teachersÕ beliefs and practices about preschool outdoor play: A qualitative study. Journal of Early Childhood Teacher Education, 36, 24Ð43. doi: 10.1080/10901027.2014.997844 Monroe, C. R. (2005). Why are Òbad boysÓ always Black? Causes of disproportionality in school discipline and recommendations for change. The Clearing House, 79 , 45Ð50. Morgan, G. G. (2003). Staffing. In D. Cryer & R. M. Clifford (Eds.), Childhood education and care in the USA (pp. 58Ð72). Baltimore, MA: Brookes Publishing. National Association for the Education of Young Children [NAEYC]. (2009). Position statement: Developmentally appropriate practice (3rd ed.). Wa shington, DC: Author. National Scientific Council on the Developing Child . (2005). Excessive stress disrupts the architecture of the developing b rain (Working Paper n o. 3). Retrieved from http://developingchild.harvard.edu/wp -content/uploads/2005/05/Stress_Disrupts_Architecture_Developing_Brain -1.pdf National Survey of Early Care and Education Pr oject Team. (2013). Number and c haracteristics of early care and education (ECE) teachers and caregivers: Initial findings from the National Survey of E arly Care and Education (NSECE) (OPRE Report #2013 -38). Washington DC: Office of Planning, Research and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Servic es. Neuman, S. B., & Cunningham, L. (2009). The impact of professional development and coaching on early language and literacy practices. American Educational Research Journal, 46 , 532Ð566. Neuman, S., & Wright , T. (2010). Promoting language and literacy development for early childhood educators. The Elementary School Journal, 111 , 63Ð86. OÕBrien, D., & Stewart, R. (1990) Preservices teachersÕ perspectives on why every teacher is not a teacher of reading: A qualitati ve analysis. Journal of Reading Behavior, 22, 101-129. Pajares, M. F. (1992). Teacher beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62 , 307Ð332. Perry, D. , Holland, C ., Darling -Kuria, N ., & Nadiv, S . (2011). Challenging behavior and expulsion from child care: The role of mental health consultation. Zero to Three, 32, 4Ð11. 97 !Phillipsen, L. C., Burchinal, M. R., Howes, C., & Cryer, D. (1997). The prediction of process quality from structural features of child care. Early Childhood Research Quarterly, 12 , 281Ð303. Pianta, R., Howes, C., Burchinal, M., Bryant, D., Clifford, R., Early, D., & Barbarin, O. (2005). Features of pre kindergarten programs, classrooms, and teachers: Do they predict observed classroom quality and teacher Ðchild interactions? Applied Developmental Science, 9, 144Ð159. Qi, C. H., & Kaiser, A. P. (2003). Behavior problems of preschool children from low -incom e families: Review of the literature. Topics in Early Childhood Special Education, 23 , 188Ð216. Richardson, V., Anders, P., Tidwell, D., & Lloyd, C. (1991). The relationship between teachersÕ beliefs and practices in reading comprehension instruction. American Educational Research Journal , 28, 559Ð586. Romano, E., Babchishin, L., Pagani, L., & Kohen, D. (2010). School readiness and later achievement: Replication and extension using a nationwide Canadian survey. Developmental Psychology, 46 , 995Ð1007. doi:10.1037/a0018880 Sanders, L. R., Borko, H., & Lockard, J. D. (1993). Secondary science teachersÕ k nowledge base when teaching science courses in and out of their area of certification. Journal of Research in Science Teaching, 30, 723Ð736. Schirmer, B., Casbon, J., & Twiss, L. (1997). Diverse lear ners in the classroom: Teacher b eliefs about learning: What happens when the child doesnÕt fit the schema? The Reading Teacher, 50 , 690Ð692. Shonkoff, J. P., & Phillips, D. A. (Eds.). (2000). From ne urons to nei ghborhoods: The science of early childhood development . Washington, DC: National Academy Press. Shulman, L. (1987). Knowledge and Teaching: Foundations of the new reform. Harvard Educational Review , 57, 1Ð21. Silverman, J. C. (2007). Epistemological beliefs and attitudes toward inclusion in pre -service teachers. Teacher Education and Special Education, 30 , 42Ð51 Snider, M., & Fu, V. (1990). The effects of specialized education and job experience on early childhood teacherÕs knowledge of developmentally appropriate practice. Early Childhood Research Quarterly, 5 , 69Ð78. Sowder, J. T., Phillip, R. A., Armstrong, B. E., & Schappelle, B. P. (1998). Middle-grade teachersÕ mathematical knowledge and its relationship to instruction. Albany, NY: State University of New York Press. Spodek, B. (1987). Thought processes underlying preschool teachers' classroom decisions. Early Child Care and Education, 28 , 197Ð208. 98 !Stevens, J. (1984). Child development knowledge and parenting skills. Family Relations, 33 , 237Ð244. Stipek, D., & Byler, P. (1997). Early childhood education teachers: Do they practice what they preach? Early Childhood Research Quarterly, 12 , 305Ð325. Tatto, M., & Coupland, D. (2003). Teacher education and teachersÕ beliefs: Theoretical and measurement concerns. Greenwich, CT : Information Age Publishing. Thomas, C. (2014). Considering the impact of pre -service teachersÕ beliefs on future practice. Intervention in School and Clinic, 49 , 230Ð236. doi: 10.1177/1053451213509490 U.S. Department of Education, National Center for Education Statistics. (2015). The Condition of Education 2015 (NCES 2015 -144), Preprimary Enrollment . Wen, X., Elicker, J., & McMullen, M. (2011). Early childhood teachers Õ curriculum beliefs: Are they consistent with observed classroom practices? Early Education and Deve lopment, 22, 945Ð969. Whitebook, M. (2003). Early education quality: Higher teacher qualifications for better learning environments ÑA review of literature. Berkeley , CA: Institute of Industrial Relations, University of California. Whitebook, M., Gomby, D., Bellm, D., Sakai, L., & Kipnis, F. (2009). Preparing teachers of young children: The current state of knowledge, and a blueprint for the future. Berkeley, CA: Center for the Study of Child Care Employment. Wilcox -Herzog, A. (2002). Is there a link between teachersÕ beliefs and behaviors? Early Education & Development, 13 , 81Ð106. Wilkins, J. L. M. (2002). The impact of teachersÕ content knowledge and attitudes on instructional beliefs and practices. Proceedings of the Annual Meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education, 24 , 10. Wilkins, J. L. M. (2008). The relationship among elementary teachersÕ content knowledge, attitudes, beliefs, and practices. Journal of Math Teacher Educa tion, 11, 139-164. doi: 10.1007/s10857 -007-9068-2 99 !CHAPTER 4 : INTEGRATIVE CONCLUSION The goals of this study were to (1) develop and validate an adequate tool to assess pre -service ECEPs knowledge of SE milestones and support strategies and (2) to build a heuristic model of ECEP preparation. High -quality care and education has been a component of early education reform over the past few years, fueling debates over the baseline educational requirements for early childhood teachers working with young children . Thus far, these debates have been based on mixed findings related to the effect of education on classroom quality and childrenÕs outcomes. However, as the shift to professionalize the early childhood field accelerates, researchers are being petitioned to go deeper and examine different aspects of the education process that may produce skilled and effective teachers of young children (Whitebook & Ryan, 2011). Acknowledging the petition, this study set out to answer questions regarding how coursework influe nces for early childhood teachersÕ development of knowledge and beliefs and how this knowledge influences ECEPsÕ classroom practices. But the lack of existing tools to assess for early childhood teachersÕ knowledge of child development was an impediment to answering these questions. Thus, an assessment tool (K -SEMS, preK version) was developed to fill this gap and help answer these substantive questions about the relations between coursework, knowledge, beliefs, and practices. The standards set out by NAYEC suggest that knowledge of childrenÕs general development and milestones alone is not enough, and that for early childhood teachers should also have knowledge of strategies to support childrenÕs development. Thus the K -SEMS includes two indices Ñthe Knowled ge of Social -Emotional Milestones Index and Knowledge of Social -Emotional Support Strategies Index. Given the importance of social -emotional development to childrenÕs school readiness and lifelong success, the importance of early 100 !childhood teachersÕ role i n this development, and lack of professional preparation dedicated specifically to this area, I focused the tool on assessing the social -emotional domain. Having established Knowledge of Social -Emotional Milestones and Support Strategies indices that are sufficiently reliable and valid for assessing pre -service early childhood teachersÕ knowledge in study 1, these were then used to test substantive questions to build a heuristic model of student preparation in study 2. Results from this study showed that w hen pre -service teachers took courses with domain -specific content, they had more accurate knowledge of childrenÕs social -emotional milestones and strategies to support children in this domain, as well as higher endorsements of expressing and supporting em otions. Further, early childhood teachers with more accurate knowledge on both indices, frequently used higher quality social -emotional practices when interacting with children in the classroom, especially when pre -service teachers had both more accurate k nowledge and higher endorsements of beliefs about emotion -instruction and modeling. While the moderate reliability of the Knowledge of Social -Emotional Milestones and Support Strategies indices might be questioned, the resultant findings provide an initial step in understanding the educational process and how coursework influences knowledge, and knowledge influences practices, elucidating the crucial role of beliefs in the relationship between knowledge and practice. This research provides evidence that tak ing courses in an early childhood program at an institution of higher education can influence pre -service teachersÕ knowledge of child development and equip them with strategies to support childrenÕs development. Further, these results help to build an ini tial model that can be tested in other institutions of higher education preparation programs. Results from the development of the Knowledge of Social -Emotional Milestones and Support Strategies indices, however, indicate the need to revise the tool to imp rove reliability. 101 !The next development phase should consider improving the quality of questions, increasing the number of items, and including items that cover a broader range of social -emotional skills and age range. Further, the Knowledge of Social -Emoti onal Milestones and Support Strategies indices only addressed knowledge of milestones and support strategies; however, there are other components of research -based child development knowledge that could be included, such as sequences of development and lea rning processes. Adding these knowledge components could enhance the tool as a more thorough assessment of knowledge of social -emotional development. This could provide even more insight into the relationship between early childhood teachersÕ coursework, k nowledge, and practices. Implications for Practice, Research and Policy, and Future Considerations This study extends existing research indicating that education, broadly defined as a degree level, influences teachersÕ classroom practices, by taking a clo ser look at education and examining the influence of courses taken on early childhood teachersÕ knowledge and beliefs, and linking those educational outcomes to classroom practices. It moves the field forward by examining coursework relevant to particular content and resultant learning, rather than simply years of education and major. Further, the development of a tool to assess early childhood teachersÕ knowledge could support the field to test and advance the initial heuristic model developed in this stud y. As the field of early care and education becomes more professionalized there will be a need for more tools to assess general knowledge of child development as well as domain -specific knowledge to ensure institutions of higher education are preparing skille d and effective teachers of young children. The Knowledge of Social -Emotional Milestones and Support Strategies indices provide a start to future development of other tools. Revised and enhanced, it could 102 !provide educators in early childhood programs with a pre -post assessment of their studentÕs knowledge. This information can then be used to revise course syllabi and guide consequent course content. Additionally, because this research shows that domain -specific course content affects early childhood teache rsÕ domain -specific knowledge and practices, programs of early childhood may consider having stand -alone courses to cover each of the main child development domains. Moving beyond elementary research of teacher education will inform the debates over baseli ne degree requirements for teachers of young children, thus future studies should examine the individual and combined aspects of education (coursework, field experiences, and reflective practices), and link them to early childhood teachersÕ competencies, c lassroom practices, and child outcomes, in order to determine the most effective professional preparation for this important workforce. 103 ! REFERENCE S 104 !REFERENCE S Whitebook, M., & Ryan, S. (2011). Degrees in context: Asking the right questions about preparing skilled and effective teachers of young children (NIEER Policy Brief , Issue 22). New Brunswick, NJ: National Institute for Early Education Research.