Abstract
Utilizing data from the National Center for Education Statistics Schools and Staffing Survey (SASS), a multilevel model (Hierarchical Linear Model) was developed to examine the association of teacher/classroom and state level indicators on reported elementary social studies instructional time. Findings indicated that state testing policy was a significant predictor of elementary teachers’ reported time spent on social studies instruction. Teachers’ perceptions of workplace autonomy and grade level were also associated with increased time on social studies. Conversely, teacher credentials, classroom socioeconomic contexts, and test design were not substantially associated with social studies instructional time. This study suggests that state policy mandates, grade-specific curricular organization, and teacher disposition have a substantial impact on the prioritization of social studies in US elementary schools.
With No Child Left Behind legislation entering its second decade and recent government rhetoric placing greater emphasis on intellectual capital in the world market, US education has entered a 21st century renaissance. In response, states have begun to develop core common standards to tackle the growing globalization phenomenon. Legislation at the state and federal level has increasingly mandated standardized accountability measures to assess students, and their teachers, in meeting these desired goals. Among elementary schoolteachers, this renewed emphasis in high-stakes testing creates an instructional-decision-making dilemma—what to teach and what not to teach? Research suggests (Center on Educational Policy, 2007, 2008; Fitchett & Heafner, 2010; Wills, 2007) that practitioners spend more time on tested subject matter than non-tested subject matter. Consequently, testing pressures, and their subsequent use for evaluating teacher performance, influence how often and in what context practitioners teach specific content (Au, 2007; Crocco & Costigan, 2007; Hargreaves, 1994). These issues of instructional time allocation affect not only the traditional “specials” of elementary education (music, art, foreign language, etc.), but also the core subject material English/language arts, math, social studies, and science; Levine, Lopez, & Marcelo, 2008; Wills & Sandholtz, 2009). Social studies, in particular, has been traditionally cast aside in many elementary classrooms in the reprioritization of tested curriculum (Brophy, Alleman, & Knighton, 2009; Houser, 1995; Thornton & Houser, 1996; VanFossen, 2005).
The marginalization of social studies within elementary school curriculum has been examined and analyzed from numerous quantitative (Fitchett & Heafner, 2010; Heafner, Libscomb, & Rock, 2006; Leming, Ellington, & Schug, 2006b; Rock et al., 2006; VanFossen, 2005) and qualitative (Au, 2009; Boyle-Baise, Hsu, Johnnson, Sierrere, & Stewart, 2008; Wills, 2007) dimensions. In the majority of these studies, the curricular prioritization of elementary social studies remains bleak. One thread of social studies marginalization research suggests that teachers who work in states where the subject is tested at the elementary level are more likely to teach it (Gradwell, 2006; Grant, 2003; Heafner et al., 2006; Lintner, 2006). However, many social educators argue that the quality of instruction within a high-stakes testing environment suffers as a result of the narrow content and pedagogical autonomy (Au, 2007, 2009; Crocco & Costigan, 2007; Rock et al., 2006). Regardless of implications and interpretations, social educators agree that elementary social studies is important for establishing a foundational understanding of one’s place and purpose within a democratic society (Alleman & Brophy, 2001; Zhao & Hoge, 2005).
Given the variability of states’ testing programs and curricula (Au, 2009), little research has analyzed the influence of teacher/school ecology factors and state policy on how much instructional time is allocated to social studies. Utilizing data from the National Center for Education Statistics Schools and Staffing Survey (SASS), we developed a multilevel model (Hierarchical Linear Model) to examine the association of teacher/classroom and state level indicators on reported elementary social studies instructional time. Three research questions guided our research design: What elementary (Grades 3-5) teacher traits are associated with a reported social studies instructional time? What classroom/school/curriculum characteristics are associated with reported elementary social studies time? Is there a significant difference in reported social studies time between elementary teachers in states that test compared to states that do not test?
The Role of Teachers in the Prioritization of Elementary Social Studies
Previous qualitative studies have documented the challenges faced by elementary teachers in their attempt to include social studies content in an ever-constrained curricular environment (Boyle-Baise et al., 2008; Costigan & Crocco, 2004; Pace, 2011; Wills, 2007; Zhao & Hoge, 2005). In this research, numerous factors were found to influence the amount of emphasis that elementary teachers spend teaching social studies. In the subsequent section, we outline three specific mechanisms that contribute to time spent on social studies: teacher preparation, curricular intensification and professional autonomy.
Teacher Preparation
Content expertise
One common theme that emerges from research of social studies marginalization is the lack of preparation many elementary school teachers possess in teaching the content associated with the multiple disciplines that comprise the social studies. Wineburg (2005) argues that it is an impossible task for preservice teachers to become experts in all disciplines that fall under the umbrella of social studies as defined by licensure standards. This formidable undertaking is realized in both the limited content efficacy of elementary teachers and continued reliance upon text-oriented curriculum (Crocco & Costigan, 2007; Passe, 2006; VanSledright, 2002; Zhao & Hoge, 2005). Even when they attempt to find time for social studies through creative, subject negotiation, teachers’ lack of understanding of social studies pedagogical content knowledge accounted for perfunctory instruction and limited time spent exploring concepts in depth (Wills, 2007).
Marginalization mindset
Preparation deficiencies have been attributed to a general apathy toward social studies because it is not regularly part of the tested curriculum (Bolick, Adams, & Willox, 2010; Owens, 1997). In their analysis of social studies preservice education, Bolick, Adams and Willox (2010, p. 17) noted that the marginalization mindset “infiltrated (teacher preparation) from preconceived subject bias to limited amounts of content exposure.” This mentality carried over into field experiences. Research indicates that it is common for preservice teachers to have almost no exposure to social studies lesson planning and instruction in student teaching (Bolick et al., 2010; Good et al., 2010; Lanahan & Yeager, 2008). To counter this phenomenon, social studies educators contend that elementary teacher preparation should place greater value on social studies content and skill development (Passe, 2006; Thornton, 2001; Yon & Passe, 1990; Zhao & Hoge, 2005). They rationalize that increased social studies pedagogy would enhance its perceived importance among current practitioners and encourage more time spent on its instruction. Promoting depth and breadth, social educators argue, teachers might be able to bridge the pedagogical divide between the expectations of teacher education and the realities of the high-stakes elementary classroom (Grant, 2006).
Integration
Given the barriers toward effective social studies preparation, the de-facto rationale for teaching social studies in many elementary classrooms is subject integration (Bolick et al., 2010; Boyle-Baise et al., 2008; Hinde, 2005; Leming et al., 2006b; McGuire, 2007; Rock et al., 2006). A few social studies educators (Holloway & Chiodo, 2009) suggest that subject area integration provides evidence that social studies is being taught in earlier grades. However, this claim is countered by others who suggest that integration dilutes social studies instruction (Brophy et al., 2009; Parker & Jarolimek, 1997; Thornton & Houser, 1996) and fails to consider discipline-specific skills such as historical investigation (Brophy & Alleman, 2008; Levstik & Barton, 2010; VanSledright, 2004, 2011). Consequently, analyzing how much specific, stand alone time social studies receives compared to other core subject areas is an important indicator of how it is prioritized in elementary classrooms (Levstik, 2008).
Intensification and Autonomy
Intensification of teacher work and inequitable prioritization of instructional time are re-occurring themes of elementary teachers’ pedagogical decision-making. Hargreaves (1994, p. 108) defines intensification as “bureaucratically driven escalation of pressures, expectations and controls concerning what teachers do and how much they should do within a school day.” Intensification is not new (Apple, 2004; Duplass, 2007; Fitchett & Heafner, 2010), but an increasing phenomenon. Given the current political and cultural atmosphere of US education, elementary school teachers are under stress to teach a rigid and prescribed curriculum. High-stakes testing and standardization rhetoric only magnify teachers’ perception of narrowing curricular options (Crocco & Costigan, 2007; Wills & Sandholtz, 2009). As control over content and instructional methods dissipate in favor of state-sanctioned mandates, teachers are more likely to perceive less autonomy in their instructional decision-making. In their qualitative analysis of two social studies classrooms, Gerwin and Visone (2006) noted that high-stakes accountability substantially influenced teachers’ content selection and pedagogy. Au (2007, 2009) points out in his meta-analyses of social studies instruction and high-stakes testing that teachers’ perception of control over content and instructional time is consistently linked to testing pressures. Conversely, elementary school teachers who perceive greater work autonomy are more inclined to resist intensification and make independent, classroom responsive decisions on how to meet the needs of learners (Gerwin, 2004; Gerwin & Visone, 2006; Pace, 2011; Wills & Sandholtz, 2009).
Maverick teaching
In an effort to inspire and enact autonomous teaching, social studies research has pointed to examples of what Grant (2007) refers to as ambitious teaching—suggesting that “. . . teaching is nuanced, complex, and contextualized both because of and in spite of state social studies tests and the consequences they hold” (p.253). In his analysis of two U.S. history teachers, Grant (2003) infers that testing does not predetermine how and what social studies is taught. Subsequent research of social studies teachers employing innovative, student-centered practices in the midst of a tested curriculum offer similar findings (Gradwell, 2006; van Hover, 2006; Yeager & van Hover, 2006). Though the examples provided tend to focus on middle and high school classrooms, this line of social studies research provides illustrative models for elementary teachers to subvert testing constraints in order to teach higher-order, engaged instruction. Furthermore, the concept of ambitious teaching finds its roots in Brophy’s (1993) research on elementary teachers who exhibited autonomous social studies pedagogical practices. Brophy coined the exemplary praxis of these elementary teachers as maverick tendencies since the exhibited traits were unique and personally associated. Brophy suggested that the autonomous nature of elementary classroom structures and curriculum flexibility enabled teachers to display maverick instructional methods. Thus, maverick teaching is indicative of ambitious and autonomous teachers.
Organizational Factors Influencing Elementary Social Studies Instructional Time
Intensification of instructional time is often mitigated by various factors beyond teachers’ perceptions (Pace, 2008; Segall, 2006). School and classroom contexts along with content and conceptual differences of the traditional elementary school curriculum influence how teachers prioritize social studies time.
School and Classroom Contexts
Socioeconomic environments
Labor characteristics and workplace demography shape the quality and quantity of elementary social studies instruction (Grant, 2007; Pace, 2011; Vogler, 2006; Wills & Sandholtz, 2009). In a nationally representative examination of teacher characteristics, Fitchett (2010) noted that the least qualified social studies teachers (as determined by academic degree in social studies-related fields and licensure) taught in significantly lower socio-economic environments than more qualified teachers. Consequently, the most at-risk learners are often receiving inadequate instruction compared to more affluent students (Boyd, Lankford, Loeb, & Wyckoff, 2005). These levels of disparity are further exacerbated by what Grumet (2010) refers to as the “audit culture” of American schooling—whereby the onus of students’ learning is becoming exceedingly tied to teacher performance. With these external pressures, teachers allocate instructional preparation and class time to content directly tied to their performance evaluation. Under this premise, teachers working in low socioeconomic schools and classrooms spend minimal time on social studies; fearing that their students need greater remediation time in tested subjects like math and ELA (Levine, et al., 2008; Pace, 2008, 2011; Wills & Sandholtz, 2009). Segall (2006, p. 113) reported in an interview with social studies practitioners of a predominately working class school in Michigan that the “push, pull…” of testing was inordinately stronger in a lower socioeconomic setting. Conversely, teachers in more affluent environments felt more comfortable exploring in-depth social studies instruction due to a perceived inevitability of high scores on tested subjects (Pace, 2011). Furthermore, differences in instructional time distributions are associated with ethnicity and school poverty ratings, whereby students of color receive more remedial time on tested subjects and significantly less time studying non-tested content (Roth, Brooks-Gunn, Linver, & Hoffereth, 2003).
Individualized student learning within classrooms
Within the often heterogeneously grouped environment of elementary classrooms, practitioners are responsible for enacting curriculum that meets the diverse learning styles of learners. According to recent data from the National Center for Education Statistics (2008), approximately 12.3% of K-12 students have an Individual Education Plan (IEP). Recent federal mandates ( “No Child Left Behind Act of 2001,” 2002 ) inclusion of exceptional children in high-stakes accountability places additional demands on school personnel to allot instructional time to tested subjects. At the elementary school level, research suggests that student achievement of exceptional children is associated with the amount of instructional time allotted by subject area (Jenkins & Jenkins, 1981). In classrooms with numerous exceptional children, social studies instruction is often truncated due to accountability pressures (Mastropieri, et al., 2005; Lintner & Schweder, 2008). In their analysis of North Carolina principal survey data on the state’s accountability policies, Lyons and Algozzine (2006) suggest that principals perceive their role of protecting instructional time for tested subjects essential while viewing the inclusion of special needs students in high-stakes testing unfavorably. These “cross-currents” (Grant, 1996) of school leadership and classroom organization compiled with state-level curricular policy suggest that elementary school teachers’ decision whether or not to teach social studies is further influenced by the diverse learning styles and individual needs of the learners.
Curricular Organization
Expanding communities
Traditional curricular organization of elementary social studies is built upon the “expanding communities” design whereby in primary grades (PK-3) content and instruction focuses on the role of self, family, and neighborhood and in intermediate grades (4-5) emphasizes state, nation, and world histories (Alleman & Brophy, 2001; Brophy & Alleman, 2008; Hanna, 1937; Levstik, 2008; Martorella, Beal, & Bolick, 2005). Though little research supports its implementation (Brophy & Alleman, 2008; Henke, Chen, & Goldman, 1999), it remains a staple of social studies curricular organization across states (Brophy & Alleman, 2008; Levstik, 2008; Thornton, 2005).
Grade-level differences
Research in marginalization indicates that grade-level curricular distinctions are associated with significant differences in social studies instructional time (Leming, Ellington, & Schug, 2006a; Shaver, 1989). Contemporary research in elementary social studies posits greater emphasis on discipline-based content as a way of improving the quality and frequency of social studies in elementary schools (Brophy & Alleman, 2008; Levstik & Barton, 2010; VanSledright, 2011). Previous findings support this claim—indicating that social studies teachers in grades 4-5 spend (on average) greater time on social studies instruction than practitioners in earlier elementary grades (Fitchett & Heafner, 2010; VanFossen, 2005). Houser (1995) and VanFossen (2005) posited that the specific content-driven aims of intermediate curriculum offers practitioners a more explicit pedagogical anchoring than the more nebulous concept-driven curriculum of primary education, thereby resulting in more specific appropriation of instructional time. In commenting on the organization of elementary social studies, Duplass (2007) noted that grade-specific curricular traditions are further compounded by the increased standardization and prioritization of NCLB-related policies. He cautions that the traditional elementary curriculum, with its lack of focus on “information knowledge” in the primary grades, contributes to a significant lack of social studies-appropriated instructional time. The pressure for students to perform in content-driven high-stakes tests in math, English/language arts (ELA) and most recently science has led teachers to explicitly re-prioritize instruction time away from social studies (Heafner & Fitchett, 2012).
State Policy Factors Influencing Social Studies Instructional Time
Interstate policy landscape
The re-authorization of the Elementary and Secondary Education Act (“No Child Left Behind Act of 2001,”) was signed into law in 2002. In alignment with previous federal policy regarding standardization (Evans, 2004; Manzo, 2005), social studies was excluded from testing and standardization mandates. Consequently, the interstate policy landscape for social studies is quite varied (Au, 2009). Standardized tests assessing elementary students’ social studies knowledge and skills have been implemented in twelve states (“Executive summary: 50-state Report Card,” 2009). However, the impact of these tests on the quality and quantity of social studies instruction has produced substantial debate.
NCLB testing effects for social studies
Numerous research studies (i.e., Au, 2007, 2009; Fitchett & Heafner, 2010; Leming et al., 2006a; Wills, 2007) offer not-so-subtle findings that No Child Left Behind and related testing accountability polices have a negative impact on how much time elementary teachers spend on social studies instruction. Researchers point out that social studies continues to lose ground to other core content (ELA, math and science) at alarming rates—approximately 30 minutes (0.5 hours) a week between 1993 and 2003 (Fitchett & Heafner, 2010; Leming et al., 2006b; Levine et al., 2008). A study conducted at the Center on Educational Policy (2008) found that social studies instruction decreased by approximately 17 minutes while reading instruction (mandatory testing under NCLB) increased by approximately 40 minutes a week. In addition, school district policies mandated a 32% decrease (on average) of social studies time to make room for tested subject matter. Findings expose a disproportionate amount of instructional time allotted to social studies compared to requisite-tested subject matter (Heafner & Fitchett, 2012).
Testing social studies?
While the loss of instructional time in elementary social studies under the NCLB-era is hardly refuted, discussion revolving around whether or not to test the elementary social studies offers substantially mixed findings (Au, 2007, 2009) As previously stated, social studies is tested at the elementary level in 12 states—down from 30 a decade ago (O’Connor, Heafner, & Groce, 2007). In a comparative analysis of elementary teachers’ instructional perceptions and attitudes of two states (one tests social studies, the other not), Heafner and colleagues (2006) indicated that teachers of the tested state spent substantial more time on social studies instruction than teachers of the non-tested state. These findings mirror additional studies suggesting social studies is being taught frequently in the states that it is tested (Grant, 2006; Holloway & Chiodo, 2009). Yet, as Au (2007, 2009) points out, there is substantial variability in the research surrounding teachers’ instructional response to mandated assessment. Citing pressures of accountability, one strand of research reports that social studies practitioners’ teaching with a high-stakes test are more likely to incorporate mechanical, instructor-centered, teach-to-the test methodology (Crocco & Costigan, 2007; Gerwin & Visone, 2006; Rock et al., 2006; Segall, 2006; Vogler, 2006). Vogler (2006) notes that testing preparation is significantly associated with textbook and lecture-based instruction. Heafner and colleagues (2006) acknowledge similar findings—social studies instruction substantially increased in a tested-state, but the quality and rational for teaching social studies was narrowly driven by curriculum mandates.
Testing contexts and structures
The nature of the qualitative research on high-stake social studies instruction is highly subjective and can only be attributed to the specific context of the given study (Au, 2007, 2009). Even in states that use effective testing practices, there is no guarantee that quality social studies teaching and learning will occur. Due to the nature of social studies knowledge and skills (i.e., Horn, 2006; VanSledright, 2004), the extraordinary variability of social studies assessment policies among states and the substantial differences in the assessments themselves as well as regional curricular differences, a national policy analysis of high-stakes testing in social studies is difficult from a qualitative standpoint.
The Influence of Instructional Time Allocation on Social Studies Instruction
Hargreaves (1994, p. 95) refers to time as a “Faustian bargain” that cripples practitioners’ instructional decision-making. Specifically, curriculum standardization and high-stakes testing have contributed to a climate of intensification (Apple, 2004; Hargreaves, 1994) in elementary classrooms whereby the priorities and accountabilities of teaching time shift. As a result, teachers’ perception of subject matter importance changes—class time defers to tested subject matter. Numerous research studies indicate that increased standardization efforts indicative of NCLB are associated with a decrease of instructional time in elementary social studies (Center on Educational Policy, 2008; Fitchett, 2010; Heafner & Fitchett, 2012; Leming et al., 2006b). Yet, what does quantity suggest about the quality of social studies instruction in the classroom? Numerous exemplars of ambitious and maverick teaching (Brophy, 1993; Gerwin, 2004; Gradwell, 2006; Grant, 2003, 2007; Holloway & Chiodo, 2009; van Hover, 2006) would indicate that some teachers counteract the implicit prescriptive demands of state curricula to enact dynamic instruction in their classrooms. Au (2007, 2009) notes that the context of many of these studies is problematic—given that they are often situated in states that explicitly test social studies and/or analyze a small sample of participants in a very limited context.
Referring back to Hargreaves (1994), “Time compounds the problem of innovation and confounds the implementation of change.” (p. 95). Research on social studies teaching (Costigan & Crocco, 2004; VanSledright, 2004, 2011; Wills, 2007) suggests that engaging, inquiry-based instruction takes time. In states where social studies is not taught, pressure is placed on elementary teachers to proportion greater instructional time based upon implicit curricular importance (Costigan & Crocco, 2004; VanSledright, 2004, 2011; Wills, 2007) noted that due to curriculum prescription and testing anxiety social studies is relegated to “specials” status; supplanting recess time and coordinated by specific fixed intervals. The lack of time implicitly encourages many elementary practitioners to regress into teacher-centered, textbook oriented instruction that offers very little substantive reflection or critical thinking. Thus, the proportion of instructional time is a substantial indicator of both the quantity and quality of social studies teaching.
Method
In this study, we examined the association of teacher, classroom, and state policy characteristics on intermediate (Grades 3-5) social studies teachers’ reported instructional time. Incorporating a nationally representative data (SASS), we disentangled the variability of testing and policy context among and within states (Au, 2009; Grant, 2006; Wills, 2007) by developing a Hierarchical Linear Model for our analysis. Our research design was guided by three research questions:
What elementary (Grades 3-5) teacher traits are associated with reported social studies instructional time?
What classroom/school/curriculum characteristics are associated with reported elementary social studies time?
Is there a significant difference in reported social studies time between elementary teachers in states that test compared to states that do not test?
Data
We utilized data from the National Center for Education Statistics Schools and Staffing (SASS) public school teacher 2007-2008 school year data set 1 . SASS data is the United States’ most representative measure of public school teacher quality, demography and professional perceptions (National Center for Education Statistics, 2007). We selected for grade 3-5, elementary school teachers (n = 1,540) 2 . Social studies research and theory indicates a shift in curricular prioritization and content emphasis in later grades (Fitchett & Heafner, 2010; Hanna, 1937; Thornton, 2005; VanFossen, 2005). Therefore, we chose to examine grades 3 to 5 because they represent a transition in the expanding communities curriculum from concept-based (Grade 3) toward content-based (Grades 4 and 5; Brophy & Alleman, 2008; Thornton, 2005; Thornton & Houser, 1996; VanFossen, 2005). Data were additionally filtered for full-time employment, elementary licensure and teaching in “self-contained” environment to avoid the organizational confounds such as subject area specialization and special education licensure, which might influence reported social studies instructional beyond the scope of this study. The resulting subsample included third grade (n = 690, 45.0%), fourth grade (n = 530, 34.4%), and fifth-grade teachers (n = 380, 24.5%) who were the singular representative from their school. However, some schools within each state were contributing multiple teachers per school. In the circumstance of teachers nested within schools, the number never exceeded five. Therefore, we did not consider it appropriate to construct a three level model that tested building-level effects at level two. Thus, we randomly selected one teacher from each school in which there was more than one respondent per building.
Research Design
A Hierarchical Linear Model (HLM) was utilized to study both classroom and statewide policy factors associated with reported social studies time in elementary grades. This model was developed to avoid both ecological fallacies and atomistic fallacies in examining marginalization theory (Hox, 2010). As put forth in the preceding theoretical framework, large-scale studies have examined social studies across time while reporting mean social studies time regardless of the individual classroom and statewide predictors such as professional attitudes, socioeconomic context of the classroom, and state testing policies (i.e., ecological fallacy). Qualitative studies have offered specific classroom context of social studies teaching in a limited environment with small sampling frames and an inability to generalize findings (i.e., atomistic fallacy). HLM, as a multilevel model, allows for more specific interpretation of how both classroom and statewide variables predict reported social studies time.
In examining teachers’ characteristics that prior research associated with elementary practitioner’s incorporation of social studies, we selected demographic, teacher quality, and school climate variables from the SASS dataset (see Table 1). To analyze teacher preparation in social studies (Bolick et al., 2010; Passe, 2006; Wills, 2007; Zhao & Hoge, 2005), we incorporated variables of social studies background in their bachelor’s and master’s degree, the number of methods courses they completed during teacher training, years experience, licensure pathway, credentials, and status as a “highly qualified teacher” as determined by NCLB (Yeager & van Hover, 2006). Characteristics also included teacher stressors associated with intensification (Apple, 2004; Hargreaves, 1994; Werner, 1988) and ambitious/maverick teaching. Autonomy was a construct of four single item response questions from the SASS (2007-2008 public school teacher survey) related to teacher control over the assignment of homework, manipulation of content, development of assessment, and teaching techniques—characteristics shared in various ambitious and maverick teaching literature (Brophy, 1993; Gradwell, 2006; Grant, 2003, 2007). Cronbach’s alpha analysis suggests this construct to be moderately reliable (α = .722). To analyze classroom and school level context associated with social studies instructional time (Segall, 2006; Vogler, 2006; Wills & Sandholtz, 2009), we included variables emphasizing classroom and school demography, curriculum structure (i.e., grade level taught), and whether the school achieved district level distinction for adequate yearly progress (see Table 1). Due to constraints of the data, we randomly selected one teacher for each school. Therefore, teacher and classroom/school variables were included at the level I of the multilevel analysis as contextual variables (Hox, 2010).
Indicator Variable Description.
In order to explore the policy level and assessment variation among states (Au, 2007, 2009; Grant, 2006), we coded states in respect to their testing of social studies at three grade level categories: elementary, middle, and secondary. Education Week’s Quality Counts report offers the most comprehensive state-by-state comparisons of specific standardization policies during a given academic year (Levstik, 2008). Using results from the 2007 report (“Executive summary: 50-state Report Card,” 2009), we examined the type of test items included as a measure of state-by-state assessment variance. We included whether the state has established a test in science, positing that science testing would have an inverse effect on the amount of time spent on social studies instruction (Heafner & Fitchett, 2012). We also incorporated the overall standardization grade as determined by Education Week’s Quality Counts 2007-2008 report. Finally, we included textbook adoption protocol (state or local) as an indicator of state’s control over the curriculum resources (Mathis & Boyd, 2009). Aligning the 2007-2008 data with state testing policy of that academic year, we coded states in which teachers were nested as testing social studies at all grade level points (n = 13, 25.5%), multiple grade levels (n = 5, 9.8%), or high school testing only (n = 4, 7.8%). The baseline comparison group was no standardized testing in social studies (n = 29, 56.9%). These variables represent level II independent variables in which teachers within schools are nested in particular states. For a dependent (criterion) variable, we assessed a single-item, opened ended question, “During the most recent FULL WEEK, approximately how many hours did you spend teaching (history/social studies) at THIS school?” (National Center for Education Statistics, 2007).
Results
As noted in teacher responses, participants spent an average of 2.520 hours of social studies instructional time per week (SD = 1.600). Weekly reported instructional time totals ranged from 0 to 15 hours per week. For the Autonomy scale scores, teachers in this sample reported an average of 11.400 (SD = 2.319), with a minimum of 4 and a maximum score of 16. This score represents a total sum score across the four items, each of which was measured on a 4 point scale. Higher scores indicate higher teacher perceived autonomy over instructional issues in the classroom. Teachers in this sample also reported an average of 2.811 children with IEPs in their classrooms (SD = 2.528). Teachers reported as few as 0 and as many as 28 children with an IEP in their classrooms. Descriptive statistics were examined and no substantial outliers were uncovered. Table 2 lists the median and quartiles boundaries for each of these variables.
Descriptive Statistics for Instructional Time, Autonomy, and Children With IEPS as Reported by Grades 3-5 Elementary Teachers (n = 1,540).
Note: Q1 = 25th percentile or 1st quartile boundary, Q3 = 75th percentile or 3rd quartile boundary.
The first step in testing our hypothesis was to fit the unconditional model. This model contained no predictor variables and was used to examine the Intraclass Correlation Coefficient (ICC), a measure of the proportion of variance in teacher reported social studies instructional hours that is between teachers within states and the proportion that is between states. The ICC was .060, indicating that approximately 6% of the variance was found to be between states and 94% between teachers within states.
Multilevel modeling produces statistical estimates of the proportion of the variance in the outcome variables that can be attributed to associations with the predictor variables at each level in the model. These values can be compared to the unconditional model (model without predictor variables) in order to estimate the variance accounted for by the predictors. These statistics are conceptually similar to r 2 statistics in single level regression models. We initially tested a model that included, based on theory, all of the level 1 and level 2 variables believed to have a potential association with our outcome variable. The initial level 1 model accounted for 3.4% of the within state variance in reported instruction time and the initial level 2 model accounted for 0.0% of the between state variance. We then examined the correlations between all variables at each level and found both multicollinearity between predictors and variables that were not related to the outcome measure. We then ran a second reduced model, retaining only those variables related to the outcome, and found that the level 1 model accounted for 3.6% of the within state variance in reported instructional time and the level 2 model accounted for 11.8% of the between state variance. Therefore, we chose to report only the most parsimonious model containing only the variables that were found to be statistically significantly associated with teacher reported social studies instructional hours. This model is reported in Table 3.
Results From Final Multilevel Model.
Note. **- p < .001, * - p < .05.
The intercept in this model can be interpreted as the reported weekly number of hours of social studies instructional time for third grade teachers with no children with an IEP in their classrooms and who work in states that do not test social studies. Autonomy was a statistically significant predictor of instructional hours and the model coefficient indicates that for every increase of one point on the autonomy scale score, we expect an increase of .064 hours of instructional time. The number of children with an IEP in the classroom was also a statistically significant predictor of instructional hours and the model coefficient indicates that for every increase of one child with an IEP, we expect a decrease of .033 hours of instructional time per week. Grade level was also a statistically significant predictor of instructional hours. Third grade was used as the baseline. The model coefficients indicate that 4th grade teachers report .408 hours (24 minutes) more weekly instructional time than third grade teachers and 5th grade teachers report .585 hours (35minutes) more instructional time per week than third grade teachers. State testing of social studies was the only state level statistically significant predictor of instructional hours and the model coefficient indicates that in states that test in every grade level, we expect .495 more hours of instructional time. This model accounted for approximately 11.80% of the between teacher variance and 9.35% of the between state variance.
In order to make these multilevel modeling results more accessible and to facilitate direct comparisons of the expected instructional time for teachers in different conditions, an expectancy table was created that includes model estimates of instructional time for teachers across the grade levels, autonomy levels, and state testing conditions (see Table 4). We chose to report these values by rounding to the nearest weekly instructional minutes as opposed to hours to make direct comparisons easier to understand. This table does not include every possible combination of teacher conditions, but does provide a general sense of the expected range of instructional minutes across teachers like those in our sample. For every cell in the table, we kept the number of children with IEPs in the classroom constant at our sample median value of 2. It is important to note that all values in the table are expected to decrease by approximately 2 minutes for each additional child with an IEP in the classroom. Therefore, these values are likely to be overestimating the amount of instructional time in classrooms with high concentrations of children with special needs. As indicated by the model coefficients, instructional time increases from 3rd to 5th grade, is higher in states that test social studies in all grades, and increases with teacher autonomy. The values range from a low of 122 minutes per week for a 3rd grade teacher with low autonomy working in a state that does not test, to a high of 198 minutes per week for a 5th grade teacher with high autonomy working in a state that tests social studies at every grade level.
Model Estimates for Weekly Social Studies Instructional Time by Grade, Autonomy Score, and Testing Status, in Minutes.
Note: Q1 = 25th percentile or 1st quartile boundary, Q3 = 75th percentile or 3rd quartile boundary.
Implications
We inferred four specific implications from these findings. First, results from this study confirm earlier studies that suggest testing and standardization influence the amount of time elementary teachers spend on social studies. Second, findings indicate that autonomous teachers spend more time on social studies instruction. Third, classroom level variables such as grade level and the proportion of special needs students significantly affected how teachers’ prioritize social studies. Finally, when controlling for these factors, teacher credentials and socioeconomic characteristics of the school were not associated with variability in time spent teaching social studies.
Implication #1: Teachers in States that Test Social Studies, Teach Social Studies
Results from this multilevel analysis corroborate previous research (Au, 2009; Heafner et al., 2006) that suggests time allocated to social studies instruction is related to the testing policy within a given state. Grades 3-5 participants were more likely to teach social studies and commit more instructional time in states that tested social studies at the elementary level. Using model estimates for weekly social studies instructional time by grade, autonomy score, and testing status, 5th grade teachers in states which test social studies devote as much as 76 more minutes per week for social studies instruction than 3rd grade teachers in states that do not test social studies. At each grade level, an estimated difference of 30 minutes per week of instructional time was associated with testing practices. However, taking into account states’ direct assessment of social studies instruction, other policy indicators were not significantly associated with reported social studies time. Moreover, the variability of tested social studies items per state, as measured by the inclusion of extended response items, was not associated with social studies instructional time.
Given that current accountability pressures often draw teachers to tested content as opposed to non-tested subjects, these findings offer a large-scale national picture of the depth and breadth of this phenomenon. Social studies, often maligned by political forces (Evans, 2004; Leming, Porter-Magee, & Ellington, 2003), has only timidly established itself in the era of No Child Left Behind. In the elementary classroom, it has frequently fallen to the wayside in competition for teachers’ attention and time as a valuable subject. However, while testing seemingly increases the amount of time spent on elementary social studies compared to non-testing, results suggest that the types of test items are not associated with more instructional time. In response to state-wide studies of elementary history/social studies teaching (Gradwell, 2006; Grant, 2006) which intimate that a particular testing protocol of open-ended questions (i.e., NY regents exam) improves the quantity and quality of instruction, our nationally-focused study suggests the inclusion of free-response items is not significantly associated with increased social studies time. Further research is needed to explore in greater context how these open-ended items improve (or fail to improve) the quality and substance of social studies instruction (Gerwin, 2004).
While our results suggest that when mandatorily assessed, social studies is more frequently taught, we caution against the notion that testing is a singular solution to marginalization issues. Numerous studies (Heafner et al., 2006; Vogler, 2006) point to a climate of rote social studies instruction in tested states. Moreover, as Au (2009, p. 53) warns, “(high-stakes testing) raises significant issues for social justice education because tests systematically push multicultural subject matter out of the school curriculum . . .” However, simply ignoring testing or wishing it away is neither proactive nor realistic. We argue that the current dialogue on statewide assessment has become an “either-or” false dichotomy (Dewey, 1938). Instead, social educators at various dimensions (practitioner, policymaker, teacher educator, and content specialist) should strive to develop and advocate sufficient assessment instruments that measure an individual learners’ competency of a state-mandated curriculum while also attempting to address the multifaceted nature of social studies knowledge.
Implication #2: Autonomous Elementary Practitioners Teach Social Studies
As Table 4 illustrates, even within non-tested states, differences between the first and third quartiles of autonomy account for approximately 11 minutes of social studies instruction per week on average. Given the average 36-week school year, differences in autonomy may account for over six hours of annual reported social studies time. This finding substantiates a recurring research theme (Brophy, 1993, Gradwell, 2006; Grant, 2003, 2006; van Hover, 2006) of ambitious or maverick social studies practitioners who believe they have the pedagogical freedom to teach in spite of, not because of, a mandated test.
Though ambitious teaching is most often evidenced in secondary classrooms, our findings suggest that a similar pedagogical ideal is possible among grade 3 to 5 teachers based on the premise that ambitious and maverick teaching are analogous. Therefore, we encourage elementary socials studies methods instructors to not only champion greater content area competence, but also a gatekeeping disposition (Thornton, 2001, 2005) in order to counteract curricular intensification that undermines and/or dictates social studies teaching. Thornton’s (2005) concept of gatekeeping suggests that social studies practitioners and elementary educators have creative control over how to enact the curriculum (Brophy, 1993; Pace, 2011). This teaching epistemology requires critical inquiry, reflection, and pedagogical content knowledge of the teacher (Thornton, 2001). Teachers who incorporate a gatekeeping ethos and exhibit maverick tendencies are more likely to design meaningful, engaged social studies instruction regardless of the curriculum’s testing requirements. Practitioners both survive and subvert schooling institutionalization (Apple, 2004; Crocco & Costigan, 2007; Hargreaves, 1994) by exposing learners to interactive, effective pedagogy that navigates the standard course of study without falling into rote instructional practices (Grant, 2003).
Implication #3: Grade Level and Classroom Contexts Effect Time Spent on Social Studies
In addition to teachers’ reported professional autonomy and state testing policy, grade level was significantly related to elementary social studies instruction. Similar to previous findings (Fitchett & Heafner, 2010; Houser, 1995; VanFossen, 2005), elementary teachers in later grades (Grades 4 and 5) reported greater social studies time than in earlier grades (Grade 3), a difference of 30 to 40 minutes per week. Given the size of the coefficients (see Table 3), differences in grade level were the most substantial level 1 predictor of reported social studies time. We posit that these findings reflect differences between 3rd grade and 4-5th grade curriculum traditions. Teachers at the primary level either lack explicit pedagogical direction in which to teach social studies (Brophy et al., 2009; Duplass, 2007) or fall back upon curriculum integration that dilutes social studies instruction (Alleman & Brophy, 1993; Brohpy & Alleman, 2008; Boyle-Baise et al., 2008; Hinde, 2005; Thornton & Houser, 1996). Conversely, 4th and 5th grade teachers’ decision to report greater time in social studies might be symptomatic of greater content emphasis in later grades (Brohpy & Alleman, 2008; Brophy et al., 2009; Evans, 2004). Findings also call into question the utilization of the traditional expanding communities curriculum (Brophy & Allen, 2008; Hanna, 1937). We suggest social studies professionals advocate for greater specificity in how the subject is included in the primary grades and consider emphasizing content-based, discipline-specific curriculum in social studies standards.
Social studies instructional time was inversely related to the number of students with IEPs in a respondent’s class. This finding illustrates a significant inequality in social studies instruction. As noted earlier, for each additional child with an IEP in the classroom, teachers at all grade levels spend up to two minutes per week less on social studies instruction. From the findings, we conclude that the clustering of students with IEPs is detrimental to overall student access to content. Moreover, social studies’ perceived lack of importance is a common theme within social studies marginalization studies (Heafner et al., 2006; Rock et al., 2006); whereby, instruction is neglected or co-opted for designated pullout time. Given the overall paucity of research on special education within the field, future research is suggested to investigate the context of social studies practice within elementary classrooms.
Implication #4: School Demography and Teacher Credentials Have a Minimal Effect on Time Spent on Social Studies
Numerous studies have explored the marginalized role of social studies not only at the elementary school level, but also in teacher education (Bolick et al., 2010; Good et al., 2010; Mathis & Boyd, 2009; Meuwissen, 2005; Passe, 2006; Zhao & Hoge, 2005). Our findings indicate that when taking into account state testing policy, teacher credentials were not significantly associated with change in reported social studies instructional time. Results from our large-scale study also diverge from qualitative and quantitative studies that suggest social studies instruction is marginalized in low socioeconomic, high-risk school settings (Pace, 2008, 2011; Roth et al., 2003). We inferred from these results that testing is so pervasive that accountability pressures overshadow a teacher’s official licensure, social studies pedigree, and various building level demographics.
Limitations
We recognize that our model is limited in analyzing the amount of time elementary teachers report on social studies instruction. We are unable to explicitly investigate how teachers interact with the subject matter as demonstrated in numerous qualitative studies (Au, 2007, 2009; Gradwell, 2006; Grant, 2003, 2006; Pace, 2011; van Hover, 2006; Yeager & van Hover, 2006). Yet, these previous studies have provided a substantial framework and context for elementary social studies research at the micro-level and a rationale for exploring marginalization at a macro-level. Richardson (2006) in acknowledging the importance of surveys in social studies research suggests: Social studies educators may find that they gravitate toward qualitative research because such methods provide a descriptive glimpse of classroom practices about which educators have personal experience and might offer more direct advice about how to improve instruction. However, survey methodology is likely to remain valuable because it is an efficient way of collecting information . . . and offer(s) greater generalizability to large populations than non-representative samples (p. 159).
In an effort to inform large-scale policy initiatives regarding social studies education, our study offers a national view of how elementary school teachers prioritize social studies instruction given their workplace disposition, curriculum, and state testing policies. Furthermore, we believe that macro-level studies are necessary to address large-scale educational policy decisions that affect social studies.
Conclusion
In this study, we examined to what degree accountability measures between states and classroom level indicators are associated with reported instructional time in social studies. Results suggest that a confluence of testing policies, grade level differences, and teachers’ perceptions of autonomy increase the emphasis of social studies teaching among 3 to 5 grade elementary teachers. Perhaps just as substantial, our findings indicate that the effects of testing policy overshadow various teaching credentials and school level socioeconomic factors. While this analysis cannot fully account for the complex mechanisms contributing to social studies prioritization, we contend that our model offers policy and pedagogical implications for elementary social studies. Namely, state testing policies that include elementary social studies are associated with greater time in social studies. In order to preserve its place at the curricular table, social studies professionals should consider advocating for a greater presence on state-wide assessment measures. In addition, increased time spent on social studies in later grades questions the purpose and development of elementary social studies in earlier grades and deserves further analysis at the classroom and curricular level. Lastly, teacher autonomy, as a significant predictor, suggests that elementary practitioners who are provided training and opportunity to explore their profession without the constraints of intensification are more likely to teach social studies. Therefore, a balance of promoting autonomy, redefining curricular organization, examining classroom milieu, and maintaining a foothold in a test-driven environment is necessary to improve the prioritization of social studies in elementary classrooms.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a University of North Carolina Charlotte Faculty Research Grant (fund number 1-11196).