The vision of quality science instruction described in Chapter 2 is not standard fare in U.S. classrooms. Answering the question, “How do we achieve these rigorous and ambitious goals for students’ science learning?” requires exploring the associated questions, “Where are we starting from?” and “How might we get from here to there?”
This chapter concerns what answers research provides to the question, “Where are we starting from?” To develop a broad picture of current science instruction in the United States, the committee reviewed four surveys that gather national-level data on teachers’ beliefs about teaching and learning and their instructional practices (see Box 3-1). First, the 2012 National Survey of Science and Mathematics Education (NSSME) documents both teachers’ beliefs and self-reports about their instruction from a representative national sample in the United States. Second, the 2011 National Assessment of Educational Progress (NAEP) included a teacher questionnaire that asked about teachers’ perceptions of effective instructional practices and their own practices, and asked students to report on instruction they had received. This survey was administered to 4th- and 8th-grade students and teachers only. The other two surveys capture characteristics of science teachers’ perceptions and classroom instruction across countries that include the United States. The Teaching and Learning International Survey (TALIS) 2013 describes teachers’ beliefs about the nature of teaching and learning and self-reported instructional practices across 23 OECD countries. However, the United States did not meet the international standards for participation rates; therefore, it is not appro-
National Survey of Science and Mathematics Education (NSSME)—2012
- Conducted by Horizon Research, Inc.
- U.S. nationally representative sample at all grade levels (1,504 schools; N = 7,700 teachers)
- Response rate = 77 percent
- Teachers’ perceptions of objectives, classroom activities, and assessment
- Self-reported classroom practices
National Assessment of Educational Progress (NAEP)—2011
- Conducted by the National Center for Education Statistics
- Nationally representative sample of 4th- and 8th-grade students and other teachers
- Teachers’ beliefs about teaching
- Self-reported classroom practices
Teaching and Learning International Survey (TALIS)—2013
- Conducted by OECD
- 23 OECD countries
- Teachers’ beliefs about teaching
- School principals report on school climate, leadership, teacher evaluation and induction
- Teachers’ reports of preparation, professional development opportunities and needs, classroom practices
Trends in International Mathematics and Science Study (TIMSS)—2011
- More than 60 countries
- 4th- and 8th-grade students (more than 20,000 students in 1,000 schools in the United States)
- Self-reported classroom practices, school resources, interaction with colleagues, and perception of preparation to teach
priate to use these data in establishing an accurate picture of national trends. The 2011 Trends in International Mathematics and Science Study (TIMSS) included survey items about instructional practices in 4th- and 8th-grade classrooms.
The committee focused primarily on results of the NSSME, the only survey that included a representative national sample at all grade levels. We examined results of the other surveys when possible to confirm whether similar trends were observed. We note also that A Framework for K-12 Science Education (hereafter referred to as the Framework) and the Next Generation Science Standards (hereafter referred to as NGSS) had not
been published at the time these four surveys were conducted, so the available evidence does not align completely with the vision of science teaching and learning laid out in those documents. However, because the Framework and NGSS build upon and evolved from earlier efforts to articulate a similar vision of science instruction, much can be learned from this existing research.
Although there are many individual classrooms in which children are routinely engaged in challenging and well-supported science learning (e.g., Gallas, 1995; Lemke, 1990; Nemirovsky et al., 2009), the national-level picture is sobering. We begin with a big-picture view of instruction before turning to more fine-grained analyses of practice at the elementary, middle, and high school levels.
The NSSME documents what science teachers report about their instruction with respect to objectives, classroom activities, and assessment; it helps paint an overall portrait of contemporary science instruction. As noted earlier, the survey is not explicitly aligned with the Framework and NGSS as it was developed before the standards were available; however, some of the survey items reflect instructional approaches that are consistent with the goals outlined in the two documents.
Overall, the survey indicates that there is a gap between current science teaching and learning and the vision embodied in the Framework and NGSS (see Chapter 2). About 60 percent of science teachers in the United States indicate that they are using “reform-oriented science teaching practices,” such as “have students do hands-on/laboratory activities,” “require students to supply evidence in support of their claims,” and “have students represent and/or analyze data using tables, charts, or graphs” (Banilower et al., 2013). While these kinds of practices are not fully aligned with the Framework and NGSS, they are in keeping with the broad learning goals outlined in the two documents. At the same time, teachers’ self-reports of the classroom activities in their most recently taught lesson confirm that teachers typically engage in more traditional instructional practices, such as “teacher explaining a science idea to the whole class,” “whole-class discussion,” and “students completing textbook/worksheet problems” (Banilower et al., 2013). Specifically,
- Seventy to 80 percent of teachers said they “have students work in small groups” at least once a week.
- About 17 percent of teachers said they “require students to supply evidence in support of their claims” during all or almost all
lessons; about 60 percent of teachers do so at least once a week. Few teachers said they never do so.
- About half of teachers (44 to 58 percent) said they “have students represent and/or analyze data using tables, charts, or graphs” at least once a week, and 8 percent said they do so during all or almost all lessons. Again, few teachers said they never do so.
- Three to 9 percent of teachers said they “have students practice for standardized tests” during all or almost all lessons. Interestingly, the percentages decrease from elementary to high school (9 percent of elementary, 6 percent of middle, and 3 percent of high school teachers). About one-fifth of teachers (19-23 percent) said they do this activity at least once a week.
Although teachers’ estimates of the frequency of different instructional practices offer some insight into what is happening in science classrooms, they do not provide information about the quality of implementation of those practices. Few large-scale observational studies of science classrooms provide assessments of the quality of instruction. The few that are available suggest that science lessons are often inadequate.
One such study, conducted by Weiss and colleagues (2003) included classrooms sampled from elementary, middle, and high schools clustered by feeder pattern in 31 sites involving 93 U.S. schools.1 The researchers focused on two primary goals of science and mathematics instruction: helping students develop conceptual understanding and deepening their ability to engage in processes of science and mathematics. With these goals in mind, the investigators developed a five-level scale of lesson quality:
- ineffective instruction (Level 1), characterized by “passive learning” or by “activity for activity’s sake”;
- partial presence of elements of effective instruction (Level 2);
- beginning stages of effective instruction (Level 3);
- accomplished instruction (Level 4); and
- exemplary instruction (Level 5).
1Weiss and colleagues (2003) selected a nationally representative set of 40 middle schools and then randomly selected an elementary school and a high school in the feeder pattern for each of these middle schools. Each set of three schools constituted a site. Two randomly selected science teachers/classes were selected for observation in each school in a given site. Observations were completed in 31 sites (93 schools) and form the basis of the reported findings. Observations were carried out in 2001 prior to full implementation of the No Child Left Behind Act and may not reflect the impact on science instruction of this legislation and the required large-scale assessments in mathematics and reading.
In Level 5 lessons, the teacher clearly articulated the instructional objectives; engaged students intellectually with science or mathematics content; portrayed the disciplines as dynamic bodies of knowledge; and provided a climate that encouraged students to generate ideas, questions, and conjectures. Teachers often invited students to interact with the content through multiple pathways, including direct experience with natural phenomena and real-world examples. These lessons were characterized by intellectual rigor, constructive criticism, and challenging of ideas. Teachers frequently used questioning strategies to elicit students’ level of understanding of the targeted concepts and adjusted instruction accordingly, building on what students knew to advance their thinking. They probed students for elaboration, explanation, justification, or generation of new questions or conjectures. These effective teachers also presented relevant and accessible examples and demonstrations and engaged students in laboratory activities, coupled with discussion of or writing about their observations or ideas to promote sense making (see Boxes 3-2 and 3-3). Because the study predates the Framework and NGSS, the characteristics of high-quality lessons do not exactly match what one would expect to see in lessons aligned with the NGSS. For example, they do not consistently integrate all three dimensions discussed in Chapter 2: science practices, crosscutting concepts, and core disciplinary ideas. Nonetheless, these prototypes provide importance guidance for considering what teachers need to know and be able to do to implement high-quality science instruction.
Accomplishing this kind of instruction entails considerable knowledge and skill. Not only would the teacher in this example need to be deft at managing a classroom in which students were engaged in laboratory activities, but she would also need to be able to anticipate the likely predictions that students would generate and what those predictions would signify with respect to students’ understanding. Leading a discussion is not a natural act, but instead takes considerable experience with hearing what students say, capitalizing on their emergent ideas, and selecting some comments for further collective discussion while being respectful of the broad array of student contributions (Brookfield and Preskill, 2005; Engle and Conant, 2002; Engle et al., 2014). We return to a discussion of the knowledge required for this kind of teaching in Chapter 5.
In contrast to high-quality instruction, lessons categorized as Level 1 (ineffective) included “passive learning” and “activities for activity’s sake.” Ineffective or less effective lessons tended to portray science as a static body of factual knowledge, and procedures in which the students engaged were not intellectually rigorous (Weiss et al., 2003). Teacher questioning tended to evoke only yes/no or fill-in-the-blank responses from students that failed to promote conceptual engagement or the develop-
An Exemplary Science Lesson
In an example lesson classified by Weiss and colleagues (2003) as Level 5, “High-Quality, Reform-Oriented Instruction,” a high school biology class was in the middle of a unit on cells. In the previous lesson, students had conducted a membrane lab using starch or sugar solutions and dialysis tubing, and the goal of this lesson was to help them learn about molecule size and transport across cell membranes. The teacher opened the lesson by asking the students, in their lab groups, to predict what they expected to happen with their lab and to use the concept of particle size to explain why. After they had made their predictions, the groups examined their data and discussed whether their predictions were right or wrong. The teacher then led the entire class in a discussion about what had happened in the experiment. Students suggested hypotheses, and the class discussed methods for testing them. As needed, the teacher chimed in with suggestions about lab techniques that would enable the students to test their ideas and prodded the groups to make sure that they conducted enough tests to explain fully what had happened. The teacher skillfully guided the students as they finished making observations and analyzing the data, asking questions that pushed students to examine their results and to provide evidence for their conclusions.
After the groups had finished all of their tests, the teacher assigned them to write a story about a paramecium living in the local freshwater river that traveled to the ocean. In their stories, the groups were instructed to use a list of eight vocabulary words related to transport across a membrane. The students spent the remainder of the class period working on their stories, an activity that allowed them to reflect on what they had learned about transport across a membrane and apply it to organisms living in their local river. This was a critical component of the lesson as it allowed the students to make sense of the lab results. Throughout the lesson, all of the students were engaged in meaningful investigation of important science content, and the teacher did a masterful job of guiding the class. Students were generating and debating hypotheses, and were given the tools they needed to test their ideas. Writing their stories allowed the students to make sense of the data and conclusions drawn from the lab investigation. The students had clearly taken ownership of their learning, and the teacher pushed and challenged all students to engage with the content.
ment of understanding. In some classrooms, the teacher both asked and answered the questions. These lessons did not provide sufficient time or support for students to discuss, reflect on, and make sense of laboratory activities, lectures, or demonstrations or to connect new information to existing knowledge (see Box 3-3).
Weiss and colleagues (2003) categorized only 15 percent of the science and mathematics lessons they observed as high quality (Levels high 3 through 5), 27 percent as medium quality (Levels low 3 and solid 3), and
An Ineffective Science Lesson
In an example lesson classified by Weiss and colleagues (2003) as Level 1, a 9th-grade biology class was near the end of a unit on evolution. The teacher opened the class by asking the students to complete a worksheet that referred to facts from their textbook. Without looking in the book, students had to decide whether the statements were true or false and correct the false ones. Then, the teacher asked them to check their work against the book, working in small groups to reach consensus on the answers, and to document where in the book (what page and paragraph) they found each answer. However, about half the students did not try to answer the questions. When they had finished the worksheet, the students copied from the board a timeline of evolution that focused on bacteria. Then the teacher announced the answers to the worksheet questions. Some students raised their hands and asked about items they did not understand, in which case the teacher asked the class to explain the answer, but he rarely gave students the time to speak before answering himself. The teacher then read through each worksheet problem one more time and asked students to identify the page and paragraph where they had found the answer.
Next, the teacher gave a lecture based on the chapter students had just read. He began by asking students to look at the inside of the textbook’s back cover, which showed a chart of the evolution of all life and when each life form was found, explaining that this chart summarized the material they were about to cover. The rest of the lecture consisted of a series of names of organisms and time frames of their existence. The focus was on lists of facts taken from the book; at several points, the teacher read straight out of the textbook or asked students to do so. The teacher instructed students to take notes in a two-column format in which one column was titled “Main Themes” and the other “Detail.” Only a few students adhered to this format, and the teacher never followed through or helped identify the main themes. The teacher’s questions rarely required higher-order thinking and never drew on previous knowledge or real-world connections, and the teacher never offered enough wait time for students to consider an answer.
59 percent as low quality (Levels 1 and 2). These findings are echoed in the TIMSS 1999 video study, which compared 8th-grade science lessons in the United States with those in four other countries that outperformed the United States on the 1999 TIMSS assessment. In 44 percent of U.S. lessons, there were weak or no connections between learning activities and science ideas. Perhaps more worrisome, 27 percent of U.S. lessons included no science ideas at all (Roth and Garnier, 2007). Results such as these highlight the urgency of creating substantial learning opportunities for teachers.
One aspect of classroom instruction that is important for supporting
the learning goals in the Framework and NGSS is providing students with opportunities to make sense of investigations and discuss their emerging ideas. This kind of systematic sense making is supported by verbal prompts from teachers or varied opportunities for student talk. In the Weiss et al. (2003) study, the authors examined teacher questioning and discourse in general. In their analyses, fewer than one in five lessons incorporated questioning that was likely to move student understanding forward (i.e., finding out what students know, pressing for reasoning, encouraging self-monitoring of one’s thinking)—even when the rest of the lesson was otherwise well designed. Many incidents were cited of teachers asking low-cognitive-demand, “fill-in-the-blank” questions in rapid-fire sequence, with the focus on correct responses (often single words or phrases) rather than on student understanding. The authors conclude that questioning was “among the weakest elements of [science] instruction” (p. 71). These findings are similar to those of Bowes and Banilower (2004), who analyzed lessons from classrooms where teachers had been supported for years through well-funded professional development initiatives. Their data showed that fewer than half of the lessons, even those of teachers who had received the most professional development, were likely to be rated as adequate in the areas of questioning and sense-making opportunities.
In a study of classrooms in a large school district in the eastern United States that included data from observations of 55 elementary classrooms, 37 middle school science classrooms, and 29 high school science classrooms (Corcoran and Gerry, 2011), fewer than one-third of these observations showed students engaged in any type of higher-order thinking. Qualitative reports on these classrooms indicated that although the lessons appeared to be well organized, students were often disengaged, and didacticism dominated instruction.
Disparities in Instruction
Differences are seen across different student groups and communities in the type and quality of instruction available to students. In the survey conducted by Banilower and colleagues (2013), classes with high-achieving students were more likely than classes consisting mainly of low-achieving students to stress reform-oriented objectives and instructional practices.
In the observational study conducted by Weiss and colleagues (2003), the quality of lessons varied across different communities and student populations. Lessons in rural schools were less likely than those in suburban and urban communities to receive high ratings, and lessons in classes that were “majority minority” scored lower than lessons in other classes.
TABLE 3-1 Science Classes in Which Teachers Report Engaging in Various Activities at Least Once a Week, by Grade Level
|Percentage of Classes|
|Explain science ideas to the whole class||87||89||88||96||95|
|Engage the whole class in discussion||90||91||90||92||83|
|Have students work in small groups||65||79||72||79||83|
|Require students to supply evidence in support of their claims||46||62||54||64||61|
|Do hands-on laboratory activities||54||55||55||62||70|
|Have students represent and/or analyze data using tables, charts, or graphs||42||46||54||54||58|
|Have students read from a science textbook or other material in class, either aloud or to themselves||39||55||48||56||37|
|Have students write their reflections in class or for homework||38||48||44||44||21|
|Focus on literacy skills (e.g., informational reading or writing strategies)||45||51||48||44||25|
SOURCES: Banilower et al. (2013, Table 5.12, p. 76), Trygstad (2013, Table 18, p. 12).
Finally, science lessons in classes comprising low- and middle-ability students were less likely to receive high ratings than lessons in classes comprising students of high or heterogeneous ability. Again, these results clearly point to the need to create learning opportunities for all teachers. Table 3-1 provides an overview of teachers’ reported practices in elementary, middle, and high school science classrooms from the NSSME.
Elementary Science Instruction
Most elementary students do not receive daily science instruction: only 19 percent of grades K-2 classes and 30 percent of grades 3-5 classes
receive science instruction on all or most days every week of the school year (Banilower et al., 2013; Trygstad, 2013). Elementary students receive less instruction in science than in reading or mathematics and less than students at higher grade levels (Banilower et al., 2013; Dorph et al., 2007; Smith et al., 2002). On the days when science instruction is provided, on average it accounts for only 19 minutes per day at the K-3 level, compared with 54 minutes per day in mathematics and 89 minutes per day in language arts. Although the average rises to 24 minutes per day in grades 4-6—compared with 61 minutes in mathematics and 83 minutes in language arts per day (Banilower et al., 2013, p. 54)—this is still less than a half hour for science learning on those days when science instruction is offered. No differences were reported by teachers in time spent on science across different student groups or different schools in the Banilower et al. (2013) study. However, a study in California found further that science was most likely to be sacrificed in elementary schools struggling to remedy weak performance results in mathematics and English language arts—the same schools most likely to enroll low-income students, African American and Latino students, and English language learners (Dorph et al., 2007). The California study encompassed surveys of elementary school teachers, together with interviews with district and county officials and surveys of and interviews with science program staff and professional development providers.
In the limited time accorded to science in the elementary grades, what is the nature of instruction that students experience? Teachers’ self-reports on the 2012 NSSME indicate that the instructional activities most frequently used in elementary school science lessons are conducting whole-class discussion, the teacher explaining science ideas to the class, and having students work in small groups (see Table 3-1). Teachers also reported employing three activities that are consistent with the Framework and NGSS—having students perform hands-on/laboratory investigations, requiring them to supply evidence in support of their claims, and having them write reflections on their science learning—but with somewhat less frequency. In addition, there were some notable differences in class activities between the upper- and lower-elementary school grade bands. Grades 3-5 classes were more likely than grades K-2 classes to engage in reading about science (55 versus 39 percent). They also had more opportunity than lower-elementary students to engage in the science practice of supporting their claims with evidence (62 versus 46 percent) and to write their reflections (48 versus 38 percent). Formal assessment also receives more emphasis at the upper-elementary level, where students are more likely to be given tests and quizzes as well as to practice for standardized tests. Table 3-1 compares the frequency of various instructional activities within the elementary grade bands and the middle and high school levels.
Other studies indicate that elementary science instruction tends to focus on activities that are connected only loosely (if at all) to science ideas and are selected primarily to be fun and motivating for students (Dorph et al., 2011; Roth et al., 2006). Often, activities progress from topic to topic, with few attempts to help students make connections between them; the goal is to sustain students’ attention rather than to engage them deeply in scientific practices or model building. The observational study by Weiss and colleagues (2003) discussed above lends support to this lack of focus on disciplinary core ideas, as the investigators judged that only one-third of the lessons they observed were likely to have a positive impact on students’ understanding of science concepts. In addition, the researchers found that the greatest weakness of elementary science lessons was in the area of giving students the time and structure needed for sense making and wrap-up.
In addition to the lack of adequate time in the school week and day for elementary school science instruction to achieve the vision of the Framework and NGSS, elementary school teachers lack appropriate technology, curriculum, and instructional materials to support instruction aligned with the vision. As reported in the 2012 NSSME (Banilower et al., 2013), median per pupil spending per year for scientific equipment (e.g., microscopes), consumable supplies (e.g., chemicals), and science instructional software was $1.55 in elementary schools, compared with $3.13 in middle schools, and $6.11 in high schools. Reflecting these low budgets, elementary science teachers are less likely than their middle and high school counterparts to have access to various instructional resources. Although most have access to the Internet, nongraphing calculators, and personal computers, fewer than half have access to other scientific resources (e.g., microscopes, probes for collecting data, and classroom response systems or “clickers”). Perhaps more important, only about one-third of elementary science teachers reported having adequate facilities, equipment, consumable supplies, and instructional technology for science instruction (see Table 3-2). Perhaps because only one-third of the elementary teachers view the available instructional technology as adequate, only 22 percent indicated that they had used it in their most recent science lesson (Banilower et al., 2013).
Curriculum materials are an important source of support for science teachers, and nearly 70 percent of elementary teachers responding to the NSSME reported that their classes use commercially published textbooks or modules as the basis for instruction (Banilower et al., 2013). Of these classes, more than half use these instructional materials for 50 percent or more of their science instructional time. However, much elementary science instruction appears to be pulled together from multiple sources, with 40 percent of grades K-2 classes and 23 percent of grades 3-5 classes
TABLE 3-2 Classes with Adequate Resources for Science Instruction by Grade Range
|Percentage of Classes Where Adequate|
|Facilities (e.g., lab tables, electrical outlets, faucets, sinks)||31||57||71|
|Equipment (e.g., microscopes, beakers, photogate timers, Bunsen burners)||37||47||60|
|Consumable supplies (e.g., chemicals, living organisms, batteries)||34||39||59|
|Instructional technology (e.g., calculators, computers, probes/sensors)||34||37||48|
SOURCE: Banilower et al. (2013, Table 6-23, p. 106).
using noncommercially published materials most of the time. Additionally, teachers in elementary classes using commercially published materials frequently supplement them with other materials, and do not always use the commercially published materials as designed. As professionals, it is important that teachers use their professional discretion in selecting and adapting curriculum. However, given the fact that many elementary teachers have not had an opportunity for substantial engagement in science content and practices, this finding suggests the need for significant opportunities for elementary teachers to enhance their content knowledge as well as their pedagogical content knowledge.
Middle School Science Instruction
Most middle schools have dedicated science teachers, and students participate in science class daily or every other day. About a third (31 percent) of middle schools use block scheduling, allowing time for laboratory investigations to extend beyond the 50-minute class period that is typical of daily scheduling in U.S. middle schools (Banilower et al., 2013). As at the elementary level, the most frequent instructional techniques reported by teachers are the teacher explaining science ideas, whole-class discussions, and students working in small groups (Banilower et al., 2013; see Table 3-1). However, middle school science teachers were more likely than elementary teachers to report that at least once a week their students were asked to (1) supply evidence in support of their claims (64 versus 54 percent); (2) engage in hands-on/laboratory activities (62 versus 55 percent); (3) represent and/or analyze data using tables, charts, or graphs
(54 versus 44 percent); and (4) read from a science textbook or other material (56 versus 48 percent). Reflecting the increasing emphasis on testing and accountability at higher grade levels, middle school science teachers also are more likely than elementary teachers to give tests and quizzes, including short-answer tests and tests requiring constructed responses.
Differences in self-reported instructional activity are nonetheless insufficient indicators of enhanced instructional quality at the middle school level, or more specifically, of the degree to which middle school science instruction is consistent with the vision expressed by the Framework and NGSS. Two observational studies of science classrooms conducted since 2000, although predating the release of the current standards, suggest that middle school students may have limited experience of high-quality science instruction.
In their observational study of schools in 31 nationally representative sites, Weiss and colleagues (2003) found that middle school science lessons were weaker than those at the elementary and high school levels. Specifically, 78 percent of lessons were rated as Level 1 or 2 (ineffective or incorporating only some elements of effective instruction), 16 percent were rated of medium quality, and only 7 percent were rated of high quality. A common weakness across the observed lessons was a lack of time and structure for sense making.
Similar observations about a lack of support for making sense of natural phenomena come from TIMSS video analyses of recordings from 8th-grade classrooms in five countries (Roth et al., 2006). During “practical” (i.e., laboratory) activities, students in all countries, including the United States, were more likely to observe phenomena than to construct models or conduct controlled experiments. In four other high-achieving countries, students typically concluded practical activities by discussing the results and drawing conclusions, but in U.S. science lessons, this was the rare exception.
In all four higher-achieving countries, science lessons focused on high content standards and expectations for student learning, but each country used a slightly different instructional approach. In the Czech Republic, for instance, instruction was dominated by regular discussion of science content among students and their teachers. Teachers engaged students in whole-class discussions, presentations, and oral quizzes, focusing on rigorous science content. In contrast, students in the Netherlands tended to learn science independently, both when reading and writing answers at their seats and when conducting individual practical activities. Whole-class discussions often focused on homework review. In Japan, students were regularly pressed to draw connections between ideas and evidence. They conducted practical activities and collected and interpreted the resulting data to reach a main idea or conclusion. And in Australia, stu-
dents regularly drew connections among ideas, evidence, and real-life issues. As in Japan, they conducted practical activities and collected and interpreted the resulting data to reach a main idea or conclusion, but they also discussed real-life issues to support the development of science ideas (Roth et al., 2006).
In contrast to these approaches, U.S. science lessons were dominated by activities with less attention to the science content, and even less attention to the links between the activities and science ideas. Relative to the other nations, important ideas in science played a less central role and sometimes no role at all. In fact, in 27 percent of U.S. lessons, students engaged in activities and followed procedures with no mention of even a single science idea (e.g., “A complete circuit is needed to light the light bulb.”). Instead, instruction involved students in such activities as games, puzzles, dramatic demonstrations, and outdoor excursions without explicit connections to science ideas. The American tendency to teach science through “activity without understanding” has been identified in other studies as well (see, for example, Corcoran and Gerry, 2011).
Roth and colleagues (2006) found that in higher-achieving countries, teachers more commonly used activities to develop science ideas and organized lessons in coherent ways. The contrast between U.S. lessons and those in higher-achieving countries highlights the need for teachers to develop the knowledge required to organize specific science content so that students can see and make the links between science ideas and lesson activities. The knowledge demands of teaching are examined in Chapter 5.
As noted above, compared with elementary schools, middle schools provide more time for science learning and spend about twice as much per pupil for science equipment and supplies (Banilower et al., 2013). Reflecting this higher spending, middle school teachers’ access to instructional resources for science teaching is greater than that of elementary teachers, although less than that of high school teachers (Banilower et al., 2013). As shown in Table 3-2, the majority (57 percent) of middle school teachers indicated that their facilities were adequate, and about half viewed their equipment as adequate, while only about 40 percent viewed their consumable supplies and instructional technology as adequate.
As at other levels, middle school science classes do not incorporate instructional technology to a great extent (Banilower et al., 2013). Only 30 percent of middle school teachers reported that they had used instructional technology in their most recent lesson. This limited usage may be linked to teachers’ perceptions, as captured by the survey, that the available instructional technology is inadequate. It is possible that the use of instructional technology would increase if better technology were available. However, research on technology integration in middle school sci-
ence classrooms suggests that robust outcomes for lesson design, instruction, and student learning are more likely where teachers experience extended professional learning opportunities and other supports (Penuel et al., 2009; Yerrick and Johnson, 2009). In a review of 43 studies of technology integration across all levels of K-12 schooling, Gerard and colleagues (2011) found that outcomes were strongest (with a few exceptions) where supports extended over more than one school year. They report, “The studies suggest teachers needed support to distinguish effective ways to use new technologies, especially when the goal was to support inquiry learning” (p. 434).
Most middle school teachers (80 percent) use commercially published textbooks or modules as the basis for instruction (Banilower et al., 2013), and about half use these texts or modules for 50 percent or more of their science instructional time. However, middle school teachers are more likely than elementary teachers to supplement these materials with other resources or to skip parts they deem unimportant.
High School Science Instruction
Like middle schools, high schools provide more time for science learning than elementary schools, and about one-third (34 percent) offer block scheduling, allowing extended time for laboratory investigations (Banilower et al., 2013). All states and districts require high school students to participate in at least 1 year of science classes, and 64 percent require students to complete 3 years of high school science (Banilower et al., 2013).
As at the elementary and middle school levels, the most frequent instructional approaches in high school are the teacher explaining science ideas to the whole class, students working in small groups, and whole-class discussions (see Table 3-1). Relative to elementary and middle school teachers, however, high school teachers are more likely to ask students, at least once a week, to do hands-on laboratory investigations (70 versus 62 percent in middle school and 55 percent in elementary school) and to represent or analyze data using tables, charts, or graphs (58 versus 53 percent in middle school and 44 percent in elementary school) (see Table 3-1).
Relative to the middle and elementary levels, high school teachers less often ask students to read from a science textbook or other material (see Table 3-1). Only half as many high school teachers as middle and elementary school teachers ask students to write their reflections (see Table 3-1). In addition, high school classes are slightly less likely than middle school classes to require students to support their claims with evidence. These data suggest that the weaknesses identified at the elementary and middle school levels, including limited use of science practices to
support conceptual understanding and lack of time and support for sense making, are present as well at the high school level.
In their observational study, Weiss and colleagues (2013) found that high school science lessons were weaker than those in elementary school, although stronger than those in middle school. Specifically, 66 percent of high school science lessons were rated as ineffective or marginally effective (Levels 1 and 2 on the 5-point scale), compared with 54 percent of elementary and 78 percent of middle school lessons (see Boxes 3-1 and 3-2 for examples of exemplary and ineffective lessons, respectively).
High schools invest more heavily than middle and elementary schools in resources for science instruction—twice as much as middle and about four times as much as elementary schools (see Table 3-2). As at the middle and elementary school levels, most high school teachers report that their classes have access to the Internet, personal computers, and nongraphing calculators. However, high school teachers have greater access to more sophisticated scientific equipment, including microscopes, probes for collecting data, and graphing calculators (Banilower et al., 2013). This greater access to scientific equipment is reflected in higher percentages of high school teachers, relative to elementary and middle school teachers, who rate their facilities, equipment, consumable supplies, and instructional technology as adequate (see Table 3-2). Although the 48 percent of high school teachers rating their instructional technology as adequate is greater than the corresponding percentage of middle school teachers (37 percent), it is still less than half, and this may explain, in part, why only about a third of high school teachers reported using instructional technology in their most recent lesson.
Most high school teachers (77 percent) use commercially published textbooks or modules as the basis for instruction, relying more than teachers at lower levels on textbooks rather than modules (Banilower et al., 2013). During their science classes, high school teachers use textbooks and modules less extensively than teachers at lower levels: fewer than one-third use them for 50 percent or more of their science instructional time. Like middle school teachers, high school teachers often supplement textbooks and modules with other resources or skip parts they deem unimportant.
Summary of Science Instruction across Levels of Schooling
The vision of science teaching and learning portrayed by the Framework and NGSS will likely present a substantial challenge for many teachers, especially at the elementary level, but also at the middle and high school levels. Although the available research suggests that the classroom environment for learning is moderately well organized and characterized
by a climate that is generally positive and respectful toward students, other common themes point to potential priorities for teacher learning and support:
- Although students frequently engage in “active work,” it is often procedural and does not involve authentic forms of scientific practice or reasoning.
- Far too few teachers in American classrooms help students link activity to substantive science ideas.
- Teacher questioning and tasks in general do not demand much from students intellectually; instruction is frequently aimed at the recall and reproduction of textbook explanations.
- Big-picture science ideas for students to develop understandings of or for teachers to organize units around are rare.
The committee does not wish to imply that instructional excellence and innovation do not exist in U.S. schools. Recall that in their observations of science classrooms, Weiss and colleagues (2003) found compelling examples of excellent instruction, albeit in only 15 percent of the classroom sample. Furthermore, an extensive descriptive literature portrays quality science instruction. For example, some elementary teachers integrate science into their curriculum, support meaningful science learning, and find ways to engage in their own professional learning—all despite working in teaching contexts and with curricula that rarely support such integration (Banilower et al., 2013; Dorph et al., 2011; Gallas, 1995; National Research Council, 2007). Similarly, some middle and high school teachers focus on fewer topics, exploring them with their students through investigations and providing time and structure for sense making (see Box 3-1). While excellence in science teaching is not yet widespread, then, it is important to remember that there are teachers in today’s schools who engage students in meaningful science learning. Indeed, their instruction inspired the committee’s notion of what Shulman (1986) has called the “images of the possible.” Yet even for these teachers, integrating the three dimensions described in the Framework (science practices, crosscutting concepts, and disciplinary core ideas) and creating coherent progressions that support students’ learning over months and years may represent a large change.
Certain factors exacerbate the disparity between vision and reality. At the elementary level, science is not taught much. With double periods of mathematics and language arts, there simply is not room in the school day for teaching science. At the middle and elementary school levels, teachers are underprepared to teach deep content and to focus on core ideas—they may not understand these ideas themselves. In high school, teachers too often are siloed in their own classrooms and certainly in their
own departments—an arrangement that again is antithetical to the notion of core ideas and of one learning experience serving as the basis for the next. In addition, high school science teachers often are uncomfortable supporting students in writing about science—or even reading work outside of texts. All of these problems are more pronounced and more challenging in schools that serve English language learners, students from underresourced homes, and students with disabilities. Finally, elementary, middle, and high school teachers have infrequent opportunities to interact with and learn from one another with respect to articulating students’ experiences across grade levels.
It is important to understand not only the characteristics of current dominant instructional practice but also science teachers’ perceptions of effective teaching and learning (i.e., what they think they should do to best help students learn science in their classrooms). Efforts to reform science instruction will depend on working closely with educators to alter or expand their current perceptions and aspirations.
Responses to the NSSME suggest that teachers regard organizing information, making clear presentations, and organizing for effective delivery as more important than student-oriented activity to effective science teaching (Banilower et al., 2013). More than 85 percent of teachers agreed that (1) students should be told the purpose for a lesson as it begins, (2) most class periods should include review of previously covered material, (3) most class periods should give students the opportunities to share their thinking/reasoning, and (4) most class periods should conclude with a summary of the key ideas addressed in that lesson. Three of these four statements (1, 2, and 4) appear to indicate that teachers view clear and well-organized representation of information as important to effective science instruction. Yet while this is an essential aspect of effective instruction, this view does not begin to include features of instruction that are more student oriented, including attention to the quality of student engagement and discourse.
Teachers’ opinions about ability grouping vary considerably by grade range, with 65 percent of high school science teachers, 48 percent of those in the middle grades, and 32 percent at the elementary level indicating that students learn science best in classes with other students of similar ability. On other statements in the survey, teachers’ opinions are largely consistent across grade ranges (Banilower et al., 2013, p. 21). For example:
- More than 75 percent of teachers at each grade range agreed that it is better to focus on ideas in depth, even if doing so means
covering fewer topics; this is one of the central tenets of calls for reform in science instruction. Although current practice does not reflect this emphasis on depth over coverage, it appears that teachers are ready to embrace this aspect of the new vision.
- Roughly 40 percent of science teachers at each grade level agreed that teachers should explain an idea to students before having them consider evidence for that idea.
- More than 50 percent indicated that laboratory activities should be used primarily to reinforce ideas that students have already learned. It is heartening that teachers appear to appreciate the importance of laboratory activities. However, this finding suggests that teachers need to consider the advantages of integrating scientific practices throughout all aspects of instruction, not merely as part of reinforcement.
- From 70 to 85 percent of science teachers at the various grade ranges indicated that students should be given definitions for new vocabulary at the beginning of instruction on a science idea. Taken together with using laboratory activities to illustrate or reinforce ideas, this view of instruction is aligned more with the conventional view of effective teaching focused on conveying final forms of knowledge to students than with the vision embodied in the Framework and NGSS. Those perceptions accord with instructional practices that are reported by many researchers.
In short, most U.S. teachers think of organizing information, making clear presentations, and organizing for effective delivery as important aspects of teaching that support student learning. A significant number of U.S. science teachers hold pedagogical perceptions that are aligned with a conventional view of teaching. Equally important, however, is that many teachers appreciate the importance of covering a smaller number of ideas in depth and the valuable role that laboratory activities can play.
A notable gap exists between the reality of current teaching practices and the vision of science learning that emerges from research on learning and teaching, as crystalized in the Framework and NGSS. Current science instruction places greater emphasis on ensuring that the learning environment is organized than on students’ sense-making activities (Weiss et al., 2003). Although teachers across grade levels report some use of such practices as “having students do hands-on/laboratory activities,” “requiring students to supply evidence in support of their claims,” and “having students represent and/or analyze data using tables, charts, or
graphs,” they spend most class time explaining science ideas or leading whole-class discussions (Banilower et al., 2013).
Activities for students sometimes include science practices but are rarely sequenced and integrated in ways that support focused learning of key science ideas (Roth et al., 2006; Weiss et al., 2003). Students rarely have time to make sense of the findings of their investigations or to engage in reflection on and revision of their understanding (Banilower et al., 2013; Weiss et al., 2003). In some cases, science lessons include no science ideas at all (Roth et al., 2006)—for example, a lesson on electric circuits in which students focus on lighting a light bulb with no mention of the idea of complete and incomplete circuits. Scientific investigations often involve performing iterative, dynamic, and inefficient activities, and the theorizing and interpretive work involved in moving from data to explanations and claims too often is missing from science classrooms (Windschitl et al., 2008).
Conclusion 1: An evolving understanding of how best to teach science, including the NGSS, represents a significant transition in the way science is currently taught in most classrooms and will require most science teachers to alter the way they teach.
It is critical, however, to resist the temptation to blame teachers for the current state of science teaching practices, which reflect the varied and underconceptualized support teachers receive from schools and districts. In addition to being prepared as generalists, elementary teachers have very limited time to plan and deliver science instruction, while teachers at all levels receive little time, structure, and support for their own learning, whether through traditional professional workshops or through teacher study groups or one-on-one coaching. Finally, resources for science are limited, and many teachers, especially at the elementary level, view the available equipment, supplies, facilities, and instructional technology as inadequate. These issues related to supports for high-quality science teaching are taken up in detail in the next three chapters.
Banilower, E.R., Smith, P.S., Weiss, I.R., Malzahn, K.A., Campbell, K.M., and Weis, A.M. (2013). Report of the 2012 National Survey of Science and Mathematics Education. Chapel Hill, NC: Horizon Research.
Bowes, A.S., and Banilower, E.R. (2004). LSC Classroom Observation Study: An Analysis of Data Collected Between 1998 and 2003. Chapel Hill, NC: Horizon Research.
Brookfield, S.D., and Preskill, S. (2005). Discussion as a Way of Teaching: Tools and Techniques for Democratic Classrooms (2nd ed.). San Francisco, CA: Jossey Bass.
Corcoran, T., and Gerry, G. (2011). Science Instruction in Newark Public Schools. Consortium for Policy Research in Education Research Report RR-71. New York: Teachers College, Columbia University.
Dorph, R., Goldstein, D., Lee, S., Lepori, K., Schneider, S., and Venkatesan, S. (2007). The Status of Science Education in the Bay Area: Research Brief. Berkeley: Lawrence Hall of Science, University of California. Available: http://lawrencehallofscience.org/rea/bayareastudy/pdf/final_to_print_research_brief.pdf [January 2015].
Dorph, R., Shields, P., Tiffany-Morales, J., Hartry, A., and McCaffrey, T. (2011). High Hopes-Few Opportunities: The Status of Elementary Science Education in California. Sacramento, CA: The Center for the Future of Teaching and Learning at WestEd. Available: http://www.wested.org/resources/high-hopes-mdash-few-opportunities-full-report-the-status-of-elementary-science-education-in-california [January 2015].
Engle, R.A., and Conant, F.R. (2002). Guiding principles for fostering productive disciplinary engagement: Explaining an emergent argument in a community of learners’ classroom. Cognition and Instruction, 20(4), 399-483.
Engle, R.A., Langer-Osuna, J.M., and de Royston, M.M. (2014). Toward a model of influence in persuasive discussions: Negotiating quality, authority, privilege, and access within a student-led argument. The Journal of Learning Sciences, 23(2), 245-268.
Gallas, K. (1995). Talking Their Way into Science: Hearing Children’s Questions and Theories, Responding with Curricula. New York: Teachers College Press.
Gerard, L.F., Varma, K., Corliss, S.B., and Linn, M.C. (2011). Professional development for technology-enhanced inquiry science. Review of Educational Research, 81(3), 408-448.
Lemke, J. (1990). Talking Science: Language, Learning, and Values. New York: Praeger.
National Research Council. (2007). Taking Science to School: Learning and Teaching Science in Grades K-8. Committee on Science Learning, Kindergarten through Eighth Grade. R.A. Duschl, H.A. Schweingruber, and A.W. Shouse (Eds.). Board on Science Education, Division of Behavioral and Social Sciences and Education. Washington, DC: The National Academies Press.
Nemirovsky, R., Rosebery, A.S., Solomon, J., and Warren, B. (2009). Everyday Matters in Science and Mathematics: Studies of Complex Classroom Events. Mahwah, NJ: Lawrence Erlbaum Associates.
Penuel, W.R., McWilliams, H., McAuliffe, C., Benbow, A., Mably, C., and Hayden, M. (2009). Teaching for understanding in Earth science: Comparing impacts on planning and instruction in three professional development designs for middle school science. Journal of Science Teacher Education, 20(5), 415-436.
Roth, K.J., and Garnier, H.G. (2007). What science teaching looks like: An international perspective. Educational Leadership, 64(4), 16-23.
Roth, K.J., Druker, S.L., Garnier, H.E., Lemmens, M., Chen, C., Kawanaka, T., Rasmussen, D., Trubacova, S., Warvi, D., Okamoto, Y., Gonzales, P., Stigler, J., and Gallimore, R. (2006). Teaching Science in Five Countries: Results from the TIMSS 1999 Video Study. NCES 2006-011. U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office.
Shulman, L.S. (1986). Paradigms and research programs in the study of teaching. In M.C. Wittrock (Ed.), Handbook of Research on Teaching (3rd ed., pp. 3-36). New York: MacMillan.
Smith, P.S., Banilower, E.R., McMahon, K.C., and Weiss, I.R. (2002). The National Survey of Science and Mathematics Education: Trends from 1977 to 2000. Available: http://2000survey.horizon-research.com/reports/trends.php [June 2015].
Trygstad, P.J. (2013). 2012 National Survey of Science and Mathematics Education: Status of Elementary School Science. Chapel Hill, NC: Horizon Research.
Weiss, I., Pasley, J., Smith, S., Banilower, E., and Heck, D. (2003). Looking Inside the Classroom: A Study of K-12 Mathematics and Science Education in the United States. Chapel Hill, NC: Horizon Research.
Windschitl, M., Thompson, J., and Braaten, M. (2008). How novice science teachers appropriate epistemic discourses around model-based inquiry for use in classrooms. Cognition and Instruction, 26(3), 310-378.
Yerrick, R., and Johnson, J. (2009). Meeting the needs of middle grade science learners through pedagogical and technological intervention. Contemporary Issues in Technology and Teacher Education, 9(3), 280-315.