Coleman McTigue Smolkin Elementary Teachers Use of Graphic Representations in Science

Elementary Teachers’ Use of Graphical Representations in Science Teaching

Julianne M. Coleman • Erin M. McTigue • Laura B. Smolkin

Published online: 15 July 2010

The Association for Science Teacher Education, USA 2010

Abstract The purpose of this study was to obtain data on United States K-5 elementary school teachers’ self-reported instructional practices with graphical representations. Via an electronic survey, 388 elementary teachers, from throughout the US, reported about their teaching of the interpretation and the production of graphics within science. The main findings indicate that: (1) pointing to or referring to graphical representations in books (92% of respondents) was the most frequently used instructional practice across the disciplines and grade levels; (2) five of nine graphical representations (over 90%) were more frequently used in science instruction than in other content areas, and (3) students’ graphical productions involving drawings, labeling, and oral and written explanations were very infrequent. The findings indicate that while teachers may tacitly use graphics within science instruction, they may not be explicitly teaching about this visual form of communication.

Keywords Visual literacy _ Graphics _ Science elementary _ Multiliteracies

Iconography comes upon us like a thief in the night—powerful and remarkably efficacious, yet often so silent that we do not detect the influence. Pictorial imagery catches us unawares because, as intellectuals, we are trained to analyze text and to treat drawings or photographs as trifling adjuncts. Thus, while we may pore over our words and examine them closely for biases and hidden meanings, we often view our pictures as frills and afterthoughts, simple illustrations of a natural reality or crutches for those who need a visual guide. We are most revealed in what we do not scrutinize (Gould, cited in Mishra 1999).

Visual representation has played a role in scientific communication since the

fifteenth century (Mishra 1999), and its place in the development of scientific

knowledge is increasingly acknowledged (e.g., Lynch and Woolgar 1990; National

Research Council [NRC] 2007). Science textbooks, a key element of science

instruction in the United States (Weiss et al. 2001) have become increasingly

graphics-laden (Martins 2002; Moss 2001; Roseman et al. 1999; Walpole 1998–

1999), and we are just beginning to learn how children approach such illustrated

science texts (Hannus and Hyona 1999; McTigue 2009; Stylianidou et al. 2002;

Walpole 1998–1999). A recent study examining released state science test items in

the United States, indicate that 52% of questions contain graphics and 79.5% of the

included graphics contain information that is essential to correctly answering the

question (Yeh and McTigue 2009). Moreover, teachers are advised and/or expected

to address graphical representations and models of various types in their science

instruction (e.g., American Association for Advancement of Science [AAS] 1993;

NRC 1996, 2007; Stylianidou et al. 2002). However, with the exception of a very

few studies (e.g., Smolkin and Donovan 2004), there is little known about what

elementary teachers actually do with graphics in their science instruction. The study

described in this paper represents a starting point for the examination of elementary

teachers’ pedagogical use (or nonuse) of visual and graphical information in

elementary science teaching. It seeks as well to understand whether teachers view

graphics, in Gould’s (Mishra 1999) words as, ‘‘frills and afterthoughts’’ or whether

they use them in ways that increase children’s abilities in scientific communications.

This work provides insights into the role of visual communication within

elementary science classrooms.

Within the following section which reviews the related literature that guided the

creation of our survey, we begin with the larger picture to summarize research on the

classification of graphics and on the use of graphics within the field of science. Next,

we summarize information on students’ abilities with graphics. Finally, we summarize

current teaching standards and resources for educators in teaching graphics.

Review of Related Literature

Organizational Schemes for Graphical Representations

Determining exactly what defines a graphical representation is not an easy task. In

order to ask teachers about their use of graphics, we needed to provide operational

definitions to insure that we were effectively communicating. Numerous theorists

and researchers (e.g., Doblin 1980; Hunter et al. 1987; Kress and van Leeuwen

1996) have attempted to create typologies for visual representations; this has led to a

flurry of confusing terminology. Confusing terminology limits potential communication

and advancement of graphics. Some individuals distinguish between

figurative and non-figurative representations (e.g., Doblin 1980; Petterson 2002),
with non-figurative representations including labels, letters, and verbal descriptions;

as opposed to figurative representations that rely upon pictures, schematics, and

various symbols. Within the studies typically concerned with figurative representations,

one important and helpful distinction (Doblin 1980; Goodman 1968; Vekeri

2002), is the separation of visual depictions into notational and non-notational

representations. Non-notational representations (what many would consider

pictures—paintings, drawings, photos) provide a complex, polysemic visual field

that mimics reality; notational representations seek to reduce reality in some way to

produce ‘‘a one-to-one correspondence between elements and their referents’’

(Vekeri 2002, p. 263). Vekiri suggested that within this notational world, there exist

four graphical categories: diagrams, maps, graphs (line, bar, and pie), and charts

(which would include both matrices and graphic organizers of various types). For

the purposes of this paper, we will focus upon notational representations.

Visual Representations in Science Instruction

Due to the abstract nature of many science principles (e.g., gravity), graphical

representations can play a powerful role in illustrating and explicating science to

novices by making concepts more concrete through the use of visual examples. For

example, in science, awareness of scale can be simply too large (e.g., continental

drift) or too small (e.g., bacteria) to observe in public elementary school classrooms

and teachers must rely on visual depictions of the target. Additionally, the

organization of elementary schools and the certification of elementary teachers

results in most elementary school science being taught by generalist teachers in

classrooms ill-equipped with scientific tools.

Historically, within the realm of science teaching, visual aspects have maintained a

minor role relative to written and oral/verbal forms of communication (Walker 1993;

Trumbo 1999). Through a meta-analysis of the use of graphics in schools, Winn

(1987, 1994) concluded that relatively little attention was given to the visual form of

communication. He termed this pattern of favoring the written word over the visual

form in schools as a verbal bias and he warned that this neglect of visual processing

could result in students failing to fully develop their abilities in visual processing (Leu

2000). As such, students may come to disregard graphics, rather than exploiting them

to their full communicative potential (Schnotz et al. 1993). The situation of students

disregarding graphics is problematic if we recognize the importance of graphics in

scientific communications (e.g., AAS 1993; Lemke 1990).

Children’s Abilities with Science-Related Visual Representations

Although there is relatively little research on children’s comprehension of scientific

graphical representations as compared with adults’ comprehension, what is known

suggests that the effective use of graphics must to be taught. We consider the

‘‘effective use of graphics’’ to be multifaceted and include being able to read a

graphic; to locate specific information within a graphic; to a create graphics to

organize information; and to communicate to others through the use of graphics.

Within the realm of interpreting graphics, Hannus and Hyona (1999), after working
with 10 year olds, suggested that science texts would benefit from specific verbal

cues, added into the textbase that provide guidance of how and when, to examine

the referent graphic. McTigue (2009), experimenting with such verbal cues

embedded in the textbase (e.g., look at the diagram now to examine the direction of

the bloodflow within the body), found them to be more effective with middle school

students who had higher levels of background knowledge. Her findings supported

those of Stylianidou et al. (2002), who conducted interviews with middle-schoolaged

children and documented the difficulties experienced by these students in

comprehending the graphics found in typical science textbook material. They

concluded that teachers ‘‘need to spend time and effort talking through the meaning

of the images’’ (p. 257). This conclusion, that instruction and experiences with

scientific models are essential to children’s interpretational development, has

recently been reinforced by the National Research Council (2007) as well as current

cognitive developmental research (e.g., Szechter and Liben 2004).

Teaching Expectations of the Science Education Community

The merit of multimodal communication in science has been affirmed through the

development of graphical learning goals in science by two important organizations

in the United States. The American Association for the Advancement of Science

(AAAS) sponsored Project 2061 and its companion report, Benchmarks for Science

Literacy (1993); the National Research Council (NRC), a part of the National

Academies, produced the National Science Education Standards (NSES). As

summarized in the following two sections, both organizations convey the

expectation that teachers will be instructing children in graphical literacy.

Benchmarks for Science Literacy

According to the Benchmarks for Science Literacy, in kindergarten through grade 2,

students are expected to create drawings of a target object or concept and correctly

represent the salient features. By the end of fifth grade, students are expected to

create sketches that explain either ideas or procedures as well as make use of

numerical data to describe and/or compare objects or events. By the end of eighth

grade, students are expected to be proficient in both the interpretation and

production of simple tables and graphs, which includes identifying relationships;

they are also to comprehend text materials that contain a variety of graphs,

diagrams, charts and tables, and symbols of science.

National Science Education Standards

NSES also reflects an emphasis on communication. For example, within the

‘‘Science as Inquiry’’ strand for K-4 students, communication about investigations

and supplying explanations, seen as fundamental abilities in scientific inquiry, are to

be delivered in three modes: oral, written, and graphical (drawn). This attention to

drawing is reinforced, for example, in discussions of the physical science standards

for this grade range: ‘‘initial sketches and single-word descriptions lead to
increasingly more detailed drawings and richer verbal descriptions’’ (NRC 1996,

p. 123). NSES also emphasizes attention to drawing within earth and space science

content standards, recommending that children draw sketches of the moon and that

older children in this age range can learn to search for patterns of meaning by

‘‘recording data and making graphs and charts’’ (p. 126). As children progress into

middle schools, NSES standards increase expectations for children’s use of graphic

representations. Students, as they progress from grades five through eight, are

expected to use ‘‘the language of science’’ which includes ‘‘writing, labeling

drawings, completing concept maps, developing spreadsheets, and designing

computer graphics’’ (NRC 1996, p. 144). It is clear from these documents, then,

that teachers are expected to include a visual component in their teaching of science.

NRC’s (2007) recent volume, Taking science to school, continues to stress the

importance of instruction: ‘‘development [in graphical skills] is significantly

enhanced by prior knowledge, experience, and instruction (p. 159, emphasis

added).

Assistance for Teachers’ Graphical Practices in Science Instruction

Beyond the Benchmarks and Standards, teachers have other textual sources that

encourage them to include graphical representations in science instruction. These

include teachers’ manuals accompanying textbooks and trade publications.

Perhaps because of their own notable insecurities regarding their science teaching

abilities (Weiss et al. 2001), elementary teachers have relied heavily on textbooks

for their science instruction, with 85% of fifth through eighth grade teachers

typically employing a single textbook for their instruction, while 64% of

kindergarten through fourth grade teachers use a science textbook. Recent research

(e.g., Carneiro and Freitas, cited in Freitas 2007) suggests that, among textbooks,

teachers favor graphics-heavy volumes over those with fewer pictures.

Although science textbooks frequently are attacked in terms of comprehensibility

and accuracy of representation (e.g., Best et al. 2005; Hubisz 2000; Kesidou and

Roseman 2002), they nonetheless provide guidance for teachers’ use of graphics in

science instruction. For example, in a recent Scott Foresman kindergarten teachers’

manual, teachers are instructed to have children create a class mural for their living

things unit. In line with NSES and Benchmark suggestions, teachers are told that

each child can draw and then label (or dictate, depending on abilities) the names and

needs of various living things (Cooney et al. 2006a, b, p. 25).

In addition to ideas in textbooks, trade publishing houses that focus on teachers

have produced texts to support the incorporation of visual representations. In our

work with teachers, to provide a common language with which to discuss graphics,

we have focused on one of the only currently available teacher texts regarding

visual literacy: the Stenhouse offering, I see what you mean (Moline 1995). This

simply-presented text, by Australian educator Moline, has been continuously

available for teachers for more than a decade and provides additional support

through an inexpensive video as well as a related website (Moline 2006). Moline’s

typology of graphics (see Table 1) is congruent and supported by more recent

research in the field, such as Vekeri (2002). These categories of graphical
communication are made more explicit for teachers through numerous examples of

children’s work. Although this text does not reach the level of depth such as the

work in semiotics by Kress and Van Leeuwen (1996) or Roth’s theory of

representation through inscription (e.g., Roth and McKinn 1998), it provides a

framework for static graphics (e.g., flow charts) with familiar vocabulary.

Use of Graphics and the Role of Visual Literacy in Elementary Science

Teaching

Although the corpus is limited, studies reporting teachers’ graphical practices

suggest that when teaching science, teachers may not be guiding children

particularly well in terms of graphical interpretations (obviously a problem

throughout all science instruction, not just graphical interpretation). For example,

when observing primary grade teachers reading aloud from a science trade book,

which displayed a multi-graphic layout including a cross-sectional diagram and a

map, Smolkin and Donovan (2004) found that teachers rarely guided the students

through navigation of the multiple graphics. Instead, teachers focused students’

attention on just the most salient graphic—a single map and did not address the

meaning of the cross-sectional diagram at all.

What actually occurs today in elementary science classrooms regarding the

teaching of graphical representations is unknown. Due to the rapidly changing and

increasing emphasis on visual and graphical communication (Pozzer-Ardenghi and

Roth 2005), current research is needed to determine how much and in what manner

teachers are actually using graphics within their teaching. For our research, we

posed the following questions:

(1) What types of graphical representations are used most frequently by K-5

elementary teachers in science compared to the other content areas?

(2) Within science instruction, do teachers report particular types of graphical

representations as used more frequently than others?

(3) What science teaching practices and activities related to the interpretation of

graphical representations do teachers report as frequently used across the K-5

grade levels?

(4) What teaching practices and activities involving the production of graphical

representations do teachers report as frequently used across the elementary

grades?

Method

Participants

We chose a national sample to give breadth to our survey and to eliminate specific

patterns of instruction that might be associated with particular state standards. To

obtain a random sampling of K-5 elementary teachers across the United States, we

employed the services of an educational marketing firm based in Northern Virginia.

This firm compiles national databases of teachers for the specific uses of educational

research and marketing; former customers include the International Reading

Association. The information in the database has been gathered voluntarily from

school districts across the United States.

The sampling parameters set for this study included classroom teachers in

kindergarten through fifth grades. Although sixth grade has been commonly
considered as elementary school, current trends find sixth grade classes located in

middle schools where teaching is far more departmentalized; this departmentalization

would have prohibited our determining whether teachers were more inclined to

use graphics in science instruction instead of other subjects in their curricula.

According to the National Center for Educational Statistics (2001), 65% of

elementary schools include kindergarten through grade five.

Opting for the Electronic Survey

Various studies have promoted the use of electronic or Web-based surveys as a

viable way to conduct research (e.g., Couper et al. 2001). We elected to use an

electronic mode for several reasons primarily because of the costs incurred by

printing (Cobanoglu et al. 2001). Given the importance of color as a distinguishing

feature in many of the images we sought to examine (e.g., Stone et al. 2006; Tufte

1990) and given the length of the survey with all images embedded, printing costs

seemed prohibitive. Although some research suggests that overall response rates for

e-mail surveys are somewhat lower than paper and pencil surveys (Anderson and

Gansneder 1995), some studies have had response rates as high as 70% (Yun and

Trumbo 2000). There has been some suggestion that, given concerns with lower

response rates for web surveys, a key issue in interpretation of results will be sample

representativeness (Dillman 2000). Concerned that our response rate might also be

limited, we sampled a substantive portion of the population, hoping minimally to

produce a representative sample. To this end, we gathered descriptive information to

compare to nationally reported demographics.

Demographics of Respondents

We were informed that 5,000 electronic survey questionnaires were sent to

prospective participants across the United States. Of the 405 returned surveys, 388

were usable. The response rate was low; however, the respondents represented a fairly

even distribution across the elementary grades: kindergarten teachers represented

17.5% of the response sample, first grade teachers 14.9%, second grade teachers

18.3%, third grade teachers 18%, fourth grade teachers 20.3% and fifth grade 17.5%.

As described above, to further address sample representativeness, we conducted an

analysis of the information regarding their demographics and education. The majority

of the respondents identified themselves as female (91.9%); 8.1% of the respondents

were male. Nationally, 91% of elementary teachers are female and 8% are male

(National Education Association 2006) which does not differ significantly from the

percentages obtained in this study (x2 = 0.05, p = 0.82). In fact, the proportion of

males to females in this study is statistically identical to the national percentages of

males to females (National Education Association 2006). When asked about their

educational level, over half (56.7%) the teachers indicated they held a master’s degree

or higher. There is no statistically significant difference in the proportion of

respondents holding a master’s degree or higher in the sample and the proportion

nationally (x2 = 0.002, p = 0.96) as reported by the National Education Association

(2006). The majority of the other respondents held a bachelor’s degree (43.3%) and
4.4% held a specialist’s degree of some type. Only one respondent (.3%) indicated

having a doctoral degree. The concordance between the sample in this study and the

national percentages is especially important to help establish representativeness of the

sample, given a possible low response rate. Comparing the population in this study to

national statistics for education level (x2 = 0.002, p = 0.96) and gender (x2 = 0.05,

p = 0.82), the used sample in this study suggests it is representative of elementary

teachers across the United States. Due to the principles of computer generated random

sampling, we did not deem it necessary to calculate the geographical locations of the

teachers.

Materials

Development of the Survey Instrument

In developing the survey, we established three major research domains: Teacher

Demographics and Education, School and Student Demographics, and Teacher

Instructional Practices. We were strongly influenced by two literacy survey studies,

research conducted by Baumann et al. (2000) and by Commeyras and De Groff

(1998). The Bauman et al. study has been identified as an exemplary model of survey

research involving literacy (Duke and Malette 2004), and we determined to replicate

and/or modify appropriate aspects of this survey as we constructed this instrument.

Teacher Demographics and Education: For the first domain, we designed nine

survey items to obtain information about the participants’ educational background,

teaching background, and gender. In particular, this domain included the adaptation

of Baumann’s et al.’s survey items and question stems.

School and Student Demographics: For the second domain, we designed six survey

items to obtain information about the types of students in the participants’

classrooms, the types of communities in which the participants’ schools were

located, and the size of the schools where the participants worked. We again

adapted survey items utilized in the Baumann et al. study and included modified

aspects of the Commeyras and De Groff (1998) study because of its similarities in

research design and general procedures. These modified aspects related to question

stems, question formats, and to the order in which the survey items were presented.

Teacher Instructional Practices For the third domain of the survey, we relied upon

Moline’s (1995) classification system (See Table 1) as the basis for the construction

of the survey items for reasons we have noted earlier. This domain consisted of two

parts. The first part contained the survey items, which focused upon the different

types of graphical representations and the frequency with which they were used in

each of the typical content areas in elementary classrooms: reading, math, social

studies, and science. We were most interested in the reports of frequent use because

research (e.g., Ainsworth 1999) indicates that students need multiple exposures to

graphical representations to fully exploit their potential and because research
additionally indicates that repeated practice will likely lead to idea retention (Brown

et al. 1981). The second part of this domain consisted of questions regarding the

various instructional practices and activities that teachers may or may not have used

involving graphical representations. Part Two of the domain, Teacher Instructional

Practices, then, gathered information regarding the frequency (frequently, sometimes,

rarely, never) with which participants used different instructional activities

and practices involving graphical representations in science. In our conceptualization

of teaching practices involving graphical representations in science, we recognized

that messages, whether text or graphic, must be interpreted to be comprehended.

Survey Construction

The survey was constructed at the online survey website, (A text version of this

survey can be found in the ‘‘Appendix’’). This on-line format allowed us to provide

full color exemplars of each target graphical representation, which would have been

cost prohibitive in a paper and pencil survey.

Pilot Testing

Two pilot tests were run, the first among a sample of former and current classroom

teachers and the second among a sample of educational researchers and methodologists.

Revisions from this feedback included (1) revising the wording of several

questions, (2) reordering the items presented in Teachers’ Instructional Practice

domain, (3) addressing several technical issues, and (4) modifying the estimated

completion time. This process was important to establishing content validity.

Data Collection Procedures

Our first step in data collection was to send an e-mail message to the individuals on the

random sample list of general elementary teachers supplied by the educational

marketing firm to request their participation in the survey. Directions for participation

in the survey, including the URL ink to the survey, were incorporated in that first

e-mail message and prospective participants were informed that they had 1 week to

participate in the survey. Three weeks later, we sent a follow-up e-mail message again

requesting participation in the survey to those who had not yet completed it. Once the

follow-up email was sent, prospective participants had an additional 7 days to

complete the survey. The on-line survey was closed after the 7 days and was no longer

available to the participants. In total, the on-line survey was available to prospective

participants for two-one-week periods, a total of 14 days. According to the

educational marketing firm, this is standard practice for the use of online surveys,

although a longer time period may result in greater response rates.

Data Analysis

The data analysis in this study consisted of two phases. The first phase was

completed by the on-line survey company, SurveyMonkey.com. As participants
completed the survey, their responses were collected by the online survey software

data collection tool, which computes basic statistical analyses and reports response

rates by survey items as well as frequencies by survey items.

During the second phase, the data was exported from SurveyMonkey.com and

imported into SPSS. Incomplete surveys were removed from further analysis. Within

SPSS, statistical analyses were performed. The data was analyzed using simple

descriptive statistics (e.g., summary statistics by survey items and frequencies).

Results

Graphics Use in Science Instruction

Table 2 reports teachers’ responses to the question about their use of graphical

types, frequency, and the discipline within which teachers were most likely to use

particular graphics. Of the 14 graphical representations contained in the survey,

eight were reported by teachers across the grades as more frequently used in science

instruction than in the content areas of social studies, math, or reading/language arts

instruction. These included flow diagrams (96.7% of respondents reporting their

usage in science), picture glossaries (94.6%), cross-section diagrams (92.5%), web

diagrams (92.3%), cutaway diagrams (92.1%), tables (87.9%), tree diagrams (81%),

and scale diagrams (78.2%). Five of these eight graphical representations were

reported as used in science instruction by more than 90% of survey participants.

Teacher Practices for Assisting Children in Interpreting Graphics

Table 3 reports teachers’ practices designed to develop children’s ability to interpret

graphics. We begin this section by noting grade level trends in terms of graphical

interpretations. For virtually all queries, kindergarten and first grade teachers

provided very different responses in terms of frequency of use than third through

fifth grade teachers. Second grade teachers’ responses were sometimes similar to

kindergarten and first grade teachers, and, at other times, more resembled the upper

grade teachers’ responses. These grade level differences are particularly notable for

Items 3 (internal structures), 5 (explaining hidden processes from analytic

diagrams), 6 (written text from graphical representations), and 8 (captions for

graphical representations).

Pointing

The most frequently used instructional practice was the teacher pointing to the

graphical representations in the text (65% overall of respondents indicated that they

engaged in this behavior). This practice is reported as frequently used by

kindergarten teachers (26% of the time) and first grade teachers (38.6%). However,

the use of this practice of pointing is more evident in the older grades: second

(80.4%).third (86.5%) fourth (74.6%) and fifth (72.6%) grade teachers.

Tables

The second most frequently used graphical practice was the interpretation of

tables, with 45% of all teachers reporting having children link information in

cells, row headings, and column headings to interpret graphical meaning. Again,

kindergarten (14.5%) and first grade (34.9%) teachers reported using this practice

less often than their upper grade peers. A higher percentage of third grade

teachers (60.7%) reported this as a frequent practice than teachers at any other

grade level.

Analytic Diagrams

The third most commonly reported practice for interpretation relates specifically to

Moline’s (1995) analytical diagram category, which includes cutaways and crosssections,

with 33% of all teachers reporting having children explain internal and

external structures. Attention to interpreting these analytical diagrams increases

across the elementary grades, from a low of 3.6% of kindergarten teachers to a high

of 46% of fifth grade teachers reporting this as a frequently used practice. However,

when cutaways depict hidden processes, such as the workings of a steam engine, as

opposed to the naming of internal and external structures, teachers reported having

students explain those processes far less regularly. Only 17% of all teachers

indicated they did so frequently, while 47% indicated that they did so rarely or

never.

Teacher Practices for Assisting Children in Producing Graphics

Table 4 presents results for teachers’ practices related to the production of graphics.

Unlike practices related to interpretation of graphics (Table 3), there is less grade

level variation. Of note, drawing and labeling graphical representations (frequently

used by on only 6% of respondents) is a far less common practice in kindergarten

through second grades than in third through fifth grades.

Synthetic Diagrams

The two practices teachers reported using frequently when having children produce

graphics both fall into Moline’s (1995) category of synthetic diagrams (making
connections among parts; flow diagrams and tree diagrams). Forty-two percent of all

teachers reported having students organize information into meaningful sequences

(graphical example was a flow diagram). Although this instructional activity was

less frequently reported as used by kindergarten (21.6%) and first grade (34.9%)

teachers, approximately half of the teachers of the upper elementary grades (third

grade, 54.5%), fourth grade (46.8%), fifth grade (53.8%) reported this as a frequent

practice.

The second most frequently reported practice for producing graphics also

involves a synthetic diagram, in this case, organizing information hierarchically in

tree diagrams. Again, approximately half of third (49.1%), fourth (48.3%), and fifth

(52.3%) grade teachers reported having their students create such diagrams

frequently. Overall, 41% of all teachers report this as a frequent practice in

graphical production.

Drawing and Labeling Details

The least frequently employed productive practice was drawing and labeling details

of a graphical representation. Whereas 6% of teachers using this practice frequently,

73% of respondents indicated that they rarely or never did so.

Discussion

Teacher Practices with Graphics

Grade Level Differences

Although one might naturally expect an increase in work with graphics across the

elementary grades, our results do suggest some concerns for kindergarten and first

grade. Fewer than five percent of these teachers reported employing cutaways and

cross-sections to explain hidden processes. This finding suggests that explanations

of how things work in the world are not commonly occurring in these early

childhood grades. Current high quality information books written for young children

books share the common trait of containing complex graphics including cutaways

(Kurkjian and Livingston 2005). If teachers are not addressing the use of such

graphics then they are likely not reading such informational books, or they are

disregarding such graphics. Considering teachers’ low self confidence for science

teaching (e.g., Weiss et al. 2001), this finding may not be surprising, but it does

suggest that the groundwork for explanations, critical in science education (NRC

1996, 2007) is not being laid. For example, while kindergarten students may not be

ready to understand a cutaway representation of a brain, they can understand and

produce cutaways that are more concrete—such as considering the inside and

outside structure of an apple. Practice with this type of representation will prepare

them for more complex versions of cutaways and cross-section.

Pointing

By far, teachers’ most commonly reported instructional graphical practice (92%

reported doing so frequently or sometimes) is pointing at them in books. This

reinforces the findings of Weiss et al. (2001) survey showing the high presence of

text materials used in U.S. elementary school science instruction and brings into

question the numbers of teachers actually engaging in discovery related practices.
The use of graphics would depend greatly on the approach of teaching science. If

science instruction is text-based, with students predominantly reading about science,

then students would need to be able to interpret graphics but would have less

impetus to produce graphics. In a more inquiry based approach, which replicates real

science, then students would need to interpret and produce graphics in order to

organize and communicate their findings. Unfortunately, some interpretations of

‘‘hands-on’’ and ‘‘inquiry based’’ science teaching over-rely on the doing of science

rather than the thinking about science. In such cases, the misinterpretation of ‘‘inquiry

based’’ science does replicate real science because students do not frequently engage

in talking, reading, writing and drawing about science (Hand and Prain 2006).

While pointing at diagrams does draw children’s attention to them and create

some awareness, research suggests this is an insufficient instructional practice.

Peeck (1993), reviewing established research findings on pictorial representations,

indicated that simply drawing students’ attention to pictures does little to support

the processing of the representations. Given these findings, Peek’s advice was that

instructors ‘‘tell the student to do something with the illustration’’ (p. 235) and

suggested that related tasks yield an examinable product, such as having students

label features in an illustration. Stern et al. (2003) paralleled Peeck’s stance,

positing that simply viewing graphics does not grant the learner a deeper

understanding of graph design. Any learning derived from the particular graphic

will remain situated in the particular text, not transferred to other texts. In contrast,

when actively creating a graphic (a practice 73% of teachers indicated they rarely or

seldom offered for students, see below), learners build awareness of the conventions

of graphics and can apply this knowledge to a novel graphic (Moline 1995; Wheeler

and Hill 1990).

One potential reason for the high frequency of pointing is that teachers can

determine to which aspects of the graphical representation they will attend (as in the

Smolkin and Donovan (2004) example noted earlier in this paper). It is possible,

given teachers’ low confidence in their science knowledge (Weiss et al. 2001), that

pointing enables teachers to address only those aspects of the graphical representation

with which they themselves feel comfortable.

Drawing and Labeling

Although Peeck (1993) and Stern et al. (2003) suggested creating graphics to

increase student understanding, the results of our study indicate that production

through drawing was an infrequent graphical practice (never or rarely done, 73% of

responding teachers). This finding is particularly odd in light of the benchmarks and

standards which specifically encourage such practices.

We have pondered why drawing and labeling appears to be an uncommon practice.

One possibility relates to teachers’ own understanding or a lack of awareness of the

importance of drawing as a form of communication (verbal bias); another relates to

their own ease, or comfort, with drawing. Anning (1997) has commented that for

teachers, ‘‘drawing is a minor mode of communication, certainly secondary to writing

and speech ….Indeed in many classrooms drawing is more likely to be caught than

taught’’ (p. 219). Anning’s assertions are echoed in survey data examined by

Chapman (2005). Chapman noted, ‘‘course requirements in art are minimal for

[elementary] teachers’’ (p. 120). Reviewing a 2002 survey by NCES, Chapman

indicated that only about 10% of elementary classroom teachers have strong

qualifications and interests in art. Whatever the cause of infrequent drawing, its lack

suggests that children are not engaging in the creation of visual records seen by AAS

and NRC as important to science understanding.

Reports of low frequency of use in labeling drawings may also be linked to other

findings regarding writing as a mode of explanation. When asked if they had

children convert graphical information into written texts, 49% of the respondents

indicated that they never or rarely did so. When asked if they had students create

captions, which is generally seen as involving less writing than a ‘‘text,’’ the

percentage of teachers indicating they never or rarely did so increased to 63%.

These findings suggest that the teachers’ view of student communications may favor

an oral mode. Our finding that 65% of teachers indicated that they frequently or

sometimes had students explain concepts or objects depicted in graphical

representations lends support to this assumption. However, we also note that only

22% of all teachers reported having their students frequently explain concepts or

objects within graphical representations, while 35% of all teachers indicated that

they rarely or never did so. In short, teachers do not appear to be strongly

promoting, either visually, textually, or orally the production of explanations.

The Content of Explanations

Still, if 65% of the teachers are having their students create oral explanations

frequently or sometimes (as opposed to the little used written or graphical/visual

explanations), the next question becomes: what are children explaining? When we

look at our results, we see a distinction regarding analytic diagrams between having

children ‘‘explain’’ structure and having them explain processes. While 71% of

respondents indicated they frequently or sometimes had children explain (possibly

meaning describe or identify) internal and external structure with cutaways and

cross sections, that percentage drops to 53% for explaining hidden processes

associated with such diagrams. Newton et al. (2002)’s study of elementary teachers

in Years 3 through 6 in British classrooms found very few instances in which

teachers, in their science instruction, focused on causes and reasons for events,

which would have entailed explanations of processes. Instead, teachers’ discourse

focused on facts and descriptions for virtually 40% of total lesson time; explanatory

discourse occurred only 9% of the total lesson time. Causal explanations of

unobservable processes, then, seem to garner less teacher attention than do

‘‘explanations’’ we might more commonly think of as descriptions.

Conclusions

Our examination of U.S. elementary classrooms suggests that, according to selfreport,

teachers are not likely employing graphical representations to their fullest

potential when teaching science. Their instruction in the use of graphics is limited

by both frequency and depth of instruction. There are various reasons why teachers

may not be more explicit in their work with graphical representations. Teachers may

underestimate the complexity of graphical images and therefore feel little need to

explicitly teach the decoding of graphics (Kress and van Leeuwen 1996). They may

not be cognizant of their own learning and, therefore, find it difficult to explain

graphical representations to young learners (Henderson 1999).

Regardless of the reason, this work indicates that elementary teachers associate

particular graphical forms with science instruction but are doing relatively little to

build their students’ ability to interpret or produce these important scientific

communicative skills. In order for students to become scientifically literate, they

need to know the facts, and process of science as well as the ability to effectively

communicate science. Using graphical representations are critical to being

scientifically literate. As such, teachers should not trust that students will tacitly

gain this knowledge but instruction in this area should be explicit and embedded

within authentic science inquiry.

Limitations

Although our intention was to construct a study with high levels of methodological

worthiness, there are certainly limitations to this starting point study. Most notable

was the very low response rate (7.75%). We believe that our use of an external

company for the lists of teachers compounded difficulties in response rate. For

example, the company never reported to us how many of the original 5,000 e-mail

messages bounced back with incorrect addresses. We had no way of knowing

exactly how many individuals received the e-mail message inviting them to

participate. Then, too, we conjectured that many e-mail recipients, having received

the message and being instructed to click on a link to complete the survey, hesitated

to do so because of the heavy amount of virus-associated spam circulating at that

time. Because the company controlled the national sample of teachers, we had no

way to contact them outside of the company’s parameters. However, given that this

study represents a starting point for research on graphical practices in elementary

science teaching, we believe that the representativeness of the sample allows us to

draw some tentative conclusions regarding elementary teachers’ practices with

graphics. Additionally, it would have been informative if we had asked teachers

more general questions about their approach to teaching science. For example, in

inquiry based curriculums, students may be using graphics in more authentic

manners than in textbook driven curriculums. It would provide insight to have more

information about the overall teaching practices. We recommend this for future

research.

References

Ainsworth, S. E. (1999). The function of multiple representations. Computers and Education, 33(2/3),

131–152.

American Association for Advancement of Science, P. (1993). Benchmarks for science literacy. Oxford:

Oxford University Press.

Anderson, S. E., & Gansneder, B. M. (1995). Using electronic mail surveys and computer-monitored data

for studying computer-mediated communication systems. Social Science Computer Review, 13(1),

33–46.

Anning, A. (1997). Drawing out ideas: Graphicacy and young children. International Journal of

Technology and Design Education, 7, 219–239.

Baumann, J. F., Hoffman, J. V., Duffy-Hestor, A., & Ro, J. M. (2000). The First R yesterday and today:

U.S. elementary reading instruction practices reported by teachers and administrators. Reading

Research Quarterly, 35(3), 338–377.

Best, R. M., Rowe, M., Ozuru, Y., & McNamara, D. (2005). Deep-level comprehension of science texts:

The role of the reader and the text. Topics in Language Disorders, 25(1), 65–83.

Brown, A. L., Campione, J. C., & Day, J. D. (1981). Learning to learn: On training students to learn from

text. Educational Researcher, 10(2), 14–21.

Chapman, L. H. (2005). Status of elementary art education: 1997–2004. Studies in Art Education, 46(2),

118–137.

Cobanoglu, C., Warde, B., & Moreo, P. J. (2001). A comparison of mail, fax, and web survey methods.

International Journal of Market Research, 43, 441–552.

Commeyras, M., & De Groff, L. (1998). Literacy professionals’ perspectives on professional

development and pedagogy: A United states survey. Reading Research Quarterly, 33(4), 434–472.

Cooney, T., Cummins, J., Foots, B., Flood, J., Goldston, M. J., Gholston-Key, S., et al. (2006a). See

learning in a whole new light: Volume 1 (Grade K Teacher’s Edition). Illinois: Pearson Education.

Cooney, T., Foots, B., Flood, J., Goldston, M. J., Gholston-Key, S., Lapp, D., et al. (2006b). See learning

in a whole new light: Volume 1 (Grade 3 Teacher’s Edition). Illinois: Pearson Education.

Couper, M. P., Traugott, M. W., & Lamias, M. J. (2001). Web survey design and administration. Public

Opinion Quarterly, 64, 464–494.

Dillman, D. (2000). Mail and internet surveys: The tailored design method (2nd ed.) . New York: Wiley.

Doblin, J. (1980). A structure for nontextual communication. In P. Kolers, M. Wrolstad, & H. Bouma

(Eds.), The processing of visible language (Vol. 2, pp. 89–111). New York: Plenum Press.

Duke, N., & Malette, M. (2004). Literacy research methodologies. New York: Guilford Press.

Freitas, C. A. (2007). Talked images: Examining the contextualised nature of image use. Pedagogies, 2,

151–164.

Goodman, N. (1968). The languages of art. Indianapolis: Hackett Publishing Company.

Hand, B., & Prain, V. (2006). Moving from border crossing to convergences of perspectives in language

and science literacy research and practices. International Journal of Science Education, 28(2–3),

101–107.

Hannus, M., & Hyona, J. (1999). Utilization of illustrations during learning of science textbook passages

among low- and high-ability children. Contemporary Educational Psychology, 24, 95–123.

Henderson, G. (1999). Learning with diagrams. Australian Science Teachers’ Journal, 45(2), 17–25.

Hubisz, J. L. (2000). Report on a study of middle school physical science texts. The Physics Teacher, 39,

304–309.

Hunter, B., Crismore, A., & Pearson, P. D. (1987). Visual displays in basal readers and social science

textbooks. In D. Willows & H. A. Houghton (Eds.), The psychology of illustration: Vol. 2.

Instructional issues (pp. 116–135). New York: Springer-Verlag.

Kesidou, S., & Roseman, J. E. (2002). How well do middle school science programs measure up?

Findings from project 2061’s curriculum review. Journal of Research in Science Teaching, 39,

522–549.

Kress, G., & van Leeuwen, T. (1996). Reading images: The grammar of visual design. London:

Routledge.

Kurkjian, C., & Livingston, N. (2005). Learning to read and reading to learn: Informational series books.

Reading Teacher, 58, 592–600.

Lemke, J. (1990). Talking science: Language, learning, and values. Norwood, NJ: Ablex.

Leu, D. J. (2000). Our children’s future: Changing the focus of literacy and literacy instruction. The

Reading Teacher, 53, 424–430.

Lynch, M., & Woolgar, S. (Eds.) . (1990). Representation in scientific practice. Cambridge, MA: MIT

Press.

Martins, I. (2002). Visual imagery in school science texts. In J. Otero, J. A. Leon, & A. C. Graesser

(Eds.), The psychology of science text comprehension (pp. 73–90). Mahwah, N.J.: Lawrence

Erlbaum Associates.

McTigue, E. M. (2009). Does multimedia theory extend to middle school students? Contemporary

Educational Psychology, 34, 143–153.

Mishra, P. (1999). The role of abstraction in scientific illustrations: Implications for pedagogy. Journal of

Visual Literacy, 19(2), 139–158.

Moline, S. (1995). I see what you mean. York, ME: Stenhouse Publishing.

Moline, S. (2006). I see what you mean. Retrieved June 26, 2008, from http://k-8visual.info/.

Moss, G. (2001). To work or play? Junior age nonfiction as objects of design. Reading, 35, 106–110.

National Center for Educational Statistics (2001). Fast response survey system, survey of classes that

serve children prior to kindergarten in public schools: 2000–2001. U.S. Department of Education.

National Education Association, The (2006). National teacher day spotlights key issues facing profession:

NEA addresses top five teaching trends and outlines ‘portrait of american teacher’. Retrieved April

4, 2006, from http://www.nea.org/app/search/performSearch.do.

National Research Council. (1996). National science education standards. Washington, D.C: National

Academy Press.

National Research Council. (2007). Taking science to school: Learning and teaching science in grades

K-8. Washington, D.C.: National Academy Press.

Newton, L. D., Newton, D. P., Blake, A., & Brown, K. (2002). Do primary school science books for

children show a concern for explanatory understanding? Research in Science and Technological

Education, 20, 227–240.

Peeck, J. (1993). Increasing picture effects in learning from illustrated text. Learning and Instruction, 3,

227–238.

Petterson, R. (2002). Information design: An introduction. Philadelphia: John Benjamin.

Pozzer-Ardenghi, L. L., & Roth, W. M. (2005). Making sense of photographs. Science Education, 89,

219–241.

Roseman, J., Kesidou, S., Stern, L., & Caldwell, A. (1999). Heavy books light on learning: AAAS project

2061 evaluates middle grades science textbooks. Science Books and Films, 35, 243–247.

DMU Timestamp: May 11, 2020 21:16