Students' Science Websites:
What Do they Tell Us About Science Teaching
by Victoria Deneroff
Information Science 246
Prof. Howard Besser

As computer technology fast becomes routine in K-12 schools in the U.S., studies of effects of computer use in schools have begun to accumulate. School being a complex place, the answers are not clear-cut or simple: The evidence shows that computers can  either positively or negatively affects on student achievement, depending on how they are used. In this paper I sample the current research on the effects of technology in classrooms and attempt to develop a protocol for analyzing student work posted on student web sites on science topics. The results of this analysis provide a way of thinking about students' classroom use of computers and its meaning in the broader context of the cultural institutions of schooling.

The late 80's and early 90's were full of euphoria about the wonderful possibilities of computer technology to in helping "fix" the many problems of American education. Some envisioned the end of schools, with students sitting at computers programmed to teach them the skills they seemed unable to learn in classrooms (especially inner-city classrooms). Some of these visions went so far as to propose the end of teaching and schools as we know them. Less radical visions still included computerized tutorials to drill students and free teachers from the drudgery of routine tasks, allowing them to interact with students in more meaningful and creative ways. Surely the students of the year 2000 would be sophisticated technology users.

Predictions to the contrary, computer use by students in the course of normal classroom routine and activity is relatively rare, despite the fact that more than half of American classrooms are now equipped with at least one computer. (Fatemi, p. 1) Anecdotes abound of computers sitting unused in schools. In a survey of 15,000 teachers (1,407 of whom responded) 53 percent of all teachers replying said they used software for classroom instruction, and 61 percent said they used the Internet for instruction. However, 77% of those who used software and 88% of those who used the Internet viewed it as supplementary to the curriculum ( Education Week's 1999 National Survey of Teachers' Use of Digital Content, Key Results ). The fact that less than 10% of the teachers responded leaves the result in question, but it is clear that a small but substantial proportion of teachers use computers for instruction.

Since the 1980's Americans have felt compelled to have computers in their children's classrooms, going so far as to measure of the quality of schools by the student-to-computer ratio, but the results in terms of student achievement are ambiguous (Viadero, p. 1). Until recently there have been no links to computer use and gains on standardized tests, but then the standardized test as indicator of student achievement is itself suspect. An administrator at an urban school says, "My gut feeling tells me something significant is going on here" (Viadero, p. 1), when discussing the effect of a large investment in state-of-the-art computerized classrooms on the achievement of poor and minority students. But hunches and intuitions are not generally accepted by policymakers.

Education Week reports a study by Harold Wenglinsky of the ETS showing that 8th graders whose teachers used computers mostly for "simulations and applications" - generally associated with higher-order thinking - performed better on the 1996 NAEP mathematics test than students whose teachers did not. Meanwhile, 8th graders whose teachers used computers primarily for "drill and practice" - generally associated with lower-order thinking - actually performed worse than students who used no computers at all. (Wenglinsky's report cautions that the results should be taken as correlative rather than causal, as it may be that higher achieving students use computers differently than low-achievers.) Low-income and black students were the least likely to have teachers who use technology to its full advantage, demonstrating once again the unfortunate rule of thumb in American education that those that have, get. (Archer, p. 1):  The ETS study concludes:

All of this suggests that computers are neither cure-alls for the problems facing schools, nor mere fads that have no impact on student learning. Rather, when they are properly used, computers may serve as important tools for improving student proficiency in mathematics, as well as the overall learning environment in the school. ( Wenglinsky, Summary of Findings)

Computers and Social Interaction
Wenglinsky's results are consistent with a number of qualitative studies which have found the use of computers is not a matter of just placing hardware in schoolrooms or libraries. Introduction of computer-based curricula can have a profound impact on the social interactions of students and teachers. For those who, like myself, believe that learning is socially mediated, the important question regarding computers in schools is how they affect the quality and character of relationships and transactions between people.

The seminal Computers and Classroom Culture by Janet Schofield reports the findings of her two-year ethnographic study of computer use in an ethnically and economically diverse urban high school. Schofield studied two teachers' classes involved in a study of an artificially intelligent geometry program (Geometry Proofs Tutor or GPTutor). Over two years the study team observed 8 geometry classes where each student individually used the tutor and 7 traditional geometry classes for comparison; in addition 7 computer science classes taught by 5 different teachers and 4 business classes each taught by a separate teacher. The two male geometry teachers were highly traditional in their classes that did not have computers, using whole class instruction followed by individual seatwork under the supervision of the teacher. During seatwork time, if a student needed help, the teacher would speak in a loud voice, as if directing his remarks to the entire class. When using GPTutor, however, the same teachers developed a different teaching style. The teacher's attention directed his attention to individual students, and used a quiet tone of voice, making help a personal matter. When interviewed, the geometry teachers said the use of GPTutor made it possible to give assistance to the students who were struggling, because they knew the advanced students could forge ahead with the computer problems (Schofield 1995).

Prof. Fred Erickson remarked a couple years ago that we (educational researchers) know quite a lot of what teachers and students do in classrooms, and some things about what teachers experience as they work, but very little about how students experience school (Erickson, personal communication). Schofield's study is one of the few to try to document their point of view. She notes that students reported being highly motivated to use the computer, and attributes this to several factors, one being a decrease in the amount of humiliation they felt about giving wrong answers, a finding which was true of both the GPTutor classes and the computer science classes. For examples, when Computer Science students were attempting to run buggy programs, failure was the norm; it was expected that it would require several trials to arrive at a solution. The teacher provided a role model, himself sometimes struggling to debug programs as he tried to help students. (The Computer Science teachers were uniformly male.)

The documentation of the changed role of the teacher, his shift from pivotal source of information to troubleshooter and advisor, is probably the most important finding of this study. Peers became an important source of knowledge. When asked in an interview, "What's the best thing about computer science, "Donna replied, "You learn a lot more in a social environment with other students helping you out instead of just the teacher... That's the best part." (Schofield, p. 83)

The redistribution of expertise from one person to many is highly significant and more than just a matter of motivation. Most traditional teachers spend about 40% of their time establishing and maintaining control. Many of a teacher's actions and statements have nothing to do with content, even when the words are about the subject. For example, the following excerpt of dialog from a science lesson is ostensibly about energy, but the subtext is power and threat to the teacher's authority:

Teacher: No! Some light is not hot at all. When I turned on these fluorescent lights today, I haven't roasted yet.

Student: The bulb has heat

Eric:    Yeah, but when the bulb is on you get - the bulb gets hot.

Teacher: And essentially - most energy from the sun comes here in the form of light and not heat.

Eric:  So the ground can't be creating heat. Because if the ground wasn't dark, then it wouldn't absorb the light, and the light, is heat, so it's not creating it.

Teacher: No. Light is not heat. The light is light energy.

Eric: Yeah, and heat is heat energy.

[Students laugh.] (Lemke, p. 30)

At the end of this sequence, the teacher reasserts his authority by writing a rule about the transformation of energy on the blackboard. It is unlikely that many of these students have learned much about energy - their attention has been diverted to the power struggle between Eric and the teacher, which is undoubtedly much more interesting and makes more sense to them.

In the classrooms that used computers in Janet Schofield's study, such interactions virtually disappeared. Teacher time was spent coaching and tutoring at the request of students who were having difficulties. Students had the power to initiate interactions, and could avoid them if they wished. Paradoxically, the students spent more time on task.

Interviewer: How did using the computers change the way you behaved in class?

Diane: Well, we didn't talk as much. On the computer you really concentrated on the screen - didn't have time to talk to the person next to you (Schofield, p. 39).

In the GPTutor classes, students were often observed working before the bell rang, starting before the teacher had said a word.

Increased motivation and decreased discipline problems are merely the most obvious benefits of computer-based curricula. Schofield's observation of Computer Science classes points toward a striking characteristic, alluded to above, common to many other classroom situations where power is more equally distributed between teacher and students: students begin to use each other as resources for learning. This sane phenomenon is readily apparent in inquiry science classes, classes in which groups of students work independently, with the teacher as advisor, on original research projects. One of the requirements of true inquiry is that it be an "authentic" problem, one that is meaningful to the students, and whose answer may be unknown to the teacher.

Roth and Bowen's study of an 8th grade environmental science project (without computers) in a Canadian private school, is a benchmark in the literature on inquiry. The goal was to create a learning environment in which students were encouraged  "to take charge of their own learning, to frame problems for inquiry, to design data collection procedures, and to interpret and present their data in a convincing fashion." (Roth & Bowen, p. 120) When veteran teachers viewed videotapes and transcripts of the students at work, they were surprised by the quality of the student work, and impressed that science inquiry allowed a teacher to gain a deeper grasp of what students actually had learned from the experience.

Students in Roth and Bowen's study worked collaboratively in 3-person small groups to make sense of their data. The following is excerpt of conversation between students trying to understand a graph they have constructed from data they collected about the relationship of light intensity (measured in foot-candles) and growth of brambles; the task is to write a description of it:

Ron: In pattern A

Ellen: Let's, now, we can conclude

Ron: We can conclude

Ellen: That, ... which one is pattern A?

Ron: Pattern A.

Ellen: That one of the readings could be a fluke.

Ron: No, no.

Ellen: The reading

Ron: That's pattern A, right along here. So as the amount of candles

Ellen: Increases

Ron: The percent of brambles will stay the same

Ellen: With the pattern we concluded that if the amount of foot candles is higher

Ron: Exceeded, exceeded

Ellen: What do you mean exceeded, is exceeded by what?

Ron: It has more

Theo: Exceeded is, there is a greater amount of

Ellen: If the amount is ... there will be a higher density of brambles

Ron: No it's flat, it's wrong, look at the graph

Ellen: Right.

Ron: The density of brambles will stay the same, 'cause look that's what we concluded.

Ellen: Will get greater and then even out.

There will be a higher density

Ron: And eventually even out. (Roth & Bowen, pp. 105-6)

In contrast to the dismal exchange between Eric and his teacher recorded by Lemke, Ellen, Ron and Theo finish each other's sentences and hash out meaning the meaning of science concepts among themselves. Theo is even able to give Ellen a quick vocabulary lesson in context. Roth and Bowen consider the students to be jointly constructing a "conceptual environment" in which knowledge and intellectual resources (for instance graphing skills) are distributed and connected by the students as they construct their knowledge (Roth & Bowen, p. 121).

Reiser et al. report using a computer program, BguILE, for middle and high school science inquiry. The program provided students with authentic data concerning biology topics. Groups of 3 students were led by the program to construct an explanation for, in one case, an evolutionary event on the Galapagos Islands which actually occurred. The program asks students to be accountable for evidence and to be explicit about cause and effect in their explanations. As in other computer and inquiry-based approaches, the teacher becomes a guide. Notice in the following conversation that the teacher is not drilling or quizzing students to see if they can give the correct answer, but asking genuine questions.

Ms. Patrick: A., how do you know that the young ones died off? ... (turning attention to whole class) I want to hear how she came up with that because she didn't give me any evidence to support that.

Amalia: Because we went into graphs. We asked them to show us all the dead birds and the graph was mostly all young.

Ms. Patrick: Ok, so you did look at a graph of all dead ones. Did you look at a comparison of dead ones versus live ones?

Amalia: Yes.

Ms. Patrick: So you know when you looked at the comparison it was the low weight birds that were dead?

Amalia: Yes.

And were you sure they were all fledglings or could they have been low weight adults?

Amalia: No, we checked that.

Ms. Patrick: How did you check that?

Amalia: We looked at the profiles.

(Reiser et al., p. 26)

The students have finished looking at a great deal of evidence (The program has about 1,000 pages of data students can access.), and have come to a conclusion. Organizing data from different sources is difficult for kids, and these students have obviously been sifting through many data sets, and carrying out conversations about how to make sense of it.

Conclusion: This small set of studies has in common several characteristics which lead to the conclusion that computer-based instruction   (1) has real potential, if used in the right way, to increase students understanding of school subjects, including but not limited to mathematics and science. These dimensions are:

Looking at the products which result from computer use according to these dimensions may provide evidence of how effectively computers are being used by students and teachers.

Students' Science Websites and the Construction of Meaning
Rationale: One fairly common way in which teachers use computers is to assign their students, either singly or in groups, to create a Web page. Most people who have surfed the Web looking for content have run across these, and some can be quite elaborate and informative, even though librarians warn patrons not to use students' Web sites as authoritative references.

The creation of a web page is itself an authentic task which captures the interest of most students, but high interest in web design does not in and of itself insure a high level of competence in the subject matter. (2) Are the student web sites created on science topics likely to be the result of students actually learning science? Are they constructing meaning along with their web pages?

This is an major, important, really crucial question. We have a pretty good idea the computer is a potentially valuable tool for increasing student achievement, but it is a complex tool and requires careful and skillful use to be effective. Used badly, it can cause harm. The landscape of American education is littered with good reforms that yielded poor results, precisely because guidelines for implementation were not established or not followed or not understood by the users (Often for very good reasons, this is not an indictment.).

Methodology: 6 student web sites were chosen randomly through Web surfing. A Google search for "classroom web pages" yielded some 16,000 hits, yet only a few of these met the requirements of the study, which were: The web site had to have been produced by K-12 students as a class project, be about science, and the server had to be working or accessible. Several of the sites contained multiple examples of student work.

The protocol for consisted of scoring the student pages according to five of the six dimensions listed above. It was not possible to infer the teacher's role from the web sites, and scoring some sites on some dimensions was a stretch. A scale of 0 to 4 was used, with 0 being not present and 4 being highest. The ratings were based solely on my judgment, based on 12 years of science teaching experience at elementary, middle and high school levels. There is not enough data to make any meaningful statistical analysis, but I think the rating process helped push me to think about the sites in useful ways.

The Sites:
Following is a listing of the web sites that were found, and a brief description, followed by my rating. You are invited to click on the link and investigate the site yourself.


Dimension Score Rationale
Authenticity: 2 Kids like animals; not tied into their lives.
Collaboration: 1 Since each student studied a different species, can infer some distribution of expertise
Complexity: 1 Many different categories of information collected, but no attempt to assemble total picture.
Rigor: 3 The web pages display science content appropriate for 5th grade
Professional Community 0 Students did not work "like scientists" at all.


Dimension Score Rationale
Authenticity: 0 There appears to be no connection to the real world of these students
Collaboration: 4 Students seem to have collaborated in groups of 4
Complexity: 0 Students do not look at variables; they appear to be answering questions out of a textbook.
Rigor: 1 Content is skimpy beyond vocabulary words.
Professional Community 0 Students did not work like scientists.


Dimension Score Rationale
Authenticity: 4 Connected to students' experiences and concerns
Collaboration: 4 Students worked in a group to gather and analyze data
Complexity: 4 Variables included temperature, pH, CO2, and oxygen levels
Rigor: 3 Appropriate level for high school biology
Professional Community 4 Students used scientific tools and language fluently.


Dimension Score Rationale
Authenticity: 3 Photos show students' identification with project
Collaboration: 2? By inference students collaborated to build and launch rockets.      
Complexity: 3 By inference, rocket launching is a complex activity that requires a great deal of problem-solving and managing variables.
Rigor: 0 No scientific knowledge is evident on the web page, although it may have been part of the classroom instruction.
Professional Community 3 Rocket launching is an engineering problem and requires students to take on problem-solving roles


Dimension Score Rationale
Authenticity: 4 Students studied their home environment
Collaboration: 2-4? Probably some? Much? collaboration required
Complexity: 2? Students collected data on multiple variables for the Watershed Project, but it is unclear that they tried to analyze this data. It was done with other schools in conjunction with a university.
Rigor: 3 Appears appropriate for high school
Professional Community 4 Infer that students worked as a scientific community.
TOTAL 15-17+


Dimension Score Rationale
Authenticity: 2 Students appear to have chosen a topic that interested them. It's possible that the experimental protocols  were devised by each child.
Collaboration: 0 Individual projects
Complexity: 0?


If students were required to design their own experiments, this is a complex task, and might get a 4.
Rigor: 1 Very low level of content.
Professional Community 2? If students were required to design their own experiments, they were introduced to the ways scientists think about experiments. No collaborative work and very little use of scientific tools.

2 of the 6 sites scored in the upper range on my rudimentary best-use-of-technology scale. Both were high schools, and both involved students carrying out water quality testing. 2 of the sites did poorly, one an elementary school and the other a middle school. 2 of the sites, middle schools, were somewhere in the middle.

Based on this pilot study, it appears that some of the sites reflect K-12 students doing good science. It is not surprising, in view of the generally poor quality of science education in the U.S., that other sites present questionable science activities and projects. I am not sure that the poor quality sites were without value; it may be that the act of creating a web page motivates students to learn more than they would otherwise. Even the pages created by students answering questions out of the textbook probably involved collaboration and discussion among group members. They may be an improvement on traditional teacher-directed lessons, although I cringe at the thought that many hours of science class time my have been devoted to learning web authoring skills. Putting it all together, we see again that the mere fact that students did their assignments on the computer does not guarantee they learned much. On the other hand, when science projects were carefully designed, the results were impressive.

Technology itself cannot transform education. Since the time of John Dewey educators have demonstrated repeatedly that good education rests on certain basic principles, the same ones that form the dimensions of my rating scale (although the need to understand professional practice may be a bit unique to science). So far all attempts to incorporate these principles into the mainstream of education in the U .S. have been watered down or abandoned, many times because they were inconsistent with institutional and cultural practices. Janet Schofield tells how some change advocates hope technology will be a "Trojan horse" for the reform agenda, but she doubts that such an approach will succeed (Schofield, p. 219).

Computers do have the potential for making school reform easier and even necessary. Middle class children with home computers will only put up with the drudgery of print media for so long, and as teachers use the Web at home to download lesson plans, they begin to question the limitations of textbooks. Not too long ago, I listened to a PTA mother bitterly complain about a history teacher who had given her son and several other students the task of copying a major portion of the U.S. Constitution as punishment for missing a homework assignment. Her son had toiled for days writing it out word for word, but all the other students had downloaded the Constitution from a web site, printed it out and turned it in, which the teacher accepted. As computers become embedded in our culture, they will be impossible to exclude from schools.


Archer, J. (1998) The link to higher scores. Technology Counts ' 98: Putting School Technology to the Test. Education Week Special Report, October 1998.

Becker, H.S. (1972) A school is a lousy place to learn anything in. American Behavioral Scientist, 16(1): 85-105.

Bruner, J. (1996) The Culture of Education. Cambridge, Massachusetts: Harvard University Press.

Cuban, L. (1984) How Teachers Taught: Constancy and Change in American Classrooms 1890-1980. 2nd. New York: Longman.

Dewey, J. (1916) Democracy and Education: An Introduction to the Philosophy of Education. New York: The Free Press.

Education Week's 1999 National Survey of Teachers' Use of Digital Content. (1999) Technology Counts '99: Building the Digital Curriculum. Education Week Special Report, September 1999.

Erickson, Frederick (2000). Personal communication.

Fatemi, E. (1999) Building the digital curriculum: summary. Technology Counts '99: Building the Digital Curriculum. Education Week Special Report, September 1999.

Lemke, J. (1990) Talking Science. Norwood, New Jersey: Ablex Publishing Corporation.

National Commission on Mathematics and Science Teaching for the 21st Century (2000). Before its too Late: A Report to the Nation from the National Commission on Mathematics and Science Teaching for the 21st Century. Washington: United States Department of Education.

Reiser, B.J., Tabak, I., Sandoval, W.A., Smith, B.K., Steinmuller, F., and Leone, A.J. (in press) BguILE: strategic and conceptual scaffolds for scientific inquiry in biology classrooms. In: Carver, S.M. and Klahr, D. (Eds.), Cognition and Instruction: Twenty-Five Years of Progress. Erlbaum: Mahwah, N.J.

Roth, W-M & Bowen, G. M. (1995) Knowing and interacting: a study of culture, practices and resources in a grade 8 open-inquiry science classroom guided by a cognitive apprenticeship metaphor. Cognition and Instruction, 13(1), 73-128.

Schofield, J.W. (1995) Computers and Classroom Culture. Cambridge: Cambridge University Press.

Viadero, D. (1997) A tool for learning. Technology Counts '97: Schools and Reform in the Information Age. Education Week. Special Report, November 1997.

Wenglinsky, Harold (1998). Does it Compute? The Relationship Between Educational Technology and Student Achievement in Mathematics. Princeton, N.J.: Educational Testing Service.


1. Roth and Bowen do not report the use of computers. However the type of data analysis they used would have been less cumbersome if the students had access to them. I include Roth and Bowen because their paradigm for curriculum is consistent with the best of computer-aided instruction and is the current standard for the type of inquiry to which computers lend themselves so well. Others agree with the value of using computers for inquiry: "It is impossible for me to imagine how school leaders who are focused on more authentic ways of doing math and science, who are developing rich environments for learning, can achieve that without technology," - Linda Roberts, the adviser on technology to former Secretary of Education Richard W. Riley. (quoted in Technology Counts,  p. 1) As an experienced science teacher myself, I would say it's not impossible, but a lot harder.

2. The evidence from cognitive psychology that skills do not usually transfer across domains is overwhelming and by now a truism. This means that because a student uses higher-level skills to create a web site it does not naturally follow that those skills will be available to the student to understand the theory of evolution. The people who told you in high school that geometry was required because it teaches you to think were wrong.