Student Epistemological Beliefs and Conceptual Change Activities: How Do Pair Members Affect Each Other?

Student Epistemological Beliefs and Conceptual Change Activities: How Do Pair MembersAffect Each Other?Author(s): Mark WindschitlSource: Journal of Science Education and Technology, Vol. 6, No. 1 (Mar., 1997), pp. 37-47Published by: SpringerStable URL: http://www.jstor.org/stable/40186411 .

Accessed: 12/06/2014 14:48

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Springer is collaborating with JSTOR to digitize, preserve and extend access to Journal of Science Educationand Technology.

http://www.jstor.org

This content downloaded from 62.122.76.45 on Thu, 12 Jun 2014 14:48:55 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=springer

http://www.jstor.org/stable/40186411?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Journal of Science Education and Technology, Vol. 6, No. 1, 1997

Student Epistemological Beliefs and Conceptual Change Activities: How Do Pair Members Affect Each Other?

Mark Windschitl1

This study examined the relationships in achievement between members of dyads who were paired according to epistemological maturity and engaged in a photosynthesis simulation exercise. This study also examined the relationship between individual students' epistemological maturity and their conceptual understanding of photosynthetic processes. The findings indicated that individuals' conceptual understanding posttest scores could be predicted, in part, by their partners' posttest scores. This relationship between posttest scores was negative. Observations of selected dyads revealed that the epistemologically mature pair members consistently articulated to their partners the relationships they perceived in the simulation environment and verbally hypothesized to their partners more often than the less epistemologically mature members. The less epistemologically mature members of the observed pairs were relatively passive recipients of their more verbal partners' reasonings. They did not appear less methodical, in fact, they tended to focus on the step-wise completion of tasks in the simulation's written guide albeit with few expressions about why or how phenomena took place. Concerning the second research question, posttest scores of individuals could be predicted by the epistemological maturity of the individual. This positive relationship would account for the higher performance on posttest scores by the more epistemologically sophisticated subjects, and, if the epistemologically mature members did play a more active intellectual role in the dyad relationships, it would explain the negatively associated posttest scores of paired subjects. KEY WORDS: Dyads; epistemological maturity; photosynthetic processes.

INTRODUCTION

When assigning students to work in pairs during laboratory exercises, instructors generally intend that members of these dyads facilitate each other's learning through meaningful dialogue and shared insights. In these cases, learner characteristics, such as ability or degree of prior knowledge, can affect not only the individual's learning but also the achievement of other group members (Carter and Jones, 1994; Webb, 1992). One non-cognitive learner charac- teristic, epistemological maturity, indicates learners' beliefs about knowledge and its acquisition. As a motivational factor, epistemological maturity may affect

how learners interact with complex learning tasks as well as with dyad partners engaged in the same tasks. This, in turn, may influence achievement for both individuals. Epistemological maturity is not a measure of ability with a domain nor is it the capacity to think abstractly, although it has been linked to achievement in complex learning environments (Windschitl and Andre, 1996). Because a growing number of science students are paired for work on computer-based tasks, and because there is a general concern about students influencing one another's achievement in group learning situations, this study was designed to examine the relationships in achievement between dyad members who engaged in a photosynthesis simulation exercise. These dyad members were paired according to epistemological maturity. This University of Washington, Seattle, WA 98915.

37

1059-O145/97A)3OO-0037$12^<V0 O 1997 Plenum Publishing Corporation



38 Windschitl

study also examined the relationship between individual students' epistemological maturity and their achievement in conceptual understanding of photosynthetic processes.

The Role of Student Epistemological Beliefs

Studies have suggested that motivational beliefs such as goals, values, self-efficacy, and control beliefs, play a role in conceptual development (Dweck, 1986; Entwistle, Entwistle, and Tkit, 1993; Perkins and Sim- mons, 1988; Pintrich, Marx, and Boyle, 1993; West and Pines, 1983). Scientists, as members of a learning community, internalize the goals of systematizing knowledge and refining theories, but this is not the orientation of most students in the classroom. Theo- ries that base conceptual development on parallels between scientists and students neglect, among other things, differences in motivation. For students, goals such as making friends, out-performing others, or seeking rapid closure to academic tasks can be primary motivations in the classroom environment. These goals are not directly associated with learning and may result in superficial processing of ideas and low levels of intellectual engagement. Instruction that emphasizes accumulation of factual information ac- commodates this weak learning approach by empha- sizing the acquisition of incremental, decontextualized knowledge (Entwistle, Entwistle, and Tkit, 1993).

A related influence on learning, epistemic motivation, has been defined as one's beliefs about knowledge itself and the process of building knowledge (Boyle, Magnusson, and Young, 1993). Kruglanski (1989, 1990) suggested that a learner's motivation toward knowledge as an object (epistemic motivation) influences knowledge acquisition. In research on postsecondary students' beliefs about the nature of knowledge and learning, Schommer (1993a, 1993b) described four dimensions of belief. Two dimensions deal with knowledge itself: belief in the simple nature of knowledge versus the belief that knowledge is complex, and, belief that knowledge has a strong right/wrong duality versus belief that knowledge is con- text dependent. TWo other dimensions describe individuals' beliefs about learning: belief in the quick acquisition of knowledge versus the belief that knowledge has to be refined over time, and, belief in an in- nate ability to learn versus belief in an ability to "learn how to learn." Researchers have described ways in which these types of beliefs can affect learning. Dweck

(1986) suggests that learners who believe intelligence can be cultivated orient themselves toward developing that quality and seek challenging intellectual situations as well as deeper understandings. Learners who believe intelligence is a fixed trait tiy to gain favorable judgments of that trait and tend to avoid academically challenging situations. Windschitl and Andre (1996) found that learners with high epistemological maturity (as measured using Schommer's dimensions of belief) performed better in an exploratory simulation exercise than in a didactically controlled exercise. Learners with low epistemological sophistication performed better in a more controlled, guided simulation exercise than in an exploratory exercise.

The aspect of epistemological maturity used for this study is "belief in the complexity of acquiring knowledge." It is based one of Schommer's epistemological dimensions: "knowledge as simple versus knowledge as complex." The items used to measure this construct are derived from a longer epistemological instrument (Schommer, 1993a).

Dyad and Small Group Influences on Learning

In primary and secondary settings, generally positive learning results have been noted for dyads and larger groups (Johnson, Johnson, and Smith, 1991; Webb, 1987; Webb, Ender, and Lewis, 1986). In a study of peer collaboration and learning about photosynthesis, Lumpe and Staver (1995) found that students working in groups developed more scientifi- cally correct conceptions than those students working alone. The study revealed that both consonant and dissonant peer interactions helped students develop clear, coherent conceptions.

Research with post-secondary students on group composition and achievement has been less positive. One study suggested that college students are relatively homogeneous in terms of ability and/or are less likely to be attuned to the academic shortcomings of their group members, precluding some of the positive influences that more capable group members could have on less capable members (Hooper, Ward, Han- nafin, and Clark, 1989). While some studies of under- graduates have examined the effects of pairing learners of different academic abilities, others have looked at the effects of personality on individuals' per- formances in pairs. These studies found that individuals high in sociability work best in pairs and those who have high levels of test anxiety work best individually



Epistemological Maturity and Development 39

(Reid, Palmer, Whitlock, and Jones, 1973; Sutter and Reid, 1969). The picture here is not entirely dear because variables related to sociability, such as propen- sity toward conversation in group work, have been associated with negative (off-task) behavior rather than positive behavior during computer-based instruction (Carrier and Sales, 1987). Slavin (1992) summa- rized the current state of research this way: "Clearly, much work needs to be done to fully develop a theoiy to account for cooperative learning effects on achievement and to understand the conditions under which each of the several motivational and cognitive perspectives has explanatory value."

RESEARCH QUESTIONS

Epistemological beliefs have been shown to influence conceptual development and the potential for conceptual change; there is also evidence that learners, when placed in dyads or larger cooperative groups can have their attitudes and achievement influenced by others. In this study, students enrolled in a college biology class were assigned to one of five types of instructional conditions for a photosynthesis simulation exercise. Based on the results of an epistemological survey, some subjects were placed with a partner who had similar epistemological beliefs, while other subjects were placed with a partner who had significantly dissimilar epistemological beliefs, and still other subjects worked as individuals. All subjects participated in a sequence of plant physiology lessons culminating in the dyad exercise with a computer-based photosynthesis simulation. The following two questions were investigated in this study:

• Question 1: Can students' posttest scores be predicted (in part) by their partners' epistemological maturity scores and by their partners' posttest scores?

• Question 2: Are epistemological maturity scores significant predictors of individuals' posttest scores?

METHODOLOGY

Sample

The sample was composed of approximately 255 biological sciences majors at a larger midwestern uni-

versity. The subjects were enrolled in a second-level biology laboratory class in the fall of 1995.

Design

Five types of subject conditions were arranged to ensure a suitable number of different pair types for analysis of research question one. These pairings were not intended to be used as discrete categories of analyses. Placement in these pairs was based on subjects' scores on a test of epistemological maturity (Schommer, 1993a; Windschitl and Andre, 1996) which was administered during the first week of the experiment:

• Some subjects worked as individuals (single) on the photosynthesis simulation exercise.

• Both subjects placed in each high-high pair scored between one-half and one full standard deviation above the mean for epistemological maturity.

• Both subjects placed in each low-low pair scored between one-half and one full standard deviation below the mean for epistemological maturity.

• One subject (referred to as low-mixed in later analyses) placed in each mixed pair scored below the mean for epistemological maturity and was paired with an individual who scored above the mean. The more epistemologically sophisticated partners in these pairings were part of the fifth subject condition.

• One subject (referred to as high-mixed in later analyses) placed in each mixed pair scored above the mean for epistemological maturity and was paired with an individual who scored below the mean.

The minimum difference between the epistemological scores of the mixed pair members was one standard deviation.

All subjects' epistemological scores were placed in a single pool to identify their respective places within the normal distribution. Then, within class sections, subjects in particular epistemological ranges were randomly assigned to partners (as described above) who fell into the appropriate region of the epistemological distribution for that pairing type. Within class sections it was possible to have different subject pair-types as well as subjects working individually.



40 Windschitl

MATERIALS

The biology laboratory course was offered in conjunction with a lecture that was held twice a week. The lecture/lab combination covered plant systems and plant physiology during the first half of the semester, and animal systems and animal physiology during the second half of the semester.

All procedures were tested with a small group of students in a pilot study in order to estimate the time necessary to complete the simulation exercise and to determine the clarity of the written materials.

Computer Simulation. A computer simulation that modeled the photosynthetic and respiratory processes in plants was used for this study. The simulation presented the user with two "windows," one of which showed a dynamic line graph of O2 and CO2 production over time as influenced by conditions sur- rounding a hypothetical plant. This graphical information resulted from conditions depicted in the second window which displayed the hypothetical plant and several variables in the plant's local environment that could be adjusted by the user (Figure 1). Examples of these adjustable factors were light intensity, light wavelength, temperature, humidity, water intake, and CO2 concentration.

Written Simulation Guides. Each subject was given her/his own written guide to use as a reference and to record predictions, data, results, and conclusions. The first part of the guide was designed to ac- quaint the students with the software by directing their attention to certain areas on the screen and prompting them to manipulate environmental variables in order to make some elementary observations. The main body of the exercise was composed of three parts entitled: Respiration versus Photosynthesis, What Affects the Rate of Photosynthesis?, and limiting Factors to Photosynthesis. The guide was adapted from parts of a written ancillary provided with the software, and modified for use in this experiment (Mintz, Rosenfeld, and Rosenstein, 1993).

Epistemological Survey. The students' epistemological maturity was one independent variable and was determined by an adaptation of an instrument developed by Schommer (1993a). A previous use of the original 63-item instrument (Windschitl and An- dre, 1996) indicated that a set of 21 questions could be derived from the original instrument; the theoretical construct measured with this shortened ver- sion was "belief in the complexity of acquiring knowledge." For the purpose of this study, the more epistemologically mature individual believes that

Fig. 1. Sample of simulation screen containing animated pictorial displays of physiological phenomena (used with permission by Logal™ software).




knowledge is continuously refined by an effortful in- tegration of information from various sources. The unsophisticated individual believes that knowledge is the accretion of unambiguous facts. Typical items (rated on a five point agree-disagree Likert-type scale) are:

• You should try to integrate new ideas in a textbook with knowledge you already have about a topic, (positive)

• Being a good student generally involves memorizing facts, (negative)

• Wisdom is not knowing the answers, but knowing how to find the answers, (positive)

• Most words have one clear meaning, (negative)

• If a person can't understand something within a short amount of time, they should keep on trying, (positive)

• If professors would stick more to the facts and do less theorizing, one would get more out of college, (negative)

• It's a waste of time to work on problems which have no possibility of coming out with a clear-cut and unambiguous answer, (negative)

• The best thing about science courses is that most problems have only one right answer, (negative)

Pretest. The pretest was a nine-item multiple choice instrument composed of questions used in a previous study on alternative conceptions (Amir and Tkmir, 1990). These questions were reviewed by the author, the coordinator of biology laboratory courses, and a botany specialist. Items dealt with topics such as limiting factors to photosynthesis, the cycling of O2 and CO2 in nature, and respiration in plants.

Posttest. Four items on the posttest solicited subjects' opinions about the simulation's worth and their reactions to working with a partner (for those who were pair members). An example of one of these multiple choice questions was: "How much do you feel you learned about photosynthesis from the use of the computer simulation?" The remainder of the posttest instrument included 14 conceptual questions about photosynthesis and respiration, some of which were similar to pretest items. Question formats included textual multiple choice, multiple choice re- quiring the interpretation of graphical information, and free-response.

Procedure

Students participated in their regular class sections; there were 13 sections of approximately 18 students each. During Week 1, subjects completed the epistemological survey and the pretest (correct answers to the pretest were not disclosed to the subjects until after the posttest was administered). The subjects were then given a demonstration of the photosynthesis simulation. This provided subjects with a uniform exposure to the simulation's various displays and control mechanisms. The goal was to have students spend their time the following week "using the simulation" as a learning tool, rather than spend time "learning how to use" the simulation.

In Week 2, subjects met during their regular class time in a computer facility adjacent to their biology laboratory classroom. Students participated in a 15-minute discussion that covered the concepts of limiting factors to photosynthesis and the relationships between respiration and photosynthesis. Then, students were each given a written simulation guide, seated at computer stations with their partners or individually, and allowed to proceed.

Throughout the exercise, subjects were prompted to make predictions about phenomena by writing statements or completing graphs in the written guide (Figure 2). Accompanying the graph shown in the guide are the instructions:

... use the empty graph below to draw a rough prediction line of what you would expect to see hap- pen to the photosynthetic rate as you increase light intensity. Label the line predicted- -Repeat the simulation for a light intensity of 300 units and wait a few moments for the values to sta- bilize. Read the digital counter and record the reading as a dot on the graph, then continue to increase the light intensity at steps of 300 units. Record your data on the graph and draw the actual curve . . .

Not all sections of the guide were this directive, much of the guide allowed more exploratory options for the students. Students were prompted to create and test hypotheses in the simulated environment, and to make generalizations about the results. The simulation exercise took about three hours. After each class had finished the exercise, there was a sum- maiy discussion of factors affecting photosynthesis, the characteristics of plant respiration, and the nature of limiting factors in plant environments.

Informal observations of four dyads were conducted in order to add perspective to the quantitative



42 Windschitl

Fig. 2. Sample graph from students' written guide.

analyses; these were unobtrusive, 30-minute sessions conducted by the author. The biology laboratory coordinator was asked to furnish the names of dyad members for observation without revealing to the author whether the members were rated as high or low in epistemological maturity. One observed pair was mixed, two pair were low-low, and one pair was high-high.

During Week 3, students took the posttest. Tb add motivation, students were informed that their posttest score would count as a class quiz grade.

RESULTS

Internal Consistencies

The epistemological survey had an internal consistency estimate of .70 (using Cronbach's alpha). The pretest internal consistency had an alpha value of .45. The internal consistency estimate of the posttest was .47. The low internal consistencies of the pre- and posttests may be attributable to the small number of items in each instrument (6 in the pretest and 9 in the posttest). Because the item topics were specifically identified as candidates for misconception, there may have been significant random guessing on both tests. These internal consistencies, while low, were determined to be sufficient for experimental purposes. The low internal consistencies increase the error variance and make Type II errors more likely, but they decrease the chance of a Type I error. Thus, if significant effects are found, they are less likely to be due to chance.

Pretest

The mean score for all subjects on the pretest was 2.26 out of 6 possible (SD = .37) or 38% correct. The

average subscore for the limiting factors questions was 18% and the average subscore for the respiration/photosynthesis questions was 50%. The total pretest score was used as a covariate for later analyses.

Posttest Results and Subject Condition Comparisons

The mean score for all subjects on the posttest was 6.39 out of 9 possible (SD = 1.59) or 71% correct. Several of the posttest items were similar to pretest items and were used for direct comparisons of pre- and post-conceptions. The pretest question "When does respiration occur in plants?" was answered correctly by only 59% of respondents, but in the posttest, when asked to select a response to complete the phrase "Respiration occurs-", 86% of the subjects correctly chose "all the time in both plants and animals." When asked to select a diagram correctly depicting the gasses exchanged between plants and animals, only 23% answered correctly on the pretest while 71% of respondents chose the correct answer in the posttest. When asked to examine the graph shown in Figure 3 and select the correct response to "What happens at light intensity 'X1?", 69% answered correctly in the pretest and 80% answered correctly in the posttest (that "Oxygen is being taken up and given off at the same rates"). Subjects showed substantial improvement between the pretest (M = 38% correct) and posttest (M = 71% correct).

The pre- and posttests contained subsections dealing with the concepts of limiting factor and res-

An experiment was carried out on a plant in a closed chamber. The purpose of the experiment was to investigate the effect of light intensity on photosynthesis by measuring the changes in concentration of O2 in the chamber. The results are shown in the graph below. Examine the graph below and answer the questions below.

Fig. 3. Diagram presented with pretest question 28, "What happens inside the plant at light intensity 'X'?"




piration/photosynthesis relationships. A comparison of these pre- and post-subsections between students of high epistemological maturity (falling above the epistemological mean) and subjects of low epistemological maturity (falling below the mean) is shown in Thble I.

The order of pair-type results with respect to total posttest means was: high-high subjects (M = 6.77, SD = 1.44), high-mixed subjects (M = 6.68, SD =

1.57), single subjects who fell above the epistemological mean (M = 6.58, SD = 1.28), single subjects who fell below the epistemological mean (M = 6.21, SD = 1.86), low-low subjects (M = 6.15, SD = 1.68), and low-mixed subjects (M = 5.95, SD = 1.65).

The primary research question was: Can students' posttest scores be predicted (in part) by their partners' posttest score? A multiple regression analysis was used to answer this question. Students were linked with their partners as single units of analysis; to avoid violating assumptions of independence of data points, only one individual of each pair was selected (randomly) to have his/her posttest score re- gressed on. Variables were entered in the following order: pretest score of individual, pretest score of partner, epistemological score of individual, epistemological score of partner, product of epistemological scores (to account for interactions), and the posttest sore of the partner. The final stage of the regression (Tfcble II) indicates that partners' posttest score was a significant negative predictor (P = -.24, p < .05), and that the remaining variables were not significant predictors. The remaining variables were: pretest score of individual (P = .11, /1 = .30), pretest score of partner (p = -.05,/? = .58), epistemological score of individual (p = -.30, p = .87), epistemological score of partner (P = -.45, p -.83), and product

of epistemological scores (P = .67, p = .81). Higher posttest sores of individuals, then, were indicative of lower posttest scores for their partners.

These statistical results were supplemented by the author's observations of four dyads. The observations provided further insight into the types of interactions that took place between dyad members. All four dyads were observed at the same point in the exercise. During the observation time, students were manipulating variables such as light intensity, local concentration of CO2, and humidity, in order to determine the effects on the rate of photosynthesis.

The first of four dyads observed was a mixed pair. Hicia was rated high in epistemological maturity and Cole was rated low. Throughout the exercise, THcia played an active role not only in judging trends in the graph readings of photosynthetic rates, but also in summarizing the results of experimental trials. She shared approximately four times as many comments with Cole as he did with her. Cole played the role of results recorder and made periodic, direct observations of what was transpiring on the screen, however, he made far fewer speculative or summary statements than THcia.

The second pair, Dustin and David, were both rated low in epistemological maturity. Interaction between these two individuals was restricted to reading steps out loud from the simulation guide, and making direct observations about the photosynthetic rate in response to changing environmental variables. They offered no "what if" or "I wonder why" statements during the 30 minute observation, and, except for the cursory reading of the steps from the simulation guide, they did not interact with each other. They did complete the exercise methodically and accu-

Table L Means and Standard Deviations of Pre- and Posttest Scores Grouped by Level of Subjects' Epistemological Sophistication0

Low epistemological subjects High epistemological subjects All subjects

Pre Post Pre Post Pre Post

M SD M SD M SD M SD M SD M SD

Limiting factor subtest .20 (.26) .56 (.29) .17 (.28) .66 (.25) .18 (.27) .61 (.27)

Resp/Photo subtest .49 (.30) .75 (.21) .51 (.32) .82 (.19) .50 (.31) .79 (.20)

Total .35 (.21) .67 (.19) .34 (.23) .75 (.16) .34 (.22) .71 (.18)

"Means are expressed as proportion correct, i.e., .20 = 20% correct.



44 Windschitl

Table II. Regression on Posttest Score Using Pretest Score, Pretest Score of Partner, Epistemological Score, Epistemological Score of Partner, Interaction of Epistemological Scores, and Posttest Score

of Partner as Predictors11

PREDICTOR B Beta t p

Pretest score .17 .11 1.10 .30 Pretest score of partner -.07 -.06 -.55 .58

Epls. belief -1.49 -.30 -.17 .87

Epls. belief of partner -1.92 -.45 -.21 .83

Epls. belief X Epls. belief .56 .67 24 .81 Posttest score of partner -.27 -.24 -2.23 .03* Constant 12.65 .37 .71

"Final solution.

bp < .05.

rately, but it is likely that they could have done

equally well working as individuals. The third pair, Brad and Chris, were both rated

as high in epistemological maturity. They conducted an on-going conversation which was saturated with

questions to each other, verbalized hypothesizes, and

interpretive statements about phenomena portrayed in the simulated environment. The following segment was typical of their dialogue, and took place while the Brad and Chris were adjusting CO2 levels in the simulation and observing a developing line graph showing changes in CO2 level, O2 level, and photosynthetic rate.

Brad: I can't decide on what the normal rate of photosynthesis should be. Chris: I think it should probably be around 100 or 106 [expressed in per cent of initial rate] Brad: If there is no carbon dioxide there won't be any photosynthesis. If I increase the CO2 level, well, do you think it [photosynthetic rate] will be linear, exponential? Basically there are three possibilities. Chris: Yeah. Brad: [increases CO2 level] Why was there an initial

jump? Chris: When it [O2 level] was at zero? Brad: Maybe there was some oxygen left in the leaf, isn't respiration continuous so it's always making carbon dioxide? Chris: Yeah, it always gives it off, I don't understand that jump either. Maybe that's normal and it will die out later.

Although Brad appears to be more active in the above interaction, Chris assumed the more active role in a subsequent part of the exercise. It is difficult to follow this dialogue precisely as it relates to the simulation activity, however, the trend of these dyad members was clearly to interact with each other rather than to simply read instructions from the written guide or make elementary observations.

The fourth observed pair had the only student rated low in epistemological maturity (Joanne) who

consistently attempted to engage her partner in dia-

logue. Joanne did not verbalize hypotheses, infer- ences, or summary statements, however she did ask her partner several questions about photosynthesis facts and about how to construct her own graph. She was paired with another low-rated student, Ttent, who's repertoire of comments during the observation

generally consisted of "yeah", "nope", and "I sup- pose." Their interactions were generally one-sided and limited to direction-following.

Although generalizations are hard to support from the small number of observations, there were some trends. The students with higher epistemological maturity engaged in more dialogue (or spoke more often to their partners), asked more hypothetical questions, and appeared to play a directive role in the simulation exercise. The students with lower epistemological maturity played more passive roles, even in the case of Dustin and David where neither individual "took the lead" in initiating productive dialogue. The primary concern of the low epistemology students appeared to be following the directions and recording results; they engaged in little conversation. These tendencies, related to epistemological maturity, are connected to the second research question.

The second research question was: Are epistemological maturity scores significant predictors of in- dividuals1 posttest scores? In order to account for variance associated with prior knowledge, pretest score was entered first, followed by epistemological score (Thble III). In the final regression stage, epistemological belief was a significant predictor of posttest score (P = .18 p > .01). Interestingly, pretest score was not a significant predictor of posttest score (P = .06, p = .32).

Subjects' attitudes toward the use of simulation was positive (Tkble IV). When asked "How much do you feel you learned about photosynthesis from the use of the computer simulation?", 62% responded

Table IH Regression on Posttest Score Using Pretest Score and Epistemological Score as Predictors"

PREDICTOR B Beta t p

Pretest score .08 .06 1.10 .32

Epistemological belief .80 .18 2.92 .01* Constant 3.19 3.00 .01

°Final solution. bp < .01.




Table IV. Student Responses to Opinion Questions Concerning Simulation Exercise

Number of Percent of Questions respondents respondents

How much do you feel you learned about photosynthesis from the use of the

computer simulation? I learned a lot 57 22.4 I learned a moderate amount 157 61.6 I learned a little 32 12.5 I learned nothing at all 0 0.0

Missing responses 9 3.5

How did you feel about using the simulation as a class activity? I enjoyed using it as a class activity 160 62.7 It was O.K. as a class activity 76 29.8 I didn't care one way or the other 7 2.7 I disliked it as a class activity 3 1.2


In the future, when using a simulation like Photosynthesis, would you prefer to work with a partner or by yourself?1

I don't have a strong preference 102 51.8 I prefer working by myself 17 8.6 I prefer working with a partner 76 38.6


Did working with a partner help your understanding of photosynthesis?1 Helped understanding a lot 16 8.1

Helped understanding moderate amount 74 37.6

Helped understanding a little 77 39.1 Did not help understanding at all 21 10.7 Hindered understanding 6 3.0


"Responses only from those subjects who worked with a partner.

that they learned "a moderate amount" and 22% responded that they learned "a lot." Approximately 63% of respondents claimed that they enjoyed using the simulation as a class activity, 30% indicated that it was "O.K." as a class activity, and only 1% disliked it as a class activity. When asked about working with a partner or working alone in the future, 52% had no strong preference, 39% preferred working with a partner, and 9% preferred working by themselves. When asked if their partners helped their understanding of photosynthesis, 46% indicated that their partners helped their understanding "a moderate amount" or "a lot", and 39% indicated that their partners helped their understanding "a little."

DISCUSSION

In this study, individuals' posttest scores could be predicted, in part, by their partners' posttest

scores. The relationship between individuals' posttest scores and their partners' posttest scores was negative. Given that pretest scores for partners were accounted for in the analysis, some significant interaction is likely to have occurred between paired subjects during the course of the experiment.

Observation of the subjects during the experiment revealed that the epistemologically mature pair members were more verbally active, tending to ex- press what they were thinking, and were consistently articulating to their partners what perceived relationships were evident in the simulation. Also, the more epistemologically sophisticated members tended to hypothesize and to suggest "what if" scenarios more often. The less sophisticated members of the observed pairs were often (although not always) passive recipients of their more verbal partners' reasonings, this more passive member often responded with a simple "yeah" or "okay" to initiatives and conclusions proposed by the more sophisticated member.



46 Windschitl

The less sophisticated members did not lack a methodical approach, in fact, they tended to focus on the step-wise completion of tasks in the written guide with few expression about why or how phenomena took place. It would be reasonable to assume that if

pairs of subjects had one markedly more active member, that this active member would be more likely to

play an assertive role intellectually as well as conver-

sationally, and that this active member would cogni- tively process information about the simulation exercise more deeply than the passive member. Be- cause the second regression analysis indicated a positive relationship between epistemological belief and posttest score, it is likely that the more epistemologically sophisticated members in many pairs assumed this active role. This explanation would account for the higher performance on posttest scores by the more epistemologically sophisticated subjects as suggested in the second experimental question and also for the negatively associated posttest scores of paired subjects.

The analysis conducted for the first experimental question did not indicate a significant relationship between subjects' epistemological maturity and subjects' posttest score; this may have been due to the method of analysis. The units of analysis included one students from each pair, and all students who worked as individuals were removed from the analysis. This reduced the number of subjects from 255 to 95, reduced the power of the test, and may have con- tributed to the lack of significance in the regression results.

Analysis of the second research question indicated that epistemological maturity was a significant predictor of posttest score. The epistemological survey measured the learners' belief in the complexity of acquiring knowledge. Even though the instrument attempted to measure beliefs, many of the item statements could be strong indicators of behaviors, specifically behaviors such as implementing learning strategies. In the photosynthesis simulation exercise, subjects were asked to speculate about phenomena, make predictions about relationships between variables, test these predictions, and draw conclusions from their hypothesis testing. This instructional model emphasized student exploration and did not provide answers or simple, overly-structured approaches to "finding" answers. This exercise appar- ently favored the more epistemologically mature subjects and was not as productive a learning environment for less epistemologically mature students.

Less epistemologically sophisticated students tend to believe that knowledge is simple and certain. An instructional approach that provides rigidly structured paths to specific conclusions, or simple dispensation of answers may be more consistent with these students' approaches to learning. Epistemologically less mature students are likely to find an approach that

emphasizes higher order thinking skills and self-con- struction of knowledge less compatible with their beliefs of knowledge attainment. Thus, the conditions of this simulation exercise may have provided an ac- commodating intellectual environment for the more

epistemologically sophisticated student, and may have induced more positive motivation for learning in these students.

There is some question as to whether epistemological maturity leads to greater conceptual development through the stimulation of deeper, more reflective processing of information, or through a direct influence on the quality of choices the learner makes within learning exercises. When one considers the constellation of personal and contextual variables associated with leaning situations, and adds to that the complexity of dyadic interactions, it becomes clear that there are many ways to account for differences in learning.

It is worth restating that the construct of epistemological maturity is not a measure of ability or achievement. More mature beliefs about the nature of knowledge and its acquisition (that knowledge is continuously restructured, created from multiple perspectives) may correlate with other variables related to learning, but the nature of these relationships is still under investigation.

CONCLUSIONS

There appears to be support for the notion that epistemological maturity influences learning, and that this maturity may be manifested in dyadic interactions. It is less certain whether epistemological maturity causes deeper, more reflective intellectual processing or causes selection of more sophisticated decisions in the transaction with the external instructional environment. The somewhat polarized scores of dyad partners suggest that a "zero-sum" situation occurs between learners. Some behaviors in high- achieving learners may be related to lower levels of learning for their partners. Observations suggest that some partners are intellectually more active in re-




sponse to a simulated environment, and that these students tend to be more epistemologically sophisticated. Any claim, however, that behaviors related to high epistemological maturity are directly causal in lowering the scores of learning partners is un- founded. Future research might include studies similar to this one in which other potentially influential variables are accounted for; measures of sociability and academic motivation are two such variables. Fur- thermore, techniques such as path analysis can offer a more complete picture of the relationships among variables associated with dyad work. Detailed de- scriptions of the processes of learning behaviors, in- cluding more qualitative reports of dyad work, are also needed.

The general framework of instruction for these paired students was effective in cultivating an understanding of photosynthesis and respiration. This sup- ports the idea that computer-based simulations offer suitable cognitive environments with which to test learners' conceptual development, general response to computer-based instruction, and interactions with others.

REFERENCES

Amir, R. and Tamir, P. (1990, April). Detailed analysis of miscon- ceptions as a basis for developing remedial instruction: The case of photosynthesis. Paper presented at the American Edu- cational Research Association Annual Meeting, Boston.

Boyle, R. A., Magnusson, S. J., and Young, A. J. (1993, April). Epistemic motivation and conceptual change. Paper presented at meeting of National Association for Research in Science Teaching, Atlanta, Georgia.

Carrier, C. A. and Sales, G. C. (1987). Pair versus individual work on the acquisition of concepts in a computer-based instructional lesson. Journal of Computer- Based Instruction, 14(1), 11-17.

Carter, G. and Jones, M. G. (1994). Relationship between ability- paired interactions and the development of fifth graders' concepts of balance. Journal of Research in Science Teaching, 31(8), 847-856.

Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist, 41(10), 1040-1048.

Entwistle, N., Entwistle, A., and Tait, H. (1993). Academic understanding and contexts to enhance it: A perspective from research on student learning. In X Duffy, J. Lowyck, and D. Jonassen (Eds.), Designing Environments for Constructive Learning, pp. 331-357. Springer-Verlag: Berlin.

Hooper, S., Ward, T, Hannafin, M., and dark, H. (1989). The effect of aptitude composition on achievement during small group learning. Journal of Computer-Based Instruction, 16 (3), 102-109.

Johnson, D. W, Johnson, R. T, and Smith, K. A. (1991). Active Learning: Cooperation in the College Classroom. Interaction Book Company: Edina, MN.

Kruglanski, A. W. (1989). Lay Epistemics and Human Knowledge: Cognitive and Motivational Bases. Plenum: New York.

Kmglanski, A. W. (1990). Lay epistemic theory in social-cognitive psychology. Psychological Inquiry, 1, 181-197.

Lumpe, A. T and Staver, J. R. (1995). Peer collaboration and con- cept development: Learning about photosynthesis. Journal of Research in Science Teaching, 32 (1), 71-98.

Mintz, R., Rosenfeld, S., and Rosenstein, M. (1993). Explorer™ Photosynthesis. Logal Software. East Arlington, Massachu- setts.

Perkins, D. N., and Simmons, R. (1988). Patterns of misunder- standing: An integrative model for science, math, and programming. Review of Educational Research, 58(3), 303-326.

Pintrich, P. R., Marx, R. W., and Boyle, R. W (1993). Beyond cold conceptual change: The role of motivational beliefs and classroom contextual factors in the process of conceptual change. Review of Educational Research, 63(2), 167-199.

Reid, J. B., Palmer, R. I., Whitlock, J., and Jones, J. (1973). Com- puter-assisted instruction performance of student pairs as related to individual differences. Journal of Educational Psychology, 69, 65-73.

Schommer, M. (1993a). Comparisons of beliefs about the nature of knowledge and learning among postsecondary students. Re- search in Higher Education, 34(3), 355-369.

Schommer, M. (1993b). Epistemological development and academic performance among secondary students. Journal of Educational Psychology, 85(3), 406-411.

Slavin, R. E. (1992). When and why does cooperative learning increase achievement? Theoretical and empirical perspectives. In R. Hertz-Lazarowitz and N. Miller (Eds.), Interaction in Cooperative Groups: The Theoretical Anatomy of Group Learn- ing, pp. 145-173. University Press: Cambridge.

Sutter, E. G., and Reid, J. B. (1969). Learner variables and inter- personal conditions in computer-assisted instruction. Journal of Educational Psychology, 60, 153-157.

Webb, N. (1992). Testing a theoretical model of student interaction and learning in small groups. In R. Hertz-Lazarowitz and N. Miller (Eds.), Interaction in cooperative groups: The theoretical anatomy of group learning 102-119. University Press, Cam- bridge.

Webb, N. (1987). Peer interaction and learning with computers in small groups. Computers in Human Behavior, 3, 193-209.

Webb, N. M., Ender, P., and Lewis, S. (1986). Problem-solving strategies and group processes in small groups learning computer programming. American Educational Research Journal, 23, 243-261.

West, L., and Pines, L. (1983). How rational is rationality? Science Education, 67(1), 37-39.

Windschitl, M. and Andre, T (1996). Using computer simulations to enhance conceptual change: The roles of constructivist instruction and student epistemological beliefs. Paper presented at annual meeting of American Educational Research Asso- ciation, New York, New York.



Documents

Student Epistemological Beliefs and Conceptual Change Activities: How Do Pair Members Affect Each Other?