The general aim of science education is to effect behavioral changes in learners such that they exhibit evidence of increasing scientific literacy. Various constructs that can be evaluated via observation of specific behaviors (which a student should exhibit after instruction) have been delineated and are targeted by science educators (Collette & Chiapetta, 1984; National Science Teachers Association, 1982). At this time, it appears there is a dearth of information regarding the determinants of children’s intentions to engage in science learning behavior. Science educators need to know what determines a youth’s intentions and behavior with regard to engaging in science learning activities. Knowledge of the determinants would inform science educators as to how they might better ensure that young people will participate in science activities. More relevant and controlled experiments could then be conducted.
Various authors (Haladyna, Olsen, & Shaughnessy, 1983; Koballa & Crawley, 1985; Linn, 1987; Zeidler, 1984) have recently called for sound, theoretical approaches to science education research, particularly in the area of attitudes. Such approaches are the ones most likely to make a significant and positive impact on learner behaviors. Along this line of thought, it is reasonable to assume that an understanding of the determinants of science learning behaviors is needed in order to develop an effective plan for changing the target behaviors.
Theory of Reasoned Action
There is substantial evidence to suggest that the construct most closely associated with the determination of behavior is behavioral intention (Ajzen & Fishbein, 1980; Fishbein & Ajzen, 1975). Fishbein and Ajzen have developed the “theory of reasoned action” which deals with behavioral intention and its determinants. The theory has recently generated considerable discussion among science educators (National Association for Research in Science Teaching, 1988) and has been used in science education research (Koballa, 1986, 1988; Ray, 1989a, 1989b, in press; Stead, 1985).
The theory of reasoned action has proven useful in understanding the precursors of intention. The model is founded on the assumption that most socially relevant actions are under volitional control, and a person’s intention to perform a behavior is the immediate determinant of the action (Ajzen & Fishbein, 1980). The paradigm’s main use is the prediction and understanding of behavior and its determinants. It is theorized ” . . . that people use the information available to them in a reasonable manner to arrive at their decisions” (Ajzen & Fishbein, 1980, p. 244) regarding particular behaviors. The information a person has serves as the basis of his or her beliefs.
The model can be represented by three equations (Fishbein & Ajzen, 1975):
(1) B » BI = w1(Ab) + w2(SN)
(2) Ab = Σ bi*ei
(3) SN = Σ nbj*mcj
The model displays that the intention (BI) to perform a certain behavior (B) is a function of the weighted (w1) attitude toward performing a behavior (Ab) and the weighted (w2) subjective norm (SN) (eq. 1). Ergo, a behavioral intention consists of a personal and social component.
Attitude toward the behavior is the individual’s feelings about performing the behavior. It is a function of the beliefs (bi) about expected consequences or outcomes of the behavior and the evaluations (ei) of these outcomes. Attitude toward behavior can be measured directly by asking the person to rate the performance of the behavior on a scale, or it can be measured indirectly by summing salient attitudinal belief scores (i.e., bi*ei scores) (eq. 2).
Subjective norm is the person’s perception of the pressures by important others, or referents, to perform the specific behavior. It is a function of normative beliefs (nbj), or who might want them to perform the behavior, and the motivations to comply (mcj) with the referents. As for the personal component, subjective norm can be measured directly, or indirectly by summing salient normative belief scores (i.e., nbj*mcj scores) (eq. 3).
The theory of reasoned action also considers the construct of external variables (e.g., demographic variables). The theory contends that external variables are related to behavior only if they are related to one or more of the variables within the theory (Ajzen & Fishbein, 1980). The authors of the theory have explained a number of ways that the relation between an external variable and behavior can be mediated (Ajzen & Fishbein). Figure 1 symbolizes the model.
Figure 1. The Theory of Reasoned Action. Intention is the immediate determinant of Behavior, while Attitude Toward the Behavior and Subjective Norm combine as the determinants of Intention.
The theory of reasoned action has been used to investigate a wide variety of behaviors such as fogging, voter decision making, and preservice elementary teachers’ intentions to use hands-on science activities with students (Bowman & Fishbein, 1978; Hom & Hulin, 1981; Koballa, 1986, 1988; Lin, 1987; Riddle, 1987; Stead, 1985). The theory might also be successfully used to generate valid and reliable baseline information and understanding concerning the determinants of children’s intentions to engage in science learning behaviors.
Selected External Variables
A growing phenomenon in educational choices is home education, with estimates of up to one million students being involved (Knowles, 1988). The position of some (e.g., National Education Association, 1988) is that children will not become properly educated if they are educated by their parents who are not state certificated teachers. That is, they need teachers formally prepared in pedagogical method and content. On the other hand, evidence indicates that the academic achievement of the home educated is generally equal to or better than that of their conventional school peers (Ray,
1988). More specifically, there is evidence that the home educated are not at a cognitive disadvantage in terms of their science education (Quine & Marek, 1988; Wartes, 1989). Since the goal of science educators is to assure the scientific literacy of all citizens, it would be helpful to gather information on this significant group of science learners at home.
Knowledge and understanding of young students is particularly important because many lose interest in science or their perceptions of science degenerate during grades 4 to 8 (Linn, 1987; Yager & Penick, 1986). In a similar vein, Talton and Simpson (1986) pointed out that “As more is understood concerning how adolescent students form attitudes toward science, policy changes can be made that will affect curriculum and instruction in science in American schools” (p. 366). Understanding the effects of grade level (particularly grades 3 to 8) on intentions to do science may lead to more logical ways of changing behavior. Grade level would be considered as an external variable in the theory of reasoned action and relevant data were collected. (Its effect will not be analyzed in this report, however.)
Science educators need information that would be useful in changing science learners’ ideas (Linn, 1987). The procured baseline knowledge and understanding about beliefs and intentions related to science learning behavior could then be used to develop and execute experimental research. That is, results of initial research might be used, for example, to inform a behavioral change method such as Hovland’s persuasive communication approach (Martin, 1985; Shrigley, 1983) or Lewin’s group dynamics approach (Shrigley). As Shrigley noted, ” . . . science educators interested in modifying attitudes can expect an accompanying change in behavior” (p. 431).
This study was designed to be exploratory, to generate baseline information, and to use correlational analyses. The study progressed in three stages. Stage I involved (1) selection of students, (2) elicitation interviews with students, and (3) construction of the instruments. Stage II was a pilot study including (1) development of procedures for administering the instruments and (2) review/revision of instrumentation after collecting data from a sample. Stage III involved (1) final data collection and (2) the analysis of data.
The population consisted of grades 3 to 8 public, private, and home school children in the mid-Willamette Valley region of Oregon. Selection by school district administrators and random selection via multistage cluster sampling (Borg & Gall, 1983) was used to determine the classes of public school students that were used in the study. One Christian school in the region readily agreed to participate. Home school support group networks were used to solicit a sample for participation in the study. A sample of 226 subjects (n = 187 public, 22 private, 17 home) was used for investigating the laboratory behavioral intention and 231 subjects (n = 190 public, 24 private, 17 home) were used for investigating the non-laboratory behavioral intention in Stage III of the study. (These two intentions are discussed later.) About 20% (n = 90; 72 public, 12 private, 6 home) of these students were randomly selected and interviewed in Stage I in order to elicit their salient attitudinal and normative beliefs. The responses of the public school students were quantitatively and subjectively analyzed according to the protocol of Ajzen and Fishbein (1980) in order to categorize them for later development of the instrument. The responses of the private and home school students were subjectively compared to those of the public school students in order to determine whether adjustments should be made in terms of which items should be included in the final instrument. A group of classes and students, completely separate from the primary sample, was randomly selected from the involved public schools for the pilot study.
The behavioral intention items for the laboratory and non-laboratory instruments were worded with special attention given to the elements of action, target, context, and time. The laboratory intention was “I plan to do the science projects and science experiment my teacher asks me to do.” The non-laboratory intention was “I plan to do the science reading and science writing my teacher asks me to do.” The instruments not only stated the preceding intentions, but also defined for the subjects the differences between laboratory and non-laboratory science activities. Students were randomly assigned to either the laboratory group or the non-laboratory group. The procedures of Ajzen and Fishbein (1980) were followed for elicitation sessions with students and for developing the instruments. Validity was assured via adherence to the theory of reasoned action and the instrument construction procedures proposed by the theory’s authors (Ajzen & Fishbein). The content validity of the item scales was also evidenced by their correspondence with the behavioral criteria of action, target, context, and time (Ajzen & Fishbein; Koballa, 1986). It was assumed that the subjects were willing and able to report accurately on the concepts represented by the items on the instruments; this enhances the instrument’s reliability. Familiar and standard methods of calculating instrument reliability (e.g., test-retest, split-half, and Cronbach’s alpha) were not appropriate for use in this study. The readability of the instruments was ascertained via the Fry method (Minnesota Educational Computing Consortium, 1982; Klare, 1974-1975) and adjusted to increase their readability for the subjects. Stage II was a pilot study, involving 102 students, during which procedures for administration of the instrument were developed. The final instruments were administered in Stage III to 377 students during a three-week period in the fall of the year. Alpha was set at .10 for statistical analyses in this exploratory study and Statgraphics, version 2.1 (Statistical Graphics Corporation, 1986) was used for executing the analyses.
Concerning laboratory science, eleven behavioral beliefs accounted for 79.7% of the public school subjects’ responses (Table 1) and six normative beliefs accounted for 79.1% of
their responses (Table 2). Concerning non-laboratory science, eight behavioral beliefs accounted for 80.5% of the public school subjects’ responses (Table 3) and seven normative beliefs accounted for 89.7% of their responses (Table 4). (The only apparent difference in elicited beliefs between
types of schooling was that the private and home school students mentioned God two to three times more frequently, with respect to normative beliefs, than did the public school students.) With
laboratory behavioral intention as the dependent variable and attitude toward behavior and subjective norm as independent variables, the adjusted coefficient of multiple determination (R2 = .14) was significant, F(2, 219) = 18.56, n = 222, p = .00 (Figure 2). With non-laboratory behavioral intention as the dependent variable and attitude toward behavior and subjective norm as independent variables,
BELIEF that Doing Science Projects and Science Number Experiments: Freq. % Cum. %
1 causes me to learn 31 19.6 19.6
2 causes me to have fun with science
and enjoy science 21 13.3 32.9
3 allows me to do science activities
and experiments 12 7.6 40.5
4 causes me to get good or better
grades 11 7.0 47.5
5 causes me to have to do work on
science assignments 11 7.0 54.4
6 causes me to be confused and not
understand science 8 5.1 59.5
7 causes me to think of how a science
experiment will turn out 7 4.4 63.9
8 causes me to be interested and not
bored 7 4.4 68.4
9 causes me to have an experiment work
well 6 3.8 72.2
10 causes me to dissect animals and
touch their parts 6 3.8 75.9
11 causes dangerous things to happen 6 3.8 79.7
various idiosyncratic beliefs 32 20.2 100.0
Table 1. Modal Salient Behavioral Beliefs for Laboratory
the adjusted coefficient of multiple determination (R2 = .27) was significant, F (2, 226) = 42.89, n = 229, p = .00 (Figure 3). The relative weight of the attitude toward behavior was greater (though not statistically analyzed) than the relative weight of the subjective norm for both laboratory and non-laboratory intentions. All of the correlations between the various other adjacent constructs in the theory were significant for both of the studied behaviors.
A three-way analysis of variance was executed to determine the effect of type of school on intentions. (The other two factors are not reported on in this paper, and there was no significant interaction between type of school and the other two factors.) In terms of laboratory science, there was a significant relationship between the external variable of type of school and the group of behavioral beliefs/evaluations of outcomes, F (2, 191) = 3.38, p = .04. A multiple comparison test revealed that the mean for home school students was greater than the means of both public and private school students (Table 5). In terms of non-laboratory science, there was no significant relationship between type of school and the other constructs in the theory of reasoned action, F (2, 196) = .76, p = .47.
Normative BELIEF Cum.
Number (or Referent) Freq. % %
1 parents 41 37.3 37.3
2 teacher 18 16.4 53.6
3 brothers or sisters 8 7.3 60.9
4 grandparents 7 6.4 67.3
5 friends 7 6.4 73.6
6 God 6 5.4 79.1
various idiosyncratic beliefs 23 20.9 100.0
Table 2. Modal Salient Normative Beliefs for Laboratory
Doing Science Reading and Science Cum.
Number Writing: Freq. % %
1 causes me to learn 22 19.5 19.5
2 causes me to have fun with science
and enjoy science 17 15.0 34.5
3 allows me to become someone who uses
science in their job 14 12.4 46.9
4 causes me to get good or better
grades 10 8.8 55.8
5 causes me to think about science
activities and experiments 9 8.0 63.7
6 causes me to do questions that are
difficult and hard to understand 7 6.2 69.9
7 causes me to do more writing 6 5.3 75.2
8 causes me to have to do work on
science assignments 6 5.3 80.5
various idiosyncratic beliefs 22 19.5 100.0
Table 3. Modal Salient Behavioral Beliefs for Non-Laboratory
Normative BELIEF Cum.
Number (or Referent) Freq. % %
1 parents 36 37.1 37.1
2 teacher 12 12.4 49.5
3 grandparents 11 11.3 60.8
4 relatives 11 11.3 72.2
5 brothers or sisters 7 7.2 79.4
6 friends 5 5.2 84.5
7 God 5 5.2 89.7
various idiosyncratic beliefs 10 10.3 100.0
Table 4. Modal Salient Normative Beliefs for Non-Laboratory
Figure 2. Laboratory behavioral intention using the theory of reasoned action. All correlations were significant, p < .05.
Figure 3. Non-laboratory behavioral intention using the theory of reasoned action. All correlations were significant, p < .05.
Type of School Σ b*e mean Multiple Comparison
Home 26.3 *
Public 19.1 *
Private 14.4 *
Table 5. Means and post hoc multiple comparison for beliefs about and evaluations of outcomes for laboratory science activities. Mean for home school students was significantly higher than the other two, p < .05.
Young science learners’ beliefs, which theoretically determine their intentions and actual science learning behavior, were identified. The theory of reasoned action explains that students’ beliefs determine their attitude toward the specified behaviors and their subjective norm with respect to the specified behaviors. Six of the beliefs (e.g., causes me to learn and causes me to have fun) about outcomes of doing laboratory science were identical or very similar to beliefs about outcomes of doing non-laboratory science. The six normative beliefs (e.g., parents and teacher) for laboratory were included in the seven for non-laboratory.
This study has provided theory-generated baseline information regarding the beliefs of laboratory and non-laboratory science learners. Home educators, elementary school teachers who include science in their classrooms, and science teachers could be informed of the cores of salient attitudinal beliefs and normative beliefs of these young science learners. Researchers might use this information to design experimental studies to test the effects of influencing salient beliefs and, ultimately, actual science learning behavior.
Research might also be directed toward using the theory of reasoned action or other theoretical models to understand why the external variable of type of school had a significant impact on students’ beliefs. Investigators might explore the environment of the home school in order to determine why home educated students appear to have such a positive attitude toward learning science and appear to achieve quite well in science (e.g., Quine & Marek, 1988; Wartes, 1989). For example, Keeves (1975) identified the home, the school, and the peer group as the three major classes of non-personal variables which influence student achievement. Schibeci (1989) considered Keeves’ variables and presented a model suggesting that “home background” influences three sets of student variables (i.e., entry behaviors, early outcomes, and final outcomes). These three sets of student variables are related to initial science-related attitude, general attitude, and final science-related attitude, respectively. It is notable that within the home education milieu Keeves’ three variables (home, school, and peer group) are, to a great degree, collapsed into one. Keeping this fact in mind, Coleman and Hoffer’s (1987) construct of social capital, and its relationship to physical capital and human capital, might be germane.
Coleman and Hoffer (1987) explained that physical capital takes the form of tools, machines, and other productive equipment (e.g., computers and microscopes in the schools). Human capital has to do with developing skills and capacities in people to make them more productive (e.g., skills and knowledge acquired by an individual in school). Social capital is even less tangible than human capital and it “… exists in the relations between persons” (Coleman & Hoffer, p. 221). Coleman and Hoffer presented trust as a form of social capital. “A group within which there is extensive trustworthiness and extensive trust is able to accomplish much more than a comparable group without that trustworthiness and trust” (Coleman & Hoffer, p. 221). They give evidence that even if families possess high levels of human capital, the children may be at an academic disadvantage if there is little social capital in the family. This low level of social capital might be caused by the physical absence of family members (i.e., a “structural deficiency”) or the absence of strong relations between children and parents (i.e., a “functional deficiency”).
Coleman and Hoffer (1987, p. 225) developed a two-dimensional (social capital and human capital) matrix that is helpful in describing families. A third factor, physical capital, can be added to their matrix to form a three-dimensional space useful for describing children’s learning environments (Figure 4). A child learning science could be described as having a niche within this environment of physical, human, and social capital; that is, an educational resource niche.
4Figure 4. The Educational Resource Niche. This three-dimensional model describes a learner’s niche in the environment of physical, human, and social capital. Absence of capital = 1; presence of capital = 2, 4, or 6.
A learner’s niche in this environment of capital might be used to predict his or her science academic achievement. If it is postulated that there is a hierarchy of importance from social capital down to human capital down to physical capital with respect to effect on academic achievement, then Figure 4 and the findings of this study might be used to explain the apparent high science achievement, and academic achievement in general, of the home educated. Figure 4 shows that numerical values might be assigned to the presence or absence of the three hierarchically arranged types of capital. For example, the presence of social capital would be denoted by a 6 and the absence thereof by a 1, while the presence of human capital would be worth a 4 and the absence thereof a 1. Now the effects of the different types of capital might be assumed to be additive. For example, a home school family in which there is social capital (value = 6) but a lack of human capital (value = 1) and physical capital (value = 1) would still be expected to perform better (total capital = 8) than a conventional school student who experiences physical capital (value = 2) and human capital (value = 4) but a lack of social capital (value = 1) (total capital = 7). Social capital, or strong relations between parents and children, might explain the positive relationship between home schooling and students’ positive attitudes toward doing laboratory science. This three-dimensional model might help explain why home school students could achieve well even if their parents are not professionally trained teachers, have relatively low levels of income, and lack resources at home such as computers and sophisticated laboratory equipment.
The preceding model is speculative and needs further exploration. It could be used in further research to generate and test hypotheses regarding the home educated (and others) and their performance in science and other academic and affective areas of development.
In general and as expected, attitude toward behavior and subjective norm predicted a significant amount of variance in intention for both laboratory and non-laboratory behaviors. However, the amount of variance explained in this study was not as high as in other investigations using the theory of reasoned action (e.g., Bowman & Fishbein, 1978; Hom & Hulin, 1981; Koballa, 1986, 1988; Lin, 1987; Stead, 1985). It is pertinent to note that no research other than this has employed the theory of reasoned action to study the intentions of children as young as those in this study (i.e., grades 3 to 8). Further research might investigate whether limitations to the theory are peculiar to its application to children.
Consistent with the majority of other research that has employed the theory of reasoned action, attitude toward behavior had a greater relative weight than subjective norm in terms of predicting behavioral intention for both laboratory and non-laboratory. This suggests that (for the types of subjects and behaviors herein described) future experimental research should focus on attitude toward behavior in attempts to change the beliefs, attitudes, intention, and behavior of science learners.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall, Inc.
Borg, W. R., & Gall, M. D. (1983). Educational research: An introduction (4th ed.). New York, NY: Longman.
Bowman, C. H., & Fishbein, M. (1978). Understanding public reaction to energy proposals: An application of the Fishbein model. Journal of Applied Social Psychology, 8, 319-340.
Coleman, J. S., & Hoffer, T. (1987). Public and private high schools: The impact of communities. New York, NY: Basic Books, Inc., Publishers.
Collette, A. T., & Chiapetta, E. L. (1984). Science instruction in the middle and secondary schools. St. Louis, MO: Times Mirror/Mosby College Publishing.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley Publishing Company.
Haladyna, T., Olsen, R., & Shaughnessy, J. (1983). Correlates of class attitude toward science. Journal of Research in Science Teaching, 20, 311-324.
Hom, P. W., & Hulin, C. L. (1981). A competitive test of the prediction of reenlistment by several models. Journal of Applied Psychology, 66, 23-39.
Keeves, J. P. (1975). The home, the school, and achievement in mathematics and science. Science Education, 59(4), 439-460.
Klare, G. R. (1974-1975). Assessing readability. Reading Research Quarterly, 10, 62-102.
Knowles, J. G. (1988). Introduction: The context of home schooling in the United States. Education and Urban Society, 21(1), 5-15.
Koballa, T. R., Jr. (1986). Teaching hands-on science activities: Variables that moderate attitude-behavior consistency. Journal of Research in Science Teaching, 23, 493-502.
Koballa, T. R., Jr. (1988). The determinants of female junior high school students’ intentions to enroll in elective physical science courses in high school: Testing the applicability of the theory of reasoned action. Journal of Research in Science Teaching, 25, 479-492.
Koballa, T. R., & Crawley, F. E. (1985). The influence of attitude on science teaching and learning. School Science and Mathematics, 85, 222-232.
Lin, C. D. (1987). A study of the factors that influence industrial education instructors to use computers. Unpublished doctoral dissertation, Oregon State University, Corvallis.
Linn, M. C. (1987). Establishing a research base for science education: Challenges, trends, and recommendations. Journal of Research in Science Teaching, 24, 191-216.
Martin, R. E. (1985). Is the credibility principle a model for changing science attitudes? Science Education, 69, 229-239.
Minnesota Educational Computing Consortium. (1982). School utilities, volume 2, version 1.1. St. Paul, MN: Author.
Staff. (1988, June). Attitude networking group report. National Association for Research in Science Teaching News, 30(2), 16.
National Education Association. (1988, December). The 1988-89 resolutions of the National Education Association. NEA Today, 17-24.
National Science Teachers Association. (1982). Science-technology-society: Science education for the 1980s. Washington, D.C.: Author.
Quine, D. N., & Marek, E. A. (1988). Reasoning abilities of home-educated children. Home School Researcher, 4(3), 1-6.
Ray, B. D. (1988). Home schools: A synthesis of research on characteristics and learner outcomes. Education and Urban Society, 21(1), 16-31.
Ray, B. D. (1989a, March/April). The determinants of grades 3 to 8 students’ intentions to engage in laboratory and non-laboratory science learning behavior. Paper presented at the Annual Conference of the National Association for Research in Science Teaching, San Francisco, CA.
Ray, B. D. (1989b, April). Determinants of student intention to engage in laboratory vs. non-laboratory science learning behavior. Paper presented at the National Convention of the National Science Teachers Association, Seattle, WA.
Ray, B. D. (in press). The determinants of grades 3 to 8 students’ intentions to engage in laboratory and non-laboratory science learning behavior. Journal of Research in Science Teaching.
Riddle, P. K. (1980). Attitudes, beliefs, behavioral intentions, and behaviors of women and men toward regular jogging. Research Quarterly for Exercise and Sport, 51, 663-674.
Schibeci, R. A. (1989). Home, school, and peer group influences on student attitudes and achievement in science. Science Education, 73(1), 13-24.
Shrigley, R. L. (1983). The attitude concept and science teaching. Science Education, 67, 425-442.
Statistical Graphics Corporation. 1986. Statgraphics, version 2.1 [Computer program]. Rockville, MD: Author.
Stead, K. (1985). An exploration, using Ajzen and Fishbein’s theory of reasoned action, of students intentions to study or not to study science. In R. P. Tisher (Ed.), Research in Science Education, Volume 15. Selections of Papers from the Annual Conference of the Australian Science Education Research Association (16th, Rockhampton, Queensland, Australia, May 1985). (ERIC Document Reproduction Service No. ED 267 974)
Talton, E. L., & Simpson, R. D. (1986). Relationships of attitudes toward self, family, and school with attitude toward science among adolescents. Science Education, 70, 365-374.
Wartes, J. (1989). Washington Homeschool Research Project report from the 1988 Washington homeschool testing. Author: Washington Homeschool Research Project, 16109 N.E. 169th Pl., Woodinville WA 98072.
Yager, R. E., & Penick, J. E. (1986). Perceptions of four age groups toward science classes, teachers, and the value of science. Science Education, 70, 355-363.
Zeidler, D. L. (1984). Thirty studies involving the “Scientific Attitude Inventory”: What confidence can we have in this instrument? Journal of Research in Science Teaching, 21, 341-342.