Learning preferences/instruction in General Chemistry.

Abstract

Correlations between students' auditory, visual and tactile-kinesthetic learning preferences, traditional and non-traditional course components and levels of success in General Chemistry have been assessed. Tactile-kinesthetic learners showed greatest perseverance and tendency for involvement in the non-traditional course components studied, computer-aided instruction and peer-led-team-learning. Even though there were correlations between this involvement and success in the course, these students appeared to remain at a disadvantage by the traditional measures of written examinations.

Introduction

Students do not all learn in the same ways. Gardner [1], Guilford [2], Sternberg [3], and others [4] have posited theories suggesting that intelligence, once characterized strictly as "book smarts", is actually an aggregate measure of the strength of a variety of components. These theories fall under the broader heading of Multiple Intelligences. Howard Gardner [1], a leader in multiple intelligence theory, has identified at least eight different components to intelligence (see Fig. 1). See issue's website <http://rapidintellect.com/AEQweb/sum2003.htm>

The levels of these components vary between individuals. In classroom and laboratory situations, the learner will demonstrate these as preferences for, or tendencies to gravitate toward, particular kinds of instruction or assessment. Gardner refers to each of these preferences separately as intelligences [4]. These leamer preferences are, to some extent, based upon the learner's successes in prior learning experiences where instructional strategies employed complimented their preferences. As an example, application of Gardner's theory to the learner profiled in Figure 1 would likely lead to an assessment that the learners preference is predominantly "tactile/kinesthetic." According to the characteristics of this intelligence, this student would probably learn best when information is presented to him/her via manipulatives and other concrete objects or physical representations. That is, this learner is more likely to grasp and internalize the salient characteristics of concepts that are presented to him/her as a physical model (the more hands-on, the better). By further examining the learner's profile, we would expect that instructional strategies that are predominantly visual and/or auditory (like lecture/discussion) would be less likely to result in the student learning efficiently and correctly internalizing the characteristics of content.

Similar predictive scenarios may be drawn when the learning preferences of students are known and responded to via the intentional use of a variety of instructional strategies aimed at supporting predominant learning preferences. When learners experience success via instructional strategies that support their dominant intelligence, the learners' awareness of their own learning preferences is strengthened. The result of this experience may be that learners will recognize, gravitate to, and benefit from similar instructional experiences that are supportive of their learning preference. It is important to note that nearly all learners have some measure of all the intelligences listed, and that most are capable of learning something from any clear instructional strategy or supplemental learning device. The extent of a learner's success, however, is largely dependent upon the degree to which the dominant instructional strategy used supports his/her dominant intelligence.

Traditionally, science instruction at the high school and introductory college levels has consisted of a two-pronged approach. First, the initial presentation of conceptual information happens in a lecture/discussion session. In this setting, the student predominantly receives auditory and visual instruction. Second, the initial presentation is accompanied by a related laboratory experience designed to practically represent salient aspects of the conceptual material. The laboratory exercises are largely tactile/kinesthetic and traditionally engage the learner in a "cookbook" activity. This approach will only optimize the learning experience of the learner whose intelligence profile includes relatively high levels of auditory, visual, and tactile/kinesthetic components. In research prior to this [5], we have found that tiffs is not the typical profile of chemistry students at our institution. Rather, they tend to be higher in one of the three and lower in the other two. One way to address the students' needs for variety in introductory level college chemistry is to understand the relationships between student learning preferences and available instructional alternatives.

The variety of active learning components in general chemistry courses has grown to be exceptionally rich during the last decade. Active learning strategies that involve cooperative[6] and collaborative[7] learning, integrated lecture and laboratory [8], computer assisted learning [2], and websites[9] with interactive applets, animations and videos all contribute to a wider learning experience than traditional textbook reading, lecture, written homework and recitation sections. Success in retention and student performance has been documented for these various strategies. While each appears to be advantageous for a larger percentage of the students in chemistry classes than the traditional structure in that they engage a wider variety of students, we have found that mandatory implementation of one or more of these programs in a course creates problems for a small fraction of students.

Our students experienced a bit of that difficulty as we attempted to include a variety of course components in our General Chemistry courses at Coastal Carolina University a few years ago. With mandatory participation in Peer-Led-Team-Learning (PLTL) also known as Workshop Chemistry [6], Computer Aided Instruction (CAI) with Jim Spain's ChemSkill Builder 2000 assignments [10], in-class process workshops [11], written textbook problems, etc. we found that many students were overburdened with redundant study by different methods even though the success rate (%ABC) had shown remarkable improvement from about 30% for lecture to about 75% as these components were added over the course of three years. Some of the students, however, complained that they felt there was too much redundancy in the requirements of the course and that their time might be better spent by concentrating on just a few of the options for supplemental learning. In a simple response to these students we began our use of "Cafeteria-Style Grading." This allowed students to choose from a variety of supplemental course components to be included for their overall course grades [12]. Pursuant to making some of the supplemental components optional for student's grades, improvement was seen in student satisfaction accompanied by continuing increases in overall student achievement.

The continued increases in student achievement accompanied by increased levels of learner satisfaction compelled us to try to understand the students' success and why they chose to engage in particular components. We began by assessing individual student learning preferences in an attempt to see correlations with individual components. A simple study of students' preferences of different course components (lecture, laboratory, PLTL and CAI) compared to their learning preferences (visual, auditory, and tactile/kinesthetic) did reveal some correlations [5]. These and additional correlations between preference, participation and course grades are presented here.

Correlations of Jungian personality type and student success in general chemistry have recently been studied showing higher and lower success rates for only particular personality types[13]. It was noted that relationships between the instructors' personalities and those of their students were also significant. In some of the active-learning components that we discuss here, and in many of the recently developed strategies, the personality type of the lecture instructor is either not directly relevant (PLTL, CAI, Laboratory) or strategically modified by the program itself. In the long term we plan to investigate the relationships between various psychological factors in learning and their relationships to a broad range of pedagogical strategies, materials and structures that we will call course components. This preliminary study focuses on learning preferences of auditory, visual and tactile/kinesthetic and the course components of CAI, PLTL, lecture and laboratory. A broader study of more specific components of non-traditional lecture and laboratory components with a broader range of learning styles is under way. In this way we hope to better advise students (and instructors) about the choices of components available to them and their likelihood of engagement and success with these choices in their cafeteria selections.

Methods

The questions about selection of course component and course grade along with the 30-item instrument for assigning learning preference was provided by Dr. Alison Morrison-Shetlar[14] and is reproduced in Figure 2. See issue's website <http://rapidintellect.com/AEQweb/sum2003.htm> On the same survey, students were asked to report what course components in which they were participating, whether they "liked" or "disliked" them, and the grade they anticipated receiving in the course. The three instructors of the General Chemistry II course at CCU gave this survey to their students in the spring of 2001. Students in all sections were distributed between common laboratory sections and workshops and used the same CAI assignments. The data for the three sections were combined and assessed as follows. For each student the total point score in each of the three preference categories was obtained by assigning 1 point to each response of the ten possible in the category. This category score was divided by the total number of learning style preference responses to permit determination of a normalized score in each category for each student and to reflect a distribution of learning preferences within each student. For example, if a student gave five responses in the auditory preference, three in the visual preference and two in the tactile/kinesthetic preference, the percentage scores for that individual would be 50% auditory, 30% visual and 20% tactile/kinesthetic. The preference for course components was tallied by assigning a +1 value to a check indicating the student "liked" the component, a 0 value for no response and a -1 value if the student "disliked" the component. The learning preferences and lecture formats used by the three instructors were not a part of this study.

Course component selection was converted to a four point numerical scale that reflects level of involvement in the optional components. Response c in survey question 1 was assigned the point value of 4, with responses b = 3, a = 2 and d = 0 to reflect heavier participation in the active learning components with higher values. Course grades were assigned point values of A = 4, B = 3, C = 2, D = 1 and F = 0. Correlations between course component and learning style values for individual students were determined by the Kendall's [] test using PSI-Plot software. In this non-parametric measure of correlation, a value of +1 is perfect correlation, and a value of -1 is perfect anticorrelation. Therefore measured absolute values of [] greater than the threshold-tabled values are considered statistically significant. Tau values for all combinations of component involvement, grade, learning preference and course preferences and their associated probability values are presented in Table 1A. Correlations with values of 0.10 and greater were selected for presentation in Table 1B. At the 90% confidence level with n = 78, a value of 0.0989 is significant [15]. At the 99% confidence level, a value of 0.1795 is the minimum level for significance.

In evaluation of the learning styles preferences for the 78 students in the General Chemistry courses using Shetlar-Morrison's instrument, average values of auditory = 26.4%, visual = 40.1% and tactile/kinesthetic = 33.5% were obtained as illustrated in Figure 3. Table 1A shows the negative correlation between these learning styles. See issue's website <http://rapidintellect.com/AEQweb/sum2003.htm> It is notable that the least negatively correlated pair, the auditory-tactile/kinesthetic with the value of -0.2 is not so negatively correlated as the other two comparisons. Tau values of both the auditory-visual comparison (-0.37) and tactile/kinesthetic-visual (-0.47) indicate a greater degree of negative correlation. This suggests that it is more likely to find a combination of auditory and tactile/kinesthetic preferences within this group of students and less likely to find a combination of visual and tactile/kinesthetic preferences.

The correlation in participation in optional course components and course grade is 0.29 with a probability value of 0.00026. This is consistent with our observations during the implementation of these components since 1997. The -0.129 correlation (albeit with relatively high probability of 0.103) for the auditory preference correlation with involvement in course components suggests that either this preference is more attuned to success by the traditional lecture method, or that students with this preference are simply not attracted to PLTL and CAI. By comparison, the "r value of 0.157 for the students with a kinesthetic preference and their involvement of the active course components--PLTL and CAI -suggests that these students are either attracted to interactive learning or less successful in the traditional lecture format. If this positive correlation is linked with the anti-correlation of -0.15 that is seen in the kinesthetic learners' course grades, this suggests that even though course grades increase with overall participation, the kinesthetic learners continue to be at a disadvantage in the course. It is also important to note that there are negative correlations between the attraction to CAI and PLTL when compared to both lecture and laboratory. The negative [] values range from -0.21 for the laboratory/PLTL comparison, to the value of -0.34 for the CAI/lab comparison. The grade disadvantage may be in part due to the fact that the course grades continue to be heavily based upon traditional lecture and laboratory evaluation methods--written examinations and quantitative laboratory reports.

Correlations between learning preference and component preference were not so clear. The [tau] value of 0.11 with a probability value of 0.16 (0.11/0.16) shows a weak positive correlation between the auditory preference and whether the student "liked" lecture. A similar correlation is seen between the visual preference and responses about liking workshop (0.11/0.17). The correlation between the auditory preference and liking all the components (0.145/0.067) is apparently contradictory with the negative correlation in component involvement and auditory preference. This may be related to the satisfaction that auditory learners have with the required lecture and that it suits their learning needs, so that their additional participation is not necessary. The negative correlation of the visual preference and liking all the components (-0.213/0.007), appears to reflect a bias in these components for auditory and tactile/kinesthetic involvement. This is weakly supported by the correlation between the tactile/kinesthetic preference with liking all the components (0.1/0.2). The remaining value on the table shows a weak correlation between liking workshop and liking of all the components (0.13/0.11) Discussion. The correlations in Table 1A can be divided into three sets--those that are clearly significant with [] values well beyond the [+ or -] 0.1795 value given for the 99.0% p for n = 78, those with marginal significance, beyond the [+ or -] 0.0989 value given for the 90.0% p for n = 78, and lastly those with ||[tau| < < 0.0989 that indicate no statistical significance for correlation. The first two sets are selected for examination in Table 1B. See issue's website <http://rapidintellect.com/AEQweb/sum2003.htm>

The most reliable values are the anti-correlations between learning preferences. Auditory and visual preferences don't tend to coexist in our students; neither do auditory and tactile/kinesthetic or visual and tactile/kinesthetic. It is clear that our students do not have strongly diversified intelligence profiles, lending credence to the practice of offering high quality, diverse instructional options to insure their success. This is supported by the positive correlations between course grade and involvement in alternative course components. Students who chose to be involved in the alternate components anticipated higher course grades with a +0.29 correlation value.

Finally, those who like the traditional lecture and laboratory components do not like the non-traditional alternatives of CAI and PLTL. It appears that those who are more successful academically in our introductory chemistry classes tend to be those who are strong in the traditional settings of school (specifically those mentioned earlier, lecture/discussion followed by laboratory experiences). Although these students appear not to favor involvement in PLTL and CAI, this may be because they are adept at the "school game" which consists of lecture and lab. To them, it is possible that venturing into the realm of unfamiliar instructional alternatives represents academic danger rather than opportunity. Results from this study suggest that diversity in student learning preferences can be accommodated in general chemistry courses by providing a choice of traditional and non-traditional course components and cafeteria-style grading. Further inquiry into the identification of characteristics contributing to student learning (for example, more of Gardner's areas of multiple intelligence, Jungian personality type profiles, and students' social demographics) may provide greater insight into the best practices in the implementation of active learning components into general chemistry courses at the college level.

Notes

(1.) Gardner, H. Educational Leadership, 1999, 57, 12-17.

(2.) Guilford, J., The nature of human intelligence. New York: McGraw-Hill, 1967.

(3.) Sternberg, R., Linn, R. (Ed.), Intelligence: Measurement, theory and public policy (pp. 119-146) Chicago: University of Illinois Press. 1989.

(4.) Eggen, P.; Kauchak, D., Educational psychology: Windows on classrooms, 5th ed. Columbus, Ohio: Merrill Prentice-Hall, 2001

(5.) Goodwin, J. A.; Smith, D. W.; Gilbert, B. D. ACS National Meeting, Chicago, Illinois, August 30, 2001.

(6. a) Woodward, A.; Gosser, D.; Weiner, M Journal of Chemical Education. 1993, 70, 651. b) Gosser, D.; Roth, V.; Gafney, L.; Kampmeier, J.; Strozak, V.; Varma-Melson, P.; Radel, S.; Weiner, M. The Chemical Educator March, 1996. http://journals.springer-ny.com.chedr c) Gosser, D.; Roth, V. Journal of Chemical Education, 1998, 75, 185.

(7.) Wolfskill, Troy; Hanson, David Journal of Chemical Education. 2001, 78, 1417.

(8.) Apple, Thomas; Cutler, Alan Journal of Chemical Education. 1999 76 462.

(9.) Tsai, Chin-Chung Journal of Chemical Education. 2001, 78, 970.

(10. a) Spain, J. D.; Peters, H. J ChemSkill Builder 5.1, Electronic Homework Systems, Inc., 1997. b) Spain, James D. Journal of Chemical Education. 1996, 73, 222.

(11.) Hanson, David M.; Wolfskill, Troy Journal of Chemical Education. 2000, 77, 120.

(12.) Goodwin, J. A.; Gilbert, B. D. Journal of Chemical Education. 2001, 78, 490-494

(13.) Clark, G J. and Riley W. D. Journal of Chemical Education. 2001, 78(10) 1406-1411.

(14.) Morrison-Shetlar, A.; Marwitz, M. Teaching Creatively: Ideas in Action Outernet Publishing LLC. 2001.

(15.) Conover, W. J. Practical Nonparametric Statistics 3rd ed. Wiley, New York., 1999.

John Goodwin, Coastal Carolina University

Doug Smith, Coastal Carolina University

Goodwin is an Associate Professor of Chemistry with a PhD in Inorganic Chemistry. In addition to the focus on learning preferences in General Chemistry course components, his research interests in chemical education include development of electronic learning objects and systematic methods for their effective adaptation by instructors. Smith is an Assistant Professor of Education with a PhD in Teacher Education. His research and writing interests in science education reflect the relationships between student learning preferences and best practice in science instruction.