The role of learning styles in student achievement in an MBA economics course.

Englander, Fred ; Terregrossa, Ralph A. ; Wang, Zhaobo 等

INTRODUCTION

This paper examines the effect of learning style preferences in explaining variations in student achievement in an MBA class in economics. To the extent that particular learning style preferences are found to have a systematic influence on student achievement, then an instructor is in a position to improve the learning performance of his/her students by adapting the mix of teaching methods and strategies that he/she employs in order to achieve a closer match between those teaching methods and the learning style preferences of that instructor's students. An OLS regression model is developed in this paper to relate student performance in the three examinations of the course (two mid-term exams and a non-cumulative, final exam) to the students' pattern of learning style preferences as measured by eighteen variables which measure learning style preferences according to the Dunn and Dunn (Dunn, 2000) learning style model (DDLSM). Also included in the OLS regression model are a series of dummy variables to control for the gender of the student, which exam the observation measures and from which of the two sections the observation is taken.

It should be noted that despite the many studies that have been done of the DDLSM, this is the first effort known to the authors to apply a regression model to the task of explaining variations in student achievement in a graduate course. The DDLSM breaks down the eighteen learning style preference variables into five categories, labeled as the environmental, emotional, sociological, physiological and psychological categories. The present paper also utilizes a restricted least squares regression methodology in order to examine which of these categories of learning style variables have a significant impact on student achievement and ranks the categories by their order of importance in explaining variations in student achievement.

LITERATURE REVIEW

Coffield, Moseley, Hall and Ecclestone (2004) analyze seventy-one different learning style models--theoretical perspectives which explain the methods, patterns, processes, personality traits or other individual characteristics which influence how people learn. Coffield et al. (2004), classify the DDLSM as one of the "major" learning style models. This view is echoed by another critic of the DDLSM, Ivie (2009), and more neutral analysts of learning style theory and practice (e.g., Arthurs (2007) and Hawk & Shah (2007)).

While the research of some scholars has cast doubt on the value of an emphasis on learning styles as a productive strategy to promote learning (see, for example, Coffield, et al. (2004) and Pashler, McDaniel, Rohrer and Bjork (2008)), the research of other scholars has provided support for the basic hypothesis that attention to student learning styles offers a productive path to enhanced teaching and learning effectiveness (see, for example, Sternberg, Grigorenko and Zhang (2008) and Naimie, Siraj, Piaw, Shagholi and Abuzaid, (2010)). Moreover, the particular perspective on learning styles embodied by the DDLSM has also been subject recently to contrary evidence regarding the value and validity of that learning styles model. The research of Kavale and LaFever (2007) and Ivie (2009) has been critical of many of the claims of success made on behalf of the DDLSM, while the research of Lovelace (2005) and Terregrossa, Englander and Wang (2010) has supported the DDLSM's internal validity and pedagogical effectiveness. (It is noted that the latter study analyzed the influence of learning style preferences on the performance of undergraduate students and, unlike the present analysis, also considered the relationship between learning style preferences and student behavioral choices such as alcohol consumption, internet use, employment hours, fitness activities and course study hours.)

The central tenet of the DDLSM is that the best method of teaching is the method that most closely matches the way the student learns. The DDLSM is composed of a combination of environmental, emotional, sociological, physiological and psychological categories. Each category is composed of several elements. The Productivity Environmental Preference Survey (PEPS) (Dunn, Dunn, & Price, 2006) was used to identify students' preferences for each of the elements in the environmental, emotional, sociological, physiological, and psychological categories.

The environmental category includes preferences for sound, light, temperature, and design, or formal versus informal seating. The emotional category includes propensities for motivation, persistence, conformity (responsibility), and structure. The sociological category reflects whether a student prefers to learn with an authority figure present and also reflects whether the student prefers working with other students. This category also considers whether a student prefers to learn in a routine manner or in a variety of ways. The physiological category includes preferences for perceptual modality (auditory, visual, tactile, and kinesthetic), intake (snacks or drinks), time of day varying energy levels throughout the day, and mobility. The psychological category refers to the way that a student absorbs and processes new and difficult information. Within the psychological category, students' learning styles are hypothesized to be analytic, global, or indifferent. The analytical and global styles correlate with preferences for sound, light, design, persistence and intake (identified as the five discriminating elements--i.e., the learning style preferences which are most relevant to categorizing a student as being an analytic or global learner). It is noted that these five discriminating elements are variables that are also components of the other learning style categories.

METHODOLOGY

Sample and Data Collection

Data for the thirty-six students were drawn from two sections of the same MBA course in economics (covering both microeconomic and macroeconomic topics, roughly the same topic agenda as what is typically covered in an undergraduate, two semester introductory sequence in the standard business curriculum). These two sections were taught in the Fall 2003 and Fall 2004 semesters, respectively. There are three exam observations, stacked, provided for each of the thirty-six student subjects, leading to a total of 108 observations. The dependent variable of the OLS regression was the number of correct answers that each student recorded in each of the three exams given over the course. The multiple choice questions from each of the three exams for each section course were identical and were selected by the instructor from the test bank provided by the publisher of the textbook used for the course. Both sections were taught by the same instructor utilizing the same syllabus. The MBA program is given at a small to medium size campus at a university in the northeast. The business program is accredited by the AACSB.

The students in both sections were requested to complete a PEPS survey in order to determine their learning style preferences. The PEPS (Dunn, Dunn & Price, 2006) instrument, based upon the DDLSM, was used to identify students' preferences for the eighteen elements comprising the five categories of the learning style model. The PEPS, designed to identify how college students and other adults learn and perform in their academic and occupational pursuits, is a self-report composed of 100 questions that can be completed in approximately twenty minutes. Each question is designed to identify an individual's preferences for eighteen separate elements which, in turn, are included among the environmental, emotional, sociological, physiological, and psychological categories. For example, to determine a preference regarding sound, an environmental factor, students are asked to answer whether they strongly disagree, disagree, are uncertain, agree, or strongly agree to a series of statements, such as:

1. I can block out noise or sound when I work.

2. I prefer to work with music playing.

3. Noise or extraneous sound usually keeps me from concentrating.

4. I can block out most sound when I work.

In a similar manner, preferences regarding all environmental, emotional, sociological, physiological, and psychological elements are identified.

The OLS Regression Model

The OLS regression model explaining variations in student exam performance for students taking the graduate economics class can be stated as:

E[X.sub.ij] = [a.sub.0] + [18.summation over (b=1)] [a.sub.b] LS[V.sub.bi] + [22.summation over (c=19)] [a.sub.c] [Dummy.sub.ci] + [e.sub.ij] (1)

where i is the student index, i = 1 to 36 and j =1 to 3, referring to the three tests. E[X.sub.ij] is the ith student's score on the jth test. LS[V.sub.bi] is the bth learning style variable for ith student. There are a total of four Dummy variables: Dummy19 is the gender binary variable, set equal to 1 for males; [Dummy.sub.20] is the first exam binary variable, set equal to 1 for the first exam; [Dummy.sub.21] is the second exam binary variable set equal to 1 for the second exam; [Dummy.sub.22] is the section binary variable, set equal to 1 for students taking the course in the Fall 2003 semester and set equal to 0 for students taking the course in the Fall 2004 semester; and [e.sub.ij] is a stochastic error term.

With regard to the model's control variables, a dummy variable accounting for gender is included on the basis of the research of Kane and Spizman (1999), Durden and Ellis (2003) and Krohn (2005) indicating that males tend to perform better, ceteris paribus, in economics courses than females. Dummy variables are also included to control for possible differences in the difficulty level of the three exams. The final exam is the reference exam. There is also a dummy variable included in the model to account for possible differences in the semester in which the course was taught. This latter dummy variable would be significant, for example, if the admissions standards of the host university had changed between the cohort taking the course in Fall 2003 versus the cohort taking the class in Fall 2004. College instructors might also argue that students in one class may do better than students in a comparable class, even if the same quality of students in another class, because a better rapport might develop between the instructor and the class in one section than another. Also, between two classes of roughly equally able students, a more productive student-to-student dynamic may develop over the semester which may allow an instructor to facilitate more learning in one class over the other.

The results of this regression model are presented in Table 1. Focusing first on the learning style variables, it may be observed that student exam performance has a statistically significant (at the 0.05 level) and positive relationship to the tactile, noise and persistence learning style preferences. A negative and significant relationship is seen between exam performance and the motivation learning style preference variable. These results may be interpreted as an indication that for the particular sample of thirty-six graduate students taking the MBA economics class, learning the course material would be easier to the extent that such material was presented in a manner that took advantage of those students' preference for greater persistence in the learning process and for a greater use of teaching methods which took advantage of the students' preference for utilizing tactile techniques in the processing of the concepts at hand. Likewise, the results provide evidence that teaching approaches which avoided a reliance on kinesthetic methods would likely be more productive.

With respect to the control variables, student exam performance was found to be significantly and negatively related to the gender variable. This indicates that, everything else being equal, female students performed better in the MBA economics course than male students. This result is contrary to the earlier research findings of Kane and Spizman (1999), Durden and Ellis (2003) and Krohn and O'Connor (2005). The coefficients for the test dummy1 ([Dummy.sub.20]) and test dummy 2 ([Dummy.sub.21]) variables are positive, statistically significant in the case of the test dummy1 variable. This suggests that the first exam may have been less difficult than the reference exam, the final. However, it is also possible that such a finding could be caused by students generally tending to do less well during final exams. It may be argued that students are under more pressure during final exams when they are facing exams in all of their courses within a short (one week in the case of the university whose students are being analyzed in this study) time interval. This greater pressure on students could lead to weaker performance on final exams, irrespective of the inherent difficulty level of the final exam itself. The SectionDummy1 variable was not found to be significantly related to student achievement.

The standardized (beta) coefficients in column (6) of Table 1 allow an analysis to be made of the relative contribution of the explanatory variables in explaining variations in student exam performance. These results indicate that among the independent variables in Table 1 that were found to be statistically significant, the variables which had the greatest relative impact on student exam performance were, in descending order, the preference for persistence, motivation, for tactile learning, the male gender, the TestDummy1 variable and the preference for noise.

It should be noted that the adjusted R-squared for the overall regression model was just over 0.30. That is, these explanatory variables explain approximately thirty percent of the student variation in exam performance for this MBA economics course. The value of the F-Statistic for the overall equation, 3.11, is statistically significant, evidence that the regression equation as a whole explains a significant portion of the student to student variation in exam performance.

Teaching Prescriptions Based on Regression Results

The statistically significant relationships between student performance in the MBA economics course and several of the learning style preference variables would offer an instructor opportunities to tailor his/her teaching approaches in order to make the teaching styles more congruent with the learning styles of the particular group of students that the instructor is facing. For example, the significantly negative coefficient associated with the learning style preference for motivation indicates that the MBA students analyzed in this study are not internally motivated to study the principles of economics. This lack of internal motivation to learn the principles of economics may be because the students are majoring in alternative fields of study in the MBA program, for example, management, accounting, marketing and finance, and may not appreciate the importance and relevance of the principles of economics in their major field of study. Therefore, to enhance both the instructor's teaching effectiveness and the students' achievement, it is incumbent upon the instructor to exogenously motivate students to learn the principles of economics. Consistent with the DDLSM, the instructor may more effectively motivate students to learn by appealing to their learning style strengths.

For the students in this sample, the instructor could appeal to the students' positive preference for persistence. The instructor should explain that to successfully complete the requirements in their major field of study, the students need to understand the principles of economics and demonstrate how economic concepts are relevant to their area(s) of interest. For example, the instructor could demonstrate how microeconomics provides the methodology necessary for identifying and analyzing the costs and benefits associated with any business activity, the first step in making informed business decisions. The instructor could also emphasize the importance of utilizing macroeconomics to analyze current conditions and form expectations of future market conditions (domestic and global) to adequately prepare for and adapt to the ever changing business environment. In this way, students realize that to be successful in their respective fields of study, they must understand the principles of economics. Such a strategy may help them to be motivated to do well in the course.

The significantly positive coefficient associated with students' preference for background noise indicates that student achievement increases when background noise exists while students are learning. An instructor wishing to adapt his/her teaching approach in order to utilize this pattern of learning style preferences could survey the students early in the semester to determine what type of acceptable background noise the students actually prefer. For example, if the students indicate a preference for background music while learning, then the instructor could allow classical, rock and roll, soft rock, pop or rap music to be played during class depending upon the tastes of the students and the instructor. It may be difficult to get all students to agree to a specific type of music, but the choice of music should be a consensus choice. In other words, the choice of music cannot in any way detract from the ability of any student's ability to learn.

The significantly positive coefficient of the tactual variable indicates that student achievement in economics may be enhanced if students are given the opportunity to manipulate learning resources, e.g., to transfer what they are learning to another medium. One potential teaching method that provides students the opportunity to manipulate the learning resources is the team learning method. For example, the team learning method could be used to teach students about the economic causes and consequences of the Industrial Revolution. First, divide the class into teams composed of an approximately equal number of students and have the teams physically move their seats to form circles. The instructor would explain that the task for each team is to identify at least three main causes of the Industrial Revolution as it occurred in England in the 1700's, identify any parallels to the economy of the United States in the 21st Century and make recommendations regarding the government's role in promoting economic growth, stability and equity. Students are instructed to use their textbooks, the internet or any other available or academically appropriate source of pertinent information. Each group must record the information to a sheet of paper, a text document or a Power Point presentation. Then each group must present the information to the class, either by a designated member of the team, or by some or all of the team members. The instructor should roam the classroom to answer any questions. The instructor would then divide the chalk board so that each team could summarize its findings and recommendations. Finally, students are instructed to grade each team's effort, but no team would be allowed to grade their own work. Utilization of this suggested teaching method provides students with the opportunity to manually manipulate learning resources while learning more about an important historical and contemporary economic issue.

Relating Student Achievement to Learning Style Categories

The restricted least squares regression method is employed to determine which categories have the greatest impact on student achievement. The hypothesis tested is whether or not a particular category, when considered as a subset of the total learning style elements, has a significant impact on student achievement. The specific question addressed is whether it is important to focus on the environmental, emotional, sociological, physiological, or psychological categories and the relative attention to be given to each category. To answer this question, we utilize the joint F-test for the restricted least square regression method, an approach suggested by Pindyck and Rubinfeld (1998). That is, the partial correlation coefficients in the linear regression model reported in Table 1 provide evidence regarding which learning style elements have a statistically significant effect on student achievement. However, the partial correlation coefficients do not indicate the statistical significance of the alternative categories, or group of elements. The restricted least squares regression method is used to determine which category, or subset of elements, has a statistically significant effect on student achievement. Also, by examining the F-statistics, such as those calculated through the use of equation (2) (Pindyck & Rubinfeld, 1998, p. 130), associated with each category and the p-value associated with those Fstatistics, it is possible to measure the relative importance of the alternative learning style categories.

[F.sub.q,N-k] = ([R.sup.2.sub.UR] - [r.sup.2.sub.R])/q/(1 - [r.sup.2.sub.UR])/(N - k) (2)

where [R.sup.2.sub.UR] is the R-square estimated from equation (1); q is the number of learning style variables in a particular category; [R.sup.2.sub.R] is the R-square from the modified equation (10 without those q learning style variables; N is the total number of observations and k is the number of remaining independent variables including the intercept.

There are two important reasons to examine the significance of the alternative categories. First, this addresses the question of which categories have a significant impact on student achievement. Obviously, for a professor to incorporate the principles of learning style theory into the teaching approach, he or she requires accurate information regarding on which categories to focus. Second, this empirical methodology allows for an objective evaluation of the efficacy of implementing learning style principles in general, and the usefulness of the Dunn and Dunn model (Dunn, 2000) in particular. That is, this approach allows us to assess the significance of the emotional, physiological and psychological categories that Hawk and Shaw (2007) identify as unique to the Dunn and Dunn learning style model.

The results of the restricted least squares regression are presented in Table 2. These results offer evidence that four of the five learning style related categories, i.e., the environmental category, the emotional category, the physiological category and the psychological category, had a statistically significant impact on student exam performance. Only the sociological learning style preference category failed to demonstrate a significant relationship to the dependent variable. It is also noted that a separate category established to account for the collection of control variables (i.e., gender, the dummy variables accounting for the different exams and the dummy variable accounting for the two sections of the course) also was found to have a significant relationship with student exam performance. Among the four significant learning style preference categories, the F-statistics for each category indicate the relative importance of each of the four categories. These results, arrayed in descending order of magnitude, indicate that greater emphasis that should be given to the psychological category, the emotional category, the physiological category and the environmental category. It should also be stressed that elements within the aforementioned four learning style categories have been recognized by Hawk and Shah (2007) as being unique to the DDLSM in their comparison of the relative advantages and disadvantages of the major, competing learning style models. That these categories are found to have the greatest weight in explaining variations in student performance is, it may be argued, strong evidence supporting the predictive validity of the DDLSM.

SUMMARY AND CONCLUSIONS

This paper has analyzed the performance of thirty-six students taking the first economics course in the MBA curriculum in terms of the relationship between their performance and the learning style preferences of those students. The learning style profile of these students was measured with the PEPS learning style instrument (Dunn, Dunn & Price, 2006) which was designed to be consistent with the DDLSM. Among the eighteen different learning style variables that were related to student performance in the three exams given in the graduate economics course, performance was found to be significantly and directly related to student preferences for persistence, noise and tactile modes of learning. Student performance was also found to be inversely related to the learning style preference for motivation. Among the control variables also considered in this analysis, student performance was directly and significantly related to the dummy variable representing the first exam. This result suggests that the first exam may have been relatively less difficult than the reference exam, the final. Student performance was also negatively related to the gender variable, suggesting that female students performed at a higher level than the male students who were enrolled. This result is surprising in the light of the earlier research that has been done explaining variations in college economics courses. It is possible that part of this difference may result from the fact that the earlier research in question examined the influence of gender in student performance at the undergraduate level.

This paper also used a restricted least squares regression method to analyze the performance of these MBA economics students relative to the categories of learning styles formulated by the DDLSM. That is, the eighteen learning style preference variables have been aggregated into five groupings of learning styles elements: environmental, emotional, sociological, physiological and psychological. It was determined that all of these categories had a significant relationship to student performance except for the sociological category. A review of the F-statistics and p-value associated with each of those F-statistics allows a ranking to be made of the relative importance of those learning style categories. In the case of the student subjects studied in the present analysis, it was found that among the variables having a significant link to student performance, the categories with the greatest relative importance, in descending order, were psychological, emotional, physiological and environmental. Learning style elements within those four categories were identified as being unique to the DDLSM, in the comparison of the relative strengths and weaknesses of various high-profile learning style theories performed by Hawk and Shah (2007). The present authors believe that the results of this paper, whereby elements within those particular learning style categories that have been identified as being distinctive to the DDLSM in the Hawk and Shaw (2007) analysis of major learning style models are found to have a significant and relatively strong link to student performance, offers evidence supporting the predictive validity of the Dunn and Dunn approach.

LIMITATIONS OF THIS STUDY

It is difficult to generalize these results to the overall group of MBA students (or graduate students, more broadly) based on such a small sample size. On the other hand, a key aspect of utilizing a learning style model such as the DDLSM is the recognition that one should be wary of trying to generalize the pattern of learning preferences observed with one group of students to another group of students. The basic approach that has been presented in this study is to diagnose the learning style profile of a particular group, knowing that the pattern is very likely to change from one group of students to another, and then crafting teaching methods which are congruent with the particular group at hand.

RECOMMENDATIONS FOR FUTURE RESEARCH

One useful extension of the present study would be to analyze the learning style profile of graduate students with the use of the more recently developed Building Excellence (BE) survey, rather than the PEPS instrument utilized in this study. Recent research (Terregrossa, Englander, Wang, & Wielkopolski, 2011) indicated that the BE, also consistent with the DDLSM, may be a more effective instrument in diagnosing learning style preferences and explaining variations in student performance. Earlier regression-based learning style research (e.g., Terregrossa, Englander, & Wang, 2010) has also demonstrated that student performance may be influenced by the age of students, the extent of their labor market efforts and measures of their academic ability (such as grade point average or performance on standardized tests). It would be useful to examine the extent to which such control variables influence student performance among graduate students. The impact of undergraduate major on performance would also be of interest.

REFERENCES

Arthurs, J. (2007). A juggling act in the classroom: Managing different learning styles. Teaching and Learning in Nursing, 2, 2-7.

Coffield, F. J., Moseley, D. V., Hall, E., & Ecclestone, K. (2004). Should we be using learning styles? What research has to say to practice. London: Learning and Skills Research Centre.

Dunn, R. (2000). Capitalizing on college students' learning styles: theory, practice, and research. In R. Dunn and S. A. Griggs (Eds.), Practical approaches to using learning styles in higher education (pp 3-18). Westport, CT: Bergin & Garvey, Inc.

Dunn, R., Dunn, K., & Price, G. E., (2006). Productivity environmental preference survey. Lawrence KS: Price Systems, Inc.

Durden, G., & L. Ellis. 2003. Is class attendance a proxy variable for student motivation in economics classes? An empirical analysis. International Social Science Review, 78, 42-46.

Hawk, T. F., & Shah, A. J. (2007). Using learning style instruments to enhance student learning. Decision Sciences Journal of Innovative Education, 5, 1-19.

Ivie, S. (2009). Learning Styles: Humpty Dumpty revisited. McGill Journal of Education. 44 (2), 177-192.

Kane, J., & Spizman, M. (1999). Determinants of student retention of microeconomics principles concepts. SUNY-Oswego Economics Department Working Paper 1999-01.

Kavale, K., & LeFever, G. (2007). Dunn and Dunn model of learning style preferences: Critique of Lovelace meta-analysis. The Journal of Educational Research, 101(2), 94-107.

Krohn, G., & O'Connor, C. (2005). Student effort and performance over the semester. Journal of Economic Education, 36, 3-28.

Lovelace, M. (2005). Meta-analysis of Experimental research based on the Dunn and Dunn model. The Journal of Educational Research, 98(3), 176-183.

Naimie, Z., Siraj, S., Piaw, C, Shagholi, R., & Abuzaid, R. (2010). Procedia Social and Behavioral Sciences, 2, 349-353.

Pashler,H., McDaniel, M., Rohrer, D., & Bjork, R. (2008). Learning styles: Concepts and evidence. Psychological Science in the Public Interest, 9(3), 105-119.

Pindyck, R. S., & Rubinfeld, D. L. (1998). Econometric models and economic forecasts (4th ed.). New York City: McGraw-Hill Irwin.

Sternberg, R., Grigorenko, E., & Zhang, L (2008). Styles of learning and thinking matter in instruction and assessment. Perspectives on Psychological Science, 3(6), 486-506.

Terregrossa, R., Englander, F., & Wang, Z. (2010). How student achievement is related to behaviors and learning style preferences. International Journal of Education Research, 5 (2), 94-108.

Terregrossa, R., Englander, F., Wang, Z., & Wielkopolski, T. (2011). How college instructors can enhance student achievement: Testing a learning styles theory. International Journal of Education Research, forthcoming.

About the Authors:

Fred Englander is a Professor of Economics at Fairleigh Dickinson University in Madison, New Jersey. He received his Ph.D. in economics from Rutgers University. He has published articles in the Southern Economics Journal, the Business Ethics Quarterly, Science and Engineering Ethics and the Journal of Education for Business.

Ralph A. Terregrossa is an Associate Professor of Economics at St. John's University in New York City. He holds a Ph.D. in economics from Binghamton University. He has published articles in The Quarterly Review of Economics and Finance, International Advances in Economic Research and Educational Review.

Zhaobo Wang is an Associate Professor of Production and Operations Management at Fairleigh Dickinson University, Madison, New Jersey. He received a Ph.D. in operations research from Rutgers University. He has published articles in the Journal of Educational and Behavioral Statistics and the Journal for Economic Educators.

Fred Englander

Fairleigh Dickinson University

Ralph A. Terregrossa

St. John's University

Zhaobo Wang

Fairleigh Dickinson University

Table 1

Student Achievement OLS Regression Results

Parameter Estimates

(1)                     (2)        (3)       (4)
Variable           Parameter   Standard
                   Estimate      Error    t Value

Intercept            -0.003     19.590         0
Tactile               0.348      0.088      3.94
Motivation           -0.552      0.166     -3.32
Persistence           0.729      0.235      3.11
Dummy-Test!           2.917      0.972      3.00
Noise                 0.281      0.119      2.37
Dummy-Gender         -3.592      1.562      -2.3
Kinesthetic          -0.256      0.155     -1.65
Dummy-Test2           1.528      0.972      1.57
Intake               -0.085      0.068     -1.25
Temperature          -0.100      0.081     -1.23
Several ways          0.080      0.080      1.00
Authority figure     -0.109      0.110     -0.99
Design                0.094      0.102      0.92
Alone-peers           0.099      0.114      0.87
Dummy-Section1        0.890      1.236      0.72
Auditory              0.045      0.066      0.69
Responsible          -0.057      0.090     -0.64
Light                -0.050      0.111     -0.45
Mobility              0.023      0.083      0.28
Time-of-day          -0.012      0.075     -0.15
Visual                0.012      0.083      0.14
Structure             0.003      0.070      0.04

R-squared = 0.4456   S.E.E. = 4.126   F-Value = 3.1 and P-value <.0001
Adj R-squared= 0.3022   Coefficient of Variation = 15.402

Parameter Estimates

(1)                    (5)            (6)         (7)
Variable                      Standardized   Variance
                   Pr > |t|      Estimate    Inflation

Intercept            1.000              0           0
Tactile              0.000          0.525       2.723
Motivation           0.001         -0.549       4.180
Persistence          0.003          0.646       6.637
Dummy-Test!          0.004          0.280       1.333
Noise                0.020          0.247       1.670
Dummy-Gender         0.024         -0.360       3.764
Kinesthetic          0.103         -0.229       2.960
Dummy-Test2          0.120          0.147       1.333
Intake               0.216         -0.130       1.661
Temperature          0.221         -0.199       3.979
Several ways         0.319          0.119       2.157
Authority figure     0.324         -0.138       2.948
Design               0.359          0.144       3.725
Alone-peers          0.387          0.144       4.185
Dummy-Section1       0.473          0.088       2.302
Auditory             0.493          0.083       2.231
Responsible          0.525         -0.112       4.690
Light                0.651         -0.060       2.716
Mobility             0.784          0.031       1.933
Time-of-day          0.878         -0.021       2.739
Visual               0.888          0.017       2.311
Structure            0.971          0.004       2.038

R-squared = 0.4456   S.E.E. = 4.126   F-Value = 3.1 and P-value <.0001
Adj R-squared= 0.3022   Coefficient of Variation = 15.402

Table 2

Joint F-Test of the Alternative Learning Style Categories

Strand        Dummies   Environmental   Emotional   Sociological

Restriction   4         4               4           3
R Square      0.3344    0.3658          0.2907      0.4313
F -stats      4.262     3.059           5.937       0.726
p-value       0.0031    0.0199          0.0002      0.5389

Strand        Physiological   Psychological   Whole

Restriction   7               5               23
R Square      0.3003          0.2564          0.4456
F -stats      3.182           5.802
p-value       0.0044          0.0001