Colloquium: Considering learning style in computer-assisted learning

By Roy B. Clariana

British Journal of Educational Technology, 28 No 1 1997, pp. 66-68.

Dr Roy B. Clariana currently works for Jostens Learning Corporation (of San Diego, California, USA); a leading provider of instructional software) as an educational consultant. He also maintains a relationship with the University of Colorado Graduate School of Education, Department of Instructional Technology, as an Honorarium Faculty member. Address for correspondence P.O. Box 488, Almont, CO 81210, USA also: rclariana@aol.com

Introduction

Learning style has been defined by Keefe (1979 ) as "the characteristic behaviors of learners that serve as relatively stable indicators of how they perceive, interact with, and respond to the learning environment." For example, Kolb's (1976) Learning Style model describes two bipolar dimensions, abstract conceptualization (AC) versus concrete experience (CE) and reflective observation (RO) versus active experimentation (AE). Matching instructional delivery to student learning style preference should positively impact student achievement (Carrier and Sales, 1987; Stice and Dunn, 1985). Research relating learning style preference and achievement should inform guiding principles for developing truly individualizing computer-assisted learning (CAL).

Method and results

Thirteen- and fourteen-year old students (n = 23) received approximately 30 minutes of CAL per day each day for five months (about 50 hours). The Kolb Learning Style Inventory (LSI) was administered pre and post study, as was a standardized mathematics test. In a stepwise multiple regression of all variables, with mathematics posttest as dependent variable, mathematics pretest entered the equation first as expected (t = 0.0001, multiple r = 0.75). The pre-survey CE dimension of the LSI entered the equation at step 2 (t = 0.0001, multiple r = 0.90). No other variables entered the equation (pin limits set at alpha = .05), suggesting that the learning style preference CE relates to increased mathematics achievement in this CAL environment. Unexpectedly, the learning style dimensions changed during this 5-month period towards the CE and AE dimensions, especially for the high-ability group (see Table 1). Additional investigations were undertaken to consider learning-style change in CAL.

The second survey consisted of students (n = 30) aged nineteen to twenty-one years old involved in a remediation project. These students received computer mathematics instruction for about three hours per week over a five-week period of time (about 15 hours). Kolb LSI data were collected at the beginning and end of the project. Mathematics standardized test scores were used to form high and low ability groupings. Again, a change towards the CE and AE dimensions were observed (refer back to Table 1).

Table 1: Learning Style Preference change expressed as effect size (change).

	Study 1	Study 2	Study 3
From AC to CE (Low ability)	0.19	0.24	0.03
From AC to CE (High ability)	0.64	0.22	0.44
From RO to AE (Low ability)	0.16	0.15	0.38
From RO to AE (High ability)	0.55	0.40	0.45

(note: no average change observed from CE to AC or from AE to RO)

A third survey consisted of adult education majors (n = 41) enrolled in a microcomputer course. Students spent about 2 hours per week with hands-on instruction with microcomputers and about 1 hour per week in lecture and demonstration (about 15 hours). The pre to post Kolb LSI data covered a period of five weeks. Course mid-term grades were used to form high and low ability groupings. Again. a change towards the CE and AE dimensions was observed (refer back to Table 1).

The high-ability groups consistently exhibited larger effect size changes than the low ability groups, except for the AC to CE dimension in Study 2 that was about equal for low and high ability learners. Also, the longer duration study showed generally larger effect sizes (Study 1 compared to 2 and 3, Table 1).

Discussion

Previous research indicates that on average, primary school teachers and students prefer the AC and AE learning style dimensions (Kolb. 1981). In this study involving three very diverse CAL experiences and different learner populations. a general shift occurred in learning style towards CE and (more) AE. The magnitude of the shift appears to vary with learner ability and extent of exposure to CAL.

If a learning style shift does actually happen with exposure to CAL. what are the possible implications for schools? The following characteristics are associated with these LSI dimensions. Learners would be less passive and more active, there would probably be less reflection and more action. They would tend towards convergent thinking at the expense of divergent thinking. Back-paging to previous content or review, if available, may be reduced. The tendency to guess the responses to questions in a trial-and-error manner may increase and would probably be rewarded. Thus, rather than pondering at length on a screen as with difficult print text, a learner would be likely to press return and hope the meaning would eventually become evident. This would increase overall instructional risk-taking behaviours resulting in a tendency to push on or forge ahead in a lesson. There is some face validity to support this in that teachers often comment that some normally reticent students (many with learning difficulties) come out of their shell and begin to work with CAL. These behaviours may be positive characteristics when the instructional methodology employed is mastery-based (with low chunk size and plenty of feedback), but may be inappropriate with other approaches.

Additionally, if these behaviours transfer from the computer to the classroom, some of these behaviours may be viewed as negative by teachers and some as positive. At any rate, these characteristics contrast with many behaviours that occur in the traditional classroom and this contrast should be further discussed and examined.

References

Carrier C and Sales G (1987) The effect of learning style and type of feedback in computer-based instruction. A paper presented at the annual meeting of the American Educational Research Association. April 198 7, Washington DC.

Keefe, W (1979) Learning Style: An Overview in Student Learning Styles National Association of Secondary School Principals.

Kolb D A (1976) Learning Style Inventory: Technical Manual MCEer, Boston, Mass.

Kolb D A (1981) Learning styles and disciplinary differences in Chickering and associates (eds) The Modern American College ]osey-Bass. San Francisco.

Stice C F and Dulm M B (1985) Learner styles and strategy lessons: A little something for everyone. Paper presented at the Annual Meeting of the Southeastern Regional Conference of the International Reading Association. Nashville. TN, November 1985 (ERIC Document Reproduction Service number ED 271 721).

.