Education-line Home Page

"I did it my way…". The impact of learning styles and strategies on students’ success in quantitative research methods in educational sciences

Juhani Rautopuro and Pertti Väisänen

Paper presented at the European Conference on Educational Research, University of Hamburg, 17-20 September 2003


In this paper the impact of two important factors – the learning styles and strategies – on study success in applied statistics is examined. The main question was to determine the effect of students’ learning styles and learning strategies on their course performance in Quantitative Research Methods course. Also, the effects of students’ learning styles and learning strategies on course success in different learner groups that were based on their prior knowledge in mathematics and statistics as well as their beliefs and perceptions of themselves as learners of statistics was examined. The research was conducted in the spring 2003 in the Faculty of Education, at the University of Joensuu, Finland. Some results will be presented and their implications for the improvement of instruction discussed.


Problems of teaching and learning statistics are widely recognised and commonly shared by statistics educators and researchers. Those problems are both cognitive and non-cognitive in nature. Such factors as deficiencies in the mathematical background of students, prior beliefs, misconceptions and strongly-held but incorrect intuitions of statistics and pre-dispositions against statistics, caused by intuition or prior learning experiences in mathematics, have been brought up in previous research (Batanero et al.1994; Galagedera 1998; Garfield 1997). Negative emotions (e.g., fears), negative attitudes and beliefs and lack of motivation (Gal & Ginsburg 1994) as well as test anxiety, anxiety about mathematics (Townsend et al. 1998) or statistics anxiety (Birenbaum & Eylath 1994; Onwuegbuzie & Seaman 1995), weak self-confidence in learning mathematics or statistics and low self-efficacy (Forte 1995; Townsend et al. 1998) may interfere with learning, too. Inadequacies in prerequisite mathematics skills and abstract reasoning are also part of the statistics learning problems (Garfield & Ahlgren 1988). On the other hand, previous research has shown how students’ prior mathematics skills positively influence success (Elmore et al. 1993; Galagedera 1998) and also how positive attitudes towards statistics appears to contribute to success in statistics courses (Elmore et al. 1993).

While research has informed us about these numerous factors that result in success and failure in statistics courses, unanswered questions still remain. In research conducted in educational psychology many factors which affect learning have been suggested. For instance, the importance of motivation and learning strategies as well as learning styles to academic performance has been demonstrated. Research has indicated that learning in, at least partially, matched conditions (i.e. teaching using instructional styles and materials structured to suit students’ thinking and learning styles) is significantly superior to the teaching in mismatched conditions (Dunn, Beaudry & Klavas 1989; Grigorenko & Sternberg 1997; Sternberg et al. 1999). Students may also enjoy learning more when they are taught in a way that takes into account their learning style preferences (Dunn et al.1989). In opposition, a mismatch between teaching and learning styles cause learning failure, frustration and demotivation (Peacock 2001; Reid 1995). Therefore, we believe that an understanding of styles component in human behaviour could improve teachers’ teaching and, thus, student learning. Felder (1995, 27), however, warns that a learning style a learner likes may not be the best one for learning.

This study is important because – as far as we know – there are only a few reported studies (e.g., Bell 1998; Onwuegbuzie 2001) and theoretical discussions (e.g., Snee 1993) of the impact of learning styles and learning strategies (e.g., Schutz et al. 1998) on statistics learning. There is also a need for more research about the styles of learning because the scope of learning style conceptualisations, theoretical approaches and ways of measurement used in research vary widely (see e.g., Curry 1991). Those few studies reported in the domain of statistics learning can only slightly touch this complex research area. Onwuegbuzie and Daley (1997), for example, suggest that learning styles play an important role in research methodology studies. Particulary, there exists evidence that graduate students who prefer to learn in co-operative learning groups tend to obtain lower levels of performance in research methodology courses in which all assignments are carried out and graded individually than do their counterparts who have more individualistic orientations (Onwuegbuzie 2001). Bell (1998) found some interesting differences between visual, auditory and tactile learners in an business statistics course. According to the study, visual learners were significantly better in their success than the other two groups. However, the correlation coefficients between learning style scores on each of the scales were rather low. Schutz and others (1998) examined motivation and learning strategies used by students in an introductory statistics course and provided mixed support for the use of deeper-level processing strategies.

Various definitions of learning styles and related terminology are presented in literature including at least four different theoretical approaches to learning styles (Ediger 1995; McKeachie 1996; Schmeck 1988; Svinicki 1995). In our study personality centred approach (e.g., Grigorenko & Sternberg 1995) to learning styles was used. A learning style inventory on the basis of previous research on learning styles (e.g., Dunn et al. 1989; Dunn et al. 1995; Reid 1987) was constructed. Learning style was defined as natural, habitual, and preferred way(s) of absorbing, processing, and retaining new information and skills and is comprised of both biological and developmental characteristics that make the identical instructional environments effective for some learners and ineffective for others.

When, according to some definitions, learning styles specifically deal with characteristic styles or strategies of information processing in a learning situation (see Messick 1984; 60–67; Schmeck 1983, 233–234), a concept near to learning style, cognitive style, refer to the characteristic and consistent way of an individual’s information processing. It develops in harmony with personality trends, and as such it also contains typical modes of thinking (Messick 1984, 61). Saracho (1997) suggests that cognitive styles identify the ways individuals react to different situations and they include stable attitudes, preferences, or habitual strategies that distinguish the individual styles of perceiving, remembering, thinking and problem solving. Although a number of cognitive styles - or thinking styles - (see, e.g., Sternberg & Zhang 2001) have been identified and studied over the years and some of those style dimensions have connection to learning, this study focused on learning style conceptualisations.

According to Reid (1987), learning styles are variations among learners in using one or more senses to understand, organise, and retain experiences. Reid categorised styles of learning into six types: Visual (these learners prefer seeing things in writing), Auditory (prefer listening), Kinesthetic (prefer active participation/experiences), Tactile (prefer hands-on work), Group (prefer studying with others), and Individual (prefer studying alone). This study is focused on the first four learning styles defined by Reid (1987). However, in this study the kinesthetic and tactile styles were combined to one learning-style group. Dunn’s Learning Style Inventory obtains a profile of each student in four major areas one of which is called perceptual preferences including auditory, visual, tactile, and kinesthetic styles that are quite similar to Reid’s conceptualisation.

Learning strategies, on the other hand, can be defined as behaviours, steps, operations, or techniques employed by learners to facilitate the acquisition, storage, retrieval, and use of information. (e.g., Oxford 1990). Research suggests that learning styles and strategies do not function separately from each other but learning style determines strategy use (Ehrman & Oxford 1990). While learning styles are generalised ways of processing information, learning strategies are more task specific. Furthermore, learning strategies can be classified into cognitive, affective and metacognitive categories. Our measurement of learning strategies included all those categories. According to literature, affective strategies include techniques for more effective time management and stress management; test anxiety reduction; and increased positive self-talk, self-monitoring, and self-coaching activities (e.g., Doyle & Garland 2001; Meichenbaum1977; Reid 1996).

Research Problems

The research questions of the study have been defined as follows:

1. What kind of groups of students on the basis of their background as statistics learners (maths school grade, elementary statistics course exam score, scores in an entry level diagnosis of statistical knowledge, self-confidence in learning math and statistics) can be identified?

2. Are there any differences in course success between different learning style groups (i.e. visual learners, auditory learners, kinesthetic-tactile learners) as a whole and in different learner groups of statistics?

3. What is the impact of learning strategies on course success as a whole and in different learner groups of statistics?



Approximately 200 students enrolled in the course ‘Quantitative Research Methods in Education 1’ (2 cu) given in two campuses (Joensuu and Savonlinna) at the Faculty of Education, University of Joensuu in the spring term 2002. The teachers of the course were the authors of the present paper. This subject level course was subsequent to the elementary statistics course and contained statistical inference and most common statistical tests (parametric and nonparametric) as well as construction of scales and estimation of reliability. Statistical analyses and computer labs were carried out by using SPSS V11.0 statistical package. Although the contents of the courses and contact teaching hours as well as assignments of course exams were very similar at the campuses they were not identical. All students enrolled in the course took part in the study, however, some students filled the questionnaires incompletely and were thus omitted from the data. The respondents (N = 186) were mainly from teacher education degree programs. In all 56 came from classroom teacher education program, 55 from kindergarten teacher education program, 31 studied home economics and handicrafts, and the remaining 44 were mainly students to become teachers of special education or studied science of education. The number of female students was 157 (84.4 %) and 29 were males (15.6 %).

Data Collection and Measurements

The data for this study were collected in the beginning and at the end of the course. First, in the beginning of the course a diagnostic test (scoring 0–25 points with a mean score 11.62 and standard deviation 3.45 in the data) to measure students’ prior knowledge in statistics was conducted. The diagnostic test consisted of four assignments measuring the student’s skills to apply the concepts and methods that were introduced them in their earlier elementary statistics courses. The first two assignment dealt with interpretation of measures of central tendency and variation (arithmetic mean, median, standard deviation and inter quartile range). Also, some statistical inference concerning group differences on the basis of these statistics had to be done. The third and fourth assignment measured the students’ understanding about linear association described by scatter plots and correlation coefficients.

As the mean value of diagnostic test tells, the results were not very encouraging. Although the students already had passed one elementary course in statistics they still had difficulties and misconceptions concerning measures of dispersion, for example, standard deviation and standard error were totally mixed with each other as well as the range and inter quartile range were mixed. Also, the linear association on the basis of a scatter plot was very hard to realise. Instead of analysing the graph the students started to invent their own explanations outside the information presented. One remarkable point was also that the students had some very curious visions about the absolute value of the correlation coefficient. This meant in practice that many students thought that positive correlation is always stronger than negative one, for example.

Second, a pre-course questionnaire to measure students’ background in mathematics and statistics, their learning styles, and perceptions of statistics as a discipline and themselves as statistics learners was administered. Third, at the end of the course students’ learning strategies were measured. In addition, the measure for academic performance was the participants’ course grade in final examination (scoring 0–30 points with a mean score 16.20 and standard deviation 5.60).

Students’ background in mathematics was operationalised as (a) the level of course taken in upper secondary school mathematics (basic course or advanced course) and (b) upper secondary school grade (scaled 4–10) using a weight of 1,25 in the case of an advanced course (scaling 4–12,50). Students’ background in statistics was operationalised as the course grade in elementary statistics (1–3).

The scales constructed were: First, a 40-item learning style inventory was designed to identify each student’s general learning style, i.e., to find out whether a student prefers to learn in a visual, auditory, kinesthetic–tactile etc. way. We constructed a learning style inventory on the basis previous research on learning styles (e.g., Dunn et al. 1989; Reid 1987), learning style inventories used in the research on statistics learning (Bell 1998) and inventories found in the internet (e.g., For the purpose of this study 22 of the items were utilised to determine each student’s preferred learning style. Principal components analysis, accounting for 40 % of the variance of variables, was used to explore the dimensions of learning styles. Three sub-scales were computed based on the principal components. Each student was assigned into a learning style group in which a student scored highest. Subsequently, 94 of the students were identified as kinesthetic-tactile learners, 43 visual learners, and 21 auditory learners. The reliability estimates (Cronbach’s alpha) for the sub-scales were as follows: kinesthetic-tactile style .78, visual style .69, and auditory style .62.

Second, A 40 item Likert type (keyed 1–5) inventory "Students Perceptions of Statistics as a Discipline and Themselves as Statistics Learners" developed by the authors was also administered in the beginning of the course. This inventory was based on previous studies on statistics learning (e.g., Galagedera et al. 2001; Townsend et al. 1998), our own studies (e.g., Haapala et al. 2002), on Schommer’s (e.g., Schommer 1994) model of Epistemological Beliefs (particularly on two of the belief dimensions: Certain Knowledge and Simple Knowledge), and research about students’ beliefs and views of mathematics reviewed in Pietilä (2002). Using factor analysis with a principal axis as an extraction method six factor solution (Varimax rotation), accounting for 53 % of the common variance, was deemed to be most interpretable. The factors were labelled as ‘Perceived Mathematics Ability’ (with a reliability estimate of .92 basing on the covariance matrix of the factor analysis), ‘Positive Attitudes’ (.84), ‘Self-confidence in Learning Statistics’ (.82), ‘Resistance’ (.76), ‘Statistics Learned as Routine Calculations’ (.72), and ‘Statistics as an Exact Discipline’ (.71).

Third, a 55-item inventory to measure the participants’ course-specific cognitive and affective learning strategies was administered in the end of the course. Learning strategies inventory was developed on the basis of literature on cognitive and affective learning strategies and statistics learning (e.g., Pintrich et al. 1991; Schutz et al. 1998) as well as our experience as statistics teachers. Using Factor Analysis with principal axis as an extraction method a seven factor solution (varimax rotation), accounting for 43 % of the common variance in the variables, was extracted. The resulting learning strategy factors were labelled as ‘Reluctance vs. Purposeful Efforts’ (reliability estimate . 88), ‘Reflection’ (.86), ‘Rote Learning & Memorizing’ (.87), ‘Elaboration’ (.76), ‘Peer-Orientation’ (.91), ‘Self-monitoring’ (.75), and ‘Organisation of Time and Materials’ (. 69).

Sample items of each of the scales are displayed in Table 1 and are reverse scored where necessary.

TABLE 1. Sample Items Comprising Sub-scales

I Learning Style Inventory

Kinesthetic-tactile Learner (9 items)
- "I like "hands on" learning better than learning from a lecture or textbook."
- "I learn better by doing than observing or reading a book."

Visual Learner (9 items)
- "I learn better by reading a book than by listening a lecture."
- "When I read a book I create pictures in mind that illustrate a phenomenon."

Auditory Learner (3 items)
- " I learn better in class when I listen to someone."
- " When I read a book or my notes I can benefit from reading out loud."

II Students Perceptions of Statistics as a Discipline and Themselves as Statistics Learners

Perceived Mathematics Ability (11 items)
- "When solving mathematical problems, I feel nervous and uncertain." (rec)
- " Mathematical thinking and reasoning is easy for me."

Positive Attitudes (11 items)
- "I think it is useful in everyday life to know some statistics."
- "I’m eager to learn statistics."

Self-confidence in Learning Statistics (6 items)
- "I’m skilful and capable enough to manage the course well."
- "For me this course will be easy and don’t demand me to make any particular effort."

Resistance (7 items)
- "It is boring and frustrating to study statistics."
- "If this course were optional, I’m quite sure not to take it."

Statistics Learned as Routine Calculations (3 items)
- "Statistics can be learned by learning the rules, algorithms and formulae by heart ."
- "Statistics can be best learned by doing calculation exercises"

Statistics as an Exact Discipline (2 items)
- "Statistical significance tests give us certain and true knowledge about phenomenon or confirm scientific facts."
- "Statistics is an exact science and gives us certain knowledge about the phenomenon we examine."

III Learning strategies

Reluctance vs. Purposeful Efforts (20 items)
- "Several times when reading for the course exam, I was so lazy and bored that I gave up although I had not reached my aims."
- "When reading the course material I really tried to clarify for myself those concepts and things that were difficult to understand."

Reflection (8 items)
- "After classes I usually reviewed what I was taught."
- "Throughout the course I tried to stay mentally in touch with the course material by reviewing the materials or in my thoughts."

Rote Learning & Memorisation (9 items)
- "I tried to learn things by heart and develop different mnemonics for the test."
- "I studied the course concepts and definitions in verbatim in order to remember and master things in the course exam."

Elaboration (6 items)
- "I tried to relate the material to what I had previously learned."
- "I tried to relate the material to real life problems and my personal experiences."

Peer-Orientation (5 items)
- "I tended to co-operate with my fellow students when doing the course work and preparing for the course exam."
- "I studied the course content together with my fellow students."

Self-monitoring (5 items)
- "When I had solved a problem, I tried to check out whether my solution were reasonable or possible in reality."
- "When interpreting statistical tests, I could go through analytically step by step."

Organisation of Time and Materials (3 items)
- "I found it hard to concentrate in studying the course contents because I had other course at the same time."
- "When reading the course material and my notes, I summarised the main things."


1. Statistics Learner Groups

Cluster analysis was used to classify the students into different learner groups. The students’ perceptions of themselves as statistics learners, the basis of their mathematics abilities (upper secondary school grade and the level of course in maths) and their level of prior knowledge in statistics (course grade in ES and scores in a pre-course diagnostic test) were used to construct the distance matrix for the analysis. This distinction was based on previous studies which have indicated that success is positively affected by students’ real mathematics skills and perceived mathematics abilities, prior statistics knowledge, positive attitudes towards statistics and self-confidence in learning statistics (Elmore, Lewis & Bay 1993; Galagedera’s 1998; Galagedera et al. 2000; Schutz et al. 1998; Townsend et al. 1998; Väisänen & Ylönen 2003).

On the basis of the cluster analysis two learner groups ("favourable background" – "unfavourable background") could be detected. From the sub-scales of students’ perception of themselves as statistics learners only two items (perceived mathematics ability and self-confidence in learning statistics) differed seemingly significantly in these two learner groups. The difference in students’ mathematical background and the level of prior knowledge is also statistically significant between learner groups. The final results of the cluster analysis are presented in Table 2.

TABLE 2. Final Cluster Centres in Different Learner Groups.

Due to many problems in effective testing of the cluster solution the F tests shown in table 2 should be understood as descriptive ones. The observed significance levels are not exact and thus cannot automatically be interpreted as tests of the hypothesis that the cluster means are equal (Dillon & Goldstein 1984).

The difference in course success between these two learner groups was statistically significant (t = 4.25, df = 115, p = .000) as suggested by previous research reviewed above. The mean scores at the final exam of the course in the "favourable background" -group was 18.73 (standard deviation 5.22) and 14.55 (standard deviation 5.42) in the "unfavourable background" -group, respectively. The difference between group means in these learner groups was also practically significant.

2. The Impact of Learning Styles on Course Success

Three learning style groups were formed up, kinesthetic-tactile, visual, and auditory learners. The correlations between different learning style sub-scales varied from .18 to .23. Due to the large sample size these correlations are statistically significant (p < 0.05) but the practical significance is unsubstantial (the coefficient of determination varies from 3% to 5%).The effect of the learning styles on course success was first analysed by one-way analysis of variance. The group means and standard deviations are shown in Table 2.

TABLE 3. Descriptive Statistics of the Course Success in Different Learning Style Groups.

As we can see from Table 3 the mean values in course success between different learning style groups don’t vary very much and the difference between groups is not statistically significant (F (2/156) = .691; p = 0.503). The correlations between the learning style sub-scales and course success were also computed. However, none of the sub-scales had statistically or practically significant association with the course success; the correlation coefficients varied from -.05 to .05.

The effect of learning styles to learning outcomes was analysed also in different learner groups. As observed in whole data no significant difference, however, was detected either in the clustered data.

3. The Impact of Learning Strategies on Course Success

The association between learning strategies and learning outcomes (course success) was analysed by using multiple regression analysis. The learning strategies were used as independent variables and the analysis was carried out for the whole data (Table 4) and also separately in both learner groups (Table 5).

TABLE 4. Results of the Regression Analysis (Enter Method) on Whole Data (N = 153)

Independent Variables

Coefficient (std.error)

Reluctance vs. Purposeful Efforts
Rote Learning & Memorisation
Organisation of Time and Materials

16.24 (.38), p = .000
1.82 (.41), p = .000
.87 (.42), p = .041
-1.47 (.41), p = .001
-.99 (.45), p = .028
.02 (.41), p = .921
2.01 (.45), p = .000
.13 (.47), p = .788


F = 9.34, p = .000
R-square = .311

When we examine the results in Table 4 we can see that the effect of sub-scales "reluctance vs. purposeful efforts", "reflection" and "self-monitoring" have statistically significant and positive association to course success. The effect of sub-scales "rote learning" and "elaboration" is also significant but negative. These results are expected as far as we consider the positive associations. It is also natural that rote learning and memorisation of the course material is bound to weaken the course success. In order to learn and apply statistical inference a student has to understand the main concepts and their relationships. Instead, the sign of the sub-scale "elaboration" was unexpected. Because the large group of students, however, was quite heterogeneous, the results were also analysed separately in different learner groups. The idea was examine the effect of learning strategies on course success for students with different backgrounds.

When we analysed the descriptive statistics (mean, median and standard deviation) of learning strategies in different learner groups we found out that the most obvious differences between groups were in sub-scales "self-monitoring", "rote learning and memorising" and "reluctance vs. purposeful efforts". In average, the students with favourable background scored higher in purposeful efforts and self-monitoring. In opposite, the students in unfavourable background seemed to rely more on rote learning and memorising.

TABLE 5. Results of the Regression Analysis (Enter Method) in Different Learner Groups

Favourable Background (n = 58)

Unfavourable Background (n = 59)

Coefficient (std.error)

Coefficient (std.error)

Reluctance vs. Purposeful Efforts
Rote Learning & Memorisation
Organisation of Time and Materials

17.60 (.59), p = .000
2.61 (.73), p = .001
1.18 (.52), p = .035
-1.12 (.60), p = .069
.29 (.69), p = .671
.20 (.60), p = .734
2.36 (.61), p = .000
1.44 (.74), p = .059

15.73 (.84), p = .000
1.62 (.70), p = .024
.20 (.85), p = .815
-2.46 (1.03), p = .021
-.90 (.79), p = .261
.01 (.90), p = .914
1.54 (.96), p = .116
-.72 (.85), p = .399


F = 6.29, p = .000
R-square = .473

F = 2.04, p = 0.07
R-square = .233

When we compare the results of the regression analyses in two learner groups (Table 5) we can see a few similarities and some interesting differences, too. The sub-scale "reluctance vs. purposeful efforts" is the only scale that has statistically significant and positive relationship to course success in both groups. However, the effect of this sub-scale is greater in the group of students with favourable background. In the "favourable background" -group sub scales "reflection" and "self-monitoring" turned out to be statistically significant predictors for the course success, too. The effect of sub-scales "organisation of time and materials" and "rote learning and memorisation" was close to significant. The last sub-scale mentioned was the only one that had a negative regression coefficient. The coefficient of determination (R-square) was also remarkably higher in the favourable background -group.

No other sub-scales, in addition to the sub-scale "reluctance vs. purposeful efforts" mentioned earlier, that were statistically significant in the "favourable background" -group were significant in the "unfavourable background" -group. In addition to this, the sub-scale "rote learning and memorisation" was a statistically significant (with negative coefficient) predictor to course success in this group.

Furthermore, a quick look at the differences of strategies used in different learning style groups showed statistically significant differences in two out of seven sub-scales of learning strategies. These were "reflection" (F = 6.20, p = .003) and "rote learning and memorising" (F = 3.28, p = .041). The "auditory learning style" –group had the highest mean scores (.47) in sub-scale "reflection" while the "kinesthetic-tactile" -group (-.22) had the lowest mean scores. The students in the auditory-group had also the highest mean score in rote learning and memorisation (.20) while the visual-group had the lowest scores (-.38).


On the basis of our results we can conclude that the learning styles don’t seem to have any effect on course success in applied statistics courses in education. This result was contradictory to Bell’s (1998) study which showed that visual learners were significantly better in their success than auditory and tactile learners in an business statistics course. We could expect from previous research that learning in, at least partially, matched conditions were significantly superior to that in mismatched conditions (Dunn, Beaudry & Klavas 1989; Grigorenko & Sternberg 1997; Sternberg et al. 1999). However, our results indicate that the instruction given in our two courses didn’t favour any specific learning style students generally use in their studying. This may indicate that we have succeeded in presenting new information and materials in a variety of modes and using variety of activities in an attempt to balance our teaching style to students learning styles as Peacock (2001) also has suggested. This balancing is important because research suggests (see Peacock 2001) that mismatch between teaching and learning styles causes learning failure and frustration, and this has implications for both learners and teachers. He further suggests that learners should take more responsibility for their own learning. That is, they should try to meet their own needs through their own efforts when studying in and outside the class. On the other hand, teachers should help students to identify their learning styles and to become more flexible in using them because research has shown the efficiency of versatile learning styles (Kinsella 1995). In addition, most people have learning style preferences, but individual preferences differ significantly and the stronger the preference, the more important it is to provide compatible instructional strategies, especially among less academically successful students whose preferences may be quite different from successful students (see Rochford 2003).

On the other hand, the learning strategies indeed influenced on course achievement both in the whole data and in different learner groups constructed on the basis of variables measuring students’ cognitive and affective entry level in statistics. Overall, the learning strategies had more effect on learning outcomes in the group of students with favourable statistics background. Purposeful working as well as reflection and self-monitoring of one’s learning process are connected to good course results in this learner group. Despite, in the "unfavourable background" -group only purposeful working seemed to be a way to improve achievements although the effect is not as strong as in "favourable background" -group. On the other hand, rote learning and memorisation is the "best way" to weaken the results in statistics courses. In addition to this, the use of different learning strategies is more efficient among the students with favourable statistics background. This suggests that a student should have a firm background in prior knowledge in statistical and mathematical concepts introduced in previous statistics courses and high-school mathematics. Good mathematical and statistical self-confidence are also required to the efficient use of learning strategies. Efficient learning strategies connected with solid background and high self-confidence predict good course achievements.

Although there exists some evidence that graduate students who prefer to learn in co-operative learning groups tend to obtain lower levels of performance in research methodology courses than do their counterparts (Onwuegbuzie 2001), our results showed no effect for the learning strategy scale ‘Peer-Orientation’ on course success.

Although the measurements proved to be quite reliable in this study we should be careful in generalising the results due to a quite small sample of students and restriction of the study to one Faculty (Education) only. In addition, locating students to one learning style group may be questionable, particularly when they receive even scores on two or more style preferences. Therefore, these results are preliminary and should be confirmed in further research conducted also in other disciplines. Interesting would also be to examine the impact of cognitive styles (e.g., Wholist vs. Analytic information processing or Field-dependency vs. Field-independency) on learning outcomes in statistics because previous research has suggested that cognitive styles may be connected to learning outcomes and peoples’ occupation. For instance, Witkin and Goodenough’s (1977) review on a large body of data indicates that Field-dependent persons have an interpersonal orientation (teachers, for example), whereas the orientation of Field-independents is relatively impersonal.

To conclude, although our study indicated no differences in course success between the learning style groups, we can recommend to the teachers of statistics to design their training program to include auditory, visual, and kinesthetic learning which are the most basic learning styles. As to an auditory learner, (s)he must hear the facts and understand the logic of a concept. This type of individual absorbs words and sounds and delights in details, statistics, and facts. His notes usually are recorded in outline or narrative form. This person learns a great deal from audiotapes. A visual learner needs to see the big picture and won't listen if (s)he doesn't see the concept being presented. If you were to look at this individual's notepad, you would see her/his notes drawn as pictures to illustrate the concept. This type of learner benefits greatly from visual aids such as overhead transparencies, slides, and videos. Finally, a kinesthetic learner learns best by doing. Hands-on practice is the best way for this individual to master a concept. (S)he enjoys small group exercises and is generally a "people person." During breaks, this student moves freely around the room, meeting and speaking with the other students. This individual's notes will resemble a doodle pad. (See Peacock 2001.)


Batanero, C., Godino, J., Green, D. & Holmes, P. 1994. Errors and difficulties in understanding introductory statistical concepts. International Journal of Mathematical Education in Science and Technology 25 (4), 527–547.

Bell, J. A. 1998. Problems in statistics: Learning style, age, and part-time students. Education 118 (4), 526–528.

Birenbaum, M. & Eylath, S. 1994. Who is afraid of statistics? Correlates of statistics anxiety among students in educational sciences. Educational Research 36, 93–98.

Curry, L. 1991. Patterns of learning style across selected medical specialties. Educational Psychology 11 (3/4), 247–277.

Dillon, W.M. & Goldstein, M. 1984. Multivariate analysis. Methods and applications. New York. John Wiley & Sons

Doyle, M. S. & Garland, J. C. 2001. A course to teach cognitive and affective learning strategies to university students. Guidance & Counselling 16 (3), 86–91.

Dunn, R., Beaudry, J. & Klavas, A. 1989. Survey of research on learning styles. Educational Leadership, 46(6), 50–58.

Dunn, R., Dunn, K., & Price, G. 1989. Learning style inventory manual. Lawrence, KS: Price Systems.

Dunn, R., S. Griggs, J. Olson, M. Bassly, and B. Gorman. 1995. A meta-analytic validation of the Dunn and Dunn model of learning-style preferences. Journal of Educational Research 88 (6), 353–62.

Ediger, M. 1995. Teaching on the higher education level. College Student Journal 29, 308–311.

Ehrman, M. E. & Oxford, R. L. 1990. Adult language learning styles and strategies in an intensive training setting. Modern Language Journal, 74, 311–327.

Elmore, P. B., Lewis, E. L. & Bay, M. L. 1993. Statistics achievement: A function of attitudes and related experience. Paper presented at the annual meeting of the American Educational Research Association, Atlanta, GA.

Felder, R. M. 1995. Learning and teaching styles in foreign and second language education. Foreign Language Annals 28 (1), 21–31.

Forte, J. 1995. Teaching statistics without sadistics. Journal of Social work Education 31 (2), 204–218.

Gal, I. & Ginsburg, L. 1994. The role of beliefs and attitudes in learning statistics: Towards an assessment framework. Journal of Statistics Education (electronic journal) 2 (2).

Galagedera, D.1998. Is remedial mathematics a real remedy? Evidence from learning statistics at tertiary level. International Journal of Mathematical Education in Science & Technology 29 (4), 475 – 480.

Galagedera, D., Woodward, G. & Degamboda, S. 2000. An investigation of how perceptions of mathematics ability can affect introductory statistics performance. International Journal of Mathematical Education & Technology 31 (5), 679–689.

Garfield, J. 1997. Discussion about the paper of S.D. Moore "New pedagogy and new content". International Statistical Review 65 (2), 137–141.

Garfield, J. & Ahlgren, A. 1988 Difficulties in learning basic concepts in probability and statistics: implications for research. Journal for Research in Mathematics Education 19, 44–63.

Grigorenko, E. L. & Sternberg, R. J. 1995. Thinking styles. In D. Saklofske & M. Zeidner (Eds.), International handbook of personality and intelligence. New York: Plenum, 205–230.

Haapala, A., Pietarinen, J., Rautopuro, J. & Väisänen, P. 2002. How to overcome stumbling blocks in learning applied statistics – the effect of concept mapping. Education-line:

Kinsella, K. 1995. Understanding and empowering diverse learners in ESL classroom. In J. M. Reid (Ed.) Learning styles in the ESL/EFL classroom. Boston: Heinle & Heinle, 74–86.

McKeachie, W. J. 1996. Learning styles can become learning strategies. Teaching and Learning Exchange 3 (3), 10 –11.

Meichenbaum, D. 1977. Cognitive behaviour modification. New York: Plenum.

Messick, S. 1984. The nature of cognitive styles. Problems and promise in educational practice. Educational Psychologist 19 (2), 59–74.

Onwuegbuzie, A. J. 2001. Relationship Between Peer Orientation and Achievement in Cooperative Learning. Journal of Educational Research 94 (3), 164 – 170.

Onwuegbuzie, A. J. & Seaman, M. 1995. The effect of time and anxiety on statistics achievement. Journal of Experimental Psychology 63, 115–124.

Oxford, R. L. 1990. Language learning strategies: What every teacher should know. New York: Newbury House/Harper & Row.

Peacock, M. 2001. Match or mismatch? Learning styles and teaching styles in EFL. International Journal of Applied Linguistics 11 (1), 1–20.

Pietilä, A. 2002. Luokanopettajaopiskelijoiden matematiikkakuva (in english: Pre-service elementary teachers’ views of mathematics: The role of mathematics experiences in forming the views of mathematics. Doctoral dissertation. University of Helsinki. Faculty of Education. Department of Teacher Education. Research Reports 238.

Pintrich, P. R., Smith, D. A. F., Garcia, T. & McKeachie, W. J. 1991. A manual for the use of the Motivated Strategies for Learning Questionnaire. Ann Arbor, MI: National Center for Research to Improve Postsecondary Teaching and Learning.

Reid, J. M. 1987. The learning style preferences of Esl students. TESOL Quarterly 21 (1), 87–111.

Reid, J. M. 1995. (ed.) Learning styles in the ESL/EFL classroom. Boston: Heinle & Heinle.

Reid, R. 1996. Research in self-monitoring with students with learning disabilities: The present, the prospects, the pitfalls. Journal of Learning Disabilities, 29, 317–331.

Rochford, R. A. 2003. Assessing learning styles to improve the quality of performance of community college students in developmental writing programs: a pilot study. Community College Journal of Research & Practice 27( 8), 665–677.

Saracho, O. N. 1997. Teachers and students’ cognitive styles in early childhood education. Westport, CT: Greenwood Publishing Group, Inc.

Schmeck, R. 1998. Strategies and styles of learning: an integration of various perspectives. In R. Schmeck (Ed.) Learning strategies and learning styles. New York: Plenum Press, 317–348.

Schommer, M. 1994. Synthesizing epistemological belief research: tentative understandings and provocative confusions. Educational Psychological Review 6, 293–319.

Schutz, P. A., Drogosz, L. M., White, V. E. & Distefano, C. 1998. Prior knowledge, attitude, and strategy use in an introduction to statistics course. Learning and Individual Differences 10 (4), 291–307.

Svinicki, M. 1995. Using cognitive theories to improve teaching. The Teaching Professor 9 (4), 3–4.

Snee, R. D. 1993. What’s missing in statistics education? American Statistician 47 (2), 149–154.

Sternberg, R. J., Grigorenko, E. L., Ferrari, M. & Clinkenbeard, M. 1999. A triarchic analysis of an aptitude-treatment interaction. European Journal of Psychological Assessment 15, 1–11.

Sternberg, R. J. & Zhang, L. F. (Eds.). 2001. Perspectives on thinking, learning, and cognitive styles. Mahwah, New Jersey: Erlbaum.

Townsend, M., Moore, D., Tuck, B. & Wilton, K. 1998. Self-concept and anxiety in university students studying social science statistics within a co-operative learning structure. Educational Psychology 18 (1), 41–53.

Väisänen, P. & Ylönen, S. 2003. The impact of students’ real and perceived mathematics ability on their learning process and performance in introductory statistics. A paper presented at the 11th biennial ISATT conference, 27 June – 2 July 2003, Leiden, The Netherlands.

Witkin, H. & Goodenough, D. 1977. Field dependence and interpersonal behavior. Psychological Bulletin 84, 661–689.


This document was added to the Education-line database on 16 September 2003