Attitudes, Motivation, and Second Language Learning: A Meta‐Analysis of Studies Conducted by Gardner and Associates

The initial impetus for this investigation was a discussion between the two authors about comments and criticisms in the literature concerning the role of attitudes and motivation in second language learning. It was noted that some researchers seemed to feel that such variables were important, while others felt that they were not, and some even felt that the relationships reported in the literature were too inconsistent to draw any firm conclusions. The senior author suggested that someone should do a meta‐analysis to see exactly what the empirical data revealed and challenged the junior author to do so. In the following weeks, the enormity of the task became apparent. There were many articles that were identified with many different conceptualizations and measurement strategies, many different cultural settings, many different ages, and so forth, and it was obvious that a multitude of decisions would have to be made. To make the task a bit simpler, it was decided to start with research conducted by Gardner and associates simply because it was readily available and used a fairly standard set of concepts and measurement operations. Much to our surprise, there were more data here than we initially supposed, and in the end it seemed prudent to limit one study to this data set and plan a second meta‐analysis to consider research by other investigators. Such an approach, we felt, would permit greater flexibility than trying to organize all studies into one format. Meta‐analyses are enormous undertakings, however, and to date we have only managed to complete this one. The second study is underway, delayed somewhat by the vast distance now separating the two authors, let alone a number of logistical issues that make it difficult to identify a common set of constructs in the relevant literature. A major purpose of this investigation was to estimate the magnitude of the contributions that motivation and attitudes make to achievement in the second language in the research conducted by Gardner and his associates. This meta‐analysis investigates the relationship of second language achievement to the five attitude/motivation variables from Gardner's socioeducational model: integrativeness, attitudes toward the learning situation, motivation, integrative orientation, and instrumental orientation. Over the years, some interesting questions have been raised in the literature concerning the socioeducational model of second language acquisition and the role played by attitudes, motivation, and orientations in second language achievement. For example, some researchers have proposed specific hypotheses speculating whether the role of attitudes and motivation in language achievement may vary as a function of certain qualities of the learning situation (Is the language available outside of the classroom?), and /or qualities of the learner (Do attitudes and motivation play a stronger role depending upon the age of the learner?). We soon realized that these questions could be answered by this meta‐analysis, and thus the focus was expanded to include two questions: (a) What is the best estimate of the correlations in the population between various aspects of second language achievement and the five attitudinal/motivational characteristics in Gardner's model? and (b) Are there other variables, such as the availability of the language in the community or the age of the learners, that influence the magnitude of these associations? These relationships were examined in studies conducted by Gardner and associates using the Attitude/Motivation Test Battery, and three achievement measures including self‐ratings, objective tests, and grades. In total, the meta‐analysis examined 75 independent samples involving 10,489 individuals. The results demonstrated that the correlations between achievement and motivation are uniformly higher than those between achievement and integrativeness, attitudes toward the learning situation, integrative orientation, or instrumental orientation, and that clearly the population correlations are greater than 0. In general, neither the availability of the language nor age had clear moderating effects on these relationships. The present article first appeared in Language Learning, 53 (1), 2003, 123–163