To assess the effect of the different variables of theoretical and practical interest to second language research, 920 Native Mexican children, aged five through twelve, learning Spanish as a second language were tested on a Spanish language achievement test. The variables measured for investigation were: order of presentation of the different parts of the test; age, sex and native languages of the children; whether the father, mother and siblings speak Spanish; proximity of the domicile to a Spanish speaking community; amount of attendance at school; and whether the parents needed to speak Spanish to travel for work. The variables which had a statistically significant effect on test scores were: order of presentation of the parts of the test; native language of the children; father speaks Spanish; attendance at school; and need for the parents to speak Spanish to travel for work. All other variables in this study were not statistically significant. Discussion of the results includes possible reasons for the findings.

No abstract

Operational definitions of the differences between “achievement” and “proficiency” in foreign language performances are given. The development of different methods of teaching foreign languages and the subsequent tests related to each method are reviewed and discussed, current attempts to measure “proficiency” in foreign languages are seriously questioned as being inadequate. To eliminate some of the most glaring inadequacies in language testing, new, experimental approaches are being conducted by investigators from different disciplines. The new approaches are reviewed, the underlying rationale for each approach is discussed, and a plea is made for increased interdisciplinary experimentation in all areas of language testing.

No abstract