Zachary Duvall and Dr. Laura Smith, German and Slavic Studies

With increasingly difficult economic circumstances, ever more evidence is needed to validate the costs of study abroad. This study iss part of a larger project conducted by my mentor, Dr. Laura Smith and her colleague Dr. Kelling to assess linguistic and cultural changes by students during study abroad.

While immersion programs like study abroad are assumed to be the best means of acquiring a language, less is known about how students actually learn the grammar of a language while abroad and whether students make any improvements at all.

In light of the culture of assessment at universities across the U.S., this project will shed light onto the student language learning experience on German study abroad. I analyzed how students make grammatical progress on study abroad programs and whether students took more risks in attempting difficult language structures by the end of an in-country program. Answers to these questions were intended to help students, faculty and administration alike determine what the benefits of study abroad are, potentially lending support to the department’s goal to encourage students to have an experience abroad.

To carry out the study we collected two timed writing samples from each of the 33 participants— initially about one week before departure and finally during the last week of their stay in Berlin, which included living with native German host families and having formal German language instruction. We typed and digitized the pre- and post-test writing samples including all spelling and grammatical errors. Subsequently, we identified and defined grammatical functions to target in the analysis. After reading and discussing the samples, we selected 61 grammar functions divided into seven categories: adjectives, nouns/pronouns/articles, prepositions, spelling, verbs, syntax/style, and word choice.

We then counted and categorized all grammatical errors from the writing samples. To score the subjective nature of and account for some overlap within the functions, we created a 17-page scoring guide defining every possible “mistake” and “attempt”. We also measured average sentence length, total number of words in each writing sample, and total number of sentences as a base measure. Percentages of errors were calculated as a function of both attempts at a particular grammatical structure and attempts are compared with word counts for each writing sample.

We ultimately ran several statistical analyses to compare the scores, errors, and attempts from the pre- and post-tests as a function of overall word counts in each sample to determine what if any changes students have made in their grammatical competence and risk taking. We hypothesized that the analysis would show significant progress in the ratio of attempts to errors. Our hypothesis also posited that in case the findings differ from expectations, it will be due to students making more attempts at complex grammar functions and longer answers thereby increasing the possibility for errors.

The hypotheses proved true for all three levels – beginning, intermediate and advanced, but most notably in the beginning, secondly intermediate and only somewhat in the advanced. We were able to present the results at the 2011 GLAC (German Linguistics Annual Conference), where it was well received and where we were given good feedback on how to further the research. Using their suggestions and some other of our ideas, the study will continue and be expanded in the future.