<p> It has perhaps never been clearer that in order to effectively communicate with global governments and develop reasoned foreign policy, the United States Intelligence Community requires the support of trained linguists. The development of foreign language proficiency is a complex process requiring a significant investment of time and resources. For learners involved in intensive foreign language training within the United States Government (USG), the Department of Defense (DoD) has developed various Language Difficulty Categorization (LDC) frameworks aimed at standardizing the amount of time learners are given to meet established proficiency criteria. Despite the widespread adoption of LDC frameworks over the past 60 years, few empirical studies have examined the systematicity in proficiency patterns for languages grouped within the same difficulty category. By situating the analysis within the framework of a logic model, data-mining techniques were used to statistically model, via path analysis, the relationships between program inputs, activities, and outcomes. </p><p> Two main studies comprised the investigation. Study 1 employed a contrastive-analytic approach to examine the coherence with which both cognitive (e.g., general aptitude, language-specific aptitude, and average coursework outcomes) and non-cognitive (e.g., language preference self-assessment scores) variables contributed to the development of foreign language achievement and proficiency outcomes for three languages grouped within the same category. For Study 1, only learners who completed the entire foreign language-training program were included in the analysis. Results of Study 1 found a great deal of coherence in the role that language-specific aptitude and 300-level average coursework grades play in predicting end-of-program proficiency outcomes. To examine the potential hidden effects of non-random attrition, Study 2 followed the same methodological procedures as Study 1, but it imputed missing coursework and proficiency test score data for learners who attrited (that is, “dropped out”) during the intensive foreign language-training program. Results of the imputation procedure confirmed that language-specific aptitude plays a robust role in predicting average coursework outcomes across languages. Study 2 also revealed substantial differences in the role that cognitive and non-cognitive variables play in predicting end-of-program proficiency outcomes between the observed and imputed datasets as well as across languages and skills.</p><p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:10790201 |
| Date | 28 July 2018 |
| Creators | Masters, Megan Christina |
| Publisher | University of Maryland, College Park |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.0018 seconds