Language Learners in Spain: The Influence of Home and State Languages on Assessment Performance

• In Event: Validity and Fairness Issues in Assessing Second-Language Test-Takers

Abstract

In many countries, including Spain, a large proportion of students with an immigrant background speak a language at home different from the language of instruction in school and in which they are assessed by national and international agencies. These Language Learner (LL) students face serious obstacles to succeeding in school. On average, across OECD countries, those students underperformed students without an immigrant background by 57 points in the 2009 edition of PISA. If LLs happen to live in an area where they are exposed to two different languages, those difficulties can be considerably greater.

That is the situation in Spain, where there are several states with their own language, and different linguistic models concerning student instruction. In some states, students are taught mostly in their own language; in some others, both languages can be used. Consequently, the scenario is different regarding their participation in educational assessment programs, with states where students can answer in any of the two languages, and others administering the tests exclusively in the state language. In addition, among the native students in the states with their own language, there are some students, ranging from 27 to 57%, that do not speak at home the state language used in the aforementioned PISA edition. Those students are known as Potential LLs (PLLs).

In PISA 2009, 14 out of 17 Spanish states oversampled students, four with a state language. The results showed an increased likelihood of scoring in the bottom quarter of the reading performance distribution for immigrant students speaking another language at home. The differences in performance between native students speaking the language of assessment at home and native students speaking another language were null. However, a closer look at the performance of students in states with their own language shows that these differences amount up to 30 score points in some states.

The goal of this study is to examine the performance of LLs in the different scenarios that the Spanish context affords: the traditional Spanish LLs (SLLs) living in a state where the only language spoken is Spanish, the SLLs in states with a state language, and the PLLs among the non-immigrant students. To do so, extensive attention will be paid to (1) the distribution of the estimates of the reading proficiency of those LLs, namely the mean and standard deviation of those estimates, the gap between the highest and lowest achievers, the performance by proficiency level, and the resilience rate, and (2) the relationship of these estimates to different variables, since the results show that performance differences remain, in Spain and other Mediterranean countries, after accounting for differences in social, economic and cultural status.

This study provides evidence that can be used to construct a validity argument for test scores from these increasingly influential educational assessments. It also provides evidence that may help to resolve the current heated debate in Spain about the effectiveness of different linguistic models in the states with their own language.

Author

• MARIA JOSE NAVAS, Universidad Nacional de Educacion a Distancia