Cathy's McNamara Abedi

McNamara, Chapter 2 “Communication and the design of language tests”

This chapter describes the activities of designing tests and making meaning of test scores particularly in language and language use. It goes on to talk about discrete point testing or individual points of knowledge and the evolvement of psychometrics that have to do with the measurement of cognitive abilities leading up to reliability. Test writers of the earlier days, l960’s, influenced today’s test writers. Integration of knowledge refers to the relevant systematic features of language. Integrative tests soon became important features of language testing such as in oral interviews. Later a new view emerged under the guidance of John Oller, in the l970’s, proposed the Unitary Competence Hypothesis, also known as pragmatic test. However, cloze tests proved to be similar to discrete point testing. Hyme’s theory of communicative competence, especially in the social structure, showed there was more to knowing a language. Cultural aspects play an important role during the process. Out of Hyme’s theory came performance tests in language and close attention to social roles in real settings. Test materials soon came to reflect communication in settings such as hospital or government run meetings.

This particular reading assignment helped me to see how testing of language and language use has evolved over time. At the same time I’m thinking about our Yup’ik community and how children learn a language. I remember how my mother-in-law, as I was learning Nelson Island dialect, would sometimes do a recast where she would rephrase what I was trying to say. Her nature of delivery and tone of voice was such that I never felt threatened. I knew that she understood what I meant and I would replay the conversation in my mind over and make a mental distinction between the word or phrase and how it’s said where I grew up. Other times, I make sure I remember the phrase so I don’t make the same mistake.


Abedi – “The No Child Left Behind Act and English Learners: Assessment and Accountability Issues”

This article describes issues that concern No Child Left Behind (NCLB) Act and how limited English proficiency (LEP) are affected. It describes six LEP assessment issues in relation to AYP reporting. The issues include 1) Inconsistency in LEP classification across and within states, 2) Sparse LEP population, 3) Lack of LEP subgroup stability, 4) Measurement quality of AYP instruments for LEP students, 5) LEP baseline scores, and 6) LEP cutoff points. The chapter goes on to discuss the disaggregated progress reports by subgroup and NCLB requirements as well consequences for schools enrolling LEP students. Its stated “LEP students exhibit substantially lower performance than non-LEP students in subject areas high in language demand.” (Abedi, p. 10-11) Suggestions as to and effective education system for LEP students are given. The three interactive components are 1) classification, 2) instruction, and 3) assessment. The three have in common the language factors or barriers. Interactive model is explained and described in the chapter.

It never occurred to me how complex LEP really was until I read this article. There are so many issues related in and around LEP, NCLB, and AYP. I thought it was interesting to know that LEP, across the country, can have a broad definition. And there are factors, both common and uncommon, that affect the LEP students all over. It’s not surprising, though, especially in knowing the diversity of our regions, our cultures, and our languages. I also found it interesting to know that “…academic achievement tests that are constructed and normed for native English speakers have lower reliability and validity for LEP populations (Abedi, Leon, & Mirocha, 2003).” (Abedi, 2003) If I understand this statement correctly, it means that the academic achievement tests do not align with LEP students. That brings a lot of questions to mind. One of them being, then why are the LEP students tested using the same forms?

I had a question about baseline scores. What do they mean?