Conducted by doctoral students, the study analyzed 3-minute tapes of conversation between 16 Title VII first grade Mexican American students and teacher aides in the Weslaco Independent School District, Texas, to determine the percentage of standard vs. non-standard language, measure the percentage of code switching by aide, show the percentage of language interference by students and aides, and determine whether an adequate questioning technique was used. Usng the Linguistic Analysis Inventory Chart and Question Asking Skills Inventory, a count of errors made by aides and pupils was determined based on the "total words spoken" and "total sentences spoken," a ratio of teacher-error vs. child-error was determined, and a percentage of each category of questions asked (cognitive, convergent, divergent, and evaluative) was determined to show the strengths or weaknesses in overall questioning during an oral language development lesson. Findings indicated that: students were weak while aides were strong in their structural grammar based on a standard vs. non-standard comparison; the percentage of code switching by aides (6%) was not significant; aides spoke 59% of the time while students spoke 41% of the time; the students' and aides' percent of errors were not significant; and 88% of the questions asked by aides were at the recall level. (NQA)