I have to develop a research design for a pretend study. It will be quantitative to study the effects of linguistic modification on nursing exams for one year. The original exam and the modified exam will be randomly assigned to students (native English speakers and English as a second language students). There would be scores from four groups to compare:
native English speakers recieving the original exam
English as a second language receiving the original exam
native English speakers recieving the modified exam
English as a second language recieving the modified exam
I have no idea what kind of quantitative analysis to use for this. Would this be correlational? If you have a better idea please tell me as I am stuck.
Assessment strategies in nursing education need to respect the diverse student population. Multiple-choice examinations are a popular assessment tool because they reflect the type of questions that make up the national licensure exam. However, small qualitative studies have shown that linguistic bias is a determent to English as a second language (ESL) and culturally diverse students' success in nursing programs and in passing the national licensure examination. This study will quantitatively determine if there is a difference between successes on linguistic modified exams and ESL students and the non-modified exams. The study will also show whether the native English speaking (NS) fairs better on the modified exam.
One exam will be the regular exam given in class
the other will be "modified" so there is no language bias in the exam
Example: regulare exam might say "the patient was drinking gatoraid for his clear liquid diet" and ESL and culturally diverse students might not know what gatoraid is
whereas a native who grew up here would know gatoraid is a clear liquid.
I just provide with you some of my thoughts. It might help you.
First of all, it seems to me that you cannot conduct a correlation analysis. It looks like the NS students would perform equally on both the modified exam and non-modified exam (since they have no problems with language), while ESL students would perform differently on the two type of exams.
Secondly, I do not think you can just do one single quantitative analysis. However, you may conduct several hypothesis tests. For instance,
1) Determine if there is a difference between successes on ...
The solution determines whether correlational study is a good choice.