Wednesday, July 11, 2007

Experimental Research Question


I .Potential experimental research question

Can increasing the number of quizzes help students of secondary school X to achieve better performance in English language studies?

The above question can be classified as Experimental Research, as it is a statement of hypothesis on the casual relationship between 2 variables:

1. number of quizzes done by students; and

2. the performance of students in English language studies (as measured by the score in year-end final examination).

II. How could data for the variables be collected?

Step 1: Each grade is treated as an independent population. During the second semester, every two weeks, a fraction (say, 50%) of the students are randomly selected to take part in a quiz on English language studies. Therefore, by the end of the semester, a student may have taken at most 20 tests, or at least, no test at all. Each student represents a data point.

Step 2: Measurement: Assume the score of final examination can measure a student's performance in English language studies.

Step 3: Computation of correlation coefficients: The 2 variables [number ofQuizzes taken, score of final examination] are used to compute the correlation coefficient for each grade. The closer the correlation coefficient to +1, the higher the strength of the relationship between 2 variables.

III. Provide argument how answer to that question can contribute to improvements in some aspect of education/teaching and learning?

If positive correlation (>0.5) is found, school may consider to employ the use of a large number of quizzes to improve the performance of students in English language studies.

Posted by Cherry

No comments: