Sixth session

Section 1: Introduction

Research questions

Look at these research questions:
1. How can X be used to teach Y?
2. Does X influence Y?
To answer the first research question, you conduct a descriptive qualitative study because you want to describe the use of X in the teaching of Y. To answer the second research question, you conduct a quantitative study, i.e. experimental research. It is possible to conduct a mixed-method study like the above, but it better for a beginning researcher to focus on one research methodology.

One of you states two research questions but there is only one objective of the study. If you state two research questions, there should be two objectives of the study.

Definition of key terms
Please do NOT use a dictionary to define the terms in section 1. You should write an operational definition of the terms. You can read what an operational definition is here. You should:
avoid using a dictionary to define the terms. The definitions in the dictionaries are written for laypeople rather than researchers. Read a lot of books and journals to find out how experts and researchers define those terms.


Section 3: Methodology

Instruments

Many students include ESL Composition Profile developed by Jacobs et al as one of the research instruments in their proposal. This is simply WRONG. The ESL Composition Profile contains the criteria of scoring essays, and it is not the tool that you use to collect data. The instrument that you use to collect the numeric data in the form of scores is of course a WRITING TEST. Please read this.

A checklist that you use in observation results in some data in the form of numbers, so you cannot use a checklist in a qualitative study. Instead, you should use field notes to obtain data in the form of words.

Sampling

If you conduct experimental research, you usually do not use simple random sampling. It is more complicated if you use simple random sampling because you select the potential subjects one by one from several classes. As a result, the experimental group may consist of students from two or more classes and it will be harder for you to group them into a single class due to time constraints. The same applies to the control group. The more appropriate type of sampling is cluster random sampling, in which you list the classes and then draw two classes as the sample rather than drawing individual students for this purpose.

After you draw two classes as the sample, many of you who are going to conduct experimental research do not explain how you are going to assign the two classes into an experimental group and a control group. Please explain random sampling AND random assignment.