The assignment instructions are as follows:
1. Identify all the variables necessary to answer your research questions (remember which one is dependent (effect), which is independent (cause), also you might have two or more independent variables).
2. When you are building your research question, you need imply cause-effect relationships. What is the effect/impact of …….. on……..? How …. will affect to ……? If you didn’t make correction for your research questions, you have to do to build your model and to show your variables.
3. Some of you will only have one item/question to measure dependent variable, but some of you may have more if you are creating composite measures or scales. Specify whether the measure is a dependent or an independent variable.
Give a brief definition of the variables (concepts) you are going to use.
Define your variables in full sentences. Focus on how variables were defined in literature.
A brief statement explaining why the variable is relevant to the study. What are the indicators and dimensions (aspects) of your concepts? (Review chapter 6, and lecture)
Don’t forget to specify whether the measure is a dependent or an independent variable.
Provide the operational definitions of the variables, meaning the exact questions from your questionnaire and their answer choices. Write in full sentences (how you define your variables and how you measure them).
4. Identify control variables that could be used to ensure causality in your relationship. Often used as controls are gender, race, age, industry type, firm size, region of the country, income, education, or something more specific to your study.
It may help to see what previous literature has used for control variables. If we did not include control variables, we could end up with spurious relationships, like when size of the company and revenue.
In other words, you need to identify other variables that could possibly be influencing your dependent variable, other than your independent variable. You need to use “control” variables in your study, making sure that the relationship you are testing is not spurious.