1.Read the assessment brief and rubric.
2.Read the article provided.
3.Answer the three (3) questions.
4.You will be required to locate two (2) studies to support your responses to the questions.
5.You will be peer-reviewed using the rubric to evaluate your response according to the criteria and standards.
Instructions:
Requirements:
Read the following article:
Bono, JE & McNamara, G 2011, ‘Publishing in AMJ—part 2: Research design’, Academy of Management Journal, vol. 54, no. 4, pp. 657-660.
Respond to the following three discussion questions. You should commit approximately the same word count to each question.
Question 1:
“Type of Data”
What is the difference between cross sectional data and panel data? What type of data does AMJ recommend and why? Give an example of a peer reviewed article that addresses a change question with balanced panel data.
Question 2:
“Variables”
Explain why it is necessary for researchers to use control variables. What are the three conditions needed for a control variable to be included in a study?
Question 3:
“Mediator Variable”
What is a Mediator variable? Explain and give an example of a study which uses a mediator variable to explain a relationship between a dependent and independent variable.
1.This is an individual assessment task that accounts for 40% of your total grade.
2.Your written submission should not exceed 3,000 words and must be submitted to Turnitin within the Canvas BUSM4448 course site.
3.You must include a correctly formatted Reference List, which adheres to the Harvard Style of referencing. The reference list does not count toward the word count.
4.Requests for an extension to the deadline must be accompanied with documentary evidence (i.e. medical certificate) to support the request. A decision on any request for an extension is at the discretion of the Course Coordinator.
5.Late submissions will attract a penalty of 10% per day.
Further Notes
Question 1:
“Type of Data”
What is the difference between cross sectional data and panel data?
What type of data does AMJ recommend and why?
Give an example of a peer reviewed article that addresses a change question with balanced panel data.
Question 2:
“Variables”
Explain why it is necessary for researchers to use control variables.
What are the three conditions needed for a control variable to be included in a study?
Question 3:
“Mediator Variable”
What is a Mediator variable?
Explain and give an example of a study which uses a mediator variable to explain a relationship between a dependent and independent variable.
Question 3 Sample response
A mediator variable can explain why a set of variables are related, especially, the relationship between a dependent and interdependent variable (Bono & McNamara 2011). While a new area of inquiry may initially focus on establishing a causal link between two variables, Bono and McNamara (2011) claim that once an association is understood, it is then necessary to define the extent to which one variable influences another. The more mediators that are known to apply to an area of enquiry, the more multiple mediators may need to be considered.
For example, XX study of XX uses a mediator variable XX to explain the relationship between XX as a dependent variable and XX as an independent variable.
Question 2 Sample Response
The proper inclusion of key control variables allows more complete conclusions to be drawn by researchers. Bono and McNamara (2011) argue that it is not practical to include all control variables but that key controls should be included in the design stage and they should be determined by three conditions established by Becker (2005) and James (1980).
1.Preceding research or a clear theoretical link connects the dependent variable with the control variable.
2.The hypothesised independent variable correlates with the control variable.
3.The control variable is not the central variable in the study, either hypothesised or as a mediator.
Failure to address these three conditions could result in ‘omitted variable bias’, superfluous variables or bias related to hypothesised variables.
Question 1 Sample Response
Bono & McNamara (2011) recommend the use of panel data, or longitudinal or experimental data, over cross-sectional data when a research question needs to address change or causality. This is because causality or change research questions usually include more than a single instance measurement or attempts to explain how change in one variable influences other variables (Bono & McNamara 2011). AMJ therefore, argues that using cross-sectional data for any research question implying change or causality between variables may not be ‘capable of testing the question of interest’ (Bono & McNamara 2011, p. 658) or the answer may be skewed. This further highlights the need for the research design to match the research question.
The following study by XX demonstrates how balanced panel data has been used to describe a change question.
