Welcome to our community

Be a part of something great, join today!

Levels of Measurement, Basic OLS Regression Questions

lomaxbridges

New member
Apr 17, 2019
2
Hello soon to be saviors, 😊


I have two really simple questions that I have already answered but the teacher wants more info. I am really stumped and I am not looking for the answers so much as an explanation on how to better answer the questions. I will copy and paste the problems and my answers (in bold black) as well as their comments (in red) below. I also have attached a pic of the ols regression table I am working from. ANY help or comment is welcome. Please help!


Robin Anderson

OLS Regression.jpg

Examine the variables inModel 5, Table 2:

Identify the variables by name and specify the level of measurement (nominal, ordinal, interval, ratio) under each of the categories below.

You have several variables that are under Independent and Control-please choose one or the other or explain why you believe we could consider a variable either or both. Note that Continuous is not synonymous with interval/ratio.


Independent Variable(s):
Name Level of Measurement

Course Type Nominal
Student GPA Numeric (continuous)
Year in school (senior) Nominal
Online courses taken Numeric (continuous)discrete
Like working with others Ordinal
Instructor interaction Ordinal
Gender (male) Nominal
Credit hours Taken Numeric (continuous)
Hours worked Per Week Numeric (continuous)

Dependent Variable:
Name Level of Measurement

Student scores Numeric (continuous)

Control Variable(s):

Name Level of Measurement

Student GPA Numeric (continuous)
Year in school (senior) Nominal
Online courses taken Numeric (continuous)
Like working with others Ordinal
Instructor interaction Ordinal

Yes, but note we are treating them as interval/ratio in the analysis

Gender (male) Nominal
Credit hours taken Numeric (continuous)discrete
Hours worked per week Numeric (continuous)


In a couple sentences describe how you differentiated between the IV’s and CV’s.



Independent variables are the variables that are changed or controlled during the experiment.

In general, this is correct but note that this isn’t an experiment where researchers manipulate the variable. It is a quasi-experiment. These variables vary across individuals but not because of anything the researcher does. He or she does not assign the students to a classroom or online. All of the predictor variables vary across individuals, not just the IVs. So you need another way to evaluate what the IVs are in this case.



10. Summarize what the results presented in Table 2, model 5 suggest about the association between course delivery method and student performance? What additional insights can we draw about this relationship when we compare Model 5 with Model 1?Explain your answer.

In model 5, the p-value for course type is not significant after adjusting the controlled variables which suggest that there is no significant association between course type and student scores. Student GPA and Instructor interaction are the only two independent variables that are associated with the student scores.

Be more specific about what it means when a coefficient on a variable is reduce (in this case to non-significance) when another variable(s) is included in subsequent models. Review your notes on mediation and apply to the variables in the article.
 

lomaxbridges

New member
Apr 17, 2019
2
I figured out the first part but I still need help with number 10 if anyone is there lol