Explore BrainMass
Share

Statistics: multilinear regression in SPSS

This content was STOLEN from BrainMass.com - View the original, and get the already-completed solution here!

Making Predictions Using Regression

Develop a hypothetical multiple regression (prediction) equation to predict something in your area of professional or personal interest. Please complete the following:

a. Identify the Variables: First you should identify the dependent (criterion) variable that you are interested in predicting. What variable do you plan to predict? Next, choose two variables (called independent) that you will use to predict your chosen dependent criterion variable. List and describe your two chosen independent predictor variables, and why you feel that they are appropriate for predicting your dependent variable. (Follow the example provided)

b. Quantify Your Predictor/Independent Variables: Fill out the following table to quantify each of your two independent or predictor variables. If you find that you have chosen a variable that cannot be quantified, select a different one and update your solution above. Fill out this table.
(Follow the example provided)

c. Correlation and Prediction: In order to use a certain variable to predict another variable, there must be a strong correlation between those two variables. Why is this true? Which of your two independent predictor variables do you think has the strongest correlation with your dependent criterion variable? Explain. (Follow the example provided)

d. Create the Prediction Multiple Regression Equation: Using your dependent criterion variable as "y," and your predictor independent variables as x1and x2, create a pretend multiple regression prediction equation. Next, choose any values for your independent variables x1and x2, and predict the corresponding value for the dependent variable "y." Use the standard equation Ŷ = a + b1x1 + b2x2. (Follow the example provided)

Follow this example
Help and Explanations
For example, I might be interested in predicting a person's BMI or Body Mass Index. Therefore, BMI would be my dependent criterion "y" variable that I will be predicting.

Next, I will choose two independent variables, (x1) hours of weekly exercise and (x2) ounces of meat eaten per week.

Predictor Name
Min value
Max value

1. hours of exercise
0
28

2. oz. meat eaten per week
0
84

Why did I choose these two predictor independent variables to predict my dependent variable of BMI?

In this case, I feel that each of these independent variables, exercise and meat are both highly correlated with BMI. If I had to guess, I would say that eating meat was the most correlated.

Next, I will create the multiple regression prediction equation, which will have the form:

Ŷ = a + b1x1 + b2x2

In my example, BMI is my "y" and is what I am predicting.

My x1 is an independent variable that represents the hours of weekly exercise and my x2 is an independent variable that represents ounces of meat eaten per week. The values of my constant "a," and my coefficients "b1" and "b2" are all numbers and are in fact weights for my independent variables.

For example, my regression prediction equation might be:
Ŷ = 25 - .3(x1) + .7(x2)

NOTE: The numbers you will use here are "invented." But, for the Assignment, you will use SPSS to generate the correct numbers.

Using my prediction equation:
Ŷ = 25 - .3(x1) + .7(x2)

I will use the following values
(1) The person exercises 6 hours per week (x1 = 6)
(2) The person eats 2 oz. of meat per week (x2 = 2)

Now, I will plug these values into my hypothetical regression equation to come up with the value of the BMI (dependent variable):

Ŷ = 25 - .3(6) + .7(2)

Now solving for y (BMI), I get a final value of: Ŷ = 24.6

In conclusion I predict a BMI of 24.6, assuming a person exercises 6 hours per week and eats 2 oz. of meat per week.

© BrainMass Inc. brainmass.com October 25, 2018, 9:37 am ad1c9bdddf
https://brainmass.com/statistics/correlation-and-regression-analysis/statistics-multilinear-regression-spss-580885

Attachments

Solution Preview

See attached

Making Predictions Using Regression

Develop a hypothetical multiple regression (prediction) equation to predict something in your area of professional or personal interest. Please complete the following:
I am the CEO of a manufacturing firm, and I want to see if my employees are correctly paid. What I want to check is whether the annual salary for my employees is based on performance ratings and years of service. In order to verify this, I am collecting the data shown in the table in SPSS for my employees.

a. Identify the Variables: First you should identify the dependent (criterion) variable that you are interested in predicting. What variable do you plan to predict? Next, choose two variables (called independent) that you will use to predict your chosen dependent criterion variable. List and describe your two chosen independent predictor variables, and why you feel that they are appropriate for predicting your dependent variable. (Follow the example provided)
We are trying to make a multilinear regression model, with one dependent variable and two independent variables:
• = the personnel's salary - the dependent variable
• = average performance over the past 3 years rating ...

Solution Summary

Predictions Using Regression

$2.19
See Also This Related BrainMass Solution

Research Analysis Set of Hypothesis

Utilizing any available disclosed database for SPSS, develop a researchable set of hypotheses related to the database. Clearly define quantitatively analyzable hypotheses, analyze your data with SPSS, and write up the results in a full quantitative report format. Draw appropriate conclusions about your hypotheses in your writeup. This assignment does not need to be related to any earlier individual projects. This is a standalone assignment. Your paper should include:
Title Page
Abstract
Brief introduction to the research problem or problems - State your research questions and hypotheses within this section but you do not have to develop a literature review.
Methods - You will not generally have access to the methodology used to acquire the data you have chosen but should use your early SPSS programs to determine a variety of information about your subjects. Explain the details of the variables you have selected to examine. Indicate the statistical test or tests you used to make your inferences about your hypotheses.
Results - Include key graphs from your SPSS output. Follow APA format when citing statistical results and presenting figures or tables.
Discussion and Conclusion
References
Make sure you submit your SPSS output file in addition to the report in a Word document.

View Full Posting Details