The Power of Prediction

The Power of Prediction

Linear Regression In Action

Critical Research Points

Conclusion

What It All Means...

Beveridge, S., & Fabian, E. (2007) Vocational rehabilitation outcomes: Relationship

between individualized plan for employment goals and employment outcomes.

Rehabilitation Counseling Bulletin, 50(

4), 238-246.

Salkind, N. J. (2011).

Statistics for people who (think they) hate statistics.

Thousand Oaks, CA: Sage Publications, Inc.

Tabachnick, B. & Fidell, L. (1989).

Using multivariate statistics.

(2nd edition).

New York: HarperCollins

**Rose Willis**

Cheryse Jackson

Brittney Sears

Maristella Vismele

Charles Carter

Cheryse Jackson

Brittney Sears

Maristella Vismele

Charles Carter

Special Thanks To Our Very Own, Dr. Kri Watson

Description

Given data on relationships between specific X (independent/predictor) variables and Y (dependent) variables, LINEAR REGRESSION allows for the prediction of an unknown Y based on a known X.

Formula

Selected Research Article:

Vocational Rehabilitation Outcomes:

Relationship between Individualized Plan for

Employment Goals and Employment Outcomes

Examine the relationship between attaining a job congruent with the Individualized Plan for Employment (IPE) vocational rehabilitation goal and vocational rehabilitation outcomes.

Purpose of the Study

Independent Variables:

Disability category

Gender

Race

Age

Job satisfaction

Dependent Variable:

Congruent employment

Sample Size:

171 vocational rehabilitation clients served by the Maryland

State Department of Education Division of Rehabilitation

LinearFormula:

Y= variable to predict

X= variable used to predict Y

a= the intercept

b= slope

Attempts to determine the strength of the relationship between one dependent variable (usually Y) and a series of other changing variables (known as independent variables).

Regression

Inferential Statistics:

Y=bX+a

• Regression analysis is used when

you want to predict a continuous dependent variable from a number of independent variables.

• Casual relationships among the variables cannot be determined. While the terminology is such that we say X "predicts" Y, we cannot say that X "causes" Y (Tabachnick & Fidell, 1989).

• Regression analysis is used with naturally-occurring variables, as

opposed to experimentally

manipulated variables.

Research Question

For those clients who obtained a successful employment outcome congruent with their vocational rehabilitation goal, did their weekly wages differ significantly from those of clients who did not obtain an outcome congruent with the vocational goal?

There is a positive relationship between weekly wages and obtaining an employment

outcome congruent to the

vocational rehabilitation goal.

Hypothesis

Regression Process:

STEP 1: Make a scatter plot

STEP 2: Conduct a regression analysis to visually assess if there is a linear relationship between variables

STEP 3: Draw a regression line (diagonal) on scatter plot, noting degree of error

STEP 4: Interpret analysis

Disability

Gender

Race

Age

Y=bX+a

Y= variable to predict

X= variable used to predict Y

a= the intercept

b= slope

Independent Variables:

Dependent Variable:

DORS- Congruent Employment

*Table 3 identifies values of Y regressed variables (Beveride & Fabian, 2007).

**Linear Regression:**

**Introduction to**

Linear Regression

Linear Regression

Conclusion Cont.

• The analysis showed that persons with a physical disability

earned higher weekly wages ($539.29) than participants with

sensory ($463.47) and mental disabilities ($352.75).

• The other independent variables in the weekly wages linear

regression model were not statistically significant; however,

race and gender did affect the participant’s weekly wages

(Beveridge & Fabian, 2007).

o In terms of:

Race: Caucasians earned $470.65 a week, compared

to weekly earnings of $394.21 for African Americans.

Gender: Men earned $452.43 a week, and

women earned $429.26.

References

• The purpose of the research question was to determine

whether obtaining a vocational rehabilitation employment outcome congruent with the vocational rehabilitation goal increased the participant’s weekly wages.

• The data shows a positive relationship between weekly

wages and obtaining an employment outcome congruent

with the vocational rehabilitation goal.

• Participants who obtained a congruent VR employment outcome earned a higher weekly wage than those not

congruent.

• The education variable indicated, the more education

obtained increased the participant’s weekly wages. (Beveridge & Fabian, 2007)

• The independent variables used in regression can be either continuous

or dichotomous (Tabachnick & Fidell, 1989).

• Independent variables with more than two levels can also be used in regression analysis, but they first

must be converted into variables that have only two levels.

Analysis

Analysis Cont.

Analysis Cont.

Questions

Comments

Table 5 identifies the impact of X predictor/IV (which is employment/wages) on Y/DV, which is congruent employment (Beveridge, S. & Fabian, 2001.

(Salkind, 2011).