Running Head: CORRELATION & MULTIPLE REGRESSION ASSIGNMENT

Correlation & Multiple Regression Assignment

Name of Student
Name of Course
Course Instructor
Date
 Correlation & Multiple Regression
Assignment

Part 1: Logical Arguments for Independent Variable Selection

The purpose of this assignment is to test the relational competence model of Spitzberg
and Hecht’s (1984) in a public speaking setting. In this part of the assignment, we make a
selection of different independent variables. The relational competence model posits that skills,
knowledge and motivation of the students impact on their interpersonal communication
competence satisfaction.

For the knowledge component, we select the ACT score and affective learning measure
has been chosen as the independent variables. The ACT score would reflect the performance of
each of the students in terms of their knowledge. Secondly, the affective learning measure would
represent the satisfaction of the students towards the fulfillment of the positive communication
expectations. Thus, these two have been selected as the measures of the knowledge component.

For the skills component, we have selected hours/wk completing homework and the
critical thinking measure variables as the independent variables. Both of these variables would
measure the skills level of the students. The critical thinking measure would allow the
informative speaker to talk consistently and develop new information ideas during the public
speaking session and the hours/wk completing homework is associated with the student’s ability
to complete their homework. The less time they take the higher is the level of their skills.

Lastly, for the motivation component we have chosen the previous public speaking
experience and communication apprehension measure as the independent variables. The
communication apprehension measure would determine the feelings of the students in the recent
communication events. The previous public speaking experience would also have a positive
influence on the motivation level of the students. Six independent variables have been selected to
predict the dependent outcome variable (first informative speech grade).

Part 2: Results of Multiple Regression Analysis

First, we have generated the correlation for all the independent and the dependent
variable. These are shown in table 1 below:

Correlations
Informati AC Critic Previou Communicatio
ve T Affective Hours/wk for al s n
Speech Sco Learning Homework Think Experie Apprehension
Grade re Measure Completion ing nce Measure
Pearso Informative 1.000 .01 .067 .264 .052 .049 .127
n Speech Grade 3
Correla ACT Score .013 1.0 -.114 .023 .131 -.017 -.010
tion 00
Affective .067 -.11 1.000 .161 .217 .204 -.213
Learning 4
Measure
Hours/wk for .264 .02 .161 1.000 .125 .097 -.042
Homework 3
Completion
Critical .052 .13 .217 .125 1.000 .326 -.399
Thinking 1
Previous .049 -.01 .204 .097 .326 1.000 -.433
Experience 7
Communicatio .127 -.01 -.213 -.042 -.399 -.433 1.000
n 0
Apprehension
Measure
Sig. (1- Informative . .40 .118 .000 .180 .192 .012
tailed) Speech Grade 9
ACT Score .409 . .022 .341 .010 .379 .426
Affective .118 .02 . .002 .000 .000 .000
Learning 2
Measure
Hours/wk for .000 .34 .002 . .013 .043 .227
Homework 1
Completion
Critical .180 .01 .000 .013 . .000 .000
Thinking 0
Previous .192 .37 .000 .043 .000 . .000
Experience 9
Communicatio .012 .42 .000 .227 .000 .000 .
n 6
Apprehension
 Measure

Table 1: Correlations

It can be seen that the correlation between the informative speech grades is significantly
positive with hours per week for homework completion and communication apprehension score.
There is also significant correlation among the independent variables such as between ACT score
and affective learning measure, affective learning measure and hours/wk homework completed,
critical thinking and hrs/wk homework completed and many others

The model summary for the multiple regression models is shown in table 2 below:

Model Summaryb
Adjuste
dR Std. Error of the
Model R R Square Square Estimate Durbin-Watson
1 .321a .103 .086 5.42733 1.669

Table 2: Model Summary

The correlation between all the independent and dependent variable is 32.1% and the
percentage variation in the informative speech grade is only 8.6%. In order to check whether the
model is fitted we need to look at the sig value of the F test in ANOVA table shown below:

ANOVAb
Sum of
Model Squares df Mean Square F Sig.
1 Regression 1047.169 6 174.528 5.925 .000a
Residual 9101.892 309 29.456
Total 10149.062 315

Table 3: ANOVA
 The p value is 0.000, which is lower than the significant value of 0.05, thus the overall
model is fitted. However, in order to determine the factors that significantly impact on the
informative speech grade of the students we need to look at the coefficients table 4 below.

Coefficientsa
Standardize
Unstandardized d
Coefficients Coefficients Collinearity Statistics
Std. Toleranc
Model B Error Beta t Sig. e VIF
1 (Constant) 73.904 4.11 17.974 .00
2 0
ACT Score .010 .080 .007 .123 .90 .959 1.043
2
Affective .282 .400 .040 .705 .48 .890 1.124
Learning 2
Measure
Hours/wk for .189 .042 .250 4.542 .00 .961 1.040
Homework 0
Completion
Critical .044 .041 .067 1.092 .27 .772 1.295
Thinking 6
Previous .344 .244 .086 1.406 .16 .773 1.293
Experience 1
Communicatio .072 .022 .211 3.338 .00 .729 1.371
n 1
Apprehension
Measure

Table 4: Coefficients

If we look at the p values of the individual independent variables from each of the three
components of knowledge, skills and motivation, then among all the six variables only hours/wk
homework completed and the communication apprehension measure impact on the students
informative speech grades significantly. The discussion and implications of these results are
performed in the next part.

Part 3: Discussion & Implications of the Results

 First, if we talk about the knowledge cluster, then neither ACT score nor affective
learning measure tends to impact significantly on the informative speech score of the students.
This means that the knowledge cannot be measured by the ACT score of the students. Students
with high ACT score might have lower level of knowledge. Similarly, the affective learning
measure would represent the satisfaction of the students towards the fulfillment of the positive
communication expectations however; this satisfaction does not necessarily mean a higher level
of knowledge of the students. Thus, both of them do not measure the knowledge cluster.

Second, if we talk about the skills cluster than hours/wk of homework completed only
impacts significantly on the informative speech score of the students. The less time they take the
higher is the level of their skills. Critical thinking can be good for writing abilities but not for the
communication of public speaking skills of the students. Thus, skills are measured only by the
hours taken per week for completing their homework.

Lastly, for the motivation measure only communication apprehension score has a
significant impact on the informative speech score of the students. The communication
apprehension measure would determine the feelings of the students in the recent communication
events. Positive feelings motivate the students to communicate well in their own public speaking.
Previous experience does not motivate the students in their current speeches and thus has
insignificant impact on the informative speech score.