Scribd
Upload
15 views4 pages

Multiple Linear Regression

The document presents a regression analysis examining the relationship between experience, gender, and job performance, indicating a strong correlation (Multiple R = 0.74). It reveals that 47.94% of the variation in job performance can be explained by these factors, with significance F showing a highly significant influence. The regression model suggests that job performance decreases by 1.017 for every unit increase in experience, with different baseline performances for male and female employees.

Uploaded by

ranamaglacion
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views4 pages

Multiple Linear Regression

The document presents a regression analysis examining the relationship between experience, gender, and job performance, indicating a strong correlation (Multiple R = 0.74). It reveals that 47.94% of the variation in job performance can be explained by these factors, with significance F showing a highly significant influence. The regression model suggests that job performance decreases by 1.017 for every unit increase in experience, with different baseline performances for male and female employees.

Uploaded by

ranamaglacion
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Respondents Experience Gender Job Performance

1 2 0 2.54 Experience Gender


2 1 0 3.57 Experience 1
3 1 1 2.51 Gender 0.034815531 1
4 2 0 2.12
5 2 1 2.43 Gender Dummy variable: k-1
6 1 0 3.11 0 male
7 2 0 2.06 1 female
8 1 1 3.63
9 1 0 2.93
10 2 1 1.57
11 1 0 2.91 SUMMARY OUTPUT Multiple R indicates the correlation of ge
12 1 0 2.31 The Multiple R is 0.74, therefore, gender
13 1 1 2.46 Regression Statistics
14 2 0 1.68 Multiple R 0.740810363
15 1 0 3.91 R Square 0.548799994 will be used for Simple Linear Regression (only 1 indep
16 1 0 3.57 Adjusted R Square 0.479384608 will be used for MLR. Although the correlation of expe
Standard Error 0.510704824 values, so, that's why the R square is adj
Observations 16 Interpretation: 47.94% of the total variat
r value: 52.06% of the total variation in job perfo
ANOVA
0 – 0.19 – very weak relationship df SS MS F
0.20 – 0.39 – weak relationship Regression 2 4.12409132 2.0620457 7.90602817
0.40 – 0.59 – moderate relationship Residual 13 3.39065243 0.2608194
0.60 – 0.75 – strong relationship Total 15 7.51474375
0.80 – 1.00 – very strong relationship
Coefficients Standard Error t Stat P-value
Intercept 4.177961165 0.3914074 10.674201 8.3967E-08
X1 Experience -1.016504854 0.26388682 -3.852049 0.00199975
Dummy Variable D1 Gender -0.234854369 0.27562079 -0.852092 0.4095863

Regression model:
icates the correlation of gender and experience towards job performance.
R is 0.74, therefore, gender and experience has a strong relationship to job performance.

ar Regression (only 1 independent variable)


gh the correlation of experience & gender is small (r = 0.035), there will still be possible effects of their
t's why the R square is adjusted. So, we use Adjusted R Square for MLR.
: 47.94% of the total variation in job performance can be explained by gender and experience.
total variation in job performance can be explained by other variables.

Significance F
0.005667599 Significance F is 0.0057 which is less than .01, so the influence of gender and
experience to job performance is highly significant. (if < .05, significant)
If the Significance F is not significant (greater than .05), there is no multiple
linear regression model. So, test separately (SLR).
Lower 95% Upper 95%Lower 95.0% Upper 95.0%
3.33237688 5.023545 3.332377 5.023545
-1.586597663 -0.44641 -1.5866 -0.44641
-0.830296888 0.360588 -0.8303 0.360588
ŷ = 4.178-1.017X1-0.235D1

Remember that gender is a dummy variable.


Your scoring for male is 0. So, for example is the employee is male, then D1 = 0:

ŷ = 4.178-1.017X1-0.235(0)
ŷ = 4.178-1.017X1 regression model for male

Your scoring for female is 1. So, if the employee is female, then D1 =1:

ŷ = 4.178-1.017X1-0.235(1)
ŷ = 4.178-1.017X1-0.235

Re-arrange:
ŷ = 4.178-0.235-1.017X1
ŷ = 3.943-1.017X1 regression model for female
Interpretation:

If the employee is male and experience is 0, job performance is 4.178.


If the employee is female and experience is 0, job performance is 3.943.
For every 1 unit increase in experience, regardless of gender, job performance decreases by 1.017

You might also like