You are on page 1of 14

VI.

Regression

Variables Variables
Model Entered Removed Method
1 Mean Anger: All
. Enter
itemsa
a. All requested variables entered.
b. Dependent Variable: Mean Delinquency: All items

Model Summary

Std. Error of the


Model R R Square Adjusted R Square Estimate
1 .338a .114 .113 .3644
a. Predictors: (Constant), Mean Anger: All items
a.
For this data, R has a value of .338 which represents the simple correlation between mean anger
and mean delinquency. The value of R 2 is .114 which tells us that mean anger can account for 11.4% of
the variation in mean delinquency.
The F-ratio is 167.917 which is significant at p<0.001. Thus, our regression model results in
significantly better prediction of mean delinquency. In short, the regression predicts the dependent
variable significantly well.
b.
When anger is constrained (set to zero), we predict that delinquency would have a score of .201.
Nevertheless, increasing anger by 1 unit, we would expect an increase of 0.358 unit on one’s delinquency
score.
c.
Mean delinquency = bo + b1
= .157 + .201 x 1
= .358
d.
The probability of the given data is .000 which is significant at p<.001. We can conclude that
mean anger can predict mean delinquency.
e. R = k (N-1)
= 1 (1306-1)
= 0.000
Thus, there is no effect since the value of the effect size is 0.

2.

Model Summary

Durbin-
Std. Error Change Statistics Watson
Adjusted of the R Square F Sig. F
Model R R Square R Square Estimate Change Change df1 df2 Change
1 .480a .231 .229 .4853 .231 193.299 2 1290 .000 1.788
a. Predictors: (Constant), Mean Overall Strain with Bullying,
Mean Anger: All items
b. Dependent Variable: Mean Delinquency:
Subscale2
ANOVAb

Sum of Mean
Model Squares df Square F Sig.
1 Regression 91.062 2 45.531 193.299 .000a

Residual 303.855 1290 .236

Total 394.917 1292

a. Predictors: (Constant), Mean Overall Strain with Bullying, Mean Anger: All items

b. Dependent Variable: Mean Delinquency: Subscale2


Coefficientsa
Standardize 95%
Unstandardize d Confidence Collinearity
d Coefficients Coefficients Interval for B Correlations Statistics
Upper
Std. Lower Boun Zero- Partia
Model B Error Beta t Sig. Bound d order l Part Tolerance VIF
1 (Constant) .132 .034 3.871 .000 .065 .199
Mean
Anger: All .179 .025 .210 7.259 .000 .130 .227 .389 .198 .177 .711 1.406
items
Mean
Overall
Strain .410 .036 .333 11.508 .000 .340 .480 .446 .305 .281 .711 1.406
with
Bullying
a. Dependent Variable:
Mean Delinquency:
Subscale2

Correlations
Mean Mean Overall
Mean Anger: Delinquency: Strain with
All items All items Bullying
Mean Anger: All items Pearson Correlation 1 .338** .539**
Sig. (2-tailed) .000 .000
N 1314 1305 1301
Mean Delinquency: All Pearson Correlation .338** 1 .426**
items Sig. (2-tailed) .000 .000
N 1305 1309 1296
Mean Overall Strain with Pearson Correlation .539** .426** 1
Bullying Sig. (2-tailed) .000 .000
N 1301 1296 1304
**. Correlation is significant at the 0.01 level (2-tailed).

a. Based on the data, anger is a significant predictor of delinquency since it has a value of 0.338. Also,
mean overall strain with bullying is a significant predictor of delinquency since it has a value of 0.426.
b. When both anger and strain is constrained (set to zero), we predict that the delinquency would
have a score of ____. Nevertheless, increasing anger by 1 unit while controlling for the effect of strain,
we would expect an (increase, decrease) of ___ unit on one’s delinquency. In addition, increasing strain
by 1 unit while controlling for the effect of anger, we would expect an (increase, decrease) of ___ unit on
one’s delinquency.
Explain the beta coefficient of anger on this two-predictor model with the beta coefficient in
the one-predictor model. What happened? Explain.
“While controlling for the effect of the other predictor” means we remove the effect of other
variables from the relation between the two or more variables. This implies that we keep the effect of
other variables brought explicitly in the model constant.
c. Express the equation of the line (regression line; your two-predictor model). What does this
model say about the propensity to commit delinquency among adolescents? Does it make sense?
Explain.
d. Does the equation of line provide predictions that are beyond random chance? (Ergo, is the
regression line significant?)
e. Effect size of the model: How much of the variance in delinquency could be explained/accounted
for by the model (or equation line)? (Compare the R squared of this model and the previous model.
Would you consider the increase significant? Would you say the increase is necessary? Would you
say the increase is substantial?

Total Variance Explained

Extraction Sums of Squared Rotation Sums of Squared


Initial Eigenvalues Loadings Loadings
Com
pone % of Cumulativ % of Cumulativ % of Cumulativ
nt Total Variance e% Total Variance e% Total Variance e%

1 7.844 25.303 25.303 7.844 25.303 25.303 3.364 10.853 10.853

2 3.157 10.185 35.488 3.157 10.185 35.488 3.040 9.808 20.661


3 1.410 4.549 40.037 1.410 4.549 40.037 2.700 8.710 29.371

4 1.311 4.230 44.267 1.311 4.230 44.267 2.408 7.768 37.139

5 1.230 3.969 48.237 1.230 3.969 48.237 2.206 7.115 44.254

6 1.052 3.395 51.631 1.052 3.395 51.631 1.787 5.763 50.017

7 1.037 3.344 54.976 1.037 3.344 54.976 1.537 4.959 54.976

8 .946 3.051 58.026

9 .886 2.858 60.884

10 .870 2.807 63.691

11 .796 2.567 66.258

12 .721 2.327 68.586

13 .711 2.293 70.879

14 .693 2.235 73.114

15 .667 2.152 75.266

16 .656 2.115 77.381

17 .625 2.017 79.399

18 .605 1.951 81.350

19 .592 1.910 83.260

20 .564 1.819 85.079

21 .536 1.730 86.809

22 .518 1.672 88.481

23 .499 1.610 90.091

24 .477 1.538 91.628


25 .447 1.441 93.069

26 .430 1.387 94.456

27 .402 1.296 95.752

28 .392 1.265 97.017

29 .377 1.217 98.234

30 .317 1.023 99.258

31 .230 .742 100.000

Extraction Method: Principal


Component Analysis.

Rotated Component Matrixa

Component

1 2 3 4 5 6 7

Done things that bother


-.076 .369 .043 .600 .092 -.080 .033
others

Pushed and shoved


-.006 .319 .031 .715 .033 .053 .000
others

Started a fight over


.156 .131 .141 .633 -.005 .158 .089
nothing

Hit or threatened to hit


fellow students at .140 -.040 .134 .686 -.014 .224 .091
school

Joined a gang, a
.195 -.141 .502 .144 .099 .399 .052
fraternity, a “clan”
Intentionally broken,
damaged property .018 -.060 .125 .348 .087 .601 .139
belonging to others

Said mean or nasty


things purposely to
-.047 .429 .097 .434 .023 .327 -.015
insult or embarrass
others

Stolen money or other


things from your
.021 .461 .001 .051 .095 .609 -.023
parents or other
members of the family

Took things from other


.068 .341 .009 .088 -.048 .637 .177
people without asking

Skipped school without


.170 .285 .426 .110 -.186 .080 .377
an excuse

Cheated on school tests .016 .612 .118 .032 .085 .073 .224

Avoided paying for


tricycles, jeepneys, .084 .276 .135 .072 .160 .137 .627
buses

Avoided paying the


food ordered in the .076 .039 .130 .078 .086 .079 .789
canteen, store

Gambling activities -.066 .275 .543 .181 .217 .066 .250

Drank alcoholic drinks


.087 .257 .735 .076 .125 .032 .093
(e.g. beers, wine. liquor)

Disobeyed parents .007 .657 .169 .209 .011 .069 -.102

Disobeyed teachers .164 .559 .121 .319 .120 -.084 .184

Lying to authorities
such as parents and .126 .608 .029 .190 -.072 .155 .102
teachers
Smoked cigarettes .310 .111 .680 .104 .079 -.044 .101

Smoked marijuana .643 .080 .468 .006 .151 .020 -.077

Sniffed glue or rugby .788 .146 .105 .052 .055 -.008 .010

Used other types of


.747 .028 .157 .003 .255 -.080 .049
drugs (e.g. shabu)

Reprimanded by police,
.599 .054 .038 .102 .359 .139 .210
barangay officials

Carried a gun/knife and


.762 .070 .085 .083 .172 .133 .091
other harmful weapons

Made vandals on walls,


fences, chairs, or any
.273 .420 .142 .064 .133 .103 .085
other property owned
by others

Watched/listened to
pornographic videos .151 .425 .354 .009 .302 .168 .112
and materials

Gone to night clubs or


.267 .180 .405 -.008 .361 .243 -.065
pubs

Extorted money from


.386 .095 .270 .098 .401 .276 .117
others

Had sex with someone


.311 .054 .141 .011 .804 .008 .102
of the opposite sex

Had sex with someone


.303 .088 .133 -.001 .773 .028 .065
of the same sex

Ran away from home .200 .015 .356 .191 .331 -.032 .095

Extraction Method: Principal Component Analysis.

Rotation Method: Varimax with Kaiser Normalization.

a. Rotation converged in 9 iterations.


Reliability

Scale: Factor 1

Case Processing Summary Reliability Statistics

N % Cronbach's
Alpha N of Items
Cases Valid 1316 89.0
.826 5
Excludeda 163 11.0

Total 1479 100.0

a. Listwise deletion based on all variables in the


procedure.
Scale: Factor2

Case Processing Summary Reliability Statistics

N % Cronbach's
Alpha N of Items
Cases Valid 1309 88.5
.769 8
a
Excluded 170 11.5

Total 1479 100.0

a. Listwise deletion based on all variables


in the procedure.

Scale: factor3

Case Processing Summary Reliability Statistics

N % Cronbach's
Alpha N of Items
Cases Valid 1309 88.5
.764 7
a
Excluded 170 11.5

Total 1479 100.0

a. Listwise deletion based on all variables


in the procedure.

Scale: factor4
Case Processing Summary Reliability Statistics

N % Cronbach's
Alpha N of Items
Cases Valid 1313 88.8
.730 5
Excludeda 166 11.2

Total 1479 100.0

a. Listwise deletion based on all variables


in the procedure.

Scale: factor5

Case Processing Summary Reliability Statistics

N % Cronbach's
Alpha N of Items
Cases Valid 1316 89.0
.738 3
Excludeda 163 11.0

Total 1479 100.0

a. Listwise deletion based on all variables


in the procedure.
Scale: factor6

Case Processing Summary Reliability Statistics

N % Cronbach's
Alpha N of Items
Cases Valid 1312 88.7
.583 3
Excludeda 167 11.3

Total 1479 100.0

a. Listwise deletion based on all variables


in the procedure.

Scale: factor7

Case Processing Summary Reliability Statistics

N % Cronbach's
Alpha N of Items
Cases Valid 1314 88.8
.468 2
a
Excluded 165 11.2

Total 1479 100.0

a. Listwise deletion based on all variables


in the procedure.

a. We were able to produce 7 factors.


b. These factors represent interdependent variables that are grouped into categories. These factors were
classified into types with similar characteristics or behavior.
c. The subscales 1, 2, 3, 4 and 5 has the Cronbach’s Alpha has a value of 0.826, 0.769, 0.764, 0.730 and
0.738 respectively. These subscales are considered to be reliable since they exceeded 0.7. On the other
hand, subscales 6 and 7 has a value of 0.583 and 0.468, respectively, are not considered to be reliable
since they did not exceed 0.7.
d. Report the result of your factor analysis
e. The subscale’s Cronbach’s Alpha are 0.826, 0.769, 0.764, 0.730, 0.738, 0.583 and 0.468. Subscales 1 to
5 are above 0.70 (70%) which is good. This means that most of the items are of the same facet or
construct. Therefore, it is internally consistent and are closely related as a group. However, two subscales
(6 and 7) are below 70% which means that they are not as closely related to the other subscales.

You might also like