You are on page 1of 8

Factor Analysis

KMO and Bartlett's Test


Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .545
Bartlett's Test of Sphericity Approx. Chi-Square 457.143
df 210
Sig. .000

- The KMO value is 0.545. This shows that the sample size is not very sufficient
for conducting the factor analysis.
- Ideally this value should be >=0.7 to represent an adequate sample size.
- The significance value obtained in the Bartlett’s test is 0.000.
- Since this value is very low,i.e, close to 0, it means that the statements can be
correlated to form certain factors and hence this data is suitable for conducting
factor analysis.

Communalities
Initial Extraction
Master Bedroom 1.000 .673
Living Room 1.000 .636
Curb Appeal 1.000 .536
Yard 1.000 .729
Kitchen 1.000 .696
Basement 1.000 .678
Garage 1.000 .769
Dining Room 1.000 .640
Crime Rate 1.000 .417
Neighborhood prestige/ 1.000 .717
Reputation
Proximity School 1.000 .739
Proximity to entertainment 1.000 .759
(Bar, Restaurant, other)
Commute 1.000 .601
Privacy 1.000 .730
Traffic Rating 1.000 .601
Current listings 1.000 .593
Track Record 1.000 .759
Referral 1.000 .637
Neighborhood Expertise 1.000 .528
Photo of Agents 1.000 .688
Year of experience 1.000 .716
Extraction Method: Principal Component Analysis.

Communalities
Communality gives the variance accounted for a particular variable by all the factors. The higher
the value of the communality for a particular variable after extraction, the higher is the amount of
variance explained by the extracted factors. As seen in the last column, all the statements have a
communality >0.4.
If any statement has communality less than 0.4, then we can remove that statement
Total Variance Explained
Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings
Componen % of Cumulative % of Cumulative % of Cumulative
t Total Variance % Total Variance % Total Variance %
1 2.700 12.856 12.856 2.700 12.856 12.856 2.089 9.948 9.948
2 2.508 11.941 24.797 2.508 11.941 24.797 1.818 8.656 18.603
3 1.981 9.434 34.232 1.981 9.434 34.232 1.743 8.299 26.903
4 1.646 7.839 42.071 1.646 7.839 42.071 1.734 8.256 35.159
5 1.495 7.120 49.191 1.495 7.120 49.191 1.647 7.843 43.002
6 1.342 6.389 55.579 1.342 6.389 55.579 1.641 7.815 50.816
7 1.145 5.453 61.032 1.145 5.453 61.032 1.625 7.738 58.554
8 1.026 4.884 65.917 1.026 4.884 65.917 1.546 7.362 65.917
9 .976 4.648 70.564
10 .897 4.269 74.834
11 .772 3.675 78.508
12 .652 3.106 81.615
13 .611 2.910 84.525
14 .591 2.817 87.342
15 .526 2.505 89.847
16 .496 2.363 92.210
17 .413 1.964 94.175
18 .367 1.746 95.921
19 .340 1.619 97.540
20 .274 1.304 98.844
21 .243 1.156 100.000
Extraction Method: Principal Component Analysis.

The above table shows that out of the 21 statements, 8 are able to explain 65.917 % of the
variation in the variables. This means that the 21 variables can be reduced to 3 factors.

For the 1st factor the eigen value as obtained from the graph  2.8
2nd factor  2.5
3rd factor  2.0

Component Matrixa
Component
1 2 3 4 5 6 7 8
Kitchen .562 -.225 -.248 -.163 -.462 -.089 .018 -.143
Living Room .535 .281 .096 -.144 -.253 -.288 -.295 -.080
Master Bedroom .492 .016 .217 .316 -.014 -.095 .392 -.347
Garage .490 -.247 .030 -.081 -.373 -.481 -.115 .277
Crime Rate .468 -.168 .214 .269 .038 .213 -.064 -.019
Proximity School .462 -.144 .277 -.150 .325 .361 -.322 -.256
Curb Appeal .421 -.140 -.286 -.165 -.277 .123 .229 -.293
Commute .417 -.279 .127 -.318 .384 .134 .158 .204
Neighborhood Expertise .378 -.296 -.338 .183 .377 .070 -.030 -.051
Dining Room .224 .696 -.048 -.027 .029 .017 -.083 -.307
Basement .113 .613 -.089 -.145 .352 -.129 .312 -.152
Yard .061 .592 .222 -.024 .354 -.388 -.209 -.074
Neighborhood prestige/ .335 .520 -.335 -.080 .022 .040 -.112 .449
Reputation
Track Record .340 .056 -.621 -.330 .273 .133 -.065 .222
Current listings .332 .250 .546 .252 -.087 .056 -.195 .099
Referral .126 .178 .539 .194 -.124 .241 .203 .383
Year of experience .337 -.125 -.207 .649 .220 -.152 -.156 .164
Photo of Agents .142 .260 .353 -.582 -.132 .104 .292 .153
Proximity to entertainment .057 .439 -.373 .454 -.136 -.048 .431 .107
(Bar, Restaurant, other)
Privacy .166 .328 -.133 .171 -.261 .689 -.055 .031
Traffic Rating .322 -.348 .241 -.020 .323 -.195 .402 .114
Extraction Method: Principal Component Analysis.
a. 8 components extracted.

Rotated Component Matrixa


Component
1 2 3 4 5 6 7 8
Basement .730 -.066 -.133 .017 .180 .171 -.181 -.171
Yard .722 .023 .003 -.416 -.013 .002 .173 .063
Dining Room .712 .105 -.054 .139 .117 -.290 .004 .025
Referral -.054 .743 -.186 -.130 -.020 .136 -.031 -.107
Current listings .202 .656 .046 -.118 -.125 -.049 .239 .177
Privacy -.001 .466 -.002 .329 .332 -.460 -.287 .023
Crime Rate -.050 .425 .297 .214 -.027 .142 .067 .272
Year of experience -.001 .169 .797 -.049 .101 .103 .139 -.101
Photo of Agents .154 .223 -.726 .083 .149 .223 .080 .055
Neighborhood Expertise -.030 -.125 .543 .216 .235 .269 -.075 .195
Curb Appeal -.012 -.096 -.031 .712 .084 .045 .098 .029
Kitchen -.118 -.061 .021 .653 .099 .003 .490 .044
Master Bedroom .313 .311 .218 .456 -.335 .320 .056 -.065
Track Record .072 -.292 .108 .184 .775 .111 .004 .099
Neighborhood prestige/ .283 .169 .023 -.041 .723 -.116 .194 -.181
Reputation
Traffic Rating -.039 .087 .077 .058 -.091 .755 .057 .039
Commute -.073 .083 -.062 .086 .295 .605 -.006 .354
Garage -.182 .056 .054 .135 .057 .197 .815 -.066
Living Room .363 .140 -.037 .170 .080 -.123 .637 .162
Proximity School .121 .204 .108 .155 .059 .121 -.045 .792
Proximity to entertainment .269 .168 .232 .204 .163 -.087 -.140 -.714
(Bar, Restaurant, other)
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.a
a. Rotation converged in 17 iterations.

Component Transformation Matrix


Component 1 2 3 4 5 6 7 8
1 .247 .352 .257 .526 .304 .321 .451 .271
2 .772 .246 -.225 -.147 .247 -.367 -.059 -.270
3 .051 .582 -.340 -.313 -.537 .227 .090 .311
4 -.029 .404 .763 -.066 -.295 -.155 -.143 -.342
5 .338 -.191 .320 -.404 .188 .504 -.431 .328
6 -.260 .404 -.109 .274 .296 -.262 -.644 .334
7 .054 .073 -.258 .372 -.119 .567 -.354 -.570
8 -.394 .329 -.067 -.471 .577 .207 .196 -.309
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.

From the above cluster we can group them into 3 -3 pair for giving the name to each entity so,

1st Pair Basement, Yard and Dining room =Essential Space

2nd referral, Current Listing, Privacy =Trust /Security

3rd Year of experience, Crime rate, Photo Of agents= trust


4th

Cluster Analysis

Final Cluster Centers


Cluster
1 2
Master Bedroom 4 4
Living Room 3 4
Curb Appeal 3 4
Yard 4 4
Kitchen 3 4
Basement 3 4
Garage 3 4
Dining Room 3 4
Crime Rate 3 4
Neighborhood prestige/ 3 4
Reputation
Proximity School 3 4
Proximity to entertainment 3 3
(Bar, Restaurant, other)
Commute 3 4
Privacy 3 3
Traffic Rating 3 4
Current listings 4 4
Track Record 3 4
Referral 4 3
Neighborhood Expertise 3 4
Photo of Agents 3 3
Year of experience 4 4

From the above cluster 1 we can say that there is strong Agreement 1, 4,16,18,21 And from rest
statement we can say that it shows strong disagreement in cluster 1

Were as in cluster 2 we can say that there is a strong Agreement from statements
1,2,3,4,5,6,7,8,9,10,11,13,15,16,17,19,20 and rest shows strongly disagreement in cluster 2
ANOVA
Cluster Error
Mean Square df Mean Square df F Sig.
Master Bedroom 6.968 1 .760 100 9.166 .003
Living Room 18.737 1 .981 100 19.093 .000
Curb Appeal 12.513 1 1.053 100 11.879 .001
Yard .047 1 .998 100 .047 .829
Kitchen 25.034 1 .900 100 27.829 .000
Basement 5.752 1 1.217 100 4.725 .032
Garage 15.029 1 .977 100 15.387 .000
Dining Room 3.356 1 1.259 100 2.666 .106
Crime Rate 9.634 1 1.050 100 9.179 .003
Neighborhood prestige/ 7.329 1 .906 100 8.086 .005
Reputation
Proximity School 9.071 1 .801 100 11.328 .001
Proximity to entertainment .466 1 1.118 100 .417 .520
(Bar, Restaurant, other)
Commute 16.159 1 .742 100 21.792 .000
Privacy .225 1 1.021 100 .220 .640
Traffic Rating 5.170 1 .978 100 5.283 .024
Current listings 2.472 1 1.129 100 2.190 .142
Track Record 34.701 1 .962 100 36.083 .000
Referral 3.334 1 1.075 100 3.100 .081
Neighborhood Expertise 13.546 1 1.137 100 11.908 .001
Photo of Agents 7.985 1 1.353 100 5.903 .017
Year of experience .255 1 .826 100 .309 .580
The F tests should be used only for descriptive purposes because the clusters have been chosen to maximize the
differences among cases in different clusters. The observed significance levels are not corrected for this and thus
cannot be interpreted as tests of the hypothesis that the cluster means are equal.

The F tests should be used only for descriptive purposes because the clusters have been chosen
to maximize the differences among cases in different clusters. The observed significance levels
are not corrected for this and thus cannot be interpreted as tests of the hypothesis that the cluster
means are equal.
In the last Column of the Anova table we have the significance values
Wherever the Significance value is less than 0.05 it means that there is a significant difference in
the perception of customers for that particular statement.

For Statement 1,2,3,56,7,8,9,10,11,13,15,17,18,19,20 the significance value is less than 0.05


which means that there is a significant difference in the perception of customers from the 2
Clusters with respect to these statements.

You might also like