QMB 3600 Project C: Analysis of Tourist Attraction Factors in Zones

This project applies multiple regression analysis to investigate the factors influencing tourist attraction to specific zones. The study examines various independent variables, including the number of trips generated, number of households, workers in the zone, adults in the zone, jobs in the zone, shopping m2, and crime rate, along with binary and categorical variables such as nature of people generated, season, and color. The initial analysis identifies and addresses multicollinearity by removing correlated variables. Further, the analysis identifies outliers and removes them to improve the model's accuracy. The final regression output reveals the significance of several variables, including the number of trips generated, number of households, jobs in the zone, crime rate, and season. The model achieves an adjusted R-squared value of 0.67, indicating that 67% of the variation in tourist attraction can be explained by the model. The model is statistically significant with an overall p-value less than 0.05. The model is then used to predict the number of tourists attracted based on a specific zone, highlighting the impact of factors like crime rate and season on tourist numbers, and providing a practical application of the model for business decisions.

2
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
NUMBER OF TOURIST TRIPS ATTRACTED TO DIFFERENT ZONES IN US
Introduction
Tourism is a major factor in human socio-economic engagements. Apart from the revenue they
generate, they improve learning of other people’s way of life and also the creation of
employment. Studying the aspect of tourism is important for the betterment of the society.
Number of tourist attracted statistics
Multiple regression model
Dependent was categorically chosen to analyze the number of tourist trips attracted to different
zone. The independent variables were chosen to ascertain some of the reasons why tourist are
attracted to some zones compared to others. Some of the variables chosen, were the availability
of jobs, households, number of adults etc.
The researcher confident interval in this study is 95% giving us the alpha of 0.05
The prediction is that tourists are attracted to areas where there are plenty of jobs, large number
of adults who are working to offer services and to areas where there are low crime rates.
Independent quantitative variables
Number of trips generated: these are the number of trips generated from a certain zone due to
people moving out. It is important to compare those who move out as some are moving in
Number of households: number of households can be used to predict the population of an areas,
this is important as this population can offer the services required by the tourists
Workers in the zone: these are different service providers
Adult in the zone: this is important to our study because it enable us to predict the labour force
when need be if there is influx of tourists
Jobs in the zone: apart from just visiting a place for pleasure, some tourist visit to work,
availability of jobs hence serve well.
Shopping: these are the different category of the goods available for shopping. People who like
shopping can be attracted to varieties of thing to shop.
Crime rate: some tourist consider the crime rate before they can visit a place
Binary categorical variables
Nature of people generated: these are people leaving a certain zone to be either tourist
somewhere or people are attracted and are leaving.
Season: this is whether the season is summer or winter

3
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
Categorical variable with at least 3 variables
Colour: these are zones with majority as Latino, Blacks, White or Others
Original data
Number
of trips
attracted
Numb
er of
trips
genera
ted
Num
ber of
house
holds
Wor
kers
in
zone
Adul
ts in
zone
Jobs in
zone
Shopping
m2
Cri
me
Rat
e
Nat
ure
of
peo
ple
gen
erat
ed
Sea
son
Col
our
1063 1144 946 946 1892 3032 125716 20 0 0 1
678 421 310 310 620 3807 67182 21 0 0 2
553 823 337 674 1348 3498 188545 15 0 0 2
1768 1216 904 904 1808 4703 133366 32 1 1 3
1045 1168 838 838 2514 5926 187937 23 0 0 1
1406 1104 538 1076 2152 5935 198630 14 1 1 3
1528 1521 690 1380 2070 4514 127228 17 1 0 2
744 1946 921 1842 3684 5995 86105 15 0 0 4
577 938 875 875 1750 3064 16370 23 1 1 3
718 1023 805 805 1610 3558 179430 39 0 0 2
1084 268 230 230 460 4851 142533 38 0 0 1
892 565 474 474 1422 3142 191692 12 1 1 1
514 1066 710 710 1420 3017 48221 32 1 1 3
106 483 289 289 578 5228 39930 27 1 0 2
22 998 610 610 1220 3221 10322 26 0 0 4
449 257 157 157 471 3308 116098 12 0 1 4
233 800 565 1130 1695 3837 31795 18 1 0 1
35 1295 927 927 2781 5912 11855 21 0 0 2
1463 948 704 1408 2112 5175 95402 22 1 0 1
685 865 695 695 2085 4739 89118 33 1 1 3

4
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
Multicollinearity
Multicollinearity exist where two variables are correlated. From our data above, the number of
worker and Jobs in the zone are correlated because people who are working will be obviously
engaged. We will therefore remove the variable “Worker” as it may interfere with our analysis.
Once we remove, the multicollinearity, we remain with the following data
Number
of trips
attracte
d
Numbe
r of
trips
generat
ed
Number
of
househol
ds
Adul
ts in
zone
Jobs in
zone
Shoppin
g m2
Cri
me
Rate
Nature
of
people
generat
ed
Engag
ed
Colo
ur
1063 1144 946 1892 3032 125716 20 0 0 1
678 421 310 620 3807 67182 21 0 0 2
553 823 337 1348 3498 188545 15 0 0 2
1768 1216 904 1808 4703 133366 32 1 1 3
1045 1168 838 2514 5926 187937 23 0 0 1
1406 1104 538 2152 5935 198630 14 1 1 3
1528 1521 690 2070 4514 127228 17 1 0 2
744 1946 921 3684 5995 86105 15 0 0 4
577 938 875 1750 3064 16370 23 1 1 3
718 1023 805 1610 3558 179430 39 0 0 2
1084 268 230 460 4851 142533 38 0 0 1
892 565 474 1422 3142 191692 12 1 1 1
514 1066 710 1420 3017 48221 32 1 1 3
106 483 289 578 5228 39930 27 1 0 2
22 998 610 1220 3221 10322 26 0 0 4
449 257 157 471 3308 116098 12 0 1 4
233 800 565 1695 3837 31795 18 1 0 1
35 1295 927 2781 5912 11855 21 0 0 2
1463 948 704 2112 5175 95402 22 1 0 1
685 865 695 2085 4739 89118 33 1 1 3
Outliers
Outlier is data that differ significantly from the rest. In the above table, we can see that there are
two number of trips attracted which differ significantly from the rest. Value 22 and 35. The
following table has no outliers.

5
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
Numbe
r of
trips
attracte
d
Number
of trips
generate
d
Number
of
househol
ds
Adult
s in
zone
Jobs in
zone
Shoppin
g m2
Crim
e
Rate
Nature
of
people
generate
d
Seaso
n
Colou
r
1063 1144 946 1892 3032 125716 20 0 0 1
678 421 310 620 3807 67182 21 0 0 2
553 823 337 1348 3498 188545 15 0 0 2
1768 1216 904 1808 4703 133366 32 1 1 3
1045 1168 838 2514 5926 187937 23 0 0 1
1406 1104 538 2152 5935 198630 14 1 1 3
1528 1521 690 2070 4514 127228 17 1 0 2
744 1946 921 3684 5995 86105 15 0 0 4
577 938 875 1750 3064 16370 23 1 1 3
718 1023 805 1610 3558 179430 39 0 0 2
1084 268 230 460 4851 142533 38 0 0 1
892 565 474 1422 3142 191692 12 1 1 1
514 1066 710 1420 3017 48221 32 1 1 3
106 483 289 578 5228 39930 27 1 0 2
449 257 157 471 3308 116098 12 0 1 4
233 800 565 1695 3837 31795 18 1 0 1
1463 948 704 2112 5175 95402 22 1 0 1
685 865 695 2085 4739 89118 33 1 1 3

6
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
Excel data regression output
R Square 0.641413
Adjusted R
Square 0.238002
Standard
Error 400.6401
Observations 18
ANOVA
df SS MS F
Significa
nce F
Regression 9
229689
8
255210
.9
1.5899
75 0.262415
Residual 8
128410
0
160512
.5
Total 17
358099
8
Coefficie
nts
Standa
rd
Error t Stat
P-
value
Lower
95%
Upper
95%
Lowe
r
95.0
%
Upper
95.0%
Intercept -530.156
728.91
24
-
0.7273
3
0.4877
63 -2211.03
1150.7
19
-
2211.
03
1150.7
19
X Variable 1 0.959304
1.0177
22
0.9425
99
0.0434
76 -1.38757
3.3061
74
-
1.387
57
3.3061
74
X Variable 2 1.079052
1.3379
63
0.8064
88
0.0432
74 -2.0063 4.1644
-
2.006
3 4.1644
X Variable 3 -0.69156
0.4323
91
-
1.5993
8
0.1484
03 -1.68865
0.3055
37
-
1.688
65
0.3055
37
X Variable 4 0.264314
0.1538
16
1.7183
73
0.0240
51 -0.09039
0.6190
15
-
0.090
39
0.6190
15
X Variable 5 0.002896
0.0030
58
0.9469
62
0.3713
8 -0.00416
0.0099
48
-
0.004
16
0.0099
48
X Variable 6 -10.6024
16.755
58
-
0.6327
7
0.0445
45 -49.2408
28.036
08
-
49.24
08
28.036
08
X Variable 7 -18.7354 418.88
31
-
0.0447
0.9654
21
-984.682 947.21
07
-
984.6
947.21
07

7
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
3 82
X Variable 8 407.7466
629.15
5
0.6480
86
0.0350
83 -1043.09
1858.5
81
-
1043.
09
1858.5
81
X Variable 9 -176.027
269.54
88
-
0.6530
4
0.5320
44 -797.607
445.55
39
-
797.6
07
445.55
39
Significant variables
Here we can see that;
Number of trips generated, Number of households, Jobs in the Zone, Crime rate and Season have
a p-value less than 0.05 which was our level of significant hence they are statistically significant.
The table below shows data with significant variables only
Number
of trips
attracted
Number
of trips
generate
d
Number of
households
Jobs in
zone
Crime
Rate
Season
1063 1144 946 3032 20 0
678 421 310 3807 21 0
553 823 337 3498 15 0
1768 1216 904 4703 32 1
1045 1168 838 5926 23 0
1406 1104 538 5935 14 1
1528 1521 690 4514 17 0
744 1946 921 5995 15 0
577 938 875 3064 23 1
718 1023 805 3558 39 0
1084 268 230 4851 38 0
892 565 474 3142 12 1
514 1066 710 3017 32 1
106 483 289 5228 27 0
449 257 157 3308 12 1
233 800 565 3837 18 0
1463 948 704 5175 22 0
685 865 695 4739 33 1

8
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
Excel regression output
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.51910
8
R Square 0.69473
Adjusted R
Square 0.67491
Standard
Error
466.906
1
Observations 18
ANOVA
df SS MS F
Significa
nce F
Regression 5
96498
2.9
19299
6.6 0.8853 0.010134
Residual 12
26160
16
21800
1.3
Total 17
35809
98
Coeffici
ents
Standa
rd
Error t Stat
P-
value
Lower
95%
Upper
95%
Lowe
r
95.0
%
Upper
95.0%
Intercept -192.193 612.46 - 0.7590 -1526.64 1142.2 - 1142.2

9
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
5 0.3138 59 54
1526.
64 54
X Variable 1
0.03134
9
0.6018
82
0.0520
85
0.0593
18 -1.28004
1.3427
37
-
1.280
04
1.3427
37
X Variable 2
0.56418
6
0.9481
73
0.5950
24
0.0028
8 -1.5017
2.6300
76
-
1.501
7
2.6300
76
X Variable 3
0.15229
1
0.1264
57
1.2042
96
0.0251
69 -0.12323
0.4278
17
-
0.123
23
0.4278
17
X Variable 4 -1.0636
15.389
48
-
0.0691
1
0.0460
39 -34.5944
32.467
21
-
34.59
44
32.467
21
X Variable 5
129.987
4
238.26
94
0.5455
48
0.5953
72 -389.157
649.13
18
-
389.1
57
649.13
18
Interpretation of the result
The adjusted R square from the output is 0.67 meaning that 67% of the number of trips attracted
to a certain zone can be well explained with the model. Since the overall p-value of our model is
also less than 0.05, we can say that our model is statistically significant.
We can also see that the coefficient of the Crime rate is negative. This means that every time
there is increase in number of crime rate, then there will be decline of the total number of people
attracted to a certain zone by 1.0636 times the total crime rate. The rest of the coefficients are
positive meaning they are directly proportional to total number of tourist received.
The model
y=−192.193+ 0.0313 x 1+0.5642 x 2+0.1523 x 3−1.064 x 4 +129.99 x 5
Where;
x1 is number of trips generated
x2 is number of households
x3 is jobs in zone

10
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
x4 is crime rate
x5 season
For example, for a particular period, the number of trips generated in a zone was 1123,
household were 986, jobs in the zone is 820, crime rate 12 and season is winter denoted by 0, we
then have;
y=−192.193+ 0. 0313× 1123+0.5642 × 986+0. 1523× 820−1. 064 × 12+ 129.99 × 0
y=511
Therefore, the total number of people attracted to that area will be 511
When we have summer which is denoted by 1, then the number of tourist attracted to that zone at
that period will increase by 130, hence total number becomes 641 tourists.
Conclusion
Based on the last Excel output, I can say that in the end, the model is good. The model we
obtained can explain 67% of the total number of trips made by tourist attracted to a certain zone.
Generally, the model is statistically significant because it has an overall p-value of 0.01 which is
less than 0.05, the confidence level.
The model obtained is strong to make future prediction. Apart from its ability to explain the
changes in number of tourist attracted to certain area, most of the independent variables are
statistically significant making it stronger.
The model can be used to make business decision because considering the factors within a
certain zone like crime rate or season, you can use these factors to determine the possible number
of tourists who are likely to be attracted in that zone at a particular hence helping in proper prior
business planning.
Some of the variable were not statistically significant because of their minimal effect they have
on the number of tourist visiting a region. For example, number of adults in the zone is not
directly linked to the number of tourist visiting a certain zone.
Outliers can bring a wrong impression on the central tendency of a given set of data, but since
outliers were eliminated in the course of the work, they can have very little effect on the final
model since the model is statistically significant.
The topic about the number of tourist attracted to a certain region is important academically
since it helps the learner to understand what they have learned in classes better. Carrying out a
research and analyzing data give the learner a real world experience.
Statistics is a very important discipline in our day to day lives. The ability of applying statistics
to obtain the significant factors influencing some movements among many other factors is vital
as it make work and learning easier.

11
FACTORS AFFECTING TOURIST ATTRACTION TO CERTAIN ZONES
What I have learned is applicable in my career in that, in future I will be a businessman and in
business, the decisions made are very important. These decision affects all aspects of business,
therefore when statistics is applied in the decision making process, one can make bold decision
based on scientific evidence.
References
https://www.analyticsvidhya.com/blog/2015/08/comprehensive-guide-regression/
https://towardsdatascience.com/statistical-significance-hypothesis-testing-the-normal-curve-and-
p-values-93274fa32687
https://www.tutorialspoint.com/tourism_management/
tourism_management_factors_affecting.htm
https://www.brainscape.com/flashcards/tourism-factors-affecting-tourism-6149350/packs/
8462399
https://www.guru99.com/what-is-data-analysis.html