Lab Report 1
CEGE 3201
Chimdi Onyiuke
onyiu004
Lab 1
CEGE 3201
Chimdi Onyiuke
Vocab
1. OTAZ (Origin Transportation Analysis Zone): This is the ID number of the origin location. eg, 5,
6, 900, etc.
DTAZ: (Destination Transportation Analysis Zone): This is the ID number of the destination
location. eg, 7, 28, 680, etc.
2. mode or primary mode: This means the different types of ways to move around from origin
to destination. eg, bus, car, bike, etc.
3. access: This is the mode used to enter the primary mode if applicable. eg, walk, school bus.
egress: This is the mode used to exit the primary mode if applicable. eg, light rail, commuter rail
4. OrigAct: This is the activity category at trip origin. eg, work, chauffeur, etc.
DestAct: This is the activity category at trip destination. eg, shopping, discretionary, etc.
Ohome-Oother: This tells us if we are doing activity at a specific trip origin. eg, 1=yes, 0=no.
Dhome-Dother: This tells us if we are doing activity at a specific trip origin. eg, 1=yes, 0=no.
5. HHWEIGHT: This is the household expansion factor.
Analysis
1. Pivot table
Primary Mode Count of primary mode
0 10
Auto 69303
Bicycle (skateboard, scooter) 1150
Bus 1379
Commuter Rail 19
Light Rail 175
School Bus 2292
Taxi (Ambulance, dial-a-ride, private
bus) 122
Walk 4786
Grand Total 79236
Pie Chart
FREQUENCY OF DIFFERENT PRIMARY MODES
6% 0%
0%
3%
0%
0%
2%
1%
0
Auto
Bicycle (skateboard, scooter)
Bus
Commuter Rail
Light Rail
School Bus
Taxi (Ambulance, dial-a-ride, private bus)
Walk
88%
The most common mode is Automobile (Cars). It isn’t what I expected because Minnesota is a
very bus friendly state so would have thought a lot more people take the bus.
I think it might be common because it is sometimes convenient and faster to drive because you
have more control over where you are going and how you go.
I think driving might become less popular if parking prices go up relative to more bus routes and
closer bus stops being developed. This will give people more options and it will be cheaper and
more convenient.
2.
TAZ with primary mode Count of primary mode
1238 34
Auto 28
Bicycle (skateboard, scooter) 1
Walk 5
1280 216
Auto 78
Bicycle (skateboard, scooter) 18
Bus 34
Light Rail 1
School Bus 3
Taxi (Ambulance, dial-a-ride, private
bus) 1
Walk 81
1299 51
Auto 26
Bicycle (skateboard, scooter) 1
Bus 7
School Bus 1
Walk 16
1319 127
Auto 86
Bicycle (skateboard, scooter) 8
Bus 5
Light Rail 1
Taxi (Ambulance, dial-a-ride, private
bus) 2
Walk 25
DIFFRERENT MODES USED TO CERTAIN ZONES
1238 Auto
1238 Bicycle (skateboard, scooter)
1238 Walk
1280 Auto
1280 Bicycle (skateboard, scooter)
6% 7%
1%0%
0% 0% 1280 Bus
1%
2%
1280 Light Rail
1280 School Bus
18%
1280 Taxi (Ambulance, dial-a-ride, private
20% bus)
1280 Walk
1299 Auto
4% 1299 Bicycle (skateboard, scooter)
1299 Bus
4%
0%
2% 8% 1299 School Bus
0%
6% 0% 1299 Walk
1%
0%
1319 Auto
19%
1319 Bicycle (skateboard, scooter)
1319 Bus
1319 Light Rail
1319 Taxi (Ambulance, dial-a-ride, private
bus)
1319 Walk
The mode shares are not that different from the Twin Cities population. Most frequent mode is
still Auto like in twin cities. The only difference is that in zone 1280, there was a bit more
walking than driving.
I noticed that the least frequent mode is Taxi, this I expected because I feel like people don’t
really use taxis to get to campus. At least people I know. I didn’t expect bicycle (skateboard,
scooter) to be more than the school bus. But when it thought about it made a little sense
seeing as I see it around campus a lot and I ride a bike.
I think the difference in mode shares might be due to accessibility and convenience. For
example, if I need to make a trip that I can use a bus bike or walk. I would want to know if I can
park my bike or if the bus stop is far or how long it would take and how I would get there. These
factors influence choice of mode.
3.
Mode Percent
Drive alone 76.6
Carpool 9.0
Transit 5.2
Work at home 4.6
Walk 2.8
Other 1.2
Bike 0.6
https://www.bts.gov/content/commute-mode-share-2015
https://www.bts.gov/content/commute-mode-share-2015
In the United States, auto is still the mode with the highest percentage. And in some cases, the
percentages in the US and the University and Twin Cities are very similar. Like the percentage of
auto in the Twin Cities is about 88% and in the US is about 85%. There isn’t much difference in
percentages.
4. Discretionary travel is defined as all trips other than home-based work and home-based
school. So, I think discretionary will produce the most trips because I feel like we travel more
for activities outside school and work, like shopping, leisure, etc.
Different activities Count of destAct
College 359
Discretionary 12412
Eating Out 4114
Home 27569
Maintenance 4806
School 2687
Shopping 8249
Work 11147
Number of trips for different activities
0%
16%
17%
College
Discretionary
6%
11%
Eating Out
Home
Maintenance
School
4%
Shopping
Work
7%
39%
Discretionary produced the most trips as I expected. I also see a high number of trips for work
and school which I expect.
Lab 2
CEGE 3201
Chimdi Onyiuke
Example
Coefficient Sign Prediction Estimate p-value
Intercept + 1441.5332 <2e-16
Population Over 18 + 2.8151 <2e-16
Households that are + -3.4089 0.2261
not families and with
more than one
person
Renter Occupied + 1.3449 0.0386
Housing
Multiple R-squared 0.2187
p-value <2.2e-16
R-Command summary(lm(Otrips~POPOVER18+HHTYPE3+RENTEROCC,
tazData))
Assignment
1)
Coefficient Sign Prediction Estimate p-value
Intercept + 967.06621 <2e-16
Vacant housing - -4.17243 0.0349
Total Employment + 1.75529 <2e-16
Households with + 11.35917 <2e-16
children and under 65
Renter Occupied + 0.03118 0.9605
Housing
Households that are + 13.81990 1.13e-08
not families and with
more than one
person
Multiple R-squared 0.3557
p-value <2.2e-16
R-Command > m1 <- lm(formula= Otrips ~ VACANT + TOTAL_EMP + HHTYPE1 +
RENTEROCC + HHTYPE3, data = tazData)
> summary(m1)
d) Vacant housing, total employment, households with children and under 65, and other
households that are not families and with more than one person are all significant. This means
that the predictors with positive coefficients will have more trips generated when their number
increases and the predictors with negative coefficients will have less trips generated when their
number decreases.
e) 3826.574 trips will be generated.
2)
Coefficient Sign Prediction Estimate p-value
Intercept + 128.06210 <4.79e-11
Vacant housing - -0.33774 0.349
Total Employment + 0.22541 <2e-16
Households with + 1.98864 <2e-16
children and under 65
Renter Occupied + -0.36829 0.0011
Housing
Households that are + 4.61594 <2e-16
not families and with
more than one
person
Multiple R-squared 0.315
p-value <2.2e-16
R-Command > m2 <- lm(formula= Otrips_Work ~ VACANT + TOTAL_EMP +
HHTYPE1 + RENTEROCC + HHTYPE3, data = tazData)
> summary(m2)
b) I am choosing the same sign for all of them because I think the predictors we are comparing
won’t change much in terms of work.
e) The estimates are smaller than the ones for the general origin trips.
f) Total employment, households with children and under 65, renter occupied housing, and other
households that are not families and with more than one person are all significant. My sign
prediction for the renter occupied housing coefficient was wrong. It changed from positive to
negative. This means that when there is more renter occupied housing, less work trips are
made.
3)
Coefficient Sign Prediction Estimate p-value
Intercept + 10.765372 0.16957
Vacant housing - 0.017228 0.90425
Total Employment + 0.010935 0.01475
Households with - -0.125066 0.00183
children and under 65
Renter Occupied + -0.107497 0.01832
Housing
Households that are - 1.148459 5.88e-11
not families and with
more than one
person
Multiple R-squared 0.02189
p-value 4.392e-13
R-Command > m3 <- lm(formula= Otrips_College ~ VACANT + TOTAL_EMP +
HHTYPE1 + RENTEROCC + HHTYPE3, data = tazData)
> summary(m3)
b) I am choosing different signs for Households with children and under 65 because first
children will not be making a lot of college trips and also depending on the age under 65 I think
there will be less trips as well. For households that are not families and with more than one
person.
e) The estimates are significantly smaller than the ones for the general and work origin trips.
f) Total employment, households with children and under 65, renter occupied housing, and other
households that are not families and with more than one person are all significant.
Lab 3
CEGE 3201
Chimdi Onyiuke
Example
1.
Coefficient Estimate p-value
Intercept 2.926e3 <2e-16
TOTAL_EMP 1.799e0 <2e-16
Multiple R-squared 0.1897
p-value <2.2e-16
R-Command m1 <- lm (formula= Otrips ~ TOTAL_EMP, tazData)
summary(m1)
Coefficient Estimate p-value
Intercept 2.918e3 <2e-16
TOTAL_EMP 1.813e0 <2e-16
Multiple R-squared 0.1897
p-value <2.2e-16
R-Command m2 <- lm (formula= Dtrips ~ TOTAL_EMP, tazData)
summary(m2)
