
# install geopy
!pip install geopy

Requirement already satisfied: geopy in c:\users\colto\anaconda3\lib\site-packages (2.4.0)
Requirement already satisfied: geographiclib<3,>=1.52 in c:\users\colto\anaconda3\lib\site-packages (from geopy) (2.0)


import pandas as pd

# Import Nominatim ("Name" in Latin) and initialize
# Though most of these services require an API key, Nominatim, which uses OpenStreetMap data, does not,
# which is why we’re going to use it here. But we still need to create a unique application name.
from geopy.geocoders import Nominatim
geolocator = Nominatim(user_agent="Wenyi Shang's mapping app", timeout=2)


# Locate a specific location
location = geolocator.geocode("East Daniel Street, Champaign")
location

Location(East Daniel Street, Campustown, Champaign, Champaign County, Illinois, 61820, United States, (40.1079572, -88.2319022, 0.0))


# Full information
location.raw

{'place_id': 27867261,
 'licence': 'Data © OpenStreetMap contributors, ODbL 1.0. http://osm.org/copyright',
 'osm_type': 'way',
 'osm_id': 745458640,
 'lat': '40.1079572',
 'lon': '-88.2319022',
 'class': 'highway',
 'type': 'residential',
 'place_rank': 26,
 'importance': 0.10000999999999993,
 'addresstype': 'road',
 'name': 'East Daniel Street',
 'display_name': 'East Daniel Street, Campustown, Champaign, Champaign County, Illinois, 61820, United States',
 'boundingbox': ['40.1079500', '40.1079691', '-88.2335000', '-88.2303450']}


print(location.address)
print(location.raw['lat'])
print(location.raw['lon'])
print(location.raw['class'])
print(location.raw['type'])

East Daniel Street, Campustown, Champaign, Champaign County, Illinois, 61820, United States
40.1079572
-88.2319022
highway
residential


possible_locations = geolocator.geocode("East Daniel Street", exactly_one=False)
for location in possible_locations:
    print(location.address)

East Daniel Street, Bloomington, Monroe County, Indiana, 47401, United States
East Daniel Street, Campustown, Champaign, Champaign County, Illinois, 61820, United States
East Daniel Street, Seniorland, Champaign, Champaign County, Illinois, 61820, United States
East Daniel Street, Collinwood, Wayne County, Middle Tennessee, Tennessee, 38450, United States
East Daniel Street, Hinton, Boone County, Missouri, United States
East Daniel Street, Greene County, Missouri, 65803, United States
East Daniel Street, Uvalde, Uvalde County, Texas, 78801, United States
East Daniel Street, North Uvalde Colonia, Uvalde, Uvalde County, Texas, 78801, United States
East Daniel Street, Albany, Gentry County, Missouri, 64402, United States
East Daniel Street, Silver City, Grant County, New Mexico, 88062, United States


Illini_Union = geolocator.geocode('Illini Union')
Illini_Union.address

'Illini Union, 1401, West Green Street, Urbana, Champaign County, Illinois, 61801, United States'


# Create a dataframe to store the geographical locations of a list of places in Champaign.
# The dataframe "Champaign_df" should contain 6 columns:
# "Place", "Address" (obtained by ".address"), "Latitude", "Longitude", "Class", "Type" (obtained by corresponding keys in ".raw")

Champaign_places = ['Foellinger Auditorium', 'Altgeld Hall', 'Krannert Center for the Performing Arts', 
                    'University of Illinois Urbana-Champaign University Library', 'Japan House']

geolocator = Nominatim(user_agent="champaign_locator")

place_list = []
address_list = []
latitude_list = []
longitude_list = []
class_list = []
type_list = []

Cham = geolocator.geocode('Champaign, Illinois')

for place in Champaign_places:
    location = geolocator.geocode(place)
    place_list.append(place)
    address_list.append(location.address)
    latitude_list.append(location.latitude)
    longitude_list.append(location.longitude)
    location_raw = location.raw
    class_list.append(location_raw.get('class'))
    type_list.append(location_raw.get('type'))

Champaign_df = pd.DataFrame({
    'Place': place_list,
    'Address': address_list,
    'Latitude': latitude_list,
    'Longitude': longitude_list,
    'Class': class_list,
    'Type': type_list
})

print(Champaign_df)

                                               Place  \
0                              Foellinger Auditorium   
1                                       Altgeld Hall   
2            Krannert Center for the Performing Arts   
3  University of Illinois Urbana-Champaign Univer...   
4                                        Japan House   

                                             Address   Latitude  Longitude  \
0  Foellinger Auditorium, 709, South Mathews Aven...  40.105941 -88.227182   
1  Altgeld Hall, 1409, West Green Street, Urbana,...  40.109328 -88.228328   
2  Krannert Center for the Performing Arts, 500, ...  40.108011 -88.222528   
3  Main Library, 1408, West Gregory Drive, Urbana...  40.104681 -88.228990   
4  Japan House, 101-111, Kensington High Street, ...  51.501350  -0.191682   

      Class              Type  
0   amenity           theatre  
1  building        university  
2   amenity       arts_centre  
3   amenity           library  
4   amenity  community_centre


!pip install folium

Requirement already satisfied: folium in c:\users\colto\anaconda3\lib\site-packages (0.14.0)
Requirement already satisfied: requests in c:\users\colto\anaconda3\lib\site-packages (from folium) (2.27.1)
Requirement already satisfied: branca>=0.6.0 in c:\users\colto\anaconda3\lib\site-packages (from folium) (0.6.0)
Requirement already satisfied: numpy in c:\users\colto\anaconda3\lib\site-packages (from folium) (1.21.5)
Requirement already satisfied: jinja2>=2.9 in c:\users\colto\anaconda3\lib\site-packages (from folium) (2.11.3)
Requirement already satisfied: MarkupSafe>=0.23 in c:\users\colto\anaconda3\lib\site-packages (from jinja2>=2.9->folium) (2.0.1)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\users\colto\anaconda3\lib\site-packages (from requests->folium) (1.26.9)
Requirement already satisfied: idna<4,>=2.5 in c:\users\colto\anaconda3\lib\site-packages (from requests->folium) (3.3)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\colto\anaconda3\lib\site-packages (from requests->folium) (2021.10.8)
Requirement already satisfied: charset-normalizer~=2.0.0 in c:\users\colto\anaconda3\lib\site-packages (from requests->folium) (2.0.4)


import folium


Champaign = geolocator.geocode('Champaign')
Champaign

Location(Champaign, Champaign County, Illinois, United States, (40.1164841, -88.2430932, 0.0))


Champaign = geolocator.geocode('Champaign')
champaign_map

Location(Champaign, Champaign County, Illinois, United States, (40.1164841, -88.2430932, 0.0))


# Add a marker
folium.Marker(location=([Illini_Union.raw['lat'], Illini_Union.raw['lon']]), tooltip = 'click me', 
              popup="Illini Union").add_to(champaign_map)
champaign_map


champaign_map.save("Data/Champaign-map.html")


# First, reload the champaign_map to drop the added Marker of Illini Union
# Then for each place in Champaign_df you created for Task 1, add it as a Marker, with location defined by the
# latitude and longitude values in the DataFrame, popup value as the place names in the dataframe



champaign_map = folium.Map(location=[Cham.latitude, Cham.longitude], zoom_start=14)

for i in range(len(Champaign_df)):
    folium.Marker(
        location=[Champaign_df['Latitude'][i], Champaign_df['Longitude'][i]],
        popup=Champaign_df['Place'][i],
    ).add_to(champaign_map)

champaign_map


# Scottish witchcraft dataset
# (http://witches.hca.ed.ac.uk/#:~:text=The%20database%20contains%20all%20people,to%20social%20and%20cultural%20history)
df = pd.read_csv('data/accused_witches.csv')
df


# Count the number of witches accused in each county
county_witches = df.groupby("Res_county")["AccusedRef"].count()
county_witches

Res_county
Aberdeen         175
Argyll             6
Ayr              153
Banff              9
Berwick          126
Bute              54
Caithness         52
Clackmannan       18
Cromarty           2
Dumfries          78
Dunbarton         25
Edinburgh        374
Elgin             28
Fife             382
Forfar            82
Haddington       543
Inverness         45
Kincardine         2
Kinross            7
Kirkcudbright     35
Lanark            77
Linlithgow       114
Nairn             55
Orkney            72
Peebles           91
Perth            109
Renfrew          124
Ross              74
Roxburgh          60
Selkirk           21
Shetland          28
Stirling          53
Sutherland        15
Wigtown           15
Name: AccusedRef, dtype: int64


# Obtain the latitude and longitude of each county by getting the median of the all latitude and longitude records
# under the name of that county. Median is chose so that incorrect record will be ignored
# (the incorrect data can only be the outliers and will not influence the median),
# and convert them to dictionaries.
county_latitude = df.groupby("Res_county")["latitude"].median()
county_longitude = df.groupby("Res_county")["longitude"].median()
county_latitude_dict = county_latitude.to_dict()
county_longitude_dict = county_longitude.to_dict()
print(county_latitude_dict, county_longitude_dict)

{'Aberdeen': 57.146262, 'Argyll': 56.4006, 'Ayr': 55.457393, 'Banff': 57.666252, 'Berwick': 55.781124, 'Bute': 55.8274, 'Caithness': 58.4389, 'Clackmannan': 56.108159, 'Cromarty': 57.6806, 'Dumfries': 55.068632, 'Dunbarton': 56.83301, 'Edinburgh': 55.92826, 'Elgin': 57.648022, 'Fife': 56.253397, 'Forfar': 56.639836, 'Haddington': 55.955755, 'Inverness': 57.479549, 'Kincardine': 56.068975, 'Kinross': 56.211141, 'Kirkcudbright': 54.835888, 'Lanark': 55.674898, 'Linlithgow': 55.9716, 'Nairn': 57.585033, 'Orkney': 58.9809, 'Peebles': 55.651467, 'Perth': 56.395704, 'Renfrew': 55.874645, 'Ross': 56.83301, 'Roxburgh': nan, 'Selkirk': 55.548073, 'Shetland': 60.305229, 'Stirling': 56.281067, 'Sutherland': 58.249999, 'Wigtown': 54.868426} {'Aberdeen': -2.136575, 'Argyll': -5.4807, 'Ayr': -4.628716, 'Banff': -2.52426, 'Berwick': -2.011552, 'Bute': -5.0936, 'Caithness': -3.0937, 'Clackmannan': -3.747183, 'Cromarty': -4.0347, 'Dumfries': -3.608237, 'Dunbarton': -4.180209, 'Edinburgh': -3.275582, 'Elgin': -3.320025, 'Fife': -3.134239, 'Forfar': -2.893115, 'Haddington': -2.783795, 'Inverness': -4.237208, 'Kincardine': -3.715198, 'Kinross': -3.425103, 'Kirkcudbright': -4.049153, 'Lanark': -3.773292, 'Linlithgow': -3.6026, 'Nairn': -3.869079, 'Orkney': -2.9605, 'Peebles': -3.191026, 'Perth': -3.435877, 'Renfrew': -4.389173, 'Ross': -4.180209, 'Roxburgh': nan, 'Selkirk': -2.839351, 'Shetland': -1.294066, 'Stirling': -4.436736, 'Sutherland': -4.499998, 'Wigtown': -4.442783}


# Create a dataframe to record the county names, number of cases, and their latitudes and longitudes
county_name = []
county_witch_num = []
county_latitude = []
county_longitude = []
for i in range(len(county_witches)):
    current_county_name = county_witches.index[i]
    county_name.append(current_county_name)
    county_witch_num.append(county_witches[i])
    county_latitude.append(county_latitude_dict[current_county_name])
    county_longitude.append(county_longitude_dict[current_county_name])
data_df = pd.DataFrame({
    'name': county_name,
    'case number': county_witch_num,
    'latitude': county_latitude,
    'longitude': county_longitude,
})
data_df['case number'] = data_df['case number'].astype(float)
data_df = data_df.dropna().reset_index(drop=True) # remove the counties with na values and reset the index
data_df


data_df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 33 entries, 0 to 32
Data columns (total 4 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   name         33 non-null     object 
 1   case number  33 non-null     float64
 2   latitude     33 non-null     float64
 3   longitude    33 non-null     float64
dtypes: float64(3), object(1)
memory usage: 1.2+ KB


# Get the location of Scotland
Scotland = geolocator.geocode('Scotland')
Scotland

Location(Alba / Scotland, United Kingdom, (56.7861112, -4.1140518, 0.0))


# Create the map of Scotland. Zoom_start is smaller than Champaign map because we want to cover a larger area
Scotland_map = folium.Map(location=[Scotland.raw['lat'], Scotland.raw['lon']], zoom_start=6)
Scotland_map


# Add circles to the map, each representing the witch accused data of a county
for i in range(len(data_df)):
    folium.Circle(location = (data_df['latitude'][i], data_df['longitude'][i]),
                  radius = data_df['case number'][i]*100,
                  tooltip = data_df['name'][i]).add_to(Scotland_map)
Scotland_map


# Recreate the map of Scottish witchcraft. First, reload the map with the location of Scottish capital "Edinburgh",
# and set the zoom_start as 7 to focus on the surrounding area of the city. Then add the data of each county into the new map,
# but the radius will be 1000*the square root (this can be obtained by importing "math" and use "math.sqrt") of the case number,
# to get a smoothed result. Besides, add popup according to the format "County_name Case Number: number"
# (e.g., Edinburgh Case Number: 374.0), and set "fill" as True. Finally display the new map.

import math

Edinburgh = geolocator.geocode('Edinburgh, Scotland')


Scotland_map = folium.Map(location=[Edinburgh.latitude, Edinburgh.longitude], zoom_start=7)

for i in range(len(data_df)):
    radius = 1000 * math.sqrt(data_df['case number'][i])
    popup_text = f"{data_df['name'][i]} Case Number: {data_df['case number'][i]}"
    folium.Circle(location=(data_df['latitude'][i], data_df['longitude'][i]), radius=radius, popup=popup_text, fill=True,).add_to(Scotland_map)

Scotland_map


# Using titles in folium
folium.Map(location=[Scotland.raw['lat'], Scotland.raw['lon']], tiles = 'cartodbpositron', zoom_start=6)


# Using external background
folium.Map(location=[Scotland.raw['lat'], Scotland.raw['lon']],
           zoom_start=4,
           tiles='http://services.arcgisonline.com/arcgis/rest/services/NatGeo_World_Map/MapServer/tile/{z}/{y}/{x}',
           attr="Sources: National Geographic, Esri, Garmin, HERE, UNEP-WCMC, USGS, NASA, ESA, METI, NRCAN, GEBCO, NOAA, INCREMENT P")


# And this not necessarily need to be real world...
folium.Map(location=[0, 30],
           zoom_start=4, min_zoom=4, max_zoom=10,
           max_bounds=True,
           min_lon=0, max_lon=70, min_lat=-40, max_lat=40,
           tiles='https://cartocdn-gusc.global.ssl.fastly.net//ramirocartodb/api/v1/map/named/tpl_756aec63_3adb_48b6_9d14_331c6cbc47cf/all/{z}/{x}/{y}.png',
           attr='Textures and Icons from https://www.textures.com/ & https://thenounproject.com/')


# US states geojson file (obtained from https://github.com/python-visualization/folium/blob/main/examples/data/us-states.json)
US_states = "Data/us-states.json"


# Include the US States boundaries in the US Map
US_map = folium.Map(location=[42, -102], zoom_start=4)

folium.Choropleth(
    geo_data = US_states,
).add_to(US_map)

US_map


# US unemployment rate data (obtained from https://www.kaggle.com/datasets/aniruddhasshirahatti/us-unemployment-dataset-2010-2020?resource=download)
US_unemployment = pd.read_csv("Data/unemployment_data_us_state.csv")
US_unemployment


# Change column name to match the geojson data
US_unemployment = US_unemployment.rename(columns={'State': 'name'})
US_unemployment


# Visualize based on unemployment rate of January 2020
US_map = folium.Map(location=[42, -102], zoom_start=4)
folium.Choropleth(
    geo_data = US_states, # Geo_data to be used
    data = US_unemployment, # Data used for visualization
    columns = ['name', 'Unemployment_Rate_Jan_20'], # First column is the key to match, second column is the value to display
    key_on = 'feature.properties.name', # The matched key in geo_data
    fill_color = 'OrRd', # Seelct a color scheme
    line_opacity = 0.2, # Select line opacity
    legend_name= 'Unemployment Rate by State in January 2020', # Choose a name for the legend
).add_to(US_map)

US_map


# Add tooltip to display the state names
tooltip = folium.features.GeoJson(
    US_states,
    tooltip=folium.features.GeoJsonTooltip(fields=['name'], localize=True)
                                )
US_map.add_child(tooltip)
US_map


# Recreate the map of US unemployment. First, create a column 'Unemployment_Rate_2020_spring' in the DataFrame, as the average
# of the unemployment rates of January, Feburary, and March. Then, reload the map using the geocode of "USA" in gelocator.
# Next, create folium.Choropleth, and set the values of the newly-created column Unemployment_Rate_2020_spring, fill_color as
# "GnBu" (green and blue), and line_opacity as 0.3. Also, change the legend name to reflect the change of data.
# Finally, add the State names as tooltip, and display the new map.

US_unemployment['Unemployment_Rate_2020_spring'] = (US_unemployment['Unemployment_Rate_Jan_20']+
                                                  US_unemployment['Unemployment_Rate_Feb_20']+
                                                  US_unemployment['Unemployment_Rate_Mar_20'])/3
USA = geolocator.geocode('USA')

US_map = folium.Map(location=[USA.latitude, USA.longitude], zoom_start=4)


folium.Choropleth(
    geo_data = US_states,
    data = US_unemployment, 
    columns = ['name', 'Unemployment_Rate_2020_spring'],
    key_on = 'feature.properties.name', 
    fill_color = 'GnBu', 
    line_opacity = 0.3, 
    legend_name= 'Unemployment Rate spring 2020',).add_to(US_map)

tooltip = folium.features.GeoJson(
    US_states,
    tooltip=folium.features.GeoJsonTooltip(fields=['name'], localize=True)
                                )
US_map.add_child(tooltip)
US_map

	AccusedRef	AccusedSystemId	AccusedID	FirstName	LastName	M_Firstname	M_Surname	Alias	Patronymic	DesTitle	...	SocioecStatus	Occupation	Notes	Createdby	Createdate	Lastupdatedby	Lastupdatedon	Family of Accused	Cases	Date
0	A/EGD/10	EGD	10.0	Mareon	Quheitt	Marion	White	NaN	NaN	NaN	...	NaN	NaN	NaN	SMD	5/15/01 11:06	jhm	8/9/02 11:40	NaN	C/EGD/21	3/4/1661
1	A/EGD/100	EGD	100.0	Thom	Cockburn	Thomas	Cockburn	NaN	NaN	NaN	...	NaN	NaN	NaN	SMD	5/15/01 11:06	jhm	10/2/02 10:32	NaN	C/EGD/111	1591
2	A/EGD/1000	EGD	1000.0	Christian	Aitkenhead	Christine	Aikenhead	NaN	NaN	NaN	...	NaN	NaN	NaN	SMD	5/15/01 11:06	jhm	10/1/02 10:48	AF/LA/150	C/EGD/1011	6/5/1628
3	A/EGD/1001	EGD	1001.0	Janet	Ireland	Janet	Ireland	NaN	NaN	NaN	...	NaN	NaN	NaN	SMD	5/15/01 11:06	jhm	10/1/02 10:49	AF/LA/151	C/EGD/1012	6/5/1628
4	A/EGD/1002	EGD	1002.0	Agnes	Hendersoun	Agnes	Henderson	NaN	NaN	NaN	...	NaN	NaN	NaN	SMD	5/15/01 11:06	jhm	10/1/02 10:50	NaN	C/EGD/1013	3/7/1628
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
4378	20/10/1637	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4379	17/2/1642	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4380	24/10/1628	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4381	25/10/1651	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4382	4/1568	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

Geocoding with GeoPy¶

Task 1¶

Making Interactive Maps with folium¶

Task 2¶

Add a Circle Marker¶

Task 3¶

Customize Map Backgrounds¶

Geojson¶

Task 4¶

Task 5¶

	name	case number	latitude	longitude
0	Aberdeen	175.0	57.146262	-2.136575
1	Argyll	6.0	56.400600	-5.480700
2	Ayr	153.0	55.457393	-4.628716
3	Banff	9.0	57.666252	-2.524260
4	Berwick	126.0	55.781124	-2.011552
5	Bute	54.0	55.827400	-5.093600
6	Caithness	52.0	58.438900	-3.093700
7	Clackmannan	18.0	56.108159	-3.747183
8	Cromarty	2.0	57.680600	-4.034700
9	Dumfries	78.0	55.068632	-3.608237
10	Dunbarton	25.0	56.833010	-4.180209
11	Edinburgh	374.0	55.928260	-3.275582
12	Elgin	28.0	57.648022	-3.320025
13	Fife	382.0	56.253397	-3.134239
14	Forfar	82.0	56.639836	-2.893115
15	Haddington	543.0	55.955755	-2.783795
16	Inverness	45.0	57.479549	-4.237208
17	Kincardine	2.0	56.068975	-3.715198
18	Kinross	7.0	56.211141	-3.425103
19	Kirkcudbright	35.0	54.835888	-4.049153
20	Lanark	77.0	55.674898	-3.773292
21	Linlithgow	114.0	55.971600	-3.602600
22	Nairn	55.0	57.585033	-3.869079
23	Orkney	72.0	58.980900	-2.960500
24	Peebles	91.0	55.651467	-3.191026
25	Perth	109.0	56.395704	-3.435877
26	Renfrew	124.0	55.874645	-4.389173
27	Ross	74.0	56.833010	-4.180209
28	Selkirk	21.0	55.548073	-2.839351
29	Shetland	28.0	60.305229	-1.294066
30	Stirling	53.0	56.281067	-4.436736
31	Sutherland	15.0	58.249999	-4.499998
32	Wigtown	15.0	54.868426	-4.442783

	State	Unemployment_Rate_Jan_20	Unemployment_Rate_Feb_20	Unemployment_Rate_Mar_20
0	Alabama	2.7	2.7	3.5
1	Alaska	6.0	5.8	5.6
2	Arizona	4.5	4.5	5.5
3	Arkansas	3.5	3.9	4.8
4	California	3.9	2.5	5.3
5	Colorado	2.5	2.8	4.5
6	Connecticut	3.7	3.9	3.7
7	Delaware	4.0	5.2	5.1
8	D.C.	5.2	2.8	6.0
9	Florida	2.8	3.1	4.3
10	Georgia	3.1	2.7	4.2
11	Hawaii	2.7	2.7	2.6
12	Idaho	2.8	3.4	2.6
13	Illinois	3.5	3.1	4.6
14	Indiana	3.1	2.8	3.2
15	Iowa	2.8	3.1	3.7
16	Kansas	3.1	4.2	3.1
17	Kentucky	4.3	5.2	5.8
18	Louisiana	5.1	3.2	6.1
19	Maine	3.1	3.3	3.2
20	Maryland	3.3	2.8	3.3
21	Massachusetts	2.8	3.6	2.9
22	Michigan	3.8	3.1	4.1
23	Minnesota	3.2	5.4	3.1
24	Mississippi	5.5	3.5	5.3
25	Missouri	3.5	3.5	4.5
26	Montana	3.5	2.9	3.5
27	Nebraska	3.9	3.6	4.2
28	Nevada	3.6	3.6	6.3
29	New Hampshire	2.6	3.8	2.6
30	New Jersey	3.8	4.8	3.8
31	New Mexico	4.8	3.7	5.9
32	New York	3.8	3.6	4.5
33	North Carolina	3.6	2.2	4.4
34	North Dakota	2.3	4.2	2.2
35	Ohio	4.1	3.2	5.5
36	Oklahoma	3.3	3.3	3.1
37	Oregon	3.3	4.7	3.3
38	Pennsylvania	4.7	3.4	6.0
39	Rhode Island	3.1	2.5	4.6
40	South Carolina	2.4	3.3	2.6
41	South Dakota	3.4	3.4	3.3
42	Tennessee	3.3	3.5	3.5
43	Texas	4.5	2.5	4.7
44	Utah	2.5	3.4	3.6
45	Vermont	2.4	2.6	3.2
46	Virginia	2.7	3.8	3.3
47	Washington	3.9	4.9	5.1
48	West Virginia	5.0	3.7	6.1
49	Wisconsin	3.5	2.7	3.4
50	Wyoming	3.7	5.8	3.7