Final Project

Introduction

Obesity is a growing epidemic that is not only posing a problem to the health and economy of Americans, but to a wide range of people around the world. Globally, there are over 1 billion overweight adults, and among them 300 million are obese; further, in America, two out of three adults are either obese or overweight, causing them to be more susceptible to chronic diseases such as type 2 diabetes, cardiovascular disease, hypertension and stroke, and others. While genes may determine body weight by controlling calorie intake, the continuing rise in obesity is likely to be attributed to the change in society and people's values towards the concept of convenience and nutrition. Modernization, urbanization, economic growth, and globalization are a few that have been listed at the top due to their drastic impact to the life-style of modern people. Modernization and urbanization transform a place from a less developed one to a wealthier and more powerful state with a greater demand for higher standard living; thus, inhabitants would be demanding new sources of stimulants, like tastier food, even if it means higher proportions of fat and sugar. In addition, economic growth allows transnational corporations (TNCs) such as McDonald's to branch out to other countries causing homogenization; in other words, globalization is causing and spreading obesity to the world.

Although obesity seems to be correlated with wealthy people, as they can afford to eat a larger range of food. However, a World Health Organization (WHO) study and a Cornell University study have respectively shown that obesity is prevalent among developing countries due to under-nutrition; a deficiency in micronutrients as a result of eating cheap (fast) foods, also the fact that low-income working class people resort to eating fast food is leading them to obesity as a result of their hectic schedule. These studies have led me to realize that I've actually seen more overweight people in poorer regions rather than in rich neighborhoods of Los Angeles, such as the downtown area and on public transportation. This intrigued me into wondering whether there is a relationship between income and obesity.

In order to commence in answering my curiosity, my group and I have come up with a hypothesis that speculates the current obesity phenomenon relating to the distribution of income and McDonald's in LA county. We believe that wealthy people in this county have a lower tendency of becoming obese due to their compliance of the Hollywood trend; while low-income people have a higher chance of becoming so since they can barely afford cheap food and they simply do not have the time to dine properly in restaurants other than fast food chains.

Method

We utilized four data frames in our project to display the correlation between income and obesity in two selected cities specifically within LA county; namely, Beverly Hills and downtown LA. This is because we believe that these two locations have the largest difference in income; thus a significantly different prevalence of obesity.

The first data frame illustrates the prevalence of obesity in LA county paying special attention to Beverly Hills and downtown LA. We downloaded a health district base map from the Los Angeles County GIS Data Portal since obesity is considered a chronic disease and would be best to be shown on a map used by the Department of Public Health to plan and manage health service delivery across the County. Further, in order to display the percentage of obesity in different regions of LA, we found statistical data from the County of LA Department of Health Services website. Since we weren't able to locate a obesity shapefile on the web, we edited the attribute table of the health district map and added a new field called "obesity percentage"; that way we could assign each percentage value to its designated location in the county. Conversely, as the health district map did not show cities, we had to download a polygon file from the UCLA GIS data site which sorted cities by their zip codes. This way we were able to locate and make Beverly Hills and Downtown LA visible on this data frame, enabling us to identify the relationship between obesity in a wealthier region (Beverly Hills) and a poorer one (downtown LA).

The second data frame portrays the locations (28 in total) of McDonald's within West LA bordered by the 405 and 10 highway and the boundaries of downtown LA. The reason why we chose locations within the periphery of 405 and 10 is that there are no McDonald's at all in the city of Beverly Hills while there are 21 just in downtown LA. All the locations were manually selected from a website called fastfoodmap.com and we have utilized every location provided there which falls into our determined boundaries. After doing so, we created an excel file and inserted the addresses of McDonald's in the following categories: addresses, city, state, and zip code. This is to allow us to use the process of geocoding which converts descriptive locations (addresses) into georeferenced locations (x,y location). The distribution of McDonald's on the map is complimented with the previously created boundaries of Beverly Hills and downtown LA to convey the notion that McDonald's target customers in the poorer region of downtown LA and not Beverly Hills.

The third data frame depicts the income distribution in the city of Los Angeles in relation to Beverly Hills and downtown LA. We extracted median household income data in each zip code from a website called "zip atlas". In order to display this data on our map, we added a new field to the attribute table on another copy of the health district base map, then matched each income value to each location just as we did for the obesity data frame. After that, we created sixteen ranges for each income value to fall into and then gave all the values a color ramp. As usual, the boundaries for Beverly Hills and downtown LA were included into this map to indicate that Beverly Hills is where people with higher income reside and downtown LA is where people with lower income situate.

The last data frame is a consolidation of all three maps into one map and this will show the obvious correlation between obesity and income in a spatial sense; also, through illustrating the relationship between McDonald's locations and income distribution, it would explain the fact that McDonald's had targeted low-income customers and this had led them to be more susceptible to obesity.

Results

As we looked at the first map, Prevalence of Obesity in Los Angeles, we were able to notice the trend as to where obesity was most and least dominant. Areas where obesity was most dominant were East LA, San Antonio, Southeast and Compton were where obesity occurs most among population; while areas of least dominance were Alhambra and Glendale. Nonetheless, when we performed a comparative analysis between the selected regions, Beverly Hills and downtown LA, it was obvious that downtown LA had the higher percentage of obesity, 21% to above 25%, where Beverly Hills fell into the 16 to 20% range. Looking at the McDonald's Locations in Selected Regions map, we noticed that there was an obtrusive comparison between Beverly Hills and downtown LA. As shown, there were 21 branches of McDonald's in the downtown region, while there was absolutely no branch in Beverly Hills, except for the 7 that were located in the out skirts of the city. For the third map, Median Household Income in Los Angeles, the focus of this map was to show the average amount of income each zip code holds. The color ramp suggested that the darker the blue was, the higher the income was for that region; Beverly Hills was one of the darkest ones with over $75000. In contrast, downtown LA had some of the lowest income values with $5000 to $24999.

Conclusion/ Discussion

In conclusion, the four maps have justified my hypothesis that wealthy people in LA county have a lower tendency of becoming obese; while low-income people have a higher chance of becoming so. The maps respectively indicated that in Beverly Hills, average income was above $75000, there was no McDonald's within the region and had a low prevalence of obesity; in the mean time, downtown LA had a low average income, simultaneously a location of 21 branches of McDonald's, also home to people contributing to the highest percentage of obesity in the county. There is definitely a correlation between income and obesity.We assume that wealthier people in Beverly Hills are less obese because Los Angeles is the home of Hollywood stars and unquestionably there are standards or status quos regarding fashion, fitness and health where these high-end people would want to comply; consequently, going to the gym and eating healthy foods with less fat and sugar have become a habit which acts as a obesity repellant for them. In contrast, low-income people are much less likely to comply to these trends due to their limited monetary resources. In addition, low-income people are limited to eating cheaper foods with less nutritious values because they simply cannot afford better quality food; also, due to the fact that many of them work in labor intensive industries such as being cleaners or construction workers; their working schedule is highly structured and inflexible and this causes them to be deprived of time to go get a decent meal and so they would go for fast food often; this is also the reason why McDonald's and other fast food chains target working class people and this explains the significant abundance of McDonald's in downtown LA in contrast to Beverly Hills. It is depressing to realize that Los Angeles is in a situation where obesity is synchronized with low-income population due to fast food chains being one of the affordable options for them; however, the gradually popularized community farms would alleviate this problem as these community-supported farms sell reasonably priced and nutritious food to the general population.

Works Cited

"WHO | Obesity and Overweight." Web. 14 Sept. 2011. <http://www.who.int/dietphysicalactivity  /publications/facts/obesity/en/>.

"Obesity In America | The Epidemic." Weight Loss Tips | Visceral Fat Loss Tips. Web. 14 Sept. 2011. <http://www.drkalsweightlosstips.com/obesity-in-america.html>.

Park, Alice. "Working Parents and Family Diets: Too Busy to Eat Right - TIME." Breaking News, Analysis, Politics, Blogs, News Photos, Video, Tech Reviews - TIME.com. 10 Sept. 2009. Web. 14 Sept. 2011. <http://www.time.com/time/health/article/0,8599,1921349,00.html>.

"Median Household Income in Los Angeles, CA by Zip Code." Zip Code, Area Code, City & State Profiles | ZipAtlas. Web. 14 Sept. 2011. <http://zipatlas.com/us/ca/los-angeles/zip-code-comparison/median-household-income.htm>.

McDonalds locations
http://www.fastfoodmaps.com/
http://www.mcdonalds.com/

Prevalence of Obesity
County of LA, Department of Health Services, Public Health
http://publichealth.lacounty.gov/ha/reports/habriefs/lahealth073003_obes.pdf

UCLA GIS data
http://gis.ats.ucla.edu/

Los Angeles County Enterprise GIS
http://egis3.lacounty.gov/eGIS/

Health Districts shapefile
http://egis3.lacounty.gov/dataportal/index.php/2010/09/28/health-district-hd/

Lab 5: Spatial Interpolation

This week’s lab required us to estimate unknown values with known values, and this process is called spatial interpolation. We were given a website for us to download rainfall data measured by a number of centers that were distributed in Los Angeles County, and the objective was to create maps that display the pattern of rainfall in the region in given criteria; specifically seasonal total, normal rainfall and the difference between these two values. To illustrate these maps, I have used two methods; namely, Inverse Distance Weighted (IDW) and Kriging. The mapping results seem to suggest that there is more rainfall in the eastern region of LA, which is where the mountain ranges are. The reason behind this higher value is the fact that the physical rise in elevation of the mountain forces air to rise and saturate, forming rain clouds; on the other hand, rainfall is particularly lowest in the northern part of LA. Personally, I believe that IDW is a better technique to display this data because it allows one to analyze the data clearly in a visual way; notice how the interpolated lines of classified rainfall regions are smoother than the ones generated by the kriging method. In addition, since we have dense values recorded from those stations, IDW would work best as this density is enough to allow this method to capture the extent of local surface variation needed for analysis.   

Quiz 2

 1) (go to attribute table of cntry02, click pop_cntry, right-click choose sort descending)
China
India
United States
Indonesia
Russia
Brazil
Pakistan
Japan
Bangladesh
Nigeria

2) (go to river's attribute table, group Amazon together, count the number of rivers)
15 rivers

3) (go to selection "attribute table of the river layer, find the 2 rivers, and select them using ctrl and mouse click, then go to selection > select by location, selection method is select features from, target layer is cities, then source layer is rivers, spatial selection method is "features within a distance of source layer" check apply a search distance, 500 km, then ok, then the map would light up in a region, then go to cities, attribute table and print scree)

4) (go to country's attribute table, sort descending, find and select Iran > select by location > select features from > target layer is cntry02 > source layer is cntry02 > spatial selection method is within a distance > apply a distance of 300km > ok > open country;s attribute table, show selected > use statistics on pop_cntry)

 total population of countries: 516490670

5)  (go to country;s attribute table, select the Y of landlocked countries, click show selected records, right click on pop_cntry, click on sort descending, look for the most and least populous countries)

Most populous country: Ethiopia 53142970
Least populous country: Vatican City 860

6) (go to cities' attribute table, select Veszprem, go to select by location, select features from > target layer is cntry02 > source layer is cities > target layer's features are within a distance of the source layer feature > apply a search of 300 km > go to country's attribute table, click selected features)

Poland
Romania
Yugoslavia
Czech Republic
Hungary
Austria
Slovakia
Croatia
Bosnia & Herzegovina
Slovenia

7) (go to attributes table of country > find and select Chad > select by location > "select features from" > target layer is cntry02 > source layer is cntry02 > spatial selection method is "Target layer features touch the boundary of the source layer feature" > OK > open attribute table of country > show selected records > exclude Chad and copy the rest of the countries that border it.

Cameroon
Central African Republic
Libya
Niger
Nigeria
Sudan

8) (ArcToolBox > Analysis Tool > Statistics > Frequency > Input is cities > Frequency fields is CNTRY_NAME > open the newly created frequency table > sort descending on frequency tab > copy the first 5)

97 Russia
93 United States
72 Thailand
67 Turkey
50 Cote D'Ivory and Poland

9) (use measure and approximate the length of the three Niles flowing in Sudan)
2838 + 322 + 614 = 3774 km


10) (open attribute tables of lakes and eyeball to see which country has the most number of lakes > select all of the lakes of one country on attribute table > go to select by location > target layer is lake, source layer is lake > look at how many lakes there are in that country, repeat for top 5 countries)

1516 Russia

1340 Canada

743 USA

219 China

168 Sweden

11)  dissolve lakes layer > input lakes > dissolve: country_name > add field area > calculate geometry in sq km > sort in descending order

443517km2- Canada
196848km2- United States
138250km2- Russia
70900km2- Kazakhstan
53530km2-United Republic of Tanzania

12) Map showing world countries and lake boundaries.

 

Join dissolved lake with country > symbology > value > area > normalization > population

Lab 4 Fire Model

      The combination of the digital elevation model (DEM) and land cover analysis used in this lab has introduced me to a new way of analyzing spatial data. Using the DEM shape file provided in the class website, as well as the GIS techniques that i have learned so far; namely, creating slope and reclassification, i was able to create the slope map which shows how susceptible certain areas of the mountain range is to fire. As you may know, the elevation of slope determines the rate of fire spread, therefore using this method to evaluate the hazard level in mountainous areas would allow better precautions to be run in the most efficient places. On the other hand, to create the Land cover map, reclassification took up a large proportion of its creation. I downloaded the land cover data from the FRAP website, and the data include a wide range of different land types, such as agriculture and urban. The reclassify tool allowed me to rank each land type according to their susceptibility to catching fire. As a result, water was classified as 1 because it is the least likely land type to catch fire, while dry wood or shrubs are given a high number as they are vulnerable to fire. Lastly, the largest map on the top left corner is the result of the slope and land cover map combined. With reclassified data, i was able to show different levels of risks within and out of the fire perimeter.

     When creating these maps, it was inevitable to encounter all sorts of problems as this was my first trial of these ArcGIS functions. Some minor problems include the fact that i have always forgotten to check the spatial analyst box under customize; thus, my tools always failed to work. Also, i had a hard time choosing from the vast variety of color ramps; picking one that suits my personal liking and one that will suffice to convey the purpose of the maps was troublesome. A considerably major challenge i encountered was the part where i had to decide what number to assign to each land type in order to reclassify; i first followed the general rankings of the land types on the tutorial which ranked them as light to heavy hazard, and i assigned my ten land types in numeric order. However, i found out later that these numbers can be overlapped according the my own preference and of course basing on the reasonable assumptions that suggest which land cover types fall into which risk level. This lab was a fun experience in the course of this class.

Lab 3 - Landfill Project

Research Brief of Final Project - "Does Income Influence the Prevalence of Obesity"

1) Project Topic
"Does Income Influence the Prevalence of Obesity in LA?" Since the fast food industry seems to be more prominent in the city of LA, i believe that the industry must have an impact on the population. Further, i realized if a poorer person eats fast food often due to the need to accommodate their hectic schedule and monetary limits with fast dining time/cheap food, wouldn't it lead to obesity? In addition, poorer people eat fast food because they may not be aware of the health consequences they may cause, and that they are cheap; on the other hand, wealthier people may be more cautious in what they eat, and so they wouldn't mind paying more for equally efficient prepared foods but of healthier quality. Thus the topic arose.

2) The Progress on Data Collection

We will use data from "County of LA, Department of Health Services, Public Health" which will show distribution of obesity around LA. Secondly, we would use Google map to locate the 5 major fast food franchises which are McDonalds, KFC, Burger King, Taco Bell and Carl's Junior. Lastly, we would utilize the income distribution information from the USGS website.

3) Propose Methods
There will be 3 maps:
1) map showing distribution of income/household
2) map showing distribution of fast food
3) map showing the prevalence of obesity

These 3 maps would show the relationships between income level and obesity in LA.

Example of an obesity map:

  Source:
County of LA, Department of Health Services, Public Health
http://publichealth.lacounty.gov/ha/reports/habriefs/lahealth073003_obes.pdf

Lab Quiz - Geocoding (Marijuana Dispensaries in LA)

I am against the decision to require marijuana dispensaries in the city of Los Angeles to be at least 1000 feet away from places where children congregate. This is because from the 50 dispensaries I have found using Google map, at least half of them are located within the 1000 feet buffer of schools, parks and libraries. If the policy was passed, it would require the state to utilize a lot of budget to eradicate those dispensaries that are already in the area; given today's economy state-wise, it is not a good idea to do so. Furthermore, even though these dispensaries are within close proximity to where children congregate, but in order for a person to access these dispensaries, they would need to have proof of doctor's prescription and a cannabis card; thus, children wouldn't be able to acquire marijuana as they wish. I believe that if the city council were to pass this new policy, the policy would have to apply to all other addictive products as well, such as cigarettes and alcohol, since these addictive products are even easier to be attained by the general population.

Recent scientific discoveries have found out more positive effects of marijuana, it not only treats multiple diseases, It also elevates people's mood. Hopefully as marijuana's benefits become more well-known by people, it would be acknowledged by the city council.

p.s. handed in on time through email.