Taylor,+K

DATA MANAGEMENT



PROPOSAL

Sunshine is said to improve ones mood. If an area is lacking in sunshine, does that mean people's moods would simply deteriorate? Alaska is known for its constant sunshine in the summer, and constant darkness in the winter.

When looking at the graph provided above, it is evident that significantly more suicides occured in Alaska, than the states. I am curious to learn whether or not Alaska's suicide rates have any correlation with the amount of sunlight. //**From 1990 to 2009, does the increased amount of sunlight during Alaska's summer months decrease the amount of suicides in that region?**//

VARIABLES Amount of Sunlight //in Alaska (between 1990 & 2009) - INDEPENDANT VARIABLE// & Suicide Rates //in Alaska (between 1990 & 2009) - DEPENDANT VARIABLE//

HYPOTHESIS

I do believe that weather affects ones mood. I think people have much more energy and enthusiasm when the sun is out. I assume that incessant bad weather, or lack of sun, could worsen a (depressed) person's situation, potentially resulting in suicide.

Considering the extended amount of sunlight in Alaska vs. the extended amount of darkness throughout the year//, **I think i will find an increase in suicide rates during the months of decreased sunlight in Alaska (between the years 1990 and 2009).**//

I also believe that **the regions within Alaska that recieve longer periods of sun** (eg - Northern Alaska with 24 hours of sunlight in the summer) **will show lower suicide rates, when compared to those with smaller periods of sunlight**. (eg - Northern Alaska, with constant darkness in the winter.)


 * OVERVIEW/ BACKGROUND INFORMATION**

One reason why I chose this topic, was to learn whether or not my hypothesis was correct. I had always assumed that more depressions and suicides would occur during the 'dark months' of Alaska, and when there is very little sun. However, as I researched more before beginning this project, I noticed articles and graphs displaying how more suicides have occurred during the sunny months. I chose this topic so i could find evidence proving one argument to be true.

//[The most northern part of Alaska "has a two-month winter period in which the sun doesn't rise.// //In the summer, all of Arctic Alaska gets 24-hour sunlight for at least one day at the solstice. [This region] has continuous daylight for 85 days. South of the circle, every town has a night every day, even if it's quite brief.// //Long dawns and dusks can make the day appear longer than it actually is. Even as far south as Anchorage, it's possible to read a newspaper outdoors at 2 a.m. -- two hours after sunset -- in the days around the summer solstice."] ([|www.alaska.com])//

//["The// Arctic Circle //marks the southern limit of the area where the sun does not rise on the winter solstice or set on the summer solstice. Continuous day or night ranges fromthe one day at Arctic Circle to six months at the North Pole. The imaginary line of the Arctic Circle extends in an arc across the upper third of Alaska, at approximately 66 degrees 33' north of the equator. This is the latitude at which both summer and winter solstices occur.// //"Solstice" means the point at which the sun is at its greatest distance from the equator. In the summer, this occurs on June 20 or 21, making that the longest day of the year."] ([])//



 RAW DATA & INFORMATION
 * **__Anchorage (//southcentral//), Alaska__ ** || 
 * (Southcentral Alaska) Latitude 61.17 north, longitude 150.02 west. ** Also near this latitude: Bethel, Eagle River, Girdwood, Glennallen, Palmer, Wasilla, Valdez, Lake Clark National Park, Wrangell St-Elias National Park; Whitehorse, Yukon Territory; Oslo, Norway; Stockholm, Sweden; St. Petersburg, Russia. || [[file:Anchorage (southcentral), Alaska.docx]] ||
 * **__Juneau, Alaska __** || 
 * (Northern Panhandle of Southeastern Alaska.) Latitude 58.37 north, longitude 134.58 west. ** Also near this latitude: Glacier Bay National Park, Kodiak; London, Berlin, Moscow. || [[file:Juneau, Alaska.docx]] ||
 * **__Nome (//western//), Alaska __** || 
 * <span style="color: #4f81bd; font-family: 'Arial','sans-serif'; font-size: 10pt; mso-fareast-font-family: 'Times New Roman'; mso-themecolor: accent1;">(Western Alaska.) Latitude 64.52 north, longitude 165.45 west. **<span style="font-family: 'Arial','sans-serif'; font-size: 10pt; mso-fareast-font-family: 'Times New Roman';"> Also near this latitude: Galena, Fairbanks, St. Lawrence Island. || [[file:Nome (western), Alaska.docx]] ||


 * **This set of raw data represents the suicide rates of Alaska, between the years 1990-2008.**[[image:urban_vs_rural.gif align="center" caption="Urban vs Rural Suicide Rates in Alaska"]]




 * This graph** **displays the number of suicide //attempts// within the years 1994 and 1998. As you can see, a significant amount of people who attempted suicide occured within the ages of 15 and 30, most being Alaska natives.** ||
 * **Here is a link to an article written by 'The Associated Press' on May 29, 2009. It discusses the increasing suicide rates in Alaska. "Preliminary numbers from the state Bureau of Vital Statistics indicate that the Alaska rate of 24.6 suicides per 100,000 people last year was the highest in more than a decade, and a 25 percent increase since 2005", and that the underlying reason to this is unknown. (Perhaps it is connected to the amount of sunlight vs darkness.)**

[|Unknown factor behind Increasing Suicide Rates (ALASKA) - ARTICLE] ||


 * **I have found a website that contradicts my hypothesis. Lee Bowman’s article, “The dark side of more sunlight: Higher Suicide Rates”, explains how regions with more sunlight will have higher suicide rates. (Alaska with 135 suicides in 2006, and 19.9 suicides per 100,000)**

[|more sunlight = more suicides? ARTICLE] ||


 * **Here is yet another article written about increasing rates of suicide due to more sun. Kate Ravilious writes her article "Too much Sunshine Spurs Suicides in Arctic, Study Says" for National Geographic News. (May 8, 2009)**

["There is so much light that people can't take it," said lead study author Karin Sparring Björkstén of the Karolinska Institutet in Sweden... When they looked at the data on a month by month basis, it became clear that suicides were far more frequent in summer, particularly in the north, where the extended summer days are most pronounced.]

[|Too Much Sunshine Spurs Suicides in Arctic, Study Says - ARTICLE] ||

<span style="color: #000000; font-family: 'Times New Roman',Times,serif; font-size: 14px; line-height: 115%; mso-themecolor: accent1; msothemecolor: accent1;"> __**<span style="color: black; font-family: 'Kristen ITC'; font-size: 14pt; line-height: 115%; mso-themecolor: text1;">Sampling Techniques **__


 * __Stratified Sample__** - A population may include groups of member that share common characteristics such as age or gender. This was prevalent within a few of my graphs, such as 'suicide and age in Alaska', 'Rural and/or Urban Suicide rates', and 'Suicide and Gender/ Race'. I chose to use raw data that categorized the suicidal population of Alaska because it allowed me to get a better idea as to who is more likely to commit suicide, when during the year, and where in that region. From this information i can then refer it back to my independant variable; the amount of sunlight, and see if there really is any correlation between the two.

__**<span style="color: black; font-family: 'Kristen ITC'; font-size: 16pt; line-height: 115%; mso-themecolor: text1;">Bias **__

<span style="color: black; font-family: 'Kristen ITC'; font-size: 14pt; line-height: 115%; mso-themecolor: text1;"> Suicide is a very sensitive topic. When asked personal questions about the suicide, family members for example, may not be willing to provide the underlying reasons as to why the suicide originally occured. Therefore, the set of data could not accurately represent that of the entire population being surveyed.
 * __Sampling Bias__** - When a sample or sampling frame does not reflect the characteristics of the population. This existed within the information i found when one variable in a set of data, for example Race or Gender, does not represent that of all suicide cases in Alaska. Many suicides were committed by females //and// males, and a large amount of varying races.
 * __Response Bias__ -** When the participants deliberately give false or misleading answers. Respondants may want to influence results or they may be afraid or embarassed to answer sensitive questions honestly.


 * __ ONE VARIABLE ANALYSIS __**
 * MEAN, MEDIAN MODE & STANDARD DEVIATION[[file:ONE VARIABLE CALCULATIONS.docx]]**

I completed a one variable analysis through the use of **mean, median and mode** as well as //**standard deviation**//. I compared the suicide rates in Alaska for the summer months (April-Sept) with the suicide rates of the winter months (Oct-March). From these calculations I noticed that there was a higher mean (average) of suicides committed in the summer months. In other words, there was a __higher amount__ of suicides occurring within the months of **more** sunlight.

Also, there was a smaller standard deviation with the amount of suicides committed in the winter than the summer. That meant that the data was more __compact__, revealing that //fewer suicides// actually occur in the winter months (with very little sunlight) in comparison to the summer months (with extreme amounts of sunlight) with a higher, more spread out standard deviation.

__**<span style="color: black; font-family: 'Kristen ITC'; font-size: 14pt; line-height: 115%; msoansilanguage: EN-US; msobidifontfamily: Arial; msobidilanguage: AR-SA; msofareastfontfamily: 'Times New Roman'; msofareastlanguage: EN-US;">TWO VARIABLE ANALYSIS **__

=
Determining a relationship between independant and dependant variables can be noticed if a pattern exists. __Independant (explanatory) variable__ - amount of daylight __Dependant (response) Variable__- suicide rates in Alaska between 1990 and 2009. When one variable changes, it is proportional to the changes of another. For example, i hypothesized that suicide rates would increase as the amount of sunlight decreased in Alaska. To take a better look at the variables given, I have created a scatter plot including the information i recieve from both. From here a positive, weak correlation was shown through the completion of a correlation coefficient.======

=
<span style="color: black; font-family: Arial; font-size: 10pt; line-height: 115%; mso-ansi-language: EN-US; mso-bidi-language: AR-SA; mso-fareast-font-family: 'Times New Roman'; mso-fareast-language: EN-US; msoansilanguage: EN-US; msobidilanguage: AR-SA; msofareastfontfamily: 'Times New Roman'; msofareastlanguage: EN-US;">For my topic, it is possible to introduce both variables; I chose to complete a //linear regression// involving amount of sunlight in Alaska, and suicide rates in that same region. The independant (amount of sunlight) and dependant (suicide rates between 1990 and 2009) variables was labelled on the 'x' and 'y' axis' and the linear line was calculated. ======

=
<span style="color: black; font-family: Arial; font-size: 10pt; line-height: 115%; mso-ansi-language: EN-US; mso-bidi-language: AR-SA; mso-fareast-font-family: 'Times New Roman'; mso-fareast-language: EN-US; msoansilanguage: EN-US; msobidilanguage: AR-SA; msofareastfontfamily: 'Times New Roman'; msofareastlanguage: EN-US;"> ======

=
<span style="color: black; font-family: Arial; font-size: 10pt; line-height: 115%; mso-ansi-language: EN-US; mso-bidi-language: AR-SA; mso-fareast-font-family: 'Times New Roman'; mso-fareast-language: EN-US; msoansilanguage: EN-US; msobidilanguage: AR-SA; msofareastfontfamily: 'Times New Roman'; msofareastlanguage: EN-US;"> ====== ****The Correlation Coefficient results (positive, weak) did not necessarily match with the graph displayed. The graph showed a positive, strong correlation. However, due to the outlier (175, 47) – it skewed the results of the correlation coefficient. If I had removed the outlier before calculating, the results would have been more accurate, in regards to the relationship shown on the graph.

=
<span style="color: #4f81bd; font-family: 'Kristen ITC'; font-size: 16pt; line-height: 115%; mso-themecolor: accent1; msothemecolor: accent1;"> <span style="font-family: 'Times New Roman','serif'; font-size: 10pt; line-height: 115%; mso-ansi-language: EN-US; mso-bidi-font-size: 11.0pt; mso-bidi-language: AR-SA; mso-fareast-font-family: 'Times New Roman'; mso-fareast-language: EN-US;">**People may suffer from SAD (Seasonal Affective Disorder). This is a depression that occurs each year at the same time, usually starting in fall or winter and ending in spring or early summer. In Alaska, those suffering with SAD may form a more severe case due to the extended amount of sun, and extended amounts of darkness. Thus, potentially resulting in suicide.** <span style="font-family: 'Times New Roman','serif'; font-size: 10pt; line-height: 115%; mso-ansi-language: EN-US; mso-bidi-language: AR-SA; mso-fareast-font-family: 'Times New Roman'; mso-fareast-language: EN-US;"> //more information on SAD -// [|**S.A.D - INFO**] ====== <span style="color: #4f81bd; font-family: 'Kristen ITC'; font-size: 26pt; line-height: 115%; mso-themecolor: accent1;"> <span style="color: #4f81bd; font-family: 'Kristen ITC'; font-size: 16pt; line-height: 115%; mso-bidi-font-family: Arial;">

CONCLUSIONS After researching my topic, Suicide Rates vs. Amount of Sunlight in Alaska (1990-2009), I was surprised to learn that my hypothesis was deemed incorrect. Through use of graphs, archives and articles I was able to see that there were higher suicide rates within the months of increased sunlight. When looking at annual precipitation levels, they too correlated with the suicide rates, when being compared. As the levels of precipitation //increased//, the amounts of suicides committed //decreased//. Also, through the calculations made for one variable data, I could see that this was occurring again. Like I explained under the ‘One Variable Analysis’ section, “there was a higher **mean** (average) of suicides committed in the summer months. In other words, there was a __higher amount__ of suicides occurring within the months of **<span style="font-family: 'Calibri','sans-serif'; mso-ascii-theme-font: minor-latin; mso-bidi-font-family: 'Times New Roman'; mso-bidi-theme-font: minor-bidi; mso-hansi-theme-font: minor-latin;">more ** sunlight.

Also, there was a smaller standard deviation with the amount of suicides committed in the winter than the summer. That meant that the data was more __compact__, revealing that //<span style="font-family: 'Calibri','sans-serif'; mso-ascii-theme-font: minor-latin; mso-bidi-font-family: 'Times New Roman'; mso-bidi-theme-font: minor-bidi; mso-hansi-theme-font: minor-latin;">fewer suicides // actually occur in the winter months (with very little sunlight) in comparison to the summer months (with extreme amounts of sunlight) with a higher, more spread out standard deviation” I read a few articles contradicting my hypothesis. One article stated “suicide rates are consistently highest between May and July, a Scripps review of national suicide statistics for the two most recent years available, 2005-2006, showed. Contrary to popular belief, December, with the shortest days and presumed holiday-related stress and depression for many, is the month with the lowest suicide rate.” Another wrote, “ There is so much light that people can't take it. When they looked at the data on a month by month basis, it became clear that suicides were far more frequent in summer, particularly in the north, where the extended summer days are most pronounced.” When looking at all the various types of data found, it appears much more likely for people in Alaska to commit suicide during the months of prolonged sun. It appears to be that my hypothesis and assumptions on this topic have been incorrect all along. Therefore, **There is an increase in suicide rates within the months of increased sunlight.**

=
<span style="font-family: 'Times New Roman','serif'; font-size: 10pt; line-height: 115%; mso-fareast-font-family: 'Times New Roman';"> //Alaska Suicide Attempts//. (n.d.). Retrieved December 7, 2009, from []======

The Anchorage Daily News. (2009). //Daylight and Darkness//. Retrieved December 4, 2009, from []

=
Minister responsible for, S. (2001). //Table 1.4.19.1 Suicide (ICD-9 E950-E959), crude count and confidence interval, by sex, Canada, provinces, territories and health regions, 1996//. Retrieved November 9, 2009, from []======