IntroductionDid you know that nearly every 40 seconds one decides to take their own life, that is nearly 3,000 deaths a day (Suicide Data, 2015)? Did you know that approximately 1.3 million adults (anyone 18 or older) attempted suicide in 2013 (U.S. Department of health and human services, 2014)? Did you know that in the same year 1.1 million people also reported making suicide plans (U.S. Department of health and human services, 2014)? But what is it that can make so many people take or consider such a drastic decision? There are many causes of suicide some including psychological, social, biological and economic factors, but one significant one is income (Machado, 2015). Income is something that we don’t have control over, even if we may feel like we do, the same goes for our mental state. One minute everything can change, like if someone were to lose their job they would not have the same income for a bit and have a difficult time paying for what they would normally be able to afford, causing stress and possibly putting one in an unhealthy mental state and possibly leading to suicide (Boseley, 2015). My research question is: does a lower average annual income result in higher suicide rates, vice versa? To answer this I will be studying anyone who has been a Canadian resident from any time in between 1976-2013. The two variables that I will be studying are suicide rates per 100,000 people and average income in Canada in from 1976 – 2013, in Canadian dollars. A census was used to collect the data as the surveys used to collect the data by the Canadian government were mandatory to respond to. I chose this topic as I feel there is a positive correlation between both variables and an external factor like society causes both to change (Penner, 2017). I’d also personally like to know if these two variables have a causation factor. I also find this topic to be very interesting due to its psychological aspect when it comes to learning how one’s financial state can affect their social and emotional state (Debt stress affects health, fuels depression, 2015). I’ve also read about the fact that they both are related but I’d like to see it for myself and see what the relationship is and what causes it (Debt stress affects health, fuels depression, 2015)The general idea in society is that most think that those who have a very low income will commit suicide (Penner, 2017). I believe people who are interested in topics related to suicide will find this interesting. In these last couple of years to present time, suicide has been taken seriously by a lot more people, so to find a possible cause or relationship (income) may be a way to prevent suicide, or just make one aware of something that may directly or indirectly be a factor in suicide. I know that my data is from a reliable source as it is from Statistics Canada, a Canadian government website. The main source of data for the suicide rates in Table 1 is “The Canadian Vital Statistics Death Database”. They collect the number of deaths and cause of death data annually from registries all over Canada. The population for this census was anyone who was a Canadian resident, who may have passed anywhere in the world (“Age-standardized Suicide,” 2015). The main source of data for suicide rates in Table 2 is Health Canada. The type of survey is a census and type of data is cross-sectional, this means that the sample is representative of the population. It is also compulsory for everyone to report any death under the Vital Statistics Act. The population for this census was anyone who was a Canadian resident, who may have passed anywhere in the world (“Suicides and Suicide,” 2017). The source of the total average income (Table 3) is the Income Statistics Division of Service Canada, who then collected the data from various sources for different years: Survey of Consumer Finances (1976-1992), the Survey of Labour and Income Dynamics (1993-2011) and The Canadian Income Survey (2012-2015). This data was collected from everyone with an income in Canada. This data is also good, as Statistics Canada’s data indicator rated data from 1993-2015 as excellent meaning it’s pretty accurate. The past data has not been rated. (“Income Statistics,” 2017).HypothesisI hypothesize that the average annual income and suicide rates will have a strong negative correlation. Lower income will result in higher suicide rates, therefore I believe that the years with a lower average income will result in higher suicide rates and in the years with a higher average income, there will be lower suicide rates for those years. I predict such a relationship because when one has a lower income, they may have an increased amount of stress because they may not have enough to provide for themselves or their family, not be able to afford basic needs like food, shelter, and clothing. They may even take loans, this leads to debt and stress when it comes to paying off a debt, leading to some believing a solution is taking their own life (Debt stress affects health, fuels depression, 2015). My predictions are based on things I’ve heard in the media or news reports, where one kills himself due to not being able to provide for themselves or their family with their income. I’ve also seen research papers and reports that support my hypothesis. An article by the Business Insider, “‘Keeping Up With The Joneses’ Could Lead To Suicide”, states that location (where one lives or works), an income less than $34,000 and unemployment could increase the chances of suicide. If one were to live in a wealthy community with a low income they wouldn’t be able to keep up with the wealthy lifestyle (ex. cost of mortgage/rent too high to afford). It also states that those with an annual income of less than $34 000 are 50% more likely to commit suicide. Those who are unemployed are 72% more likely to commit suicide and those who are retired or on leave from work are more likely to commit suicide (Woodruff, 2012). AnalysisOne variable analysisAverage income in Canada (1976-2013)For the average income in Canada between 1976 and 2013 the mean is $69,797, the median is $67,800 and the mode is $66,400. The range is $17,900, the maximum average income was in 2013 and was $81,400, while the minimum average income value was in 1993 and was $63,500. The mean and median are close in value but the mean is slightly higher than the median, this is due to outliers increasing the value the of the mean. Looking at Figure 4, the most significant outliers can be found when 1994×1998 and in the years of 1984 and 1985. The distribution (Figure 3) is skewed right distribution, this is supported by the box and whisker plot (Figure 2) which shows us that 75% of the data is found when 63500×73375(Q3=73375). This tells us that the data is heavily skewed to one side and the most appropriate measure of central tendency would be the median. The standard deviation is 5302 which means the data is dispersed about the mean because the standard deviation value is great. It also tells us that most of the average income values are 5302 less than and greater than the mean, meaning most points will be approximately between $64,500 and $75,099. The IQR gives us 50% of the data, as it is Q3-Q1= IQR = 73375 – 65400 = 7975. Canada’s suicide rates (1976-2013)For the suicide rate, the mean is 12.56, the median is 12.81 and the mode is 11.30. The mean and median are close in value, but the mean is slightly smaller in value than the median because of the mode, which is acting as an outlier. The mode is weighed heavily because by definition the mode is the most repeated/frequent value in the data set, and so it affects the mean and brings it’s value down. In this data set, the most appropriate measure of central tendency is the median, as it’s not as affected by other values like the mean is. The distribution type is “normal distribution” because as we can see, the histogram (Figure 6) has a “bell-like” shape and in the box and whisker plot (Figure 5), the box is in the middle of the graph, Q1 and Q3 are found in between 11.4 and 13.6 on the graph. Also, we know that the distribution is normal because the mean and median are close in value (mean=12.56, median=12.81), this means the data is roughly symmetrical, making the histogram (Figure 6) bell-shaped and the distribution of the data normal. The standard deviation tells us if most of the data is clustered or dispersed about the mean. The greater the value of standard deviation, the more dispersed about the mean the data is, the smaller the value, the more clustered about the mean the data is. In this case, the data is clustered about the mean, this means that there is little dispersion because the standard deviation value is so small, it is 1.40. In other words, most of the data is 1.40 greater than or less than the mean. 68% of the data falls between 11.16 and 13.96, it’s a very small range containing a great portion (68%) of the data points. The IQR gives us 50% of the data, as it is Q3-Q1= IQR = 13.6 – 11.4 = 2.2, so 50% of the data is between 13.6 and 11.4. Two variable analysisSuicide rates vs. Average income in Canada (1976-2013)According to the coefficient of correlation (r=-0.785), there is a strong negative linear correlation between suicide rates and average income in Canada from 1976 – 2013. This means that as the average income increases, suicide rates decrease, vice versa. The negative correlation is supported by the fact that the “r-value” is negative and the line of best fit on the scatterplot (Figure 1) is negative. We know that the correlation is strong because the r value is close to -1. The coefficient of determination (r2) measures the strength of the relationship between both variables (ie. tells us the strength of the correlation between two variables). In this case, the r2 value is 0.616, so approximately 62% of the variation in Canada’s average income is due to the variation in Canada’s suicide rates. This means that 38% of the variation (100%-62%=38%) is caused by external factors (discussed in conclusion).y=–20.7100,000 x+27 This equation is the equation of the line of best fit in Figure 1. The line of best fit goes through as many points on the scatterplot as possible or through the center of data points (Line of best fit, n.d). The line of best fit measures the strength between two variables. It also tells us their coefficient of determination and coefficient of correlation. For example if the data points are close to the line of best fit, then the coefficient of correlation (r) will be close to 1 or -1, and the correlation will be strong, but if the points are dispersed and scattered around the line, then the coefficient of correlation will be close to 0 and the correlation will be weak. If the line of best fit it negative then the coefficient of correlation will be negative, same as if the line of best fit was positive. The y-intercept is the value at which x=0. The equation allows us to predict points that aren’t on the graph (extrapolation). If you plug a point into the equation and the value you get is 0, then the point will be on the line of best fit, if the value is less than zero then the point is below the line and if the value is greater than zero, then the point is above the line of best fit (Re: Detecting whether a point is above or below a slope, 2013). In this particular equation, the numerator is so small compared to the denominator because the scale on the x-axis is increasing by 5,000, while on the y-axis the scale is increasing by 2. The equation is negative because the line of best fit is negative and because the correlation is negative. Canada’s suicide rates (1976-2013)Looking at Figure 7, we can see that there is a strong negative linear correlation (r= -0.885) because the r value is very close to -1, this means the correlation is strong. The correlation is negative because the line of best fit is negative and the r-value is negative. The negative correlation means that as years go by (i.e. time increases), suicide rates decrease. The coefficient of determination (r2= 0.783) tells us that 78% of the variation in years is caused by the variation in suicide rates. Average income in Canada (1976-2013) By analysing Figure 4 we can see that there is a strong positive correlation (r=0.801) between “year” and “average income”, this means that as the years go by (i.e. time increases), the average income in Canada increases. This is true because the r-value is 0.801, which is a positive value, which means the correlation is positive and because it is close to 1 the correlation is strong. The r2 value is 0.642, this tells us that 64.2% of the variation in years is due to the variation in the average income in Canada. Discussion There is a strong negative linear correlation (r=-0.785) between suicide rates and average income in Canada from 1976 – 2013. Therefore as the suicide rates decrease, the average annual income increases, vice versa. The reason why a lower income increases one’s risk of suicide is that of debt, that can be caused by a low income. The debt causes stress and/or a poor mental health (Debt stress affects health, fuels depression, 2015). A severe case of poor mental health results in one taking their own life (Debt stress affects health, fuels depression, 2015). Another survey of 1,300 people who had debt anywhere from $10,000 to $49,999 showed that 28% of respondents had suffered from diagnosed depression, which was contributed to by debt worries, of which 18% said that their debt lead to suicidal thoughts (Penner, 2017). Overall, based on the research I’ve done, a lower income results in increased suicide rates. Other studies showed: If one makes less than $32,000 their suicide risk increases by 50% (Sanburn, 2012)An unemployed individual’s suicide risk is 72% higher than that of someone who is employed (Sanburn, 2012)In the research I did, I also came across the fact that if one lives in a wealthy neighborhood it increases their risk of suicide. A study showed that two people with the exact same income but who lived in two different neighborhoods (one neighbourhood had a higher average income), the one who lived in the wealthier neighborhood had an increased suicide rate by 4.5% (Sanburn, 2012). The reason behind this is the idea of being like others or wanting what others have and you don’t (Sanburn, 2012). Also if one had an income of $102,000 their suicide rate only slightly decreases than that of one who makes less than $34,000 (suicide rate for one who makes less than $34,000 increases by 50%) (Sanburn, 2012). Therefore, a very low (approx$34,000) or very high (approx$102,000) income increases an individual’s risk of suicide. Other factors like society, neighbours and etc, affects the correlation between both variables (Sanburn, 2012). BiasThe survey in which they collected the data for suicide rates was mandatory to fill out, but there is still bias because let’s say for example someone died in 2012, there could still be an ongoing case in court for their death deciding whether it was murder or suicide, so in that case it would not be recorded for now, because the case may be closed as suicide 10 years later, but it will currently affect the data and not make it 100% accurate. Another bias is what if a death was said to be suicide when it was really murder, vice versa, then this would affect the data. Overall, suicide rates can’t be 100% accurate. The data discussed was obtained from the same source: (Income statistics by economic family type and income source, Canada, provinces and selected census metropolitan areas, 2017). I got my data from Stats Canada, they graded the data for average income in Canada as “excellent”, “very good”, “good”, “acceptable” and “use with caution”, each of which was given a coefficient of variation. They graded data from 1993-2013 as excellent, with a coefficient of variation of only 0%-2%. Data from anytime before 1993, was not graded as it isn’t 100% accurate and they state that they estimated these values from the Survey of Consumer Finances (SCF). I was unable to find out why exactly the data was not graded, what factors affected it, but we can assume it is because the government system wasn’t as strict and good when it came to reporting your income tax anytime before 1993, as compared to now. Further AnalysisIn Figure 4 we can see great “dips” in the graph in the 1980s and 1990s. It is in these years where the most significant outliers can be found. The outliers occur when 1994x1998and in the years of 1984 and 1985. These outliers are the result of the unemployment rates in those years, because if we have many people with an income of zero, then the annual average income will be brought down in value. In Figure 8 we can see that there is a strong negative linear correlation (r=-0.872) between average annual income in Canada and unemployment rate (in percent) in Canada from 1980 – 2000 because the coefficient of correlation is close the -1 and because both the r-value and line of best fit are negative. This means that as the average annual income increases, the unemployment rate decreases, vice versa. In the 1980s we can see the unemployment rate hit 11.3% and 10.4% in 1984 and 1985, this means that the outliers in Figure 4 (x=1984, 1985) were outliers because of the high unemployment rates in those years. The other significant outliers in Figure 4 were when1994x1998 and in these years the unemployment rates were at a high of 10.4% and low of 8.3%. These unemployment rates were caused by the recession period in the early 1980s, it caused the unemployment rates to increase at a fast rate, going from 7.5% in 1980 to 12.0% in 1983 (The sharp increase in service sector employment in 2007 offset the decline in manufacturing, 2015). The recession continued in the 1990s resulting unemployment rates to rise as high as where 11.4% in 1993 (The sharp increase in service sector employment in 2007 offset the decline in manufacturing, 2015). Other factors that would come into play would be genetics, as they have an impact on one’s mental health. According to “Is Suicide Genetic” written in 2009 by McGrath, Suicide itself is not genetic, but genetics can trigger something that may lead to suicide. For example, one may inherit a genetic disease like cystic fibrosis (CF), CF is such a terrible disease that one may feel that they are better off ending their life than dealing with the CF. Diseases like depression, bipolar disease, schizophrenia and some types of anxiety can be caused by some genetic influences as well. It would also be interesting to research what caused these people to commit suicide(ex. What they left in their suicide notes or what was assumed the cause of their suicide) as this would make my research more accurate because for example if the individual was having trouble paying bills before their death, it could be assumed that they were having trouble with money and this leads them to take their own life. If I had more time, I would expand on my research by researching topics like inflation, wages and general things going on in the economy (ex. Stock market) in those years and how they would impact an individual’s income. For the following example I got all my information from an article by The Globe and Mail, written by Mahboubi in 2017: on January 1st, 2018 Ontario increased its minimum wage to $14 an hour from $11.40 an hour. And on January 1st, 2019 Ontario will increase its minimum wage to $15. There have been mixed responses from the public about the change in the minimum wage, mostly about how there aren’t many benefits in this change of wage. It is said that this great jump in the minimum wage will actually leave many unemployed and increase poverty, it’ll be the low-income families that will greatly and negatively be affected by this change. I could not find exactly how it is expected to/has affect(ed) the population’s income. Not only this but small businesses are being forced to close their doors because they can’t afford to pay their employees (Savory, 2017). Even big businesses like Tim Hortons, can’t afford to give their employees paid breaks (Saltzman, 2018). This just shows how much a change in wage is affecting businesses and individuals’ income. ConclusionBased on the research I’ve completed about suicide rates and average income in Canada from 1976 – 2013, suicide rates and income do have a strong negative linear correlation. This is proven by the coefficient of correlation which has value of -0.785, this relationship is strong as the r-value is close to -1 and negative as the value is negative. Looking at Figure 1 also proves that the correlation is negative and strong, we can see that the line of best fit is negative and the data points are closely scattered around the line of best fit. My findings do match my hypothesis, because in my hypothesis I hypothesized that income and suicide rates would have a strong negative correlation. My hypothesis was based on the research I did on this topic. One of my sources stated that if one were to have an annual income less than $34,000 than they’re 50% more likely to commit suicide (Woodruff, 2012). The same source also stated that an unemployed individual is 72% more likely to commit suicide (Woodruff, 2012). There are external factors that affect this correlation such as unemployment rates. In conclusion, as suicide rates increase, the annual average income decreases. Therefore, suicide rates and annual average income in Canada have a strong linear correlation.