Can you please help with my homework i have COVID

This week, we learned about correlations.

For the discussion board this week:

1) Read the journal article Correlation Does Not Imply Causation.

2) Discuss why correlation is not the same as causation.

3) Provide an example that would demonstrate that correlation is not the same as causation (Don't use the murder rates and ice cream example since I already did in the lecture!). If you are having difficulty, try doing some research, look at journal articles, google correlations in criminal justice, see what you can find.

Inside the numbers

Crim 4830 Lecture 11 Correlations

Purpose of Correlations Use when we want to see how 2 variables are related - how one variable changes if the other variable is changed Correlations say nothing about causality! Both variables must be continuous! We are calculating bivariate correlations

Correlation Coefficient The correlation coefficient reflects a linear relationship between 2 variables Linear means that for every value on 1 variables there is 1 and only 1 value that corresponds to it on the other variable

What correlations really look at is the amount of variability that is shared by 2 variables - how much they overlap Correlation Coefficient Correlation

Ranges from -1.0 to +1.0 Correlation Coefficient The number tells us how strong the relationship is The closer the number is to 1.0 the stronger the relationship between the variables (regardless of the negative sign!) For example, .81 is stronger than .61 and -.85 is stronger than .51

Correlation Coefficient .80 - 1.0 - very strong relationship .50 - .80 - strong relationship .40 - .50 - moderate relationship .20 - .40 - weak relationship .00 - .20 - weak or no relationship A strong relationship does not necessarily mean it is significant or meaningful!

The negative/positive sign tell us what direction the relationship is in Correlation Coefficient Positive Correlation - when both variables change in the same direction - meaning as one increases the other does and vice versa Negative Correlation - when variables change in opposite directions - meaning if one variable increases the other decreases

Correlation Coefficient For example: -.70 and +.70 are equal in strength but one is a positive correlation and one is a negative correlation -.85 would be stronger than +.65 - the negative does not mean less in this case

Correlation does not imply causation Journalists breach the simple rule too frequently By Kelly Toughill A bortion reduces crime. Hire more cops to cut the murder rate. Canada needs more prisons to contain a growing population of convicts. These are all number stories rooted in crime statistics. Each of them ran as front- page, top-of-the-hour, twitter-worthy news. Not one of them is probably true. Reporters find narrative in numeri- cal data by looking at changes over time, or by comparing one set of numbers with another. A big change in a steady number usually means there has been a newswor- thy event - an earthquake, a stock market crash, a serial killer. A consistent change over time in one direction usually signals a newsworthy trend: a recession, the obesity epidemic, global warming. Finding the change is the first step in re- porting a number-based story. The next step is figuring out what is driving the change; this is where well-meaning and not-so-well meaning sources most often lead reporters astray. And this where crime stories most often go wrong. Correlation does not equal causation. The phrase is drilled into anyone who stud- ies statistics. Just because two numbers move at the same time doesn't mean that one causes another. Sometimes both num- bers are being driven by a third factor. More sunscreen is sold during months when air conditioners are in use. That doesn't mean sunscreen causes air conditioners to switch on. Both are symptoms of a third factor: summer. Sometimes numbers move at the same time for no obvious reason. And some- times, one number really does drive an- other, even if it's hard to prove. The cor- relation between smoking and lung cancer was spotted decades before scientists could prove the causal link. Here are a few tips to avoid being led astray when writing about crime numbers: 1) Don't compare data from differ- ent agencies. Statistics Canada reported a homicide of two people for every 100,000 Toronto residents in 2007, yet Toronto police reported a rate of three homicides per 100,000 - a rate that is 50 percent higher. Both could be right, because they use different boundaries to calculate population. 2) Try to examine the quality of the data. The most reliable crime statistics in Canada come from the Incident-based Uniform Crime Reporting Survey, which is administered by the Canadian Centre for Justice Statistics. The methodology is sound, but it is based on how many crimes are substantiated by local police forces. That can vary wildly from town to town, and even from month to month. The sad truth is that sometimes police ignore crime, and that skews the statistics. And of course not all crime is reported to police; reporting rates vary by location, by victim group and by the type of crime. 3) Don't buy the conclusion just because the data is good, which brings us back to abortion, beat cops and prisons. The abortion-reduces-crime hypothesis was first published in an academic journal in 2001, but hit the news when one of the authors included it in his bestseller, Freak- onomics. According to economist Steven Levitt, crime fell after abortion was legal- ized in the United States. The theory was that unwanted children grew up to commit more crimes and that legalizing abortion reduced the number of unwanted children. The claim attracted predictable attention from both sides of the abortion debate, but it took a while for economists to join in. In 2005, two bank economists criticized Levitt's methodology, suggesting that he should have used crime rates, instead of absolute numbers, and that he failed to include a key formula in his calculations. Two years later researchers tried to du- plicate Levitt's research in England and Just because two numbers move at the same time doesn't mean that one causes another. 37 MEDIA

Wales, and failed. They found a superfi- cial correlation between abortion rates and crime, but it disappeared when they looked at the age of those committing crimes. By then, the equation was widely accepted. Abortion reduces crime. The summer of 2005 was traumatic for Toronto residents, who faced killing after killing in what became known as the "sum- mer of the gun." A year later the murder rate had plum- meted, and media began to wonder why. Several stories focused on an anti-gang task force set up by Toronto police, and a shift of 450 uniformed officers to the street. (To be fair, Chief Bill Blair didn't try to take the credit. He told reporters that many factors contributed to the 25 per cent drop in the homicide rate.) There would be few budget battles at city hall if simply hiring more police could stop murders. But drug trends, gang dis- putes and many other things drive homi- cide rates. The truth is that a one-year drop in homi- cides is almost meaningless. The homicide rate is volatile; it jumps up and down all the time. The only way to determine a trend is to look at long-term numbers. Chief Blair surely knew this when he refused to take credit for the drop. Two years later, homi- cides were worse than ever; 84 people were killed in Toronto in 2007, compared to 79 during the year that included the infamous "summer of the gun." The most recent tweaking of crime sta- tistics has been in relation to prison con- struction. Prime Minister Stephen Harper has called for a massive expansion of the prison system, even though the over- all crime rate is falling. So why the need for more beds? Apparently because of an anticipated crime wave, and a belief that criminals will be sentenced to longer pris- on terms. As of our deadline, there were no numbers to back up that multi-billion- dollar bet. Kelly Toughill is an associate professor and di- rector of the school of journalism at the Univer- sity of King's College in Halifax. TOUGH ON CRIME: Conservative MP, and former police chief, Julian Fantino looks on as Minister of Justice and Attorney General of Canada Rob Nicholson speaks about Bill S-10 during an Ottawa news conference. The Conseratives claim their so-called tough on crime agenda will lead to safer strees. Opposition politicians criticize the proposed laws for being "dumb on crime", and for lacking a price tag. PHOTO CREDIT: THE CANADIAN PRESS/Adrian Wyld SPRING 2011 38

Copyright of Media is the property of Media Magazine and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use