Statistics can be misleading. Reading statistical blurbs at the bottom of a news channel's broadcast or in a newspaper can lead people to believe that by printing numbers and associating them with a phenomenon means that the statistic is true. This is especially the case when it comes to social statistics. The truth is that statistics only tell part of the story. By truly understanding relationships between variables, you may be able to decipher those numbers. This article explains one type of such relationships: positive correlations.
In psychology, relationships between variables are called correlations. It's basically a statistical measurement of that relationship. A negative correlation suggests that when one variable changes, another directly and negatively changes with it. The opposite effect, when one variable increases as another variable does, is a positive correlation. A denotation of +1 between two variables shows a perfect positive correlation ratio.
A true positive correlation will absolutely show some positive relation between one variable to another. One statistic that has proven to be true in many studies is the relationship between education and income. Studies have shown that the higher level or number of years a person has in education, the higher average income that person will earn over a lifetime. This study has shown, in some cases, a positive correlation of 0.79 or +0.79. This means that 79 percent of the time, the higher level an education a person in the study had, the higher average income she had versus someone with a lower-level education. It isn't absolute as a +1. There could always be that random person who never graduated from high school but went on to become a famous rock star or successful inventor.
You've probably seen headlines in papers that say one phenomenon is absolutely linked to another. A headline that reads "Study Says If You Love Red, You Love Chocolate" may suggest that all things colored red are totally linked to chocolate. You may even begin to believe that if you like chocolate, then red must be your favorite color. Some may even go as far as thinking that owning a red car means you are into chocolate. But that headline is just an assumption that chocolate and red have a positive correlation.
This may or may not be true, depending on the true details of the study. Let's say a survey was done among 100 people asking them two questions: which color do they prefer among red, yellow or blue; and if they love chocolate. Twenty-four people say blue, 31 say yellow and 45 say red. Twelve of the blue people like chocolate (50 percent), as do 10 of the yellow people (about 33 percent) and 34 of the red people (75 percent). Does this mean that most people who like red love chocolate?
Not really. The people who "prefer" red may not even really care about the color all that much; it just happened to be one of the three choices. Those red people could really like mauve. They may even like chocolate, but not all that much. The same could be said about the yellow people, who didn't mind chocolate but didn't exactly love it. We'll never know because the choice was black and white: either you love chocolate or you don't. This is an example of a false positive correlation. Statistically, given the choices, people who selected red among three colors also selected a preference for chocolate.
The potential for positive correlation in psychology is great. By seeing a strong positive correlation between two variables in a study, psychologists may be able to discern life-changing information for that population or particular phenomenon. Let's say that there was a study among 10,000 middle-aged men in Spain and depression. Let's say the study found that 75 percent of the men who were clinically depressed for at least six months out of the year more than the other men also lived in Northeast Spain, psychologists could recognize that positive correlation. They may go to that region of Spain and do further studies and see what separates that region from the other areas. They could consider cultural significance of their population, the climate, diet--any number of variables. This positive correlation found in the original study helps these psychologists pinpoint a cause of depression and could help them find a solution to reducing it.
The size of the sample and the demographic of that sample must always be considered when a positive correlation is found. Political statistical findings during an election year should be seen with caution. If a poll says that Americans who make more than $100,000 prefer one candidate over another 87 percent of the time, you could assume that the well-to-do like that candidate. But how many people were polled? A hundred? A thousand? And where was this poll done? Online through a website visited by mostly rich people? The middle of Rodeo Drive in Beverly Hills? Or how about in an area where the preferred candidate has many constituents? This could be a false positive correlation that doesn't necessarily represent the rest of the country's true preference.
How to Calculate Correlation
The correlation coefficient, or r, always falls between -1 and 1 and assesses the linear relationship between two sets of data points...
What Is a Scatter Graph?
In the course of research, college students sometimes need to know if one variable they are studying is influenced by another. Businesses...
The Psychological Effects of Video Game Playing
It's hard to believe the video game industry is nearly 30 years old. Most of us can think back to the days...
Types of Dual Diagnosis
Types of dual diagnosis include chemical dependency and eating disorders. Learn how dual diagnosis disorders are often only treated for one-half of...