Tipping the scale to positive correlation

1. Tipping the scale to positive correlation

This short lesson works with correlation & its sign. By the end of the lesson you'll intuitively understand how a scatter plot pattern represents a correlation value.

2. Correlation tips from one to negative one

Recall that correlation is a measure between two variables. Specifically, it is a descriptive value for a relationship between the variables. Correlation ranges from negative one to positive one. When the value is negative that means the variables have an opposite relationship. As one variable increases the other will decrease. The converse is also true. Positive correlation means as one variable increases so does the other. Therefore, logically correlation near 0 means a weak or no relationship.

3. Lifeguards for Ice Cream

It's also important to note that a correlational relationship is not the same as a causal relationship. Just because one value increases doesn't mean it *causes* the other value to increase or decrease. For example there is a positive correlational relationship between ice cream consumption and drowning each year. For each month, as ice cream consumption goes up, so do the number of swimming deaths. Therefore, the correlation is positive. However, ice cream doesn't really cause drowning. The real increase is caused by summer months. People are naturally more exposed to swimming when it is hot out. And during these same hot summer months, naturally people enjoy ice cream more often. So the two values move upwards together but clearly do not cause each other.

4. Correlation Visualized

Correlation also can be seen with an X-Y or scatter plot. Selecting a variable as X & another as Y will plot points on the visual. This dummy data has age as the X or horizontal data point. The vertical, Y axis represents income for these people. Notice how there is a distinct pattern sloping upwards from left to right? It's very obvious so you wouldn't be surprised to know that these 9 data points have a 0.99 positive correlation. The opposite also holds true, a negative correlation will slope downwards from left to right. And you may also guess that correlations near 0 will be represented on an XY plot as just a flat cluster of dots.

5. Let's practice!

Go find out if there is a relationship among the dating profiles for age, volunteering & roommates.