The answer to ts question is....Trend Line.
A line of best fit or a trend line.
Correlation is an estimate of a linear relationship between two variables and takes no account of non-linear relationship. If the relationship is quadratic and the domain is symmetric about some point, the correlation will be zero. It is, thus possible for the points on the scatter plot to lie exactly on a parabola while the calculated correlation is zero. In such a case, it is easy to make a prediction despite no correlation.
Data can be correlated (meaning there is an indication of a relationship) either positively or negatively. The datasets of two variables (x,y) which have a negative correlations, when plotted, will show a negative trend, that means with increasing values of x, there will be, generally, decreasing values of y. An example of negative correlation, would be the more hours someone exercises, the less they weigh, if weight loss is measured as a negative number and weight gain as a positive number. In this case x= hours exercised, y = final weight - original weight. For presentation purposes, we frequently define our variable to show positive correlations. As per the above example, I could have defined y = original weight - final weight, which would show a positive correlation and plot as an upward trend. It would not change the absolute value of correlation just the sign. You may check wikipedia under correlation to get more understanding.
If measurements are taken for two (or more) variable for a sample , then the correlation between the variables are the sample correlation. If the sample is representative then the sample correlation will be a good estimate of the true population correlation.
Experiment used to quantify a trend
A "trend" is a mathematically provable correlation between reported crime activity and whether it rising or falling. The "trend' refers to whether it is going up or down.
A correlation exists in a scatter plot if there is a general trend in the outputs as inputs increase. If the outputs generally increase in value, then there is a positive correlation. If the outputs generally decrease in value, then there is a negative correlation.
By finding a correlation trend by means of line of best fit.
The graph follows a very strong downward trend. Would have helped if you specified which correlation coefficient; there are different types.
I believe you are asking how to identify a positive or negative correlation between two variables, for which you have data. I'll call these variables x and y. Of course, you can always calculate the correlation coefficient, but you can see the correlation from a graph. An x-y graph that shows a positive trend (slope positive) indicates a positive correlation. An x-y graph that shows a negative trend (slope negative) indicates a negative correlation.
Easier to reply when full details are given...
Trend correlation
One variable must react to the other. If represented by a graph, a visible trend will appear and a "trend line" should easily be visualized and determined.
D. Trend correlation
Specify what you mean by "trend". The correlation remains the same. Specify what you mean by "increase". Multiplying will result in a difference in variance, addition will not.
There would be no definite correlation. It would just be a random correlation that would be all over the graph because there is no trend in hair color and weight. Your weight doesn't determine your hair color.