Vaccination Investigation
When conducting experiments, there are various ways in which you can record and present data, depending on the audience and the overall message behind the experiment itself. When presenting data, the standard that many choose to follow is to publish relevant data on a graph. Usually, graphs are used to display relationships between variables.
Graphs can be displayed in many ways, but they generally take the form of a linear model, a proportional model, and/or an inverse proportional model. A linear model is displayed on a scatter plot and the line of best fit for all of the data points is generated to create a straight line that displays the average line according to the data. This line, as was aforementioned, is the average of the values and can also produce the next prospective point on the graph according to the trend it is averaging out.
A proportional model also creates a line that has proportional variables. That is, if the independent variable is doubles, so does the dependent variable. It takes the general form y=Ax, where A is the proportionality constant and the model does not have a y-intercept- it starts from (0,0).
An inverse proportional model does create a line to better fit a graph and be as close to the actual plotted data as possible. What sets this apart and makes it different from a linear model is that a linear model has a y-intercept and can be displayed as y=mx+b as its general form whereas an inverse proportional model is demonstrated as y=A/x. This can better predict the next point and leaves the least amount of room for error. So, if a variable is doubled, the other variable is halved.


Comments
Post a Comment