In many science experiments, there is a main dependent variable. This is a variable that you can measure at one time to determine the value of another variable. In the example given above, the main variable would be the temperature of the water. You could ask for the average temperature and then compare it to the average for that day.
There are several ways to set up the dependent variables. You could use a repeated-measures design. The reason that this type of design is called repeated-measures is because you would want to look at the same data set (the temperature) over again. With a repeated-measures design, the effect of the independent variable is linearly regressed. Let us take the same temperature measurement over three days. That measurement will be linearly regressed three days later.
One problem with using the repeated-measures design in an ocean study, for instance, is that when you average the three measurements, you get a value that does not tell you what the actual temperature was on any particular day. Oceanographers determine the temperature using a variety of methods, including charts and buoys. The best method may not always give you the correct value.
Let us say, instead, that we want to know what the temperature was on, say, the day before the balloon flight. To determine this, we could use a continuous-measures model. The problem with continuous-measurement models is that you cannot let the dependent variable change because then you are measuring what did not change. If the balloon flies in one direction, you know it was warm on the day it flew, but if it flies in the other direction, you do not know.
Many oceanographers believe that the Pacific Ocean is relatively warm, roughly 4 degrees Celsius warmer than the North Pacific. To measure this, they will use a buoys float. If the float rises to the surface of the water, it indicates warmer air, which can sink into the water, thus warming the ocean. However, the buoys could rise to the surface of the water and then drop, causing a drop in the temperature. This could skew the results, leading to the conclusion that the Pacific Ocean is too warm, and that the North Pacific is too cold. It is difficult to control this kind of error because there are so many uncertainties in measurements of ocean temperatures.
Oceanographers cannot control the wind, the temperature of the sea water, or the rain. All these factors have an effect on how the oceanographer will measure the temperature of the ocean. Temperature measurements are therefore quite prone to errors. Additionally, balloon flights are not easily controlled by meteorologists, who can decide where and when the balloon flight will occur. Thus, the ballooning process also contributes to uncertainty in the results.
Scientists study oceanographic data to learn about the climate and weather patterns. Oceanographers must therefore make certain measurements of temperature in order to study these climate and weather patterns. Unfortunately, they cannot make balloon flights every day to monitor temperature measurements. But luckily, they can make special measurements known as “wind profiles” in order to study temperature patterns. These measurements allow the oceanographers to determine which parts of the ocean are warmer and which parts are colder.
The problem with studying temperature in one area of the ocean, like the Pacific Ocean for example, is that there are many different factors that can lead to different results. For instance, the winds, clouds, waves, and currents can all change the temperature of the area. Thus, scientists will often combine their results from many different measurements in order to determine what is the dependent variable in a particular experiment. The uncertainty of the results comes from these multiple independent variables.