Foundations of Quantitative Research in Political Science
- Students often have difficulty grasping the difference between random sample and random assignment. This is completely understandable because both words have the word random in it. However, understanding the difference between the two concepts is essential. This video will address exactly this issue. By the end of this video, you should be able to distinguish the difference between a random sample and a random assignment. Say we want to run an experiment to see whether a treatment we're interested in has some effect. As an example, let's say that a politician wants to publish an ad, and we want to know whether the ad will increase support among voters for that politician. The first step is defining the population. In this case, the population would be all voters. The second step is drawing a random sample from the population. Drawing a random sample assures that we have a sample that is representative of the population. Remember that this allows us to make inferences about the population from our sample. Third, to find out if a treatment has an effect, we want to randomly assign some units to the control group and some units that the treatment group. The module on experiments explains why randomizing the treatment is desired. But just to recap, it's because, as can be seen here, it creates a treatment and a control group that are very similar to each other. If the treatment has any effect, we can be fairly sure that it's due to the treatment itself and not because the treatment and control groups are different. Once we have our treatment and control groups, we want to give the treatment group the treatment, in this case show them the ad, and not give the control group the treatment, so not show them the ad. After this has been done, we want to measure support for this politician in both the treatment and the control group. When we compare the outcomes of the treatment and control groups, if we find any difference in support for this politician we can be fairly confident that the ad had this effect. So notice that a random sample are the units that we want to collect data on while random assignment decides which of these units get the treatment and which don't.