Topic: The Size of Social Groups
Statistical Concepts Covered: In this applet, you’ll expand on your knowledge of correlations by using them in regards to regression and making predictions.
In this applet we will utilize the data collected from a study by Dunbar (2003) to test the social brain hypothesis, the belief that the human brain, specifically the neocortex, has increased in size to accommodate the cognitive demands of being in large social groups. In order to look at human evolution in this way, we will explore how humans, hominoids, and simians compare in terms of neocortex ratio and social group size. Simians are “higher” primates, such as apes and monkeys, while hominoids are those resembling humans (e.g., Neanderthals), but are not what we refer to as humans. Neocortex ratio is the ratio of the size of the neocortex to the rest of the brain. A 1:1 ratio (represented as simply 1 on the X-Axis in the graphs in this applet) would mean that the neocortex makes up half of the brain, while a 2:1 ratio (represented by 2) would mean that the neocortex is twice the size of the rest of the brain.
Statistical Lesson. A topic related to correlation is regression. Regression is used to make predictions based on the relationship between two or more variables. If we find that there is a relationship between two variables, we might want to use one of those variables to predict another. For example, we find that there is a relationship between study hours and exam grade, meaning as we increase the number of study hours, we increase the chances of a higher exam grade. Using this correlation, we decide to use the number of hours we study to predict what our exam grade might be. This is the process of regression, modeling the relationship between two variables so we can use one of the variables to predict the other variable. Using some math we can create a regression line that is the best fit for our data, which means we try to find the line that is closest to our current data. We can then use that line to predict about future values. No need to worry about the math involved in creating this line, however, as this line will be provided for you in this applet. You will simply need to use it to make predictions and determine what it tells us about the relationship between our two variables. Keep in mind that even though we can make predictions using a regression line, this does not mean that prediction is a “sure thing”. Predictions are just estimates that have the possibility of error and do not imply cause and effect between our variables.