You are familiar with the relative measures of association, but as I've mentioned before, there are absolute measures of association as well which serve a slightly different purpose. In this lecture, I will show you how we can calculate the risk difference and the incidence rate difference. Again, you obviously need two groups of individuals. One exposed to a certain factor, and one unexposed. Let's stick to the example we used previously. Two groups of students attending classroom A and attending classroom B. Four out of the 10 students who were exposed to the boring lecturer, and two out of 10 students who were not exposed, fell asleep within one hour. The risk or cumulative incidence of sleeping was 0.4 among the exposed and 0.2 among the unexposed during the study period. This time, I would like you to calculate the risk difference. Conveniently, the name of the measure suggests a way to calculate it. The risk difference is simply the numerical difference of the risks in the two groups. In other words, the risk among the exposed minus the risk among the unexposed. Based on the previous calculations, 0.4 minus 0.2 which is equal to 0.2 over a one hour period. How would you interpret this number? The key value of the risk difference and of the incidence rate difference is zero. When the risk of the disease among the exposed is equal to the risk among the unexposed, the risk difference is zero. Compare these with the ratios where the value indicating no difference between the two groups is one. If the risk difference is higher than zero, it means that the risk of disease among the exposed is greater than the risk among the unexposed. In contrast, when the risk of disease among the exposed is smaller than the risk among the unexposed, the risk difference is a negative number. Focusing on the concept of association, we would say that the risk difference of zero means that the exposure is not associated with disease. A positive risk difference means that the exposure is associated with an increased risk of the disease, and the negative risk difference that the exposure is associated with a decreased risk of the disease. In our example with the risk difference was 0.2 over a period of one hour, we can be more specific and say that there was a 0.2 or 20% excess risk of sleeping in the classroom in those exposed to the boring lecturer compared to those unexposed over the one hour study period. An equivalent expression is that there were 20 more cases of sleeping per 100 people in those exposed to the boring lecturer compared to those unexposed over the one hour study period. Similarly, by subtracting the incidence rate among the unexposed from the incidence rate among the exposed, you can calculate the incidence rate difference. In our example, the numerical value of the incidence rate difference would be expressed as cases of sleeping per person hour over the study period. As you can see, the two absolute measures of association I presented quantify the actual absolute differences between the groups. This can be very informative when considering the impact of a factor at the population level.