Slide 1Studies seek to describe, explain, or predict. This video is about the first of those three goals, and about a distinction that trips up a lot of students.

Slide 2It's important to understand which goal you are pursuing with your work before you can ask a focused research question and design a study.

Slide 3Every study uses an element of description, but description itself can be the goal. Let's say you recruit a sample of 100 people who suffer from the same disorder and conduct a trial to estimate the effect of a new drug on a clinical outcome. When you summarize what you know about these 100 people at the time they were recruited, for instance the average age of the group, you're describing your sample. Descriptive summaries appear in nearly every research article, usually as Table 1.

Slide 4But we must distinguish between the use of descriptive statistics, like the mean age of those 100 people, and descriptive research questions. Descriptive research aims to characterize, to describe, a population of interest. This is done through quantitative estimation methods, where you recruit or construct a representative sample and use the data from the sample to estimate how common something is in the population.

Slide 5The core challenge to this type of descriptive research is recruiting a large enough sample that reflects the population of interest well enough to say something valid about the group, with sufficient precision to be informative.

Slide 6Descriptive research is important in every discipline, but especially so in epidemiology. Epidemiologists ask descriptive research questions about people, place, and time. Here's how one team defines descriptive epidemiology. It seeks to characterize the distributions of health, disease, and harmful or beneficial exposures in a well-defined population as they exist, including any meaningful differences in distribution, and whether that distribution is changing over time. That's a mouthful. Let's break it down.

Slide 7To characterize the distribution of something means to describe how often it occurs. The focus of descriptive studies is frequently a health status, referred to as an outcome, like cancer. Or potential exposures, such as environmental carcinogens. But description extends to knowledge, attitudes, and practices, to what people know, think, and do. For instance, a study might ask, what percentage of women of reproductive age in Nepal use a modern method of contraception?

Slide 8To ask a good descriptive research question, you must be clear about who makes up your group of interest, your target population. Your target population is a specific group that you want to make a claim about, such as women of reproductive age in that country. Typically you'll recruit a sample of individuals from this group and use the data you collect to make an inference from this limited sample to the larger group.

Slide 9In addition to describing the group overall, there are often theoretical or programmatic reasons to quantify variation, or differences, in an outcome or exposure across subgroups, across geographies, and across time. For instance, you might want to stratify an analysis by ethnicity and race to quantify disparities in outcomes or in access to services. Doing this over time tells you whether conditions are improving, worsening, or staying the same.

Slide 10Here's an example to tie it all together. The Demographic and Health Surveys have answered this question, and many others, in more than 90 countries. The DHS Program designs, conducts, and analyzes large, nationally representative surveys on population, health, HIV, and nutrition. In 2011, DHS researchers surveyed a random sample of 10,826 households across the country and interviewed 12,674 women between the ages of 15 and 49 about their health behaviors and preferences.

Slide 11They estimated that 43 percent of married women reported using a modern method of contraception. Stratified by age, modern method use was twice as common among married women ages 35 to 49 compared to married women ages 15 to 24.

Slide 12Description is essential to science, and to decision-making related to needs and resources. The result from Nepal suggests that more than half of married women of reproductive age were not using a modern method of contraception. But description only gets you so far. In the next video, we ask why.