There are many ways of categorizing scientific studies. One way is to look at the times at which data were collected. Another is to look at how many variables were studied. Together, these allow a study to be called cross-sectional or longitudinal, and, within longitudinal, to be classified as repeated measures or time series. These three different types of study incur different costs and allow different sorts of conclusions to be drawn.


Cross-sectional studies look at only one time point. Longitudinal studies can be repeated measure or time series. Both look at multiple time points, but repeated measure studies usually look at more variables, while time series looks at more time points (usually at least 50) and very few variables (often only one).


One example of a cross-sectional study would be the relationship between demographic variables (such as age, ethnic group and income) and vote in the 2008 elections. An example of repeated measures would be tracking people's weights over time, and looking at their eating habits, demographics and other variables as well. An example of time series would be looking for seasonal effects in stock market prices over the course of many years.

Advantages and Disadvantages

Cross-sectional studies usually allow a larger sample for the same cost than repeated measures, but do not allow the researcher to look for changes over time. Repeated measures studies do look at changes over time, but are often complex and costly, and are often affected by attrition. Time series studies allow intensive investigation of long-term trends, but do not allow investigation of a large number of variables.

Combinations of Types of Studies

Although any one study can only be in one category, research projects can use two or even all three types to explore different aspects of a phenomenon. For example, studies of people's weights could include investigation of long-term trends and seasonal effects (time series), investigation of the effect of different diets (repeated measures) and complicated relationships among many variables at a given time (cross-sectional).