Pooled cross-sectional analysis combines, or pools, multiple cross-sectional studies. Researchers may study multiple cross-sections at a single point in time, such as comparing cross-sections of the populations of multiple countries, or study a cross-section over time by collecting data at multiple time points, such as studying a sample over a period of 10 years.
There are different approaches to pooled cross-section analysis. A study of two or more cross-sections taken at a single time point allows researchers to examine differences across space. Such a study, for example, might compare economic measures such as gross domestic product and unemployment across 30 countries for a single year.
In contrast, multiple measures on a single cross-section let a researcher examine changes over time. For example, an economist could examine changes in gross domestic product and unemployment rate in a single country over a period of 10 years. This approach is sometimes referred to as a cohort study or as a time series study.
Most researchers conduct pooled cross-sectional analyses for one of two reasons: to increase the sample size, or number of subjects studied, beyond what would be available in a single cross-sectional analysis, or to examine the effects of change over time by studying a sample at multiple time points. Regardless of the reason, pooled cross-sectional analysis allows a researcher or graduate student to test complex hypotheses from a variety of academic disciplines, including business, education and the social sciences.
Pooled cross-sectional analysis requires a computer with a statistical software program. Popular statistics programs include SAS, SPSS and Stata. Spreadsheet programs such as Excel are not necessarily designed for complex statistical analysis; however, a number of programs are available online to enhance Excel's data analysis capabilities.
A pooled cross-sectional analysis often requires more time for data collection and analysis, especially if the researcher studies the same population over time. Depending on the study's size, additional research workers may be needed for data collection.
An additional risk, especially with an analysis of a cross-section over time, is that subjects may drop out of the study as time goes on, potentially distorting the results, as these subjects may not be representative of the group as a whole.