Secondary Data Analysis

An important option for the study of life-span development is the use of secondary data analysis to study the changes that occur in individuals across the life span. Secondary data analysis refers to the use of existing data that have been collected by another researcher for a specific primary research pursuit. It also refers to the analysis of large-scale data that are collected by research organizations for the use of the research community. In both situations, the current user of the data did not design or collect the data and thus is using it in a secondary capacity. There are many advantages to using these data sets for pursuing research on life-span development, as will be detailed in this entry along with information on how to access and analyze this type of data.

Large-scale, longitudinal secondary data may be of particular interest to life-span researchers because they provide data on individual changes across multiple developmental periods. Hence, questions can be examined such as whether poverty in early childhood relates to college attendance in adolescence or whether the parent–child relationship in adolescence is predictive of taking care of an aging parent. If the secondary data set is also a nationally representative data set, then researchers can generalize their results more broadly.

Large-scale, longitudinal data sets are large both in terms of sample size or number of participants as well as the number of times data have been collected on these participants. These two characteristics of the data give researchers the statistical power to test even the subtlest of influences on individuals’ development across time. As an example, the Early Childhood Longitudinal Study–Kindergarten cohort started collecting data on approximately 20,000 children across the United States in 1998 as they entered kindergarten and ended after they entered eighth grade. A researcher who is interested in examining how self-concept may relate to children’s achievement would be able to test this relation across development even though it has a relatively small influence on young children. There are many more questions that can be answered using this data set regarding parents, teachers, curriculum, and state policies, not to mention ethnicity/race, gender, and area of the country in which the family resides. Thus, not only do the researchers have statistical power, but they also have the power of explanation such that many questions and alternative explanations can be tested.

Not all secondary data sets are large and nationally representative; there are many that are smaller and are collected at only one time period (cross-sectional). The advantage of using these smaller scale secondary data sets is the reduced cost of data collection for the researcher. Collecting data is time-consuming and expensive; therefore, it is a considerable advantage when there are data that are already available that can answer the questions of interest.