What are Longitudinal Data?

A dataset is longitudinal if it tracks the same type of information on the same subjects at multiple points in time. For example, part of a longitudinal dataset could contain specific students and their standardized test scores in six successive years.

Student Name	Grade 1 (2001) Raw Score	Grade 2 (2002) Raw Score	Grade 3 (2003) Raw Score	Grade 4 (2004)  Raw Score	Grade 5 (2005) Raw Score	Grade 6 (2006) Raw Score
Mike	339	350	361	366	381	390
Jasmine	332	343	350	351	351	355
Thomas	360	380	400	420	430	438

The primary advantage of longitudinal databases is that they can measure change. So we can estimate, for example, the effect of various factors on improvement in student achievement. We can also estimate the overall effectiveness of individual teachers by examining the performance of successive classes of students they teach, as well as examine the extent to which teacher effectiveness changes with experience or the composition of their class.

The longitudinal data extend into the past as well as the present. So we can evaluate the effect of a specific policy by looking at, say, student performance or teacher turnover before as well as after the policy was introduced. Longitudinal data also allow us to use sophisticated analytic strategies to measure the impact of various policies with reasonable precision.

CALDER is capitalizing on the richest source of information about schools, teachers, and students in the United States—state administrative longitudinal databases.  Several features of CALDER’s data enable the Center’s experts to conduct first-rate research:

1. The state databases at the core of our work are census files, which means that the Center has access to ALL districts, schools, teachers, and students in a state. Census files do not restrict CALDER’s research to the imperfect assumptions and margins of error in sampling a population.
2. Having access to multiple state databases allows the Center to replicate studies across states to study policies in comparative perspective and test the robustness of its findings.
3. Data span multiple years of observation, enabling CALDER’s experts to study both contemporary and past policies relative to a baseline.

Detailed demographic information on students enables CALDER to distinguish patterns in racial, social, and achievement segregation, as well as patterns in teacher quality at the district, school, and classroom levels