1	SECONDARY DATA ANALYSIS October 2004
2	What is Secondary Data Analysis? Use of data collected by someone else, often for a different purpose What is usually meant is secondary analysis of quantitative data, but not always See http://www.soc.surrey.ac.uk/sru/SRU22.html http://www.sagepub.com/book.aspx?pid=4854 Or could be archival (documents, historical materials, records, etc.) Meta-Analysis
3	Reasons to do Secondary Data Analysis (SDA) Costs (including time) Bigger and More Representative Samples Standards for surveys and questionnaires have gone way up This is also true of qualitative work
4	Reasons Not to Do Secondary Data Analysis (SDA) Doesn’t fit the question you care about. Questions have to come first.
5	Sources of Secondary Data National Center for Education Statistics http://nces.ed.gov/ Archives (e.g., University of Michigan, University of Wisconsin) http://www.icpsr.umich.edu/ http://dpls.dacc.wisc.edu/archive.html Many others: Census Bureau, Health surveys, NLSY, etc.
6	Types of Secondary Data Cross-Sectional Longitudinal i. Panel Study ii. Repeated Design
7	Types of SDA Designs Trend – comparisons of cross-sectional data at two (really three) or more points in time Cohort – common life event Panel – follows the same people over time Event History – examines rates of change in people’s transitions (e.g., job changing, marital dissolution) Time Series – broader patterns of behavior (e.g., voting)
8	Things to Watch Out For Weighting Estimation SPSS (e.g.) not always adequate Alternatives – WesVar, Sudaan, AM Getting Overwhelmed and/or Distracted with “One More Variable” Changing definitions over time (item comparability) e.g., educational attainment or race in the Census
9	For More Information: Kiecolt, K . Jill and Laura E. Nathan. 1985. Secondary Analysis of Survey Data. Thousand Oaks, CA: Sage Publications.