Metadata Driven Survey Research Jeremy Iverson

Open Standards

Less effortMore and better documentationEasier to use dataIncreased transparency

1. Don’t Repeat Yourself

Metadata Driven Survey Design Metadata Survey instruments Paper questionnaires Statistical source code Paper documentation Web documentation

The Data Lifecycle Documentation Lost

2. Make documentation a byproduct of the process

3. Make Metadata Accessible

Metaphors

Repositories Question Banks Classification Banks Variable Banks Concept Banks

Harmonization Share items across multiple waves of a study Or across multiple studies Easily discover and compare studies or datasets that use the same elements

Answerable Questions What studies explore these concepts? What questionnaires use this question? What datasets have this variable?

3. Make Metadata Accessible

4. Increase Transparency

Social Scientific processes should be reproducible. Science

"It is the policy of the American Economic Review to publish papers only if the data used in the analysis are clearly and precisely documented and are readily available to any researcher for purposes of replication."

By Federal Reserve [Public domain], via Wikimedia Commons

“Materials sufficient for replicating results reported in this article will be [have been] deposited with the ICPSR publications-related archive” - Gary King’s replication standard “Materials sufficient for replicating results reported in this article will be [have been] deposited with the ICPSR publications-related archive” - Gary King’s replication standard

4. Increase Transparency

5. This is Possible Today

Define the metadata and relationships Automate everything else with a tool

DDI 3 Tools ColecticaQuestasy DDA Metadata Editor Canada RDC Tools STARDAT Custom tools

Metadata Driven Survey Research Don’t repeat yourself Document the whole lifecycle Make metadata accessible Increase transparency

Thank you Jeremy Iverson