1. New Infrastructure for Harmonized Longitudinal Data with MIDUS and DDI 3.2
Project that MIDUS is working on with Colectica using DDI 3.2 to deliver harmonized datasets.
IASSIST Toronto

2. Overview of Presentation
Background on MIDUS
Importance of DDI for
Harmonization
Facilitating discovery and complex analysis
Current Project Goals
Implementation of Project Goals
Upgrading MIDUS from DDI 3.1 to 3.2
Building on the MIDUS-Colectica Repository

3. Background on Baseline: 1995-96 Harvard University
MacArthur Foundation
N=7,108
Ages 25-74
Satellite studies of stress, cognition

6. MIDUS: Unique Characteristics
Multiple waves (9-10 year interval)
Multiple samples/cohorts
1 National (MIDUS Core)
2 Milwaukee
3 Japan (MIDJA)
4 National (MIDUS Refresher)
5 Milwaukee (Refresher)
Multidisciplinary design
Aging as integrated bio-psycho-social process
Result
N=11,500
34,000 variables

7. MIDUS: Unique Characteristics
Multiple waves (9-10 year interval)
Multiple samples/cohorts
Multidisciplinary design
Wide use of MIDUS – Open Data philosophy
#1 data download at NACDA
Top 10 data download at ICPSR
530+ publications

8. Status of Current DDI Efforts
MIDUS Metadata Repository

9. Moving Forward
National Institutes of Aging project: “Facilitating Secondary Analysis and Archiving of MIDUS through DDI”
Timeline:
The proposal is in response to RFA-AG “Secondary Analyses and Archiving of Social and Behavioral Datasets in Aging”.

10. Current Project goals Under a DDI 3.2 rubric…
1. Harmonization (internal, post-hoc)
Clarify related nature of longitudinal and cross-cohort survey variables (RepresentedVariable)
Provide information/procedures for reconciliation
2. Custom Data Extract (CDE)
Allow researchers to focus on variables of interest
Facilitate accurate merges across numerous datasets

11. Harmonization Concordance table
MIDUS P1 concordance table (Google Spreadsheet)
Includes “Comparability notes” and “Comparability class”
Example: Variable A1PA30 “time since last BP test”
Comparabililty notes:
“M1 is not directly comparable with M2, MKE, MR, MKER, M3: M1 responses were coded as number of months, while other waves broke out number (amount) and unit (days, weeks, months, years) into 2 separate variables.”
Offer code/algorithm for reconciliation
Also includes notes and descriptions of how the variables differ.
READ example.
We’re making this table available to our users and researchers as an Excel spreadsheet, so it can be used “manually,” but it is large and unwieldy. Ideally this information will be integrated in our DDI codebooks and the Colectica repository – aside from the goal of harmonizing, it also accomplishes a much more mundane goal of identifying the same variables across datasets, facilitating longitudinal and cross-cohort analysis.

12. Custom Data Extract Customized dataset
Search variables, use shopping basket
Include variables from across all MIDUS projects
Merge different datasets
Different formats (csv, SPSS, SAS, Stata)
Associated DDI codebook

13. Development Milestones
1. Metadata Quality Report 2. Harmonization 3. Web-based Discoverability 4. Data Extraction

14. Step 1. Metadata Quality Report
Compare the harmonization spreadsheet to the Repository
Check for:
Missing information
Inconsistent labels
Inconsistent data types
Update the metadata to improve quality

15. Step 2. Harmonization Use the harmonization spreadsheet
Create a RepresentedVariable for each row
Store these in the repository

16. Step 3. Web-based Discoverability
Build on top of Colectica Portal
Searching and information retrieval out-of-the-box
Add cross-reference tables for easy discoverability
Choose variables or groups of variables to include in the data extract

17. Step 4. Data Extraction Store master data in Colectica Repository
Based on a user’s selected variables, generate:
Datasets
CSV, R, SAS, SPSS, Stata
HTML and PDF codebooks
DDI XML

18. Progress  Complete Metadata Quality Report Complete Harmonization
Upcoming
Web-based Discoverability
Data Extraction
Approximately 3800 RepresentedVariables

19. Acknowledgement
This research project is supported by a grant from the National Institute on Aging (R03-AG046312).

20. NADDI 2015 – April
“Research Data Management: Facilitating Discoverability using Open Metadata Standards”

21. NADDI 2015 – April 8 - 10 “Research Data Management:
Facilitating Discoverability using Open Metadata Standards”
University of Wisconsin - Madison

22. Thank you midus.wisc.edu www.colectica.com
Jeremy Iverson – Colectica
Barry Radler – UW-Madison
Dan Smith – Colectica
midus.wisc.edu