Data Management in the Heartland Network

1
Data Management in the Heartland Network

• Background
• Centralized staff
• Protocols developed and implemented primarily
  in-house
• 12 of 15 network parks are fairly small and
  represent places and events of historic and
  cultural significance
• Integrated with prototype

2
What Have We Learned?

• Data management is hard work!

3
Revisiting Priorities

• Data mining
• NPSpecies and other service-wide databases
• Data management planning
• ATR/COR on protocol development projects
• GIS/GPS support to monitoring projects

• In
• Data management needs assessment with project
  leads
• Managing data work flow and security
• Database development and maintenance
• Writing data management SOPs

4
Establishing Clear Roles and Responsibilities

• Our approach
• DM is shared responsibility among Project Manager
  and Data Manager
• Project Manager is expected to have working
  knowledge of relational databases and ability to
  independently conduct basic data queries and
  exports

• Implications
• Platform and design complexity are constrained
• Increases risk of data loss
• Creates some redundancy in database skills
• Increases PI interaction and ultimately results
  in a better customized product

5
Managing Work Flow

• Idealized protocol development time line

Final version DB
Peer review
Finalize SOPs
Objectives
Pilot sampling
T 1 yr
T 2 yr
Internal review
Methodology
Final protocol
Preliminary analyses
Beta version of DB
Write DM SOP
Database development times should be short, for
each version, about 120 days. Keep PIs' involved
and the initial database design simple. Add to
the complexity later.
6
Minimizing Complexity By Considering Return on
InvestmentThe Subprime Syndrome
Efficiency/Usability Ratio to Development/Maintena
nce Effort
High
Low
Low
Database Complexity
High
7
Get The Data In!
Find a balance between data security, ergonomics,
efficiency, and data management capacity
8
Get The Data Out!
Find a balance between automated data analyses
and simple data export routines
9
Inspirational Message Albeit Poorly Crafted
Yes You Can!

• Data management is hard work!