Holding ourselves accountable for data management: actions in Alaska

class: center, middle
# Holding ourselves accountable for data management: actions in Alaska

.pull-left[
.large[McCrea Cobb]  
Refuge Inventory and Monitoring Program

.large[Erik Osnas]  
Migratory Bird Management 
]

.pull-right[
.large[Amy Pocewicz]  
Science Applications

.large[Ryan Wilson]  
Fisheries and Ecological Services
]

???

---
class: center, middle
# How do we improve data management?

???
My coauthors and I are members of the Alaska Data Stewardship Team, which was founded a few years ago to focus on regional data management issues. During one of our first meetings, we discussed why aren't we meeting our data management vision for success.

Being a group of technically-minded people, many of us focused on technical capacity issues as the problem.

While these factors are arguably important, we kept returning to the human dimension side of the problem: people know that they should be doing it but simply aren't doing it because they don't see value, or they see value but it is a lower priority than other work at hand.

It was then ointed out that we are not dealing with a **technical problem**. What we were facing is an **adaptive challenge**.

So what's the difference and why is that important?

---
.pull-left[
## Technical problems

- **Easy** to identify
- Solved **quickly** by edict 
 
 
- Quick and easy solution solved by outside experts or authority

- People generally receptive to the change 
 
 
- Solutions are quickly implemented by edict 
]

.pull-right[
## Adaptive problems

- **Difficult** to identify (easy to deny)
- Solution requires a **change in values, roles, and work approaches**

- People facing the problem need to do the work to solve it

- People are resistant to even acknowledging the challenge

- Solutions are complex, unclear and require time and new discoveries
]

.smallest[
Adopted from R. A. Heifetz and D. L.Laurie. 2001. The Work of Leadership. Harvard Business Review. Cambridge, MA.
]

???
Here are the differences outlined.

Most notably, unlike technical problems, adaptive challenges are difficult to even identify and solving them requires a change in values, roles, and work approaches.

While technical problems can be quickly solved by experts or authority external to the problem, solving adaptive challenges requires that the people that are facing the challenge do that work to solve it.

People are generally resistant to even acknowledging adaptive challenges and the solutions are complex and require time and new discoveries.

So how does this apply to data management?

---
.pull-left[
## Everybody* plays a role in data management
 
 
.large[It's not just a biologist or data manager problem.]

.large[Success requires implementing accountability across roles and organizational levels.]

.smaller[*Ok, maybe there are probably some exceptions.]
]

.pull-right[
![:scale 90%](images/teamwork.jpg)
]

???
The bottom line is that tackling our data management challenges will require a group effort.

FWS employees across all programs and grade-levels have roles and data management will not be solved solely by those collecting data or technical IT staff.

Therefore, we need accountability across roles and organizational levels, or as we say, from "program directors to data collectors".

---
class: hide-logo, middle, center, inverse
# Lessons from Alaska: 
## Four actions to increase accountability

???
Ok, many of you may be saying "That's great. We know that. But I'm not a program director, so what can I do to improve accountability?".

In my remaining time, I'll share four actions from Alaska to consider.

---
# 1. Start with yourself

.pull-left[
![:scale 100%](images/be_the_change.jpeg)
]

.pull-right[
.large[
Familiarize yourself with laws and policies related to FWS data.

Learn about your available systems, guidance, and tools for managing data and metadata.

Apply these to your work.
]
]

???
The first step to any change, and arguably the easiest, is to **change yourself**. For data management, that means becoming familiar with laws and policies. Participate on a regional or national data management team. Learn about what data management systems (ServCat) and tools (mdEditor) are available. Read the guidance. Then apply them to your data.

I can say that being a member of our regional Data Stewardship Team, I have learned a lot about data management that I now apply to my work and share with others.

---
# 2. Adopt data management workflows

.pull-left[
.large[
Incorporate data management into **all** project and funding workflows.
 
Expect (require?) data management best practices. 
 
Leverage for change where you can.
]
]

.pull-right[
![:scale 85%](images/data_lifecycle.png)
]

???
#2 Adopt data management workflows.

Secondly, incorporate data management into **all** your project and funding workflows.

What do I mean by that? Consider the data life cycle. Perhaps the easiest place to advocate for data management is in the planning stage.

If you're responsible in allocating resources and funding, **require a data management plan and metadata with project proposals**. Once funding has been allocated, an opportunity to account for data management best practices has been lost. In Alaska, The Fisheries and Ecological Service Program is starting to require data management plans with project proposals. No data management plan, no funding.

If you're leader, expect your staff to be following data management best practices. When asking staff for data (accessing), rather than telling them to send it in an email, indicate that you want a hyperlink to a repository. When appraising your employees, refer to our existing data systems to measure productivity. Completing a survey is not just collecting the data. These leadership actions all demonstrate and increase the value of our data management systems.

Finally, not all of us are supervisors and have the ability to enforce accountability. But, if your job provides a desired service for people producing data, you can use that carrot to leverage for data management best practices. In Alaska, refuge biometricians and data managers that receive requests for technical support now require that the project has been documented in the refuge survey database "PRIMR" and that all associated data have been archived with metadata. Our message is: we're here to help, but you need to meet us halfway.

---
# 3. Engage leadership

.pull-left[
.large[
Seek out data management advocates.

Present "non-invasive" actions that leaders could act upon.

Include leaders in data management groups and advertise successes.
]
]

.pull-right[
![:scale 100%](images/dst_meeting.jpg)
]

???
Thirdly, seek out senior leaders that are advocates of data management. They're out there. Those that understand the added value of good data management. These people can not only advocate for data management but can provide critical feedback on proposed approaches to meeting your goals.

Don't be shy about presenting some "non-invasive" actions for leaders to act upon. Propose small changes that together  message the importance of data management. In the Refuge program in Alaska, we met with Refuge supervisors and provided them with a list of proposed actions, most of which they are adopting. This also resulted in increased engagement from the Deputy Chief and a future memo to Refuge staff regarding the importance of data management and clarifying expectations.

Finally, take opportunities to include leaders in data management groups. Their input is extremely value. This photo shows our Regional Director in Alaska providing opening remarks during a face-to-face meeting of the Data Stewardship Team. Our team now has a Regional Directorate Team representative to serve as a liaison.

---
.pull-left[
# 4. Measure success

.large[
 
Quantify the current state of data management

Develop and use tools to measure progress
]
]

.pull-right[
 
![:scale 100%](images/measure_success.png)
]

???

Our last action is to measure the success of your data management.

Ask yourself: am I or is my team following data management best practices? How do you know?

Understanding whether you're following best practices requires an assessment of the current state of affairs. For example, how many data sets have metadata. Once you have this baseline, you can then start to measure progress.

To meet this need in Alaska, our team is working on a **regional project inventory**, essentially a list of projects collecting data. This inventory is the first step toward understanding what data are being collected.  That information will provide the foundation to measure whether metadata for these data have been created. Over time, we can then assess whether our actions are making a difference.

---
class: hide-logo, inverse
# Summary

.large[
Data management is an **adaptive challenge**.

1. Everybody is accountable so **start with yourself**.

2. **Incorporate data management** into all project and funding workflows.

3. **Leadership engagement is critical** for instituting accountability.

4. **Understand the current state** of your data management **so that you can assess the performance and success** of your strategies.
]

???
Data management = **adaptive challenge**, therefore there is no easy technical solution from an outside expert. It will require changes by those experiencing the challenge.

- **Everybody is accountable**, but the easiest place to start is with yourself.

- **Incorporate data management** into all project and funding workflows.

- **Leadership engagement is critical** for developing accountability. Foster that support.

- Finally, **Understand your current state** of data management. Think about **how you will assess performance and success** of your strategies.

---
class: inverse, hide-logo