Surveys and Statistics: Lessons for Avoiding Survey Design Missteps – Carrie Iwema, Andrea Ketchum, Melissa Ratajeski
- surveyed folks about data management practices
- questions included things like demographic questions, type of research (human, bench, animal, or any combination)
- Issues
- specific questions vs trends
- is it compatible to do an environmental scan AND a statistical analysis at the same time?
- inconsistent wording of questions
- survey length
- coding issues
- each option with Check All That Apply questions needed to be coded as individual variables
- knowledge/attitude probing – explanations of “why” should be broken into separate questions than the Yes/No binary
- Focused recruitment helps separate out issues possibly with ambiguous dept names
- data visualization methods in stats maybe unfamiliar to librarians
- things to remember
- statistician – consult BEFORE survey; be clear on expectations
- input & output – even with piloting, they only tested if questions made sense, not how the data would come out
Limited Perfectionism, or How I Learned to Stop Worrying and Love the Epic Fails – Robert Johnson
- failure refers to a specific instance with a clearly defined set of expectations (ex: a class)
- in a job with multiple projects, tasks, and evolving roles, does the failure really permeate throughout or is it limited to a context?
- consider the stakeholders (internal/external) and their response to success, failure, or something else
- emotional component should be put aside
- consider the other aspects: what were your initial goals? did you meet those? goals can change and rescope
- does the institution gain value from this project?
- his project: integrate library resources into EMR/EHR for a hospital that just rejoined the university
- failures issues/causes
- administrative shakeups
- technological challenges
- physical distance/restricted access
- other wins
- shared costs
- got hospital to buy resources
- establish relationships and services
- put goal in the context of the world you are living in
- his project: librarian rounding
- buy-in from nurse manager but not physicians, unaware of hierarchies and relationship structures – internal politics
- did develop great relationships and opportunities to present to nurses, social workers, etc
- got referral for a more amenable unit for rounding
- his project: OBGYN grand rounds
- he dropped the ball and so did the counterpart in the unit by not having time to follow up/getting distracted with other things
- never say never – be open to reintegration
Balanced Scorecard Implementation for Health Sciences Libraries: Expectations and Surprises – Keith Cogdill
- engaged an outside consultant to help with implementation
- Balanced Scorecard identifies successes and gaps/missed items
- “balance scorecard is an endurance report with no finish” – not for those needing task completion satisfaction
- having leaders (those who are already super engaged) set targets for engagement may not be the best or most reasonable for the entire organization’s attainable goals
- used QuickScore to try and help re-integrate data; took a lot longer than anticipated and more manual entry than previously considered
- connect team objectives to key strategies
- individual plans are now also tied to balanced scorecard
- leads to more on-going attention to strategy
- enhances communication between staff and adminstration
Turning Lemons into Lemonade: Making Negative Research Results Useful – Linda Hartman
- publication bias
- not publishing due to outcome of results
- selective outcome reporting
- some outcomes not reported
- it takes longer to write up negative results than positive results
- positive results are around longer, so more likely to be cited or picked up
https://twitter.com/carolynthelib/status/600342558359867392
Carolynthelib's Blog 