Things you WON’T think… - University of Washington



Things you WON’T think… page 1

Things You WON’T Think To Do

Helpful Hints for a Successful Fellowship or Junior Faculty Career

The following suggestions reflect the collected experience and wisdom of the fellows and faculty.

GENERAL TIPS

➢ Time to read and reflect, and think are now important (and to be relished) parts of your job description. Spend time meeting people and talking about your ideas – Enjoy!

➢ Explore several ideas, then consider the “best” project as one you are interested and passionate about – being excited about it will keep you going through rough spots.

➢ When you commit to a project, you commit to a mentor as well (this is the “renting and apartment analogy”-you are renting the landlord as well as the apartment). Help your mentor by:

o get their vision of success

o agree on prioritized commitments - set timelines and adjust as necessary

o give updates and ask for feedback early

▪ doing this gives you time for course correction

o talk about styles of communications early, find out about pet peeves

o ask their preferences for decision-making (how often to meet, prefer contacting in person vs. call vs. email, how long do they need to respond to questions, etc), have your preferences in mind to discuss

➢ Make part of the project (or projects) “fail-safe”: meaning that if things don’t work out as planned, there will still be some interesting and reportable data on some aspect of the project (secondary questions, a qualitative part of a larger project, a validation of measurement tool, etc).

PLANNING

➢ Set up a timeline, but know that it will take longer than you estimate.

➢ Before collecting data, take a stab at creating some mock tables (and even sketches of graphs or figures) of your potential results (as you might present them in a paper)… it will help you focus and write the paper(s).

➢ Think about whether you can collect all the data you will need (especially for confounders, potential effect modifiers). If you can’t, you may need to find another data source.

Things you WON’T think… page 2

➢ Call the people who know the most in the country about your subject. Chances are they will be happy to discuss your planned project, and they will have a perspective you lack. Generally very senior people are glad for new blood, whereas the younger researchers maybe more guarded.

➢ Write your budget early, and have others review it.

➢ Human subjects review will take time. Call the liaison for the human subjects committee early to get a sense of what will be needed, and start on that early. They’re really helpful in making the process smoother.

CONDUCTING

➢ Pilot testing is worth it.

➢ If you aren’t sure or get stuck with some part of the analysis, ask for help. Might just save you months.

➢ Back up everything, everyday in more than one place. Jump drives are a great tool – but can be easily lost, broken or virus-infected – so don’t depend on it.

➢ Name and date your files – it will help you keep track of where you have been and if necessary you will be able to reconstruct.

➢ When performing routine calculations in a statistical package, triple-check your formulae. Double is not enough!

➢ Putting something down on paper is valuable even though it will be revised later. While conceiving a project write out the introductions; later draft up the methods sections as an ongoing diary of what you’re doing. It will also give you ideas to go back to later.

➢ As you think of what the main message of your work are, share them early amongst your colleagues-just articulating them helps to crystallize your thoughts, and you get useful feedback too.

WRITING TIPS – with assistance from Gil Welch

➢ Abstract

o Context: Motivate the reader – why should they care about your findings and conclusions?

o Objective: simple, directly related to your finding

o Methods: Explain how the finding below was obtained, not other methods in paper

o Results: Primary finding – what is the one fact you want readers to walk away with

o Conclusion: supported by the result below

Things you WON’T think… page 3

➢ Introduction

o What’s the current situation?

o What’s the problem (or gap)?

o How will this study help?

➢ Methods

o Make sure there is a method for every result

o Use subheads to force structure: Overview, Setting, Exclusions, Exposure/Intervention, Outcome, Secondary Outcomes, Analysis

➢ Results

o Short and to the point

o Driven by your figures and tables

➢ Discussion

o What’s the central finding? Restate and place in context of other work.

o Could it be wrong? Identify and deal with threats to validity. Consider alternative explanations for your findings given the study design.

o What does it mean? Put your work in perspective. Assess its generalizability and speculate about its implications. Specify what you think should happen next.

➢ Make sure labels are consistent across abstract, text, tables and figures.

o A specific thing (or idea) should always have the same name.

➢ Think like your target audience; write for them, not for yourself.

➢ Write less - focus on high-visibility components

o Readers look at abstract and title (often only part read)

o Tables and figures – communicate your major findings here

➢ Get feedback

o General reviewer - whose primary job is to determine whether your writing can be understood.

o Expert reviewer - whose primary job is to help prepare you for external review.

A “content" expert to scrutinize specific technical or theoretical issues.

A “hostile friend" who will look hard for flaws.

o Tips for eliciting feedback

▪ Get an explicit commitment

▪ Clarify the review process

▪ Simple courtesies

o Tips for receiving feedback

▪ Have a conversation

▪ Don’t be defensive

▪ Focus on understanding

▪ Judge suggestions critically

➢ It’s easier to revise than write from scratch. The earlier you start the easier it is to find time to allow yourself to step back, digest, and read your work with a fresh perspective

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download