Pleasantly surprised at how much traction this thread received. I am going to share #hiddencurriculum on dissertation advice via mini-threads, as time permits. #AcademicTwitter

As requested, grant writing and transition to post-doc/faculty will also follow... 1/ https://twitter.com/Monica_Aswani/status/1319288707271127041

The focus of this thread will be on how to conceptualize and formulate your dissertation question. #highered #dissertation #phdadvice

Again, my hope is others will chime and share their insights to help demystify. 2/

Caveat: some dissertation advice is specific to fields, institutions, etc. What follows is a mix of what I found helpful (or wish I had known in retrospect), not blanket advice. 3/

Figuring out where to start and how to narrow down your topic can be anxiety-inducing - it's the first meaningful departure from classwork to a largely self-directed project.

Don't let imposter syndrome spiral into a self-fulfilling prophecy of failure. 4/

Observe and listen to the world around you. Part of the dissertation is gaining self-awareness about how you translate intellectual stimuli that flags your curiosity into testable ?s.

Make note of what strikes you as interesting or puzzling. 5/

Research inspiration is not limited to academic sources. Front-line insights from caring for my dad, who was frequently hospitalized, have influenced my research. Translating inspiration into research does not have a defined or linear roadmap. 6/

Broadly, consider:
1) What constructs/relationships are at play?
2) What is known*? Not known?
3) Is that disconnect an addressable gap?

*What is known is a function of context: what is and is not measured and how it is constructed. And, publication bias. 7/

I struggled w/ how to operationalize advice. I was often told to read voraciously, but how did I get from there to a research question? It also felt daunting to dive into a large literature base w/out a game plan. So, I came up with one... 8/

Organize lit in a spreadsheet. Color code it based on main x/y vars, dataset, method, dir/sig of results, etc. - elucidates ways topic has been studied, how findings align, and starts constructing a narrative. Missing patterns = potential ideas. 9/

I had that idea b/c, as a 1st year, I worked on a project w/ ~100 model permutations. To understand the results, I made a spreadsheet - each color captured coefficient direction (+/-) + level of sig (1% vs. 5%). I later used arrows + varying boldness so it was more intuitive. 10/

Knowing the results is not the same as understanding what they convey. Think deeply about fundamentals. I devoted a lot of time to *really* appreciating regression. What does it mean to control for a variable - numerically, conceptually? (Adapted from late Dr. E M Foster) 11/

My visual lit review gave me a way to frame ? in the context of existing work - a dissertation storyboard, if you will. It may not resonate w/ you though.

Another idea: try to find commonalities or triangulate "Future research should..." statements. 12/

What makes a good question? Folks will have different opinions - ideally, it should be interesting no matter what you find, including null results. It should intrigue you, given the amount of time/energy you will invest trying to answer it. 13/

It should also be feasible. Not all data can, or should, be collected. If available data is too expensive, it never hurts to kindly ask if there is reduced pricing for students or reach out to researchers w/ access to discuss a potential collab. 14/

There is no one 'right' way to develop a ?. Hypotheses may not be testable due to data or other constraints. Figuring that out is a part of the process. Likewise, gaining self-awareness about how you best figure that out (i.e., Tetris-style spreadsheet) is also a part of it. 15/

I worry this thread skews more towards my random ramblings, rather than #hiddencurriculum insights.

If/how hidden curriculum becomes explicit is, I suspect, partly shaped by exposure to others' experiences though.

On deck: dissertation writing and mentoring/committee selection?