How I got over myself and listened to my supervisor

In the first year of my PhD, I took a mandatory course called ‘Communicating Your Research’. Don’t tell my supervisor (who happened to be the lecturer) but I absolutely loathed this class. Not because I thought it was badly run, but because I resented it. I felt that it the expectation of planning my ‘publication narrative’ and strategise my for my long term publication track was a tall order with little justification.

Before I elaborate, I want to declare my 1st-year naivety (and potentially arrogance). Today, I now respect the need to learn how to communicate my research to an audience beyond the academy. Dear supervisor, should you ever read this, please by all means say:

I told you so.

Presumed response of supervisor

I don’t have time


On top of classes, experiments, writing, teaching, volunteering (in the faculty), going to seminars (often mandatory in year 1), and all the other overwhelmingly time-sucking aspects of PhD life (and sacrifice of family time, social life and sleep), it was unrealistic to start crafting my research persona and communicating my research beyond the peer-reviewed literature (see footnote 1).

enlighten me

To a greater extent, I thought the core tenets of the unit, or at least the way they were demonstrated, were out of touch with the reality of being an Australian PhD student in 2018

Image credit @beneaton7 at Unsplash, Adapted by Caitlin Doogan.

I guess the biggest issue I had with the class was that it featured a whole host of jaded professors who started their research careers in a time where their focus of their PhD was to produce a thesis. Papers were a bonus. Arguably, these academics finished their PhDs, knowing they were walking into a salaried position and not facing the uncertainty of an increasingly casualised workforce or post-docs won based on h-index. Towards the end of the 1990’s the US ‘publish or perish’ mentality had gained traction in Australia. Still, it can’t be denied that the progression of their PhDs and the academy’s expectations of PhD candidates were different from today’s (Bastalich et al., 2010).

Additionally, being told to publish in an A* or Q1 journal, or in our case, one that was on the approved faculty list, was to me, a bit much to ask. For a while, I would watch other PhD students get added as fourth or fifth authors on their supervisors’ journals and wonder why that wasn’t our strategy. Today I think differently and appreciate that different disciplines have different norms, cultures and rules that relate to their publication cycle (although that didn’t really help me much considering the interdisciplinarity of my project). Regardless, I couldn’t relate to these academic and, in my opinion, they could not relate to me, nor could they justify why I should try to follow their path.

Particularly in my field, Machine Learning (ML), the publication cycle is intense and the pressure to produce papers is almost overwhelming. As someone who practices applied ML in interdisciplinary settings (and Natural Language Processing) my ability to pump out papers is hindered somewhat in comparison to my peers. What’s worse is that I get the distinct impression my lack of papers (and lack of appropriate venues for those papers) is looked down upon by my peers.

It wasn’t until my second year that I got over myself and re-developed my strategy. Three things happened to get me thinking more clearly about how best to communicate what I was doing.

Acknowledging publication cycles

I realised that the publication cycle is f#@ked (that’s the technical term). The volume of papers beings pushed out has led to some serious reproducibility issues for the discipline. I routinely read low-quality papers published in high-quality journals and conferences where I couldn’t quite work out what the contribution to the field was, or what the applied use could be. Surveys that simply quoted the abstract of papers seemed to be acceptable for ML. But in the other camp, I work in, this would be thrown out and viewed as complete garbage. To me, this race to publish and subsequent flood of sub-par article, seemed self-serving. They negated the ability to demonstrate potential impact in the real work (immediately or as part of a chain of development).

Ironically, I wrote a paper on my opinion of the over datafication of academics. Basically, we are manipulated to chase metrics rather than meaningful knowledge production. The paper is here if you are interested. Acknowledging the publication cycle and the allure of metrics was one thing, but what opened my eyes was just how detrimental that trap is. I saw the nasty side of academia – stolen ideas, predatory PhD supervisors and a person’s worth based purely on a number. ‘Stuff that’, I thought. Suddenly the guilt and frustration I felt began to melt away. Articles were nice, but no longer the goal.  

Hell hath no fury

In my field, Machine Learning (ML), the publication cycle is intense, and the pressure to produce papers is almost overwhelming. As someone who practices applied ML in interdisciplinary settings (and Natural Language Processing), my ability to pump out papers is hindered somewhat in comparison to my peers. In my first year, I got the distinct impression of my lack of articles (and lack of appropriate venues for their publication) was looked down upon by my peers.

I copped flack from the chair of my confirmation about my ‘ambitious but yet to be realised publication plan’. That really got to me, I probably haven’t let that go, but sometimes the ‘I’ll show you’ attitude is a good motivator. At the same time, one of my panellists, who didn’t even read my report, told me my research was shallow and wouldn’t be published in ML journals or accepted to conferences. Given my respect for him, the hurt I felt at the time was crippling. But then one day I saw him in his office, he looked ill, was somewhat dishevelled and was berating another student. My supervisor told me was under a lot of pressure. For some reason, that respect turned to pitty. I stopped caring about his opinion and vowed never to get like that (I have, but I’m getting better). 

I became the expert

My supervisor started recommending me for projects (in a good way), pitching me as ‘ the best in the University at this’, or ‘the absolute expert’. My other supervisor complimented my writing. Weirdly, this didn’t give me imposter syndrome. It gave me confidence. Mainly because my supervisors don’t give praise for the sake of making us feel good.

Later, when a manuscript that took 7 months to write was rejected from a journal, I felt almost none of the expected ‘sting’. Reviewer two was unnecessarily rude, and I wasn’t quite sure if that was reasonable, but that only served to motivate me to take the excellent critiques of reviewer one and re-write it. I also sought feedback from some prominent academics I had met. Previously I would have been terrified of their critiques; at that point, I practically hounded them for them. I knew i was the expert on the methodology (it was a methods paper), but I also knew I wasn’t the expert on how to communicate to that field (media and communication).

I went viral

A twitter post went viral. Back when I was doing my honours (2012), I expressed to my then supervisor that I wanted to pursue a career in bioinformatics. Being a Biomedical Science grad meant I needed to learn how to code to do this. He basically said I couldn’t compete with the ‘boys’. Gender stereotyping was on my radar at the time of the tweet, there aren’t many Australian women in my PhD program, and there are far more men than women (see footnote). I was proud to be someone who practised ML from a medical background. However, there is a bit of a negative culture. After some offensive and unprofessional comments from a course mate, I retaliated online. 

Follow me on Twitter @CaitDoogan

The post was seen 1.5 million times and to date has attracted 2,496 retweets and 362 comments. The comments are what inspired me. Short sharp and to the point, I generated discourse around women in stem and the issues we need to address collectively. I was then nominated as the women’s rep for graduate research students in the faculty and today being the first to call out the tones, statements and actions that keep women down in the discipline. People listened to me.

Take aways

When I was little, I did ballet. After a few terms, my mum got told not to bring me back. I refused to curtsey to her because I didn’t like the dried apricots that were the end of lesson reward for dancing well (seriously who incentivises a four-year-old with dried fruit). I wish my mum had used the opportunity to teach me something about humility. Instead, it was years before I did do sport of any kind, preferring to invest in academics overly. The downside of this was that throughout my teens, I defined my self worth by my exam scores. This mentality carried over to my 1st year of PhD.

I wish I had learnt early to listen.

Really this post is about thinking seriously about who you want to be as a researcher and then assessing whether a) you are communicating to the right audience, the right way, at the right time. Publications are important, but it’s ok to do other things too.

For those students doing thesis by publication, the benifit is that you potentially have increased employability. The downside is that you may pigeon hole youself. If you got this far in this post then if anything, take this away.

Your supervisor doesn’t fully understand what it’s like to be a PhD right now. They don’t get the pressures you are under and they don’t get how they might contribute to that. But they have survived the ever changing and often brutal landscape of academia. They know things you don’t. So listen to them… at least sometimes.

Footnotes

  1. Admittedly I did start this blog, but as you can see, it is not regularly updated.
  2. This is in reference to the difficulty I had developing strong bonds with others in our multicultural cohort. Cliques tend to form which can be socially isolating.

Bastalich, W., Behrend, M., Bloomfield, R., Ford, J., Loeser, C., & McCulloch, A. (2010). Communicating research: Audiences, academics and research students. 9th, 207.

  • Post
  • Block

No block selected.Open publish panel

  • Document