When I think back to my PhD now – which you can read here if you have a sudden bout of insomnia – I sometimes cringe a little … well, maybe a lot depending on what day I (try not to) think about it. This is not only because I came to hate my PhD – well, mostly – but also because I end up thinking another mistake I made at some point during the process. I’ve actually run a workshop with graduate students specifically called “learning from my mistakes” which was about identifying the various ways I would do things differently if I had a time machine – why I’d bother re-doing my PhD if I actually did have a time machine is not worth asking!
I thought it might be useful to outline the various ways I managed to screw up my PhD in this blog, primarily as a learning exercise for others about to embark on this journey or those already on their way. Now, I have to say at the start that my PhD passed with pretty minor corrections – they took me a weekend to do – so I obviously managed to pull things together at the end, but for a long time I went through the usual doctoral angst, especially the night before my viva.
Anyway, here goes with displaying my dirty pages for all to see …
Literature review
- From the start I did not have a properly conceived literature focus for my PhD; I kind of meandered from one set of literatures to another depending on who I’d spoken to in the last few weeks. This meant I missed whole areas of scholarship, which I had to then go back to as I wrote up the thesis, and ended up reading some really boring stuff I just simply did (and still do) not use.
- One key part of the literature review I didn’t think about until almost too late was the methods literature – I can’t emphasize enough how helpful it is to read this stuff (esp., I imagine, at the start).
- I also could have done with talking to more people in the field, not just my supervisor. I might then have hit upon the approach I subsequently found I should have used but didn’t until I’d finished – i.e. global commodity chains. Part of the reason for this was my assumption that I had to follow my supervisor’s approach or ideas – don’t make this mistake! As long as you can justify what you’re doing then don’t have to ape your supervisor.
Conceptualization & operationalization
- It’s hard to appreciate sometimes how difficult it is to translate abstract concepts (e.g. “knowledge”) into examinable objects of study. For example, I sometimes ended up using rather ‘fuzzy concepts‘ (e.g. tacit knowledge) which I did not operationalize as clearly as I would have liked. It definitely needs careful thought.
- A side issue here is that my own conceptual fuzziness led to a sense of uncertainty about what I was doing, which probably made me sound uncertain when talking to informants.
Distractions
- Although this is not always a bad thing, I did end up getting rather distracted by numerous other ideas, events, literature, writing, etc. during my PhD. All these things have since helped me develop my thinking in areas that my PhD just did not cover, but I was not always as focused on the task at hand as I could have been. Ultimately this is a toss-up between ‘mistake’ and being useful in the long-run … just be careful!
Sampling
- One of the major mistakes I made was assuming that there would be plenty of objects of study out there for me; for context, I wanted (and did) study biotech innovation, but I did so by sampling people involved in ‘successful’ (i.e. marketed) biotech products. I started by focusing on pharmaceutical products and then quickly realized that UK firms had produced less than a handful of these – I ended up having to expand my sample somewhat as a result.
- It would probably be a good idea to think about at least two sampling frames when starting out, just in case the first turns out to be nonsense … or at least make sure that your research design is flexible.
Fudging
- One key issue that came out during my pilot phase – ALWAYS do a pilot phase! – was that I had fudged, for want of a better word, the difference between two sets of informant. Fudging this distinction led to some difficulties in comparing the data I had collected.
- What I had basically done was conceptualize innovation as a system with all sorts of social actors contributing to it, but I then treated each social actor the same. Big mistake – I couldn’t identify distinct social actor groups using the same sampling frame so I had to reconsider how I would relate the findings from each group together. In the end I had to create distinct questions for each and compare their responses.
Hesitancy / procrastination
- Personally I don’t like phoning people up – not sure why, but there you go. So, it probably wasn’t a great idea to use (structured) telephone interviews as my key methodology!
- Anyway, I could have got the data collection out of the way much faster if I had just got over my hesitancy and got on with things. I found myself procrastinating about calling people up – sometimes out of the blue but rarely – even though when I did call them people were generally very generous with their time (and patience!).
- So, don’t be afraid to contact people – especially directly. Most people are friendly and if you show some interest in their life or work, I’m guessing most people are happy to share.
Data overload & analysis
- I found myself rather overwhelmed with all the data I ended up collecting. Alongside phone interviews, I was also collecting secondary data on the UK biotech industry and I ended up compiling a massive database of information that I had no clear idea about how to analyze when I started out.
- This is not always a bad strategy, obviously, but it meant that I did end up with a lot of data I just didn’t use – so, having a clear analytical goal or objective in mind is useful. My major mistake was not having this clear objective in mind when I started collecting data and therefore wasting on collecting data that didn’t make it into the final thesis.
- When it came to the analysis phase – WARNING, analysis takes longer than you think! – I had so much data I was trawling through it trying to find patterns, correlations, or what-have-you that would have been much easier to analyze if I’d clearly thought about these at the outset, especially the analytical framework that reflected my theoretical underpinnings (e.g. scale, social group, etc.).
- Another mistake I made was in the presentation of this analysis which started out as incredibly reader-unfriendly tables of statistical correlations. Thinking about the presentation of your findings is critical because you don’t want to piss off your readers (i.e. examiners) any more than necessary.
Writing up
- During the final phase of my PhD (i.e. writing up) I was working in fits and starts. I had moved to a full time job and was doing the analysis and writing up in the evenings and at weekends – not the nicest experience I can tell you. Basically it meant I ended up working 2-3 hours in the evening most weekdays. But it was never enough concentrated time to get my writing on.
- Another thing I really didn’t think about until this point was the need to return to the literature – you must keep up with reading anything new that comes out, of course, but it is also worthwhile returning to the original literature you based your proposal etc. on in the first place. Re-reading stuff obviously takes time so that added to my evening workload.
- What I found worked for me in the end was booking off two periods of annual leave – two-weeks each – and sitting down for those four weeks and writing every day. I managed to get half my thesis written in that way and it meant I could concentrate on bringing it all together as a (semi-)coherent whole.
That’s about it! As I mentioned above, I would do everything differently if I was forced to do everything again – I guess that’s the point of a PhD. The great thing about academia is that we get a chance to do exactly that once we move onto other projects. I found my PhD a great training experience for understanding how to (not) go about doing research – every mistake or error taught me what not to do in the future, so it wasn’t all a bad experience (just at the time!).
Although I imagine the PhD experience is very different in other disciplines, it’s worthwhile bearing in mind in all fields that your PhD does not have to be perfect, it merely has to pass.