I find it suprising that academics are really strong advocates of ‘evidence based practice’ and seek to account for and justify every methodological design decision they make, but don’t seem to apply the same standard to the ‘project methodologies’ they tell/require their PhD students to use to manage a research ‘project’.
I say this because in my PhD programme, I was asked to provide a ‘timetable for completion’ (one page) as part of my ethics application and will need to provide a revised one for confirmation panel. The specification of ‘one page’ led me to produce a Gantt chart for my ethics application and I am advised by my supervisor to do the same again for confirmation panel.
It isn’t just my university, this sort of thing is in lots of text books. For example:
There are many ways of planning a project and presenting the plan. One technique is to make use of a table, which sets out the tasks and planned dates for their completion. A better approach is through the use of a Gantt chart through which you not only specify tasks but whether they are going to be completed in sequence or in parallel (page 50)
Develop a project schedule: a project schedule such as a Gantt chart provides a list of the main project tasks and the dates for their completion. (page 64)
Gray, D.E. (2014), Doing research in the real world, 3rd edition, Sage Publications, London
But, whether I choose to present it as a Gantt chart or not, a ‘timetable for completion’ or project schedule just seems like the product of lots of guesses – yes guesses or vague aspirations not intelligent estimates. There are simply too many unknowns and uncertainties.
Writing an accurate timetable for completion requires me have some degree of certainty about:
- what the tasks (and all the sub-tasks within a task) to be done are – as a novice researcher, I don’t know every task to be done. A simple note on ‘read literature on xxx’ is only experienced as a series of sub-tasks as I am in the process of doing it. New tasks also come up such as the need to write a blog post on project management(!) or the opportunity to write a journal article. These ‘unpredicted’ tasks immediately alter a precious timetable to the point it wasn’t worth having one in the first place. My original list of tasks (and sub-tasks) are simply informed guesses based on the chapter headings in standard research text books – design, collect data, analyse data, write up thesis.
- the amount of time (hours) that each task will take – again, I don’t know how long it takes to search for and review literature on a particular topic. I don’t know how long it takes to transcribe or analyse a research interview. I’ve never done it before so why would I, so I guess.
- how much time I can dedicate to doing my research work each week – I don’t know this – I am juggling part-time study with a job, running a household and having social/family time. I don’t know how tired I will be after a day at work. I don’t know what jobs will ‘intrude’ into any study space I try to create. So I guess.
- the duration each task will take – i.e. the amount of time needed (as stated above a guess) times by amount of time I can give each week (as stated above also a guess) therefore equals a guess. When it comes to fieldwork any form of real estimate also requires you to know about the availability of your participants for interview (again a guess, you don’t know when their holidays are or what their other demands are).
It doesn’t take much of a hunt on Google to find out that I am not some strange, isolated voice questioning the value of Gantt (and gantt like) charts.
For example, blogger Chris Burnley writing about agile processes in software development concludes:
Gantt charts have no place in an agile process; they are the tool of bluff used by the project manager that really doesn’t understand software development.
Source: Objectopia https://objectopia.com/2009/11/17/the-arrogance-of-gantt-charts/
So it seems that in asking me to create a ‘timetable for completion’, the university puts me in the position of needing to bluff – of pretending I understand the research process, my time availability and that of my participants enough to ‘plan’ it.
If you look into their background, Gantt charts (the idea of representing a timetable on a single page and then reviewing progress against that plan) were introduced by Henry Laurence Gantt (1861-1919). Note the dates, this guy was a contemporary of Taylor, in fact he was a ‘disciple’ of Taylor and close friends with the Gilbreths (the pioneers of time-motion studies). Gantt is included in a book called ‘False Prophets: the gurus who created modern management and why their ideas are bad for business today’ (Hoopes, 2003). From what I can gather from the text available through google books, Gantt introduced gantt charts to measure the production of rifles in an armoury – he could count how many could be made in a day and then could multiple up to see when/if he could meet his orders. If he needed to complete orders more quickly he needed to make sure more rifles were made in a day. Gantt later transferred to the shipbuilding industry – ships take longer than a day to make so he had to introduce the number of rivets driven in a day as a way of tracking progress!
[Those with access through journal paywalls may also be interested in the article Wilson (2003), Gantt Charts: a centenary appreciation, European Journal of Operational Research, 149 (2), 430-437)]
This means that Gantt charts emerged in the era of ‘classical management’ – a time when we believed things were predictable and controllable; a time of production (making things) rather than a time of knowledge work; a time when we believed that the workforce could be motivated entirely by pay and reward; a time when organisations themselves were seen as machines that pursued pre-selected goals. In short, in an era of positivism. As a tool (technology) that was developed in that era, the Gantt chart embeds all the assumptions of its age.
So, if we question those assumptions. If we now accept that things aren’t predictable and controllable; if we are doing something that is knowledge rather than production based; if we have a much more humanistic approach to the people around us; if we realise that positivism isn’t the most appropriate epistemology for what we are doing – why, why, why do we carry on using a tool that makes us default back to those old assumptions.
This is particularly problematic for me as someone doing action research. Action research was born out of the critique and rejection of classical management by Lewin and others. I’m in a bizarre scenario where my research methodology is incommensurable with a project methodology I am being advised to use.
I mentioned that software developers have completely ‘overhauled’ their ways of managing projects away from the waterfall approach (predict everything up front and then deliver) to an agile approach. Agile approaches don’t dictate timescales and schedules up front – there is more of a test, try, learn ethos to them. An agile approach isn’t tool-free – you just use different ones to the more ‘traditional’ waterfall approach based on that classical view of organisations/management. For example, the kanban board is a way of visualising work and workflow, without the big focus on time and deadlines (see for example this article on kanban boards). This can be done with cards or post-it notes but there are some software applications (like trello.com) that mimic the physical kanban board. The whole thing is much more flexible – it is much more easy to bring in the un-predicted and flex priorities as the work flows – rather than feeling like you’ve ‘failed’ to meet the guesses in your Gantt chart. This is what I have started using to help me manage how I focus my time, track what I have yet to do and celebrate the progress I have achieved in my PhD journey. As a task comes more into focus, I expand my understanding of it when it is happening and can create the sub-tasks there and then. I can add in un-predicted things and change how to focus my time really easily. It’s flexible style fits much more readily with the flexible design of an action research project so it feels as if it ‘clashes’ less.
A long while ago, I wrote a post titled ‘If not a project, then what?‘, in some ways this post is a re-expression of the struggle I had then. That post ended with a description of a ‘newer’ wave of research into project management – research that recognises complexity and the need for ‘reflexive’ project managers. This is the sort of evidence that academics could be drawing on to adapt their own (and their students) project management practices – it seems to be a mirroring of the distinction between protocol-driven research designs and flexible ones – a protocol-driven research design may well work best with ‘traditional’ project management tools; a flexible research design surely needs to use tools that helps the researcher manage the flexibility and adaptation.
But as things stand, I still have to produce that ‘timetable for completion’ – a document that is no use to me whatsoever either as an end product or the process of production. Wouldn’t it be better if I am asked about my ‘project management approach’ and what justification I have for my choices. Afterall, that’s what we have to do with every other choice we make as part of our research design – why should this be any different?