Research on a shoe-string
26 March 2013 3 Comments
Emily conducted her PhD at the University of Sydney on promoting fruit and vegetable consumption to Australian young adults. She graduated in 2012.
Her honours, masters and PhD projects had a combined budget of less than $400.
Emily is in the process of writing her first set of internal grant applications as an academic staff member, and is interested in the process of developing projects in the context of conducting research on a shoe-string.
I’ve been going through my paperwork from my student days recently. In the process, I found my funding requests for my PhD research. Not including conference travel, my research expenses for my PhD were $375.95.
That included a 1-month subscription to Thinkstock to allow me to buy high quality images for use in an online intervention to increase fruit and vegetable consumption, and the purchase of the domain name that I used for the intervention website. My Honours and Masters projects, and the research I’ve been running for the year since completing my PhD, have all been conducted at zero cost (except for my time).
This means that in the last 6 years I’ve spent an average of $62.60 a year on research costs.
At research institutions, developing, submitting, and ultimately receiving, competitive grants is a key indicator of productivity and performance for academic staff. This means that obtaining a Category 1 grant (e.g. ARC Discovery or NHMRC Project Grant) is central to my career development.
Assuming that I want to progress in my career (spoiler alert: I do!) then I would be expected to apply for a faculty-level internal grant ($$), a university-level seed grant ($$$), then a Category 1 Grant ($$$$$$).
As a freshly minted academic staff member, I’m starting small with the preparation of a faculty-level grant (the maximum budget is $18,000, with all funds to be spent in a year). In the process of preparing this grant, I’ve had to think about spending about 47 times more on research than I have ever before. Obviously, I don’t need to ask for the whole amount, but spending months putting together a request for $62.60 in funding would be colossal waste of time for everyone involved!
So, given my history of tiny research spending, it’s not surprising that I’m finding the process of writing this internal grant (or, more precisely, its budget) challenging.
Over the years, I’ve become accustomed to designing research so that it has no (or very few) costs attached to it. But grants require expenses, so I’ve been spending the last couple of months trying to come up with a project that actually requires some funding. One of the difficult things with this is that most of the advice on how to write budgets assumes that you know what you want to spend money on, and – if anything – you want to spend more money than is allowed.
While trying to find ways to frame research questions and write budgets when I’m not used to having money to spend, I’ve reflected on the things that I have done over my research career that allowed me to spend very little on research.
To those of you who know me, it won’t be a surprise that most of these relate to the ways we can effectively use technology to facilitate research.
Below, I’ve listed ways to abandon everything I’ve learnt and increase your research costs (some tongue-in-cheek). Hopefully, this is useful for those of you out there who are struggling to create your budget!
1. Don’t use electronic data collection! Make sure all of your research materials are paper-based to maximise the amount of time needed for data entry.
After spending hours entering paper-based surveys into SPSS for my Honours research, I’ve ensured that all of my data collection has been electronic. Electronic data collection (either online or in-person) means that research data is entered directly into the study database. This is faster and less prone to error than manual entry, but it has one major downside. It’s too cheap! If you’re trying to get your budget up, you’ll have to budget for data entry time for a research assistant (or assistants), as well as all of the printing and dissemination costs for your paper based materials.
2. If you must use electronic data collection make sure to use expensive (preferably custom-built) software.
There are lots of excellent open-source programs out there for data collection. I use the open-source survey software LimeSurvey for online surveys, and for completing in-person data collection when a computer is available. Limesurvey allows for conditional questions, a range of response formats, and is completely free. It’s so flexible that I now use it to help run a lot of my intervention studies (a colleague used it to deliver a web-based Cognitive Behavioral Therapy (CBT) program for people with coeliac disease).
If you’re trying to keep your budget high, don’t even think about whether there are open-source software options that would be suitable for your project.
For the best results, include the design and development of software in all of your grants. It’s especially good to make sure that the program is so complicated that you and the research team don’t own it and can’t make changes to it on your own. Ongoing development and support costs will really help to keep your budgets high over the life of your research career.
3. Don’t automate anything.
My final PhD project involved sending an automated email to every participant once a day for 30 days (200 participants x 30 = 6000 emails). People could sign-up at different times, so I would be sending Participant 1 email #23 at the same time as sending Participant 55 email #3. For my PhD, I fully automated this process so that participants would sign-up to the study, complete the baseline questionnaire, be sent their daily emails, then complete the follow-up questionnaire. Once it was done, I’d receive email notification that their data was complete.
An earlier study in my PhD required me to send emails manually every 3 days to 100 participants for a month. I processed these by hand. The automated email system took 2 days to set up and now runs projects for several other researchers I know. The manual system took about 2 hours every morning (24 days over the life of the project) and sent me slightly insane.
The great thing about research funds is that you can use them to pay research assistants, who cost a lot of money and save your sanity without the need for automation. If you want to make sure your budget stays high, don’t automate anything. This would only take away valuable hours for your research assistants to do mind-numbing work.
Note: If you’re a grant reviewer reading this, please don’t cut my budget in half! I have finally managed to develop a project that does involve legitimate costs. It has electronic data collection, open-source software and automation, but requires a time-consuming testing procedure. I really do need all the money, I promise.