I have decided to restrict my study to Wales, due to time and financial constraints of reaching the wider UK population of Architects. There are 1,179 ARB registered Architects in Wales (2019 figure) and 642 chartered RIBA members. The question for me is how many Architects out of the total population will provide me with a reasonable representation of the overall population?
A few things to consider when determining a sample size:
- The less variable (more homogeneous) a population, the smaller the sample size needed – I will be asking mainly Architects, so a homogeneous population.
- Generally, if the nature of the topic is clear, fewer participants are needed – this is true for me, as it is made clear to participants what an Active Building is and I am asking for responses on one document – the Active Building Design Guide.
- Quality of data gained from the research – Architects tend to be good at expressing themselves and able to identify and articulate what they like and don’t like about something quite easily. Therefore, fewer participants are likely to be needed before saturation is reached – Architects don’t hold back in giving their opinion!
There are published tables which provide figures for sample sizes depending on the size of the overall population. These figures are for obtained responses, so it is common for researchers to add 10% to the published tables to account for non-responses. While this is more relevant when seeking responses to mailed surveys, it is good practice to target more participants than the minimum recommended, which also accounts for cancellation of planned focus groups.
To determine my sample size, I have been using a table originally published in 1992 in a fact sheet produced by the University of Florida, which has been used by many qualitative researchers since publication, hence widely accepted.
For my population of 1,179 Architects in Wales, the published table suggests I will need 91 participants, for a precision level of +/-10% and with a Confidence Level of 95%, i.e. to be 95% certain that if I carried out the same study with more of the population, I would get the same or similar results. To ensure I have sufficient feedback, I will target 100 participants, i.e. an extra 10%.
I will also include participants from other disciplines, such as Housing Associations, Main Contractors, and other consultants, but will not be working out a particular sample size for these, as my main target audience is architectural designers.
For the architectural students included in my study, I am using a different strategy. There are only two Schools of Architecture in Wales – one, the Welsh School of Architecture (WSA) in Cardiff, is well established and the larger of the two; the other school is in Swansea – University of Wales Trinity Saint David (UWTSD) – and was only established in 2015. The WSA course structure is well established and they have 115 places on their Architecture course. There are hence 2 issues here: firstly, it could be difficult to influence the well established course within the timescales of this D.SBE project; and secondly it would be very difficult for me to work effectively with such a large group of students – any more than about 20 students would be too many for me to work with. The small group of students at UWTSD and the relatively new course structure, which is still being developed, provides an ideal opportunity to trial my developing design guidance, before it could be rolled out to students at different institutions.
If any Architects are interested in learning more about Active Buildings and helping to contribute to my developing Active Building Design Guide, you can attend one of 3 RSAW CPD sessions which will be held in Swansea, Cardiff and Llandudno during February and March, or contact me separately to arrange a CPD session at your office: firstname.lastname@example.org.