Why you are not normal
Norman, D. 2002. The Design of Everyday Things. New York: Basic Books (2nd Ed.) chapter 6.
Pheasant, S. 1996. Bodyspace: anthropometry, ergonomics and the design of work. London: Taylor & Francis (2nd Ed.) chapters 1-6.
How would you describe yourself? Would you say you had some striking features or do you think of yourself as a average sort of person? In this session you will find out just how "average" or "special" you are. The session is about understanding what we mean by "normal". In it you are going to do a practical exercise to find out how normal you are compared with (a) the rest of your group, (b) people generally and then reflect on what this means for the way information should be presented.
By the end of this session you should be able to:
Explain what is meant by the "average fallacy".
Explain why it is unreliable to base design decisions on notions of what is normal.
Chinese Man Declared The Tallest Naturally-growing Human By Guinness World Records © Getty Images. 
It's important to know what sizes people come in when designing things like clothes, bicycles, car interiors, hand tools, anything in fact that people have to fit in, on or interact physically with. The science of establishing the size, shapes and other physical characteristics of people is known as "anthropometry" and its application in design is called "applied anthropometry" or "anthropometrics" (Pheasant and Haslegrave, 2003). Pair up with someone in your group to measure yourselves in terms of:
- shoulder height
- elbow height
- hip height
- fingertip height
- shoulder breadth
- hip breadth
- shoulder grip length
- head breadth
- vertical reach
(Its important that you do all 10 measurements, don't just do two or three).
If you want to see how these measures are defined, take a look at at the RoyMech website  (http://www.roymech.co.uk/Useful_Tables/Human/Human_sizes.html).
When you have finished, collate the figures in your group and calculate the group averages for each measurement. Calculate the averages for males and females separately first, then calculate a combined average for each measurement. Write up your results in your wiki.
How did you do?
How do you compare with the rest of the group? What do you notice about the male, female and group averages?
My measurements are slightly different when compared with the other elements of the male group. Even though all 10 dimensions are relatively close no one presented the same figures. The group was not able to compare the averages with total accuracy; nonetheless the available data shows that every person has their own particular measurements. (José)
My measurements confirmed that my height is in my long bones and not my torso. A lack of standardisation in measuring method between our groups makes the validity of any comparisons doubtful except within groups. Averaging data across the four males began to level out the big individual differences in height and reach - there was no female average as measurements were from a single individual. (David)
My measurements differ, as do others in the group. Averaging this number has proven averages can be misleading, as non of the group really fit into any of the average measurements we ended up with. Also the standardisation lead to the two groups measuring the same measurement in different way therefore resulting in a very inaccurate way to record data. We at least know none of us have abnormally large heads, arms etc. There is only one female, so an average was impossible to take, however i did measure Alex, the outcome was I cant use a tape measure accurately (mike)
Is anyone in your group average with respect to all 10 of these simple dimensions?
No one. (David)
How average is your group?
Now compare your group's averages with the table of data at the RoyMech website  (http://www.roymech.co.uk/Useful_Tables/Human/Human_sizes.html). How close are the group averages to the figures supplied by RoyMech? Is anyone in your group average with respect to all 10 of these simple dimensions?
Comparisons could not be made with the RoyMech data as these were given as population percentiles. Our male data was a group average (the group being too small to apply population statistics) while the female 'group' data was from a single individual. (David)
Using US 50%ile data no one in the group was average with respect to any of the 10 dimensions measured (David) (I think I just answered that question in the section above)
No single member of the group fits into these averages, and also as david has pointed out here, again our group is too small to draw comparison. (mike)
The fallacy of the average person
In an experiment designed to answer similar questions Daniels and Churchill (1952) categorised 4063 men (note, men only) according to 10 measurements used in clothing design. Not one was average in all 10 dimensions and fewer than 4% were average in even the first three. The experiment, which is described briefly here  (http://openlearn.open.ac.uk/mod/resource/view.php?id=159813), demonstrates clearly that there is no such thing as an average person. It is a fallacy. So far we have looked at just simple physical dimensions, but people vary in all kinds of other ways too. For example the way we see things varies depending on our colour vision and visual acuity (how many people in the group wear spectacles or contact lenses?). And it's not just how we perceive data, its also about how we process it and assimilate it and remember it. Different people presented with the same picture will observe different things in it, depending on their interests, knowledge, surroundings, previous experiences, cognitive styles, etc. The same thing happens when you present people with Web pages, magazines, catalogues and so on.
Why do you think this matters?
Well, for a start, if there is no such thing as an average person, then we can't design for the average when we want to present some complex information such as an exhibition, or a catalogue, or a research database. Secondly it means we can't rely on our own judgement about what will work.The idea that "this works for me, so it should be OK for other people because I'm pretty average" is commonly held, even among designers. But if we all differ from the average in so many ways then what works for me is quite likely not to work so well for you. So what can we do about it? Well, there are three ways out of this dilemma. Anthropometry, psychology and other sciences which systematically collect and analyse information about large groups of people using statistical techniques are one way. Another is scientific experiment in which observations are made under carefully controlled conditions. Both these methods tend to be expensive and time consuming (although they produce very reliable data when done properly). The third approach, which we are going to use in the next session, is user-centred design.
Prepare a brief presentation (maximum 5 wiki pages or Powerpoint slides, not counting references) on what you have done and what you have learned from this session.
You may find it useful to illustrate your report with some of pictures taken during the activity on the anthropometric measurements activity page.
To prepare for the next session read:
Norman, D. 2002. The Design of Everyday Things. New York: Basic Books (2nd Ed.) chapter 7.
Krug, S. 2000. Don’t Make Me Think: A Common Sense Approach to Web Usability. Indianapolis, Indiana, USA: Circle.com Library,New Riders Publishing. Chapter 9.
Buxton, W. 2007. Sketching User Experiences. San Francisco: Morgan Kaufman pp. 77-115.
Follow this link to the Photography Resources in a Digital Age Contents page to return to the module contents.
Back to the MA: Photographic History Main Page to return to the Course contents.