Earlier this year we interviewed Mark Baskinger, associate professor at the School of Design of the Carnegie Mellon University. In the interview Mark talks about drawing ideas and shares his thoughts about the differences between industrial designers and interaction designers and how interaction designers can use sketching to communicate their designs better.

Mark Baskinger teaches industrial design with an emphasis on form and interaction and conducts Drawing Ideas workshops

Note: the first minutes of the interview we had minor problems with the audio. No worries, this will go away…

In Johnny TV Features we’ll share with you interesting videos that we come across, enriched with our healthy opinion. This time we have ‘The Domestic Gubbins’, a video by Microsoft Research.

At first glance this video by Microsoft Research seems a bit whimsical or silly. But if we dig a bit deeper, we find there is a lot more to it, and there are some useful things that we can take away from it.

What is the video’s purpose?

The first thing to consider is what the video was for. What was its purpose? Once we understand this, we can go on to consider what the researchers got out of it and how this helped them in specific area that they were investigating.

So what was the video for? First, a bit of context. The Domestic Gubbins are part of a project called Objects Incognito (subtitle: Rethinking Machine Intelligence), which is “an ongoing enquiry into everyday ideas of intelligence”. We are promised a plethora of “intelligent” devices and technologies in the near future. But what is meant by intelligence, exactly, and what will it be like to live with these ubiquitous intelligent technologies? That is the question that this video attempts to answer.

Interviewing people

Why video, though? Jain and Taylor originally wanted to create the Gubbins as actual devices that they could give to people to live with and interact with. But this proved too challenging, so they decided instead to create this video, which shows how people might interact with the Gubbins. Then they showed it to people and interviewed them to find out what their thoughts and reactions were.

These interviews (excerpts of which can be seen here) provided the researchers with new insights and led them down new research paths.

How can we use this?

So how can we adapt and adopt this approach for use in our work? This research is very high-level and conceptual, whereas in our day-to-day work we usually deal with matters that are much more concrete. However, there are many situations where we would like to be able to put a product in users’ hands so that they can play with them (especially for products that are radically different from those currently in use or that address as-yet-unmet needs), but where we do not have anything close to a working prototype. In cases like this, we can use this approach to show how our new product might function and how people might use it, and then see how potential users react. We can then use the insights that we gain to guide the direction of our product.

What are the weaknesses of using an approach like this? The video itself has to strike the right balance between showing realistic scenarios of use and keeping things ambiguous enough to make viewers think and use their imaginations a bit. And when interviewing the viewers, a certain degree of finesse is needed to avoid drawing interviewees in a direction favored by the interviewer.

In conclusion, this is a very well-made video that we can borrow ideas from for certain situations (though definitely not all).

Johnny TV

This and many other UX videos are posted on Johnny TV. Should you come across a video that you think should be there, please contact us via mail or Tweet me @martinpolley.

I started wondering last evening what twitter would be like if in addition to followers we could also see who was actually being paid attention to. The groups many of us use in clients like Tweetdeck or Seesmic, for example. So in the midst all of our positive talk of transparency and authenticity, I found myself chuckling at the opacity we in fact rely on to make it through the day.There’s nothing wrong with this, and while some may see a cynical twist or twitter’s dirty little secret (nobody’s listening!), I see instead perfectly reasonable social media coping mechanisms…

Social media’s two audiences

Social behaviors are shaped and informed by design, but not explained by design. The obvious reason that none of us can see each other’s twitter usage (groups, or subsets of followers actually viewed and paid attention to) is that if designed into twitter, activity would change instantly and radically. This is not just a matter of privacy, but a deeply social matter.

Reflecting on this last night led me to thinking about the social and public space constructed across all social media. There are, in mediated social contexts, always two audiences.

• There is an audience we’ll call social, and which we describe in terms of proximity: it’s a internalized social world of friends, peers, colleagues: known individuals.
• And there is a second, anonymous public, which is not internalized but is imagined.

Any person known belongs in the social and is potentially present. Any anonymous individual, because we don’t yet know them (as soon as we do, they move to the internalized social world), is possibly present.

Potential and possible relations

Potential social relations become active relations, or interactions, when we communicate. Possible relations become actual relations, based on the action of following, when we are seen and found.

I think the doubling of audience could go far in explaining the power of social media.

We know, for example, that the probability of actually having a conversation is less in social media than it is face to face. There’s simply a lot more at our command in face to face situations by means of which to have conversation. However, face to face situations limit us, of course, to those in our presence. Social media may reduce the probability of having real conversation but increase the opportunities for creating conversation.

This seems, to me, the main reason we use social media. Not mass, but mini media. Or, “me”-dia, in the context of social, not mass audiences. The distinction between social and mass media being that relations are possible in the former, not so in the latter. (This is changing as mass incorporates social.)

The medium’s three modes: mirror, surface, window

Back then to attention, and the veil of nondisclosure from behind which we engage in social media. I like to say that the social interface has three modes: mirror, surface, and window.

• We see ourselves reflected in social media: this is it’s mirror mode.
• We consume content of all kinds off the screen — sites, apps, communication — all using the screen as a presentation layer: this is its surface mode.
• And we talk to each other through social media: this is its window mode

Modes of attention

Social presence, proximity, and attention are then each implicated in a mediated social context that has ways of seeing and ways of being seen.

Consider this, for example. We enjoy accumulating followers, seeing ourselves referred to, commented to, and otherwise being made visible. Doesn’t matter whether this involves acknowledgment, recognition, or validation; the point is that the medium does create a kind of social visibility. Call it, for simplicity’s sake, “being paid attention to.”

Well, attention doesn’t correlate with actually engaging in conversation. Many of us sometimes ignore a request for communication, for whatever reason. It’s part of daily life; in real life it’s called “civil inattention,” and is handled by acknowledging others in ways that also indicate to them “I see you, recognize you, but I’m not available to interact.” Simply put, politeness.

Now, consider the social media space. Attention paid to others may not be visible to them. But if it’s given, such as by taking any action recorded and captured by the medium and surfaced by design, then this action can have two social outcomes, not one. This is the power of the medium, and the net effect of the doubled audience mentioned above.

Social actions, social relations

One translates as the potential for further social action. The other translates into the possibility for social relation. For the social world already has relations but has activity only on the basis of user actions. And the public world has activity but lacks the connection until a relation is established.

• A social action has been made which can be picked up by any user who sees it: potential for further action
• A social action increases the user’s visibility: the possibility of being seen

The possibility of being seen is motive enough, for some. While communication is no more probable, the possibility is there. As they say of the lottery: your odds of winning increase dramatically if you buy a ticket.

The power of this second audience, the public, which creates infinite possibilities and which is motivation for much of what we do, explains a lot of how the attention economy works.

Perceived and transactional influence

Attention, interestingly, is described in economic terms: paid, spent, given, taken. Note that the first two are zero sum and involve the temporality of attention. Paying attention takes our time. The second two are non-zero sum and transactional.

Giving and getting attention is the simplest social action. Nothing yet has to be said or communicated verbally: attention can be given a person, and that in itself, is socially meaningful.

Now consider how we attend to the attention economy in social media. Brands, as well as users, watch and attend to it. Brands, as well as users, transact in it.

• Social capital, the perceived value of a brand or individual, collects attention paid and spent on that brand or person. Call this perceived influence.
• Social currency, the transacted value of a brand or individual, is attention given and taken by the brand or person by means of social actions. Call this transactional influence.

Unfortunately, perceived influence, which is just social observation, is grossly under-rated. It’s much more difficult to measure because there’s no action taken. Brands can’t see the value in it for it’s not in the numbers provided by metrics and analytics tools. For it lies behind the veil of personal social media use, in the activity of paying attention to twitter, or more specifically, to the users we actually follow.

I say this is unfortunate because i think much social action is preceded by long periods of social observation. Consider the difference it would make, to brands and to users, if all social media were split screen interfaces: what I see and what you see. Real life social situations are like this: I see you looking at me, and can see reflected in your face something of how you see me (what you think of me).

Motives explained by the social and the public

The dual public also helps to explain many of our motives in using social media. Again, our actions can lead to potential further action, and if not, are at least possibly seen. Tweets, like comments, reflect these motives.

For example:

• Tweets or comments intended to get attention from the author
• Tweets or comments soliciting or appealing for direct response
• Tweets or comments that are a direct response
• Tweets or comments that continue a conversational run or thread
• Tweets or comments intended to garner attention to their author

We could break each of these down and show that for each, the user’s motive may be to appeal to the author’s attention, to get visibility in front of the public, to solicit a response, or to respond. Tweets and comments, in other words are not just that: (Nothing is explained if we describe social action by its form of content.)

Summary

To conclude, then, I think that the fact that any use of social media can have outcomes in two distinct audiences may explain its uniqueness as a medium, and its use by brands and individuals alike. That the attention economy involves both looking and being seen, posting and responding, would explain why motives for participating in social media reflect to the “presence” of two audiences. These are properties particular to the sociality of the medium, and to the sociability of its uses.

Johnny is having a party today, so let’s give away cool prizes. We organized a simple, but cool contest. It is really easy to join and at the same time you are contributing to the community. So join, and win.

What can you win?

We thought about giving away post-it notes, sketch pencils or paper prototype material. But in the end we thought this wouldn’t do. You guys have high expectations, consume our articles like it’s candy and always yearn for more knowledge… so we decided to give away stuff that you’d actually love.

We are giving away 4x two books from Rosenfeld Media. When you join the contest you have a chance of winning one set of two books. And you can decide for yourself which books you want, which will be a hard job when you look at the collection:

From left to right: ‘Web Form Design‘ by Luke Wroblewski, ‘Card Sorting‘ by Donna Spencer, ‘Remote Research‘ by Nate Bolt & Tony Tulathimutte , ‘Mental Models‘ by Indi Young, ‘Design is the Problem‘ by Nathan Shedroff & ‘Prototyping‘ by Todd Zaki Warfel.

What do you have to do?

In order to win one of the prizes you have to do something: share your UX knowledge with the community. We want you to share your best UX tips via Twitter with other practitioners. You can do this as often as you want. The tweet has to have the following hashtags: #uxtip & #johnnyholland. That means you have 118 characters left for your tip, so keep is short and juicy.

• tweet a UX tip
• add hashtag #uxtip
• add hashtag #johnnyholland

An example:

In the end we want to end up with a huge collection of brilliant UX tips.

What are the rules?

The rules are very simple. You tweet, we decide who wins. Only people that followed the (simple) rules of the game have a chance to win. There is no possibility to discuss the outcome of the contest. We will only inform the winners of the contest. They will get an e-mail where we ask for their contact details. Winners have two weeks to reply, if that doesn’t happen… we will choose a new winner.

The contest starts on October 25 and ends November 8. The winners will be informed before November 15. You can send in as many UX tips as you want. Everybody can compete, except for the kahunas and dudes.

Special thanks to Rosenfeld Media for sponsoring this contest. Don’t forget to follow them on Twitter: @rosenfeldmedia.

When I was checking some statistics a week ago I realized Johnny was reaching a special day: his first birthday. Since it is our child, we would love to have a small party for him. So let’s start of with the speech and see what goals we reached in the past year.

Let's party (image by k bost)

Exactly a year ago we launched Johnny Holland. It was to become an open collective around interaction design, a place where everybody could talk, learn and share around this topic. And I think we reached this goal… now we have a huge group of rock stars sharing their knowledge and ideas with the audience. Within this group are people from all over the world and from different disciplines. We have well known people from our field, inventors, new talents, students and loads of other great Johnnies. The constant energy that they put into their articles is magical. And it is really important for our team of kahunas and dudes to support this energy and make it as easy as possible for the rock stars to publish their articles.

We started really small, basically me editing all the articles. Fortunately there were a lot of enthusiastic people who loved to help out make it a great place. So I would like to give a lot of credit to my fellow kahunas Steve Baty and Will Evans and the dudes Dennis Koks, Vicky Teinaki, Niklas Wolkert and Martin Polley to help make this a fun project, as well as all the rock stars that wrote articles.

Some statistics (because they rock)

Now we are 365 days further and when looking at the statistics we can say we reached some superb goals. So let’s show off a little bit:

• We have 66 registered Johnnies, and most actually wrote an article;
• A total of 209 articles was published, which is an average of four articles a week;
• About 2.000 people registered to our feed;
• We had +61.000 unique visitors that stayed for an average of 2:38 minutes;
• We allowed +850 comments;
• Fortunately we also blocked +14,500 spam messages;
• In Australia, Denmark and Sweden we were official media partner for UX events;
• We helped out with several Mozilla Labs Design Challenges;
• Saying that the iPhone isn’t easy gets you +12,400 visitors.

Most popular articles

We had some great articles, which started with Adrian Chan’s ‘A Social Interaction Primer‘. But the top 10 articles of the past year are:

1. ‘The iPhone Is Not Easy to Use‘ by Fred Beecher
2. ‘Deconstructing Analysis Techniques‘ by Steve Baty
3. ‘What is an Experience Stategy‘ by Steve Baty
4. ‘Good IxDers Borrow, Great Ones Steal‘ by Vicky Teinaki
5. ‘User Stories: A Strategic Design‘ Tool by Penny Hagen & Michelle Gilmore
6. ‘Introduction to Interaction Design: An Interview with Dave Malouf‘ by Will Evans
7. ‘Learning from Games: A Language for Designing Emotion‘ by Joe Lamantia
8. ‘Design for Interaction: Ideation and Design Principles‘ by Dan Saffer
9. ‘Selling What We Do‘ by Daniel Szuc
10. ‘Metaphor on the Brain‘ by Dave Malouf

Johnny is growing up

From day one we stated that Johnny Holland Magazine would just be the first step. After that we worked on new initiatives. In the past year this resulted in two new things. The first was UX calendar: an online calendar with all the important UX events around the globe. The second project we launched was Johnny TV (run by Martin Polley): your online source for inspiring video’s.

For the coming year we have a lot of new ideas and goals. We want to set up new initiatives, publish even higher quality articles and get a bigger audience. Your ideas, help and energy are always welcome, both for new initiatives as well as writing articles for Johnny Holland Magazine. So let us know how you want to contribute.

Looking for sponsors

We want to grow. We want to support the community. And in order to do that we need sponsors to help us pay for servers and tools. It would be super if you’d be interested in helping out.

Let’s go and have some fun for another 365 days

In our last article on Johnny Holland we provided an overview of what a ‘social life of a visualization’ might look like. Based on a person-centered social network, it showed how the identity of the visualization was important, and how having this allowed the underlying data to retain its integrity and facilitated the process of people interacting around it. Its implementation created a shared storytelling experience around visualization, and we broke this up into three phases; create, interpret and capture.  In this second article, we’ll delve more deeply into the creation phase of the ‘social life of visualization’; including its rationale and the design challenges that it represents.

The creation phase – choosing the right tool

In the creation stage of the shared storytelling experience, the initial dataset is presented as a visualization.  The problem that needs to be overcome is that people aren’t generally well versed in presenting information visually. So the purpose of this process is to help people to decide  how to visualize their data and communicate the meaning of their data to the online community without resorting to text.

While visualisation can be an ideal medium for people to tell stories about their data, the problem is that they don’t necessarily know the best visualisation technique (and by this we mean box plot, bar chart, scatter plot etc.) to use that adequately communicates what their data is about. While this is a problem that exists for the individual person when they are trying to gain some individual insight into their data, it is especially problematic when data visualization is being introduced into a social network. This is because other people need to be able to interact with the visualization, continue the shared storytelling process and add more knowledge to what was contained in the initial visualization.

So an integral part of building any interface that supports a social network for data visualization has to be including a tool that helps people to better understand the techniques they should use to visualize data. Enabling this allows them to focus on the type of story they want to tell through the data visualization, rather than becoming preoccupied with how to tell the story. So the intention of the first part of the creation process is to helps people to visualize their data so that other people within a social network can understand its intention and interact with it accordingly.

Guidelines

The theory behind this comes from the work of several important figures in the field of data visualization and visual thinking:  Tableau Software CEO Christian Chabot, ‘Back of the Napkin’ author Dan Roam, and noted visualization expert Stephen Few.

In his keynote address at InfoVis 2008, Chabot presented five flawed principles of data visualization.  Some of these are considerations for the way this particular aspect of the interface should be designed, or in particular, what it has to achieve to help users. The first of these flawed principles is that people adopt visual analytics primarily to help them see and understand new visual paradigms. The answer to this is that most people’s needs can be solved with tried and tested visualizations such as bar charts, line graphs and scatterplots.

Dan Roam’s work in The Back of the Napkin introduces a simple and straightforward methodology for visual thinking and problem solving. So some of what he talks about provides the basis for the reasons why a person might choose a particular visualization approach. Roam’s approach is to begin by thinking about what sort of question needs to be asked of the data. These ‘problems’ are clumped into those that involve:

• who and what
• how much

when

• where
• how
• why

There is then a corresponding ‘showing technique’ that equates to each of these problems, making a matrix (see below).

Dan Roam's visual thinking matrix from Back of the Napkin

There are only six of these techniques, one for each type of problem, and as Roam points out in the book, that is all that is needed. All visualization techniques are derived from the portrait, chart, map, timeline, flowchart and multiple-variable plot visualization approaches that he uses. His take is that other visualisation techniques are great but they are not necessary in this type of application.

This fits in with Christian Chabot’s second flawed principle of data visualisation; that people adopt visual analytics primarily to help them see and understand massive data. The truth is that people want to better understand small datasets more readily than large ones, and so complex visualization techniques are unnecessary for this. His fourth flawed principle is that people adopt visual analytics primarily to help them see and understand hidden insights. However the real reason that people employ visualisation techniques on their data is to save time.

Complicating this need of people is that according to Stephen Few is that they are still struggling to achieve simple tasks because existing visualisation tools complicate the task of making sense of data and effectively presenting it to others.

Goals

However the key to the creation process is to help people determine their analysis or communication goals and then suggest a visualization approach that maps most closely onto their stated objectives and is appropriate for their dataset.This is instead of forcing people to concentrate on learning the merits of different visualization approaches, and rather it helps people to focus on what they already know about their data and the context they want to present it in. This can be achieved by attempting to determine the communication or analysis goals the person has for their data visualization, including; who they will be sharing the visualization with, what kind of data they will be visualizing, and what outcomes they want the visualization to create.
Based on these factors, it is proposed that the interface would suggest a visualization approach for the data, explaining to the person why that approach is best suited to their goals. Along with this a range of other visualization approaches should be presented to the person, stressing their individual strengths and weaknesses.

Through this process, the person is required to have a good understanding of the original data to be able to choose an appropriate visualization approach that communicates the dataset in the visual medium.

How can a visualization become ‘social’?

As we discussed in our previous article, setting up the social space as an object-centered social network (e.g. Flickr) establishes the visualization as the object that interactions occur around. So while this explains why interaction will occur, it doesn’t necessarily encourage it. On the other hand, giving a visualization an identity makes it recognizable and approachable within the social space, and consequently does promote interactions.

Firstly, to understand the importance of identity to an object, consider its importance to a person within a social network; it is a way of uniquely identifying that person within the social space. It is also the most basic requirement of any social space. However social spaces aren’t always built around people. To refresh the idea of object centered sociality that we’ve discussed previously, it is an alternative to the idea that a social network is a map of relationships, and instead says that people within a social space are connected through the existence of an object. The object centered sociality theory suggests that when it becomes easy to create a digital instance of an object, the online services for networking on, through and around the object will emerge too.

How can users make their visualisations social?

Therefore just as people within social networks have identities so they can be uniquely identified, objects need identities as well. Social spaces have ways of creating these for people. One of these ways is through the creation of a profile that allows the person to provide specific information about themselves that would give other people on the system some idea about the identity of that particular person. Often used in conjunction with a profile is a profile picture or avatar that provides the particular person with a visual identity within the system. This gives other people on the system extra information about the particular person that can only be conveyed in visual form.

Identity of a dataset in Swivel

When visualizations are objects within a social system, they can have identities attached to them as well. A visualization’s identity is created by the title given to it by its creator, its description and the content of original data set as well. However, a visualization can also have visual imagery or an avatar attached to it, in order to more clearly communicate its identity within the social space.

This serves an important purpose for those people within the space that did not create the visualization, in that it adds an extra layer of identity to an object that they have no pre-existing familiarity with. Specifically it helps them to make sense of the visualization, which aids any collaboration which may occur around the object. For the creator of the visualization, it helps them to make sense of what they have created by thinking about its identity and what sort of iconography they might attach to it. An avatar also contextualizes the visualization’s place within a social space. In turn, this objectifies it and allows it to exist on its own within the social environment. It also reduces the cognitive load on other people, and allows the inherent meaning in the visualization to be communicated and consequently transferred to the community with greater ease.

The process can be achieved by integrating with the search APIs of person generated content communities to access images and media that relate to the content of the visualization. The only issue that arises from this part of the creation process is that assigning absolute meaning to media can be tricky, and often fails to communicate effectively across different cultures. People can ‘read’ images and media very differently.

Conclusion

Ultimately the creation phase is the most important in ‘the social life of visualization’ because it is where an individual idea or question about a dataset can be transformed into an object for social interaction. So this process needs to help people who aren’t visual thinkers to make that jump and set up the visualization as an object so that interactions can occur around it. It leads into the next stage which is interpret, where the interface should act so that people can drag further insight out of the data. We’ll talk about this in a lot more detail in the next article.

Background
In 2008 the Australasian CRC for Interaction Design (ACID) was approached by Deloitte Digital for their expertise in data visualization which was being developed through the Loupe Project. Deloitte Digital was preparing its accounting firm in Australia for the introduction of XBRL (eXstensible Business Reporting Language) which would see a significant change in the way business reporting was conducted. Rather that sending multiple reports to different agencies, XBRL would produce one set of data that agencies could draw upon for their own purposes when needed. As part of this change, Deloitte has released an online accounting platform called Accounts IQ which will change the relationship between accountant and client to become an ongoing conversation online. This process needs visualization to make complex business data more easy to understand for the client, and an interface to make this conversation process a good user experience. The Social Life of Visualization is the outcome of our research into this solution for Deloitte.

Transformation is the act of taking a set of values from a dataset, processing them in some way (depending on the aims of the research) and arriving at a new set of values with the goal of revealing some aspect of the data from a new perspective.

(This article is the fourth part in the Deconstructing Analysis Techniques series.)

This technique is characterised by the fact that the values are changed; that someone looking at the new values will be unable to work backwards to the original values; and that for each original data point there is a single, new data point.

In mathematical parlance, (and you can skip this part if you like) the difference between a manipulation technique and a transformation technique is that manipulated data sets are congruent with the original, whereas transformed data only maintains cardinality (i.e. the same number of elements).

So, what does that all mean? We’re talk here about analysis methods like:

• Scaling – taking one set of data and massaging them to fit a distribution or ‘shape’ of values.
• Moving averages – taking a number of consecutive values and averaging them as way of ‘smoothing’ the last value in the series.
• Weighted averages – calculate an average value where more importance – ‘weight’ – is given to some values.
• Weighted indexes – calculate an indexed score (against a baseline) where more importance – ‘weight’ is given to some values.
• Seasonal adjustments – an adjustment made to a data point to account for cyclical peaks and troughs to highlight the ‘real’ shift
• Differences – a method of looking at the changes between one value and the next.

Now, initially, most of these methods may feel pretty technical, quantitative and removed from standard design research analysis. However, they form a powerful collection of analysis methods that will better equip you in undertaking design research. They also represent fairly low-level mathematical/quantitative methods and are available in a standard spreadsheet program. More importantly, used properly, these methods – and transformation techniques generally – open up new avenues for understanding the people who will use the services and products we design.

Used properly, these methods – and transformation techniques generally – open up new avenues for understanding the people who will use the services and products we design.”

Scaling

In “Deconstructing Analysis Techniques” we used the example of fitting test scores to a pre-determined probability distribution – scaling – as the example for Transformation techniques.

When we measure a population characteristic – such as height, or a test score – we create a sample set of data for that characteristic (unless we are measuring the entire population). There are times when the raw distribution (the frequency of occurrence for each value in our data) of results is not what we’re after. We may wish to compare the shape and attributes of two separate samples – two groups of test participants, for example – and so we transform the two sets of data so that they share a common mean (the average value for the data set).

Usually this is done to bring both sets of data to what is known as a ‘normalized’ distribution with a mean of 0. Of course, in our test/exam result example, we want to adjust the scores so that the class as a whole receives a pre-determined number of A, B, C, D & F. What we’re doing here is to adjust the overall shape of the data. (In these cases a plot of the raw data will look different to the scaled data.) When graphed the scaled data will look roughly bell-shaped, with the middle – or ‘hump’ – representing average performance, and the two thin tails representing high-performance (at the top end) and failure (at the bottom end).

Moving Averages

A moving average is used to smooth out day-to-day fluctuations with time series data. It is, literally, the average of the previous x days’ worth of data. A good example would be the number of page views received by a site. Each day the data will jump up and down, creating a sense of “noise” that makes analysis difficult, and, when a small number of observations are looked at in isolation, can create a false impression. A moving average is useful in time-series or longitudinal studies where we measure the value of a characteristic for a single object (person, server, site etc) over time.

One rather well-publicised and important example of this is the series of global temperature readings that have been used by both sides of the climate change debate. Skeptics of global warming point to a recent period of observations (2002 – 2007)   which show a decline in global average temperatures. When the same data is looked at using a moving average, smoothing out the peaks and troughs, a clear upward movement is seen.

The choice of time period to use when calculating a moving average is based on the specific circumstances of the data. However, common sense is usually all that’s required. For example, when looking at Web traffic, a moving average calculated over 7 days is sufficient to counter spikes that occur during a given week. You might also calculate a moving average over a month if fluctuations occur over a longer cycle.

Weighted Average

Weighted averages aim to address one of the criticisms of a moving average – and other types of averages – that being all values in the average are treated equally. It is often the case that one observation is more significant or important that another.

Let’s say for example we’re measuring the time to complete a task in a user evaluation session. We have representatives from each of our personas (or other audience segments): 2 primary personas, 3 secondary personas, and one tertiary persona. In this case, the performance of the two primary persona representatives is far more significant than that of the tertiary participant.

When we calculate the mean time-to-complete value, we can weight the results so as to reflect the relative importance of each participant. We may assign (and the exact values will vary for you) a weighting as follows:
Primary: multiply by 9
Secondary: multiply by 3
Tertiary: no multiplier

What we’re essentially saying is that our secondary personas are three times more important than our tertiary persona; and that our primary persona are three times more important than our secondary. We could just as easily use a factor of 2 (instead of 3) leading to values of 4, 2 & 1 in the example above; what matters is that we use weighted averages to adjust the dataset to account for the relative importance of some measurable data set by some exogenous variable.

Weighted Index

An indexed value is one measured in terms of some baseline figure. The aim is to convey movement around a starting point when there is no way to specify a zero.

An example of an index might be a satisfaction score. Since satisfaction a largely subjective measure, there is no way to define a zero point. Instead we typically measure a ‘pre’ figure and map that over time. Common values for an index are zero and 100. The choice is arbitrary and is typically chosen for clarity in communication.

Indexes are often calculated as an aggregate of a number of measurements. But it is also the case that we sometimes need to treat the data we receive from one group as being more important than another. This is where a weighted index comes in handy. A weighted index  – like our weighted average – treats different values as more or less important.

So, if it is common practice to design a product or service to better meet the needs of our primary audience segments; it also makes sense for our satisfaction index to put more stock in the satisfaction of our primary segments. We do this by applying a weighting (some multiplier) to each piece of data collected based on its relative importance.

We could easily do the same with responses to a question like “Would you recommend this service to a friend?”

This technique provides us a with a convenient way to build positive bias – towards the needs of our important audience segments – directly into our research methods.

Seasonal Adjustments

Some of the things we observe in design research are subject to cyclical variations. We may not, however, want to include a change in our data due to “seasonal” fluctuations, instead wanting to identify “real” changes (in frequency of use, for example).

In order to look at the real changes in our observed data we need to account for the seasonal variability first.

A familiar example might be to look at the number of page views or unique visits received by a site. We might see a big lift in traffic between Sunday & Monday; and a big drop between Friday & Saturday. In order to tell whether an observed drop in traffic on some Saturday is “normal”, we need to look at the regular pattern of changes and “adjust” the Saturday figure.

One way to do this is to calculate the average drop in traffic over time (between Friday & Saturday) and then apply this to the current observation for Friday. This as a predictor or estimator for the current Saturday, which we can then compare against the actual observed data. The average difference acts as our seasonal adjustment.

The Adaptive Path Aurora concept uses a scenario where a farmer shows that their farm will still have rain, using seasonal adjustment. See video

Differences

There are times when what we’re interested in knowing is not the raw value of an observation but the change between one observation and the next.

The calculation (transformation) is simple: for each pair of observations, subtract one from the other. Of more interest is why we would want to know such a thing.

Consider a test of a new design in which we test first the time to complete a task with the current design; and then the same task with a new design. Across all participants in the test the raw observations (i.e. time to complete) is far less interesting than the change in that time as a result of the new design. (Note that we may wish to express that change as a percentage rather than a raw value.)

We can use the same technique to highlight the variability of some observation over time. For example, we may be tracking the number of connections or ‘friends’ a person has in some social network to understand the relationship between the current number of connections and the rate at which new connection requests come in. To identify the number of new connection we simply calculate the difference between successive observations.

Summary

Although primarily applied to quantitative data, transformation techniques are useful in a wide range of design research activities beyond the quantitative.

Transformation of our research data can act as a way of reducing noise and bringing into sharp relief characteristics of the underlying user behaviour. The act of transforming removes us from the raw, original data, but in doing so we can gain the opportunity to uncover meaningful insights hidden from us otherwise.

Recently I came across an article in the Wall Street Journal about online ratings. The article, which surveys a number of online properties, cites the tendency to 4.3: On the Internet, Everyone’s a Critic But They’re Not Very Critical. The article’s authors pretty much capture what many of us get intuitively about why online ratings really don’t work, but I thought I’d break this down from a social interaction design perspective to get at some of the causes of this. First and foremost is the fact that most online systems built to capture user tastes, preferences, and interests engender bias. And online media amplify bias, for a number of reasons.

This bias originates with the user’s intention, which goes unknown and is not captured in the rating system itself. The reasons a user may have for rating something can be many: a mood, attitude, a personal interest, a habit of use, interest in getting attention, building a profile, promoting a product, and so on.

Amplified distortions

Social media, because they provide indirect visibility in front of a mediated public, amplify any distortion baked into the selection itself (a selection being the act of rating something). This amplification is explained in part by the de-coupling of selective acts (rating) from consequences and outcomes.

Selections are de-coupled from personal consequences, which excuses a certain lack of accountability and responsibility. Selections are de-coupled from their context of use, which range from personal utility to social promotion. And selections are de-coupled from social implications, which removes the user from his or her contribution to a social outcome (eg, highly-rated items look popular).

Consider the reasons a user may have for making a selection (rating something). They include:

• personal recollection (like favoriting)
• to inform a recommendation engine (so that it can make better personal recommendations)
• because the item is a favorite (sharing favorites)
• because the social system has no accountability
• because it always creates the possibility of recognition for the user
• because it promotes the item
• because it’s nice (socially; possibly karmic)
• because it’s a gesture about how the user felt

Social selections are thus encumbered by ambiguity: of intent, of meaning, of relevance, and of use.

Can these be addressed and resolved by better system design? Or can they only be resolved by social means?

Considerations

It might be possible to couple ratings with outcomes. This would involve new sets of selections and activities made available to other users and used to create consequences. Users would then consider these consequences when making a rating selection.

Contexts of use could be distinguished, so that users rate with greater purpose. This would involve creating new views of rated content, such as “rate your favorite item this wk,” “rate your favorite genre,” “rate your personal favorite,” “rate which you think is the best,” and so on. Each of these distinctions, if followed by users (!) would specify the selection by means of a different social purpose.

Reduce ambiguity

It might be possible to reduce ambiguity by means of some cross-referencing achieved by algorithms and relationships set up in the data structure. Without detailing these, they would probably include means by which to distinguish:

• the bias of the user him or herself, measured in terms of personal tastes
• the domain expertise of the user, as demonstrated by ratings provided by the user on other items and in which categories/genres/domains
• the social communication and signaling style of the user, which would reveal some of his/her relation to the social space
• use by other users and the public, as a measure of relevance

Cross references could then be applied when aggregating ratings, used to filter and sort the ratings sourced for averaged results. Theoretically, the system would be able to identify experts, promoters, favoriters, and others by their practices.

Social solutions

Social solutions might be created to supply distinctions among the different kinds of social capital involved in ratings. Such as:

• the user’s expertise (domain knowledge)
• trust capital, or the user’s standing within his/her social graph
• credibility capital, or the user’s believability, as measured in loyalty perhaps
• reputation capital, or the tendency of the user’s ratings to be referred to and cited beyond his/her immediate social graph

Finally, ratings systems can diversify possibilities for making selections, and separate communication from ratings selections so that ratings are used less for visibility and attention-seeking reasons (eg users who rate a lot).

There are too many kinds of socially-themed activities and practices in which ratings play a part for me to delve into this here. But each theme could be examined for the social benefits of ratings, for how they attribute value to the user, add value to content, and distinguish social content items to result in shared social and cultural resources. Those distinctions could be used to isolate different rating and qualification systems so that they are tighter and less biased.

In my previous incarnation as a philosopher, I spent a lot of effort trying to argue for a different, phenomenological approach to the sciences of cognition – the very sciences at the root of the study of human-computer interaction. I find myself turning back to that train of thought in light of recent discussions I’ve had around establishing a methodology for user experience design.

One thing that American philosopher Richard Rorty really liked about his student Robert Brandom‘s book Making It Explicit was that the word “experience” was not to be found anywhere in the index. Rorty was one of the most important philosophers of language of his time – as Brandom is today – and their pragmatist approach is extremely influential in contemporary philosophical circles.

Rorty’s praise – which Brandom would no doubt have appreciated – gives you some idea of how far contemporary trends in interface design, which regard the design task as enhancing or enabling certain sorts of user experience, are from the mainstream philosophical conceptions of what such users are and what they are doing when they engage with texts and symbols everywhere, including online.

Balancing Intuition and Evidence

Perhaps like all forms of design, in practice user experience design rarely resembles the execution of a method, so much as it resembles the practice of an art. There is a heavy reliance on intuition, and when a designer does choose to refer to some piece of shared knowledge, that knowledge usually takes the form of a pattern (in the architectural sense) rather than empirical studies or a unified theoretical framework. (There’s no problem with stealing from other intellectual traditions – as Vicky Teinaki suggests – while theoretical frameworks are regarded simply as large pattern repositories.)

The famous Sidney Harris cartoon - must we be more explicit?

Still, it’s not uncommon for designers to worry that such a heavy dependency on intuition is problematic, or at least risky. As the profession matures, and businesses establish a more critical engagement with agencies offering UX design services, clients are becoming more confident in demanding details of the design process itself, in the hope of ensuring that an agency is designing for that client’s customers and not simply to satisfy the tastes of its own designers.

In this way, there is a growing sense that user experience design should be – not merely intuitive – but evidence-based. The emphasis on user testing in all its forms is a manifestation of this.

Partially because the way we test our designs resembles a traditional psychological or ethnographic study, it’s often assumed that the kind of evidence required is statistical evidence. I want to suggest that this may actually be a misunderstanding – itself a result of a naturalistic bias inherent in our society – and to suggest an alternative view.

Finding a methodology that is true to practice

This tension between the design process itself and the requirement that its products have statistical validity is felt most keenly when trying to articulate a UX methodology, for example, in the context of a pitch, or to formalise the design process. It quickly becomes clear that any method that lives up to this requirement is one that no UX project could ever actually live up to.

On the one hand, no design firm has the time or the resources to conduct a large-scale psychological or ethnographic study of the users whose experience they are purporting to represent. In a world in which quantitative analysis is king: the rapid, qualitative psychological studies that are standard among designers appear as nothing more than arbitrary, and certainly insufficiently large, sample populations.

On the other hand, there is the hunch that this lack of statistical rigor may well be a “good thing”. Incomplete data leaves room for designers; for the speculative leaps that are what make design feel like design. And anyway, the alternative threatens to make the process into something akin to “design by committee” – where the committee is a population in the hundreds… or hundreds of thousands!

Proving the pudding

At this point, it’s important to distinguish between imposing the requirement of statistical validity on the design process, and imposing it on the products of that process, i.e. on the design itself. I’m not trying to argue that user experience design should be immune to criticism, nor that the only criticism a user experience designer should listen to is that of another designer. The ultimate test of any design is how it actually performs in the wild, in front of those hundreds of thousands, and every effort should be made to identify design flaws in advance of a product’s release by exposing it to tens – or hundreds – of potential users in a controlled setting. (Of course, the test setting needs to controlled, not to resemble the target audience, which is “in the wild”, but to minimise the assumptions built into the inferences derived from the test results; to maximise their scalability.)

Eye tracking - tells you what is wrong, not right

Let’s spend another moment on user testing, because it will help clarify the kind of validity that I think is appropriate for user experience design. The thing about user testing – which as I’ve said, should be grounded in statistical evidence – is that it can tell you that a component of a design is broken, but not what is broken about it.

Take the case of eye-tracking. An eye-tracking study can reveal that the current design inhibits the completion of a particular step or objective. It does so by illuminating that a statistically significant section of the population tested took an inordinate amount of time to complete that step or objective (where inordinate might mean “above average” or “above a certain threshold”).

But the crucial thing is to realise that this is all it reveals. In order to diagnose the problem, a different sort of mode of interrogation is required. It’s necessary to watch one particular case, or a few, and develop a sense of what is going on.

It’s tempting to say that this sense is an “interpretation” of the data, but that word is loaded up with all sorts of connotations, and these get in the way of capturing the richness of the experience of watching another human being interact with something and ultimately understanding the nature of that interaction.

Diagnosing a usability problem is an achievement

An interpretation is commonly parsed as a point of view on something. Moreover, all points of view are necessarily subjective, so an interpretation is just my point of view – it has no objective validity. What that misses is that while this point of view I am offering is mine now, it’s something I’ve achieved, namely by watching another human being whose behaviour I didn’t predict or initially understand. In diagnosing a problem with our design on the basis of watching someone else interact with it, I take my best stab at adopting the point of view of the person I’m watching.

Now, it may be that I fail to adopt their point of view entirely, and I may even be wrong in my diagnosis of the problem, but clearly, it isn’t just my point of view that’s involved here. At the very least, the diagnostic point of view I come to adopt has its origins (and so, some of its validity) in the behaviour of the test subject.

What’s more, there’s obviously a skill involved here – an ability to effectively, and reliably, adopt other points of view, other ways of engaging with an interface. This is the skill that I suggest is at the heart of all user experience design.

But a UX designer has to do more than this. After all, our brief is to design for all users in one fell swoop. Thus, to develop a unified design – a design that works – a UX designer has to discover that configuration of design elements that has a relatively stable meaning across the diverse range of potential modes of engagement that can be adopted by users. User testing, therefore, keeps a user experience designer honest. Testing exposes a design to an actual diversity of actual modes of engagement, and the point is to ensure the designer hasn’t become parochial or staid in their approach, favouring one or a few modes over all others.

Psychology with a sample space of one

Edmund Husserl - 1859-1938.

The picture I have painted above of user experience design is remarkably similar to that of phenomenological psychology, as championed by the early twentieth-century German philosopher, Edmund Husserl. As I have tried to do above with regard to design, Husserl counterposed his phenomenological understanding of psychology as an “eidetic” science, or science of essences, against an empirical psychology grounded in statistical data. An essence for Husserl is an invariant structure in our engagement with things. It is, if you like, the flip-side of understanding how a certain perceptual structure might be misunderstood. Once you have exhausted the ways something can be misunderstood, or misperceived, what you have left is its essence. The crucial point is that Husserl regarded these essences as a perfectly valid form of evidence on which to base inferences about the world and our behaviour within it. In fact, he regarded it as the only philosophically respectable form of evidence.

That might be going too far, and Husserl was probably also overreaching in jumping straight from endorsing the practice of criticising assumptions by exercising the skill of misperceiving, to asserting the existence of an essence of every perception. But what is clear, however, is this: if what I have said above about user experience design is correct, the methodology of user experience design might share a great deal with that of phenomenology.

An association with phenomenology might go some way toward alleviating that anxiety associated with trying to establish a methodology for UX design. I’ve heard phenomenological psychology described as “psychology with a sample space of one”. Among phenomenologists, that’s not a cause for embarrassment. It’s a source of pride. And that’s because to them it’s a reaffirmation of both the possibility of adopting perspectives other than our own and our responsibility to do so as reflective human beings. What UX designer wouldn’t want to be associated with that?

…

In case you want to learn more about phenomenology:

I can recommend the New World Encyclopedia article on Eidetic Reduction – this is the technical term for the “skill of misperceiving” I describe above – I’d never used this resource before, but it has a better article than Wikipedia.

Also, have a look at Hubert Dreyfuss with Bryan Magee on Husserl, Heidegger and Existentialism (70s BBC show), which gives a nice overview of the tradition in the comfort of a soft, beige couch!

Husserl image via Wikipedia