Interaction Design - Science topic

Byungwook Choi Careful here. If you hypothesized an interaction effect, you should keep the interaction term in the analyses whether statistically significant or not. Changing the model at this point is referred could be construed as HARKing (hypothesizing after the results are known) and is generally considered an unethical research practice.

Remember, nonsignificant results are important too!

The term "Graphic Design" is general and refers to visual communication applied to analog and / or digital media.

This has certain reminiscences of print, therefore, in the field of interaction design, the term "Visual Design" is often used.

Does it have to be a software app? An alternative would be to do this within HTML. It would not be too challenging for your student to create a basic HTML page using javascript to accomplish what you are looking for. Furthermore, if you are at a University you could probably have someone in IT or a student set this up on a University server.

I do not understand what do you mean by "Hepatology students"? Are these graduate students doing post-graduate training in Hepatology? Why Hepatology students in particular? what is your control group? Are these questions part of a questionnaire? What is the aim of your research? How did you develop these questions? What is the problem, and what are your research questions?

Hepatology is a straight forward subject and is much easier compared to cardiology, or neurology. So you need a control group to compare.

I can answer from an educator's perspective. As an educator who designs and implements educational technologies (Ed Tech) in higher education, I think interaction design is an important part of whether a particular technology will see success (or not) in the classroom. From that point of view, we need user-friendly software which will enable all users (non-programmers) implement successful interaction design. If anyone, anywhere can make good interaction design, the field of interaction design will be able to collect lots of data from different parts of the world, perform analytics to determine what the good and bad parts of interaction design are, I think the entire field of interaction design will be able to evolve faster. This is a challenge and an opportunity at the same time.

you can read this research I hope it will be useful for you and good luck

The reference David Eugene Booth gives has a section (6–3.6) titled, "The Assumption of No Interaction in a Two-Factor Model." (There are a couple of other places they speak on interactions, e.g., pp. 229).

In a 'round-about way Montgomery is stating that fitting a no-interaction model is fine as long as the residuals do not suggest that interaction terms are needed for the model to conform to assumptions. E.g., if a residual vs. fitted plot gives you funky patterns, you might need to consider an interaction term (or transformation).

My assumption is that this could extend to lower- vs. higher-interaction terms and I suggest: study the residuals. However, it may be just as efficient and, perhaps, better to first think about the context and previous work of the system under study. I.e., Does this k-way interaction make sense? Is there enough data to support such parameter estimation? What evidence is there from previous studies? Etc.

Interpretation beyond a two-way interaction becomes difficult and pretty much impossible for any humans I know in the 4th plus.

Need help

If I am employing moderation in secondary time series analysis can I need to proceed with the methodology of time series?

I need some liturature regarding its methodology

Actually my all variables (quantitative) are of mixed order can I still procedd with ARDL approach irrespective of moderation analysis?

If someone guide me / suggest some relavent stuff I shall be very grateful for this favor.

we knew the CR and AVE for the measurement model ( Dependent and independent variables)only

Brian R Johnson has bring a very broad view to the subject. I agree with the definition of technology, which is changing very fast. Before all these cool stuff, the interactivity of the space was going hand in hand with the adaptive space, which can respond to the changing needs of the user. So, it will be different from now on, also, regarding the discussion of AI. It is useful to limit the scope of the discussion when trying to delve deep into something.

If you have continuous predictors then generally statistical power to detect interactions is very low so you'll struggle to detect interactions with 220 observations.

This is because statistical power for the interaction depends on the co-occurence of extreme values of both var 1 and var 2. If most observations are average - as say if data are approximately normal - on var 1 and var 2 these values (low-low, low-high, high-low and high-high) are are rare and statistical power is consequently poor.

The best thing is to take a 3D Plot as an output and study the pattern. Things will be much more clear since it is the influence of the two variables which are varied to look at their influence on the response.

nice thinking

Dear Mi Sc ,

Simply put, serendipity is when the connection between things, leads to something positive.. So if serendipity is simply a series of positive connections that can be engineered, then serendipity can be calculated. In fact, I would go as far as saying that serendipity is an algorithm.

Regards,

Shafagat

Dear Ms. Janki Dodiya,

There are a few usability metrics methods that can be used to help map not only users' mental model, but also to understand the relation between users' expectations and affordances. I often like to use Think-aloud Protocol for digital experiences and Task-Flow observation for physical experiences.

Hope these are useful. Let me know if you would like to discuss further about the subject.

All the best,

Adriano Renzi

Tehrim Khaliq Indeed , smell based impairment is very rare work. I mean with the accessibility and web accessibility terms how to facilitate interaction and knowledge acquiring tasks for different types of the disabilities.

I have worked in this track for time, If you're interested please refer to my profiel and feel free to contact me.

Regards

Samaa

I believe your approach has been quite appropriate and extensive, considering how this is usually done.

The ultimate evidence of the effectiveness of guidelines is their successful application by practitioners. However this application is often ill defined. Who do you envision to be using your guidelines, and how do they go about it in practice. If you know this, you can set up an experiment simulating this practice and check whether your guidelines lead to an improvement. This is a lot easier if the intended use is heuristic evaluation than if the intended use is design. So many things influence design that is merely impossible to establish the positive impact of a certain set of design guidelines.

An alternative, as you suggest, is to compare your guidelines to generic criteria for guidelines such as understandability, comprehensiveness etc. I would say guidelines need to be decisive (thus help to distinguish good from bad alternatives), specific (it should be very clear what they apply to and what not), practical (it should be easy to apply the guidelines) and robust (they should be based on the best evidence available). These are properties that can be evaluated by experts from practice or academia, and thus form the basis for a heuristic evaluation of your heuristics.

For each of these approaches, I think the most important thing is to have a clear idea of your users of the guidelines whether they are designers or evaluators and a model of their practices.

have you seen this article?

Regards.

R & D is the process by which a company works to obtain new knowledge that it might use to create new technology, products, services, or systems that it will either use or sell. The goal most often is to add to the company’s bottom line (https://www.shopify.com/encyclopedia/research-and-development-r-d). The end product of R & D is innovation. It is not just taking surveys and interviewing customers, etc. It is a process that always seeks to improve upon the existing technology, products and services, to excite customers. R & D in IT has to do with coming out with new ways of resolving IT issues in organizations through the improvements in softwares, etc. Surveys, etc. may serve as some of the tools that can be employed to achieve it. It is in two folds: RESEARCH and DEVELOPMENT. Without the development part, it is not complete. After the research, the development follows, and it is the development that produces the new technology, products and services, and new ways of doing things.

My dear friend,

My new book has published. It is about design

as you are in this field i hope you will like it

Please read it and share it with everyone. It is talking about prosumer also for the first time in this book we talk about magic and its power in product design. It is talking also about future of consumers .I request you put the link of book on your page and your school website for your students.

Name: Everyone Is a Designer

Author: Mohsen Jaafarnia

Publisher: MJ

Ghochan, 2017

In Persian, Chinese and English

Topic: Industrial Design

Jaafarnia, Mohsen (2017). Everyone Is Designer. Ghochan, Iran : MJ Publication. ISBN: 978-600-04-7870-4

In my experience the most accurate technology is UWB. RSSI based systems are not so accurate.

The problem you try to solve is unsolvable. There is a rigorous theorem known as Earnshaw's theorem saying that you can't make a permanent magnet float in a stationary magnetic field. To do what you want to do you have to play a few tricks. Either you create a dynamic feedback mechanism which changes the magnetic field in time, or you try to make a diamagnet float (that works) or you make the permanent magnet spin (there is a toy that you can purchase that works this way.) Again a permanent magnet in a stationary magnetic field can't work for fundamental reasons.

Prof. Ian, exactly that what I meant

I don't understand your question. My understanding of the use of covariates is to remove nuisance variance that you cannot control experimentally. For example, if you're studying assertiveness in negotiations and you know that men and women differ in self-reported assertiveness (Cohen's d in one meta-analysis was > 1.0). You don't want to use only men or only women and you don't care to study this phenomenon (by making it a factor), so you use sex as a covariate.

I would generally urge researchers to avoid covariates unless they are necessary because any covariance between the covariate(s) and the outcome will decrease the effect size. Standard treatments of ANCOVA/MANCOVA simply tell practitioners to pick uncorrelated covariates, but the reality of that most variables are correlated in social sciences.

But I do have this answer about the 11 Likert items.  If they form a scale (i.e., they collectively indicate a latent variable) and you're interested in the latent variable measured by the scale, you're much better off using the scale score. The only advantage of dummy-coding each item would be if you were interested in the specific effects of each item.  For example, maybe the 11 items are each different symptoms of depression and you wanted to know the relationship of each symptom to the treatment.  Then it might make sense to do as you suggest. I think it might be just as easy to correlate each item with the outcome.

Above I said "it might make sense..." because if your items do collectively indicate a latent variable, you are almost always better off using the scale score. In order to find an effect for one item, you would have to find that the item measured something unrelated to the latent variable (i.e., that the scale was constructed in a shoddy manner), or that one or more items were biased. Biased items are actually fairly rare. A common example is about the frequency of crying on depression scales. Men and women of the same level of depression have different probabilities of admitting to crying (at least in Western cultures). Thus such items are biased. But a common finding of item bias research is that most items are not biased. I just helped a student examine the MCMI-2 for racial bias and she found just two items with both practical and statistical significance.

BTW, I see zero advantage to a median split unless it's just a convenience (after the analysis) to make graphing the relationship easier. I don't have the statistical equations handy, but it's well known that dichotomization reduces true-score variability of a measure without producing much benefit (e.g., Hunter's meta-analysis text treats this extensively).

I agree with the above but the interaction has to be inviting and user friendly like wise for the instructional system if you do not have a user interface the instruction become boring and people loose interest. The average attention span of a person is 15 minutes, and you only have the first 5 minutes of interest, that is how long it will take a regular person to process and decide

If you thinking for a marker-based approach, one possible option is the Vuforia library and use it with Unity 3D. This facilitates maker-based AR and you have the option of creating your own maker. With regard to the content, you can import custom pre-defined objects and have it fused on top of the marker. If you are planning to display dynamic content, then I presume you should have an API to access dynamic content and assign the pose of the marker to the content at frame rate.. 

Here are some examples for Vuforia with Unity 3D. 

https://youtu.be/TljI56icYjM 

https://youtu.be/tA6p1GDdryo 

Check out the book, Measuring the User Experience (Second Edition), by William Albert and Thomas Tullis (ISBN: 978-0-12-415781-1 - http://measuringuserexperience.com/). It may shed some light on your question. You might also read, Time Perception during Retrospective and Prospective Paradigms with a Distraction, a Master's Thesis by Kristi L. Lontz.  

From my research for my dissertation into using online 3D simulations for learning, I have read many studies that show a correlation between how deeply immersed into the simulation a learner becomes and how quickly time passes according to the learner's perception. This, according to many of the authors, is an indicator for how good a simulation is. The belief being a learner would not "lose track of time" and spend hours in the simulation if it sucked.

You might also be interested in Scott Brown's paper, Time perception and attention: The effects of prospective versus retrospective paradigms and task demands on perceived duration (DOI: 10.3758/BF03198848). This study looked at two groups - one which kept track of their time and one which did not keep track of their time.

Hi Christian, 

I guess the most complete recent work you can find on the issue is the project of Matthias Schröder, which you can find here:

There you have PDFs and videos of his PhD research.

I  agree with Mr. Falgueras too. The reason Interaction Design rarely appears in other context than computer-related technology might be that ID of non-digital artifacts is considered part of product design/industrial design even though it's ID as well. Don Norman's "The Design of Everyday Things" should be mentioned here. Apart from that, ID is not interchangeable with HCI due to their different point of views.

Thanks Stephan! I will check it out and let you know if I have any questions. 

All the best,

Raphael

Michael,

You don't describe how you assess outcomes/efforts in stage 2 such as "feeling pressured", "becoming more aggressive near the end of the game". I am working on HCI using the humanities -- maybe you can apply my ideas in your research.

You could measure behavior using means of the humanities defined in my research as modeling dimensions. For example, dimension 'space' measuring physical proximity between persons could influence behavioral aspects in games and be implemented using infrastructure of RoboCup (Simulation League). See section 3.5 "Summary of significant modeling dimensions", section 6.4 "Implementations of agent-specific modeling dimensions", section 6.5 "Implementations of simulation-wide modeling dimensions" for detailed information (see below).

Cameron Tonkinwise has recently published an article reviewing Dunne & Raby's Speculative Everything book. I think you may find his criticisms on their work very valuable:

Tonkinwise, C. "How We Intend to Future: Review of Anthony Dunne and Fiona Raby, Speculative Everything: Design, Fiction, and Social Dreaming." Design Philosophy Papers 12.2 (2014): 169-187.

There are some interesting discussions in the pattern community between the difference of pattern languages and project languages. The latter ones are the part of the pattern language that works together for a specific project (as also initially was intended by Christopher Alexander). Also their implementations and the way they work together are specific for this one project.

One paper touching this topic is published in the PLoP'13 proceedings: http://www.hillside.net/plop/2013/papers/proceedings/papers/motohashi.pdf

Hi Jérôme!

I think that research on action or "dynamic" games for people with motor disabilities is very interesting, specially if it tries to avoid the usage of additional expensive hardware.

In fact my research group and I are working on the development of GNomon, a framework that enables the creation of dynamic and engaging one-switch video games for children with severe motor disabilities. The framework supports the selection of multiple elements, which can freely move around the screen, with just a single switch. It is based on a custom implementation of the NOMON interaction modality. I attach the work-in-progress on GNomon and our ongoing research.

Have a nice day and good luck!

Hi Helen,

great problem to solve. The critical questions to ask are quite multidisciplinary ones and not restricted to one subject area. Most likely, flexible and open approaches are needed, e.g. see our short book on people with complex conditions  (the general issues tackled in the book are much broader than the schooling question): "Educating children with complex conditions" or the work with clinical populations. To achieve inclusion with using software technology cannot only be a matter of word size or screen resolution but rather of soft factors to be defined in a SMART way.  The applied aspect to use non-material resources in a way to help improving inclusion I like in particular.

From an applied point of view, to consider the cultural perspective seems crucial, as, studying politeness in two cultures, we found telling differences and similarities when analysing forms of address, e.g. initiating the first contact between strangers might well have some likeness to situations when getting immersed in computerised communication scenarios.

Journal of Pragmatics, 43 (15). pp. 3807-3821.

If of interest, let me know. Sounds a great project for the elderly.

Kind regards,

Winand.

By mid 2015 there will be a special flights based data gathering for virtual reality examination / simulation around bridge design.

Riccardo,

A master student of Eindhoven University of Technology, Guus Baggermans, did a work on vending machines a few years ago. The reflection was about how the design of such machines actually do not care for the user. Therefore, though making and reflection upon making, he proposed a 'Friendly vending machine'. Most of his work o this project can be found at http://www.guusbaggermans.nl/friendlyvending/ .

If you need more input, please contact me.

Regards

Pierre

It might be an idea to make this question more precise: most important health technology at what time and for what population. Sanitation was historically most important (see cholera epidemic in London and John Snow) but soap is a great technology against all kinds of skin diseases. In some developing countries these technologies still play a role, as does adequate nourishment including before birth. As for the most important technology to improve health in let's say the G8 countries , it might be techniques and policies to increase the level of exercice and to decrease the BMI. 

GIVEN: A + B + AB + AD + ABD

Let's put these terms in the conventional form: A = X1, B = X2; D = X3; and D is dummy variable. The general regression model is given by:

(1)   Y = B₀ + B₁X₁ + B₂X₂ + ... + B_mX_m + e

In the present case, there are two variables (X₁ and X₂) which are continuous. The model for these continuous variables may be represented, thus:

(2)   Y = B₀ + B₁X₁ + B₂X₂ + e

What to do with D = dummy variable? Can the dummy be combined with continuous (quantitative variable)? Generally, dummy variable is considered "nominal" and nominal data is used only for identification and may not be subjected to mathematical operation, such as addition, subtraction, multiplication, and division. In this case the D variable (possibly comes from the demographic information section of the survey) Should be used only for that purpose: identification. If one wants to know the effect of gender then test it separately by bifurcating the sample into its dichotomous categories, i.e. male and female. Under this approach, equation (2) becomes:

(3)   Y_male = B₀ + B₁X₁ + B₂X₂ + e

(4)Yfemale = B₀ + B₁X₁ + B₂X₂ + e

As for probability treatment for male and female, one needs to defined a category of interest, i.e. MALE, then all male will be score 1 and female = 0. From there fin p and q. This binomial distribution takes us away from the subject of interaction effect measurement.

Compare the two equation's plots as suggested in DUMMY VARIABLE below. Thus, you will be working with the form of equation (2); active variables are quantitative and continuous. The dummy variable is used for intra-group comparison and testing the significance of the difference among sub-group in the sample.

LATENT MODERATED STRUCTURE (LMS) MODELING: Now from equation (2), we may measure the interaction among X₁ and X₂, thus:

(5)   Y = B₀ + B₁X₁ + B₂X₂ + e

The test statistics is given by:

(6)   t = (B₁ + B₂) / sqrt (A/B)

(6.1)   A = n₁SE₁² + n₂SE₂²  ; and

(6.2)   B = n₁ + n₂ - 2

A more robust modeling is discussed in the Ozer article (attached below), the model is given as:

(7)   Y = B₀ + B₁X₁ + B₂X₂ + B₃(X₁ - X^₁)(X₂ - X^₂) + B₄X₁² + B₅X₂² + e

See attached article by Moosbrugger.

DUMMY VARIABLE: In Latent Moderated Structural (LMS) modeling, the dummy variable, i.e. male or female, for instance, brings in the concept of "ordinal interaction." Suppose we separate the same into two categories classified its dummy (sex) variable, we would have an estimated function for male and another estimated function for female. If we plot both function we can see the two lines. Visually and mathematically, we can see and check if the two lines are perfectly parallel or will cross one another. if the two lines are perfectly parallel, there is no interaction among the dummy variable. if the line crosses, there is an interaction between the sexes. We call it "gender effect." if the dummy is age then "age effect;" if the dummy is education then "education effect," etc. The answer to the question of whether to include B*D depends on above explanation on how to treat dummy variable and whether you are seeking the dummy effect in the model. Here, if D is included, it will not be D, but Pr(D) because D is a dummy variable which needs to be quantified first. One solution may be just pre-treat the dummy variable first and then proceed to equation (2). if one wants to compare the effect on the basis of dummy variable then compare the two groups, as in (3) and (4) above.

REFERENCES: Attached are some articles on effect measurement modeling. A link to LISREL9.0 student version---a software for effect modeling is also provided below.

I wrote a handful of papers (and a Ph.D.) about the design of digitally mediated public services and how you can empower citizens through by improving the design of the digital interfaces that mediate such services. So yes, if the above few lines match your definition of public services, I would say it can. I'm assuming you mean public services as opposed to "survives".

At PhD level and PG level already some universities introduced Research Methodology  in their curriculum ,try  to adopt  it .Sop at the beginning level their thinking about research  certainly developed. Also allow the interaction of  GG & PG students while presentationof seminars and projects.

Dear,

I believe that the use of appropriate and decorative objects and colors by themselves enhance any ambience.

However, I truly believe that the use of  Feng-Shui  improves the "MOVEMENT OF ENERGY". In Brazil this concept is gaining importance.

Best Regards,

Vanessa

I think a central notion of Value-Centered Design is User Experience.  Values are extracted from UX by wondering about the impacts of the tested product on the user's routine. (Was it worth using? How do you feel regarding others when using it? ...) I like the approach of Camara (link attached) on this question.

Hi Teisanu,

maybe check the proceedings of this year's CHI Play conference?

Cheers

Some more facts:

1. You can use pupil diameter as a measurement for cognitive load (task difficulty).

2. Some folks use pupil diameter as a measurement for emotions while solving a task.

(But, first of all it would be better to think about a research question, then getting the design and after that the data.)

Search for Matthias Rauterberg...

Hello, you could look into psychological studies that investigate the role of metaphors in group development. One paper I am aware of is:

Srivastva & Barret (1988) The Transforming Nature of Metaphors in Group Development; Human Relations, 41(1), p. 31-63.

It is not much but you can use it as a starting point. When you google the title you may find more papers that address this issue. You may also get a few ideas for relevant keywords you could use for your research.

Another issue may be a bit of a long shot, but you could also try papers that deal with the impact metaphors have on organizations. In the end a design team is a part of an organization and the metaphors they use do not stay within the group. They shape the in-group communication as well as the communication of the group with the outside. Here is one paper I know:

Hamburger & Itzhayek (1998) Metaphors and Organizational Conflict; Social Behavior and Personality, 26(4), p. 383-398

Hope this helps, Jan

https://www.youtube.com/watch?v=T4xRPyId_DU, openCV

hope this helps

Modules from picture seem to be modules from Hoperf Electronic, RFM01 (RX) and RFM02 (TX).

Look at these links and compare them:

They have datasheets there with application examples for connection with uCont.

I think Chalkup can help you do this really well!

https://www.chalkup.co/tour/messaging

You can share files, write in LaTeX. In addition to messaging there are also discussion features for brainstorming or having Q/A.

I would be happy to give you a tour and show you how to use it for professor-student communication and live office hours.. etc. (also it's free!)

Send me a message!

My colleague Nadia Berthouze from UCLIC, University College London might be the right person to contact... http://www.ucl.ac.uk/uclic/people/n-berthouze

I guess its patented by samsung, their own creation. Thats all i can say as of now.

Hi Nathalie. In fact the same that as been happening to other sectors might be necessary to retail... mass customization (of the the experience). The customers still have the need at least in some products to go to the store but the way they might use it, the store, may come to be very diferent in the future.

Our paper on metrics for arm fatigue was accepted for publication at CHI '14. You can find more details here: http://hci.cs.umanitoba.ca/projects-and-research/details/ce

What about work with dogs from the Max Planck Institute for Evolutionary Anthropology? Or consider variables used with human infants- i.e, approach, gazing, vocalizing etc.

Dear Qurat ul ain Nazir, You are most welcomed. I will try to trace some more publications but may be later. Meanwhile, try to search using any search engine. Good Luck!

IIir - interesting take on football efficiency

Hi Nasim, I used to work at the German Naval Medical Institute some 15 years ago - so my view is a bit dated, and most of my work was classified. However, in terms of environmental factors, you always have (unsorted, incomplete list, numbers just for reference):

1) lighting conditions. Whenever people have to go outside, you want their vision to be dark-adapted. This has various implications for lighting conditions and screen color sets.

2) sleep deprivation. In all shift systems you have to deal with fatigued personnel; in addition you may have extreme workload during port phases.

3) boredom and monotony.

4) Socio-cultural factors, such as command hierarchies, group dynamics, a distinct group culture growing over time that may have very different values from those people might have at home.

5) interfering personal equipment, like gloves, cold protection, other protective clothing (e.g. leather in some engine rooms) etc.

6) varying climatic conditions - heat, wind (also indoors due to air conditioning), moisture.

7) The influence of motion is overrated, at least in large ships which move slowly enough to be considered stationary. You may have to consider kinetosis (seasickness), though, which has some specific effects. On small vessels, you also may have to consider the ship's rolling; angles of some 30° may not be uncommon.

As for observations, nothing beats going aboard. If you look into questionnaire or psychometric test studies, you'll need a trained interviewer to administer them. A colleague of mine got very useful results for hardware ergonomics by installing a surveillance camera in a control room (which requires everyone's agreement of course).

Hope this helps, ;-) Bernard

The web. There are amazing GUI toolkits for the web now, there's the canvas tag, there's WebGL and Three.js. If you need to interface with non-JavaScript code, you can (a) run a web server that exposes a REST API to the code --- I've wrapped code in python and then exposed it with bottle.py, (b) run a web socket server and deal with asynchronous message passing --- this is similar to (a) but more efficient, (c) use Chrome's proprietary Native Client (NaCL), (d) compile your C/C++ code to JavaScript with emscripten ( http://emscripten.org ), or (e) embed Chrome into your app and expose some functions and variables directly ( https://code.google.com/p/chromiumembedded/ or the commercial http://awesomium.com/ ).

I've done (a), (b), and (c). My current favorite is (b), since it's pretty simple to set up. I haven't tried options (d) or (e) yet.

How about the courtroom, where formal argument is aimed at persuading either a judge or a jury? That setting would seem to be the prototype of a persuasive environment.

I would like to add two additional challenges faced in doing in the wild studies.

First, the logs make it very clear what people are doing, but they cannot tell you why. If possible, it is always great to get a little qualitative information to ground the insights pulled from the logs.

Second, a hypothesis driven approach to analysis is not always best. It is important to be open to discovery. If you are releasing a novel system, and not a redesign of a current system, you need to be open to seeing how people appropriate new technology for uses never imagined during their development. A simple example used by Schön describes how when 3M release Scotch tape as a way to repair books, people began to use it to remove lint. If you are doing a longitudinal study, do you have the ability to see this kind of appropriation?