Personas

Personas: From Theory to Practice
Question: The article mentions that some practitioners go to great lengths to represent their user as accurately and detailed as possible during their design process. Some even create "posters, websites, and real-size cardboards" of their poersona. Do we need to go that far in creating our personas? What is the benefit of going to such lengths?

Discussion: The more detail and effort you go to, the more concrete your persona will become. This will help you better design, especially if you are on a design team.
You could overdo it and waste time, money, and resources that aren't necessary.
What about recording the persona on a video? Either just describe the persona or hire an actor to portray the persona?
The more real it feels to you, the more detailed your design will be. Maybe not a life-size cardboard cutout, but detailed.

Q: The section "What Shapes a Persona?" quotes a designer saying, "Every product we build is a product we build for ourselves to solve our own problems." The figure (Figure 2) shown in this section also shows that, in practice, the designer's ideas play a large role in shaping a persona. To what extent do designers put themselves into a shaped persona?

D: Can you entirely separate yourself from a persona? Maybe not entirely.
It's hard to pull yourself out of it.
You are not the user.

Q: Do you get better Instructional Design when your group has concrete, defined personas, or can the design be just as good, or better, when given more creativity using no personas, (or unspoken personas)?

D: How do you know when you have the right persona? Your user analysis and research should lead you to the right persona.
If you only have one persona, are you leaving out a lot of your audience? Using multiple personas can help. You also have to realize that a business model is ok with leaving a few users behind if they don't fit your personas. This is contrary to what we do in education.

Q: Is there a time when we would choose not to use a persona?

D: Maybe not, but there are certainly times when it is difficult to pin down who your persona should be. When a large, diverse group of people are required to use a system. Mass use of a product creates a side range of users.

Child Personas: Fact or Fiction?
Q: Can a ten-year old child explain how they think to an adult? If not, are personae an adequate substitute? If not, what other methods could be brought into the process in order to improve the project?

D: You can have very good results, but there are some ways to make it more successful. If they have a relationship with the interviewer and are comfortable giving their opionions, you will have better response. If you didn't know the kids already, you would benefit from warming them up to you.
An unbiased, uninvolved interviewer (or at least someone the kids think don't care whether they liked it or not) might help because the kids would be willing to say what they really felt without worrying about offending the creator.
It seems more important to observe kids because they may not have the words to express their thoughts and they may have more surprising interactions with the program.

Q: Do you think we need to be more careful about putting our own views into child-personas than we need to be with adults?

D: I think so. I think if "you are not the user" as an adult, you are REALLY not the user when the user is a child. They are different than you were as a kid. You don't remember things as well as you may think. Designers may not be used to working with children.
I think it is a mistake not to use a child-persona. I think a child-persona will really help and can be created.

The Origin of Personas
Alan Cooper (1999, 2004) "The Inmates are Running the Asylum"

• Personas Goal: "Develop a precise description of our user and what he wishes to accomplish"
• User is a resiurce, but won't know solution!
• "Make up pretend users and design for them"
• Failure = design for borad, abstract groups
• Success = design for a single (archetype) user

Personas: Definition
Description of a specific, fictitious person
-Written in the form of a narrative
-Represents gathered info about a group of users with common characterisitcs (single users are too quirky!)

• Usually given both a name and a face
• May contain personal information (family members, friends, occupations, possessions) make the persona more "real"
• Focuses on the goals, needs, and frustrations of the persona when using the product being designed

-3 to 7 personas usually created for a project (three is probably good for our capstone project)

• Some advocate using on primary persona

Personas: Key Considerations
"Pretend" but not "made up"
-Based on data with users

• interviews - phone or face-to-face interviews can be better than email surveys because people self-correct in email, but when you are talking to them you get them more free-flowing
  
• Observations - watch them use the system
  

Presented as a story about a believable person
-Project team should refer to the persona by name

• Stop talking about abstract "users"!

Focused on enabling effective design decisions
-Should explicitly define the needs, goals, and frustrations of the persona

• Designers should be able to infer what features are needed and how they should be designed

What are personas good for?
Assisting communication
-Easier to talk about "James" and his needs
-User is too abstract -> doesn't drive decisions

Informs design decisions
-What does James need to do with the new system?
-How do you meet James's goals?
-How do you resolve James's frustrations?

Supports design evaluation
-Where will you trip up James?
-Will he know what to do? How to interact with the system?
-Will he even use the system?

Personas: Drawbacks?
Bad personas won't help you

Some consider them too "artsy"

User interviews can be costly
-Recruiting users
-Conducting interviews
-Transcribing protocols
-Time to analyze data, extrat themes
-Some estimates (Forrester Research): $47,000 for commercial apps

Creating Personas
Interview potential users. Take good notes! Or record it.

Identify key observations ("factoids")
-10-12 per interviewee is typical

Sort individuals into groups based on observations

Cluster key observations from multiple interviewees
-Look for patterns/themes
-Typically, 3-4 characteristics from each person are relevant to the group

Interview Data
-Look for common goals
-Look for common frustrations
-Look for common perspectives, approaches

• Technophlie vs. Technophobe?!

Observational Data
-How do users interact with existing technology?
-Do they take shorcuts?
-Frustrations? How quickly do they opt-out?
-Do they know how they use things? (Are they actually doing what they said they do when you interviewed them?)

Persona Template
Name
Role
Daily Tasks/Relevant
Experience
Likes
Dislikes
Goals? Needs?
Frustrations?
What interview questions should we ask?

Usability, HCI, Task Analysis

Usability

1. (Jennifer Hogle) According to Smith and Ragan (2005), the goal of instructional design is to create instruction that is "effective", "efficient", and "appealing". On page 27 of the Leventhal & Barnes chapter, usability is defined as, "the extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use." What comparisons can be made between the usability and instructional design? How can understanding the process of instructional design help us in understanding usability?

*If you have bad ID, there probably wouldn't be good usability.
*If there is good ID, there is probably going to be good usability.
*Usability is a subset of ID. ID is the bigger umbrella. It's possible that in you ID process you would decide not to use computers at all. (Of course, usability can refer to non-computer based tools too.) But usability, although part of the whole ID process in some aspects, is more focused on the final product. It seems like a smaller part of the entire ID.
*Is there ever a time when good ID will violate good usability?
Maybe when it is done on purpose -- making something hard so that the user has to struggle a little to learn more. (Cognitive Load) But this would be intrinsic cognitive load, not extrinsic.

2. (Joanna Gibb) Nielsen & Shackrel are similiar in that their definition of 'usability' is a sytem; is valid or usable if it is found useful. How many systems are usable, easy to learn, but have no use once they have been learned? How useful will the final project be once it's completed? Easy to access only, or worthwhile and applied once it's been presented?

*There are times when it seems a system has been created that really is of no use to the the user. They did not task match. There wasn't the need for it. They didn't use participatory design.

HCI

1. (Kevin Dolan) The study and research of HCI is evident back to the 1980's - even as a multidisciplinary science and investment topic. Can we determine that there is measurable progress in the evolution of HCI over the years; i.e., is the software and computer development world improving the effort to connect humans and computers or are we just getting better at paying lip service to this area of interaction?

*Companies are having to pay attention to what customers want and need and want. If they don't pay attention to HCI they don't sell as much and people will go to another product.
*Web 2.0 has made user wants and needs more obvious. People are "talking" online about their opinions, wants, needs, etc. and they can't be as easily ignored or unknown.
*It seems that education is a niche that gets ignored. HCI principles aren't used as well because of limited funding and because they don't feel like they have to. Things are also designed top-down and then mandated to the users. They aren't driven by marketing, but the irony is they are in fact wasting money if they make a system that people will really use.

2. (W Scott Slade) If there is too much fragmentation, too many theories, too many methods, too many application domains and too many systems, is simplification possible?

3. (Shannon Ririe) Is there/should there be a committee or organization that defines what HCI is or isn't?

*Remember: You are not the user! You can't ever think like a novice again. It is really important to involve your user in the design process.

Tasks

Usability for Ed Tech
•Not just whether system is used (cf., Eason)
•System Functions should include
–Outcome of use(!)
•Task Characteristics should include
–Desirable difficulty
•Task match will need to include relevance to mental models/cognitive processes

Usability Engineering
•Works best early in design/development
•Good interface won’t solve all your problems
•HCI Usability Engineering builds on foundation of:
–Task
–User
–Context of use

Developing Useful Systems
•Slightly more than 30%of the code developed in application software development ever gets used as intended
•Likely because developers do not understanding what users need

Maybe also because the basic tasks were made usable and the rest was not

Maybe because the designer got excited after the basic user needs were designed and added a whole bunch of other stuff

Maybe the extra/advanced stuff too difficult to find users to test

The questions is do you want it this way so that as users become more proficient they can move on? or does the extra stuff get in the way?

Useful Systems Support Tasks
•Task ≠ User can use the system
•Tasks:
–Specific
–Observable
•Will you know if you’ve successfully accomplished it?
–Reflects end-goal of a user session
–Relate to key aspects of system components
•What do you think are the key parts of the system?
•What will be frequently used?

–Hugh Beyer and Karen Holtzblatt, "Contextual Design: A Customer-Centric Approach to Systems Design," ACM Interactions, Sep+Oct, 1997, iv.5, p. 62.

What is a Task Analysis?
•Analysis of how a task is performed
–Detailed description of behaviors in interface
•Highly detailed
•Step-by-Step
•Procedural (ignore mental processes for now)

How is Task Analysis Useful?
•Specify problems/gaps in process
•Highlight unnecessary or inconsistent steps
•Specifies procedural aspects of key tasks
•Requires concrete analysis of user actions

For next week:
•DUE:HCI Exercise: Task & Task Analysis Exercise - Something in education - Keyboarding for kids? - goal is to login and practice a lesson
•Read:2 articles on personas
•Post:2 questions per article on WebCTdiscussion area
–Due by 12 noon on day of class

First Day of Class

Class Objectives
Define human-computer interaction
Define usability and its relation to HCI
Identify important considerations for education technology

What is Human-Computer Interaction?
What you're doing with the system
Your experience with how those interactions are going
How you are going to accomplish your goals with the software
Depends on your user goals
Visual appeal, layout, design, perceptual aspects
Bulk and speed of the program
Design so it runs the way we want it to so it is functional and quick enough
The use of the user's time
Does the interface match what the user is used to and wants and can utilize?
Usability

What is usability? Is it the same thing as HCI?
Usability is only part of the entire picture of HCI
It's how we access a tech tool in the first place so we can interact with it at all
Our standards of usability are rising. We no longer are willing to use things that aren't user friendly.

Educational Technology: Do our definitions need revising? (If so, how?)
Your target audience might be a lot more specific (ie. are you trying to reach teachers, students, what age or level?)
Motivation and attention-getting becomes more necessary
Training people to use the technology - how quickly can they learn and be trained?

Formal Definition of HCI:
A discipline concerned with the design, evaluation, and implementation of interactive computing systems for human use and with the study of the major phenomena surrounding them

HCI vs. CHI
Some think we should put the Human first, not the computer, but sometimes you'll still see CHI. :)

Usability:
Usability is typically the goal human-computer interaction methods
Technology is usable if:

• appropriate for the target users
• allows users to accomplish their goals

Usability ≠ user friendly

HCI & Educational Technology
•Usability not necessarily end-goal
–Goals may be educator’s (not student’s)
–Learning processes?
–Learning outcomes?
•Learner Characteristics Complicate Usability
–Prior Knowledge
–Personalized technology

Desirable Difficulty
–Desirable Difficulty
•Learning can be triggered by impasses (VanLehn1991; 1995)
–The “Assistance Dilemma” (Koedinger& Aleven, 2007)
•Key problem for interactive educational technology.
•When to let students struggle, when to provide support
•How much struggle is productive may be personal

Prior Knowledge
•Expertise Reversal (Kalyugaet al., 2003)
–Experts: Visual representations
–Novices: Need plenty of textual instruction
•Expert knowledge structures (e.g., Chi et al, 1981)
–Experts have conceptual knowledge organization
–Integrate incoming information
•Self-Regulated Learners
–Hypermedia requires self-regulation (e.g., Azevedoet al, 2004)

Personalization
•Increases HCI & usability demands
•Intelligent tutors
–immediate feedback
–custom selection of content
–customized hints and help messages
•Automated knowledge analysis
–recommended materials
–provide customized prompts

HCI & Ed Tech: What Doesn't Change?
•Importnce of
–task
–users
–scenarios of use
•Useful methods for analysis
–cognitive walkthroughs
–heuristic analyses
–learner interviews & tests

For Next Week:
•2 chapters (one is short)
•Post questions on WebCT discussion area
–Due by 12 noon on day of class