UX Research Methods Midterm
Empirical Methods
Observing and learning from users as they work with a product before, during, and after the design and development process - usability/ user testing - field testing - A/B testing
Methods of Card Sorting
Open Card Sorting: Cards showing content with no pre-established groups. (good for new products or when you need to redesign) Closed Card Sorting: Cards showing site content with an established initial set of primary groups. (good for when you are adding new content to existing structure) Hybrid card sorting: predefined categories but also allows for opportunities to create new categories
Focus Group
Posing questions to a group to find out what they think, feel, do, know, believe and expect - up to 12 user - typically structured discussion - range from 1-2 hours Advantages- controlled environment, attain consensus from multiple people at once, more data for less time & $$ Disadvantages- lack of context, introduction of biases & influences, destructive group dynamics, descriptive data
Interviews (generative research)
Posing questions to an individual or group to find out what they think, feel, do, know, believe and expect. Sessions may be structured, semi-structured, or unstructured. Range from 15min- 1 hr Advantages: Individual thinking about topic without the influence of others, probe deeper, no location barriers Disadvantages: Lack of context, interviewer/interviewee relationship, based on what is asked and what is said (bias/ participant honesty)
Launch & Maintenance (stage 4 of product development lifecycle)
Product ships, and you receive post release feedback through UX evaluations and tech support
Generative Research Methods
- Inquiry (focuses on what people say) - Field studies (focuses on what people do) - Immersion (you become the user) - participatory (focuses on what people make)
Phases of an interview
- Introduction (<5 min) - Warm up - Body of the session - cooling off - wrap up General info --> specific info --> general info
Types of of standardized usability test questionnaires
- Post-Task: contextual assessments at the end of each task or scenario (Single Ease Question (SEQ) most popular) - Post-study: administered at the end of the study (Software usability scale (SUS) is most popular)
Number of scale steps
- more scales steps are better than fewer - *Aim for 7 steps - odd number of steps provides a neutral point - even number of steps forces respondents to either express a positive or negative attitude
Information Architecture
- structure, content, information categorization, navigation - good IA helps users find information and accomplish their tasks with ease
Participatory Design
Actively involves the user in the design process to help ensure that the product designed meets their needs and is usable - Collaging (to solicit memories, communicate experiences through pictures and words) - Card Sorting (to see how people organize information) - Velcro Modeling (lo-fidelity 3-D constructs that participants build) Focuses on what people make!
Standardized usability questionnaire
-Designed to assess participants' satisfaction with the perceived usability of products during or immediately after usability testing - Designed for repeated used - development of standardized questionnaires requires measurements of its reliability, validity, and sensitivity
Working with human subjects
-UXPA code of conduct - Make sure participants are comfortable - Respect privacy, confidentiality and anonymity
Use Cases
-User perspective of product behavior as it responds to product requests - Sequence of steps to complete the task including user goals - Explain how the product should behave
Preparing to conduct an interview
1) Identify objectives of the study 2) select type of interview (structured, semi-structured, unstructured) 3) Data analysis type 4) Write the questions 5) Test the questions
Field Study process
1) Introduction- build rapport & trust, sign consent form, get familiar with environment, give participant chance to vent any frustrations with product or job 2) Inquiry- ask for demonstrations not explanations, observe social and physical environment, record tools used, sequences in which actions occur, methods of organization, types of interactions 3) Wrap up- ask questions that were saved that would have interrupted work flow, summarize insights to clarify & correct, schedule follow up if needed 4) Data analysis- Affinity diagram
What makes up a good user experience?
1) Utility (has value) 2) Usability (easy to use) 3) Desirability (look and feel is pleasing) 4) Positive brand experience
Contextual inquiry
4 parts: -context- work with participants individually, observe participant in action, ask participant to think out loud and ask clarifying questions - partnership- develop rapport with participant, participant becomes the "master" and you are the "apprentice" - interpretation- share your interpretations and verify correctness - focus- learn what's important to participants
User-Centered Design
A framework outlining design and development phases, focused on satisfying user needs, throughout the product life-cycle
Personas
A persona is a user archetype you can use to help guide decisions about product features, navigation, interactions, and even visual design. It's a brief sketch of each type of user--> focus on the major needs and expectations of the most important user groups Based on user research data and should describe real people
Discussion guide
A script for the moderator to follow, providing a consistent framework and a schedule for the inquiry session Tips for writing: - order matters (good flow) - non-directed (no answer implied) - open ended - unambiguous
Evaluative Methods
Assessing the degree which users can use the product, the efficiency of the product, and user satisfaction of the product. Methods: - Inquiry - Inspection (based on expert judgement) - Empirical (user testing)
Card Sort
Card sort is a method that involves writing objects that are in a product on cards and asking users to sort the cards into meaningful groups. Need at least 20 participants and between 30-50 cards
Practices for Developing Personas
Conduct Research: Find out who are your users and why they are using the system? Translate User Research: Look for themes/characteristics that are specific, relevant, and universal to the system and its users. Refine: Combine and prioritize the rough personas. Separate them into categories, you should have 3-5 personas. Make them realistic: Develop the appropriate descriptions of each personas background, motivations, and expectations.
Scenarios
Description of stories and context: - Prototype built with words - Stories describing how the user will use the product - Describes goals and how a user can achieve them with the product
Design & Implementation (stage 2 of product development lifecycle)
Designers and engineers put together rough concepts and designs using evaluation and feedback to make sure they are on track. During this stage you can identify usability problems
Inspection Methods
Diagnostic techniques to decide whether product design elements follow established usability standards and guidelines. - Guidelines and Standards Review - Expert Reviews -Heuristic Reviews
User Experience
Every aspect of the user's interaction with a product, service, or company and how the interaction impacts the user's perceptions of the whole.
Questionnaires
Form design to obtain information from respondents- based on self reported data items in a questionnaire can be open ended or multiple choice/ closed ended measure users' perception of the system and their interaction with it for evaluation purposes- can measure satisfaction, ease of use, engagement, and enjoyment *important to balance positive statements with negative bc people are more likely to agree with a statement than disagree with it
Evaluative Inquiry Methods
Gathering information about user likes, dislikes, needs, and understanding of the product by either talking to them, observing them using the product, or having them answer questions verbally or in written form Methods: - Interviews - Focus groups - Field research - Surveys
Generative Research
Generative research is a more focused on gaining a deeper understanding of user needs and desires. (who, what, where, how and when) *Helps to: - Identify problem or opportunity - inspire new ideas - leads to concept exploration and ideation
User Requirements (stage 1 of product development lifecycle)
Identifying who the user is and what they are going to use the product for- also identifying things your users didn't even know they needed
Generative Inquiry Methods
Inquiry focuses on what people say! Posing questions to an individual or group to find out what they think, feel, know, believe, and expect. Inquiry Methods: - Interviews - Focus Groups - Surveys
Generative Method Comparison
Interviews -Collecting rich data -Sensitive topics Focus Groups -Need data from several participants in a short period -Cover few questions without a lot of depth from each participant -Topics are not sensitive or likely to be influenced by others Surveys -Need large number of responses -Questions can be answered adequately from selection -Cover large geographic region in short period of time
types of rating scales
Likert scales (scale of 1-5) semantic differential scales (uses opposites)
Translation
Merging user insights and interaction principles into practical requirements that inform design
Immersion
Researcher becomes the user adopting their lifestyle and activities for a period of time in order to understand their logic & emotions Focuses on what you do! provides empathy and context
Task Flows
Show how a user travels through the system while performing a specific task
Product Development Lifecycle
Stages are iterative and you should evaluate at every stage Stages: - User Requirements - Design & Implementation - Evaluation - Launch & maintenance
Evaluation (stage 3 of product development lifecycle)
Summative evaluation is empirical (objective & subjective). During this stage there is active involvement of users to evaluate designs
When to use interviews, focus groups, surveys
Surveys are better when need info from large sample of population interviews and focus groups are better for detailed information gathering via probing focus groups are better than interviews when you need multiple points of views in a short period of time focus groups are better for brainstorming interviews and surveys are better than focus groups when you want to cover more topics without influence
Task Analysis
Systematic method for identifying interactions with a product, analysis of how a task gets done
Types of Dendrogram
The AAM (actual agreement method) dendrogram shows the % of participants who agree with the grouping The BMM(best merge method) dendrogram shows the % participants who agree with parts of the grouping
Dendrogram
The dendrograms present data clusters that represent participant agreement about the card categories.
Similarity Matrix
The similarity matrix is a simple representation of pair combinations, intended to give you a quick insight into the cards your participants paired together in the same group the most often. type of cluster analysis- quantifies the data by calculating the strength between pairs
Field Study
The systematic study of behaviors as it occurs in the natural environment - Field ethnography - video ethnography - self reporting Especially useful when you can't talk to the users (Ex: Watching a Dr. in surgery) Consideration: - gaining stakeholder support - bias introduced by investigator (simplification bias) and participant (translation bias)
Ways to communicate results from research:
These methods are translation deliverables: - task flows - personas - use cases - scenarios - storyboarding - ... and many more
Design
Translating user needs into real concepts and going through iterations
Digital Product Development Roles
UX researchers focus on generative (understanding users) and evaluative (user testing) research. UX designer- design and prototyping UX strategy- indicating KPIs and business goals
Storyboarding
Visual representation of how the user interacts with the product--> shows key moments in action
When should you conduct an interview?
When you want individual information, or when there is sensitive information
When to use a card sort?
When you want to understand users' mental model (how they group things) When you need feedback about content, terminology, and organization of a product