User:VirpiRoto/UX evaluation

User experience (UX) evaluation means investigating how a person feels about using a system (product, service, non-commercial item, or a combination of them). It is non-trivial to evaluate user experience and come up with solid results, since user experience is subjective, context-dependent and dynamic over time. Laboratory experiments may work well for studying a specific aspect of user experience, but holistic user experience is optimally studied over a longer period of time with real users in a natural environment.

Detailed guidance for user experience evaluation is hard to give, since there are many different dimensions to consider when choosing the evaluation approach: In all cases, however, there are certain aspects of user experience that evaluators are interested in (measures), and certain procedures and techniques used for collecting the data (methods). When investigating user experience evaluation methods, we can identify methods for emotion assessment and overall UX assessment. The measures and methods for these two evaluation types are described below. Episodic UX can be evaluated with either approach, depending on the case.
 * Goal: Summative (score) or formative (areas for improvement)
 * Approach: Objective or subjective
 * Data: quantitative or Qualitative
 * Granularity: Momentary, episodic, or overall UX
 * Setup: Lab or field

Emotion evaluation
When investigating momentary user experiences, we can evaluate the level of positive affect, negative affect, joy, surprise, frustration, etc. The measures for emotions are bound to the methods used for emotion assessment, but typical emotion measures are e.g. valence and arousal. Objective emotion data can be collected by psychophysiological measurements or by observing expressed emotions. Subjective emotional data can be collected by using self-report methods, which can be verbal or non-verbal.

Examples of emotion evaluation methods:
 * Psychophysiological emotion measurements aim to identify emotions from physiological changes in muscles (e.g. face), pupils, skin, heart, brains, etc.
 * Expression observers monitor person's facial and other gestures or the tone of voice to identify emotions manually
 * Think aloud protocol can be used for reporting emotions (real-time verbal self-report)
 * PANAS (retrospective verbal self-report)
 * Geneva emotion wheel (retrospective verbal self-report)
 * Emotion Slider (continuous non-verbal self-report)
 * Sensual Evaluation Instrument (snapshot non-verbal self-report)
 * PrEmo, a new version of EmoCards for assessing emotion (snapshot non-verbal self-report)

Overall UX evaluation
In contrast to identifying a momentary emotion, overall UX evaluation investigates how a person feels about a system as a whole, typically after using it for a while. Many of the overall UX evaluation methods are suitable also for evaluating episodic UX, i.e., assessing how a person feels about a specific interaction episode or after executing a task.

There is no agreement on the exact measures for evaluating the overall UX with a system, largely because different products aim at different kinds of experiences. However, there are some high-level constructs of user experience that can be used as the basis for defining the user experience measures, for example:


 * 1) Utility: Does the user perceive the functions in the system as useful and fit for the purpose?
 * 2) Usability: Does the user feel that it is easy and efficient to get things done with the system?
 * 3) Aesthetics: Does the user see the system as visually attractive? Does it feel pleasurable in hand?
 * 4) Identification: Can I identify myself with the product? Do I look good when using it?
 * 5) Stimulation: Does the system give me inspiration? Or wow experiences?
 * 6) Value: Is the system important to me? What is its value for me?

Since the importance of the above user experience constructs is different to different people, it is an interesting option to define the overall UX measures together with each study participant. Another option to evaluate overall UX is to use simply a scale from positive to negative, without further consideration of the user experience constructs.

Overall UX assessment is methodologically different from objective emotion assessment, but similar to subjective emotion assessment. Generic subjective user experience evaluation methods include interviews, questionnaires, story-telling, and often, a combination of these. An individual method can collect data about a set of specific constructs of user experience, such as usability testing is to collect data about usability construct.

Examples of overall UX evaluation methods (excluding traditional usability methods):
 * Diary methods for self-reporting experiences during field studies
 * Experience Sampling Method (ESM) for self-reporting during field studies
 * Day Reconstruction Method (DRM) - story-telling to reveal the meaningful experiences during field studies
 * AttrakDiff questionnaire for overall UX evaluation
 * Ladder interviews e.g. to find out attitudes or values behind behaviour or experience