Human Computer Interaction

1.5 Understanding HCI Research Methods and Measurement

HCI research requires both rigorous methods and relevance. It’s frequently tempting to lean more heavily towards either. Another fields of research do focus more about theoretical results than you are on relevance. However, HCI research should be practical and highly relevant to people, organizations, or design. The study needs so that you can influence interface design, development processes, user training, public policy, or anything else. Partly because of the philosophies from the founders from the field, HCI has already established a historic concentrate on practical results that improve the caliber of existence (Hochheiser and Lazar, 2007). What is the tension sometimes between researchers and practitioners? Absolutely. But all HCI research should a minimum of consider the requirements of both audiences. Simultaneously, the study methods used (whatever the source discipline) should be rigorous and appropriate. It’s not sufficient to build up a brand new computer interface without researching the requirement for the interface and without following track of user evaluations of this interface. HCI researchers are frequently placed ready of evangelism where they have to get out there and convince others of the requirement for an emphasis on human users in computing. The only method to support statements on the significance of users and human-centered design is by using solid, rigorous research.

For this reason interdisciplinary focus and also the historic growth and development of the area, there are various methods to measurement and research presently used in the area of HCI. Several researchers, all focusing on HCI-related topics, frequently disagree on which “;real HCI research” means. You will find major variations in how various leaders within the field see the presence of HCI. Remember that, being an HCI investigator, you might encounter individuals who don’t much like your research methods, aren’t confident with them, or just originate from another research background don’t know them. Which’s OK. Consider it as being another chance to become an HCI evangelist. (Note: So far as we all know, the word “;interface evangelist” was initially accustomed to describe Bruce Tognazzini. But we actually believe that the word pertains to many of us that do HCI-related work.) Since the aim of this book is to supply a guide that introduces the readers towards the group of generally recognized empirical research practices within the concept of HCI, a main real question is, therefore, how can we execute measurement in the area of HCI research? Exactly what do we measure?

In the past of HCI research, measurement took it’s origin from standards for human performance from human factors and psychology. How quickly could someone develop a task? The number of tasks were completed effectively, and the number of errors were created? These are the fundamental foundations for calculating interface usability and still relevant today. These metrics are extremely much with different task-centered model, where specific tasks could be separated out, quantified, and measured. These metrics include task correctness, time performance, error rate, time for you to learn, retention with time, and user satisfaction (see Chapters 5 and for additional info on calculating user satisfaction with surveys). These kinds of metrics are adopted by industry and standards-related organizations, like the National Institute of Standards and Technology (within the U . s . States) and also the Worldwide Organization for Standardization (ISO). While these metrics continue to be frequently used and well-recognized, they’re appropriate only in situations where using computers could be damaged lower into specific tasks which themselves could be measured inside a quantitative and discrete way.

Shneiderman has described the main difference between micro-HCI and macro-HCI. The written text in the last paragraph, improving a person’s experience using well-established metrics and methods to enhance task and time performance, might be considered micro-HCI (Shneiderman, 2011). However, most of the phenomena that interest researchers in a broader level, for example motivation, collaboration, social participation, trust, and empathy, possibly getting societal-level impacts, are challenging measure using existing metrics or methods. A number of these phenomena can’t be measured inside a laboratory setting while using human factors psychology model (Obrenovic, 2014 Shneiderman, 2008). And also the classic metrics for performance might not be as appropriate when using a brand new technologies are discretionary contributing to enjoyment, instead of task performance inside a controlled work setting (Grudin, 2006a). In the end, how can you measure enjoyment or emotional gain? How can you measure why individuals use computers once they don’t need to? Job satisfaction? Sense of community? Mission in existence? Multimethod approaches, possibly involving situation studies, observations, interviews, data logging, along with other longitudinal techniques, might be most suitable for being aware of what makes them new socio-technical systems effective. For example, the study section of Computer-Supported Cooperative Work (CSCW) highlights the sociological perspectives laptop or computer usage greater than the mental perspectives, having a focus more about observation within the field, instead of controlled lab studies (Bannon, 2011).

That old ways of research and measurement are comfy: hypothesis testing, record tests, control groups, and so forth. They are available from the proud good reputation for research, and they’re easily understood across a variety of academic, scientific, and research communities. However, they alone aren’t sufficient methods to measure all today’s phenomena. Exactly the same pertains to the “;old standard” measures of task correctness and time performance. Individuals metrics may measure “;how frequently?” or “;how lengthy?” although not “;why?” However, they’re still well-understood and well-recognized metrics, plus they allow HCI researchers to speak their leads to other research communities in which the cutting-edge tools and research methods might not be well-understood or well-recognized.

You might be unable to use experimental laboratory research to understand why people don’t use technology. If you wish to examine how people use portable or mobile technology for example smartphones and wearable computing, you will find limitations to studying that inside a controlled laboratory setting. If you wish to study how people talk to reliable partners, decide to perform transactions with someone they don’t know on another continent (as frequently happens with Ebay), or decide to collaborate, you have to find new methods for research and new types of measurement. These aren’t research questions that may be clarified with quantitative measurements inside a short-term laboratory setting.

Consider Wikipedia, a collaborative, open-source encyclopedia. Presently, greater than 5 million articles appear in British on Wikipedia, with approximately 70,000 active contributors (https://www.wikipedia.org), who spend their very own time creating and editing Wikipedia records. What can cause them to do this? Exactly what do they get free from the knowledge? Clearly, task and time performance wouldn’t be appropriate metrics to make use of. What metrics ought to be used? Pleasure? Emotion? A sense of community? Lower bloodstream pressure? It isn’t really a phenomenon that may be studied inside a controlled laboratory setting (Menking and Erickson, 2015). The concept of HCI has started to use more research methods in the social sciences, so we let the readers to begin with a couple new information approaches that aren’t even just in this textbook! Take note that individuals using their company disciplines, along with your “;home discipline,” will most likely challenge the suitability of individuals research methods!