http://prettytallstyle.com/birkof/7052 Authour:Jason Last & Shelley Simmons
Edited by Christine Keene
Developing a deep understanding of the audience’s mental models or constructs. Insight into how the participants make sense of the domain or subject under consideration for an intervention will be instrumental in solution design.
Card sorting is a participatory, user-centered technique used to elucidate the attitudes, values, desires and/or behaviors of participants as they relate to the domain under study.
Because card sorting is a broadly applicable tool for understanding, it can be used at the front or back end of the design process.
Problem/insight identification, knowledge acquisition, participatory design activities, and evaluation/optimization of a prototype. This method can “provide insight into users’ mental models, illuminating the way that they often tacitly group, sort and label tasks and content within their own heads.” (Rosenfeld & Moreville, 2002).
Card sorting is also useful to:
- Make a structured survey more engaging for the respondent;
- Understand how (and why) individuals break down a big concept, problem or relationship into component parts, revealing a mental pattern;
- Understand what is most important to users, the relative value of items, that which is prioritized and how to best organize or structure a solution; and
- Understand the similarities and differences between multiple user groups as they relate to research questions.
Depending on the nature of the research—whether the technique is used for exploratory or evaluative purposes—the research design may vary slightly. Most often, card sorting projects are “open sort”, in which participants are given a variety of items (images or words) on individual cards and are asked to repeatedly categorize the items according to criteria supplied by the researcher, or according to criteria that make sense to the participant. Open sorts allows for participant-driven discoveries, limiting any preconceived barriers by the designer.
In some instances, it may make sense for the researcher to make the project a “closed sort”, asking participants to sort the items into a pre-determined set of categories (i.e., in a structured survey). A closed sort assumes that the researcher knows the categories that matter to the participants and/or to the research project—but often, this is not the case.
1) http://nkbosna.ba/?myravey=site-de-rencontre-lesbienne-quebec-gratuit&3ca=29 Domain – Determine the conceptual area that you want to research. Depending on the nature of the project, the domain could be something broad (i.e., “Family” or “Work”) or something very specific (i.e., a particular website or application).
2) sites de rencontre disons demain Items – For exploratory research projects, researchers might create an item list by examining past research/literature related to the domain, by brainstorming using a mind-mapping or free association exercise, or through other observational techniques. Items can be concrete, abstract or both, depending on the research objectives. Relevant research suggests that “the maximum number of entities which is conveniently manageable…is about 20 or 30” (Rugg & McGeorge, 2005), though online software makes it possible to include many more.
3) http://www.capacuras.com/?rtyt=rencontre-juif-toulouse&b87=54 Demographics – Online card sorts that use analysis software allow for much larger samples than would be possible or recommended for in-person card sorting exercises. However, the literature suggests that a relatively small sample for in-person card sorts can offer significant results: “As few as 25-30 participants will likely yield results similar to those of several hundred, provided these participants are representative of [your target group] and are familiar with the domain being questioned” (Wood & Wood, 2008).
4) follow url Materials – Print items on individual standard filing cards (one unique item per card). Number each item card for recording purposes. To prepare for the circumstance in which a participant isn’t familiar with the item on the card, provide a brief definition or description on the reverse side of each item card (Rugg & McGeroge, 2005).
5) my messenger rencontre Session – While online card sorting research makes sense for projects that come further along in the design process, the card sorting technique is best applied at the front end of the design process through an in-person session. In-person card sorting allows the researcher to probe respondents on their sorting decisions and gain valuable insights that would be tedious using online software.
The first step in conducting a card sorting exercise is to ensure that the respondent (or respondents, since group card sorting is also possible, although it comes with the large risk of “group think” bias) is clear on the instructions and rules of the exercise. It is important to ask the respondent to avoid creating categories that are based on more than one criterion: for example, “’big and expensive’ should be sorted once for ‘big’ and once for ‘expensive’” (Rugg & McGeorge, 2005).
The researcher should shuffle the deck of cards and lay them out on a large, flat surface. He/she should then ask the respondent to sort the items into groups based on one criterion. ( http://fitbodz.com.au/?frimer=site-de-rencontre-musulman-en-cote-d%27ivoire&e70=f3 Note: the researcher might supply the criterion or ask the respondent to select it him/herself. Use consistency in application across all respondents.)
Once the respondent has sorted the items into groups, the researcher should ask him/her to give each group or category a name and probe the respondent’s rationale. For example: Why were the items grouped this way? Why is this item in this category and not another? The researcher must never criticize the respondent’s decisions—the point of the exercise is to understand an individual’s view, not to reinforce a single “right” view.
The researcher would then re-shuffle the cards and follow the same exercise, changing the criterion upon which the sort is based each time. The literature suggests that a respondent can accomplish a single sort of 100 items in just under an hour, so the researcher would do well to contain the sorting of 20-30 items to four or five single-criterion sorts in one in-person session (Hinkle, 2008).
http://inter-actions.fr/bilobrusuy/3019 Card Sorting Exercise: Participants identified personal core values
6) go to link Record – Because it would be tedious and time-consuming to record each sort on-the-fly, a couple of options are suggested: a) while the researcher is probing the respondent on his/her decisions, have a research assistant record the criterion used, the category names and the card numbers grouped into each category; and/or b) take pictures or record the results of each sort for analysis later on (Rugg & McGeorge, 2005).
7) follow url Analyze – “Domain complexity, time, effort, and ease of use influence what tool you use to analyze card sorting results” (Hinkle, 2008). Options range from intuitive (e.g., simply adding up the frequency with which items are placed in particular categories) to statistical (performing a cluster factor analysis using software). Many online card sorting applications will cluster results automatically, which is helpful to support research for evaluating a prototype or to gain insight into user priorities in a specific domain, such as a website structure. If the research takes place at the beginning of the design process, a basic frequency count may suffice to reinforce insights gathered through speaking with the respondent about the results during the session.
Arranging the sorts on large whiteboards may make it easier to compare sorts and identify emerging patterns. If it is a large card sort, data can be entered into a spreadsheet using the number code on the back of each card. Research by Spencer (2009) suggests that the types of feedback from the card sort analysis would include the following: content that participants haven’t understood well; content that could belong to more than one area; alternative paths to content (for example, a list of all “how-to” articles could be created); and how different types of participants see information.
https://www.cedarforestloghomes.com/enupikos/11037 Open Card Sort
Users can make their own category headings instead of fixed heading, as provided in a closed card sort. It’s useful when creating either new or existing sites.
http://celebritysex.cz/?triores=online-dating-study-race&0f0=26 Closed Card Sort
Users are given a set of existing headings or “primary groups”. It is more useful when building on an existing site. (Spencer)
Card Sort Software
Online software allows for the flexibility to conduct remote sorts, and can be easier to get participants for the sort. The limitation is that you may not get the observation qualitative data. Some of these programs include:
Reverse card sort or “tree testing”
To test your newly created structure, a low-tech Tree Testing method could be used to test how people find information in an existing structure. Once a structure has been created and you need:
- To decide between a few potential structures
- Need quantifiable results to prove that the new structure is better than the old
Cataldo, E.F., Johnson, R.M., Kellstedt, L.A., & Milbrath, L.W. (1970). Card sorting as a technique for survey interviewing. Public Opinion Quarterly, 34 (2), 202 – 215.
Coxon, A.M. (1999). Sorting data: Collection and analysis. Sage University Paper Series on Quantitative Applications in the Social Sciences, 07-127. Thousand Oaks, CA: Sage.
Hinkle, V. (2008). Card-sorting: What you need to know about analyzing and interpreting card sorting results. Usability News, 10 (2), 1 – 6.
LeCompte, M.D., & Schensul, J.J. (2010). Designing & conducting ethnographic research: An introduction (2nd ed., Vol. 1). Lanham, MD: AltaMira Press.
Paul, C.L. (2008). A modified delphi approach to a new card sorting methodology. Journal of Usability Studies, 4 (1), 7 – 30.
Rosenfeld, L. and Morville, P. (2002). Information Architecture for the World Wide Web: Designing Large Scale Web Sites. Sebastopol, CA: O’Reilly & Associates.
Ross, J. (2011, June). Comparing user research methods for information architecture. In UX Matters. Retrieved from http://www.uxmatters.com/mt/archives.
Rugg, G., & McGeorge, P. (2005). The sorting techniques: A tutorial paper on card sorts, picture sorts and item sorts. Expert Systems, 22 (3), 94 – 107.
Spencer, D. (2009). Card sorting: Designing usable categories. Brooklyn, NY: Rosenfield Media.
Spencer, D., & Warfel, T. (2012). Card sorting: A definitive guide. In Boxes and Arrows. Retrieved from http://www.boxesandarrows.com/view/card_sorting_a_definitive_guide.
Usability Body of Knowledge. (2010). Methods: Card sorting. Retrieved from http://www.usabilitybok.org/card-sorting
Wood, J.R., & Wood, L.E. (2008). Card sorting: Current practices and beyond. Journal of Usability Studies, 4 (1), 1 – 6.