Cross-cultural web usability model

Different cultures consume information in different ways. This is why a website that’s designed for a western audience might not necessarily work well for an eastern audience, and vice versa. Alexander, Murray, and Thompson propose design guidelines that make it much easier to create and evaluate usable cross-cultural websites.

Chinese and American websites don’t look the same. Their writing systems differ of course, but cultural factors also influence how people from different cultures browse the web and consequently how web pages are designed.

Over the years the design community has come up with design guidelines that aim to help designers incorporate cultural factors into website designs. However, these guidelines are often too high-level to be of practical use. Moreover, they’re often based only on cultural theories and lack usability tests that support their claims.

The authors therefore propose a new cross-cultural web usability model that addresses these issues by incorporating design elements, cultural factors, and HCI (human-computer interaction) factors.

The study appears to lean heavily on reviews of existing literature, which includes quantitative work that has been done previously by the paper’s three authors (Alexander, Thompson, & Murray, 2017).

Previously the authors studied differences in design attribute usage between cultures by evaluating Australian, Chinese, and Saudi Arabian web pages. That study provided quantitative evidence that many design elements are indeed culturally specific.

The table below lists prominent design elements, i.e. design elements that appear at least 40% more often in websites from exactly one or two of the three countries.

According to Hofstede, Hofstede, and Minkov (2005) cultures can be distinguished by ranking them among four factors on a 100-point scale:

• Tolerance towards large power distances within a society;

• Propensity towards individualism or collectivism;

• A society’s tolerance for risk and uncertainty, as opposed to desire for predictability;

• Whether a members of a society focusses on long-term goals.

Another widely used model is that of Hall and Hall (1990), which distinguishes between:

• High-context cultures, which communicate implicitly with heavy reliance on context, and low-context cultures, which do not;

• Monochronic or polychronic time perception, i.e. whether individuals prefer single-tasking or multi-tasking.

The corresponding values for the Australia, China, and Saudi Arabia are listed in the table below:

HCI factors describe how cultures process information with regards to time, context, and mental aspects. These factors include:

• Speed: how long information is presented

• Density: the number of elements at a single point in time

• Frequency: the number of elements per time unit

• Sequentiality: the order in which information is presented

• Redundancy: how often information is repeated

Combining findings from the literature review with correlations between prominent design elements, cultural factors, and HCI factors results in a more general model that will also work for countries that are not Australia, China, or Saudi Arabia.

The model is primarily presented in the form of two tables.

The first table describes correlations between HCI factors and cultural factors:

The second table provides advice on how to employ web features for audiences with particular cultural characteristics:

The tables don’t make that much sense on their own. Fortunately the authors also provide some examples that should make it a bit clearer.

Navigation and layout

Cultures with a short-term orientation are more focussed on completing single tasks and therefore dislike information complexity. This corresponds with navigation menus which show only a few items at a time and hide most items within deeply nested hierarchies.

Users from long-term-orientation and low-uncertainty avoidance cultures on the other hand, will want to see as much information as possible. They’re better at efficiently filtering large amounts of information and thus don’t like it if a user interface hides information from them.

Similar patterns can be seen in footers: high-uncertainty avoidance cultures, which dislike ambiguity, prefer fat footers with redundant information in order to avoid overlooking potentially important information.

Links

Websites from high-power distance, low-uncertainty avoidance, and long-term orientation countries like China make heavy use of external links, which often open in new browser tabs (side note: The article only mentions browser windows, but nowadays most browsers default to opening new tabs instead.). These make it easier to access larger amounts of information and emphasises relationships between websites (of organisations).

Presenting those same links to users from cultures that do not share these cultural characteristics is a good way to confuse them.

Multimedia, colour, and text

Web pages typically consist of text and visual images. The ratio between those two depends on two cultural characteristics:

• High-context cultures don’t need that much text to get their point across and instead make heavy use of images, animations, scrolling, colours, and bold text (side note: This also happens to be a pretty accurate description of GeoCities websites from the 90s – these were predominantly made by people from low-context cultures however. 🤷‍♂️).

• Attention-grabbing elements are especially important if a website intends to target a low-uncertainty avoidance audience, because they’re less likely to scan web pages linearly and might therefore expect interesting content to appear pretty much anywhere.

• Users from high-uncertainty avoidance cultures prefer softer colours over brighter colours, so as not to be distracted from their goal(s).

Visual representation

Humans are automatically attracted to images of humans, so website designers often try to incorporate lots of them.

• Cultures that tolerate large power distances typically prefer images of strong leaders or figures of authority, whereas low-power distance cultures prefer images of ordinary people;

• Collectivistic cultures prefer images of groups and political leaders that promote group characteristics, whereas individualistic cultures would rather see images of young individuals.

The last two tables in this article can tell you what you should and should not do when creating designs for culture-specific web pages, provided that you know the appropriate cultural factors, which serve as input values:

• The first four cultural factors take the form of numbers in the range 0–100. A value 0–40 corresponds to “Low”, 41–60 to “Medium”, and 60–100 to “High”.

• High-context and low-context are pretty straightforward, as they map to “High” and “Low” respectively.

• The last cultural factor, time perception, is a bit confusing: monochronic time perception (single-tasking) maps to “High”, while polychronic time perception (multi-tasking) maps to “Low”.

1. Alexander, R., Thompson, N., & Murray, D. (2017). Towards cultural translation of websites: A large-scale study of Australian, Chinese, and Saudi Arabian design preferences. Behaviour & Information Technology, 36(4), 351–363.

2. Hofstede, G., Hofstede, G. J., & Minkov, M. (2010). Cultures and organizations: Software of the mind. New York, NY: McGraw–Hill Education.

3. Hall, E. T., & Hall, M. R. (1990). Understanding cultural differences: Germans, French and Americans. Yarmouth, ME: Intercultural press.