Maps for Decision Making

Literature Review

Millennials are a mobile audience

This investigation focuses on the millennial generation, people roughly between the ages of 21 and 34. I chose to focus on this age group in the United States because, despite being the most diverse generation (Krueger, 2013), members of this cohort appear to share many preferences and tolerances that are markedly different from past generations. Many inherit high debt from college expenses, and quickly look for employment in cities and urban areas, hoping to both earn and save money soon after graduation (Fulton, 2012). According to a 2013 summary of the preferences of Millennials, the American Public Transportation Association (APTA) concluded that many in this generation are moving to so-called millennial “hot-spots;” cities such as Boston, Chicago, San Francisco, Seattle, Portland and Washington D.C. Additionally, more young adults delay the move out to the suburbs until their mid to late 30’s, if at all (APTA, 2013).

This marks a distinct shift in preferences from suburban, drivable communities towards walkable, urban neighborhoods, according to the 2013 Urban Land Institute’s report on urban growth patterns in the United States (Krueger, 2013).Generally speaking, young adults prefer living close to downtown areas, as well as easy access to local businesses through a variety of public transportation modes. Whereas in previous generations, owning a car was seen as a “coming-of-age” rite, to Millennials in the city, cars are viewed as a “hassle” (Filisko, 2012). Others attribute this transportation preference to Millennials wanting to spend a minimum amount of time commuting, unless they can simultaneously multi-task, either through texting, exercising, or socializing (Filisko, 2012). Moreover, by the time they decide to move to a city, Millennials have yet to reach their full earning potential, and many are not yet ready to have children, so an emphasis on living near good school systems is not a high priority (Hudson, 2013).

Decision-making strategies can be complex and diverse

Many cognitive and behavioral psychologists study and propose theories about human decision-making processes. Thomas Princen, Associate Professor at the University of Michigan School of Natural Resources and Environment, for one, resists the idea that humans are naturally short-term thinkers. Instead, he claims that there is both biological and physiological evidence that our ancestors exhibited both long-term and short-term planning abilities, citing such examples as the use of fire for cultivating future crops, and the propensity for hunter-gatherer societies to raise families in multi-generational households (Princen, 2009). But in modern, consumerist societies, he argues, there is a shift in focus towards celebrating abundance and concern for the “new,” and thus, the immediate often overrides the ability to plan for the future. Because the future is cognitively flexible (“anything could happen”), people can be manipulated into focusing solely on the immediate (Princen, 2009). A goal of this investigation is to see how design can help shift a user’s focus towards long-term planning.

In addition to the situational differences among recent graduates, each person may approach long-term decision making in a variety of ways. Some may try to gather all possible information before choosing the option that offers the greatest potential. Others may follow what Princeton University professor of psychology Daniel Kahneman describes as Prospect Theory — where large decisions are fragmented into smaller ones, and each is evaluated based on contextual gains and losses (2003). Patricia Wright advocates for designers to better respond and accommodate this diversity in decision-making strategies:

“...people may differ in being visualizers or verbalizers (Mayer and Massa, 2003), in being convergent or divergent thinkers, and in the knowledge and expectations they bring to the decision-making task. It means being responsive to differences in age and language fluency, for these are known to change people’s choices about the way information is communicated (Wright et al., 2008)”

— (Wright, 2009, p. 207).

In addition to diverse cognitive strategies, users approach decision-making with a variety of emotional responses. According to Harvard professor of psychology Daniel Gilbert, people are often bad at predicting their emotional response to decisions because they fail to account for what he calls their “psychological immune system.” This, he says, is a process by which people quickly adapt to concrete decisions (Gilbert & Ebert, 2002). Instead, decision-makers try to protect themselves against sub-optimal outcomes by purchasing escape clauses, return options, or otherwise providing themselves an “out” (Gilbert & Ebert, 2002). The disadvantage to such escape clauses is that the decision-maker will then spend more time considering potential alternates to the one they already have, wasting time that could be used to make other future decisions.

Official & Unofficial Data Sources

Academic geographer Michael Peterson argues that although any map requires a critical “leap of faith” in order to be useful, the abundance and ease of creation for online maps means there is reason to be skeptical of any map’s information source (Peterson 2009). The sources of location-specific information cover a wide spectrum. On one end, official sources provide quantitative data and statistics about a city. Real estate agencies collect and publish housing rental prices and market activity data. Regional and local transit authorities, as well as Google and other aggregated sensory systems make available transportation data. Government agencies publish weather and environmental data, and federal and state forums collect and issue crime statistics.

Without proper expressions of certainty in information quality, the technology used to display this data can falsely present the user with what appears to be a complete picture. In his book, Picturing the Uncertain World (2009), author Howard Wainer describes several scenarios where readers draw incorrect and often costly conclusions from data, due to over-reliance on the “objectiveness” of the data visualization.

At the other end of the spectrum, user-generated information about a particular place is used to inform decisions. Michael Goodchild, a research professor at the University of California, Santa Barbara, named this particular data “Volunteered Geographic Information,” or simply “VGI” (2007). VGI is the result of large-scale engagement from social media users in the creation of online, geo-tagged content, often taking the form of images (uploaded to Flickr or Facebook, for example), or digital narratives as tweets or other social network postings. This practice creates digital translations of physical spaces as well as the human activity that occurs in those spaces. Such user-generated data proves helpful when visualized in map form, as seen on the opposite page in such cases as disaster response [1] , determining what citizens name regions of a city [2], and tracking the riots that occurred after the 2012 NCAA Men’s Basketball championship in Kentucky [3].

While these previous examples suggest audiences of local citizens, sociologists and academic researchers, they each demonstrate a core value of VGI as one that embraces a dynamic, “bottom-up” approach to digital knowledge creation of place. This contrasts with traditional “top-down” knowledge platforms from officials or experts that some argue to be static, authoritative, and reductively definitive (Sui, Elwood and Goodchild, 2013). I believe, when thoughtfully referenced and visualized, this citizen-generated content represents a potentially rich source of information for people relocating to a particular city.

While digital interfaces make it easy to assume that geo-tagged content is extensive, comprehensive and diverse, authors Graham and Zook (2013) claim there is still vast unevenness in indexed content about the world’s cultures. Studies that analyze Flickr photos near the US-Mexican border by Watkins (2012), and geo-referenced content of multi-lingual regions in Ontario and Quebec, Canada by Graham and Zook (2013), demonstrate that both cultural and linguistic differences — even within close geographic proximity — create varying amounts of information and fundamentally different representations of place. As a result, they claim, “... places are increasingly defined by dense and complex layers of representation that are created, accessed and altered via digital technologies and often opaque lines of coded algorithms,” (2013).

Such linguistic and cultural variations, combined with user-centered, computational search algorithms may prevent users from accessing content from a variety of information sources.As both official and user-generated geographic data becomes more readily accessible to digital users, it will be evermore crucial for users to assess information quality. Despite its benefits, VGI still poses significant limitations in accurately representing activity in physical places. Authors such as Crampton (2013), Graham (2011) and Hale (2012) advocate for a cautious approach to VGI data analysis. As one example, computer algorithms that assign geo-tags can be inaccurate or incomplete, referencing the location of a user’s profile instead of the actual location where digital content is created. Thus, in aggregate, the information may not always be an accurate proxy for the activities of a particular place (Hale, Ga ney and Graham, 2012).Hollenstein and Purves (2010) argue that people think and reason about geographic locations in vague terms. While our technology is capable of recording information in precisely calculated distances and geo-spatial coordinates, the human mind conceptualizes space in less-defined, often hierarchical ways.

These vaguely expressed relationships often take the form of “near the park,” “downtown,” or “on Wade Avenue” in online annotation. While these definitions may come into conflict with official or more precisely defined legal boundaries in various media, they are still crucial for communicating geographic information at multiple scales through language (Hollenstein & Purves, 2010). Hollenstein and Purves continue to explain that there are areas of agreement and disagreement among citizens and officials with regard to what areas are called: “Typically, there will be locations that are clearly agreed upon as being part of a place, perhaps situated towards the center of an associated region.

However, there will often be uncertainty and disagreement on locations towards the fringe that might be less characteristic,” (2010). These names and designations, they say, are in constant transition. Hence, for purposes of investigating a city remotely, a user must have the opportunity to combine a range of precise and vague descriptions. Thus, digital map interfaces that attempt to present multiple sources of information must present the user with the necessary tools to aggregate, assess, and verify such language and knowledge. In the case of deciding where to live, users need to be able to compare abstract, official data with a variety of citizen-generated knowledge.