Metro Detroit: residential location and neighbourhood perceptions

Abstract Metro Detroit has long been one of the most socially divided metropolitan regions in the United States. While the north-western suburbs are among the wealthiest in the nation, the region’s urban core has struggled with the nation’s post-industrial transition. Despite benefiting from various intra-regional ties, people from Metro Detroit’s suburbs often share negative stereotypes about the city at the region’s core and attempt to distance themselves from it. This article analyses surveys from 140 residents of the Metro Detroit region in order to understand better intra-regional perceptions of Detroit proper. Rather than the hypothetical neighbourhoods of varying racial diversity often found in perceptions research, surveys asked residents about real places in their home region. Asking about actual locations helped answer the research question behind this article: is residential location associated with neighbourhood perceptions? While race continues to influence neighbourhood perceptions, findings show that residential location is also a significant factor.


INTRODUCTION
New regionalist claims of structural developments leading to greater regional cooperation may not be substantiated in metro regions such as Detroit (Indergaard, 2015). Some large social institutions may push for intra-regional cooperation, but the social actors on the ground may also resist attempts to link their fates with those across socially diverse regions. Detroit's 1967 race riots and decades of racial politics continue to influence local perceptions about city and suburbs. Suburban counties also maintain longstanding divisions across socioeconomic lines (Sugrue, 2005). Indeed, recent years have seen a decline in socially beneficial regional cooperation across Metro Detroit. 1 . The decision by some suburban municipalities to stop participating in the regional bus system, for example, has negatively impacted the lives of more than 100,000 residents of Detroit proper -two-thirds of the city's employed residents -who commute to work in the suburbs (Urban, 2015). Facing fiscal challenges, suburban municipalities have also united against Detroit proper in competition for declining state and federal funds (Indergaard, 2015). At the same time, various efforts within the city are attempting to overcome the problems that contributed to America's largest municipal bankruptcy (Schindler, 2016).
Against this backdrop, it is important to understand what influences intra-regional perceptions of Metro Detroit. Assessing perceptions is an important step in understanding residential segregation across metropolitan regions (Krysan & Bader, 2009). Interestingly, the foundational work in this area, as well as numerous studies since, drew upon participants from the Metro Detroit region (Farley et al., 1978). The special relevance of Metro Detroit to neighbourhood perceptions research stems from the fact that it has consistently been ranked as one of America's most segregated metropolitan regions throughout the growth of this area of research.
The general research question behind this article is: is residential location associated with neighbourhood perceptions? This article analyses responses from 140 surveys completed by residents of Metro Detroit about their views on the region they call home. Race continues to be an important factor in perceptions about the region's urban core, but it is not the sole factor. Though participants from different races often expressed different views, several (partial) associations between perceptions and residential location were identified. The major lesson from this article is that scholars should be more willing to consider residential location in their analysis of perceptions of real neighbourhoods. Residential location and race are often significantly associated, especially in metropolitan regions with declining urban cores such as Detroit. By controlling for race, however, this study shows that residential location can be a significant factor of its own.

RESEARCHING NEIGHBOURHOOD PERCEPTIONS
Over the past four decades, scholars have developed a substantial body of literature on neighbourhood perceptions. The significance of such research for scholars and practitioners lies in the belief that perceptions may help explain differences in lived experiences across metropolitan areas. Most notably, perceptions have been linked to residential segregation according to race (Iceland & Wilkes, 2006;Krysan & Bader, 2009). Remarkably, much of the research has not asked about perceptions of real places known to participants. According to Bruch and Mare (2006), the use of hypothetical vignettes to assess participants' perceptions can present unrealistic (racial) scenarios. Crowder and Krysan (2016, p. 21) have further identified the absence of 'meaningful information on how individuals perceive actual neighbourhoods that might represent residential options' as a major gap in the literature.
In addition to considering real neighbourhoods with socially meaningful histories that are known to participants, perceptions studies can be improved by expanding the frame of analysis. Studies have primarily focused on the role of race in determining which areas people consider desirable (Charles, 2006;Krysan & Bader, 2007;Sampson & Sharkey, 2008). While race remains an undeniably important consideration, it is often complicated by other variables, such as those pertaining to social class (Bruch & Mare, 2006). Accordingly, it is important to consider other variables when looking at neighbourhood perceptions (Iceland & Wilkes, 2006;Swaroop & Krysan, 2011). One variable often associated with race, yet largely ignored within the literature, is residential location (Bader & Krysan, 2015). A primary aim of this article is to contribute to the literature by assessing the extent to which residential location may be related to perceptions of real places across the Metro Detroit region.

METHODOLOGY AND CASE STUDY OVERVIEW
The data presented in this article stem from a larger quantitative project aimed at understanding how race, social class and other variables are associated with Metro Detroiters' intra-regional perceptions and activities. Data collection occurred in public spaces (i.e., parks and pavements) across Metro Detroit. As indicated in Figure 1, participants were recruited from nine socioeconomically diverse municipalities within the region's three counties (i.e., Macomb, Oakland and Wayne). In order to ensure familiarity with the neighbourhoods in question, only residents of the region were eligible to participate. A total of 140 face-to-face surveys were completed at a reported response rate of 57%.
In addition to questions about their background and frequency of travel to Detroit proper, participants were asked to respond to a variety of statements using a five-point Likert scale. Such questions assessed their perceptions of both urban and suburban areas within the region. These included statements pertaining to safety (e.g., I feel safe driving in Detroit proper) and residential desirability (e.g., Detroit proper is a good place to live). Results of bivariate analysis are provided Regional StudieS, Regional Science to show what most influences regional perceptions of Detroit. Since race is often linked to social class and residential location, multivariate analysis was conducted to control for the effects of race.

DATA
The purposive sampling technique used in this study resulted in a diverse sample across several demographic variables. Important for this study, the sampling technique provided participants with diverse residential backgrounds. At the time of the survey, participants resided in 36 different municipalities across Metro Detroit. While the residential location of participants was not intended to mirror the region's actual makeup, the general tendencies are reflected within the population. A demographic summary of participants is displayed in Table 1.
As discussed above, race has undoubtedly been linked to neighbourhood perceptions. While this article focuses on the potential impact of residential location, its purpose is not to dismiss the effects of race. In fact, given the link between residential location and race in segregated regions like Metro Detroit, the analysis controlled for race whenever testing for associations with residential location. Table 2 provides results from Fisher's exact test of association. The test was used as an alternative measure of association for chi-square to counteract the relatively small sample size (n = 140).
In addition to the values from the complete cross-tabulations, Table 2 includes the partial associations that resulted when controlling for race. Since 93% of valid responses indicated that participants were either black or white, the analysis was limited to those who self-reported as either black or white. Removing the small number of other responses helped combat difficulties (i.e., low expected frequencies) when testing for association.

DISCUSSION
At first glance, it seemed that residential location was the most important factor associated with intra-regional perceptions of Metro Detroit. On controlling for race through the process of elaboration, however, many of those associations were found to be spurious. Table 3 shows some of these spurious relationships. In this sense, the study supports previous findings that linked race with neighbourhood perceptions. This is not an unimportant finding, as it speaks to the First, the findings support other social research that challenges the popular notion that America had become a post-racial society. Second, this supports Indergaard's (2015) suggestion that new regionalism might not apply in a region like Metro Detroit, where many suburbs have a 'sense of racial estrangement and competition with Detroit' proper (p. 138). The region's complicated racial history continues to impact residents' views and lived experiences. While findings show that race continues to be an important factor influencing perceptions across the Metro Detroit region, they also provide a new perspective by revealing a relationship between residential location and perceptions. The one association that was maintained for both racial groups was between county of residence and the statement that residents of Detroit proper work as hard as residents of suburban Detroit. For this question, participant responses varied significantly depending upon their county. Among blacks, residents of Macomb County -destination of the newer wave of intra-regional outmigration from the city proper -were least likely to agree with this statement. For whites, Oakland County was the only one to have a majority of residents disagree with the statement regarding Detroiters' work ethic. The finding that whites from the wealthiest county in the region do not perceive the (predominantly black) residents of the urban core to be hard workers may come as little surprise to social scientists familiar with the region's discriminatory past. In addition to being a choice location for Detroiters with higher socioeconomic status, the western part of Detroit proper was home to the most blatant and combative racist practices of the post-war era, including the construction of a wall to divide white and black neighbourhoods. As the region began to suburbanize, Oakland County drew primarily  upon those previously living in western Detroit proper (Sugrue, 2005). Asking participants about real places in their home region has allowed this article to put their perceptions into context and identify a relationship between their residential location and those perceptions. In addition to the relationship between county and perceived work ethic of the urban core's residents, multiple perceptions were partially associated with residential location. When significant association varied according to county, Wayne County tended to have the more favourable perceptions of Detroit proper. Not only did whites from Wayne County agree more strongly with the statement that they feel safe driving in the urban core, blacks from the county were more agreeable with the statement that it is a good place in which to work. One potential explanation for these more favourable views is that Detroit proper is part of Wayne County. Being more closely tied to Detroit proper (e.g., through county governance, taxes etc.) may lead residents of Wayne County to look at the urban core less harshly than those from the region's north.
Partial associations resulting from significant differences between urban and suburban residents were also identified. It may seem unsurprising that residents of Detroit proper tended to have more positive perceptions, but it should not be taken for granted. For decades, people with the financial means to leave the city have done so. Many of those remaining in Detroit proper have been those who are essentially trapped within the declining city (Sugrue, 2005). In contrast to suburban blacks, blacks residing in the region's urban core were significantly more likely to agree that they feel safe walking in Detroit proper. Similarly, white residents of Detroit proper felt safer driving the urban streets than did their suburban counterparts. Lacking inherent causality, it is unclear if they chose to live in Detroit proper because they were not afraid or if their experience living in the city has caused them to reject the stereotype that Detroit is wildly dangerous for all. Either way, the experiences of white participants currently residing in Detroit proper have not resulted in perceptions that the city is unsafe.
These partial associations show that residential location can be significantly related to neighbourhood perceptions. It is not confined to spurious relationships. Though the chosen sampling technique does not promote generalizations, these findings can be valuable to scholars interested in neighbourhood perceptions across diverse metropolitan regions.

CONCLUSIONS
This article has shown that residential location is, at least partially, associated with intra-regional perceptions of Metro Detroit's urban core. By considering the importance of residential location in the formation of neighbourhood perceptions, this article makes an important contribution to the existing body of literature. Even though many social scientists have discussed how people and places influence each other (e.g., Gieryn, 2000), perceptions researchers have largely failed to acknowledge that participants' physical location within real places can impact their perceptions. One possible explanation for the relative neglect of residential location in neighbourhood perceptions research is that researchers take for granted the link between race and residential location. From the United States to France to South Africa, studies continue to emphasize this link (Iceland & Wilkes, 2006;Préteceille, 2016;Seekings, 2008). If race is assumed to explain away the effect of residential location, then there is no need for the latter. As discussed above, however, this assumption does not always hold. At the very least, this article shows that future research should look more closely at the relationship between residential location and other key factors.
In order to understand properly (and combat social inequalities tied to) the relationship between race, residential location and perceptions, it is important that researchers and practitioners take into account the local social conditions (Bader & Krysan, 2015;Swaroop & Krysan, 2011). Accordingly, future research should also continue to look at how perceptions of real places are related to behaviours. Such endeavours can strengthen the social applicability of the scholarship by expanding beyond the home ownership/rental patterns often discussed by neighbourhood perceptions scholars. In addition to residential segregation, it is important for scholars to consider other forms of social segregation that are beyond the scope of this article, such as social class. If people are unwilling to travel to certain areas within their region because of negative perceptions, as appears to be the case in Metro Detroit, then there are opportunities for those areas to be lifted out of the socioeconomic basement. By identifying areas where people with significantly more negative perceptions of Detroit proper reside, this study can assist practitioners in Metro Detroit who aim to improve perceptions in order to promote greater intra-regional social integration and cooperation. While this study focused on a specific metropolitan region, these lessons may be applied to other regions with declining urban cores surrounded by relatively steady suburbs.