Soc Indic Res DOI 10.1007/s11205-014-0810-8
Comparing Subjective and Objective Quality of Life Criteria: A Case Study of Green Space and Public Transport in Vienna, Austria Eva Haslauer • Elizabeth C. Delmelle • Alexander Keul Thomas Blaschke • Thomas Prinz
•
Accepted: 30 October 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract A future-oriented spatial planning has to face the challenges of integrating ecological, social, and economic aspects of living. This is often seen as a principle to also guarantee a certain level of Quality of Life (QoL). QoL can be assessed subjectively, according to individual perceptions, and objectively, via secondary data sources. This paper is concerned with how well these two approaches may agree with one another, and in particular, enables the spatial mis-match between perceived satisfaction and objectively measured results to be identified. The case study of two fundamental QoL dimensions is examined in the city of Vienna, Austria: public transport and green space availability. Areas of general agreement discordance are mapped within a geographic information system and characteristics of residents living in places with a mis-match between satisfaction and GIS-derived measurements are assessed. Results show that while the objective and subjective measurements are largely in congruence with one another, some variations do exist, thus stressing the spatial heterogeneity in residential QoL perceptions. Keywords Quality of Life Geographic-Information-Systems Social-scientific survey Green space Public transport
E. Haslauer T. Blaschke Interfaculty Department of Geoinformatics-Z_GIS, University of Salzburg, Schillerstraße 30, 5020 Salzburg, Austria E. Haslauer (&) T. Blaschke T. Prinz Research Studios Austria, Studio iSPACE, Schillerstraße 25, 5020 Salzburg, Austria e-mail:
[email protected] E. C. Delmelle Department of Geography & Earth Sciences, University of North Carolina at Charlotte, 9201 University City Blvd, Charlotte, NC 28223, USA A. Keul Department of Psychology, University of Salzburg, Hellbrunnerstraße 34, 5020 Salzburg, Austria
123
E. Haslauer et al.
1 Introduction Quality of Life (QoL) is a multifaceted term that incorporates such notions as a good life, valued life, satisfying life, and happy life (McCrea et al. 2006). For city residents, QoL may be impacted by features of the urban environment at a number of different scales or locals: for instance, where one lives, works, or plays (Marans 2012, McCrea et al. 2006). Determining how a city fares in terms of the QoL of its residents is a crucial piece of information to administrators and planners charged with identifying aspects of the city that are in need of improvement (Tesfazghi et al. 2010, Lee 2008). Urban QoL assessments are one tool that may provide such information to urban leaders. These instruments can take one of two fundamental forms: objective and subjective. Objective approaches to measuring urban QoL primarily utilize secondary spatial data sources to quantify how segments of the city fare according to a select set of indicators, while the subjective approach enables residents to individually determine how they perceive certain aspects of their living environment to be (Marans 2012). This latter approach better captures the heterogeneity of perceptions that likely exists across a population (Lu 1999). Coupled together, a potential dichotomy emerges between how individuals within a population perceive their urban environment, and how this environment is deemed via objective measurements. What is limited in our state of knowledge is an understanding of how well individually collected perceptions of QoL correspond with or contrast from objective measurements of the urban environment. While there is some evidence that objectively measured indicators are weakly related to subjective responses at the individual level (Berhe et al. 2013, McCrea et al. 2006), the purpose of this paper is to take a spatial perspective to determine if there are certain locations throughout the city that exhibit a particularly high agreement or discordance between objective and subjective measurements. In particular we introduce a framework to spatially examine the cohesion between a social survey of residential perceptions and Geographic Information System (GIS)-based measurements of two fundamental QoL indicators: green space availability and access to public transit. A case study for the city of Vienna, Austria is performed. The framework further enables the analysis of population characteristics in locations where a disagreement between measurements is present. QoL in a modern scientific use is furthermore a broad-band concept encompassing material and social conditions, individual ambitions, experiences and evaluations, which often lead to confusion in the theoretical foundation and operationalization of empirical research. Originating in classical philosophy with its ,,good life‘‘-question (Aristoteles, Stoa), the concept is linked with Bentham’s utilitarism underlining individual decisions leading to joyful consequences. Social welfare is either postulated as a measurable human need under extrinsic forces or, in contrary, as result of immeasurable intrinsic personal utilities. Since objective conditions do not necessarily cause subjective wellbeing (Veenhoven 2000), which is indicated by a wellbeing paradox and a dissatisfaction dilemma, economy and politics had to learn from social sciences that subjective wellbeing implies a cognitive processing of emotions, i.e. is itself a meta-construct with different dimensions, and will give different results according to its moderating reference level, adaption, social comparison etc. For the empirical study of green space availability and access to public transit this has three main consequences: (a) Subjective QoL assessment about green space proximity in the city may not simply correlate with objective green QoL indicators from a GIS database. (b) Likewise, subjective QoL assessment about public transport quality in the city may not simply correlate with objective public transport indicators from a GIS database. (c) With the expectations of subjective mis-match effects in (a) and (b), the relation of green space and public transport may be dependent on socio-demographic
123
Comparing Subjective and Objective Quality of Life Criteria
moderating variables, or even still undetected hidden variables. This makes the empirical analysis not a simple procedure, but requires multidimensional thinking taking into account domain-specific differences. The research questions addressed in this paper are: (1) How is the general perception of the public transport connection and the proximity to green spaces in the city of Vienna? (2) Does satisfaction with public transport correspond with actual public transport quality, analyzed in terms of travel time and service frequency? (3) Does satisfaction with proximity to public green spaces agree with the measured green shares in a neighborhood? (4) Are there discernible socio-demographic characteristics of neighborhoods that contain a ‘mis-match’ between perceived and measured scores? The remaining structure of the paper is as follows: in Sect. 2, a brief review of the literature is performed followed by a description of the study area, data, and methodology in Sect. 3. Results are presented in Sect. 4 and conclusions are summarized in Sect. 5.
2 Background In terms of urban QoL, subjective indicators are often derived from surveys and represent residential perceptions, whereas objective indicators are mostly derived from evident facts or secondary data, e.g. demographic or economic data (Tesfazghi et al. 2010). Objective indicators, in more detail, can refer to environmental or external conditions typically measured at some aggregate spatial scale (Lee 2008, Lotfi and Kooshari 2009, Berhe et al. 2013). These may include variables such as population or housing density, crime rates, education quality and availability, mobility potential, healthcare options, recreational opportunities, green space, pollution, amongst many other factors. Subjective indicators are typically assessed on a Likert-scale and cover similar topics as the aforementioned objective indicators. As compared to objective variables, subjective indicators are often critiqued as being incomparable, unstable, unintelligible, and are often not related to objective perceptions or given facts (Santos et al. 2007). They are thus less often used by urban leaders in helping to shape policy. The use of GIS for the collection of objective indicators dates back to 1997 when Lo and Faber stated, that computer-assisted approaches are best to analyze urban structures, since the urban landscape can thus be characterized in more detail than with census data alone. This topic is also tackled by Martı´nez (2009), who notes, that GIS can be utilized to collect and analyze urban indicators, which in turn can support the observation and recording of inequalities and deprived areas, define priorities, and transfer resources. He further states, that GIS-based indicators are applicable to define priority areas for policy makers to target where first and how much to invest. The use of these indicators can assist the identification of areas with a concentration of needs. Other GIS-based research on the assessment of objective indicators has also incorporated remote sensing or satellite data (Lo and Faber 1997; Li and Weng 2007); has sought to determine an optimal, or parsimonious selection of variables (Galster et al. 2005); or has mapped the intra-urban spatial distribution of QoL including its change over time and the optimal weighting of indicators (Delmelle et al. 2013; Saitluanga 2014; Rinner 2007). Several studies have developed integrated approaches to looking at QoL from both an objective and subjective perspective. For example, in one such study undertaken in the city of Salzburg, Austria, residential satisfaction surveys were linked with a GIS analysis to uncover three relevant areas that influenced QoL: (1) QoL and satisfaction with one’s apartment rises significantly with his or her living duration, (2) QoL is strongly related to the satisfaction with
123
E. Haslauer et al.
one’s own apartment, green spaces, neighborhood, security, and proximity to public transport, and (3) QoL correlates negatively with population density (Keul and Prinz 2011). This project was strongly based on the Detroit Area Studies (DAS) that started as a student survey practicum in 1952 at the University of Michigan, was repeated regularly, and finally went international, triggered by Inglehart’s World Values Survey (Marans and Kweon 2011). For the DAS 2001 survey interviews were conducted with 315 adults, and a shorter version of the interview questionnaire was mailed to an adult sample throughout the region. The mail survey yielded 4,077 (56.4 %) responses. DAS 2001 viewed overall QoL as a subjective phenomenon and a composite of people’s assessments of various domains of their lives—the individual’s family life, health, job, friends, overall standard of living, the use of leisure time, and the amount of time to do the things you want to do. A final question about satisfaction with life as a whole was also asked as summary measure. Evaluations of the DAS showed moderate but significant relationships between people’s feeling about their home, neighborhood and community and QoL in both settings. Satisfaction with the home is the strongest predictor of QoL, community satisfaction the least important. In addition to crime and physical stressors (noise, crowding, excessive traffic, poor maintenance), negative ratings of services also contributed to neighborhood dissatisfaction. For the region as a whole, less than a third of the respondents evaluated public transportation serving their neighborhoods favorable while another third evaluated it negatively. As stated previously, this paper analyzes two fundamental components of the multidimensional QoL construct: availability of green space and access to public transit. Previously literature has underscored the importance of both of these dimensions in contributing to urban QoL (Garhammer 2008; Osius et al. 2001; Deutsche Umwelthilfe 2004; Prinz et al. 2007; Berhe et al. 2013). There is a significant body of literature that has pointed to the positive impact of urban green space and contact with nature on the direct physical and psychological health of residents (Kaplan 1985, 2001; de Vries et al. 2003; Ulrich et al. 1991). Common green spaces further encourage the social integration of residents (Kweon et al. 1998). Thus, it is the purview of many cities to provide residents green spaces and thus understanding where green space availability is both lacking and where residents perceive their green space access to be poor are critical in properly planning for this amenity. The second dimension, public transit access is crucial to the QoL of urban residents on several fronts. Perhaps most importantly, public transportation holds the potential to link all residents to employment and services throughout a city. The notion of social exclusion, a multidimensional construct relating to the inability of individuals to fully participate in society, has an important transportation component. Those without the means for a private vehicle necessarily rely on public transport options to provide mobility (Kenyon et al. 2002) and thus, cities strive to provide an equality level of service to all urban residents, both to and from public transportation. A disparity between perceived and measured access may emerge as individuals have personal mobility constrains that prevents them from walking a certain distance to reach a stop, for example.
3 Methodology and Data 3.1 Methodology In order to identify potential mismatches between self-reported/subjective QoL and GISbased analyzed/objective measures of the aforementioned dimensions, an integrated analysis is performed that blends qualitative and quantitative responses (Fig. 1).
123
Comparing Subjective and Objective Quality of Life Criteria
The paper at hand analyzes the distribution of green spaces, the use of public transportation modes, the satisfaction with the proximity to green areas and with the public transport connection, and the travel time as such within the city. 3.2 Subjective Assessment The data source for the assessment of subjective indicators is a social-scientific survey regarding ‘‘Life and QoL in Vienna’’ authored by the University of Vienna and the City of Vienna in 2008. The survey incorporates 135 questions covering seven general QoL dimensions: Living, Education, Work and Employment, Security, Health, Mobility, and Participation. The survey is part of a larger policy initiative by the Vienna City Government to help identify pending challenges for the city. Interviews were conducted with residents aged 15 years or older via the telephone. Interviewees were chosen by Random Digit Dialing, and included also cell phone numbers. The average time of an interview was 40–45 min. Due to the large number of participants some of the questions were split, i.e. only half of the participants were asked. Nevertheless, all samples are representatively distributed over the whole sample area. The interviews were mainly conducted in neighborhoods of the core area of the city, but also in some outer neighborhoods. The interviewers talked to 8,400 individuals on the phone in German language, and did native face-to-face interviews with 300 individuals who included immigrants from Turkey and the former Yugoslavia. They were asked the whole set of questions, i.e. the questions were not split. The results refer to the residential population since the survey is based on a personal sample and is not referring to households. The final sample set was weighted according to district, age, gender, education, and form of living. The samples were not weighted by nationality as it was aimed to produce a representative sample regarding the aforementioned criteria. Regarding the districts, the sampling was disproportional in order to get representative results that included smaller districts. This disproportionality was then removed again during evaluation, so the results are representative for the whole sample area. It has to be noted that the population with non-Austrian citizenship is underrepresented since the survey was largely conducted in German language. The response rate was 61.8 % of all contacted phone numbers. For this particular paper, the two primary subjectively answered responses incorporated into the analysis are: ‘‘How satisfied are you with your access to green space’’ and ‘‘How satisfied are you with the public transit?’’. The residential evaluation was done on a Likert Scale from 1 (very satisfied) to 5 (unsatisfied). Three additional questions were also subsequently analyzed including: ‘‘Would the additional construction of green space enhance your QoL?’’, ‘‘Which kind of transportation mode do you use?’’, and ‘‘Would a better public transport connection enhance your QoL?’’. 3.3 Objective Indicators For all analyses, the entire city is divided into equal-area cells of 250 m 9 250 m (vector format). This cell size was chosen to match the 2009 population distribution dataset for the city of Vienna. In order to compare the subjective responses mentioned above, objective measurements of urban green and recreation space is derived from a multi-purpose land-use map based off of 2007 aerial imagery. This dataset has a resolution of 500 m and contains records on forests, tree-covered areas, pastures,
123
E. Haslauer et al.
Fig. 1 Integrative approach
nearly-natural green areas, fields, arable land, natural and artificial water bodies (public); the dataset was resampled to a resolution of 250 m to match the analysis extent of this study. It has to be noted, that this was done just to allow for a comparison of the population dataset and the land-use map. The authors want to mention particularly that there could exist a misleading of land-use information due to the resampling of the original land-use map (which had a resolution of 500 m 9 500 m) to a resolution of 250 m 9 250 m. The first indicator derived from this source captures the share of green spaces (also including water bodies) in each cell. Given the city’s radial structure, with a dense urban core, green space precipitously declines away from the center city. In all, the measured, average green space across the city for each cell is approximately 66 % coverage, with maximal green space coverage of approximately 90 % appearing in three of the outermost districts, and the lowest coverage (approximately 20 % green) in the most central districts. In order to objectively estimate our second QoL dimension, quality of public transportation, two metrics are utilized: average travel time throughout the city and the average number of boardings per minute for all modes. For the first measure, service is assessed by calculating the travel time from each cell to seven a priori defined stops. The travel time consists of walking time, waiting time, and driving time and is calculated in minutes. To calculate the averaged travel time per cell, the authors looked at the ten nearest public transport stops around each neighborhood block. For each stop the sum of walking, waiting and driving time was calculated. The minimum of these values specified the ‘‘most suitable’’ public transport stop in terms of travel time. This stop was selected to represent the shortest averaged travel time from a neighborhood to the seven centers. The values were averaged for raster cells afterwards. The defined stops, serving as start—respectively endpoints for the travel time calculation are: Kagran, Stephansplatz, Su¨dtirolerplatz, Wien Westbahnhof, Mariahilferstrasse, Wien Mitte, and Floridsdorf. The districts of Vienna and the seven stops are shown in Fig. 2.
123
Comparing Subjective and Objective Quality of Life Criteria
The bus and subway stops represent the most important public transport stops of seven important functional centers in the city area regarding leisure facilities, working places, etc. The definition and selection of these stops was done by the city government of Vienna a priori and not by the authors themselves. The datasets were spatially analyzed using ESRI Arc Map 10/10.1. For each interview there was also given information about the neighborhood block in which the interviewee was living at the time when the survey was conducted. So the answers given in the interviews could be geo-referenced by first aggregating them to neighborhood blocks (weighted mean) and afterwards to equal-area cells. Results per cell therefore represent the mean weighted values of all given responses regarding travel times and departures. To analyze green spaces and public transport travel time, the city area was classified into four zones of green shares, according to their objective measurement results: zone 1 (0–20 %), zone 2 (20.1–50 %), zone 3 (50.1–80 %), and zone 4 (80.1–100 %), and into four zones of travel time: zone 1 (0–20 min.), zone 2 (21–30 min.), zone 3 (31–40 min.), and zone 4 (41–130 min.). Finally, to analyze socio-demographic characteristics of neighborhoods, data provided by the City Government of Vienna includes age-classes and countries of birth of the residential population in 2009.
4 Results and Discussion 4.1 Satisfaction with Green Spaces The agreement between the aggregated subjective responses and the four calculated zones of green spaces described in the previous section is displayed in Fig. 3. The figure shows that the highest levels of reported green space satisfaction (1 on the Likert Scale) are in Zone 4; the zone with the highest calculated green space. Individual satisfaction then gradually declines with declining shares of green space coverage as computed by the GIS. Thus, a high degree of agreement exists in the initial comparison of indicators. This is further reiterated in the Table 1 which indicates the weighted average of responses for each of the four zones. Residents were also asked if the construction of additional green spaces would enhance the quality of their lives. These responses are similarly linked to the four zones. Again, responses are very consistent with the four green space zones; in zone 1, half of the respondents answered affirmatively to the query, while less than one tenth of respondents in zone 4 answered yes to the same question. Table 2 lists the respondent proportions for each of the four zones. 4.2 Satisfaction with Public Transit In terms of public transit satisfaction, some basic modal split statistics are first examined (Table 3). In Vienna, nearly 75 % of queried residents have access to a personal vehicle, but nearly 75 still make use of public transport several times a week. Public transport is particularly important when commuting to the central city given the scarcity of parking spaces. Although of all people that were asked have a car available in their household, also of all people make use of public transport several times a week. Above all, public transport is especially popular when driving to work or to the inner city. In the survey, the following
123
E. Haslauer et al.
Fig. 2 The 23 districts of Vienna and the seven centres for calculating the travel time
Fig. 3 Grades of satisfaction concerning access to green space in the four zones of green shares
answers were given to the question about the transport mode people use several times a week (Table 3): To analyze the average travel time calculated from the previously mentioned seven stops within the city to each cell in the city-area, the area was divided into zones of averaged travel time (Fig. 4). The inner city is entirely located in zone 1. This zone extends to the north across the river Danube into the 22nd district. Zone 2 borders zone 1 and
123
Comparing Subjective and Objective Quality of Life Criteria Table 1 Weighted mean grading of satisfaction with the access to green spaces Zone (%)
Weighted number of people asked
Weighted mean grading
SD
B20
2,272
2.18
1.17
20.1–50
2,644
1.65
0.93
50.1–80
2,536
1.34
0.71
294
1.25
0.66
80.1–100
reaches into all other districts. Zone 3 borders zone 2 and covers a large area from the northern border of the city to the southern border. Zone 4 covers the fringe districts in the east and west. Results of the survey question regarding satisfaction with public transit in Vienna are subsequently analyzed in companion with the travel time map. Together, the results suggest that the public is overwhelmingly supportive of the transit system regardless of their mean travel time though satisfaction does diminish slightly from the central city to the outer-most zones (Table 4). The distribution of satisfaction scores according to the four zones is displayed in Fig. 5. As can be seen, the highest satisfaction exists in zone 2, while the highest dissatisfaction occurs in the zones with the highest average travel times. Interestingly, zone 1, the most central locations in the city, shows a large degree of variability in responses whereas in zone 3 no significant trend can be recognized, and finally in zone 4, dissatisfaction with transit is the majority response. Turning now to a second measured assessment of public transport, the service frequency per zone, or the average departures per minute for a 20 operating hour duration for three transport modes: bus, train, and subway (Table 5). In this case, locations with the fewest departures per minute (0.13 and 0.17) do not necessarily correspond with negative transit perceptions. In fact, locations with the fewest departures per minute (and long travel times C45 min.) had high average satisfaction scores (\1.5) and areas with the highest average transport boardings (0.44) and low travel times (B30 min.) had poor satisfaction scores ([3.5). Thus, some inconsistency exists between how public transport is measured and corresponding satisfaction scores. Finally, consistent with the first results, responses to the query on whether or not better transport would enhance QoL, ‘‘yes’’ responses dominate the outermost zones, while only 8 % of residents in the first zone responded positively (Table 6). A synthesis of these two QoL dimensions is performed by analyzing the share of green spaces within each travel time zone, and then, segments of the city that score highly in both dimensions, in one, but not the other, or are low in both dimension are summarized in Table 7. The indicator ‘‘Green spaces related to public transport travel time’’ provides information about the travel time to green spaces. The following areas are distinguished: areas of short travel time and high proportions of green space (??), areas with long travel time and low proportions (-), areas of short travel time and low proportions (?-), and areas of long travel time and high proportions of green spaces (-?). Within both the inner belt and the inner part of the 2nd district are the lowest average public transport time and proportion of green space (?-). Small areas of short travel time and high proportions of green spaces (??) can be found in the area between the old and the new part of the Danube River; these relatively few areas thus
123
E. Haslauer et al. Table 2 Responses to: ‘‘Would the additional construction of green spaces enhance your quality of life?‘‘ Zone (%)
Weighted number of people asked
Proportion of ‘‘yes’’-answers to: ‘‘Would additional construction of green spaces enhance your quality of life?’’ (%)
B20
2,272
50
20.1–50
2,644
33
50.1–80
2,536
16
294
8
80.1–100
score highly in both QoL dimensions. Small shares of green space corresponding with long public transport travel times (-) are found in the south of the 23rd district and in the west of the 22nd district. And finally, long travel times coupled with high shares of green space (-?) are found in the areas furthermost in the east and west. Most of the 23 Viennese districts have an averaged share of green spaces of 26–75 % and an averaged public transport travel time to seven stops of 31–60 min. Only four districts have a very short averaged public transport travel time—below 30 min—but they are also combined with a low share of green space. The areas with the highest shares of green spaces per cell also, expectedly, have with the longest travel times ([60 min.) given their distance from the urban center where development is less dense. In these areas however, residential satisfaction scores were slightly higher for the public transit dimension (1.3 average score), than with the green space satisfaction score (1.5 average). 4.3 Spatial Discrepancies in Objective and Subjective Indicators While overall, there is much agreement between subjectively reported satisfaction with green space and the objectively derived assessment of this indicator, we can identify some areas of the city where these two indicators diverge. This is mapped in Fig. 6; areas where reported satisfaction and the measured indicator are in agreement are colored in black and white, while disagreement is registered in the white cell with dots and the colored white and black bands cells. In the first type of cell, a discordance exists between high satisfaction with low (B20 %) measured green space, and in the second type, the opposite is true: satisfaction scores are low (an average of 3.5 or worse), while average green shares are more than 80 %. Contrary to initial expectations (see consequence a in the Introduction section), the subjective green QoL assessment measured via ‘‘satisfaction with access to green spaces’’ varies proportionally to the GIS shares of adjacent green areas. People are very well aware about abundance or lack of green space and consequently wish to have more green in ill-supported urban environments. Subjective QoL and objective transport reality do not show polarization (see consequence b), but a pragmatic overlap of daily public transport and car use. Public transport is used independently from travel time, but personal public transport satisfaction drops from the inner city core to outer zones. People are logical in their evaluation whether public transport improvement would enhance their QoL. To further probe some of the characteristics of areas with a high discrepancy, sociodemographic characteristics of these cells is examined in relation to birth country of origin and age distribution. We find that areas with poor satisfaction, but highly measured areas
123
Comparing Subjective and Objective Quality of Life Criteria Table 3 Modal split (more than one answer selectable)
Transportation mode used several times a week
Share (%)
Public transport
74
Car: driver
39
Car: passenger
13
Fig. 4 Visualization of four travel time zones in Vienna
Table 4 Weighted mean of the satisfaction with public transit in Vienna Zone (min) B20
Weighted number of people asked
Weighted mean grading
SD
979
1.21
1.79
21–30
4,206
1.33
0.66
31–40
2,216
1.6
0.92
2.17
1.17
41–130
825
have an older population (ages 41–55) as compared to the population living in areas with high satisfaction and low score cells (age 26–40), though their ethnic makeup is quite similar (Table 8).
123
E. Haslauer et al.
Fig. 5 Grades of satisfaction concerning public transit in the four zones of public transport travel time
Table 5 Service frequencies per zone Zone characteristics
Mean service frequency (departures/min)a
Minimum value (departures/min)
Maximum value (departures/min)
Satisfaction B1.5, avg. travel time B30
0.47
0.14
1.57
Satisfaction B1.5, avg. travel time 31–45
0.35
0.06
1.2
Satisfaction B1.5, avg. travel time [45
0.17
0.01
0.57
Satisfaction 1.5–3.5, avg. travel time B30
0.47
0.14
1.57
Satisfaction 1.5–3.5, avg. travel time 31–45
0.33
0.06
1.2
Satisfaction 1.5–3.5, avg. travel time [45
0.13
0.02
0.43
Satisfaction [3.5, avg. travel time B30
0.44
0.15
1.42
Satisfaction [3.5, avg. travel time 31–45
0.21
0.11
0.57
Satisfaction [3.5, avg. travel time [45
0.29
0.38
0.18
a
Averaged for 20 working hours
Table 6 Answers given to: ‘‘Would a better public transport connection enhance your quality of life?’’ Zone (min)
B20
Weighted number of people asked
Share of ‘‘yes’’-answers to: ‘‘Would a better public transport connection enhance your QoL?’’ (%)
979
8
21–30
4,206
13
31–40
2,216
30
825
48
41–130
123
Comparing Subjective and Objective Quality of Life Criteria Table 7 Green spaces related to public transport travel time Zone
Share of green spaces B25 %
Share of green spaces 26–75 %
Share of green spaces [75 %
Travel time B30 min
1st, 5th, 6th, 7th, 8th district
4th, 9th, 12th, 15th district
–
Travel time 31–60 min
–
2nd, 3rd, 6th, 10th, 11th, 20th, 21st district
16th, 17th, 18th, 19th district
Travel time [60 min
–
–
13th, 14th, 22nd, 23rd district
Fig. 6 Visualization of satisfaction with proximity to green spaces and actual green shares per raster-cell
The same analysis is now done for the public transport dimension. Figure 7 is a map illustrating the spatial discrepancies between the objective and subjective indicators. These datasets were again analyzed in relation to socio-demographic data of age-classes and countries of birth. The dominating age-classes and countries of birth are shown in Table 9. It should be noted that in both this case and the previous example only those cells are considered in which inhabitants were asked about their satisfaction with public transport (3,407 out of 6,957 cells). In this case, again locations with a higher dominating age cohort exhibit greater dissatisfaction when in fact the measured variable is relatively good. For public transport, this discordance exists in areas with a much older population ([65 years old) which would make intuitive sense as mobility declines with age.
123
E. Haslauer et al. Table 8 Characteristics of zones Zone characteristics
Inhabitants in relevant cells*
Dominating age-class
3 dominating countries of birtha
Satisfaction B1.5, avg. green share B20 %
237,567
26–40
1. Austria, 2.Other EU-countries, 3.Western Balkan countries
Satisfaction B1.5, avg. green share 20.1–80 %
421,913
41–55
1. Austria, 2.Other EU-countries, 3.Western Balkan
Satisfaction B1.5, avg. green share [80 %
5,681
41–55
1. Austria, 2.Other EU countries, 3.Other countries
Satisfaction 1.5–3.5, avg. green share B20 %
413,930
26–40
1. Austria, 2.Western Balkan, 3.Other EU countries
Satisfaction 1.5–3.5, avg. green share 20.1–80 %
463,854
26–40
1. Austria, 2.Other EU-countries, 3.Western Balkan
Satisfaction 1.5–3.5, avg. green share [80 %
13,144
41–55
1. Austria, 2.Other EU countries, 3.Other countries
Satisfaction [3.5, avg. green share B20 %
4,124
26–40
1. Austria, 2.Western Balkan, 3.Other EU-countries
Satisfaction [3.5, avg. green share 20.1–80 %
49,059
41–55
1. Austria, 2.Other EU-countries, 3.Western Balkan
Satisfaction [3.5, avg. green share [80 %
795
41–55
1. Austria, 2.Other EU countries, 3.Other countries
* i.e. cells, where interviews were conducted a
For reference: the order of country of birth for all residents of Vienna is as follows: 1. Austria, 2. Other European countries, 3. Western Balkan, 4. Turkey, 5. Other countries, 6. Asia, 7. Africa, 8. Near and Middle East, 9. Other European countries
5 Conclusion This paper provided an assessment and analytical framework for spatially identifying areas of agreement and discordance between objectively and subjectively measured QoL indicators. The case of two fundamental urban indicators, public transit quality and green space availability for the city of Vienna was investigated. Subjective responses to a QoL survey were geocoded and compared to objective measurements obtained via a geographic information system. For the latter, green space was obtained from classified areal imagery and summarized for 250 m pixels throughout the city and public transport quality was assess through two measurements: the average travel time to seven destinations throughout the city and the average number of boardings for all modes of public transport. In regards to green space perception and measurement, on average, residential population of green space throughout the city garnered an average score of 1.72 (with 1 being the most satisfied option, and 5 being the least). To compare the subjective and objective responses, the city was divided into zones of 0–20, 20.1–50, 50.1–80 and 80.1–100 % green shares. Exhibiting a large degree of accordance, the highest satisfaction scores were most prevalent in areas with the highest concentration of green space, while satisfaction scores progressively declined with falling shares of measured green space. Thus, for this indicator, we found an overall high degree of agreement between the objective and subjective. However, upon closer inspection of potential spatial mis-matches, in small areas throughout the city, locations of high green shares were aligned with low average satisfaction ratings, notably in the northern most part of the city, while the city center, which
123
Comparing Subjective and Objective Quality of Life Criteria
Fig. 7 Visualization of perceived public transport connection and actual travel time
Table 9 Characteristics of zones of travel time Zone characteristics
Inhabitants in relevant cells*
Dominating age-class
3 dominating countries of birth
Satisfaction B1.5, avg. travel time B30
771,480
26–40
1. Austria, 2. Western Balkan countries, 3.Other EU countries
Satisfaction B1.5, avg. travel time 31–45
332,381
41–55
1. Austria, 2.Other EU-countries, 3.Western Balkan
Satisfaction B1.5, avg. travel time C45
40,972
41–55
1. Austria, 2.Other EU countries, 3.Other countries
Satisfaction 1.5–3.5, avg. travel time B30
360,000
26–40
1. Austria, 2.Other EU-countries, 3.Western Balkan
Satisfaction 1.5–3.5, avg. travel time 31–45
99,201
41–55
1. Austria, 2.Other EU-countries, 3.Western Balkan
Satisfaction 1.5–3.5, avg. travel time C45
8,294
41–55
1. Austria, 2.Other EU countries, 3.Other countries
Satisfaction [3.5, avg. travel time B30
7,493
[65
1. Austria, 2.Western Balkan, 3.Other EU-countries
Satisfaction [3.5, avg. travel time 31–45
2,736
[65
1. Austria, 2.Other EU-countries, 3.Western Balkan
Satisfaction [3.5, avg. travel time C45
576
[65
1. Austria, 2.Western Balkan, 3.Other EU-countries
* i.e. cells, where interviews were conducted
123
E. Haslauer et al.
has a measurably small green share percentage, registered extremely positive perceptions. Such a discrepancy might be attributable to self-selection realities; those wishing to reside in a more urban setting may be quite satisfied with less green space as urban amenities are more important to these residents, whereas those residing on the outskirts of the city may have higher expectations for green space. In terms of socio-demographic characteristics of neighborhoods possessing a mis-match between perceived and measured green space, residents in a younger age cohort, 26–40 years old, tended to represent those living in neighborhoods where green space was measured to be low, but satisfaction was ranked highly. This could be a reflection of younger residents living in the more urban areas and placing a premium on urban amenities rather than green space. This furthermore seems to imply a higher cognitive flexibility of younger residents whereas older city residents exhibit a more critical tendency, even in ample green areas since a slightly older age group, 41–55 years old, characterized the opposite discrepancy. For the transport dimension, the analysis similarly revealed declining satisfaction scores with rising travel times throughout the city. Spatial discrepancies according to this dimension existed in isolated cells in the inner city and in the outer fringe of the city region. The socio-demographic results showed that elderly residents tended to evaluate public transport more poorly than the measured results indicated, while those aged 41–55 gave more positive assessments in areas with worse measurements. Habituation effects may work here—service frequency is learned and only delays cause emotional reactions (not analyzed here). Summing up the results of this representative, but still item-selective survey, it can be said that Vienna residents guess their urban green share QoL quite accurately and public transport is attractive enough to be widely used although on the outskirts with lower satisfaction. The results of this study underscore that the needs and perceptions of urban residents are not spatially uniform throughout a city. Further analysis about gender, education, and household/family composition as mediators may be interesting to follow. Policies aimed at enhancing the QoL of residents must consider the micro-scale intraurban variations in these responses to adequately meet the needs of residents. The analytical framework put forth in this paper enables potential spatial mismatches between subjective and objective indicators to be identified. Future research should evaluate other QoL dimensions, and could incorporate a wider range of socio-economic characteristics of residents in ‘mis-matched’ areas. A confirmatory statistical approach could also help isolate statically significant characteristics of the population in these areas. Finally, similar types of analyses could be performed in other geographic to get a coherent picture of similarities as well as contrasts in city structures. Acknowledgments The authors want to thank Mr. Helmut Augustin and Mr. Rainer Hauswirth from the City Government of Vienna, Department of City Planning (Magistrat Abteilung 18) for providing us with the survey and socio-demographic data of Vienna.
References Berhe, R. T., Martinez, J., & Verplanke, J. (2013). Adaption and dissonance in quality of life: A case study in mekelle, Ethiopia. Social Indicators Research, 114, 1–20. De Vries, R. A., Verheij, P. P., Groenewegen, P., & Spreeuwenberg, P. (2003). Natural environmentshealthy environments? An exploratory analysis of the relationship between greenspace and health. Environment and Planning A, 35, 1717–1731.
123
Comparing Subjective and Objective Quality of Life Criteria Delmelle, E., Thill, J.-C., Furuseth, O., & Ludden, T. (2013). Trajectories of multidimensional neighborhood quality of life change. Urban Studies, 50, 923–941. Deutsche Umwelthilfe (2004) Nachhaltigkeitsindikatoren – Leitkategorie ‘‘Wohlbefinden’’ (Sustainability indicators—category ‘‘Wellbeing‘‘). http://www.duh.de/251.html. Accessed 15 May 15 2013. Galster, G., Hayes, C., & Johnson, J. (2005). Identifying robust, parsimonious neighborhood indicators. Journal of Planning Education and Research, 24, 265–280. Garhammer, M. (2008) City ranking and quality of life in European cities—10 theses and an indicator system for monitoring integrated Urban development. http://www.opus-bayern.de/ohm-hochschule/ volltexte/2008/15/. Accessed 12 April 2013. Kaplan, R. (1985). Nature at the doorstep. Residential statisfaction and the nearby environment. Journal of Architect Planning Research, 2, 115–127. Kaplan, R. (2001). The nature of the view from home. Psychological benefits. Environment and Behavior, 33, 507–542. Kenyon, S., Lyons, G., & Rafferty, J. (2002). Transport and social exclusion: investigating the possibility of promoting inclusion through virtual mobility. Journal of Transport Geography, 10, 207–219. Keul, A. G., & Prinz, T. (2011). The Salzburg quality of urban life study with GIS support. In R. W. Marans & R. Stimson (Eds.), Investigating quality of urban life: Theory, methods, and empirical research (pp. 273–293). Dordrecht: Springer. Kweon, B.-S., Sullivan, W., & Wiley, A. (1998). Green common spaces and the social integration of innercity older adults. Environment and Behavior, 30, 832–858. Lee, Y. J. (2008). Subjective quality of life measurement in Taipei. Building and Environment, 43, 1205–1215. Li, G., & Weng, Q. (2007). Measuring the quality of life in the city of Indianapolis by integration of remote sensing and census data. International Journal of Remote Sensing, 28(2), 249–267. Lo, C. P., & Faber, B. J. (1997). Integration of landsat thematic mapper and census data for quality of life assessment. Remote Sensing Environment, 62, 143–157. Lotfi, S., & Kooshari, M. J. (2009). Analyzing accessibility dimension of urban quality of life: Where urban designers face duality between subjective and objective reading of place. Social Indicators Research, 94, 417–435. Lu, M. (1999). Determinants of residential satisfaction: Ordered logit vs. Regression models. Growth and Change, 30, 264–287. Marans, R. W. (2012). Quality of urban life studies: An overview and implications for environmentbehaviour research. Procedia—Social and Behavioral Sciences, 35, 9–22. Marans, R. W., & Kweon, B. S. (2011). The Quality of Life in Metro Detroit at the beginning of the millennium. In R. W. Marans & R. J. Stimson (Eds.), Investigating quality of urban life (pp. 163–184). Dordrecht: Springer. Martı´nez, J. (2009). The use of GIS and indicators to monitor intra-urban inequalities. A case study in Rosario, Argentina. Habitat International, 33, 387–396. McCrea, R., Shyy, T. K., & Stimson, R. (2006). What is the strength of the link between objective and subjective indicators of urban quality of life? Applied Research in Quality of Life, 1, 79–96. Osius, N., Trojan, A., Flesch-Janys, D., Berger, J., Kappos A. D. (2001). Indikatoren fu¨r eine nachhaltige und gesunde Stadtentwicklung (Indicators for a sustainable and sound urban development).-Documentation of a workshop in the project ‘‘Nachhaltige Entwicklung in Metropolregionen’’ (Sustainable development in metropolitan regions). University of Hamburg, Germany. Prinz, T., Keul, A., & Lang, S. (2007). Ra¨umlich differenzierte Lebensqualita¨t—GIS-gestu¨tzte Analyze von 293 QOL Interviews in der Stadt Salzburg (Spatially explicit Quality of Life—GIS based analysis of 293 interviews in the city of Salzburg). In J. Strobl, T. Blaschke, & G. Griesebner (Eds.), Angewandte geoinformatik 2007 (pp. 589–594). Heidelberg: Wichmann Verlag. Rinner, C. (2007). A geographic visualization approach to multi-criteria evaluation of urban quality of life. International Journal of Geographical Information Science, 21, 907–919. Saitluanga, B. (2014). Spatial pattern of urban livability in Himalayan Region: A case of Aizawl City, India. Social Indicators Research, 117, 541–559. Santos, L. D., Martins, I., & Brito, P. (2007). Measuring subjective quality of life: A survey to porto’s residents. Applied Research in Quality of Life, 2, 51–64. Tesfazghi, E. S., Martinez, J. A., & Verplanke, J. V. (2010). Variability of Quality of Life at Small Scales: Addis Ababa, Kirkos Sub-City. Social Indicators Research, 98, 73–88. Ulrich, R. S., Simons, B. D., Losito, E., Fiorito, M. A., Miles, M. A., & Zelson, M. (1991). Stress recovery during exposure to natural and urban environments. Journal of Environmental Psychology, 11, 201–230. Veenhoven, R. (2000). The four qualities of life.- Ordering concepts and measures of the good life. Journal of Happiness Studies, 1, 1–39.
123
