The Dictionary of Human Geography (198 page)

Read The Dictionary of Human Geography Online

Authors: Michael Watts

BOOK: The Dictionary of Human Geography
12.2Mb size Format: txt, pdf, ePub
survey analysis
The various procedures involved in the collection and analysis of data from individuals, almost invariably using some sort of Questionnaire. As such they are a type of extensive research design. (NEW PARAGRAPH) A survey involves several stages. The first is definition of the research problem, including the formulation of hypotheses and identifica tion of the needed information. The second includes determining the population to be studied, which includes deciding what form of sampLing is needed. The next stage involves deciding how the hypotheses will be tested (including the analytical techniques to be employed), and is followed by development of a questionnaire (which should include pre test stages and pilot investigations). (NEW PARAGRAPH) After administration of the questionnaires the data are prepared for analysis: quantitative data are readily dealt with; qualitative infor mation (such as reported occupations and responses to open ended questions) has to be handled through the development of coding schemes, increasingly through the use of sophisticated computer software for textual analysis. To ensure statistically reliable results, it is unusual for the number of separate cat egorical codes (such as social class) to exceed ten, and is typically five or under. The data are then usually entered into a computer database and checked for consistency and gross errors (?cleaning? the data set) before the analyses are conducted, although increasingly the responses are entered directly to a computer by the interviewer (whether at a face to face interview or in an interview by telephone). (NEW PARAGRAPH) The analysis of surveys consists of four quantitative elements that need to be under taken simultaneously: (NEW PARAGRAPH) Evaluating the size effects between vari ables and by doing so taking account of other variables what Rosenberg (1978) calls the logic of survey methodology; (NEW PARAGRAPH) Testing as part of confirmatory data analy sis whether the observed effects could have occurred by chance, or whether the lack of an effect is due to inadequate statistical power; (NEW PARAGRAPH) Taking account quantitatively of the com plex design of the survey; and (NEW PARAGRAPH) Dealing with non response and missing data. (NEW PARAGRAPH) To appreciate the issues concerning size of effects and the interplay between variables, consider the following 2 x 2 table in which there is a single outcome (happiness) and explanatory variable (age), each being meas ure on a binary scale: The odds ratio calculated as (A/B)/(C/D) gives the degree of association between the vari ables: if it is 1 there is no relationship; greater than 1 means that younger people are unhap pier; less than 1 suggests that younger people are more likely to be happy. However, these results should not be taken at face value and there needs to be model elaboration (Davis, 1986) to see how the relation changes as account is taken of other variables. This can be clearly seen by examining the following set of 2 x 2 tables: In (a) there is a clear relationship in that the odds ratio of 2.1 suggest that younger people are twice as unhappy as the old. However, in (b) and (c) when the same data are disaggre gated to examine the relation for men and women separately, each odds ratio is exactly (NEW PARAGRAPH) 1 showing that there is no relationship between age and happiness; the apparent relationship is an artefact of the relation ships between age and gender (i.e. they are confounded) and gender and happiness. In (a) the aggregate relation shows no effect as there is an odds ratio of 1, but in (b) and (c) when the analysis is done separately by gender, the odds ratios are 1.79 for males and 2.25 for females. In this case, the effect between age and happiness has been masked by not taking account of gender. The analysis can of course be extended to more than three variables and to include continuous variables (see categorical data analysis) but the underling logic of model elaboration remains the same. (NEW PARAGRAPH) Analysis also has to guard against the Type I error of finding a relation when it does not exist, and the Type II error of not finding a genuine relation (cf. sampling). For the for mer confirmatory data analysis is used to test for the significance of a relation and to exam ine confidence intervals; the key here is the absolute sample size. Thus the estimated odds of 2.1 for the relation in the first aggregate analysis has 95 per cent confidence intervals that lie between 1.38 and 3.21, so there is little chance that the aggregate relation is really 1, which would indicate no effect. When no sig nificant relationship is found this may be an outcome of too small a study. If this is the case we can perform a retrospective power analysis (Cohen, 1988) to see if the sample size was large enough to detect an effect. It would have been better, however, to do an initial power analysis before sampling. (NEW PARAGRAPH) Many surveys have a complex sampling design involving clustering, stratification and disproportionate sampling. These attributes should be taken into account during analysis so as to avoid biased estimates of standard errors and increased likelihood of Type I errors. Clustering, or multi stage selection of sample units, typically generates dependency so that there is less information than appears. muLti LeveL modeLs estimate and correct for the degree of dependency even when there are more than two sampling stages. Their results can also be substantively interesting, finding for example that members of the same house hoLd tend to vote together (Johnston, Jones, Sarker, Burgess, Propper and Bolster, 2005). Stratification involves grouping the sampling frame into believed to be homogeneous groups and thereby reducing standard error. Often there is disproportionate sampling of strata so that having grouped primary sam pling units into strata based on percentage ethnic minority population, ethnic areas are over sampled. This requires at the analysis stage that the over sampled ethnic areas are down weighted to their correct population proportion. Sturgis (2004), using data from the 2000 UK Time Use Survey, shows how these factors should be incorporated into the estimates and illustrates the threats to infer ence if ignored. LongitudinaL data anaLysis faces its own particular analytical problems arising from the analysis of repeated measures over time. (NEW PARAGRAPH) Non response is a growing problem with surveys. The best approach is to ensure detailed follow up so that the issue is minim ized at the design and collection phase; this is a problem where doing nothing is doing some thing, and where prevention is better than any cure. We can distinguish between full, or unit, non response and item non response, the latter being when the respondent has only answered some questions. For the former, differential weighting can be used to reduce bias by boost ing the effective size of subgroups (such as young men) that are under represented in the survey, but their relative size is known from other large scale surveys or censuses. There is a danger of increasing standard errors, how ever, when the variance of the weights is large. Most software for Quantitative anaLysis automatically excludes the entire respondent when any values are missing and this is known as complete case analysis. If the data are miss ing completely at random (MCAR: Rubin, 1976) this will not bias the results if the obser vations are excluded but it will reduce the effective sample size. If the data are missing at random (MAR) but the ?missingness? depends on recorded information, then com plete case analysis can be used, but the deter minants of the ?missingness? must be included in the analysis to avoid non response bias. This suggests that at collection phase, vari ables that should be easier to collect are obtained alongside those that are thought to be difficult (e.g. income may be difficult so also collect information about property value). If the ?missingness? depends on unobserved predictors, even after accounting for informa tion in the observed data, then the data are said to be not missing at random (NMAR). In this case, complete case analysis is likely to produce biased results. (NEW PARAGRAPH) There are two main approaches to ?missing ness?, either explicit modelling of the under lying mechanism generating the missing data or some form of imputation (or ?guessing?) to replace the missing values. A number of ad hoc procedures can be used for the latter, such as carrying the last observation forward, creating an extra category for the missing observation, or replacing missing observations by the mean of the variable used. All can give unpredictable results. Consequently, the only practical, gen erally applicable, approach for substantial datasets is multiple imputation whereby each missing value is replaced by several (typically less than five) imputed values which come from an imputation model which also reflects sam pling variability. A sensitivity analysis can then be undertaken to investigate the robustness of the estimates to differential ?missingness?. (NEW PARAGRAPH) Survey analysts are aware of criticisms that see quantitative approaches as imposing meaning on people?s attitudes and behaviours. Consequently, as Marsh (1982) argued in her defence of the survey method, researchers have been highly attentive to just such issues, and developments continue to be made. One set of issues relates to whether respondents, particularly from different cultures, under stand questions in different ways, or if researchers mean one thing and respondents think they mean something else. Analytical approaches to this treat survey questions as a function of the actual quantity being measured along with an element of interpersonal incom parability that is potentially different for each respondent. The new idea is to use anchoring vignettes as a common reference point (King, Murray, Salomon and Tandon, 2004; King and Wand, 2007) to measure directly and then ?subtract off? the incomparable portion. Respondents are asked for their own response (NEW PARAGRAPH) to the concept being measured along with assessments, on the same scale, for each of several hypothetical benchmark situations described in the vignettes. Interpersonal incomparability is the only reason the response can differ, as the vignettes are literally the same. Statistical models have been designed to require only a small random sub sample to correct the respondent?s reply for the personal element. kj (NEW PARAGRAPH) Suggested reading (NEW PARAGRAPH) Groves, Fowler, Couper, Lepkowski, Singer and Tourangeau (2004); King, Honaker, Joseph and Scheve (2001); Little and Rubin (2002); Skinner, Holt and Smith (1989). Excellent prac tical advice on missing data can be found at http://www.lshtm.ac.uk/msu/missingdata/index. html and the Anchoring Vignettes website is at http://gking.harvard.edu/vign/. (NEW PARAGRAPH)
surveying
To survey is to assess or study a place or population, the salient features of which might then be mapped or recorded. From a cadastral (land ownership) perspective, surveying is to apply the principles of math ematics (geometry and trigonometry) to deter mine points on the Earth?s surface delimiting a land boundary. Alternatively, a physical sci entist might apply the principles of physics, chemistry or biology to survey a site, whereas a polling company uses techniques of statistical inference to survey a population by means of a sample (see survey analysis). Geographical research includes not only mathematical/ scientific conceptions of surveying but also qualitative methods. rh (NEW PARAGRAPH) Suggested reading (NEW PARAGRAPH) Aldridge and Levine (2001). (NEW PARAGRAPH)
sustainability
Sustainability, like sustain able development, is becoming increasingly difficult to invoke with any critical weight. In fact, it would be fair to say that in critical circles, and among those interested in progres sive or radical environmental politics in par ticular, use of the word will almost certainly elicit a cringe. This may point to an abiding cynicism, but it probably also reflects the pro liferation of this term as a form of discursive gloss over disparate material and political pro jects, including no shortage of mobilizations in corporate ?greenwash? campaigns. Indeed, while sustainability as a buzzword does much work to enhance environmental awareness, it may just as easily be viewed as evidence of an increasingly promiscuous convergence of capital accumuLation and certain kinds of environmentaLism (Katz, 1998). To this may be added the concern that this word and much more so sustainable development has become hopelessly co opted by an instru mentaList and administrative connotation that takes from it any edge as a challenge to prevailing ways of thinking about and relating to one another, and to the non human world. It is in fact difficult to argue that that any serious progress is being made in the name of this term when, by any reasonable definition and notwithstanding rosy portraits of dematerializing industrial economies, the global political economy and ecology what Luke (2005) provocatively calls a system of ?sustainable degradation? is characterized by more and more aggregate material and energy throughput, and by greater and greater social inequality (Harper, 2004) (cf. poLiticaL ecoLogy). (NEW PARAGRAPH) There are nevertheless good reasons to take this word and some of what it conveys ser iously, and in particular to differentiate the word from the term ?sustainable develop ment?. For one thing, sustainability is much less easily and intuitively grafted on to deveL opment orthodoxy aimed at sustaining little more than economic growth. Sustainability continues to function more as an ambiguous mantra than as a new paradigm of a post coLoniaL development agenda, the latter a critique levelled at the institutionalization of sustainable development (Escobar, 1995). Instead, sustainability has been more success fully mobilized in ways that challenge conven tional development paradigms including, for example, in the notion of sustainable liveli hoods and in directing attention towards ?satisficing?, or meeting basic needs (Sneddon, (NEW PARAGRAPH) 2000). (NEW PARAGRAPH) Sustainability is also deployed more con cretely in scientific and technical parlance. This includes efforts to develop sustainability indicators (O?Riordan, 2004) to be used as benchmarks, fixed goals by which the abstraction and obfuscation so typical of sus tainability discourses might be reined in. It also includes the use of the term in policy oriented or more applied ecological sciences (e.g. conservation biology) seeking to develop and apply notions of specifically ecological sustainability, particularly when both human and non human systems exhibit uncertain behaviour. As thin as this literature typically is in conceptualizing human behaviour, impor tant principles such as adaptive management and precautionary action have taken hold (Walters and Holling, 1990: Walters, Korman, Stevens and Gold, 2000) (cf. ecology). (NEW PARAGRAPH) Attempts to define and implement prin ciples of sustainability in planning and devel opment policy have sometimes been pursued through the so called ?three pillars' of sustain ability, namely economic, social, and environ mental or ecological. This is, for instance, a feature of local and regional planning for sus tainability as pursued in the UK (Haughton and Counsell, 2004). True, the condition of sustainability in this context is still tethered (legislatively) to the maintenance of economic growth. But one advantage of at least recog nizing disparate connotations of sustainability, in terms of these three pillars or otherwise, is that it leaves open the possibility that trade offs must be made, and that not all efforts to achieve sustainability can be achieved via the win win optimism that has been a predomin ant gloss on sustainable development since the Brundtland Commission, and certainly since the 1992 Rio Summit (Adams, 1995). Critical work remains to be done on the governme ntalizing dimensions of particular sustaina bility programmes, specifically the ways in which new political subjectivities and modes of governance arise around the institutio nalization of sustainability (see governmen tality). This comprises one way to bring politics into discussions and analyses con cerning sustainability. (NEW PARAGRAPH) In fact, this speaks to a bigger problem of politics when it comes to sustainability and sustainable development. Seemingly endless rounds of defining the terms leads to an over riding idealism in policy and academic litera tures that can actually obscure attention to changing genealogies, as predominant con notations evolve shaped in part by prevailing power relations. Put another way, and in the spirit of Michel Foucault, tracking the chan ging meaning of the term must always be situ ated in relation to the capacity of power to produce these changing meanings. And this is what frustrates many of a critical bent in encountering this word; it seems to preclude or leave unexamined in most iterations ques tions of power and politics. Revisiting the three pillars noted above, for instance, it is not clear where politics enter and how. (NEW PARAGRAPH) Thus, as opposed to more and more attempts to define the term and pin it down, it might be more useful to consider what ques tions it invokes. One of these, as Drummond and Marsden (1999) argue, is why sustaina bility literature is so much characterized by ?line drawing' exercises rather than more critical analyses of systemic tendencies for lines to be transgressed. This echoes early critics of both sustainable development and sustainability, who argued that both require direct challenges to capitalism itself as an inherently unsustainable form of economic, social and political organization constituted by and productive of profound social inequal ities, and predicated on the mobilization of energy and raw materials increasingly com modified for the purposes of an expanding and inherently expansionist economy (Redclift, 1987; O?Riordan, 1991; Benton, 1994). (NEW PARAGRAPH) A second question concerns the ways in which sustainability needs to be operational ized as disparate challenges in relation to the so called environmentalism of the poor versus the environmentalism of the rich (Martinez Alier, 2002). This would allow affluence to be challenged, while recognizing that the poverty and environmental degradation nexus requires distinct approaches (including not just policies aimed at fostering socially equit able and environmentally benign growth policies, but redistribution via genuine third world debt relief and reparations for colo nial plunders). (NEW PARAGRAPH) A third question concerns examination of real world trade offs and complex political ecological dynamics involved in the institution alization of specific programmes aimed at enhancing sustainability, moving past mantras to ask the kinds of hard questions that lend themselves to social science. What, for instance, are the scaled social and environmen tal implications of reforestation programmes, particularly vis a vis reinforcing logging pres sure in faraway places (Robbins and Fraser, (NEW PARAGRAPH) ? What happens to local level social relations, property rights and land use prac tices under the influence of international fair trade and organic standardization regimes (Mutersbaugh, 2004)? (See also forestry.) What are the social and environmental effects (NEW PARAGRAPH) again across scales introduced by regimes such as ?food miles? that stigmatize distance travelled, particularly as these effects ripple through complex commodity chains linking First World markets and Third World agricul tural systems (Friedberg, 2004)? While such critically minded, theoretically informed and empirically oriented engagements with real world instantiations of sustainability oriented policies and programmes might seem to take some of the wind out of sustainability's sails, they also help ensure that sustainability con veys more than a lot of hot air. sp (NEW PARAGRAPH) Suggested reading (NEW PARAGRAPH) Adams (2001); Dobson (1998); Ostrom, Burger, Field, Norgaard and Policansky (1999); Redclift (1987). (NEW PARAGRAPH)

Other books

The Make-Believe Mystery by Carolyn Keene
Black Swan by Chris Knopf
Around the World Submerged by Edward L. Beach
Collateral Trade by Candace Smith
Innocent in Las Vegas by A. R. Winters, Amazon.com (firm)
American Blood by Jason Manning
Zizek's Jokes by Slavoj Zizek, Audun Mortensen