A Conceptual Poverty Mapping Data Model

July 21, 2017 | Autor: Felicia Akinyemi | Categoria: Human Geography, Geomatic Engineering, Data Format
Share Embed


Descrição do Produto

Transactions in GIS, 2010, 14(S1): 85–100

Research Article

A Conceptual Poverty Mapping Data Model Felicia Akinyemi Centre for GIS and Remote Sensing National University of Rwanda Butare, Rwanda

Abstract GIS is increasingly used in poverty mapping but there is no generic data model for database development. Examples exist already of industry-specific models. Having such a data model eases the complexity of incorporating spatial data in poverty assessments. This article raises awareness about the need for a generic poverty data model for use in poverty mapping. It seeks to stimulate a lively debate that will lead to the development and adoption of such a data model. The ultimate goal will be to get to some level of standardization for common data types that would facilitate spatial data use in poverty assessment and sharing among poverty projects. This article is a first step at developing a data model for poverty mapping at a conceptual level. Handling multidimensional social problems, such as poverty, using a spatial framework can be challenging because of the myriad of poverty indicators in use. Employing the entity-relationship approach, a conceptual model is developed in the current article that identifies the key thematic layers, entities, and relationships. The conceptual model produced is useful for modeling the content of the database for use in assessing and monitoring poverty.

1 Introduction Poverty1 reduction is of global concern; the 2015 deadline by which extreme levels of poverty worldwide are to be halved (the first Millennium Development Goal – MDG) is rapidly approaching. The MDGs are a set of development strategies to which many countries are signatories, although implementation, particularly in Sub-Saharan Africa, is currently well behind schedule. A rich diversity of information is required in order to plan and manage development strategies for alleviating poverty effectively. A large Address for correspondence: Felicia Akinyemi, Centre for GIS and Remote Sensing, National University of Rwanda, Loiret, P.O. Box 212 Butare, Huye, Rwanda. E-mail: felicia.akinyemi@ gmail.com © 2010 Blackwell Publishing Ltd doi: 10.1111/j.1467-9671.2010.01207.x

86

F Akinyemi

amount of this information is spatial in nature, and therefore lends itself to presentation in the form of maps (Ezigbalike 2002). Increasingly in many countries, poverty and welfare indicator maps are used to inform decision making and in designing interventions from local to national levels. They are the basis for allocating social assistance (Ahnaf and Hartanto 2009, Buys et al. 2006, de la Espriella 2009, Rasano et al. 2008). In Kenya, division-level thematic maps of natural resources, livestock, etc. were produced from a spatial database comprising poverty related datasets. These maps were useful in analyzing the role of livelihood assets in explaining variations in poverty levels (see Kristjanson et al. 2005). In Mexico, poverty maps were the framework for selecting locations for on-farm work using innovative breeding techniques for maize with the aim of helping poor, small-scale farmers. Poverty maps have also been effective in targeting postharvest technologies in southern Mexico (CIESIN 2006). Combining poverty indicators with geo-referenced datasets such as soil fertility, slope, land use, water access points, etc. helps to highlight areas where, for example, poverty overlaps with other socio-economic and environmental challenges (Bedi et al. 2007). Cartographic representations of welfare indicators ranging from nutritional status to educational level can be used in combination with geospatial information on environmental conditions to design programs that address problems across a range of geographic scales, including at a local level (FAO 2003). Location is a powerful determinant of poverty. Spatial patterns of inequality between and within countries have become an important focus of the development community, and research on patterns of poverty and inequality across districts, municipalities, and communities has accelerated over the past decade. With spatial variables increasingly recognized as determinants of poverty (Bedi et al. 2007, Hyman et al. 2005), the role of Geographic Information Systems (GIS) in poverty assessment has increased in importance, particularly as a means of generating explanatory variables and because of its data integration and spatial analysis capabilities. There is a broad consensus among poverty researchers around the view that poverty is multidimensional. Households in poverty show consumption deficits often linked to restricted access to basic services, limited networks and access to economic opportunity. Typically households in poverty show deficits along many dimensions of well-being at the same time (Barrientos 2010). Capturing the multidimensionality of poverty requires the use of a multitude of different indicators (Akinyemi 2007, Bedi et al. 2007, Davis 2003, Latifa et al. 2008). With a focus on capturing different poverty dimensions, measures are developed for use with differing assumptions and data requirements. This situation poses a challenge to managers of poverty reduction programs (PRPs) who have to decide on the indicators to use, with the final choice often influenced by the quality of the available data. Data modeling is required to better capture the components, processes, and meanings of poverty assessment as with any other geographic phenomenon, and translating the knowledge into a GIS (Glennon 2010). Already, about 32 different industry-specific ArcGIS data models have been designed and implemented. Examples of available data models are census-administrative boundaries, energy utilities, forestry, geology, health, land parcels, national cadastre, transportation, groundwater, water utilities (for details, see http://support.esri.com/datamodels). These ArcGIS data model templates implement what users have found to be widely adopted best practices for building systems that really work (Arctur and Zeiller 2004). Much work has been done on developing different © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

Poverty Data Model

87

aspects of data modeling, e.g. object-oriented (Balram and Dragicevic 2006), spatiotemporal (Camossi and Bertolotto 2010, George et al. 2009). Examples of spatial data modeling applications are flow modeling within GIS (Glennon 2010), land administration domain model (ISO 2010, Uitermark et al. 2010), hydrological/watershed modeling (Bollaert 2006, Gosain et al. 2010), artisanal fisheries management (Vales 2007). This article describes a GIS data model that has been designed to aid schema creation and management of a spatial database. The data model consists of both spatial and non-spatial datasets that are used for assessing poverty levels in PRPs. This data model can be modified for use in analyzing other poverty related problems, such as food security, etc. This work is an attempt at putting together content for such a data model from looking at real life examples. The article outline is as follows: a background to poverty assessment is given describing the types of poverty measures in use. This is followed by an overview of data modeling and a description of the EntityRelationship approach utilized in designing the data model. The article presents the poverty data model design and details the content of the database at a conceptual level. Lastly, the types of spatial representation suitable for poverty related entities are explored.

2 Background to Poverty Assessment There is no definition of poverty that enjoys general consensus, the term poverty is varied in definition and reflects the different causes of poverty observable from place to place. Poverty is increasingly conceived as a latent concept that has never been defined precisely neither has there been a single, commonly agreeable proxy indicator to gauge it (Waglé 2008). There is the conventional economic laissez-faire perspective that attributes poverty to the personal failings of the individual and the structural poverty perspective that views poverty as the inevitable outcome of an unfairly structured political and economic system that discriminates against disadvantaged groups (Townsend 1993, Wratten 1995). Perspectives about the nature of poverty and the policy responses that follow these perspectives are central in deciding how best to study, measure and analyse the phenomenon. Each perspective invariably leads to an alternative set of policy prescriptions based on differing measures and demand different kinds of information. There are four major classes of poverty mapping measures identified on the basis of data sources, assumptions, and the statistical routines utilized. These are econometric, social, demographic, and vulnerability-based measures (Davis 2003). These classes of poverty measures are discussed briefly.

2.1 Econometric Measures The most popular econometric measure is the Foster-Greer-Thorbecke class of decomposable poverty indices (FGT indices) which addresses poverty in its three dimensions of incidence, intensity and severity (degree) among a given population. This is the basis for calculating the poverty headcount, poverty gap and poverty severity indices (see Foster et al. 1984). The FGT index has been used in several studies to generate overall poverty indices at national, sub-national, and/or socio-economic levels of interest (see Baker and Grosh 1994, Kanbur 1987, Osberg and Kuan 2008). A most widely used © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

88

F Akinyemi

econometric based technique for poverty mapping is the statistical small-area estimation (SAE) technique. The SAE combines survey and census data to estimate welfare or other indicators for administrative units such as municipalities and districts. A set of explanatory variables, common to both survey and census, are used to estimate model parameters. It is assumed that the relationship defined by the model using the original sample holds for the larger population as well (see Elbers et al. 2002). The computed estimates of poverty levels are then represented graphically as maps. These maps are used for devising policy strategies for given localities (see Bedi et al. 2007 and Sachs 2005).

2.2 Social Measures Social measures use non-monetary indicators and aim to capture household well-being as measured by the quality of and access to social services such as education, health, nutrition, water, housing, and neighborhood quality. According to the unsatisfied basic needs (UBN) approach, poverty is linked to a state of necessity, a deficiency of income, and a deprivation of the goods and services necessary to sustain life at a minimum standard (see Deutsch and Silber 2005). Various indices are in use based on the UBN approach for mapping poverty at disaggregated geographic units. The basic differences between these indices are the types of variables and weighting methods used. Basically, a UBN index selects certain needs considered essential, identifies a minimum criterion for satisfying each need (e.g. the corresponding national average value is sometimes used), and computes the poverty level. The most widely used of these UBN based indices are the human development indices (HDIs) developed by the United Nations Development Program (UNDP). The HDIs are computed as the weighted average of longevity, knowledge, decent standard of living, and social exclusion (this last one only in case of OECD countries, see UNDP 2005). These indices are computed annually for every country and are generally available for use.

2.3 Demographic Measures Demographic measures focus on gender issues, health and age structure of households, child nutritional status, and household size. In addition, various child outcome indicators are important such as calorie intake, low height for age, low weight for age, low weight for height, body mass index, and low birth weight. These child outcome indicators can be used as good indicators of development in a region and are easy to interpret as measures of poverty especially when based on anthropometric measurements such as low height for age. Anthropometric measures are recommended for use as the best general indicators of constraints to the welfare of the poorest, including dietary inadequacies, infectious diseases, and other environmental health risks (UN 1992).

2.4 Vulnerability Measures This class of measures is concerned with the level of household exposure to shocks, using indicators such as environmental endowment and hazard, physical insecurity, political change related to empowerment, governance, participation, transparency of the legal system, structural inequities, and skewed processes that become impediments to human © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

Poverty Data Model

89

well-being (Henninger 1998). Vulnerability can be defined as the probability that individuals, households or communities will be in poverty in the future. Vulnerability is not only a dimension of poverty, but can also be a cause of poverty and its persistence (Barrientos 2010). Increasing decentralization of governmental authority in many developing countries is creating demand for the information that poverty and welfare maps provide. Citing the case of Ghana, Dery and Dorway (2007) noted that district assemblies and nongovernmental and community-based organizations are asking for more and more georeferenced information on the location of the poor and the magnitude of poverty so that they can set priorities, target interventions, empower local communities and improve their understanding of the causes and effects of poverty. Although provincial governors are no doubt interested in knowing how poor their provinces are relative to other provinces, they are probably more interested in knowing which parts of the province are the poorest so as to focus development efforts where most needed. Similarly, a district commissioner would be interested in sub-district poverty rates so as to help set poverty reduction priorities within the district (Bedi et al. 2007, Elbers et al. 2007).

3 Data Modeling The term modeling is used in several different contexts, both in the world of computer science as well as in GIS. Consequently, an effort is made to clarify its meaning in the context of this article. Data modeling is an abstraction process where the essential elements are emphasized and the non-essential ones eliminated with regard to a specific goal (Bédard and Paquette 1989). The aim is to model real world entities and the relationships in a way that maximizes benefits while utilizing a minimum amount of data (Kufoniyi 1997). Data modeling enhances application maintainability and future systems may re-use parts of existing models, which should lower development costs. Since a data model can be used by multiple people or organizations, it allows for a specific application, the use of the same datasets and attributes, thus making interoperability easier to achieve. A good data model design is crucial to making better decisions based on available data in GIS while supporting existing standards (ESRI 2005). A major advantage is that it gives access to the elements that are relevant to a GIS analysis. For GIS to adequately support poverty assessment, which is the first step to reducing poverty prior to policy formulation, there is the need for data modeling.

3.1 What is a Data Model? Data models are simplified views of a part of reality. They are a description of the rules by which data is defined, organized, queried, and updated within an information system (ESRI 2006). That is, data models are built according to certain rules to facilitate the implementation of a database (Bédard and Paquette 1989). A data model is a set of expectations about data – a template into which the data needed for a particular application can be fitted (Goodchild 2005). A data model is the product of the database design process which aims to identify and organize the required data logically and physically. Database design is the information system planning activity where the contents of the intended database are identified © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

90

F Akinyemi

Figure 1 Levels of data view in database development (adapted from Akinyemi 1999)

and described. Database design is usually divided into three major activities (Elmasri and Navathe 1989), namely: conceptual, logical and physical modeling (see Figure 1). In Figure 1, the design and implementation of the database are shown as the two basic phases involved in the development of a database. The focus of this article is limited to the conceptual level of database design. This is because there is as yet no generic poverty data model in place. A conceptual data model ensures that the database content is (1) understandable by the proposed users of the database or system; and (2) sufficiently structured for a programmer or analyst to design the data files and implement data processing routines to operate on the data. Conceptual modeling is the representation of the functional application requirements and information system components at an abstract level, i.e. a description of what is to be included in an information system rather than how the intended information system will work. The data model shows what data are to be contained in the database and how the items in the database will be related to each other. For example, a data model might specify that a customer is represented by a customer name, credit card number; that a product has a product code and price, and that there is a one-to-many relation between a customer and a product. Data modeling requires the use of rules to create the model and to communicate this model, i.e. a language using a well-defined set of symbols (literal and/or graphical) with associated meanings. In the next subsection, the EntityRelationship model (ER model) is examined. The ER model is one of the most widely used methods for developing data models today (Ullman 1988). © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

Poverty Data Model

91

3.2 Entity-Relationship Modeling The ER model adopts the more natural view that the real world consists of entities and relationships (see Chen 1976). It involves identifying, classifying, describing and relating parts of the real world to organize the information into a formal structure amenable to a computer form. Thus, it is useful to perceive the reality as containing entities or objects, attributes or characteristics (properties) of the objects, and relationships between entities. Entities and relationships are a natural way to organize physical things as well as information . . . The ER concept is the basic fundamental principle for conceptual modeling (Winslett 2004). ER modeling activities are to: (1) determine what entity types are involved; (2) determine which entity types are related; and (3) refine the definition of the relationships (see http://www.inf.unibz.it/~franconi/teaching/2000/ct481/er-modelling/). Through the use of an entity-relationship diagram (or ER diagram), the logical relationships of entities are represented to create a database. In basic ER symbology, entities are portrayed as rectangles and relationships as diamonds with the lines connecting the rectangular boxes defining the relationships between the entities. For example, 1:n, m:n and 1:1 relationship mappings are distinguishable in the ER diagram. An entity is a thing that can be distinctly identified. A relationship is an association between two types of entities; usually identified by a verb or a preposition (e.g. Road to cross River). A relationship has a cardinality giving the number of times (minimum and maximum) the relation can occur between two specific entities (occurrences). For example, if we say that a Road crosses a River a minimum of 0 times and a practical maximum of 5; and on the other hand that a River can be crossed by a minimum of 0 Roads and an unknown maximum of N, this leads to a relation to cross with a cardinality of 0,5 in one direction and 0N in the other direction. An attribute is a characteristic of an entity type or a relationship (Bédard and Paquette 1989, see Table 1).

3.3 Spatial ER Data Modeling The need to accommodate the specific intricacies of modeling spatial data for different applications has necessitated further modifications of ER symbology over the years (see Armstrong and Densham 1990, Bédard and Paquette 1989, Hadzilacos and Tryfona 1997, Malinowski and Zimányi 2007, Shashi et al. 1997, Tryfona and Jensen 1999). Goodchild et al. (2007) offer a unified perspective on these geographic representations. The industry-specific ArcGIS data models provide a starting point for each user-specific design project and incorporate industry-standard design techniques from information technology (IT), database management systems (DBMS), GIS and various standard organizations. Each data model includes commonly adopted spatial representations (for example, points, lines, and polygons), classifications, and map layer specifications that can be implemented in any GIS. It is important to note that the analysis possible in a GIS is dependent on the mode of data modeling and structuring in the database. Throughout the model creation process, the designer should consider the typical queries the model should be able to answer (Glennon 2010). In this study, Calkins’ extension to basic entity-relationship models is used (Calkins 1996). This extension handles spatial elements, multiple spatial representations, temporal representations, and coordinate and topological attributes of spatial entities. The spatial relationships are defined by three relationship symbols. The traditional diamond © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

92

F Akinyemi

Table 1 Basic entity/relationship (ER) concepts Concept

Informal definition

Examples

Symbol

Entity

A distinguishable object, person, concept or event about which we want information Entity whose existence depends on another entity. i.e. entity type Y is a subtype of entity type X if and only if every Y is necessarily an X Describes the association interconnecting entities

River, Road, Census block, Parcel

Rectangle

School, hospital are subtypes of facility

Double Rectangle

Cardinality e.g. One to many relationship (LocalAuthority_ Neighborhood) Census Block has literacy rate, average household size, male, female, average income

Diamond

Subtype

Relationship

Properties or attributes

A piece of information that describes an entity

Ellipse

symbol can be used for normal database relationships. An elongated hexagon and a double elongated hexagon, represent spatial relationships. The former defines it through topology (e.g. connectivity and contiguity), whereas the latter uses x,y coordinates as well as related spatial operations (e.g. coincidence, containment and proximity) for defining spatial relationships (see Calkins 1996 for details).

4 Conceptual Poverty Data Model A GIS database is founded upon geographic representations (Arctur and Zeiller 2004). The spatial representation of the data in the GIS is often based on the information requirements, the cartographic display needs, the implicit map scale of the database and other factors important to the application. Thus, the locational attributes of spatial data (e.g. for settlement, region, etc.) are formally expressed by means of the geometric features of points, lines or areal units (polygons) in a plane, or, on a surface. In the Social Sciences, spatial samples are mostly limited to data for areal units (Anselin (1992). Frequently, the geographic representations will be pre-determined to some degree by the available data sources. This spatial referencing of observations is also the salient feature of a GIS, which makes it a natural tool to aid the analysis of spatial data. Data for poverty assessment are generally socioeconomic in nature and sources for such data are censuses, household surveys, etc. These socioeconomic datasets are linked © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

Poverty Data Model

93

to geographic constructs for representation in a GIS. For example, census divisions (census blocks, wards, enumeration areas) or administrative units have spatial representations as area (polygon) features in a GIS. Also, demographic data as attributes have to be linked to the spatial representations of poverty. This in a way makes such socioeconomic datasets spatial. Identifying the typical datasets that can be used for poverty mapping is most crucial in facilitating the incorporation of spatial data in poverty analysis. Managers of PRPs are saddled with the responsibility of designing appropriate strategies and they need to identify necessary datasets to be included in the application. Spatial datasets for poverty mapping are identified prior to the design of the conceptual model.

4.1 Identifying Spatial Data for Poverty Assessment It is evident from the earlier discussion on classes of poverty measures that both spatial and non-spatial datasets are needed for poverty assessment. In seeking to identify relevant datasets for poverty assessment, Akinyemi (2007) surveyed recent poverty mapping projects and studies. A major finding was that a broad range of spatial data is in use for poverty assessment and policy formulation. The types of data in use differ within and between the major classes of poverty mapping measures examined. In other words, datasets in use are as diverse as the poverty dimensions being examined and the measures utilized (see Akinyemi 2007 for details). In addition to socioeconomic and demographic datasets, the common spatial datasets for poverty mapping are land cover, normalized differential vegetation index (NDVI), rainfall data, soil fertility and quality. Utilizing measures of distance and physical accessibility such as travel times to markets and distances to major towns and facilities, etc. are also increasingly important in poverty mapping. This is because income generation for small-scale farmers, for example, often depends on distances to markets and associated transport costs ( Jacoby 2000 and Van De Walle 2002; both cited in Hyman et al. 2005).

4.2 The Conceptual Poverty Model for Spatial Data In conceptual modeling, all entities represented as spatial and non-spatial datasets, their attributes and the relationships between them are identified. Based on the range of spatial datasets identified in poverty mapping programs, key thematic layers to be represented in the database are identified. These are legal and administrative areas, census and statistical boundaries, environmental based datasets, transportation, location of infrastructure important to the community, geographical names and land (see Figure 2). Once the key layers in the poverty mapping application have been identified, these layers are further decomposed into basic entities. Working at a conceptual modeling level, the spatial representation of the entities, attributes and relationships, etc. are illustrated in Figures 3 and 4. Figure 3 shows the poverty data model developed in this study using an ER model diagram. It shows the spatial representations of the entities in the thematic layer categories. For readability, only attributes for State and LocalAuthority are shown in Figure 3. Usually attributes should be included in the ER diagram for evaluation purposes. For better understanding, Figure 4 shows how attributes are related to entities in the model, for example, for Neighbourhood and Household. © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

94

F Akinyemi

Figure 2 Key thematic layers for poverty spatial data modeling

For more details of the entities and their attributes, see Table 2. Table 2 gives a breakdown of the major categories, entities, attributes and means of spatial representation. The datasets shown in Figures 3 and 4 and Table 2 are not exhaustive. More can be added depending on the poverty dimensions and the types of queries that are envisaged.

5 Discussion The design of a database is carried out at different levels representing the conceptual, the logical, and the physical views of reality. The data model developed in this study is limited to the conceptual design view. As the most abstract stage of a database design phase, working at the conceptual level helps to unambiguously and rigorously describe the data to be included in the database. Creating such a database framework deconstructs the phenomenon to its basic conceptual elements and offers a formalization of a phenomenon’s ontology. From a pragmatic GIS perspective, the database framework offers a mechanism to add meaning to primitive GIS elements such as points, polylines, and polygons, relate their attributes, facilitate data collection and sharing, and build analytical and representation functionality. Sometimes, it happens that some order and relationships identified in the data model may be previously unrealized (Glennon 2010). The focus in this article has been to describe the content of the poverty information system rather than how the information system will be implemented in a software environment. In developing this poverty data model, the study identifies the key thematic layers, basic entities, attributes and relationships for use in poverty mapping applications. In representing the entities in the data model in a GIS, it is important to note that the spatial representation of an entity in a database is relative. Goodchild et al. (2007) noted that discrete objects and continuous fields are the only possible bases for spatial representation. Most poverty mapping applications use administrative area for spatial representation. An administrative area is a discrete object and is typically represented as a polygon feature in a GIS. It can be defined at different administrative hierarchical levels © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

Poverty Data Model

95

Figure 3 Conceptual poverty data model

– national, state, district, etc. at which data is aggregated to aid the understanding and management of activities geographically. An entity may be represented by one of several corresponding spatial entities. For example, a city in a small-scale database could be a point, whereas the same city is a polygon in a large-scale database (Calkins 1996). The properties of the entities do not have a geographic representation in the data model. As a result, these are perceived as table objects. In the GIS environment, these table objects are related to the spatial objects components of the entities through © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

96

F Akinyemi

Figure 4 Sample attributes non-spatial data (adapted from Akinyemi 2001)

relationships. In the poverty data model, the relationships are defined in three ways. These are relationships represented in the database (e.g. spatial relationship between State and LocalAuthority, non-spatial relationship between Building and Household), relationships represented by topology (e.g. adjacency relationship between Roads and Parcel ), and relationships derived using spatial operations (e.g. coincidence between LocalAuthority and Wetlands, proximity between Roads and DistanceAccessibility). Typically, the information needed to develop a conceptual model comes from the user need assessment. Need assessment is the first step in implementing a successful GIS (Gosain et al. 2010). Assessment in this article was accomplished by examining the content of several poverty mapping studies and projects. This approach was used as poverty applications can be complex because of the multidimensional nature and interconnected data. Moreover, this article is an initial attempt aimed at stimulating a lively debate on the need to develop a generic data model for use in poverty assessment. When there is no proper understanding of an application, the data model designers may need to first develop a preliminary model based on their perception of the requirement rather than that of the users (see Gosain et al. 2010). Ultimately, the data model must meet the requirements of its users to be successful. Once identified, entities are mapped to feature classes using the conceptual schema in the next design stage, i.e. logical design where the data structure is defined. Data © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

© 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

Wetland Soil Aridity zone Natural park Meteorological station Road Railroad Parcel Building Household

Environment

Geographic names

Facility (subtypes)

Land

Name

Landfill School Hospital Water sources Well Urban center Distance accessibility

Facility

State (ST) Local authority (LA) Neighborhood

Administrative and statistical (census)

Transportation

Entity

Thematic category

Table 2 Data list of poverty data model

STID, Name LAID, Name NID, Name, LAID, Po-PovertyHeadcount, P1- PovGap, P2-PovertySeverity, LowHeightAge, LowWeightAge, LowWeightHeight, CalorieIntakeRequired, QualityOfLife, PopulationGender, PopDensity, AveHouseholdSize, FemaleHeadedHousehold, FloodRisk Type, MajorFauna, MajorFlora, TypeSoil, AgricPotential, Fertility, Type, BiodiversityImportance, EndemicSpecie AmphibianSpecies AmountRainfall, Temperature, Humidity Name, Pavement, NumLanes, NumAccidents, QualityRoad, SpeedLimit Name, Capacity PID, TransactionID, Ownership, Value, Zone, Use BID, Address, Type, Ownership, HousingQuality HID, HouseholdSize, DependencyRatio, QualityOfLife, PerCapitaIncome, HouseholdTotalIncome Name, LocationAddress, NeighborhoodName, StateFacilityID, NationalFacilityID, OfficialStartDate, OfficialEndDate, Status, LAID Size, OwnershipType, LandfillCover AverageEnrolment, AverageDistance, Status, DiseaseIncidence, WaterTable, Pollution, WaterPtManagement, WaterQuality WellType, CasingType, DepthToBedrock, DepthToGrdwater, Depth LAID, Name, Population DistanceToUrbanCenter, DistanceToNearestRoad, DistanceToMajorRiver, TravelTimeToMajorMarket, DistanceToNearestSchool, DistanceToNearestWaterPoint, TravelTimeToNearestMajorForest /National park, DistanceToNearestHospital ObjectID, Name, AlternateName, NamingAuthorityID, NameStyle

Attribute

Table

Point Point Point Point Point/Polygon Spatial Operation (e.g. Proximity) Table

Point

Polygon Polygon Polygon Polygon Point Line Line Polygon Polygon Table

Polygon Polygon Polygon

Spatial representation

Poverty Data Model 97

98

F Akinyemi

structure is the logical arrangement of data as used by a system for data management and is the schematic representation of how the data will be stored for retrieval (AGI 2006 cited in Bollaert 2006). For example, the attributes of the entities may be held in a relational database management system linked to the GIS.

6 Conclusions This article describes the design of a poverty data model with the aim of aiding schema and spatial database creation for poverty mapping. The design of the data model is at a conceptual level. The purpose of the conceptual data model is to describe the data to be included in the database at an abstract level. The poverty data model developed consists of both spatial and non-spatial datasets for use in assessing poverty levels. Key thematic layers in the poverty data model were identified. For each thematic layer, the entities, their properties, relationships between the entities are spatially represented in the data model for use in a GIS. The thematic layers and the entities were selected on the basis of the types of spatial datasets commonly used for poverty mapping as evident from several previous studies. This work is an initial attempt at putting together content for a data model from looking at real life examples. This data model can be modified for use in other poverty reduction programs.

Acknowledgements The useful comments of the anonymous reviewers are highly appreciated.

Note 1 The term poverty is used in this article in a broad sense to include poverty, food-security and other related themes.

References Ahnaf A and Hartanto W 2009 The use of population census data for various program targeting. In Proceedings of the Twenty-fourth Association of National Census and Statistic Directors of America, Asia and Pacific Census Conference, Hong Kong, China: 1–9 Akinyemi F O 1999 Database design and development for census mapping in Nigeria. In Balogun O Y and Soneye A S O (eds) Cartography in the Service of Government. Lagos, Nigerian Cartographic Association: 191–202 Akinyemi F O 2001 Geographic targeting for poverty alleviation in Nigeria: A GIS approach. In Proceedings of the Twenty-first International Cartographic Association Conference on Mapping the Twenty-first Century (ICC 2001, Vol. 2), Beijing, China, 1259–70 Akinyemi F O 2007 Spatial data needs for poverty management. In Onsrud H J (ed) Research and Theory in Advancing Spatial Data Infrastructure Concepts. Redlands, CA, ESRI Press: 262–77 Anselin L 1992 Spatial Data Analysis with GIS: An Introduction to Application in the Social Sciences. Santa Barbara, CA, University of California: National Center for Geographic Information and Analysis Technical Report No 92-10 © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

Poverty Data Model

99

Arctur D and Zeiller M 2004 Designing Geodatabases: Case Studies in GIS Data Modeling. Redlands, CA, ESRI Press Armstrong M P and Densham P J 1990 Database organization strategies for spatial decision support systems. International Journal of Geographical Information Systems 4: 3–20 Baker J L and Grosh M E 1994 Measuring the Effects of Geographic Targeting on Poverty Reduction. Washington, DC, World Bank Living Standards and Measurement Survey, Working Paper No 99 Balram S and Dragicevic S 2006 Modeling collaborative GIS processes using soft systems theory, UML, and object-oriented design. Transactions in GIS 10: 199–218 Barrientos A 2010 Social Protection and Poverty. Geneva, United Nations Research Institute for Social Development Social Policy and Development Program: Paper No 42 Bédard Y and Paquette F 1989 Extending entity/relationahip formalism for spatial information systems. In Proceedings of the Ninth International Symposium on Computer-Assisted Cartography (AutoCarto ‘89), Baltimore, Maryland: 818–27 Bedi T, Coudouel A, and Simler K (eds) 2007 More Than a Pretty Picture: Using Poverty Maps to Design Better Policies and Interventions. Washington, DC, World Bank Bollaert M J 2006 Development of a Geographic Data Model for Hydrological Modelling. Unpublished MSc Thesis in Hydrology, University of KwaZulu-Natal, South Africa Buys P, Chomitz K, Dasgupta S, Deichmann U, Larsen B, Meisner C, Nygard J, Pandey K, Pinnoi N, and Wheeler D 2006 The economics of decentralized poverty-environment programs: An application for Lao PDR. Journal of Policy Modeling 28: 811–24 Calkins H W 1996 Entity relationship modeling of spatial data for geographic information systems. International Journal of Geographical Information Systems 10: 1–19 Camossi E and Bertolotto M 2010 A Multi-granular 3D Spatial-Temporal Model Tailored to Environmental and Geographic Areas Monitoring. Dublin, University College Dublin School of Computer Science and Informatics Technical Report No. 3 Chen P 1976 The entity-relationship model: Toward a unified view of data. ACM Transactions on Database Systems 1(1): 9–36 CIESIN 2006 Where the Poor Are: An Atlas of Poverty. Palisades, NY, Columbia University, Center for International Earth Science Information Network Davis B 2003 Choosing a Method for Poverty Mapping. Rome, FAO Dery B B and Dorway A 2007 District-Based Poverty Profiling, Mapping and Pro-poor Planning as a Monitoring and Evaluation Tool. Bamako, Communicances Deutsch J and Silber J 2005 Measuring multidimensional poverty: An empirical comparison of various approaches. Review of Income and Wealth 51: 145–74 Elbers C, Lanjouw J, and Lanjouw P 2002 Micro-level Estimation of Welfare. Washington, DC, World Bank Policy Research Working Paper No. 2911 Elbers C, Fujii T, Lanjouw P, Ozler B, and Yin W 2007 Poverty alleviation through targeting: How much does disaggregation help? Journal of Development Economics 83: 197–213 Elmasri R and Navathe S B 1989 Fundamentals of Database Systems. Redwood City, CA, Benjamin/Cummings Publishing de la Espriella C 2009 A technique for small-area poverty analyses. Urban Studies 46: 2399– 421 ESRI 2005 Introduction to ArcGIS Data Models. Redlands, CA, ESRI Press ESRI 2006 GIS dictionary. WWW document, http://resources.arcgis.com/glossary/term/373 Ezigbalike D 2002 Justification for SDIs. Unpublished paper presented at the Joint FAO-GISD (Geographic Information for Sustainable Development) Regional Meeting (GISD ‘02), Asmara, Eritrea FAO 2003 The State of Food Insecurity in the World: Monitoring Progress Towards the World Food Summit and Millennium Development Goals. Rome, United Nations Food and Agricultural Organization Foster J, Greer J, and Thorbecke E 1984 A class of decomposable poverty measures. Econometrica 52: 761–5 George B, Kang J M, and Shekhar S 2009 Spatio-temporal sensor graphs (STSG): A data model for the discovery of spatio-temporal patterns. Intelligent Data Analysis 13: 457–75 Glennon A 2010 Creating and validating object-oriented geographic data models: Modeling flow within GIS. Transactions in GIS 14: 23–42 © 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

100

F Akinyemi

Goodchild M F 2005 GIS, spatial analysis and modeling overview. In Maguire D J, Batty M, and Goodchild M F (eds) GIS, Spatial Analysis, and Modeling. Redlands, CA, ESRI Press: 1– 17 Goodchild M F, Yuan M, and Cova T 2007 Towards a general theory of geographic representation in GIS. International Journal of Geographical Information Science 21: 239–60 Gosain A K, Sandhya R, Puja S, and Anamika A 2010 Integration of bio-geospatial database for selected watersheds in Himalayan region. Current Science 98: 183–91 Hadzilacos T and Tryfona N 1997 An extended entity-relationship model for geographic applications. SIGMOD Record 26(3): 24–9 Henninger N 1998 Mapping and Geographic Analysis of Human Welfare and Poverty: Review and Assessment. Washington, DC, World Resources Institute Hyman G, Larrea C, and Farrow A 2005 Methods, results and policy implications of poverty and food security mapping assessments. Food Policy 30: 453–60 ISO 2010 Geographic Information – Land Administration Domain Model (LADM): Draft International Standard. Geneva, Switzerland, ISO Technical Committee ISO/TC 211 on Geographic Information/Geomatics Kanbur S M R 1987 Transfer, targeting and poverty. Economic Policy 4: 111–98 Kristjanson P, Radeny M, Baltenweck I, Ogutu J, and Notenbaert A N 2005 A livelihood mapping and poverty correlates at a meso-level in Kenya. Food Policy 30: 568–83 Kufoniyi O 1997 Vector GIS for topographic information production. In Ikhuoria I A (ed) Cartography and GIS for Sustainable Development. Lagos, Nigerian Cartographic Association: 37–55 Latifa A, Aswatini, and Romdiati H 2008 Population and social demographic poverty: A case study in the border areas of East Kalimantan and North Sulawesi. Journal of Indonesian Social Sciences and Humanities 1: 175–91 Malinowski E and Zimányi E 2007 Logical representation of a conceptual model for spatial data warehouses. GeoInformatica 11: 431–57 Osberg L and Kuan X 2008 How should we measure poverty in a changing world? Methodological issues and Chinese case study. Review of Development Economics 12: 419–41 Rasano C, de Perez G, and Perez R A 2008 Analyzing Urban Poverty: GIS for the Developing World. Redlands, CA, ESRI Press Sachs J 2005 The End of Poverty: Economic Possibilities for Our Time. New York, Penguin Press Shashi S, Coyle M, Goyal B, Liu D R, and Sarkar S 1997 Data models in Geographic Information Systems. Communications of the ACM 40(4): 103–11 Townsend P 1993 The International Analysis of Poverty. London, Harvester Wheatsheaf Tryfona N and Jensen C S 1999 Conceptual data modeling for spatio-temporal applications. GeoInformatica 3: 245–68 Uitermark H, van Oosterom P, Zevenbergen J, and Lemmen C 2010 From LADM/STDM to a spatially enabled society: A vision for 2025. In Proceedings of the Annual Bank Conference on Land Policy and Administration, Washington, DC: 1–13 Ullman J D 1988 Principles of Database and Knowledge-Base Systems: Vol. 1, Classical Database Systems. New York, Computer Science Press UN 1992 Second Report on the World Nutrition Situation: Vol. 1, Global and Regional Results. Geneva, United Nations Administrative Committee on Coordination, Subcommittee on Nutrition UNDP 2005 Human Development Report 2005. New York, Oxford University Press Vales M E 2007 A Database Based Information System for Artisanal Fisheries Management: A Case Study of Moma-Angoche in Mozambique. Unpublished Master’s Thesis, University of KwaZulu-Natal, South Africa Waglé U R 2008 Multidimensional poverty: An alternative measurement approach for the United States? Social Science Research 37: 559–80 Winslett M 2004 Peter Chen speaks out on paths to fame, the roots of ER model in human language, the ER model in software engineering, the need for ER databases and more. SIGMOD Record 33(1): 1–9 Wratten E 1995 Conceptualizing urban poverty. Environment and Urbanization 7: 11–36

© 2010 Blackwell Publishing Ltd Transactions in GIS, 2010, 14(S1)

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.