Glossary
Alpha levelA probability threshold used for evaluating a null hypothesis.
Alternative hypothesisThe alternative hypothesis (Ha) describes the spatial pattern that the test is designed to detect. This may be a specific alternative, such as direct overlap of objects or clustering, or it may be the omnibus "not the null hypothesis".
AsynchronousSpace-time data is asynchronous when the position or attributes of objects change at different times. Compare to synchronous.
AutocorrelationHaving a self-similarity or relationship. SpaceStat can test for spatial autocorrelation using the Moran and local Moran statistics.
AxesIn a scatterplot, vertical and horizontal lines that go through the origin (0,0). Compare to bounds.
BinsTo make a histogram of a data distribution, SpaceStat groups the distribution into a set number of classes or bins. Each bin is a bar on the histogram, the height of the bar indicates the number of locations with data in the bin's range.
BoundsThe edges of the plots. Elsewhere, the bounds are often termed the x and y axes, but in SpaceStat we are reserving the term axes for the lines or cross-hairs on a scatterplot that go through the origin.
CategoricalSee qualitative color scheme or string data.
Center containmentThis is used on the map properties dialog to specify how selection should occur in the map. The selection rectangle selects those objects for which it contains the center.
Class patternSee fill pattern.
ClassificationClassification is the process of reducing the dimensionality of complex data by grouping values into set classes for display. Compare to divergent or sequential gradient color.
ContiguitySee spatial weights.
Data typeData can either be numeric (decimal or integer) or string (text-based).
Data viewA window that holds a list of all of the datasets and geographies in the current SpaceStat project. This window is usually docked at the far left side of the application.
Divergent colorThis is a color scheme for numeric data that is continuous, rather than classified. Use divergent color when you want to emphasize an internal threshold, such as the mean of a z-score standardized dataset. For divergent color, you specify the minimum, diverge, and maximum colors and SpaceStat ramps between them to provide intermediate colors for intermediate values.
DockSome views in SpaceStat are dockable, which means you can attach them to the edges of the windows that contain them. For instance, you can dock (or undock) the session log and the data view in the application window. You can also dock toolbars and the query window in the map to which it refers.
Equal intervalA classification method used for coloring objects in the map. The equal interval method divides the data values into a set number of classes dividing the range of the data evenly between classes.
Exact timeThis setting controls whether the step size is set in the minimum time units (say 365 days rather than a year) or as specified (so a year rather than 365 days). Exact time is in the time units, not as specified.
Fill patternSpaceStat draws polygons in a map using color and fill pattern. The fill can be solid fill, dotted, or hatched patterns.
First order neighborsSee neighbor rings.
FIPS IDFIPS stands for Federal Information Processing Standard. Many geographic datasets use the "FIPS ID" as the unique identifier of the spatial units (states, counties, etc.) in the dataset.
GeographyIn SpaceStat, a geography is the collection of spatial information (the location of a point or the vertices of a polygon) referring to a specific region and valid for a specified time. A single geography can be assembled from more than one imported shapefile, representing states of the objects (movement, aggregation, splitting) over time. We sometimes refer to "parent" geographies, and these are simply the geographies from which a subset geography is created, or into which new datasets are added (e.g., a weight set).
GlobalIn spatial statistics, a statistic that assesses a characteristic about an entire dataset rather than individual locations within a dataset. In map properties, a global fill pattern is the same fill for all categories or classes rather than custom fill patterns for individual classes.
GroupTo combine two or more categories into a single category, used in qualitative color schemes for maps.
HandleThe bar at the edge of docked windows and toolbars on which the user can click to reposition window or toolbar.
Heteroskedascitya way to describe a data sample or data-generating process in which the errors display non-constant variance
ID fieldID fields are required for matching datasets together when importing time series or time slice information. The ID can be in integer or string format, but there must be a unique value for each object.
InclusiveA time stamp is inclusive if the time stamp or range of time the data includes the start or end time. A time stamp is non-inclusive if it is valid up to but not including the start or end time.
Interquartile rangeThe interquartile range is represented as the box in the box plot. It is the difference between the values at the 25th and 75th percentiles of a data distribution.
LocalIn spatial statistics, a statistic that assesses a characteristic individual locations within a geography rather than an entire region as a whole. Compare to global.
Lower tail p-valueA P-value obtained by comparing the test statistic to the end of the reference distribution where the statistic's values are lowest. It provides an estimate of how likely values lower than the test statistic are, based on the null hypothesis.
MBRSee minimum bounding rectangle.
MergeYou can join a DBF to an imported shapefile or join several shapefiles into a single space-time dataset using SpaceStat.
Minimum bounding rectangleThe smallest rectangle that can include the entire object (polygon or line). This is used in the map properties dialog to set how selection on a map occurs.
Missing value codeWithin SpaceStat, missing values are indicated by dashes. However, when importing or exporting shapefiles or DBFs, missing values are represented by a code, often something that is recognizably different from the actual data (such as -9999).
Monte Carlo randomizationSee randomization.
Natural breaksA classification method used for coloring objects in the map. Jenk's natural breaks optimizes the breaks between classes for a set number of classes by minimizing the sum of squares error term.
Nearest neighborA way to define spatial weights for point datasets. You can define a set number of neighbors for each point location and then specify how SpaceStat should weight those neighbors.
Neighbor orderSee neighbor rings.
Neighbor relationshipsSee spatial weights.
Neighbor ringsThis is a way to define spatial weights for polygon datasets. The first neighbor ring of a polygon is those polygons that touch its borders. The second neighbor ring of a polygon is those polygons that touch its first neighbor ring. This concept is also known as neighbor "order" with first order neighbors touching, second order being your first order neighbors other first order neighbors (minus the first order ones), etc.
Non-inclusiveA time stamp is non-inclusive if it is valid up to but not including the start or end time. A time stamp is inclusive if the time stamp or range of time the data is valid includes the start or end time.
Null hypothesisThe null hypothesis (Ho) describes the spatial pattern expected when the alternative hypothesis is false (e.g. CSR).
Null spatial modelThe null spatial model is a mechanism for generating the reference distribution. This may be based on distribution theory, or it may use randomization (e.g. Monte Carlo) techniques.
Numeric dataNumeric data consist of numbers where the difference between the numbers has meaning. Numeric data can be integers or decimals.
Object overlapThis is a method for setting how objects will be selected in the map. For overlap, the selection rectangle just has to cross into an object to select it. This takes more calculation than Minimum bounding rectangle or center containment and is therefore slower.
OutlierIn the box plot, an outlier is any points more than 1.5 times the interquartile range from the quartiles. In spatial analysis, it is a location significantly distinct from its neighbors (such as a high-low or low-high location identified in a Local Moran analysis).
P-valueThe probability that the observed test statistic was drawn from the null distribution, or the probability that the null hypothesis is true given the observed statistic.
Qualitative color schemeThis is a color scheme for categorical data, where the data are arranged in different classes but not ranked in any way. A qualitative scheme has a different fill color and/or pattern for each unique value.
QuantileA classification method used for coloring objects in a map. The quantile method divides the data values into a set number of classes, allocating the mapped objects evenly within the classes.
Queen adjacencyPolygons are neighbors by the queen criterion if they share any common border at all, whether it is a single point (like chess squares that only touch at a vertex) or a longer shared border (such as chess squares that share an entire side). Compare to rook adjacency.
RampA gradient created between two or three colors in map color schemes.
RandomizationA procedure for obtaining a reference distribution for the evaluation of a test statistic.
Reference distributionThe reference distribution is the distribution of the test statistic when the null hypothesis is true.
Rook adjacencyPolygons are neighbors by the rook criterion if they share a common border larger than a single point. Compare to queen adjacency.
Self-intersectingA polygon is called "self-intersecting" when two or more of its borders intersect anywhere except their endpoints. Make sure to prepare your data with a GIS data editor so that it contains no self-intersecting polygons.
Sequential gradient colorThis is a color scheme for numeric data that is continuous, rather than classified. For sequential gradient color, you specify the minimum and maximum colors and SpaceStat ramps between them to provide intermediate colors for intermediate values.
Significance levelA probability threshold used for evaluating a null hypothesis.
Spatial autocorrelationThe time stamp is the time (an instant or an interval) that an object or attribute is valid. The time stamp for an object is specified when the data is imported.
Spatial data formatThe type of spatial data: points, lines, polygons, or raster format. SpaceStat currently accepts points, lines, and polygons but not raster (image) data.
Spatial weightsWeights are often used in spatial statistics to define which locations have influence on other locations. Often near neighbors or adjacent polygons are given high weights and distant locations are given weights of zero.
SplitTo break apart a category into two or more groups, used in qualitative color schemes for maps.
SPSA *.sps file is a project file that was created in SpaceStat 2.0.
StandardizeYou can standardize data or spatial weights in SpaceStat. Standardization of data makes datasets with different ranges or variability more comparable. Standardization of spatial weights makes locations with different numbers of neighbors more comparable.
Step sizeThis setting is used to control the jumps in an animation. For synchronous data that changes only at a specified interval, it might be easiest to set the step size to that interval.
STISSpace-Time Intelligence System, an information system that goes beyond a Geographic Information System (GIS) by including time as a dimension of the data rather than an attribute. SpaceStat 2.0 is built on the space-time platform first introduced in STIS. Projets created by this software can also be opened in SpaceStat 2.0.
String dataString or text data can consist of letters and/or numbers. If it is numbers, differences between the values do not have mathematical meaning (they can be descriptors for different categories).
STSAn *.sts file is a project file created our space-time intelligence system. It may contain all of the space-time datasets, maps, plots, and analysis results you created in working with STIS. SpaceStat 2.0 and 2.1 can open and use these projects as well.
SubsetA subset geography is a portion of another geography that has been split off. Subset geographies may or may not have all of the datasets of their parent.
Synchronize viewsYou can time-sync views in SpaceStat so that they animate together.
SynchronousSynchronous space-time data is where objects' positions or attributes change at the same times for all objects in the dataset. Compare to asynchronous.
Test statisticThe test statistic quantifies a relevant aspect of spatial pattern (e.g. spatial clustering, overlap).
Text dataSee string data.
Time seriesFor time series data, the attributes of objects vary over time asynchronously. Time slice data is for data that varies synchronously (perhaps sampled regularly at all locations). There may be several rows of data for a single object in a geography. Compare to time slice.
Time sliceTime slice data is referenced to a particular time or time interval on import, no time information is contained in the file itself. Compare to time series.
Time stampA spatial pattern that arises when the value of a variable at one location is related to its value at nearby locations. Positive spatial autocorrelation means that nearby areas have similar values, indicating spatial clustering.
Time-linked viewSee synchronize views.
Title barThe top bar on a window. The bar often has a title for the window, minimize, maximize, and close buttons on it. Docked windows do not have exposed title bar.
Upper tail p-valueA P-value obtained by comparing the test statistic to the end of the reference distribution where the statistic's values are highest. It provides an estimate of how likely values greater than the test statistic are, based on the null hypothesis.
Z-scoreA form of data standardization that takes the raw data value, subtracts the mean and then divides this quantity by the standard deviation. A z-score of zero indicates that the raw value is at the dataset mean, positive above the mean, and negative below.
© 2014 BioMedware