Perform basic descriptive statistical analyses, modifications or tabulation of data in the dedicated utility working with large numerical amounts and featuring multiple algorithms for specific operations. Output data is accessible for further updating and management.
EpiData Analysis is based on the same principles as EpiData Entry. If you properly define, document and verify data with EpiData Entry, the definitions are also available in EpiData Analysis. E.g. specified legal values with attached text labels(1 = No 2= Yes) or definitions of missing values.
When reading data EpiData Analysis will do some control based on variable definitions, e.g. all dates are verified.
EpiData Analysis is suitable for small as well as rather large datasets. simple datasets like one questionnaire as well as datasets with many or branching dataforms. The principle of EpiData is rooted in the simplicity of the dos program Epi Info v6, which has many users around the world.
EpiData Analysis is used for: Basic descriptive Analysis of quantitive data, Defining and modifying data, Editing / correcting data already entered, Graphing Data, Asserting that the data are consistent across variables and printing or listing data for documentation of error-checking and error-tracking.
v2.2 [Mar 12, 2009]
- Can read EXP formatted data files (new form of EpiData XML data structure)
- Odds Ratios are now Standard Mantel Haenzel with Breslow CI's
- For Exact Odds Ratios option /oex should be used - to get same results as previously
- RunCharts added counts and skipped points
-Histogram Now groups data into bins.
- /CLOSE added to Generate command.
- Added: /FONTSIZE and /SUBFOOT added to graph commands.
- Xbar-S Charts - Common Sigma-bar shown Instead of varying by subgroup size
- Documentation in commands.htm clarified and extended
- Run Charts - expected runs for observations: 31-40 added
- Documentation in commands.htm clarified and extended
- Table Statistics extended with Confidence Intervals for Gamma Coefficient
- Added: Graph dialog moved XLabel to "Variables" tab page.
- Add: Error is printed in PChart if Denominator is larger than Nominator.
- SPC template extended with Xbar-R
- Gchart - value of first data point indicated on X-axis
- A first b165 downloaded April 1st-Apr 2nd did not have this feature
- Using freeze: /F= with a date type x-axis variable now uses data point no. from /F
- /f freeze option functioning in pchart, uchart, xbar and gchart
- See all changes below in builds since last public release
- Result Variables for Means command
- Default windowing changed to width 700px
- Added result variables for means
- Changed default decimal count for MEAN vector created by aggregate.
- Option /adj documented for lifetable
- Xbar adapted to accomodate where a given x-value contains only one observation
- Option /noXlabel /noYlabel added to all graphs and graph dialog
- Graph dialog rewritten
- SPC command options extended - see command reference
- SPC graph commands rewritten to a new common internal form
- Added SPC template (spc.htm) to graph sub menu.
- SPC command syntax simplified (no need for time variables)
- Added preliminary SPC template to testdata (spc.htm)
- New definitions of set commands, can be focused to each command
- Added preliminary SPC template to testdata (spc.htm)
- First implementation of 2 additional SPC warning tests (test4 and test5 in Ichart)
Bugs Fixed :
- Mismatch of /subfoot and /Noinf in SPC graphs leading to non-showing of estimates for periods (solved)
- Graph dialog reshaped to show weight variable always (was sometimes hidden)
- Removed Sigma3 from legend in runchart command.
- Freq sorted incorrectly by default, and using /SA or /SD.
- Variable name with "if" in, cause pre-processor to incorrectly insert "set option" text the wrong place.
- Labelvalue command gave no error with incorrect construction.
- Use of /W in tables command deleted wrong variable in some instances
- TAble dialog /RR option adapted (but still not working with /FV)
- SPC run charts - fixed error, expected runs included median
- Frequency table sorting not working
- Gamma coefficient not printed in certain tables
- Selection of RR in table dialog
- Internal error in handling number of records
-Could lead to failure in merge (and other special cases)
- The X-axis on dotplots only used 75 percent of the space
- Incorrect MEDIAN in Runchart with few decimal points in measurement.
- /SAVE in graphs did not respect given path
- XBar failed with subgroup count > 20 due to factorial miscalculation.
- The ½ point offset for sigma/center lines in SPC charts was incorrectly calculated in some special cases.
- Fixed: Issue where a value of 0 (zero) in the denominator of PChart would crash program.
- Table crashed if multiple valuelabels existed with same labels, but different value.
- Incorrect MEDIAN in Runchart with few decimal points in measurement.
- /SAVE in graphs did not respect given path
- XBar failed with subgroup count > 20 due to factorial miscalculation.
- The ½ point offset for sigma/center lines in SPC charts was incorrectly calculated in some special cases.
- Date variable accepted for aggregating in xbar spc charts
- Sparse data table problem with correct cell allocation (stratified tables)
- Other minor bugs fixed - see Mantis database
- No data exception with bar and histogram
- Reference to old variable names after close (fixed)
- Other minor bugs fixed - see Mantis database
- Text on interface of SPC submenu adapted
- Issue with exit
- Parsing of certain statements
- Several minor issues (aggregate variable naming,percentiles ...)
- Several minor issues - see Mantis system
- Fixed width of browser columnds to accomplish dates and labels (incl. update)
- Fixed menu issue when using EpiData Analysis in Linux under "Wine"
Mall Creeps
