RSS Matters

Link to the last RSS article here: Moderators and Mediators  - Ed.

Using S-Plus Graphics in Excel; SPSS News

By Dr. Mike Clark, Research and Statistical Support Services Consultant

We were running a bit behind this go around so here is something put together quickly for those that use excel, possibly for initial data entry, but still might want a quick graph S-Plus style.  If S-Plus is installed on a machine with Excel, an S-Plus drop down menu is made available when one is running Excel.  Essentially you can take what you have in Excel and still have the S-Plus functionality for graphics.

Here is Excel but with the menu for S-Plus: 

We will be using data from regarding SPSS stock prices since 1993.  To create the graph, specify the column name and which cases to include for each variable and separately, as in the following.  Separate one variable from another with a comma.

After that, there is the option to specify a conditioning variable should you need to, and then you just have to note with is the very first case of the data.  Now you have the S-Plus graphical options at your disposal.

Once you have the graph you then have some options you can play around with to get the look you want.

The finished graph shows that maybe SPSS isn’t doing too hot these days.  Time for a new release?  Yeah why not?  The following is from spss.com:

New Version of SPSS for Windows

"The new version of SPSS for Windows is a major upgrade. SPSS 13.0 includes an array of features that people will use often in their daily work."

— Bob Muenchen
   Manager, Statistical Consulting Center
   The University of Tennessee Office of Information Technology

Graphics and output

• SPSS 13.0 provides you with better reporting capabilities through improvements to the presentation graphics system (added in SPSS 12.0). These enhanced reporting capabilities give you the ability to:
  • Include three new chart types in your work
    • Population pyramids, also called mirror plots or dual charts, for creating graphs that clearly show the distribution between two groups
    • 3-D bar charts for creating charts that highlight differences across combinations of categories
    • Dot charts, also called dot density charts, for displaying individual observations on a continuous scale using a dot or other symbol. This enables you to represent the frequency distribution of your data while displaying all individual points or observations.
  • Better explain your data when you use new chart display features/options in your graphs
    • Create paneled charts for displaying rows and variables from more than one chart in a table-like arrangement of sub-charts, showing differences among groups, and visualizing effects of conditioning variables. Paneled charts are available for most SPSS graphs.
    • Include error bars on categorical charts for displaying information about the variability of measurement. You can use error bars to represent confidence level, standard deviation, or standard error spread measures. Add error bars to nearly any categorical chart, including bar charts, area plots, and line plots
    • Use the sort categories display option to automatically reorder categories in differing order (descending or ascending) or by different sort methods (value, including dates, label, or summary statistic)
    • Drag data value labels to any position on your chart, add connecting lines, and match font color to subgroup
    • Add diagonal reference lines to compare your data to a diagonal line. Also, drag horizontal and vertical reference lines to any position on the axis
  • More easily work with templates and the Chart Editor
    • Browse the template settings in the new tree-view layout; then explicitly select the settings you want to save in your chart templates
    • Quickly modify your charts using new features in the Chart Editor interface. You can automatically add titles, footnotes, and gridlines; drag annotations to any position on the chart; and more intuitively select objects.
  • Improve the presentation quality of your results. Pivot table output is now available for a number or procedures, including AREG, CURVEFIT, KAPLAN-MEIER, MULT RESPONSE, NLR, and CNLR.

Data and output management

• More powerful data management in SPSS 13.0 allows you to:
  • Ensure data containing text strings of more than 256 bytes is not truncated or lost when working with open-ended question responses, data from other software that allows long data strings (up to 32,767 bytes), and other types of long text strings
  • Easily work with dates and times in SPSS using the Date and Time Wizard. You can calculate with dates and times, bring date/time data from a variety of sources to SPSS, create a date/time variable from a string containing a date variable, and parse individual date/time units.
  • More quickly and easily understand wide and long datasets using splitter controls in the Data Editor
  • Automatically convert string variables to numeric variables using the improved autorecode command
    • A recode template enables you to append, for example, new products to an existing scheme
  • Save aggregated values directly to your active file—in just one mouse click—using the improved aggregate command
  • Create custom programs in SPSS 13.0, even if you have little or no experience using syntax in SPSS. SPSS 13.0 provides Output Management System (OMS) functionality through the interface.
  • Conveniently share data between SPSS 13.0 and SAS ® 9.0 because SPSS 13.0 can read/write SAS 9 files
  • Run the majority of your syntax jobs uninterrupted, even if you encounter an error in your job. The new INSERT command enables you to specify interactive syntax rules that tell SPSS 13.0 to bypass errors.
  • Keep a log of your work while maintaining a high level of performance. SPSS 13.0 provides improved system performance by optimizing the way SPSS writes syntax to journal files.
  • Use the HOST command to enable applications to “escape” to the operating system and execute other programs in sync with the SPSS session
  • Obtain flexibility for your version-dependent applications because automation scripts in SPSS 13.0 can recognize configuration parameters, including UI language, options installed, and the SPSS version being used
  • Prevent syntax jobs from breaking when you create a common, or main project, directory that enables you to include transformations for multiple projects using a new Change Directory command

Output export enhancements

• With just a few mouseclicks, you can export tables and charts directly from SPSS 13.0 to Microsoft® PowerPoint®

"Overall, I think the new feature set is a major improvement in an already outstanding product. "

— James W. Golden, PhD
   Associate Professor
   Department of Criminal Justice
   University of Arkansas at Little Rock

New SPSS add-on module: SPSS Classification Trees™ 13.0

This new add-on module creates classification and decision trees directly within SPSS to identify groups, discover relationships between groups, and predict future events. With SPSS Classification Trees, you can:

• Use classification and decision trees for segmentation, stratification, prediction, data reduction and variable screening, interaction identification, category merging, and discretizing continuous variables
• Visually explain categorical results. Highly visual trees enable you to present results in an intuitive manner—so you can more clearly explain categorical results to non-technical audiences.
• Use results to enhance existing and new data. You can identify a particular subset of the data via the tree and run further analysis, segment and group cases directly within the data, and create predictive values and probabilities within your dataset. Write information from the tree model directly to your data or create XML models for use in SPSS Server 13.0 to score other data files. You can also generate selection or classification/prediction rules in the form of SPSS syntax, SQL statements, or simple text (through syntax).
• Perform analysis using one of four established tree-growing algorithms: CHAID, exhaustive CHAID, classification & regression trees (CRT), and QUEST. You can try different types of tree-growing algorithms and find the one that best fits your data.

SPSS Complex Samples™ 13.0 enhancements

Use predictive analytics for survey research. SPSS Complex Samples 13.0 extends this module, which was new in SPSS 12.0, enabling you to:

• More accurately analyze and predict numerical outcomes from your complex sample designs. Use the new complex samples general linear model (CSGLM) to build linear regression, analysis of variance (ANOVA), and analysis of covariance (ANCOVA) models to predict numerical outcomes.
• More accurately analyze and predict categorical outcomes from your complex sample designs. Use the new complex samples logistic regression (CSLOGISTIC) statistic for binary and multinomial outcomes.

SPSS Tables™ 13.0 enhancements

Better control how you display your data using expanded category options in SPSS Tables by:

• Sorting categories by any summary statistic in your table
• Hiding the categories that comprise subtotals-you can remove a category from the table without removing it from the subtotal calculation

SPSS Categories™ 13.0 enhancements

A new procedure in SPSS Categories greatly enhances this add-on module's functionality for multiple correspondence. This procedure enables you to:

• Reveal the underlying relationships between more than two nominal variables when similar categories are grouped close to one another in a chart. Market researchers typically refer to these types of analyses and charts as "perceptual maps."
• Use the new multiple correspondence analysis functionality for object weighting, data binning, supplementary object inclusion, and correlation matrix
• Use the expanded multiple correspondence analysis functionality for missing data, diagnostics, plot, output, and export

SPSS Regression Models™ 13.0 enhancements

New options available for multinomial logistic regression (MLR) in the SPSS Regression Models add-on module enable you to:

• More quickly reach results when you have a large number of predictors by using Score and Wald methods for stepwise selection in MLR
• Assess your model fit using Aikaike information criterion (AIC) and Bayesian information criterion (BIC; also called Schwarz Bayesian Criterion, or SBC)

SPSS Server 13.0 enhancements

SPSS Server continues to help your organization increase its productivity by giving you the ability to:

• Sort and aggregate data within the database prior to retrieval
• Increase performance for select users when you assign temporary disk location and user priority through new administrative tools
• Achieve high-performance sorting using SyncSort® (must be purchased separately) with dynamic optimization, parallel processing (with multiple processors), and patented performance algorithms
• Use XML models created through SPSS to score individual or new cases and data.

Return to top