Computer currency. (SPSS 7.0 for Windows; SNAP Professional)(Evaluation)

Marketing research has been the province of specialists who own sophisticated tools and know how to employ them. Over time, survey research and statistical analysis software have become much easier to use. Tools have advanced to the point where they are more affordable and within the reach of smaller organizations without a dedicated marketing research department. In this issue we examine two advances, SPSS 7.0, a recently updated release of the familiar statistical package, and Snap, a survey research tool.

Despite the advances, field researchers still go out to consumers to collect marketing data. However, now they are much more likely to carry notebook computers than traditional notebooks. We also look at an update to the familiar LapLink program used for transferring data from a portable to a mainframe system.

SPSS 7.0: the latest Windows 95 update of a pillar of research

If one is looking for the convenience and ease of use of spreadsheets, the graphics of a word processing packages, and the power of sophisticated mainframe statistical packages, one can find all of these in SPSS 7.0 for Windows. It is one the newest programs for basic as well as advanced statistical needs. Although SPSS has been around as long as I can remember, the program has undergone some major improvements.

The initial improvement toward greater user friendliness resulted from the introduction of SPSS for Windows; now the SPSS 7.0 for Windows has received a major facelift. It continues to be one of the most comprehensive, versatile, and user-friendly statistical software. It now provides high-quality ready-to-paste tables and graphics.

SPSS Inc. provides users a comprehensive set of statistical (data analysis) tools for analyzing virtually unlimited amount of data. These programs are invaluable for professional researchers, business analysts as well as students.

SPSS programs

SPSS offers a variety of statistical programs, some are grouped in bundled and others sold as stand-alone programs. An example of latter is CHAID (Chi-square Automatic Interaction Detection) that provides a tree-like output useful for segmentation purposes. Most programs, however, are available in bundled packages. As an overview (not a detailed exhaustive listing), the contents of major bundled packages are described next.

Base system. The base system includes a wide variety of simple to moderately complex statistical programs as well as presentation quality graphics. These include: descriptive, crosstab, T-tests, ANOVA (one-way and simple factorial), correlations, regression, non-parametric, multiple response analysis, and a chart facility that includes most standard types of charts, as well as some statistical charts such as P-P plots, Q-Q plots and time series graphs.

Advanced statistics. This bundle is comprised of general linear models, loglinear analysis, multivariate analysis of variance, constrained nonlinear regression, probit regression, cox regression, and actuarial survival tables.

Professional statistics. This group is made up of cluster, discriminant, factor, multidimensional scaling, proximity and reliability analysis.

Other programs. In addition to above procedures, SPSS has other optional programs. These are: SPSS CHAID, SPSS exact tests, SPSS categories (conjoint, optimal scaling and correspondence analysis), mapinfo, neural connection, trends (forecasting), and AllCLEAR III.

Installation

The programs come on disks or CD and disks. I received the Base program on a CD and other programs on disks. The installation is easy through the "Start, Run" command in Windows 95. The menu guides the user through complete installation. When adding additional programs, it is not essential to reinstall the programs that have already been installed.

Data compatibility

SPSS can read data from spreadsheet files (Lotus, Excel, SYLK), database files or text files, etc. For raw data entry, the DATA Editor looks like a spreadsheet and can be used for entry, browsing, editing, etc.

Data editor

The data editor has the appearance of a spreadsheet but is well suited to the specialized task of analysis. For example, "Data, View, Value Labels" command (or clicking the label button on menu bar) switches the display of data from raw values to value labels. The cell values can be changed by retyping the new values. The variable names, labels, values, etc, can be changed just as easily through the "Data, Define Variable" command. One of the features of SPSS is the ease in which values can be recoded. In survey research it is common to have variables that are reverse scored. These can be changed through the menu "Transform, Recode" command.

The editor allows for automatic movement of cursor to the next field. However, if one is to make a mistake in the middle of a case, it is not possible to invoke the automatic cursor movement to the next field in the middle of a case. That case has to be reentered. The data editor does not allow one to specify the range of permissible values so as to prevent entering an out-of-range data value. This is not a major problem, as one can (and should) check the out of range values after data entry using the "Descriptives" command.

Data transformation

SPSS provides a large set of transformation options. Included in these …