Our strength is the analysis and reporting of survey data

We became famous as a crosstabs and banner tables package StatPac became famous as a crosstabs and banner tables survey software package. Researchers usually bought the software for that purpose but quickly discovered that the other survey analysis procedures in StatPac had equally impressive features. StatPac is generally regarded as a "power-house" of survey software.

 

You might be surprised to learn that StatPac survey software was designed by it's users. We released the first version over 30 years ago. It was primitive by today's standards, but when a user made a suggestion or noted a limitation, we listened and the software was improved. After three decades of enhancements, our survey software has evolved so it does pretty much everything that researchers want. Nowadays, we don't receive many suggestions, but we still encourage users to tell us how to make the software better.

 

When you try the demo, please let us know of anything that appears to be missing or could be improved. 

General survey analysis features

Easy to learn programming language

A programming language is used to design tables and analyses. Usually, user's only have to use a handful of keywords to get the results they want. While there are many keywords, most users will never need more than a few. StatPac uses a programming language to design analyses

 

Commands you'll probably find useful

 

Commands you might never use (but they're available if you need them)

Average Dummy Let Run Sum
Count Footnote Merge Sort Weight
Difference Lag Normalize Stack Write

Most common statistical analyses procedures

Nearly all surveys use four types of analyses: frequencies, descriptive statistics, crosstabs, and listing open-ended text The four most common analyses are frequencies, descriptive statistics (mean, median, standard deviation, etc.). crosstabs and banner tables, and listing open-ended text. The keywords to run these analyses are frequencies, descriptives, banners, and list. They can be abbreviated as FR, DE, BA, and LI).

 

FREQUENCIES provides counts and percents for categorical data. You get clearly labeled tables and presentation-quality graphics. Show cumulative percents and confidence intervals (sampling error). Set the  percentage base, labeling, missing cases and other output options. The procedure handles multiple response and codes open-ended questions. Several display formats are available.

 

DESCRIPTIVES include measures of central tendency (mean, median and mode), measures of dispersion (variance and standard deviation), confidence intervals (to assess reliability), measures of normality (skewness and kurtosis), and quartiles.

 

BANNERS displays the relationships among several variables by combining multiple crosstabs into one table. You have complete control over the format and labeling of the table. Print counts, percents, totals, means, and standard deviations anywhere on the table. Use significance testing to identify differences between subgroups.

 

LIST displays open-ended text. Surveys usually have questions that ask for an open-ended response. The LIST command lets you display those responses verbatim.

 

There are other kinds of basic analyses (e.g., t-tests and correlations) but they are used far less frequently.

Multivariate analyses are available

StatPac doesn't stop there. Advanced users will find a complete selection of multivariate analyses techniques: analysis of variance (ANOVA), linear and non-linear regression, stepwise multiple regression, probit and logistic regression, canonical correlation, principal components analysis, factor analysis, cluster analysis, stepwise discriminant function analysis, and perceptual mapping.

 

Don't get worried. Most of our users never need or use advanced multivariate statistics. Just know they're there if you do need them.

There's only one way to find the best survey software for you

It seems like no matter how often we warn people, most purchases of survey software are based on the hype they read on the publisher's Web sites, and then after the purchase, they're disappointed.

 

Please, do your homework. Pick a few packages you think might work for you and then try them. Not just a cursory look, but actually try them with one of your surveys. Give the software a complete workout. It will take a few days, but it's the only way to know if the software will do the job for you. You can get a copy of our demo survey software here.