Apple eNews - Finding Statistical Order with SPSS

By Michael Suh
“All of us in psychology and in science who have remained committed to the Mac platform have just been very ecstatic that SPSS brought it back,” says Dr. David Fresco. He’s talking about SPSS 10.0 for Macintosh, the first Mac version of SPSS’s top-selling statistical software since 1995. For Dr. Fresco, a clinical psychologist and postdoctoral fellow at the Adult Anxiety Clinic of Temple University (AACT), the return of SPSS to the Mac is a welcome event for at least two reasons:

	•	he conducts “easily 90%” of his statistical research on SPSS and
	•	his computing platform of choice since the 80’s has been the Macintosh (his first Mac was the Plus)

Developing a better understanding of Generalized Anxiety Disorder (“the tendency to worry chronically”) is a major focus of Dr. Fresco’s research—as is developing a treatment “from the gound up” that will help those afflicted with the disorder.

“We’re beginning to wonder whether anxiety in and of itself is a risk factor for depression,” Dr. Fresco says. “So if we can nip the anxiety in the bud, we may help individuals experience less depression in their lives.”

Statistical analysis plays a significant role in almost all of Dr. Fresco’s research, and he uses SPSS to engage in the various forms of statistical interpretation and analysis he needs to conduct. “Sometimes it’s as simple as looking at the relationship between two variables, using what’s called a ‘correlation coefficient,’” he explains. “As one changes, how does the other change?”

He also uses another statistical procedure, “regression analysis,” to find out “how well a number of factors predict some other outcome or a score on something else.” SPSS 10.0 performs these tasks “very well,” he says. “It’s a very feature-rich statistical package.”

One of Dr. Fresco’s primary responsibilities at the AACT is to conduct data analysis for the various studies.

Recenty, he put SPSS through its paces by analyzing the data from another large-scale study to determine the effectiveness of various treatments for Social Anxiety Disorder. Patients who participated in the study were randomly assigned one of four kinds of treatment: psychotherapy, medication, psychotherapy with medication, or a pill placebo. After receiving their assigned treatment for a period of 12-52 weeks, it was then discontinued.

“We analyzed data using SPSS to determine which individuals, across the course of treatment, were getting better and which individuals were not,” Dr. Fresco explains. Those who showed improvement after treatment were then monitored for another year without treatment to determine which of them maintained their gains and which ones deteriorated. “We used SPSS to address all those questions,” Dr. Fresco says. “And we used SPSS to determine which treatment seemed to be providing the highest rate of improvement.”

“I find it an easy statistical package to use,” says Dr. Fresco. He particularly appreciates how easy it is to import data from other packages. “Getting the data from a database management product into the statistics product is often half the battle” he says. And once the data are in SPSS, he values the flexibility of being able to employ “both the point-and-click approach to statistical anaylsis” and the ability to use SPSS’s own syntax to write programs that will conduct the analyses.

Dr. Fresco especially likes look and feel of SPSS 10.0. “There was attention to detail,” he says, “in making it look and conform to all the standards of the Mac interface and in making it feel like a Mac product.

The statistical research that SPSS and the Mac help him conduct plays an integral part in Dr. Fresco’s science. “It’s very important to demonstrate things in an empirical way. Obviously, statistics is nothing more than a tool, and we’re only kind of scratching at the surface of truth, but it’s a very appealing way to approximate truth.”