5
My undergraduate degree was in political science with a minor in history and a second minor in secondary education. Since 1971, I had enrolled in graduate courses at several of the CUNY colleges to fill some of my evening time. Computer science at City College, political science at Hunter College, and social studies at Lehman College. The computer science courses at City College were all mathematics and had little relationship to anything I was doing. I enjoyed the courses at Hunter and Lehman and managed to accumulate enough credits by June 1972 for a master’s degree. For fall 1972, I decided to enroll in another master’s degree program, this one at Bernard Baruch College in public administration. Two colleagues of mine at Lehman, John Antonaccio and Miriam Ramos, were enrolling in the same program. John was the assistant business manager and Miriam was the executive assistant to the dean for adult and continuing studies. The three of us were all thinking that a degree in public administration would help our careers. At Baruch, one of the faculty, Fred Lane, became a mentor for me and guided me well through some of my future choices. We have remained friends for five decades. The program at Baruch was very flexible and I took courses in higher education administration, operations research, and systems analysis. It turned out to be a very valuable degree as it gave me a good sense of administration in general as well as honing my analytical skills. I completed the program in 1975 with a master’s in public administration (MPA).
During my first year at Baruch, important things were going on in the country as well. In November 1972, George McGovern, the Democratic presidential nominee lost to Richard Nixon. McGovern ran as a “peace” candidate but failed to generate large-scale support for his views. He promoted a campaign message of “coming together” and I attended one of his events at a standing-room only rally held at Madison Square Garden. In the election, he only carried the state of Massachusetts and the District of Columbia. On January 23, 1973, the United States and North Vietnam signed the Paris Peace Accords ending any further direct United States military involvement in Vietnam.
On January 22, 1973, the U.S. Supreme Court issued a 7–2 decision in favor of “Jane Roe” (Norma McCorvey) that held that women in the United States have a fundamental right to choose whether or not to have abortions without excessive government restriction and struck down Texas’s abortion ban as unconstitutional. This landmark decision that protected a pregnant woman’s liberty to choose to have an abortion, prompted an ongoing national debate in the United States about whether and to what extent abortion should be legal, who should decide the legality of abortion, and what the role of religious and moral views in the political sphere should be. Roe v. Wade reshaped American politics, dividing much of the United States into abortion rights and anti-abortion groups, and activating major movements on both sides.
Meanwhile back at Lehman, it was in 1973 that I learned the Statistical Package for the Social Science (SPSS), one of the most useful tools ever developed for the social sciences and for the teaching of research methods. My interest in it at the time, had little to do with research and everything to do with reporting student data to the CUNY Central Office. For years, CUNY colleges were required to complete paper documents referred to as Form A (undergraduate students) and Form B (graduate students) every semester. They began as seven- and five-page documents respectively, providing summary information on students. These documents not only could be used for institutional research but were also the basis for the following year’s college budgets. They were expected to be accurate although there was no real way of checking this. By 1973, Form A had ballooned to 35 pages and the reporting was getting more complicated. Student characteristics (gender, age, ethnicity, years in program, major, etc.) were reported in two and three-dimensional tables. Every semester I found myself rewriting the program to generate this report. I dreaded getting the memo each August with specifications/rules for reporting these data. I had first been introduced to SPSS in spring 1972 while taking a course in political science at Hunter College. My professor, Ken Sherrill, required a statistical analysis project using a national database of voting behavior. He taught us the basics of SPSS in order to complete the assignment. Later in spring 1973, I was taking a course in research methods at Baruch College when the instructor, James Guyot, also introduced SPSS, and I made the connection between it and the Form A and Form B reports.
The project to build SPSS was started at Stanford University in the 1960s by three graduate students, Norman Nie, Dale Bent, and C. Hadlai Hull. Although the initial work on SPSS was done at Stanford University, in 1969, Nie joined the University of Chicago’s National Opinion Research Center. The University of Chicago recognized SPSS as an important intellectual property and encouraged Nie’s continuing development of the software system. Nie was successful in recruiting Hull to join him at the University of Chicago by encouraging him to take a position as the head of the University’s Computation Center. Bent, a Canadian, decided not to join Nie and Hull in Chicago, and returned to Canada where he had an academic appointment at the University of Alberta. (About SPSS, Inc., 2009)
The speed with which SPSS caught on was nothing short of amazing. Prior to its advent, social science researchers doing data analysis were typically using manual equipment and calculators. Frequently, these researchers relied on other experts who knew the specifics of statistical procedures and had great dexterity in using manual equipment, since it was one thing to add a sum of numbers on a calculator, and it was quite another to do an analysis of variance. Nie and Hull decided to give the software away for the cost of making a duplicate of the program which initially consisted of a tray of punched cards. However, Nie, Bent, and Hull came to realize that SPSS users would need instructions or a manual to use the software properly. The original version of the SPSS manual, published in 1970, has been described as one of “sociology’s most influential books,” allowing ordinary researchers to do their own statistical analysis (Wellman, 1998). The manual became a best seller as tens of thousands of social scientists took charge of their own research destinies, learning the basics as well as the nuances of data analysis. SPSS provided the classic environment for learning by doing. The software, with the aid of the manual, was very user friendly and a conscientious student learned the statistical procedures as well as the SPSS coding to conduct the analysis.
It is difficult to assess SPSS’s impact on the entire field of data analysis, but in less than a decade it became the mainstay of faculty teaching introduction to statistics in all of the disciplines. Furthermore, it allowed for collecting larger sample sizes because the drudgery of running statistical routines disappeared once the data were converted into electronic form. It was SPSS that changed the way statistical research was conducted. Students and researchers could now collect sample sizes numbering in the thousands. Government agencies could collect sample sizes in the hundreds of thousands and make the data available in SPSS format free to researchers and set the standard for a number of other statistical packages such as SAS and Stata. Agencies such as NASA, the Census Bureau, as well as corporate America became regular users of the software. Versions of SPSS would also be made available for a number of computer hardware platforms (IBM, Univac, and Control Data Corporation). A whole new style of research evolved based on collecting large quantities of data without necessarily having specific hypotheses or research questions and later searching the data for relationships and patterns. This approach has evolved into the present era of “big data”.
I was determined to learn all I could about SPSS and its basic statistical routines. Form A and Form B were crosstabulations or contingency tables that easily fit SPSS output formats. My plan, back in1973, was to no longer write source programs to produce Form A and Form B, but instead to develop a student data file with all of the pertinent variables or data elements and run SPSS routines which would save hours and hours of programming effort. I taught one of the associate registrars how to read the SPSS output and she was ecstatic at learning this new tool. For the rest of my career, I continued to use SPSS. Wherever I went and whatever I did, it became a basic tool for me as an administrator, instructor or researcher. After leaving Lehman, I delivered several papers in the later 1970s, on the use of SPSS for management data reporting at information technology and college administration conferences.
In 1975, Nie and Hull incorporated SPSS into a company and began to sell SPSS commercially. To its credit, SPSS, Inc. always provided student versions at substantial discounts. In 2003, SPSS, Inc. also developed predictive analytics products that were especially popular for marketing applications. In 2009, SPSS, Inc. was acquired for $1.2 billion by IBM which continues to market it under the SPSS...