A Five-Decade Retrospective

“My 55-year path to retirement at WSU.”

After mentioning to a graduate survey design class in the late 1990’s that I had been on the WSU sociology faculty since 1969, I was taken-aback by one student’s quick response, “That’s before I was born!” Mentioning that date again in a recent seminar drew a response that my arrival preceded the birth of a student’s grandparents. It’s been a wonderful 55 years at WSU, but I think it is time to retire, despite continuing to enjoy getting up each morning to see what’s new in survey methodology.

In 1969, I was hired as a half-time assistant professor in the former College of Agriculture’s Department of Rural Sociology to research and help rural communities. At the same time, I was also hired half-time in the former College of Sciences and Arts sociology department to do research and teach students about the ways communities influence people’s lives.

Starting my WSU professorial life was challenging. I was assigned to an unused office in the Department of Agricultural Education in Wilson Hall and given a hand-me-down manual typewriter by the rural sociology departmental secretary who had just obtained a new one. I began teaching Sociology 101 to 180 students in a packed Todd Hall classroom with only a blackboard and chalk to elaborate verbal instructions. The textbook had been selected before I arrived and was not one I had seen before my arrival in Pullman.

I learned quickly to survive the fumes emitted in a closet-size room when cranking out handouts and tests on the spirit duplicator invented in 1923. My contact with distant friends and colleagues was through the U.S. Postal Service and expensive long-distance calls. Personal computers did not exist, and use of the mainframe required formal submission of requests. Dare I say that things moved a bit slowly in academia at that time.

In contrast the direction of my career changed quickly and unexpectedly after my first semester. Students marched on the Office of WSU President Glenn Terrell, demanding that that he do something about the Cambodian invasion that was expanding the controversial Vietnam War. They also went on strike against attending classes. The protest soon wound down, partly because there was little of significance that the President could do about the unpopular war. However, those concerns were quickly replaced by a set of 11 demands from Black students to address racism on campus that received strong support from students and many faculty.

Observing these protests, my sociology chair went to President Terrell offering to do telephone surveys of students to find out what they were thinking.  It was reasoned that knowledge could help University administration understand and make appropriate decisions to resolve the crisis.

I was asked to create a data collection facility in the g­raduate college’s new Social Research Center (SRC) using telephone and mail surveys. Suddenly I had appointments in three different colleges. And I was tasked with conducting surveys by methods that some survey designers considered inadequate.

The surprising outcome of this work was that these methods worked far better than any of us expected, obtaining very high response rates and data quality. We were asked to expand our survey work to the public and other groups where good results were also obtained. These results fueled the demand for providing survey help to MA and PhD students as well as faculty and groups outside the University. (Dillman, 1978)

The science of surveying is straightforward. It is possible to estimate with known statistical precision the attitudes and behaviors of millions of people by successfully surveying only a few hundred or thousand randomly selected people provided four sources of potential error—coverage, sampling, measurement, and nonresponse—are kept low. (Dillman, 2000). The biggest impact from creation of the SRC mail and telephone capability beginning in 1970 was that it became possible to do community, state, and regional surveys.  At that time, in-person interviews were considered by methodologists the only acceptable way to do sample surveys, which was too costly and slow to do in-person interviews for local issues.

Developing the WSU Social Research Center (later to become the Social and Economic Sciences Research Center or SESRC) led to us advising and conducting surveys for clients both inside and outside the University. Today the SESRC employs over twenty staff members and dozens of student employees trained in the specialized technical skills necessary for conducting a wide variety of local, state, and national surveys. Nearly $80 million in grants and contracts have been received over the last 55 years making it possible to identify and contribute to solving practical problems for local, state, and sometimes national organizations.

The SESRC became my research laboratory that allowed the near constant testing of innovative ideas that the changing times required. These efforts could not have happened without the presence of SESRC staff with specialized skills for implementing surveys and Department of Sociology graduate students who helped me design studies. Details of our joint work is described in “Fifty Years of Survey Innovation” (Dillman, 2022) and include these decade-by-decade efforts:

1. 1970s: Creation and publication of detailed procedures for conducting successful mail and telephone mail surveys consistent with the available technology (Dillman, 1978).
2. 1980s: Researching the likelihood that new information technologies would change rural communities (Dillman, 1985) and developing a compelling case that survey methods would also be significantly changed by the availability of personal computers and internet connectivity.
3. 1990s: Working with U.S. Census Bureau under loan from WSU through the Intergovernmental Personnel Act to provide leadership for trying to make the 2000 Decennial Census Questionnaire and implementation methods respondent-friendly. This experience contributed to the development of effective internet survey methods (Scheaffer and Dillman, 1998) and revision of the 1978 book that advocated the use of mixed-mode surveys relying on multiple contact and response methods to improve survey quality (Dillman 2000).
4. 2000s: Conducting dozens of SESRC experiments and consultations financed by the National Science Foundation and U.S. Department of Agriculture that led to understanding why survey questions presented visually instead of aurally produce different answers and adoption of the need to write questions the same across survey modes (Christian and Dillman, 2005), thus providing the conceptual basis for the creation of web-push survey methods as a replacement for telephone and in-person interviews (Smyth, Dillman and Christian, 2010).
5. 2010s: Publication and dissemination of detailed methods for doing web-push surveys that improved survey response rates from various kinds of respondents, while keeping measurement the same across modes (Dillman, Smyth, and Christian, 2014). Web-push methods are now being used for government censuses throughout the world and as a replacement for telephone and in-person interviews for other general public surveys. (Dillman, 2017).

Each of the early accomplishments—mail and telephone data collection, understanding the potential of the internet, being able to explain why persistent differences occurred in how people answered verbally vs. visually delivered questions—provided building blocks for finding effective ways of combining different modes of contact and response to improve response rates and data quality. The resulting web-push methodology provided a much-needed replacement for telephone and in-person survey methods previously considered essential for survey data collection.

Remaining a WSU professor for so many years was not something I planned or thought much about. It happened because there was always something else that needed to be tested, reported, and integrated to produce quality surveys. And I was conscious of how discoveries constantly suggested new possibilities. Pursuing these issues has involved publishing nearly three hundred articles and reports. It also resulted in publishing four editions of what became known as the Total Design Methodology (TDM) book on survey design that integrated the results of individual experiments. (Dillman, Smyth and Christian, 2014).

As work on developing web-push methods slowed down and the pandemic left me at home for an extended time, I used the opportunity to write papers on topics I had not previously found time to investigate. These efforts included a paper on why survey response theories need to change (Dillman, 2020) and directly testing the relative power of different theories for explaining survey response rates in populations with low education levels (Greenberg, and Dillman, 2023). In addition, the stay-at-home period provided the opportunity to work virtually with surveyors in many other countries as they tried to adapt web-push methods as a replacement for in-person and telephone surveys. The adoption of modern technologies and methods takes time, and I appreciate having university support to do the outreach necessary for encouraging adoption of these methods.

The joy of working with sociology graduate students and SESRC staff to invent modern survey data collection and observing those methods being adopted by others has been the most satisfying aspect of my WSU career. Being loaned to the U.S. Census Bureau Director’s office as a senior survey methodologist was a key development in getting to understand the larger survey world. Methods I worked on improved response and have continued to be used during the 2010 and 2020 Censuses. The Census appointment also connected me to dozens of other government and private organization surveys in ways that changed their methods and produced breakthroughs in how visual and oral communication produce different answers to survey questions (Dillman, 2020, Dillman under review).

Being part of the WSU sociology department has been more than research and teaching. In 2012 I sensed that the history of the department was being lost to the digital age, as old paper files were being discarded. A graduate student shared the same concern and together we created Sociology News, now in its tenth year of publication. It helps us keep track of both student and faculty accomplishments, with room for reports from former faculty and students that give perspective on departmental goals and accomplishments, which seem likely to otherwise be lost.

I am grateful for a stimulating career at WSU, having appointments from two or sometimes three different colleges or offices during each of my 55 years. Those appointments have helped me serve WSU’s land-grant mission by simultaneously teaching, researching, and turning that research into practical recommendations.

I cannot imagine a more exciting time to be a sociologist and survey methodologist. The bases of social interaction at the core of sociological thinking are changing dramatically from in-person to electronic. The influence of geography on interaction concepts is neither settled nor understood. And the speed of change continues to accelerate.

When reflecting on my years at WSU I am struck with how the survey design issues in need of investigation kept changing, forcing me to learn new ways of thinking. My advice to PhD students is that the half-life of the knowledge gained in graduate school, and my classes, will dissipate far more quickly that happened for me, requiring them to constantly update themselves to be effective scientists and practitioners into the mid-21st century, regardless of the career track they pursue.  -Don A. Dillman