By Lyle Durbin
For many years, public opinion researchers have relied upon the telephone as a means to gather a random probability sample for general population studies. Until recently, this method had many advantages – for the most part, every household had exactly one telephone line, and these telephone numbers were assigned in such a way that a random selection could be easily obtained to cover any desired geography. Before recent technological innovations such as caller ID, answering machines and cell phones, people would generally answer their home phone, and they were fairly likely to cooperate with an interview. Thus, a perfect sample could be obtained that was projectable to the entire population, and telephone interviewing reigned supreme. Only two alternatives existed – mail surveys and in-person interviews. While perhaps appropriate for the U.S. census, these methods were generally less appealing to most researchers due to a lack of timeliness, cost factors and response rates.
Over the last decade, many technological changes have come about that challenge the traditional stronghold of telephone interviewing. Caller ID and voicemail have lowered response rates. Cell phone only households have surged in popularity, especially among specific demographics within the population (young, no kids, minorities, etc.). The Internet has given rise to alternative forms of communication, with email, texting, and social media replacing conversations previously conducted via the phone. However, the need for research has only grown as advertisers and communicators need more insight into the ever more complicated landscape they must contend with. With this, new forms of research have emerged to challenge the traditional telephone survey – Internet surveys, social media research, address-based sampling, etc. Are any of these alternatives the magic bullet that will replace telephone interviewing?
Let’s start with Internet surveys. This is a cheaper (albeit not as cheap as some believe) and more timely alternative to telephone interviewing. This is an appropriate and useful method when trying to reach a very specific target audience that would be too costly to find via a phone sample (25-34 year old plumbers), or when time is of the essence. However, those selected to participate in online surveys come from opted-in panels. These are convenience samples characterized by a nonsystematic approach to selection as opposed to a probability sample which accurately reflects the true microcosm of the population at large. It is more difficult to confidently make projections to the overall population when recruiting survey participants from online panels.
Social media research involves a more passive approach to gathering opinions. You’re listening to the conversation, rather than asking specific questions. This is a very valuable tool to measure the level and direction of public sentiment, but is unable to replace the directed feedback telephone surveys provide. As with online surveys, a sizable, unique segment of the population is ignored (those without online access or participation).
Finally, many researchers are falling back to traditional methods of mail and in-person interviews, given the challenges faced by the other alternative methodologies. These methods are effective in providing a representative population, but still have cost and time constraints.
So no, I do not expect telephone surveying to disappear in the near future. Other methodologies are now available to augment and assist the telephone study, but certainly not to completely replace this method. While not as effective as in years past, this still remains an effective, relevant and necessary tool to research public opinion.