These are excellent techniques.  They have been used for decades in market and consumer research, are very effective in political polling, and are being used with greater frequency (albeit a low frequency) with Organizational Surveys.  There is much to recommend about telephone survey techniques but some of the advantages are more apparent than real.

For example, it is assumed that telephone surveys produce results that are equivalent to other data collection methods, they provide results in a shorter period of time, and are less expensive than other administration methods for Organizational Surveys.  These assumptions are not necessarily true.

The use of Interactive Voice Response (IVR), and live interviewer telephone surveys is not recommended.  With all its benefits as a proven technique, telephone surveys have some significant problems when it comes to Organizational Surveys or employee attitude surveys. Telephone surveys are not necessarily the fastest for the reporting of results.  Also, they are not as cost-effective as other methods of data collection.  The most significant problem, however, is that attitude survey results from telephone surveys do not compare with results for the same questionnaire that are collected by other means.

Published research over the past decade and our own experience lead us to conclude that IVR phone survey results cannot be compared to survey results from hard copy and online surveys. Research and our own experience demonstrate that results from paper and pencil surveys, PC-based administration, e-mail administration, network administration, internet administration, and scannable fax-back forms are largely comparable to one another.

IVR phone surveys or live operator telephone interviews would be acceptable if all data were collected in this manner and no comparisons were made to results from other modes of administration.  Some research suggests that phone surveys produce more favorable results than other methods.  However, there is no consistency to the differences, thus attenuating the utility of such comparisons.  Significant experience of one of our senior consultants suggests no consistency at all.  An enormous amount of time, effort, and money went into a project to collect national norms with a telephone survey.  It was abandoned when it became apparent that comparing the data to paper surveys was an impossibility..

It is not uncommon to see results vary for telephone surveys vs. paper surveys by 15 to 20 percentage points.  The variation also seems to be inconsistent.  If the results were always 15 points higher or 15 points lower, then adjustments would enable comparisons to be made.  Telephone surveys do not have a consistent effect on these differences.

If your Organizational Survey used a telephone survey to collect its data, it could never compare the results with any confidence to other survey data collected under different techniques.  Our Vice President, William E. Dodd, Ph.D., had first-hand experience with this problem while working for a major survey consulting organization in New York City.    Recently published research, as well as anecdotal reports, confirm this phenomenon.

If your organization chooses to use a telephone survey, we recommend that you not compare the results with data gathered with other techniques, and not change the telephone survey in the future.  Even if your organization so chooses, it will be almost impossible not to make the comparisons, and not make decisions based on unreliable comparisons.

One method would be to deliver the survey as an e-mail attachment and have the employee fill it out on the computer.  If your employees are reachable by e-mail, then they are reachable by CyberSurvey(tm).  At the end of the questionnaire, the employee 'SENDS' the survey, and the response file is sent automatically.  This is a very efficient process and requires almost no setup on the part of the client organization.

Another method is to distribute the survey on floppy diskettes and mail or ship them to the employees.  CyberSurvey(tm) runs from the floppy drive and creates an encrypted response file on the diskette.  The employee mails or ships the diskette.  CyberSurvey(tm) can also run on a network, corporate Intranet, or on the Internet.  It can be hosted on your server or on ours.

A mixed mode survey administration may be employed.  [SEE maXimumSurvey(tm)]  For example, you might use the Internet for headquarters employees, e-mail for most of the Field offices, and fax-back scannable forms for the remainder of the Field offices. The scannable fax-back forms represent a fallback choice for anyone who prefers a paper and pencil survey.

The selection of primary and backup administration modes should be made with the Organizational Survey project manager and the Steering Committee.  You may find that an Intranet survey is better for some units, while an e-mail survey is better for others.  There may be some units for which a paper and pencil survey in a group setting is the most expedient.

Another important decision is the location of the host server for the Organizational Survey.  Hosting the survey on your survey consultant's server will be easier for them to administer and less costly for you.  It also provides a better perception of the protection of anonymity and confidentiality.  However, if your consultant hosts the survey on their server, the respondents may be subject to the reliability and the varying speeds of the World Wide Web.  This is not a serious issue, but it does create a few more HELP DESK calls than when it is hosted on your Intranet.  Help desk services can be provided by your survey consultant or you.

Hosting the survey on your Intranet server may guarantee more reliable and faster connections with fewer incident reports and HELP DESK queries.  However, there may be a greater cost to you as a result of training supplied by your survey consultant and the need to coordinate their access and maintenance activities with your IT unit. If you host the survey on your server, it will require the dedication of internal IT support to make any survey successful.  Finally, there may be some impact on the perception of anonymity and confidentiality.

For all of the above reasons we usually recommend hosting the Organizational Survey on your survey consultant's server.