7.2.3 Survey administration

Email survey

As discussed in [Berge and Collins, 1996], email surveys make it "easy to census the entire subscription list of a discussion or distribution list by sending a single message to the mailing list" rather than relying on a sampling technique. This was the technique used for this research. Psyche has associated with it two electronic mailing lists, Psyche-L and Psyche-D. Psyche-L is used for distribution of the ASCII version of Psyche, while Psyche-D is used for discussion based around the themes of the journal. The instrument was originally distributed to Psyche-L on September 14, 1995. In order to increase the response rate, it was redistributed to both Psyche-L and Psyche-D on February 16, 1996. The instrument was also posted on a Web site so that people could download the text, complete it and send it in. Use of a Web forms mechanism to gather responses was not adopted due to lack of time to set up the form and data-handling. Other than the redistribution to the mailing lists (which is by definition non-targeted), there was no specific follow-up of respondents.

Print survey

Those surveyed were provided with:

• a generic covering letter (it was not possible to gain access to the membership databases to create a personalised letter) including contact information and the information required by Monash University's Standing Committee on Ethical Research on Humans (SCERH)
• the survey instrument, printed double-sided on a single A4 sheet
• a reply-paid envelope for return of the survey
• a bookmark (generously provided by RAECO Library Services) as a token appreciation for the respondent's time.

The mailout of the print survey was governed by the procedures of the different societies/associations.

The American Psychological Association (APA) does not make its mailing lists available to any organization or individual.They provided mailing labels for one time use after approving the materials submitted. A random sample of 2000 labels was ordered from the total membership of approximately 85,000. There was no subgroup that corresponded adequately to the readership of Psyche . Assuming a 10% response rate, 2000 surveys would provide around 200 responses (compared with the 190 responses from the U.S. A. in the email survey). The surveys were sent out from Australia in February 1997.

The U.K. British Psychological Society (BPS) does not divulge members' addresses to any third party.They require the materials to be supplied to their head office where they handle the insertion and postage for a fee.The BPS allows members to nominate one of a number of interest areas as part of their membership profile. The closest match to the Psyche population was deemed to be Cognitive Psychology. A random sample of 360 labels was ordered from this section. Assuming a 10% response rate, this would provide around 36 responses (compared with the 36 responses from the U.K. in the email survey). The materials were sent from Australia to the U.K. in December 1996. It was agreed with the BPS office that they were not to be posted out to members until mid-January 1997, thus avoiding the Christmas rush.

The Australian Psychological Society (APS) uses a similar system. A random sample of 220 labels from the entire membership was ordered. Assuming a 10% response rate, this would provide around 22 responses (compared with the 22 responses from Australia in the email survey). The materials were sent to the APS head office for distribution in February 1997.

Note that all the above procedures made it impossible to send out follow-up letters. In other words, the mechanisms to improve response rate were little better than for the email survey (none). The only difference was the ability to include the bookmark with the survey as a minor inducement to feel some sense of reciprocal obligation.