|Page tools: Print Page Print All|
SURVEY FAX DESIGN STANDARDS
Fax machines have been a useful addition to most businesses since the early 1980s. Most fax machines produced before 1990 used rolls of low quality paper, and produced poor/unclear transmissions. The top and bottom of pages may have been distorted, and vertical lines and spots from telephone line static were often visible on faxed images (Walker, 1994).
In addition, early fax technology was limited by both the quality of the images produced by the receiving fax machine, and by the number of respondents who had access to fax technology. For these reasons researchers were hesitant to fax survey forms, as the fax received by the respondent may have been very hard to read. Furthermore, if a poor quality form were to be completed and returned by fax, the quality of the image would diminish further (Walker, 1994).
Almost all fax machines manufactured since the early 1990s use regular A4 paper and produce clear reproductions, free from many of the flaws of earlier models. Modern machines automatically remove distortions and "noise", and transmissions are considerably faster. Faxes may also be used in conjunction with software designed to send faxes automatically, retry busy numbers and log the outcome of transmissions.
These improvements to the quality of faxed data, as well as the increasing number of Multi-function Peripherals (MFP), or combined fax, printer, scanner and photocopy devices, have made the fax machine an increasingly common business tool (Cullem, 2001).
Studies using identical short black and white questionnaires have found that there are no significant differences in data quality between the responses obtained by fax compared to mail (Vazzana & Bachmann, 1994). Similarly, there is no significant difference in levels of item non-response between fax and mail or fax and email forms (Cobanoglu, Warde & Moreo, 2001). Respondent error rates for fax responses are comparable to mail responses for both open-ended and checklist questions (Dickson & MacLachlan, 1992; Dickson & MacLachlan, 1996). Respondents also voice positive attitudes about fax surveys and letters (Lensing et al., 2000; Rosen & Clayton, 1992).
While fax surveys have lower overall non-response rates and faster return times than mail surveys (Dickson & MacLachlan, 1996; Shannon & Arbet, 1998; Schreckhise, 2006; Au, 1994), and email surveys (Cobanoglu, Warde & Moreo, 2001), fax technology should be used in conjunction with other survey methods (Dickson & MacLachlan, 1996; Rosen & Clayton, 1992). For example, fax advance notice for Touchtone Data Entry (TDE) surveys have been found to be more effective than mail postcards (Rosen & Clayton, 1992) and phone calls (Chan, 1997), and fax reminder letters have been found to be more effective at increasing response rates in mail surveys than mail-out reminder letters (Au, 1994; Shannon & Arbet, 2004).
Where mixed mode surveys are to be conducted, the fax and mail forms should be as similar as possible to ensure consistent reporting from respondents, and to minimise mode effects (Dickson & MacLachlan, 1992). As very few changes to the letter standards are required to conform to fax typography, reminder and pre-approach letters can easily be sent by fax. The changes required to adapt a colour questionnaire to fax, however, are substantial, and we are unaware of any research into the relative effectiveness of colour forms versus fax forms.
However, there is research comparing colour and black and white mail out questionnaires, and it is clear from this research that respondents prefer coloured forms (Phipps, Robertson & Keel, 1991). Interestingly, there are no significant differences in response rates between colour and black and white forms (Crittenden, Crittenden & Hawes, 1985; Lagarce & Washburn, 1995; Wheildon, 1990), although colour forms allow greater design flexibility in terms of distinguishing note boxes from answer boxes and so on.
Consequently, the quality of data obtained by colour forms will be superior to data from black and white forms, in terms of a reduction in non-sampling error. Where possible, colour forms should be used for mail-out surveys. Therefore, including fax in the initial despatch for mixed mode surveys that also include a colour mail out form is not recommended, as common content cannot be assured. However, fax forms may be used in the following cases:
Use of Fax Technology at the ABS
The ABS regularly faxes black and white self-administered forms to business survey respondents. While there were some concerns about the speed and quality of fax transmissions in the past, technological advancements in fax machines and survey design software have made fax a feasible despatch method for black and white forms. In addition to improved graphics, the most significant improvement to modern fax technology is the ability to send bulk faxes electronically, rather than one by one manually. However, as faxes are always black and white, the superior quality of coloured mail out questionnaires makes them the preferred mode of distribution for paper-based self completed forms.
The ABS is not alone in increasing the use of fax technology. The U.S. Bureau of the Census and the U.S. Bureau of Labour Statistics regularly utilise fax technology to send advance notices and non-response follow up letters. However, the effects of the long term use of fax for surveys forms are unknown. Many researchers express the concern that the positive effects of fax will be negated once the novelty of the mode wears off (Rosen & Clayton, 1992; Vazzana & Bachmann, 1994; Walker, 1994). For this reason, the use of fax transmissions should be regularly reviewed.
The ABS currently uses an automated fax system to send and receive faxes. This system is a Paperless Fax Image Reporting System (PFIRS) called a "fax gateway", and is designed to integrate fax technology with computer software and existing databases. The fax gateway can send and receive faxes automatically, and does not require the incoming faxes to be printed in order for the data to be recorded.
The ABS does not expect respondents to have a PFIRS in place. The majority of respondents receive all fax transmissions as printed paper copies, not as digital files. These paper faxes can be filled out by hand and returned by fax. For this reason, it is vital that all faxes sent are of the highest possible quality, to ensure the paper reproduction received by the respondent is clear (Dickson & MacLachlan, 1992).
All mail out questionnaires have a freecall telephone number printed on the cover page. Respondents are able to call and request a new form which, if requested near to the close off date, can be faxed. The majority of quarterly and annual collections have a "fax-friendly" or black and white .pdf version of the form designed for this purpose. The .pdf fax-friendly forms contain fields that can be edited electronically, so that the following details can be entered prior to faxing:
When requested, fax-friendly forms can be sent directly from the computer of the operator who answered the respondent's call. The respondent should be told how many pages the form is, as many respondents may be reluctant to be sent a particularly long fax. For this reason the use of fax should be limited to respondents who request a fax version, or who agree to be sent a fax during IFU.
In addition to the freecall number, all business survey forms are printed with a fax number. Respondents may return surveys by fax for the cost of a local call (about 25c), and are encouraged to do so towards the end of the survey cycle. Currently, the ABS PFIRS is run using a fax gateway which receives all incoming faxes, and sends them directly to a work group database. The database receives the incoming faxes as an image file, and the data is then read by Optical Character Recognition (OCR) or manually keyed.
Fax gateways must have a high capacity to ensure respondents are not constantly getting a busy signal when they attempt to fax back their form. The alternative is to use an external system, which stores incoming faxes in memory offsite. This is not recommended, as it may lead to confidentiality problems.
The future of fax technology at the ABS
Integrated OCR/form design software which enables PFIRS (e.g. Teleform) has been on the market for some time, and similar systems (e.g. electronic fax machines) can send and receive faxes as email attachments using programs such as Microsoft fax (Cobanoglu, Warde & Moreo, 2001). Currently, however, the ABS does not have OCR capability for mail out forms returned by fax, and this data must be recorded manually. Returned "fax friendly" forms, however, with the inclusion of barcodes, should enable OCR and will consequently be significantly less costly and resource intensive to process.
In 2007 the ABS received approximately 45,000 completed forms by fax. Of these, only twenty per cent were fax friendly forms. The other eighty per cent (approximately 36,000) were mail out forms returned by fax, which were then manually keyed.
Due to the additional expense of manually keying data, the use of fax friendly forms should be encouraged for respondents who intend to return their form by fax. Similarly, respondents should be discouraged from faxing a mail out form, and instead be offered a fax friendly copy. Pending further improvements to fax friendly forms, the relevant data will be collected using OCR. This would greatly reduce the need for manually entering data.
Any fax friendly forms which are unable to be identified by OCR must be entered into the system manually. Consequently the effectiveness of ABS fax-OCR facilities and fax friendly form design should be reviewed regularly to ensure that manual data entry is kept to a minimum. Typography
While fax machines are no longer affected by many of the flaws of older models, there are still two major limitations:
Faxed information sheets and pre-approach, induction, cover, reminder and other letters should conform to letter standards with particular emphasis on minimising the use of bold text. Only key words, such as headings and the return fax number should be in bold.
Furthermore, as fax machines can only send and receive black and white images, lines which in a colour form would be solid colour or white must be changed to black, and shading should not be used. Fax friendly forms must conform to the black and white form design standards but, to aid form reading by OCR, lines on fax friendly forms may be in 1 point for answer boxes and 1.5 point for totals boxes.
As respondents receive all pages of the survey as single sheets of paper (unlike the booklet format of mail out surveys) it is important to ensure the reference number is printed on every page of the form. This will ensure that a returned form can be processed even when some pages are not returned (e.g. if the cover page is missing). The editable reference number field on .pdf fax-friendly forms will automatically copy the reference number to every page of the form. It is therefore particularly important that the reference number is entered accurately.
Just as with mail surveys, response rates for fax surveys are increased by freecall numbers, advance notice, cover and reminder letters (Au, 1994). For this reason, appropriate IFU procedures should be followed.
Advance notice (or new-on letters for new respondents for continuing surveys) and reminders for outstanding surveys are usually issued to ABS survey respondents in conjunction with IFU telephone contact. If fax advance notice and reminders are used, they should conform to the ABS letter standards with the modifications specified in the section on typography. In addition, the following amendments should be made to fax reminder letters:
All fax transmissions should be addressed to the intended respondent of the survey. The cover page should include a salutation with the respondent's name, and should conform to the ABS letter standards. If the name of the respondent is unknown, the fax should be directed to an appropriate respondent. For example "Attention: Chief Financial Officer" should be printed in 12 point bold on the top of the cover page.
It is possible to personalise salutations for electronic faxes by overlaying data from an external database or file. Provided this technology is available, the respondent's or business's name can be included in all faxes using PIMS. In line with ABS letter standards, this function should be used to include a name in all salutations.
While mail surveys include a correctly addressed return envelope, a fax survey relies on the respondent to accurately dial the return fax number. For this reason, the covering letter should include a modified confidentiality statement which stresses the importance of correctly entering the return fax number to ensure the confidentiality of the data. In addition, the covering letter should include instructions directing the receiver to notify the ABS if a fax is received in error.
An example covering letter for a form dispatched by fax is below. This letter contains a personalised salutation, instructions for those who receive a form in error, and instructions for the respondent which outline the importance of correctly entering the return fax number:
Attached is a copy of the survey form for the [Survey name] for [reference period].
This survey is being conducted by the ABS under the authority of the Census and Statistics Act 1905. The Act ensures that the information you provide is treated as strictly confidential and will be used for statistical purposes only.
The ABS will maintain security and confidentiality of your data, once we receive it. A free fax number is provided on the front of the survey form for the return of your survey. Care should be taken to enter the fax number correctly to ensure your confidential information is not misdirected.
Please complete this form and return it to the ABS by dd/mm/yyyy
If you want to send the data by fax but have particular concerns, or need help completing this form please contact us. We can often advise on ways of making this task easier for you.
[Name of ABS officer]
Fax transmission is limited to respondents with access to a fax machine. As few households have fax machines, household surveys are not ideal for fax data transmission (Au, 1994). However, fax can be suitable for businesses, as a great majority of businesses have fax capability.
Providers who prefer returning their survey by fax rather than mail, and have fax capability, can request communications be sent by fax. Faxes should only be sent to new survey respondents after initial contact is made by phone (Walker, 1994; Klassen & Jacobs, 2001) or mail. Ongoing survey respondents should be notified by their usual contact method that a fax option is available if they provide a valid fax number (Schreckhise, 2006). However, fax may not be a suitable mode for all collections and should not be offered unless a fax friendly form is available.
All respondents who select fax as a contact method should be told when to expect faxes, and asked to check that the machine is left on. In addition, respondents should be asked to check that there is adequate paper and ink to receive fax transmissions. Fax can be a useful addition to ongoing surveys, as advanced notice letters and reminders for regular reporting can be faxed to businesses automatically at preprogrammed intervals. However, faxes longer than one or two pages should never be sent without the respondent's prior permission.When should faxes be sent?
In a study of a survey with monthly reporting by fax, the U.S. Bureau of the Census found that reminder faxes should be sent to respondents who had not reported after the third week of the month (Appel, Petunas & Russell, 1995). Similarly, Rosen and Clayton (1992) found that non-response prompt faxes should be sent a minimum of 3 days prior to the cutoff date. If they are sent any later the response rate drops significantly (Rosen & Clayton, 1992). Both these studies were conducted with sub-annual collections using a single reminder letter.
The implication of this research on the use of fax reminder letters at the ABS is that, regardless of the frequency of the collection, the final reminder should be dispatched early enough to ensure the respondent has ample time to complete and return the form before cutoff. For sub-annual collections the third reminder letter should be faxed to respondents 3 to 7 days before the final cutoff date, while the final fax reminders for annual collections may be dispatched 7 to 14 days before the cutoff date.
In mixed mode surveys, the fax reminder should be sent a few days later than the mail reminders. This increases the likelihood that mail and fax respondents will receive the reminder on the same day. Similarly, where fax is an initial dispatch method in a mixed mode survey the fax form should be dispatched a few days after the mail dispatch to ensure all respondents receive the form within the same time frame.
Faxes sent at the request of the provider (e.g. during IFU) should be dispatched as soon as possible after the request is made. This will ensure that the fax arrives quickly, and that the provider is prepared for the transmission (i.e. the fax machine is on and there is plenty of paper and ink).
While mail is the preferred mode of dispatch for self completed forms, there are circumstances in which the use of fax may be suitable. Fax has the advantage of being faster than mail, and as such it is suitable for fax-friendly forms to be dispatched to respondents, with their permission, near the end of the IFU period. Respondents who intend to return their form by fax should be sent a fax-friendly form, as a mail out form returned by fax cannot be read by OCR and is, consequently, much more costly to process.
Appel, M.V., Petkunas, T.F. & Russel, C.E. (1995), "Experiences with fax data reporting and questionnaire distribution", 1995 Proceedings of the Section on Survey Research Methods, American Statistical Association, pp 544 - 548
Au, A.K.M. (1994), "The appropriateness of facsimile as a survey medium", Journal of International Marketing and Marketing Research, Vol 19, pp 143 - 149
Cobanoglu, C., Warde, B. & Moreo, P.J. (2001), "A comparison of mail, fax and web-based survey methods", International Journal of Marketing Research, Vol. 43, pp441 - 455
Chan, Y.I. (1997), "Nonresponse follow-up in establishment surveys: A split-half experiment", 1997 Proceedings of the Section on Survey Research Methods, American Statistical Association, pp988 - 993
Crittenden, W.F., Crittenden, V.L. & Hawes, J.M. (1985), "Examining the Effects of Questionnaire Colour and Print Font on Mail Survey Response Rates", Akron Business and Economic Review, Vol 16, pp51 - 56
Cullen, S. (2001), "Think You Don't Need an MFP?", Office Solutions, Vol. 18 (6), pp34 - 38
Dickson, J.P. & MacLachlan, D.L. (1992), "Fax Surveys?", Marketing Research, pp26 - 30
Dickson, J.P. & MacLachlan, D.L. (1996), "Fax Surveys: Return patterns and comparison with mail surveys", Journal of Marketing Research
Dickson, J.P. & MacLachlan, D.L. (1992), "Fax Surveys?", Marketing Research,Vol 33, pp108 - 113
Klassen, R.D. & Jacobs, J. (2001), "Experimental comparison of Web, electronic and mail survey technologies in operations management", Journal of Operations Management, Vol. 19, pp713 - 728
Lagarce, R. & Washburn, J. (1995), "An Investigation into the Effects of Questionnaire Format and Colour Variations on Mail Survey Response Rates" Journal of Technical Writing and Communication, Vol 25 (1), pp57 - 70
Matchett, S.D., Creighton, K.P. & Landman, C.R. (1994), "Building integrated systems of CASIC technologies at The Bureau of the Census", 1994 Annual Research Conference and CASIC Technologies Interchange, pp573 - 595
Phipps, P.A., Robertson, K.W. and Keel, K. (1991), "Does Questionnaire Color Affect Survey Response Rates?", Proceedings of the Survey Research Section, American Statistical Association, pp484 - 489
Rosen, R.J. & Clayton, R.L. (1992), "An Operational Test of Fax for Data Collection", American Statistical Association, 1992 Proceedings of the Section on Survey Research Methods, pp602 - 607
Rowe, E.G. & Appel, M.V. (1993), "Paperless Fax Image Reporting System (PFIRS)", American Statistical Association, 1993 Proceedings of the Section on Survey Research Methods, pp168 - 172
Schreckhise, W.D. (2006), "The costs and benefits of mail, fax, and mixed-mode surveys in organizational research", Public Administration Quarterly, Vol. 29, Iss. 4, pp494 - 512
Shannon, D.M. & Arbet, S.E. (2004), "Using FAX Technology to Survey Professionals and Organizations", Journal of Experimental Education, Vol. 62, Iss. 3, pp259 - 269
Tse, A., Ching, R., Ding, Y.B., Fong, R., Yeung, E & Au, A. (1994), "Miscellany: A comparison of the effectiveness of mail and facsimile as survey media on response rate, speed and quality", Journal of the Market Research Society, Vol 36 No 4, pp349 - 355
Vazzana, G.S. & Bachmann, D. (1994), "Fax Attracts: New technological advances make fax a promising alternative to mail surveys", Marketing Research, Vol 6 No 2, pp19 - 25
Walker, J.T. (1994), "Fax Machines and Social Surveys: Teaching an Old Dog New Tricks", Journal of Qualitative Criminology, Vol 10 No. 2, pp181 - 188
Wheildon, C. (1990), Communicating or Just Making Pretty Shapes, Newspaper Advertising Bureau of Australia, Sydney