Determinants of industrial mail survey response: A survey‐on‐surveys analysis of researchers' and managers' views

This paper compares the perceptions of marketing research executives and industrial respondents regarding factors impacting on industrial mail survey response. Following a review of the different research traditions to the study of mail response rates, a “survey‐on‐surveys “ approach is introduced and used to study the potential impacts of different design and implementation factors on survey response. The results of the study provide empirical evidence on the perceived influence of a wide variety of response‐inducing factors in industrial mail surveys, and are used to generate managerial recommendations for the conduct of the latter. Directions for future research are also identified.     

WEB‐BASED AND MAIL SURVEYS: A COMPARISON OF RESPONSE,DATA, AND COST

The mail survey is the most common data collection technique used by logistics researchers today. Many researchers have noted that response rates to mail surveys have been declining. The Internet offers logistics researchers several potential advantages over traditional mail surveys. This research compares these two methods across response rates, speed of response, consistency of results, and cost. The research also provides unique insight regarding the ability of electronic methods to gain interest and subsequent participation among potential respondents thereby helping in theory testing.     

Response rates in international mail surveys: Results of a 22-country study

Researchers embarking on their first international mail survey find very little guidance in the present academic literature. In 1988, two articles were published in the fall issue of the Journal of International Business Studies that claimed that: “at the moment the crossnational researcher has very little evidence upon which to base his judgments about [mail] survey design” [Jobber and Saunders (1988) An experimental investigation into cross-national mail survey response rates, Journal of International Business Studies19, 488] and “Literature concerned with response rates from industrial samples drawn from multiple countries is nearly absent” [Dawson and Dickinson (1988) Conducting international mail surveys: the effect of incentives on response rates within an industrial population, Journal of International Business Studies19, 492]. Unfortunately, not much has changed since. Still, very little is known about how respondents from different countries react to mail surveys. This article intends to fill part of this gap by describing the results of a large scale international mail survey in 22 countries. Response rates are shown to vary considerably across countries in a way that contradicts much of the earlier (American) research on this subject. Several explanations for these differences in response rates are put forward.     

Sociodemographic factors and mail survey response

Effects of sociodemographic factors on mail survey response rate, response speed, and data quality are summarized. Consistent with previous reviews, evidence for effects of education is strongest with possible effects of age and gender. Evidence for effects of other sociodemographic factors is either ambiguous or absent. Speed and quality of response are also associated with educational level and possibly age. Sociodemographic factors are briefly discussed in light of three theories of mail survey response behavior. © 1996 John Wiley & Sons, Inc.     

Determinants of mail‐survey response: Survey design factors and respondent factors

With the use of a hierarchy‐of‐effects model, the decision to complete a mail survey was modeled as a process moving through several stages. A set of variables thought to influence the survey‐completion decision process was examined. These variables include survey design factors, which were manipulated by the researchers, and respondent factors, such as attitudinal and personal constructs. Based on the research findings, the hierarchy‐of‐effects model is an appropriate way to model the mail‐survey‐response process. The data indicate that among the survey design factors included in the study, the monetary incentive had the most effect on the decision process, with a pervasive impact throughout the process. Importantly, respondent factors such as attitude toward research were of critical importance in moving respondents through the survey‐response process. Research into the response‐rate phenomenon is likely to benefit by considering how both survey design and respondent factors affect specific stages of the response process and enhance participation in surveys. © 2002 Wiley Periodicals, Inc.     

The impact of telephone notification strategies on response to an industrial mail survey

The mail questionnaire is a popular method of gathering data among marketing academics. However, response rates from industrial populations are often low. A number of alternative strategies designed to enhance response from industrialists are evaluated. The results show that satisfactory response rates can be achieved by the integration of telephone and mail contacts within a survey design without loss of response quality nor distortion of sample composition.     

Do Certified Mail Third-Wave Follow-ups Really Boost Response Rates and Quality?

The Total Design Method (TDM) of designing and implementing mail surveys has been shown to achieve high response rates. One key step in the TDM is sending a third-wave of surveys by certified mail. However, little research exists to verify the effectiveness of this step in improving response rates and quality relative to its increased expense, although this is the 50th anniversary of certified mail in the U.S. Perhaps as a result, scholars rarely use certified mail third-waves or omit third-wave mailings altogether. This article presents the results of two experiments that we embedded in two large-scale organizational mail surveys. Both studies reveal that sending a third-wave of questionnaires significantly increases the rate of response over the first two waves of mailings. Also, the results of Study 1 show that sending a certified mail third-wave has no appreciable effect on response quality. Study 2 shows that a certified mail third-wave does not significantly increase response rate or quality compared to sending the third-wave by regular, 1st-class mail.     

Reliability and validity effects under mail survey conditions

In searching for means to reduce error in mail surveys researchers have focused largely on one component of total survey error—sampling error; means have been sought for its reduction by increasing mail survey response rates. The second component of total survey error, nonsampling error (e.g., response bias), has received relatively little attention. This study focuses on the second type of survey error through the development of indices of retest reliability and covergent validity. These indices were applied in a field experiment involving manipulations of survey administration conditions previously linked to response rate effects. The results indicate that in addition to influencing response rates, survey sponsorship, cover letter message, question order, and notification method can substantially affect the retest reliability and covergent validity of data obtained from mail questionnaires.     

Raising Response Rates in Mail Surveys of Small Business Owners: Results of an Experiment

Mail surveys of small business owners have notoriously low response rates, creating the potential for substantial error in surveys of this population and diminishing the credibility of research conducted on small firms. The author recently carried out an experiment as part of a larger project involving 16,000 small business owner/members of the National Federation of Independent Business (NFIB). The experiment's purpose was to ascertain survey treatments that might enhance mail survey response among small business owners. Results showed that none of the six treatments examined improved response rates enough to warrant its routine use over the alternative, nor did any combinations of treatments help. The implication is that commonly used treatments, for example prenotification, often are unproductive.     

Research note: A cost‐benefit analysis of letter prenotification and follow‐up

Conflicting findings for two of the most commonly used, cost‐incurring mail survey inducement techniques encouraged a re‐examination of the effectiveness of letter prenotification and letter follow‐up. An expanded cost‐benefit analysis, which considered response rate, time, quality and cost, was conducted on a sample of university academics in a 2 × 2 experimental design with control. Both letter prenotification and letter had no significant effect on response rate. Cost‐benefit analysis showed the control group was the most cost‐effective, followed by the follow‐up group, the interaction group and the prenotified group; although this ordering depends on the researcher's priorities. The results suggest that the use of prenotification letters to improve response from organisational respondents may be misguided and questions the use of follow‐up letters.     

HANDLING NONRESPONSE IN LOGISTICS RESEARCH

Assessing and controlling for nonresponse bias is critical for the validity and reliability of survey‐based logistics research. In order to gain a better understanding of how researchers assess and report nonresponse bias in their studies, we analyzed articles published from 1998 through 2007 in three top logistics journals (International Journal of Physical Distribution and Logistics Management, Journal of Business Logistics, Transportation Journal). We describe and explore how nonresponse has been handled in these articles. Our findings show that first, survey response rates have declined over time; this influences the generalizability of survey results in case of nonresponse. Second, an average of 44% of published mail survey articles does not mention tests for detecting nonresponse bias.     

Manipulating dissonance to improve mail survey response

Despite the large amount of research on mail questionnaires, there has been little effort toward the development of a questionnaire response theory that might guide the design of mail survey research. The authors conceptualize mail questionnaire response as a series of responses to a set of stimuli rather than a single decision to respond or not respond. A review of the mail survey response literature is provided and cognitive dissonance theory is used to explain why particular techniques are effective. Four types of nonrespondents are identified and suggestions are provided for the use of incentives, prenotification, follow-up contacts, type of postage, and cover letter treatments.     

The effects of the researcher's physical attractiveness and gender on mail survey response

This study examines how using a photo in the cover letter of a mail survey affects mail survey response. A 2 × 2 factorial design was used that manipulated the physical attractiveness and gender of the supposed researcher that was depicted in the cover letter photo. In addition, a control group was used that received a cover letter without a photo. The results revealed the following: (1) a photo of the survey researcher does not enhance the response rate to the survey; (2) in the case of an attractive male researcher, a photo may dampen the response rate; and (3) within the factorial design, the only variable to have an effect on the survey's response rate was the gender variable: Photos depicting female researchers produced a higher response rate than photos depicting male researchers. This study also illustrated that a crude pretest of the survey procedures with students can, in some areas, produce fairly accurate predictions of the results of a field experiment. © 2008 Wiley Periodicals, Inc.     

An assessment of organic apparel,environmental beliefs and consumer preferences via fashion innovativeness

The topic of organic apparel has been widely discussed among academics and practitioners in recent years. While numerous studies have been done on the topic, few studies to date have assessed the topic of fashion innovativeness and its influence on attitudes towards organic apparel. In addition, an evaluation of consumer preferences for organic apparel from a conjoint analysis perspective has not been implemented. Thus, the purpose of the present study was to identify which organic apparel attributes are most important to high fashion innovativeness and low fashion innovativeness groups. Furthermore, the variables of environmental beliefs and attitudes towards purchasing organic apparel were assessed. An online survey was developed to measure the variables, including a full profile discrete choice design used to measure attribute preferences for t‐shirts. The data were analyzed using a multinomial logit model and desirability indices. The results indicated that the low fashion innovativeness group preferred organic and eco‐friendly apparel more than the high fashion innovativeness group. In addition, when examining high fashion innovativeness and low fashion innovativeness groups overall, the preferred t‐shirt was Dri‐Fit, Cotton Jersey Knit, Made In America, Eco‐Friendly and $25.00. However, when examining high fashion innovativeness and low fashion innovativeness groups separately for the preferred t‐shirt, differences appeared in Sustainable. The results suggest that high fashion innovativeness and low fashion innovativeness groups may desire different organic apparel attributes when considering organic apparel.     

A NOTE ON MAIL SURVEYS AND RESPONSE RATES IN LOGISTICS RESEARCH

Logistics researchers frequently use mail surveys to collect data. This paper documents the decline in response rates to surveys reported in JBL, and presents a content analysis of mail survey results reported in two leading logistics journals, the Journal of Business Logistics and the International Journal of Physical Distribution and Logistic Management. One interesting finding is the more questionnaires mailed out in a given study, the lower the response rate.     

Consumer involvement in health research: a UK scoping and survey

Consumer involvement or patient and public involvement (PPI) in health research is a UK policy imperative and a prerequisite for many funders. PPI in research is defined as research carried out with or being carried out by the public (or service users), rather than research on patients and public as subjects or participants. Despite the clear policy driver, there is relatively little empirical evidence on the extent, processes and impact of user involvement in research. This paper aims to add to the international evidence base on PPI in research by providing a key overview of current trends and impacts. In order to understand the current extent and variation of PPI in research, a scoping exercise and survey were carried out on selected UK studies. Six research topic areas (cystic fibrosis, diabetes, arthritis, dementia, intellectual and developmental disabilities, and public health) were selected to ensure a range of designs, study populations and histories of PPI in research. A total of 838 studies (non‐commercial studies and not older than 2 years) were contacted. The response rate for the scoping was 38% and the survey 28%. In the scoping, 51% of studies had some evidence of PPI and in the survey 79%. The most common PPI activity was steering committee membership and reviewing patient information leaflets. There appeared to be some blurred roles with patients participating as research subjects as well as carrying out patient involvement roles. A major finding was the limited amount of available information about PPI in publicly accessible research documents. We suggest that the invisibility of this type of involvement and the lack of routinely collected information about PPI results in a lack of shared understanding of what optimal PPI in a study should look like, with important implications for practice. Furthermore, without a framework to review PPI it is difficult to know if different approaches to PPI have a different impact on key outcomes of the research.     

SMS-based surveys: strategies to improve participation

The advertising industry is increasingly using mobile technology to communicate and research. This paper examines the use of the short messaging service (SMS) on mobile phones to recruit samples for probability web and telephone surveys. The influence of topic salience, sponsor identity and repeated contacts on decision to participate in the survey is tested through an experimental design. Results indicated that sponsor identity and repeated contacts have an effect on decision to participate, but topic salience did not. The majority of respondents preferred the web survey alternative over telephone survey mode, and the main advantage of using SMS is the high speed of response. However, the method appears to elicit higher levels of participation from male and younger members of the population.     

Consumers’ attitudes,trust and willingness to pay for food information

Market failure can be corrected using different regulatory approaches ranging from high to low intervention. Recently, classic regulations have been criticized as costly and economically irrational, and thus, policy makers are giving more consideration to soft regulatory techniques such as information remedies. However, despite the plethora of food information conveyed by different media there appears to be a lack of studies exploring how consumers evaluate this information and how trust towards publishers influence their choices for food information. In order to fill such a gap, this study investigates questions related to topics that are more relevant to consumers, who should disseminate trustful food information, and how communication should be conveyed and segmented. Primary data were collected both through qualitative (in‐depth interviews and focus groups) and quantitative research (web and mail surveys). Attitudes, willingness to pay (WTP) for food information and trust towards public and private sources conveying information through a new food magazine were assessed using both multivariate statistical methods and econometric analysis. The study shows that consumer attitudes towards food information topics can be summarized along three cognitive‐affective dimensions: the agro‐food system, enjoyment and wellness. Information related to health risks caused by nutritional disorders and food safety issues caused by bacteria and chemical substances is the most important for about 90% of respondents. Food information related to regulations and traditions is also considered important for more than two‐thirds of respondents, whereas information about food production and processing techniques, lifestyle and food fads are considered less important by the majority of respondents. Trust towards food information disseminated by public bodies is higher than that observed for private bodies. This behaviour directly affects WTP for food information provided by public and private publishers when markets are shocked by a food safety incident. WTP for consumer association (€1.80) and the European Food Safety Authority (€1.30) are higher than WTP for the independent and food industry publishers that cluster around zero euro. Furthermore, trust towards the type of publisher also plays a key role in food information market segmentation together with socio‐demographic and economic variables such as gender, age, presence of children and income. These findings invite policy makers to reflect on the possibility of using information remedies conveyed using trusted sources of information to specific segments of consumers as an interesting soft alternative to the classic way of regulating modern food markets. JEL codes: D12, D18, D89, Q18.     

Toward a conceptualization of mail survey response behavior

A conceptual framework for understanding and researching mail survey response behavior is proposed and developed around the constructs of cooperation, trust, and commitment. Survey response is viewed as the cooperative exchange of information by respondents in return for an opportunity to shape their environment. Issues related to the valuation of the survey exchange process are examined within the proposed framework. Current survey inducement techniques are evaluated in terms of their relationship to the three central constructs and their effects on the decision to participate in a mail survey. © 1996 John Wiley & Sons, Inc.     

Consumer Participation: The Case of Home Meal Preparation

Consumer participation (CP) has become an important topic in marketing literature with the emergence of service‐dominant logic (SD logic), where consumers are viewed as co‐creators or coproducers of value. However, most studies of CP have largely focused on its economic implications for the supplying firm and thus not much attention has been paid to the factors explaining this phenomenon. This study explains the CP process in the context of home meal preparation by proposing and testing a model of how convenience orientation and knowledge influence both a motivational (involvement) and a behavioral component (time use) of a CP process. The results show that high involvement is a main driver of CP behavior. Convenience‐oriented people are less involved and spend less time on CP activities. Knowledge is suggested to have a positive influence on involvement with CP, but it may not necessarily positively impact the time people use in CP because skills may bring about efficient performance.