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.     

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.     

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.     

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.     

Comparing the generalizability of online and mail surveys in cross-national service quality research

To compare the generalizability of online and mail surveys in a cross-national service quality study, the authors use G-theory and find a comparable level of generalizability, though online surveys benefited from considerably lower costs. This article contributes to the current comparison of the response quality between online and mail surveys. Furthermore, the authors illustrate how G-theory can be used to compare online and mail surveys and take data collection costs into account. Important implications include the process and results of comparing two survey modes and the effects for service research. Martin Wetzels now also works at Maastricht University     

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.     

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.     

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.     

PRODUCT RETURNS PROCESSING: AN EXAMINATION OF PRACTICES OF MANUFACTURERS,WHOLESALERS/DISTRIBUTORS,AND RETAILERS

Few research studies have published specific empirical data regarding the reverse logistics practices of companies. This multi‐stage study employed interviews, site visits, and a mail survey to collect responses from 230 members of the Warehousing Education and Research Council (WERC) regarding their reverse logistics practices. Results suggest that in spite of the growing importance of reverse logistics, few executives have product return processing as their primary responsibility and often undertake this activity along with other job responsibilities. Most firms handle the product returns process themselves and typically within the same facilities that handle forward logistics. Returning items directly to stock, repackaging and returning to stock, and selling as scrap, were the three top disposition options employed by firms. Results indicate that, contrary to general understanding, the majority of retailers and wholesalers reported a recovery rate of over 75% of product cost. Several hypotheses developed from the published literature on reverse logistics were tested. In many instances, these hypotheses were formulated on anecdotal information or single case studies and had not been empirically tested prior to this research being conducted.     

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.     

Latent Information in the Pattern of Missing Observations in Global Mail Surveys

Incomplete responses are quite common in global mail surveys. A review of empirical research published in the last decade revealed that this problem is pervasive, and missing responses to the extent of 15% of total responses are not unusual. Most researchers use simple methods like deletion or substitution to deal with this problem. However, these methods ignore the basic philosophical assumptions of the field and do not consider the information hidden in the pattern of the missing observations. The exploratory study reported here analyzed two existing data sets which had been previously collected with mail survey to study technology transfer issues. Both data sets had a significant amount of missing responses. Our study results reveal that the pattern of missing responses was not random in nature, and there was a significant association between socio-demographic characteristics of the respondents and the number of missing responses.     

A REFINED VIEW OF GROUNDED THEORY AND ITS APPLICATION TO LOGISTICS RESEARCH

Logistics research with an objective to construct theory and develop deeper insights into logistics social phenomena and test those theories relies on both qualitative and quantitative methods. Grounded theory is one powerful qualitative research tradition for the theory building objective, yet it is also one that is often misunderstood and misapplied. The result of this misunderstanding can, and has in other disciplines such as marketing and management, resulted in weaker grounded theory studies than ought to be and research claiming to emerge from grounded theory but does not. The article offers a review of the foundations of grounded theory and insights to important aspects of two similar, but different, approaches to grounded theory that most researchers ignore, but are critical for researchers to understand. We argue that (1) logistics needs more qualitative research, (2) grounded theory offers the potential for a unique and specific kind of insight as compared to other traditions, (3) thus specific tenants of grounded theory must be followed, anything does not go, and finally (4) being specific in the application of grounded theory means knowing the differences between the Glaser and Straus views and understanding that a choice must be made as to how to proceed with grounded theory research based on that knowledge.     

A SUMMARY AND ANALYSIS OF MULTI‐ITEM SCALES USED IN LOGISTICS RESEARCH

Academic disciplines are strengthened through many years of theoretical research leading to significant knowledge advancements in the field. One primary technique for measuring concepts important for theory development is the use of multi‐item scales. This research documents the multi‐item scales utilized throughout the history of leading logistics journals. The test results for validity and reliability with respect to each scale are disclosed. Results indicate a need for logistics researchers to continue developing new measures and refining existing measures by employing rigorous methods to establish scales having highly acceptable psychometric properties.     

THE INTELLECTUAL FOUNDATION OF THE JOURNAL OF BUSINESS LOGISTICS AND ITS EVOLUTION BETWEEN 1978 AND 2007

Many logistics and supply chain management researchers have so far studied the nature of logistics and supply chain management research in terms of its domain and scope, its epistemological assumptions, and its evolution. However, the knowledge repository on which the scientific research community draws, that is, its intellectual foundation, has not yet been studied. Studying the intellectual foundation of research provides an unbiased and comprehensive picture of the development, dissemination, and utilization of its knowledge. In this article, we identify the most contributive works—in terms of citations received—that have been used in 497 articles published in the Journal of Business Logistics (JBL) between 1978 and 2007. By means of citation and co‐citation analysis, the intellectual structure of research in JBL is revealed and transformations therein are explored. Overall, the most frequently‐cited literature can be classified into six themes: physical distribution; inventory models; customer service; interorganizational relationships; competitive strategy; and empirical methodologies for socio‐scientific research. Furthermore, we determined a development in citation frequencies to these themes: literature related to physical distribution and inventory management declined over the three decades under study, whereas literature related to competitive strategy and empirical methods gained in importance. This development indicates a shift from an operational focus to a prioritization of managerial issues. Moreover, our results demonstrate a shift towards more relational and institutional research in logistics (management), which has been typically linked with the notion of supply chain management since the 1990's.     

Expanding Horizons and Deepening Understanding via the Use of Secondary Data Sources

While primary data analysis has been popular in logistics and supply chain research, secondary data methods have been overlooked. These methods, however, have the potential to generate a variety of important opportunities to expand the horizons of logistics and supply chain research. In this article, we emphasize the use of secondary data analysis and how it can address contemporary challenges in logistics and supply chain research. Our review of the logistics and supply chain literature identifies six important methodologies that can be useful for secondary data generation and analysis. We discuss how these methods can help effectively address various logistics research questions.     

THE “WHITE SPACE” OF LOGISTICS RESEARCH: A LOOK AT THE ROLE OF METHODS USAGE

This paper examines the use of various research methods by contemporary researchers. The previous six years of articles in a leading logistics journal are examined and classified. The analysis seeks to identify and discuss changes and trends that are occurring in the use of such methods, as well as considering why the current conditions are in place, and what, if any, opportunities exist for logistics researchers given this situation.     

Absolutely, Positively Hearing the Voice of the Customer: Exploring Overnight Delivery in Survey Research

Recently, marketers have increased their research budgets and used many methods to capture the voice of the customer. Yet surveys, which are used regularly in marketing research, are still tainted by nonresponse rates despite efforts to reduce them. We study overnight delivery (OD) as a survey method for reducing nonresponse. Precontact, a commonly used survey practice, is also considered. Three primary results emerge: (1) overall, OD increases response rates, but increases response times; (2) similarly, precontact increases response rates, although it generates increased response times; and (3) OD combined with precontact provides the greatest rate of response.     

A COMPARISON OF THE ECONOMIC ORDER QUANTITY AND QUICK RESPONSE INVENTORY REPLENISHMENT METHODS

When determining order quantity, logistics managers can choose between ordering the Economic Order quantity (EOQ) or a Quick Response (QR) quantity. QR is a general term that collectively describes several rapid‐replenishment inventory methods such as Just‐In‐Time or Continuous Replenishment. The QR order quantity is defined as the minimum inventory needed to support operations until the next delivery. The EOQ and the QR methods are substantially different because the EOQ minimizes the joint cost of ordering and holding inventory whereas the QR method minimizes only the cost of holding inventory. The goals of this research are to compare the costs of the two methods and to propose rules that help managers select the more appropriate method to use in specific situations.     

A review of research methodologies in international business

What is common practice in international business (IB) research methodology? To address this question, we surveyed 1,296 empirical articles published in six leading international business journals from 1992 to 2003. The study uncovers state-of-the-art approaches in research methodologies in IB in terms of five major aspects: data collection methods, sample sources including sampled countries and subjects, sampling methods, sample sizes, and response rates. The results indicate that (1) mail questionnaire surveys dominate empirical research, (2) 60.9% of the studies use a one-country sample (88.9% from western countries), (3) 33.7% of the studies are based upon sample frames provided by third parties, and (4) the median sample size is 180 with an average response rate of 40.1%. Suggestions and recommendations are also provided to improve the methodological rigor of IB research.     

Reducing uninformed responses: The effects of product‐class familiarity and measuring brand knowledge on surveys

Consumers often give answers to survey questions about which they are uninformed. Although research has documented this phenomenon and examined the conditions under which it occurs, the current research project is the first to examine how survey researchers might reduce uninformed response bias through the use of additional questions added to surveys that measure knowledge of brands for which consumers might be uninformed. In the current study, adding brand knowledge questions pertaining to an unknown (fictitious) brand reduced the likelihood of respondents providing uninformed responses when later asked to evaluate that brand. However, the effects of asking for brand familiarity on uninformed response rates differed by consumers' level of product‐class familiarity and the relative placement of the brand knowledge and brand attitude questions on the survey. Brand knowledge questions can suggest the existence of brands, placing them within a knowledge schema, thus leading to more uninformed responses from consumers low in product‐class familiarity. These results, coupled with recent other research findings suggest that researchers must consider the effects of related knowledge on what have been previously considered completely uninformed responses. © 2007 Wiley Periodicals, Inc.