Volume 1 - Marketing Scales Handbook

VOLUME I

During the course of our academic careers we, like so many others, have wanted to engage in research using multi-item, summated ratings scales to measure marketing-related constructs. More recently in our work with doctoral candidates we have recognized the need for some central source of measures to aid students in their dissertation research. Up until now the conscientious researcher would either perform a tedious literature review in search of appropriate scales or would instead have to engage in the equally tedious job of scale development.

Therefore, this sort of handbook has been needed in our field for some time, but appears to be particularly welcomed in the 1990s. Scale development and use in marketing research though somewhat limited during the 1970s, increased dramatically in the 1980s. Therefore, after a decade of heavy scale use, the job of compiling those uses did appear to be overwhelming. In fact, some organizations approached for support thought it was a worthwhile goal but one that could not be accomplished. Luckily, we were not overly discouraged by their lack of support and set our minds on seeing the project through.

However, we did have to set limits on what we would attempt to do in a first volume. The two major limits had to do with the domain of published literature we would review: time frame and number of journals. We decided to cover the 10-year period from 1980 to 1989 for seven marketing-related journals: Journal of Marketing, Journal of Marketing Research, Journal of Consumer Research, Journal of the Academy of Marketing Science, Journal of Advertising, Journal of Advertising Research, and Industrial Marketing Management. Our logic was that if scales were of reasonable quality and relevance they would have been reported in one of these journals. Ultimately, Industrial Marketing Management was dropped from the domain because so few articles reported the use of scales and most of the scales were already being reported in another journal in the domain. Admittedly, there are two or three other journals that we would have liked to cover. We plan to include them in future volumes.

Not every scale that was used and discussed in the domain was written up. Scales were included only if sufficient information was available from known published sources or we were able to get necessary information from authors. The most critical information we required for including scales was the items themselves and an indication of reliability. We focused our efforts on simple multi-item scales, typically of the Likert or semantic differential variety. Also, preference was given to measures with three or more items. Though some two-item scales are included here, they are primarily from articles that used other, longer scales as well. So, if some scales appear to be missing it may be because they were too short or we were not able to secure enough critical information. Of course, there may be some innocent oversights as well. We suspect that errors and misstatements are present and will be found by users. We encourage people to let us know about them because we will make corrections for future printings and editions.

What to call scales was a problem. Our guiding goals were consistency and clarity. Researchers have not been consistent in referring to the same constructs with the same names and in many cases the names appeared to us to be very awkward if not misleading. So, we took it upon ourselves to look at every use of a scale, examine its items, compare the names that had been used, and then assign a name. We tried to make the names short and to have the first word or two provide the general idea of the construct being measured. In many cases descriptors were put at the end of scale names to indicate their specific applications; e.g., Opnion Leadership (Clothing).

The majority of the write-ups are at the "scale" level rather than the "use" or "construct" level. Consequently, for many constructs we report multiple scales and for many scales we report multiple uses. Where there were multiple uses of the same scale, they were written up together so the reader can easily determine most of what is known about the scale's psychometric characteristics and use in research. Because in many cases similar though not exactly the same versions of scales were used in more than one study, our rule of thumb was to write up scale uses together if the measures shared half or more of the items in common.

The major exception to this rule is for attitude-toward-the-ad, -brand, and -product scales. These write-ups include discussions of a variety of different scales that appear to measure the same construct but may not have items in common. This approach seemed to be the most parsimonious way to present these commonly assessed constructs that have been measured numerous ways.

We attempted in many cases to contact authors when the information available in known published sources was insufficient to fully describe a scale. We wrote to some authors two or three times, urging them to share the information with us so their work could be included and accurately reported. Though many were eager to comply, others chose not to. If possible, we still included the nonresponding authors' work, but in many cases critical information was lacking and we reluctantly decided to eliminate discussion of scale usage by those authors.

The write-ups are mostly descriptive, but we made some evaluation of quality as well. Though we tried not to be too critical of a scale's quality, it would have been inappropriate for us not to provide some word of caution to novice users about a scale with poor psychometric characteristics. If anything, we held back from making more negative comments that are probably justified. So, readers are urged to not rely exclusively on what we say about a scale, but to judge the quality of a measure by using whatever criteria are relevant to them.

We provided several means to assist users in finding scales of interest. First, the Table of Contents lists every scale included in the book alphabetically by section. Also, there are two indexes. One consists of key words and synonyms related to the discussion of particular scales. The other is an author index listing names of researchers who are linked with the development and use of specific scales. As shown in the Table of Contents, the scales are divided into three major groupings: consumer behavior-related scales, salesforce and organizational scales, and advertising-related scales. Measures within each major section are alphabetized for easy usage.

Finally, with the publication of this compilation, the time for developing or using "quick and dirty" scales out of ignorance or convenience should be past. Previously, researchers commonly used whatever scales were convenient to use and not necessarily the ones most appropriate and psychometrically sound for a situation. Too many times yet another scale was developed when several "good" ones were already available. Our intent is to make it considerably easier for researchers to determine whether a measure exists and, if more than one exists, to provide the information from which a logical selection can be made. By funneling the efforts of academia into industry, we hope to make the practical applications of the former more obvious to the latter. We trust that it is not too naive to hope that as researchers make better use of scales, the quality of market research will ultimately improve.

Comments: Dr. Gordon C. Bruner II

Office of Scale Research