Loyalty - Attitude, Behavior, and Good Science: A Third Take on the Neal-Brandt Debate
Grisaffe, Doug, Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior
This article first appeared as a Walker Information white paper. It is reprinted here to increase its availability in the hope of stimulating continuing dialogue on this topic. It is reprinted by permission of Walker Information who copyrighted it in 2001.
In the interest of academic fairness, full discussions from each party represented in this "debate" are available in white paper form on their respective web sites. Grisaffe's paper can be accessed at www.walkerinfo.com/resources, then click on "white papers", then click on "Loyalty-- Attitude, Behavior, and Good Science...." Brandt's paper can be accessed at www.burke.com, then click on "search" and type into the search box "Attitude does matter by D. Randall Brandt". Neal's paper can be accessed at www.sdrnet.com, then click on "Analytical Services", then click on "Loyalty Modeling" then click on "A Rebuttal".
DEBATING LOYALTY AND LOYALTY MEASUREMENT
In the June 5, 2000 issue of Marketing News, William Neal, respected authority on marketing research says categorically, "Loyalty is a behavior." He says, "If I purchase in a product category 10 times in one year, and I purchase the same brand all 10 times, I am 100% loyal. If I purchase the brand only five out of 10 times, I am 50% loyal." Neal also says it is "ridiculous" to attempt to measure loyalty with three questions-- overall satisfaction, recommend intent, and repurchase intent. These three questions, says Neal, will likely correlate at least .80. Measuring intent to recommend and intent to continue in addition to measuring overall satisfaction is tantamount to "measuring the same thing two more times," according to Neal (an expanded discussion can be found on SDR's website).
Naturally these statements cry for rebuttal by Burke, Inc. because Burke uses exactly those three questions in their approach to loyalty research. Replying to Neal in the August 14, 2000 issue of Marketing News (with an expanded discussion on Burke's website), D. Randall Brandt, a respected authority himself, states, "we take a position that is strongly opposed to the one offered by Mr. Neal." Unlike Neal's behavior-- only view, Brandt states his firm's position-- loyalty is "reflected by a combination of attitudes and behaviors." Brandt goes on to defend the three specific items by noting that while correlated, the measures are not redundant. Scoring highly on one does not necessarily mean scoring highly on all. But, says Brandt, scoring highly on all is an indication of being a "secure customer." Brandt says the three items can serve as leading indicators of a variety of actual behaviors surrounding loyalty (e.g., repeat purchase, customer retention) once an association has been established empirically.
So we have competing opinions about the nature of loyalty. We also have competing opinions about appropriateness (or lack thereof) of measurement with the three items: satisfaction, recommend and continue. I'm compelled to chime in with a third perspective on some of the points raised by Neal and Brandt. I suggest that (a) previous literature in our field, (b) specification of causal relationships, and (c) scientific principles related to measurement and modeling, can help to shed some light on the debate.
WHAT DOES THE LITERATURE SAY: LOYALTY AS BEHAVIOR ONLY, OR ATTITUDE AND BEHAVIOR?
First, let's consider the nature of loyalty. Is it attitudinal and behavioral as described by Brandt, or is it behavioral only as argued by Neal? As Brandt has pointed out, the attitude and behavior perspective seems to have prevailed in the literature as early as the 1970s. Indeed in 1969, George S. Day, a pillar in our field, argued that loyalty involved both attitude and behavior. Other early theorists also promoted this view (e.g., Richard Lutz and Paul Winn). The classic text is probably Jacoby and Chestnut (1978), Brand Loyalty: Measurement and Management, published by Wiley. …