Contacter nous

Publications

A Method for Handling a Large Number of Attributes in Full Profile Trade-Off Studies

Full profile conjoint- or choice-based trade-off studies have traditionally been limited to six attributes. Full profile studies allow for the estimation of interaction terms and generally present more realistic choices to the respondent than partial profile or self-explicated approaches. However, clients often want to test a long list of potential product features that may or may not be included in…

Read More

A Taxonomy and Assessment of Current Market Research Conferences in North America

Do you find this issue’s column title impressive? Ostentatious?  Incomprehensible?  Your answers (multiple responses allowed) will indicate which conferences you should, and should not, attend.  I’ll explain further down. Marketing research conferences can be divided into two broad classes: Vacuous and Substantive. Please note I’ve excluded invitation-only conferences from this taxonomy.  I’ve also excluded seminars and workshops. Almost all marketing…

Read More

A User's Guide to Conjoint Analysis

Conjoint Analysis is the most powerful and important family of analytic techniques in all of marketing research. How else can you answer, all at once, questions of this strategic and tactical magnitude: What price will maximize my profits? What features should my product have? How many of these will I sell? Who will buy them? Why will they buy them?…

Read More

Abbreviated Task Sets

The author examines four commercial data sets to determine how few choice tasks and/or how few respondents are required for generating reasonably accurate disaggregate utilities when Hierarchical Bayes utility estimation is employed.  The author demonstrates that, in carefully  designed and analyzed studies, as few  as one to four choice tasks  per respondent can yield accurate disaggregate choice models. To continue…

Read More

An Examination of the Components of the NOL Effect in Full-Profile Conjoint Models

The existence of the number of levels effect (NOL) in conjoint models has been widely reported since 1981 (Currim et al.). Currim et al. demonstrated that the effect is, for rank-order data, at least partially mathematical or algorithmic. Green and Srinivasan (1990) have argued that another source of this bias may be behavioral. Although NOL can significantly distort study findings,…

Read More

Are Methodologists Becoming Irrelevant?

I imagine that, as dinosaurs slowly slipped into extinction, they were the last ones to notice. It might be the same with marketing research methodologists today. Being a methodologist of sorts myself, it took an unfortunate and seemingly unrelated event to bring the possibility of my professional demise into focus. Recently, the AMA in a rather clumsy way, purged Marketing…

Read More

Bring Your Survey Design Out Of The Dark Ages

Take a questionnaire written last week and place it side by side with one written 20, 30 years ago.  Chances are they will look identical.  Same logic.  Same skip patterns.  Same batteries and scales.  Same limitations.   Even though today’s questionnaire is most likely being programmed on the web, with all the new question formats and controls web surveys offer.  Yet…

Read More