Balázs Kovács, Michael T. Hannan
Sociological Science, May 13, 2015
DOI 10.15195/v2.a13
Abstract
A general finding in economic and organizational sociology shows that objects that span categories lose appeal to audiences. This paper argues that the negative consequences of crossing boundaries are more severe when the categories spanned are distant and have high contrast. Available empirical strategies do not incorporate information on the distances among categories. Here we introduce novel measures of distance in conceptual space and derive measures for typicality, category contrast, and categorical niche width. Using the proposed measurement approach, we test our theory using data on online reviews of books and restaurants.
A general finding in economic and organizational sociology shows that objects that span categories lose appeal to audiences. This paper argues that the negative consequences of crossing boundaries are more severe when the categories spanned are distant and have high contrast. Available empirical strategies do not incorporate information on the distances among categories. Here we introduce novel measures of distance in conceptual space and derive measures for typicality, category contrast, and categorical niche width. Using the proposed measurement approach, we test our theory using data on online reviews of books and restaurants.
- Citation: Kovács, Balázs, and Michael T. Hannan. 2015. “Conceptual Spaces and the Consequences of Category Spanning.” Sociological Science 2: 252-286.
- Received: July 21, 2014
- Accepted: September 24, 2014
- Editors: Olav Sorenson
- DOI: 10.15195/v2.a13
Four notes on this interesting paper:
a. While the measurement strategy presented in the paper is quite compelling, the overall claim to progress is a stretch. The motivating observation is that the prior literature has made an “assumption” that “lacks credulity”—i.e., “that all kinds of spanning are expected to have the same consequences (pp. 252-3).” But I cannot think of a paper that relies on such an assumption even implicitly. And to the contrary, note how (i) Phillips-Turco-Zuckerman 2013 [PTZ henceforth] is expressly about how different kinds of spanning have different consequences; and (ii) measures of categorical “coherence” (see Zuckerman ASR 2004 and Zuckerman and Rao ICC 2004) effectively take distance among categories into account: such distance is reflected in the lack of overlap in coverage such that a wide-spanning object will look especially incoherent.
b. On p.255, the paper notes that one reason audiences downgrade category-spanning offers is that skill-development usually requires specialization and so audiences usually expect category-spanners to be low performers (The correct cite is Zuckerman et al. 2003, not Phillips & Zuckerman 2001). But the theory and discussion of results revolve solely around the idea that audiences are confused by category-spanners. Two reasons it is problematic to ignore the performance story: (i) PTZ show that audiences in a service market do not lower their evaluations when they are confused about an unusual category combination as long as it doesn’t raise questions about performance; (ii) It is curious that consumers *who have already selected and consumed a product/service* would downgrade it based on how it is categorized. One might think when an audience is confused and therefore exhibits “lower attention” and even “aversion or repulsion” to offers (p.265), this would affect which offers the audience may consider, not its ratings after direct experience of performance.
c. This same issue reappears in a different way in the paper’s treatment of “middle-status conformity” (p.280). The paper suggests that some actors (or only “organizations”?) need not be clearly categorized because their “individual identity” is so “strong” and “visible.” But as long as an actor is in competition with others (status does not imply a lack of competition), audiences will still try to compare its offers with others. In existing theory (see PTZ for a revised version), “pressures” for conformity are “weaker” not because high-status actors are less confusing but because high status (and the recognition of high capability and commitment it implies) necessarily entails category membership. Accordingly, while the paper suggests that high-status actors must remain “consistent with (their) individual identity,” recent research suggests instead that high-status actors can be quite inconsistent (Sgourev & Althuizen ASR 2014). The reason is that membership norms are used to screen on *minimal* performance but nonconformity often leads to *higher* performance.
d. On p.256, the paper cites Zuckerman (1999) as one of “many… studies (that) collect overall assessments of objects without knowing the evaluator’s conceptual focus.” In fact though, there is no basis for wondering whether analysts (or more importantly, investors; the DV is market price) focus on individual business segments or the entire firm. They *must* focus on the entire firm because their orientation to the firm is as an indivisible, tradable asset. See the intro of PTZ for a quick take on this important issue.