You Can’t Get There from Here: A Response to Lohr and Lepczyk

The following comments were submitted by Frank Hamilton, president of the Animal Coalition of Tampa, Martha Girdany of the Kauai Community Cat Project, and myself, in response to Conservation Biology’s publication of “Desires and Management Preferences of Stakeholders Regarding Feral Cats in the Hawaiian Islands.”

Unfortunately, our critique of this badly flawed work was rejected by the journal. As editor-in-chief Mark Burgman explained, “the reviewers and handling editor have substantial concerns … the reviewers noted important and consistent concerns, the most significant of which is that the methodological issues raised in the comment were not sufficient to warrant publication.” Not surprisingly, my co-authors and I strongly disagree, and regret that Cheryl Lohr and Christopher Lepczyk were not required to defend their work (a trivial undertaking if, as the reviewers suggest, our concerns were off-base or overblown).

One often hears that science is self-correcting. The present case, however, supports the assertion, made in a 2012 Atlantic article, that self-correcting science is largely a myth.

•     •     •

In “Desires and Management Preferences of Stakeholders Regarding Feral Cats in the Hawaiian Islands,” authors Cheryl Lohr and Christopher Lepczyk [1] report, based on their analysis of survey results, that “live capture and lethal injection was the most preferred technique and trap-neuter-release was the least preferred technique for managing feral cats” in the Hawaiian islands. As we will demonstrate, however, a variety of flaws with the authors’ survey, sampling, and analysis undermine these claims. The study’s shortcomings, both technical and philosophical, are too numerous to address here; we will focus our attention, therefore, on the factors that contribute most significantly to Lohr and Lepczyk’s results, conclusions, and recommendations.

Among these factors is measurement error, which “is often the result of poor question wording or design and other aspects of questionnaire construction.” [2] Lohr and Lepczyk used inconsistent wording, for example, in their questionnaire (which was also the foundation of Lohr’s PhD work, for which Lepczyk served as Chair), referring to Felis catus variously as “feral cats,” “un-owned cats,” “cat (wild),” and “free-roaming cats.” [3] Farnworth et al., [4] whose work the authors cite, suggest that even cat owners, who, on average, “found nonlethal methods significantly more acceptable for stray cats than non-owners did,” felt differently when the same cats were described as “feral” (as distinct from “stray” or “companion”). By choosing the terms they did, Lohr and Lepczyk created a survey likely to elicit responses favoring lethal methods.

Contrary to what Lohr [5] suggests, the sampling methods (e.g., “direct solicitation at conferences and expos, Internet searches, assessment of organization membership lists, and snowball techniques”) she and Lepczyk used do not conform to “the tailored design method” described by Dillman et al., [2] who warn of the risks associated with such non-random sampling. And their relatively small sample sizes (22­–483) and large nonresponse rate (20.0­–54.8%) preclude the kinds of generalizations the authors make.

Far more problematic, however, is the lack of representativeness associated with Lohr and Lepczyk’s sample. Only “if all members of the population have an equal chance of being selected in the sample,” explains Babbie, [6] can “descriptions and explanations derived from an analysis of the sample… be assumed to represent similar ones in the population.” Lohr and Lepczyk [1] explain that their 6 pre-identified stakeholder groups “have been involved in previous cases of conservation conflict in Hawaii,” although it’s clear from Lohr’s dissertation [5] that these previous cases involved only feral pigs (Sus scrofa) and axis deer (Axis axis). Domestic cats (Felis catus), by contrast, are considered “pet animals” under Hawaii law [7] and are owned by 16.2% of households in the state. [8]

Furthermore, 82.5% of Lohr and Lepczyk’s “random residents” sample “lived in a rural area or small town” whereas “only 10% of Hawaii’s population live in rural areas with fewer than 50,000 residents.” [5] And “considerable differences exist between urban and rural residents… in their animal rights and welfare orientations” [9, as cited in M. J. Manfredo, Pierce, Fulton, Pate, & Gill, 1999]. Lohr [5] also acknowledges that 24% of “random residents” indicated that they hunted at least once per year, more than 34 times the expected rate (0.7%) based on hunting licenses purchased in 2009.

Because Lepczyks and Lohr’s sample was not representative of the target population (either at the stakeholder level or, more broadly, the state level), their results, conclusions, and recommendations are at best questionable. We worry, however, that the sophisticated nature of the authors’ analysis techniques give their results an air of credibility and therefore also warrant a response.

Lohr and Lepczyk [1] explain that survey response data were converted “into weights… via the linear weight variable slope model” described by Alfares and Duffuaa (2009) and then “entered into the [consensus convergence model] described by Regan et al. (2006).” However, whereas Alfares and Duffuaa [10] asked respondents to rank a list of criteria “in the order of importance,” Lohr [3] asked her respondents to rate each one using a Likert scale and allowed them to “use the same number more than once.” We are unaware of others applying Alfares and Duffuaa’s model to ratings. If, as appears to be the case, Lohr and Lepczyk are the first, we would expect them to acknowledge as much and demonstrate the validity of their novel approach. A similar issue arises from the authors’ application of the “adapted original consensus convergence model (CCM)” described by Regan et al., [11] which is also based on weights derived from rankings, not ratings.

More troubling, however, is the extent to which Lohr and Lepczyk dismiss the very premise of CCM. As Regan et al. [11] explain, the process is an iterative one in which each participant “has opinions about the expertise and rationality of other members in the group” and “agrees to the formal consensus process… [and] to compromise their views in light of the views of others.” Lohr and Lepczyk [12] acknowledge that “incorporating the views of many stakeholders [via CCM] will not be beneficial if the conflict to be addressed involves two entrenched and opposing points of view.” Indeed, Lohr [5] suggests as much in her dissertation: “the opinions held by animal rights advocates versus the opinions held by environmentalists regarding the management of feral cats could be described as polar opposites and entrenched.” It’s difficult to reconcile this acknowledgement that CCM is inappropriate for their research topic with the authors’ subsequent use of the method.

Surveys have no mechanism for securing the kind of agreements described by Regan et al. And the “transactional approach” Lohr and Lepczyk [12] see as a “limitation” to be overcome is in fact an indispensible component of Regan et al.’s CCM. Once again, the authors are applying methods developed by others in ways different from their original intent without acknowledging as much, and without providing evidence that such an application is valid.

In light of their finding that “live capture and lethal injection was the most preferred technique and trap-neuter-release was the least preferred technique for managing feral cats,” it’s curious that Lohr and Lepczyk [1] failed to acknowledge survey results that revealed vastly different attitudes. [13, 14] Of particular interest is a 2011 Ward Group survey of Hawaii residents revealing that 85% of respondents support trap-neuter-return, compared to just 16% who support lethal control methods. [15]

Nevertheless, Lohr and Lepczyk [1] suggest that there would have been even greater support for lethal methods had respondents been better informed about the “environmental impacts of feral cats,” and in particular “that cats spread diseases such as toxoplasmosis.” However, the authors have no way of assessing respondents’ knowledge of the role domestic cats play in toxoplasmosis transmission—which, in any case, has often been exaggerated and misrepresented in both the mainstream media [16, 17] and scientific literature. [18-20]

We agree that Lohr and Lepczyk’s survey results would likely have been different had her respondents been better informed, but suggest that support for non-lethal methods would have increased, not decreased. We assume that few, if any, respondents were aware of the unpalatable techniques (e.g., feline distemper, poisoning, hunting and trapping, and dogs) and enormous costs (e.g., nearly $2,200, in 2013 dollars, for each of the 635 cats eradicated from Ascension Island) involved with successful eradication programs, for example. [21, 22] Recent modeling work reveals relatively modest costs by comparison, but suggests that the cost of lethal control is 4.5–9 times greater than costs associated with trap-neuter-return. [23] One must also consider the various costs (e.g., ecological, economic, etc.) associated with the imbalance that would likely result from the island-wide removal of cats and the subsequent increase in non-native predators, such as black rats (Rattus rattus). [24, 25]

Lohr and Lepczyk suggest that their work has revealed what residents of Hawaii want in terms of “feral” cat management goals (i.e., a reduction in abundance) and their preferred technique (i.e., live trapping followed by lethal injection). As we have demonstrated, however, their poorly constructed survey, non-representative sampling, and flawed analysis preclude any such claims. Furthermore, we consider the undertaking of this work to be misguided at best, and, in light of the obvious policy implications, its publication irresponsible.

Literature Cited

1. Lohr, C.A. and C.A. Lepczyk, Desires and Management Preferences of Stakeholders Regarding Feral Cats in the Hawaiian Islands. Conservation Biology, 2013: p. n/a-n/a.

2. Dillman, D.A., J.D. Smyth, and L.M. Christian, Internet, Mail, and Mixed-mode Surveys: The Tailored Design Method. 3rd ed. 2009: John Wiley & Sons, Inc.

3. Lohr, C.A. and C.A. Lepczyk, Desires and Management Preferences of Stakeholders Regarding Feral Cats in the Hawaiian Islands (Supporting Materials). Conservation Biology, 2013: p. n/a-n/a.

4. Farnworth, M.J., J. Campbell, and N.J. Adams, What’s in a name? Perceptions of stray and feral cat welfare and control in Aotearoa, New Zealand. Journal of Applied Animal Welfare Science, 2011. 14(1): p. 59–74.

5. Lohr, C.A., Human dimensions of introduced terrestrial vertebrates in the hawaiian islands, in Natural Resources and Environmental Management2012, University of Hawaii at Manoa.

6. Babbie, E.R., The Practice of Social Research. 12th ed. 2009: Cengage Learning.

7. HRS-0711, in Hawaii Revised Statutes2014.

8. n.a., A Quantitative Survey of Cat Owner and Cat Feeder Attitudes, 2011, Ward Group (prepared for the Humane Society of the United States).

9. Richards, R.T. and R.S. Krannich. The ideology of the animal rights movement and activists’ attitudes toward wildlife. in Transactions of the North American Wildlife and Natural Resource Conference 56. 1991.

10. Alfares, H.K. and S.O. Duffuaa, Assigning Cardinal Weights in Multi-Criteria Decision Making Based on Ordinal Ranking. Journal of Multi-Criteria Decision Analysis, 2009. 15: p. 125–133.

11. Regan, H.M., M. Colyvan, and L. Markovchick-Nicholls, A formal model for consensus and negotiation in environmental management. Journal of Environmental Management, 2006. 80(2): p. 167–176.

12. Lohr, C.A., L.J. Cox, and C.A. Lepczyk, Patterns of hypothetical wildlife management priorities as generated by consensus convergence models with ordinal ranked data. Journal of Environmental Management, 2012. 113(0): p. 237–243.

13. Chu, K. and W.M. Anderson, Law & Policy Brief: U.S. Public Opinion on Humane Treatment of Stray Cats, 2007, Alley Cat Allies: Bethesda, MD.

14. Karpusiewicz, R. Poll: Americans Favor No-Kill Animal Shelters. 2012.

15. Gibson, I., Hawaii Case Study: Developing Productive Partnerships to Protect Cats and Wildlife, in The HSUS Cats Outdoors Conference2012: Marina del Rey, CA.

16. Hamilton, J. A Parasite Carried By Cats Could Increase Suicide Risk. Shots (NPR Health Blog), 2012.

17. Choi, C.Q. Mind-Controlling Parasite May Increase Brain Cancer Risk. LiveScience, 2011.

18. Gerhold, R.W. and D.A. Jessup, Zoonotic Diseases Associated with Free-Roaming Cats. Zoonoses and Public Health, 2012

19. Fredebaugh, S.L., et al., Prevalence of antibody to Toxoplasma gondii in terrestrial wildlife in a natural area. Journal of Wildlife Diseases, 2011. 47(2): p. 381–92.

20. Jessup, D.A. and M.A. Miller, The Trickle-Down Effect. The Wildlife Professional, 2011. 5(1): p. 62–64.

21. Ratcliffe, N., et al., The eradication of feral cats from Ascension Island and its subsequent recolonization by seabirds. Oryx, 2010. 44(01): p. 20–29.

22. Bloomer, J.P. and M.N. Bester, Control of feral cats on sub-Antarctic Marion Island, Indian Ocean. Biological Conservation, 1992. 60(3): p. 211–219.

23. Zawistowski, S., Simulating different approaches for managing free-roaming cat populations, in 2013 National Council on Pet Population Research Symposium Presentations: CATS: The Ins and Outs: Improving their Future Through Research2013, Society of Animal Welfare Administrators Tempe, AZ.

24. Courchamp, F., M. Langlais, and G. Sugihara, Cats protecting birds: modelling the mesopredator release effect. Journal of Animal Ecology, 1999. 68(2): p. 282–292.

25. Fan, M., Y. Kuang, and Z. Feng, Cats protecting birds revisited. Bulletin of Mathematical Biology, 2005. 67(5): p. 1081–1106.

Share and Enjoy:
  • Facebook
  • Twitter
  • Reddit
  • StumbleUpon
  • Technorati
  • Digg
  • Sphinn
  • Mixx
  • Google Bookmarks
  • Blogplay
  • LinkedIn
  • Wists
  • email
  • Print
  • RSS