ghs in appellants' favor. Under the LanhamAct, in camera inspection is not sufficient; the district courtmust view the marks "as they appear in the marketplace." E& J Gallo Winery v. Gallo Cattle Co., 967 F.2d 1280, 1291(9th Cir. 1992).[15] Appellants presented evidence of actual confusion, thefourth factor. Both Ratzenberger and Wendt stated in theirdeclarations that they have been approached by members ofthe public who commented on the similarity between theappellants and the robots at the Cheers airport bars: "Theusual comment is some variation on `Hey George, I just hada drink with you in Kansas City.' " They also submitted evi-dence of consumer confusion to the district court prior tosummary judgment in the form of survey evidence. The courtrejected this evidence as "not a good survey. " For reasonsdetailed below, this evidence should not have been excluded.Sufficient evidence exists by which a reasonable jury mightinfer actual consumer confusion.[16] The fifth factor, marketing channels used, weighs inthe appellants' favor. The allegation is that Host is appropriat-ing appellants' likenesses because the target audience of theCheers bars are customers who are fans of the televisionseries. Such a similarity in marketing channels suggests thatthere is at least a likelihood of consumer confusion.[17] The sixth factor, likely degree of purchaser care,weighs in favor of appellants as well. Consumers are notlikely to be particularly careful in determining who endorsesor is affiliated with an airport bar in which they might pur-chase only a single beverage. They will be even less likely toscrutinize the source of the animatronic figures which are notfor sale, but are used instead to attract patrons to the bars. 12364This low degree of care makes confusion of sponsorshiplikely. See White, 971 F.2d at 1400 ("consumers are not likelyto be particularly careful in determining who endorses VCR's,making confusion as...
