(dissenting).
I agree, not without misgivings, that there is a sufficient evidentiary basis in the administrative record to support the conclusion by the Federal Communications Commission that WICU-TV, Channel 12, Erie, Pennsylvania, projects a Grade B contour over Meadville. Thus I agree that under 47 C.F.R. § 76.92 (1975), WICU is entitled to signal priority over WFMJ-TV, Channel 21, Youngstown, Ohio, on the Meadville Master Antenna, Inc. (MMA) cable system. I do not agree, however, that the Commission’s action on MMA’s waiver petition can be approved.
When WICU requested signal priority, MMA petitioned for a waiver of compliance with the non-duplication rules. See 47 C.F.R. § 76.97. That petition put in issue both the signal strength and the signal quality of WICU’s transmission. On May 6, 1969 the Commission, without a hearing, denied the petition for a waiver. 17 F.C. C.2d 506 (1969). On September 2, 1970 it denied a petition for reconsideration. 25 F.C.C.2d 315 (1970). On petition for review this court held that the Commission’s brief and conclusory disposition of the waiver petition provided an insufficient basis for judicial review. We remanded to the Commission with directions to “set forth in detail its conclusions and the reasons for them on the issues of signal strength, as well as on black and white and color signal quality.” Meadville Master Antenna, Inc. v. FCC, 443 F.2d 282, 291 (3d Cir. 1971). Thus we required an adequate administrative record and statement of reasons both on the issue of WICU signal strength (the Grade B contour issue) and on the issue of signal quality.
The Commission has frequently taken the position that signal strength contours are the beginning and end of the priority inquiry for purposes of the non-duplication rule. See, e. g., Shen-Heights TV Association, 11 F.C.C.2d 814 (1968); Bluefield Television Cable, 10 F.C.C.2d 731 (1967); Mission Cable TV, Inc., 4 F.C.C.2d 236 (1966). Indeed, in this case the Review Board relied on those cases. 47 F.C.C.2d 147 (1974). But in Meadville Master Antenna, Inc. v. FCC, supra, we rejected such an approach as inconsistent with the public interest. Signal strength contours are expressions of the median of measurements on a grid over an entire geographic area. See 47 C.F.R. §§ 73.683-.686. Particularly in mountainous terrain like that surrounding Meadville, signal strength contours may tell nothing at all about signal quality in inhabited regions. Thus in our prior decision we instructed the Commission to take into account actual black and white and color signal quality of *222the competing broadcasters. 443 F.2d at 291.
When we remanded in 1971 the administrative record contained the 1966 Taylor Report. Our remand specifically directed the Commission to respond to that report. 443 F.2d at 290. It did so by holding that conditions had changed since 1966, and that a new evidentiary hearing was required. The majority opinion approves this treatment of the Taylor Report. Although MMA argues that this Commission holding was inconsistent with our prior opinion, I do not quarrel with the majority’s treatment of the reopening issue. Certainly the public interest permitted, if it did not require, the consideration of changed conditions.
But I cannot approve of the Commission’s treatment of the evidence presented in the reopened hearing. In that hearing MMA supported its waiver petition with evidence of a random sampling of its subscribers conducted in January 1973 by one Moodey, an assistant professor of sociology trained in empirical methods. The survey included a sampling directed to 418 out of about 8000 MMA subscribers, of whom 288 responded. The survey showed that viewers expressed a preference for WFMJ’s black and white picture quality by a 68% to 32% margin, and its color quality by an 81% to 19% margin. Particularly significant are the responses to questions 4 and 5 of the survey:
Question 4. Which of the stations has the best quality black and white picture of the items listed:
WICU Erie Channel 12 WFMJ-TV Youngstown Channel 21
(1) The least snow 36% (31) 64% (56)
(2) The sharpest picture 36% (33) 64% (59)
(3) The least double images 33% (24) 67% (49)
(4) The greatest contrast in the picture 36% (30) 64% (54)
Question 5. Which of the stations has the best quality color picture of the items listed:
WICU Erie Channel 12 WFMJ-TV Youngstown Channel 21
(1) The least snow 28% (40) 72% (104)
(2) The sharpest picture 26% (39) 74% (110)
(3) The least double images 26% (34) 74% (96)
(4) The most natural flesh tones 26% (37) 74% (105)
(5) The most consistent color quality 25% (36) 75% (106)1
When asked to indicate its degree of approval or disapproval of Commission action requiring MMA to omit the WFMJ signal the sample responded:
Strongly approve 0% (0)
Approve 8%' (19)
Do not care 23% (56)
Disapprove 30% (71)
Strongly disapprove 39% (95)2
Although the Moodey survey strongly reinforced the findings of the 1966 Taylor Report, the Review Board cavalierly discounted its significance. Inexplicably, however, the Review Board made no finding that the opinion sampling was unfairly or inadequately conducted. The majority opinion justifies the rejection of the Moodey survey. But neither the Administrative Law Judge nor the Review Board explained why the survey results could not be afforded “significant weight.” The suggestion that the failure of some persons included in the opinion sampling to express a preference should be given evidentiary significance in denigrating the weight to be given to the clear majority preference of those who responded seems to me completely inconsistent with the statistical assumptions which underlie the assumed validity of opinion surveying. Even the Commission did not, as does the majority opinion, convert a majority preference into a minority preference by treating the non-responses as *223equivalent to negative votes against the WFMJ signal. If the Commission had demonstrated why the Moodey poll was statistically insignificant or tainted I might be willing to approve its rejection. But as I read the decision under review, the Commission concluded that viewer preference as to signal quality was irrelevant to the waiver issue. I am simply at a loss to understand how that can be so.
The Communications Act of 1934 stipulates, and the Supreme Court has held, that the public interest is the polestar of Commission regulation. See 47 U.S.C. § 303; National Broadcasting Co. v. United States, 319 U.S. 190, 215-16, 63 S.Ct. 997, 1009, 87 L.Ed. 1344, 1362-1363 (1943). The non-duplication rule reflects one aspect of the public interest — the protection of local television service from unfair competition by CATV systems which might import identical signals from distant markets. See, First Report and Order, 38 F.C.C. 683, 697-715 (1963); Second Report and Order, 2 F.C. C.2d 725, 734-45 (1966); Meadville Master Antenna, Inc. v. FCC, supra, 443 F.2d at 284 n.2; Titusville Cable TV, Inc. v. United States, 404 F.2d 1187, 1190 (3d Cir. 1968). But the viewing public also has an interest in receiving the highest quality signal available. That being so, its perception of signal quality cannot be dismissed as irrelevant. See Cable TV, Inc. v. FCC, 428 F.2d 672, 680-82 (9th Cir. 1970).
The summary rejection of the Moodey survey evidence convinces me that the Commission’s action on remand was nothing more than a thinly-disguised effort to disregard this court’s prior mandate directing it to take into account not only predicted signal strength contours but also actual signal quality. The Commission’s present approach overemphasizes the guarantee of freedom from competition for local stations to the detriment of the competing interest of the viewing public. See Titusville Cable TV, Inc. v. United States, supra. The Commission’s inflexible application of its non-duplication rule in this case, which results in the exclusion of a local signal in favor of a second, prima facie inferior local signal,3 is in my view “arbitrary, capricious, an abuse of discretion, [and] otherwise not in accordance with law.....”5 U.S.C. § 706(2)(A). I would remand with instructions that the Commission weigh viewer preferences as to signal quality and explain why such preferences should not, in this case of marginal application of the non-duplication rule, control disposition of MMA’s waiver petition.
. The parenthetical values represent the raw responses to each question.
. The results of the Moody poll correspond with those of a viewer poll conducted in 1969 by MMA. At that time MMA mailed postcards to its 6,865 subscribers soliciting responses indicating that the reception of the WICU signal was inferior to WFMJ. Of 2,597 replies, 2,588 containing 4,040 signatures certified the inferiority of the WICU signal. See 443 F.2d at 290-91 n.24. The survey was not conducted consistent with statistically approved methods, and the solicitation itself was biased. See id.
. Erie is 30 miles from Meadville while Youngstown is 45 miles away.