Within each box, indicating a specific market type/system location, we further consider the two or three most likely lineups of available local signals. While we have not reported every combination which can occur, the cases tabulated are representative of the majority of signal patterns to be encountered and they cover a degree of variation sufficient to include most other possibilities. 8 CABLE PENETRATION At the time Comanor and Mitchell's research was under taken virtually no reliable statistical information was available to quantify the effects on cable penetration of the number, types and quality of local signals available, the additional cable signals provided, the price of cable service and the incomes of potential subscribers. That study provided estimates of most of these variables by use of multiple regression analysis on a randomly selected sample of 149 systems drawn from the Television Factbook. The authors noted that these systems were largely outside of the top 100 markets or in areas of quite poor reception, or both. Projection of penetration in the major markets under the thenproposed FCC rules (allowing four distant independent signals) was recognized as subject to considerable error. Since publication of the Comanor-Mitchell paper the measurement of factors determining penetration has been advanced considerably by Park in his study "Prospects for Cable in the 100 Largest Television Markets." Park uses statistical techniques closely related to those employed earlier. He improves on the ComanorMitchell study in three major ways: First, all 63 cable systems analyzed by Park had at least three A-contour, good reception-quality signals available off the-air. Second, all data were verified with system operators by telephone interview, insuring greater accuracy than available from only published sources. Third, two improved measures of signal quality were incor- porated into the analysis. each transmitter was explicitly included, and UHF signals were measured separately to account for more rapid signal attenuation with distance and the absence of UHF tuners in some households. The complete penetration equation as estimated by Park measures the effects of the following variables: .....number of off-the-air VHF signals, with separate ....number of off-the-air UHF signals, by the same ..number of cable signals, by the same categories .....color set penetration ..annual subscriber price .annual family income Park's research is particularly appropriate to the present assessment of the effect of alternative copyright fee schedules on the viability of cable systems in the major markets. In projecting penetration rates for the systems studied here the average figure predicted by Park's equation has generally been used, since this represents the central experience to be expected in the major markets. In addition, a selected number of intermediate sized systems have been analyzed using penetration rated 33% greater than predicted on average. Such increased penetration is definitely atypical, and would be expected to occur in only about one out of ten market situations, because of factors not fully accounted for in the penetration equation. 11 |