In television audience research in The Netherlands (and similarly in other European countries) viewing behaviour is registered by the meter on a Per second basis. A change in channels is noted from the moment a viewer watches a channel for fifteen seconds or longer. Audience behaviour during channel changes lasting less than fifteen seconds is attributed to the last channel the viewer watched for fifteen seconds or longer. As consequence, in the case of a series of quick channel changes, a delay in the registration of channel changes occurs. Assuming that relatively more switching occurs during commercial blocks than during programs, it can be expected that audience ratings for commercial blocks would benefit more from a persistence threshold than audience ratings for programmes. A study was carried out to determine the effects of a fifteen second threshold on the viewing behaviour reported, and found this threshold has no effect on audience rating for commercials. The only effect was a very small increase in the audience ratings for the ten national Dutch channels in relation to the other, foreign channels.