An experiment to determine the relative effectiveness of newspaper advertising schedules as affected by advertisement size and frequency of insertion