Looking at real life example, we examine statistics related to Nelson rating. There is also an in-depth question on surveys.
1) Does a rating of 8.4% have twice the number of households as a rating of 4.2%?
Yes. This is because a percentage was calculated, and when we have percentages, the data can be compared in a linear fashion. We can say that since 8.4% is double 4.2% the proportional number of households is also double.
2) Percentage of household:
Our sample os 5000 households. There are 108,400,000 households in the country. To get our percentage we would divide
5000/108400000 and multiply it by 100 to get our percentage = 0.00461% of households. That is a very small percentage, but it is too expensive to poll any more people.
3) Nominal - this is a category of data where there is no data - the category of data is verbal in nature. So there are are several nominal categories - "Program Name" is one of them "Network" is another.
4) Ordinal data: - This is measurements with ordinal scales that are ordered in the sense that higher numbers represent higher values. You rank the data. So in this case, it is easy to see, since the rank is listed right in the title "rank" or "rank last week".
How can the data be ordered? Well, we can ordered it from 1, 2, 3... Also, this type of scale doesn't comment on the relative distance or variation between the two ranked options. So we can even set an arbitrary way to order the data, you can start the scale at 5, and order them this-way.
5) Interval data: On interval measurement scales, one unit on the scale represents the same magnitude on the trait or characteristic being measured across the whole range of the scale.
I would say the ...