One of the points that appears in the chapter is that data contains a limited amount of information. You can summarize it but it is difficult to make it more accurate after the fact.
The concept of significant digits or figures states that you cannot add digits after the initial values. For example if you have a value of 3.5 you cannot arbitrarily make it 3.500. The value that you took is actually be between 3 and 4. It is closest to the middle but how close? The researcher may not know if it is 3.6 or 3.4. So looking at the raw data it would be unreasonable to assume that if you saw a value of 3.5 to assume that it is 3.500000000. You need to need the precision of the experiment and how many significant digits there are.
We can tie this concept back to the chapter by looking at the amount of information in the different scales. An ordinal scale will not have as much information in it as an interval scale. There is a hierarchy in that you can go downward from an interval scale to an ordinal scale but not vice versa.
Comments
Yes, let's hear it for levels of significance. It doesn't look like engineering calculators get used much in social science research. I guess it's a good thing we're not designing and testing flight-critical systems here - ouch!
I'm glad you gave the example of carrying out the decimal point. The other issue is when people make the mistake of trying to claim that p <.000000001 is a more significant result!