Rounding with Decimal Numbers
Rounding with decimal numbers is really different than rounding with whole numbers. The biggest issue is deciding place value is the significant place to use for rounding, and this decision generally based on context. Often the context involves some sort of measuring device, and the precision of the measuring device is what determines the place value to which we will round. The best way to illustrate this is through an example. Suppose we wanted to compare the average weight of several sample boxes of cereal coming off the assembly lines at two different factories. The scale at Factory A give us the weights to the nearer thousandth of an ounce. The scale at Factory B give us the weights to the nearer tenth of an ounce. The greater precision of the scale at Factory A would be of no use to us, since our purpose is to compare the average from Factory A to the less precisely weighed average from Factory B. We would then round all the weights of the Factory A cereal boxes to the nearer tenth before computing and comparing the averages. The context determines that it makes sense to round these measurements to the nearer tenth.
To show a visual model of what happens when we round to the nearer tenth, consider the following. A box of cereal has a reported weight of 12.342 ounces. The goal is to round this weight to the nearer tenth. The 12 whole ounces are not involved in the rounding operation, so we will ignore them temporarily. We can represent the fractional part of an ounce by using base-10 blocks(Figure 7.8). We would represent 342 with three faces, four columns, and two cubes. The rounding represent question is, "Is this closer to three faces or to four faces?" Seeing the four columns and two cubes makes it easy to see that this is closer to three faces the than to four faces. If we had any amount less than five columns, the nearer face would be three faces. Five columns would be exactly halfway between three faces and four faces, and by rounding convention we round "up" to four faces. Clearly, any amount greater than five columns means that the nearer amount is four faces.
An alternative way to represent this same problem is on a numberline. Since the decimal numbers are used to represent amounts between the whole numbers, our numberline in this case needs to show the region between 12 whole ounces and 13 whole ounces. If we place 12.342 on the numberline it is, again, clear that this number is closer to 12.3 than it is to 12.4. In Figure 7.9 the x marks the spot for 12.342.
The visual model of the numberline helps us to see why 12.3 is the better tenth to round to, rather than 12.4. The same model can help us to see that 12 would be the better whole number to round to, rather than 13. If we had a close-up view of this numberline in which the hundredths were shown, we could see what hundredth would be the better hundredth to round to(12.34 or 12.35). The point is, when we place a number on a numberline we are always placing it between two place-value digit marks, and the rounding question is simply a question of which of these two marks is closer. The only arbitrary rule involved with rounding is what to do if the two marks are an equal distance from our number. In that case, the convention (rule) is round to the higher mark,
Whether we are representing rounding on a numberline or showing rounding with base-10 blocks or some other tool, the visual model of the number makes it easier to understand what rounding is all about. Typically, people who have difficulty understanding rounding were given too few opportunities to actually see models to represent the numbers. Often rounding is taught only as a series of steps of a procedure or a set of rules to follow. Without the benefit of visual models, rounding can be difficult to understand.