Criteria: DCP Aspect 2

Having collected the data the next stage is to process it, if there is no processing then you can't award any marks. This is a problem if students are designing there own experiment and simply want to graph the raw data without processing, they haven't done anything wrong but can't get any marks for processing. That's why I give many other examples with straight forward processing so when they don't need to worry about DCP in design experiments. I still give a grade though but explain why it can't be more than 4.

It's a good idea to display the process the data in a separate table to the raw data. I like students to find the average in the raw data table since it makes more sense to see the 5 values and the average in the same row. This average value along with the independent variable is then taken on to the processed data table. The actual calculations in excel are of course all being done in the one big table, it just when its taken to word that it's split up.

The table should include the quantity/units in the header if the uncertainty has been processed then there is no need to put an uncertainty in the header since it will be in a new column. Uncertainties don't have uncertainties of course.

Here is the sort of thing we are looking for:

In addition to the table there should be an explanation of what calculations they have done. I don't think they need to say things like "I squared t by multiplying it by itself" but probably should say how they processed the uncertainties.

Note In the table above the significant figs in average time isn't consistent with the uncertainty, this is OK since this column is just being used to find the uncertainty in t² .

 

Checklist: DCP Aspect 2

The data should be processed in some way, for example averaging, squaring or finding the sine.  Processed data should be displayed in a table separate to the raw data table.

According to the IB clarifications of the IA criteria: "It might be that the data is already in a form suitable for graphical presentation, for example, light absorbance readings plotted against time readings. If the raw data is represented in this way and a best-fit line graph is drawn and the gradient determined, then the raw data has been processed." I think this becomes a bit confusing since the graph is assessed in aspect 3 so how can you award marks for the graph in aspect 2 and 3. I like to keep it simple and make a clear difference between processing and presenting so I only use examples where there is a clear "processing" stage.

The table must have headers that include units and uncertainties

Unless the column is a column of uncertainties or some interim stage of the processing.

Calculate uncertainties in the repeated measurements by finding the ½(max value – min value) in the spread of data.

If the dependent variable has not been measured repeatedly then the estimated value for the measurement are is used. If the uncertainty from the spread of data is smaller than the measured uncertainty (sometimes it's zero) then take the measurement uncertainty. There is no need to calculate standard deviation or  any other statistical analysis but you can do it if you really want to.

Calculate the uncertainties in processed data by calculating the ½(max value – min value) e.g. if uncertainty in time is 0.2 then uncertainty in t² is ½[(t+0.2)² –(t-0.2)²]

For a more detailed discussion of uncertainties see the section on uncertainties. The more traditional method to propagate errors is by taking % errors but I find my students understand the ½(max-min) method much better. If they have a range of values the max and min functions in excel can be used rather than adding and subtracting the uncertainty from the average value.

The number of decimal places in each column must be consistent with each other and the uncertainty.

A simple rule is to always quote the uncertainty to 1 significant figure than reduce the number of decimal places in the data to the same number of decimal places as the uncertainty. The uncertainty won't always be the same number of decimal places for all values but that's ok.

Any calculation must be explained.

Within reason.