Key Points for ISAs

1
Key Points for ISAs

• In the last ISA a number of weakness were shown
  by pupils.

2
Tables
Independent variable
Dependent variable
B
A
Temp of water /oC
Time to react /s
Heading and units
21 34 56 74 97
Range 21-97oC 76oC
The range is good if it has an appreciable effect
on the outcome (the dependent variable)
3
Conclusion

• This is how B is affected by A

4
Repeating results

• This improves the reliability of the results.
• Repeated results are not needed if there is a
  clear pattern.
• If results are repeated then the results table
  should show them and the mean.
• Anomalous results should be ignored and the
  others averaged.

5
Sensitivity

• This is the smallest change that the measuring
  equipment can measure.
• More sensitive equipment (or more divisions) will
  give more precise results.

6
Reliable

• Repeating measurements and comparing to other
  groups who have used a similar range improves
  reliability.

7
Preliminary Tests

• These are carried out to determine a suitable
  range for the independent variable.

8
Identifying Overcoming Errors

• E.g. Systematic

Zero error zero input give a reading
A 0 2 4 6 8
B 8 12 17 22 34
Every result affected systematic error
Adjust every result by deducting 8
9
Overcoming Errors (Random)

• E.g.

Here, there is a clear pattern but 27 does not
fit, therefore this is a random error. It is an
anomalous point.
A 2 4 6 8 10
B 5 10 27 20 25
How can we overcome this?
By repeating the measurement ignore the
anomalies when averaging or drawing graphs.
10
Errors
Possible Systematic error e.g. no batteries but
voltmeter has a reading
Random error
11
Graphs (good use of graph paper)
Pencil
Dependent variable e.g. time /s
Linear scale Do not place numbers on here
randomly
Units
Independent variable e.g. temperature /oC
12
Graphs

• Continuous variables give line graph
• Categoric variables give bar chart
• Continuous variable, use linear scale and create
  a line graph, therefore it is better than a bar
  chart.
• Anomalous results should be ignored on the
• graph.