|
|
For entry
into the higher level, it is not necessary to include trivial data, diagrams
and charts. But you should still look
at the tips, available on this site, which relate to the intermediate level data handling coursework.
Further
sheets on standard deviation, hypothesis testing and the normal distribution
are available at www.funkyfactsheets.co.uk
The report
at this level should read like a cross between a write-up for a science
experiment and an English essay.
SPP –
Specify the problem and planning
v
You
should choose a more demanding project so that you can use more advanced statistical
techniques.
v
You
should state your aims about using the more advanced techniques very clearly
and give valid reasons in your plan.
v
You
should explain any limitations that may arise.
v
During
the project you might refine your planning in light of your findings so far.
CPR – Collect Process and
Represent
v
You
should collect data that is relevant to the problem and takes variability and
bias into account.
v
You
should represent and analyse the data using box plots or cumulative frequency
diagrams for example.
v
You
should analyse the data using techniques such as standard deviation,
correlation etc.
v
Accuracy
and no redundancy. Your calculations
should be accurate, your diagrams correct and no redundancies.
v
Your
explanations should link all the diagrams and calculations together and should
relate back to the original problem.
IDR – Interpret and discuss
results
v
You
should make comments and correctly summarise the data you have collected.
v
You
should make your comments using the statistics, such as standard deviation.
v
Your
comments need to be relevant to the original hypotheses and you need to show
how significant they are.
v
Consider
your conclusions and how realistic they are.
If possible, suggest improvements.
Contents -
The Plan
1. Introduction. What are you going to compare. What types of variables you are looking
for.
2. Any obstacles? Outliers and what are you going to do with them? Anomalies, obvious wrong recordings, missing
values.
3. What do you expect in terms of patterns, trends, relationships, spread.
4. How are you going to get your sample? Why are you choosing this type of
sampling method and size? Growth Spurts.
5. List your hypotheses.
Get the Data
1.
Explain how you got it and record it. The data, available here,
may be recorded in the appendix.
2.
Work out all the statistics
you are going to use. These may all be
worked out using excel but you will need to explain your understanding of
measures of location and spread.
3.
Show the maths involved, although the computer will do all
the working out.
4.
Use excel to draw scatter graphs and the line of best fit
5.
Spot the anomalies and outliers, remove them and recalculate
any necessary statistics.
Analyse, Compare, interpret and Discuss
1.
Notice
the difference between the means when the outliers have been removed and
comment on the effect of large or small numbers appearing as outliers.
2.
Compare
the means with the medians as measures of location.
3.
Discuss
positive and negative skew as determined from the box plot.
4.
Discuss
the significance of the slope and the intercept of your lines of best fit.
5.
If you
have removed any outliers show them on the box plot.
6.
Test
your hypotheses
Conclusion
1.
What
would you have done if you had more time.
2.
How and
why would you extend your survey across the country and what about kids height
and weight across the world?