Flashcards in Maths Deck (19):
Percentage error =
-------------- X 100
Half the smallest graduation e.g. on a ruler with 1mm graduations it would be 0.5mm
Margin of error?
Uncertainty value double e.g. On a ruler with 1mm graduations the margin of error would be 1mm
You unintentionally avoid areas of your potential sample e.g. Muddy areas.
Even without bias it may not have been representative as it is impossible to test every single thing.
Why won't things be representative?
- sampling bias
Ways to avoid sampling bias?
- random sampling e.g. Random coordinates.
- using a larger sample, less chance of chance effect the results.
- analysis of data collected e.g. Stats tests, mode, median, mean and standard deviation.
Measuring variation from the mean in a sample. Smaller standard deviations results more clustered data which means less extreme values so it is more reliable. Standard deviation is a good measure of the reliability of the mean.
Normal distributions of standard deviation?
Standard deviation valve is smaller for A than B so the smaller the spread of data around mean A which is therefore a more reliable mean.
Binomial distribution, comparing two standard deviations?
Standard deviation value for A is smaller than B so smaller spread of data around mean A which is therefore a more reliable mean. No overlap between standard deviations means there is a significant difference between the two means.
Rate of reaction =
Null hypothesis: no significant difference in NUMBER
Appropriate table: normal two headed table
How to present data: bar chart
Student T test?
Null hypothesis: no significant difference in MEAN
Appropriate table: table with mean column
How to present data: bar chart with standard deviation bars
Null hypothesis: no significant difference in CORRELATION
Appropriate table: pairs of data in table
How to present data: scatter gram
At p = 0.01 to p = 0.05
Reject the null and accept the experimental as there is a significant difference
At p = 0.06 and above
Accept the null hypothesis and reject the experimental hypothesis as there is not a significant difference
What are the three stats tests?
Chi squared , student t test and spearmans rank.
Volume of solute needed =
Concentration you want
------------------------------- X total volume you want
Concentration you've got