Reporting And Critical Evaluation Of Biological Research- Data Analysis Flashcards
What happens once raw data is collected?
the information is processed to interpret what it means.
How can data be processed?
Via graphs, mean, median, mode, standard deviation and range. Data should be presented in a clear, logical manner suitable for analysis
Were in a project would analysed data be presented?
In the results section
In the results What is useful to measure the variability between?
Between individual replicates of a data set and The reliability between independent replicates of an experiment/ study
What does the measurement of variability within and between a data set inform?
The reliability
What is standard deviation?
Standard Deviation is one way of telling how much each measurement of a data set differs from the mean of the data set. The greater the SD of a data set, the more variable the data (and the less reliable) formula: s=_/E(x-x’)^2/n
What do measurements of variation about the mean include?
-SD
-Standard error of mean
-Confidence intervals
What are error bars used to present?
Measures of variation about the mean (e.g. SD) on a graph.
What do smaller error bars mean?
The smaller the error bars the less variable the data and therefore the more reliable the data set is
What do larger error bars mean?
The larger the error bars the more variable the data and therefore the less reliable the data set is
Why will studies often involve several sets of data?
They are looking to see if the data for each set is different.
If there is a difference in means of control and treatment group how likely is it that this difference is due to chance alone?
Statistically tests are used to determine weather the differences between the means are likely or unlikely to have occurred by chance (you do not need to know specific statistical tests for the AH course)
What can interpreting error bars give an indication of?
Whether or not a difference is more or less likely to be significant.
What does it mean if error bars overlap?
The data is less likely to be significant
What does it mean if error bars don’t overlap?
The data is more likely to be significant
