IB Data Analysis Practice 1

In this activity, you will practice creating the most common graph used in IB Biology, the scatter
plot. Additionally, you will conduct a correlation test and use standard deviation to depict the
degree of uncertainty in the data. It may be helpful before, during, or after completion of this
activity to examine the explanation of IB Lab Standards that outlines the specific requirements
for each component of the Data Collection Standard. They can be found here.

Part A: Creating a Table

Create a table like the one below in Excel.
Tables should be arranged so that rows/columns are labeled. All rows/columns
must have units, uncertainties, and units.
All tables must have a title that specifically indicates what is being presented
(this one is done for you).
Tables are generally arranged in this manner so to make calculations easier in future
steps.

*The mass for each bird sample (1-5) is being measured per beak length (12-16mm) so the
numbers under (12-16mm) are the mass measurements for each bird.

Part B: Calculating Mean and Standard Deviation.

Excel makes it extremely easy to calculate mean and standard deviation very quickly for large
data sets. Calculate both mean and standard deviation of these in Excel (don’t use a
calculator) by watching the following video:
Calculating mean, median, mode, and standard deviation in Excel.

Part C: Creating an X-Y Scatter Plot Graph.

 An X-Y Scatter Plot graph allows two numerical data groups to be compared with one another.
For example, how does the mass of birds compare with their beak length. In this example, and
for nearly every data collection opportunity, there will be multiple trials for each variable. In this
case there are five different beak lengths identified and five bird mass samples for each bird
beak length. Using the calculated mean, create an X-Y scatter plot graph for the sample data
using the following video:
How to create an X-Y scatter plot graph

Part D: Creating Error Bars on an X-Y Scatter Plot Graph.

Anytime data is collected for a number of samples of the same category, there will almost
always be variation in said data. By using standard deviation, one can calculate the standard of
the mean as a way to indicate the reliability of measurements as an indicator of the true mean
for the entire population. Using this value and a confidence interval of 95% can be used to
show, with 95% confidence, what values would the true mean would be expected to fall within.
Using the calculated standard deviation, standard error of the mean and 95% confidence
interval add error bars to your graph after watching the following videos (Google Sheets can
not add customized error bars and thus must be created in Excel. Your Google Sheets
file can be downloaded on a Mac as an Excel file, add error bars, and then re-upload to
Google Drive as a Google Sheets file):
How to calculate Standard Error of the Mean & 95% Confidence Interval
How to create Excel Error Bars using 95% Confidence Interval

Part E: Creating a Linear Regression Correlation Test

A correlation test can indicate the correlation or degree of relatedness between two data sets.
As discussed in class, this is merely a suggestion and does not necessarily prove two data sets
are correlated. However, the test can be useful and will be used throughout IB Biology. For
example, a correlation test could indicate the relatedness between bird beak length and mass.
A correlation test can be accomplished in Excel or Google Sheets by adding a line of linear
regression; it is also possible to add lines of polynomial regression which may be useful for data
sets that both increase and decrease rather than just one. Using the same graph and data, add
2
a linear regression line and R value to indicate the correlation between bird beak length and
2
mass in which the R value is the correlation value by watching the following video:
How to create Excel Linear Regression

Part F: Turn-In through schoology

Insert Data Tables & Graphs Below:

Assessment: Analysis
IB Mark Analysis Descriptor

5-6 The report includes sufficient relevant quantitative and qualitative raw data that
could support a detailed and valid conclusion to the research question.
Appropriate and sufficient data processing is carried out with the accuracy
required to enable a conclusion to the research question to be drawn that is
fully consistent with the experimental data.
The report shows evidence of full and appropriate consideration of the impact of
measurement uncertainty on the analysis.
The processed data is correctly interpreted so that a completely valid and detailed
conclusion to the research question can be deduced.

3-4 The report includes relevant but incomplete quantitative and qualitative raw data
that could support a simple or partially valid conclusion to the research
question
Appropriate and sufficient data processing is carried out that could lead to a
broadly valid conclusion but there are significant inaccuracies and
inconsistencies in the processing
The report shows evidence of some consideration of the impact of measurement
uncertainty on the analysis
The processed data is interpreted so that a broadly valid but incomplete or limited
conclusion to the research question can be deduced

1-2 The report includes insufficient relevant raw data to support a valid conclusion to
the research question
Some basic data processing is carried out but is either too inaccurate or too
insufficient to lead to a valid conclusion
The report shows evidence of little consideration of the impact of measurement
uncertainty on the analysis
The processed data is incorrectly or insufficiently interpreted so that the conclusion
is invalid or very incomplete

0 The student’s report does not reach a standard described by the descriptors above.