Laboratory Skills 5 : Graphs, variables & distribution

Question 1

What does a basic graph ( x-y scatterplot) include?

Answer 1

• Title

Axes with :

• Label, Units, Scale
• X-axis: independent variable
• Y-axis: dependent variable

Trendline:

• To make trend more visible
• To allow interpolation (and extrapolation)

Key- If more than one data series

Question 2

what kind of line is drawn to the axes when reading a graph?

Answer 2

Dotted lines

Question 3

How do you determine trend lines?

Answer 3

• Can draw a best straight line by eye
• Or a curve by eye
• Computer can fit a “best” straight line to the data
• Computer can also fit curves to the data, but these curves are somewhat arbitrary

Question 4

What are anomalies and outliers?

Answer 4

In physics,:

• data is often very tight
• If one observation is way out from all the others you can describe this as an anomaly
• It may be reasonable to ignore this data point

In biology and medicine:

• data is often very variable
• There is no good reason to ignore observations because they are a bit further out

Question 5

What are Figure legends?

Answer 5

• In published papers, the title is placed in the “figure legend” not on the graph itself
• The figure legend will also contain a very brief explanation of the graph, and any information you need to interpret the graph

Question 6

Name other types of graphs

Answer 6

• Column scatter graph
• Column bar graph
• Tree Heights
• Histograms (only used if the y axis shows frequency, not actual numbers)

Question 7

What happens with graphs with Logarithmic scales ?

Answer 7

Sometimes the dependent variable may change in an exponential or a logarithmic way

This is better presented using a log scale

Question 8

What are dose response curves?

Answer 8

Dose response curves to drugs are often more informative with drug concentration on a log scale

Question 9

What can be used instead of calculating logarithms

Answer 9

Semi - log graph paper

Question 10

What are the two types of data?

Answer 10

1. Numerical (quantitative) data

2. Categorical data

Question 11

Explain numerical data

Answer 11

• Continuous- can take any value
• Discrete- can only take specific values, e.g.
  Something you count can take the values 1, 2, 3
• Should be displayed on a quantitative axis
• A trendline is often appropriate

Question 12

Explain Categorical data

Answer 12

E.g. male/female, treatment condition 1/condition 2/ condition 3

Note that the numbers
here don’t mean anything numerical. We could have used a, b, c or Derek, Jean & Flora

• Should never be displayed on a quantitative axis
• Often presented as a bar chart
• A trendline is never appropriate

Question 13

What is a distribution?

Answer 13

• If you collect a lot of data points, you can plot a more detailed histogram
• The distribution of the data will start to look as if it approximates a smooth mathematical curve

Question 14

Explain a symmetrical distribution

Answer 14

A graph where the mean, median and mode are all directly in the middle
bell shaped curve

Question 15

Explain a Skewed - Right Distribution

Answer 15

The graph has a positive skew or right skew as the mean is to the right of the median

For a skewed distribution, the median may give a more reasonable measure of “average” than the mean

Question 16

Explain a Skewed - Left Distribution

Answer 16

This has a negative skew or left skew as the mean is to the left of the median

Question 17

What is a Normal Distribution?

Answer 17

The “normal distribution” is a particular mathematical distribution with two parameters, the mean and the standard deviation.

Question 18

What is the charcteritics for experimental observations to fit to a normal istribution?

Answer 18

• There is a mathematical explanation behind this (the Central Limit Theorem)
• If a variable is affected by a lot of different random factors
• Each has a small effect
• And their effects are additive
• The distribution will approximate to a normal distribution