Agricultural Statistics

What a great visual to help you decide how to visualize your data.   By studying this “visualization” you can see that tables and graphs are primarily used to summarize data and to find relationships.

Tables vs Graphs

Tables:

• Verbal representation
• Read the information in rows or columns

Graphs:

• Visual representation
• See patterns or relationships

Neither one is better than the other – they each have their own merits and purposes.  It is up to you as the researcher to decide which is more appropriate for your story.

When would you use a Table?

• Look up individual values
• Compare pairs of related values
• Need precision
• Multiple sets of values in different measures
• Show summary and detailed information

When would you use a Graph?

• Show relationships among and between sets of values by giving them shape
• Patterns, trends and exceptions are more easily seen rather than read
• Series of values – seen as a whole

Different types of Tables

1. Data table
   • Show rows and columns of data
   • Very difficult if not impossible to see any trends or relationships by looking at raw data
2. Contingency Table – or a Crosstab(ulation) Table
   • Can show the relationship between two variables
   • Variables MUST be categorical!
3. Summary or Aggregate Tables
   • Show descriptive statistics such as: mean, minimum, maximum, standard deviation, standard error, etc…
   • Can also group these by a categorical variable

Anatomy of a Table

Remember that a table should stand on its own!!!

Highlights of the anatomy of a table.  If you are publishing, please check with the publication you are submitting to.  The guidelines listed below have been pooled from a number of different sources and are meant to be used as a teaching tool and guide only.

TITLE:  Should be clear and concise
Also known as the HEADING – according to the Operations Manual for the Canadian Journal of Plant Science, Canadian Journal of Soil Science, and the Canadian Journal of Animal Science.

• Capitalize the heading in sentence format with no period at the end
• Do not indent the second and any subsequent lines
• No units of measure in the title

COLUMN TITLES: Visible and concise
Also known as COLUMN HEADINGS 

• Capitalize only the first word
• Units of measure in parentheses on the last line of the subheading
• If several headings share the same UOM, place below the headings, centred

LINES:  to separate different parts of the table

BODY: 

• headings within the body of the table need to be italicized
• centre entries under the column headings
• centre data within the columns on decimal point, dashes, etc..

FOOTNOTES: used to clarify information in the table and should always appear at the bottom of the table!

• Footnotes start with the letter a as a superscript
• Each footnote is on a separate line
• Asterisk – * to designate statistical significance

Examples of Tables

Data Table

Crosstabulation Table

• Select Add/Remove Data
• Under Geography – select all provinces
• Select Apply at the bottom of the page
• This will create a Crosstab table of geography vs Quarter/year

Research results table

Designing a table

Items to think about when you are designing a table.  A statistical package may not always provide you with the ideal table 🙂

1. if you are comparing categories, these should be presented vertically in columns rather than rows.
2. Row entries of data should not be random – order them by importance or alphabetically.
3. If you are presenting more than one level of categories, arrange the hierarchy to emphasize the categories you think are most important

Example:

Steer weight                                                   Heifer weight
1981     1991     2001    2011                          1981    1991    2001    2011

Versus

1981                                                      1991
Steer weight     Heifer weight          Steer weight    Heifer weight

Summary

• Often used to present a lot of data
• Audience will glaze over the table and may not remember the message behind it.
• Not recommended to use a table to show patterns, trends, or interactions between values – this may be easier to see and remember by using a more visual object
• Remember who your audience is!!

 

 

 

Data visualization can mean many different things to different people.  To me, it all comes down to the purpose.  WHY?  and to WHOM?

Our first data Viz meeting, held on October 17 was a quick review of why we use Data visualizations, and a review of the general principles of visualization.  We then reviewed some of the graphic design principles.  Some of these come naturally to folks, while others may be a challenge.  I want to highlight the principles here and will provide a link where you can download the powerpoint that was used during the session.

Four main purposes of Data Visualization:

1. Analysis
2. Communication
3.  Monitoring
4. Planning

Remember your audience will play a major role in helping you define how you present your data and/or your results

Five General Principles behind Data Visualization:

1. Show the data
2. Simplify – you want to keep the message simple and remove all the flowery bits of your visualizations
3. Reduce the clutter – do you REALLY need all those grids or ticks on your graph??
4. Revise your visualizations – creating a graph or a table should be viewed as part of your writing.  You don’t write something and leave it.  You write, you revise, you may write again and revise again.  Treat you visualizations in the same manner.
5. Be HONEST!  This may sound funny – but let’s face it, there are times where the visualizations we create may have an element of exaggeration added in.  Those y-axes – where do they start?  at 0 or somewhere else?  Are we exaggerated the differences between those lines?

The attached powerpoint presentation also demonstrates some of the graphic design principles that we should be aware of as we create tables and graphs.  Please review these as you work on your visualizations.