1301.6.55.001 - Tasmanian Statistical News, Mar 2010  
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 09/03/2010   
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product

STATISTICAL LITERACY


UNDERSTANDING STATISTICS

Making sense of statistics is vital for informed decision-making. To build your statistical know-how check out Understanding Statistics on the ABS website at www.abs.gov.au/understandingstatistics. There are some excellent resources contained within these pages, some are very basic, whilst others go in to more depth. There are quizzes, video tutorials and online presentations. A great resource for anyone wanting to know more about statistics and the ABS website.


UNDERSTANDING STATISTICAL CONCEPTS

In today's information-rich society, we encounter statistical information on a daily basis, ranging from unemployment rates, retail figures and cancer rates, to football ladders and cricket scores. Statistics tell interesting stories and enable us to make sense of the world. Statistics are essential for research, planning and decision-making purposes.

There are several concepts that recur throughout the literature on statistical literacy. These fall into four key areas and can be considered in a practical manner as ‘criteria’ on which to base statistical literacy:

  • Data awareness
  • The ability to understand statistical concepts
  • The ability to analyse, interpret and evaluate statistical information
  • The ability to communicate statistical information and understandings

In this issue, we will focus on understanding statistical concepts. There are three basic forms of statistical representation: tables, graphs and maps.


Tables

The ABS presents much of its information in the form of tables where data is presented in rows and columns. By following some simple principles, information can be quickly understood from a table.
  • Headings
    Look at the heading at the top of the table. There will be an overall description of the information presented in the whole table. By examining these headings it will be apparent fairly quickly whether the table will be useful to you.
  • Rows and columns
    The data in a table is presented in rows (horizontal ) and columns (vertical). The intersection of each column and row is called a cell. Each row and column has a heading.
  • Row headings
    Row headings give an explanation of the data contained in each row of a table. The row label is often broken down into sub-headings to give more detailed information. It is important to remember that the top heading (the one furthest to the left hand side of the page) is the heading for all sub-headings indented underneath it.
  • Column headings
    These are also broken down into sub-headings. The hierarchy is shown by horizontal lines drawn over columns to which the heading immediately above the line applies.
  • Footnotes
    Footnotes are used to add comments and/or explanation to the data held in a table. Always read the footnotes carefully to ensure you are interpreting the data accurately.
  • Sampling error
    Sampling error is related to the error that occurs because we are testing a sample of a population, and the sample’s characteristics may differ slightly to the characteristics of the entire population.

Once you are clear on what the table is about, proceed to the actual figures and extract the information you require. Over time, you will be able to interpret tables more quickly and accurately. It is good practice, however, to be methodical and follow the steps above any time you use a table.

Why is there variation in the data?

Figures will often vary slightly from table to table due to randomisation. Introduced random error is a technique that was developed to avoid identification of individuals. Prior to the 2006 Census, the confidentiality technique applied by the ABS was to randomly adjust cells with very small values. For the 2006 Census, a new technique was developed which slightly adjusts all cells to prevent identifiable data being exposed. These adjustments result in small introduced random errors, but do not impair the value of the table as a whole.

Tables which have been randomly adjusted will be internally consistent, however comparisons with other tables containing similar data may show minor discrepancies. This is the case for both customised tables and standard products. These small variations can, for the most part, be ignored.


Graphs

Information that is presented visually is often easier to understand. Graphs are an ideal way to show trends or differences between data. It is important to interpret scales accurately to avoid misinterpretation of the data. The ABS uses various types of graphs to illustrate different types of data:
  • Line graphs
    Line graphs are generally used to show time series data, to capture movement and trends over time.

UNEMPLOYMENT RATE, Tasmania

Graph: Graph Unemployment rate, Tasmania


  • Dot charts
    Dot charts are used to show comparative values clearly. They allow a comparatively large number of categories to be displayed and have best impact when the values are ranked in descending order.

NEW RESIDENTIAL BUILDING APPROVALS, top 10 contributors to the state total,
by local government area, Tasmania, 2007-08
Graph: Graph New residential building approvals, top 10 contributors to the state, by local government area, Tasmania, 2007-08


  • Bar or column graphs
    Bar or column graphs are used when comparing data values is important, but with a limited number of categories.

TYPE OF INTERNET CONNECTION, Tasmanian dwellings(a),
by statistical division, August 2006
Graph: Graph Type of internet connection, Tasmanian dwellings, by statistical division, August 2006


  • Age pyramids
    Age pyramids are used when representing the age structure of a population.

POPULATION CHANGE, Tasmania, 1996-2006

Diagram: Graph Population change, Tasmania 1996-2006
Source: Census of Population and Housing, 1996, 2006
ABS data available on request


Maps

Maps provide a simple visual comparison between geographic areas. Legends must be interpreted accurately to fully understand information presented in this way. The map below is based on the 2006 Census. It shows the unemployment rate by local government area, identifying areas with the highest rate of unemployment in red, grading to areas with the lowest rate in yellow. In 2006, the highest rate of unemployment (10.3%) was in the George Town Local Government Area (LGA), followed by Kentish (9.6%) and Break O'Day (9.2%). The lowest rates of unemployment were in King Island (2.2%), Flinders (3.8%) and Circular Head (4.2%) respectively.

Insets can draw out finer detail, such as rates of unemployment around the Hobart area, where the highest rate of unemployment was found in the Derwent Valley (9.0%) and the lowest was in Kingborough (4.5%).


UNEMPLOYMENT RATE, by local government area, August 2006
Diagram: Map Unemployment rate, by local government area, Tasmania, August 2006
Source: Census of Population and Housing, 2006
ABS data available on request




For further explanation of terms see Statistical Language! (ABS cat. no. 1332.0.55.002)

In upcoming issues of Tasmanian Statistical News we will discuss other statistical literacy concepts in more detail. Meanwhile, if you would like to know more about statistical literacy and its relevance to you, check out the article: What is statistical literacy and why is it important to be statistically literate? as featured in Tasmanian State and Regional Indicators (ABS cat. no. 1307.6) or visit the Understanding Statistics portal on the ABS website.