1301.6.55.001 - Tasmanian Statistical News, Mar 2011  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 08/03/2011  Final
   Page tools: Print Print Page Print all pages in this productPrint All



    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:

    1. Data awareness
    2. The ability to understand statistical concepts
    3. The ability to analyse, interpret and evaluate statistical information
    4. The ability to communicate statistical information and understandings

    In this issue, we will focus on understanding statistical concepts, and examine the difference between censuses and surveys. All censuses and surveys are based on a population. A population is an entire group with at least one characteristic in common. For example, every person in Australia on Census Night or, another example is every agricultural property in Australia. We can measure certain characteristics within this population by using a census or a survey.


    A census collects information from every unit in a population. As a result, data is truly representative of the whole population and detailed accurate data can be made available right down to small areas.

    However, there are huge resource costs arising from the expansive coverage so the number of questions asked is usually kept to a minimum. Processing the data takes time and usually only main results are released. More detailed results are not as readily available and need additional time and money to provide.

    • Accuracy – whole population counted
    • Detail – reliable statistics for small sub-groups

    • Cost – expensive for large populations
    • Speed – long collection, analysis and publication time-frame


    In a survey, only part of the total population is selected. The survey data is then used to make inferences about the whole population, providing a reasonably reliable picture of that population. Most ABS findings are based on surveys, covering such topics as household expenditure, employment, crime, etc.

    ABS surveys are generally conducted on a nationwide basis and produce data at a national, state and sometimes regional level, but detailed data for small areas are not available.

    Costs are generally much lower than a census so more questions, or more detailed questions can be asked, and results can be available far more quickly.

    • Cost – fewer questionnaires to send and chase up
    • Speed – less data to process

    • Accuracy – whole population not represented
    • Detail – estimates for small groups or areas usually unreliable