Skip to main content

Statistics, data and test instruments: Stats basics

Locate test instruments and critiques of measures

Banner image source: Pixabay 1, 2, 3, 4, 5 licensed under a CC0 Public Domain licence.

Contact for support

Faculty of Arts, Business, Law and Education

ablelibrarians-lib@uwa.edu.au

Faculty of Engineering and Mathematical Sciences

emslibrarians-lib@uwa.edu.au

Faculty of Health and Medical Sciences

hmslibrarians-lib@uwa.edu.au

Faculty of Science    

scilibrarians-lib@uwa.edu.au    

School of Indigenous Studies

ablelibrarians-lib@uwa.edu.au

For further contact information visit the UWA Library Contact us page.

Statistical methods

Statistics explained online

Wolfram Math World                            Statistics at Square One book cover

Wolfram MathWorld provides links to probability and statistics web resources.

The online textbook Statistics at Square One is a useful resource, particularly the chapter "Study design and choosing a statistical test".

Stats basics

There are two main types of statistics.

1. Descriptive statistics such as "The median number of friends on Facebook is 342" which describe a population

Graph distribution of number of facebook friends. High peak around 300, trails down to almost 0 around 2000

Image source: http://blog.stephenwolfram.com/2013/04/data-science-of-the-facebook-world/

2. Inferential statistics which seek to answer questions about set of data, such as "Does X Brand of anaesthetic cause less complications than Y Brand?"

In simple terms, this is done by mathematically comparing groups within the data (eg; XBrand or YBrand) and seeing if the Effect (statistical difference between groups of data) is larger than the Error (differences through random chance).

 

Business, Medicine, Psychology, and Math/Statistics students all use data and statistics in their studies.

 

In almost all cases data is only gathered from a sample population. How large and representative that sample is affects if statistics can be used to make statements or even predictions beyond the sample. Good statistics can significantly empower an argument or decision.

Statistics can be used to :

  • Predict an outcome
  • Test a hypothesis
  • Compare information
  • Provide information in a structured way

Types of statistics in health and medical sciences

There are a number of different types of statistics:

  • Incidence

         The number of specified new events
         e.g., persons falling ill with a specified disease 

  • Mortality

         An estimate of the proportion of the population that dies during a specified period

  • Survival

         The proportion of the population that is alive at a specific period in time despite the diagnosis of their disease/condition

  • Prevalence

         The number of cases of a disease existing in a given population at a specific period of time (period prevalence)

(all definitions taken from Stedman's Medical Dictionary)