New Page 1

Navigational Menu

Once you have collected a set of quantitative data, you need a plan for how to analyze it. The plan you devise is dependent on answers to three questions:

What is the level of measurement of the variables I am working with?
- Nominal/Crude Ordinal
- Good Ordinal/Interval/Ratio
How many and what kind of samples am I working with? Sample size is also a consideration.
- Single sample
- Two samples
  - Independent
  - Dependent
- More than two samples
  - Independent
  - Dependent
What kind of statements about the data do I want to be able to make?
- Describe the sample without generalizing to the larger population
  - A single variable
    - Distribution
    - Central Tendency
    - Variation
  - Relationship between two variables
  - Relationship between more than two variables
- Generalize from the sample or samples to the population from which it was/they were drawn. This is called the process of statistical inference or hypothesis testing and relies on the laws of probability to determine the risk of making an inaccurate generalization.
  - A single variable
    - Distribution
    - Central Tendency
    - Variation
  - Differences between two samples
    - Independent or Dependent?
  - Differences between three or more samples
    - Independent or Dependent?
  - Relationship between two variables
  - Relationship between more than two variables

The answers to these three questions determine which statistical techniques you should use. Below are links to pages which describe each of the questions in more depth.

There are several on-line flow-charts, tables, interactive webpages to help researchers decide which statistical test is appropriate. Some of these are:

http://bmj.com/statsbk/13.shtml

http://research.med.umkc.edu/tlwbiostats/choosetest.html

http://trochim.human.cornell.edu/selstat/ssstart.htm

OVERVIEW OF STATISTICAL THINKING How to approach data analysis--what are the basics and what can statistics do for you?

OVERVIEW OF STATISTICAL THINKING
How to approach data analysis--what are the basics and what can statistics do for you?