go to the School of Psychology home page Go to the UNE home page
Chapter 7 - Analysing the Data Part IV - Analysis of Variance Chapter 1 - Behavioural Science and research Chapter 2 - Research Design Chapter 3 - Collecting the Data Chapter 4 - Analysing the Data Part I - Descriptive Statistics Chapter 5 - Analysing the Data Part II - Inferential Statistics Chapter 6 - Analysing the Data Part III - Common Statistical Tests Sources of data Samples and populations Sampling methods Quantitative and qualitative approaches Questionnaire design Coding of data Readings and links


Chapter 3: Collecting the Data


Coding of data

The final phase in the data collection process is to convert the observations and measurements you have made about a construct into a case X variable matrix suitable for SPSS to understand. This is referred to as coding. For our purposes, a printout of the database with an explanation of the codes is called a "data coding sheet". There is more detail about this process in the SPSS Screens & Output booklet, but some points will be reiterated here.

In SPSS (but not necessarily so in other analysis packages) each row in the database represents a different example or case of the thing being measured. In the behavioural sciences these objects are usually people, but they might also be rats. The columns in the database represent the variables that have been measured about each object. For each person we might measure their gender, their birth order, their age, their IQ, their level of sensation seeking, their frequency of marijuana use. Each of these things is a variable and until some particular analysis is undertaken they are all dependent variables.

For some variables such as age, or IQ, or a Likert scale response, we can simply enter the number we measured. For gender or ethnicity or religious affiliation we have to invent a coding system. Such a system is therefore completely arbitrary. For gender we might use a 0 for female and a 1 for male, or we could use a 2 for female and a 1 for male. The main thing to remember is that we need to enter numbers. SPSS will allow you to enter "male" and "female" (by changing the Type of variable from Numerical to a String), but you are very limited in the sorts of analyses you can then conduct. For religious affiliation we might decide to use a 1 for Catholic, 2 for Anglican, 3 for United, 4 for other, and 5 for No religious affiliation.




© Copyright 2000 University of New England, Armidale, NSW, 2351. All rights reserved

UNE homepage Maintained by Dr Ian Price
Email: iprice@turing.une.edu.au