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


Types of questions

The construction of a new survey needs to consider how certain information is to be gathered. Many psychological scales use Likert type rating scales. But then you have to decide how many anchor points to provide (as low as two - e.g., "Yes/No" or "True of me/Not True of me" to seven, eight or nine). You have to decide whether to measure frequency ("How often in the past 3 months have you . . . ") or intensity (e.g., "How strongly do you agree or disagree with the following statements?"). You have to decide whether to use closed and open-ended questions. There are also other types such as the semantic differential.

One broad division of the types of questions you should be familiar with is the distinction between "demographic" questions and scale questions. Items that ask for factual information such as your gender, how old you are, how many dependents you have, or your income, are called demographic items. The responses to these items can be evaluated against some objective truth and they are rarely combined. Demographic questions include such things as gender, age, ethnicity, religion, smoker or non-smoker, or occupation.

Scale questions are the ones that have the most psychological interest. These items ask for your self-evaluation on some topic, such as what you would do in a certain situation, or what you have done in the past, or how strongly you agree or disagree with a series of statements. They cannot be so readily gauged against objective truth or there is no right or wrong answer and they are usually combined into a total score on a construct.

When several items are used to measure the same psychological substance or construct, we have a scale or index. For example, if we want to measure extraversion, there would be several items asking things about enjoying parties, entertaining people, telling jokes etc which can be added up to form a total score for extraversion. So the 20 items that make up the Arnett Sensation seeking scale all form a scale because you add all the responses together to get a total score on this construct for each person. Often these scales have subscales

Another consideration when constructing scales is whether to include some reverse-worded items or not. The AISS includes such items and they are included to combat some people’s tendency to respond to the items in the same way. Some people tend to agree with statements that are put to them and so that if you have all the items worded positively a bias might develop. A few reverse worded items ensure that each person stops to think about what the item is actually asking for.

Finally the survey needs to be set out clearly and unambiguously. Respondents need to know what is being asked of them. You need to take into account the amount of time you are asking people to spend on your survey.




© 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