Section 4: Methods of data collection
From Research Methods in Psychology
Contents |
[edit] WHAT ARE YOU STUDYING?
Finally, we get to the actual measurement process. In this section we will consider a range of different methods of measurement. Each of these methods is a different way of trying to find out about a person's mental processes.
We will now consider some of the methods of collecting data. Each of these methods may be used in conjunction with any of the experimental or non-experimental research methods discussed in Chapters 3 and 4. In psychology, we are always interested in what is happening inside people’s brains. We want to be able to look inside their heads and find out what they know, and what they want, and what they believe. We can’t look inside their heads directly, so instead we have to take a different approach. The first method we could use is asking them – we ask them what they know, and they tell us. Psychologists call this a self-report, carried out with an interview or questionnaire method. The second method we could use is to watch them, and see what it is that they do. Psychologists call this the observational method. The final thing we could do is monitor their bodies for a biological response. Psychologists call these physiological methods.
[edit] Self-Reports: Questionnaires and Interviews
|
Crucial Concept: Questionnaires and interviews are usually about asking people questions to find out what they think about something, or know about an issue. |
Self-report methods of data collection involve asking people questions. We can do this by using questionnaires (which are usually paper based) or by using interviews. There is something of a fuzzy line between these two methods - in a telephone interview, for example, may be just a questionnaire done over the telephone.
In this section, we will have a look at some of the issues that arise when we use surveys or questionnaires – that is, when we ask people questions. Many of the issues that we will cover occur for both types both interview and questionnaire methods, and so we consider them together. At the end of this section, we will compare the two methods, to help you decide which you should be carrying out.
Designing a good questionnaire or interview schedule is surprisingly difficult, and we will only look very briefly at some of the issues. If you are going to use a questionnaire or interview in a research project, you should be sure to thoroughly read some good information. A start would be Oppenheim (2000), or Fink (1995). In the box section, we look at just one of the problems in designing appropriate questions.
|
Many polls in popular newspapers are of this type, but even polls with more serious aims can have the same problems. The Marie Stopes International organisation recently carried out a survey to find out about parental attitudes to children’s sex education. For one part of the survey, parents responded on a six-point scale (from “strongly agree” to “strongly disagree”)
If you wanted to carry out a similar survey to that carried out by Marie Stopes International, and you were of the opinion that parents did not need any outside help with sex education, would you have asked those questions? If not, what would you have asked. Whose questions are more likely to get to the truth? Yours, or the original questions. |
[edit] Observation
|
Crucial Concept: Data collection through observation involves monitoring and systematically recording the behaviour of participants. |
Observation is a very direct approach to collecting data. In a survey, we ask people what they do in different situations. In a way, we are saying ‘Watch yourself and then tell us what you did’ - thus a survey may be considered to be ‘indirect observation.’ Therefore, when we, the researchers, watch people to see what they do, we can call this ‘direct observation.’
Students are often misled into thinking that observation is a tool only of social psychologists who are carrying out experiments in social settings, or of developmental psychologists who place one-way mirrors in nurseries. Whilst observation certainly does include that type of research, you should realise that it covers a range of other areas of psychology.
Observations may also be used as the basis of surveys, or to back up and validate surveys. The supermarket Tesco sells a large number of pork pies, but in their surveys, no one admitted to buying them. Observational research revealed that it was middle-aged men bought them but did not want to admit in front of their wives that they ate pork pies.
Observational methods can be divided along two dimensions, formal vs casual, and participant vs non-participant, and we will consider these in turn.
[edit] Casual vs Formal Observation
The relationship between casual and formal observational techniques is much like the relationship between unstructured and structured interviews described in the previous section. Casual observation, like unstructured interviews, will often take place during the initial stages of a research project. The researcher will go to the situation looking to find the categories and behaviours that will be used later in the formal observation stage. In the first stage, the researcher may also be considering factors such as the best place to stand, or the least obtrusive way to dress.
Formal observation is analogous to the structured interview. When carrying out a formal observation the researcher will have a clear idea of what behaviours they are looking for, and how they will be recorded. It is common, especially in laboratory situations, to record information on video, and then to analyse it in more detail later.
[edit] Participant vs Non-Participant Observation
You can observe a group of people from two perspectives, either from outside the group or from within the group. When a researchers immerses themselves in a group so that they can observe behaviour from with the group they are known as participant observers. The participant observer has the advantage of being able to become completely involved with the group. The only way to really feel and understand what it is like to be a member of a group is to join that group. Sometimes this approach is the only way to observe the behaviour of a group. A famous example of this is Festinger, Riecken and Schacter (1956) who wanted to study a group of people who believed that the world was going to end on a particular day. He wanted to observe how they reacted when the world did not end, and joining the group was the only way that he could observe their behaviour. The main disadvantage of this approach is that it is not possible to formally record all observations and behaviours - as soon as you start to take notes, the other members of the group will realise that something is amiss. A further disadvantage may be that you might have influenced the group by being there. You may try to do nothing, but by doing nothing, you may encourage others to do nothing – you don’t know what would have happened if you weren’t there.
In non-participant observation, the researcher remains separate from the group, and observes them from a distance. This approach allows much more care to be taken in the observation of behaviours and the possibility of using video to record events so that they can be checked again for the accuracy of coding. The main disadvantage of this approach is that the individuals who are being observed may alter their behaviour, because they are being observed.
[edit] The Reliability of Observations
Researchers working with observational data must go to great lengths to ensure that their data are reliable. They must decide beforehand (often in the casual observation stage) precisely what they will be recording, and define these as accurately as possible. They can choose to describe very small units of behaviour (child x moved towards child y; child y smiled at child x) or larger units of behaviour (child x and child y played together in the sandpit for 8 minutes). Two types of reliability can be checked. They have confusing titles: inter-observer reliability and intra-observer reliability.
Inter-observer reliability concerns the agreement between two separate observers, using the same scheme. If two observers, with the same checklists, looking for the same behaviours observe the same group of people, they should agree on what it is that they saw. If they disagree, then the checklist and the criterion should be revised. Intra-observer reliability concerns an agreement of observers with themselves. If you are observing a group over a long period, you might find that your criteria shift slightly. If possible, you should record the behaviours on video, and then recheck earlier versions to ensure that you still agree with your earlier ratings.
[edit] Ethical Issues in Observational Research
We will cover ethical issues in much more detail in Chapter 5, however it is worth mentioning ethical issues in this section, to consider when it is acceptable to observe people without their knowledge
The British Psychological Society Code of Conduct and Ethical Principles (BPS, 2000) states “observational research is only acceptable where those observed would expect to be observed by strangers … particular account should be taken of … the possibility of intruding upon the privacy of individuals who, even while in a normally public space, may believe they are unobserved” (p. 12).
This means that if people are in a public space, surrounded by strangers, it is OK to observe their behaviour without their permission. An example of this would be examining whether people ate chips in the canteen. However, it is not acceptable to observe people just because they are in a public space - you would probably be unhappy about researchers observing your behaviour in a public toilet.
[edit] Physiological Methods
|
Crucial Concept: Physiological measures involve direct biological measurement of individuals. |
The third method of data collection that we will examine is physiological measurement. This involves taking measurements of physiological, or biological, functioning. Whilst most of these methods require special equipment and training, some of them are simple enough to be possible to carry out in undergraduate projects. In addition, as we have said before, you can gain marks in practical reports is by describing what you would have done, given enough time, expertise and equipment.
- Measures of Heart Rate, Blood Pressure and Heart Functioning
- Breathing and Respiration
- Brain Activity (measured using electroencephalograph (EEG), magnetic resonance imaging (MRI) or computerised automated tomography (CAT).
- Eye Movements and Responses
- Electrodermal response (also known as galvanic skin response), a measure of how much a person is sweating, and therefore how anxious they are.
- Hormonal Levels (can be measured directly, through blood, or indirectly, through saliva.)
[edit] Section Summary
In psychology, we want to know what people are thinking about. In this section we have examined three general ways of trying to find out. We can ask someone, using either an interview or a questionnaire technique. We can observe them, and look at their behaviour. We can take some sort of physiological measurement.
