1
Doing research in social psychology
Imagine you are having a debate with a friend about relationships, and about what types of people get together. Your friend claims that ‘opposite attracts’ and points out that a couple he knows are happily married but they are like chalk and cheese; she is very extroverted and likes to host parties whereas he is very introverted and likes nothing better than a quiet night in with a cup of cocoa and a good book. But no you say, surely ‘birds of a feather flock together’, and you point out a couple you know who are both introverts, and both like cycling; and what’s more you say, you once knew of a friend who was an extrovert who started dating an introvert but it never lasted – after six months they split up because they were too dissimilar.
The chances are that you will both claim you are right … but how could we test which of these two contradictory proverbs is actually correct? Well, this is what social psychologists do – they formulate interesting questions and then test them to find an answer. However, the way in which they do so can vary greatly (depending on the question) but the manner in which they do it is a careful one to ensure that the answer they arrive at is valid. There are many books that deal in depth with how social psychologists go about doing research, but the aim of this chapter is to give you a very simple overview of the area so that you are able to understand some of the different methods and terminology used in the studies described in the coming chapters.
The scientific method
Most social psychologists consider the discipline of social psychology to be a science, and to reflect this, many of their studies employ what is known as the scientific method. This involves a number of steps:
1 Observe: social psychologists observe the world around them, and when they find something interesting they put it to the test. This book is full of areas in which they have been testing over the years. In the case of relationships, the interesting question is whether birds of a feather flock together or whether opposite attracts.
2 Hypothesize: the next step is to make a prediction – called a hypothesis. This is a statement that can be measured in some way and potentially falsified. For our example, the hypothesis could be something along the lines of: ‘There will be a difference in the length of time couples stay married dependent on their levels of similarity’.
3 Test: The hypothesis is then operationalized in some way; that is to say, it is turned into something that can be tested. For example, for our hypothesis the terms need to be defined – what do we mean by ‘levels of similarity’ (e.g. will we be testing personality traits of both individuals, such as Introversion/Extraversion; Agreeableness; Openness, Conscientiousness, etc., or will we define similarity in some other way.). Once this has been done, we set about collecting our data.
4 Analyse: once the data has been collected, we analyse it to see if there is any truth in the theory that we are testing; to state it more formally, we look to see whether we can accept or reject our hypothesis. If we accept it, then we are suggesting that the data we have collected supports our theory, but if we reject it, our theory might need some modification (before we reject it totally).
5 Modify: whilst the data might support our initial idea, our hypothesis might still be too general, or perhaps too specific. We might then modify our theory to tell us more. For example, when we referred to married couples, we might only have tested heterosexual couples, and therefore be unable to say whether it applies to same-sex couples. We would therefore modify our hypothesis, or perhaps generate a number of new hypotheses, and run our study again by collecting more data to incorporate our new parameters.
6 Repeat: if we have modified our hypotheses, we would run a new study. However, even if we were happy with our study, other researchers might not be. They might see something in our research that they consider to be a methodological flaw (perhaps we only collected data in a specific area of the country and they might feel that this is not a representative sample of married couples); or they might feel that our findings are so unusual that they want to repeat our study just to check that we haven’t falsified our data. One of the strengths of the scientific method is replication. If someone else, in a different university or a different country, can take our methodology, repeat our study and find the same results, this will give our theory even more credibility. If they were to find something different, then perhaps further work would be needed on our theory.
A statement that is falsifiable through some form of testing employing the scientific method.
There are a number of different methods through which scientists test their hypotheses, and they are generally broken down into two broad categories: experimental methods and non-experimental methods. Both have their strengths and weaknesses, and many hypotheses can be tested in more than one way.
Experimental methods
The experiment is a traditional method employed by scientists and it is usually conducted within a controlled ‘laboratory’ environment. It consists of a situation where the researcher manipulates something (called an independent variable) and observes its effects on something else (called the dependent variable). For example, if we were interested in whether viewing violence on television increased aggressive behaviour, we might create an experiment to test our idea. We could assign participants to one of two different conditions: in one they watch a television clip containing violent acts, and in one they watch a television clip containing no violent acts. After they have watched their clip, any subsequent acts of aggression are measured (for example, we might ask them to wait in a room that contains a punch bag and secretly monitor if, and how often, they hit the bag whilst waiting).
| Key idea: Independent variable (IV) |
The variable or variables that are manipulated to see whether they produce an effect on one or more dependent variables.
| Key idea: Dependent variable (DV) |
The variable which is expected to change as a result of the manipulation of an independent variable.
When designing our experiment, there are a number of elements that we need to consider:
1 Who are our participants? These are the people who take part in our experiment (or our research in general). They can be drawn from the general population, or they can have specific characteristics (e.g. children). Generally, participants are randomly allocated to different experimental conditions to equalize any individual differences exhibited by specific individuals (...