PSYC 201
Psychology and science
Goals of the scientific method
Organization and description:
Collect information about a given topic or question.
Among the pieces of information that be useful…
- What exactly is the behavior or topic that one is interested in?
- When does it occur?
- How often does it occur?
- Does it happen to everybody or just to certain types of people?
Ex. Workplace violence. What pieces of information should we record about this behavior?
- How many cases are there a year?
- What do these cases have in common?
- Is there a common profile of the killers? Age, Gender, type of job, etc
- Is there a common profile of the victims? Supervisors, co-workers, males or females
Question: Who are we supposed to be describing; groups of people or individuals?
Prediction:
Is it possible to know the circumstances in which the behavior will occur again. Are there things that one can measure which predict it's occurrence in the future. This information might be valuable.
Ex. Workplace violence
- Do the working conditions of postal employees predict the behavior?
- Is there some combination of employer style and employee personality that predicts the behavior?
- Is it possible to predict which employees are most likely to act in this way? If it is, then it may be possible to intervene.
Explanation (understanding): It’s useful to know that a particular behavior is going to occur in a particular situation, but it’s often more useful to be able to explain why it occurred. Is it possible to determine the cause of the behavior? It may be possible to prevent some kind of negative behavior from occurring again.
If you understand why a behavior occurs it may be possible to do something to prevent it from happening.
ex. Is there a theory in psychology that can account for the behavior? Learned helplessness, models of stress and coping.
In general, the only kind of study that can get at whether one variable causes a change in another variable is an experiment. In an experiment the researcher changes the conditions in one particular way, while they hold everything else constant. The variable being changed or manipulated is the independent variable. The variables being held constant are called control variables. The researcher obtains scores for the variable they’re trying to explain. This variable is called the dependent variable.
When the conditions are changed in just one way and then at a later point the researcher observes a change in the variable they’re interested in, the only possible explanation for the change in the dependent variable is the change in the independent variable. The researcher can say that the independent variable caused a change in the dependent variable.
Intervention (creating change)
- Can a researcher use the information about the behavior they’re studying to change the behavior in a particular way.
- Ex. Are there particular changes in the work environment that could be put in place to reduce the incidence of workplace violence?
The scientific method
- Not a specific formula or recipe for producing information. It’s a set of tools, it’s a strategy for producing and evaluating information.
- It’s more of a frame of mind than a recipe.
- One way to get a sense of what the scientific method is, is to contrast it with non-scientific approaches.
- General approach:
Nonscientific: intuitive. People’s everyday decision making is prone to a number of different biases. Availability heuristic. People don’t know that their decisions are based on incorrect information.
Ex. Not in book. Are there more words that begin with the letter R or words that have the letter R as the third letter? More with R as the third letter! The Availability Heuristic (Tversky and Kahneman, 1973). It’s easier to think of words that begin with R so we assume that there are more of them.
Non-scientific – Authoritative: Believing something because another person says it’s the truth.
Ex. Spontaneous generation. Debate in France in the 1860s over the issue of spontaneous generation. French scientist Felix Pouchet believed that under the right conditions life could appear from dead matter – spontaneously. (a piece of cheese left in a refrigerator) Louis Pasteur did not believe this was possible, but that a living thing could only come from other living things. Both Pouchet and Pasteur were good scientists and both had data to support their positions. If anything, Pouchet’s data were a little bit better. The question was decided by a committee set up by the French Academy of Science. The people on the committee voted on who was right. And that became the official position of French science. If you wanted to be taken seriously as a scientist at that time, the committee decided what you should believe. If you didn’t believe that, you weren’t a considered to be a scientist, at least officially. Is this how scientific questions should be decided?
Scientific approach: Empirical. Decisions are based on direct, recorded observations. Intuition may be a very valuable source of new ideas, but these ideas are then tested.
- Observation:
Nonscientific: Casual, uncontrolled
Scientific: Systematic, controlled
Is it possible to be sure why a particular event is occurring? Does caffeine have an effect on memory. Could you tell from just making observations of people?
What if you do an experiment. Big dose vs no dose. Big dose – remember fewer words from a list. But you find that the Big dose group has an average age of 74. The other group has an average age of 27. The experiment is systematic, but it not well controlled.
Systematic: set up a standard condition to observe the behavior.
Control: Eliminate other possible explanations for a differences in behavior.
The variable you’re observing is called the dependent variable.
The variable you’re manipulating is called the independent variable.
Experimental condition: the presence of the explanation being examined
Control condition: the absence of the explanation being examined.
Sometimes the independent variable is not being manipulated per se by the investigator. You could say that nature has done the manipulating. Here the independent variable might be an individual difference variable. For example you could look at whether age has an effect on memory. If you compared younger adults to older adults, age would be an example of an individual difference variable.
In this example, age is a variable being controlled for.
[Spot the I.V., the D.V., the control variable. Is the I.V. manipulated or an individual difference variable?]
- Concepts
Nonscientific: Ambiguous definitions, with surplus meaning
Scientific: Clear definitions, operational specificity
Ex. The concept of intelligence. What is it? It’s an extremely difficult question. It would be very easy for two people to use the same word, but mean different things.
One way to resolve ambiguity is to provide an operational definition of the concept. An operational definition basically answers the question, “How do we measure it”? So, given the approach, intelligence is what intelligence tests measure. It’s a way for people to agree on what they’re talking about.
Problem: just because you decide to measure intelligence in a certain way, this doesn’t mean that you’re necessarily measuring intelligence. Francis Galton measured the size of people’s skulls, and said that he was measuring intelligence.
- Instruments:
Nonscientific: Inaccurate, imprecise
Scientific: Accurate, precise
Reaction time: estimate the reaction time versus have the computer measure the amount of time.
- Hypotheses:
Nonscientific: Untestable
Scientific: Testable
One of the marks of a good theory is that it’s possible to test it. There are theories where you can interpret any possible observation in terms of that theory. One criticism of Freud’s theory is that it can explain practically any behavior after the fact. If they do one thing, they’re fixated at the anal stage of development. If they do the other thing, they have an unresolved oedipal complex. It isn’t possible to think of a behavior in a situation that doesn’t fit with some aspect of his theory.
- Attitude:
Nonscientific: Uncritical, acceptable
Scientific: Critical, skeptical
Scientists don’t believe things just because someone else told them to. They aren’t very likely to believe something just because someone in authority told them to. Scientists take the testability of scientific theories seriously so that in order for a scientific theory to widely accepted it has to go through a vicious meat grinder of other scientists trying their best to disprove it. Scientists base their opinions on sound compelling arguments and data that have been collected in a systematic and controlled way.
Ex. Extrasensory perception. What would kind of data would someone have to present to make a convincing case for ESP? Do a study where all other explanations except ESP have been ruled out. Is this really possible?
Scientific Theories
Explanations in science are often presented in the form of theories.
A theory is a logically organized set of statements that serves to define events (concepts), describe relationships among these events, and explain the occurrence of these events.
A theory is an attempt to address the three basic goals of scientific research. Does it help us to organize the information we have about a type of behavior? Does it help to predict the behavior that we’re interested in? Does the theory help us to explain why the person behaves in a certain way?
The structure of scientific theories
The concepts and events that psychologists are interested in are usually described and captured in terms of variables. A variable is a way in which people, animals, or things can differ from each other.
Dependent variable: A variable that the investigator is trying to explain.
Ex. Reaction time is an example of a dependent variable a researcher might be interested in.
Independent variable: A variable being used to explain another type of event.
Ex. Age might help me to predict who has faster and who has slower reaction times.
Intervening variable: A variable that fills a logical gap between the occurrence of an independent variable and the eventual occurrence of the event measured by the dependent variable.
Ex. One theory in my field is that as we grow older our brains experience a generalized slowing in the speed of thinking. It’s like having a computer where the speed of the processor gets slower and slower over time. It can do the same stuff, it just can’t do it as fast.
The dependent variable in the theory is reaction time. That’s the variable I’m trying to explain. The independent variable is age group. Age is a proposed explanation for why the reaction times of some people are faster than others.
But how do you get from increasing age to slower reaction times? There must be something else going on here. In one version of the theory, there is an intervening variable, loss of neurons, that links the independent and dependent variables together.
Age Loss of Neurons Slower reaction time
A good theory makes it easy to see how the independent variables result in changes in the dependent variable.
What makes for a good scientific theory?
- Parsimony: The simpler the theory, the better. If two theories can both explain the same behavior, but one is a lot simpler than the other, the simpler theory is usually considered to be better. This criterion is also known as Occam’s Razor.
- Precision: How accurate or specifically stated are the predictions made by the theory? The more specific the predictions, the better the theory.
Ex. One theory says that in a certain condition, younger adults will have faster reaction times than older adults. Another one says that in this same condition, younger adults will have reaction times than are two tenths of a second faster than those of older adults. The second theory makes a more precise prediction. If that theory is correct it would lead to much more accurate predictions about when you should see important performance differences between younger and older adults.
- Testability (Falsifiability): Karl Popper. If a theory can’t be disproven, it’s not a very good scientific theory. It’s more like something you’re taking on faith.
Where do scientific theories come from?
Running debate about the relative contribution of two sources of new theories.
Induction: Make a large number of systematic observations. Propose an explanation for the occurrence of these events. General rules to explain the occurrence of the specific events. Start with empirical observation.
Deduction: A proposed explanation for one or more events. May be logically derived from pieces of information one already has access to. The hypothesis is proposed and then specific observations are made to see if they are consistent with the theory of how things work. Start with rational explanation
It’s a chicken or the egg kind of problem. Which comes first, the observation or the theory? How do you decide what to observe?
Maybe with repeated exposure to the problem we get more and more precise and comprehensive theories.
Science is a remarkable combination of both induction and deduction – of direct observation and the ability to apply our reasoning skills to generate compelling explanations for what we see.
Nomothetic vs. Idiographic Approaches
Nomothetic approach: Our job is to describe what people are like, in general. The goal is to be able to generalize across all of the people we’re studying.
Ex. 70% of depressed clients have sleep problems. 75% of depressed clients have lost weight lately or gained weight lately. 60% of depressed clients have significant problems with their cognitive function. These percentages might describe the group of depressed individuals, as a whole. And this information is very useful of knowing what kinds of behaviors are associated with clinical depression.
However, It might be that no one person looks like this “average” person. Maybe no body has all of these symptoms.
Idiographic approach: You can learn something important by learning at the data for individual people, rather than just average across large numbers of people. For example, you might learn that there are different “profiles” for a disease like depression. Some people have some combinations of symptoms are other people have other combinations. This kind of information is extremely useful for deciding which treatment strategy to use with a particular person.