Researching Abnormality 155
Replication
The process of repeating a study using the
same data collection methods under identical
or nearly identical conditions to obtain data
that should have the same characteristics as
those from the original study.Hypothesis
A preliminary idea that is proposed to answer
a question about a set of observations.Theory
A principle or set of principles that explains a
set of data.Predictions
Hypotheses that should be confi rmed if a
theory is correct.Develop a Hypothesis
After identifying a question, a researcher then forms ahypothesis, which may or
may not be the same as the original impression or hunch. A hypothesis is a pre-
liminary idea that is proposed to answer the question about a set of observations.
Hypotheses are important in part because they direct the researcher to make spe-
cifi c additional observations (which may include making precise measurements).
The most common kinds of hypotheses propose a way to understand differences in
measurements of a variable in different circumstances or to establish a relationship
between variables. To develop a hypothesis, you need to propose some specifi c re-
lationship for the variables in the question you’ve asked. For instance, a hypothesis
might address which people are likely to get depressed or why some people are likely
to get depressed. Let’s say you develop this hypothesis: Individuals who suffered a
major loss during childhood (such as the death of a parent) are more likely to get
depressed after a relationship ends during adulthood. Your hypothesis involves two
variables: one categorical (people either experienced a loss in childhood, or they
didn’t) and one continuous (depression can have different degrees of severity).
Collect Data to Test the Hypothesis
The next step involves systematically collecting data to test the hypothesis. For ex-
ample, to test your hypothesis, you might recruit people who have just broken up
with their spouse or partner. You might ask them to complete two questionnaires:
one about any major losses they experienced in childhood, and the other about
their current mood and functioning. Sometimes researchers compare two or more
hypotheses, trying to determine which one best fi ts the data.
Develop a Theory
After enough data are collected and examined, the researcher proposes a theory,
which is a principle or set of principles that explains a set of data. A theory pro-
vides an answer to the question identifi ed by the researcher. For example, you might
theorize that depression is particularly likely to arise in adults after a relationship
breaks up if they, as children, suffered a loss and there was nothing they could do
to control or manage the situation. In fact, as we saw in Chapter 2, when animals
cannot control an aversive stimulus (a shock), they learn to become helpless; rather
than try to escape from such a stimulus, they appear to give up. Such a process
could underlie some forms of depression (Miller & Seligman, 1973, 1975). That is,
according to this theory, the child forms an enduring association between loss and
a sense of helplessness, and it is the feelings of helplessness that produce depression.
Later in life, when the adult experiences a loss, this triggers the associations to help-
lessness, which in turn leads the adult to become depressed. This theory may apply
to Carlos’s case: His mother died of breast cancer when he was 6 years old, and he
remembers feeling helpless during her illness and at her death.
Test the Theory
The next step of the scientifi c method is to test the theory by collecting and examining
additional data. Your theory focuses on people who not only had a signifi cant loss dur-
ing childhood, but also felt helpless to control or manage the situation. The theory leads
topredictions, hypotheses that should be confi rmed if a theory is correct. So, for in-
stance, the theory predicts that children who felt particularly helpless after a loss would
be more likely as adults to become depressed after a relationship breaks up. Data can
then be collected to address this specifi c prediction: In addition to questionnaires about
early loss and mood and functioning, an additional questionnaire might ask about
memories of feeling helpless. You could then test people who did or did not have an
uncontrollable loss during childhood by giving them increasingly diffi cult puzzles; as
a measure of helplessness, you would assess how easily participants give up trying to
solve the puzzles. Figure 5.1 provides a summary of the steps of the scientifi c method.
Many methods can be used to test the predictions made by a theory, and we’llsee what these are in the following section.