Abnormal Psychology

Researching Abnormality 173

the people administering the study, and the cultural context writ large. Challenges to

conducting research on social factors that affect psychopathology include the ways

that the presence or behavior of the investigator and the beliefs and assumptions of a

particular culture can infl uence the responses of participants.

Investigator-Infl uenced Biases

Like psychological factors, social factors are often assessed by self-report (such as

patient’s reports of fi nancial problems), reports by others (such as family members’

descriptions of family interactions or of a patient’s behavior), and direct observa-

tion. Along with the biases we’ve already mentioned, the social interaction between

investigator and participant can affect these kinds of data.

Experimenter Expectancy Effects

Experimenter expectancy effectrefers to the investigator’s intentionally or unintention-

ally treating participants in ways that encourage particular types of responses. The ex-

perimenter expectancy effect is slightly misnamed: it applies not only to experiments,

but also to all psychological investigations in which an investigator interacts with par-

ticipants. For instance, suppose that you are investigating whether a particular symptom

of social phobia—a fear of being in social situations with strangers—is associated with

a particular symptom of OCD—recurring intrusive thoughts or images. You are investi-

gating whether these two symptoms—each of a different disorder—are correlated. You

interview participants with social phobia and ask them about the nature of their fears.

When they mention symptoms that are not of interest to you (such as their preference

to be alone), you ask about how they feel in social situations with strangers. It’s possible

that you might ask with a particular tone of voice or facial expression, unintentionally

suggesting the type of answer you hope to hear. Participants might, consciously or un-

consciously, try to respond as they think you would like, perhaps exaggerating certain

symptoms a bit. Such a social interaction can undermine the validity of the study.

Reducing Experimenter Expectancy Effects with a Double-Blind Design

To minimize the likelihood of experimenter expectancy effects, researchers often use

adouble-blind design: Neither the participant nor the investigator’s assistant (who

has contact with the participant) knows the group to which the participant has been

assigned or the predicted results of the study. That is, both are blind to the par-

ticipant’s group assignment and the predicted results. Implementing double-blind de-

signs has the following features:

• First and foremost, the assistant, who does not know the predictions, collects the

information from participants.

• The investigator eliminates cues in the environment that might allow the assistant

to infer the predictions.

• Instructions given to participants or questions asked of them are standardized, so

that all participants are treated in the same way.

• A computer or other automated device delivers instructions and stimuli and scores

the responses.

Although double-blind studies are a way to minimize experimenter expectancy

effects, participants try to make sense of what is being asked of them in a study de-

spite an investigator’s best efforts. Participants often develop their own hypothesis

about the nature of the study and may behave accordingly (Kihlstrom, 2002a).

Reactivity

Have you ever had the experience of being watched while you were doing something?

Did you fi nd that you behaved somewhat differently simply because you knew that

you were being observed? If so, then you have experienced reactivity—a behavior

change that occurs when one becomes aware of being observed. When participants

in a study know that they are observed, they may subtly (or not so subtly) change

their behavior, leading the study to have results that may not be valid. One way to

counter such an effect is to use hidden cameras, but this may raise ethical issues—

many people (rightfully) object to being “spied on.”

Experimenter expectancy effect

The investigator’s intentionally or

unintentionally treating participants in ways

that encourage particular types of responses.

Double-blind design

A research design in which neither the

participant nor the investigator’s assistant

knows the group to which specifi c

participants have been assigned or the

predicted results of the study.

Reactivity

A behavior change that occurs when one

becomes aware of being observed.