PRODUCTION OF TRUSTWORTHY RESULTS
Internal validity
We can be confident about the results of psychological research when the methods are valid. An experiment is said to have internal validity when we are confident that the results have occurred for the reasons we have hypothesized, and we can rule out alternative explanations of them. These alternative explanations (or threats to internal validity) [internal validity the extent to which the effect of an independent (manipulated) variable on a dependent (outcome) variable is interpreted correctly] can involve an experimental confound – an unintended manipulation of an independent variable [confound an unintended or accidental manipulation of an independent variable that threatens the validity of an experiment]. The risk of confounds can be reduced by better experimental design. Suppose we conduct a study to look at the effect of crowding on psychological distress by putting 50 people in a crowded room and 50 people in an open field. Having found that the people in the room get more distressed, we may want to conclude that crowding causes distress. But the participants’ body temperature (generated by having a lot of people in one room) may represent a confound in the study: it may be the heat, not the crowding, that produces the effects on the dependent variable. The experiment could be redesigned to control for the effects of this confound by using air-conditioning to keep the temperature the same in both conditions.
External validity
A study has a high level of external validity when there are no reasons to doubt that the effects obtained would occur again outside the research setting. We might, for example, question a study’s external validity if participants responded in a particular way because they knew that they were taking part in a psychological experiment. They might inadvertently behave in a way that either confirms or undermines what they believe to be the researcher’s hypothesis. In experiments we usually try to deal with this specific potential problem by not telling experimental participants about the hypotheses that we are investigating until after the experiment has finished. [external validity the extent to which a research finding can be generalized to other situations]
Internal validity
We can be confident about the results of psychological research when the methods are valid. An experiment is said to have internal validity when we are confident that the results have occurred for the reasons we have hypothesized, and we can rule out alternative explanations of them. These alternative explanations (or threats to internal validity) [internal validity the extent to which the effect of an independent (manipulated) variable on a dependent (outcome) variable is interpreted correctly] can involve an experimental confound – an unintended manipulation of an independent variable [confound an unintended or accidental manipulation of an independent variable that threatens the validity of an experiment]. The risk of confounds can be reduced by better experimental design. Suppose we conduct a study to look at the effect of crowding on psychological distress by putting 50 people in a crowded room and 50 people in an open field. Having found that the people in the room get more distressed, we may want to conclude that crowding causes distress. But the participants’ body temperature (generated by having a lot of people in one room) may represent a confound in the study: it may be the heat, not the crowding, that produces the effects on the dependent variable. The experiment could be redesigned to control for the effects of this confound by using air-conditioning to keep the temperature the same in both conditions.
External validity
A study has a high level of external validity when there are no reasons to doubt that the effects obtained would occur again outside the research setting. We might, for example, question a study’s external validity if participants responded in a particular way because they knew that they were taking part in a psychological experiment. They might inadvertently behave in a way that either confirms or undermines what they believe to be the researcher’s hypothesis. In experiments we usually try to deal with this specific potential problem by not telling experimental participants about the hypotheses that we are investigating until after the experiment has finished. [external validity the extent to which a research finding can be generalized to other situations]
No comments:
Post a Comment