1985
DOI: 10.1037/0021-9010.70.1.101
|View full text |Cite
|
Sign up to set email alerts
|

Generalizing across stimuli as well as subjects: A neglected aspect of external validity.

Abstract: In order to generalize the results of an experiment beyond the specific stimuli employed, it is necessary to consider variance due to stimulus sampling. This can be accomplished by treating stimuli as a random effect rather than the traditional procedure of treating stimuli as a fixed effect. The serious consequences of the traditional approach are illustrated using examples from applied psychology. Statistical and design considerations for generalizing the results of experiments are discussed.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
38
0

Year Published

2005
2005
2018
2018

Publication Types

Select...
6
1
1

Relationship

1
7

Authors

Journals

citations
Cited by 53 publications
(38 citation statements)
references
References 12 publications
0
38
0
Order By: Relevance
“…This appears to be the case both for complex objects (e.g., humansoften used to instantiate independent variables in social psychology, see [51]) and simpler objects (e.g., line drawings, see [50]) commonly used in cognitive psychology. Even when the objects are quite simple (i.e., have few features and potential interactions between them), Fontenelle et al [48] conclude: "when it is the intention of an experimenter to generalize results beyond the particular sample of objects employed, the statistical treatment of objects as a fixed effect is generally inappropriate. Thus, unless a researcher is willing to limit the generalizability of his or her findings severely, the effect of stimulus sampling must be considered both in the design of the experiment and in the analysis of the results."…”
Section: From Stimuli To Artifact Samplingmentioning
confidence: 99%
See 1 more Smart Citation
“…This appears to be the case both for complex objects (e.g., humansoften used to instantiate independent variables in social psychology, see [51]) and simpler objects (e.g., line drawings, see [50]) commonly used in cognitive psychology. Even when the objects are quite simple (i.e., have few features and potential interactions between them), Fontenelle et al [48] conclude: "when it is the intention of an experimenter to generalize results beyond the particular sample of objects employed, the statistical treatment of objects as a fixed effect is generally inappropriate. Thus, unless a researcher is willing to limit the generalizability of his or her findings severely, the effect of stimulus sampling must be considered both in the design of the experiment and in the analysis of the results."…”
Section: From Stimuli To Artifact Samplingmentioning
confidence: 99%
“…Wells and Windschitl [51, p. 1115] consider this neglect "a serious problem that plagues a surprising number of experiments," casting doubts on the validity of conclusions drawn from such studies. To increase the validity of experimental studies, more and more researchers call for stimuli sampling -selecting objects at random from the theoretical feature space [48], [51].…”
Section: From Stimuli To Artifact Samplingmentioning
confidence: 99%
“…It is rare in research to find this many different productand item-stimuli tested in one paper; and the range and number of products and stimuli that comprise our studies increases the external validity of our findings. Although conducting field studies is one way to assess external validity, it can also be assessed by conducting studies containing many stimuli (Fontenelle, Phillips, and Lane 1985;Wells and Windschitl 1999). …”
Section: Meta-analysis Of Studies 1-8mentioning
confidence: 99%
“…Interviews with two triads of female Hispanic students further established the Hispanic symbolism of the warm colors and mainstream symbolism of the cool colors. Further, three sets of color stimuli were chosen for stimulus sampling purposes (Fontelle et al, 1985). Each color stimulus set included a warm and a cool color: (1) red (warm) and blue (cool); (2) purple (warm) and green (cool); and (3) redpurple (warm) and blue-purple (cool).…”
Section: Product Attribute Symbolism Manipulation and Color Stimulus mentioning
confidence: 99%