Although environmental researchers use many different sizes of respondent samples, few environmental studies provide statistical rationales for selecting those sizes. This article utilizes bootstrap analysis to calculate split-block correlations of preferences of randomly selected respondents to a variety of environmental stimuli. Two analytic methods were used, raw score ratings and comparative choice. Analysis showed split-block correlations of .90 and higher could be achieved with 25 to 30 respondents with rating protocols and 10 to 15 respondents with comparative-choice protocols.
Get full access to this article
View all access options for this article.
References
1.
BockR. D.JonesL. V. (1968) The measurement and prediction of judgment and choice. San Francisco, CA: Holden Day.
2.
CochranW. G.CoxG. M. (1957) Experimental designs. New York: Wiley.
3.
EfronB.TibshiraniR. (1991) Statistical analysis in the computer age. Science, 253, 390–395.
4.
KaplanR.KaplanS. (1989) The experience of nature: A psychological perspective. Cambridge, UK: Cambridge Univer. Press.
5.
SchroederH. W. (1984) Environmental perception rating scales: A case for simple methods of analysis. Environment and Behavior, 16, 573–598.
6.
StampsA. E.III, (1991) Public preferences for high rise buildings: Stylistic and demographic effects. Perceptual and Motor Skills, 72, 839–844.
