A control group (n = 120) overranked position number one in ranking a set of five homogeneous entities. This error did not appear when two experimental groups (ns = 120) performed the same task after reading a short paragraph and long paragraph, respectively, explicating the nature of “position error” in ranking. In fact, greater randomness was obtained with the short paragraph group, suggesting that a brief set of precautionary instructions is sufficient for eliminating such an error.
Get full access to this article
View all access options for this article.
References
1.
BernardinH. J.WalterC. S.Effects of rater training and diary-keeping on psychometric error in ratings. Journal of Applied Psychology, 1977, 62, 64–69.
2.
HooverT. O.WagnerE. E.The influence of list length on ranking error. Paper presented at the 81st meeting of the American Psychological Association, Montreal, Quebec, August, 1973.
3.
PaivioA.YuilleJ. C.MadiganS. A.Concreteness, imagery, and meaningfulness values for 925 nouns. Journal of Experimental Psychology Monograph Supplement, 1968, 76, 1–25.
4.
RandT. M.WagnerE. E.Generalizability of position error in ranking. Psychological Reports, 1975, 37, 811–814.
5.
WagnerE. E.HooverT. O.The effect of serial position on ranking error. Educational and Psychological Measurement, 1974, 34, 289–293. (a).
6.
WagnerE. E.HooverT. O.Effect of terminal ranking error on meaningful occupational choice. Journal of Applied Psychology, 1974, 59, 247–248. (b).
7.
WagnerE. E.HooverT. O.The influence of technical knowledge on position error in ranking. Journal of Applied Psychology, 1974, 59, 406–407. (c).
