Bauhaus-Universität Weimar

Titel:
Experimental Psychology: A Manual of Laboratory Practice, Vol. II: Quantitative Experiments, part 2: Instructor's Manual
Person:
Titchener, Edward B.
PURL:
https://digitalesammlungen.uni-weimar.de/viewer/image/lit11940/346/
§25. The Method of Average Error : Historical 169 
above).—Wundt, then, did not find the value r in the R. Still less did 
he find it in the essay in Poggendorffs Jubelband. He found it rather 
in Eh, Ü-. which he does not cite. 
We noted above (p. 155’) that Fechner discovered, in work with av. 
error, a constant error, which he ascribed to the simple fact that the r 
of every series remains constant, affords a one-sided standard of refer¬ 
ence, while rx is variable. “ [Es kann] bei der Methode der mittleren 
Fehler ein constanter Fehler darauf beruhen, dass wir stets die Fehl¬ 
grösse, aber nicht die Normalgrösse der Abänderung bis zur scheinbaren 
Gleichheit mit der anderen unterwerfen” (EL, ii., 140). Let p and q 
stand for the errors of time and space respectively, and s for the constant 
error which we are now discussing. Then the four possibilities for 
average error are : 
cx=+p — q + s y cm=+p + q + s j 
Gi——p—ç+sj C\ v = —P-\-q~\-s' 
Whence we have : 
ZLfll cnr~Gv . 
„_fm — ci_fry—‘hi . 
v 2 ~ 2 ’ 
s_ci + civ_fn + fur 
2 ~ 2 ' 
See El., ii., 151 f. ; cf. 347 f.—Here, then, is the Wundtian s. Wundt 
was working out his method of minimal changes, with a keen eye for 
constant errors, between 1880 and 1882. When he came to revise his 
account of average error in 1887, he remembered Fechner’s f (forgetting, 
or at least neglecting to state, the place of its discussion, whether in EL, 
ii., or in the second Abh. of 1884), and incorporated it in the method, 
with a general reference to Fechner’s av. error discussions in the R. and 
in Pogg. Jubelband. His originality consists, not in the introduction of 
J as a test-value of the method: Fechner had used it in i860: but 
rather in its interpretation, as a value akin to the liminal values of 
minimal changes and right and wrong cases. 
The à of minimal changes, with which Wundt parallels the 
value s, is half the difference of Aru and Ar(. It is, therefore, 
strictly dependent upon the magnitude of the DL (Fechner, Abh., 
1884, to; et. Estel, P. S., ii., 48). On the other hand, the s' of 
the Fechnerian method is dependent, as we have seen above 
(p. 15 5 )j both upon the relative magnitude of Aru and Ar, 
and upon O’s tendencies of manipulation. There can, then, be
        

Nutzerhinweis

Sehr geehrte Benutzer,

aufgrund der aktuellen Entwicklungen in der Webtechnologie, die im Goobi viewer verwendet wird, unterstützt die Software den von Ihnen verwendeten Browser nicht mehr.

Bitte benutzen Sie einen der folgenden Browser, um diese Seite korrekt darstellen zu können.

Vielen Dank für Ihr Verständnis.