網頁圖片
PDF
ePub 版
[graphic]

a rather striking consistency in the sevominance of the last group is marked.

TABLE XIV

dif

in

of

e Absolute Length of the Total Exposure. s question. It is unsatisfactory in that in both XIII and XV. The material of the same cards used in the earlier w in the One-Group Apparatus. The from sec. to sec. for each group. nated, the tendency to a time-error, if bly have freer play.

-f the new table are for the most part further fact about the time-error that, it appears to be independent of the when the groups are equal in this reanother fact drawn from Table XIV, aratus the time-error is greater on the fferentiated by other factors. Thirdly,

the values for Table XIII show withdrawn, except the differences mum. This may be significant of that error, or, more probably, b tween work by daylight and work We shall be better able to consid

TABL

88 experiments wit 176 experiments wit

154 experiments with 66 experiments with

3. The Distribution-Error. The last of our three "errors" us. We may recall once more the in these studies. It points to a t those cards where all objective fac equalized,a tendency to mass ticular arrangement of the circles; structed with a view to filling the g irregular arrangement would allow

As in the two "errors" precedi of the facts that gave rise to the pr them to us, gathered out of all th has been technically reckoned with ation are needed to make the ne results are found in its two parts. arrangements employed and the exactly the same. The two sets that the material for the second replacing the small-difference ca Such a change as this might affect in favor of the two sets of arrang

[graphic]
[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]
[blocks in formation]

3. The Distribution-Error.

The last of our three "errors" of exp us. We may recall once more the meani in these studies. It points to a tendency those cards where all objective factors we equalized, a tendency to mass one's ju ticular arrangement of the circles; though structed with a view to filling the given ar irregular arrangement would allow.

As in the two "errors" preceding, so h of the facts that gave rise to the present enc them to us, gathered out of all the tables has been technically reckoned with. But f ation are needed to make the new table results are found in its two parts. In eac arrangements employed and the frequency exactly the same. The two sets differ, as that the material for the second set was f replacing the small-difference cards by th Such a change as this might affect the prop in favor of the two sets of arrangements i

[graphic]

efore, no longer fully comparable with nting the directions of tendency in the , as in all the similar cases throughout as a standard in whose favor all the mn should be understood as given, hat, because the very method by which in the groups, the experimenter was unthe significant differences in the arrangeever, when analyzed on this basis, were at consistencies and agreements among ily apparent. We can now understand to say to us.

TABLE XVI

Cases ency

2

6.3

[ocr errors]

f A: (1) The only lapses from consistservers; and in both these cases there iform trend. (2) With three exceptions eir difference-values. Of these threethe last furnishes but one significant

of the reckoning on this point. (3) Of 10%, some far beyond, showing the at of relative number of this factor of cases, 45 agree in tendency. (5) That

with this surprising agreement w another item to the growing array ance of some subjective factor for nature of this factor we are yet i

of this same table adds the furth inconsistent in the old are not consistent still maintain their reco

a. Analysis of the Experimenta once we are interested to enquir results. To put ourselves upon the what factors are involved in any s have used in the material for th cisely, we may ask in what way significantly. Finally, by an exp we may solve our problem.

The groups of objects in our n marked out in each corner by a c were set irregularly, with the rest objects distinguished from a homo or less irregular outline whose irre ternal arrangements. Within its ou pattern of bright and dark. Whil

by the corner circles was always th tive amounts of brightness and da wise the same, yet practically differ enter through the changing chara ideal completeness could scarcely ground by the uninitiated. The a times surprising to one whose chie as vignetted in process of construc toward the edges the central spaces toward the interior gaps appear in

It is not a very easy task to fill arrangement in such a way that stand out by contrast against the To succeed in this is to fill the area are good that some vacant patch neighbors or some section of circle than the surrounding circles; or pe unexpectedly intrusive. Now in a cannot become more thickly massed

[graphic]

therefore, no longer fully comparable wit presenting the directions of tendency in the here, as in all the similar cases through factor as a standard in whose favor all th column should be understood as givenon that, because the very method by whi ted in the groups, the experimenter was unto the significant differences in the arrange however, when analyzed on this basis, wer that consistencies and agreements amic eadily apparent. We can now understand has to say to us.

TABLE XVI

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small]

with this surprising agreement we have another item to the growing array of evid ance of some subjective factor for the n nature of this factor we are yet in the of this same table adds the further info inconsistent in the old are not consist consistent still maintain their record.

a. Analysis of the Experimental Cond once we are interested to enquire for t results. To put ourselves upon the right what factors are involved in any such ar have used in the material for these stu cisely, we may ask in what way such significantly. Finally, by an experimen we may solve our problem.

The groups of objects in our material marked out in each corner by a circle. were set irregularly, with the result tha objects distinguished from a homogeneou or less irregular outline whose irregulari ternal arrangements. Within its outlines pattern of bright and dark. While the by the corner circles was always the same tive amounts of brightness and darkness wise the same, yet practically differences, enter through the changing character of ideal completeness could scarcely be bro ground by the uninitiated. The amount times surprising to one whose chief thou as vignetted in process of construction. toward the edges the central spaces open toward the interior gaps appear in the ma

It is not a very easy task to fill an are arrangement in such a way that no sect stand out by contrast against the remain To succeed in this is to fill the area homog are good that some vacant patch will ge neighbors or some section of circles will than the surrounding circles; or perhaps a unexpectedly intrusive. Now in a given a cannot become more thickly massed witho

[graphic]

very likely that the strength of tion-error was due to its appearan knew that other factors were bei The general method already d groups that gave objective prom filling. To fulfil the conditions difficult than in the first set; and The small-difference cards were d ratus used.

A further attempt was made to It had been found that, among t judgment had been studied, hear any. So the small stopped pipe us brought into service and the error way. Incidentally our new tables ation about the effect of this facto tions that are theoretically highly upon the attention of the observ has shown them only too ready to sei to the factor of hearing should ap somewhat more than its face value.

Exp

Homo

geneous Vacant tendency geneous No Homo50 253-4

251.2

152.2

44 experiments.

288 experiments.

The per cents recorded indicate the average per Now we are ready to inspect the re come of the attempt to emphasize sec. exposure were repeated wit

« 上一頁繼續 »