ロールシャッハ・ランキング・テストに関する研究

村上 英治; 江見 佳俊

doi:10.4992/jjpsy.27.185

村上英治, 江見佳俊

著者情報

ジャーナルフリー

1956 年 27 巻 3 号 p. 185-192

DOI https://doi.org/10.4992/jjpsy.27.185

詳細

発行日: 1956 年受付日: - J-STAGE公開日: 2010/07/16 受理日: - 早期公開日: - 改訂日: -
訂正情報
訂正日: 2010/07/16 訂正理由: - 訂正箇所: 論文抄録訂正内容: 訂正前 : (1) ロールシャッハ・テストのグループ法としては, Harrower, M. R. とSteiner, M. E. のマルティプル・チョイス・メソッドが代表的であるが, 批判的研究も多いので, Eysenck, H. J. にしたがい, ランキング・メソッドを試みた。被験者群は正常者群, 精神分裂病患者群及び非行少年群であるが, 各被験者群の平均スコアーの間にはそれぞれ1%の危険率で有意差を認めることができ, このテストが一応大まかな臨床鑑別のめやすとして妥当なものであることが理解された。
(2) テストの信頼性を検討するため, 3日間隔, 1年間隔で再検査を試み, 信頼性係数を求めたところ, いずれの場合も危険率1%以下の有意水準で相関があるという結果を得, 被験者より得られた反応はでたらめ応答でなく, 相当の信頼性をもつものであることが認められた。
(3) 上述 (1) の結果は, 10, 20歳代という大まかな年齢的要因によつてのみ統制された被験者より得られた結果であるので, 年齢, 知的両要因のテスト・スコアーへの影響を検討するため中学校, 高等学校の生徒335名についての結果を考察したところ, 高等学校以上の年齢段階では年齢, 知的両要因は余りテスト・スコアーに影響しないと考えられる結果を得た。
(4) 正常者群, 精神分裂病患者群, 非行少年群のテスト・スコアーの分布を比較すると, かなりの重複部分のあることが認められるが, かくの如き重複部分を減少させ, テストの妥当性を高めるためには, 各反応の弁別能力を検討する必要があると考え, 検討したところ, 弁別能力をもつ反応が非常に少ないことが認められた。
さらに被験者によつて理解されない反応は, それらを似ている順に順位づけた場合, 後の方の番号に順位づけられ, テスト・スコアーに影響すると予想し, 検討した結果, やはりこのような反応は後の方の番号へ順位づけられやすい傾向のあることが認められ, テストの反応構成にかなりの問題のあることが理解された。したがつてこれらの不適当な反応を除外して, より弁別度の高い妥当な反応とさしかえることが, 今後これらのテストを臨床的にいよいよ有効ならしめるためには必要なように思われる。

訂正日: 2010/07/16 訂正理由: - 訂正箇所: 論文抄録訂正内容: Right : In connection with some criticisms on the multiple choice method, a typical group method of the Rorschach test, an attempt was made to increase its reliability and validity and to make its application to the clinical field more effective by adopting the ranking method as presented. by Eysenck H. J.
While Eysenck merely altered a procedure from the multiple choice method to the ranking method in order to increase the reliability, the present study was intended to comprehend more directly the psychological mechanism which takes place in selection of responses prepared for each stimulus card ; namely an effort was made, without losing the original purpose of the Rorschach test, to grasp the psychological mechanism of perception and response to a stimulus card, for the personality diagnosis. The ranking method is con-sidered as more dynamic for the item analysis than the multiple choice method which is merely to analyze the response items selected by a subject.
The results of the present study are summarized as follows :
1) The subjects consisted of a normal group, a schizophrenic group, and a delinquent group. Since one percent of significant difference was found between the mean scores of each group, the ranking method might be considered as a valid tool for the clinical diagnosis.
2) The reliability of the test was measured by the retest method, and a relatively high reliability was obtained and thus it was shown that the rating was not arbitrary.
3) No influence of the intellectual and the chronological factor on the test score was found.
4) Since some response items are not appropriate for the clinical diagnosis, a further study should be made to improve about this point.

抄録

In connection with some criticisms on the multiple choice method, a typical group method of the Rorschach test, an attempt was made to increase its reliability and validity and to make its application to the clinical field more effective by adopting the ranking method as presented. by Eysenck H. J.
While Eysenck merely altered a procedure from the multiple choice method to the ranking method in order to increase the reliability, the present study was intended to comprehend more directly the psychological mechanism which takes place in selection of responses prepared for each stimulus card ; namely an effort was made, without losing the original purpose of the Rorschach test, to grasp the psychological mechanism of perception and response to a stimulus card, for the personality diagnosis. The ranking method is con-sidered as more dynamic for the item analysis than the multiple choice method which is merely to analyze the response items selected by a subject.
The results of the present study are summarized as follows :
1) The subjects consisted of a normal group, a schizophrenic group, and a delinquent group. Since one percent of significant difference was found between the mean scores of each group, the ranking method might be considered as a valid tool for the clinical diagnosis.
2) The reliability of the test was measured by the retest method, and a relatively high reliability was obtained and thus it was shown that the rating was not arbitrary.
3) No influence of the intellectual and the chronological factor on the test score was found.
4) Since some response items are not appropriate for the clinical diagnosis, a further study should be made to improve about this point.

責任著者(Corresponding author)

訂正情報

J-STAGEへの登録はこちら（無料）