ABSTRACT
Since its publication in 1921, a large segment of Rorschach history, mainly from the late 1940s through the mid-1970s, was often marked by considerable criticism based largely on arguments that convincing evidence for the reliability and/or validity of the test had not been established. Some, such as Cronbach (1949) pointed out that many Rorschach scores do not have the psychometric characteristics that are common for most psychological tests because they are not normally distributed. Others, such as Holtzman, Thorpe, Swartz, and Herron (1961) noted that this problem is made even more complex by the fact that the test is open-ended; that is, all protocols do not contain the same number of answers. Holtzman argued that the variability in the number of responses makes systematic research concerning the test difficult at best, and the application of basic measurement principles and/or the serious application of statistical analyses virtually impossible.
