השוואת שיטות

	עיצוב ניסויי אדפטיבי של מבחן קדם-מבחן-לאחר	מערך ניסוי חסום של מדידה לפני ואחרי
תחום	תכנון ניסויים	תכנון ניסויים
משפחה	Process / pipeline	Process / pipeline
שנת המקור	2000s (integration of adaptive principles with classic pre-post structure)	1935 (blocking, Fisher); 1963 (pretest-posttest + blocking synthesis, Campbell & Stanley)
הוגה השיטה	Synthesizes Campbell & Stanley (1963) pretest-posttest framework with adaptive design methodology formalized by Chow & Chang (2000s)	Donald T. Campbell & Julian C. Stanley (systematized); blocking technique from Ronald A. Fisher
סוג	Experimental design	Experimental design
מקור מכונן	Campbell, D. T., & Stanley, J. C. (1963). Experimental and Quasi-Experimental Designs for Research. Rand McNally.	Campbell, D. T., & Stanley, J. C. (1963). Experimental and Quasi-Experimental Designs for Research. Rand McNally.
כינויים	adaptive pre-post design, adaptive pretest-posttest trial, adaptive two-period design, pre-post adaptive experiment	blocked pre-post design, RBPP design, block-randomized pretest-posttest design, randomized block pre-post control group design
קשורות	6	6
תקציר	An adaptive pretest-posttest experimental design measures all participants before and after an intervention while allowing pre-specified modifications to the trial — such as sample size re-estimation, treatment arm dropping, or randomization ratio adjustment — based on accumulated interim data. It combines the interpretive power of change-score analysis with the efficiency gains and ethical safeguards of adaptive methodology, making it particularly valuable in clinical, educational, and behavioral research where early data can inform better resource allocation.	The blocked pretest-posttest experimental design combines blocking — grouping participants into homogeneous strata before randomization — with pre- and post-intervention measurement. Blocking controls for known sources of variability (e.g., baseline ability, gender, site), while the pretest-posttest structure quantifies change scores directly. Together, they reduce error variance and increase statistical power compared to a simple pretest-posttest design, making this approach well suited to educational, clinical, and behavioral intervention studies.