TY - JOUR
AU - Friedlander, Frank
AB - Errors in research do occur. Their prevalence should be viewed with alarm rather than passive acceptance as an essential concomitant of humans conducting research. The author looks at the discussion of errors in an article by Leroy Wolins (Amer. Psychologist, 1962, 17, 657-658) and an article by Emanuel Berger (Amer. Psychologist, 1962, 17, 657). The first article suggests that we may be quite unaware of the large proportion of the iceberg (erroneous research results) which is never perceived or reanalyzed. The second article pleads for experimental evaluation of research results. It would seem as though these two articles are both positively and negatively related. On the one hand, subjective evaluation and interpretation of research results may possibly expose inadvertent errors in computations (or in design, sampling, etc.). Thus, results which do not readily blend with an already established nomological network might call for statistical recomputations. On the other hand, subjective evaluation can become quite blinding when one identifies with or strongly favors certain results. In such cases, errors in computation would never be suspected, much less discovered. The author discusses experimental errors further, taking a look at Type I and Type II bias.
TI - Type I and Type II Bias
JF - American Psychologist
DO - 10.1037/h0038977
DA - 1964-03-01
UR - https://www.deepdyve.com/lp/american-psychological-association/type-i-and-type-ii-bias-jxGANCpTy8
SP - 198
EP - 199
VL - 19
IS - 3
DP - DeepDyve
ER -