Strong Inference

Platt, John R.

doi:10.1007/978-1-4899-0084-5_21

Cited by 423 publications

(593 citation statements)

References 11 publications

Supporting

Mentioning

578

Contrasting

Unclassified

Order By: Relevance

“…When E is a prediction of H, P(E | H) = 1. Thus, according to this theorem, when P(E) is large-close to 1-observation of E will have little effect on belief in H. "Strong inference" experiments (Platt, 1964)-where different theories make contradictory predictions-are a practical application of this idea. They embody the notion that the best evidence for a theory is evidence that would be otherwise unlikely.…”

Section: Why the Use Of Good Fits As Evidence Is Wrongmentioning

confidence: 99%

“…This approach was first sketched by Bacon (1620Bacon ( /1960Urbach, 1987) If alternative theories exist and make differing predictions (e.g., one theory says a certain measurement should be zero, another theory says it should be positive), we can combine the two approaches and test a prediction of the theory and a prediction of a competing theory at the same time. When the two predictions are incompatible (nonoverlapping), this is what Platt (1964) called strong inference. (Efficient inference might have been a better name.…”

Section: Does Philosophy Of Science Support the Use Of Good Fits As Ementioning

confidence: 99%

See 1 more Smart Citation

How persuasive is a good fit? A comment on theory testing.

Roberts

Pashler²

2000

Psychological Review

757

668

View full text Add to dashboard Cite

Quantitative theories with free parameters often gain credence when they closely fit data. This is a mistake. A good fit reveals nothing about the flexibility of the theory (how much it cannot fit), the variability of the data (how firmly the data rule out what the theory cannot fit), or the likelihood of other outcomes (perhaps the theory could have fit any plausible result), and a reader needs all 3 pieces of information to decide how much the fit should increase belief in the theory. The use of good fits as evidence is not supported by philosophers of science nor by the history of psychology; there seem to be no examples of a theory supported mainly by good fits that has led to demonstrable progress. A better way to test a theory with free parameters is to determine how the theory constrains possible outcomes (i.e., what it predicts), assess how firmly actual outcomes agree with those constraints, and determine if plausible alternative outcomes would have been inconsistent with the theory, allowing for the variability of the data.Many quantitative psychological theories with free parameters are supported mainly or entirely by demonstrations that they can fit data-that the parameters can be adjusted so that the output of the theory resembles actual results. The similarity is often shown by a graph with two functions: one labeled observed (or data) and the other labeled predicted (or theory or simulated). That the theory fits data is supposed to show that the theory should be taken seriously-should be published, for example. This type of argument is common; judging from a search of Psychological Abstracts , the research literature probably contains thousands of examples. Early instances involved sensory processes (Hecht, 1931) and animal learning (Hull, 1943), but this reasoning is now used in many areas. Here are three examples.1. Cohen, Dunbar, and McClelland (1990) proposed a parallel distributed processing model to explain the Stroop effect and related data. The model was meant to embody a "continuous" view of automaticity, in contrast to an "all-or-none" view (Cohen et al.,

show abstract

Section: Why the Use Of Good Fits As Evidence Is Wrongmentioning

confidence: 99%

Section: Does Philosophy Of Science Support the Use Of Good Fits As Ementioning

confidence: 99%

How persuasive is a good fit? A comment on theory testing.

Roberts

Pashler²

2000

Psychological Review

757

668

View full text Add to dashboard Cite

show abstract

“…In contrast, accounting appears to form tribes more on theory and method than topic. Prior research (Atkinson et al 1997;Wilson 1988;Pratt 1964) suggests that greater advances are made in fields that embrace many methods and theories to explain phenomena. Although the accounting tribes are based primarily on method and theory, citations across tribes are numerous.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

“…Scientific knowledge advances most rapidly when scholars draw upon multiple theories and diverse research methods (Wilson 1998;Atkinson et al 1997;Pratt 1964). This suggests that research in management control would be better served if accounting and management scholars draw upon each other's work.…”

Section: Introductionmentioning

confidence: 99%

A Social Network Analysis of the Literature on Management Control

Euske

Hesford

Malina

2011

Journal of Management Accounting Research

View full text Add to dashboard Cite

This paper investigates the literature on management control published in accounting and management journals. Social network analysis of citation data from the 25-year period 1981-2005 enables us to examine topics and ties among researchers. Social ties have important consequences for the development of the literature, shaping topics, research methods, and the diffusion of knowledge. We observe minimal communication between the two disciplines, appearing as two distinct communities despite similar interests. This lack of communication includes citations and authoring across the two disciplines. When citations across disciplines occur, it is almost exclusively accounting authors citing management authors, not vice versa. There is virtually no joining of accounting and management scholars within social networks. Within the two broader communities, there also exist smaller research clusters. While we cannot determine the impact this has on our understanding of management control, we discuss possible reasons for this phenomenon and its potential implications for management control research.

show abstract

“…First, there is the highest quality scientific evidence, based on considerations such as the excellence of the design, the validity and reliability of measures, the size and diversity of student samples, and similar considerations of internal (scientific rigor and soundness) and external validity (generalizability to different circumstances and students). The importance of scientific theory and hypothesis testing, especially the active search for disconfirmation, cannot be overstated in judging the quality of research (see Platt, 1964). Furthermore, scientifically supported theory provides the surest path to generalization (Lewin, 1951).…”

Section: Background: Categories Of Internal and External Validitymentioning

confidence: 99%