Additive conjoint measurement and the resistance toward falsifiability in psychology

Heene, Moritz

doi:10.3389/fpsyg.2013.00246

Cited by 25 publications

(20 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Edelsbrunner and Dablander (2018) could show that psychological modeling and scientific reasoning do not always follow a logical procedure. Heene (2013) describes in a very restrictive way why no approach, neither additive conjoint measurement nor modeling of structural equations or item-response theory, can solve the problem of measurement from a purely mathematical point of view and concludes that perhaps “human cognitive abilities and personality traits are simply not quantitative” (p. 3). Here, I would add the idea that cognitive abilities and personality traits might not solely be quantitative.…”

Section: Introductionmentioning

confidence: 99%

Should We Say Goodbye to Latent Constructs to Overcome Replication Crisis or Should We Take Into Account Epistemological Considerations?

Hanfstingl

2019

Front. Psychol.

View full text Add to dashboard Cite

This paper discusses theoretical and epistemological problems concerning validity of psychological science in the context of latent constructs. I consider the use of latent constructs as one reason for the replicability crisis. At the moment, there exist different constructs describing the same psychological phenomena side by side, and different psychological phenomena that are reflected by the same latent construct. Hagger called them déjà-variables, which lead to a decreasing validity of measurements and inhibit a deeper understanding of psychological phenomena. To overcome this problem, I suggest a shift of theoretical and epistemological perspective on latent constructs. One main point is the explicit consideration of latent constructs as mental representations, which change objects and are changed by objects via assimilative and accommodative processes. The explicit orientation toward assimilation and accommodation allows the control of normally automatized processes that influence our understanding of psychological phenomena and their corresponding latent constructs. I argue that assimilation and accommodation are part of our research practice anyway and cause the mentioned problems. For example, taking a measurement is an assimilative process, and thus a high measurement error should lead to an increase of accommodative processes. Taking into account these considerations, I suggest consequences for research practices, for individual researchers and for the philosophy of science.

show abstract

Section: Introductionmentioning

confidence: 99%

Should We Say Goodbye to Latent Constructs to Overcome Replication Crisis or Should We Take Into Account Epistemological Considerations?

Hanfstingl

2019

Front. Psychol.

View full text Add to dashboard Cite

show abstract

“…The weighted and unweighted proportion of violations are high when compared to the 2% for the unweighted and 1% for weighted violations for data simulated from a Rasch model and subjected to the same checks (Domingue (2014)). Heene (2013) has warned that such results would be obtained for many data sets that fit parametric IRT models, even when violations of its assumptions are present, so the result is not surprising. None of these standardized tests satisfies the conditions of an interval scale (i.e.…”

Section: Resultsmentioning

confidence: 99%

Measuring Unobserved Variables in Educational Inequality Research: Mathematics and Language 'Abilities' in Early Childhood

Sánchez¹

2019

Preprint

View full text Add to dashboard Cite

Evidence of social inequalities in cognitive abilities in early childhood has been documented in many societies; however, three characteristics of the data used to measure cognitive constructs make it difficult to quantify inequalities across groups. First, a causal understanding of validity is not compatible with the standard validation framework, which forces researchers to think critically what it means to measure unobserved constructs. Second, test scores only provide ordinal information about individuals, they are not interval scales and require the use of suitable corresponding methods for their study. Third, measurement invariance, which causes measurement error, may make comparison of test scores across groups invalid. The paper explores these three data problems applied to standardized tests---one mathematics and two language assessments---taken by a cohort of German children. The paper proposes a comparative validation framework for researchers based on nonparametric psychometric models and the representational theory of measurement. This framework can help researchers to determine if data fit the assumptions of a measurement model, to check for various forms of measurement error, and to overcome potential issues. A comparison of competing statistical modeling alternatives reveals substantial differences: By conceptualizing ability as ordinal instead of interval and excluding items that do not fit the assumptions of measurement models, I find a reduction in effect sizes for typical covariates studied in social stratification research.

show abstract

“…As Schönemann ([87], p. 151) put it: "...it also requires a willingness to accept empirical results which conflict with traditional beliefs." We doubt that psychologists are willing to pay the price of falsification of theories being based on the claim of quantitative measures (cf., [88,89]). Yet "solving" the problem by ignoring the possible non-quantitative properties of psychological phenomena is a pleasant self-delusion at the expense of falsifiability, resting on the dubious implicit assumption that knowledge is the "...result of 'processing' rather than discovery" ( [90], p. 259).…”

Section: Discussionmentioning

confidence: 99%

Detecting Violations of Unidimensionality by Order-Restricted Inference Methods

Heene

Kyngdon

Sckopke

2016

Front. Appl. Math. Stat.

Self Cite

View full text Add to dashboard Cite

The assumption of unidimensionality and quantitative measurement represents one of the key concepts underlying most of the commonly applied of item response models. The assumption of unidimensionality is frequently tested although most commonly applied methods have been shown having low power against violations of unidimensionality whereas the assumption of quantitative measurement remains in most of the cases only an (implicit) assumption. On the basis of a simulation study it is shown that order restricted inference methods within a Markov Chain Monte Carlo framework can successfully be used to test both assumptions.

show abstract

Additive conjoint measurement and the resistance toward falsifiability in psychology

Cited by 25 publications

References 47 publications

Should We Say Goodbye to Latent Constructs to Overcome Replication Crisis or Should We Take Into Account Epistemological Considerations?

Should We Say Goodbye to Latent Constructs to Overcome Replication Crisis or Should We Take Into Account Epistemological Considerations?

Measuring Unobserved Variables in Educational Inequality Research: Mathematics and Language 'Abilities' in Early Childhood

Detecting Violations of Unidimensionality by Order-Restricted Inference Methods

Contact Info

Product

Resources

About